JPH0361802A - Method and apparatus for observation - Google Patents

Abstract

PURPOSE:To use a plurality of lights different in wavelength to obtain correct position information of a wafer by a method wherein an auxiliary optical system using a transparent wedge member and four parallel planes is placed so as to align in position a pattern on a reticle with a pattern on the wafer on an optically equal plane. CONSTITUTION:A circuit pattern on a reticle 11 is illuminated by light of g-rays while a circuit pattern image is projected on a wafer 12 by a projection lens system 13. On the other hand, an alignment mark 37 on the wafer 12 is projected by polychromatic light from a light source 29 and formed in the vicinity of a reticle alignment mark on the reticle by the projection lens system 13 and an auxiliary optical system 31. The auxiliary optical system 31 comprises a transparent wedge member 1 and parallel planes 2, 3, 4A, 4B which are mutually placed in V-shapes, and this system aligns in position the pattern on the reticle 11 with the pattern on the wafer 12 in an optically equal plane thereby facilitating observation of both patterns and enabling highly accurate alignment of the reticle 11 and the wafer 12.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to an observation method and an observation apparatus.
Observation method and observation device suitable for obtaining wafer position information by observing alignment marks formed on the wafer surface through a projection lens before projecting a circuit pattern such as I onto the wafer surface using a projection lens Regarding.

[Prior art]

2. Description of the Related Art Conventionally, in a projection exposure apparatus, alignment marks formed on a wafer surface are observed through a projection optical system to obtain wafer position information.

In a conventional reduction projection exposure apparatus, a wafer alignment mark is observed through a projection lens of the apparatus in order to obtain wafer position information. In this type of observation method using a projection lens, monochromatic light with a wavelength that is the same as or close to the wavelength of the light used for exposure is used for observation, which eliminates chromatic aberration caused by the projection lens. This is preferable because it is not affected. However, the substrate surface of the wafer is usually covered with a resist layer of a predetermined thickness, and when monochromatic light is used to observe wafer alignment marks, light beams from the top and bottom surfaces (substrate surface) of the resist layer interfere with each other. I have trouble observing the alignment marks when I wake it up.

A method to solve this interference problem is disclosed in US Patent No.
, 4,355,892. In this US patent, to reduce the effects of interference in the resist layer,
The wafer is irradiated with two monochromatic lights with different wavelengths,
The wafer alignment mark is observed through a projection lens.

[Summary of the invention]

The present invention is an improvement on the observation method and observation device described in the above-mentioned US patent, and it is possible to accurately observe an object by using a plurality of lights with different wavelengths and a projection optical system. The purpose is to provide an observation method and an observation device.

To achieve this objective, the present observation method is a method of observing the second object through a projection optical system that projects a pattern of the first object onto the second object, and the method includes: illuminating with a plurality of lights (polychromatic light) having different wavelengths; and correcting principal rays of a plurality of imaging light beams having different wavelengths regarding the second object from the projection optical system so that they become parallel to each other; After the correction step, the method includes a step of observing an image of the second object formed by the plurality of imaging light beams.

Further, the present observation device is a device for observing the second object through a projection optical system that projects a pattern of the first object onto the second object, and the second object is observed by a plurality of lights having different wavelengths. and an image forming optical system that receives a plurality of imaging light beams having different wavelengths from the projection optical system and forms an image of the second object. , the image forming optical system includes a correction means for making principal rays of the plurality of imaging light beams parallel to each other.

Since the present invention includes the above-described correction stage and correction means, it is possible to correct differences in pupil aberration of the projection optical system depending on each wavelength. Therefore, even if the second object is displaced in the optical axis direction of the projection optical system and the image of the second object is defocused,
Since the images of the second object formed by the imaging light beams of each wavelength are shifted in the same way, the second object can be observed stably regardless of the displacement of the second object.

The plurality of lights having different wavelengths used in the present invention may be supplied by a white lamp having a continuous spectrum, or by a plurality of LEDs or lasers that emit light of different wavelengths, or may be supplied by a plurality of LEDs or lasers that emit light of different wavelengths. It is possible to supply it with a laser (Zeemann laser) that emits light of Furthermore, the observation method and observation apparatus of the present invention are suitable for a projection exposure apparatus for semiconductor manufacturing, and can be used to properly detect a wafer alignment mark of a wafer, which is a projection object onto which a mask pattern is projected, through a projection optical system. It can be observed. Therefore, it is possible to accurately obtain wafer position information and precisely align the wafer with respect to the mask.

Some features and specific configurations of the present invention will become clear from the examples shown below.

〔Example〕

FIG. 1 is an explanatory diagram for explaining the basic concept of the present invention. In FIG. 1, l is a wedge-shaped transparent member, 2 is a parallel plane plate made of high dispersion glass, 3 is a parallel plane plate made of low dispersion glass, and 4A and 4B are parallel plane plates made of glass having a predetermined dispersion value. show. Furthermore, M is a mutually different wavelength λ8. It shows an imaging light beam having (light of) λ2, which is a light beam emitted from a projection optical system that has been aberration-corrected for light with a wavelength λ3 (≠λ1≠λ2) (not shown). In addition, parallel plane plate 4
For reasons described later, A and 4B are made of the same glass material and have the same plate thickness.

In FIG. 1, principal rays LI0. The inclinations of L20 differ from each other due to differences in pupil aberration of the projection optical system. In this state, if the light beam with wavelength λ1 and the light beam with wavelength λ2 are focused to form an object image, the image forming position by each light beam will be different from each other, and the principal ray L 10 +L2o of each light beam with respect to each image plane will be different. The angle of incidence is also different.

Here, the chief ray L of the light beam with wavelength λ1 and the light beam with wavelength λ2 is
1° and L20 are parallel to each other, a wedge-shaped transparent member 1 is provided in the optical path of both light beams, and the wavelength λ1. The member 1 is constructed such that the refractive index n1°n2 and the apex angle 6 (the angle formed by the light entrance surface and the light exit surface) for light of λ2 satisfy the following formula.

= sin i 3 ...(1) Here, i,, i2 are the principal ray L10 of the luminous flux of wavelength λ1 and the principal ray L20 of the luminous flux of wavelength λ2, as shown in FIG.
is the incident angle with respect to the light incidence surface of the member 1.

Furthermore, by satisfying the above equation (1), the light beam with wavelength λ and the light beam with wavelength λ2 are emitted from the light exit surface of member l at the same angle i3, as shown in FIG.

Therefore, the chief rays LIOIL2G of both light beams are parallel to each other.

Wavelength λ1. The light beams of λ2 are incident on the plane-parallel plate 2 which is inclined within the meridional plane by an angle θ with respect to the chief ray (optical axis) of each light beam. Parallel plane plate 2
Since it is made of high dispersion glass, there is a large difference in refractive index due to the difference in wavelength of incident light. Therefore, as shown in FIG. 3(A), if the wavelength of the incident light differs, there will be a difference in the amount of lateral shift of the light beam due to the prism effect.

Taking advantage of this property, the plane-parallel plate 2 has wavelengths λ1 and λ1 whose chief rays from the wedge-shaped transparent member 1 are parallel to each other. The chief rays of the light beams of λ2 are made to substantially coincide on the light exit surface of the plane-parallel plate 2. At this time, the chief ray of each luminous flux is. Two symmetrical rays with equal angles to +IJ20, for example the ray L 11 shown in FIG.
There is an optical path length difference between and L12 and L21, resulting in comatic aberration. Therefore, the quality of the image formed on the image plane by each light beam deteriorates. Therefore, in order to correct the comatic aberration by correcting the optical path length difference between the light ray L1□ and the light ray L12 that occurred on the parallel plane plate 2, and the optical path length difference between the light ray L21 and the light ray 2, On the other hand, the plane-parallel plate 3 is provided so as to form a "V" shape and is inclined by an angle θ2 in the meridional plane with respect to the chief ray of each light beam. Since the plane-parallel plate 3 is made of low-dispersion glass, the difference in refractive index due to the difference in wavelength of incident light is minute. Therefore, as shown in Fig. 3(B), in the case of low-dispersion glass, the deviation of the chief ray for wavelengths λ1 and λ2 on a plane-parallel plate of the same length and the same angle of incidence is compared to that of high-dispersion glass. It is extremely small.

In this way, the inclination angle θ1 of the parallel plane plates 2 and 3. θ2, plate thickness,
By appropriately setting the refractive index and utilizing the difference in dispersion between the high-dispersion plane-parallel plate 2 and the low-dispersion plane-parallel plate 3, principal rays of wavelengths λ1 and λ2 are formed on the light exit surface of the plane-parallel plate 3. This makes it possible to remove comatic aberration while keeping the values consistent.

By the method described above, the ray aberration within the paper plane of FIG. 1, that is, within the meridional plane, is corrected. However, the rays in the zagital plane, which is perpendicular to this cross section, are
Due to aberrations caused by the light passing through the meridional plane, an image is formed at a location shifted from the image forming position of the light ray within the meridional plane. That is, astigmatism occurs on the image plane. Therefore, a pair of parallel plane plates 4A, 4. which are rotated by 90 degrees with respect to the parallel plane plate 2.3 with the optical axis (principal ray) as the rotation axis. B is provided in the optical path of each light beam. The parallel plane plates 4A and 4B are made of the same glass material and have the same thickness, and are incorporated in the optical path so as to form a dotted line. Moreover, the absolute value of the set angle of the parallel plane plate with respect to the optical axis is equal [7]. By adjusting the parameters such as the glass material and plate thickness of the parallel plane plates 4A and 4B, and the set angle with respect to the optical axis, the projection optical system member l and the parallel plane plates 2 and 3 can be adjusted.
Astigmatism that occurs on the image plane can be eliminated by this action. - At that time, each wavelength λ1. The difference in astigmatism at λ is the parallel plane plate 4A.

This can be corrected by selecting a 4B glass material. Also, at the image plane, the wavelength λ2. If chromatic aberration of magnification occurs between λ2, if an imaging optical system is installed behind this image plane to correct it and the image is re-formed, a polychromatic light image with very high image quality can be created. You can get it. In addition, minute coma aberration and each wavelength λ,.

When correcting the deviation of the chief ray of λ2, using a parallel plane plate made of high-dispersion glass and low-dispersion glass in place of the parallel plane plate 2゜3 allows you to obtain a polychromatic image with very good image quality. Obtainable.

FIG. 4 is a schematic diagram showing an embodiment in which the present invention is applied to a reduction projection exposure apparatus for semiconductor manufacturing. In the same figure, 1
1 is a reticle as the first object, and reticle stage 3
It is placed on 0. Reference numeral 12 denotes a wafer as a second object to which 1 nodist is applied. , 13 is a reduction projection one-lens system, which projects the circuit pattern drawn on the surface of the 13 lens systems of the reticle 11 onto the wafer 12 surface. This projection lens system 13 is constructed so that the wafer 12 side is de1nocentric and the 1noticle] side is non-telnocentric. 34 is the θ-2 stage and the wafer 12
The stage drive device 401
The rotation angle in the θ direction and the position in the Z direction of Uko and C 12 are adjusted based on instructions from the controller. θ-Z stage 34
is placed on an xy stage 32 for performing step movement of the wafer 12 with high precision.
2 moves in the X-Y plane, and like the θ-2 stage 34, it is driven based on commands from the stage drive device 401. A sphere 33 (mirror) is placed, and this optical sphere 33 is monitored by a laser interferometer 35.

14 is a bending optical system; 31 is an auxiliary optical system; 9 is a projection system; The two parallel plane plates 2.3 and 2.3 are arranged at the same angle to each other with the sagittal face direction orthogonal to the inclined plane ε, that is, the observation optical system 31 is arranged with respect to the parallel plane plates 2 and 3. It has two parallel plane plates 4A and 4B which are rotated 9071 degrees around the optical axis as a rotation axis.

18, 19, and 23 show that the bending mirror 22 is a correction I-nons system. Reference numeral 36 indicates the position where the image of the alignment mark 37 is formed by the projection optical system 13 and the observation optical system 31, and the alignment mark of the first cutter 11 is provided at this position. Scope (objective 1/ns), 26 is Be Mus. 3. 27 shows a lens for illumination, 29 is a white light source, and 39 is a light source that extracts light of several wavelengths from the white light source without exposing the resist on the wafer 12. 1 filter is shown. Reference numeral 28 denotes an imaging device consisting of a CCD, which is provided at a position conjugate to the circuit pattern imaging plane of the projection lens system 13. Reference numeral 402 does not indicate the controller electrically connected to the image pickup device R1128, the heather interferometer 35, and the stage drive device 401 by a signal line. , 4o3 is a lens that illuminates a part of the illumination optical system, and illuminates the circuit pattern of one block 11 with uniform illuminance with exposure light consisting of B rays emitted by an ultra-high pressure mercury lamp or the like. Note that the light source that supplies the exposure light is K
An l-noser light source such as an rF excimer laser, or a light source that emits i-rays may be used. The controller 402 receives position information regarding the wafer 12 from the imaging device and 1. via a signal line. - Receives the stage position information from the laser interference cutting 35 and inputs a signal to the stage drive device 401 to control the drive of the stages 32 and 34. Through such a procedure, the wafer 12 is aligned with the reticle 11.

The white light emitted from the white light source 29 enters the filter 39, and becomes an illumination light flux including a plurality of lights having different wavelengths, which does not expose the resist of Ueno\12 through the filter 39. This illumination light flux is transmitted through the beam splitter 26
, and illuminates the vicinity of the alignment mark on the reticle 11 through the alignment scope 25 and the bending mirror 24, passes through the reticle 11, and is directed to the auxiliary optical system 31. The light passes through the projection lens system 13 and is irradiated onto the wafer alignment mark 37 . The reflected light reflected by the wafer 12 follows the path (optical path) of the illumination light in the opposite direction to the beam splitter 26, passes through the beam splitter 26, and is directed toward the imaging device 28, whereupon the reticle 11 and the wafer are placed on the imaging device 28. 11
Connect the 2 statues. The positional relationship between the reticle 11 and the alignment mark on the wafer 12 is then observed using the imaging device 28.

In this embodiment, the circuit pattern on the reticle surface 11 is
λ=436 nm), and a circuit pattern image is projected onto the wafer 12 by a projection lens system 13. On the other hand, the alignment mark 37 on the wafer 12 is irradiated with polychromatic light from the light source 29, and an image is formed near the reticle alignment mark on the reticle 11 by the projection lens system 13 and the auxiliary optical system 31. Then, both alignment marks are observed simultaneously using the alignment scope 25 and the imaging device 28. Although it is possible to make the auxiliary optical system 31 movable in the apparatus of this embodiment and move it according to the position of the wafer alignment mark, the following description will be made assuming that the position of the auxiliary optical system 31 is fixed for the sake of simplicity.

Although the projection lens system 13 is well corrected for aberrations at the wavelength corresponding to the g-line for projection exposure of circuit patterns,
Aberrations are not sufficiently corrected at the wavelength of light for observing alignment marks. In particular, various aberrations due to color differences, such as Hosotsu chromatic aberration, lateral chromatic aberration, chromatic spherical aberration,
Many chromatic comatic aberrations, chromatic astigmatisms, etc. remain.

For this reason, when considering the wafer surface as an object surface, the wafer 12
The upper alignment mark 37 is imaged above the reticle 11 because the wavelength of the observation light is often longer than the wavelength of the exposure light.

For example, the pattern projection magnification of the projection lens system 13 is 115.
When the magnification is multiplied, if the Hosochromatic aberration on the wafer 12 side is 300 μm, the image of the wafer 12 on the reticle 11 side is
The image is formed above the reticle 11 by 0.3×5′=7.5 (mm).

For this reason, it becomes difficult to simultaneously observe patterns such as alignment marks on the reticle 11 and patterns such as alignment marks on the wafer 12. Therefore, as shown in the above-mentioned US patent, for example, various auxiliary optical systems have been arranged between the reticle and the projection lens system to make the pattern images of the two coincide with each other for correction. however,
It is difficult to completely correct aberrations using conventional auxiliary optical systems, and it is difficult to observe both patterns well.

In this embodiment, aberrations were well corrected between the reticle 11 and the projection lens system 13 not only in the sagittal plane (direction) but also in both directions in the meridional plane (direction). In particular, by arranging an auxiliary optical system 31 that can satisfactorily correct various aberrations due to color over all the wavelengths used, the positions of the pattern on the reticle 1 and the pattern on the wafer 12 are optically matched within the same plane. This allows for good observation of both patterns and enables highly accurate alignment of the reticle and wafer.

The auxiliary optical system in this embodiment has a wedge shape arranged as shown in FIG. It is characterized by correcting pupil aberration (chromatic spherical aberration), coma aberration, and astigmatism using a transparent member 1 and four parallel plane plates 2, 3, 4A, and 4B.

Among these, each wavelength (λ5... The spherical aberration of the pupil color with respect to λ2) is corrected, and the principal rays of light of each wavelength are made parallel to each other.

Next, the chromatic aberration of magnification for each wavelength of the observation light of the projection lens system 13 is corrected by the parallel plane plate 2, which is arranged obliquely in the meridional plane with respect to the imaging light beam in the meridional plane. The principal ray of light is made to coincide with the optical axis ε of the auxiliary optical system 31. At this time, the inclination angle of the plane-parallel plate 2 with respect to the optical axis is determined depending on the amount of aberration generated in the projection lens system 13 and the refractive index, dispersion, and thickness of the plane-parallel plate 2. Although the wedge-shaped transparent member 1 and the parallel plane plate 2 are effective against clear color spherical aberration and lateral chromatic aberration of the projection lens system 13, the first
As explained in the figure, this causes coma aberration to occur in the imaging light beam. Therefore, in this embodiment, in the meridional plane, a parallel plane plate 3, which is arranged at an angle with respect to the optical axis in the same way as the parallel plane plate 2ε, is provided in the optical path of the imaging light beam, and is parallel to the projection lens system 13 and the wedge-shaped transparent member l. Comatic aberration for each wavelength of the observation light of the system consisting of the flat plate 2 is corrected.

At this time, the angle of inclination is determined mainly depending on the amount of aberration generated by the projection lens system 13 and the thickness of the parallel plane plate 3. This single parallel plane plate 3 is effective against coma aberration, but on the other hand, it causes astigmatism. Astigmatism caused by the parallel plane plate 3 and the projection lens system 13
The sum of the astigmatism of the wedge-shaped transparent member 1 and the astigmatism at the wavelength of the observation light of the plane-parallel plate 2 becomes the astigmatism of the system up to the plane-parallel plate 3.

Therefore, in this embodiment, the two parallel plane plates 4A and 4B are arranged so as to be inclined to each other in a plane (sagittal plane) perpendicular to the plane in which the parallel plane plates 2 and 3 are inclined with respect to the optical axis. Corrects astigmatism. That is, the parallel plane plate 2,
Two parallel plane plates 4A and 4B are arranged within a plane in which the optical axis of the optical axis of the auxiliary optical system 31 is rotated by 90 degrees.

When the parallel plane plates 4A and 4B have the same thickness, they may be arranged in a line-symmetrical relationship with respect to a predetermined line perpendicular to the optical axis, and when they have different thicknesses, they may be arranged in a line-symmetrical relationship with respect to a predetermined line orthogonal to the optical axis. You can place it at an angle. and two parallel plane plates 4
Comatic aberration is prevented from occurring in A and 4B as a whole. The parallel plane plates 4A, 4B are adjusted so that the astigmatism of the parallel plane plates 4A, 4B cancels out the astigmatism caused by the system before the parallel plane plate 3. For example, if the projection lens system 13 generates only comatic aberration and no astigmatism at the wavelength of observation light, the optical path length of the two parallel plane plates 4A and 4B is set to that of the parallel plane plates 2 and 3. The auxiliary optical system 31 is set to approximately 1/2 of the sum of the optical path lengths, and even if the parallel plane plate 2.3 and the parallel plane plates 4A and 4B are twisted with respect to each other, the auxiliary optical system 31
2.3.4 Parallel plane plates with four angles to the optical axis of
If A and 4B are all made equal, the alignment mark can be observed with the coma and astigmatism of the projection lens system 13 well corrected.

In addition, if astigmatism occurs in the projection lens system 13 with respect to the wavelength of the observation light, the angle that the parallel plane plates 2 and 3 and the two parallel plane plates 4A and 4B make with the optical axis is By making them different from each other according to the amount of astigmatism generated in the system before the parallel plane plate 3, it becomes possible to observe an alignment mark in which astigmatism generated in the projection lens system 13 is also corrected. That is, in this embodiment, by adjusting the inclinations of the parallel plane plates 4A and 4B, it is possible to arbitrarily change the amount of correction of astigmatism.

In this embodiment, the projection lens system 13 is effectively corrected by properly correcting coma aberration and astigmatism using the above-described configuration.
Not only in the sagittal plane (direction) but also in the meridional plane (
The alignment marks on both the reticle 11 and the wafer 12 are
At the same time, it is possible to observe it as a clear image. This enables highly accurate alignment.

In this embodiment, if some spherical aberration remains at the wavelength of the observation light from the projection lens system 13, it is preferable to correct it using the correction lens section 22.

In this case, the spherical aberration at the wavelength of the observation light from the projection lens system 13 is preferably corrected by once generating an opposite spherical aberration in the correction lens section 22. Further, the axial chromatic aberration of the projection lens system 13 can be corrected by the correction lens section 22. The lateral chromatic aberration of the projection lens is corrected by a unit including a parallel plane plate, as described above.

Further, the correction lens section 22 is disposed on the reticle II side, and has a relatively small N (for example, 0.1 or less).
In case A is used, for example, when there is a large spherical aberration of several λ, an aspherical member is installed as part of the correction lens section 22 and at a position approximately coextensive with the pupil position of the projection lens system 13. It is also possible to arrange and correct. For example, if spherical aberration occurs that is insufficiently corrected on the long wavelength side, an aspherical lens with a shape in which the negative refractive power increases toward the periphery of the lens may be used.

Note that the insertion of the auxiliary optical system 31 in this embodiment changes the conditions during alignment mark observation and pattern projection exposure, so the adjustment of the optical system as a baseline or a factor that is processed as an offset. Contains. Aberrations that can be treated by adjusting the optical system include color-based defocus (axial chromatic aberration) and field curvature, since the focus can be adjusted by adjusting the optical path length of the observation optical system. Furthermore, the shift of the image point due to the insertion of the auxiliary optical system 31, the color distortion of the projection lens system 13, etc. can be optically or electrically processed as offsets of positional deviations.

In short, if the problem is simply refocusing or a simple shift in the image position, it can be easily offset, so it is not a big problem.

In reality, when detecting an image, problems arise from spherical aberration, coma aberration, and astigmatism that impair the contrast of the image.As mentioned above, these aberrations of the projection lens system 13 are caused by the exposure light. Although the wavelength is well corrected, some wavelengths of observation light are not necessarily well corrected. However, the way these various aberrations occur at the wavelength of observation light can be treated in the area of basic third-order aberrations as a simple deviation from good aberration correction at the wavelength of exposure light. As the results of the analysis have revealed, by using the auxiliary optical system 31 having the above-mentioned configuration, it is possible to perform good aberration correction and to achieve an observation apparatus capable of observing a clear alignment mark image.

With the above configuration, in this embodiment, the projection lens system 1
The state on the wafer 12 can be clearly observed through the wafer 3.

At this time, in this embodiment, the auxiliary optical system 31 and the three mirrors 1
8, 19.23, the pattern such as the alignment mark 37 on the wafer 1 will be observed in an inverted state, but as with the offset described above, this is done by photoelectric conversion by the imaging device 28 and then signal processing. By inverting the sign, it can be observed without any problem.

In addition, the correction lens unit 22 of this embodiment not only functions to image the pattern on the wafer 12 onto the reticle 11, but also functions to image the pattern on the wafer 12 onto the reticle
2 onto the reticle surface at a predetermined magnification. For example, when using a projection optical system with a projection magnification of 5 times, the projection magnification is set to be exactly 5 times (in this case, the imaging magnification of the correction lens unit 22 itself is 11 times), thereby imposing a load on the subsequent processing device. This reduces the amount of Although the function of the correction lens section 22 is made to have an 11 times the imaging effect, it is also possible to make it have a 1 times the imaging effect.

In this embodiment, the auxiliary optical system 31 corrects pupil aberration, coma aberration, astigmatism, and spherical aberration among various chromatic aberrations mainly generated by the projection lens system 13.
The deviation of the image plane is corrected by adjusting the optical path length, and the magnification and distortion are generally corrected by offset processing. This allows for good observation of both the reticle and wafer, and enables highly accurate alignment.

In this embodiment, the aberrations of the optical system are well corrected not only in the sagittal plane (direction) as in the conventional case but also in all directions including the meridional plane (direction). By performing observation, two signals (wafer position information) in the X and Y directions can be detected. To perform two-dimensional alignment, at least one more point must be observed. This makes it necessary to match the θ direction. This can be done using only one observation system as shown in Fig. 4, but as shown in Fig. 5, two observation systems shown in Fig. 4 are arranged,
A pair of observation systems (24~29.39.24'~29'39
By observing the pair of alignment marks at ('), it is possible to align the reticle 11 and wafer 12 in the x, y, and θ directions while maintaining throughput.

Furthermore, the reticle alignment mark and the wafer alignment mark 37 can be observed without observing the reticle 11 and the wafer 12 at the same time.Next, an example will be described.

FIG. 6 is a schematic view showing a second embodiment in which the present invention is applied to a reduction projection exposure apparatus for semiconductor manufacturing, in which the same members ε as shown in FIG. A feature of this embodiment is that a reticle alignment optical system R and a wafer alignment optical system W are provided separately.

In the figure, 14 is a folding mirror, 1 is a wedge-shaped transparent member, 2.3.4A and 4B are parallel plane plates, and members 1 and 2
．． 3.4A, 4B, and 14 constitute the auxiliary optical system 317. The basic arrangement of each member of the auxiliary optical system 31 is the same as that shown in FIG. 4, and the explanation will be omitted here. Omitted.

40 is objective 1 nons, 41 is beam splitter 142
Reference numeral 43 indicates an imaging device consisting of a relay 1 non-sensor and a CCD, and an image of the alignment mark 37 of the Ueno X12 is formed on (the light-receiving surface of) the imaging device 43. Illumination light consisting of several lights having mutually different wavelengths for forming an image of the alignment mark 37 and observing the alignment mark 37 is transmitted from a white light source 46 to an illumination condenser l/
Scum 45, filter 44 that allows only non-sensitized light to pass through
through the beam brick-4], and the auxiliary optical system 31
The mark 48 (hereinafter referred to as "reference mark 48"), which serves as a reference for detecting the position of the alignment mark on the Eva 12, is white. The reference mark is illuminated with light from the light source 50 via the condenser lens 49 for illuminating the fiducial mark. The light from the reference mark 48 passes through the objective lens 47 and enters the beam splitter 41, where it is combined with the optical path of the reflected light (polychromatic imaging light flux) from the wafer 12, and is combined with the optical path of the reflected light from the wafer 12 (polychromatic imaging light flux). A reference mark 48 is imaged onto the imaging device 43 by directing the reference mark 48 toward the imaging device 43 .

The construction of the reticle alignment optical system R for setting the reticle 11 in the main body of the exposure apparatus is as follows. Illumination light having a non-exposure wavelength is emitted from the fiber 51, and enters the prism 52 to illuminate the reticle reference mark 53 fixed to the main body and the reticle alignment mark of one notch 11. The direction of the optical path of the light from these marks is changed by a mirror 54, and the light passes through an objective lens 55 and a relay lens 56 to the imaging device iff.
(CCD) 57 , and one notch alignment mark images the reference mark 53 on the imaging device 57 .

In order to supply light to the fiber 51, it is sufficient to 1. introduce a part of the light emitted by an ultra-high pressure mercury lamp or the like for projection exposure of the noticle pattern into the fiber 51;

First, the reticle 11 is set in the exposure apparatus main body using the above alignment optical system. Reticle 11 and reference mark 5
3 is illuminated with light from a fiber 51, and an imaging device 57
Images of both marks are formed thereon, and the positional deviation imprinted on the main body of the reticle 11 is calculated based on the positional relationship between the images of both marks. Based on the results, reticle stage 3
0 is driven by a drive device (not shown), and the 1 notchle alignment mark and the reticle reference mark 53 are aligned. When this alignment is performed, the center of the reticle and the optical axis AX of the projection lens system 13 coincide, and the alignment between the reticle 11 and the main body of the exposure apparatus is completed.

Next, the alignment of the reference mark 48 to the wafer alignment mark 37 will be described.

The reference mark 48 is illuminated with illumination light from a light source 50 and a lens 49, and is imaged on the imaging device 43 via the optical system (47, 41, 42). Here, an arrangement is made so that the position of the reference mark 48 and the optical system (47, 41, 4, 2) does not change over time.

The wafer alignment mark 37 on the wafer 12 is illuminated with polychromatic illumination light from the white light source 46 and the optical system (45, 44, 41, 31, 13), and the image is aberration-corrected by the auxiliary optical system 31. It is formed on the imaging device 43. The wavelength range of the polychromatic illumination light can be determined by the filter 44. Therefore, the state of the resist of the wafer 11 (
By optimizing the wavelength range of the illumination light for observation by replacing the filter 44 depending on the thickness (thickness), it is possible to always obtain a clear wafer alignment mark image without being affected by interference from the resist on the wafer 11. can.

At this time, if each wavelength of the multicolor illumination light for wafer alignment mark illumination is set to a wavelength that does not expose the resist, unnecessary patterns will be printed on the resist of the wafer 11 when observing the wafer alignment mark 37. Therefore, the same alignment mark can be used in each process. Also, even if the material absorbs most of the exposure light, such as a multilayer resist, the alignment mark 37
observation becomes possible.

Now, the images of the reference mark 48 and the wafer alignment mark 37 formed on the imaging device 43 are converted into video signals by the imaging device 43, and the positions of the respective images on the imaging device are determined by signal processing. Then, the position of the wafer 12 with respect to the virtual position of the reference mark 48 on the wafer 12 is detected from the positional relationship between the two. In this example, since the wafer alignment mark observation position and the exposure position of the wafer 12 are different, the wafer 12 is detected at a distance (baseline) from the known deviation observation position to the exposure position.
The XY stage 32 is driven by the stage drive device 401 by the amount of positional deviation (both x and y directions) of 2 added, the wafer 12 is aligned with the reticle 11, and the pattern is projected and exposed.

In this embodiment as well, a pair of reticle alignment optical systems R and a pair of wafer alignment marks are provided at different positions on the wafer and two reticle alignment marks are provided at different positions on the reticle. By arranging the alignment optical system W, it is possible to improve the accuracy and processing speed in alignment between the reticle and the wafer.

The projection lens system of the embodiments described above is a telecentric system only on the wafer 12 side, and the light source for observing the wafer alignment mark on the wafer 12 is also an incoherent light source consisting of a white light source. The present invention can also be implemented in an apparatus equipped with a telecentric projection optical system on both sides of the wafer, and an apparatus equipped with a light source that emits coherent light for observing wafer alignment marks.

FIG. 7 is a schematic diagram showing a third embodiment in which the present invention is applied to a reduction projection exposure apparatus for semiconductor manufacturing, and the same members as those shown in FIG. 6 are denoted by the same reference numerals. However, the reduction projection lens system 13 of the projection exposure apparatus of this embodiment is a telecentric system on both the reticle 11 and wafer 12 sides, and is composed of a lens system 60.degree. 61. Furthermore, a stage reference mark 60 is fixed on the XY stage 34, and serves as a reference for correcting changes over time in the baseline value used when moving the XY stage 34 from the observation position to the exposure position. It is something.

14 is a folding mirror, 1 is a wedge-shaped transparent member, 2°3.4
A, 4B indicate parallel plane plates, members 1.2.3°4A,
4B, 14 is a wedge 15 and a parallel plane plate 16. 1
7 constitute an auxiliary optical system 31 by mutually arranging them within the meridional cross section of the projection optical system.

The basic arrangement of each member of the auxiliary optical system 31 is the same as that shown in FIG. 4, so a description thereof will be omitted here.

40 is an objective lens, 41 is a beam splitter, 142 is a relay lens, 62 is a beam splitter, 163 is an erector, and 164 is a CCD.

A system for supplying illumination light for observing the wafer alignment mark 37 on the wafer 12 is configured as follows. The light source is composed of three lasers 70.71.72 which each supply laser beams with different wavelengths.

The optical paths of the laser beams from each laser 70, 71, 72 are made to coincide with each other by a bending mirror 69 and beam splitters 67, 68.

66 is a condenser lens for illumination, 65 is a diffuser plate,
The diffuser plate 65 is vibrated or rotated by a drive mechanism (not shown) to average out the speckle pattern generated on the wafer 12 because the light source is a laser, and
illuminate evenly. The polychromatic laser beam that has passed through the diffuser plate 65 is sent to the beam splitter 41, the auxiliary optical system 31. It is directed to the wafer 12 via a projection lens system 13 .

In order to efficiently use the laser beams from the lasers 70, 71, and 72, a gicroic mirror or a polarizing beam splitter is used as the beam splitters 67 and 68 used to overlap each other's optical paths.

For example, if the beam splitter 67 is configured with a gicchroic mirror, the laser beam with the wavelength λ1 from the laser 70 is transmitted, and the laser beam with the wavelength λ2 (≠λ,) from the laser 71 is transmitted.
wavelength λ3 (≠λ2≠
λ, ) is reflected, and the beam splitter 68 is configured with a polarizing beam splitter, so that the laser beam from the laser 71 is directed to the beam splitter 68 as S-polarized light, and from the laser 72. so that the laser beam becomes P-polarized ε and directed to the beam splitter 68! No Surf 1.72, mirror 69, and beam splitter 68 are arranged.

A reference mark 48, which serves as a reference for detecting the positions of the twelve wafer alignment marks 37, is illuminated by a white light source 50 and an illumination condenser lens 49, and the light from the reference mark 48 is transmitted through a beam splitter 62 to an erector. 63, and the image pickup device 6 is captured by the erector 63.
4, an image of the fiducial mark 48 is formed on the imaging device 64.

On the other hand, reflected light from the wafer 12 illuminated with polychromatic laser light enters the objective lens 40 via the projection lens system 13 and the auxiliary optical system 31. Then, by the action of the auxiliary optical system 31, pupil aberration and coma astigmatism are corrected. The light from the objective lens 40 is transmitted through a beam splitter 41, a relay lens 42, a beam splitter 62, and an erector 63.
to form an image of the wafer alignment mark 37 on the imaging device 64 .

The image of the reference mark 48 and the image of the wafer alignment mark 37 formed on the imaging device 64 are converted into video signals,
The operation of detecting the positional relationship between both marks is the same as in the previous embodiment. Further, when forming an image of the reference mark 48 and an image of the wafer alignment mark 37 on the imaging device 64, the images of both marks 48.37 may be formed simultaneously, or the images of each mark 48.37 may be formed sequentially. You may do so.

The configuration of the reticle alignment optical systems R, , R2 for setting the reticle 11 in the exposure apparatus main body is as follows. Each illumination light having a different wavelength from the exposure light emitted from the fiber 51, 51' is reflected by the prism 52, 52' toward the reticle ll side, and is reflected by the reticle reference mark 5.
3. Illuminate the 53' and reticle alignment marks.

The light from the reticle alignment mark corresponding to the reticle reference mark 53 and the reticle alignment mark corresponding to the reticle reference mark 53' is transmitted to the bending mirror 54.
54', objective lens 55.55', beam splitter 74, C via 74' relay lens 56.56'
It is directed toward an imaging device 57, 57' located further away from the CD. Then, an image of the reticle reference mark 53 and one reticle alignment mark, and an image of the reticle reference mark 53' and the other reticle alignment mark are formed on the imaging devices 57 and 57', respectively. Using these images, reticle 11
The operation for positioning is the same as in the previous embodiment. Further, in this embodiment, the pattern forming surface of the reticle 11 is located at the focal position of the objective lens 55, 55', and the light receiving surface of the imaging devices 57, 57' is located at the focal position of the relay lens 56, 56'. I have to.

Therefore, between the objective lens 55, 55' and the relay lens 56, 56', the imaging light beam forming the images of the reticle alignment mark and the reference mark becomes a parallel light beam. Then, the optical systems R1 and R2 are moved so that the mirrors 54 and 54' and the objective lenses 55 and 55' move together as one body as shown by the broken line.
A separate illumination source 77, 77' shutter 76.7
6' and filters 75 and 75' are provided.

At this time, members 54 (54') and 55 (55'
), 74 (74').

75 (75'), 76 (76'), 7
7 (77') and a member 54 (54')
), 55 (55'), 74 (74').

56 (56') and 57 (57'), TTL (Through The Le
ns) wafer alignment mark observation system will be constructed. That is, the light sources 77 and 77' supply a light beam containing light of the same wavelength as the exposure light, and the objective lens 55
．． 55' and folding mirror 54.54' to the reticle 12
By moving the wafer alignment mark 37 to a predetermined position where it can be observed, the projection lens system 13
and the wafer alignment mark 37 via the reticle 11
can be observed. Of course, other reticle alignment marks may be provided at this predetermined location so that the reticle alignment marks can also be observed at the same time.

Note that the filters 75 and 75' are for extracting only light having the same wavelength as the exposure light from the light sources 77 and 77'.
When the light sources 77, 77' are lasers or the like that emit laser light with the same wavelength as the exposure light or with a wavelength very close to the wavelength of the exposure light, the filter 7
5. It is preferable to provide a movable diffuser plate in place of 75' to average speckles when illuminating the wafer with laser light.

Although not shown in FIG. 7, when a plurality of lasers 70 to 72 having different oscillation wavelengths are used as an observation light source as in this embodiment, the optical axis is located immediately after the light exit of each laser 70 to 72. It is convenient to provide a polarizing plate that can be rotated about a rotation axis and an ND filter that has variable density so that the intensity of the laser light supplied by each of the lasers 70 to 72 can be modulated. With this configuration, each laser 70-7
It is possible to make the intensity of each laser beam almost uniform without selecting lasers so that the emission intensity of each laser beam is the same.
The observation conditions for the wafer alignment mark 37 can be changed by making the intensity of the laser beam of a predetermined wavelength weaker than the intensity of the laser beam of other wavelengths.

In each of the embodiments described above, the auxiliary optical system 31 includes the wedge-shaped transparent member 11, the parallel plane plates 2 and 3, and the parallel plane plate 4A.

4B, but in the present invention, by providing at least the wedge-shaped transparent member l, the chief rays of the imaging light beams of each wavelength are made parallel to each other, thereby improving observation performance.
Other members 2.3.4A, 4B are not necessarily required.

FIG. 8 is a schematic diagram showing a fourth embodiment in which the present invention is applied to a reduction projection exposure apparatus for semiconductor manufacturing, and is a modification of the second embodiment shown in FIG.

The only difference between this embodiment and the second embodiment is the configuration of the auxiliary optical system 31, and the other configurations are completely the same.

Therefore, in FIG. 8, the same reference numerals as in FIG. 6 are used for the same members as in FIG. 6.

In the second embodiment, the auxiliary optical system 31 was equipped with the wedge-shaped transparent member 1 and the parallel plane plates 2.3.4A and 4B, but the auxiliary optical system 31 of the present embodiment includes the wedge-shaped transparent member 1 and the parallel plane plate 4A.
, 4B, but not the parallel plane plate 2.3.

In this embodiment, although the wavelengths are different from each other due to the filter 44, the difference is relatively small. λ2 is extracted, and the wedge-shaped transparent member 1 assists in making the principal rays of the imaging light beams having different wavelengths from the projection lens system 13 parallel to each other so that the optical paths of the principal rays almost coincide. It consists of an optical system 31. Therefore, in the second embodiment, the plane-parallel plate 2 used to match the chief rays of the imaging light beams having different wavelengths and the plane-parallel plate 3 used to correct the coma that mainly occurs in the plane-parallel plate 2. can be eliminated, and the configuration of the auxiliary optical system 31 is simplified.

In this embodiment, in order to improve the aberration correction function of the auxiliary optical system 31, the wedge-shaped transparent member 1 is an achromatic member made by laminating a wedge-shaped member with high dispersion on one side and a wedge-shaped member with low dispersion on the other side. Just do it.

As described above, in the embodiments shown in FIGS. 4 to 8, the auxiliary optical system 31 and the projection lens system 13 are used to illuminate the wafer 12 with a plurality of lights having different wavelengths. It is also possible to illuminate the wafer 12 without using it. For example, the alignment mark 37 on the wafer 12 can be illuminated by directing a laser beam onto the wafer 12 at a predetermined angle of incidence from the space between the projection lens system 13 and the XY stage 34 . At this time, the mark image formed on the imaging device via the projection lens system 13 and the auxiliary optical system 31 is mainly due to the diffracted light generated at the edge of the wafer alignment mark 37.

Furthermore, when forming images of the wafer alignment mark on the imaging device using light having different wavelengths as in the above embodiment, instead of simultaneously forming images using each light, it is possible to sequentially form wafer alignment mark images using light having different wavelengths. It can also be formed on the imaging device. When such an image forming method is carried out using, for example, the apparatus shown in FIG. 7, lasers 70, 71 . '72 is sequentially oscillated to sequentially form wafer alignment mark images of three colors (wavelengths) different from each other on the imaging device 64, and each video signal of the three color images is captured and the wafer alignment mark image of each video signal is The amount of positional deviation (position data) is detected. Then, based on a plurality of pieces of position data obtained from the video signals of these three rice species (for example, by finding the average value of each data), the position information (positional shift amount) of the wafer is found. Furthermore, when obtaining wafer position information, it is not always necessary to use position data based on images of all colors; instead, it is necessary to identify data with abnormal values that are significantly different from other position data, and use this abnormal value data. It is better not to use the data in the process to obtain the wafer position information, and only use the remaining data to obtain the position information.

Although a projection exposure apparatus for semiconductor manufacturing has been exemplified as an apparatus to which the present invention is applied, the present invention is applicable not only to this type of projection apparatus but also to various other projection apparatuses. Furthermore, even when applied to a projection exposure apparatus for semiconductor manufacturing, as shown in each of the above embodiments, the wafer alignment mark can be observed and In addition to obtaining positional information, it is also possible to obtain positional information regarding the optical axis direction of the projection lens system of the wafer by observing a predetermined mark formed on the wafer. At this time, the contrast of the mark image is detected.

〔Effect of the invention〕

As described above, in the present invention, when an object such as a wafer is illuminated with a plurality of lights (polychromatic lights) having different wavelengths and is observed through a projection optical system, a plurality of imaging light beams having different wavelengths are emitted from the projection optical system. Since the chromatic aberration of the pupil of the projection optical system is corrected even if the principal rays of the images are made parallel to each other, stable observation of an object such as a wafer can be performed regardless of the displacement of the projection optical system in the optical axis direction.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram showing the basic concept of the present invention. FIG. 2 is an explanatory diagram showing the action of the wedge-shaped transparent member. FIGS. 3(A) and 3(B) are explanatory diagrams showing the action of a pair of parallel flat plates tilted with respect to the meridional in-plane direction. FIG. 4 is a schematic diagram showing a first embodiment in which the present invention is applied to a projection exposure apparatus for semiconductor manufacturing. FIG. 5 is a partial schematic diagram showing a modification of the device shown in FIG. 4. FIG. 6 is a schematic diagram showing a second embodiment in which the present invention is applied to a projection exposure apparatus for semiconductor manufacturing. FIG. 7 is a schematic diagram showing a third embodiment in which the present invention is applied to a projection exposure apparatus for semiconductor manufacturing. FIG. 8 is a schematic diagram showing a fourth embodiment in which the present invention is applied to a projection exposure apparatus for semiconductor manufacturing.

Claims (6)

[Claims] (1) A method of observing the second object through a projection optical system that projects a pattern of the first object onto the second object,
illuminating the second object with a plurality of lights having different wavelengths; and correcting principal rays of a plurality of imaging light beams having different wavelengths from the projection optical system regarding the second object so that they become parallel to each other. and, after the correction step, observing an image of the second object formed by the plurality of imaging light beams. (2) A device for observing the second object through a projection optical system that projects a pattern of the first object onto the second object,
an illumination unit for illuminating the second object with a plurality of lights having different wavelengths;
An observation device comprising an image forming optical system that forms an image of an object, the image forming optical system having a correction means for making principal rays of the plurality of imaging light beams parallel to each other. (3) The observation device according to claim (2), wherein the correction means comprises a wedge-shaped transparent member. (4) The illumination means illuminates the second object with a plurality of lights having different wavelengths from the wavelength of the light used to project the pattern on the first object. Observation device described in section 3). (5) The image forming optical system receives the plurality of imaging light beams from the wedge-shaped transparent member, and the meridional surface of the projection optical system is configured to cause principal rays of the plurality of imaging light beams to substantially coincide with each other. A first parallel flat plate tilted with respect to the optical axis within the meridional plane and a second parallel flat plate tilted with respect to the optical axis within the meridional plane to correct comatic aberration occurring in the first parallel flat plate. Observation device according to scope (3). (6) The image forming optical system is arranged such that the sagittal surface of the projection optical system is configured to correct astigmatism occurring in the system consisting of the projection optical system, the wedge-shaped transparent member, and the first and second parallel flat plates. An observation device that has a pair of parallel flat plates tilted at the same angle in opposite directions with respect to the optical axis.