JPS63224327A - Projection aligner - Google Patents

Abstract

PURPOSE:To always enable an accurate mark detection even if the mark for alignment is formed on the back of a sensitive substrate by mutual calibration by providing a focal point detection system for the surface of the sensitive substrate and a focal point adjusting system for the back of the sensitive substrate. CONSTITUTION:The focal point of an alignment system 20 which observes the back of a wafer W is adjusted automatically according to the movement of the a Z-stage 9 adjusted by the focussing operation for the surface of the wafer W which is to be exposed. Accordingly, detecting the focal point according to the picture signal by the alignment system 20 can be omitted. This can improve the precision of alignment, can make matching between alignment systems easy and an aligner which has an alignment unit which has the highly free selection of a mark position can be obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field of the Invention) The present invention relates to a projection exposure apparatus that prints a necessary pattern on, for example, a resist layer on a semiconductor wafer, and particularly relates to an improvement in a focusing device thereof. It is for.

(Background of the Invention) In a conventional step-and-repeat projection exposure apparatus, alignment marks are generally formed on the surface of a semiconductor wafer that is to be exposed, that is, on the side where a resist layer is formed and where light is irradiated. There is. Then, using this mark and an alignment mark formed on the reticle or mask, the semiconductor wafer and the reticle (or mask) are aligned.

However, in such an alignment method, the alignment marks are observed through a resist layer coated on the wafer surface, making it difficult to observe them clearly, resulting in an increase in alignment errors. Particularly in recent years, as the degree of integration of integrated circuits has improved, patterns have tended to become increasingly finer, and there has been a desire to reduce such alignment errors.

Furthermore, as the wafer surface is planarized, there is a possibility that marks on the surface may become undetectable.

Furthermore, when performing die-by-die alignment or each-shot alignment as such alignment, a device such as an alignment optical system must be provided on the reticle. Therefore, in order to prevent the optical system of the apparatus from interfering with the illumination light for exposure, it is necessary to move it in and out in steps or to set a die-by-die mark at a fixed position.

On the other hand, if such alignment devices are separated, it becomes necessary to perform matching between each alignment system.

(Object of the Invention) The present invention has been made in view of the above points, and is an alignment system that improves alignment accuracy, facilitates matching between alignment systems, and provides a high degree of freedom in selecting mark positions. An object of the present invention is to provide an exposure apparatus having a device.

(Summary of the Invention) (Examples) Examples of the present invention will be described below with reference to the accompanying drawings. FIG. 1 shows the overall configuration of an embodiment of the present invention. In this figure, a reticle R on which a pattern to be exposed is formed is held by a reticle holder 1, and this reticle holder 1 is supported by a column 2 at an appropriate position. The reticle R is provided with alignment marks sx, sy, and sθ, respectively, and the reticle holder 1 is configured to be movable relative to the column 2 by a drive unit 3.

A projection lens 4 that is telecentric on both sides or on one side is arranged below the column 2, and the wafer W is arranged at an optical position that is conjugate with the reticle R of this projection lens 4.

This wafer W is placed on a glass plate 5, and the glass plate 5 is further supported by a θ table 6. The θ table 6 is configured to be slightly rotatable about a rotation center 6a, and is driven by a θ table drive section 7, and the rotation angle is read by a θ angle reading encoder 8. The glass plate 5 described above is provided with a glass block 5A as an optical path length correction means at an appropriate position, and a fiducial mark FM is formed on the surface of the glass block 5A.

This part will be explained in detail later.

The θ table 6 is arranged on a Z table 9 that can be slightly moved in the Z direction, that is, in the vertical direction. This movement of the Z table 9 is performed by a Z table driving section 11 under the guidance of a Z direction movement guide roller 10.

The aforementioned Z table 9 is placed on the Y stage 12 via a Z direction movement guide roller 10. This Y stage 12 is capable of rectilinear movement in a direction perpendicular to the plane of the drawing, and is driven by an X stage drive section 13.

Y stage 12 is provided on X stage 14. This X stage 14 moves in the left and right directions of the figure, that is,
It is configured to be movable by an X stage drive unit 15 in directions orthogonal to the respective directions. This X stage 14 is
It is placed on a surface plate or column base 16.

At approximately the center of the upper surface of the column base 16, an objective optical system 1 is provided between the glass plate 5 and the direction of the projection lens.
7 is fixedly provided. This objective optical system 17 is
This is for detecting marks formed on the back surface of the wafer W. Furthermore, this objective optical system 17 is capable of detecting marks Sθ on the reticle R when the wafer W is not on the glass plate 5.
, 5x1SY are arranged coaxially with the optical axis AX of the projection lens 4 so that each projected image of 5x1SY can be observed simultaneously.

Further, on the bottom side of the column base 16, alignment sensor sections 20θ and 20 are provided corresponding to the viewing area of the objective optical system 17.
An alignment system 20 including X (not shown) and 20Y is provided. These alignment sensor parts 20θ, 2
0X and 20Y are marks formed on the back surface of the wafer W, reference marks FM, or marks Sθ and sX of the reticle R, respectively.
, SY photoelectrically detects the projected images and detects the deviation between these marks and a predetermined detection center. Since this alignment system 20 is fixed to a surface plate, that is, a column base 16, it is hardly affected by vibrations of the stage or the like. Therefore, alignment can be performed with high precision.

Next, a movable mirror 30 is provided on the side of the above-mentioned Z stage 9, while a fixed mirror 31 is fixed to the lower part of the lens barrel of the projection lens 4. The movable mirror 30 has a mirror 32
The light from the interferometer 34 is incident on the fixed mirror 31 via the beam splitter 33, and the light from the interferometer 34 is incident on the fixed mirror 31 via the beam splitter 33. That is, the light from the interferometer 34 including the laser light source is incident on the movable mirror 30 and the fixed mirror 31, respectively, and the coordinates of the wafer W are determined by the Y and X stages 12 and 14 using the interference of the respective reflected lights. The value is now measured.

In addition, the alignment sensor portions 20θ, 20X,
20Y are each connected to the alignment processing section 40. The alignment processing unit 40 determines the amount of position correction of the wafer W in the θ, x, and y directions based on alignment signals from the alignment sensor units 20θ, 20X, and 20Y, and also determines the amount of position correction in the θ, x, and y directions of the wafer W based on the contrast of the observed marks and patterns. It also performs arithmetic processing for focus detection.

Further, in this embodiment, an oblique incident light type focus detection device is provided which is used to align the surface of the wafer W with the imaging plane of the projection lens 4 (a plane conjugate with the reticle R) during exposure. As is well known, this includes a light projector 45 that projects a slit image or the like obliquely onto the surface of the wafer W, and
A light receiving unit 46 receives the reflected light at a fixed position, and photoelectrically detects the received slit image reflected light and outputs a focus detection signal according to the position of the wafer W in the direction of the optical axis AX of the projection lens 4. It is composed of a focus detection system 47.

Next, the drive unit 3, the θ table drive unit 7, the angle reading encoder 8, the Z table drive unit 11, the Y stage drive unit 13, the X stage drive unit 15, the interferometer 34, and the alignment processing unit 40 are all It is connected to the control device 50. This main controller 50 (a) controls the drive section 3 during alignment of the reticle R; (b) controls θ during global alignment of the wafer W;
Control by the table drive section 7, angle reading encoder 8, Y stage drive section 13, X stage drive section 15, and interferometer 54, (C) During alignment for each exposure shot (so-called die-pie-die alignment) Z table drive unit 1
1. Control by Y stage drive section 13, X stage drive section 15, interferometer 34, and alignment processing section 40; (d) Z table drive section for absolute focusing and calibration between reticle R and wafer W; ll,
It supervises the control by the oblique incidence focus detection system 47 and the alignment processing section 40, etc.

FIG. 2 shows an example of a projected image of the marks 5xSsy, sθ on the reticle R on the wafer W. In this figure, the inner frame represents a small shot size SS, and the outer frame represents a large shot size LS. Note that the shot size is the size of one pattern to be exposed. In both cases, the shot center or reticle center SC is shown together.

Mark images sxs, sys, sθS are shot size S
The mark images SXL, SYL, and SθL are the projection images of the marks SX and 5YSSθ of the reticle R at the shot size LS, and the mark images SXL, SYL, and SθL are the projection images of the marks 5xSsy, 5YSSθ of the reticle R at the shot size LS.
This is a projected image of sθ.

The x and y coordinates in this figure are running coordinates on the wafer stage.

Next, referring to FIG. 3, the glass block 5A and fiducial mark FM (fiducial mark) of the glass plate 5 shown in FIG. 1 will be explained.

The glass block 5A is provided on the glass plate 5 at a position other than the wafer mounting surface, and the surface height of the glass block 5A is higher than the surface of the wafer W placed on the glass plate 5 or the exposure surface. almost corresponds to the height of The fiducial mark FM is formed of a material such as chromium on the glass block 5A.

In FIG. 3, the glass block 5A has a circular plane, and a set of fiducial marks FMx, FMV and a focus check grating correspond to the running coordinates x, y of the wafer W shown in FIG. shaped patterns FCx and FCy are respectively formed. The fiducial mark FMx consists of two parallel lines extending in the y direction, and the fiducial mark FMy consists of two parallel lines extending in the y direction.
It consists of two parallel lines extending in the direction.

In addition, when the glass block 5A is equipped with the oblique incidence focusing device of this embodiment, the glass plate 5 can be moved up and down, so that it can be brought into exact alignment with the imaging plane of the projection lens 4.

Next, referring to FIG. 4, the alignment sensor section 2
The configurations of 0X, 20Y, and 20θ will be explained in detail.

Note that since both have the same configuration, the alignment sensor sections 20Y and 20θ will be explained as a representative.

As shown in FIG. 1, the imaging plane FP of the objective optical system 17 coincides with the back surface of the wafer W, and this imaging plane FPo
(or a position conjugate thereto) objects P1 and P2 are transferred to the image plane FP through the objective optical system 17, and an image P1,
The image is formed as P2''.

The objects PI, Pg include marks formed on the back surface of the wafer W, reference marks FMx, FMy, FCx, FCy, or marks SYL, SYS, SθL, shown in FIG.
SθS correspond to each other. These images P1°, p, l are
It is formed by light (for example, the same wavelength as the exposure light) from the light sources 205a and 205b.

The light from the light sources 205a and 205b is transmitted to the shutter 204a.
, 204b into the lens system 203a. Furthermore, a half mirror 202a and a first objective lens 20
1 a (20l b), mirror 200
a (200b) so that the optical axis is bent in the direction of the telecentric objective optical system 17. After passing through the objective optical system 17, the bent light passes through the imaging plane F P
After being reflected by objects P1 and P2 located at position o and passing through objective optical system 17 again, mirror 200a (200b)
incident on . and mirror 200a (200b)
The optical axis is bent by the first objective lens 201a (20
1b) and is reflected by the half mirror 202a. The reflected light forms an image on an aperture plate 207a (207b) having a window APY (APθ) by a lens 206a movable in the direction of arrow AY for focusing.
The image is relayed again by the imaging lens 208a and is imaged on the imaging surface of the television camera 209a (209b).

Of these, the mirror 200a (200b) and the first objective lens 201a (201b) are movable as a whole in a direction along the image plane FP except during alignment. This is an operation performed in conjunction with changing the shot sizes LS and SS shown in FIG. In addition, the aperture plate 207a (2
The window APY (APθ) in 07b) defines the detection center of the alignment sensor section 20Y (20θ). The position of this window APY (APθ) and the object P2 (PI
) and the position of the image P2''(p,') are both conjugate, and the positions of the window APY (APθ) and the light receiving surface of the television camera 209a (209b) are also conjugate. .

By each of the above parts, the alignment sensor parts 20Y, 20
θ is configured. The same applies to the alignment sensor section 20X.

Further, each of the alignment sensor sections 20Y and 20X has a pattern provided on the reticle R and a reference pattern FCx.
, FCy, etc., and check the correct imaging plane (absolute imaging plane) FP of the reticle R based on the contrast of these patterns. Furthermore, during alignment to detect the mark pattern formed on the back surface of the wafer W, the alignment signal (image signal) is taken in so as to avoid blurring as much as possible. Therefore, in this embodiment, these alignment sensor sections 20Y, 20X
corresponds to the second position detection means of the present invention.

Next, referring to FIG. 5, an example of the arrangement of marks on the imaging plane FP0 described in FIG. 4 will be described.

This figure shows shot size LS marks SX, SY, on reticle R with wafer W removed from glass plate 5,
It is a projection of Sθ.

In FIG. 5, the center of the xy coordinate system corresponding to the running coordinates of the wafer stage coincides with the optical axis AX of the objective optical system 17. That is, the objective optical system 1 with respect to the xy coordinate system
7 is fixed, and even if the glass plate 5 or wafer W moves, the origin does not move with it.

In the image field IF, which is the field of view of the objective optical system I7, as explained in FIG.
When there is θL and the alignment is well performed, these images SXL, SYL, and SθL are detected by the detection windows APX and A of the alignment sensor sections 20X, 20Y, and 20θ.
P.Y.

It comes to the center of each APθ. And stage 12.14
By moving the images SXL, SYL, etc., parts of the fiducial marks FMx and FMy corresponding to the width LS of the fiducial marks can also be arranged so as to sandwich the images SXL and SYL, respectively. Also, near each mark sy, sθ, SX of the reticle R, there is a grid pattern RFC for focus check.
is formed, which is also observed simultaneously inside each of the windows APXSAPY and APθ.

Further, in this embodiment, as shown in FIG. 6(a), on the back surface of the wafer W, there are linear alignment marks wx, wy extending in the radial direction, corresponding to each shot area E on the front surface.
, wθ are formed in advance. These marks wx,,wy,
wθ is the mark SX, SY on the reticle R shown in FIG.
, Sθ so as to match the arrangement of the projected images SXL, SYL, and SaL. And these marks wxSwy,
A grid pattern WFc for focus check on the shot area E of the wafer W is also formed in association with each of wθ, and these patterns WF,c also correspond to the respective detection windows APX, APY, of the alignment sensor sections 20X, 20Y, 20θ
Observed within APθ.

Next, the operation of this embodiment will be explained with reference to the flowchart shown in FIG. In this embodiment, as shown in FIG. 6(b), alignment marks wx, wy, w are pre-marked on the back surface of the wafer W in correspondence with each shot area E.
It is assumed that θ and a focus check pattern WFc are formed. The flowchart in FIG. 7 mainly shows the operation of absolute value calibration of the oblique incidence focus detection system.

First, the reticle R is set at a predetermined position, and the reticle R is positioned as shown in FIG. 5 by the objective optical system 17 and the alignment system 20, with the wafer W not on the stage. Next, the glass block 5A of the glass plate 5 is moved into the field of view of the projection lens 4, and positioned so that the reference patterns FCX and Fcy for focus check can be observed by the alignment system 20 via the objective optical system 17 (stage 100). And a reference pattern RFc and a reference pattern FC for focus check provided on the reticle R.
The wafer stages 12 and 14 are used to position the wafers xSFCy and xSFCy so that they do not overlap each other within the detection windows APY and APX (step 101). The state at this time is as shown in FIG. Although FIG. 8 only shows the arrangement of the alignment mark SY for the y direction and the reference mark FMy, the same applies to the arrangement of the pattern RFc and the reference pattern FCy attached to the mark SY. Here, the television camera 2 first images the pattern RFc on the reticle R.
Based on the image signal from 09a, the lens system 206a is moved in the optical axis direction so that the projected image of the pattern RFc by the projection lens 4 has the best contrast. Performing the focusing operation of the alignment system itself based on the contrast from such a pattern RFc (grid-like) is
For example, it is as disclosed in Japanese Patent Application Laid-open No. 101540/1983. By the way, at this time, the pattern RFc of the reticle R must be illuminated, and there are two methods for doing so. One is to open the shutter 2043 in FIG. 4 and turn on the light source (laser light source or optical fiber) 205a.
The other is a system that uses exposure light LB that illuminates the entire surface of the reticle R or only the mark portion, as shown by the arrow in FIG. When the light source 205a is used, the television camera 209a receives the reflected light from each of the pattern RFc and the reference pattern FCy, and when the exposure light LB is used, the transmitted light from each of the pattern RFC% and the reference pattern FCy is received by the television camera 209a. It will receive light. In such contrast-based focus detection, it is desirable to use transmitted illumination in terms of detection accuracy and S/N ratio. As described above, the pattern RFC by the lens system 206a and the television camera 209a (that is, the window APY
), a focus check is similarly performed for the reference pattern FCy (step 1).
02). Since the reference pattern FCy is moved up and down along the optical axis AX by the Z stage 9, the drive unit 11 is operated to adjust the height position of the Z stage 9 so that the contrast of the reference pattern FCy is maximized.

As a result of the above, the pattern imaging plane (Fpo) of the reticle R precisely coincides with the surface of the glass block 5A where the reference pattern FCy exists, and absolute focusing is completed. Now, in this state, since the slit image from the oblique incidence focus detection system (light projector 45, light receiver 46, etc.) is obliquely projected onto the surface of the glass block 5A, the surface of the block 5A (reference pattern surface) The height position is simultaneously detected by the focus detection section 47 (step 103).
. Therefore, the detection unit 47 detects the position of the reference pattern surface at that time.
, and store the position as offset Wk (
Step 104). The oblique incidence focus detection system can only detect the height position of the surface of the wafer W or the surface of the block 5A, and even if this focus detection system determines that the focus is in focus, it will not be accurate to It is not fully guaranteed that they match. Therefore, if the offset amount is zero, the imaging plane and the focus detection position match.

The above operations complete the calibration of the oblique incidence focus detection system, but if the detected offset amount is excessively large, the slit image of the oblique incidence focus detection system may be Adjust the angle to correct the focus detection position along the optical axis AX, and repeat step 1.
03.104 should be executed. Furthermore, if the angle setting accuracy of the parallel plane glass is sufficient, even if the angle of the parallel plane glass is adjusted so that the reference pattern surface is determined to be in focus even after step 102 is completed. good.

Next, during an actual step-and-repeat exposure operation, a sequence such as the following steps 105 to 109 can be taken, for example.

First, the X, Y stage 12.14 holding the wafer W is stepped based on design values to align the circuit pattern area of the reticle R and the shot area on the surface of the wafer W (step 105). At this time, assuming that the global alignment of the wafer W has been completed, alignment by only stepping can be achieved within 1 μm. Therefore, alignment sensor parts 20X, 20Y, 2
Images of marks WX and WYSWθ formed on the back surface of the wafer W are formed in each of the detection windows APX, APY, and APθ of 0θ. At this time, the marks WX, WY, and Wθ are illuminated by the light sources 205a and 205b, so the shutter 204a.

204b is opened. This state is as shown in FIG. 9, and although the objective optical system 17 is shown to be observing only the mark WY through the glass plate 5, the objective optical system 17 itself is not able to observe the marks wy, wx, and Wθ. At the same time, the image plane FP,
image is formed. Of course, the image of the focus check pattern WFc attached to the marks wy, wx%Wθ is also included in each detection window APX.

It is formed within APY and APθ. Furthermore, when the surface of the glass block 5A and the image plane FP are made conjugate as shown in FIG. 8, and the glass plate 5 is moved horizontally to form the image plane FP as shown in FIG. The surface is determined to be the front surface of the glass plate 5, that is, the back surface of the wafer W.

Now, before detecting the alignment mark in the state shown in FIG. 9, check the focusing state between the back surface of the wafer W on which the pattern WFc is being detected and the image plane of the television camera 209a (209b). Step 106). Here too, the lens system 206a (20
6b) Set 1[1]. When the focusing operation for the pattern WFc is completed, the amount of positional deviation between each mark wx, wySwθ and the windows APX, APY, APθ is determined based on the image signal, and the X and Y stages 12 are adjusted so that the amount of deviation becomes zero. .14 position is slightly moved (step 107).

Then, the height position of the wafer surface is detected by the oblique incidence focus detection system, and 2
The height position of the stage 9 is corrected to focus the circuit pattern image of the reticle R on the surface of the wafer W (step 10).
6). As a result, one shot right page area E on the wafer W
The wafer W is exposed to light (step 108), and the wafer W
If the exposure of the entire upper surface has not been completed, the same operation is repeated from the previous step 105 based on the determination at step 109.

As described above, in the sequence shown in FIG. 7, the mark position detection operation was performed for each shot area immediately before exposure, but only some marks on the back surface of the wafer W were detected in advance.
The shot arrangement on the surface of the wafer W may be estimated and the shot address design value during stepping may be corrected.

According to this method, an alignment mark detection operation is not required when exposing each shot area, so it is advantageous in terms of throughput.

Next, a focusing sequence according to another embodiment of the present invention will be described with reference to FIGS. 10 and 11. 10th
As shown in the figure, for example, if minute dust (for example, about 1 to 5 μm) PT is caught between the wafer W and the glass plate 5, the wafer W will be warped with respect to the plate 5. Even in this case, according to the previous embodiment, since the lens system 206a is adjusted by the focusing operation of the alignment system 20 during alignment, the image of the mark WY is formed in focus in the window APY. In other words, the conjugate image plane of mark WY is FP,
to FP,', and an image of mark WY is formed at point P2' in the image plane FP,°. However, the focus detection method of the alignment system 20 of this embodiment is based on the grid pattern W.
Since the method involves looking at changes in the contrast of Fc, there is also the problem that the focus detection operation takes time. In particular, when the dust PT gets stuck as shown in FIG. 10, the area near it will be in a large defocus state, and it will take time for the focusing operation to occur.

Therefore, in this embodiment, the lens system 206 of the alignment system 20 is adjusted based on the adjustment amount during the focusing operation by the oblique incidence focus detection system. In this way, high-speed focusing operation is possible even when the wafer W has large warpage or waviness.

First, as shown in FIG. 1O, a control section 150 is provided for detecting the amount of movement of the Z stage 9 (glass plate 5) and operating a motor 151 from a drive section 11 for moving the Z stage 9 up and down. The motor 151 moves the lens system 206a in the direction of arrow AY. The operation of this device is performed as shown in the flowchart diagram of FIG. First, after calibrating the oblique incidence focus detection system using the glass block 5A (at the completion of step 104 in FIG. 7), the control unit 150 stores the height position of the Z stage 9 as an initial value (step 160).
. During actual alignment (or exposure), the slit image projection light LF from the oblique incidence focus detection system is transferred to the wafer W.
, the reflected light LF' is photoelectrically detected, and the first
The Z stage 9 is moved until focus is determined by the focus detection unit 47 shown in the figure, and the control unit 150 reads the height position of the Z stage 9 at the time of completion of movement, and stores the amount of movement ΔZ from the initial value. In FIG. 1θ, the focus detection unit 47 determines that the focus is on the plane pw, but if there is no dust PT and the surface of the wafer W is in close contact with the plate 5, it is determined that the focus is on the plane PWz. That is, due to the warping caused by the dust PT, the focus is detected to be shifted from the initial value by the difference ΔZ between the planes PW and PWI.

Next, it is determined whether ΔZ is larger than a predetermined allowable amount DC (step 162), and if it is larger, the lens system 206a is moved by the motor 151 so that ΔZ is corrected on the window APY side (step 162). 163).

The allowable amount DG is determined depending on, for example, the amount of pattern blur allowed on the television camera of the alignment system 20, the amount of thickness unevenness (tapering) of the wafer W, and the like.

Next, a more precise focus check is performed based on the image signal of the alignment system 20 (step 1
64). However, this step 164 may be omitted.

According to this embodiment, the focus adjustment of the alignment system 20 for observing the back surface of the wafer W is automatically performed according to the movement amount of the Z stage 9 adjusted by the focusing operation on the front surface of the wafer W to be exposed. This makes it possible to omit focus detection based on the image signal from the alignment system 20, and extremely high-speed focusing and alignment can be achieved. The initial set operation in step 160 involves, for example, positioning the Z stage 9 so that the oblique incidence focus detection system focuses on the center of the wafer W, and adjusting the lens system so that the contrast detection by the pattern WFc is optimal. The height position of the Z stage 9 may be stored as an initial value while adjusting the height 206a.

(Effects of the Invention) According to the present invention, a focus detection system for the front surface of the photosensitive substrate and a focus adjustment system for the back surface of the photosensitive substrate are provided, and mutual calibration is performed. Even when a mark of

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the configuration of a projection exposure apparatus according to an embodiment of the present invention, FIG. 2 is a plan view showing the arrangement of projection positions of each pattern on a reticle, and FIG. 3 is a reference mark on a glass block 5A, FIG. 4 is a plan view showing the arrangement of the reference pattern, FIG. 4 is a diagram showing the configuration of the alignment sensor section, FIG. 5 is a diagram showing the arrangement of each pattern of the reticle observed on the image plane of the objective optical system, and FIG. FIG. 7 is a flowchart explaining the calibration operation according to the first embodiment; FIG. 8 is a diagram showing the state during calibration; FIG. 9 is a diagram showing the state at the time of alignment, FIG. 10 is a diagram showing the configuration according to the second embodiment, and FIG. 11 is a flowchart diagram explaining the focusing operation according to the second embodiment. (Explanation of symbols of main parts) R...Reticle, W...Wafer, 4...
・Projection lens, 5...Glass plate, 5A
...Glass block, 9...Z stage, 17...
・Objective optical system, 20... Alignment system, 45.4
6.47...Oblique incidence focus detection system, 206a...Focusing lens system, 209a, b...TV camera.

Claims (1)

[Scope of Claims] An apparatus for illuminating a mask having a predetermined pattern and projecting and exposing an image of the pattern onto the surface of a photosensitive substrate via a projection optical system, wherein the back surface of the photosensitive substrate is provided with a positioning plate in advance. a transparent holding means for mounting the photosensitive substrate so as to hold the back surface on which the mark is formed; a first member for detecting the surface position of the photosensitive substrate with respect to the optical axis direction of the projection optical system; a second position for detecting a mark formed on the back surface of the photosensitive substrate via the transparent holding means and detecting the position of the back surface of the photosensitive substrate with respect to the optical axis direction of the projection optical system; A projection exposure apparatus comprising: a detection means; and a reference pattern member formed on a transparent portion of the holding means for associating the first position detection means with the second position detection means.