JPS60262423A - Position detector - Google Patents

Abstract

PURPOSE:To detect a position rapidly, by moving a stage linearly in the scanning direction which intersects the X direction and by detecting registering marks in the X and Y directions simultaneously. CONSTITUTION:A control unit 17 moves an optical axis obliquely in the 45 deg. direction along the line connecting the center CC of a chip and the point of intersection P0 of line segments l1 and l2. When a registering mark M passes across the light point LA along the scanning track SC, a photoelectric signal SA has the peak value. A processing circuit 16 processes a position data from a length measuring device 3 and the signal SA to detect the peak position coordinates x0 and y0. Thus, the positional relation between the optical axis and the point P0 can be determined. The processing circuit 16 outputs to the control unit 17 while the position of the mark M (x0+DELTAx, y0+DELTAy) where DELTAx and DELTAy are predetermined offsets for the mark M is set as the target value. The control unit 17 servo controls a table 2 such that the positional data PD reaches the target value. Consequently, the optical axis of the lenses is allowed to pass the center CC of the chip correctly and a wafer is aligned precisely. According to this construction, the marks Mx and My can be detected simultaneously, and therefore the position can be detected rapidly.

Description

Detailed Description of the Invention (Technical Field of the Invention) The present invention relates to an apparatus for detecting the position of a substrate having alignment marks, and the present invention relates to an apparatus for detecting the position of a substrate having alignment marks. ) related to a position detection device.

(Background of the Invention) The miniaturization of large-scale integrated circuit (LSI) patterns is progressing year by year, and reduction projection exposure equipment has become popular as a device for printing circuit patterns that meet the demands for miniaturization and has high productivity. I've been doing it. In these conventionally used devices, a reticle pattern that is several times (for example, 5 times) the size of the pattern to be printed on the silicon wafer is reduced and projected by a projection lens, and the pattern that is printed in one exposure is This area is smaller than a square with a diagonal length of 21 mrn on the wafer. Therefore, in order to print a pattern on the entire surface of a wafer with a diameter of about 125M1, a so-called step-and-repeat method is used in which the wafer is placed on a stage, moved a certain distance d, and exposed repeatedly.

In LSI manufacturing, patterns of several layers or more are sequentially formed on a wafer, but if the overlay error (positional deviation) of patterns between different layers is not kept below a certain value, conductivity between layers or 18 The edge state is no longer the intended one, and the LSI is no longer able to perform its functions. For example, for a circuit with a minimum line width of 1 μm, at most 0.2
Only a positional deviation of 8 degrees μm is allowed.

The reduction projection exposure method uses a method of overlapping patterns,
That is, one method of overlapping the projection of a pattern on a reticle with a pattern already formed on a wafer is the through-the-lens (TTL) method. General alignment using the T'f'L method is performed by detecting a pattern already formed on the wafer, that is, two marks attached to the chip, through a projection lens. For example, the two marks are aligned with the two-dimensional alignment direction of the wafers,
It is formed of a linear pattern extending long and thin in the X direction, and a linear pattern extending long and thin in the X direction perpendicular to the X direction, each of which is provided in a different portion of the periphery of the chip. These two marks are each detected by separate alignment optical systems via projection lenses. The actual procedure for aligning the wafer is to approximately position the chip to be aligned within the projection area (image field) of the projection lens, and then
After positioning the stage in the X direction so that the position of the mark extending in the X direction is aligned by one alignment optical system, the position of the mark extending in the X direction is aligned in the X direction by one alignment optical system. It has been common practice to position the stage in the X direction so as to achieve alignment. In this way, although it is possible to obtain good results in terms of accuracy by detecting the position of Ueno in the X direction (!: y direction) using an independent dedicated mark, Since detection, that is, two position detection operations are required, there is a drawback that there is a limit to the speeding up of alignment.

(Object of the Invention) An object of the present invention is to overcome these drawbacks and provide a high-speed position detection device that does not require separate position detection operations in the X direction and the X direction of the substrate.

(Summary of the Invention) (5) The present invention provides a method for determining the position of a substrate having a mark Mx for positioning in the X direction and a mark My for positioning in the X direction perpendicular to the X direction in different parts. The detection device includes a stage (2) that holds the substrate and moves in two dimensions in the X direction and the
, or P. ), the mark Mx can be detected and the stage is positioned at a predetermined second position fit (
Mark detection means (spot lights Lkx, LAy, and 4 to 15) capable of detecting mark My when positioned at position P i or Po), and a stage that detects the first position and second position when detecting the position of the substrate A control means (control device 17. motor 20) for moving the stage in an exploration direction intersecting the X direction at a constant angle so that the mark M The technique is to provide means (alignment signal processing circuit 16) for detecting the position It(xo) of the substrate in the X direction and the position (yo) of the substrate in the X direction when the mark My is detected. This is the main point.

(6) (Embodiment) FIG. 1 is a diagram showing a schematic configuration of a reduction projection type exposure apparatus suitable for an embodiment of the present invention. Reduction projection lens (hereinafter referred to as
1 (simply referred to as a projection lens) exposes an image of a circuit pattern formed on a reticle R onto a wafer W by 115 degrees, and the father reduces the image to 1/10.

The reticle R is placed on a reticle stage (not shown), and this reticle stage is driven by a drive unit (not shown)
direction, and the θ (rotation) direction.

Then, the reticle R uses an alignment visualization (not shown) to
For example, it is aligned (positioned or positioned) at a predetermined position with respect to the optical axis AX of the projection lens 1. Further, the reticle R is illuminated with exposure light containing effective wavelengths (eg, g-line and i-line) for exposing the resist coated on the wafer W. The luminous flux EL that forms one pattern on the reticle from one illumination of this exposure light is the wafer W.
A crystal appears on the surface. On the other hand, this wafer W is placed on a stage 2 that moves two-dimensionally in the X and Y directions. stage 2
Although not shown, there is a Z stage section for moving the wafer W up and down, and a Z stage section provided on the Z stage section for moving the wafer W up and down.
The two-dimensional movement of the stage 2, which has a θ table that minutely rotates the For example, 0.
It is constantly detected with a resolution of 0.02 μm.

Next, the alignment optical system for positioning the wafer W will be explained. The laser beam from the laser light source 4 is expanded to a predetermined beam diameter by the beam expander 5, and shaped into an elongated elliptical beam by the cylindrical lens 61. This shaped laser beam is reflected by mirror 7, passes through lens 8, beam smitter 9, and lens 10, and is reflected upward from the lower surface of reticle R by mirror 11. The laser beam from the mirror 11 once converges into a slit shape, and then the reticle IR
Below the reticle l, a mirror 12 having a reflection plane parallel to the reticle l is reached, where the laser beam is reflected towards the entrance pupil 1a of the projection lens 1.

The laser beam that has passed through the projection lens 1 is turned into a long and narrow strip-shaped spot light L on the wafer W by the action of the cylindrical lens 6.
The image is formed on Ay. The wavelength of this spot light is determined so as not to expose the resist on the wafer W. Since a positioning mark (alignment mark) is previously formed on the wafer W, when the spot light LAy irradiates this mark, scattered light or diffracted light is generated from the mark.

In this example, since the mark is a diffraction grating pattern, the mark mainly reflects the specularly reflected light of the spot light LA4 (
0th-order diffracted light) and diffracted light (first-order light or higher) are generated. The optical information from these marks enters the projection lens 1 in the opposite direction, passes through the incident ff1la, is reflected by the C mirror 1211, passes through the lens 10, is reflected by the beam splitter 9, and reaches the spatial filter 13. The spatial filter 13 is conjugate with the entrance pupil 1a of the projection lens, and blocks only specularly reflected light (0th-order diffraction source) from the reduced surface of the wafer W. The diffracted light (scattered light) from the surface (mark) of the wafer W is displaced in the optical path I of the specularly reflected light depending on the spatial frequency. (9) Here, the spatial filter 13 passes only the diffracted light and scattered light, and the condenser lens 14 condenses the diffracted light and scattered light onto the light receiving element 15 as a photoelectric detector.

The light-receiving element 15 outputs a photoelectric signal SA corresponding to the intensity of the diffracted light or scattered light, and this photoelectric signal SA is input to an azimuth signal processing circuit (hereinafter simply referred to as a processing circuit) 16.

The processing circuit 16 also inputs the position information @ (time-series up/down pulse signal or parallel digital signal) pD from the measuring device 3, and calculates the generated value It of the photoelectric signal SA according to the diffracted light from the mark. (travel position) is detected. Specifically, the photoelectric signal SA is sampled by up and down pulse signals generated at the unit movement distance (0.02 μm) of stage 4, and each sampling value is converted into a digital value and stored in memory in address order. After that, the moving position of the mark is detected by predetermined calculation processing. The control device 17 controls the drive unit 20 based on the detected mark position information.

If the mark on the wafer W is attached to each of a plurality of chips on the wafer, the center of the chip and the optical axis AX can be determined by detecting the position of each center (10). and can be accurately aligned.

Figure 2 shows the circular image field if of the projection lens 1.
FIG. 3 is a plan view showing the arrangement relationship between the light beam LAY and the spot light LAY. In order to simplify the explanation, only one set of alignment optical systems is shown in Fig. 1, but in reality, another set of alignment optical systems with a similar configuration is installed in the direction perpendicular to the plane of the paper in Fig. 1. It is being The optical axis AX of the projection lens 1 is the orthogonal coordinate system x
When determined to pass through the origin of y, the spot light LAy is elongated in the X direction on the X axis as shown in Figure 2, and the spot light LAx from the alignment optical system of another 1 m is
It is formed to be elongated in the direction. Although the respective spotlights do not necessarily have to be located on the X and y axes, in order to simplify the explanation, it is assumed here that they are aligned on the X and y axes.

Spora) LAY is used to detect the position of the alignment mark extending in the X direction on the wafer W, and the spot light L
A'X is used to detect the position of an alignment mark extending in the X direction on the wafer W in the X direction.

Also, as is clear from FIG. 2, the spot light LAy,
The position of LAx is determined within the image field if and outside the rectangular pattern projection area PA inscribed therein. In addition, in this embodiment, the distance mDy from the center position of the spot light LAx in the X direction to the optical axis AX is
and the optical axis AX from the center position of the spot light LAy in the X direction.
It is assumed that the distance pliDx is known in advance at the time of manufacturing the device. Note that if the spot lights LAx and LAy do not align on the y-axis and the X-axis, respectively, the spot light L
A line segment passing through the center of x and parallel to the y axis and spot light L
Intersection point P where a line passing through the center of AY and parallel to the X axis intersects at right angles
Each distance Dy from P to the spot lights LAx and LAy,
J) It is sufficient if x is known in advance. FIG. 3 is a plan view showing the arrangement of marks on the uriever W suitable for alignment by this apparatus. A plurality of chips CP are arranged in a matrix on the wafer W, and each chip has an X-direction alignment mark Mx extending in the X-direction and an X-direction alignment mark Mx extending in the X-direction as shown in FIG. An alignment mark MY is provided at a different portion. When the center 00 of the chip CP coincides with the origin of the coordinate system xy, the mark Mx is
The mark My is provided at a position ΔX away from the X axis in the X direction around the chip CP. Furthermore, in this embodiment, a line segment 10 passing through the center of mark Mx and parallel to the y-axis, and mark My
Let P be the orthogonal intersection of the line segment l, which passes through the center of Xs and is parallel to Xs. When it is closed, the intersection point P. Distance O from to mark Mx
y is the optical axis AX of the spot light L A x (or the intersection PP
) is set equal to the distance mDy from the intersection point P. Kara mark M
The distance OX to y is determined to be equal to the distance 14IDx of the spora) LAy from the optical axis AX (or the intersection PP).

Now, FIG. 4 is an enlarged plan view of the representative mark Mx among the marks Mx and My. The mark Mx is formed as a diffraction grating pattern in which minute d elements having step edges extending in the X direction or dots are regularly arranged in the X direction at a constant pitch (13). Therefore, when the spot light LAx and the mark factory overlap, the 0th order diffracted light (regularly reflected light) and the higher order (±
1. ±2...) diffracted light is generated. This higher-order diffracted light is received by a light receiving element similar to the light receiving element 15 shown in FIG. 1, and the light receiving element outputs a photoelectric signal depending on the light intensity (or light t). The light receiving element 15 shown in FIG.
Since the spot receives the higher-order diffracted light from the mark My due to the irradiation of the spot light LAy, the photoelectric signal is hereinafter referred to as SAy. Let's call it S A x. In this embodiment, in order to increase the detection rate of the marks Mx and My, it is desirable that the length of the mark Mx (My) is made sufficiently longer than the length of the spot light LAx (LAy).

Next, FIG. 3 shows the position detection operation according to this embodiment.

The explanation will be made using Fig. 51(a) and fb) together with Fig. 4.

FIG. 5(al) and (b) respectively show +aC signals SAx, 8
This is a waveform diagram of Ay, where the horizontal axis represents the scanning position in the X direction or the X direction (14-), and the vertical axis represents the intensity of each photoelectric signal. First, the wafer W is roughly positioned on the stage 2 by an unillustrated realignment device ζ, and then placed thereon.

Furthermore, the two-dimensional positional deviation and rotation of the entire wafer W with respect to the optical axis AX of the projection lens 1 are referred to as the two-dimensional positional deviation and rotational deviation of the entire wafer W with respect to the optical axis AX of the projection lens 1.
Axis Alignment G#Global alignment is performed by scanning the mirror once. Through the above operations, the arrangement coordinate system of each chip on the wafer W and the moving coordinate system of the stage 2, that is, the xy coordinate system, are associated with each other in a unique relationship. Therefore, if the control device 17 reads the position information PD from the length measuring device 3 and positions the stage 2 by driving the drive unit 2° so that the position changes along the chip arrangement coordinates of the wafer W, the projection The optical axis AX of the lens 1 can be aligned so that it passes approximately through the center CO of one chip OP on the wafer W. Next, projection 1 of pattern area PA!
i! and its one chip CP are aligned in an n-close manner. For this purpose, the control device 17 adjusts the arrangement relationship between the chip OP and the two spot lights LA as shown in FIG.
The stage 2 is positioned so that it is shifted by a predetermined distance in an oblique direction (investigation direction) tilted by 5 degrees. In Figure 3,
By positioning the optical axis AX and the center of the chip CP,
The figure shows a state in which the two are roughly in agreement. From this state, the control device 117 moves the stage 2 linearly (stage scanning) in the exploration direction at an angle of 45° as indicated by arrow A in FIG. Specifically, the point on the chip CP through which the optical axis AX passes (the third
In the figure, approximately the center OO) and the line l□11. intersection point P. The stage is scanned along the line segment connecting the Accordingly, the mark Mx moves along the scanning trajectory (exploration trajectory) S01 so as to cross the spot light LAx at an angle of 45°, and the mark MY moves along the scanning trajectory (exploration trajectory) S02 so as to cross the spot light LAy. Move across at a 45° angle. At this time, in this embodiment, the arrangement relationship between the j marks Mx and My and the spot light LAx.

The arrangement relationship of LAy is congruent (Dy-Oy, Dx=Ox
), the time point at which the mark Mx overlaps the spot light IAx and the time point at which the mark My overlaps the spot light LAy almost coincide. However, strictly speaking, due to the accuracy of the pre-alignment and global alignment of the wafer W, the third
Since there is a slight positional deviation between the X direction and the
The time point at which LAy detects the marks Mx and My, respectively, will be slightly different in time depending on the slight positional deviation.

Now, when the mark Mx crosses the spot light LAx in the 45° direction as shown in Fig. 84, the photoelectric signal S A x is generated as shown in Fig. 5 (
As shown in a), a peak occurs depending on the intensity change of the higher-order diffracted light from the ζ mark Mx. Regarding the photoelectric signal SAy, the fifth
A similar peak waveform ζ is obtained as shown in Figure fb). The processing circuit 16 processes the stage 2 of the digit #11 PD from the sub-master unit 3.
The positional information in the X direction of 8 A x and the photoelectric signal S A x are processed to detect, for example, a position Xo in the X direction corresponding to the peak point of the photoelectric signal 8 A x. Similarly, the processing circuit 16 processes (17) the position information of the stage 2 in the X direction and the photoelectric signal SAy, and determines the position y in the X direction corresponding to the peak of the photoelectric signal SAy. Detect. The processing circuit 16 receives the photoelectric signals SAx, S
Since it has a circuit that samples and stores the magnitude of Ay for each unit of movement in the
and y. Both can be detected. This position X. is mark Mx
is the position in the X direction of the line segment l passing through the center of
o is the position in the X direction of the line segment passing through the center of mark My. Therefore, by the above position detection operation, the spot light L
The two-dimensional positional relationship of the chip OP with respect to Ax and LAy, that is, the intersection point P with the optical axis AX. This means that the positional relationship with n@ is associated with n@.

Now, the marks Mx and My are located at positions that have passed through the spotlights LAx and LAy, respectively, due to stage scanning. Therefore, the processing circuit 16 uses the value of the detected position X0I3'0 as
The offset amount of the marks Mx and My predetermined in the design, that is, the offset amount of the marks Mx and My, that is, the amount It (X Q 1'
(18) Output to the control device 17 as the positioning target position for #. Then, the control device 17 servo-controls the stage 2 so that the position information PI from the length measuring device 3 becomes the target position. As a result, the optical axis AX of the projection lens 1 passes accurately through the center 00 of the chip OP, and perfect alignment of the lens J is achieved. Thereafter, the reticle R is irradiated with exposure light, and the pattern is superimposed on the chip OP on the Rinino W and exposed.

As described above, in this embodiment, the stage is scanned at an angle of 45° with Dy-0y and Dx=Ox, so the mark Mx
, My can be detected almost simultaneously.

Therefore, there is an advantage that the scanning distance (mark search distance) for stage scanning at an angle of 45° can be shortened, and the time required for positioning can be extremely shortened.

Similarly, if Dy=Oy and Dx-Ox, it is possible to simultaneously produce marks Mx and My even if the stage scanning direction is set to a direction other than 45°.

Next ζζ A second embodiment of the present invention will be described based on FIG. The second embodiment is characterized in that the stage scanning direction in the position detection operation is made to coincide with the stepping direction for exposing the missing chip. In other words, instead of stepping the stage 2 only in the X direction and the 1y direction with respect to a plurality of chips arranged in a matrix on the wafer W, the stage 2 is stepped to expose a chip located diagonally adjacent to the chip being exposed. This allows stepping. FIG. 6 shows the distribution tf relationship immediately after exposure of a specific nap on the wafer W and the pattern projection axis PA in a superimposed manner. Chips OPl, OP,, OP on wafer W.

The mark Lvix i is shown in the same way as in Figure 3.
, My s marks Mx ta , My m, and marks Mx a , My a are provided. In this embodiment, the stage 2 is moved so that the center OC of the chip CP, which is located diagonally adjacent to the exposed chip r, coincides with the light @AX of the projection lens 1 (or the center point of the pattern projection 1 image PA). It is moved diagonally in a straight line along the scanning trajectory SCo. The distance of this scanning trajectory SCo can be determined in advance by a simple calculation because the arrangement pitch of the chips is known in the design. Each scanning trajectory 801 of the spot lights LAx and LAy when the stage 2 is moved along the scanning trajectory Sco.

80 is parallel to the scanning orbit SOo. At this time, the mark Mxs associated with the chip OP is positioned on the scanning trajectory SC1, and the mark My8 is positioned on the scanning trajectory 80. Marks attached to other chips OFl and OP are arranged in exactly the same manner. In addition, in this embodiment, the mark Mx 1 is set to the intersection Po of the line segment No. 1 passing through the center of the mark Mxa and the line segment ls passing through the center of the mark Mya.
1, My 11 position, i.e. distance Oy, Ox
However, unlike the case shown in Fig. 3, DyζOy, Dx+
It is defined as Ox. Furthermore, the center 0 of chip 0P11
The offset amount ΔX in the X direction between a line segment ll passing through 0 and parallel to the y-axis and mark Mx8, and the offset amount iY in the X direction between a line segment M passing through center 00 of chip 0P11 and parallel to the X axis and mark My8. Both are known in advance (by design).

Therefore, when the stage 2 is moved linearly along the scanning trajectory 80o for exposing the chip OP from the state shown in FIG. When positioned, the mark My a overlaps with the spot light LAY. Further, the stage 2 moves and the optical axis AX reaches a point P on the scanning orbit SOo.

When the mark Mxs is located at , the mark Mxs overlaps the spot light LAx. Therefore, the processing circuit 16 uses the spot light LAy to mark M.
Sampling of the photoelectric signal SAy is started from the time when the spot light LAy is located near the mark Mys, and sampling is stopped when the spot light LAy completely crosses the mark Mys.
The detection calculation process for the position ya in the X direction of s (scanning position 1 it) is started.

Further, the processing circuit 16 sets the position (ya + jY) obtained by adding (or subtracting) the detected position ys and the offset amount ΔY as the target stop position of the stage 2 in the X direction, and the control device 17
Output to. The same is true for detecting the mark Mxs, and the processing circuit 16 detects the mark Mxs when the spot light L A
Immediately after crossing X@, the detection calculation process for the position Xs of the mark Mx8 in the X direction is started, and the target stop position (Xs + jX) of the stage 2 in the X direction is output to the control device 17.

As described above, the spot lights LAx and LAy each have a mark (-
22) From the time when scanning of both Mx and My is completed, the stop target position of stage 2 (Xs+)X + Y a +i
It is desirable that the time required to calculate Y ) (hereinafter referred to as detection processing time) be sufficiently short. For example, the processing circuit 16
10m5ec by using a dedicated high-speed processor for
High-speed processing of orders is possible.

Now, when the target position #(3CB+) yB+jY), the stage 2 is positioned by servo control. By the above operations, the optical axis AX of the projection lens 1 (or pattern projection 1 #
! The center of PA) exactly coincides with the center 00 of the chip CP8, and the pattern projection image PA and the chip OP are superimposed closely. When this alignment is completed, the spot light LAx is located at the point Px, and the spot light LI
Ay is located at point py.

As described above, according to this embodiment, when stepping diagonally to expose the next chip, the spot light LA
Since the marks Mx and My are positioned on the respective scanning trajectories Sol and 80s of x and LAy, the stage 2 is not stopped between the mark position detection operation and the stepping for the exposure operation. The effect is that precise positioning can be achieved in an extremely short period of time.

Similarly, in this embodiment, as is clear from FIG.
Due to the arrangement relationship between x B and My B, mark MxB may be detected later than mark My8 in terms of time. Since the offset value of mark MxB is small, the detection processing time for the position (xa + ΔX) is shorter than the point PII to center 0
In this case, the optical axis AX of the projection/zoom 1 passes the center 00 of the chip OP8, or at a position 1d just before it. It is also necessary to stop stage 2. If it is not desired to stop the stage 2, the feed speed of the stage 2 may be slowed down at the time when the mark M x 11 is detected by the robot light LAx, or just before that, in accordance with the detection processing time. However, stopping or decelerating the stage 2 causes a reduction in the throughput of the exposure apparatus. Therefore, the arrangement of marks Mx and My is changed as shown in FIG.

FIG. 7 shows the mark MxB attached to chip CP8.

It is a layout diagram according to a third example of Mys. In this embodiment, the mark Mx is the same as the mark arrangement shown in FIG.
Each distance Oy, Ox of B, My B is Dy-0
y, Dx-Ox, and the intersection Po of line segments 111 and 1B
Marks Mx lI and MyB are arranged so that they coincide with the scanning trajectory SOo of the optical axis AX. That is, mark Mx
Spot light LAx (LA
y) to a position as far away as possible from the point px (Py) on the travel trajectory 5O1 (So,) (however, the position can be formed in conjunction with the chip OP).

In this way, the marks MxB and MYa are detected almost simultaneously, and the optical axis AX of the projection lens 1
The time from the intersection Po to the center CO of the chip OPB is longer than the detection processing time, and operations such as temporary stopping and deceleration of the stage 2 are not required. For this reason#
There is an advantage that the repeated operations of detection 2 alignment and exposure (25) are faster.

Next, the above 1. 21. The modified darkness common to the third embodiment will be explained as a fourth embodiment. In the valley embodiments so far, the marks Mx and My are in the form of a diffraction grating pattern, and are rectangular grating patterns in which the stepped edges of each grating that contribute to the generation of higher-order diffracted light extend in a direction perpendicular to the longitudinal direction of the mark. Ta. However, when spot light L
When scanning diagonally with Ax and LAy, the spot light LA
x, width of LAy (spot size), mark Mx, My
Depending on the width of the mark and the relationship with the angle of diagonal scanning, the relative positional relationship of the stepped edges of the grating to the diffracted light, which contribute to the generation of higher-order diffracted light, and the number of edges change during the mark set meal. SAx and SAy are modulated, and there is a possibility that the waveform of the photoelectric signal when the marks Mx and My are scanned in a direction perpendicular to their longitudinal direction is slightly distorted.

This waveform distortion causes a decrease in mark position detection accuracy. Therefore, in the fourth embodiment 1, the method (26) is to move the grating G having the step edge e extending in the direction of stage scanning in the X direction or in the X direction as shown in FIGS.
A diagonal lattice pattern arranged in the direction.

In FIG. 8, each of the marks Mx and My is an array of gratings G having edges e parallel to the scanning trajectories SO□ and SO2 of the spot lights LAx and LAy.
The shape of is a parallelogram. Moreover, Mark Mx, My
Both edges of the grating G in the detection direction (X or X direction) are determined to be parallel to the line segment IJx,ls passing through the center of the mark. In the diagonal lattice pattern of FIG. 9, the lattice G is simply changed to a rectangle. As in this embodiment, by forming an oblique grating pattern in which the direction of the edges e contributing to the generation of diffracted light of the grating G and the scanning direction of the stage 2 are made coincident with each other, the photoelectric signal can be modified as described above. It becomes possible to detect the position of the mark with extreme precision without receiving any interference. Also, when the oblique grating pattern L is rubbed, the diffracted light and scattered light generated from other patterns on the wafer W (circuit patterns, cross lines, etc.) and the diffracted light generated from the marks are filtered through the spatial filter 13.
There is also the advantage that separation becomes easier by optimally selecting the shape of the light shielding part.

As described above, in each of the first embodiments of the present invention, the photoelectric signal SAx
, SAy is sampled and digitally processed,
Although the positions of the marks Mx and My are detected, a circuit is provided that compares the photoelectric signals SAx and SAy with a predetermined slice level and binarizes them, and obtains a pulse signal when the spot light and the marks match. The same effect can be obtained by latching (temporarily storing) the position l1iffII information PD from the length measuring device 3 in response to this pulse signal. Also, the marks Mx and My do not need to be a diffraction grating pattern, and may be a simple piece of #l batter 7.

Furthermore, as shown in each embodiment, the present invention allows the position of one chip on the wafer W to be determined by the projection lens 1 without using the reticle R.
It is not limited to position detection using a single optical system.

For example, the same effect can be obtained by using an alignment optical system that detects the alignment state with the marks Mx and My on the mark wafer W provided on the reticle R, or a so-called die-pie-die alignment system. . In this case, mark illumination light (
When the wavelength of the spot light (corresponding to LA
, SO3) may be limited to a small area including the marks Mx and My. Furthermore, the present invention can be implemented not only for detecting the position of chips on the wafer W, but also for alignment of the entire wafer W (global alignment). For example, even if an off-axis two-season alignment wA microscope is used, which has detection centers in the X direction and X direction at positions separated from the optical axis AX of the projection lens 1 by a predetermined distance in the X direction and the good. In this case, two off-
Each detection center of the axis alignment microscope corresponds to the spot light L A x r LA y shown in FIG. Therefore, on the wafer W, marks for the X direction and marks for the X direction (2
9) It is sufficient to provide a mark for Then, the stage 2 is scanned diagonally, and when each detection center and the mark match, the positions of the stage 2 in the X direction and the X direction are latched. In this way, the two-dimensional position of the entire wafer W with respect to the detection center of the off-axis alignment microscope is defined. For off-axis microscopes, spot light LAx is used.

It is preferable to have an optical system that focuses an elongated spot light onto the wafer W as the detection center, similar to LAy.

Furthermore, although the spotlights LAx and LAy in each of the above embodiments are formed into elongated sheet-like spots, similar young fruits can be obtained even if they are formed into minute circular spots.

(Effects of the Invention) As described above, according to the present invention, each position of the X-direction alignment mark and the X-direction alignment mark can be detected in one scan, which is extremely fast. This is expected to have the effect of enabling accurate positioning operations. Therefore, in a step-and-repeat type exposure apparatus for manufacturing semiconductor elements, it is expected that the throughput of the apparatus will be high (30) and the productivity of semiconductor elements will be improved.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing a schematic configuration of a reduction projection type exposure device suitable for a position detection device according to the embodiment ζ of the present invention, and FIG. 2 is a plane diagram showing the arrangement of spot light within the image field of the projection lens Figure 3 shows the position detection operation according to the first embodiment. Figure 3 is a chip layout diagram. Figure 4 is an enlarged half-view showing the arrangement relationship between marks and spot lights. Section 5 shows photoelectric signals. 6 is a chip layout diagram showing the positioning operation by ZIJ in the second implementation, and FIG. 7 is a waveform diagram in 3'.
8 and 9 are enlarged views showing modified examples of marks as the fourth embodiment. Applicant: Takao Watanabe (31), agent of Nippon Kogaku Kogyo Co., Ltd.

Claims (2)

[Claims] (1) In an apparatus for detecting the position of a substrate, which has a mark Mx for alignment in the X direction and a mark My for alignment in the X direction perpendicular to the X direction in different parts, the substrate is held. a stage that moves two-dimensionally in the above direction and the X direction; when the stage is located at a predetermined first position, the mark Mx can be detected; and when the stage is located at a predetermined second position; a mark detection means capable of detecting the mark My; and a mark detection means capable of detecting the mark My in a search direction intersecting the X direction at a constant angle so that the stage passes through the first position and the second position when detecting the position of the substrate. control means for linearly moving the stage; while the stage is moving in the exploration direction, the mark detection means detects the mark M;
A position detection device comprising: means for detecting the position of the substrate in the X direction when detecting x and the position of the substrate in the Y direction when detecting the mark My. (2) The mark detection means includes a projection objective optical system, and a spot light LAx for detecting the mark Mx and a spot light LAy for detecting the mark My at a predetermined position within the field of view of the objective optical system. a spot light irradiation system that irradiates the substrate through the objective optical system, and the spot light LA;
When x irradiates HiJ mark Mx, the mark Mx
a first photoelectric detector that outputs a photoelectric signal according to optical information from the mark l)f, and when the spot light LAy illuminates the mark My, outputs a photoelectric signal according to the optical information from the mark Y; 2. The device according to claim 1, further comprising a second optical detector for detecting an IfG'F signal.