JPH0992601A - Projection aligner - Google Patents

Abstract

PROBLEM TO BE SOLVED: To ensure high overlay accuracy even when a projection aligner having a remaining asymmetric random image distortion is used mixedly. SOLUTION: A transparent correction plate 8 is interposed between a reticle R and a projection optical system PL. The correction plate 8 is disposed removably by a replacing unit 21 and replaced, as required, by another correction plate 24a-24c stocked in a stock room 23. The correction plate 8 is polished to correct random distortion in a projection aligner employing it. Furthermore, distortion of a projected image is corrected by selecting the correction plate 8 depending on the image distortion characteristics of a projection aligner 28 used in the preceding exposure step or a projection aligner 29 used in the following exposure step thus enhancing the overlay accuracy between two layers.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention transfers a mask pattern onto a photosensitive substrate during a photolithography process for manufacturing, for example, a semiconductor integrated circuit, a liquid crystal display element, an image pickup element (CCD or the like), a thin film magnetic head or the like. In particular, the present invention is suitable as a projection exposure apparatus used when performing exposure in different layers on a photosensitive substrate by a mix-and-match method.

[0002]

2. Description of the Related Art Conventionally, a projection exposure apparatus (stepper or the like) for transferring a pattern of a reticle as a mask onto a wafer coated with a photoresist through a projection optical system has been used in manufacturing a semiconductor device, for example. To be done. Generally, a semiconductor device is formed by stacking a plurality of layers of circuit patterns on a wafer in a predetermined positional relationship. Also, in recent semiconductor manufacturing plants, in order to increase throughput (the number of wafers processed per unit time), different projection exposure apparatuses are mixed on different layers on a wafer and exposure is performed by a mix-and-match method. However, strict overlay accuracy is required for the projection exposure apparatus used in such cases. Therefore, in each projection exposure apparatus, image distortion such as distortion of the projection optical system is corrected with high accuracy. However, it is difficult to completely correct the image distortion, and image distortion that cannot be corrected remains. In this case, even if the image distortion of each projection exposure apparatus is within the allowable range,
There may be unacceptable variations in image distortion between projection exposure apparatuses. In particular, when the magnification of the projection optical system and the size of the exposure field of the projection exposure apparatus are different, or when the projection exposure apparatus is manufactured in different years and the apparatus has different image distortion tolerance standards, such a situation may occur. There is a variation in image distortion, and the overlay accuracy, so-called matching accuracy, is reduced.

In order to solve these problems, Japanese Patent Laid-Open No. 62-7129 and Japanese Patent Laid-Open No. 62-24624 disclose that the superposition accuracy of images having image distortion of each device is optimized. There is disclosed a method in which the projection magnification of each device is adjusted and the exposure position is corrected so that the exposure position is optimized. Further, in Japanese Patent Application Laid-Open No. 4-127514, an anisotropic image distortion is corrected by three-dimensionally moving or tilting an optical member forming a reticle or a projection optical system to further improve overlay accuracy. A method of doing so is disclosed. Further, Japanese Patent Laid-Open No. 5-166699 discloses a method of managing distortion information of a projection optical system by a central processing unit so as to optimize distortion matching accuracy between respective projection exposure apparatuses.

[0004]

The image distortion component that can be corrected by the above-mentioned conventional technique is a translation component by exposure position correction, a point symmetry component by projection magnification correction, or an optical member of the projection optical system. There is only a predetermined symmetric component, which is a line-symmetrical component due to the correction of the inclination angle of (1), and conventionally, it has been attempted to reduce the overlay error by making full use of the correction means for only those symmetric components. Therefore, for example, a correction condition that minimizes a deviation amount of a projected image from an ideal grid point, that is, a maximum error among so-called distortion errors is calculated and corrected.

However, even if such a correction is performed, the nonuniformity of the refractive index due to the nonuniformity of the glass material which is difficult to remove in the projection optical system, or the partial deviation from the spherical surface during the glass polishing. It is unavoidable that there is a certain amount of image distortion due to a random component having no symmetry caused by the above. In the past, such a random component rarely hindered the overlay accuracy. However, as the pattern line width becomes finer in recent years, the required overlay accuracy becomes stricter. The distortion component has been an obstacle to obtaining the required overlay accuracy.

Further, when projection exposure apparatuses having different angles of view (sizes of exposure fields) are mixed, even image distortion having symmetry for each projection exposure apparatus, symmetry for projection exposure apparatuses having different field angles. There may be no image distortion. Also in this case, it is still difficult to obtain perfect matching accuracy with the conventional technique, and the same inconvenience occurs. The present invention has been made in view of the above problems, and an object thereof is to provide a projection exposure apparatus that can obtain high overlay accuracy even when used in combination with a projection exposure apparatus in which random image distortion without symmetry remains.

[0007]

In a projection exposure apparatus according to the present invention, a pattern on a mask (R) is projected onto a projection optical system (PL).
In a projection exposure apparatus that performs projection exposure on a photosensitive substrate (W) via a mask, the mask (R) and its projection optical system (PL)
And an optical member (8) for generating image distortion according to the image distortion characteristics of another exposure device (28, 29) that exposes the photosensitive substrate (W).

According to the projection exposure apparatus of the present invention, the optical member (8) can generate image distortion according to the image distortion characteristics of the other exposure apparatuses (28, 29). Therefore, for example, even when used in combination with another projection exposure apparatus in which random image distortion without symmetry remains, high overlay accuracy can be obtained. In this case, it is preferable to provide an inserting / removing means (21) for inserting / removing the optical member (8). Thereby, for example, a plurality of optical members corresponding to the image distortion characteristics of various exposure apparatuses are provided, and the optical members can be exchanged by the insertion / removal means (21) according to the corresponding exposure apparatus.

An example of another exposure apparatus is the exposure apparatus (28) used in the previous exposure step. Another example of the other exposure apparatus is an exposure apparatus (29) used in a subsequent exposure process.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An example of an embodiment of a projection exposure apparatus according to the present invention will be described below with reference to the drawings.
In this example, the present invention is applied to a stepper type projection exposure apparatus that collectively exposes a pattern on a reticle to each shot area on a wafer via a projection optical system.

FIG. 1 shows a schematic structure of a projection exposure apparatus of this example. In FIG. 1, the illumination light IL generated by a light source 1 passes through a shutter (not shown), and then a collimator lens and a fly-eye lens. Illumination uniformizing illumination system 2
The illuminance distribution is converted into a substantially uniform luminous flux. As the illumination light IL, for example, KrF excimer laser light or ArF is used.
Excimer laser light, harmonics of copper vapor laser and YAG laser, or ultraviolet emission line (g line,
i line) is used.

Illumination light IL which has passed through the illumination uniformizing illumination system 2
Is a variable diaphragm 3a for changing the illumination state of the reticle R
Incident on. The variable diaphragm 3a is a drive system (revolver) 3
Any one of the ordinary circular aperture, the small circular aperture, the deformed light source diaphragm including four small apertures decentered from the optical axis, or the ring-shaped diaphragm is rotated by b to rotate the illumination light I.
It is set on the optical path of L. As a result, the pattern of the exposure target (line width, pitch, line width, pitch, It is possible to select the illumination condition that provides the optimum image formation characteristics according to the periodicity, isolation, etc.). At the time of exposure, the illumination light IL for which a predetermined illumination condition is set by the variable diaphragm 3a further illuminates the reticle R via the relay lens 4a, the dichroic mirror 5 and the condenser lens 4b, and the reticle R is illuminated under the illumination light. The above circuit pattern is reduced and projected onto each shot area on the wafer W at a projection magnification β (β is 1/5, 1/4, etc.) via the projection optical system PL. Here, the Z axis is taken parallel to the optical axis AX of the projection optical system PL, and the X axis is taken parallel to the paper surface of FIG. 1 and the Y axis is taken perpendicular to the paper surface of FIG. 1 within a plane perpendicular to the Z axis. .

The reticle R is vacuum-sucked on the reticle holder 6, and the reticle holder 6 is placed on the reticle stage 7 via the expandable / contractible drive elements 11a and 11b. The driving of the reticle R (reticle holder 6) will be described in detail later. The position of the reticle stage 7 is measured by a laser interferometer (not shown), and the position information is sent to the main control system 25. The main control system 25 positions the reticle R via the reticle drive unit (not shown) and the reticle stage 7 based on the position information.

The wafer W is vacuum-sucked on the wafer holder 16, and the wafer holder 16 is held on the wafer stage WST. The wafer stage WST moves the wafer W by a wafer stage drive unit (not shown) including a motor and the like.
Driving in the so-called step-and-repeat method in the direction Y and the direction Y. Further, wafer stage WST is configured so that wafer W can be tilted in any direction with respect to the best image plane of projection optical system PL and that wafer W can be finely moved in the optical axis AX direction (Z direction). . Also, the wafer stage WS
T can rotate the wafer W around the optical axis AX within a predetermined range. A movable mirror 18 that reflects the laser beam from the laser interferometer 20 is fixed to the end of the wafer stage WST, and the position of the wafer stage WST in the XY plane is set by the laser interferometer 20 to, for example, 0.01 μm.
It is always detected with a resolution of about m. The position information (or speed information) of wafer stage WST is sent to main control system 25, and main control system 25 drives wafer stage WST based on this position information (or speed information).

The optical axis AX direction (Z direction) of the wafer W
As a focus position detection system for measuring the position (focus position) of the wafer W, and a light transmission optical system 14 for irradiating the surface of the wafer W with the detection light as spot light in the form of slits or pinholes from an oblique direction, and the surface of the wafer W. Of the oblique-incidence system including a light-receiving optical system 15 that collects the reflected light from the optical system, re-images the spot light on the vibrating slit, and photoelectrically converts the light flux passing through the vibrating slit to generate a detection signal. A focus position detection system (hereinafter, referred to as focus position detection system 14, 15 ") is provided. A focus signal is generated by synchronously rectifying the detection signal. In the focus position detection systems 14 and 15, the angle of parallel flat glass (plane parallel) (not shown) provided inside the light receiving optical system 15 is adjusted in advance so that the image plane becomes the zero point reference.
Autofocusing is performed so that the focus signal from the light receiving optical system 15 becomes zero. Focus position detection system 14, 1
The focus signal from No. 5 is sent to the main control system 25 via the image formation characteristic control system 22, and the main control system 25 produces a result when the focus position of the image formation surface in the image formation characteristic changes. Through the image characteristic control system 22, the angle of the parallel plate glass is finely moved so that the focus position of the wafer W follows the image plane. Alternatively, the tilt angle of the wafer W may be detected by irradiating the surface of the wafer W with a plurality of spot lights from the focus position detection system, and the tilt angle may be corrected by the auto-leveling method.

Further, a photoelectric detection system capable of measuring distortion is provided on wafer stage WST. This is disclosed in, for example, JP-A-59-94032, and photoelectrically detects the pattern on the reticle R via the projection optical system PL. this is,
This method is used in a method of directly observing an aerial image of the projection optical system PL to obtain an image forming characteristic, as opposed to a method of calculating and correcting a change in the image forming characteristic. As shown in FIG. 1, a pattern plate 17 is fixed on a wafer stage WST, and a photoelectric sensor 19 such as a silicon photodiode for detecting a light flux passing through the pattern plate 17 is arranged at the bottom of the pattern plate 17, The projection image of the reference mark on the reticle through the projection optical system PL is detected by the pattern plate 17 and the photoelectric sensor 19. The detection signal from the photoelectric sensor 19 (information regarding the projected image of the reference mark) is supplied to the main control system 25. The surface of the pattern plate 17 is installed so as to be almost at the same height as the surface of the wafer W, and photoelectric detection and the like are configured by the pattern plate 17 and the photoelectric sensor 19. A method of measuring the image forming characteristic of the projection optical system PL using this photoelectric detection system will be described below. In this measurement, the images of a plurality of reference marks formed on the reticle through the projection optical system PL are detected by a photoelectric detection system, and the intervals between the projected images of the reference marks are precisely measured to project the images. In order to detect the image formation characteristic of the optical system PL, a test reticle having an accurate reference mark position is used instead of the reticle R.

FIG. 2A shows a plan view of the pattern plate 17 corresponding to the test reticle. In FIG. 2A, the circular pattern plate 17 has a cross surrounded by a light shielding portion 32 at the center thereof. The light transmitting portion 31 including a slit is provided. This light transmitting portion has substantially the same size as the projected image of the reference mark 33 on the test reticle TR on the pattern plate 17 shown in FIG. 2B.

FIG. 2B shows the relationship between the test reticle TR and the photoelectric detection system on the wafer stage WST via the projection optical system PL. In FIG. 2B, the reference mark 33 of the test reticle TR is shown. The wafer stage WST of FIG. 1 is driven so that the light transmitting portion 31 of the pattern plate 17 crosses the image forming position in the X direction or the Y direction. The projected image of the reference mark 33 through the projection optical system PL is measured by the pattern plate 17 and the photoelectric sensor 19. For example, the wafer stage WS
By moving T in the X direction, the image of the reference mark 33 and the light transmitting portion 31 of the pattern plate 17 of the photoelectric detection system relatively move in the X direction.

FIG. 2C is a graph showing the relationship between the position of wafer stage WST and the amount of light received by photoelectric sensor 19, where the horizontal axis is the position x of wafer stage WST in the X direction and the vertical axis is the photoelectric sensor 19. The detection signal I from FIG. By moving the wafer stage WST in the X direction, as shown in FIG.
The waveform curve 34 as shown in (c) is obtained. The center x ₀ of the waveform curve 34 is measured as the center of the image forming position of the reference mark 33. At this time, since the position of the light transmitting portion 31 is precisely measured by the laser interferometer 20, an accurate image forming position of the image of the reference mark 33 of the test reticle TR is required. The measurement in the Y direction is performed in the same manner.

By performing the above-described measurement on a plurality of reference marks on the test reticle TR, the magnification error and image distortion of the projection optical system PL can be measured. There is also a method of improving the measurement reproducibility by using a mark composed of a plurality of lines instead of the single line as in this example as the reference mark. Further, although only one light transmitting portion 31 is used in this example, a plurality of light transmitting portions are provided according to the number of reference marks on the reticle,
A method of measuring multiple points at one time with one scan is also conceivable. In this case, if the distance between the light transmitting portions is strictly measured beforehand, there is an advantage that even if there is a measurement error in the laser interferometer 20, the measurement accuracy is not affected.

Further, instead of using the slit-shaped transmitting portion 31, the projected image of the reference mark is scanned by a pattern plate having a substantially square opening, and the photoelectric sensor 19 is scanned.
There is also a method of receiving light with. FIG. 3A shows a plan view of a pattern plate having a square opening. In FIG. 3A, a test reticle TR is formed in the X direction and the Y direction in a light shielding part 103 on the surface of a circular pattern plate 101. A square opening 102 having a width slightly larger than the maximum lengths of the projected image of the reference mark 33 in the X and Y directions and surrounded by sides parallel to the X and Y directions is formed. This method uses the opening 10 as a measurement reference in the X direction.
This is a method of using the edge 102a in the X direction of No. 2. Wafer stage WST is moved to X using this pattern plate 101.
When scanning in the direction, the amount of light incident on the photoelectric sensor 19 is measured as a so-called integrated amount. Figure 3
This will be described with reference to (b) and (c).

FIG. 3B shows a waveform curve 105 of the detection signal I from the photoelectric sensor 19, and FIG. 3C shows FIG.
A waveform curve 106 of a signal dI / dx obtained by differentiating the detection signal I from the photoelectric sensor 19 in (b) at the position x of the wafer stage WST is shown. In FIGS. 3A and 3B, the horizontal axis represents the position x. As wafer stage WST moves, detection signal I gradually increases as shown by waveform curve 105 in FIG. 3B, and reaches a constant value at a certain position. As the edge of the opening 102 approaches the image forming position, the rising angle of the detection signal I increases,
The angle becomes smaller and converges to 0 as the edge moves away from the image forming position. That is, from the photoelectric sensor 19,
Since a detection signal in the form of integrating the waveform curve 34 of FIG. 2C is obtained, it is necessary to differentiate this detection signal to obtain an output signal corresponding to the waveform curve 34. The waveform curve 106 of FIG. 3C is obtained by differentiating the waveform curve 105 of FIG. 3B by the position x of the wafer stage WST, and an output signal corresponding to the waveform curve 34 of FIG. 2C is obtained. ing. Therefore, the imaging position is calculated by the same method as described with reference to FIG. The method of using the large opening portion 102 as the light transmitting portion in this way has an aspect that the calculation process is complicated, but is advantageous in that the light amount may be small and the pattern plate 101 can be easily formed. There is also an advantage that marks with various line widths can be measured with one transmission part. In addition to this, it is also possible to use a method of enlarging the formed image and measuring it with a photoelectric sensor such as a CCD capable of one-dimensional or two-dimensional measurement. There is also a method of actually exposing the reference pattern, that is, measuring the image forming characteristics by test exposure.

Further, the apparatus shown in FIG. 1 is provided with a correction mechanism for correcting the image forming characteristic of the projection optical system PL. This correction mechanism is mainly composed of a first correction mechanism that corrects image forming characteristics such as symmetric image distortion, and a second correction mechanism that mainly corrects image forming characteristics such as asymmetric image distortion. . First, the first correction mechanism will be described. The image forming characteristics of the projection optical system PL include a focus position (focus position), field curvature, distortion (magnification error, image distortion, etc.), astigmatism, etc., and mechanisms for correcting them are conceivable. Now, the correction mechanism for distortion will be described. In this example, the imaging characteristics are determined based on the detection result by the photoelectric detection system including the photoelectric sensor 19 and the prediction result of the imaging characteristics of the projection optical system PL itself due to atmospheric pressure change, illumination light absorption, change of illumination conditions, and the like. to correct.

In FIG. 1, the first correction mechanism comprises a drive mechanism for the reticle R and a drive mechanism for the lens element 12 closest to the reticle in the projection optical system PL. That is, the image forming characteristic is corrected by driving the reticle stage 6 on which the reticle R is mounted or the lens element 12 in the projection optical system PL by the image forming characteristic control system 22.

First, the driving of the lens element 12 will be described. In the projection optical system PL, the lens element 12 closest to the reticle R is fixed to the support member 9, and the lens element 13 following the lens element 12 is fixed.
Etc. are fixed to the main body of the projection optical system PL. In this example, the optical axis AX of the projection optical system PL indicates the optical axis of the optical system of the main body of the projection optical system PL below the lens element 13. The support member 9 is connected to the lens barrel main body of the projection optical system PL via a drive element (two drive elements 10a and 10b are shown in FIG. 1) composed of two or more piezo elements which are expandable and contractible. ing. in this case,
By expanding and contracting the drive elements 10a and 10b, the lens element 12 can be moved parallel to the optical axis AX. Further, by providing three driving elements and independently expanding and contracting, the lens element 12 can be tilted with respect to a plane perpendicular to the optical axis AX, and by these operations, the image forming characteristics of the projection optical system PL, for example, projection. Magnification, distortion,
Field curvature, astigmatism, etc. can be corrected.

Here, the lens element 12 is the optical axis AX.
When the lens is moved in parallel in the direction of, the projection magnification of the projection optical system PL (magnification from the reticle to the wafer) changes at a change rate according to the movement amount. When the lens element 12 is tilted with respect to a plane perpendicular to the optical axis AX, one projection magnification is increased and the other projection magnification is reduced with respect to the rotation axis, so that a so-called square image is formed. It is possible to cause a trapezoidal deformation. On the contrary, the trapezoidal distortion can be corrected by the inclination of the lens element 12.

Next, driving of the reticle R will be described. As described above, the distance between the projection optical system PL and the reticle R can be changed by expanding and contracting the drive elements 11a and 11b formed of the piezo elements on the bottom surface of the reticle holder 6. Here, when the reticle R moves in parallel to the optical axis AX, it is possible to generate aberration called so-called pincushion (or barrel) distortion in the projected image. As the driving elements 10a and 10b for driving the lens element 12 of the projection optical system PL and the driving elements 11a and 11b for driving the reticle R, other electrostrictive elements or magnetostrictive elements are used in addition to the piezo element. it can.

As described above, the projection optical system PL is driven by driving the reticle R or the lens element 12.
The projection magnification or image distortion can be optimally corrected. Further, by driving these, the focus position or the tilt angle of the image forming plane changes, but the amount is the focus position detection system 1
It is fed back as offsets of 4 and 15, and is controlled so that the focus position of the surface of the wafer W always coincides with the average focus position of the image plane of the projection optical system PL.

The first for correcting the distortion of the projected image
The correction mechanism is not limited to the above-mentioned mechanism. For example, a mechanism for inserting a glass plate having a slightly delicate curvature to correct image distortion in the space between the projection optical system and the reticle, or the projection optical system. A glass plate with a variable thickness in the space between the system PL and the reticle R (for example, an optical wedge)
A mechanism or the like for inserting can also be used. In particular, this method of inserting a glass plate having a variable thickness is almost equivalent to the method of moving the reticle R up and down, but the same effect can be obtained without adversely affecting the rigidity of the reticle stage 4. Further, various methods have been proposed, such as a method of sealing a gas chamber between some lenses of the projection optical system PL to change the pressure or the composition of air, and these methods can be used as well.

These first correction mechanisms normally form an image due to changes in atmospheric pressure, absorption of illumination light of the projection optical system PL, changes in illumination conditions, etc., in addition to correction of the measured image formation characteristics. It is used to correct changes in characteristics. This will be briefly described below. First, the correction corresponding to environmental changes such as changes in atmospheric pressure will be described. The main control system 25 includes an environment sensor 2 including an atmospheric pressure sensor and a temperature sensor.
6 is supplied, and the main control system 25 calculates the amount of change in the imaging characteristic based on the information by using a coefficient obtained in advance by calculation or experiment or a table. Further, the correction amount of each correction means such as the driving elements 11a and 11b is obtained, and the result is the image formation characteristic control system 2.
2 as a control signal. Based on this control signal,
The imaging characteristic control system 22 includes drive elements 10a, 10b, 11
The lens elements 12 or the reticle R are controlled by driving a and 11b.

Regarding the absorption of the illumination light of the projection optical system PL, for example, a photoelectric sensor (not shown) on the wafer stage WST measures the amount of illumination light passing through the projection optical system PL before the actual exposure operation. The main control system 25 stores in advance a mathematical model such as a differential equation for calculating the change amount of the image forming characteristic with respect to the illumination light amount. By monitoring the illumination light amount, the image forming characteristic is changed every moment. Is calculated. Based on the amount of change, correction can be performed by the correction mechanism as in the case of the above-described change in environment. Further, regarding the change of the illumination condition, the change amount of the image forming characteristic can be calculated and corrected based on the setting information of the rotation angle of the drive system 3b of the variable diaphragm 3a.

Next, the structure and operation of the second correction mechanism will be described in detail. The second correction mechanism mainly corrects the asymmetric image distortion as described above, and corrects the imaging characteristics by distorting the projected image of the reticle R in accordance with the image distortion on the wafer W. Be seen. First, the configuration will be described. As shown in FIG. 1, a correction plate 8 made of a transparent optical member such as a glass plate is arranged in parallel with the XY plane at a substantially intermediate position between the reticle R and the lens element 12 of the projection optical system PL. The correction plate 8 is provided so as to be freely inserted and removed by the exchange device 21, and the other correction plates 24a to 24a stored in the storage 23 provided near the exchange device 21.
It is designed to be exchanged with c, etc. when necessary. These correction plates 8 and 24a to 24c are used last time by changing the shape of the projected image of the projection optical system PL.
Or, in addition to generating an image distortion according to the distortion characteristic of the projection exposure apparatus to be used next time, in the projection exposure apparatus of this example used in the present exposure process, the first correction mechanism corrects a random distortion that is difficult to correct. It is also used for the purpose. The shape of the correction plate will be described later.

Next, the operation of the second correction mechanism will be described.
4 and FIG. As mentioned above, the projection light
The distortion of the academic PL is test exposure or photoelectric
It is measured by the method using the sensor 19, and first, the first correction mechanism
Is corrected by. However, the disc that cannot be corrected
The torsion remains. Figure 4 (a) shows the residual distortion.
An example of the residual distortion
It is shown as a pattern shift on the reticle R. Immediately
The residual distortion of the projected image on the wafer.
Figure 4 shows the state converted to the amount of lateral displacement of the pattern on the roof.
It is shown in (a). In FIG. 4A, the dotted line
Ideal lattice (corresponding to a lattice-like projected image without distortion)
At each grid point of the pattern 35,
There is a gap. In this case, the direction and size of the arrow
Indicates the direction and size of the deviation of the projected image. This Figure 4
As shown in (a), the direction and size of these arrows
Are both inconsistent and randomly distributed. example
For example, four adjacent grid points P on the upper left_1,1, P_1,2, P
_2,1, P_2,2Is the point Q on the upper left side_1,1, Lower left
Point Q_1,2, Upper right point Q_2,1, And the lower right point Q_2,2Ma
It is off. If there is a gap between the positions
It has various directions. Also, the grid points of other corners
P_1,5, P_5,1, P_5,5Also point Q in the lower left _1,5,
Point Q on the upper left_5,1, And the lower point Q_5,5Deviated
You. That is, regardless of the partial area and the entire area
The direction is random. In addition, the size of these deviations is
It is undam.

Random image distortion components having no symmetry cannot normally be corrected by the conventional method. However, in this example, the random image distortion component is corrected using the correction plate (in this case, the correction plate 8).
In FIG. 4B, the correction plate 8 has a reticle R and a projection optical system PL.
FIG. 4 shows a cross-sectional view of the state of being arranged between FIG.
As shown in (b), the upper surface 8a of the correction plate 8 is formed substantially parallel to the reticle R, while the lower surface 8b is formed.
Has a local angle and is polished so that the chief ray passing through the reticle R bends in a direction to cancel the distortion. In this case, the sectional shape of the correction plate 8 is
An example of correcting the distortion of the projected image on the straight line 36 connecting the central grid point P _3,1 to the grid point P _3,5 of the ideal grid 35 of FIG. 4A is shown. Therefore, the grid point P
4 the principal ray L _3,1 in (b) through the _3,1 grid point P _3,1
It can be bent in the direction that cancels the shift in the upper left direction at, that is, in the lower right direction. Similarly, the chief ray L _3,2 ~L _{3, 5} also lattice points P _3,2 ~ passing the grid points P _3,2 ~ lattice point P _{3, 5,} respectively
It can be bent in such a direction as to cancel each distortion at the lattice points P _3,5 . The entire shape of the correction plate 8 is polished into a shape corresponding to the distribution of distortion shown in FIG.

By thus polishing the correction plate into a predetermined shape, the corresponding distortion can be corrected.
However, if an attempt is made to correct an excessively large amount of image shift by the correction plate 8, the thickness of the correction plate 8 will change greatly depending on the location, and the image plane and spherical aberration will be adversely affected. It is preferable that the correction is performed and only the residual amount is corrected by the correction plate 8.

An example of correcting random distortion in the projection exposure apparatus of this example has been described above.
Now, the final circuit pattern on the wafer W is formed by repeating a series of processes including steps such as exposure, etching, and vapor deposition. Therefore, a plurality of projection exposure apparatuses are mixedly used. FIG. 1 also shows a projection exposure apparatus 28 used in the previous exposure step and a projection exposure apparatus 29 used in the next exposure step. That is, the wafer W is processed in the order of the projection exposure apparatus 28, the projection exposure apparatus of this example, and the projection exposure apparatus 29, as indicated by the arrow.

In this case, between projection exposure apparatuses whose distortion is highly corrected by the correction plate as in this example, even if various different projection exposure apparatuses are mixed, the overlay accuracy is deteriorated due to the difference in distortion. Does not occur.
Therefore, it is not necessary to replace the correction plate 8 once inserted as it is. However, when a projection exposure apparatus that does not have a correction plate as in this example is mixed, even if the distortion of only one projection exposure apparatus is highly corrected, the overlay accuracy will not be good. However, in this example, even in such a case, it is possible to realize a highly accurate overlay accuracy by exchanging the correction plate.

Usually, the circuit on the wafer W is formed by stacking several layers of patterns. For a projection exposure apparatus that forms a pattern on the first layer (first layer) of the wafer W, there is no previous exposure step, and there is no restriction on distortion characteristics. Therefore, it is possible to always use a correction plate that matches the distortion characteristics of the projection exposure apparatus used in the next exposure step, and this improves the matching accuracy with the projection exposure apparatus used in the next exposure step. . Even in the second and subsequent layers, there may be no restriction on the distortion characteristics of the projection exposure apparatus used in the previous exposure step. In such a case, the distortion characteristics of the projection exposure apparatus used next time may not be limited. A matched compensator can be used.

Further, for a projection exposure apparatus that forms a pattern on the second and subsequent layers on the wafer W, a correction plate that normally matches the distortion characteristics of the projection exposure apparatus used when forming the pattern of the previous layer is used. Therefore, good matching accuracy with the previously used projection exposure apparatus can be obtained. Hereinafter, in FIG. 1, the projection exposure apparatus 2 used for exposing the wafer W in the previous exposure step.
An example in which a correction plate is used according to the projection exposure apparatus 29 used for exposing the wafer W in the 8th or next exposure step will be described.

First, prior to the exposure operation, data such as to which projection exposure apparatus the image distortion of the wafer W to be exposed this time should be adjusted is input from the operator or the like via the input means 27 such as a keyboard. The projection exposure apparatus selected in this case is the projection exposure apparatus 28 or the projection exposure apparatus 29,
The operator inputs the type of the projection exposure apparatus 28 or the projection exposure apparatus 29. Based on this information, the main control system 25 selects the appropriate correction plate and instructs the exchange device 21 to exchange the correction plate. The exchange device 21 takes out a specified correction plate from the storage cabinet 23 in which the correction plates 24a, 24b, 24c for exchange created in accordance with the image distortion of the projection exposure apparatus that may perform overlay exposure are stored. , The reticle R and the projection optical system PL as in the previous correction plate 8.
Place it in place between. By the exchange device 21, the correction plate is positioned within a predetermined error. It should be noted that, as described above, the correction plate on which the original ideal lattice should be projected, as described above, can be accurately polished to obtain a desired image distortion by further polishing the correction plate as much as desired. Therefore, it is possible to provide a correction plate that can be applied to a projection exposure apparatus having any distortion characteristic. As a result, it is possible to generate image distortion suitable for any projection exposure apparatus provided with a correction plate, and improve overlay accuracy. Instead of the replacement device 21, the operator may manually replace the correction plate.

Further, as described above, inconvenience occurs if all the image distortion components are provided in the correction plate. Therefore, regarding the symmetrical component of the image distortion, data is previously stored in the memory of the main control system 25 for each projection exposure apparatus. In the image forming characteristic control unit 2 based on the data.
It is desirable to perform the correction mainly by the first correction mechanism via 2.

Further, the projection exposure apparatus has an alignment function of reading the alignment mark already formed on the wafer W, and positioning the wafer W thereby to perform overlay exposure. As an example of this alignment mechanism, the alignment mark of a partial shot area (sample shot) in the wafer W is read prior to exposure to perform EGA.
A scaling error or the like caused by the process processing of the wafer W is calculated by statistical calculation such as (enhanced global alignment) method, and the magnification correction or the distortion of the symmetric component is corrected. In this case, since it is unlikely that the shot area (chip) will be distorted randomly due to the process processing, it is considered that the conventional correction of the symmetrical component is sufficient. That is, the correction can be performed only by the first correction mechanism. Further, in this case, even if the projection exposure apparatus 28 used in the previous exposure step has a variation in the symmetrical distortion component such as a magnification error, the distortion component is also included in the measurement, so that no error occurs. That is, even in such a case, the conventional correction function such as the second correction mechanism of the present example and the first correction mechanism for correcting the symmetrical component is sufficient.

Further, some projection exposure apparatuses have different exposure field sizes capable of performing exposure at one time, and it is conceivable to mix them. In this case, even if the projection optical systems having the same size of the exposure field have the same symmetric distortion, when the overlapping exposure is performed on the exposure area of, for example, 1/4 of one exposure field, the asymmetrical distortion occurs. It may cause distortion,
In such a case, it cannot be corrected by the conventional correction means.

FIG. 5 is a diagram for explaining the asymmetry of distortion in two projection exposure apparatuses having different exposure field sizes (the first projection exposure apparatus and the second projection exposure apparatus). FIG. 5A shows a distortion state in the first projection exposure apparatus, and FIG.
FIG. 6 is a diagram for explaining a method of correcting distortion in the first projection exposure apparatus in the second projection exposure apparatus. In FIG. 5A, the ideal grating 35A and the projected image 3 by the first projection exposure apparatus corresponding to the ideal grating 35A.
A so-called pincushion-type symmetric distortion is generated between 6 and 6. Then, the projected image 36 is converted into the ideal lattice 35.
As shown in the enlarged view of FIG. 5B, the second projection exposure is performed on the projection image A having the area of the upper left ¼ divided by the symmetry lines 38X and 38Y in the X and Y directions of A. The projection image 37 is superimposed and exposed by the apparatus. In that case,
The distortion between the projected image A and the ideal grating 35B in the second projection exposure apparatus has no symmetry and is asymmetric. Similarly, the remaining three projected images B to D divided into four in FIG. 5A are also asymmetric with respect to the second projection exposure apparatus. By using the correction plate of this example, such asymmetrical distortion can be corrected.

First, the projected image 37 is formed by the conventional correction mechanism similar to the above-mentioned first correction mechanism of the second projection exposure apparatus.
Distort into a rhombus like. Then, the remaining asymmetric component may be distorted by using a correction plate. FIG.
Since the four projected images A to D for superimposing (a) are each rotated by 90 °, the same correction plate may be rotated by 90 ° and inserted.

In this example, since the shape of the distortion is changed as described above, it is more reliable to perform the exposure while confirming what the actual shape of the distortion is. For this reason, it is desirable to measure the distortion state in advance by the photoelectric sensor 19 before exposure to confirm whether or not the desired distortion is achieved. This makes it possible to prevent mistakes such as mistaken replacement of the correction plate.

As described above, the method of this example is useful for mixing with a conventional apparatus that is not highly corrected by the correction plate. The device as in this example is not necessary between devices that can be corrected by the correction plate. In any projection exposure apparatus, since exposure is performed almost on the ideal grid as it is, superposition is performed with high accuracy. However, if any one of the conventional projection uncorrected devices is inserted between these projection exposure apparatuses, the overlay accuracy is lowered. However, according to the method of this example, if the image distortion of another device is used in accordance with the image distortion of that device, there is no problem in superposition. In particular, when a plurality of conventional projection exposure apparatuses are mixed, exposure is performed in accordance with each image distortion before and after a process in which overlay accuracy is not so demanded, and image distortion is different in a process in which precision is not required. Just stack them on top of each other. Further, when the image distortion characteristics of a plurality of conventional devices are similar to some extent, a method of generating and using image distortion in accordance with these average image distortion characteristics can be considered. In addition, there is also a method in which those having similar image distortion characteristics are divided into groups and used separately before and after a process in which accuracy is not so required.

According to a method of processing a correction plate that realizes an almost ideal lattice so as to obtain a desired image distortion, a correction plate that produces the same image distortion as that of a certain apparatus is a main one. The correction plate is unique to each projection exposure apparatus to which the example can be applied. Therefore, it is necessary to prepare a large number of correction plates. In order to solve this inconvenience, if the correction plate for overlay correction is inserted above or below the correction plate that realizes the original ideal lattice, the correction that causes the same image distortion as that of a certain device is generated. Since the plate is common to all devices, the number of correction plates to be prepared can be significantly reduced.

As described above, the projection exposure apparatus of recent years has a variable aperture (variable aperture 3 shown in FIG. 1 for improving the resolution.
The illumination conditions are variable due to a). When the illumination conditions change, the passage paths of the light rays inside the projection optical system differ,
For this reason, the distortion changes under the influence of subtle nonuniformity inside the projection optical system. That is, when the illumination conditions change, the distortion becomes that of another projection optical system. Even in such a case, the correction plate of this example is useful, and can contribute to the improvement of overlay accuracy even when the same projection exposure apparatus has different exposure conditions such as illumination conditions. Therefore, it is desirable to manufacture and use the correction plate by distinguishing it according to the illumination condition even with one projection exposure apparatus. Of course, if the amount of change in the random component of the distortion change due to the illumination condition is sufficiently small and can be ignored, it is not necessary to apply this example.

When the distortion characteristics of the projection optical system change due to maintenance work of the projection exposure apparatus or the like, it is necessary to take measures such as correcting the correction plate or manufacturing a new correction plate. Further, since the method of this example requires management of information between a plurality of projection exposure apparatuses, it is desirable to manage that information by a host computer that can manage the entire exposure system.

The present invention is not limited to a stepper type projection exposure apparatus, but a scanning type projection method such as a step-and-scan system in which a reticle and a wafer are synchronously scanned to sequentially transfer the pattern on the reticle onto the wafer. The same can be applied to the exposure apparatus. As described above, the present invention is not limited to the above-described embodiments, and various configurations can be taken without departing from the gist of the present invention.

[0052]

According to the projection exposure apparatus of the present invention, an optical member for generating an image distortion corresponding to the image distortion characteristics of another projection exposure apparatus is inserted between the mask and the projection optical system, so that the projection optical system can be used. The image distortion of the system can be matched with the image distortion caused by another projection exposure apparatus. Therefore, for example, even when used in combination with a projection exposure apparatus in which random image distortion without symmetry remains, there is an advantage that high overlay accuracy can be obtained.

Further, when the inserting / removing means for inserting / removing the optical member is provided, the optical member can be quickly and accurately positioned and replaced according to the corresponding projection exposure apparatus by the inserting / removing means. When the other projection exposure apparatus is the projection exposure apparatus used in the previous exposure step, the image distortion in the projection exposure apparatus used in the previous exposure step is corrected and used in the previous exposure step. Good matching accuracy can be obtained with the projection exposure apparatus.

Further, when the other projection exposure apparatus is a projection exposure apparatus used in the subsequent exposure step, image distortion corresponding to the image distortion characteristic in the projection exposure apparatus used in the subsequent exposure step may occur. Since this occurs, good matching accuracy can be obtained with the projection exposure apparatus used in the subsequent exposure process.

[Brief description of drawings]

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

FIG. 2 is a diagram for explaining an example of a method for measuring image distortion of the projection optical system in FIG.

FIG. 3 is a diagram for explaining another example of the method for measuring the image distortion.

4A is a diagram showing an example of an asymmetric residual distortion, and FIG. 4B is a sectional view showing a shape of a correction plate 8 for correcting the residual distortion.

FIG. 5 is a diagram for explaining a method of matching image distortion between projection exposure apparatuses having different exposure field sizes in the embodiment of the present invention.

[Explanation of symbols]

 R Reticle PL Projection optical system W Wafer 7 Reticle stage 8, 24a to 24c Correction plate 10a, 10b, 11a, 11b Driving element 12 Lens element 19 Photoelectric sensor 21 Exchange device 22 Imaging characteristic control system 23 Storage box 25 Main control system 28 Projection exposure apparatus used in the previous exposure step 29 Projection exposure apparatus used in the next exposure step

Claims (4)

[Claims] 1. A projection exposure apparatus for projecting a pattern on a mask onto a photosensitive substrate via a projection optical system, wherein the photosensitive substrate is exposed between the mask and the projection optical system. A projection exposure apparatus comprising an optical member for generating an image distortion according to an image distortion characteristic of the exposure apparatus. 2. The projection exposure apparatus according to claim 1, further comprising an insertion / removal means for inserting / removing the optical member into / from an imaging light beam between the mask and the projection optical system. Projection exposure system. 3. The projection exposure apparatus according to claim 1, wherein the other exposure apparatus is an exposure apparatus used in a previous exposure step. 4. The projection exposure apparatus according to claim 1 or 2, wherein the other exposure apparatus is an exposure apparatus used in a subsequent exposure process.