JP2009031169A - Position detection apparatus, exposure apparatus, and manufacturing method for device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce the influence due to the pattern or the reflectivity of a surface to be inspected, without having to drive an optical member, and to detect the positional information of the normal direction of the surface to be inspected. <P>SOLUTION: This position detection apparatus includes a transmitted light optical system 30A for forming images B1-B9 of pin holes A1-A9 on the lower surface of a reticle R; a received light optical system 30B for receiving reflected light from the lower surface and forming images C1-C9 of the pin holes; a photoelectric sensor 44 for detecting the images C1-C9 through pinholes D1-D9; a stage for scanning the reticle R; and an RAF control system 52A for determining the positions of the Z direction of a series of measuring points of the lower surface of the reticle R, based on a detected signal of the photoelectric sensor 44 obtained by the scan of the reticle R. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a position detection device for detecting position information of a surface to be measured, an exposure device provided with the position detection device, and a microdevice (electronic device) such as a semiconductor element or a liquid crystal display device using the exposure device. The present invention relates to a device manufacturing method.

  In a lithography process for manufacturing a semiconductor element or the like, a pattern formed on a reticle (mask) is transferred onto a wafer (or glass plate or the like) coated with a resist (photosensitive agent) on the surface via a projection optical system. An exposure apparatus is used. As an exposure apparatus, a step-and-repeat reduction projection exposure apparatus (so-called stepper) or a step-and-scan scanning projection exposure apparatus (so-called scanning stepper) that synchronously scans a reticle and a wafer is generally used. Has been used.

  With high integration of semiconductor elements and miniaturization of patterns, high resolution (high resolution) is required for exposure apparatuses. Further, in the exposure apparatus, exposure is performed by positioning the wafer surface within the range of the depth of focus (DOF) of the image plane of the projection optical system (the best imaging plane of the pattern) in order to prevent image blur due to defocus. There is a need to do. However, if the exposure wavelength is shortened and the numerical aperture (NA) is increased in order to increase the resolution, the depth of focus becomes narrower. Therefore, in a scanning projection exposure apparatus, in order to position the wafer surface within the range of the focal depth of the image plane of the projection optical system, the height of the wafer stage is based on the measurement value of the autofocus sensor that measures the wafer surface. By performing the control, the focus error due to the tilt or undulation in the exposed area on the wafer surface during the scanning exposure is corrected.

However, in recent years, the resolution has further increased and the depth of focus has further narrowed, and in addition to the focus error due to the wafer surface tilt and waviness, the focus error due to reticle deflection has become a problem. That is, the shape of the image plane of the projection optical system also changes depending on the deflection of the reticle lower surface (pattern surface). Therefore, in order to correct a focus error due to the deflection of the lower surface of the reticle, the measurement light is irradiated almost perpendicularly to the lower surface of the reticle through a lens driven in the optical axis direction. ) Is used (for example, see Patent Document 1).
JP 2007-48823 A

Since the above conventional autofocus sensor for a reticle includes a mechanism for driving the lens in the optical axis direction, the optical system can be miniaturized. However, since it is necessary to drive the lens during measurement, it is difficult to improve the measurement speed, and in order to improve the measurement accuracy, it is necessary to take measures against vibration when driving the lens.
In addition, since the pattern for the device is formed on the lower surface of the reticle (surface to be measured), the intensity of the measurement light varies depending on the reflectance of the formed pattern in addition to the height of the lower surface of the reticle, As a result, the measured value varies. Therefore, it is preferable that the autofocus sensor includes a mechanism that can reduce the influence of the pattern or reflectance of the test surface.

  In view of such problems, the present invention can reduce the size of the optical system without using a method of driving the optical member, and reduce the influence of the pattern or reflectance of the test surface as necessary. An object of the present invention is to provide a position detection technique capable of detecting position information in the normal direction of the surface to be measured, an exposure technique using this position detection technique, and a device manufacturing method using this exposure technique.

  A position detection apparatus according to the present invention is a position detection apparatus that detects position information in a normal direction of a test surface (Ra), and a first image of a plurality of openings (A1 to A9) on the test surface. A light transmission system (30A) that forms a light beam with a different defocus amount with respect to a predetermined reference surface (RFMa), and receives reflected light from the test surface to form a second image of the plurality of openings. The first light receiving system (30B) and the first light receiving sensor (44) for detecting the second images of the plurality of openings through the openings (D1 to D9) having a predetermined positional relationship with respect to the conjugate plane with the reference plane, respectively. ), A relative scanning mechanism (RST) that relatively scans the test surface and the first image of the plurality of apertures along the reference plane, and the test surface and the plurality of apertures by the relative scanning mechanism. The detection information of the first light receiving sensor obtained when the first image is relatively scanned. Based on the arithmetic unit for obtaining position information of the normal direction of the test surface and (52A), those with a.

  According to the present invention, for example, the position in the optical axis direction of the second image of the plurality of apertures formed by the first light receiving system by the reflected light of the light irradiated substantially perpendicularly to the test surface is the second image. And varies depending on the position in the normal direction of the test surface. Therefore, by obtaining the light amount information obtained from the first light receiving sensor when the first image of the plurality of openings sequentially passes through one measurement point on the test surface, without driving the optical member, The position information in the normal direction of the test surface can be detected by using an optical system that can be miniaturized.

  Further, by performing relative scanning between the first image of the plurality of openings and the test surface, position information can be obtained at a plurality of adjacent measurement points. For example, predetermined statistical processing is performed on the position information. Thus, the influence of the pattern or reflectance of the test surface can be reduced.

[First Embodiment]
A preferred first embodiment of the present invention will be described below with reference to FIGS. In this example, the present invention is applied to a projection exposure apparatus having an autofocus sensor for a reticle.
FIG. 1 is a diagram showing a schematic configuration of a scanning exposure type projection exposure apparatus (exposure apparatus) including a scanning stepper according to the present embodiment. In FIG. 1, the projection exposure apparatus of this embodiment includes an exposure light source (not shown), an illumination optical system IL that illuminates a reticle R (mask) with exposure light from the exposure light source, and a circuit pattern for transfer on the lower surface. A reticle stage RST that holds and moves the formed (drawn) reticle R, and a projection optical system PL that projects an image of the pattern of the reticle R onto a wafer W (photosensitive substrate) coated with a resist (photosensitive material). And a wafer stage WST that holds and moves the wafer W. Further, the projection exposure apparatus includes a reticle autofocus sensor (hereinafter, referred to as a reticle autofocus sensor) for detecting positions in the normal direction (here, parallel to the optical axis direction of the projection optical system PL) at a plurality of positions on the lower surface of the reticle R. 2) and a wafer autofocus sensor (hereinafter referred to as an AF sensor) for detecting the position (focus position) of the projection optical system PL in the optical axis direction at a plurality of measurement points on the surface of the wafer W. (60a, 60b) and a focusing mechanism (image position correcting mechanism) that focuses the surface of the wafer W on the image plane of the projection optical system PL based on the measurement results of the RAF sensor 2 and the AF sensor (60a, 60b). And a main control system 50 composed of a computer that controls the overall operation of the apparatus. In the following, the Z-axis is taken in parallel to the optical axis AX of the projection optical system PL, the X-axis is perpendicular to the plane of FIG. 1 in the plane perpendicular to the Z-axis, and the direction is parallel to the plane of FIG. Y axis is set for each. A direction (Y direction) parallel to the Y axis is the scanning direction SD of the reticle R and the wafer W during scanning exposure.

In FIG. 1, an illumination optical system IL includes a light quantity adjusting optical system, an illuminance uniformizing optical system including an optical integrator, an optical system 4 including a reticle blind, a relay lens system, and the like, a condenser lens 6, and an optical path bending unit. Mirror 8. As an exposure light source (not shown), a KrF or ArF excimer laser, an F ₂ laser, a mercury lamp (g-line, i-line, etc.) or the like can be used.

  A reticle stage RST for holding the reticle R is mounted on a guide surface substantially parallel to the XY plane on the reticle base RBS so as to be movable via an air bearing, and is driven by a drive mechanism (not shown) such as a linear motor. , Driven at a constant speed in the Y direction (scanning direction SD), and movable in the X direction with a predetermined width. On the reticle stage RST, a reference mark or the like is formed on the lower surface so as to be close to the reticle R in the Y direction, and the reticle mark plate RFM having a narrower width in the Y direction than the reticle R is held. The reticle R and the reticle mark Openings 10a and 10b for allowing exposure light and measurement light from the RAF sensor 2 to pass are formed in the reticle stage RST on the bottom side of the plate RFM. An opening 11a for allowing exposure light to pass is formed in a region including the optical axis AX of the reticle base RBS, an opening 11b is formed in a region away from the optical axis AX in the Y direction, and the reticle R and the reticle mark plate are formed through the opening 11b. The lower surface of the RFM is irradiated with measurement light from the RAF sensor 2 (details will be described later).

  Further, a laser interferometer 13 that irradiates the movable mirror 12 with a measurement laser beam is disposed so as to face the movable mirror 12 fixed on the reticle stage RST. As shown in FIG. 2B including a plan view of the reticle stage RST, the movable mirror 12 in FIG. 1 is a Y-axis biaxial movable mirror (for example, a corner reflector) 12YA, 12YB and an X-axis elongated movable mirror. The laser interferometer 13 includes a Y-axis two-axis laser interferometers 13YA and 13YB and an X-axis laser interferometer 13X.

  In FIG. 1, the laser interferometer 13 uses, for example, a reference mirror (not shown) provided on the side surface of the projection optical system PL as a reference, and at least positions Y, X of the reticle stage RST, RY, RX in the range of 0.5-. It measures with a resolution of about 0.1 nm, and also measures the rotation angle around the Z axis. The measurement value is supplied to a stage control unit in the main control system 50, and the stage control unit controls the two-dimensional position and speed of the reticle stage RST via a drive mechanism (not shown) based on the measurement value. Information on positions RY and RX in the Y direction and X direction of reticle stage RST is also supplied to RAF control system 52 that processes detection signals of RAF sensor 2 as described later.

  The projection optical system PL is a refracting system or a catadioptric system that is telecentric on both sides and has a reduction magnification such as 1/4 or 1/5, and displays an image of a pattern in an illumination area that is elongated in the X direction on the lower surface of the reticle R. Projection is performed on an exposure area on one shot area on the wafer W. A variable aperture stop (not shown) for controlling the numerical aperture NA is installed on or near the pupil plane of the projection optical system PL. In addition, through split lens barrels 55A and 55B, which are part of the barrels of the projection optical system PL, via three drive elements (piezo elements, etc.) 57A and 57B and lens frames 56A and 56B that can be expanded and contracted in the Z direction. Lens elements 58A and 58B are supported. The drive amounts of the drive elements 57A and 57B are controlled by the drive system 54 under the control of the main control system 50. By controlling the position of the lens elements 58A and 58B and / or the tilt angle with respect to the optical axis AX via the drive elements 57A and 57B, the imaging characteristics such as the field curvature and the focus position of the projection optical system PL are controlled. Can do. An imaging characteristic control mechanism of the projection optical system PL is configured including the drive elements 57A and 57B and the drive system 54. Note that the number and position of lens elements (or optical members) in the projection optical system PL driven by the imaging characteristic control mechanism are set according to the type of imaging characteristics to be controlled.

  On the other hand, wafer W is sucked and held on wafer stage WST via wafer holder WH. The wafer stage WST controls the wafer table 21 to which the wafer holder WH is fixed, the XY stage 23, and the position of the wafer table 21 in the Z direction with respect to the XY stage 23 and the tilt angles around the X and Y axes. The three Z drive units 22A, 22B, and 22C are provided. The XY stage 23 is mounted on a guide surface substantially parallel to the XY plane on the wafer base WBS so as to be movable via an air bearing, and is driven at a constant speed in the Y direction by a drive mechanism (not shown) such as a linear motor. At the same time, it is driven to move stepwise in the X and Y directions.

  Laser interferometer 24 (actually, for example, at least one axis in the Y direction and the X direction) irradiates a measurement laser beam on the reflecting surface so as to face the orthogonal reflecting surface (or a movable mirror) of the wafer table 21. Are constructed of two-axis laser interferometers). The laser interferometer 24 uses, for example, a reference mirror (not shown) provided on the side surface of the projection optical system PL as a reference, and at least the position of the wafer table 21 (wafer W) in the X direction and Y direction is 0.5 to 0.1 nm. It measures with a resolution of about, and also measures the rotation angle around the Z axis. The measurement value is supplied to a stage control unit in the main control system 50, and the stage control unit controls the two-dimensional position and speed of wafer stage WST via a drive mechanism (not shown) based on the measurement value.

  An alignment sensor ALG for detecting the position of the alignment mark on the wafer W is installed on the side surface of the projection optical system PL, and a reference mark plate in which slits and reference marks are formed in the vicinity of the wafer W on the wafer table 21. An aerial image measurement system 25 is installed that includes 25a, a lens system 25b that collects the exposure light that has passed through the slit, and a photoelectric sensor 25c that receives the collected light flux. The aerial image measurement system 25 can also measure the position of the alignment mark image on the reticle R. Detection signals from the alignment sensor ALG and the aerial image measurement system 25 are supplied to an alignment control unit in the main control system 50, and the alignment control unit aligns the reticle R and the wafer W based on the detection signals.

  In addition, the AF sensor (60a, 60b) projects a slit image obliquely to a plurality of measurement points on the surface of the wafer W and receives reflected light from the surface to re-form the slit image. This is an oblique incidence type multi-point autofocus sensor composed of a light receiving unit 60b that outputs a detection signal corresponding to the lateral shift amount (and hence the focus position of the measurement point). The detection signal of the light receiving unit 60b is supplied to the wafer side focus calculation unit in the main control system 50, and the wafer side focus calculation unit processes the detection signal to obtain the focus position and tilt angle of the wafer W surface.

  Further, as will be described later, the reticle side focus calculation unit in the main control system 50 includes a plurality of measurement points along the Y direction on the lower surface of the reticle R measured from the RAF control system 52 via the RAF sensor 2. Position information in the Z direction is also supplied. Based on this position information, the reticle-side focus calculation unit calculates the position in the Z direction of the image plane of the reduced image of the pattern on the reticle R by the projection optical system PL, and the tilt angle (or image plane around the X axis and Y axis). (Bending state). At the time of scanning exposure, the focus drive unit in the main control system 50 focuses the surface of the wafer W obtained by the wafer side focus calculation unit on the image plane of the projection optical system PL calculated by the reticle side focus calculation unit. As described above, the Z drive units 22A to 22C are driven by the autofocus method, and the curvature of field or the focus position of the projection optical system PL is controlled via the imaging characteristic control mechanism (drive system 54) as necessary. . That is, the Z driving units 22A to 22C and the imaging characteristic control mechanism (driving system 54) constitute a focusing mechanism (image position correcting mechanism). As a result, even when a slight deflection occurs on the lower surface (pattern forming surface) of the reticle R, the surface of the wafer W is focused on the image surface of the projection optical system PL, and the pattern of the reticle R can be obtained with high resolution. It can be transferred to each shot area of the wafer W.

  At the time of scanning exposure, on the wafer W via the wafer stage WST in synchronization with the movement of the reticle R in the Y direction via the reticle stage RST with respect to the illumination area of the exposure light from the illumination optical system IL. The movement of one shot area in the Y direction with the projection magnification as the speed ratio and the operation of stopping the exposure light irradiation and moving the wafer W stepwise in the X and Y directions via the wafer stage WST Repeated. By such a step-and-scan operation, an image of the pattern of the reticle R is exposed to each shot area on the wafer W.

  Next, in the present embodiment, an RAF sensor (reticle autofocus sensor) 2 for measuring the Z position (deflection etc.) of the lower surface of the reticle R, which causes a focus error of the projection optical system PL, and this detection signal The RAF control system 52 for processing will be described in detail. A detection device that detects the Z position, which is the position in the normal direction of the lower surface (test surface) of the reticle R, is configured from the RAF sensor 2 and the RAF control system 52 (calculation unit). Note that the measurement of the Z position of the reticle R by the RAF sensor 2 and the RAF control system 52 is performed, for example, after exchanging the reticle on the reticle stage RST of FIG. In the subsequent exposure process using the same reticle, autofocus control is performed based on the Z position corresponding to the position in the scanning direction of the reticle that is measured and stored first.

  2A shows the RAF sensor 2A and the RAF control system 52A disposed between the side surface of the upper end portion of the projection optical system PL of FIG. 1 and the reticle base RBS, and FIG. 2B shows the RAF sensor 2A in FIG. 2 is a plan view showing a reticle stage RST of FIG. The RAF sensor 2 in FIG. 1 is actually set along three measurement lines 49A, 49B, and 49C parallel to the Y axis on the pattern forming surface (lower surface) of the reticle R, as shown in FIG. In addition, the RAF sensors 2A, 2B, and 2C having the same configuration are used to sequentially detect the Z position, which is the position in the normal direction of the lower surface, at a large number of measurement points. The measurement lines 49A, 49B, and 49C are set at the end portion in the −X direction, the center portion, and the end portion in the + X direction of the pattern region on the lower surface of the reticle R.

  Similarly, the RAF control system 52 in FIG. 1 includes three RAF control systems 52A, 52B, and 52C (52B and 52C are not shown) having the same configuration corresponding to the RAF sensors 2A to 2C in FIG. 2B. ing. The number of RAF sensors 2A to 2C is arbitrary. Below, the RAF sensor 2A at the end portion in the −X direction and the RAF control system 52A for processing the detection signal thereof will be described.

  In FIG. 2A, the projection optical system PL is supported by a column (not shown) via a flange portion PLF. The RAF sensor 2A irradiates measurement illumination light L1 substantially perpendicularly to the lower surface Ra of the reticle R, and images a plurality of pinholes with respect to the reference surface RFMa (see FIG. 3) on the lower surface of the reticle mark plate RFM. Light transmitting optical system 30A for projecting with different defocus amounts, light receiving optical system 30B for receiving reflected light L2 from the lower surface Ra and forming a plurality of pinhole images, and light for forming these pinhole images And a photoelectric sensor 44 that receives light through a predetermined opening. Further, the RAF sensor 2A has a base monitor optical system 30C in which a plurality of pinhole images are formed by the reflected light L4 branched from the light receiving optical system 30B, and light that forms these pinhole images at a predetermined aperture. And a photoelectric sensor 48 for receiving light via the. In the present embodiment, the reference surface RFMa of the reticle mark plate RFM is substantially perpendicular to the optical axis AX of the projection optical system PL (substantially parallel to the XY plane) and has a high flatness or a flatness with the reference reticle. This is a plane in which the difference is managed, and is used as a predetermined reference plane when measuring the Z position of the lower surface of the reticle R. For example, instead of the reticle R, a lower surface of a glass substrate having a high flatness placed on the reticle stage RST can be used as the predetermined reference surface. The reference reticle is a reticle for which projection exposure is actually performed, the influence of the reticle surface accuracy on the projected image is found, and the value to be corrected by the RAF mechanism is clearly known. Further, the RAF control system 52A controls on / off of the illumination light L1 from the light transmission optical system 30A and processes detection signals from the photoelectric sensors 44 and 48.

  The light transmission optical system 30A includes a light source (not shown) such as a light source device including a light emitting diode or a halogen lamp and a wavelength selection filter, an optical fiber bundle 31 that guides illumination light L1 from the light source, and an optical fiber bundle. The condenser lens 32 that converts the illumination light L1 from the exit end 31a (see FIG. 3) 31 into illumination light having a constant illumination NA, and a large number of pinholes illuminated by the illumination light beam are formed. A first multi-point pinhole plate 33 arranged to be inclined with respect to the axis, a mirror 34 and a lens 35 that respectively reflect and collect the illumination light L1 that has passed through the first multi-point pinhole plate 33, and a lens 35 A beam splitter 36 for branching the illumination light L1 from the light source, and a mirror 37 and an objective lens 38 for reflecting the branched illumination light L1 and condensing them on the lower surface Ra. That. An aperture stop 51 is installed at the pupil-side focal position of the objective lens 38. The numerical aperture NA on the lower surface Ra is determined by the aperture stop. It is desirable that the illumination NAi formed by the condenser lens is sufficiently larger than the NA of the light transmission system. In this embodiment, the NAi of the illumination system is set to incoherent illumination (illumination σ ≧ 1) larger than the NA of the light transmission system. When the illumination σ is large, it is advantageous in securing a light amount. However, even if the illumination σ is small, the measurement itself can be performed in the same manner as when the illumination σ is large.

  The light receiving optical system 30B includes an objective lens 38 and a mirror 37 that collect and reflect the reflected light L2 from the illumination light L1 from the lower surface Ra of the reticle R, and a beam splitter 36 that branches the reflected light L2 from the mirror 37. The mirror 39 and the lens 40 that respectively reflect and collect the reflected light L2 branched by the beam splitter 36 in a direction different from that of the light transmitting optical system 30A, and the reflected light L2 from the lens 40 is branched into reflected light L3 and L4. A beam splitter 41, a second multi-point pinhole plate 42 that is arranged perpendicular to the optical axis on the optical path of the branched reflected light L3 and has a large number of pinholes, and a second multi-point pinhole plate And a relay lens system composed of lenses 43A and 43B that form reduced images of 42 pinholes. The beam splitter 36, the mirror 47, the objective lens 38, and the aperture stop 51 are shared by the light transmitting optical system 30A and the light receiving optical system 30B.

Further, the base monitor optical system 30C is arranged on the optical path of the reflected light L4 branched from the beam splitter 41 of the light receiving optical system 30B so as to be inclined with respect to the optical axis, and a third pinhole is formed. A multi-point pinhole plate 45 and a relay lens system including lenses 47A and 47B that form reduced images of a large number of pinholes on the third multipoint pinhole plate 45 are provided.
In this case, the objective lens 38, the beam splitter 36, and a part of the mirror 39 are housed in the opening 11b provided in the reticle base RBS, and the illumination light L1 and the reflected light L2 are the lower surface Ra of the objective lens 38 and the reticle R. Passes through the opening 11b.

  2B, the objective lenses 38 of the three RAF sensors 2A to 2C are arranged along measurement lines 49A to 49C on the lower surface of the reticle R, respectively. In the present embodiment, as shown in FIG. 2A, the illumination light L1 can be irradiated almost vertically on the lower surface Ra of the reticle R, so that the RAF sensor 2A can be downsized. Therefore, the three RAF sensors 2A to 2C can be compactly arranged between the upper side surface of the projection optical system PL and the reticle base RBS.

  Next, the configuration of the RAF sensor 2A will be described with reference to FIG. 3 in which the RAF sensor 2A of FIG. In FIG. 3, for the sake of clarity, the mirrors 34, 37, 39, and 46 and the aperture stop 51 in FIG. 2A are omitted, and the light transmission optical system 30A and the light reception for the beam splitters 36 and 41 are omitted. The arrangement of the optical system 30B and the base monitor optical system 30C is different from the arrangement shown in FIG.

  In FIG. 3, the optical axis AX1 of the light transmission optical system 30A is parallel to the Z axis. The surface of the first multipoint pinhole plate 33 is inclined at a predetermined angle around an axis parallel to the X axis with respect to a plane perpendicular to the optical axis AX1, and thus a plane substantially parallel to the reference plane RFMa of the reticle mark plate. A number of pinholes A1, A2,..., A9 are formed on the surface of the first multipoint pinhole plate 33 so that the pitch in the Y direction (scanning direction SD of the reticle R) is PY1. The first multipoint pinhole plate 33 forms a large number of pinholes along a straight line parallel to the paper surface of FIG. 3 by etching or the like in a light shielding film (for example, a metal film such as chromium) on the surface of a glass substrate, for example. It is a thing. In FIG. 3, for the sake of clarity, the number of pinholes A1 to A9 is nine, but the number of pinholes A1 to A9 is actually, for example, 256 points or 512 points.

  Further, the images B1 to B9 of the pinholes A1 to A9 of the first multipoint pinhole plate 33 are, for example, 1 near the lower surface Ra of the reticle R by the illumination light L1 from the lens 35 and the objective lens 38 of the light transmission optical system 30A. A reduction imaging optical system that is reduced to about / 2 or less and is formed with a spot interval (pitch) ΔY in the Y direction is configured. At this time, since the first multipoint pinhole plate 33 is tilted with respect to the optical axis AX1, the images B1 to B9 are driven by the reticle stage RST in FIG. The defocus amount is almost linearly different in the Y direction with respect to the reference surface RFMa when the lens is moved to the upper part of the objective lens 38. Note that the defocus amount of the center image B5 is set to 0 with respect to the reference plane RFMa.

  For example, as shown by images B1 and B9, the positions in the Z direction of the images B1 ′ and B9 ′ of the pinholes A1 and A9 by the reflected light L2 from the lower surface Ra of the reticle R are the same as the images B1 and B9 with respect to the lower surface Ra. It becomes a symmetrical position. In the light receiving optical system 30B of the present embodiment, since the reflected light L2 from the lower surface Ra is received, the positions of the images of the pinholes A1 to A9 viewed from the light receiving optical system 30B are symmetrical with the images B1 to B9 with respect to the lower surface Ra. Become position. Then, by using the reflected light L2 from the objective lens 38 and the lens 40 of the light receiving optical system 30B, pinhole images B1 to B9 (actually images B1 ′ to B9 ′) near the lower surface Ra of the reticle R are obtained. An enlargement image forming optical system that forms images C1 to C9 of the pinholes A1 to A9 by enlarging with the reciprocal of the reduction magnification of the light transmission optical system 30A is configured. When the lower surface Ra of the reticle R is substantially parallel to the reference surface RFMa as shown in FIG. 3, the image formation position (best focus position) along the optical axis AX2 of the images C1 to C9 is substantially perpendicular to the optical axis AX2. It is arranged on a straight line inclined around an axis parallel to the X axis with respect to a simple plane.

  In addition, the surface of the second multipoint pinhole plate 42 disposed perpendicular to the optical axis AX2 is disposed at the position of the center image C5 of the images C1 to C9. In the direction perpendicular to the optical axis AX2 (the Z direction in FIG. 3), the pitch is the same as the pitch of the images C1 to C9 and is positioned on a straight line passing through the images C1 to C9 and parallel to the optical axis AX2. Pin holes D1 to D9 are formed. In the present embodiment, the pinholes D1 to D9 are arranged on the same surface as the image (conjugate image) formed by the light receiving optical system 30B of the reference surface RFMa. Similar to the first multipoint pinhole plate 33, the second multipoint pinhole plate 42 is formed by forming a large number of pinholes in the light shielding film on the surface of the glass substrate. The magnitude may be set so that, for example, a detection signal from a photoelectric sensor 44 described later can be sufficiently obtained.

  The reflected light L3 that has passed through the pinholes (openings) D1 to D9 in the light beams forming the images C1 to C9 is pinned on the light receiving surface of the photoelectric sensor 44 via the relay lens system including the lenses 43A and 43B. Hole images E1 to E9 are formed. Corresponding to these images E1 to E9, the photoelectric sensor 44 is provided with light receiving elements F1 to F9 made of silicon photodiodes. Therefore, the photoelectric sensor 44 is a photodiode array. The detection signal FiS is read from the light receiving element Fi (i = 1 to 9) of the photoelectric sensor 44 at a series of timings set according to the speed of the reticle stage RST in FIG. The detected signal FiS is stored in the memory in the RAF control system 52A in correspondence with the Y coordinate and X coordinate of the reticle stage RST (reticle R). The detection signal FiS is calibrated in advance so as to have the same level with respect to the same light quantity.

  The level of the detection signal FiS of the light receiving elements F1 to F9 of the photoelectric sensor 44 increases as the defocus amount with respect to the pinholes D1 to D9 of the images C1 to C9 formed by the light receiving optical system 30B decreases. In the arrangement of FIG. 3, since the second multipoint pinhole plate 42 is perpendicular to the optical axis AX2, the detection signals F1S to F9S are the lower surfaces Ra of the pinhole images B1 to B9 projected from the light transmission optical system 30A. Are on the reference plane RFMa, that is, on the two-dot chain line K1 that divides the straight line connecting the images B1 to B9 and the reference plane RFMa at equal angles with each other on the lower surface Ra. It becomes the maximum level when it is in. Actually, the lower surface Ra is substantially on the straight line K1 in the range including two points of the images B1 to B9, but the lower surface Ra is substantially the straight line K1 in the entire range of the images B1 to B9. The lower the surface Ra is, the more the position is higher. The detection signals F1S to F9S obtained in the state shown in FIG. 3 (the stationary state of the reticle R) have a peak-shaped distribution in which the detection signal F5S corresponding to the central image C5 is the largest and the signals before and after the signal gradually decrease. Based on the detection signal FiS of the photoelectric sensor 44, the defocus amounts of the images C1 to C9, and hence the Z position of the lower surface Ra of the reticle R with respect to the pinhole images B1 to B9 are obtained. This makes it possible to obtain the Z position of a series of measurement points along the Y axis of the lower surface Ra.

In FIG. 3, when the second multipoint pinhole plate 42 is inclined along a two-dot chain line straight line K <b> 2 symmetric with respect to the straight line connecting the images C <b> 1 to C <b> 9, each of the lower surfaces Ra of the reticle R When the point matches the images B1 to B9, the corresponding detection signals F1S to F9S are at the maximum level. In this sense, the second multipoint pinhole plate 42 is not necessarily arranged on the conjugate plane with the reference plane RFMa, and has a predetermined angle (for example, the lower surface Ra is on the reference plane with respect to the conjugate plane with the reference plane RFMa). Or an angle that does not overlap with a straight line connecting the images C1 to C9.
In FIG. 3, the reflected light L4 branched by the beam splitter 41 causes the pinholes A1 to A9 to be reflected in the base monitor optical system 30C in the same manner as the pinhole images C1 to C9 formed by the light receiving optical system 30B. Images G1 to G9 are formed. Images G1 to G9 in FIG. 3 are images when the lower surface RA of the reticle R matches the reference surface RFMa. In this case, the image formation positions along the optical axis AX3 of the images G1 to G9 are arranged on a straight line inclined about an axis parallel to the X axis with respect to a plane substantially perpendicular to the optical axis AX3.

  Further, the surface of the third multipoint pinhole plate 45 is disposed so as to be inclined with respect to the optical axis AX3 so as to contact the positions of the images G1 to G9 when the lower surface Ra matches the reference surface RFMa. Pinholes H1 to H9 are formed on the surface at the same positions as the images G1 to G9. In other words, the third multipoint pinhole plate 45 is conjugated with the first multipoint pinhole plate 33 with respect to the light transmitting optical system 30A, the lower surface Ra (mirror surface) of the reticle R, the light receiving optical system 30B, and the beam splitter 41. Pinholes H1 to H9 are formed at positions that are tilted so as to be conjugated with each other and are conjugated with the pinholes A1 to A9. The third multipoint pinhole plate 45 is formed in the same manner as the second multipoint pinhole plate 42.

  The reflected light L4 that has passed through the pinholes H1 to H9 in the light beams forming the images G1 to G9 passes through the relay lens system including the lenses 47A and 47B, and the pinhole images I1 to I1 on the light receiving surface of the photoelectric sensor 48. I9 is formed. Corresponding to these images I <b> 1 to I <b> 9, the photoelectric sensor 48 is provided with light receiving elements J <b> 1 to J <b> 9 made of silicon photodiodes similarly to the photoelectric sensor 44. When the number of pinholes A1 to A9 of the first multipoint pinhole plate 33 is 512, the pinholes D1 to D9 of the second multipoint pinhole plate 42, the light receiving element Fi of the photoelectric sensor 44, the third The number of pinholes H1 to H9 of the multipoint pinhole plate 45 and the number of light receiving elements Ji (i = 1 to 9) of the photoelectric sensor 48 are also 512. The detection signal JiS is read from the light receiving element Ji of the photoelectric sensor 47 by the RAF control system 52A at the same timing as the photoelectric sensor 44, and the read detection signal JiS is the Y coordinate and X coordinate of the reticle stage RST (reticle R). Are stored in the memory in the RAF control system 52A. The detection signal JiS is also calibrated in advance so as to have the same level with respect to the same amount of light.

  In a state where the lower surface Ra matches the reference surface RFMa, the images G1 to G9 in the base monitor optical system 30C match the pinholes H1 to H9 of the third multipoint pinhole plate 45, so the detection signal JiS is The level is in accordance with the reflectance of the pattern formed on the lower surface Ra of the reticle R. Therefore, by using the detection signal JiS detected through the base monitor optical system 30C, a pattern (base pattern) formed on the lower surface Ra from the detection signal FiS detected through the light receiving optical system 30B as described later. Can be removed.

Next, signal processing in the RAF control system 52A when obtaining the Z position of the lower surface Ra of the reticle R using the detection signal FiS from the photoelectric sensor 44 will be described.
At this time, as exaggeratedly shown in FIG. 4A, it is assumed that the reticle R is bent, and the reticle R is moved through the reticle stage RST in FIG. 2A to the pinhole image B1 in FIG. Suppose that B9 is scanned at a speed V in the Y direction. At this time, the detection signals FiS (i = 1 to 9) output from the light receiving element Fi of the photoelectric sensor 44 in FIG. 3 are converted into the signals (A), (B), (C), (D), (E) The vertical axis of (F), (G), (H), (I) is shown. 6A to 6I, the horizontal axis represents time t from time t1 to time t9. In the following, signal processing related to three measurement points RP1, RP2, and RP3 having different Z positions on the lower surface Ra in FIG. 4A will be described. However, in actuality, the measurement points are imaged in the Y direction on the lower surface Ra. It is set to 1 / M (M is an integer) of the spot interval ΔY (see FIG. 3) of B1 to B9.

In the arrangement of FIG. 3, the detection signal FiS is maximized when each point of the lower surface Ra is actually on the straight line K1 of FIG. 3, but each of the lower surface Ra will be described below for the sake of simplicity. Description will be made assuming that the detection signals F1S to F9S are at the maximum level when the points coincide with the images B1 to B9, respectively.
First, in the state of FIG. 4A (referred to as time t1), the image B1 is formed on the lower surface Ra, and the lower surface Ra is gradually separated from the other images B2 to B9 in the Z direction. As for the signal FiS, the detection signal F1S of FIG. 6 (A) becomes maximum, and the detection signals F2S to F9S of FIGS. 6 (B) to (I) gradually decrease. That is, in the state of FIG. 4A, the image B1 is focused on the lower surface Ra, so that the corresponding pinhole image C1 of FIG. 3 is at the position of the pinhole D1 of the second multipoint pinhole plate 42. As a result, the detection signal F1S of the corresponding light receiving element F1 is maximized.

  Next, when the reticle R moves in the Y direction and the measurement point RP1 reaches the position of the image B1 as shown in FIG. 4B (time t2), the detection signal F1S in FIG. The level slightly decreases in accordance with the amount of shift of the Z position with respect to the image B1 of RP1. Thereafter, as shown in FIG. 4C (time t3), when the measurement points RP1 and RP2 reach the positions of the images B3 and B1, respectively, the detection signal F3S (FIG. 6C) and the detection signal F1S are respectively measured points. The signal corresponds to the shift amount of the Z position of RP1 and RP2. Since the measurement point RP1 matches the image B3, the level of the detection signal F3S is maximum. When the measurement point RP1 reaches the position of the image B2 between FIG. 4B and FIG. 4C, the detection signal F2S of FIG. 6B is the amount of deviation of the Z position of the measurement point RP1. (Same below).

The time required for one measurement point on the lower surface Ra to move the spot interval ΔY of the images B1 to B9 in FIG. 3 is the Z position data capturing time interval Δt shown in FIG. 6B. The time interval Δt is as follows using the velocity V in the Y direction of the reticle stage RST and the spot interval ΔY.
Δt = ΔY / V (1)
Further, the signal capture sampling interval Δts, which is the interval between the sampling points SP corresponding to the series of actual measurement points in FIG. 6A, is as follows using an arbitrary integer M. In FIG. 6A, the integer M is 2, but it is preferable that the integer M is actually set to a value such as 16, 256 that allows easy signal processing.

Δts = Δt / M (2)
Thereafter, as shown in FIG. 4D (time t4), when the measurement points RP1, RP2, RP3 reach the positions of the images B5, B3, B1, respectively, the detection signal F5S (FIG. 6E), the detection signal F3S. , And the detection signal F1S are signals corresponding to the shift amounts of the Z positions of the measurement points RP1, RP2, and RP1, respectively. Subsequently, as shown in FIG. 4E (time t5), when the measurement points RP1, RP2, RP3 reach the positions of the images B7, B5, B3, respectively, the detection signal F7S (FIG. 6G), the detection signal F5S and the detection signal F3S are signals corresponding to the deviation amounts of the Z positions of the measurement points RP1, RP2, and RP3, respectively. Since the measurement point RP2 matches the image B5, the level of the detection signal F5S is maximum.

  Thereafter, when the measurement points RP1, RP2, and RP3 reach the positions of the images B9, B7, and B5 as shown in FIG. 5A (time t6), the detection signal F9S (FIG. 6I) and the detection signal F7S are detected. , And the detection signal F5S are signals corresponding to the shift amounts of the Z positions of the measurement points RP1, RP2, and RP3, respectively. Subsequently, when the measurement points RP2 and RP3 reach the positions of the images B9 and B7, respectively, as shown in FIG. 5B (time t7), the detection signals F9S and F7S are shifted from the Z positions of the measurement points RP2 and RP3, respectively. The signal corresponds to the quantity. Since the measurement point RP3 coincides with the image B7, the level of the detection signal F7S is maximum at the time point t7.

After that, as shown in FIG. 5C (time t8), when the measurement point RP3 reaches the position of the image B9, the detection signal F9S becomes a signal corresponding to the shift amount of the Z position of the measurement point RP3. ) When the measurement point RP3 passes the image 9 as in (time t9), the Z position measurement for the measurement points RP1 to RP3 is completed.
To summarize the above description, as shown in FIGS. 6A to 6I, the detection signals FiS (i = 1 to 9) related to the measurement points RP1, RP2, and RP3 on the lower surface Ra are respectively represented by dotted straight lines 62A, 62A, A set of signals at the time of gradually shifting by the capture time interval Δt of Expression (1) along 62B and 62C. A set of detection signals FiS relating to each measurement point on the lower surface Ra in this way is hereinafter referred to as a confocal waveform. This is a signal obtained in a state in which each measurement point is substantially installed at the positions (positions in the Y direction) of the images B1 to B9 of the pinholes A1 to A9 formed by the light transmission optical system 30A of FIG. Meaning. In this sense, the light transmitting optical system 30A and the light receiving optical system 30B can be referred to as confocal optical systems, respectively.

  The confocal waveform related to the measurement point RP2 at the center of the lower surface Ra in FIG. 4A, that is, the signal obtained by arranging the detection signals F1S to F9S obtained along the dotted line 62B in FIGS. The signal level of the light receiving element F5 at the center of FIG. The horizontal axis in FIG. 7A represents the light receiving element Fi (i = 1 to 9) of the photoelectric sensor 44 in FIG. 3, and the vertical axis represents the detection signal FiS (light quantity) of the light receiving element Fi. Similarly, the detection signals F1S to F9S obtained along the confocal waveforms relating to the left and right measurement points RP1 and RP3 on the lower surface Ra of FIG. 4A, that is, the dotted lines 62A and 62C of FIGS. The signals obtained side by side are dotted line waveforms 63B and 63C in which the signal levels of the left and right light receiving elements F3 and F7 in FIG. 7A are high.

  In addition, since the Z positions of the measurement points RP1, RP2, and RP3 in FIG. 4A are equal to the Z positions of the pinhole images B3, B4, and B7, respectively, the confocal waveform of FIG. By specifying the Z positions of the pinhole images B1 to B9 corresponding to the detection signal FiS indicating the peak position, the Z positions of the measurement points RP1, RP2, and RP3 can be obtained. In this case, information on the image forming positions in the Z direction of the pinhole images B1 to B9 in FIG. 3, for example, information on the positional deviation ΔZi (i = 1 to 9) in the Z direction with respect to the reference plane RFMa is previously stored in the RAF control system. It is stored in the storage unit in 52A. In the RAF control system 52A, the peak position of the confocal waveform (waveforms 63A to 63C, etc.) in FIG. 7A is obtained for each series of measurement points on the lower surface Ra, and a pinhole image corresponding to this peak position is obtained. By obtaining the positional deviation amount ΔZi of B1 to B9 in the Z direction, the Z position of the measurement point (here, the positional deviation amount in the Z direction with respect to the reference plane RFMa) can be obtained. In the arrangement of FIG. 3, ½ of the positional deviation amount ΔZi of the images B1 to B9 corresponding to the peak position is the Z position of the corresponding measurement point. When the second multipoint pinhole plate 42 in FIG. 3 is inclined along the straight line K2, the positional deviation amount ΔZi of the images B1 to B9 relative to the peak position itself is the Z position of the corresponding measurement point. It becomes.

For example, in the case where the peak of the confocal waveform in FIG. 7A is in the middle of the adjacent light receiving elements, interpolation of the positional deviation amounts in the Z direction of the images B1 to B9 corresponding to the adjacent light receiving elements is performed. Thus, the Z position of the measurement point can be obtained more accurately.
When the confocal waveform is used in this way, as shown in FIG. 7B, the waveform 64D having a low signal level relative to the waveform 63A having a high signal level in the portion where the reflectance of the lower surface Ra of the reticle R is low. Even if is obtained, for example, the peak position and the like can be accurately specified. Therefore, even when the reflectance of the test surface gradually changes as a whole, the Z position at a series of measurement points on the test surface can be accurately obtained.

Further, in FIG. 2C, by measuring the Z position via the RAF sensors 2A to 2C at a series of measurement points along the three measurement lines 49A to 49C on the lower surface of the reticle R, the reticle R The state of curvature in the YZ plane and the XZ plane on the lower surface can also be measured. From this measurement result, the curvature of field of the pattern of the reticle R can be obtained.
Further, in the present embodiment, the detection signal JiS obtained from the photoelectric sensor 48 by the base monitor optical system 30C in FIG. 3 is also processed in the same manner as the detection signal FiS obtained from the photoelectric sensor 44, so that FIG. ), A confocal waveform in which the detection signals JiS of the light receiving elements Ji (i = 1 to 9) are arranged at each measurement point on the lower surface of the reticle R is obtained. At this time, the confocal waveform of the detection signal FiS obtained from the photoelectric sensor 44 is shown in FIG. 8A, and the confocal waveform in FIG. 8A is based on the base pattern formed on the lower surface Ra of the reticle R. It is assumed that the peak position is deviated from the original position due to the change in reflectance.

  In this case, the confocal waveform obtained via the base monitor optical system 30C in FIG. 8B corresponds to the reflectance of the base pattern. Therefore, as an example, in the RAF control system 52A, the confocal waveform of the detection signal FiS in FIG. 8A is divided by the confocal waveform of the detection signal JiS in FIG. By obtaining the confocal waveform of the standardized detection signal FiSN, the peak position of the confocal waveform and thus the Z position of the measurement point can be obtained accurately without being affected by the background pattern. Note that other calculations such as subtracting the confocal waveform of the detection signal JiS in FIG. 8B from the confocal waveform of the detection signal FiS in FIG. 8A are also possible.

Further, in order to further reduce the influence of the base pattern, the measurement point on the lower surface Ra of the reticle R is set in the X direction (scanning direction) in addition to the Y direction (scanning direction SD) as shown by the measurement region 64 in FIG. The Z position may be measured by shifting also in the direction NSD orthogonal to the direction.
9A is a view of the main part when FIG. 2A is viewed in the + Y direction, and FIG. 9B is a distribution of a large number of measurement regions 64 set on the lower surface of the reticle R of FIG. 9A. FIG. 9C is an enlarged view showing one measurement region 64 of FIG. 9B.

  As shown in FIG. 9B, the measurement region 64 is set to a width ΔMY in the Y direction along the measurement lines 49A to 49C by the three RAF sensors 2A to 2C, and is set to a width ΔMX in the X direction. ing. In this case, the width ΔMY is, for example, about the width of 512 or more measurement points at the Z position, and the width ΔMX is, for example, about 64 μm. The reticle stage RST (relatively moving mechanism) in FIG. 2A can finely move in the X direction within a predetermined range exceeding the width ΔMX, so that the measurement region 64 moves the reticle stage RST in the X direction with a step amount of about 1 μm, for example. The measurement operation of scanning the entire lower surface of the reticle R with respect to the RAF sensors 2A to 2C via the reticle stage RST after the movement to the position can be set, for example, by repeating 64 times. Since the reticle stage RST can be scanned at a very high speed, 64 scans can also be performed in a very short time.

  As shown in FIG. 6C, in the measurement region 64, along n measurement lines 66-1, 66-2,..., 66-n arranged in the Y direction (n is 64, for example). For example, 512 or more measurement points 65 are set at intervals corresponding to the sampling interval Δts in FIG. 4, and the confocal waveforms in FIGS. 8A and 8B are obtained at each measurement point 65. Therefore, in each measurement region 64, for example, a predetermined statistical process (for example, the confocal of FIG. 8B of the confocal waveform of FIG. By performing division and averaging by the waveform, etc., the influence of the underlying pattern on the lower surface of the reticle R can be reduced.

The effect of this embodiment is as follows.
(1) In the detection apparatus for the Z position of the reticle R including the RAF sensor 2A of FIG. 3 according to the present embodiment, the images B1 to B9 of the pinholes A1 to A9 on the lower surface Ra of the reticle R are different from the reference plane RFMa. A light transmitting optical system 30A formed with a defocus amount, a light receiving optical system 30B that receives reflected light from the lower surface Ra and forms images C1 to C9 of the pinholes A1 to A9, and images C1 to C9 are respectively pinholes. Based on the photoelectric sensor 44 that detects through D1 to D9, the reticle stage RST that relatively scans the lower surface Ra and the images B1 to B9 along the reference plane RFMa, and the detection signal FiS of the photoelectric sensor 44 obtained by the relative scanning. And an RAF control system 52A for obtaining the Z position (position in the normal direction) of the lower surface Ra.

  Therefore, it is possible to obtain the Z position of the lower surface Ra without using a method of driving the optical member, and the optical member does not vibrate, so that measurement can be performed with high accuracy. In addition, since the illumination light can be incident on the lower surface Ra substantially perpendicularly, the optical system can be reduced in size. Furthermore, for example, by performing measurement while shifting the measurement point in the relative scanning direction, the influence of the pattern of the lower surface Ra or the reflectance can be reduced as necessary.

(2) The reticle stage RST relatively scans the lower surface Ra and the images B1 to B9 of the pinholes A1 to A9 in the arrangement direction of the images B1 to B9. Therefore, the confocal waveform shown in FIG. 7A can be easily obtained for each measurement point. For example, even if the reflectance of the lower surface Ra gradually changes, the Z position can be measured with high accuracy.
(3) The reticle stage RST also includes a mechanism for relatively moving the lower surface Ra and the images B1 to B9 of the pinholes A1 to A9 in a direction (including an orthogonal direction) intersecting the arrangement direction of the images B1 to B9. ing. Therefore, by shifting the reticle R in the direction intersecting the arrangement direction, repeating the measurement of the Z position, and statistically processing the measurement results of the adjacent measurement points, the influence of the base pattern can be reduced.

  (4) Further, in the light transmission optical system 30A of FIG. 3, the positions of the first multipoint pinhole plate 33 in which the pinholes A1 to A9 are formed and the image of the pinholes A1 to A9 are relative to the reference plane RFMa. An imaging optical system including a lens 35 and an objective lens 38 that form the images B1 to B9 of the pinholes A1 to A9 on the lower surface Ra so as to be inclined. Accordingly, the images B1 to B9 having different defocus amounts with respect to the reference surface RFMa can be formed (projected) on the lower surface Ra with a simple configuration.

(5) Since the imaging optical system is a reduction optical system, the first multipoint pinhole plate 33 can be easily manufactured, and the optical system near the reticle R can be compactly assembled.
(6) A base monitor optical system 30C that separates part of the reflected light from the lower surface Ra in the light receiving optical system 30B to form the images G1 to G9 of the pinholes A1 to A9, and the images G1 to G9. And the photoelectric sensor 48 that detects through the pinholes H1 to H9 arranged at positions almost conjugate to the images B1 to B9, and the RAF control system 52A uses the detection signal of the photoelectric sensor 48 as the detection signal of the photoelectric sensor 44. It is corrected. Therefore, even if the base pattern is formed on the lower surface Ra, the influence of the fluctuation of the reflectance can be reduced.

In the present embodiment, measurement of the Z position can be repeated by moving the measurement point in the X direction and the Y direction by the reticle stage RST. Therefore, the base monitor optical system 30C and the photoelectric sensor 48 can be omitted when the influence of the base can be sufficiently eliminated only by the action.
(7) In the projection exposure apparatus (exposure apparatus) of the above-described embodiment, the RAF sensor is used in the projection exposure apparatus that exposes the pattern of the reticle R illuminated by the exposure light onto the wafer W via the projection optical system PL. 2A and a RAF control system 52A. The detection device detects position information in the optical axis direction of the projection optical system PL on the lower surface of the reticle R. Therefore, using this position information, even if the reticle R is bent or the like, the exposure surface of the wafer W can be focused with high accuracy on the image of the reticle R pattern. Can be transferred onto the wafer W.

(8) The projection exposure apparatus corrects the positional relationship in the optical axis direction of the projection optical system PL between the pattern image of the reticle R and the wafer W based on the detection information of the detection apparatus. To 22C and an imaging characteristic control mechanism (correction mechanism). Therefore, focusing can be performed easily and with high accuracy.
[Second Embodiment]
Hereinafter, a second embodiment of the present invention will be described with reference to FIGS. In this example as well, the present invention is applied to an autofocus sensor (RAF sensor) for a reticle. In FIGS. 10 and 12, the same or similar reference numerals are given to the portions corresponding to FIGS. The detailed description is omitted or simplified.

  FIG. 10 shows the RAF sensor 2AS and the RAF control system 52AS of the present embodiment. In FIG. 10, the RAF sensor 2AS is arranged in the Y direction on a multi-point slit plate 71 that is arranged inclined with respect to the optical axis AX1. Images b1 to b9 of a large number of slits (apertures) a1 to a9 formed at a pitch PY1 are formed on the lower surface Ra of the reticle R with a different defocus amount with respect to the reference surface RFMa at an interval ΔY in the Y direction (scanning direction SD). The light transmitting optical system 30AS to be formed, the light receiving optical system 30BS that receives the reflected light L2 from the lower surface Ra, and forms the images c1 to c9 of the slits a1 to a9, and the images c1 to c9 are conjugate with the reference surface RFMa. And a photoelectric sensor 72 for detecting through a predetermined opening (in this embodiment, the light receiving surface of the light receiving element itself) on the light receiving surface 72b disposed on the surface.

  Further, the RAF sensor 2AS is coupled to the optical axis AX3 in a conjugate state with the base monitor optical system 30CS that forms the images g1 to g9 of the slits a1 to a9 by the reflected light L4 branched from the light receiving optical system 30BS, and the multipoint slit plate 71. The photoelectric sensor 73 includes a light receiving surface 73b that is inclined with respect to the image sensor and detects the images g1 to g9 through a predetermined opening (in this embodiment, the light receiving surface of the light receiving element itself). Also in FIG. 10, the reticle R is scanned in the Y direction with respect to the images b <b> 1 to b <b> 9 by the same stage (not shown) as the reticle stage RST in FIG. 2A, and is relatively movable in the X direction. . Further, the RAF control system 52AS uses Z coordinates of the stage, X coordinates RY, RX, and detection signals of the photoelectric sensors 72, 73, and Z at a series of measurement points along the Y direction of the lower surface Ra of the reticle R. Measure the position.

  That is, the multipoint slit plate 71 is formed with a number of slits a1 to a9 that are elongated in the direction perpendicular to the paper surface of FIG. 10 (X direction). The number of slits a1 to a9 is actually 256 or 512. The photoelectric sensors 72 and 73 are TDI (Time-Delay Integration) type photoelectric sensors in which a large number of pixels (light receiving elements) are arranged along the arrangement direction of the images c1 to c9 and g1 to g9.

  FIG. 11 is an enlarged view of a main part showing the relationship between the photoelectric sensor 72 for the light receiving optical system 30BS in FIG. 10 and the image c1, etc. As shown in FIG. .. Are arranged at a pitch TDP, and slit images c1, c2,... In FIG. 10 are projected at a pitch SIP. The height of the images c1, c2,... In the best focus state is SIH, and the width is SIW. In the present embodiment, the size of the pixel fi (i = 1, 2,...) Of the photoelectric sensor 72 is set substantially equal to the size of the image cj (j = 1, 2,...) In the best focus state. Yes. Therefore, in the present embodiment, the light receiving surface itself of the pixel fi is used as an opening when detecting the image cj.

On the other hand, the pitch TDP of the array of the pixels fi is set to an integer of the pitch SIP of the image cj. Therefore, the following equation is established using the integer m. The integer m is 4, for example.
TDP = SIP / m (3)
As an example, assuming that the integer m is 4, in FIG. 11, slit images c1, c2, c3,..., Ck are respectively projected onto the pixels f1, f5, f9,. Assuming that the detection signals of the pixels fus (for example, the pixels f2 to f4) other than the pixel f (4k-3) (k = 1, 2,...) Of the photoelectric sensor 72 are used for detecting the Z position of the reticle R. Not used. In this case, if the number of images c1, c2,... Is 256, the number of pixels f1, f2,... Of the photoelectric sensor 72 is 1024, and such a TDI sensor can be easily obtained.

  FIG. 13A shows an example of the detection signals fiS (light quantity) of all the pixels fi when the images c1, c2,... Of FIG. As shown in A), slight detection is performed even in pixels (for example, f2, f4, f6, etc.) between every third pixel f1, f5, f9,... According to the defocus amount of the actual image c1, etc. A signal (light quantity) is detected.

Therefore, when obtaining the confocal waveform corresponding to FIG. 7A, as shown in FIG. 13B, only the detection signal of every third pixel f (4k−3) is used. The Z position of the reticle R can be accurately measured.
Similarly, the photoelectric sensor 73 for the base monitor optical system 30CS in FIG. 10 is also configured by arranging pixels f1, f2,... At a pitch TDP as shown in FIG. , G2,... Are formed with a pitch SIP. Therefore, in the photoelectric sensor 73, the light receiving surface of the pixel fi itself is used as an opening for detecting the image gj, and detection signals for pixels other than the pixel f (4k-3) (k = 1, 2,...) Are used. Is not used for detecting the Z position of the reticle R. Thereby, the influence of the base pattern on the lower surface Ra of the reticle R can be reduced with high accuracy. The other RAF sensor 2AS has the same configuration as the RAF sensor 2A shown in FIG.

  When the integer m in the expression (3) is 4, the RAF control system 52AS in FIG. 10 is configured as shown in FIGS. 12 (A) to (I) corresponding to FIGS. The Z position of the lower surface Ra of the reticle R is determined using only the detection signal f (4k-3) S (k = 1, 2, 3,...) Of every third pixel f (4k-3) in the photoelectric sensor 72. measure. In this case, the measurement points on the lower surface Ra of the reticle R bent as shown in FIGS. 4 (A) to 4 (E) and FIGS. 5 (A) to (D) with respect to the slit images b1 to b9 in FIG. Assume that RP1, RP2, and RP3 move in the Y direction. The RAF control system 52AS obtains the detection signal f (4k−3) S shifted by the time interval Δt along the dotted lines 62A, 62B, and 62C in FIG. A confocal waveform similar to (A) can be obtained. As an example, the Z position of the measurement point can be obtained from the known defocus amount with respect to the reference plane RFMa of the slit images b1 to b9 in FIG. 10 corresponding to the peak position of the confocal waveform.

Further, by obtaining the confocal waveform in the same manner as in FIG. 12 using only the detection signals of every third pixel in the photoelectric sensor 73 of FIG. 10, the influence of the base pattern of the reticle R can be reduced.
According to the present embodiment, the photoelectric sensor 72 includes a plurality of pixels fi (photoelectric conversion elements) arranged along the images c1 to c9 of the slits a1 to a9, and the light receiving surface of the pixel fi is for the photoelectric sensor 72. In addition to the effects of the first embodiment, the RAF sensor 2AS can be further compactly combined.

  Regarding the optical system of the present embodiment, regarding the optical axis alignment of the optical system and the mechanism for adjusting the tilt of each pattern plate, etc., an optical part and a mechanical mechanism are appropriately added to constitute the optical system. Is desirable. As a result, assembly adjustment of the RAF sensors 2A, 2AS and the like can be easily performed. In addition, although the LED and the halogen lamp are used as the light source in the present embodiment, the illumination light L1 from the exit end 31a (see FIG. 3) of the optical fiber bundle 31 described above may be used even with a laser light source or other light sources. It is possible to use an alternative configuration if possible.

In this embodiment, a step-and-scan projection exposure apparatus is described as an example. However, the present invention can also be applied to a step-and-repeat projection exposure apparatus. .
Further, the device manufacturing method of the present embodiment includes an exposure step of exposing the pattern (predetermined pattern) of the reticle R onto the wafer W (photosensitive substrate) using the projection exposure apparatus (exposure apparatus) of the above embodiment. And a developing process for developing the wafer W exposed by the exposure process. The exposure apparatus of this embodiment can transfer the pattern of the reticle R to each shot area of the wafer W with high resolution, so that micro devices (semiconductor elements, image sensors, liquid crystal display elements, thin film magnetic heads, etc.) can be manufactured with high accuracy. can do.

FIG. 14 shows an example of a technique for obtaining a semiconductor device as a micro device by forming a predetermined circuit pattern on a wafer or the like as a photosensitive substrate using the projection exposure apparatus according to the above-described embodiment. This will be described with reference to a flowchart.
First, in step 301 of FIG. 14, a metal film is deposited on one lot of wafers. In the next step 302, a photoresist is applied on the metal film on the wafer. Thereafter, in step 303, using the projection exposure apparatus according to the above-described embodiment, the image of the reticle pattern is sequentially exposed and transferred onto each shot area of the wafer of one lot via the projection optical system. Thereafter, in step 304, the photoresist of the lot of wafers is developed, and in step 305, the resist pattern is used as a mask on the lot of wafers to form a pattern on the reticle. Corresponding circuit patterns are formed in each shot area on each wafer.

Thereafter, a device pattern such as a semiconductor element is manufactured by forming a circuit pattern of an upper layer.
Further, in the projection exposure apparatus according to the above-described embodiment, for example, as shown in the flowchart of FIG. 15, a micro device is formed by forming a predetermined pattern (circuit pattern, electrode pattern, etc.) on a plate (glass substrate). The liquid crystal display element as can also be obtained.

That is, in the pattern forming process 401 in FIG. 15, a so-called photolithography process in which the mask pattern is transferred and exposed to a photosensitive substrate (such as a glass substrate coated with a resist) using the projection exposure apparatus according to the above-described embodiment. Is performed, and a predetermined pattern including a large number of electrodes and the like is formed on the photosensitive substrate.
Next, in the color filter forming step 402, a large number of sets of three dots corresponding to R (red), G (green), and B (blue) are arranged in a matrix, or three of R, G, and B are arranged. A color filter is formed by arranging a plurality of stripe filter sets in the horizontal scanning line direction. In the next cell assembly step 403, for example, liquid crystal is injected between the substrate having the predetermined pattern obtained in the pattern formation step 401 and the color filter obtained in the color filter formation step 402, and the liquid crystal panel. Assemble (liquid crystal cell).

Thereafter, in a module assembling step 404, components such as an electric circuit and a backlight for performing a display operation of the assembled liquid crystal panel (liquid crystal cell) are attached to complete a liquid crystal display element.
In addition, the present invention is not limited to the application to the manufacturing process of a semiconductor device and a liquid crystal display element. For example, the manufacturing process of a plasma display or the like, an imaging element (CCD or the like), a micromachine, MEMS (Microelectromechanical Systems: It can be widely applied to manufacturing processes of various devices such as micro electromechanical systems), thin film magnetic heads, and DNA chips. As described above, the present invention is not limited to the above-described embodiment, and various configurations can be taken without departing from the gist of the present invention.

1 is a partially cutaway view showing a schematic configuration of a projection exposure apparatus of a first embodiment. 2A is a partially cutaway view showing the RAF sensor 2A and the RAF control system 52A in the RAF sensor 2 of FIG. 1, and FIG. 2B is a plan view showing the reticle stage RST of FIG. It is a figure which simplifies and shows the RAF sensor 2A of FIG. (A)-(E) are the figures which show the change of the positional relationship of the images B1-B9 and the reticle R at the time of scanning the reticle R with respect to the image B1-B9 of a pinhole, respectively. (A)-(D) are the figures which show the change of the positional relationship of the images B1-B9 and the reticle R at the time of scanning the reticle R following the state of FIG.4 (E), respectively. (A)-(I) is a figure which shows an example of the detection signals F1S-F9S obtained corresponding to the state of FIG. 4 (A)-(E) and FIG. 5 (A)-(D), respectively. (A) is a figure which shows an example of the confocal waveform obtained from the detection signal of FIG. 6 (A)-(I), (B) is a figure which shows two confocal waveforms in case the reflectance of the lower surface of a reticle differs. It is. (A) is a diagram showing an example of the confocal waveform obtained from the detection signal FiS in FIG. 3, (B) is a diagram showing an example of the confocal waveform obtained from the detection signal JiS in FIG. 3, (C) is FIG. It is a figure which shows the confocal waveform obtained by correct | amending the confocal waveform of (A) with the confocal waveform of FIG. 8 (B). 2A is a side view of the main part showing the reticle R in FIG. 2A, and FIG. 3B is a plan view showing an example of the distribution of a large number of measurement regions set on the lower surface of the reticle R in FIG. (C) is an enlarged view showing one measurement region in FIG. 9 (B). It is a figure which shows RAF sensor 2AS of 2nd Embodiment. It is an enlarged view which shows the relationship between the photoelectric sensor 72 in FIG. 10, image c1, etc. FIG. In 2nd Embodiment, it is a figure which shows an example of the detection signal obtained corresponding to the state of FIG. 4 (A)-(E) and FIG. 5 (A)-(D). (A) is a figure which shows an example of the detection signal of all the pixels of the photoelectric sensor 72 of FIG. 11, (B) is a figure which shows the detection signal of the pixel f (4k-3) of the photoelectric sensor 72. It is a flowchart which shows an example of the manufacturing process of a semiconductor device. It is a flowchart which shows an example of the manufacturing process of a liquid crystal display element.

Explanation of symbols

  R ... reticle, PL ... projection optical system, W ... wafer, RST ... reticle stage, A1-A9, D1-D9 ... pinhole, B1-B9 ... pinhole image, 2A-2C, 2AS ... RAF sensor, 30A ... Light transmitting optical system, 30B: Light receiving optical system, 30C: Base monitor optical system, 33: First multipoint pinhole plate, 42: Second multipoint pinhole plate, 44, 48 ... Photoelectric sensor, 45 ... Third multi Point pinhole plate, 52A, 52AS ... RAF control system, a1-a9 ... slit, b1-b9 ... slit image, 71 ... multi-point slit plate, 72, 73 ... TDI photoelectric sensor

Claims (10)

A position detection device that detects position information in a normal direction of a test surface,
A light transmission system for forming a first image of a plurality of openings on the surface to be measured with a different defocus amount with respect to a predetermined reference surface;
A first light receiving system that receives reflected light from the test surface and forms a second image of the plurality of openings;
A first light-receiving sensor that detects a second image of each of the plurality of openings through an opening having a predetermined positional relationship with respect to a conjugate plane with the reference plane;
A relative scanning mechanism that relatively scans the test surface and the first images of the plurality of openings along the reference surface;
A position in the normal direction of the test surface based on detection information of the first light receiving sensor obtained when the test surface and the first images of the plurality of openings are relatively scanned by the relative scanning mechanism. An arithmetic unit for obtaining information;
A position detection device comprising:   2. The position according to claim 1, wherein the relative scanning mechanism relatively scans the test surface and the first images of the plurality of openings in an arrangement direction of the first images of the plurality of openings. Detection device.   The relative scanning mechanism includes a mechanism that relatively moves the test surface and the first images of the plurality of openings in a direction intersecting an arrangement direction of the first images of the plurality of openings. The position detection device according to claim 2. The light transmission system is
An opening member in which the plurality of openings are formed;
An imaging optical system that forms the first images of the plurality of apertures on the test surface so that the positions of the first images of the plurality of apertures are inclined with respect to the reference plane. The position detection device according to any one of claims 1 to 3, wherein the position detection device is characterized.   The position detection apparatus according to claim 4, wherein the imaging optical system is a reduction optical system. A second light receiving system for separating a part of reflected light from the test surface in the first light receiving system to form a third image of the plurality of openings;
A second light receiving sensor that detects a third image of each of the plurality of apertures through an aperture disposed at a position substantially conjugate with the first image,
6. The position detecting device according to claim 1, wherein the calculation unit corrects detection information of the first light receiving sensor using detection information of the second light receiving sensor. 7. . The first light receiving sensor includes a plurality of photoelectric conversion elements arranged along a second image of the plurality of openings,
7. The position detection device according to claim 1, wherein light receiving surfaces of the plurality of photoelectric conversion elements also serve as openings for the first light receiving sensor. In an exposure apparatus that exposes a mask pattern illuminated by illumination light onto a photosensitive substrate via a projection optical system,
A position detection device according to any one of claims 1 to 7, comprising:
The exposure apparatus is characterized in that the position detection device detects position information of the mask in the optical axis direction of the projection optical system.   9. A correction mechanism for correcting a positional relationship in an optical axis direction of the projection optical system between the image of the mask pattern and the photosensitive substrate based on detection information of the position detection device. The exposure apparatus described in 1. An exposure step of exposing a predetermined pattern on a photosensitive substrate using the exposure apparatus according to claim 8 or 9,
A developing step of developing the photosensitive substrate exposed by the exposing step;
A device manufacturing method comprising:

Images