Exposure method and projection exposure apparatus

Abstract

A projection exposure apparatus carries out scan exposure with illumination flux of slit(s) by moving a mask and a substrate in a direction of one-dimension at synchronized speeds with each other. The mask is inclined with a predetermined angle relative to the substrate in the direction of one-dimensional movement. The substrate is also moved in a direction of optical axis of projection optical system when moved in the direction of one-dimensional, such that a central part of transfer area on the substrate is located on a best focal plane of projection optical system upon scan exposure.

Description

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

FIG. 1 is a drawing to show a structure of a preferred embodiment of a projection exposure apparatus according to the present invention. In this embodiment, a projection optical system PL may be composed only of refraction elements or of a combination of refraction and reflection elements to be a one-fifth reduction system, as being double side telecentric.

Illumination light for exposure for a mercury vapor lamp 1 is condensed at a secondary focus point by an ellipsoidal mirror 2. A rotary shutter 3 is disposed at the secondary focus point to switch the illumination light between interception and transmission of light by a motor 4. The illumination light passed through the shutter 3 is reflected by a mirror 5, and enters an optical integrator or fly eye lens 7 through an input lens 6. There are numerous secondary light source images formed at the output side of the fly eye lens 7, and the illumination light from the secondary light source images is led through a beam splitter 8 into a lens system or condenser lens 9.

Movable blades BL₁, BL₂, BL₃, BL₄ of a reticle blind mechanism 10 as shown in FIG. 2 are arranged on a back side focal plane of the lens system 9. The four blades BL₁, BL₂, BL₃, BL₄ are independently moved by a drive system 50. In this embodiment, edges of the blades BL₁, BL₂ define a width of aperture AP in the X-direction or scan exposure direction, and edges of blades BL₃, BL₄ a length of the aperture AP in the Y-direction or stepping direction. The aperture AP defined by the respective edges of four blades BL₁-BL₄ is included in a circular image field IF of the projection optical system PL.

The illumination light takes a uniform distribution of illumination at the position of the blind mechanism 10. The illumination light passed through the aperture AP of the blind mechanism 10 is guided through a lens system 11, a mirror 12, and a main condenser lens 13 to a reticle R to be illuminated. Then an image of the aperture AP defined by the four blades BL₁-BL₄ of the blind mechanism 10 is focussed on a pattern plane on a lower surface of the reticle R.

The reticle R receiving the illumination light defined by the aperture AP is held by a reticle stage 14, which is movable at a uniform rate at least in the X-direction on a column 15. The column 15 is incorporated with an unrepresented column holding a lens barrel of the projection optical system PL. The reticle stage 14 may be one-dimensionally moved in the X-direction and rotated by a small amount to correct its yawing by a drive system 51. A movable mirror 31 is fixed at an end of the reticle stage 14 to reflect a measurement beam from a laser interferometer 30, so that the laser interferometer 30 may measure a position of the reticle R in the X-direction and the amount of yawing in a real time manner. A fixed mirror or reference mirror 32 for the laser interferometer 30 is fixed at the upper end of the lens barrel of projection optical system PL.

An image of pattern formed in a transfer region, for example a rectangular region, of reticle R is imaged on a wafer W after reduced to one-fifth thereof by the projection optical system PL. The wafer W is held together with a reference mark plate FM by a wafer holder 16 which is rotatable by a small amount and inclinable at an arbitrary angle. The wafer holder 16 is disposed on a Z-stage, which is movable by a small amount in a direction of the optical axis AX or the Z-direction of the projection optical system PL. The Z-stage 17 is mounted on an X,Y-stage 18 which is two-dimensional moveable in the X- and the Y-directions in the step and repeat method, and the X,Y-stage 18 is driven by a drive system 52. Further, a laser interferometer 33 measures a coordinate position and a yawing amount of the X,Y-stage 18. A fixed mirror or reference mirror 34 for the laser interferometer 33 is fixed at the lower end of the lens barrel of the projection optical system, and a moveable mirror 35 is fixed at one edge of the Z-stage 17. Since the projection magnification is one fifth in this embodiment, a moving velocity Vws of the X,Y-stage is set at one fifth of a velocity Vrs of the reticle stage in the X-direction 14 upon scan exposure.

Also in the present embodiment, there are provided an alignment system 40, which employs a TTR (Through-The-Reticle) method to detect an alignment mark or reference mark FM on the wafer W through the reticle R and the projection optical system PL, and an alignment system 41, which employs a TTL (Through-The-Lens) method to detect an alignment mark or reference mark on the wafer W through the projection optical system at a position below the reticle R. These alignment systems 40, 41 perform relative alignment of the reticle R and the wafer W before start of S and S exposure or during the scan exposure. If the reference mark FM is a light emitting type, a photoelectric sensor 42 as shown in FIG. 1 receives the light from the mark through the projection optical system PL, the reticle R, the condenser lens 13, the lens system 11, 9 and the beam spitter 8, whereby defining a position of the reticle R in the coordinate system of the X,Y-stage 18 or defining a position of detection center of the respective alignment systems 40, 41. It should be noted that these alignment systems are not always essential to the present invention.

As the length of the aperture AP of the blind mechanism 10 is made longer in the Y-direction perpendicular to the scan direction or X-direction, the number of scannings in the X-direction or the number of steppings in the Y-direction on the wafer W may be reduced. However, the length of the aperture AP in the Y-direction might better be changed by the edges of the blades BL₃, BL₄ depending on size, shape, and arrangement of chip pattern on the reticle R. A preferred example is that the facing edges of the blades BL₃, BL₄ coincide with street lines dividing the shot area on the wafer W. It is easy for this arrangement to adjust the length of aperture in correspondence with a size change in the Y-direction of shot area. If a length in the Y-direction of one shot area is over the maximum length of the aperture AP in the Y-direction, overlay exposure should be effected in the shot area to obtain seamless exposure as disclosed in U.S. Pat. No. 4,924,257. Since the method itself is not always essential to the present invention, detailed explanation thereof is omitted.

Below explained with reference to FIG. 3A are the wafer holder 16 inclinable at an arbitrary angle, and neighbors thereof. A motor 21 is provided at the Z-stage 17, on the X,Y-stage 18 to drive the Z-stage 17 in the direction of the optical axis AX. The wafer holder 16 is mounted on the Z-stage 17 with its center being supported. Leveling drive selection 20A, 20B are provided at periphery of the wafer holder 16, so that the wafer W on the holder 16 may be inclined at an arbitrary angle. The leveling mechanism is disclosed for instance in U.S. Pat. No. 4,770,531, so detailed explanation of the leveling mechanism is omitted.

A focus and leveling sensor is provided to control the inclination angle of the wafer W, which is constructed by a light projection 19A emitting an optical flux BPL of a wave length different from that of the exposure light and a light receiver 19B receiving an optical flux BRL, which is the optical flux BPL reflected by the wafer surface. A focus point, of the optical flux BPL from the focus and leveling sensor is coincident with a line including a point through which the optical axis of the projection optical system PL passes on the wafer W. An example of the focus and leveling sensor is disclosed in U.S. Pat. No. 4,558,949. Thus detailed explanation is omitted.

The leveling drive sections 20A, 20B are driven by a command from a leveling control system 53 to determine an inclination amount of wafer holder 16, based on leveling information from the light receiver 19B and information from the main control section 100. With constant feed back of leveling information from the light receiver 19B, a proper inclination angle of the wafer W may be maintained. Further, focus information could be obtained to always locate an intersection with the optical axis AX on the wafer W on the best imaging plane of the projection optical system, with the information from the focus and leveling sensor. In this case, the motor 21 is driven by a command from the Z-stage control system 54 based on position information obtained by the light receiver 19B, to drive the Z-stage 17 in the direction of the optical axis AX.

The optical flux BPL is radiated on the wafer W as a slit light SLI inclined by 45°C with respect to the rectangular illumination area AP' defined by the aperture AP of the blind as shown in FIG. 3B. By this, the position and the inclination of the wafer W in the Z-direction may be controlled without influence from the directionality of circuit pattern in the chip area CP₁-CP₄ already formed on the wafer W. Although there is shown only two points of the leveling drive section for explanation, it is no doubt that drive on three points is better.

In the focus and leveling sensor as disclosed in the above-mentioned U.S. Pat. No. 4,558,949, parallel optical fluxes occupying a determined ara are impinged on the wafer surface, and reflection optical flux from the wafer surface is photoelectrically detected, for example using a quartered photodetector, to detect the inclination or the leveling information of the wafer surface. Then, a variable field stop may be disposed inside the light projector 19A as disclosed for example in U.S. Pat. No. 4,902,900, to adjust the size and the shape of illumination area of parallel optical fluxes on the wafer surface, so that the size and the shape of the illumination area or detection area of the parallel optical fluxes on the wafer surface is desirably made almost coincident with the rectangular illumination area AP' defined by the aperture AP of the blind. By this, an average inclination of a local area in the shot region on the wafer W corresponding to the rectangular illumination area AP' may be effectively detected, the inclination amount of the wafer surface or wafer holder 16 may be controlled with a higher precision.

An operation of the preferred embodiment of the projection exposure apparatus will be below explained. The main control section 100 totally dominates the sequence and control of the operation as shown in FIG. 1. A fundamental operation of the main control section 100 is that, based on inputs of position information and yawing information from the laser interferometer 30, 33 and on inputs of speed information from a tachogenerator or the like in the drive systems 51, 52, reticle pattern and the wafer pattern in one shot area are relatively moved within a determined alignment error of the relative position while keeping a determined ratio of speed of the recticle stage 14 and the X,Y-stage 18 upon scan exposure. In addition to the fundamental operation, the main control section 100 of the present embodiment is characterized in that the best imaging plane or the best focal plane of the projection optical system PL (projection image plane of transfer region of the reticle) is inclined relative to the shot area on the wafer W, the central part of transfer region (corresponding to illumination area AP') on the shot area is always located at or near the best imaging plane or best focus position of the projection optical system PL, and thereby the focus condition of pattern image of reticle is continuously or discretely changed with correspondence to a position in the illumination area in the one-dimensional scan direction during the scan exposure, by controlling the leveling control system 53 and the Z-stage control system 54 together.

FIGS. 4A-4C schematically show a method for exposure using the projection exposure apparatus of the preferred embodiment according to the present invention. Positions 1-9 in a circuit pattern IR on the reticle R correspond to positions 1-9 on the wafer W, respectively. The wafer W is inclined relative to the pattern IR. The circuit pattern IR is displayed just above the wafer W, and a projection ratio of the circuit pattern IR is 1 on the wafer W for convenience of explanation. In the drawings, there are three optical fluxes LR, LC, LL shown out of the exposure flux defined by the single aperture AP. The optical flux LR, LL of the three are defined by the blades BL₁, BL₂ as shown in FIG. 2, and symmetrically arranged before and after the optical axis AX in the scan exposure direction. The width between the optical fluxes LR and LL corresponds to a width of the aperture AX in the X-direction, representing illumination range of exposure flux in the scanning direction. The intensity of exposure flux is uniform in this illumination range. The optical flux LC has a main light beam passing through the center of the illumination range of exposure flux. The main light beam corresponds to the optical axis AX of the projection optical system PL. The best imaging plane of the projection optical system PL is shown by a broken line BF.

The scan exposure is controlled such that the X,Y-stage 18 is driven in the X-direction and the Z-stage 17 is simultaneously driven in the direction of the optical axis AX, to always locate the approximate center in the illumination area of the wafer W (corresponding to the approximate center of illumination range of exposure flux) on the best imaging plane BF of the projection optical system PL. If the width of illumination area AP' on the wafer W is defined as D_ap, the inclination angle between the illumination area AP' on the wafer W and the best imaging plane BF as θ₁, and a width in the optical axis of the depth of focus of the projection optical system PL (DOF) as ΔZ_f, at least one of the depth D_ap of the illumination area and inclination angle θ₁ is adjusted to satisfy the following relation: D_ap.sin θ₁≧ΔZ_f. A theoretical depth of focus is normally given by an equation Δ_Zf=λ/NA², where λ is an exposure wave length, and NA is a numerical aperture of projection optical system.

A positional relation of wafer W and pattern IR to the exposure flux just after the scan exposure start is shown in FIG. 4A. Noting the position 2 in the circuit pattern IR, it is just entered within the illumination range of exposure flux. However, an image at the position 2 on the wafer W is in a condition of defocussing and the distribution of intensity of projection image has a gentle peak. FIG. 4B shows a condition after further scan exposure, in which the position 2 on the wafer W is located on the best imaging plane BF. In this condition, the image at the position 2 is in the best focus condition, presenting a steep peak in the intensity distribution of image. When the wafer is moved as shown in FIG. 4C, the position 2 is in the condition of defocussing opposite to the condition as shown in FIG. 4A, again showing a gentle peak in the intensity distribution of image.

FIG. 5 shows a distribution of exposure amount in the direction of the optical axis AX or in the Z-direction at the position 2 on the wafer W by the above-described scan exposure or uniform rate scan. The exposure amount at the position 2 is uniform in the Z-directional range of D_ap.sin θ₁ (the depth of focus DOF). FIGS. 6A and 6B show resultant distributions of intensity of image at the position 2. The intensity distributions ER, EC, EL in FIG. 6A represent intensities of images obtained from the optical fluxes LR, LC, LL, respectively. A distribution of intensity E as shown in FIG. 6B represents an integration of image intensity obtained from exposure flux of fluxes LR, LC, LL. Since the position 2 receives the optical fluxes (optical energy) while in the illumination range of exposure flux, the integrated intensity distribution E shows a gentle peak. As seen in FIG. 6B, a width in which the intensity is over an exposure amount E_th enough to effect photosensitizing of photo resist on the wafer W, i.e., to completely remove the photo resist, becomes relatively broad, accordingly.

In order to narrow the width W, there may be provided at least two peaks in the one-dimensional scan direction of scan exposure in the distribution of intensity of rectangular illumination flux. For example, as shown in FIG. 7, the reticle blind mechanism may be arranged to have such a structure that a central portion of the aperture AP is intercepted (double slit aperture). It may be achieved by providing the blade BL₄ of the four blades of the blind mechanism 10 with a Y-directionally extending interception branch to intercept light in a determined width in the X-direction at the center of the aperture AP. In case of use of such blind mechanism, the exposure amount in the direction of the optical axis AX or in the Z-direction is distributed as shown in FIG. 8 at the position 2 on the wafer W by the scan exposure or uniform rate scan. The exposure at the position 2 shows two identical intensity ranges located near the both ends of the Z-directional range of D_ap.sin θ₁ (depth of focus DOF). By this arrangement, only the optical fluxes LR, LL in the exposure fluxes as shown in FIG. 4 possibly have intensities in the distribution.

FIGS. 9A and 9B show distributions of intensity of image obtained at an arbitrary position, for example at the position 2 as above described, on the wafer W when the scan exposure of uniform rate scan is conducted by using such optical fluxes. The intensity distributions ER', EL' as shown in FIG. 9A are distributions of intensity of images given by the optical fluxes LR, LL, respectively. An intensity distribution E' as shown in FIG. 9B is an integration of the intensity distributions ER', EL'. The intensity distribution E' shows a steeper peak than that as shown in FIG. 6B. A width W' in which the intensity is over the exposure amount E_th enough to effect photosensitizing of photo resist on the wafer W to completely remove it, is narrower than the width W as shown in FIG. 6B.

Furthermore, three peaks may be employed in the one-dimensional scan direction of scan exposure in a distribution of intensity of rectangular illumination flux. For this purpose, a reticle bind mechanism is provided with blades with three slits in the aperture. FIG. 10 shows a distribution of exposure in the direction of optical axis AX at the position 2 on the wafer W by the similar scan exposure. At the position 2, three regions in the Z-direction have almost identical intensities of exposure, one near the best imaging plane BP, and two near the both ends of the Z-directional range of D_ap·sin θ₁ (depth of focus DOF). Therefore, exposure flux reaching the wafer W includes fluxes corresponding to the optical fluxes LR, LC, LL, as shown in FIGS. 4A-4C. The optical fluxes LR, LL are symmetrical with respect to the optical flux LC having the same optical axis AX of the projection optical system. In case of scan exposure with the flux with three peaks in the distribution of intensity of the illumination range, a distribution of intensity of image projected onto the wafer W shows a steeper peak than the distribution of intensity E as shown in FIG. 6B. A width of projected image will be narrower than the width W as shown in FIG. 6B, accordingly,

Comparing two slits and three slits in the one-dimensional direction of scan exposure in the intensity distribution of rectangular illumination flux if the optical intensity of illumination flux is almost identical, exposure with three slits allows faster moving speed of the X,Y-stage and gives a higher throughput. This is opposite to the successive focussing exposure method conventionally known in U.S. Pat. No. 4,869,999.

In the above examples, the blades of the bind mechanism have a light intercepting portion. In another arrangement, the same effect may be obtained by an interception member such as ND filter having dimensions and shape corresponding to a region to be intercepted at a position conjugate to the circuit pattern IR in the optical path. Furthermore, although the wafer surface is inclined before the scan exposure in the above examples, the inclination of the wafer holder 16 may be controlled together with the Z-directional position of the wafer surface by using detection information of the focus and leveling sensor at the time of scan exposure start.

Claims

1. A method for exposure in which a pattern formed in a transfer region on a mask is subject to projection exposure through a projection optical system to be led onto an area to be exposed on a photosensitive substrate, and said mask and photosensitive substrate are at least one-dimensionally, relatively scanned with respect to a projection field of said projection optical system: comprising,

limiting a width of said area of pattern image projected on said photosensitive substrate through the projection optical system to an approximately constant value in a direction of one-dimensional scan; and

inclining a local surface on said photosensitive substrate on which said pattern image is formed, relative to a best focal plane of said projection optical system in the direction of one-dimensional scan.

2. A method for exposure according to claim 1, wherein a central part of said local surface in the direction of one-dimensional scan on said photosensitive substrate, on which said pattern image is formed, substantially coincides with said best focal plane of projection optical system when scanned for exposure.

3. A method for exposure according to claim 1, wherein, defining a width of local surface on the photosensitive substrate in the direction of one-dimensional scan, on which said pattern image is formed as D_ap, an angle of inclination between said local surface and said best focal plane as θ₁, and a depth of focus of said projection optical system in a direction of optical axis as ΔZ_f, the following relation is satisfied by adjusting at least one of said width D_ap of the pattern image area and said inclination angle θ₁;

D_ap·sin θ₁≧Z_f.

4. A method for exposure according to claim 1, wherein, the order to limit said width of pattern image area in the direction of one-dimensional scan, a shape of illumination flux for exposure is made rectangular on said mask to be illuminated thereby, and an intensity distribution of said rectangular illumination flux has at least two peaks in the direction of one-dimensional scan.

5. A projection exposure apparatus comprising:

a projection optical system for projecting a pattern formed in a transfer region on a mask, onto an area to be exposed on a photosensitive substrate;

a mask stage for one-dimensionally moving said mask over a region beyond a width of said transfer region in a direction of movement;

a substrate stage for one-dimensionally moving said photosensitive substrate in the direction of one-dimensional movement of said mask stage at a speed synchronized with a movement speed of said mask stage;

illumination means for illuminating said mask with an illumination flux for exposure, having a shape between a rectangular and a slit within the projection field of said projection optical system and having an approximately constant width in the direction of one-dimensional movement;

a substrate holder the holding said photosensitive substrate on said substrate stage with a predetermined inclination angle with respect to the direction of one-dimensional movement of illuminated area formed by said illumination flux on said photosensitive substrate through said mask and said projection optical system;

holder drive means for moving said substrate holder in a direction of optical axis of said projection optical system such that a central part of said illuminated area on said photosensitive substrate is located near a best focal plane of said projection optical system; and

control means for controlling said holder drive means to maintain an imaging condition of pattern image of said mask on said photosensitive substrate with correspondence to a position in said illuminated area in the direction of one-dimensional movement while scan exposure of pattern of said mask is effected on said area to be exposed.

6. A scanning exposure apparatus in which a substrate is exposed with an energy beam by moving a mask and the substrate relative to the energy beam, comprising:

a projection system, arranged in a path of the energy beam, which is a reduction system to project a reduction image of a pattern formed on the mask onto the substrate;

an illumination system, disposed along the path of the energy beam, which distributes the energy beam within a specified rectangular region in an image field of the projection system;

a first stage, disposed on an object surface side of the projection system, which is movable in a first direction while holding the mask during scanning exposure of the substrate;

a first interferometer which measures positional information of the first stage in the first direction during the scanning exposure;

a second stage, disposed on an image surface side of the projection system, which is movable in a second direction while holding the substrate during the scanning exposure, wherein a width of the specified region in the second direction is shorter than a length of the specified region in a direction perpendicular to the second direction;

a second interferometer which measures positional information of the second stage in the second direction during the scanning exposure;

a driving unit connected to the first and second interferometers to adjust a relative relationship between the mask and the substrate based on the positional information measured by the first and second interferometers during the scanning exposure, said driving unit including a first driving system to move the first stage and a second driving system to move the second stage, wherein the first stage is moved at a first speed and the second stage is moved at a second speed which is different from the first speed and a ratio between the first speed and the second speed is determined in accordance with a reduction magnification of the projection system; and

a reflection member, disposed in the path of the energy beam within the illumination system, which reflects the energy beam, wherein the second direction is substantially parallel to a plane which includes an axis of the energy beam incident on the reflection member and an axis of the energy beam reflected by the reflection member.

7. An apparatus according to claim 6, wherein:

the projection system is a telecentric reduction system and is composed only of refraction elements;

an optical axis of the projection system is substantially in one straight line; and

the specified rectangular region extends in the direction substantially perpendicular to the second direction.

8. An apparatus according to claim 6, wherein:

the second interferometer measures rotational information of the second stage during the scanning exposure.

9. An apparatus according to claim 6, further comprising:

a first reference reflection surface for the measurement of the first interferometer, arranged on a barrel of the projection system; and

a second reference reflection surface for the measurement of the second interferometer, arranged on the barrel of the projection system.

10. An apparatus according to claim 6, wherein;

a reference member, disposed on the second stage, which defines a relationship between the positional information measured by the first interferometer and the positional information measured by the second interferometer.

11. An apparatus according to claim 6, wherein:

speed information of the first and second stages is detected during the scanning exposure.

12. An apparatus according to claim 6, wherein

the projection system is composed only of refraction elements; and

an optical axis of the projection system is substantially in one straight line.

13. An apparatus according to claim 12, wherein:

the projection system is telecentric on the image surface side and on the object surface side.

14. An apparatus according to claim 6, wherein:

the specified rectangular region extends in the direction substantially perpendicular to the second direction.

15. An apparatus according to claim 6, further comprising:

an optical integrator disposed in the path of the energy beam in the illumination system; and wherein

the reflection member is disposed between the optical integrator and the projection system.

16. An apparatus according to claim 15, wherein:

the optical integrator includes a fly-eye lens system.

17. An apparatus according to claim 6, wherein:

the axis of the energy beam incident on the reflection member is substantially parallel to the second direction.

18. A scanning exposure method in which a substrate is exposed with an energy beam by moving a mask and the substrate relative to the energy beam, comprising:

providing the mask on a first stage;

providing the substrate on a second stage;

moving the first stage in a first direction during scanning exposure;

moving the second stage in a second direction during the scanning exposure, wherein the first stage and the second stage are moved at respective speeds which are different from each other, and the respective speeds are determined in accordance with a reduction magnification of a projection system that projects a reduction image of a pattern formed on the mask onto the substrate;

distributing the energy beam within a specified rectangular region in an image field of the projection system through a reflection member of an illumination system, wherein a width of the specified region in the second direction is shorter than a length of the specified region in a direction perpendicular to the second direction, and wherein the second direction is substantially parallel to a plane which includes an axis of the energy beam incident on the reflection member and an axis of the energy beam reflected by the reflection member;

measuring, during the scanning exposure, positional information in the first direction and rotational information of the first stage using a first interferometer;

measuring, during the scanning exposure, positional information in the second direction of the second stage using a second interferometer; and

adjusting, during the scanning exposure, a positional relationship between the mask held on the first stage and the substrate held on the second stage based on the positional information and the rotational information measured by the first interferometer and the positional information measured by the second interferometer.

19. A method according to claim 18, wherein:

the projection system is a telecentric reduction system and is composed only of refraction elements;

an optical axis of the projection system is substantially in one straight line; and

the specified rectangular region extends in the direction substantially perpendicular to the second direction.

20. A method according to claim 18, further comprising:

measuring rotational information of the second stage using the second interferometer during the scanning exposure.

21. A method according to claim 18, wherein:

a first reference reflection surface for the measurement of the first interferometer is arranged on a barrel of the projection system; and

a second reflection surface for the measurement of the second interferometer is arranged on the barrel of the projection system.

22. A method according to claim 18, further comprising:

defining a relationship between the positional information measured by the first interferometer and the positional information measured by the second interferometer using a reference member on the second stage.

23. A method according to claim 18, further comprising:

detecting speed information of the first and second stages during the scanning exposure.

24. A method according to claim 18, wherein:

the projection system is composed only of refraction elements; and

an optical axis of the projection system is substantially in one straight line.

25. A method according to claim 24, wherein the projection system is telecentric on the image surface side and the object surface side.

26. A method according to claim 18, wherein:

the specified rectangular region extends in the direction substantially perpendicular to the seond direction.

27. A method according to claim 18, wherein the reflection member is disposed between an optical integrator and the projection system.

28. A method according to claim 27, wherein the optical integrator includes a fly-eye lens system.