JPH0697046A - Exposing device - Google Patents

Abstract

PURPOSE:To set the true average surface of the exposed surface of each one of a plurality of shot regions on a sensitized board in parallel to the best image- forming surface of a projection optical system at each shot region, even if the exposed surface has protrusions and recessions. CONSTITUTION:Concerning to each of nine specific shot regions, the true average surface (inclination) of its exposed surface is found by detecting shifts from the best image-forming surface P1 at each of a plurality of points in the specific shot region, using a focus detection optical system in a state that the exposed surface accords with the best image-forming surface of a projection optical system 3 by the use of a leveling optical system. Next, an inclination (correction value for leveling) is calculated at each one of all the shot regions on wafers 4B and 4C, by performing statistical operation of an inclination at each specific shot region, and leveling is performed using this correction value at each shot region.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure apparatus having a leveling mechanism, for example.

[0002]

2. Description of the Related Art In a lithography process for manufacturing a semiconductor device or a liquid crystal display device, a reticle pattern is reduced to 1/5 through a projection optical system and sequentially transferred to each of a plurality of shot areas on a photosensitive substrate. A step-and-repeat type projection exposure apparatus (stepper) is used. Normally, in order to transfer a clear image of the reticle pattern onto the exposure surface of the photosensitive substrate (shot area), the exposure surface is set parallel to the image formation surface (focal plane) of the projection optical system in the projection exposure apparatus. It has a built-in leveling mechanism. This leveling mechanism is composed of an inclination detection optical system (leveling optical system) for measuring the inclination state of the exposure surface with respect to the image plane of the projection optical system, and a leveling stage for setting the inclination state of the photosensitive substrate to a desired state. Has been done.

FIG. 7A shows a main part of a projection exposure apparatus having a conventional leveling mechanism. In FIG. 7A, the reticle 1 is placed on the reticle holder 2, and the image of the pattern of the reticle 1 is transferred to one shot area on the wafer 4 via the projection optical system 3 under the exposure light L1. It The wafer 4 is held by the upper stage (holder) 5,
Further, the upper stage 5 is mounted on the lower stage 6 via three fulcrums. As shown in FIG. 7C, two fulcrums 5a and 5b of the three fulcrums 5a to 5c of the upper stage 5 can be moved up and down, and the remaining one fulcrum 5c is It is fixed. By adjusting the amount of movement of the two fulcrums 5a and 5b on the lower stage 6 side, the tilted state of the upper stage 5 can be set to a desired state, and the upper stage 5 and the lower stage 6 allow the leveling stage to move. It is configured. The stage having the above structure is disclosed in, for example, Japanese Patent Laid-Open No. 62-274201.

The lower stage 6 is a wafer stage 7
The wafer stage 7 is mounted on the wafer stage 7 and the wafer stage 7 can be moved two-dimensionally in a plane perpendicular to the optical axis of the projection optical system 3 and a Z stage that can be finely moved in the optical axis direction (Z direction) of the projection optical system 3. It is composed by. The two-dimensional position (coordinate position in the X and Y directions) of the wafer stage 7 is constantly detected by the laser interferometer 8, and the wafer stage 7 is positioned in the X, Y and Z directions by the driving means 9.

Now, the detection light L2 from the light source 10 for the leveling optical system is converted into a parallel light flux by the projection objective lens 11, and this parallel light flux is incident on the wafer 4 obliquely with respect to the optical axis AX of the projection optical system 3. Then, almost the entire exposed surface (surface of the resist layer) in the shot area is irradiated. Further, the light (parallel light flux) reflected obliquely with respect to the optical axis of the projection optical system 3 from the wafer 4 is focused on the light receiving surface of the photoelectric sensor 13 by the condenser objective lens 12. When the tilted state (tilt angle) of the exposure surface of the wafer 4 (shot area) changes, the center of gravity position of the light amount distribution on the light receiving surface of the photoelectric sensor 13 changes (shifts). The state can be detected (for details, see Japanese Patent Laid-Open No. 58-1137).
Disclosed in Japanese Patent Publication No. 06). The photoelectric sensor 13 is shown in FIG.
As shown in FIG. 3, the light receiving element is divided into four, for example.

Further, as shown in FIG. 7C, when the exposure surface of the wafer 4 having good flatness is parallel to the best image forming surface (best focus surface) of the projection optical system 3, the wafer 4 is
The photoelectric sensor 13 is positioned such that the light reflected from the detection area (corresponding to a shot area) 14 is focused on the center of the light receiving surface of the photoelectric sensor 13. After that, the tilted state of the exposure surface of the wafer 4 is adjusted via the leveling stages 5 and 6 so that the reflected light is focused on the center of the light receiving surface of the photoelectric sensor 13. The image plane and the exposure surface of the shot area substantially coincide with each other.

Although not shown in FIG. 7A, the projection exposure apparatus has a leveling optical system and a projection optical system at a predetermined measurement point (for example, the center of the shot) in the exposure surface of the shot area. A focus detection optical system for measuring the height (position in the Z direction) of the optical axis 3 is also arranged. Therefore, by driving the wafer stage 7 (Z stage) according to the detection signal from the focus detection optical system, the exposure surface (measurement point) of the shot area is set to be substantially the same height as the image forming surface. That is, the focus detection optical system and the Z stage constitute an autofocus mechanism.

[0008]

As described above, when the flatness of the exposure surface of each shot area on the wafer 4 is good, the exposure surface is formed by the leveling mechanism into the image forming surface of the projection optical system 3. Can be parallel to. However,
As a result of various processes, unevenness with a large height difference may occur on the wafer, and the light reflected from the wafer may not correspond to the tilted state of the exposed surface in the shot area due to uneven coating of the resist. Therefore, even if the leveling mechanism is operated, the exposure surface of each shot area may not be parallel to the image plane of the projection optical system 3, which is disadvantageous.

FIG. 8A shows a state in which the sawtooth-shaped irregularities are formed on the wafer 4A. As shown in FIG. 8A, when one shot area SA on the wafer 4A is irradiated with the detection light L2, even if the upper stage 5 is moved according to the detection signal from the photoelectric sensor 13 to perform leveling, The average surface of the exposure surface in the shot area (for example, a virtual surface defined in the approximate center of the concave portion and the convex portion in the shot area) and the best image forming plane P1 of the projection optical system 3 are not parallel. In practice, it is desirable that the exposure surface (average surface) of the shot area coincides with the best image forming surface P1 as shown in FIG. 8 (b).

Further, depending on the process (layer), the projection optical system 3 when the leveling mechanism is operated due to the mutual influence of the step due to the pattern and the uneven coating of the resist.
The tilt amount of the exposure surface of the wafer 4 with respect to the image forming plane may vary depending on the position within the wafer 4, that is, each shot area, as shown in FIG. FIG. 9A shows one shot row (seven shot areas) SA _{11 to} SA ₁₇ on the wafer 4, and FIG. 9B shows the running surface (moving plane) of the wafer stage.
Each exposure surface of P2 for seven shot area SA ₁₁ -SA ₁₇ shows a tilt after the leveling operation of the (average plane) P ₁₁ ~P _17. In such a case, it is impossible to match the exposure surface with the image plane for each shot area by adding a uniform offset to the leveling result for all shot areas on the wafer as in the conventional case. Is. Further, using a means other than the leveling optical system, for example, a focus detection optical system to obtain the amount of tilt of the exposure surface with respect to the image plane for each shot area significantly increases the processing time.

In view of the above points, the present invention can set the average surface of the exposure surface of each shot area parallel to the best image forming surface of the projection optical system even if the exposure surface of the photosensitive substrate has irregularities. It is an object to provide an exposure apparatus.

[0012]

In order to solve such a problem, the present invention holds a substrate (4) in which a plurality of shot areas are two-dimensionally arranged and holds a predetermined reference plane,
That is, it can be arbitrarily tilted with respect to the best imaging plane P1 of the projection optical system 3 and can be two-dimensionally moved within a plane substantially parallel to the best imaging plane (a plane perpendicular to the optical axis of the projection optical system 3). The stage (5 to 7) and a focus detection optical system (which detects a deviation in a direction substantially perpendicular to the substrate (optical axis direction of the projection optical system 3) at at least one measurement point in the shot area with respect to the best imaging plane ( 21-24, 16-19, 25-28) and an inclination detection optical system (10, 11, 15-19, 13) for detecting the inclination state of the shot area with respect to the best image plane, and a plurality of shot areas are provided. Of the plurality of shot areas SA on the substrate (4) in which at least three shot areas are specified shot areas. (SA
_{1 to} SA ₉ ), and using the focus detection optical system in a state where the exposure surface of the specific shot area detected using the tilt detection optical system and the best image formation surface of the projection optical system are matched. First control means (20, 29) for obtaining the amount of deviation from the best imaging plane at each of a plurality of predetermined measurement points in the shot area; and specifying based on the amount of deviation at the plurality of measurement points By calculating the tilt amount of the exposure area of the shot area with respect to the best image plane and statistically calculating the calculated tilt quantity of each of the plurality of specific shot areas, each of the plurality of shot areas on the substrate is calculated. A calculation means (29) for calculating the tilt correction amount; and a tilt correction amount and a tilt detection optical system for each shot area so that the exposure surface of each of the plurality of shot areas SA and the best image formation surface substantially coincide with each other. Based on the detected signal from It was the provision of the second control means for controlling the movement of the over-di (29).

[0013]

In the present invention, the inclination state of the shot area on the substrate detected by the inclination detection optical system is corrected by a predetermined correction.
The exposure plane and the reference plane (image plane of the projection optical system) are accurately matched for each shot area. For this reason,
Inclination amount of each exposure surface of at least three specific shot areas detected by using the focus detection optical system in a state where the exposure surface of the specific shot area and the reference plane are matched by using the tilt detection optical system. Is calculated statistically to obtain the tilt correction amount for each shot area. The principle of the present invention will be described with reference to FIG.

In FIGS. 1A to 1F, a substrate (wafer) 4B having a good flatness or a substrate (wafer) 4C to be processed is a leveling stage including an upper stage 5 and a lower stage 6. The shot area on the substrate is leveled so that the shot area on the substrate coincides with the best imaging plane P1 of the projection optical system 3. The leveling stage is moved along the plane P2 by an XY stage (not shown). The plane P2 represents the running plane of the XY stage (here, it is a moving plane including the orthogonal coordinate system defined by the interferometer), and is not parallel to the best image plane P1 (that is, the plane P2). The best image forming plane P1 is inclined with respect to.

First, as shown in FIG. 1A, a wafer 4B having a good flatness is placed on a leveling stage, and the detection light L parallel to the exposure surface of the wafer 4B is detected by a leveling optical system.
2 is obliquely irradiated to the optical axis of the projection optical system 3 to detect the tilt of the exposure surface. After that, the leveling stage is driven to set the exposure surface of the wafer 4B parallel to the best image formation plane P1. Since the flatness of the exposed surface of the wafer 4B is good,
The inclination of the exposure surface can be accurately detected by the leveling optical system, and the exposure surface is set accurately parallel to the best image formation plane P1.

Next, as shown in FIG. 1B, the leveling stage is moved leftward in the plane of the drawing along the plane P2.
Then, the focus position detection optical system is used to irradiate the detection light L3 obliquely with respect to the optical axis of the projection optical system 3, and the measurement point on the exposed surface of the wafer 4B (for example, the point intersecting the optical axis of the projection optical system 3).
A probe pattern image such as a slit pattern is formed on. The reflected light from the probe pattern image is re-imaged on a predetermined light receiving surface by the focus position detection optical system, and the height of the measurement point on the exposure surface of the wafer 4B is higher than the position of the re-imaged probe pattern image. To detect Actually, for example, the deviation δ ₁₀ between the height of the measurement point and the height of the surface P2 is calculated.

Thereafter, as shown in FIG. 1C, the leveling stage is moved rightward in the plane of the drawing along the plane P2, and the focus position detection optical system is used to measure the height of the measurement point and the stage running time. A deviation δ ₂₀ from the height of the surface P2 is obtained.
Besides, by moving the leveling stage along the plane P2, various measurement points (at least three measurement points which are not on the same straight line are required on the exposure surface of the wafer 4B, and these measurement points are required). Are all set within the range of one shot area).
The deviation δ _j0 between and is _obtained . The average surface obtained from the plurality of deviations δ _j0 represents the best imaging plane P1. Therefore, by measuring the height of each measurement point of the wafer 4B having good flatness with the focus position detection optical system, the surface P2
The inclination of the best imaging plane P1 with reference to is obtained.

Next, as shown in FIG. 1 (d), the wafer 4C to be processed having a large unevenness on the exposure surface is placed on the leveling stage. Then, the detection light L2 from the leveling optical system is applied to the exposure surface of one shot area on the wafer 4C to measure the tilt state thereof, and the surface determined to be an average surface by this is determined by the projection optical system 3. It is set parallel to the best image plane P1. However, the actual average surface P3 of the wafer 4C and the best image forming surface P1 are not set to be parallel to each other due to unevenness of the exposure surface, uneven coating of the resist, and the like. That is, the amount of tilt between the true average plane P3 and the image plane P1 differs for each shot area.

Therefore, in the present invention, of the plurality of shot areas on the wafer, at least three shot areas that are not on the same straight line are selected in advance as the specific shot areas, and the plurality of selected shot areas are selected by the following procedure. For each shot area, the difference (offset amount) between the true average plane P3 and the image plane P1 is calculated. First,
The tilting state is detected for one specific shot area by using a leveling optical system, and the surface determined to be an average surface is set parallel to the best imaging plane P1 of the projection optical system 3. Then, as shown in FIG. 1E, the leveling stage is moved to the plane P.
Move it to the left in the plane of the drawing along 2. Then, the detection light L3 is radiated obliquely to the optical axis of the projection optical system 3 by using the focus position detection optical system to form a probe pattern image at a measurement point on the exposure surface of the wafer 4C. Then, the reflected light from the probe pattern image is re-imaged on the predetermined light receiving surface by the focus position detection optical system, and the deviation δ ₁₁ between the height of the measurement point and the height of the surface P2 is obtained.

Next, as shown in FIG. 1F, with the leveling stage moved to the right in the plane of the drawing along the plane P2, the focus position detection optical system is used to measure the height and the plane of the measurement point this time. The deviation δ ₂₁ from the height of P2 is calculated. In addition to that, by moving the leveling stage along the surface P2, various measurement points (at least three measurement points not on the same straight line are required on the exposure surface of the wafer 4C, both of which are one specific shot). The deviation δ _j1 from the stage running surface P2 in (set within the range of the area) is obtained, and the average surface of the specific shot area is obtained from these plural deviations δ _j1 . Then, the tilt offset amount is calculated by subtracting the tilt of the best image plane obtained from the deviation δ _j0 of the wafer 4B from the average surface inclination obtained from the deviation δ _j1 of the wafer 4C. Thereafter, the tilt offset amounts are sequentially calculated for all the remaining specific shot areas, exactly as in the above operation.

Here, the tilt offset amount for each specific shot area calculated as described above can be defined by, for example, two components in each of the X and Y directions. Therefore,
By performing statistical calculation using these plural tilt offset amounts (X, Y components) and the position of each specific shot area within the wafer (coordinate position on the orthogonal coordinate system that defines the moving position of the wafer). , Tilt offset amount (θx, θy)
And the position (x, y) of the shot area are determined.
This is because, for example, a model function (θx = ax + by + c and θy = dx + e that is assumed for the tilt offset amount is used.
y + f) may be prepared in advance, and the coefficients (a to f) of the model function may be determined by least square approximation.

Using the model function determined as described above, the tilt offset amount for each shot area on the wafer is calculated. In the actual exposure operation, the offset amount for each shot area is added to the inclination amount of the exposure surface of the shot area obtained by the leveling optical system, and the leveling stage is driven (servo control) based on this result. As a result, for each shot area on the wafer, the true average surface P3 of the exposure surfaces is set parallel to the best image formation surface P1 of the projection optical system 3.

In the present invention, the focus offset optical system may be used to detect the tilt offset amount only for at least three shot areas on the substrate. It is possible to perform high-precision leveling for all shot areas of.
Regarding a plurality of wafers stored in the same lot, for example, only for the first (first) wafer, the tilt offset amount for each shot area is calculated by the statistical calculation as described above, and two wafers are stored. For the subsequent wafers, the result of the first wafer may be used as it is.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS. In this embodiment, the present invention is applied to a projection exposure apparatus. In FIG. 2, parts corresponding to those in FIG. 7 are designated by the same reference numerals, and detailed description thereof will be omitted. FIG. 2 shows a main part of the projection exposure apparatus of the present embodiment. In FIG. 2, the image of the pattern of the reticle 1 is exposed under the exposure light L1 via the projection optical system 3 on the wafer 4 as a photosensitive substrate. Is transferred to the shot area. The wafer 4 is mounted on a leveling stage including an upper stage 5 and a lower stage 6, and the leveling stage is mounted on a wafer stage 7.

Further, the detection light L2 from the light source 10 for the leveling optical system is condensed by the condenser lens 11 on the pinhole formed in the aperture plate 15. The detection light L2 from this pinhole is the dichroic mirror 16
After passing through the second projection objective lens 17, it is converted into a parallel light flux, and this parallel light flux is irradiated to almost the entire one shot area on the wafer 4. Further, the reflected light beam from the wafer 4 passes through the first condensing objective lens 18, passes through the dichroic mirror 19, and is then focused on the light receiving surface of the photoelectric sensor 13. As the photoelectric sensor 13, a sensor that can detect the two-dimensional coordinates of the center of the light amount distribution on the light receiving surface, for example, a light receiving element divided into four is used. The tilt information from the photoelectric sensor 13 is supplied to the signal processing circuit 20.

On the other hand, the detection light L3 from the light source 21 for the focus detection optical system is applied to the slit-shaped opening pattern (slit pattern) of the slit plate 23 by the condenser lens 22. The detection light L3 from the slit pattern is reflected by the dichroic mirror 16 after passing through the first projection objective lens 24, and then is reflected by the second projection objective lens 17 at a predetermined measurement point on the wafer 4 (image field of the projection optical system 3). The image of the slit pattern of the slit plate 23 is formed at this measurement point at an oblique position with respect to the optical axis AX of the projection optical system 3.
In addition, the reflected light from the image of this slit pattern is
After passing through the condenser objective lens 18 and reflected by the dichroic mirror 19, it is condensed on the light receiving slit plate 27 via the second condenser objective lens 25 and the vibrating mirror 26, and the slit pattern is formed on the light receiving slit plate 27. Is re-imaged.

The light receiving element 2 is provided on the back surface of the light receiving slit plate 27.
8 is arranged, and the detection signal from the light receiving element 28 is supplied to the signal processing circuit 20. If the wafer 4 moves in the direction along the optical axis AX of the projection optical system 3 in this state, the position of the slit pattern image on the wafer 4 shifts to the left or right in the plane of the drawing, and therefore on the light receiving slit plate 27. The position of the slit pattern image is also shifted in the direction parallel to the light receiving slit plate 27 (vertical direction in the plane of the drawing). In addition, the vibrating mirror 26
Due to this, the slit pattern image vibrates at a constant amplitude in the moving direction with a predetermined amplitude. This applies the detection principle of the photoelectric microscope, and the signal processing circuit 20 synchronously detects the detection signal of the light receiving element 28 with the drive signal of the vibrating mirror 26, so that the wafer 4 (measurement point in the shot area) is detected. When the height of the image plane of the projection optical system 3 is matched, a so-called S-curve position detection signal that becomes a zero cross is obtained. From this position detection signal, the deviation amount δ of the height of the measurement point of the wafer 4 from the best image plane of the projection optical system 3 can be obtained.

Further, the main controller 29 includes a signal processing circuit 2
Information on the tilt state of the exposure surface of the wafer 4 detected by the leveling optical system from 0, and the deviation amount δ of the predetermined measurement point of the wafer 4 detected by the focus detection optical system from the best imaging plane.
Information is supplied. In response to this, the main controller 29
The wafer 4 is positioned (focused) by controlling the operation of the wafer stage 7 via the driving means 9, and the leveling stages 5 and 6 are servo-controlled to tilt the exposure surface of the shot area on the wafer 4. Is set parallel to the best image plane of the projection optical system 3.

Next, the procedure of the leveling operation and the exposure operation of this embodiment will be described. First, the leveling operation and the exposure operation of the wafer to be processed will be described. In this case, in order to obtain the image plane tilt of the projection optical system 3 in advance and to check the parallelism between the best image plane of the projection optical system 3 and the running surface (moving plane) of the leveling stage, FIG. 3 is shown. Preliminary steps are required.

In the preliminary step of FIG. 3, first, as an initial step, a test reticle having a predetermined pattern (for example, a wedge pattern) formed at each of a plurality of predetermined positions is placed on the reticle holder 2. Further, with the flatness and the resist coated wafer for test printing placed on the leveling stage, the Z stage in the wafer stage 7 is lowered to expose the exposed surface of the wafer to the projection optical system 3 It is arranged below the best image forming plane of.

Next, after the first exposure is performed in step 101, the Z coordinate of the Z stage is set to the upper limit Z in step 102.
Check whether the _end has been reached. When the upper limit is not reached, the process proceeds to step 103, the Z stage is moved in the direction of the projection optical system 3 by ΔZ, and the XY stage in the wafer stage 7 is driven to expose the next shot area on the wafer. After moving to the position (projection position of the reticle pattern by the projection optical system 3), step 10
Return to 1 and perform the second exposure. Since the focus detection optical system is operated when setting the Z coordinate, raising the Z stage by ΔZ means that the projection optical system 3 is accurate.
Means that the distance between the wafer and the exposed surface of the wafer is reduced by ΔZ.

Then, after repeating steps 101 to 103 until the Z coordinate of the Z stage reaches the upper limit, the process proceeds to step 104 to develop the wafer. Next, a plurality of resist images formed for each shot area on the wafer are visually observed or detected by using an alignment sensor of the exposure apparatus, and a resist image in the best image formation state is formed for each of the plurality of reticle patterns. The shot area that the user has, that is, the Z coordinate is determined as the best focus position. As a result, the best focus position (and the depth of focus) at each of a plurality of points in the image field of the projection optical system 3
Will be detected. In the next step 105, a plane defined by a plurality of best focus positions (Z coordinate values), that is, the best image plane of the projection optical system 3 is obtained.
At this time, when the image plane tilt exists, the plane obtained as described above is tilted with respect to the plane perpendicular to the optical axis AX of the projection optical system 3, so that the image plane tilt is detected. . Here, the least squares method or the like may be used when determining the best image plane, and when the projection optical system 3 has a field curvature or the like, the image plane can be obtained as an approximate plane. For details of the above measuring method, see Japanese Patent Laid-Open No. 1-18
No. 7817 and Japanese Patent Laid-Open No. 2-30112.

The above is the projection optical system 3 by test printing.
In this example, the image plane of the wafer stage 7 is obtained.
A luminescent reference mark is provided on the upper surface, an image of the reference mark is formed on the lower surface (pattern surface) of the reticle via the projection optical system 3, and light from the pattern surface is passed through the projection optical system and the reference mark. System for photoelectric detection (Japanese Patent Application No. 3-313
No. 43, etc.) can also be used to obtain the image plane. In this case, the best focus position at one point in the image field of the projection optical system is obtained based on the change of the photoelectric signal when the reference mark is moved in the optical axis direction (Z direction) of the projection optical system, By performing the measurement at a plurality of points in the image field, the image forming plane of the projection optical system 3 and further the image plane inclination (field curvature) can be obtained.

Next, at step 106, the offset amount of the tilting amount of the leveling stage is calculated so that the exposure surface of the wafer and the image forming surface are accurately aligned with each other, and the offset amount is set in the internal storage unit of the main controller 29. The offset amount is added to the existing target tilt amount for the leveling mechanism. And
The main controller 29 controls the tilting states of the leveling stages 5 and 6 according to the target tilt amount to which the offset amount is added, so that the exposure surface of the wafer on the leveling stage becomes the best image forming surface of the projection optical system 3. They are set in parallel (step 107).

After that, in step 108, a super flat wafer having extremely good flatness is placed on the leveling stage, the leveling mechanism is operated in step 109, and then in step 110, the super flat wafer is detected using the focus detection optical system. Measure the position of the surface of the wafer. Specifically, while the leveling mechanism is operating, that is, the best image plane of the projection optical system 3 and the surface of the super flat wafer are aligned, the XY stage of the wafer stage 7 is moved by a predetermined amount. , The optical axis A of the projection optical system 3
A plurality of measurement points in a predetermined area (corresponding to a shot area) on the wafer with respect to a reference plane perpendicular to X (for example, a stage running surface) (same positions and numbers as a plurality of measurement points in a specific shot area described later). It is preferable that each deviation amount) is obtained, and an average surface from these deviation amounts is obtained. This average surface is calculated by, for example, statistical calculation processing (least squares method or the like), search method, or the like. Present (Stage 5
Since the upper wafer is a super flat wafer, the surface obtained here represents the leveling matching surface (best imaging surface). The tilt amount of the leveling mating surface from the stage running surface is stored in the storage unit of the main controller 29 as a reference tilt amount. This completes the preliminary process.

Next, in step 111 of FIG. 4, the wafer 4 to be processed is placed on the leveling stage,
In step 112, it is determined whether or not the process is a leveling abnormality. The process in which the leveling abnormality appears is a process in which the average surface of the wafer 4 detected by the leveling optical system is inclined with respect to the actual average surface of the wafer 4 (for example, the wafer 4A in FIG. 8 is formed. Process), and the existence of such a process is empirically known. In the case of the process in which the leveling abnormality does not occur, the pattern exposure is executed while performing the leveling for each shot area in step 114 while the leveling mechanism is operated in step 113.

On the other hand, in the process in which the leveling abnormality occurs, the process proceeds to step 120, where a specific shot area is selected and a measurement point (position and number) by the focus detection optical system in the specific shot area is specified. Here, at least three shot areas are selected as the specific shot areas from the plurality of shot areas on the wafer 4, but the arrangement (position, number) of the specific shot areas on the wafer 4 is the exposure surface within the wafer. It may be determined so that the change in the tilt amount (tilt offset amount) between the image forming surface and the image forming surface can be known. In this embodiment, nine shot areas SA ₁ to SA ₁ as shown in FIG.
The SA ₉ shall be selected. However, when selecting the specific shot areas, it is necessary to select all the specific shot areas so that they are not aligned on the same straight line. Further, the number of measurement points in the specific shot area may be at least three,
Also here, when there are three measurement points, it is necessary to specify so that the measurement points do not line up on the same straight line. Further, the positions of the measurement points in the specific shot area may be designated at positions that show an average surface of the exposure surface well, for example, four corners of the specific shot area. At this time, information about the step structure (concavo-convex) in the shot area is input to the main control device 29 in advance, or the operator inputs the information to the main control device 29 via an input device (keyboard or the like). ,
It is advisable to set each position of a plurality of measurement points according to this step structure.

Next, the process proceeds to step 121, and it is selected whether the processing is performed for each wafer or the measurement result of the wafer at the top of the lot is also used for the second and subsequent wafers. Step 129 if measurement is performed only on the wafer at the top of the lot
If the process is performed for each wafer, the process proceeds to step 122. In step 122, one of the nine specific shot areas, for example, the specific shot area SA ₁ (FIG. 10) is moved to the exposure position. In the next step 123, the leveling mechanism is operated, that is, the tilting state of the exposure surface of the shot area SA ₁ is detected by using the leveling optical system, and the exposure surface and the image plane of the projection optical system 3 are made to coincide with each other. Servo control the leveling stage.

Thereafter, in step 124, main controller 29 uses the focus detection optical system while moving the XY stage to determine the amount of deviation from the stage running surface at each of a plurality of designated positions within the specific shot area. measure.
Further, main controller 29 calculates the average surface of the specific shot area on the basis of the plurality of deviation amounts, and then obtains the inclination amount of the specific shot area with respect to the best image forming surface of projection optical system 3, and this inclination is calculated. Information about the amount is stored in a storage unit inside the main controller 29 together with the position information (coordinate value) of the specific shot area (step 125). Step 12
At 6, it is determined whether or not the above measurement has been completed for all the specific shot areas. Here, since the measurement of the specific shot area SA ₁ is just completed, the process returns to step 122, and the operation of the next specific shot area S is performed by the same operation as above.
Calculate the average face of A ₂ . Hereinafter, step 122-
When 126 is repeatedly executed and the measurement in all the specific shot areas is completed, the process proceeds to the next step 127. Here, the average surface (inclination amount) calculated for each specific shot area can be decomposed into components in the X and Y directions, and the inclination amount can be obtained by adjusting the inclination amount in the X direction (X component) and the Y direction. It may be stored by being decomposed into the inclination amount (Y component).

In the next step 127, a linear component is extracted and interpolated by a statistical calculation process using the X- and Y-direction positions (coordinate values) and inclination amounts for each specific shot area obtained by the above process. By performing the above, the tilt amount is calculated for each of all the shot areas on the wafer 4. Specifically, the coefficient of the model function prepared in advance is determined by the least squares method using the coordinate values in the X and Y directions of the nine specific shot areas and the X and Y components of the tilt amount. Here, as an example of the model function, the X component of the tilt amount in an arbitrary shot area on the wafer 4 is θx, and the Y component thereof is θy.
Then, the following Equation 1 (determinant) can be considered.

[0041]

[Equation 1]

That is, the coordinate values (x, y) of the specific shot area and the X and Y components (θx, θy) of the tilt amount are substituted into the above equation, and the parameters a to f are determined using the least square method. Good. The main controller 29 has six parameters a-
After f is determined, the tilt amounts of all the shot areas on the wafer 4 are calculated using Equation 1. Here, the tilt amount (θx, θy) for each shot area, which is obtained from Equation 1, is precisely the exposure surface of the shot area with respect to the best image plane when the leveling stage is driven according to the detection signal of the leveling optical system ( It represents the amount of deviation (difference in inclination) of the average surface). Therefore, main controller 29 adds the tilt amount (θx, θy) for each shot area to the reference tilt amount stored therein as a tilt offset amount (leveling correction amount), and sets the target tilt amount for each shot area. Set.

Thereafter, the main controller 29 sets the leveling stages 5 and 6 for each shot area according to the target inclination amount.
By controlling the tilted state of (3), the exposure surface of each shot area on the wafer 4 is set parallel to the image plane of the projection optical system 3 (step 128). As a result, in step 114, the image of the pattern of the reticle 1 is clearly transferred onto the exposed surface of the shot area having the unevenness on the wafer 4. In the above description, the tilting of the leveling stages 5 and 6 is controlled according to the target tilt amount. However, in step 127, only the leveling correction amount (θx, θy) for each shot area is calculated and stored, In the actual exposure operation, the tilting of the leveling stage may be controlled so that the detection signal of the leveling optical system has an offset by the correction amount (θx, θy).

There is also a method of using the detection signal of the leveling optical system instead of performing the statistical processing by using the inclination amount for each specific shot area, and an example thereof will be described with reference to FIG. Steps 111 to 121 are the same as in the above embodiment. After moving one specific shot area to the exposure position in step 122 ′, step 12
At 3 ′, the focus detection optical system is used to obtain the deviation amount from the running surface of the wafer stage 7 for each of the plurality of measurement points in the specific shot area. In step 124 ', the tilt amount of the specific shot area is calculated from these deviation amounts.

At the next step 125 ', the leveling stage is driven based on the tilt amount calculated as described above,
The exposure surface of the shot area and the image formation surface of the projection optical system are matched. Then, the detection signal of the leveling optical system under this condition is read and stored (step 12).
6 '). Further, in step 127 ′, it is judged whether or not the reading of the detection signal as described above in all the specific shot areas is completed, and if it is completed, step 12 is executed.
Go to 8 '.

In the next step 128 ', the same processing as the statistical processing performed using the inclination amount in the above-mentioned embodiment is performed.
This is performed with respect to the output of the leveling optical system, for example, the output voltage in each direction of pitching and rolling. As a result, the target output of the leveling optical system is set for each shot area on the wafer 4. Then, in the actual exposure operation, main controller 29 servo-controls the inclination of leveling stages 5 and 6 according to the target output value of the leveling optical system set for the shot area to be exposed (step 129 ′). As a result, the exposure surface and the image plane of the projection optical system 3 can be set in parallel for each shot area.

Here, if the above-described calculation of the leveling correction amount is performed for all the wafers in the same lot processed in the process in which the leveling abnormality is estimated to occur, the throughput is lowered. Therefore, an example of a method of reducing the number of times of measuring the correction amount of the leveling mechanism will be described with reference to FIG. 6 below. If the leveling correction amount is not calculated for each wafer in step 121 in FIG. 4, the operation proceeds to step 129 in FIG. 6 and the correction amount of the leveling mechanism is calculated using the first wafer in the lot. Determine whether to do or not. When the correction amount is calculated using the first wafer, the process proceeds from step 130 to step 131, and at step 131, step 122 of FIG.
The same process as in .about.128 is performed to obtain the correction value of the target tilt amount for the leveling mechanism. Then, in step 132, the leading wafer is exposed.

For the second and subsequent wafers, the process proceeds from step 130 to step 132, and exposure is performed with the leveling mechanism corrected for each shot area based on the correction amount obtained for the first wafer. On the other hand, if the correction is not performed for the leading wafer in step 129, the process directly proceeds to step 132 and the exposure operation for the wafer is started without calculating the correction amount of the leveling mechanism. For example, for each of a predetermined number of wafers from the beginning, exposure is performed while calculating the above correction amount and performing the correction for each shot area, and for subsequent wafers, a plurality of corrections are performed for each shot area. The leveling mechanism may be corrected based on the average value of the amounts (or at least one of the plurality of correction amounts).

Of the plurality of specific shot areas on the wafer, the average surface is selected for the specific shot areas for which the deviation from the stage running surface at each of the plurality of designated positions cannot be measured in step 124 described above. Since it cannot be calculated accurately, it goes without saying that the above step 127
In the (statistical calculation process), the data in the specific shot area is rejected, and only the data in the remaining specific shot area is used. At this time, for example, a shot area adjacent to the rejected specific shot area is designated as a substitute shot, steps 122 to 126 are executed for this substitute shot, the average surface (tilt amount) is obtained, and the obtained data is obtained. May be used in step 127.

As described above, according to this embodiment, the focus detection optical system is used to measure the state of the unevenness of the surface of the wafer. Therefore, the average of the exposed surface of the wafer detected by the leveling optical system is measured. Even if the target surface intersects the true average surface and the angle thereof varies depending on the shot areas, the correction value of the tilt amount for the leveling mechanism can be set. Therefore, accurate leveling can be performed even when unevenness is present on the surface of the wafer. Also, the best image plane of the projection optical system is tilted (including field curvature)
However, the exposure surface of the wafer can always be set accurately and parallel to its best imaging plane.

A sensor capable of measuring multiple points may be used as the focus detection optical system. What is a sensor that can measure multiple points?
For example, an image field of a projection optical system, that is, at least one elongated slit image is projected along a diagonal line of a shot area, and an image re-formed by reflected light is divided into a plurality of images by a one-dimensional or two-dimensional image sensor. It is possible to measure the heights of a plurality of exposure regions at one time by receiving light. According to this, the measurement time can be significantly shortened.

By the way, in the above embodiment, at least 3
Although one shot area is selected as the specific shot area, the selection of this specific shot area is shown in FIG.
1 and FIG. 12 will be described in more detail. It should be noted that the nine shot areas shaded in FIG. 11 are specific shot areas. Usually, the main controller 29, before the first print (exposure of the first layer), based on the external dimensions of the wafer 4, the size of the shot area, the step pitch, etc., the optimum shot map corresponding to the external shape of the wafer 4. Create the data. This shot map data represents how the shot areas should be arranged in the wafer from the outer shape of the wafer 4, is automatically determined by calculation, and is stored in the memory as the stepping coordinate position of the wafer stage 7. be registered. Further, main controller 29 calculates the coordinate position of the center point of the wafer, the boundary position of the forbidden zone, etc. based on the determined shot map data and stores it in the memory.

FIG. 11 schematically shows an example of shot map data determined corresponding to the outer shape of the wafer.
One of the rectangles in FIG. 11 represents a shot area. The forbidden band means a region where the tilt detection by the leveling optical system is prohibited within a constant width (about 1 to several mm) from the outer peripheral edge ED of the wafer 4, as shown in FIG. The boundary line determined by main controller 29 is line LL in FIG. 11, and is defined by the radius from the center of wafer 4. Further, regarding the linear notch (orientation flat) portion OF, a forbidden zone (boundary line LL) having a constant width along the orientation flat portion OF.
Is set. Furthermore, the forbidden band is generally uneven resist coating, non-uniformity of various processes, or wafer edge ED.
It is set because a chip that is a non-defective product can hardly be obtained due to a defect of the device. Therefore, the width of the forbidden band may be fixed to a substantially constant value, but the forbidden band width may be input so as to be slightly changed according to the chip size.

Second print (exposure of the second layer)
Then, based on the result of the wafer global alignment, the offset amount from the mounting position of the wafer on the upper stage (holder) 5 at the time of first printing is obtained, and this offset amount is used as the wafer center point and the forbidden band at the time of first printing. It is inserted as a correction value when the boundary line position is calculated. By doing so, the deviation between the shot map (matrix in FIG. 11) and the wafer outer shape in the X and Y directions, which occurs during the second printing, is corrected, and the state during the first printing is always maintained.

In step 120, the main controller 29 (or operator) determines a large number of exposure shots (shot areas where the reticle pattern is actually exposed) on the wafer based on the shot map data and the boundary line LL. At least three specific shot areas are selected from the above. That is, the measurement area set in the exposure shot (the area in which each of the plurality of designated positions where the amount of deviation from the stage running surface is measured by the focus detection optical system in step 124 is set) is inside the boundary line LL. To select at least three exposure shots as specific shot shots so that all shot shot shots do not lie on the same straight line among the shot shots existing in the levelable area and not located at the outermost periphery of the wafer. become.

Here, the above-mentioned measurement area (area in which a plurality of measurement positions by the focus detection optical system can be designated) is set to an area including at least a part of the exposure shot, and its average surface within the exposure shot. Corresponds to the area to be accurately aligned with the reference plane (the best image plane of the projection optical system), and its size and position are determined based on, for example, the level difference (unevenness) in the exposure shot and the size of the circuit pattern area. It is what is done. FIG. 12 is a partially enlarged view of the wafer 4 shown in FIG. 11, in which an area AR indicated by a dotted line in the exposure shot in the drawing represents a measurement area, and a black circle represents the center position of the exposure shot. In FIG. 12, the five exposure shots SA _{21 to} SA _{25 in} which at least a part of the measurement area AR is outside the boundary line LL (in the forbidden zone) do not satisfy the above condition, and the target is selected when the specific shot area is selected. Will be excluded from. That is, in FIG. 12, exposure shots SA _{21 to} SA
_The remaining seven exposure shots except ₂₅ are selected for the specific shot area.

Next, referring to FIG. 13, a method of setting the measurement area in the exposure shot will be briefly described. FIG. 13 shows one exposure shot SA on the wafer.
The intersection RC of the one-dot chain line is a rectangular measurement area (dotted line) AR
Represents the center position of. In FIG. 13, measurement area A
R is eccentric to the center CC of the exposure shot SA in the Y direction by a predetermined amount, and its size (length of each side) is Lx,
It is defined as Ly. That is, main controller 29 (or the operator), according to the leveling area determined according to the level difference in the exposure shot SA, the size of the circuit pattern area, etc., the eccentricity amount (offset amount) Ox, Oy of the center RC with respect to the shot center CC, And size (side length) L
By setting x and Ly to predetermined values, the exposure shot SA
The size and position of the measurement area AR inside can be arbitrarily determined.

In FIG. 13, 16 measurement points (circles in the figure) for the focus detection optical system are set in the measurement area AR. In particular, 16 measurement points are 4 by 2 x 2 = 4
Two groups (circle of two-dot chain line in the figure) GA, GB, G
It is divided into C and GD, and four groups GA to GD are set at the four corners of the measurement area AR, respectively. FIG.
In this case, the main controller 29 uses the focus detection optical system to
After measuring the deviation amount from the stage running surface at each of the measurement points, the four deviation amounts for each group are averaged to obtain one deviation amount, and the four deviation amounts are further calculated. The average surface of the specific shot area, that is, the measurement area AR is calculated based on the (average value). At this time, the main controller 29 (or the operator) sets the number (m × n) of measurement points in each group and the distances (pitch) Px and Py between the measurement points to predetermined values, whereby the measurement area AR It is possible to arbitrarily determine the number and positions of measurement points in the group. Here, it is premised that four groups are respectively set at the four corners of the measurement area AR, but the setting position of each group may be set as a parameter like the pitches Px, Py and the like.

Here, the number of groups set in the measurement area AR is not limited to four, but any number may be set as long as it is three or more, and the position of each group in the measurement area AR may be arbitrary. . Further, the number of measurement points in each group may be one or more, and the number and position of the groups may be different from each other. Furthermore, the measurement area A
The position and size of R, or the number and positions of groups and measurement points therein may be set individually for each of a plurality of specific shot areas, or may be set to be the same for all specific shot areas. Alternatively, main controller 29 may automatically determine the type of wafer (steps in the exposure shot, size of circuit pattern, etc.). Further, theoretically, all of the measurement area AR may be set outside the exposure shot, but in reality, it is desirable that the measurement area AR include at least a part of the exposure shot. Furthermore, the measurement area AR is
It may be set equal to or more than the exposure shot (area). Especially when the measurement area AR is made larger than the area of the exposure shot, the measurement area AR is set across the specific shot area and at least one exposure shot adjacent to the specific shot area. AR
It is necessary to determine its size and position so that is set inside the boundary line LL. In the above-described embodiment, the leveling optical system irradiates the parallel light flux onto almost the entire surface of the exposure shot, whereas the measurement area AR is set to a partial area within the exposure shot, but it does not depend on the size of the exposure shot. The measurement area AR may be set equal to or larger than the illumination area of the light flux from the leveling optical system on the wafer.

By determining the measurement area AR and the measurement point as described above, it is possible to accurately calculate the tilt amount of the specific shot area (measurement area AR) with respect to the best image plane of the projection optical system. By the way, in the above-mentioned embodiment, the formula 1 is used as the model function in step 127, but the following formula 2 may be used.

[0061]

[Equation 2]

That is, the equation 2 is the one in which the parameter d in the equation 1 is d = b. Hereinafter, the reason why this equation 2 can be used will be briefly described. Now, applying leveling means measuring the inclination of the wafer surface. That is, if the wafer surface is represented by z = W (x, y), leveling is ∂W.
It means to measure / ∂x and ∂W / ∂y. If the wafer surface obtained from this result (that is, the wafer surface recognized by the leveling optical system) is z = W '(x, y),
The residual tilt amounts θx and θy are θx = ∂W / ∂x−∂W ′ / ∂x = ∂ (W−W ′) / ∂x θy = ∂W / ∂y−∂W ′ / ∂y = ∂ ( WW ′) / ∂y. Although θx and θy are the differences between the slopes of W and W ′, they can also be interpreted as the slope of the curved surface W ″ = WW ′. Here, since both W and W ′ are “normal curved surfaces”, the curved surface W ″ is also “normal curved surfaces”. The "normal curved surface" includes two planes and straight lines that intersect with a general quadric surface, but the curved surface W '' is not such a
This means that the curved surface is not so discontinuous.
Therefore, conversely, using the focus detection optical system, θx and θy are obtained from the deviation amounts at a plurality of points in the shot area, and θx = ∂W ″ / ∂x = ax + by + c θy = ∂W ″ / ∂y = dx + ey + f Then, this simultaneous differential equation is actually the solution W ″ = W−
It is natural to think of having W '. Then, θx = ∂
W ″ / ∂x = ax + by + c is partially differentiated by y, and θy = ∂
If W ″ / ∂y = dx + ey + f is partially differentiated by x, d =
It can be seen that ∂ ² W ″ / ∂y ∂x = ∂ ² W ″ / ∂x∂y = b.

By the way, although a super flat wafer is used in the above-mentioned embodiment, for example, a reference member which is provided on a part of the upper stage 5 and has a reflecting surface having good flatness in substantially the same plane as the wafer surface is used. You may do it.
Further, in the apparatus having the above-mentioned structure, for example, JP-A-61-444
The enhanced global alignment (EGA) as disclosed in Japanese Patent Publication No. 29, Japanese Patent Laid-Open No. 62-84516, etc. has become mainstream. The EGA method refers to at least three of a plurality of shot areas formed on a wafer, for example, seven shot areas located near the outer periphery of the wafer, before performing overlay exposure on one wafer. (Sample shot) is designated, and the coordinate position of two sets of alignment marks (in the X and Y directions) associated with each shot area is measured by the alignment sensor (sample alignment). Thereafter, based on these mark positions (measured values) and design values, error parameters relating to the array characteristics of the shot areas on the wafer, that is, the offset of the wafer center position (X and Y directions) and the degree of expansion and contraction of the wafer (X , Y direction), the remaining rotation amount of the wafer, and the orthogonality of the wafer stage (orthogonality of the shot array), a total of six parameters are determined by statistical calculation (such as the least square method).
Then, the design coordinate values are corrected for all shot areas on the wafer based on the determined parameter values, and the wafer stage is stepped according to the corrected coordinate values. . Therefore, the arrangement of the specific shot areas in the above-described embodiment on the wafer and the arrangement of the sample shots in the EGA may be the same or part of them may be common. At this time, each coordinate value of the sample shot may be corrected using the leveling correction amount for each specific shot area.

In the above embodiment, the leveling correction amount is obtained for nine specific shot areas. However, the shot area that cannot be measured or the shot area whose measured value has an extreme offset is rejected.
You may make it use only the remaining shot area.
For the rejected shot area, the leveling correction amount detected by using the focus detection optical system may be directly used instead of using the target tilt amount. Alternatively, a shot area near the rejected shot area may be selected as a substitute shot and the measurement value of the shot may be used. Further, when some of the tilt amounts of the nine specific shot areas cannot be measured for some reason and the correction amount cannot be calculated, or the correction amount cannot be calculated due to an abnormal measured tilt amount, or When an abnormal correction amount is obtained, the leveling operation of each shot area on the wafer may be corrected by using the correction amount calculated last time (for example, for the previous wafer).

[0065]

According to the present invention, since the measurement result of the leveling optical system is statistically processed and corrected by using the focus detection optical system, unevenness or the like exists on the exposed surface of the photosensitive substrate and Even in the case where the shot areas on the substrate have different irregularities or the like, the true average surface of the exposure surface can be set parallel to the best image forming surface of the projection optical system. Further, since the correction amount for each shot area is measured and statistically calculated for each photosensitive substrate or each lot, there is an advantage that it is possible to deal with a change in unevenness of the exposure surface due to process nonuniformity.

[Brief description of drawings]

FIG. 1 is a diagram illustrating the principle of a leveling operation of an exposure apparatus according to the present invention.

FIG. 2 is a configuration diagram showing a main part of a projection exposure apparatus according to an embodiment of the present invention.

FIG. 3 is a flowchart showing a preliminary process of a correction operation of the leveling mechanism of the embodiment.

FIG. 4 is a flowchart showing a correction operation of the leveling mechanism.

FIG. 5 is a flowchart showing another example of the correction operation of the leveling mechanism.

FIG. 6 is a flowchart showing a part of the operation when the correction of the leveling mechanism is selectively performed on the wafer to be processed in the embodiment.

7A is a configuration diagram showing a main part of a conventional projection exposure apparatus, FIG. 7B is a front view of the photoelectric sensor of FIG. 7A, and FIG.
FIG. 8 is a perspective view showing the leveling optical system of FIG.

FIG. 8A is a diagram showing a state in which an average surface detected by a wafer leveling optical system is set parallel to the best imaging plane, and FIG. 8B is a true average surface of the wafer. FIG. 3 is a diagram showing a state in which is set parallel to the best image plane.

9A is a diagram showing one shot row on a wafer, and FIG. 9B is a diagram showing a tilt amount for each shot region.

FIG. 10 is a diagram showing an arrangement of specific shot areas selected from a plurality of shot areas on a wafer in the embodiment.

FIG. 11 is a diagram schematically showing an example of shot map data determined corresponding to the wafer outer shape.

12 is a partially enlarged view of the wafer shown in FIG.

FIG. 13 is a diagram showing an example of measurement areas and measurement points set in one shot area on a wafer.

[Explanation of symbols]

 1 Reticle 3 Projection Optical System 4 Wafer 5 Upper Stage 6 Lower Stage 13 Photoelectric Sensor 15 Aperture Plates 16, 19 Dichroic Mirror 17 Second Projection Objective Lens 18 First Focusing Objective Lens 20 Signal Processing Circuit 23 Slit Plate 24 First Projection Objective Lens 25 Second condensing objective lens 26 Vibrating mirror 27 Light receiving slit plate 28 Light receiving element 29 Main controller

Claims (1)

[Claims] 1. A substrate, in which a plurality of shot regions are arranged two-dimensionally, is held and can be arbitrarily tilted with respect to a predetermined reference plane, and is two-dimensionally moved in a plane substantially parallel to the reference plane. A possible stage, a focus detection optical system for detecting a shift in a direction substantially perpendicular to the substrate at at least one measurement point in the shot area with respect to the reference plane, and an inclination state of the shot area with respect to the reference plane. In an exposure apparatus that includes a tilt detection optical system that detects, and transfers a pattern formed on a mask by arranging each of the plurality of shot areas on the reference plane, at least one of the plurality of shot areas on the substrate. Three
One shot area is selected as a specific shot area, and in a state where the specific shot area and the reference plane are matched by using the tilt detection optical system, the focus detection optical system is used to pre-set the specific shot area in advance. First control means for obtaining a deviation amount from the reference plane at each of a plurality of defined measurement points; and an inclination amount of the exposure surface of the specific shot area based on the deviation amount at the plurality of measurement points. Calculating means for calculating the tilt correction amount of each of the plurality of shot areas on the substrate by statistically calculating the calculated tilt quantity of each of the plurality of specific shot areas; Based on the tilt correction amount and the detection signal from the tilt detection optical system, for each shot area, the exposure surface of each shot area and the reference plane are substantially coincident with each other. Exposure apparatus characterized by comprising a second control means for controlling the movement of the stage.