JPH1145846A - Scanning type exposure method and aligner - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain good imaging characteristic by measuring the shape of a pattern surface of a reticle, when a gap between a stage of the reticle side and a projection optical system is narrow. SOLUTION: In the state that a part of a pattern of a reticle R is projected on a wafer W via a projecting optical system PL under illumination light for exposing, the reticle R and the wafer W are synchronously scanned with respect to the projection optical system PL, and exposure is performed. A surface shape detecting system 30 is arranged at a position which is deviated fr to the scanning direction to a space between the projecting optical system PL and a reticle stage RST, and the surface shape of a pattern surface of the reticle R is measured via the surface shape detecting system 30 during the approach-rum period of scanning exposure and the like. The changing amount of imagery characteristics like an image surface position is calculated from the result of the measurement, and imagery characteristics of the projection optical system PL or forcus position of the water W is so controlled that the changing amount is corrected.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure method used for transferring an image of a mask pattern onto a substrate during a lithography process for manufacturing, for example, a semiconductor device, a liquid crystal display device, or a thin film magnetic head. More specifically, the present invention relates to a scanning type exposure method and apparatus such as a step-and-scan method for performing exposure by synchronously scanning a mask and a substrate.

[0002]

2. Description of the Related Art In manufacturing a semiconductor device or the like, in addition to a batch exposure type projection exposure apparatus such as a stepper, a scanning exposure type projection exposure apparatus such as a step-and-scan method (scanning type exposure apparatus) ) Is also being used. In the projection optical system of this type of projection exposure apparatus, a resolution close to the limit is required. Therefore, factors that affect the resolution (for example, atmospheric pressure, environmental temperature, etc.) are measured, and an image is formed in accordance with the measurement result. A mechanism for correcting characteristics is provided. Also, the numerical aperture of the projection optical system is set large to increase the resolution,
As a result, the depth of focus is quite shallow,
An oblique incidence type focus position detection system (AF sensor) measures the focus position of the unevenness on the surface of the wafer as a substrate (position in the optical axis direction of the projection optical system), and projects the wafer surface based on the measurement result. An autofocus mechanism for adjusting to the image plane of the optical system is provided.

However, in recent years, an imaging error due to deformation of a reticle as a mask cannot be ignored. That is, if the pattern surface of the reticle bends almost uniformly toward the projection optical system, the average position of the image plane also decreases, and defocus occurs when the wafer focus position is the same. Further, when the pattern surface of the reticle is deformed, the position of the pattern on the pattern surface in the direction perpendicular to the optical axis of the projection optical system may also change, and such a lateral displacement of the pattern also causes a distortion error. .

[0004] Such reticle deformations can be classified according to factors: (a) deflection due to its own weight, (b) deformation during polishing of the glass substrate itself of the reticle, and (c) when the reticle is forcibly sucked and held on the reticle holder. In addition, deformation or the like that occurs in accordance with the flatness of the contact surface between the two may be considered. Since the state of such a reticle deformation differs for each reticle and further for each reticle holder of the exposure apparatus, in order to accurately measure the amount of deformation of the reticle, the reticle must be actually mounted on the reticle holder of the projection exposure apparatus. It is necessary to carry out the measurement in a state where it is adsorbed and held.

As a method of measuring the amount of deformation of the pattern surface of the reticle held by suction on the reticle holder, a method of first performing test printing via a projection optical system can be considered. That is, images of a plurality of predetermined patterns formed on the pattern surface are projected onto a wafer for evaluation via a projection optical system, and the projection is repeated while changing the focus position of the wafer in a stepwise manner. The best focus position of the image of each pattern is determined from the condition that the contrast of the subsequent resist pattern image is the highest. At this time, the amount of deformation of the pattern surface can be calculated to some extent quantitatively from the shift amount of the best focus position.

[0006]

As described above, in order to obtain higher imaging performance in a projection exposure apparatus, it is desirable to measure the shape of the pattern surface not only on the wafer but also on the reticle side. However, the method for performing the test print takes time and reduces the throughput of the exposure process. Furthermore, since a pattern for transfer is formed on the reticle for actual exposure, the test printing method cannot be applied. Therefore, in order to quickly measure the surface shape of the reticle, a position sensor similar to the oblique incidence type AF sensor for detecting the focus position of the wafer may be arranged on the reticle stage side.

In this case, the pattern surface of the reticle is a lower surface,
That is, since the surface is on the side of the projection optical system, the oblique incidence type position sensor may be arranged in the space between the reticle stage and the projection optical system or in the vicinity thereof. However, a dustproof film (pellicle) may be provided on the reticle via a metal frame to prevent foreign matter from adhering to the pattern surface. In this case, since there is a restriction that the oblique incident light is not blocked by the metal frame, it is impossible to irradiate the reticle pattern surface with the detection light at an extremely shallow angle (large incident angle) from the position sensor.

In particular, in a scanning exposure apparatus, the reticle stage needs to have sufficient rigidity so as not to be deformed even when subjected to stress during acceleration / deceleration for synchronous scanning. In many cases, a structure having a sufficient thickness up to the limit of almost contacting the surface is adopted.
Furthermore, since the design of the projection optical system is easier if the space between the reticle and the projection optical system is narrower, the space between the projection optical system and the reticle becomes more and more as the accuracy of the projection optical system increases. It tends to be less. Therefore, it is difficult to dispose a reticle position sensor between the projection optical system and the reticle.

In addition, a reticle position sensor is required to have extremely high stability. This is because if the measured value of the position sensor changes with time, it is determined that the surface shape of the reticle has changed, and erroneous imaging characteristics are corrected. In view of the above, the present invention has a narrow space between the stage on the reticle side and the projection optical system, even if it is difficult to install a sensor for measuring the shape of the pattern surface of the reticle in that space, It is an object of the present invention to provide a scanning type exposure method capable of measuring the shape of the pattern surface and thereby obtaining good imaging characteristics. Still another object of the present invention is to provide a scanning exposure apparatus capable of performing such a scanning exposure method.

[0010]

In the scanning exposure method according to the present invention, the pattern of the mask (R) and the substrate (W) are moved synchronously so that the pattern of the mask (R) and the substrate (W) are projected.
In the scanning exposure method of transferring onto the substrate via L), the shape of the pattern surface of the mask (R) is measured prior to scanning and exposing the substrate with the pattern image of the mask, Based on the measurement result of the pattern surface, at least one of the imaging characteristic of the projection optical system (PL) and the position of the substrate (W) is corrected.

According to the present invention, for example, the shape of the pattern surface of the mask can be measured when at least a part of the pattern surface of the mask is located outside the area to be transferred by the projection optical system. The sensor for measurement (30) can be arranged at a position distant from the projection optical system in the scanning direction. Therefore, even if the space between the projection optical system and the stage on the mask side is narrow, the sensor (30) can be easily installed, and the mask (30) can be easily used by using the sensor (30).
Can be measured. By correcting the distortion of the projection optical system or the focus position of the substrate according to the amount of deformation, good imaging characteristics can be obtained.

It is desirable that the shape measurement of the pattern surface of the mask (R) is performed when the mask is at the approach start position or when the mask is at the approach position. That is, for example, if the surface shape is measured once when the pattern surface of the mask is in the approaching start position or in the approaching section after the replacement of the mask, the correction can be performed during the subsequent exposure. Further, if the surface shape can be measured at the approach position, it is not necessary to extend the movement stroke of the mask for the surface shape measurement. The surface shape of the mask may be measured at the time of scanning exposure, at the time of deceleration after the end of scanning exposure, or at the time of stop after deceleration.

Next, the scanning exposure apparatus according to the present invention moves the mask (R) and the substrate (W) in synchronism, so that the pattern on the mask is projected onto the substrate via the projection optical system (PL). A scanning exposure apparatus that has a detection point outside a transfer target area (44) by a projection optical system (PL) and measures a shape of a pattern surface of a mask; A correction system (15, 19, 23, 23) that corrects at least one of the imaging characteristic of the projection optical system and the position of the substrate based on the measurement result of the shape measurement system.
26).

According to the scanning exposure apparatus of the present invention,
The shape measuring system (30) can be arranged at a position away from the projection optical system in the scanning direction, and the shape of the pattern surface of the mask can be easily measured by this shape measuring system. Further, by scanning the mask along the shape measurement system, the shape of the entire pattern surface of the mask can be measured. Thereby, the scanning exposure method of the present invention can be performed.

In this case, the reference surface (10) for calibrating the shape measuring system (30) is placed on the mask stage (7, RST) for moving the mask (R), and substantially the same as the pattern surface of the mask (R). It is desirable to form them at the same height. At this time, when measuring using the shape measuring system (30), first measure the position of the reference plane (10), and then measure the surface position of the mask (R) based on the measured value. I do. That is, by measuring the difference between the reference plane (10) and the plane position of the mask (R), the stability of the measurement value of the shape measurement system (30) can be determined by measuring the reference plane (10) and then measuring the mask (R). It suffices if R) is maintained for a short time until the measurement of R) is completed. Therefore, the shape measurement system (3
Even when the measured value of 0) changes over time during a predetermined time,
Accurately measure changes in imaging characteristics due to mask shape,
Corrections can be made.

Also in this case, it is desirable that the shape measurement of the pattern surface of the mask is performed when the mask is at the approach start position or when the mask is at the approach position.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. In this example, the present invention is applied to a case where exposure is performed using a step-and-scan type projection exposure apparatus. FIG. 1 shows a projection exposure apparatus of the present embodiment. In FIG. 1, illumination light IL for exposure emitted from an exposure light source 1 is shaped including a relay optical system, a fly-eye lens for uniformizing the illuminance distribution, and the like. Optical system 2
Incident on. As the illumination light IL, i-line of a mercury lamp (wavelength 365 nm), KrF (wavelength 248 nm),
Alternatively, excimer laser light such as ArF (wavelength 193 nm) or a harmonic of a YAG laser can be used. The exit surface of the shaping optical system 2 corresponds to a pupil position with respect to the surface on which the reticle to be transferred is disposed, and a turret plate 3 on which various aperture stops for changing illumination conditions for the reticle are provided. Have been. On the periphery of the turret plate 3, there are a normal circular stop, a variable circular stop capable of changing the coherence factor (σ value) of the illumination system, a ring stop for annular illumination, and a four-division stop for modified illumination. The desired aperture stop can be selected by rotating the turret plate 3 with the drive motor 3a.

The illumination light IL that has passed through one aperture stop in the turret plate 3 further passes through a reticle R via an optical system 4 including a field stop (reticle blind) and the like, a mirror 5 for bending the optical path, and a condenser lens 6. Illuminate a slit-shaped illumination area on the pattern surface (lower surface). The image of the pattern in the illumination area of the reticle R is projected through a double-sided telecentric projection optical system PL at a projection magnification β (β is 1
/ 4 or 1/5) and projected onto a slit-shaped exposure area on the wafer W coated with the photoresist. Hereinafter, Z is parallel to the optical axis AX of the projection optical system PL.
An axis is taken, and an X axis is taken along a scanning direction of the reticle R and the wafer W during scanning exposure (that is, a direction parallel to the plane of FIG. 1) in a plane perpendicular to the Z axis. The description will be made taking the Y axis along the scanning direction (that is, the direction perpendicular to the paper surface of FIG. 1).

The reticle R is held on the reticle holder 7 by vacuum suction at a four-point support as an example, and two ribs 8A and 8B are provided at the bottom of the reticle holder 7 in parallel so as to increase rigidity to withstand high-speed movement. Are fixed on the reticle stage RST, and the reticle stage RST is continuously moved in the Y direction on the reticle base 9 by a linear motor.
Fine movement in the direction, Y direction, and rotation direction. Reticle holder 7
The movable mirror 11m fixed to the upper end of the reticle and an external laser interferometer 11 allow the reticle holder 7 (reticle R)
The dimensional position is measured, and the reticle stage drive system 12 controls the operation of the reticle stage RST based on the measured values and control information from the main control system 14 that controls the overall operation of the apparatus.

A reference member 10 made of a glass substrate having a good bottom surface (hereinafter referred to as a "reference surface") is fixed to a region adjacent to the reticle R in the X direction on the reticle stage RST. . The reference plane of the reference member 10 is
In design, the height is set to be the same as the pattern surface of the reticle R, and it is almost the same size as the slit-shaped illumination area for the reticle R.
An evaluation mark for measuring imaging characteristics such as distortion and image plane is formed.

FIG. 3 (a) is a plan view showing the reference member 10. In FIG. 3 (a), the reference surface (bottom surface) of the reference member 10 has, for example, two rows in the X direction for a cross-shaped evaluation. Mark FRM _1,1 , ..., FRM _1,5 , FRM _2,1 , ..., FRM _2,5
Are formed. Each evaluation mark may be a two-dimensional mark, and may be formed, for example, from two line-and-space patterns whose arrangement directions are orthogonal. In addition, the arrangement may be such that the arrangement is substantially evenly distributed over the entire reference plane. In this example, the reference member 10 is used so that the imaging characteristics can be efficiently evaluated without changing the reticle R for actual exposure to a test reticle. In this example, the reference surface of the reference member 10 is used for calibration of a detection system for measuring the shape of the pattern surface of the reticle R described later.

Returning to FIG. 1, the wafer W is held by suction on a wafer holder (not shown).
The sample stage 23 is fixed on the wafer stage WST. The sample stage 23 controls the focus position (position in the Z direction) and the tilt angle of the wafer W, and the wafer stage WST continuously moves the sample stage 23 in the X direction by, for example, a linear motor system, The sample stage 23 is moved stepwise in the Y direction. A moving mirror 25 m fixed to the upper end of the sample table 23 and an external laser interferometer 25
The two-dimensional position of the sample stage 23 (wafer W) is measured, and the wafer stage drive system 26 controls the operation of the wafer stage WST based on the measured values and control information from the main control system 14.

At the time of scanning exposure, the reticle R is + X with respect to the projection optical system PL via the reticle stage RST.
In synchronization with the scanning at the speed VR in the direction (or -X direction), the sample stage 23 (wafer W) moves in the -X direction (or + X direction) via the wafer stage WST at the speed β · VR (β
Are scanned at the projection magnification). When the exposure of one shot area is completed, the next shot area moves to the scanning start position by the stepping of wafer stage WST, and the exposure of each shot area is sequentially performed by the step-and-scan method.

In order to align the surface of the wafer W with the image plane of the projection optical system PL by the auto-focus method at the time of scanning exposure, a plurality of projections of the surface of the wafer W from the projection optical system 27 below the side of the projection optical system PL. A slit image is projected obliquely to the measurement point. These measurement points are also arranged in the slit-shaped exposure area and in the pre-read area on the near side in the scanning direction with respect to this exposure area. In the light receiving optical system 28 symmetrically arranged on the projection optical system 27, the wafer W
The main control system 14 and the wafer stage drive system 26 receive the reflected light from the surface of the main control system 14 and generate a focus signal corresponding to the amount of the lateral shift thereof.
To supply. An oblique incidence type focus position detection system (hereinafter referred to as an “AF sensor 2”) is provided by the projection optical system 27 and the light receiving optical system 28.
7, 28 "). When the focus position of the wafer W changes, the lateral shift amount of those slit images also changes, so that the focus position at the corresponding measurement point can be detected from the focus signals. Of the sample table 23 so that
The position of the direction and the inclination angle are controlled by a servo system.

Further, the projection optical system PL of this embodiment has a built-in mechanism for correcting the imaging characteristics. That is, the projection optical system PL
A lens frame 20 in which a lens 21 is housed is disposed on a lens barrel 18 of the main body via three driving elements 19 such as a piezo element which can be extended and contracted. And the main control system 14
Controls the amount of expansion and contraction of the drive element 19 via the imaging characteristic control unit 15 and finely adjusts the position and the tilt angle of the lens 21 to thereby provide a predetermined distortion (including a magnification error) of the projection optical system PL and the like. In a predetermined range. Although this example shows an example in which only one lens is driven, by driving a plurality of predetermined lenses in the projection optical system PL, a plurality of other imaging characteristics (for example, an image plane) It is desirable to correct curvature, coma, etc.).

Normally, when the exposure is continuously performed, the projection optical system PL is changed due to a change in the atmospheric pressure around the exposure apparatus, a change in the ambient temperature, and accumulation of thermal energy of the exposure light in the projection optical system PL. It is known that the imaging characteristics (distortion, image plane position, etc.) gradually change. Therefore, the main control system 14 of the present embodiment receives detection signals from a sensor (not shown) for measuring the atmospheric pressure or the like and a sensor (not shown) for continuously monitoring the light amount of the light beam branched from the illumination light IL. The main control system 14 predicts the amount of change in the imaging characteristics from the detection signals, and cancels out the amount of change in the imaging characteristics control unit 15 or the drive mechanism (in the Z direction) of the sample table 23 in the Z direction. That is, the imaging characteristic of the projection optical system PL is corrected via the defocus amount correction mechanism). In this example,
By using the imaging characteristic control unit 15 and the sample table 23, errors in the imaging characteristics caused by the deformation of the reticle R are also corrected.

The reference mark plate 22 is fixed near the wafer W on the sample table 23. The surface of the reference mark plate 22 is held at the same height as the surface of the wafer W. In the light-shielding film on the surface of the reference mark plate 22, as shown in FIG. A slit 22x extending in the Y direction and a slit 22y extending in the X direction are formed for measuring imaging characteristics.

FIG. 3C shows the bottom of the reference mark plate 22.
FIG. 3 shows a detection system provided inside the sample stage 23;
In (c), the light for exposure passed through the slit 22x
The bright light IL passes through the condenser lens 29A in the sample stage 23.
The light is received by the photoelectric detector 29B, and the slit shown in FIG.
Similarly, a photoelectric detector is fixed to the bottom of the
The detection signal of the photoelectric detector is a signal of the main control system 14 in FIG.
It is supplied to the processing unit. In this case, the example of FIG.
If the evaluation mark FRM_1,1 X coordinate of the projected image of
The wafer stage WST
By driving the slit of the fiducial mark plate 22
22x for the evaluation mark FRM_1,1 Crosses the projected image of
Thus, the detection signal of the photoelectric detector 29B is
Is sampled in correspondence with the X coordinate of. Then, for example
If the detection signal is binarized with a predetermined threshold,
Of the evaluation mark FRM as the coordinates of the middle point of the_1,1 X
Coordinates can be detected. Similarly, using the slit 22y
With the evaluation mark FRM _1,1 Can be detected.

In order to detect the position (best focus position) of the image plane of the projected image of the evaluation mark FRM _1,1 , the reference mark plate 22 is driven via a driving mechanism in the Z direction in the sample table 23. May be changed by a predetermined step amount, and the contrast of the detection signal obtained when the projected image is scanned by the slit 22x may be detected.
In this case, the focus position of the reference mark plate 22 when the contrast is highest is the best focus position of the projected image. The image plane can also be detected by detecting the best focus positions of the evaluation marks having various image heights.

In addition to the sensor for detecting a projected image through such a slit, a sensor for detecting a projected image via a knife edge, or a CCD-type imaging of the projected image via a relay optical system. A sensor or the like for imaging with an element can also be used. Next, a mechanism for measuring the amount of deformation of the reticle R will be described. In the projection exposure apparatus of FIG. 1, the reticle holder 7 is provided with ribs 8A and 8B for increasing rigidity, and a lens driving mechanism is provided at the upper end of the projection optical system PL. The space between the reticle stage RST) and the projection optical system PL is considerably narrow. Therefore, in the present example, the flatness and the inclination angle of the pattern surface of the reticle R on the bottom side of the reticle stage RST in a region deviated in the scanning direction with respect to the space between the reticle stage RST and the projection optical system PL. A surface shape detection system 30 for detecting the surface shape is provided.

In the projection exposure apparatus of this embodiment, after each reticle is loaded onto the reticle holder 7, the shape of the pattern surface is measured once, and the shape is stored. Therefore, it is not necessary to measure the shape of the reticle during the scanning exposure for irradiating the exposure illumination light IL, and during the scanning exposure, the shape of the stored pattern surface is called according to the X coordinate of the reticle stage RST. Variations in the imaging characteristics can be corrected based on the shape. Thus, the projection optical system P
Since there is no need to measure the shape of the reticle R on the optical axis AX of L, it is not necessary to dispose the surface shape detection system 30 in a narrow space between the reticle stage RST and the projection optical system PL. .

FIG. 2 is a side view of the surface shape detection system 30 of FIG. 1 as viewed in the -X direction (scanning direction). As shown in FIG. Light DL emitted from the light sources 31A to 31C
Respectively illuminates the slit plates 33A to 33C, and the detection light DL passing through the slits formed in the slit plates 33A to 33C is substantially in the same plane as the pattern surface of the reticle R via the objective lenses 34A to 34C, respectively. Project the slit images on the measurement points 41A to 41C arranged in a line in the Y direction. The measurement points 41A to 41C are the reticle R
Are set outside the transfer target area (projection field of view on the reticle R side) of the projection optical system PL with respect to the movement direction of. The detection light DL reflected at the measurement points 41A to 41C is applied to the vibration slit plate 37 via the objective lenses 35A to 35C, respectively.
The slit image is re-imaged on the apertures A to 37C. The vibration slit plates 37A to 37C vibrate in a one-dimensional direction via a vibrator (not shown) by the drive detection unit 39, and the detection light DL that has passed through each opening of the vibration slit plates 37A to 37C is
The light enters the corresponding photoelectric detectors 38A to 38C, and the detection signals of the photoelectric detectors 38A to 38C are supplied to the drive detector 39. Therefore, the surface shape detection system 30 of this example has a configuration in which three oblique incidence type position detection systems are arranged in parallel.

In this case, when the positions in the Z direction of the measurement points 41A to 41C on the pattern surface of the reticle R change, the positions of the slit images re-imaged on the vibrating slit plates 37A to 37C shift laterally in the vibration direction. Therefore, the drive detection unit 39
Then, the detection signals of the photoelectric detectors 38A to 38C are synchronously rectified with, for example, a drive signal of the vibrator, so that
The positions in the Z direction of 1A to 41C are detected with a resolution of, for example, about 100 nm, and the detection results are supplied to the calculation unit 13 in FIG. The X coordinate and the Y coordinate of the reticle stage RST (reticle R) measured by the laser interferometer 11 are also supplied to the calculation unit 13. The positions in the Z direction at the measurement points 41A to 41C indicate the surface shape of the pattern surface in the non-scanning direction when the reticle R is at the X coordinate. And
In FIG. 1, the reticle stage RST is moved in the X direction above the surface shape detection system 30 to detect the Z direction positions of the entire surface of the pattern surface of the reticle R and the entire surface of the reference surface of the reference member 10 in the Z direction. It is configured to be able to be detected by the system 30.

In this embodiment, the positions in the Z direction of the three measurement points on the pattern surface of the reticle R are detected. However, if more detailed positional information is required, the oblique incidence type detection system May be increased to increase the number of the measurement points.
Next, a method of calibrating the measured value of the surface shape detection system 30 of the present example will be described. Surface shape detection system 30
Is used to measure the relative position of the reticle R with respect to the projection optical system PL to evaluate the effect on the imaging characteristics. Therefore, if the measured values are not stable, the imaging characteristics become unstable. However, the surface shape detection system 30 is a sensor that requires extremely high accuracy of about 100 nm, and it is very difficult to maintain the stability at that level for a long period of time, and this leads to an increase in manufacturing cost. Therefore, the surface shape detection system 30 is calibrated by using the reference surface of the reference member 10 for measuring the imaging characteristics provided on the reticle holder 7. Since the image plane can also be detected by the projected image of the evaluation mark on the reference plane, the surface of the wafer W is adjusted to the image plane, and the measured value of the surface shape detection system 30 with respect to the position of the reference plane at that time. By calibrating to a reference value (for example, 0), the pattern surface of reticle R and wafer W
The in-focus state with the surface can be maintained.

More specifically, in FIG. 1, the reticle stage RST is driven to
0 is moved to the illumination area of the illumination light IL for exposure, and the reference member 10 is irradiated with the illumination light IL. At this time, a reticle R for actual exposure may be placed on the reticle holder 7. Then, images of evaluation marks FRM _{1,1 to} FRM _2,5 of reference member 10 shown in FIG. 3A are projected onto wafer stage WST via projection optical system PL. Thus, while changing the focus position of the reference mark plate 22 as described above, predetermined (three or more) images of the evaluation marks are scanned by the slits 22x (see FIG. 3C), and the detection signals are detected. The best focus position of each image is obtained from the contrast, and the optimum image plane (matching plane) is determined from these best focus positions by, for example, the least square method.

When the focus position of the reference mark plate 22 is changed in this manner, the main control system 14 uses the oblique incidence type AF sensors 27 and 28 at measurement points near the images of the plurality of evaluation marks. And the focus signal at the best focus position of the plurality of images is obtained. Further, the main control system 14 obtains, as offsets, focus signals of the respective measurement points of the AF sensors 27 and 28 on the optimum image plane, and supplies these offsets to the wafer stage drive system 26. Instead of supplying the offset of the focus signal, for example, the incident angle of the detection light from the irradiation optical system 27 or the position of the slit image re-imaged in the light receiving optical system 28 is shifted so as to cancel the offset. You may.

Thereafter, when the surface of the wafer W is in the exposure area at the time of exposure, the wafer stage drive system 26 sets the value obtained by subtracting the offset from the focus signal supplied from the AF sensors 27 and 28 to 0. Then, the focus position and the tilt angle of the sample table 23 are controlled. As a result, the surface of the wafer W is accurately fitted to a plane approximating the image plane of the reference plane of the reference member 10 by the projection optical system PL, that is, an optimum image plane (matching plane).

Immediately after the optimum image plane is measured using the reference member 10 (or immediately before the measurement), the reference surface of the reference member 10 is moved to the detection area of the surface shape detection system 30 as shown in FIG. To measure the surface position. In FIG.
The reference member 10 is moved in the + X direction with respect to the projection optical system PL, and the three measurement points 41A to 41C at the center in the X direction of the reference surface (lower surface) of the reference member 10 are respectively shown in FIG. The detection lights DLA to DLC from the light sources 31A to 31C of the surface shape detection system 30 of FIG. 1 are projected, and the Z-direction positions ZA ₀ , ZB ₀ , and ZC at the measurement points 41A to 41C in the arithmetic unit 13 in FIG.
₀ is detected. Next, as shown in FIG. 6, a part of the pattern surface of the reticle R for actual exposure is moved to the detection area of the surface shape detection system 30, and the surface position of the pattern surface is measured. Here, as an example, it is assumed that the shape of the pattern surface of the reticle R is almost uniformly deformed in the scanning direction.
In order to perform the scanning exposure on the first shot area on the wafer, the surface position of the reticle R is measured while the reticle R is stopped at the approach start position (acceleration start position).

In FIG. 6, a reticle R is a projection optical system P
The reticle R has moved to the + X direction side with respect to L, and is located on the + X direction side of the illumination area 44 by the illumination light for exposure. The X coordinate of the reticle stage RST at this time is x _1. Although the shape of the illumination area 44 is set by the field stop in the optical system 4 in FIG. 1, the illumination area 44 has not yet been irradiated with illumination light in the state of FIG. And
When the exposure is started, the reticle R starts running in the −X direction, reaches a predetermined scanning speed and is synchronized with the wafer W, and then the end of the pattern area of the reticle R becomes the illumination area 4.
4, irradiation of illumination light (scanning exposure) is started, and after the pattern area passes through the illumination area 44, the reticle R
Is started, and the reticle R stops. afterwards,
At the time of exposure to the next shot area, the reticle R is scanned in the + X direction with respect to the illumination area 44, and thereafter the reticle R is alternately scanned in the opposite direction.

In this example, at the approach start position of the reticle R shown in FIG. 6 (the X coordinate of the reticle stage RST is x ₁ ), three measurement points 41A to 4A at the end of the pattern surface of the reticle R.
1C, the light sources 31A to 31A of the surface shape detection system 30 in FIG.
The detection lights DLA to DLC from 31C are projected, and the calculation unit 13 in FIG. 1 calculates the position Z in the Z direction at the measurement points 41A to 41C.
A (x ₁ ), ZB (x ₁ ) and ZC (x ₁ ) are detected. Thereafter, the arithmetic unit 13 subtracts the position of the reference surface of the reference member 10 in the Z direction from the position of the pattern surface of the reticle R in the Z direction to obtain the differences ΔZA (x ₁ ), ΔZB (x ₁ ), and ΔZC (x ₁ ). And supplies these differences to the main control system 14 as the position of the pattern surface in the Z direction. In the main control system 14, the supplied positions in the Z direction ΔZA (x ₁ ), ΔZB (x ₁ ), ΔZ
From C (x ₁ ), the amount of change in the position of the image plane of the projection optical system PL with respect to the previously determined mating plane of the wafer W is calculated.

FIG. 4 shows an example of the deformation of the pattern surface of the reticle R. FIG. 4 is a view of the reticle R viewed in the + X direction (scanning direction). In this case, the solid line indicates a state where the pattern surface 40 of the reticle R held on the reticle holder 7 matches the reference surface, and the surface of the wafer W matches the image surface of the pattern surface 40 by the projection optical system PL. And the two-dot chain line indicates the pattern surface 40 in a state where the reticle R is bent by its own weight.
A is shown. The amount of deflection of the pattern surface 40A from the reference surface is determined by the above-described Z-direction positions ΔZA (x ₁ ) and ΔZB.
(X ₁ ) and ΔZC (x ₁ ). Therefore, the main control system 14 controls the projection optical system P for the pattern surface 40A.
The amount of change of the image plane 42A due to L is calculated from the projection magnification β of the projection optical system PL and its position in the Z direction. As a result, the amount of curvature of field and the amount of change ΔZ in the average focus position of the image plane 42A are calculated. Therefore, the main control system 14 first receives the image plane via the imaging characteristic control unit 15 in FIG. Correct the curvature. Here, it is assumed that the curvature of field can be changed to some extent by controlling the driving element 19. At this time, since the average focus position also changes, the main control system 14 calculates the amount of change ΔZ ′ of the remaining focus position, and sends the target value of the focus position on the surface of the wafer W to the wafer stage control system 26. Is changed by −ΔZ ′. Thereby, the curvature of field and the defocus due to the deflection of the pattern surface of the reticle R are corrected, and the wafer W
Is precisely aligned with the actual image plane with respect to the pattern surface of the reticle R.

When the distortion also changes due to the deformation of the pattern surface of the reticle R, the distortion is corrected via the imaging characteristic control unit 15.
Further, when the amount of deformation of the pattern surface of the reticle R greatly differs depending on the position in the scanning direction, the X coordinate of the reticle stage RST changes by a predetermined amount and x _i (i = 2, 3,...)
, The positions ZA (x _i ) and ZB in the Z direction at three measurement points on the pattern surface of the reticle R via the surface shape detection system 30.
It is desirable to detect (x _i ) and ZC (x _i ) and obtain the difference from the position of the reference plane. Then, an average plane of the entire pattern plane is determined, and the image plane of the average plane may be used as the mating plane of the wafer W. However, the imaging characteristic control unit 15 performs the operation according to the X coordinate of the reticle stage RST. The correction amount of the field curvature and the correction amount of the target value of the focus position by the sample table 23 may be changed.

According to this embodiment as described above, the projection optical system P
After determining the image plane of the reference plane of the reference member 10 via L,
In a very short time, the amount of displacement of the pattern surface of the reticle R with respect to the reference surface in the Z direction is measured. Therefore, even when the measured value of the surface shape detection system 30 gradually changes with time, the position of the pattern surface of the reticle R can be measured stably and with high accuracy without being affected by the change with time. Therefore, an inexpensive sensor can be used as the surface shape detection system 30.

The measurement of the shape of the pattern surface of the reticle R by the surface shape detection system 30 may be performed at the time of deceleration at the end of the scanning exposure of the reticle R or at the time of stopping after the end of the scanning exposure. Next, an example of the entire operation in the case of performing the exposure while correcting the imaging characteristics by measuring the amount of deformation of the pattern surface of the reticle R with the projection exposure apparatus of the present embodiment will be described with reference to the flowchart of FIG. explain. First, in step 300 in FIG. 7, the operator determines whether or not to perform calibration of the imaging characteristics of the projection optical system PL, that is, whether to perform so-called lens calibration. Lens calibration is performed as necessary due to the stability of the imaging characteristics of the projection optical system PL. If the lens calibration is to be performed, the process proceeds to step 301, where the reference member 10 is controlled under the control of the main control system 14 as described with reference to FIG.
After measuring the imaging characteristics (distortion and image plane) of the projection optical system PL using the evaluation marks of
In step 02, the main control system 14 corrects the imaging characteristics so as to be optimal with respect to the reference surface of the reference member 10.

Next, after the lens calibration is completed, or when the lens calibration is not executed in step 300, the process directly proceeds to step 303,
A reticle (referred to as a reticle R) that is actually used for exposing a circuit pattern is mounted on a reticle holder 7 shown in FIG. Next, proceeding to step 304, the center of the reference member 10 in the X direction is set to the surface shape detection system 3 as shown in FIG.
Then, the surface shape detection system 30 measures the position in the Z direction (position in the optical axis AX direction) of the reference surface of the reference member 10 and moves the measured value to the surface shape detection. The reference value of the system 30 is stored in the calculation unit 13.
Since the width of the reference member 10 in the scanning direction is as narrow as the illumination area, the error due to the position in the scanning direction is considered to be negligible. However, in order to increase the position accuracy of the reference plane,
By changing the position of the reference member 10 in the scanning direction, the surface shape detection system 3
The position measurement in the Z direction may be performed a plurality of times via 0, and the average value of the measurement values may be stored as the reference value of the surface shape detection system 30.

Next, the process proceeds to step 305, where the shape of the pattern surface of the reticle R for actual exposure is measured. Here, assuming that the entire shape of the pattern surface of the reticle R is measured in order to increase the measurement accuracy, as shown in FIG. 1, a reticle is transmitted via a reticle stage RST based on a measurement value of a laser interferometer 11. While moving R in the X direction, the position in the Z direction is measured at three measurement points in the non-scanning direction (Y direction) on the pattern surface of the reticle R via the surface shape detection system 30 at predetermined intervals in the X direction. I do. Then, the calculation unit 13 in FIG. 1 supplies a difference obtained by subtracting the reference value stored in step 304 from each measurement value to the main control system 14 as a position in the Z direction of the pattern surface of the reticle R.
Thus, the three-dimensional surface shape of the pattern surface is measured.

Next, the routine proceeds to step 306, where the main control system 1
As described with reference to FIG. 4, reference numeral 4 calculates the amount of change in the imaging characteristic caused by the deformation of the pattern surface of the reticle R with respect to the reference surface, and calculates the correction amount for the change. In the scanning exposure method, since the exposure area has a slit shape, finer correction can be performed by changing the correction amount according to the position (X coordinate) of the reticle stage RST in the scanning direction. For example, when the reticle R is being deformed such that it is twisted, it is desirable to determine the correction amount according to the X coordinate of the reticle stage RST, such as changing the inclination angle of the surface of the wafer W according to the amount of twist. Of course, the X coordinate of the reticle stage RST in this case takes into account the difference between the measurement position (the position of the measurement point of the surface shape detection system 30) and the exposure position (the position of the optical axis AX).

Then, the process proceeds to a step 307, wherein an exposure operation for printing a circuit pattern of the semiconductor element on the wafer is started. That is, for example, one lot of wafers
3 and scan exposure is performed on the shot area of each wafer. Immediately before this scanning exposure, the main control system 14
The correction amount for canceling the fluctuation amount of the imaging characteristic due to the atmospheric pressure fluctuation and the absorption of the illumination light of the projection optical system PL is added to the correction amount of the imaging characteristic obtained in step 306 to obtain a comprehensive correction. Find the quantity. Then, based on the total correction amount, the main control system 14 drives the imaging characteristic control unit 15 and the sample table 23 in accordance with the X coordinate of the reticle stage RST to correct the imaging characteristics. Scan exposure is performed on the shot area.

For example, if exposure is to be continued after the exposure operation for one lot of wafers is completed, the operation proceeds from step 308 to step 309 to determine whether or not to replace the reticle. In the case where exposure is continuously performed using the same reticle, since the shape of the reticle does not need to be measured again, the process returns to step 307, and the exposure operation is performed similarly.
If the reticle is to be replaced in step 309, the process returns to step 300. At this time, if not much time has passed since the previous calibration (lens calibration) of the imaging characteristics of the projection optical system PL, the change factor is only the surface shape of the reticle. What is necessary is just to start from loading of a reticle. Furthermore,
If the elapsed time is sufficiently shorter than the time when the measurement value of the surface shape detection system 30 changes with time, step 30
The measurement of the position of the reference plane by the surface shape detection system 30 of 4 can also be omitted.

As described above, according to the present embodiment, the shape of the pattern surface of the reticle R is measured using the surface shape detection system 30 with reference to the position of the reference surface of the reference member 10, and the imaging characteristics are determined in accordance with the measurement result. Is corrected, the shape of the pattern surface can be measured with high accuracy even when the measurement value of the surface shape detection system 30 changes with time, and as a result, the imaging characteristics are maintained in a desired state with high accuracy. Exposure can be performed.

In the case where a reticle which has been used before as a reticle and in which measurement data of the surface shape is stored is used and the accuracy required for the exposure process is not so high, It is also possible to omit the step of measuring the surface shape of the reticle in step 305 to increase the throughput. However, even with the same reticle, the position of the reticle when mounted on the reticle holder 7 is slightly changed due to the positioning accuracy of the reticle loader, and foreign matter is caught between the reticle holder 7 and the reticle. Therefore, it is desirable to measure every time. Further, when it is determined that the residual error is large even if the imaging characteristics are corrected due to a foreign matter or a manufacturing error of the reticle, or when the correction amount is insufficient, a warning is issued to the operator and the exposure is stopped. It is desirable to do.

In this example, the reticle R was measured only once when the reticle R was mounted on the apparatus, on the assumption that the shape of the reticle R was kept constant after being mounted on the reticle holder 7. However, when the shape of the reticle R changes with the passage of time, or when the reticle R needs to be re-measured due to absorption of illumination light and expansion, etc., step 30 is performed again.
It is desirable to return to 4 or 305 and measure the surface shape again. Further, in the description so far, only the shape change of the reticle R is considered, but the posture of the reticle stage RST may change depending on the position of the reticle stage RST, and the posture of the reticle R may also change. In this case, since the problem is unique to the apparatus irrespective of the reticle R, the amount of change in the attitude of the reticle R is measured in advance for each projection exposure apparatus, and the correction value of the amount of change is determined according to the coordinates of the reticle stage RST. What is necessary is just to incorporate it into the correction value of the image formation characteristic and add it to the correction component due to the shape of the reticle. Further, when the attitude of the reticle stage RST changes depending on the scanning speed and the acceleration of the reticle stage RST, it is desirable that the measurement be performed in advance and corrected.

In the above embodiment, the surface shape detection system 30 is arranged on the bottom surface side of the reticle stage RST. However, the surface shape detection system 30 may be arranged obliquely above the reticle R. However, since the pattern surface of the reticle R is the lower surface, it is more accurate to dispose the surface shape detection system 30 on the bottom surface side of the reticle stage RST in a state where the reflected light from the upper surface of the reticle R is less affected. Can be detected.

Further, in the above embodiment, the projection optical system P
The imaging characteristics were corrected by driving the lens in L and controlling the attitude of the sample stage 23 on the wafer stage WST. However, an attitude control mechanism was provided on the reticle stage RST side, and the reticle was detected by the surface shape detection system 30. The position of the reticle may be controlled while performing the feedback of the position of the pattern surface. In this case, it is desirable to detect the positions of the measurement points of the surface shape detection system 30 not at one line but at at least three points not on one straight line so that a plane can be determined.

It should be noted that the present invention is not limited to the above-described embodiment, and it is needless to say that various configurations can be adopted without departing from the gist of the present invention.

[0056]

According to the scanning exposure method of the present invention, the shape of the pattern surface of the mask is measured before the scanning exposure, and the image forming characteristic is corrected based on the measurement result of the pattern surface during the scanning exposure. Even when the distance between the optical system and the stage on the mask side is very small, for example, there is an advantage that a sensor for measuring the shape of the pattern surface of the mask can be easily arranged at a position deviated in the scanning direction with respect to the projection optical system. is there. Therefore, by measuring the amount of deformation of the mask, it becomes possible to calculate and correct the amount of change in the imaging characteristics caused by the deformation.

The shape measurement of the pattern surface of the mask is performed by
When the mask is at the approach start position or when the mask is at the approach position, the sensor for measuring the surface shape can be reliably disposed at a position deviated in the scanning direction, and the mask for measuring the surface shape can be used. There is no need to extend the travel stroke. Next, according to the scanning exposure apparatus of the present invention, the scanning exposure method of the present invention can be performed. At this time, when a reference surface having substantially the same height as the pattern surface of the mask is formed on the mask stage for moving the mask, and the calibration of the shape measurement system is performed using the reference surface, the Since the stability of the shape measurement system requires only a short time from the measurement of the reference plane to the end of the measurement of the mask, the shape of the mask can be changed even if the measurement value of the shape measurement changes with time (poor stability). Can be measured with high accuracy, and the amount of change in the imaging characteristics caused by the change in the shape can be corrected.
Therefore, the configuration of the shape measurement system can be simplified.

[Brief description of the drawings]

FIG. 1 is a partially cut-away configuration diagram showing a projection exposure apparatus used in an example of an embodiment of the present invention.

FIG. 2 is a side view showing a configuration of a surface shape detection system 30 in FIG.

3A is a plan view showing the arrangement of evaluation marks on the reference member 10 in FIG. 1, FIG. 3B is a plan view showing the reference mark plate 22 in FIG. 1, and FIG. FIG. 3 is a cross-sectional view showing a detection system at the bottom of FIG.

FIG. 4 is an explanatory diagram showing a change in an image plane due to deformation of a pattern surface of a reticle R.

FIG. 5 is a perspective view showing a simplified arrangement when measuring the position of the reference surface of the reference member 10;

FIG. 6 is a perspective view showing a simplified arrangement when measuring the shape of the pattern surface of the reticle.

FIG. 7 is a flowchart illustrating an example of an overall operation in the case where exposure is performed while correcting for changes in imaging characteristics due to the surface shape of the reticle in the embodiment of the present invention.

[Explanation of symbols]

 Reference Signs List 3 turret plate R reticle PL projection optical system W wafer 7 reticle holder 10 reference member RST reticle stage 13 calculation unit 14 main control system 15 imaging characteristic control unit 19 drive element 22 reference mark plate 23 sample table WST wafer stage 30 surface shape detection System 41A, 41B, 41C Measurement points by surface shape detection system

Claims (5)

[Claims] 1. A scanning exposure method for transferring a pattern of the mask onto the substrate via a projection optical system by synchronously moving the mask and the substrate, wherein the substrate is scanned with a pattern image of the mask. Prior to exposure, the shape of the pattern surface of the mask is measured, and at least one of the imaging characteristic of the projection optical system and the position of the substrate is corrected based on the measurement result of the pattern surface during scanning exposure. A scanning type exposure method characterized by the above-mentioned. 2. A method for measuring a shape of a pattern surface of the mask, comprising:
2. The scanning exposure method according to claim 1, wherein the method is performed when the mask is at the approach start position or when the mask is at the approach position. 3. A scanning exposure apparatus for transferring a pattern of the mask onto the substrate via a projection optical system by synchronously moving the mask and the substrate. And a shape measuring system for measuring the shape of the pattern surface of the mask, and correcting at least one of an imaging characteristic of the projection optical system and a position of the substrate based on a measurement result of the shape measuring system. A scanning type exposure apparatus, comprising: 4. A method in which a reference surface having substantially the same height as a pattern surface of the mask is formed on a mask stage for moving the mask, and the shape measurement system is calibrated using the reference surface. The scanning exposure apparatus according to claim 3, wherein: 5. The shape measurement of the pattern surface of the mask,
5. The scanning exposure apparatus according to claim 3, wherein the scanning is performed when the mask is at the approach start position or when the mask is at the approach position.