JP2013106014A - Deformable reflective optical element, driving method thereof, optical system, and exposure equipment - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to deform a reflective surface of a reflection member into a various shapes with a small force by using a mechanism easy to install.SOLUTION: A deformable mirror 50, which reflects illumination light IL, comprises: a mirror 22 composed of a concave mirror which reflects illumination light IL; a convex mirror post 24 disposed on a rear face 22e of a reflective surface 22d of the mirror 22; a load supply system 58 which displaces a tip of the mirror post 24 in a direction crossing the direction in which the mirror post 24 extends; and a position sensor 60 which measures a displacement of the tip of the mirror post 24.

Description

  The present invention relates to a reflective optical element whose reflective surface can be deformed, a method for driving the reflective optical element, an optical system having the reflective optical element, and an exposure apparatus including the optical system. The present invention further relates to a device manufacturing method using the exposure apparatus.

  For example, in a lithography process, which is one of the manufacturing processes of a semiconductor device, a pattern formed on a reticle (or a photomask or the like) or a pattern generated by a spatial light modulator or the like is photo-transmitted via a projection optical system. A batch exposure type exposure apparatus such as a stepper or a scanning exposure type exposure apparatus such as a scanning stepper is used to transfer the resist onto the surface of a wafer (or glass plate or the like) coated with a resist.

  The projection optical system mounted on these exposure apparatuses is assembled and adjusted so that various aberrations are within a predetermined allowable range. At this time, for example, even if rotational aberrations such as distortion aberration and magnification error and low-order aberration components remain, these aberrations are not affected by normal imaging characteristic correction mechanisms (for example, mounted on the projection optical system) This can be corrected by a mechanism that controls the position and tilt angle of the predetermined lens and / or mirror in the optical axis direction. On the other hand, when a non-rotationally symmetric aberration component such as astigmatism on the optical axis (hereinafter referred to as “center astigmatism”) remains, The correction is difficult.

  Therefore, in order to correct such a non-rotationally symmetric aberration component, the shape of the reflecting surface of the mirror is deformed by deforming the convex portion on the side surface of the predetermined mirror in the projection optical system by a small amount with an actuator. Such a correction mechanism has been proposed (see, for example, Patent Document 1). According to this correction mechanism, the reflecting surface can be deformed with a relatively small force.

JP 2007-266511 A

The conventional non-rotationally symmetric aberration component correction mechanism mechanically deforms the convex portion on the side surface of the mirror in the projection optical system, and therefore there is a relationship between the deformation amount of the convex portion and the deformation amount of the reflecting surface of the mirror. It was a complicated relationship. Therefore, it is difficult to deform the reflecting surface into a target shape with high accuracy, and it is difficult to deform the reflecting surface into various complex shapes.
In order to control the shape of the reflecting surface of the mirror to various shapes, it is preferable to increase the number of convex portions on the side surface of the mirror. However, due to the structure of the lens barrel of the projection optical system, it has been difficult to add a deformation mechanism to each of the many convex portions on the side surface of the mirror.

  In view of such circumstances, an aspect of the present invention has an object of enabling a reflective surface of a reflective optical element to be deformed into various shapes using a mechanism that is easy to install.

  According to the first aspect of the present invention, a deformable reflective optical element is provided. The reflective optical element includes a reflective member that reflects light, a convex member provided on the back surface of the reflective surface of the reflective member, and a direction in which the convex member intersects the direction in which the convex member extends. And a displacement mechanism for displacing.

According to the second aspect, an optical system including a plurality of optical elements is provided. This optical system includes the deformable reflective optical element of the present invention, and a control device that corrects aberration by displacing the convex member via the displacement mechanism of the reflective optical element.
Further, according to the third aspect, in the exposure apparatus that illuminates the pattern with the exposure light and exposes the substrate with the exposure light through the pattern and the projection optical system, the exposure apparatus including the optical system of the present invention is provided. Is done.

  Moreover, according to the 4th aspect, the drive method of a reflective optical element is provided. In this driving method, the reflective optical element is constituted by a reflective optical element including a combination of the plurality of convex members and the displacement mechanism of the present invention, and at least one of the plurality of convex members is used. One convex member is displaced through the displacement mechanism corresponding to the convex member, and the first step of deforming the reflective surface of the reflective member and the displacement of the at least one convex member are generated. Among the plurality of convex members, a convex member different from the at least one convex member corresponds to the convex member so as to correct the force causing the position change of the reflecting member. And a second step of displacing via the displacement mechanism.

  According to a fifth aspect, there is provided a device manufacturing method comprising: forming a pattern of a photosensitive layer on a substrate using the exposure apparatus of the present invention; and processing the substrate on which the pattern is formed. Is provided.

  According to the aspect of the present invention, a mechanism that is easy to install is used by displacing the convex member provided on the back surface of the reflective surface of the reflective member in a direction intersecting the direction in which the convex member extends. The reflecting surface of the reflecting member can be deformed into various shapes with a small force.

It is sectional drawing which shows the main-body part of the exposure apparatus of an example of embodiment. It is sectional drawing which shows the deformable mirror 50 in FIG. It is sectional drawing which follows the AA line of FIG. 3A is a cross-sectional view showing an X-axis actuator 59X and a bias portion 63X in FIG. 3, and FIG. 4B is a cross-sectional view showing a modification of the method for supporting the mirror post 24. FIG. (A) is a diagram showing a first example of the input load on the back surface of the mirror, (B) is a diagram showing the force acting on the hold block of the mirror, (C) is a diagram showing the distribution of deformation of the mirror reflecting surface, (D) is a diagram showing the distribution of deformation after correcting the refractive power and the tilt angle. (A) is a diagram showing a second example of the input load on the back surface of the mirror, (B) is a diagram showing the force acting on the mirror hold block, (C) is a diagram showing the distribution of the deformation amount of the mirror reflection surface, (D) is a diagram showing the distribution of deformation after correcting the refractive power and the tilt angle. (A) is a diagram showing a third example of the input load on the back surface of the mirror, (B) is a diagram showing the force acting on the mirror hold block, (C) is a diagram showing the distribution of deformation of the mirror reflecting surface, (D) is a diagram showing the distribution of deformation after correcting the refractive power and the tilt angle. It is a figure which shows an example of the error in the case of expressing the deformation amount of a reflective surface with a Zernike coefficient. (A) is a figure which shows an example of the some load point of a mirror back surface, (B) is a figure which shows the state which applied the unit load to a certain load point, (C) is a figure which shows the 1st set of load sets. . (A) is a figure which shows a 1st set of load sets and the deformation amount of the mirror reflective surface corresponding to this, (B) is a figure which shows several sets of load sets. It is explanatory drawing of the method of obtaining the deformation | transformation amount of arbitrary reflective surfaces from the combination of several sets of load sets. (A) is a flowchart which shows an example of operation | movement of the preparation process of an exposure process, (B) is a flowchart which shows an example of an exposure process. (A) is sectional drawing which shows the actuator and bias part of a modification, (B) is sectional drawing which shows the support method of the mirror post of a modification. It is a flowchart which shows an example of the manufacturing process of a semiconductor device.

Hereinafter, an exemplary embodiment of the present invention will be described with reference to FIGS.
FIG. 1 shows an exposure main body of the exposure apparatus EX of the present embodiment. The exposure apparatus EX is, for example, a scanning exposure type projection exposure apparatus composed of a scanning stepper. In FIG. 1, an exposure apparatus EX includes an exposure light source (not shown) that generates exposure illumination light IL (exposure light) and an illumination optical system ILS that illuminates a reticle R (mask) with the illumination light IL (in FIG. 1). And a projection optical system PL that forms an image of the pattern of the reticle R on the surface of the wafer W (substrate). Further, the exposure apparatus EX includes a main control system 5 including a reticle stage 15 that holds and moves the reticle R, a wafer stage 32 that holds and moves the wafer W, and a computer that controls the overall operation of the apparatus. (See FIG. 2).

  Hereinafter, a plane that takes the Z axis parallel to the optical axis AX1 of the partial optical system on the reticle R side of the projection optical system PL (first imaging optical system G1 described later) and is perpendicular to the Z axis (in the present embodiment, substantially horizontal) ), The Y axis is parallel to the paper surface of FIG. 1, and the X axis is perpendicular to the paper surface of FIG. The scanning direction of reticle R and wafer W during scanning exposure is a direction (Y direction) parallel to the Y axis, and the pattern surface of reticle R and the surface of wafer W are substantially parallel to the XY plane. The rotation directions (inclination directions) around the axes parallel to the X axis, the Y axis, and the Z axis are also referred to as the θx direction, the θy direction, and the θz direction.

  First, the exposure main body including the projection optical system PL, the reticle stage 15 and the wafer stage 32 is supported by a frame mechanism. The frame mechanism includes a base member 1 made of a frame caster installed on the floor surface, for example, three (or four, etc.) first columns 2 installed on the upper surface of the base member 1, and the first of these. A second column 4 is provided on the upper surface of the column 2 via, for example, active vibration isolators 3A and 3B (actually three or four are disposed). The projection optical system PL is mounted on a U-shaped opening at the center of a flat support plate 4a provided at the bottom of the second column 4.

  An exposure light source (not shown) installed in the vicinity of the frame mechanism is an ArF excimer laser light source (oscillation wavelength 193 nm). In addition, a KrF excimer laser light source (wavelength 248 nm), or a solid-state laser (YAG laser or semiconductor laser) Etc.) can be used. The illumination light IL emitted from the exposure light source enters the illumination optical system ILS. The illumination optical system ILS illuminates a slit-like illumination area elongated in the X direction (non-scanning direction) of the pattern surface (lower surface) of the reticle R with illumination light IL with a substantially uniform illuminance distribution.

  The illumination optical system ILS disposed in the sub-chamber 14 includes a light amount distribution setting mechanism (including a spatial light modulator or a diffractive optical element) as disclosed in, for example, US Patent Application Publication No. 2003/025890. Illumination uniformity optical system (not shown) including an optical integrator (such as a fly-eye lens or rod integrator), a variable field stop (not shown) such as a reticle blind, and lenses 11, 13 and a mirror 12 are included. Condenser optics etc. are included. Further, according to the illumination conditions such as normal illumination, dipole illumination, quadrupole illumination, or annular illumination, the light amount distribution setting mechanism has a light amount of illumination light IL on a pupil plane (not shown) in the illumination optical system ILS. The distribution is switched to a distribution having a large light amount in a circular area centered on the optical axis, two areas sandwiching the optical axis, four areas sandwiching the optical axis, or a ring-shaped area.

  The illumination light IL that has passed through the reticle R passes through the projection optical system PL, and the X of one shot region on the surface of a wafer (semiconductor wafer) W that is a disk-shaped substrate coated with a photoresist (photosensitive agent). An image obtained by reducing the pattern in the illumination area of the reticle R with a projection magnification β (for example, 1/4, 1/5, etc.) is formed in the exposure area that is elongated in the direction. The projection optical system PL of the present embodiment is a catadioptric optical system. The projection optical system PL is placed on the support plate portion 4a by the flange portion 44a. The optical paths of the illumination light IL of the illumination optical system ILS and the projection optical system PL are almost hermetically sealed, and in these optical paths, a gas having a high transmittance with respect to light in a substantially vacuum ultraviolet region (hereinafter referred to as “purge gas”). The dry air, nitrogen, or rare gas (such as helium) is supplied through the supply pipe 20A and the exhaust pipes 21A and 21D.

  The reticle R is held on the upper surface of the reticle stage 15 via a reticle holder (not shown), and the reticle stage 15 is movable on the upper surface parallel to the XY plane of the reticle base 16 at a constant speed in the Y direction. And it is mounted in a state that can be displaced in the X direction, the Y direction, and the θz direction. The reticle base 16 is fixed to the upper end of the second column 4. At least the position of the reticle stage 15 in the X and Y directions and the rotation angle in the θz direction are measured by a laser interferometer 17, and based on this measured value and control information from the main control system 5, a drive including a linear motor and the like. An apparatus (not shown) drives the reticle stage 15.

  On the other hand, the wafer W is held on the upper surface of the wafer table 31 via a wafer holder (not shown), and the wafer table 31 is fixed to the upper surface of the wafer stage 32. The wafer stage 32 is mounted on an upper surface parallel to the XY plane of the wafer base 33 so as to be movable at a constant speed in the Y direction and to be movable in steps in the X and Y directions. The wafer base 33 is placed on the base member 1 via active vibration isolators 38A and 38B. At least the position in the X direction and the Y direction of the wafer stage 32 and the rotation angle in the θz direction are measured by a laser interferometer 34, and a drive including a linear motor or the like is based on the measured value and control information from the main control system 5. An apparatus (not shown) drives the wafer stage 32.

  In addition, a focus / leveling mechanism for adjusting the Z-direction position (focus position) of the wafer table 31 (wafer W) and the inclination angles in the θx direction and the θy direction is incorporated in the wafer stage 32. Yes. A plurality of wafers W measured by an oblique incidence type multi-point focus position detection system (autofocus sensor) composed of a projection optical system 35A and a light receiving optical system 35B disposed on the lower side surface of the projection optical system PL. Based on the information of the focus position at the measurement point, the focus / leveling mechanism continuously focuses the surface of the wafer W on the image plane of the projection optical system PL during exposure. The projection optical system 35A and the light receiving optical system 35B are attached to a sensor column 36 attached to the bottom surface of the flange portion 44a of the projection optical system PL.

  In addition, a shearing interference type or point / diffraction / interference type (PDI type) wavefront aberration measuring device 39 disclosed in, for example, US Pat. No. 6,573,997 is provided above the wafer table 31. It has been. Information on the wavefront aberration of the projection optical system PL measured by the wavefront aberration measuring device 39 is supplied to the imaging characteristic control system 6 of FIG. The imaging characteristic control system 6 sequentially predicts the amount of fluctuation of the wavefront aberration of the projection optical system PL based on, for example, the integrated energy of the illumination light IL during normal exposure.

  At the time of exposure, after aligning the reticle R and the wafer W using an alignment system (not shown), the wafer stage 32 is stepped in the X direction and the Y direction so that the shot area to be exposed on the wafer W becomes the exposure area. Move to the front. Thereafter, the reticle R and the wafer W are projected in the Y direction via the reticle stage 15 and the wafer stage 32 while exposing the shot area of the wafer W with an image of the projection optical system PL of the pattern in the illumination area of the reticle R. Scanning exposure is performed in which the projection magnification of the optical system PL is synchronized with the speed ratio. By repeating the step movement and scanning exposure by the step-and-scan method, the image of the pattern of the reticle R is exposed on the entire shot area of the wafer W.

  Next, the configuration of the projection optical system PL of the present embodiment will be described in detail. In FIG. 1, a projection optical system PL composed of a catadioptric optical system includes a refractive first imaging optical system G1 that forms a first intermediate image of a reticle R pattern, a mirror 22 composed of a concave mirror, and two negative refractions. A second imaging optical system G2 that is composed of force lenses L8 and L9 and forms a second intermediate image that is approximately the same size as the first intermediate image; and light from the second intermediate image on the wafer W. A refractive third imaging optical system G3 that forms a final image of the pattern of the reticle R and a deformable mirror 50 are provided. The deformable mirror 50 includes the mirror 22 and can control the shape of the reflecting surface of the mirror 22. Further, the projection optical system PL has a reflection surface A that deflects the light from the first imaging optical system G1 toward the second imaging optical system G2, and the light from the second imaging optical system G2. An optical path bending mirror FM on which a reflecting surface B deflected toward the image optical system G3 is formed is provided. The first intermediate image and the second intermediate image are formed between the reflecting surface A and the second imaging optical system G2 and between the second imaging optical system G2 and the reflecting surface B, respectively.

The first imaging optical system G1 and the third imaging optical system G3 have an optical axis AX1 parallel to the Z axis, and the optical axis AX2 of the second imaging optical system G2 is orthogonal to the optical axis AX1. And parallel to the Y axis. Further, the optical axis AX1 and the optical axis AX2 intersect at an intersection line C (strictly speaking, an intersection line of the virtual extension surfaces) C of the two reflection surfaces A and B of the optical path bending mirror FM.
The first imaging optical system G1 includes a plane parallel plate L1, lenses L2, L3, L4, L5, L6, and L7 arranged in this order from the reticle R side. The second imaging optical system G2 is configured by disposing negative lenses L8 and L9 and a mirror 22 in order from the reticle side (that is, the incident side) along the light traveling path. The third imaging optical system G3 is configured by arranging lenses L10 and L11, an aperture stop AS, and lenses L12 and L13 in order from the reticle side along the light traveling direction. The arrangement surface of the aperture stop AS is the pupil surface of the projection optical system PL or a surface in the vicinity thereof, and the reflection surface 22d of the mirror 22 is disposed substantially in the vicinity of the pupil surface of the second imaging optical system G2. That is, the reflecting surface 22d of the mirror 22 is substantially conjugate with the pupil surface of the projection optical system PL. In addition, a parallel plane plate for correcting imaging characteristics is arranged in the vicinity of the pupil planes of the first imaging optical system G1 and the second imaging optical system G2 (a plane conjugate with the pupil plane of the projection optical system PL). Also good. The configuration of the projection optical system PL is arbitrary.

In this embodiment, synthetic quartz or fluorite (CaF ₂ crystal) is used as the optical material of all refractive optical elements (lens components) constituting the projection optical system PL. Moreover, the optical path bending mirror FM and the mirror 22 are formed by depositing a metal film such as aluminum or a dielectric multilayer film on the reflective surface of a composite material of silicon carbide (SiC) or SiC and silicon (Si) as an example. It is formed. At this time, it is preferable to coat the entire mirror 22 with silicon carbide or the like in order to prevent degassing. Further, as the material of the mirror 22, a low expansion material such as Corning's ULE (Ultra Low Expansion: trade name) or beryllium (Be) may be used. When beryllium is used, it is preferable to coat the entire mirror 22 with silicon carbide or the like.

  In addition, the plane parallel plate L1 and the lenses L2 to L7 of the first imaging optical system G1 are respectively divided into cylindrical divided lens barrels 41A through annular lens frames 42A, 42B, 42C, 42D, 42E, 42F, and 42G. , 41B, 41C, 41D, 41E, 41F, and 41G, and the divided lens barrels 41A to 41G have, for example, bolts (not shown) facing flange portions (not shown) in a state of maintaining airtightness along the optical axis AX1. Are fixedly connected to each other. The lens frames 42B to 42G and the like are formed with a plurality of openings for allowing the purge gas to flow (the same applies hereinafter).

  Similarly, the lenses L10, L11, L12, and L13 of the third imaging optical system G3 are respectively divided into cylindrical divided lens barrels 41H, 44, 41I, and 41J via annular lens frames 42H, 42K, 42I, and 42J. Is held in. The aperture stop AS is held in the divided lens barrel 41K sandwiched between the divided lens barrels 44 and 41I, and the divided lens barrels 41H, 44, 41K, 41I, and 41J are connected in a state of maintaining airtightness. The split lens barrel 44 is provided with a flange portion 44a. A cylindrical split lens barrel 43 having an opening in the + Y direction is connected between the split lens barrels 41G and 41H, and an optical path bending mirror FM is fixed to a protrusion in the split lens barrel 43 via a holding frame 43a. Yes. The first partial barrel 7 is constituted by the divided barrels 41A to 41K, 43, and 44.

  Further, under the control of the imaging characteristic control system 6 in FIG. 2, for example, the lens frames 42A to 42E are driven to finely move the plane parallel plate L1 and the lenses L2 to L5 in the Z direction, θx direction, and θy direction. Accordingly, a rotationally symmetric imaging characteristic correction mechanism (not shown) for correcting relatively low-order aberrations with rotational symmetry such as distortion and coma aberration of the projection optical system PL is provided. As such a rotationally symmetrical imaging characteristic correction mechanism, for example, a mechanism disclosed in US Patent Application Publication No. 2006/244940 can be used.

  The lenses L8 and L9 of the second imaging optical system G2 are held in the cylindrical divided lens barrels 45 and 41L via the holding frames 46A and 46B, respectively. The deformable mirror 50 including the mirror 22 is The divided lens barrel 41L is connected and held. The deformable mirror 50 is also a correction mechanism for correcting wavefront aberration including non-rotationally symmetric aberrations such as center ass and higher order aberrations as the imaging characteristics of the projection optical system PL. In this embodiment, the wavefront aberration is expressed by a Zernike polynomial coefficient (hereinafter referred to as Zernike coefficient Zi). The imaging characteristic control system 6 shown in FIG. 2 has a wavefront aberration (for example, fifth order) that excludes rotationally symmetrical and relatively low-order aberrations from the predicted fluctuation amount of the imaging characteristic of the projection optical system PL during normal exposure. Information on aberrations represented by the ˜81st order Zernike coefficient Zi) is supplied to the aberration control system 9 of FIG. In the aberration control system 9, the reflecting surface of the mirror 22 of the deformable mirror 50 is deformed so as to correct the wavefront aberration. A storage device 10 in which information for driving the deformable mirror 50 is stored is connected to the aberration control system 9.

  The deformable mirror 50 is connected to the divided lens barrel 41L, and is divided so as to cover the cylindrical divided lens tube 47 in which the mirror 22 is housed via the hold block 23 and the + Y direction opening of the divided lens tube 47. A partition plate 53 fixed to the end surface of the lens barrel 47, a drive system 54 provided on the outer surface 53a of the partition plate 53 to deform the reflection surface of the mirror 22, and a partition plate 53 fixed to cover the drive system 54. Housing 55. The drive system 54 applies loads in the X direction and the Z direction to the tips of the plurality of rod-shaped mirror posts 24 connected to the back surface 22e of the mirror 22, respectively. In the present embodiment, since the tip is displaced in the X and Z directions in proportion to the load in the X and Z directions applied to the tip of the mirror post 24, the load applied to the tip is displaced. Is equivalent to

  As shown in FIG. 2, the partition plate 53 is fixed to the divided lens barrel 47 through a plurality of bolts 40A, and the housing 55 is fixed to the partition plate 53 through a plurality of bolts 40B. Further, as shown in FIG. 3 which is a cross-sectional view taken along the line AA in FIG. 2, three convex holding portions 22a, 22b and 22c are formed at equal angular intervals on the side surface of the mirror 22 whose outer shape is substantially cylindrical. The holding portions 22a to 22c are held by hold blocks 23A, 23B, and 23C fixed to the inner surface of the divided lens barrel 47, respectively. The mirror 22 is stably and kinematically supported by the three hold blocks 23A to 23C. The hold blocks 23A to 23C are typically represented by one hold block 23 in FIG.

Further, a plurality of convex portions 25 are formed integrally with the mirror 22 on a surface (back surface) 22e opposite to the reflecting surface 22d of the mirror 22, and extend substantially in the Y direction at the tip portions of the plurality of convex portions 25, respectively. The connecting portions 24a and 24b of the rod-shaped mirror post 24 are connected.
Note that the convex portion 25 may not be formed integrally with the mirror 22 but may be screwed to the back surface of the mirror 22 or fixed with an adhesive or the like.
The columnar tip portions 24e of the plurality of mirror posts 24 protrude to the outer surface 53a side of the partition plate 53 through openings 53b provided in the partition plate 53, respectively. The mirror post 24 is made of metal as an example, and the mirror post 24 is integrally formed from the front end portion 24e to the connecting portions 24a and 24b. Two hinge portions 24c and 24d (narrow in diameter for easily displacing the tip portion 24e by elastic deformation in the X direction and Z direction between the tip portion 24e of the mirror 22 and the connecting portions 24a and 24b ( 4) is formed. As a result, the displacement of the tip 24e increases when a load is applied to the tip 24e in the X direction and the Z direction. Therefore, as described later, the displacement of the tip 24e is measured by the position sensor 60, and the load is calculated from this displacement. The measurement accuracy when obtaining is improved. For example, when measuring the displacement of the distal end portion 24e with higher resolution (accuracy), the hinge portions 24c and 24d are not necessarily provided.

  Further, in the vicinity of each opening 53b of the outer surface 53a of the partition plate 53, an X axis is applied to the tip 24e of the mirror post 24 in the X and Z directions by applying a load in the X and Z directions. And Z axis actuators 59X, 59Z, a load applying system 58, and X axis and Z axis bias portions 63X for applying forces (preload or bias force) in the X direction and Z direction to the tip 24e of the mirror post 24, A bias system 62 made of 63Z is fixed. A drive system 54 is configured including a plurality of load applying systems 58 and bias systems 62 installed corresponding to the plurality of mirror posts 24. In this embodiment, when a load in the X direction and the Z direction is applied to the tip 24e of the mirror post 24, a large stress acts on the corresponding convex portion 25 of the mirror 22 due to the lever principle, and this stress causes the mirror 22 to The reflecting surface 22d is deformed as indicated by a dotted line B2. At this time, information on the conversion relationship between the load applied to the tip portions 24e of the plurality of mirror posts 24 (or displacement of the tip portions 24e) and the deformation amount of the reflecting surface 22d (for example, expressed by wavefront aberration) is obtained. This information is stored in the storage device 10. In addition, the height from the back surface 22e of the mirror 22 to the front-end | tip part 24e of the mirror post 24 is several 10 mm as an example.

  Further, a two-dimensional lattice pattern 61 is formed on the end face of the tip portion 24e of each mirror post 24, and the lattice pattern 61 in the X direction and Z direction is respectively formed on the inner surface of the housing 55 at a position corresponding to each tip portion 24e. A position sensor 60 for measuring the displacement is fixed. The position sensor 60 includes, for example, a reference member 60a (see FIG. 4A) on which a pattern corresponding to the lattice pattern 61 is formed, and detects the displacement of the lattice pattern 61 relative to the reference member 60a using detection light. It is an optical sensor. However, as the position sensor 60, another type of sensor such as a capacitance type or a magnetic type can be used. Detection signals from the plurality of position sensors 60 are supplied to a detection signal processing system 57. In the detection signal processing system 57, predetermined reference positions in the X direction and Z direction of the tip 24e of the j-th (j = 1, 2,...) Mirror post 24 detected by the position sensor 60 (for example, in the movable range). From the displacements ΔXMj and ΔZMj from the center position), the loads fXMj and fZMj in the X and Z directions applied to the tip 24e are obtained, and the information on the loads thus obtained is supplied to the actuator control system 56.

  The aberration control system 9 is the j-th component based on the wavefront aberration to be corrected (information on the target deformation amount of the reflecting surface of the mirror 22) supplied from the imaging characteristic control system 6 and the information on the conversion relation in the storage device 10. Target loads fXTj and fZTj in the X direction and Z direction of the tip 24e of the mirror post 24 are calculated, and information on the calculated target loads fXTj and fZTj is supplied to the actuator control system 56. The actuator control system 56 has a corresponding load application system so that the measured loads fXMj and fZMj of the tip 24e of the j-th mirror post 24 supplied from the detection signal processing system 57 become the target loads fXTj and fZTj. 58 operations are controlled. Assuming that the number of mirror posts 24 is n (n is an integer of 2 or more), the degree of freedom of displacement (direction in which a load is applied) of the tip 24e of each mirror post 24 is 2, and therefore the mirror 22 as a whole. The degree of freedom of deformation of the reflecting surface is 2n.

  As shown in FIG. 3, the plurality of mirror posts 24 are arranged at, for example, 16 locations (n = 16) centered on the position on the optical axis of the back surface 22 e of the mirror 22. Hereinafter, the center of the tip 24e when the tip 24e of each mirror post 24 is at the reference position is referred to as a load point Pj (j = 1, 2,..., 16). The load point on the optical axis is P8. As an example, the load points P2, P8 are arranged on a straight line parallel to the Z axis, and the load points P6, P7, P8, P9, P10 are arranged on a straight line C3 parallel to the X axis, and the load points P1, P4, P8, P13, and P16 are arranged on a straight line C1 parallel to a straight line rotated 30 ° clockwise with respect to the Z axis. The load points P3, P5, P8, P12, and P15 are arranged on a straight line C2 that is parallel to a straight line rotated 30 ° counterclockwise with respect to the Z axis, and the straight line connecting the load points P8 and P11 is the X axis. Is a straight line rotated 30 ° clockwise with respect to a straight line parallel to. Further, the load points P3, P5, P9, P10, P13, P14, and P16 are symmetrical with the load points P1, P4, P7, P6, P12, P11, and P15 with respect to a straight line connecting the load points P2 and P8. The number n of mirror posts 24 (number n of load points Pi) and the arrangement thereof are arbitrary.

  Here, the configuration of one mirror post 24 in FIG. 2 and the corresponding X-axis actuator 59X and bias unit 63X will be described with reference to FIG. Note that the Z-axis actuator 59Z and the bias part 63Z are configured by rotating the X-axis actuator 59X and the bias part 63X by 90 ° around the mirror post 24, respectively, and thus the description thereof is omitted.

  4A, a square prismatic convex portion 25 is formed on the back surface 22e of the mirror 22, and connecting portions 24a and 24b that sandwich the convex portion 25 in the X direction and the Z direction are formed at the lower end of the mirror post 24. Has been. The connecting portions 24a and 24b are provided with hinge portions 24a1 and the like by notches, and the tips of the connecting portions 24a and 24b are opened to the outer surface of the convex portion 25 with the holding tool (not shown) open. By closing the tip, the mirror post 24 can be connected to the convex portion 25 via the connecting portions 24a and 24b.

  The actuator 59X is coupled to a base member 66 fixed to the partition plate 53, and to a side surface of the base member 66 via a hinge portion 68a so as to be rotatable around an axis parallel to the Z axis within a predetermined angle range. And an L-shaped displacement converter 68. As an example, the displacement converter 68 and the base member 66 are made of metal and are integrally formed. In addition, you may connect the displacement conversion part 68 to the base member 66 via a rotating bearing. Further, the actuator 59X is fixed to the base member 66, and a driving element 67 such as a piezoelectric element or an ultrasonic motor that displaces one end 68b of the displacement converting portion 68 in the Y direction, and the displacement converting portion 68 in the X direction. A rod-like connecting member 69 that connects the other displaceable end portion 68c and the tip end portion 24e of the mirror post 24; As an example, the metal connecting member 69 is coupled to the distal end portion 24e by a screw portion, and the end portion 68c of the displacement converting portion 68 and the coupling member 69 are coupled by a bolt. Since the tip 24e of the mirror post 24 is also displaced in the Z direction, the connecting member 69 is formed with a hinge 69a having a small diameter and having flexibility in the direction perpendicular to the X axis.

  Further, the bias portion 63X is provided between the support member 64 fixed to the partition plate 53, and the front end portion of the support member 64 and the front end portion 24e of the mirror post 24, and the front end portion 24e faces the support member 64 side. A coil spring 65 that constantly applies a pulling force. For example, when the displacement portion of the drive element 67 and the end portion 68b of the displacement conversion portion 68 are coupled, the bias portion 63X (bias system 62) can be omitted.

  By the actuator 59X and the bias portion 63X, the tip 24e of the mirror post 24 can be displaced to an arbitrary position between the state A1 rotated counterclockwise and the state A2 rotated clockwise shown by a dotted line. In accordance with the displacement (load), stress acts on the convex portion 25 of the mirror 22 according to the lever principle. By using the lever principle as described above, even when the load generated by the actuator 59X is small, a large stress can be applied to the mirror 22, and the reflecting surface of the mirror 22 can be easily deformed greatly.

As shown in FIG. 4B, a groove portion 25S having a predetermined depth may be formed around the convex portion 25 on the back surface 22a of the mirror 22. As a result, even when the load (displacement) in the X direction (or Z direction) with respect to the tip 24e of the mirror post 24 is the same, the stress applied to the convex portion 25 of the mirror 22 increases, and the deformation amount of the reflecting surface of the mirror 22 is reduced. There are cases where it can be increased.
Next, in order to give a desired deformation amount (deformation distribution) to the reflection surface 22d of the mirror 22, the distribution of the load applied to each load point Pj (the front end portion 24e of the mirror post 24) on the back surface 22e of the mirror 22 is shown. An example of the input load will be described. First, on the mirror reflecting surface, as shown in FIG. 5C (the interval between contour lines is 0.001 times the wavelength. The same applies hereinafter), 1λ (λ is illumination light) with a fifth-order Zernike coefficient (Z5). An example of the input load when a deformation amount corresponding to the wavelength of IL) is given is the distribution shown in FIG. In FIG. 5A, the radius of the circumference E1 corresponds to a load of 1 kgf (1 kgf), and the vector E2 represents the magnitude and direction of the load applied to the load point P4 (the other load points). The same applies to vectors). Further, as shown in FIG. 5B, this input load is the sum of the forces E3, E4, E5 (each having a magnitude of 4 gf) acting on the hold blocks 23A, 23B, 23C that hold the mirror 22 (the resultant force). Is set to 0, and the sum of moments acting on the mirror 22 by the forces E3, E4, and E5 (total moment) is set to 0. As a result, even when the reflecting surface of the mirror 22 is deformed, a force that translates the mirror 22 in the X and Z directions perpendicular to the optical axis and a moment that rotates the mirror 22 around the optical axis do not act. The mirrors 22 can be stably held by the hold blocks 23A to 23C. In order to prevent the mirror 22 from causing lateral shift with respect to the hold blocks 23A, 23B, and 23C, it is preferable that the forces E3, E4, and E5 acting on the hold blocks 23A, 23B, and 23C are 0, respectively. As described above, even when the forces E3, E4, and E5 acting on the hold blocks 23A, 23B, and 23C are generated, these forces E3 to E5 are suppressed sufficiently smaller than the holding forces of the hold blocks 23A to 23C. Therefore, the mirror 22 does not cause a lateral shift.

The deformation of the mirror reflecting surface in FIG. 5C is easily shown in FIG. 5D by correcting the refractive power and tilt angle of the mirror 22 using a rotationally symmetrical imaging characteristic correction mechanism. Thus, it can be converted into deformation centered on the optical axis. Therefore, the deformation amount of the mirror reflecting surface in FIG. 5C is substantially equivalent to the deformation amount in FIG.
In addition, as shown in FIG. 6C, an example of an input load when a deformation amount corresponding to 1λ with a 10th-order Zernike coefficient (Z10) is given to the mirror reflecting surface is shown in FIG. 6A. It becomes. As shown in FIG. 6B, this input load is also set so that the resultant force on the mirror 22 is zero and the resultant moment is zero. The deformation amount of the mirror reflecting surface in FIG. 6C is substantially equivalent to the deformation amount in FIG. 6D by correcting the refractive power and the tilt angle.

  Further, as shown in FIG. 7C, an example of an input load when a deformation amount corresponding to 1λ is given to the mirror reflecting surface by a 14th-order Zernike coefficient (Z14) is a distribution shown in FIG. It becomes. As shown in FIG. 7B, this input load is also set so that the resultant force acting on the mirror 22 is zero and the resultant moment is zero. The deformation amount of the mirror reflecting surface in FIG. 7C is substantially equivalent to the deformation amount in FIG. 7D by correcting the refractive power and the tilt angle.

  In this way, in order to give the reflecting surface 22d of the mirror 22 a deformation amount corresponding to 1λ with Zernike coefficients Zi (Z5 to Z35) from the fifth order to the 35th order, the load point Pj on the back surface of the mirror 22 is applied. FIG. 8 shows the shape error of the reflecting surface 22d when an optimum input load is applied in which the resultant force is zero and the resultant moment is zero. In FIG. 8, the horizontal axis represents the Zernike coefficient Zi, and the vertical axis represents the shape error expressed by RMS (root mean square) (λ).

Next, an example of the operation of exposing the wafer while correcting the wavefront aberration of the projection optical system PL using the deformable mirror 50 will be described with reference to the flowcharts of FIGS. First, in step 102 of the preparation process of FIG. 12A, the n load points Pj on the back surface of the mirror 22 (tip portions 24e of the mirror post 24) are calculated in the calculation unit in the imaging characteristic control system 6 of FIG. Is the amount of change in the Zernike coefficient Zi (i = 1 to 81) from the first order to the 81st order when the unit load in the X direction and the Z direction is applied to each. Z _{a, b} (a = 1 to 81, b = 1 to 2n). The degree of freedom of the load application method is 2n. When the amount of change is represented by a matrix, the following equation is obtained. This matrix is stored in the storage device 10. The amount of change Z _{a, b} can be obtained by computer simulation. However, the amount of change Z _{a, b} is shown in FIG. 1 when a unit load is actually applied from the drive system 54 to each load point Pj. You may obtain | require from the variation | change_quantity of the wavefront aberration measured by the wavefront aberration measuring apparatus 39. FIG.

また、説明の便宜上、ミラー２２の裏面の荷重点Ｐｊを図９（Ａ）の荷重点Ｐ１〜Ｐ７として、図９（Ｂ）に示すように、ある一つの荷重点（例えばＰ７）にある方向の単位荷重を付与するものとする。

Further, for convenience of explanation, the load point Pj on the back surface of the mirror 22 is set as load points P1 to P7 in FIG. 9A, and as shown in FIG. 9B, the direction at a certain load point (for example, P7). A unit load of

In the next step 104, as shown in FIG. 9C, the direction and magnitude are such that the resultant force is 0 and the resultant moment is 0 at a plurality of load points (for example, load points P1 to P6) other than the load point P7. Apply a certain load. A set of loads having such a load distribution that the resultant force is 0 and the resultant moment is 0 is called a load set. The load set in FIG. 9C is the first load set LS1. For convenience of explanation, the coordinate system in which the load point Pj in FIG. 9A is arranged is (x, y), and the x-direction coordinate of the i-th load point Pi is x _i , and the y-direction coordinate is y _i. The angle formed by the load acting on the load point Pi with respect to the x direction is φ _i . At this time, moments m _i of origin around due to the load acting on the load point Pi is as follows.

また、各荷重点Ｐｊに２つの自由度で作用する荷重をｆ_iとすると、全部で自由度が２ｎの荷重によるミラー２２に対する合力（ｘ方向の合力Ｆ_x、ｙ方向の合力Ｆ_y）及び合モーメントＭは次のように表される。

Further, if the load acting on each load point Pj with two degrees of freedom is f _i , the resultant force to the mirror 22 by the load having a total degree of freedom of 2n (the resultant force F _{x in} the x direction, the resultant force F _{y in the} y direction) and The resultant moment M is expressed as follows.

このとき、ｉ番目の荷重点Ｐｉに対する荷重が１のときに、式（１２）の合力及び合モーメントが０となる組み合わせを求めるものとする。このために式（１２）の左辺のＦ_x，Ｆ_y，Ｍに０を代入すると次式が得られる。なお、ｆ_c,iは、ｉ番目の荷重点Ｐｉの荷重が１のときの他の荷重点Ｐｊの荷重である。

At this time, when the load on the i-th load point Pi is 1, a combination in which the resultant force and the resultant moment of Expression (12) are 0 is obtained. For this purpose, the following equation is obtained by substituting 0 into F _x , F _y , and M on the left side of equation (12). Note that f _{c, i} is a load at another load point Pj when the load at the i-th load point Pi is 1.

式（１３）を変形すると次式が得られる。

When the equation (13) is transformed, the following equation is obtained.

By solving this equation for the vector F _i , the load f _{c, i and} thus the first load set LS1 can be obtained. However, if n is an integer greater than or equal to 3, the matrix M _i is an irregular matrix of 3 rows × (2n−1) columns and has no inverse matrix. That is, in the matrix M _i , the number of columns is larger than the number of rows, so there are infinitely many solutions. For this reason, the following Moore-Penrose general inverse matrix is used as an example. The matrix M ^T _i is a transposed matrix of the matrix M _i .
M ⁺ _i = M ^T _i (M _i M ^T _i ) ⁻¹ (15)
The vector F _i is expressed as follows using the general inverse matrix of Equation (15).

式（１６）で表されるベクトルＦ_iは式（１４）を満たし、かつ次式により定められるベクトルＦ_iのノルムを最小にする解である。

The vector F _i represented by the equation (16) is a solution that satisfies the equation (14) and minimizes the norm of the vector F _i defined by the following equation.

すなわち、式（１６）で表されるベクトルＦ_iは、ｉ番目の荷重点Ｐｉに単位荷重を与えたときに、合力及び合モーメントが０になるように、かつ全体として最小の力となるように他の荷重点Ｐｊに与えられる力（ステップ１０４で他の荷重点に付与される荷重）の解である。

That is, the vector F _i represented by the equation (16) is such that when the unit load is applied to the i-th load point Pi, the resultant force and the resultant moment become zero and the minimum force as a whole. Is the solution of the force applied to the other load point Pj (the load applied to the other load point in step 104).

In the next step 106, when the unit load is applied to the i-th load point Pi and the load of the equation (16) is applied to the other load point Pj, the deformation amount of the reflecting surface of the mirror 22 is shown in FIG. As shown in FIG. 4, the amount of change Z _{a, b} (Z = 1 to 81, b = i) of the Zernike coefficient Zi is calculated. The amount of change Z _{a, b} may be obtained by measurement by the wavefront aberration measuring device 39 in FIG. 1, or may be obtained by multiplying the equation (10) by a load vector F _i having 2n elements. In the next step 108, the first load set LS <b> 1 and the deformation amount of the reflecting surface are stored in the storage device 10. In the next step 110, by repeating steps 102 to 108, as shown in FIG. 10B, other sets of load sets LS2, LS3,... And reflection surfaces of the mirrors 22 corresponding to these load sets are obtained. A deformation amount (change amount Z _{a, b of} Zernike coefficient Zi) is obtained and stored in the storage device 10.

In this case, there are 2n degrees of freedom in applying the load to the n load points Pj, but in this embodiment, the resultant force in the X direction (translation force in the X direction) and the resultant force in the Z direction (in the X direction) Since three conditions that the translational force in the Z direction) and the resultant moment about the optical axis are 0 are added, there are (2n−3) sets of load sets that are independent from each other. When the loads f _{a, b} (a = 1 to 2n, b = 1 to 2n−3) of all the load sets are represented by a matrix, the following matrix W _loadset is obtained.

また、各荷重セットに関してステップ１０６で求められるミラー２２の反射面の変形量（ツェルニケ係数Ｚｉの変化量Ｚ_a,b）を行列で表すと、式（１９）のツェルニケ感度レート行列Ｒ_set->Zが得られる。なお、ツェルニケ感度レート行列Ｒ_set->Zは、式（１０）の行列Ｒ_F->Zに式（１８）の行列Ｗ_loadsetを掛けても得られる。このツェルニケ感度レート行列Ｒ_set->Zは記憶装置１０に記憶される。これによって準備工程が終了する。

Further, when the deformation amount of the reflecting surface of the mirror 22 (change amount Z _{a, b of the} Zernike coefficient Zi) obtained in step 106 with respect to each load set is represented by a matrix, the Zernike sensitivity rate matrix R _{set-> in the} equation (19). _Z is obtained. The Zernike sensitivity rate matrix R _{set-> Z} can also be obtained by multiplying the matrix R _{F-> Z} of equation (10) by the matrix W _loadset of equation (18). This Zernike sensitivity rate matrix R _{set-> Z} is stored in the storage device 10. This completes the preparation process.

Next, in the exposure apparatus EX of the present embodiment, when the exposure process of FIG. 12B is started, the reticle R is loaded onto the reticle stage 15 in step 112, and illumination is performed with the illumination optical system ILS in step 114. Conditions are set, and alignment of the reticle R is performed. In the next step 116, the unexposed wafer W coated with the photoresist is loaded on the wafer table 31, and the alignment of the wafer W is performed. In the next step 118, the imaging characteristic control system 6 calculates the variation amount of the imaging characteristic of the projection optical system PL based on the integrated energy of the illumination light IL and the like. In step 120, the imaging characteristic control system 6 calculates a target deformation amount of the reflecting surface of the mirror 22 for correcting the fluctuation amount of the imaging characteristic, and supplies the calculation result to the aberration control system 9. In the next step 122, the aberration control system 9 determines the weight s such as each load set LSi that defines the load applied to the load points of the plurality of mirror posts 24 on the back surface of the mirror 22 in order to realize the target deformation amount. _i (i = 1 to 2n−3) is calculated. As shown in FIG. 11, when the target deformation amount is expressed by the Zernike coefficient Zi, the weights a, b, etc. of the load sets LS1, LS2, etc. for realizing the deformation amount are determined. Means.

In this case, if the Zernike sensitivity rate matrix R _{set-> Z of} the equation (19) and the vector S having the weight s _i as elements are used, the deformation amount of the reflecting surface of the mirror 22 (deformation converted into Zernike coefficients Z1 to Z81). The vector Z representing the quantity Z _a (a = 1 to 81) satisfies the following expression (20) and an equivalent expression (21).

If Expression (21) can be solved for the vector S (weight s _i ), the weight of each load set LSi is obtained. However, the Zernike sensitivity rate matrix R _{set-> Z} in equation (21) is an irregular matrix of 81 rows × (2n−3) columns and does not have an inverse matrix. In the matrix R _{set-> Z} , in the normal state where n is less than 42, the number of rows is larger than the number of columns, so there is no exact analytical solution, and there is always an error in the solution. Therefore, in the present embodiment, the weight vector S of the plurality of load sets is calculated so as to minimize the error in the surface shape of the generated mirror 22. For this purpose, the following diagonal matrix W _RMS is assumed. The diagonal component of the diagonal matrix is a conversion coefficient from each Zernike coefficient Zi to a surface shape error (error represented by RMS (root mean square)).

そして、式（２１）の両辺に左から式（２２）の対角行列Ｗ_RMSを掛けることによって、次式が得られる。
Ｗ_RMSＺ＝Ｗ_RMSＲ_set->ZＳ （２３）

Then, the following equation is obtained by multiplying both sides of equation (21) by the diagonal matrix W _RMS of equation (22) from the left.
W _RMS Z = W _RMS R _{set-> Z} S (23)

Here, using the vector Z _RMS and the matrix R _{set-> RMS} whose elements are errors represented by the RMS of the Zernike coefficient Zi, the equation (23) becomes as follows.
Z _RMS = R _{set-> RMS} S (24)
However, the vector Z _RMS and the matrix R _{set-> RMS} are defined as follows.
Z _RMS ≡W _RMS Z (25)
R set- _{> RMS} ≡W _RMS R set- _{> Z} (26)
Expression (21) is an expression expressed using the Zernike coefficient Zi, but expression (24) is an expression expressed using an error expressed by the RMS of the Zernike coefficient Zi.

Next, when the Moore-Penrose general inverse matrix is applied to the matrix R _{set-> RMS} in the equation (26), a matrix R ⁺ _{set-> RMS} expressed by the following equation is obtained. The matrix R ^T _{set-> RMS} is a transposed matrix of the matrix R _{set-> RMS} .
R ⁺ _{set-> RMS} = ( ^RT _{set-> RMS} R _{set-> RMS} ) ^{-1 RT} _{set-> RMS} (27)

By applying Expression (27) to Expression (24), the aberration control system 9 calculates the vector S (weight s _i ) as follows.
S = R ⁺ _{set-> RMS} Z _RMS (28)
Since the vector S obtained by this equation is an approximate solution, when the vector S is substituted into the equation (24), the following error e _RMS occurs on the left side.
Z _RMS + e _RMS = R _{set-> RMS} S (29)

Each component of the error e _RMS is an _RMS of the surface shape error represented by the Zernike coefficient Zi. Due to the property of Moore-Penrose's general inverse matrix, the vector S in Equation (28) is calculated so that the square sum of each component of the error e _RMS is minimized. Since the square sum of each component of the error e _RMS is the square of the RMS of the shape error of the reflecting surface of the mirror 22, the shape error (RMS) of the reflecting surface is minimized by calculating the vector S from the equation (28). Thus, the weights s _i (i = 1 to 2n−3) of the plurality of load sets LSi can be calculated.

In the next step 124, the aberration control system 9 uses the vector S composed of the weights s _i (i = 1 to 2n−3) of the load set LSi obtained in step 122 and the matrix W _loadset of Expression (18). From the following equation, a load applied to the tip portions 24e of the n mirror posts 24 on the back surface of the mirror 22 in the X direction and Z direction (or a driving amount formed by displacement of the tip portions 24e in the X direction and Z direction) f _i ( i = 1 to 2n) is calculated. A vector having the load f _i as a component is F. This means that a weight for each mirror post 24 is calculated by a weighted sum of a plurality of load sets.
F = W _loadset S (30)

In the next step 126, the aberration control system 9 supplies the 2n loads f _i to the actuator control system 56 of FIG. As a result, loads (and thus displacements) set in the X direction and Z direction are applied to the tip 24e of each mirror post 24, respectively, and the load acts on the mirror 22 as stress by the lever principle, and the reflection of the mirror 22 occurs. The surface 22d is deformed into a target shape. Accordingly, since the fluctuation amount of the imaging characteristics of the projection optical system PL is corrected with high accuracy including rotationally asymmetric components and higher-order components, the pattern image of the reticle R is transferred to the wafer via the projection optical system PL. It is formed on the surface of W with high accuracy.

  In this state, in the next step 128, irradiation of the illumination light IL is started, and an image of the pattern of the reticle R is exposed to each shot area of the wafer W by the step-and-scan method. Thereafter, in step 130, the exposed wafer W is unloaded. In the next step 132, when the next wafer is exposed, steps 116 to 130 are repeated. Then, when there is no unexposed wafer in step 132, the exposure process is completed. At this time, since the imaging characteristics of the projection optical system PL are corrected with high accuracy, the image of the pattern on the reticle R can be exposed to each shot area of the wafer W with high accuracy.

The effects and the like of this embodiment are as follows.
(1) The deformable mirror 50 of the present embodiment includes a mirror 22 made of a concave mirror that reflects the illumination light IL, a convex mirror post 24 provided on the back surface 22e of the reflection surface 22d of the mirror 22, and a mirror post 24. A load applying system 58 (displacement consisting of actuators 59X and 59Z) applies a load in the X direction and the Z direction substantially orthogonal to the direction in which the mirror post 24 extends to the front end portion 24e of the lens to displace the front end portion 24e in the X direction and the Z direction. Mechanism).

  According to the present embodiment, a load is applied to the tip 24e of the convex mirror post 24 provided on the back surface 22e of the mirror 22 in a direction substantially perpendicular to the longitudinal direction (substantially Y direction) of the mirror post 24 ( By applying the displacement, the reflecting surface 22d of the mirror 22 can be deformed with a small load by the lever principle. Further, even when the number of mirror posts 24 is large, the same number of load applying systems 58 as the mirror posts 24 can be easily installed on the back surface 22e of the mirror 22. Then, by increasing the number of mirror posts 24, the reflecting surface 22d of the mirror 22 can be easily deformed into various shapes.

(2) The deformable mirror 50 also includes a position sensor 60 that measures the displacement of the tip 24e of each mirror post 24 in the X and Z directions. Therefore, since the load applied to the tip 24e from the displacement of the tip 24e in the X direction and the Z direction can be measured, by feeding back this measured value, a target load (on the tip 24e of each mirror post 24 ( (Or target displacement) can be given with high accuracy.
Each mirror post 24 may be provided with a load sensor (load cell or the like) instead of the position sensor 60, and the load may be directly measured.

(3) Further, since the plurality of mirror posts 24 are provided on the back surface of the mirror 22 and the load applying system 58 is provided for each of the plurality of mirror posts 24, the reflection surface 22d of the mirror 22 can be freely deformed. High degree. Note that only one mirror post 24 may be provided on the back surface of the mirror 22.
(4) Further, each mirror post 24 has a load (displacement) in the X direction and the Z direction, and a Y direction (normal direction of the back surface 22e of the mirror 22) substantially parallel to the longitudinal direction of the mirror post 24. A load (displacement) may be applied. In this case, if the number of mirror posts 24 is n, the degree of freedom in the load direction (displacement direction) is 3n, so that the reflecting surface of the mirror 22 can be deformed with more degrees of freedom.

Each mirror post 24 is connected to the convex portion 25 on the back surface of the mirror 22. For example, by forming the convex portion 25 high, a load is applied to the tip portion of the convex portion 25 from the load applying system 58. The tip may be displaced.
(5) In addition, the projection optical system PL of the present embodiment is a catadioptric optical system including lenses L8 and L9 and a mirror 22 (a plurality of optical elements), and the reflecting surface of the mirror 22 among these optical elements is deformed. A possible deformable mirror 50 is provided. Accordingly, the deformable mirror 50 deforms the reflecting surface of the mirror 22 so as to correct the variation amount of the imaging characteristics of the projection optical system PL (for example, non-rotationally symmetric aberration and higher order aberration), thereby projecting. The imaging characteristics of the optical system PL can always be maintained in the target state.

  (6) The exposure apparatus EX of the present embodiment illuminates the pattern of the reticle R with the illumination light IL (exposure light), and the wafer W (substrate) through the pattern and the projection optical system PL with the illumination light IL. In the exposure apparatus that performs exposure, the projection optical system PL includes a deformable mirror 50. Accordingly, the imaging characteristics of the projection optical system PL can always be maintained with high accuracy in a target state, so that the pattern image of the reticle R can be exposed to each shot region of the wafer W with high accuracy.

  (7) Further, when the deformable mirror 50 includes a plurality of combinations of the mirror post 24 and the load applying system 58 (actuators 59X and 59Z), the driving of the present embodiment for driving the deformable mirror 50 In the method, at least one of the plurality of mirror posts 24, for example, the mirror post 24 at the load point P7 is displaced by applying a load via the load applying system 58 corresponding to the mirror post 24, and the reflecting surface of the mirror 22 is thereby displaced. The step 102 is deformed. Further, the driving method corrects the force (the resultant force in the X direction, the Z direction, and the resultant moment around the optical axis) that causes the position change of the mirror 22 caused by the displacement of the mirror post 24 at the load point P7. The step 104 includes, for example, a step 104 in which the mirror post 24 at the load points P1 to P6, which is different from the mirror post 24 at the load point P7, is displaced by applying a load via the load applying system 58 corresponding thereto.

According to this driving method, even if a load is applied to the plurality of mirror posts 24, the force for displacing the mirror 22 as a whole does not act. Therefore, for example, the mirror 22 can be stably held by the three hold blocks 23A to 23C.
In step 104, the mirror post 24 at the load points P1 to P6 may be displaced so as to correct at least one of the resultant force in the X direction and the Z direction of the mirror 22 and the resultant moment around the optical axis. Good.

  (8) In addition, the driving method causes, for example, a change in the position of the mirror 22 that occurs when the mirror post 24 at the load point P7 is displaced under a plurality of different conditions (a displacement amount and / or a displacement direction corresponding to the load). You may have the step which memorize | stores the information (synthetic force, resultant moment) regarding force according to the conditions. In this case, in step 102, a predetermined condition is selected from the plurality of different conditions according to the target deformation amount of the reflecting surface of the mirror 22, and the mirror post 24 at the load point P7 is assigned a corresponding load. The reflecting surface of the mirror 22 may be deformed by being displaced via the system 58. In step 104, for example, the mirror posts 24 at the load points P1 to P6 that are different from the mirror post 24 at the load point P7 are selected based on the information about the force that causes the position change of the mirror 22 corresponding to the predetermined condition. It may be displaced via a load applying system 58 corresponding to. Thereby, the freedom degree of a deformation | transformation of the reflective surface of the mirror 22 improves.

In the present embodiment, the following modifications are possible.
First, the X-axis actuator 59X in FIG. 4A transmits the displacement of the drive element 67 to the tip 24e of the mirror post 24 via the displacement converter 68. On the other hand, the load (displacement) from the drive element 71 having the same configuration as that of the drive element 67 is passed through the connecting member 69 as shown by the X-axis actuator 59XA of the first modification of FIG. You may transmit directly to the front-end | tip part 24e of the mirror post 24. FIG. In FIG. 13A, in which parts corresponding to those in FIG. 4A are denoted by the same reference numerals, a drive element 71 having a tip that can be displaced in the X direction is fixed to a base member 70 fixed to the partition plate 53. ing. A connecting member 69 having a hinge portion 69 a is disposed between the tip portion and the tip portion 24 e of the mirror post 24.

  Further, a support member 64 fixed to the partition plate 53, and a compression coil provided between the front end portion of the support member 64 and the front end portion 24e of the mirror post 24, and biases the front end portion 24e toward the drive element 71. An X-axis bias portion 63XA is constituted by the spring 65R. The configuration other than this is the same as that of the actuator 59X of FIG. 4A. Even when the actuator 59XA of the first modification is used, the load (displacement) in the X direction can be easily applied to the tip 24e of the mirror post 24. Can be given. For example, when the displacement portion of the drive element 71 and the mirror post 24 are coupled, the bias portion 63XA can be omitted.

  Further, in the example of FIG. 4B, the mirror post 24 is connected to the convex portion 25 on the back surface of the mirror 22. In addition to this, as shown in the main part of the second modification of FIG. 13B, a recess 22f having a substantially square cross-sectional shape is formed on the back surface 22e of the mirror 22, and a rod shape connected to the recess 22f. The tip 24Ae may be displaced by applying loads in the X and Z directions to the tip 24Ae of the mirror post 24A. In FIG. 13B, a connecting portion 24Aa composed of two rods arranged in the X direction via a hinge portion 24Aa1 is formed at the lower end of a rod-like mirror post 24A. A connecting portion 24Ab composed of two rods arranged in the Z direction is also formed at the lower end via a hinge portion (not shown).

In FIG. 13B, for example, by inserting the leading ends of the connecting portions 24Aa and 24Ab into the recesses 22f in a state where the leading ends of the connecting portions 24Aa and 24Ab are narrowed, the mirror posts 24A are connected to the connecting portions 24Aa, 24Aa, It can couple | bond with the back surface 22e of the mirror 22 via 24Ab. Further, the hinge parts 24Ac and 24Ad are also formed on the mirror post 24A, but the hinge parts 24Ac and 24Ad can also be omitted.
The cross-sectional shape of the tip portions 24e and 24Ae of the mirror post 24 or 24A is not limited to a circular shape, and the cross-sectional shape of the tip portions 24e and 24Ae is, for example, triangular, and the cross section is displaced from three directions. Also good. Furthermore, the cross-sectional shape of the tip portions 24e and 24Ae may be a square or a rectangular shape such as a quadrangle, or a polygonal shape, and the cross-section may be displaced from a plurality of directions.

  Next, when a semiconductor device (electronic device) is manufactured using the exposure apparatus EX of the above-described embodiment, as shown in FIG. 14, this semiconductor device performs step 221 for performing device function / performance design, this design. Step 222 for producing a mask (reticle) based on the steps, Step 223 for producing a substrate (wafer) as a base material of the device, and exposure of the reticle pattern onto the resist-coated substrate (photosensitive substrate) by the exposure apparatus EX A process of developing the exposed substrate to form a resist pattern, a substrate processing step 224 including heating (curing) and etching of the developed substrate, a device assembly step (dicing process, bonding process, packaging process, etc.) 225) including the processing process, and the inspection step 226, etc. It is manufactured Te.

  In other words, the device manufacturing method forms the pattern of the photosensitive layer on the substrate (wafer) using the exposure apparatus of the above embodiment, and processes the substrate on which the pattern is formed (step 224). ). According to this manufacturing method, since the exposure apparatus can reduce various aberrations including non-rotationally symmetric aberration, a device having a fine pattern can be manufactured with high accuracy.

  In the present invention, for example, the aberration correction of the projection optical system of the immersion type exposure apparatus disclosed in US Patent Application Publication No. 2007/242247 or European Patent Application Publication No. 1420298 is performed. It can also be applied to. The present invention can be applied not only to a scanning type exposure apparatus but also to a batch exposure type exposure apparatus (such as a stepper).

  The present invention can also be applied to the case where aberration correction of a projection optical system of a projection exposure apparatus using extreme ultraviolet light (EUV light) having a wavelength of about several nm to 100 nm as exposure light is performed. When EUV light is used as the exposure light, the projection optical system is composed of a plurality of mirrors (concave mirrors, convex mirrors, plane mirrors, etc.) excluding a specific filter and the like. It can be used to hold at least one of the plurality of mirrors.

In addition, the present invention is not limited to application to an exposure apparatus for manufacturing a semiconductor device, for example, an exposure apparatus for a display device such as a liquid crystal display element formed on a square glass plate or a plasma display, It can also be widely applied to an exposure apparatus for manufacturing various devices such as an image sensor (CCD or the like), a micromachine, a thin film magnetic head, and a DNA chip. Furthermore, the present invention can also be applied to an exposure apparatus when manufacturing a mask (photomask, reticle, etc.) on which a mask pattern of various devices is formed using a photolithography process.
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.

  EX ... exposure apparatus, R ... reticle, W ... wafer, ILS ... illumination optical system, PL ... projection optical system, 6 ... imaging characteristic control system, 9 ... aberration control system, 22 ... mirror (concave mirror), 24 ... mirror post , 39 ... Wavefront aberration measuring device, 47 ... Split barrel, 50 ... Deformable mirror, 54 ... Drive system, 58 ... Load applying system, 60 ... Position sensor, 62 ... Bias system

Claims (17)

A reflective optical element that reflects light,
A reflective member that reflects light;
A convex member provided on the back surface of the reflective surface of the reflective member;
A displacement mechanism for displacing the convex member in a direction intersecting a direction in which the convex member extends;
A deformable reflective optical element comprising:   The deformable reflective optical system according to claim 1, wherein the displacement mechanism displaces the convex member in two directions orthogonal to each other within a plane orthogonal to the direction in which the convex member extends. element.   The deformable reflective optical element according to claim 1, wherein the displacement mechanism is capable of displacing the convex member in a direction in which the convex member extends.   The deformable reflective optical element according to any one of claims 1 to 3, wherein the displacement mechanism includes a number of actuators according to a degree of freedom of displacement of the convex member.   The deformable reflective optical element according to any one of claims 1 to 4, wherein the convex member is a rod-shaped member. A convex portion is formed on the back surface of the reflecting member provided with the convex member,
The deformable reflective optical element according to any one of claims 1 to 5, further comprising a clamp portion that connects the convex member and the convex portion on the back surface.   The deformable reflective optical element according to claim 1, comprising a plurality of combinations of the convex member and the displacement mechanism. In an optical system including a plurality of optical elements,
The deformable reflective optical element according to any one of claims 1 to 6,
An optical system comprising: a control device that corrects aberration by displacing the convex member via the displacement mechanism of the reflective optical element. The reflective optical element includes a plurality of combinations of the convex member and the displacement mechanism,
The control device displaces at least one convex member among the plurality of convex members via the displacement mechanism corresponding to the convex member in order to correct the aberration, A convex shape different from the at least one convex member among the plurality of convex members so as to correct a force that causes a change in the position of the reflecting member caused by the displacement of the at least one convex member. The optical system according to claim 8, wherein the member is displaced through the displacement mechanism corresponding to the convex member. In an exposure apparatus that illuminates a pattern with exposure light and exposes the substrate through the pattern and the projection optical system with the exposure light,
An exposure apparatus comprising the optical system according to claim 8 or 9.   The exposure apparatus according to claim 10, wherein the optical system is the projection optical system. A method for driving a reflective optical element, comprising:
The reflective optical element includes the reflective optical element according to claim 7,
A first step of displacing at least one convex member among the plurality of convex members via the displacement mechanism corresponding to the convex member, and deforming the reflective surface of the reflective member;
A convex shape different from the at least one convex member among the plurality of convex members so as to correct a force that causes a change in the position of the reflecting member caused by the displacement of the at least one convex member. And a second step of displacing the member via the displacement mechanism corresponding to the convex member. Storing the information on the force causing the position change of the reflecting member that occurs when the at least one convex member is displaced under a plurality of different conditions according to the condition,
In the first step, a predetermined condition is selected from the plurality of different conditions according to a target deformation amount of the reflecting surface of the reflecting member, and the at least one protruding member is moved to the protruding shape. Displacement via the displacement mechanism corresponding to the member, to deform the reflective surface of the reflective member,
In the second step, a convex member different from the at least one convex member is selected as the convex member based on information on a force that causes a change in the position of the reflective member corresponding to the predetermined condition. The method of driving a reflective optical element according to claim 12, wherein the displacement mechanism is displaced via the displacement mechanism corresponding to the above.   14. The driving method for a reflective optical element according to claim 12, wherein the plurality of different conditions include a displacement amount or a displacement direction of the convex member.   15. The driving of a reflective optical element according to any one of claims 12 to 14, wherein the information regarding the force causing the position change of the reflective member includes a rotational force around the optical axis of the reflective member. Method.   The method of driving a reflective optical element according to claim 12, wherein the first step obtains a weight of a condition selected from the plurality of conditions. Forming a pattern of a photosensitive layer on a substrate using the exposure apparatus according to claim 10 or 11,
Processing the substrate on which the pattern is formed.

Images