JP2006245145A - Optical characteristic measuring method and apparatus, and exposure method and apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED To obtain an even function aberration of a projection optical system in a short time.
When projecting an image of a predetermined measurement pattern via a projection optical system PL, a 0th-order light L (0), 1st-order diffracted light L (+1), L (−) using a variable aperture stop AS. The diffracted light other than 1) is shielded. The projection image is received through the slit 122, and first the peak position of the light quantity is obtained in the lateral direction. Then, after positioning the slit 122 at the peak position, the slit 122 is scanned in the Z direction to detect the change in the light quantity. measure. A predetermined even function aberration of the projection optical system PL is obtained from the position where the light quantity reaches a peak in the Z direction.
[Selection] Figure 3

Description

  The present invention relates to an optical characteristic measurement technique for measuring an optical characteristic of a test optical system, and an exposure technique using the optical characteristic measurement technique, for example, a lithography process for manufacturing a device such as a semiconductor element and a liquid crystal display element. Among them, it is suitable for use in measuring optical characteristics of a projection optical system of a projection exposure apparatus used for transferring a mask pattern onto a substrate.

  Conventionally, in a lithography process for manufacturing a semiconductor element or the like, a reticle (or photomask or the like) pattern as a mask is coated with a wafer (or glass plate) coated with a photosensitive material as a substrate via a projection optical system. Etc.) A batch exposure type projection exposure apparatus (stepper or the like) and a scanning exposure type projection exposure apparatus (scanning stepper or the like) to be transferred thereon are used. As the integration degree and fineness of semiconductor elements and the like are further improved, the accuracy of imaging characteristics such as various aberrations required for the projection optical system of the projection exposure apparatus is also increasing.

  In addition, the imaging characteristics of the projection optical system gradually vary due to changes in environmental conditions such as the thermal energy of the exposure light accumulated in the projection optical system and the atmospheric pressure and temperature around the projection optical system by continuing the exposure. . Therefore, in the projection exposure apparatus, the distortion is controlled by controlling the posture of some lens elements constituting the projection optical system or by controlling the atmospheric pressure in a space (lens chamber) between predetermined lens elements. An imaging characteristic control mechanism that can control predetermined imaging characteristics such as aberration and spherical aberration to a predetermined state is provided. In order to control the imaging characteristics of the projection optical system using this imaging characteristics control mechanism, the imaging characteristics are measured with high accuracy while the projection optical system is mounted on the projection exposure apparatus (on-body). There is a need to. Further, for example, it is necessary to measure the imaging characteristics of the projection optical system with high accuracy on-body during assembly adjustment or maintenance of the projection exposure apparatus.

  As a conventional method for measuring imaging characteristics, a test reticle on which a predetermined pattern is formed is used to expose a wafer with a projected image of the pattern and measure the resist image obtained by developing the wafer. A method of calculating image formation characteristics based on this (hereinafter referred to as “baking method”) has been mainly used. However, this baking method requires a long time for measurement.

Therefore, without actually exposing the wafer, the aerial image of the measurement mark of the test reticle illuminated by the exposure light is projected via the projection optical system, and the aerial image (projected image) is measured. Based on this, a method of calculating the imaging characteristics of the projection optical system (hereinafter referred to as “aerial image measurement method”) has been proposed (see, for example, Patent Document 1). As another example of this aerial image measurement method, a plurality of diffraction gratings are illuminated, and the image intensity of each diffraction grating obtained through the projection optical system is measured at a plurality of focus positions of the projection optical system, A method for obtaining the wavefront aberration of the projection optical system based on the result has also been proposed (see, for example, Patent Document 2).
JP-A-10-170399 JP 2001-57337 A

As described above, when measuring the aberration of the projection optical system on-body using the aerial image measurement method, conventionally, for example, the slit is scanned in a direction perpendicular to the optical axis of the projection optical system at one or a plurality of focus positions. Therefore, it is necessary to measure the light intensity distribution of the aerial image of the measurement mark and to perform Fourier analysis on the measurement result. Therefore, there is a problem that it takes time for measurement.
In general, when an aberration function indicating wavefront aberration on the exit pupil (or pupil plane) of the projection optical system is W (ρ, θ), the aberration function W (ρ, θ) is expressed in a polar coordinate format, and ρ is It is a normalized position (radial radius) in the radial direction of the exit pupil (or pupil plane) of the projection optical system, and θ is an angle. The aberration function W (ρ, θ) is series-expanded by using a completely orthogonal system polynomial in which the radial ρ and the angle θ are separated, for example, Zernike's Polynomial represented by the following equation: Is possible.

Here, Z _i is a coefficient representing the magnitude of the aberration represented by the i-th order Zernike polynomial fi (ρ, θ) among the various aberrations of the projection optical system. Hereinafter, for convenience of explanation, the coefficient Z _i (i is an integer of 1 or more) is also regarded as an aberration (or aberration amount) represented by an i-th order Zernike polynomial. At this time, an aberration expressed by a Zernike polynomial in which the radial function becomes an odd function is called an odd function aberration, and an aberration expressed by a Zernike polynomial in which the radial function becomes an even function is called an even function aberration.

  In general, in a projection optical system that projects an aerial image of a pattern on the object plane onto the image plane, if there is an odd function aberration in the projection optical system, the diffracted light of each order from the pattern is combined. Each image position is shifted laterally. Further, when there is even function aberration in the projection optical system, the imaging positions of the diffracted lights of the respective orders are shifted in the optical axis direction of the projection optical system. In other words, for example, when an image of a measurement pattern is formed via a projection optical system, the image is shifted in the horizontal direction due to an odd function aberration (lateral aberration) of the projection optical system, and the even function of the projection optical system Aberration (longitudinal aberration) reduces the contrast of the image.

Thus, even function aberration is a reduction in contrast. Therefore, when measuring even function aberration using a conventional aerial image measurement method, for example, at many focus positions, the slit is perpendicular to the optical axis of the projection optical system. It is necessary to repeat the scanning operation in the direction, and there is a problem that the measurement time is particularly long.
In view of such a point, the present invention has a first object to provide an optical characteristic measurement technique capable of measuring optical characteristics such as aberration of a test optical system in a short time by an aerial image measurement method.

It is a second object of the present invention to provide an optical characteristic measurement technique that can measure even function aberration of a test optical system in a short time by an aerial image measurement method.
The present invention also provides an exposure technique that can measure the optical characteristics of the projection optical system in a short time by the aerial image measurement method in a state where the projection optical system is mounted on the projection exposure apparatus (exposure apparatus). The purpose of 3.

  The aberration measurement method according to the present invention is an optical property measurement method for measuring optical properties of a test optical system (PL), except for diffracted light of three orders from each of a plurality of marks (2A to 2E) having different pitches. A first step of controlling the numerical aperture of the test optical system to shield the diffracted light of the light, and the light intensity distribution of the image formed by the diffracted light of the three orders passing through the test optical system And a third step for obtaining optical characteristics of the optical system to be measured based on the light intensity distribution measured in the second step.

According to the present invention, by measuring an image of three orders of diffracted light (three-beam interference) from a mark to be measured, the light intensity distribution of the image becomes the light intensity distribution of the image of the order of diffracted light. Become. Therefore, the optical characteristics of the optical system to be measured can be measured in a short time without performing complex Fourier analysis using the aerial image measurement method.
In the present invention, as an example, the three orders of diffracted light are 0th order diffracted light and ± 1st order diffracted light. As a result, the light intensity distribution of the obtained image becomes a fundamental wave component, so that the measurement time can be shortened most.

Further, the second step includes a step of detecting a specific position in a direction perpendicular to the optical axis of the optical system to be detected in the light intensity distribution of the image by the optical system to be detected for each of the plurality of marks, And detecting a change in light intensity of the image by the optical system under test in a direction parallel to the optical axis.
In the present invention, since the three-beam interference is used, the light intensity distribution of the aerial image becomes a frequency component of a predetermined order substantially as it is. Therefore, once a specific position (for example, the peak position of the light intensity distribution) is detected in a direction perpendicular to the optical axis of the optical system to be detected, the optical axis is then simply moved in parallel to the optical axis. It is possible to measure the change in light intensity at the specific position in the direction parallel to. Further, the optical characteristic in which the light intensity changes in a direction parallel to the optical axis is particularly an even function aberration. Therefore, especially the measurement time of even function aberration can be greatly shortened.

Further, as an example, the optical characteristic of the test optical system obtained in the third step is the first aberration of the test optical system, and the second step is the first aberration of the test optical system. A step of measuring a light intensity distribution of an image of the plurality of marks by the optical system to be measured while changing a second aberration different from the aberration. Thereby, a predetermined even function aberration can be easily measured.
Further, as an example, the change in the second aberration of the optical system to be detected is caused by the displacement of a predetermined optical member (13 ₁ , 13 ₂ ) in the optical system to be detected or the defocus of the measurement position of the image. Given.

  As an example, the plurality of marks having different pitches are included in marks having pitches of six or more types arranged along two different directions. As a result, when the optical characteristics of the measurement target are represented by wavefront aberration and the wavefront aberration is represented by a Zernike polynomial, 0θ, 1θ, and cos2θ components of the wavefront aberration represented by the Zernike polynomial up to the 16th order can be measured. It is.

As another example, the plurality of marks having different pitches are included in marks of six or more types of pitches arranged along six different directions. As a result, wavefront aberrations represented by Zernike polynomials up to the 37th order can be measured.
An exposure method according to the present invention is an exposure method in which a second object (W) is exposed via an exposure beam via a first object (R) and a projection optical system (PL). It has a measurement process for measuring the optical characteristics of the projection optical system, and a correction process for correcting the optical characteristics of the projection optical system based on the measurement results of the measurement process. According to the present invention, the optical characteristics of the projection optical system can be corrected in a short time. As a result, it is possible to transfer the pattern of the first object onto the second object with high accuracy using the projection optical system whose optical characteristics are adjusted to an appropriate state.

  Next, an optical characteristic measuring apparatus according to the present invention is an optical characteristic measuring apparatus for measuring optical characteristics of a test optical system (PL), and a substrate (RFM) on which a plurality of marks (2A to 2E) having different pitches are formed. ), The illumination system (14, 12) for illuminating the substrate, and the optical system under test for shielding diffracted light other than the three orders of diffracted light from each of the plurality of marks having different pitches A space for detecting each of the images formed by the optical system to be detected using the variable aperture stop (AS) for sequentially controlling the numerical aperture of the aperture and the diffracted light of the three orders of the marks having different pitches. An image detection system (59) and an arithmetic unit (50) for obtaining optical characteristics of the optical system to be detected based on the detection result of the aerial image detection system are provided. According to the present invention, the optical characteristics of the optical system to be measured can be measured in a short time by the aerial image measurement method.

  In the present invention, as an example, the aerial image detection system has slit-like opening patterns (9A to 9F) for detecting images by the test optical system, and the slit-like opening pattern is used as the optical axis of the test optical system. A stage (WST) that moves in a direction parallel to the optical system, and the computing device is a spatial image detection system obtained when the slit-shaped aperture pattern is moved in a direction parallel to the optical axis of the optical system to be tested. Based on the detection result, the optical characteristics of the optical system to be detected are obtained. This makes it possible to measure the even function aberration of the test optical system in a short time.

  The exposure apparatus according to the present invention illuminates the first object (R) with an exposure beam, and exposes the second object (W) with the exposure beam via the first object and the projection optical system (PL). In the apparatus, the optical characteristic measuring apparatus of the present invention, a first stage (RST) for holding a substrate on which a plurality of marks having different pitches are formed together with the first object or instead of the first object, and A second stage (WST) that holds the light receiving part of the aerial image detection system together with the second object, and measures the optical characteristics of the projection optical system using the optical characteristic measuring device. Thereby, the optical characteristics of the projection optical system can be measured on-body in a short time using the aerial image measurement method.

According to the present invention, by using three-beam interference, the optical characteristics of the projection optical system can be measured in a short time by the aerial image measurement method.
Further, in the present invention, when there is a step of detecting a change in the light intensity of the image by the test optical system in a direction parallel to the optical axis of the test optical system at a specific position, the even function aberration is particularly reduced for a short time. Can be measured.

  Further, according to the exposure method and apparatus of the present invention, the optical characteristics of the projection optical system can be measured in a short time with the projection optical system mounted on the exposure apparatus.

Hereinafter, an example of a preferred embodiment of the present invention will be described with reference to the drawings.
FIG. 1 shows a schematic configuration of a projection exposure apparatus 10 of this example. The projection exposure apparatus 10 corresponding to the exposure apparatus of the present invention is a step-and-scan type scanning exposure type projection exposure apparatus, that is, a scanning stepper.
In FIG. 1, a projection exposure apparatus 10 includes a light source 14 (exposure light source) that generates a laser beam LB, an illumination optical system 12 (illumination unit), a reticle stage RST that moves while holding a reticle R as a mask, and a projection optical system. PL, a wafer stage WST that holds and moves a wafer W as a substrate (or photoconductor), a control system that controls these, and the like are provided. The parts other than the light source 14 and the control system are actually housed in an environmental chamber (not shown) in which environmental conditions such as the internal temperature are controlled with high accuracy and are maintained constant.

In this example, an ArF excimer laser light source (oscillation wavelength 193 nm) is used as the light source 14. The light source 14 is controlled to turn on / off the laser emission, the center wavelength, the spectral half-value width, the repetition frequency, and the like by a main controller 50 that is a computer that controls the overall operation of the apparatus. As an exposure light source, KrF excimer laser (wavelength 248 nm), F ₂ laser (wavelength 157 nm), harmonic generator of YAG laser, harmonic generator of solid-state laser (semiconductor laser, etc.), or mercury lamp (i-line etc.) ) Etc. can also be used.

  The illumination optical system 12 includes a beam shaping optical system 18 that shapes the cross-sectional shape of the laser beam LB supplied from the light source 14, a fly-eye lens 22 as an optical integrator (a homogenizer or a homogenizer), an illumination system aperture stop plate 24, A relay optical system including a first relay lens 28A and a second relay lens 28B, a fixed reticle blind 30A, a movable reticle blind 30B, a mirror M, a condenser lens 32, and the like are provided. As the optical integrator, an internal reflection type integrator (for example, a rod integrator) or a diffractive optical element may be used. A large number of microlenses constituting the fly-eye lens 22 each focus the laser beam LB from the beam shaping optical system 18 on the focal plane on the emission side, and a secondary light source (surface light source) is formed on the focal plane. . Hereinafter, the laser beam LB emitted from the secondary light source formed by the fly-eye lens 22 is referred to as “illumination light IL” as an exposure beam (exposure light).

The light source 14 and the illumination optical system 12 are also used as an illumination system for a later-described aerial image measurement. In the illumination optical system 12, an illumination system aperture stop plate 24 made of a disk-like member is disposed in the vicinity of the exit-side focal plane of the fly-eye lens 22. The illumination system aperture stop plate 24 is provided with an aperture stop (small aperture) made up of, for example, a normal circular aperture, and an aperture stop (small size) for reducing the σ value that is a coherence factor made up of a small circular aperture at substantially equal angular intervals. σ stop), an annular aperture stop for annular illumination (annular aperture stop), and a modified aperture stop (for example, for dipole illumination or quadrupole illumination) in which a plurality of apertures are decentered for the modified light source method. ) And the like are arranged. The illumination system aperture stop plate 24 is rotated by a drive device 40 such as a motor controlled by the main control device 50, and any one of the aperture stops is on the optical path of the illumination light IL by this rotation operation. Selectively set.

A beam splitter 26 having a low reflectance and a high transmittance is disposed on the optical path of the illumination light IL emitted from the illumination system aperture stop plate 24, and further, relays are provided on the rear optical path with reticle blinds 30A and 30B interposed therebetween. Optical systems (28A, 28B) are arranged.
The fixed reticle blind 30A is disposed on a surface slightly defocused from the conjugate plane with respect to the pattern surface of the reticle R, and the fixed reticle blind 30A is formed with a rectangular opening that defines the illumination area IAR on the reticle R. Has been. Also, in the vicinity of the fixed reticle blind 30A, there is a movable reticle blind 30B having an opening whose position and width are optically corresponding to the scanning direction at the time of scanning exposure and the non-scanning direction orthogonal thereto. Has been placed. At the start and end of scanning exposure, the illumination area IAR defined by the fixed reticle blind 30A is further restricted by the movable reticle blind 30B according to instructions from the main controller 50, so that unnecessary portions (reticles) The exposure of the portion other than the portion to be transferred such as the circuit pattern on R is prevented. In this example, the movable reticle blind 30B is also used for setting an illumination area when performing aerial image measurement, which will be described later, as necessary.

On the other hand, an integrator sensor 46 including a condenser lens 44 and a light receiving element is disposed on the optical path of the illumination light IL reflected by the beam splitter 26 in the illumination optical system 12.
The laser beam LB emitted from the light source 14 at the time of exposure becomes illumination light IL in the illumination optical system 12, and the illumination light IL is bent vertically downward by the mirror M and then passes through the condenser lens 32. A slit-like illumination area IAR elongated in the non-scanning direction of the pattern surface (lower surface) of the reticle R is illuminated with a uniform illuminance distribution.

  On the other hand, a part of the illumination light IL reflected by the beam splitter 26 is received by the integrator sensor 46 via the condenser lens 44, and the photoelectric conversion signal of the integrator sensor 46 passes through the peak hold circuit and the A / D converter. It is supplied to the main controller 50 through the signal processing device 80 having the same. In this example, the measurement value of the integrator sensor 46 is used not only for exposure amount control on the wafer W but also for calculation of the irradiation amount for the projection optical system PL. This irradiation amount is not only the wafer reflectivity (which can also be obtained based on the output of the integrator sensor 46 and the output of the reflectivity monitor (not shown)) and the imaging characteristics due to the absorption of illumination light of the projection optical system PL. It is also used to calculate the amount of change.

In this example, the main controller 50 measures the irradiation amount of the illumination light IL at predetermined time intervals based on the output of the integrator sensor 46, and the measurement result is stored in the memory 51 (storage device) as an irradiation history. It has come to be remembered.
Under the illumination light IL, an image formed by the telecentric projection optical system PL on both sides (or one side on the wafer side) of the pattern in the illumination area IAR of the reticle R is coated with a photoresist as a photosensitive material. And projected onto an exposure area IA on one shot area of the wafer W. The exposure area IA is conjugate with the illumination area IAR, and the projection optical system PL converts the pattern image (first surface) of the reticle R (first object) to the upper surface (second object) of the wafer W (second object). Surface). The projection magnification of the projection optical system PL is, for example, a reduction magnification such as 1/4 or 1/5. In the following description, it is assumed that the projection magnification of the projection optical system PL is 1/4. Although the projection optical system PL of this example is a refraction system, a catadioptric system or the like can also be used as the projection optical system PL. As shown in FIG. 3, a variable aperture stop AS for controlling the numerical aperture NA of the projection optical system PL is disposed in the vicinity of the pupil plane PP of the projection optical system PL.

  Hereinafter, the Z-axis is taken in a direction parallel to the optical axis AX of the projection optical system PL, the X-axis is taken in a direction perpendicular to the paper surface of FIG. 1 within a plane perpendicular to the Z-axis, and the direction parallel to the paper surface of FIG. A description will be given taking the Y axis. In this example, the scanning direction of the reticle R and the wafer W during scanning exposure is a direction parallel to the Y axis (Y direction), and the illumination area IAR on the reticle R and the exposure area IA on the wafer W are not in each case. This is an elongated region in the scanning direction (X direction).

The projection optical system PL of this example is provided with an imaging characteristic control mechanism for controlling (correcting) the predetermined imaging characteristics.
2 is a cross-sectional view showing a part of the imaging characteristic control mechanism of the projection optical system PL in FIG. 1. In FIG. 2, for convenience of explanation, the optical axis is configured so as to constitute the projection optical system PL. Of the many lens elements arranged along AX, only eight lens elements 13 ₁ , 13 ₂ ,..., 13 ₈ are shown. In this case, some of the lens elements 13 ₁ , 13 ₂ ,..., 13 ₈ , for example, the lens elements 13 ₁ , 13 ₂ , are respectively arranged in the direction of the optical axis AX by a plurality of drive elements (for example, piezoelectric elements) 20 It is configured so that it can be finely driven in an inclination direction with respect to the XY plane. A clean gas such as nitrogen is supplied between the lens elements via a pressure adjusting mechanism 41 from a gas supply mechanism (not shown).

  In this example, the drive voltage (drive amount of the drive element) applied to each drive element 20 is controlled by the imaging characteristic correction controller 78 in accordance with a command from the main controller 50 in FIG. As described above, the image formation characteristic control mechanism is configured including the drive element 20 and the image formation characteristic correction controller 78. This corrects the imaging characteristics of the projection optical system PL, such as field curvature, distortion, magnification, coma aberration, astigmatism, spherical aberration, and the like. The number of movable lens elements may be arbitrary. However, in this case, since the number of movable lens elements corresponds to the types that can correct the imaging characteristics of the projection optical system PL, excluding the focus, the movable lens elements can be adjusted according to the types of imaging characteristics that need to be corrected. The number should be determined.

  Returning to FIG. 1, the reticle R is fixed on the reticle stage RST by, for example, vacuum suction (or electrostatic suction). Reticle stage RST is two-dimensionally (X direction, Y direction, and rotational direction (rotation angle θz) around the Z axis) in the XY plane on reticle base RBS by reticle stage drive system 56R including a linear motor or the like. ) It can be driven minutely and can move on the reticle base RBS at a scanning speed designated in the Y direction.

  A movable mirror 52R that reflects a laser beam from a laser interferometer (hereinafter referred to as “reticle interferometer”) 54R is fixed on the reticle stage RST. The position of the reticle stage RST in the XY plane is the reticle stage RST. For example, the interferometer 54R always detects with a resolution of about 0.1 to 1 nm. In other words, the moving mirror 52R is actually composed of two Y-axis moving mirrors for measuring the position in the Y direction at two locations, and the X-axis moving mirror, and the laser interferometer 54R also corresponds thereto. And a three-axis laser interferometer.

  Position information of reticle stage RST from reticle interferometer 54R is sent to stage controller 70 and main controller 50 via this. The stage controller 70 controls the movement of the reticle stage RST via the reticle stage drive system 56R according to an instruction from the main controller 50. Alternatively, the end surface of reticle stage RST may be mirror-finished to form the reflection surface of movable mirror 52R described above.

  In addition, a reticle fiducial mark plate (hereinafter referred to as a “reticle mark plate”) as a mark forming member in which an aerial image measurement reference mark (measurement pattern) is formed in the vicinity of the end in the −Y direction of the reticle stage RST. The RFM is abbreviated to be aligned with the reticle R. This reticle mark plate RFM (described later in detail) is made of the same glass material as that of reticle R, for example, synthetic quartz, fluorite, lithium fluoride, and other fluoride crystals, and is fixed to reticle stage RST. . Reticle stage RST has a movement stroke in the Y direction such that the entire surface of reticle R and the entire surface of reticle mark plate RFM can cross at least optical axis AX of projection optical system PL. In addition, openings for passing illumination light IL are formed in reticle stage RST below reticle R and reticle mark plate RFM, respectively. Also, a rectangular opening at least larger than the illumination area IAR, which is a passage for the illumination light IL, is formed in a portion almost directly above the projection optical system PL of the reticle base RBS (portion centered on the optical axis AX). ing.

  Also, above the reticle R, a mark on the reticle R or on the reticle mark plate RFM and a reference mark on a later-described reference mark plate (not shown) on the wafer stage WST are simultaneously provided via the projection optical system PL. A pair of TTR (Through The Reticle) type reticle alignment microscopes (hereinafter referred to as “RA detection system” for convenience) (not shown) (not shown) using light having an exposure wavelength for observation is provided. The detection signals of these RA detection systems are supplied to the main controller 50 through an alignment controller (not shown). A configuration equivalent to that of the RA detection system is disclosed in, for example, JP-A-7-176468.

  In FIG. 1, wafer stage WST includes XY stage 42 and Z tilt stage 38 mounted on XY stage 42. The XY stage 42 is levitated and supported above the upper surface of the wafer base 16 by an air bearing (not shown) with a clearance of about several μm, for example. Further, the XY stage 42 is configured to be capable of two-dimensional driving in the Y direction which is the scanning direction and the X direction orthogonal thereto by a linear motor (not shown) constituting the wafer stage driving system 56W. A Z tilt stage 38 is mounted on the XY stage 42, and the wafer holder 25 is fixed on the Z tilt stage 38. The wafer W is held by the wafer holder 25 by vacuum suction or the like.

  As shown in FIG. 2, the Z tilt stage 38 is placed on the XY stage 42 by three Z position driving units 27A, 27B, and 27C (however, the Z position driving unit 27C on the back side in FIG. 2 is not shown). Supported by a point. These Z position driving units 27A to 27C have three actuators (for example, a voice coil motor) that independently drive the respective support points on the lower surface of the Z tilt stage 38 in the optical axis direction (Z direction) of the projection optical system PL. 21A, 21B, 21C (however, the actuator 21C on the back side of the paper in FIG. 2 is not shown), and the Z position drive units 27A, 27B, 27C of the Z tilt stage 38 are supported by the actuators 21A, 21B, 21C at the respective support points. It includes encoders 23A to 23C (however, encoder 23C on the back side in FIG. 2 is not shown) that detects a driving amount in the direction (displacement from the reference position).

  In this example, the actuators 21A, 21B, and 21C control the position of the Z tilt stage 38 (wafer W) in the optical axis AX direction (Z direction), the rotation angle θx around the X axis, and the rotation angle θy around the Y axis. To do. The stage controller 70 in FIG. 1 adjusts the position and leveling amount (rotation angle θx) of the Z tilt stage 38 so that the upper surface of the wafer W is focused on the image plane of the projection optical system PL during exposure. , Θy) is calculated, and the actuators 21A to 21C are driven using the calculation result. In FIG. 1, the linear motor and the like for driving the XY stage 42 and the Z position driving units 27A to 27C in FIG. 2 are collectively shown as a wafer stage driving system 56W.

In FIG. 1, a laser interferometer (hereinafter referred to as “wafer interferometer”) is placed on the Z tilt stage 38.
The movable mirror 52W that reflects the laser beam from 54W is fixed. The position of the Z tilt stage 38 (wafer stage WST) in the XY plane is always detected by the wafer interferometer 54W with a resolution of, for example, about 0.1 to 1 nm. Actually, on the Z tilt stage 38, a movable mirror having a reflective surface orthogonal to the scanning direction (Y direction) and a movable mirror having a reflective surface orthogonal to the non-scanning direction (X direction) are provided. The wafer interferometer is also provided with a plurality of axes in the X direction and the Y direction, respectively, and the position of the Z tilt stage 38 in the direction of five degrees of freedom (the position in the X direction, the Y direction, and the rotation angles θx, θy, θz). Measurement is possible. Position information (or speed information) of wafer stage WST is supplied to stage controller 70 and main controller 50 via this. Stage control device 70 controls the position of wafer stage WST in the XY plane via wafer stage drive system 56W in accordance with an instruction from main control device 50. Note that the end surface of the Z tilt stage 38 may be mirror-finished to form the reflective surface of the movable mirror 52W.

Further, the projection exposure apparatus of the present example is provided with an aerial image measurement device 59 (aerial image measurement system) used for measuring the imaging characteristics (optical characteristics) of the projection optical system PL. A part of the optical system constituting the aerial image measurement device 59 is arranged inside the Z tilt stage 38.
FIG. 3 is a partially cutaway view showing the aerial image measuring device 59. In FIG. 3, the aerial image measuring device 59 is a stage side component provided on the Z tilt stage 38, that is, a pattern forming member. And a relay optical system comprising a slit plate 90, lenses 84 and 86, a mirror 88 for bending an optical path, a light transmitting lens 87, and components outside the stage provided outside the wafer stage WST, that is, a mirror 96, a light receiving lens 89, And an optical sensor 94 (photoelectric sensor) composed of a photoelectric conversion element.

  More specifically, the slit plate 90 is provided on the upper surface of the end portion of the Z tilt stage 38 of the wafer stage WST, and is projected from above in a state of covering the opening with respect to the protruding portion 58 formed with an opening in the upper portion. It is inserted. The slit plate 90 is configured by forming a reflective film 83 also serving as a light shielding film on the upper surface of a rectangular flat glass substrate 82 parallel to the XY plane, and a slit shape having a predetermined width of 2D is formed on a part of the reflective film 83. The opening pattern (hereinafter referred to as “slit”) 122 is formed. 3 representatively shows one of a plurality of slits (see FIG. 9) provided in the slit plate 90. As the material of the glass substrate 82, here, synthetic quartz or fluorite having good ArF excimer laser light transmission property is used.

  In the state of FIG. 3, the measurement mark PM (measurement pattern) formed on the reticle mark plate RFM is located in the illumination area of the illumination light IL, and the image of the mark is slit by the projection optical system PL. Projected onto the plate 90. A part of the imaging light beam made of the illumination light IL passes through the slit 122. In the Z tilt stage 38 below the slit 122, a relay optical composed of lenses 84 and 86 is interposed via a mirror 88 that horizontally bends the optical path of illumination light IL (imaging light beam) incident vertically downward through the slit 122. A system (84, 86) is arranged. In addition, illumination light relayed by the relay optical system (84, 86) is almost outside the wafer stage WST on the side wall on the + Y direction side of the Z tilt stage 38 behind the optical path of the relay optical system (84, 86). A light transmission lens 87 that transmits light in the + Y direction is fixed.

  The optical path of the illumination light IL sent out of the wafer stage WST by the light sending lens 87 is 90 ° vertically upward by a mirror 96 having a predetermined length in the X direction and inclined at an inclination angle of 45 °. It can be bent. On the bent optical path, a light receiving lens 89 that is larger than the light transmitting lens 87 is disposed, and an optical sensor 94 is disposed above the light receiving lens 89. The light receiving lens 89 and the optical sensor 94 are housed in the case 92 while maintaining a predetermined positional relationship, and the mirror 96 is also fixed to the case 92 via a support member (not shown). The illumination light IL reflected upward by the mirror 96 is condensed on the light receiving surface of the optical sensor 94 by the light receiving lens 89. The case 92 is fixed to the vicinity of the upper end portion of the support column 97 implanted on the upper surface of the wafer base 16 via the mounting member 93.

  As the optical sensor 94, a photoelectric conversion element (photoelectric sensor) capable of accurately detecting weak light, for example, a photomultiplier tube (PMT, photomultiplier tube) or the like is used. The photoelectric conversion signal PS from the optical sensor 94 is sent to the main controller 50 via the signal processing device 80 of FIG. The signal processing device 80 can be configured to include, for example, an amplifier, a sample hold circuit, an A / D converter, and the like. The arrangement and shape of the actual plurality of slits represented by the slit 122 will be described later.

  According to the aerial image measurement device 59 configured as described above, the measurement mark PM (or the mark formed on the reticle R) formed on the reticle mark plate RFM (or the reticle R) is projected via the projection optical system PL. When measuring the projection image (aerial image) obtained in this way, the slit plate 90 is illuminated by the illumination light IL (imaging light beam) transmitted through the projection optical system PL. The illumination light IL that has passed through the slit 122 of the slit plate 90 is guided to the outside of the wafer stage WST through the lens 84, the mirror 88, the lens 86, and the light transmission lens 87. The light guided to the outside of the wafer stage WST is received by the optical sensor 94 via the mirror 96 and the light receiving lens 89, and a photoelectric conversion signal (light quantity signal) PS corresponding to the amount of received light is received from the optical sensor 94. It is output to the main controller 50 via the signal processing device 80 of FIG.

  In this example, since the measurement of the projected image (aerial image) of the measurement mark is executed by the slit scan method, the light transmitting lens 87 is in the X direction and Y direction with respect to the light receiving lens 89 and the optical sensor 94 at that time. Will move in the direction. Therefore, in the aerial image measuring device 59, the diameter of the light receiving lens 89 is set larger than the diameter of the light transmitting lens 87 so that all the light passing through the light transmitting lens 87 moving within a predetermined range is incident on the light receiving lens 89. Has been. In the aerial image measurement device 59 of this example, the moving unit including the slit plate 90 provided in the Z tilt stage 38 and the fixed unit including the optical sensor 94 provided in the case 92 are mechanically separated. Yes. The moving part and the fixed part are optically connected via the mirror 96 only when the aerial image is measured. Thereby, a decrease in measurement accuracy due to heat generation of the optical sensor 94 is suppressed.

  In the aerial image measuring device 59, the optical path between the light transmitting lens 87 and the light receiving lens 89 may be connected by a flexible optical fiber cable. Further, for example, when the heat generation of the optical sensor 94 is small, or when the influence of the heat generation can be reduced by the cooling mechanism, the optical sensor 94 and the light receiving lens 89 are fixed to the side surface in the + Y direction of the projection optical system PL, for example. It is also possible. Furthermore, the optical sensor 94 can be provided inside the wafer stage WST (Z tilt stage 38). The aerial image measurement and the aberration measurement method performed using the aerial image measurement device 59 will be described in detail later.

Returning to FIG. 1, an off-axis alignment system ALG as a mark detection system for detecting an alignment mark on the wafer W or a predetermined reference mark is provided on the side surface of the projection optical system PL. In this example, an image processing type alignment system, a so-called FIA (Field Image Alignment) system is used as the alignment system ALG. An imaging signal from the alignment system ALG is supplied to an alignment control device (not shown). Based on the imaging signal and the position information of wafer stage WST, which is the output of wafer interferometer 54W at that time, the alignment control device aligns the alignment mark in the stage coordinate system defined by the measurement value of wafer interferometer 54W or The coordinate position of the reference mark is calculated, and the calculation result is supplied to the main controller 50. Based on the coordinate position, main controller 50 calculates the interval (baseline) between the center of the projected image of the pattern of reticle R and the detection center of alignment system ALG, and the array coordinates of each shot area on wafer W. And so on.

  Further, as shown in FIG. 1, the projection exposure apparatus 10 of this example is provided with an oblique incidence type multi-point focal position detection system (60a, 60b) comprising an irradiation system 60a and a light receiving system 60b. The irradiation system 60a projects a plurality of slit images obliquely with respect to the optical axis AX on the test surface that is the surface of the wafer W or the surface of the slit plate 90, and the light receiving system 60b reflects the reflected light from the test surface. It receives light and re-images those slit images. Then, the light receiving system 60b supplies a detection signal corresponding to the lateral shift amount of the re-imaged slit images to the stage controller 70. In the stage control device 70, as an example, these detection signals are converted into defocus amounts, and the defocus amount in the Z direction with respect to the image plane of the projection optical system PL of the test surface from the plurality of defocus amounts, and X An inclination angle about the axis and the Y axis is obtained. The detailed configurations of the multipoint focal position detection system (60a, 60b) and the similar multipoint focal position detection system are disclosed in, for example, Japanese Patent Application Laid-Open No. Hei 6-283403. Detailed description is omitted.

  During normal exposure, the stage controller 70 uses the detection result of the multipoint focal position detection system (60a, 60b) to automatically focus the surface of the wafer W on the image plane of the projection optical system PL. The position and tilt angle of the Z tilt stage 38 are controlled via the wafer stage drive system 56W by the focus method and the auto leveling method. Further, based on a command from the main controller 50 during exposure or aerial image measurement, the stage controller 70 projects the surface of the wafer W or the surface of the slit plate 90 via the wafer stage drive system 56W. It is also possible to defocus the image plane by an amount designated in the Z direction.

  In addition, an environmental sensor 81 for detecting atmospheric pressure fluctuations and temperature fluctuations is provided in the vicinity of the projection optical system PL in FIG. The measurement result by the environment sensor 81 is supplied to the main controller 50. Further, in the memory 51 connected to the main control device 50, for example, information on predetermined higher-order aberrations previously measured at the time of assembly adjustment of the projection optical system PL and the like and an aberration measurement method described later are obtained. Information on aberrations of the projection optical system PL is stored.

  Next, the scanning exposure operation in the projection exposure apparatus 10 of this example will be briefly described. First, main controller 50 sets an illumination condition optimal for exposure using reticle R based on an operator's instruction. Next, the alignment of the reticle R and the alignment of the wafer W are performed using the reticle alignment microscope and the wafer side alignment system ALG. Thereafter, the next shot area to be exposed on the wafer W is positioned on the front side of the optical axis AX by stepping the wafer stage WST. Then, the irradiation of the illumination light IL is started, and in synchronization with the movement of the reticle R in the Y direction at the speed Vr with respect to the illumination area via the reticle stage RST, the exposure area via the wafer stage WST. Thus, one shot area on the wafer W moves in the Y direction at a velocity β · Vr (β is the projection magnification of the projection optical system PL). In this way, the stepping operation between shots and the synchronous scanning operation for each shot are repeated, and the pattern image of the reticle R is transferred to all shot regions on the wafer W by the step-and-scan method.

  By the way, in the above-described scanning exposure operation, in order to transfer the pattern of the reticle R onto the wafer W with high resolution and high accuracy via the projection optical system PL, the imaging characteristics of the projection optical system PL are in a predetermined state. It needs to be adjusted. For this purpose, it is necessary to measure the imaging characteristics with high accuracy. In the following, it is assumed that the imaging characteristics of the projection optical system PL to be measured and adjusted are predetermined aberrations.

In addition, assuming that wavefront aberration is used to classify the aberration, an aberration function indicating the wavefront aberration on the exit pupil (or pupil plane) of the projection optical system PL is W (ρ, θ). In this case, ρ is a normalized position (radial radius) in the radial direction of the exit pupil of the projection optical system PL, and θ is an angle, the aberration function W (ρ, θ) is the above ( As shown in equation (1), series expansion can be performed using i-th order (i = 1, 2,...) Zernike's Polynomial fi (ρ, θ) and its coefficient Z _i . The Zernike polynomials may also be referred to as Fringe Zernike polynomials or Zernike circle polynomials.

As an example, the Zernike polynomials fi of the first order to the 37th order are illustrated together with the corresponding coefficients Z _i as shown in Table 1 below.

As shown in Table 1 above, the Zernike polynomial fi (ρ, θ) of each order is represented by the radius vector (ρ).
The radial function, which is a function of, and the angle (θ) function are expressed in a separated form. Zernike polynomials can be classified into those whose radial function is an odd function and those that are an even function. For example, for f7 and f8 shown in Table 1, the radial functions are both 3ρ ³ −2ρ and an odd function, and f5 and f6 are both even and the radial functions are ρ ^2. Yes. The function of the angle θ in each order Zernike polynomial is sin (mθ) (m is an integer of 1 or more) or cos (mθ) (m is an integer of 0 or more). When the integer m that defines the angle mθ is an odd number, the corresponding radial function is an odd function, and when the integer m is 0 or an even number, the corresponding radial function is an even function. . As described above, an aberration corresponding to a Zernike polynomial whose radial function is expressed by an odd function is called an odd function aberration, and an aberration corresponding to a Zernike polynomial whose radial function is expressed by an even function is called an even function aberration.

The aberrations of the projection optical system PL can be classified into lateral aberrations, which are lateral shifts of the image, and longitudinal aberrations, which are changes in the contrast of the image. The former lateral aberration is an odd function aberration, and the latter longitudinal aberration is an even function aberration. It is. Accordingly, spherical aberration, defocus, and the like among the aberrations of the projection optical system PL are even function aberrations, and the coma aberration is an odd function aberration. Further, the coefficient Z _i of the i-th order Zernike polynomial represents the aberration expressed by the i-th order Zernike polynomial. In this example, the aberration Z ₉ and Z ₁₆ , which are spherical aberrations (even function aberrations), and coma aberration (odd function aberrations) are performed by the imaging characteristic control mechanism of the projection optical system PL including the imaging characteristic correction controller 78 of FIG. A plurality of aberrations including aberrations Z ₇ , Z ₈ , Z ₁₄ , Z ₁₅ , and aberrations Z ₂ , Z ₃ , which are distortion (odd function aberration), can be corrected. In this example, the above-described aerial image measurement device 59 is used to measure these odd-function aberration and even-function aberration. Hereinafter, aerial image measurement by the aerial image measurement device 59, measurement of aberration of the projection optical system PL, and the like will be described in detail.

  FIG. 3 shows a state in which the aerial image of the measurement mark PM formed on the reticle mark plate RFM is measured using the aerial image measuring device 59. Instead of the reticle mark plate RFM, it is also possible to use a test reticle dedicated to aerial image measurement, or a reticle R used for manufacturing a device on which a dedicated measurement pattern is formed. Here, on the reticle mark plate RFM, a line-and-space pattern (hereinafter referred to as a ratio of the width of the line portion to the width of the space portion (duty ratio)) having a periodicity in the Y direction at a predetermined location is 1: 1. It is assumed that a measurement mark PM composed of “L & S pattern” is formed. Such a measurement mark PM is actually one of a plurality of marks provided on the reticle mark plate RFM.

  Here, a method of aerial image measurement using the aerial image measurement device 59 will be briefly described. For example, as shown in FIG. 4A, the slit plate 90 is formed with a slit 122 (opening pattern) having a predetermined width 2D extending in the X direction. In the measurement of the aerial image, the main reticle 50 of FIG. 1 drives the movable reticle blind 30B via a blind driving device (not shown). As shown in FIG. 3, the illumination area of the illumination light IL of the reticle R It is limited only to a predetermined area including the measurement mark PM.

  In this state, when the illumination light IL is irradiated onto the reticle mark plate RFM, as shown in FIG. 4A, the light diffracted and scattered by the measurement mark PM (illumination light IL) is projected onto the projection optical system PL. And a spatial image (projected image) PM ′ of the measurement mark PM is formed on the image plane of the projection optical system PL. At this time, wafer stage WST is set at a position where the aerial image PM ′ is formed on the + Y direction side (or the −Y direction side) of slit 122 on slit plate 90 of aerial image measurement device 59. And A plan view of the slit plate 90 at this time when viewed from the projection optical system PL side is shown in FIG. If the projection magnification of the projection optical system PL is ¼, the pitch (period) of the spatial image PM ′ is ¼ of the pitch of the L & S pattern of the measurement mark PM. In the following description, it is assumed that the line width and pitch of each measurement mark and the like indicate the line width and pitch of the aerial image, respectively.

  When the main controller 50 drives the wafer stage WST (scanning mechanism) in the + Y direction as indicated by the arrow F in FIG. 4A via the wafer stage drive system 56W, the slit 122 becomes a space. The image PM ′ is scanned in the Y direction. During this scanning, light (illumination light IL) that passes through the slit 122 is received by the optical sensor 94 via the optical system in the wafer stage WST, the mirror 96, and the light receiving lens 89, and the photoelectric conversion signal PS is converted into a signal processing device. 80 to the main controller 50. Main controller 50 acquires light intensity distribution information corresponding to aerial image PM ′ based on the photoelectric conversion signal. Note that the aerial image PM ′ and the slit 122 may be relatively scanned in a direction perpendicular to the slit 122. Therefore, the aerial image PM ′ side may be moved by moving the reticle mark plate RFM via the reticle stage RST (scanning mechanism) in FIG.

FIG. 4B shows an example of a photoelectric conversion signal (light intensity signal) PS obtained in the above aerial image measurement. In this case, the aerial image PM ′ is averaged due to the influence of the width (2D) of the slit 122 in the scanning direction (Y direction). Therefore, if the transmittance distribution in the scanning direction (here Y direction) of the slit 122 is p (y), the light intensity distribution of the aerial image is i (y), and the observed light intensity signal is m (y). The relationship between the intensity distribution i (y) of the aerial image and the observed intensity signal m (y) is expressed by the following equation (2). In the equation (2), the unit of the intensity distribution i (y) and the intensity signal m (y) is the intensity per unit length, and the u axis is the same coordinate axis as the y axis.

  However, the function p (y) of the transmittance distribution of the slit 122 is expressed by the following equation (3).

That is, the observed intensity signal m (y) is a convolution of the function p (y) of the slit 122 and the light intensity distribution i (y) of the aerial image.
Therefore, in terms of measurement accuracy, the width (hereinafter simply referred to as “slit width”) 2D of the slit 122 in the scanning direction (Y direction) is preferably as small as possible. As in this example, when a photomultiplier tube (PMT) is used as the optical sensor 94, even if the slit width becomes very small, if the scanning speed is slowed and the measurement takes time, the light intensity (light intensity ) Can be detected. However, in reality, since there is a certain restriction on the scanning speed at the time of aerial image measurement from the viewpoint of throughput, if the slit width 2D is too small, the amount of light transmitted through the slit 122 becomes too small and measurement is difficult. turn into. In this example, the slit width 2D is set to about 200 nm or less, for example, about 100 to 150 nm. Note that the aerial image PM ′ may be scanned using a pinhole instead of the slit 122. In the case of a pinhole, there is no directionality, but the amount of light is reduced. Therefore, pinholes can be used particularly when the pitch of the aerial image PM ′ is large. When using a pinhole, in order to increase the amount of received light, the diameter is set to about twice the slit width 2D, that is, about 400 nm or less, for example, about 200 to 300 nm.

  As described above, the light intensity distribution in the aerial image (projected image) PM ′ of the measurement mark PM can be measured by the aerial image measuring operation using the aerial image measuring device 59 (first step). Information on the measured light intensity distribution is supplied to the main controller 50 of FIG. Since the information on the light intensity distribution includes information on the horizontal image formation position (lateral shift) and amplitude (contrast) of the aerial image PM ′, the main control device 50 (arithmetic unit) determines that information. Can be used to determine odd-function aberration and even-function aberration (second step). Furthermore, using the light intensity distribution information, it is possible to calibrate the best focus position with respect to the projection optical system PL, and it is also possible to obtain the positions of a predetermined mark image in the X direction and the Y direction.

In this example, it is assumed that the aberration represented by the Zernike polynomial up to a predetermined order exceeding 9th order is measured. For this purpose, as the aerial image PM ′, it is necessary to measure aerial images of various periodic marks having different directions and pitches as described later. For this purpose, it is necessary to form slits (opening patterns) arranged in a plurality of directions on the slit plate 90 as well.
FIG. 9 shows an arrangement of a plurality of slits as an opening pattern formed on the slit plate 90 of this example. In FIG. 9, the slit plate 90 has a slit width 2D extending in the Y direction and a length L. And a slit 122a that extends in the X direction and is formed by rotating the slit 122b by 90 °. Further, on the slit plate 90, six aberration measuring slits 9A, 9B, 9C, 9D, 9E, and 9F (slit-like opening patterns) each having a width 2D and a length L1 are formed. In this case, the slits 9A and 9D are arranged on the extensions of the slits 122b and 122a, respectively, and the remaining four slits 9B, 9C, 9F, and 9E are located at the positions of the four apexes of a substantially square, and a set of slits. 122b and 9A and another set of slits 122a and 9D substantially constitute two adjacent sides of the square. In addition, a pin hole 123 having a diameter of about 4D is formed at substantially the center of the square. The illumination light that has passed through the slits 122a, 122b, 9A to 9F, and the pinhole 123 is received by the optical sensor 94 in FIG. In this case, in order to individually detect the illumination light that has passed through the slits 122a and 9D, the slits 122b and 9A, the slit 9B, the slit 9C, the slit 9E, the slit 9F, and the pinhole 123, for example, a slit is formed on the bottom surface of the slit plate 90. A liquid crystal panel as a selection member may be provided so that only illumination light that has passed through the selected set or one slit or pinhole may enter the optical sensor 94.

  In this example, the slit width 2D of each slit image is about 100 nm, the length L of the slits 122a and 122b is about 8 μm, and the length L1 of the slits 9A to 9F is about 3 μm. These slit widths and lengths are values at the stage of the projected image. The distance between the slit 122a and the slit 9D and the distance between the slit 122b and the slit 9A are about 1 to 2 μm. In this case, the slits 122a and 122b are used for calibration of the best focus position and position measurement of the mark image. Further, when the slits 122a and 122b are used, the aberration measuring slits 9D and 9A on the extensions are also used at the same time. By simultaneously using the aberration measurement slits 9D and 9A for the slits 122a and 122b in this way, a sufficient amount of light can be secured when measuring the best focus position and the image position, and measurement reproducibility is improved. To do.

  Further, the longitudinal directions (arrangement directions) of the aberration measurement slits 9A, 9B, 9C, 9D, 9E, and 9F in FIG. 9 are respectively 90 °, φ5, φ6, and 0 ° counterclockwise with respect to the X axis. , Φ7, and φ8. In this case, each of the slits 9A to 9F is scanned relative to a corresponding aerial image in a direction (measurement direction) orthogonal to the longitudinal direction. In other words, the measurement directions of the six slits 9A, 9B, 9C, 9D, 9E, and 9F are 0 °, φ1 (= φ5-90 °), φ2 (= φ6) counterclockwise with respect to the X axis, respectively. -90 °), 90 °, φ3 (= φ7 + 90 °), and φ4 (= φ8 + 90 °). As an example, the angle φ1 is 30 °, the angle φ2 is 45 °, the angle φ3 is 120 °, and the angle φ4 is 135 °. Thus, by using the slit plate 90 of this example, the aberration as the optical characteristic of the projection optical system PL can be quickly measured on-body.

When the aberration to be measured is 0θ, 1θ, and cos2θ components up to the coefficient Z ₃₇ , the two slits 9A and 9D on the slit plate 90 in FIG. 9, that is, the measurement direction is relative to the X axis. Two slits (opening patterns) that intersect at 0 ° and 90 ° may be used.
Next, in order to theoretically and quantitatively represent the odd-function aberration and the even-function aberration, the following calculation is performed. The aerial image to be measured in this case is defined as an aerial image PM ′ of the measurement mark PM in FIG. Here, in the aerial image PM ′ of the measurement mark PM, the complex amplitude distribution in the Y direction in FIG. 4A is o (y), and the spatial frequency spectrum is O (s) (s is the spatial frequency axis) (The coordinates above). Of the spatial frequency components included in the periodic pattern of the spatial image PM ′ of the measurement mark PM, if the two spatial frequency components are f ′ and f ″, respectively, the spectra O (f ′) and O (f ″) An intensity distribution i (y) of the aerial image PM ′ of the measurement mark PM is obtained by integrating an interference fringe generated by beats multiplied by a certain weight at the entire spatial frequency. This weight is referred to as the cross modulation coefficient T (f ′, f ″). The cross modulation coefficient T (f ′, f ″) is defined by the following equation (4).

In this equation, F is a pupil function at the exit pupil (or pupil plane) of the projection optical system PL ( ^* indicates a complex conjugate), and σ (ξ, η) is an effective light source. Note that ξ and η are orthogonal coordinate axes on the exit pupil of the projection optical system PL. Therefore, the imaging formula of the measurement mark PM by partial coherent illumination is expressed by the following formula (5).

[Measurement of odd-function aberration (lateral aberration)]
In general, the absolute value of the aberration amount of the odd function aberration and the image shift amount at the measurement point in the pupil coincide with each other if the unit is a phase. Conversion from the unit of length to the unit of phase is expressed by the following equation.
Image shift amount (phase) = (image shift amount (length) / mark pitch) × 360 ° (K1)
Note that the amount of aberration at the measurement point in the pupil is determined by the effective illumination size (the size of the secondary light source on the pupil plane of the illumination optical system), and an example of the secondary light source is a small annular zone.

When the measurement point in the pupil is indicated by polar coordinates (ρ, θ), θ coincides with the measurement mark arrangement direction (repetition direction and the opposite direction), and the radius ρ is the mark pitch P and the projection optical system. It is as follows using the numerical aperture NA and the exposure wavelength λ.
ρ = λ / (P · NA) (K2)
Here, the imaging formula (5) when there is an odd function aberration is specifically calculated. As an example, it is assumed that the odd-function aberration is a coma aberration Z ₇ represented by a seventh-order Zernike polynomial f7 (ρ, θ). The Zernike polynomial f7 (ρ, θ) indicates a phase delay on the exit pupil as follows.

f7 (ρ, θ) = (3ρ ³ −2ρ) cos θ (A1)
In this case, for convenience of explanation, the position y in the equation (5) is replaced with x ′. As shown in FIG. 4A, the pitch P _h of the aerial image PM ′ of the measurement mark PM is 1 / f _h (P _h = 1).
/ F _h ) and 50% duty (the line width and the space width are the same). Aerial image P
The intensity I _hN (x ′) of the Nth-order harmonic component (N is an odd number) when the spatial frequency component corresponding to the pitch P _{h of} M ′ is the fundamental wave component is expressed by the following equation. In this case, f _h is the fundamental spatial frequency of the aerial image PM ′.

In this equation, c ₀ and c _N represent the amplitudes of the 0th and Nth order Fourier spectra, respectively. The angle φ _hN is an Nth-order harmonic aberration (lateral shift) and is expressed by the following equation.

Here, the phase of the fundamental wave is considered in an example of coherent illumination (coherence factor (σ value) is 0). When the measurement mark PM is an L & S pattern with a 50% duty and a pitch _Ph of 1 / f _h as described above, the complex amplitude distribution o (x ′) is expressed by the following equation. However,
The values of Fourier coefficients c _0, c _1, c _3, c ₅ ,... Are 1/2, 1/4, −1/12, 1/20,.

Here, let us consider the case of coherent illumination where the measurement mark PM is 0.5 μm L & S pattern (aerial image line width is 0.5 μm) and the numerical aperture NA of the projection optical system PL in FIG. 3 is 0.78. Then, the harmonics contributing to image formation are up to the third order. Furthermore, in coherent illumination, the amount of aberration on the exit pupil only gives at most one phase difference for each diffracted light. Further, since the seventh-order Zernike polynomial f7 representing the coma aberration Z ₇ is an odd function as can be seen from the equation (A1), the aberration amount at the spatial frequency f _h is exp (iφ ₁ ), and the aberration at the spatial frequency 3f _h . When the quantity is exp (iφ ₃ ), the equation (8) can be expressed by the following equation (9).

Considering the case where a primary (fundamental) component is constructed from each Fourier component of equation (9), it is clear that the primary component is composed only of beats of the 0th and 1st components. . Considering the beat of the zeroth-order component and the first-order component, the intensity distribution of the first-order component is expressed by the equation (10), and the phase φ _h1 of the first-order component is expressed by the equation (11).

  In coherent illumination, the intensity distribution of the primary component is as follows.

This equation indicates that the phase of the aerial image changes by φ ₁ due to the aberration Z ₇ and an image shift occurs. Specifically, the sensitivity (Zernike sensitivity) which is the amount of change in phase with respect to the change in aberration is examined. In this case, when considering an aerial image of a 0.5 μm L & S pattern (pitch is 1 μm) under an exposure wavelength λ of 193 nm and a numerical aperture NA of 0.78, the sine of the diffraction angle of the first-order diffracted light is 0. Since 193, the radius ρ = 0.247 on the exit pupil of the primary component. Therefore, from the equation (A1), the value (Zernike sensitivity) for the first-order component of the seventh-order Zernike polynomial f7 (ρ, θ) is as follows.

f7 (ρ = 0.247, θ = 0) = − 0.449 (A2)
For example, when an aberration of 20 mλ occurs, the aberration Z ₇ = 20 (mλ), and therefore the phase φ ₁ is as follows. Note that λ is 360 °.
φ ₁ = f ₇ × Z ₇ = −0.449 × 20 (mλ) = − 3.23 ° (A3)
This phase change is as follows because the lateral shift amount (image shift amount) of the aerial image is 1 μm per pitch.

Image shift amount = 1 (μm) × (−3.23 / 360) = − 8.97 (nm)
Therefore, the image shift amount of the aerial image PM ′ shown in FIG. 4A is measured, and is expressed by a seventh-order Zernike polynomial using the measured value and the Zernike sensitivity that can be obtained in advance by calculation (optical simulation). The coma aberration Z ₇ (lateral aberration amount) can be measured. In this case, it is necessary to prepare a plurality of pitches (line widths) or periodic directions of measurement marks to be measured depending on the order of aberrations up to the Zernike polynomial.

[Measurement of even function aberration (longitudinal aberration)]
Here, the imaging formula (5) when there is even function aberration is specifically calculated. As an example, the even-function aberration is assumed to be aberrations Z ₉ , Z ₁₆ , and Z ₂₅ represented by the 9th, 16th, and 25th order Zernike polynomials f9, f16, and f25 in Table 1. The 9th-order Zernike polynomial f9 (ρ, θ) important in this example is shown below. The 16th and 25th order Zernike polynomials f16 (ρ, θ) and f25 (ρ, θ) are shown in Table 1.

f9 (ρ, θ) = (6ρ ⁴ −6ρ ² +1) (A4)
As shown in FIG. 4A, the pitch P _h of the aerial image PM ′, which is the projection image of the measurement mark PM, is 1 / f _h and is 50% duty (the line width and the space width are the same). Suppose there is. The intensity distribution I _{hN — even} (y) of the Nth-order harmonic component (N is an odd number) when the spatial frequency component corresponding to the pitch P _h of the aerial image PM ′ is the fundamental wave component is expressed by the following (15) It is shown by the formula. In this case, f _h is the fundamental spatial frequency of the aerial image PM ′.

Therefore, when the amount of aberration at the frequency Nf _h of the even function aberration of the projection optical system PL is expressed as exp (iφ _{N — even} ) with respect to the 0th-order light, the aforementioned intensity distribution I _{hN — even — coh} (in coherent illumination) y) is expressed by the following equation (16) from the above equation (15).

  As is apparent from the above equation (16), the phase difference of the Nth-order harmonic component of the fundamental wave component does not change due to the even function aberration, and the amplitude of the component changes. Therefore, in the aberration measurement method of this example, the amount of even function aberration of the projection optical system PL is calculated based on the amplitude of the spatial frequency component of a predetermined order (Nth order). In this example, even function aberration may be obtained based on the amplitude of even-order, for example, second-order harmonic components. However, since the second-order harmonic component does not exist in principle in the spatial frequency component of the spatial image PM ′ of the measurement mark PM that is the L & S pattern with a duty of 50%, the harmonic components of other orders It is necessary to use a predetermined beat component as the second harmonic component.

Next, a method for measuring even function aberration of the projection optical system PL of this example will be described in detail. In this measurement method, the actual aberration of the projection optical system PL, for example, another even function aberration is measured while changing the amount of a predetermined even function aberration. As an example, here, the low-order even function aberration Z ₉ is changed, and the value of Z ₉ when giving the peak of the amplitude of each frequency component is used as an evaluation value for other aberrations, for example, higher order even function aberrations such as Z _16. Shall be measured. The aberration Z ₉ can be controlled by the imaging characteristic control mechanism of the projection optical system PL including the imaging characteristic correction controller 78 of FIG. 2 as described above.

Here, a method for measuring the aberration Z ₁₆ will be described. FIG. 5 corresponds to the phase delay due to each aberration (Z ₄ , Z ₉ , Z ₁₆ , Z ₂₅ ) of the even function aberration at the normalized position on the exit pupil (or pupil plane) of the projection optical system PL. The values of Zernike polynomials f4, f9, f16, and f25 in Table 1 are shown as phase delay levels. 6 shows the phase lag level when the amount of aberration Z ₉ is changed with the amount of aberration Z _{16 being} the same as in FIG. As shown in FIG. 5, the phase lag level for Z ₁₆ represented by a solid line has an extreme value when the coordinate value of the pupil position is ± 0.525, and the value is 0.447. Yes. As a pattern for generating the first-order diffracted light located at the coordinate value 0.525 of the pupil position, when the optical system conditions are NA = 0.78, σ = 0 (coherent illumination), λ = 193 nm, Since pitch = λ / (NA × 0.525) = 471 nm, an L & S pattern having a line width of 235 nm can be used. The amplitude of the spatial image of the pattern, the aberration of the projection optical system PL When only Z _16, maximized when Z ₁₆ is zero. This is clear from the equation (16). Further, since the phase lag level for the aberration Z ₁₆ is −1 on the optical axis (pupil position = 0), the phase lag at an extreme value of ± 0.525 due to the aberration Z ₁₆ in FIG. The phase difference between the level (± first-order diffracted light phase) and the phase delay level on the optical axis (zero-order light phase) is 1.447.

At this time, in FIG. 6, it is assumed that there is only an aberration amount of Z ₁₆ expressed by a coefficient (Zernike coefficient) multiplied by a 16th-order Zernike polynomial (currently defined as “C” because of an unknown number). If you, using the phase difference in FIG. 5 (1.447), the phase difference B between the zero-order light on the first-order diffracted light and the optical axis is located in the pupil coordinate 0.525 due to Z ₁₆ are the following Thus, C × 1.447.

B = C × 1.447 (A9)
Next, in FIG. 5, the phase difference between the first-order diffracted light located at the pupil coordinate 0.525 caused by Z ₉ and the zero-order light on the optical axis is approximately −1.2. Accordingly, in FIG. 6, when the aberration amount of Z ₉ expressed by the Zernike coefficient is generated only by V, the first-order diffracted light positioned at the pupil coordinate 0.525 caused by Z ₉ and 0 on the optical axis. The phase difference A with the next light is as follows.

A = −V × 1.2 (A10)
Then, the phase difference A cancels out the phase difference B caused by Z ₁₆ in the equation (A9) (A = −B), and when the phase difference between the first-order diffracted light and the zero-order light becomes zero, The amplitude of the aerial image of the L & S pattern having a width of 235 nm is maximized. Further, the phase difference A can be arbitrarily set by changing the aberration amount V of Z ₉ . As described above, the aberration amount V of Z ₉ for canceling out the phase difference B by the phase difference A is approximately 1.2C as follows.

V = −B / (− 1.2) = C × 1.447 / 1.2≈1.2C (A11)
In this case, when the aberration amount V of Z ₉ is set so that the amplitude of the aerial image of the L & S pattern having a line width of 235 nm is maximized, the aberration amount V includes information on the aberration amount C of Z ₁₆ . In other words, the value of Z ₁₆ can be converted from the value of Z ₉ . In this case, from the viewpoint of improving the measurement accuracy, the value of Z ₁₆ is converted from the value of Z ₉ that maximizes the amplitude of the fundamental wave component of the aerial image when Z ₉ is sequentially changed. Is desirable.

FIG. 7 shows a calculation result when the amplitude of the fundamental wave component of the aerial image intensity is calculated by simulation when Z ₉ is changed. In this simulation, the optical system conditions were NA = 0.78, σ = 0.1, λ = 193 nm, and the measurement mark was an L & S pattern with a line width of 235 nm. FIG. 7 also shows changes in the amplitude of the fundamental wave component when the amount of aberration of Z ₁₆ is 0, 20, 40, 60, 100 (mλ). Here, the reason why the amplitude value of the fundamental wave component is sequentially measured while changing Z ₉ is to determine the peak position of the amplitude with high accuracy. The peak position of the amplitude is determined by the amount of aberration of Z ₉ (hereinafter referred to as “Z ₉ converted value”). The peak position of the amplitude changes in proportion to the amount of Z _16.

FIG. 8 shows the correlation between the amount of aberration of Z ₁₆ in FIG. 7 and the peak position of the amplitude (Z ₉ converted value). As shown in FIG. 8, the amount of aberration of Z ₁₆ and the peak position in terms of Z ₉ are almost completely proportional, and the correlation coefficient R ² = 1. The ratio of the change of the peak position (Z ₉ converted value) to the change of Z ₁₆ is calculated as 0.9157 in terms of so-called Zernike sensitivity. This sensitivity varies depending on NA, σ, wavelength, and measurement mark.

As described above, the aerial image measurement is executed by setting Z ₉ as the first aberration of the projection optical system PL to a plurality of amounts, and the measured aerial image, that is, the fundamental wave included in the light intensity signal. A position where the magnitude of the component is maximum and a position where the magnitude of the harmonic component of the predetermined order is maximized are calculated in terms of Z _9, and the difference between these positions is used as an evaluation amount to determine Z as the second aberration. ₁₆ can be calculated.

In the case of measuring aberrations Z ₂₅ and Z ₃₆ of other even function components, three or more types of measurement marks are used, and Zernike sensitivities at these measurement marks are obtained with respect to Z ₁₆ , Z ₂₅ and Z ₃₆ , and three types are measured. By measuring the peak position δ _m (m = 1, 2, 3,...) Of the amplitude in terms of Z ₉ with the above marks and solving a predetermined simultaneous equation, each aberration of Z ₁₆ , Z ₂₅ , Z ₃₆ The amount can be determined.

  When obtaining the amplitude of the fundamental wave component (fundamental frequency component), for example, a spatial image of the measurement mark whose period direction is the X direction is scanned in the X direction by the slit 9A in FIG. It takes time to acquire the corresponding one-dimensional aerial image information and then to Fourier transform this to calculate the amplitude of the fundamental wave component. In particular, for example, when the scanning in the X direction and the calculation using the Fourier transform are repeated at each position in the Z direction while gradually changing the position of the slit plate 90 in FIG. It becomes considerably long and the throughput of the exposure process decreases.

Therefore, in this example, the measurement time of the aberration of the projection optical system PL is shortened by detecting the aerial image of the measurement mark in a three-beam interference state. This will be described with reference to FIGS.
10 (A), (B), (C), (D) and FIGS. 11 (A), (B), (C), (D) each have a line width of 478 nm as a measurement mark on the wafer. The light intensity distribution of the aerial image of the line and space pattern with a duty ratio of 1: 1 is shown. The measurement mark is a mark for a normal reticle formed on a chromium film, the exposure wavelength is 193 nm, the detection slit width is 150 nm, the numerical aperture NA of the projection optical system is 0.92, and the illumination optical system The numerical aperture iNA is 0.92 × 0.045 (σ value is 0.045).

10A and 11A are aerial images at the best focus position, FIG. 10B and FIG. 11B are aerial images when 0.3 μm is defocused, and FIG. 10C. FIG. 11C shows the aerial image when 2.0 μm is defocused, and FIG. 10D and FIG. 11D show the aerial image when 2.34 μm is defocused.
10A to 10D show the light intensity distribution of the aerial image when the numerical aperture NA of the projection optical system is kept at the maximum value (0.92) (the horizontal axis indicates the position (μm)). The wave line A represents the aerial image itself, the sine wave curve A1 represents the fundamental wave component (fundamental frequency component), the straight line DC represents the DC component, and the other plurality of curves AH represent the harmonic components. As in this example, the measurement mark having a duty ratio of 1: 1 (the width of the light transmitting space is equal to the width of the non-light transmitting line) generates even-order diffracted light other than the zero-order light. do not do.

  Under this condition, if there is no deviation in the telecentricity of the projection optical system, the diffracted light passing through the projection optical system is up to the 0th order light, ± 1st order light, and ± 3rd order light. is there. In this state, the Fourier spectrum of the aerial image includes not only the fundamental wave but also harmonics up to the second, third, fourth, fifth, and sixth harmonics. It is only necessary to measure the amplitude of the fundamental wave by measuring the intensity at one point of the aerial image. For example, in the aerial image A in FIG. 10A, the top of the mountain (high level portion) is broken into two. Furthermore, it is asymmetric due to the effect of coma. The cause of the crack at the top of the peak depends on the phase of the third harmonic being inverted.

Therefore, as an example, the position where the amplitude of the aerial image is maximized is assumed to be the position where the amplitude of the fundamental wave component is maximized. At this time, if the amplitude of the sine wave of the fundamental wave component A1 is first maximized from the right end of FIG. 10A, that is, the signal intensity of the fundamental wave component A1 at a position of approximately −0.478 μm is measured. Good.
Further, in FIG. 10B showing the aerial image when 0.3 μm is defocused, the phase of the third-order component is reversed at this time, and the amplitude of the aerial image A near −0.478 μm is rather increased. On the other hand, the amplitude of the fundamental wave component A1 decreases. This means that the defocus amount when the amplitude of the aerial image A is maximized is different from the defocus amount when the amplitude of the fundamental wave component A1 is maximized. Further, in either of FIG. 10C given 1 μm defocus and FIG. 10D given 2.34 μm defocus, the intensity information at one point of the five-beam interference aerial image A is used. It is difficult to measure the defocus position when the amplitude of the fundamental wave component A1 is maximized.

  Therefore, in this example, when the even function aberration of the image of the measurement mark that is in an interference state of five or more light beams is measured, the variable aperture stop AS of the projection optical system PL shown in FIG. A three-beam interference state of first-order diffracted light and ± first-order diffracted light is set (first step). Thereafter, the light intensity distribution of the aerial image is measured using the slits 9A to 9F in FIG. 9 (second step), and the aberration of the projection optical system PL is obtained based on the measurement result (third step). Here again, as in the above-described example, the case where the maximum numerical aperture NA of the projection optical system PL is 0.92, the numerical aperture iNA of the illumination optical system is 0.92 × 0.045, and the exposure wavelength λ is 193 nm is considered. In this case, the variable aperture stop AS of FIG. 3 is used to reduce the numerical aperture NA of the projection optical system PL to 0.56, resulting in a three-beam interference state.

The condition for the three-beam interference is expressed by the following equation, where NA _aft is the numerical aperture NA after focusing and P is the mark pitch.
λ / P + iNA ≦ NA _aft ≦ 3λ / P-iNA (17)
Now, the spatial image A, fundamental wave component A1, and harmonic wave component AF when the numerical aperture NA _aft is reduced to 0.56 are shown in FIGS. The harmonic component AF has only the second harmonic component. Third-order and higher-order harmonic components are also observed, but this is for simulation including non-linearity of the electrical system, and is optically up to the second-order harmonic. The second harmonic has the same phase as the fundamental frequency component, and the peak positions match. Further, as shown in FIGS. 11B, 11C, and 11D, the amplitude of the second harmonic component is hardly changed by defocusing. This is because the second harmonic component is formed by two-beam interference caused by ± first order diffracted light. Incidentally, in the state where the numerical aperture iNA = 0 of the illumination optical system, in principle, the change in amplitude is eliminated by defocusing. Further, since the image is formed by the two-beam interference, the aberration is influenced only by the portion of the projection optical system that is exposed to the ± first-order diffracted light.

  Here, in order to measure the amplitude of the fundamental wave component A1 as in FIG. 10A, it is assumed that the light intensity in the vicinity of −0.478 μm from the right end of FIG. Although the position of the peak position of the aerial image A is displaced by an odd function of aberration, the lateral shift amounts due to the aberration in the wafer surface of the first harmonic and the second harmonic coincide with each other. This is because the amount of misalignment due to both aberrations is determined by the amount of aberration in the portion where the ± first-order diffracted light strikes. Accordingly, the fundamental wave component and the second harmonic component are shifted by the same amount due to the presence of the odd function aberration, so that the peak position of the aerial image formed from these and the peak position of the fundamental wave component and the second harmonic component are Match. Since the aerial image is composed of a fundamental wave and a second harmonic component, and these three peak positions coincide with each other, it goes without saying that the aerial image has a symmetrical shape with respect to the peak position.

  Accordingly, the peak position of the fundamental wave component A1 can be easily estimated by binarizing the aerial image A with respect to the position in the lateral direction with an appropriate threshold and detecting the center of the edge of the high level portion, for example. By measuring the light intensity by moving the light receiving surface of the aerial image measuring device 59 of FIG. 3 parallel to the optical axis of the projection optical system at one point of the peak position thus obtained, refer to FIG. As described above, it is possible to measure another even-function aberration amount that gives the amplitude intensity peak of the fundamental wave component. Another even function aberration amount that gives the maximum amplitude of the fundamental wave component is not necessarily the same as another even function aberration component that gives the maximum amplitude of the second harmonic component, but the numerical aperture iNA of the illumination optical system Is small (generally iNA is about 0.1 or less), it is considered that the aberration within the area occupied by the respective diffracted lights in the pupil plane of the projection optical system is uniform. In this case, since the amplitude of the second harmonic component is considered to be constant, it is possible to measure the aberration amount of another even function aberration component that gives the maximum value of the amplitude of the fundamental wave component.

On the other hand, when the higher order even function aberration than the defocus aberration is not negligible and the numerical aperture iNA of the illumination optical system at the time of measurement is relatively large, the amplitude of the second harmonic component is maximized. There is a high possibility that one even function aberration amount is different from the even function aberration amount that maximizes the fundamental wave component.
This phenomenon will be described with reference to FIG. FIG. 12 shows an example of a wavefront represented by an even function. In FIG. 12, coordinates (ξ, η) indicate a pupil coordinate system, and a direction orthogonal to this plane indicates an aberration amount in an arbitrary unit. Further, the places through which the 0th-order light and the ± 1st-order diffracted light pass are indicated by circles representing the numerical aperture iNA of the illumination optical system. It is assumed that the wavefront in the circle can be approximated by a plane. In this case, the amplitude of the second harmonic formed by the two-beam interference is maximized when the circular portions indicating the passing portions of the ± first-order diffracted light are parallel to each other. Therefore, if another even-function aberration is given with a low-order even-function aberration of the same level as in FIG. 12, the relative inclination of these two circular portions can be controlled. There is a possibility that the amplitude of the wave can be increased. On the other hand, the state in which the amplitude of the fundamental wave component is maximized is such that the average aberration amount in the circular portion through which the ± first-order diffracted light of FIG. 12 passes matches the average aberration amount in the circular portion through which the zero-order light passes. When another even aberration is given, there is no guarantee that the amplitude of the second harmonic becomes maximum. Thus, the measurement error of the even function aberration amount caused by the size of the numerical aperture iNA of the illumination optical system and the presence of higher order even function aberration is confirmed by simulation, and the numerical aperture iNA of the optimum illumination optical system is determined in advance. The size is desirable.

Paying attention to the above points, another even-function aberration amount that maximizes the amplitude of the fundamental component by changing the other even-function aberration while measuring the light intensity at one point of the aerial image. That is, the even function aberration amount at the aberration measurement point on the pupil plane of the projection optical system PL determined by the mark of the pitch to be measured can be measured. A method for giving another even function of a predetermined amount at the time of measurement is the movement of the lens elements 13 ₁ , 13 _2, etc. of the projection optical system PL in FIG. 2, or the light receiving part (slit) of the aerial image measuring device 59 in FIG. Given by the defocus of the plate 90).

  Next, FIG. 13 shows an example of a measurement pattern used in combination with the slit plate 90 (slits 9A to 9F) of FIG. 9 when measuring aberrations represented by coefficients up to the 37th order of the Zernike polynomial. The measurement pattern in FIG. 13 is formed on the pattern surface of the reticle mark plate RFM in FIG. 3, but the measurement pattern is formed on a test reticle loaded on the reticle stage RST instead of the reticle R. You may keep it.

  In FIG. 13, the attached mark area 61 of the reticle mark plate RFM includes an X-axis mark 62X for measuring the best focus position and a Y-axis mark 62Y (focus mark) and an X-axis for measuring the position of the projected image. A mark 63X and a Y-axis mark 63Y (image position mark) are formed. The size of the attached mark region 61 is approximately 60 μm square when represented by a projection image by the projection optical system PL, and the arrangement of the marks 62X, 62Y, 63X, and 63Y represents the arrangement in the state of the projection image. 9 is scanned in the X direction with respect to the images of the X-axis marks 62X and 63X, and the slit 122a in FIG. 9 is scanned in the Y direction with respect to the images of the Y-axis marks 62Y and 63Y. By measuring the light intensity distribution of the aerial image, it is possible to measure the defocus amount of the projected image of the reticle mark plate RFM and the position of the image of the attached mark region 61 in the X direction and the Y direction.

  Further, five L & S patterns (line and pace patterns) 2A, which are formed at a pitch gradually decreasing in the X direction and have a duty ratio of 1: 1 in an area close to the attached mark area 61 of the reticle mark plate RFM. A first periodic mark 1A including 2B, 2C, 2D, and 2E is formed, and the periodic mark 1A is formed on the −X direction side with a pitch that gradually decreases in the X direction, and has a duty ratio of 1: 1. A first periodic mark 5A including L & S patterns 6A, 6B, 6C, 6D, 6E, and 6F is formed. That is, the periodic direction of the periodic marks 1A and 5A (periodic marks) is the X direction, and the periodic marks 1A and 5A are arranged in parallel to the X axis.

  As an example, the line width (1/2 of the pitch) and the number of line patterns of the L & S patterns 2A to 2E and the L & S patterns 6A to 6F are the values shown in the upper and lower stages of Table 2 (A) below (1 nm), respectively. The following fractions are rounded off). The length of the L & S patterns 2A to 2E and 6A to 6F in the direction orthogonal to the periodic direction is set to 4 μm, that is, about 30% longer than the lengths (3 μm) of the slits 9B, 9C, 9E, and 9F in FIG. Has been. Note that the length, line width, and pitch of the pattern used in this example are all represented by values in the aerial image by the projection optical system PL.

  From Table 2 (A), the line widths of the L & S patterns 2A to 2E gradually decrease from the maximum 1080 nm to 120 nm (240 nm in pitch), and the line widths of the L & S patterns 6A to 6F gradually increase from 1320 nm to 120 nm (pitch). (240 nm). The smallest pitch among the L & S patterns 2A to 2E and the smallest pitch among the L & S patterns 6A to 6F are set equal to 240 nm. The number of line patterns of the L & S patterns 2A to 2E is set so that the lengths (and thus areas) of the L & S patterns 2A to 2E in the periodic direction (measurement direction) are substantially equal to each other. That is, the lengths in the periodic direction (measurement direction) of the L & S patterns 2A to 2E are commonly set to approximately 10 μm.

  In FIG. 13, second periodic marks 1B and 5B each having a shape obtained by rotating the periodic marks 1A and 5A counterclockwise by an angle φ1 are formed in parallel above the periodic mark 1A, and below the periodic mark 1A. Third periodic marks 1C and 5C having a shape obtained by rotating the periodic marks 1A and 5A counterclockwise by an angle φ2 are formed in parallel. In addition, fourth periodic marks 1D and 5D each having a shape obtained by rotating the periodic marks 1A and 5A by 90 ° clockwise are formed in parallel between the periodic marks 1A and 5A, respectively. Further, fifth periodic marks 1E and 5E each having a shape obtained by rotating the periodic marks 1A and 5A counterclockwise by an angle φ3 are formed in parallel above the periodic mark 5A, and the periodic marks 5A are respectively formed below the periodic mark 5A. Sixth periodic marks 1F and 5F having a shape obtained by rotating the marks 1A and 5A counterclockwise by an angle φ4 are formed in parallel. The angles φ1, φ2, φ3, and φ4 in this example are 30 °, 45 °, 120 °, and 135 °, respectively, similarly to the corresponding angles in FIG.

  In the reticle mark plate RFM of FIG. 13, periodic marks 3A, 3B, 3C, 3D, 3E, and 3F are formed at positions obtained by translating the periodic marks 1A, 1B, 1C, 1D, 1E, and 1F in the + Y direction. Periodic marks 7A, 7B, 7C, 7D, 7E, and 7F are formed at positions where the marks 5A, 5B, 5C, 5D, 5E, and 5F are translated in the + Y direction, respectively. The periodic direction of the periodic marks 3A to 3F is parallel to the periodic direction of the periodic marks 1A to 1F. The periodic marks 3A to 3F have a duty ratio of 1: 1 at a pitch gradually decreasing with respect to the periodic direction. L & S patterns 4A, 4B, 4C, 4D, and 4E are formed. Further, the periodic direction of the periodic marks 7A to 7F is parallel to the periodic direction of the periodic marks 5A to 5F, and the duty ratio is 6: 1 in the periodic marks 7A to 7F at a pitch gradually decreasing with respect to the periodic direction. L & S patterns 8A, 8B, 8C, 8D, 8E, and 8F are formed. Line widths of L & S patterns 4A to 4E and L & S patterns 8A to 8F (1/29 of the pitch and the number of line patterns are the values shown in the upper and lower stages of Table 2 (B), respectively. Rounded).

  From Table 2 (B), the line widths of the L & S patterns 4A to 4E gradually decrease from the maximum 1620 nm to 180 nm (360 nm in pitch), and the line widths of the L & S patterns 8A to 8F gradually increase from 1980 nm to 180 nm (pitch). 360 nm). The smallest pitch among the L & S patterns 4A to 4E and the smallest pitch among the L & S patterns 8A to 8F are set equal to 360 nm.

  The periodic marks 1A to 1F and 3A to 3F in FIG. 13 include five L & S patterns 2A to 2E and 4A to 4E, respectively. Accordingly, when the measurement pattern of FIG. 13 is illuminated with the illumination light IL of FIG. The ± first-order diffracted lights from the periodic marks 1A, 1B, 1C, 1D, 1E, and 1F and the periodic marks 3A, 3B, 3C, 3D, 3E, and 3F are each of six on the pupil plane of the projection optical system PL. It passes through 10 positions (5 on each side) arranged along the direction. Therefore, the position through which the ± 1st-order diffracted light passes is the phase measurement point.

  On the other hand, since the periodic marks 5A to 5F and 7A to 7F in FIG. 13 include six L & S patterns 6A to 6F and 8A to 8F, the ± first-order diffracted lights from the periodic marks 5A to 5F and 7A to 7F are , Each of which passes through 12 positions (six on one side) arranged along six directions on the pupil plane of the projection optical system PL. The position where the ± first-order diffracted light passes is the phase measurement point. Therefore, when phase information is measured using the periodic marks 5A to 5F or 7A to 7F, aberration information represented by higher order Zernike polynomials can be obtained.

  Specifically, by using marks of six or more pitches arranged along two different directions, when the wavefront aberration is represented by a Zernike polynomial, the wavefront aberration represented by the Zernike polynomial up to the 16th order is represented. Among them, 0θ, 1θ, and cos2θ components can be measured. Further, by using marks of six or more pitches arranged along six different directions, wavefront aberrations expressed by Zernike polynomials up to the 37th order can be measured.

  In FIG. 13, the minimum pitches of the L & S patterns 2A to 2E in the -Y direction side periodic marks 1A to 1F and the L & S patterns 6A to 6F in the periodic marks 5A to 5F are the periodic marks 3A to 3F on the + Y direction side. The L & S patterns 4A to 4E and the L & S patterns 8A to 8F in the period marks 7A to 7F are set smaller than the minimum pitch. In this case, since the diffracted light from the periodic mark having a smaller pitch passes through the outer region on the pupil plane of the projection optical system PL, the periodic marks 1A to 1F on the −Y direction side and Of 5A to 5F and the + Y direction side periodic marks 3A to 3F and 7A to 7F, a periodic mark whose minimum pitch is closer to the minimum pitch of the reticle pattern to be transferred may be used. Accordingly, it is possible to measure an aberration close to the aberration at the time of transferring the reticle pattern to be actually exposed via the projection optical system PL.

  Specifically, for example, a case where diffracted light generated from the L & S patterns 2A to 2E in the period marks 1A to 1F in FIG. 13 is detected will be described. In this case, if the periodic marks 1A to 1F in FIG. 13 are regarded as the projected images, first, the slit 9A in FIG. 9 is shown at the end of the projected image of the periodic mark 1A in the periodic direction (here, the X direction) as shown by a dotted line. 3 is driven, and the images of the L & S patterns 2A to 2E in the periodic mark 1A are scanned by the slit 9A in the measurement direction parallel to the periodic direction by driving the wafer stage WST of FIG. Measure specific position). This peak position corresponds to obtaining the peak position of the fundamental wave component A1 as described above. At this time, the variable aperture stop AS of the projection optical system PL in FIG. 3 is driven to shield diffracted light other than the 0th order light and the ± 1st order diffracted light (the same applies hereinafter). Thereafter, the slit plate 90 is moved in the Z direction to measure a change in light intensity, and a predetermined aberration is measured (details will be described later).

  Next, after moving the slit 9B to the end in the periodic direction of the projected image of the periodic mark 1B in FIG. 13 (here, the direction intersecting at an angle φ1 with respect to the X axis), the L & S in the periodic mark 1B is moved by the slit 9B. The images of the patterns 2A to 2E are scanned in the measurement direction parallel to the periodic direction to measure the peak position of the light intensity distribution, and then the change in the light intensity is measured by moving the slit plate 90 in the Z direction. Similarly, after the slits 9C to 9F are sequentially moved to the ends in the periodic direction of the projected images of the periodic marks 1C to 1F in FIG. 13, the L & S patterns 2A to 2E in the periodic marks 1C to 1F are respectively moved by the slits 9C to 9F. The image is scanned in a measurement direction parallel to the periodic direction to measure the peak position of the light intensity distribution, and then the slit plate 90 is moved in the Z direction to measure the change in light intensity.

Similarly, the peak positions are obtained by scanning the slits 9A to 9F in the measurement direction which is the periodic direction with respect to the projected images of the periodic marks 5A to 5F, the periodic marks 3A to 3F, and the periodic marks 7A to 7F in FIG. After that, by moving the slit plate 90 in the Z direction, it is possible to measure the aberration passing through the corresponding measurement point of the pupil plane.
When the even function aberration of the projection optical system PL is measured using the aberration measurement marks 1A to 1I (L & S patterns 2A to 2E) of FIG. 13 and the slit plate 90 of FIG. 9, the measurement is performed in the following order. To do.

1) A measurement mark to be actually measured is selected from the many measurement marks in FIG.
2) Depending on the pitch P of the measurement mark, the variable aperture stop AS of FIG. 3 is driven so as to satisfy the expression (17), and ± 3rd order diffracted light L (+3), The diffracted light of L (−3) or more is shielded, and only the 0th-order light L (0) and ± 1st-order diffracted lights L (+1) and L (−1) are allowed to pass through the projection optical system PL.

3) The peak position (specific position) in the direction perpendicular to the optical axis AX of the light intensity distribution of the spatial image of the measurement mark is measured. For example, if the number of measurement marks (measurement direction is X direction) is an odd number, after the aerial image is scanned in the X direction by the corresponding slit 9A in FIG. 9 to obtain a signal corresponding to the light intensity distribution The center position of the light intensity distribution is measured.
4) The slit 9A is positioned at the peak position of the light intensity distribution of the aerial image measured in 3).

FIG. 14 shows a state where the slit 9A is positioned at the peak position in the X direction of the light intensity distribution of the aerial image B of the measurement mark. Note that the peak position need not coincide with the optical axis AX of the projection optical system PL.
5) While measuring the light intensity detected through the measurement slit 9A, the slit 9A is scanned in the direction perpendicular to the image plane of the projection optical system PL, that is, in the direction parallel to the optical axis AX (Z direction). . Even during scanning in the Z direction, the aerial image measurement device 59 in FIG. 3 continuously captures a light intensity detection signal (light intensity signal).

  6) From the relationship between the Z-direction position of the obtained slit 9A and the light intensity signal, the Z-direction position of the slit 9A that gives the maximum light intensity signal is obtained. At this time, the defocus amount of the slit 9A is calculated by subtracting the position of the reference surface that has been measured in advance from the position in the Z direction. From this defocus amount, a predetermined (another) even function aberration amount is obtained using the relationship corresponding to FIG. 7 and FIG.

That is, the obtained light intensity signal becomes an intensity signal curve B1 as shown in FIG. 15, and the amount of aberration is determined from the deviation amount ΔEA of the peak position of the curve B1 from the reference position. Further, the horizontal axis of FIG. 15 indicates the controllable even function aberration amount EA. When the controllable even function aberration is a defocus aberration, the defocus amount is converted into an aberration amount.
7) The obtained even function aberration amount is set as the even function aberration amount at the measurement point on the pupil plane of the projection optical system PL corresponding to the measurement mark.

The above is a case where measurement is performed by changing the even function aberration by defocusing. However, a method of measuring this by changing the other even function aberration by moving the lens element of the projection optical system PL in FIG. This is the second method.
The even function aberration measurement method of this example has been described above. For aberration measurement of odd function components, a method of obtaining the phase of the fundamental wave component of each aerial image by Fourier transform is desirable from the viewpoint of accuracy and measurement time. Since the aerial image can be Fourier-transformed with respect to the odd function component, it is not necessary to reduce the numerical aperture NA of the projection optical system PL to the three-beam state.

  Note that it is desirable to adjust the projection optical system PL of FIG. 1 based on the aberration amount measured by the aberration measurement method of each of the above embodiments, and ideally, the aberration of the projection optical system PL is set to 0. Actually, some aberrations may remain even after the projection optical system PL is adjusted. Therefore, when the projection exposure apparatus of the above embodiment is operated, after the projection optical system PL is adjusted, the residual aberration of the projection optical system PL is again measured as the initial aberration amount by using the aberration measurement method of the above embodiment. It may be left. In this case, when the aberration fluctuation amount of the projection optical system PL is periodically measured by the aberration measurement method of the above-described embodiment and the aberration changes due to a change over time, the main controller 50 performs the operation shown in FIG. The imaging characteristics of the projection optical system PL may be adjusted via the imaging characteristic correction controller 78 so that these aberration amounts return to the initial aberration amounts. In addition, it is desirable from the viewpoint of measurement stability that periodic aberration measurement is performed using a measurement pattern formed on the reticle mark plate RFM or the like.

In each of the above embodiments, the case where the reduction system is used as the projection optical system has been described. However, the present invention is not limited to this, and the projection optical system may be an equal magnification or an enlargement system. Either a catadioptric system or a reflective system may be used.
Further, for example, for a semiconductor device, a step of designing the function / performance of the device, a step of manufacturing a reticle based on the design step, a step of manufacturing a wafer from a silicon material, and the projection exposure apparatus (exposure apparatus) of the above-described embodiment Thus, the wafer is manufactured through a step of transferring a reticle pattern to a wafer, a device assembly step (including a dicing process, a bonding process, and a package process), an inspection step, and the like.

  In each of the above embodiments, the case where the present invention is applied to a scanning exposure type projection exposure apparatus has been described. However, the present invention is not limited thereto, and the mask pattern is transferred to the wafer while the mask and the wafer are stationary. The present invention can also be applied to a static exposure type (collective exposure type) projection exposure apparatus such as a stepper. The present invention can also be applied to the case where the aberration of the projection optical system is measured by an immersion type exposure apparatus disclosed in, for example, WO 99/49504.

  Further, the present invention is not limited to an exposure apparatus for manufacturing a semiconductor device, but is used for manufacturing a display including a liquid crystal display element, a plasma display, and the like. An exposure apparatus for transferring a device pattern onto a glass plate and a thin film magnetic head. The present invention can also be applied to an exposure apparatus for transferring a device pattern used in the above to a ceramic wafer, and an exposure apparatus used for manufacturing an imaging device (CCD, etc.), an organic EL, a micromachine, a DNA chip, and the like. In addition to microdevices such as semiconductor elements, circuits for glass substrates or silicon wafers are used to manufacture masks used in optical exposure equipment, EUV exposure equipment, X-ray exposure equipment, and electron beam exposure equipment. The present invention can also be applied to an exposure apparatus that transfers a pattern.

  In addition, this invention is not limited to the above-mentioned embodiment, Of course, a various structure can be taken in the range which does not deviate from the summary of this invention.

  According to the exposure method and apparatus of the present invention, the optical characteristics of the projection optical system can be measured in a short time with the projection optical system mounted on the exposure apparatus. Therefore, the mask pattern can be transferred onto the substrate with high accuracy by correcting the optical characteristics of the projection optical system in accordance with the measurement result.

It is the figure which partly cut off showing the schematic structure of the projection exposure apparatus of an example of embodiment of this invention. FIG. 2 is a partially cutaway view showing an imaging characteristic control mechanism of projection optical system PL of FIG. 1 and a Z tilt stage of wafer stage WST. It is a figure which shows the internal structure of the aerial image measuring device with which the projection exposure apparatus of FIG. 1 was equipped. (A) is a figure which shows an example of the slit on the slit board 90 of FIG. 3, (B) is a figure which shows an example of the photoelectric conversion signal obtained in the case of an aerial image measurement. It is a figure which shows the phase delay level of the some even function aberration component with respect to the normalized pupil position. By adjusting the aberration Z ₉ is a diagram for explaining when measuring other aberrations Z _16. When the aerial image has a different aberrations Z _16, is a graph showing the calculation results of the change in the amplitude of the fundamental wave component contained in each aerial image when each changing the aberration Z _9. It is a graph showing the correlation between the peak position of the fundamental wave component in the aberration Z ₁₆ and Z ₉ terms. It is an enlarged plan view which shows arrangement | positioning of the actual slit on the slit board 90 of FIG. It is a figure which shows the change of the light intensity distribution of the aerial image of the measurement mark at the time of changing the defocus amount in the state where the numerical aperture NA of the projection optical system PL is not reduced. It is a figure which shows the change of the light intensity distribution of the aerial image of the measurement mark at the time of changing the defocus amount in the state which narrowed the numerical aperture NA of the projection optical system PL. It is a figure which shows an example of the wave aberration in the pupil surface of projection optical system PL. FIG. 4 is an enlarged plan view showing an example of a measurement pattern on the pattern surface of the reticle mark plate RFM of FIG. 3. It is a figure which shows the state which scans a slit in a direction parallel to the optical axis of a projection optical system with respect to an aerial image. In FIG. 14, it is a figure which shows an example of the change of the light intensity signal at the time of changing the controllable even function aberration amount.

Explanation of symbols

  9A to 9F: Slit, 1A to 1F, 3A to 3F, 5A to 5F, 7A to 7F ... Periodic mark, 2A to 2E, 4A to 4E, 6A to 6F, 8A to 8F ... L & S pattern, 12 ... Illumination optical system, DESCRIPTION OF SYMBOLS 14 ... Light source, 50 ... Main controller, 51 ... Memory, 59 ... Aerial image measuring device, 78 ... Image formation characteristic correction controller, 80 ... Signal processing apparatus, 90 ... Slit plate, 123 ... Pinhole, PL ... Projection optical system AS ... Variable aperture stop, R ... Reticle, RST ... Reticle stage, RFM ... Reticle mark plate, W ... Wafer, WST ... Wafer stage

Claims (11)

In an optical property measurement method for measuring optical properties of a test optical system,
A first step of controlling the numerical aperture of the test optical system in order to shield diffracted light other than diffracted light of three orders from each of a plurality of marks having different pitches;
A second step of measuring a light intensity distribution of an image formed when the three orders of diffracted light pass through the optical system to be tested;
And a third step of obtaining an optical characteristic of the optical system to be measured based on the light intensity distribution measured in the second step.   The optical characteristic measurement method according to claim 1, wherein the three orders of diffracted light are zero-order diffracted light and ± first-order diffracted light. The second step includes
Detecting a specific position in a direction perpendicular to the optical axis of the test optical system of a light intensity distribution of an image by the test optical system of each of the plurality of marks;
The optical characteristic measurement method according to claim 1, further comprising: detecting a change in light intensity of the image by the test optical system in the direction parallel to the optical axis at the specific position. The optical characteristic of the test optical system obtained in the third step is a first aberration of the test optical system,
The second step includes a step of measuring a light intensity distribution of an image of the plurality of marks by the test optical system while changing a second aberration different from the first aberration of the test optical system. The optical characteristic measuring method according to any one of claims 1 to 3, further comprising:   5. The change in the second aberration of the test optical system is given by displacement of a predetermined optical member in the test optical system or defocus of a measurement position of the image. Optical property measurement method.   6. The optical characteristic according to claim 1, wherein the plurality of marks having different pitches are included in marks having six or more types of pitches arranged along two different directions. Measurement method.   6. The optical characteristic according to claim 1, wherein the plurality of marks having different pitches are included in marks having pitches of six or more types arranged along six different directions. Measurement method. In an exposure method of exposing a second object with an exposure beam via a first object and a projection optical system,
A measurement step of measuring the optical characteristics of the projection optical system by the optical characteristic measurement method according to any one of claims 1 to 7,
An exposure method comprising: a correction step of correcting an optical characteristic of the projection optical system based on a measurement result of the measurement step. In an optical property measuring device that measures the optical properties of a test optical system,
A substrate on which a plurality of marks having different pitches are formed;
An illumination system for illuminating the substrate;
A variable aperture stop for sequentially controlling the numerical aperture of the optical system to be shielded in order to shield diffracted light other than diffracted light of three orders from each of the plurality of marks having different pitches;
An aerial image detection system that detects each of the images formed by the test optical system using the diffracted light of the three orders of the plurality of marks having different pitches;
An optical characteristic measuring apparatus comprising: an arithmetic unit that obtains optical characteristics of the optical system to be detected based on a detection result of the aerial image detection system. The aerial image detection system has a slit-like opening pattern for detecting an image by the test optical system,
A stage that moves the slit-shaped opening pattern in a direction parallel to the optical axis of the optical system to be tested;
The arithmetic unit calculates optical characteristics of the test optical system based on a detection result of the aerial image detection system obtained when the slit-shaped opening pattern is moved in a direction parallel to the optical axis of the test optical system. The optical characteristic measuring apparatus according to claim 9, wherein the optical characteristic measuring apparatus is obtained. In an exposure apparatus that illuminates a first object with an exposure beam and exposes the second object with the exposure beam via the first object and a projection optical system,
The optical property measuring device according to claim 9 or 10,
A first stage for holding a substrate on which the plurality of marks having different pitches are formed together with the first object or instead of the first object;
A second stage that holds the light receiving unit of the aerial image detection system together with the second object,
An exposure apparatus for measuring an optical characteristic of the projection optical system using the optical characteristic measuring apparatus.

Images