JPH08213306A - Position detecting device and projection exposure apparatus including the device - Google Patents

Abstract

(57) [Summary] [Object] To provide a position detecting device capable of improving detection sensitivity to defocusing with a simple configuration when incorporated as a focus position detecting system of an alignment device. [Structure] Pattern 17a of AF phase pattern plate 17
An image of the above is projected on the wafer 44 through the plane-parallel plate 19, the dichroic mirror 5, the illumination relay lens 6, and the like,
The image of the pattern on the wafer 44 is transferred to the first objective lens 8 and the second objective lens 8.
It is re-formed on the image sensor 21 via the objective lens 9, the dichroic mirror 10, the plane parallel plate 20, and the like. Parallel plane plate 19 (20) and dichroic mirror 5 (10)
Coma aberration is generated by the coma aberration imparting optical system consisting of
The asymmetry of the image on the image pickup device 21 is detected, and the defocus amount of the wafer 44 is obtained from the asymmetry.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a position detecting device for detecting the height (position in the normal direction) of a surface to be inspected, and a projection exposure apparatus equipped with such a position detecting device.
For example, a focus position detection system of an autofocus type alignment apparatus for detecting the position of a periodic alignment mark on a substrate to be inspected such as a semiconductor wafer, and a projection exposure apparatus equipped with such a focus position detection system. It is suitable to be applied to.

[0002]

2. Description of the Related Art When manufacturing a semiconductor device, a liquid crystal display device, or the like, a stepper or the like is used to print a pattern of a reticle as a mask on each shot area on a wafer (or glass plate, etc.) coated with a photosensitive material. Exposure apparatus is used. In such an exposure apparatus, in order to maintain the overlay accuracy between layers when forming a multi-layer circuit pattern on a wafer within a predetermined allowable range, alignment positions attached to each shot area on the wafer are aligned. An alignment device is provided for detecting the position of the mark (wafer mark) for use and accurately aligning the corresponding shot area with the exposure position based on the detection result.

However, since the surface of the wafer has unevenness due to warpage during manufacturing or through various processes, the alignment apparatus has conventionally been used to detect the height of a wafer mark to be detected. A focus position detection system is provided to detect whether or not it matches the in-focus point of the objective lens,
The wafer mark is focused by the autofocus method based on the detection result.

First, as a conventional alignment apparatus,
JP-A-4-65603, JP-A-4-273246
As disclosed in Japanese Unexamined Patent Application Publication No. 2003-264, etc., an enlarged image of a wafer mark illuminated by light having a wide wavelength band from a halogen lamp or the like is formed on a detection surface of an image sensor through an imaging optical system (alignment microscope). 2. Description of the Related Art There is known an image pickup type alignment apparatus which forms a mark and performs image processing on the obtained image pickup signal to detect a mark position. According to the imaging method, the influence of thin-film interference can be reduced by broadband illumination, and the asymmetric mark can be subjected to image edge selection processing and the like, so that highly accurate alignment is possible.

Next, as a focus position detection system for autofocus, a light for focus detection (hereinafter referred to as "AF detection light") is directly passed through an image forming optical system of an alignment apparatus.
The TL (through-the-lens) method can perform focus detection with high accuracy. As such a focus position detection system of the TTL system, as disclosed in, for example, Japanese Patent Laid-Open No. 3-42611, a predetermined pattern is projected on a surface to be inspected, and telecentricity of AF detection light with respect to the surface to be inspected. There is a so-called lateral deviation method in which an image of the pattern is formed by breaking the property, and the focus position of the surface is detected from the lateral deviation amount of the image when the surface is vertically displaced. As a method close to the lateral shift method, for example, Japanese Patent Laid-Open No. 5-19042.
As disclosed in Japanese Patent No. 4 publication, a predetermined pattern is obliquely projected on a surface to be inspected, and the oblique incidence for detecting the focal position of the surface to be inspected based on the lateral shift amount of the image of the pattern to be re-imaged. There is an AF method.

Further, as disclosed in, for example, Japanese Unexamined Patent Publication No. 2-99907, a so-called vertical shift method utilizing the fact that the contrast of the image of the pattern due to the AF detection light decreases with the defocus of the surface to be inspected. Is also known. The focus position detection system described above detects the focus of the surface to be inspected by detecting the amount of optical fluctuation such as the position of the image of a predetermined pattern and the amount of light by the AF detection light.

[0007]

However, in the conventional focus position detection system, both the so-called vertical deviation method for detecting the change in the contrast of the formed image and the so-called lateral deviation method for destroying the telecentricity of the AF detection light are combined. Since it is a method that depends on the numerical aperture of the image optical system relatively greatly, it is difficult to increase the detection sensitivity, which is the amount of change in the detection signal with respect to the amount of defocus (defocus amount), and as a result, the detection resolution of defocus is detected. , And the detection accuracy cannot be improved. In particular, in the alignment microscope, there is a particular restriction that the numerical aperture of the imaging optical system cannot be made very large, so that the conventional TTL focus position detection system also has a disadvantage that the detection sensitivity is low.

In view of the above point, the present invention has an object to provide a position detecting device which, when incorporated as a focus position detecting system of an alignment device, can improve the detection sensitivity for defocusing with a simple structure. And A further object of the present invention is to provide a projection exposure apparatus equipped with such a position detection device.

[0009]

A first position detecting device according to the present invention is a device for detecting a position (height) of a surface to be inspected (surface of 44) in a normal direction, as shown in FIG. 1, for example. In (1), a detection optical system (8, 9) for forming an image of a pattern formed on the surface to be inspected or an image of the pattern to be inspected consisting of a pattern projected on the surface to be inspected, and an image of the pattern to be inspected Optical system that gives coma aberration to
9, 5; 10, 20), and the position of the test surface is detected from the asymmetry of the image of the test pattern.

In this case, it is desirable that the coma-aberration giving optical system generate coma along the measuring direction of the asymmetry of the image of the pattern to be measured. Further, an observation optical system (1 to 11) for forming an image in a predetermined observation visual field including the inspection pattern on the inspection surface is provided, and the coma aberration is given to the outside of the optical path of the image-forming light flux by the observation optical system. It is desirable to arrange an optical system.

Next, the second position detecting device of the present invention is, for example, as shown in FIG. 6, a predetermined measurement in the position detecting device for detecting the position of the surface to be inspected (the surface of 44) in the normal direction. Illumination systems (1 to 3) for illuminating the pattern plate (17) on which the test pattern (17a) arranged in the direction is formed, and a defocused image of the test pattern is projected onto the test surface. A projection system (6 to 8), a light beam splitting system (30) that splits a light beam from the test surface into two, and a detection optical system that forms an image of the test pattern from one light beam from the light beam splitting system. (8, 9, 33), the observation optical system (8, 9) that forms an image on the surface to be inspected from the other light beam from the light beam splitting system, and the pattern to be inspected in the detection optical system. Coma aberration imparting optical system for generating coma aberration in an image (33)
And, and the position of the surface to be detected is detected from the asymmetry of the image of the pattern to be measured in the measuring direction.

In the projection exposure apparatus of the present invention, as shown in FIG. 1, for example, a projection optical system (43) for projecting an image of a mask pattern (41) onto a photosensitive substrate (44) and a photosensitive substrate (44). ) Is moved in a direction perpendicular to the optical axis of the projection optical system, and an observation optical system (8 to 11) that forms an image of the alignment mark (48) on the photosensitive substrate (44). ), Position detecting means for detecting the position in the optical axis direction (Z direction) in the vicinity of the alignment mark (48) on the photosensitive substrate (44), the photosensitive substrate (44) and its observation optical system. And a height adjusting means (45) for adjusting the relative position of the optical element in the direction parallel to the optical axis of the photosensitive board (45) via the height adjusting means (45) based on the detection result of the position detecting means. 44) After focusing on the observation optical system, A projection exposure apparatus for aligning a photosensitive substrate (44) on the basis of the position of the image of the mark for location registration (48).

In the present invention, the position detecting means is a mark (4) for alignment on the photosensitive substrate (44).
8), or a detection optical system (8, 9) that forms an image of a pattern to be detected that is formed by a pattern for position detection projected on the photosensitive substrate (44), and coma aberration is generated in the image of the pattern to be detected. Optical system for giving coma aberration (19, 5; 10, 2
0) and the position of the surface of the photosensitive substrate (44) in the optical axis direction is detected based on the asymmetry of the image of the test pattern.

[0014]

According to the first position detecting device of the present invention,
Due to the coma aberration imparted by the coma-aberration optical system, the light intensity distribution of the image of the pattern to be measured becomes asymmetric in the predetermined direction. For example, the degree of asymmetry of the left and right edges of the image in the predetermined direction is sensitively changed by the displacement (change in height) of the surface to be inspected in the normal direction. In this case, by appropriately setting the generation amount of coma aberration, the state of the test pattern, and the like, it is possible to control the detection sensitivity, which is the change amount of the asymmetry of the image of the test pattern with respect to the displacement of the test surface, and The linearity of the degree of asymmetry with respect to the displacement of the surface to be inspected can be secured sufficiently wide. Therefore, when applied to the focus position detection system, the detection sensitivity and the detection range for the defocus amount of the surface to be inspected are improved.

Further, according to the second position detecting apparatus of the present invention, since the defocused image of the test pattern is projected on the test surface, the test pattern observes the image of the test surface. It does not become an obstacle when doing. Furthermore, the amount of change in the height of the surface to be detected can be detected with high sensitivity based on the degree of asymmetry that occurs in the image of the pattern to be measured due to the addition of coma. Further, according to the projection exposure apparatus of the present invention, the photosensitive substrate (4) is provided via the height adjusting means (44) so that the asymmetry generated in the image of the pattern to be inspected due to the addition of coma aberration becomes a predetermined state.
After adjusting the height of 4), the position of the alignment mark (48) on the photosensitive substrate (44) is detected by the observation optical system, so that the photosensitive substrate (44) can be focused by the autofocus method. Done.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Various embodiments of the present invention will be described below with reference to the drawings. [First Embodiment] The first embodiment will be described with reference to FIGS. In this embodiment, the present invention is applied to an off-axis type alignment system of a projection exposure apparatus for manufacturing semiconductor devices. FIG. 1 shows a main part of the projection exposure apparatus of this example. In FIG. 1, the reticle 41 held on a reticle stage 42 is exposed under an illumination light L1 for exposure from an illumination optical system (not shown). An image of the pattern through the projection optical system 43 is projected and exposed on each shot area of the wafer 44. Here, the Z axis is taken parallel to the optical axis of the projection optical system 43, the X axis is taken parallel to the paper surface of FIG. 1 in the plane perpendicular to the Z axis, and the Y axis is taken perpendicular to the paper surface of FIG.

The wafer 44 is held on a Z stage 45 which positions the wafer 44 in the Z direction.
Are placed on an XY stage 46 that positions the wafer 44 in a direction parallel to the X axis (X direction) and in the Y direction. Each shot area of the wafer 44 is positioned based on the wafer mark attached to each shot area. Figure 2 (a)
2A shows a certain shot area 47 on the wafer 44. In FIG. 2A, an X-axis wafer mark 48 formed of an uneven pattern arranged adjacent to the shot area 47 at a predetermined pitch in the X direction, and Wafer marks 49 for the Y-axis are formed of a concavo-convex pattern arranged in the Y-direction at a predetermined pitch. In this example, the wafer mark 4 for the X axis is used.
8 is the detection target.

Returning to FIG. 1, an off-axis imaging system alignment system 50 is arranged in the lateral direction of the projection optical system 43. The imaging system alignment system is FIA
It is also called a (Field Image Alignment) type alignment system. In the alignment system 50, the illumination light emitted from the light source 1 such as a halogen lamp for position detection enters the optical bandpass filter 2 and the photoresist on the wafer 44 is not exposed by the optical bandpass filter 2. The illumination light L2 in the visible range is selected. The illumination light L2 is condensed on the field stop 4 by the condenser lens 3, and the illumination light L2 that has passed through the field stop 4 is reflected almost entirely by the dichroic mirror 5 and then converted into a substantially parallel light flux by the illumination relay lens 6. Half prism 7
Leading to. The light flux reflected by the half prism 7 illuminates an observation region including the wafer mark 48 for the X axis on the wafer 44 via the first objective lens 8.

Under the illumination light L2, the arrangement surface of the field stop 4 and the surface of the wafer 44 are substantially conjugate with each other, and the wafer 4 is
The surface of No. 4 has a Fourier transform relationship with the arrangement surface of the light source 1. That is, the illumination of the illumination light L2 on the wafer mark 48 is almost Koehler illumination. The illumination light L2 reflected from the wafer 44 is emitted from the first objective lens 8
Then, the light flux that has been condensed by the half prism 7 and returns to the half prism 7 passes through the second objective lens 9 and reaches the dichroic mirror 10. Then, the dichroic mirror 10
The illumination light L2, which is reflected almost entirely by, is divided into two by the half prism 11, and each is an X-axis image pickup device 12X and a Y-axis image pickup device 1 which are each composed of a two-dimensional CCD or the like.
It is incident on the imaging surface of 2Y. Then, at the time of focusing, the image pickup surfaces of the image pickup devices 12X and 12Y are respectively conjugate with the surface of the wafer 44, and the image of the wafer mark 48 is formed on the image pickup surface of the X-axis image pickup device 12X.

FIG. 2B shows the image pickup surface 12 of the image pickup device 12X.
The state of Xa is shown in FIG.
An image 48A of the wafer mark 48 is formed at the center of 2Xa. In this case, the wafer mark 4 on the wafer 44
The measurement direction on the imaging surface 12Xa of FIG. 2B corresponding to the measurement direction (X direction) of 8 is the X direction, and the index marks 24A and 24 are provided on both sides of the wafer mark image 48A in the X direction.
B is formed. The index marks 24A and 24B can be omitted when the pixels of the image sensor 12X are used as a reference. Then, the image is converted into an image pickup signal along the scanning line 25 parallel to the X direction on the image pickup surface 12Xa, and the obtained image pickup signal is processed to thereby generate the index marks 24A, 2A.
The position of the wafer mark 48 in the X direction with reference to 4B is detected. On the other hand, the position in the Y direction of the wafer mark 49 for the Y axis shown in FIG.
It is detected by processing the Y image pickup signal.

Next, the focus position detection system for performing the autofocus by the alignment system 50 will be described. First, a light source for focus detection consisting of a light emitting diode (hereinafter,
Illumination light (AF detection light) L3 from the “AF light source” 15 illuminates the AF phase pattern plate 17 via the condenser lens 16. The wavelength band of the illumination light L3 is
The dichroic mirrors 5 and 10 are deviated from the wavelength band of the position-detecting illumination light L2, and have dichroic mirrors 5 and 10 each having a wavelength selectivity that reflects the illumination light L2 and transmits the illumination light L3. Further, the AF phase pattern plate 17 is one in which a concavo-convex phase pattern 17a is formed at a predetermined pitch along a predetermined measurement direction on a transparent substrate such as a glass substrate. The difference in optical path length between the adjacent concave and convex portions of the phase pattern 17a is set to ¼ of the central wavelength λ ₀ of the illumination light L3 (that is, 90 ° in phase).

The illumination light L3 that has passed through the phase pattern 17a of the AF phase pattern plate 17 passes through the field stop 18, and then passes through the plane parallel plate 19 that is inclined at 45 ° with respect to the optical axis, and then the parallel light. It reaches the dichroic mirror 5 in a state where the plane plate 19 is rotated by 90 °. The parallel plane plate 19 and the dichroic mirror 5 correct astigmatism in the image of the phase pattern 17a to generate coma.
The coma-aberration giving optical system is constructed.

The illumination light L3 transmitted through the dichroic mirror 5 enters the half prism 7 as a substantially parallel light flux through the illumination relay lens 6, and the light flux reflected by the half prism 7 passes through the first objective lens 8 and the wafer. An image of the phase pattern 17a of the AF phase pattern plate 17, that is, a projected image 2 of the focusing mark 2 on the wafer mark 48 of 44.
6 is formed. That is, under the illumination light L3, the arrangement surface of the AF phase pattern plate 17 and the surface of the wafer 44 are substantially conjugate. Further, since the field stop 18 is arranged close to the AF phase pattern plate 17, the area of the projected image 26 of the focusing mark projected on the wafer 44 by the field stop 18, that is, the observation for focus detection. The field of view is set. Further, the projected image of the phase pattern has a dark portion at a portion corresponding to the step portion, and thus the projected image 26 also has a light-dark one-dimensional grid pattern in which the width of the dark portion is narrower than the width of the bright portion.

The illumination light L3 reflected from the projected image 26 of the focusing mark returns to the half prism 7 through the first objective lens 8 and is transmitted through the half prism 7.
3 passes through the second objective lens 9 and reaches the dichroic mirror 10 which intersects the optical axis at 45 °. Then, the illumination light L that has almost all transmitted through the dichroic mirror 10
3 is the dichroic mirror 10 90 ° around the optical axis
After passing through the plane-parallel plate 20 in the rotated state, the two-dimensional C
An image of the projected image 26 of the focusing mark is formed on the image pickup surface of the image pickup device 21 for focus detection such as a CD. In the focused state, the surface of the wafer 44 and the image pickup surface of the image pickup device 21 are conjugated under the illumination light L3.

Further, the dichroic mirror 10 and the plane-parallel plate 20 constitute a second coma-aberration giving optical system for correcting astigmatism and generating coma for the projected image 26 of the focusing mark. ing. In this example, the first
The coma aberrations generated on the wafer 44 by the coma aberration imparting optical systems 19 and 5 and the second coma aberration imparting optical systems 10 and 20 are set to be maximum in the measurement direction S shown in FIG. There is. As the optical system for generating the coma aberration, only one of the first coma aberration imparting optical systems 19 and 5 or the second coma aberration imparting optical systems 10 and 20 may be used. Further, since the illumination light L2 for position detection is reflected by the dichroic mirrors 5 and 10, no coma aberration occurs in the image of the wafer mark 48 captured by the image sensor 12X.

FIG. 3 shows an observation visual field surrounding the wafer mark 48 in FIG. 1. In FIG. 3, the pitch P _WM in the X direction is shown.
A projection image 26 of the focusing mark, which is an image of the phase pattern 17a in FIG. 1, is formed on the wafer mark 48 formed in Step 1. The projection image 26 extends in the dark portion 2 along the measurement direction S.
6a and bright portions 26b are arranged at a pitch P _AF to form a periodic pattern. The measurement direction S intersects the X direction, which is the measurement direction of the wafer mark 48, at approximately 45 °, and the direction perpendicular to the measurement direction S is the non-measurement direction T. Then, an image in the circular observation field 27 surrounding the wafer mark 48 is picked up by the image pickup device 12X for position detection shown in FIG.
Further, the image of the projection image 26, which is limited to the substantially rectangular area, is captured by the image sensor 21 for focus detection in FIG. 1, and the scanning direction of the pixels of the image sensor 21 is set to the direction corresponding to the measurement direction S. ing.

Further, as an example, the pitch P _WM of the wafer mark 48 is set to about 4 to 16 μm, and the pitch P _AF of the projected image 26 of the focusing mark is also set to the wafer mark 48.
It is set to the same level as the pitch P _WM of. Further, the numerical aperture of the imaging optical system including the first objective lens 8 and the second objective lens 9 is set to about 0.2 as an example. In this example, only the illumination light L2 is incident on the image sensor 12X, and only the illumination light L3 is incident on the image sensor 21, so that the focusing mark is projected on the image capturing surface of the image sensor 12X for position detection. The image of image 26 is not formed. However, on the image pickup surface of the image pickup device 21 for focus detection, an image of the wafer mark 48 in the projected image 26 by the illumination light L3 is formed.

In FIG. 1, the image pickup signal V _AF from the image pickup device 21 is supplied to a signal processing device 22, and the signal processing device 22 integrates the image pickup signal in a non-measurement direction perpendicular to the measurement direction. The asymmetry of the captured image is calculated based on ΣV _AF as described later, the defocus amount of the wafer mark 48 is calculated from this asymmetry, and the defocus amount is output to the stage control system 23. In the stage control system 23,
The Z stage 45 is driven in the Z direction by the amount of defocus. This allows the wafer mark 4 to be
Focusing of 8 is performed.

Next, the operation of detecting the position of the wafer mark 48 in this example will be described in detail with reference to FIGS. First, in FIG. 1, the position of a mark (not shown) for search alignment on the wafer 44 is detected, and the detection result is subjected to an offset correction stored in advance as a design value, whereby the wafer mark 48 to be measured. The rough position of is calculated. By driving the wafer stage 46 based on the calculated position, the wafer mark 48 is positioned within the observation visual field of the first objective lens 8. This is search alignment. After that, autofocus is performed using the projected image 26 of the focusing mark projected on the wafer 44.

Before that, in the present example in FIG. 3, in order to reduce the influence of the image of the wafer mark 48 at the time of focus detection,
The width _HAF of the projected image 26 of the focusing mark in the non-measurement direction T is set to be a natural multiple of the pitch 2 ^1/2 · P _WM of the wafer mark 48 in the non-measurement direction T. Therefore, natural number m
To establish the following relationship: H _AF = m ・ 2 ^1/2・ P _WM (1)

From this relationship, even if the position of the wafer mark 48 is slightly shifted in the X and Y directions, the image pickup signal V _AF output from the image pickup device 21 for focus detection in FIG. _By integrating only _AF , the influence of the image of the wafer mark 48 becomes almost DC level in the measurement direction S. Therefore, the focus position (Z coordinate of the surface of the wafer 44) can be accurately detected without being affected by the pattern of the wafer mark 48.

Next, in FIGS. 4A to 4C, the integrated signal ΣV _AF obtained by integrating the image pickup signal V _AF output from the image pickup device 21 for focus detection in FIG. 1 in the non-measurement direction T is measured. FIG. 4A is a diagram plotted with respect to the direction S, and FIG. 4A shows a case where the Z coordinate of the surface of the wafer 44 is shifted upward by ΔZ from the focal point Z _{0 with} respect to the image sensor 12X. Also,
4B shows the case where the Z coordinate of the surface of the wafer 44 is at the focal point Z ₀ , and FIG. 4C shows that the Z coordinate of the surface of the wafer 44 is shifted downward from the focal point Z ₀ by ΔZ. It shows the case. S ₀ indicates the center position of the projected image. 4 (a)-
As shown in (c), due to the influence of coma aberration, the wafer 4
Integral signal ΣV according to the change of the defocus amount on the surface of No. 4
The degree of asymmetry of _{AF in} the measurement direction S changes, and in the in-focus state shown in FIG. 4B, the projection magnification from the wafer 44 to the image sensor 21 is β, and the integration signal ΣV _AF shows the measurement direction. The signal becomes a symmetric signal that changes to S with the pitch β · P _AF .

In order to quantify the degree of the asymmetry, first, in FIG. 4 (a), the value of the left and right edge portions (falling portions) of one cycle (pitch β · P _AF ) of the integrated signal ΣV _AF is V. Let _iL and V _iR (i = 1, 2, 3, ...). Next, as an index of the asymmetry of the image pickup signal, the value V of the left and right edge portions
The difference ΔV _i between _iL and V _iR is _calculated . However, the difference ΔV _i is standardized by using the difference between the maximum value V _max and the minimum value V _{min in} the part of the integrated signal ΣV _AF except for both ends. Therefore, the following equation is established.

ΔV_i= (V_iR-V_iL) / (V_max-V_min) (2) Note that the difference ΔV_iIs not always (V_max-V_min) Standardize
do not have to. Next, the image is picked up by the image sensor 21 of FIG.
Pitch β of the projected image 26 of the focusing mark
_AFLet n be the number of cycles with n as the unit (n is an integer of 2 or more)
And the difference ΔV _iIntegrated signal ΣV_AFN (n is 2 or more
(Integer of) Focusing marks averaged over a period
A quantity ΔIA indicating the degree of image asymmetry of the projected image 26 of
Meru. Therefore, the sum symbol Σ is from 1 to n for the subscript i.
The following relations are established to represent the integration in.

ΔIA = (1 / n) ΣΔV _i (3) FIG. 5 shows the amount ΔIA indicating the degree of asymmetry as the wafer 4
4A and 4B are graphs plotted against the defocus amount ΔZ of the surface of FIG. 4, FIG. 5A shows the characteristics in the case of ideal image formation without coma aberration, and FIG. The characteristic in the case of the image formation with is given. Furthermore, FIG.
In FIG. 1, the flat straight line 28A is the AF light source 15 of FIG.
Shows the characteristic when there is no eccentricity, and the curved line 28B shows the characteristic when the AF light source 15 has a predetermined eccentricity.
In FIG. 5B, the approximate straight line 29A passing through the origin is AF
The characteristics when the light source 15 for AF has no eccentricity, and the approximate straight line 29B passing through the position deviated from the origin shows the characteristics when the light source 15 for AF has a predetermined eccentricity. In FIG. 5B, the slope tan φ of the approximate straight lines 29A and 29B is substantially constant regardless of the eccentricity of the AF light source 15.

As can be seen from FIG. 5B, when coma is imparted to the image of the projected image 26 of the focusing mark as in this example, regardless of whether the AF light source 15 is decentered or not, (3 ) Amount ΔIA indicating the degree of asymmetry expressed by the equation
Changes substantially in proportion to the defocus ΔZ. Therefore, focusing is performed by calculating the amount ΔIA and controlling the Z stage 45 so that the calculation result becomes a predetermined value (for example, 0).

The change rate ΔIA / ΔZ of the amount ΔIA with respect to the defocus ΔZ, that is, the slope tan φ in FIG. 5B is
Although it is proportional to the amount of coma given, the amount of coma is
This can be controlled by changing the thicknesses of the dichroic mirrors 5 and 10 and the plane-parallel plates 19 and 20 which form the coma aberration imparting optical system. Further, the rate of change ΔIA / ΔZ can be regarded as the detection sensitivity to defocus, and the detection sensitivity corresponds to the numerical aperture of the imaging optical systems 6 and 8 for projecting the AF phase pattern plate 17 onto the wafer 44. It also depends on the σ value of the illumination system, which is the ratio of the numerical apertures of the illumination systems 15 and 16 of the AF phase pattern plate 17, and the step of the AF phase pattern plate 17.

In addition to increasing the coma aberration, it is effective to reduce the σ value of the illumination system in order to improve the detection sensitivity to defocus. The unevenness of the pattern 17a of the AF phase pattern plate 17 causes a phase difference of 90 ° (2n + 1) (n = 0, 1, 2, ...) At the center wavelength λ ₀ of the focus detection illumination light L3. When held, the change rate ΔIA / ΔZ (detection sensitivity) becomes maximum. The detection sensitivity obtained in this way is much more sensitive than in the case of the conventional lateral displacement method, and is almost the same as the numerical aperture of the imaging optical system (alignment microscope) including the first objective lens 8 and the second objective lens 9. Not restricted. The numerical aperture of the alignment microscope is, for example, about 0.2, but in this example, defocus can be detected with high sensitivity by appropriately combining the amount of coma aberration to be applied and the σ value of the illumination system.

Another feature of the focus detection method using coma aberration of this example is that the linearity of the change rate ΔIA / ΔZ with respect to the defocus ΔZ is extremely wide when the numerical aperture of the alignment microscope is about 0.2. It is good and can sufficiently cover the focus detection range practically required. The pitch of the projected image 26 of the focusing mark projected on the wafer 44 has little relation to the value of detection sensitivity and linearity.

After the focusing is performed as described above, the X coordinate of the wafer mark 48 is detected by processing the image pickup signal of the image pickup device 12X, and then the Y coordinate of the wafer mark 49 for the Y axis is also detected. To be done. Then, the final alignment (fine alignment) of the corresponding shot area is executed based on the X and Y coordinates detected in this way.

[Second Embodiment] A second embodiment will be described with reference to FIGS. 6 and 7. This embodiment is different from the first embodiment in that the light source for position detection and the light source for focus detection are shared by one light source. In FIG. 6, parts corresponding to those in FIG. The detailed description thereof will be omitted.

In the alignment system 50A shown in FIG. 6, the illumination light L2 selected through the optical bandpass filter 2 from the illumination light emitted from the light source 1 including a halogen lamp for position detection and focus detection is used. The illumination light L2, which is condensed on the AF phase pattern plate 17 by the condenser lens 3 and has passed through the phase pattern 17a of the AF phase pattern plate 17, passes through the field stop 4 arranged with a defocus amount Δd. , Enters the half prism 7 via the illumination relay lens 6. Then, the light flux reflected by the half prism 7 illuminates a predetermined observation region including the wafer mark 48 for the X axis on the wafer 44 via the first objective lens 8.

In this case, the field stop 4 with respect to the illumination light L2.
And the surface of the wafer 44 are substantially conjugate with each other, and the pattern 17a of the AF phase pattern plate 17 is defocused by Δd from the conjugate plane with the wafer 44. As a result, the pattern 17a does not become an obstacle when observing the wafer mark 48. The illumination light L2 reflected from the wafer 44 is condensed by the first objective lens 8 and returns to the half prism 7, and the light flux transmitted through the half prism 7 passes through the second objective lens 9 and the beam splitter 3
It reaches 0. Then, the light flux transmitted through the beam splitter 30 is divided into two by the half prism 11, and the X-axis image pickup device 12X and the Y-axis image pickup device 12 are respectively divided.
It is incident on the image pickup surface of Y. Then, at the time of focusing, the image sensor 1
The image pickup surfaces of 2X and 12Y are respectively conjugated with the surface of the wafer 44, and the image of the wafer mark 48 is formed on the image pickup surface of the X-axis image pickup device 12X.

On the other hand, the light flux reflected by the beam splitter 30 once forms an image of the pattern 17a of the AF phase pattern plate 17 on the intermediate image plane 31, and the light flux from the intermediate image plane 31 is a visual field for blocking stray light. After passing through the diaphragm 32, the image of the pattern 17a is re-formed on the image pickup surface of the image pickup device 21 for focus detection through the relay lens system 33 having a coma aberration generating function. That is, at the time of focusing under the illumination light L2, the intermediate image plane 31 and the image pickup surface of the image pickup device 21 are respectively conjugate with the phase pattern plane of the AF phase pattern plate 17, and the arrangement plane of the field stop 32 is the wafer. It is conjugated with the surface of 44. In this case, the distance from the arrangement surface of the field stop 32 to the intermediate image plane 31 is the defocus amount ΔD, the projection magnification by the illumination relay lens 6 and the first objective lens 8 is β ₁ , the first objective lens 8 and the second objective lens 8. When the projection magnification by the lens 9 is β ₂ , the defocus amount ΔD is equal to the AF phase pattern plate 1
It is expressed as follows using a defocus amount Δd of 7.

ΔD = (β ₂ / β ₁ ) ² Δd (4) In this example, the phase pattern 1 of the AF phase pattern plate 17 is
A direction corresponding to the measurement direction S with respect to the image of the pattern 17a projected on the image pickup surface of the image pickup device 21 in the relay lens system 33 with a coma aberration generating function, with the arrangement direction (pitch direction) of the 7a being the measurement direction S. The coma aberration is generated so as to be maximum in (this is referred to as the measurement direction S on the imaging surface). Further, the image pickup signal is read out along the measurement direction S on the image of the pattern 17a.

The projection area (focus detection area) of the pattern 17a is not limited to the inside of the wafer mark 48, and the image of the pattern 17a may be projected on a predetermined area including the wafer mark 48. . By thus widening the focus detection area, the detection error is reduced by the averaging effect. Therefore, there are no particular restrictions on the pitch of the projected image of the focusing mark (the image of the pattern 17a or the like) or the projected area.

Next, the focus detection operation of this example will be described. In this example, as shown in FIG. 6, since the pattern surface of the AF phase pattern plate 17 is defocused by Δd from the conjugate plane with the wafer 44, it is projected onto the wafer 44 when the wafer 44 is focused. The image of the phase pattern 17a of the AF phase pattern plate 17 is blurred, and the illuminance distribution in the field of view of the objective lenses 8 and 9 becomes substantially uniform. Even at this time, the image of the wafer mark 48 is clearly formed on the image sensor 12X for position detection, and the blurred image of the pattern 17a becomes substantially uniform background light.

On the other hand, when the wafer 44 is in focus, the image of the pattern 17a is clearly formed on the image pickup device 21 for focus detection, but the image of the wafer mark 48 is blurred and becomes a substantially uniform background light. . Further, a coma aberration is generated in the image of the pattern 17a on the image sensor 21 for focus detection along the direction corresponding to the measurement direction S (the arrangement direction of the light and dark patterns and the scanning direction), but the coma aberration is generated in the relay. Lens system 33
Is located outside the optical path of the image-forming light beam for position detection, the wafer mark 48 on the image sensor 12X for position detection is
There is no coma in the image. As a result, the position detection and the focus detection can be independently and highly accurately executed without affecting each other.

Next, by integrating the image pickup signal V _AF output from the imaging device 21 in the present embodiment in a non-measurement direction, the integrated signal [sigma] v _AF is obtained, the integrated signal [sigma] v _AF wafer 4
As shown in FIGS. 4A to 4C, the defocus amount ΔZ of 4 changes as in the first embodiment. Therefore, as in the first embodiment, the defocus amount ΔZ can be detected from the asymmetry of the integrated signal ΣV _{AF in} the measurement direction S.

Besides, the asymmetry is quantified as follows.
There is also a method. That is, as shown in FIG.
Minute signal ΣV_AF1 cycle in the measurement direction S (for two peaks)
Left and right edge width a_iLAnd a_iR(I = 1, 2, 3,
…) Is measured. For this purpose, the integrated signal ΣV_AFof
Maximum value V in the part excluding both ends of the whole_maxAnd the minimum value V
_minAverage value V_me(= (V_max+ V_min) / 2)
Therefore, the integrated signal ΣV_AFIs the average value V_meCross
The width of time can be measured. Next, the integrated signal ΣV_AF
As an index of asymmetry in the measurement direction S of
Width a_iLAnd a_iRDifference Δa_iTo calculate. Therefore,
The following equation holds.

Δa _i = a _iR −a _iL (5) Next, the number of cycles in units of the pitch (two peaks) of the image of the pattern 17 a picked up by the image pickup device 21 is n (n is 2 or more). The difference Δa _i as an integral signal ΣV
Averaged over n (n is an integer of 2 or more) cycles of _AF ,
An amount ΔIB indicating the degree of asymmetry of the image of the pattern 17a is obtained. Therefore, the sum symbol Σ is 1 to n for the subscript i.
The following relations are established to represent the integration up to.

ΔIB = (1 / n) ΣΔa _i (6) Similar to the amount ΔIA represented by the equation (3) in the first embodiment,
The amount ΔIB indicating the degree of asymmetry expressed by the equation (6) is also
It changes substantially in proportion to the defocus amount ΔZ of the wafer 44. Therefore, the autofocus can be performed as in the first embodiment. The change rate (detection sensitivity) and linearity of the amount ΔIB with respect to the defocus ΔZ are the same as the amount ΔIA in the first embodiment.

As described above, also in this example, focus detection can be performed with high detection sensitivity by adding coma aberration and detecting the asymmetry of the image pickup signal. Furthermore, since the illumination light for focus detection is illumination light for the same wide band as the illumination light for position detection (for alignment), the thin film interference due to the photoresist on the wafer 44 is less likely to be adversely affected during focus detection. There are advantages. Further, since the light source is commonly used for focus detection and position detection, the entire optical system is simplified.

Next, a specific configuration example of the relay lens system 33 having a coma aberration imparting function will be described with reference to FIG. The relay lens system 33 in FIG. 7A relays the image on the intermediate image plane 31 onto the image pickup device 21 by the first relay lens 34A and the second relay lens 34B. In this case, for example, decentering the first relay lens 34A on the front side with respect to the optical axis AX2 up to that point causes decentration coma aberration in the image for focus detection on the image sensor 21. Note that in FIG. 7A, the first relay lens 34
With the optical axis of A aligned with the optical axis AX2, the relay lens system 33 may be decentered as a whole so that the light flux from the intermediate image plane 31 passes outside the optical axis of the relay lens system 33. . At this time, coma aberration depending on the image height is generated on the image pickup device 21. In any of these cases, the amount of coma aberration generated can be controlled by the amount of eccentricity of the lens.

FIG. 7B shows another example of the configuration of the relay lens system 33. In FIG. 7B, the image on the intermediate image plane 31 is formed on the image pickup device 21 by the two relay lenses 35A and 35B. It relays and the 1st relay lens 35
A pair of zone plates 36A and 3 is provided near the pupil plane (Fourier conversion plane) F1 between A and the second relay lens 35B.
6B is arranged in close proximity. At this time, for example, by decentering one of the zone plates 36A from the optical axis AX2 and relatively shifting the zone plates 36A and 36B, the amount of coma produced on the image sensor 21 can be controlled.

Next, in the embodiment of FIG. 6, since the position detection and the focus detection are performed independently, the AF phase pattern plate 17 is defocused from the conjugate plane with the wafer 44.
The F phase pattern plate 17 may be arranged on the conjugate plane of the wafer 44 (that is, Δd = 0 in FIG. 6). In this case, in the X-axis image pickup device 12X of FIG. 6, the area of the projected image 26 of the focusing mark in FIG.
An image that is spread out so as to cover the entire surface of 8 is observed. In order to perform the position detection and the focus detection with high accuracy in this state, respectively, as in FIG.
The measuring direction S of (the image of the pattern 17a) is intersected at 45 °, and the measuring direction S is measured by the relay lens system 33 of FIG.
The maximum coma aberration is generated in the direction corresponding to.

[0057] Also, the width H _AF of non-measurement direction T which integrates the image signal by the image sensor 21, of the first embodiment (1)
Impose conditions on the formula. As a result, focus detection is performed without being affected by the wafer mark 48. On the other hand, in this example, an image of the projected image 26 of the focusing mark is also formed on the image sensor 12X for position detection. Therefore, in order to remove the influence of the image, in FIG. 3, the image pickup signal output from the image pickup device 12X is output in the Y direction in the region of the width H _{WMX in} the Y direction perpendicular to the measurement direction (X direction) of the wafer mark 48. We will integrate. The width H _WMX is set so that an integral number of bright and dark patterns of the projected image 26 of the focusing mark are included in this width. The pitch of the projected image 26 in the Y direction is 2
^Since it is ^1/2 · P _AF , its width H _WMX may be set as follows using a natural number m ′.

H _WMX = m′2 ^1/2 · P _AF (7) As a result, the pattern of the AF phase pattern plate 17 is used even when the common light source is used and the AF phase pattern plate 17 is not defocused. The position of the wafer mark 48 in the X direction can be accurately detected without being affected by the image of 17a. Further, even if the positional relationship between the wafer mark 48 and the image of the pattern 17a (projected image 26) is deviated as a result of the search alignment, by integrating the image pickup signal,
The effect of the image of the pattern 17a is constant (DC component), and the position detection accuracy is maintained high.

[Third Embodiment] A third embodiment will be described with reference to FIG. The present embodiment is different from the above-described embodiments in that the wafer mark itself is regarded as a focusing mark, and in FIG. 8, parts corresponding to those in FIG. 1 are designated by the same reference numerals and detailed description thereof will be omitted. In the alignment system 50B of FIG. 8, the illumination light L2 selected from the illumination light emitted from the light source 1 including a halogen lamp for position and focus detection through the optical bandpass filter 2 is selected.
Is condensed on the field stop 4 by the condenser lens 3, and the illumination light L2 passing through the field stop 4 enters the half prism 7 via the illumination relay lens 6. And
The light beam reflected by the half prism 7 passes through the first objective lens 8 and the wafer mark 4 for the X axis on the wafer 44.
A predetermined observation area including 8 is illuminated.

In this case, the field stop 4 with respect to the illumination light L2.
And the surface of the wafer 44 are substantially conjugate with each other, and the field stop 4 sets a position on the wafer 44 and an observation field of view for focus detection. Illumination light L2 reflected from the wafer 44
Is condensed by the first objective lens 8 and returns to the half prism 7, and the light flux transmitted through the half prism 7 reaches the beam splitter 30 via the second objective lens 9. Then, when the wafer 44 is focused, the beam splitter 30
The light flux transmitted through forms an image of the wafer mark 48 on the image pickup surface of the image pickup device 12 for position detection. That is, when the wafer 44 is focused, the image pickup surface of the image pickup device 12 is conjugate with the surface of the wafer 44.

On the other hand, the light beam reflected by the beam splitter 30 once forms an image of the wafer mark 48 on the intermediate image surface 31, and the light beam from the intermediate image surface 31 passes through the relay lens system 39 having a coma aberration generating function. The image of the wafer mark 48 is re-formed on the image pickup surface of the image pickup device 21 for focus detection via the above.
That is, at the time of focusing under the illumination light L2, the image pickup surface of the image pickup element 12, the intermediate image surface 31, and the image pickup surface of the image pickup element 21 are
Each is conjugated with the surface of the wafer 44.

The relay lens system 39 is composed of a first relay lens 37A, a coma aberration generating optical system 38, and a second relay lens 37B, and the coma aberration generating optical system 38.
Is composed of, for example, a pair of zone plates 36A and 36B shown in FIG. Then, by decentering the optical members in the coma aberration generating optical system 38, coma aberration is generated in the image on the image sensor 21. In this case, the direction in which the coma aberration is maximized (that is, the measurement direction S) is the direction corresponding to the X direction which is the pitch direction of the wafer mark 48.

In this example, the defocus amount of the wafer 44 is detected by first determining the asymmetry of the image of the wafer mark 48 on the image pickup device 21 in the measurement direction S. After that, the defocus amount is corrected via the Z stage 45, and then the position of the wafer mark 48 is detected based on the image of the wafer mark 48 on the image sensor 12 for position detection. At this time, since the wafer mark 48 itself is used as the focusing mark, the optical system is simple, and the SN ratio of the focus detection signal and the position detection signal is high, which results in high accuracy. Autofocus and position detection (alignment) can be performed.

Although the position of the mark to be detected is detected by the image pickup method (FIA method) in the above-described embodiment, the mark to be detected is irradiated with a coherent laser beam from two directions in addition to that. Even when using a so-called heterodyne interference type alignment system that detects the position of the test mark based on the phase of the heterodyne beam returned from the test mark, it is possible to apply the present invention to perform stable autofocus. You can

As described above, the present invention is not limited to the above-described embodiments, and various configurations can be adopted without departing from the gist of the present invention.

[0066]

According to the first position detecting device of the present invention,
By generating coma aberration in the image of the test pattern and determining the asymmetry of the image of the test pattern that changes depending on the position (height) in the normal direction of the test surface, The height is detected. Therefore, the height detection sensitivity can be increased by simply increasing the coma aberration amount with a simple configuration, and for example, the detection sensitivity to defocus (defocus) when used as a focus position detection system can be increased. There are advantages.

In this case, when the coma-aberration giving optical system produces coma aberration along the measuring direction of the asymmetry of the image of the pattern to be measured, the asymmetry becomes the largest in that measuring direction, so that the height change ( The detection sensitivity to defocus etc.) is the highest. Further, when an observation optical system for forming an image in a predetermined observation visual field including the pattern to be inspected on the surface to be inspected is provided and the coma aberration imparting optical system is arranged outside the optical path of the image-forming light flux by this observation optical system, Since no coma aberration is generated in the image observed by the observation optical system, when applied to an alignment device, for example, the position of the alignment mark on the surface to be inspected can be detected with high accuracy. Further, since it is a method of generating coma, when focus detection is performed by the TTL method via the objective optical system of the observation optical system,
Even if the numerical aperture of the objective optical system is small, there is an advantage that defocus can be detected with high detection sensitivity.

Next, according to the second position detecting device of the present invention, the height of the surface to be inspected is detected based on the asymmetry of the image of the pattern to be inspected which is caused by coma aberration. With the configuration, it is possible to enhance the detection sensitivity to the height change only by increasing the coma aberration amount. Further, since the test pattern is projected on the test surface in a defocused state, there is an advantage that the test pattern is hardly influenced by the observation image of the test surface.

Further, according to the projection exposure apparatus of the present invention, the position of the alignment mark on the photosensitive substrate can be detected while the photosensitive substrate is focused on the observation optical system by the autofocus system. . Since a device that detects the defocus amount based on the asymmetry of the image obtained by adding coma aberration is used as the position detection means in that case, the detection sensitivity to defocus can be easily improved.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a projection exposure apparatus according to a first embodiment of the present invention.

2A is an enlarged plan view showing a wafer mark attached to a shot area on the wafer of FIG. 1, and FIG. 2B is an image of the wafer mark 48 on an image pickup surface of an image pickup device 12X of FIG. It is an enlarged image.

FIG. 3 is an enlarged view showing a projected image of a focusing mark projected on a wafer mark in the first embodiment.

FIG. 4 is a diagram showing how the degree of asymmetry of an image formed on an image sensor changes according to the defocus amount of a wafer in the first embodiment.

FIG. 5A is a defocus amount ΔZ in the case of ideal image formation.
And (b) is a diagram showing the relationship between the defocus amount ΔZ and the amount ΔIA when coma aberration is applied.

FIG. 6 is a configuration diagram showing an alignment system of a projection exposure apparatus according to a second embodiment of the present invention.

7 is a configuration diagram showing a specific example of a relay lens system 33 having a coma aberration imparting function of FIG.

FIG. 8 is a configuration diagram showing an alignment system of a projection exposure apparatus according to a third embodiment of the present invention.

[Explanation of symbols]

 1 Light source for position detection 2 Optical bandpass filter 4,18 Field stop 5,10 Dichroic mirror 6 Illumination relay lens 7,11 Half prism 8 First objective lens 9 Second objective lens 12X, 12Y Image sensor for position detection 15 AF light source 17 AF phase pattern plate 19, 20 Parallel plane plate 21 Focus detection image sensor 22 Signal processing device 23 Stage control system 41 Mask 43 Projection optical system 44 Wafer 45 Z stage 48 Wafer mark

Claims (5)

[Claims] 1. An apparatus for detecting a position of a surface to be inspected in a normal direction, wherein an image of an inspection pattern formed of a pattern formed on the surface to be inspected or a pattern projected on the surface to be inspected is formed. A detection optical system that forms the image, and a coma-aberration giving optical system that generates a coma aberration in the image of the test pattern, and detects the position of the test surface from the asymmetry of the image of the test pattern. Characteristic position detection device. 2. The position detecting device according to claim 1, wherein
The position detecting device, wherein the coma-aberration giving optical system generates coma along the measurement direction of the asymmetry of the image of the test pattern. 3. The position detecting apparatus according to claim 1, further comprising an observation optical system that forms an image in a predetermined observation visual field including the inspection pattern of the inspection surface, the observation optical system. The position detecting device, wherein the coma-aberration giving optical system is arranged outside the optical path of the image-forming light flux. 4. A position detecting device for detecting a position of a surface to be inspected in a normal direction, and an illumination system for illuminating a pattern plate on which an inspected pattern arranged in a predetermined measuring direction is formed, and the inspected pattern. Of the defocused image on the surface to be inspected, a light beam splitting system for splitting the light beam from the surface to be inspected into two, and an image of the pattern to be inspected from one light beam from the light beam splitting system. A detection optical system that forms an image, an observation optical system that forms an image on the surface to be inspected from the other light beam from the light beam splitting system, and a coma aberration in the image of the pattern to be inspected in the detection optical system. And a coma-aberration giving optical system for detecting the position of the surface to be detected from the asymmetry of the image of the pattern to be measured in the measurement direction. 5. A projection optical system for projecting an image of a mask pattern onto a photosensitive substrate, a substrate stage for moving the photosensitive substrate in a direction perpendicular to an optical axis of the projection optical system, and alignment on the photosensitive substrate. Optical system for forming an image of a mark for use, position detection means for detecting a position in the optical axis direction in the vicinity of the alignment mark on the photosensitive substrate, the photosensitive substrate and the observation optical system And height adjusting means for adjusting the relative position in the direction parallel to the optical axis, and the observation optical system for observing the photosensitive substrate via the height adjusting means based on the detection result of the position detecting means. In the projection exposure apparatus, which aligns the photosensitive substrate based on the position of the image of the alignment mark by the observation optical system after focusing on the system, the position detecting means is provided on the photosensitive substrate. For alignment Of the mark, or a detection optical system that forms an image of the pattern to be inspected, which is formed by a pattern for position detection projected on the photosensitive substrate, and a coma-aberration giving optical system that generates comatic aberration in the image of the pattern to be inspected. A projection exposure apparatus configured to detect the position of the surface of the photosensitive substrate in the optical axis direction based on the asymmetry of the image of the test pattern.