JPS63175421A - Alignment apparatus - Google Patents

Abstract

PURPOSE:To reduce an optical amount loss and to improve an alignment accuracy of an alignment apparatus by disposing a mirror at the pupil of an eye in a projection optical system, and introducing an illumination light for an alignment. CONSTITUTION:Illumination lights LA, LB are reflected by optical reflection means 40 provided at the pupil of an eye in a projection optical system to be irradiated on a substrate W. Reflected lights from marks WA, WB on the substrate are incident through marks RA, RB on a mask R to signal detecting means, which obtains photoelectric signals regarding the respective marks. In this case, the marks RA, RB on the mark R are merely transmitted and illuminated by the reflected light from the substrate W, and the mask marks RA, RB are detected only by the transmitted and illuminated image. Since the photoelectric signal obtained in this manner is obtained without necessity of optically dividing means, the loss of the illuminated light is reduced, thereby accurately aligning with the photoelectric signal.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an exposure apparatus used, for example, in the manufacture of integrated circuits, and particularly relates to a mask on which a pattern to be projected is formed and a semiconductor wafer. The present invention relates to an improvement of an alignment device that performs alignment with a projection target substrate.

[Prior Art] As a conventional alignment device used in an exposure device or the like, there is one shown in FIG. 13, for example.

In this figure, a double-sided telecentric projection lens 10 is arranged below a reticle R on which a circuit pattern to be exposed and projected is formed, and further below it a wafer W as a substrate to be projected is placed at a conjugate position. It is located.

The wafer W is placed on a stage 12 that is movable at least in the illustrated X and Y directions, and the movement of the stage 12 aligns the reticle R and the wafer W.

Marks RM%WM for alignment are formed on the reticle R and the wafer W, respectively.

Next, the illumination light outputted from the alignment light source 14 passes through the objective lens 16, is converted into a parallel beam, and enters a half mirror 18 arranged above the reticle R. The illumination light reflected by this half mirror 18 passes through a portion of the reticle R near the mark RM and passes through the projection lens 1.
0, passes through this point, and enters the mark WM portion of the wafer W.

Next, the reflected light from the illumination light scattered by the edge portion of the mark WM formed on the surface of the wafer W is transmitted to the projection lens 10.
Since the light is telecentric on both sides, the reflected light enters the reticle R again, and the reflected light that passes through this enters the half mirror 18 described above. The reflected light from the mark WM that has passed through the half mirror 18 is incident on the objective lens 20 together with the reflected light from the mark RM, where it is converted into parallel light beams, and is incident on the spatial filter 22 .

This spatial filter 22 has a pupil position Pa of the projection lens 10.
The specularly reflected light is cut by this spatial filter 22, and only the scattered light is incident and imaged on the CCD camera 26 via the objective lens 24.Such scattered light Due to the incidence of the reticle R
A dark field image of the marks RM and WM on the wafer W is observed by the television camera 26.

Then, the mark RM of the reticle R in such an observation image
The stage 12 is moved in accordance with the overlapping state (or predetermined arrangement state) of the marks WM and the marks WM on the wafer W, and their alignment is performed.

In the above alignment apparatus, light having the same wavelength as the exposure light is used as alignment illumination light output from the light source 14.

[Problems to be Solved by the Invention] In the apparatus shown in FIG. is mark R
It is not a simple composite image of the image of M and the image of mark WM.

Since the mark WM on the wafer W is illuminated with uniform illumination light that has passed through the transparent portion of the reticle R, the image of the mark WM detected by the television camera 26 is a simple image.

On the other hand, regarding the image of the mark RM, a direct reflection image formed by the light directly reflected by the mark RM is formed, but at the same time, the illumination light irradiated on the wafer W is reflected and transmits and illuminates the reticle R. Since the illumination light is reflected and illuminates the reticle R, a transmitted illumination image of the mark RM is also formed. Furthermore, a reflected projection image in which the image of the mark RM is reflected by the wafer W and back-projected onto the mark RM is also formed.

That is, the image of the mark RM observed by the television camera 26 is actually the above-mentioned direct reflection image or transmitted illumination image.

This is a dark field image in which three reflected projection images are superimposed. This is because the projection lens 10 is telecentric on both sides, and if the projection lens is telecentric only on the wafer side, the observed image of the mark RM will be a superimposed image of the transmitted illumination image and the reflected projection image. In any case, the image of the mark RM on the reticle R is formed by overlapping the transmitted illumination image and the reflected projection image, so it may not be a clear image. In particular, if the reticle R and the wafer W are not accurately conjugated with respect to the projection lens 10, the bright and dark edges of the image of the mark RM will become unclear.

In addition, if light with the same wavelength as the exposure light is used as illumination light for alignment, if a resist with high absorption for light of that wavelength, such as a dyed resist, is coated on the wafer, the surface of the sea urchin It may not be possible to obtain enough reflected light to illuminate the marks on the reticle.

In such a case, it may not be possible to observe the image with a television camera in a good manner with high contrast, and as a result, alignment may become impossible.

This invention was made in view of the problems of the prior art, and it improves the contrast of mark images on a mask (reticle) and uses light of the same wavelength as the exposure light as illumination light for alignment. It is an object of the present invention to provide an alignment apparatus that can obtain a sufficient amount of scattered light from a projection target substrate and thereby perform highly accurate alignment.

[Means for Solving the Problems] The present invention provides a light reflecting means disposed at a pupil position of a projection optical system such as an exposure apparatus to irradiate incident light onto a substrate;
An optical means for causing the illumination light to enter the light reflecting means, and simultaneously observing the marks on the substrate and the mark on the mask, and moving relative to each mark to obtain a photoelectric signal regarding each mark. The technical point is that it is equipped with a signal detection means.

[Operation] According to the present invention, illumination light is incident from the pupil position of the projection optical system. Therefore, the number of optical elements through which the illumination light passes is reduced, and a decrease in the amount of light reaching the substrate is prevented.

The illumination light is reflected by a light reflecting means provided at the pupil position of the projection optical system, and is irradiated onto the substrate.

The reflected light from the marks on the substrate enters the signal detection means via the marks on the mask. The signal detection means obtains a photoelectric signal regarding each mark. At this time, the mark on the mask is only transmitted and illuminated by the light reflected from the substrate, and the mask mark is detected only from the transmitted illumination image.

According to one embodiment, the detection signal is scanned in the photoelectric conversion means. According to another aspect, the detection signal is scanned by movement of the substrate.

Since the photoelectric signal obtained by these operations is obtained without the need for a light splitting means, the loss of illumination light is further reduced.

Therefore, alignment based on the photoelectric signal obtained by the signal detection means is performed with high precision.

[Embodiments] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. Note that the same reference numerals are used for the same parts as in the prior art described above.

First, a first embodiment of the present invention will be described with reference to FIGS. 1 to 6.

FIG. 1 shows the configuration of the first embodiment. It is designed to be incident on the pattern area.

The exposure light that has passed through the reticle R further passes through the projection lens 34, which is telecentric on both sides.
The wafer W placed on a stage 36 that is movable in this direction is irradiated with light.

Of the above components, the reticle R has a circuit pattern area RP in the center, as shown in FIG. It is made of light shielding material.

Further, on the wafer W, seven cross-shaped alignment marks WA and WB are formed on the surface of the wafer W on the street line WS between the pattern portions WP (see FIG. 5), corresponding to the alignment marks RA and RB, respectively. It is formed as an uneven surface.

In this embodiment, in addition to the exposure system described above, two sets of alignment systems A and B as described below are provided. The only difference between the two is the direction in which they are arranged.

Hereinafter, alignment system A will be explained, and regarding alignment system B, corresponding elements will be denoted by B.

In FIG. 1, laser beams LA and LB are irradiated from the sides of the projection lens 34, and condensing lenses 38A and 38, respectively.
B't-The light is focused and incident.

The directions of incidence of these laser beams LA and LB on the projection lens 34 in the XY plane are as shown in FIG. The incident laser beam LA.

The LB enters a minute plane mirror 40 placed at the center inside the projection lens 34, is reflected there, and is irradiated onto the wafer W.

A sectional view taken along the arrow IV--IV line in FIG. 3 is shown in FIG. 4. In FIG. 4, the projection lens 34
has concave lenses 34C and 34D, each supported by a support 34B within a lens barrel 34A. A translucent member 34 is placed between these concave lenses 34C and 34D.
E is provided, and the above-mentioned mirror 40 is attached to this member 34E. The position of this mirror 40 is at or near the pupil position of the projection lens 34, and its area is determined to be sufficiently small with respect to the pupil diameter.

Further, an opening window 34FA is provided on the side of the above-mentioned lens barrel 34A.
is provided, thereby allowing laser light LA to enter the mirror 40 from the outside.

That is, the laser light L output from the laser light source 42A
A is made to enter the mirror 48A via the shutter 44A and the beam expander 46A, respectively.

The laser beam LA reflected by this mirror 48A is converted into a parallel beam by a lens 50A, and then...
, the above-mentioned condenser lens 3 via the aperture 52A.
It is designed to be incident on 8A. The laser beam LA is directed to the mirror 40 at the pupil position by the condensing lens 38A.
to form a spot.

Among the above-mentioned parts, the aperture 52A has a numerical aperture (
This is for adjusting the aperture angle α) of the luminous flux. In other words, the laser beam LA whose optical path is changed by the mirror 40 and illuminates the wafer W in parallel illuminates only the street line WS of the wafer W. This is because if the wavelength of the laser beam LA is the wavelength at which the resist layer formed on the wafer W is sensitive, all the locations illuminated by the laser beam LA will be exposed.

Next, laser light LA illuminates the wafer W.

LB are alignment marks WA and WB on wafer W, respectively.
The light beams are incident on the mirror 40 from different directions, as shown in FIG. 3, so that the light beams are incident on the mirror 40 respectively. The laser beam reflected by the wafer W then enters the projection lens 34 again and enters the vicinity of the alignment marks RA and RB of the reticle R.

As shown in FIG. 1, there is a reticle mark RA between the above-described condenser lens 32 and the reticle R.

Mirrors 54A and 54B are arranged at positions corresponding to RB, respectively. The laser light reflected by this mirror 54A (the part of the reflected laser light from Ueno\W that has passed through the reticle R) is reflected by the objective lenses 56A and 58.
The light passes through each of the marks A and enters the CCD camera 60A, and the marks RA and WA are imaged on the imaging surface thereof.

Mark RB and mark W detected by mirror 54B
The image of B is similarly detected by the CCD camera 60B.

Next, video signals for images of each mark from the CCD cameras 60A and 60B are connected to a waveform processing device 62 so as to be inputted thereto, and the processing output side of this waveform processing device 62 is connected to a control device 64. It is connected.

This control device 64 includes an interferometer 66 that measures the coordinate position of the stage 36 described above using interference of laser light;
A drive device 68 such as a motor that drives the stage 36 is connected to each stage.

The control device 64 includes a waveform processing device 62 and an interferometer 66.
It has a function of outputting a control signal for driving the stage to the drive device 68 based on an input signal from the , and aligning the reticle R and the wafer W with respect to the projection lens 34 .

Next, the overall operation of the first embodiment will be explained.

The laser light LA output from the laser light source 42A is incident on the mirror 40 in the projection lens 34 from the direction shown in FIG. 3 by the optical means shown in FIG. 4, and is reflected as shown in FIG. Then, the alignment mark WA on the street line WS of the wafer W is illuminated.

Next, the laser light scattered by the edge portion of the alignment mark WA and the laser light (including specularly reflected light) reflected by the surface other than the mark WA are transmitted to the projection lens 34.
The alignment mark RA of the reticle R is illuminated by passing through the entire KOIi surface. At this time, since the projection lens 34 is a telecentric optical system, the specularly reflected light from the wafer W does not return toward the reticle R by the mirror 40, and only the scattered light passing around the mirror 40 moves toward the reticle R.
will be illuminated.

The laser beam that has passed through the reticle R is then transferred to the mirror 54A.
The optical path is changed by , passes through objective lenses 56A and 58A, and images are formed on CCD camera 60A.

FIG. 6 shows such an alignment mark RA.

The imaging state of WA is shown. Of the optical image shown in this figure, arrow F1
When the scanning signal in the direction F4 is extracted by the waveform processing device 62, the result is as shown in FIG. 7(A).

That is, since the alignment mark RA of the reticle R has a light shielding property as described above, the amount of light is small in the portion between the edges RL and RR. Edge WL of alignment mark WA on wafer W.

In WR, the laser beam LA is scattered, so the amount of light is large. Therefore, the edges of each mark and the changing points of the graph correspond as shown. Therefore, in the waveform processing device 62, the edges RL, RR, WL of each mark are
, WR are determined, and further, (RL+RR)/2-(WL+WR)/2 is calculated as the amount of deviation between each mark and sent to the control device 64. Note that these deviation amounts are determined in each of the image areas 11 to I4 and sent to the control device 64.

In this way, by measuring the amount of deviation at four locations on the mark image, it is possible to check the deviation in four items: rotation θ in the X, Y, and Z directions, and magnification in Figure 1. becomes.

Next, the control device 64 outputs a control command signal to the drive device 68 in accordance with the amount of manual deviation. That is, while monitoring the stage position based on the output of the interferometer 66, the stage 36 is moved by the amount of shift by the drive device 68.

As described above, the pattern formed in the circuit pattern region RP of the reticle R is exposed using the alignment-completed metal material 30 for masking and cloud-kissing.

Note that when the surface of the wafer W easily reflects the incident laser beam LA, the graph in FIG. 7(A) becomes as shown in FIG. C). In either case, the mark RM on the reticle R is a simple transmitted illumination image, so the signal level is always at a minimum and the contrast with the mark RM is clear.

The same applies to the operation of alignment system B.

Next, a second embodiment of the present invention will be described with reference to FIGS. 8 to 11.

In the first embodiment described above, the alignment marks RA and RB of the reticle R have a light-shielding property, but in this embodiment, as shown in FIG. 9, they have a light-transmitting property. That is, the alignment marks RC and RD are formed by etching a light shielding film such as chromium in a cross shape.

As mentioned above, the alignment mark RC of the reticle R,
In connection with the fact that the RD has translucency, as shown in FIG. 8, a photoelectric element 70A such as a photomultiplier is used instead of a CCD camera.

70B are used, the outputs of which are each connected to a waveform processing device 72. In addition, the photoelectric element 70A,
70B may be arranged in a conjugate imaging relationship with the reticle or wafer, or may be arranged conjugately with the pupil of the projection lens 34.

Next, the operation of the second embodiment will be explained. first,
Similarly to the first embodiment described above, light scattered by the edge portions of the alignment marks WA and WB on the wafer W enters the alignment mark portions on the reticle R.

As mentioned above, the alignment mark RC.

RD has translucency. Therefore, for example, if the marks WA and RC are observed to overlap as shown in FIG. 10, and the mark WA moves in the direction of the arrow F5 due to the movement of the stage 36, the edges WL and W of the marks WA are
The larger the overlap between R and mark RC is, the more the photoelectric element 7
The output of 0A also increases. In this state, if the mark WA is placed separately, the edge of the mark WA can be observed through the mark RC.

FIG. 11 shows the position of the stage 36 and the photoelectric element when the mark WA is moved as described above. The change in the output of the OA is shown. As shown in this figure, when edges WL and WR of mark WA pass mark R'C, the output signal reaches its peak.

The alignment position corresponds to the midpoint of the rising and falling positions of the signal waveform. For example, the position of the stage 36 at the midpoint between the rising position P1 of the peak PA and the falling position P2 of the peak PB corresponds to the alignment position.

The amount of positional deviation of the stage 36 with respect to such an alignment position is determined by the waveform processing device 72, and 1.
It is output to the control device 64. The control device 64 moves the stage 36 and aligns the reticle R and the wafer W in the same manner as in the first embodiment.

In any of the above embodiments, the laser beams LA and LB, which are illumination lights for alignment, are transmitted through the projection lens 34.
However, when projecting the circuit pattern on the reticle R,
Such mirror 40 blocks the exposure light.

FIG. 12 shows the light shielding rate of the mirror 40 on the pupil plane 34S of the projection lens 34. As shown in this diagram,
0% and 5% of the pupil size.

If 8% and 10% light shielding is performed by the mirror 40, the MTF data values will be as shown in the table below as an example.

This table shows MTF in %. As is clear from this table, if the shielding rate is approximately 8% or less, even if the center of the pupil position is shielded, the resolution of each L/S (line and space line width) will not be affected practically. Therefore, the circuit pattern can be exposed well.

As explained above, according to the embodiments of the present invention, a sufficient amount of illumination light is incident on the alignment mark on the wafer, and since no light splitter or the like is used, light loss is also reduced. This has the effect that mark images can be observed clearly and alignment accuracy is improved.

In addition, if the spot of the illumination light condensed on the mirror 40 is made sufficiently small with respect to the spot of the mirror located at the pupil position of the projection lens, the spot of the illumination light condensed on the mirror 40 can be made small enough so that the position of the mirror 40 that receives the specularly reflected light is slightly tilted. Even if the reticle moves upward, there is no possibility of it leaking toward the reticle, and alignment problems can be reduced.

Furthermore, if the configuration is equipped with both a scanning photoelectric detector and a normal photoelectric detector, and the former detector is used on a rough surface with a bright background such as AJ2, the image of the alignment mark on the reticle will be It is effective to use the latter detector for objects that can be clearly recognized and are dark because the inconvenience of the alignment mark on the reticle becoming invisible does not occur.

Furthermore, as shown in Figure 3, by detecting multiple alignment marks by entering illumination light from different directions, the processing time required for alignment can be shortened.
, Y, θ (rotation) 1 magnification information can be obtained, which is even more effective.

Note that the present invention is not limited to the above-mentioned embodiments; for example, the shape of the alignment mark may be appropriately set as necessary, and does not need to be cross-shaped.

In addition, the illumination light for alignment may be made to enter the projection lens from one direction or from multiple directions, for example three directions. You may use the same light source, or you may divide the same light source. The wavelength of this illumination light may be the same as the wavelength of the exposure light, or may be another wavelength that does not cause chromatic aberration with respect to the projection optical system, or a wavelength near these wavelengths.

Furthermore, the configuration, arrangement, etc. of each optical element can be changed as necessary.

[Effects of the Invention] As explained above, according to the present invention, a mirror is disposed at the pupil position of the projection optical system to introduce illumination light for alignment, so the loss of light amount is reduced. This has the effect that alignment accuracy can be improved.

[Brief explanation of the drawing]

FIG. 1 is a configuration diagram showing the configuration of the first embodiment of the present invention, and FIG.
The figure is a plan view showing an example of a reticle, FIG. 3 is an explanatory diagram showing the direction of incidence of laser light on the projection lens, and FIG. FIG. 5 is an explanatory diagram showing the incident position of the laser beam on the wafer, FIG. 6 is an explanatory diagram showing the overlapping state of alignment marks, and FIG. 7 is an explanatory diagram showing an example of the waveform of the detection signal. 8 is a configuration diagram showing a second embodiment of the present invention, FIG. 9 is a plan view showing an example of a reticle of the second embodiment, and FIG. 10 is a diagram showing the overlap of alignment marks in the second embodiment. 11 is a diagram showing an example of the photoelectric detection signal of the second embodiment, FIG. 12 is an explanatory diagram showing the light shielding state of the pupil position of the projection lens, and FIG. 13 is a diagram showing the conventional device. It is a block diagram which shows an example. [Explanation of symbols of main parts] 34... Projection lens, 34A... Lens barrel, 34FA.
...opening window, 36...stage, 40...mirror,
60A, 60B...CCD camera, 62.72...
Waveform processing device, 64...control device, 66...interferometer, 68...drive device, ? OA, 70B...Photoelectric element, LA, LB...Laser light, R...Reticle, RA
. RB, RC, RD, WA, WB... alignment mark, W... wafer.

Claims (4)

[Claims] (1) Observe the optical image of the mark formed on the substrate to be aligned by the projection optical system and the mark formed on the mask to be aligned using illumination light for alignment, and then observe the photoelectric signals of both. an alignment device for aligning a mask and a substrate using a light reflecting means disposed at or near a pupil position in the projection optical system for irradiating incident light onto the substrate; an optical means for making the illumination light incident on the means; simultaneously observing the mark on the substrate and the mark on the mask, and moving relative to each mark to obtain a photoelectric signal regarding each mark; An alignment device comprising a signal detection means. (2) The alignment device according to claim 1, wherein the optical means has a numerical aperture for illumination light incident on the light reflecting means so that only the mark portion on the substrate is illuminated. . (3) The alignment device according to claim 1 or 2, wherein the optical means makes illumination light enter the light reflecting means from a plurality of different directions. (4) The alignment apparatus according to any one of claims 1 to 3, wherein the wavelength of the illumination light is a wavelength that does not cause chromatic aberration in the projection optical system or a wavelength close to the wavelength.