HK1130091B - Exposure device of projector, exposure method and device manufacturing method - Google Patents

Description

Projector exposure apparatus, exposure method, and device manufacturing method

The present application is filed as a divisional application entitled "projector exposure apparatus, exposure method, and device manufacturing method" in original application No. 200480031414.5 (international application No. PCT/JP2004/015853, international application date 2004, 10/26).

Technical Field

The present invention relates to an exposure technique used in a photolithography (L: lithography) process for manufacturing various devices such as a semiconductor integrated circuit (LSI) device, an imaging device, a liquid crystal display, and the like, and more particularly, to an exposure technique for illuminating a mask pattern with light of a predetermined polarization state. The present invention also relates to a device manufacturing technique using the exposure technique.

Background

In forming a fine pattern of an electronic component such as a semiconductor integrated circuit or a liquid crystal display, an exposure transfer method is used in which a pattern of a reticle (or a photomask) of a mask (to be formed) is enlarged by 4 to 5-fold degrees so as to reduce the size of the pattern on a wafer (or a glass plate) of a substrate (a photosensitive body) to be exposed by a projection optical system. In the exposure transfer, a projection exposure apparatus of a stationary exposure type such as a stepper (stepper) or a scanning exposure type such as a scanning stepper (scanning stepper) is used. The resolution of the projection optical system is proportional to the exposure wavelength divided by the Number of Apertures (NA) of the projection optical system. The Number of Apertures (NA) of the projection optical system is obtained by multiplying the sine (Sin) of the maximum incident angle of the illumination light for exposure to the wafer by the refractive index of the medium through which the light beam passes.

Therefore, the exposure wavelength of the projection exposure apparatus is becoming shorter in order to cope with the miniaturization of semiconductor integrated circuits and the like. Although 248nm of KrF excimer laser (eximer laser) is the mainstream of the exposure wavelength at present, 193nm of ArF excimer laser with shorter wavelength is coming into practical use, but 157nm F with shorter wavelength is also used₂Laser and Ar with a wavelength of 126nm₂A projection exposure apparatus using a so-called vacuum ultraviolet exposure light source such as a laser beam has been proposed. In addition to the reduction in the wavelength, the increase in the resolution is also possible due to the increase in the number of openings of the projection optical system (increase in NA), and development is being made to increase the NA of the projection optical system further, and the NA of the projection optical system at the forefront of the present time is about 0.8.

On the other hand, in order to improve the resolution of a transferred pattern, a so-called phase Shift grating (Shift grating) method, a so-called super-resolution technique such as ring illumination, dipole illumination, and quadrupole illumination, which controls the distribution of the incident angle of illumination light to the grating to a predetermined distribution, is also put into practical use in a projection optical system using the same exposure wavelength and the same NA.

Among these, the ring-shaped illumination is an illumination in which the incident angle range of the illumination light on the grating is limited to a predetermined angle, that is, the distribution of the illumination light on the pupil plane of the illumination optical system is limited to the ring-shaped region centered on the optical axis of the illumination optical system, whereby the resolution and the depth of focus can be improved (see, for example, japanese patent laid-open No. 61-91662). On the other hand, when the two-pole illumination and the four-pole illumination are not only in the incident angle range but also in the case of a pattern having a specific directivity on the grating, the incident direction of the illumination light is also limited to the direction corresponding to the directivity of the pattern, so that the resolution and the depth of focus are greatly improved (see, for example, japanese patent laid-open No. 4-101148, U.S. patent No. 6233041 corresponding thereto, japanese patent laid-open No. 4-225357, or U.S. patent No. 6211944 corresponding thereto).

Further, attempts have been made to optimize the polarization state of the illumination light with respect to the pattern direction on the grating, thereby improving the resolution and the depth of focus. This method is to enhance Contrast (Contrast) of a transferred image by linearly polarizing illumination light having a polarization direction (electric field direction) in a direction orthogonal to a periodic direction of a pattern, that is, in a direction parallel to a longitudinal direction of the pattern. For example, see patent document 1 and non-patent document 1).

In the annular illumination, attempts have been made to match the polarization direction of illumination light with the circumferential direction thereof in the annular region where the illumination light is distributed on the pupil plane of the illumination optical system, thereby improving the resolution, contrast, and the like of a projected image.

[ patent document 1 ] Japanese patent laid-open No. Hei 5-109601 publication.

[ non-patent document 1 ] Thiothy A.Brunner, et al: "High NA Lithographiming at Brewster's Angel", SPIE

(U.S. Vol.4691, pp.1-24 (2002)).

In the case of performing the annular band illumination as in the above-described conventional technique, when the polarization state of the illumination light is linearly polarized in the pupil plane of the illumination optical system so as to substantially coincide with the circumferential direction of the annular band region, the loss of the illumination light amount increases, which causes a problem of lowering the illumination efficiency.

In the detailed description, the illumination light emitted from the recently mainstream banded KrF excimer laser light source is linearly polarized light of the same type. When the light is guided to the grating in the polarization state as it is, the grating is illuminated with the same linearly polarized light, and thus, it is not possible to realize linearly polarized light that matches the circumferential direction of the annular zone region of the pupil plane of the illumination optical system.

Therefore, in order to realize the above-described polarization state, it is necessary to use a method of selecting a desired polarization component from illumination light composed of arbitrary polarization by using a polarization selection element such as a polarization filter (filter) or a polarization beam splitter (beam splitter) in each portion of the annular region after converting linearly polarized light emitted from a light source into arbitrarily polarized light at one time. In this method, only the energy contained in the predetermined linearly polarized light component, that is, almost half of the energy of the arbitrarily polarized illumination light (energy) can be used as the illumination light to the reticle, and there is a problem that the loss of the illumination light amount is large, and further the loss of the exposure capability (power) to the wafer is large, and the throughput (throughput) of the exposure apparatus is reduced.

Similarly, in the case of multipole illumination using dipole illumination or quadrupole illumination, there is a problem that illumination efficiency is lowered when the polarization state of illumination light in the dipole or quadrupole region is set to a predetermined state at the pupil plane of the illumination optical system.

Disclosure of Invention

The present invention has been made in view of the above problems, and a first object of the present invention is to provide an exposure technique capable of reducing light loss when a mask such as a reticle is illuminated with illumination light in a predetermined polarization state.

Further, a second object of the present invention is to provide an exposure technique capable of reducing a decrease in the amount of illumination light when the polarization state of illumination light in the field of an annulus, a dipole, a quadrupole, or the like on the pupil plane of an illumination optical system is set to a predetermined state, and as a result, improving the resolution without substantially decreasing the throughput.

Further, the present invention also provides a device manufacturing technique for manufacturing a high-performance device with high throughput by using the exposure technique.

The following symbols with reference symbols after the respective members of the present invention correspond to the embodiments of the present invention. Therefore, the reference numerals are merely examples of the members, and the members are not limited to the configurations of the embodiments.

The first projection exposure apparatus according to the invention is a projection exposure apparatus comprising an illumination optical system (ILS) and a projection optical system (25), wherein the illumination optical system (ILS) irradiates a first object (R) with illumination light from the light source (1), the projection optical system (25) projects a pattern image on the first object onto a second object (W), the light source generates illumination light in a substantially single polarization state, and the illumination optical system has a plurality of birefringent members (12, 13) arranged along the proceeding direction of the illumination light, and the direction of the phase axis of at least one birefringent member among the plurality of birefringent members is different from the direction of the phase axis of the other birefringent member, in the illumination light, the specific illumination light irradiated to the first object in a specific incident angle range is made into a polarized light state having S-polarized light as a main component.

A first illumination optical device according to the present invention is an illumination optical device for irradiating a first object (R) with illumination light from a light source (1), and has a plurality of birefringent members (12, 13) arranged along the proceeding direction of the illumination light, and the direction of the phase advancing axis of at least one birefringent member among the plurality of birefringent members is different from the direction of the phase advancing axis of the other birefringent member, and in an illumination optical device in a substantially single polarization state supplied from the light source, the specific illumination light irradiated to the first object in a specific incident angle range is made to be light in a polarization state having S-polarization as a main component.

In the present invention, for example, by setting the distribution of the plurality of birefringent members to a predetermined distribution, the polarized light state of the illumination light emitted from the light source passing through the plurality of birefringent members can be made to be the main component, for example, in the annular region centered on the optical axis, the polarized light state in the circumferential direction centered on the optical axis can be made to be the main component. When the exit surfaces of the plurality of birefringent members are arranged at positions close to the pupil surface of the illumination optical system, for example, the illumination light (specific illumination light) in the annular band-shaped region can illuminate the first object in a predetermined polarization state in which S-polarization is the main component, with almost no light loss.

In this case, the illumination light applied to the first object may be limited to the specific illumination light by the light flux limiting members (9a, 9 b). Thus, the first object is illuminated under the condition of illumination by the approximate annular band. In the ring-shaped illumination, when the illumination light on the first object is approximately S-polarized, the projection image of the line and space (line and space) pattern arranged at a fine pitch in an arbitrary direction on the first object is imaged by the illumination light in the longitudinal direction parallel to the line pattern in the main polarization direction, and thus the imaging characteristics of contrast, resolution, and depth of focus can be improved.

Further, the light flux restriction member may restrict the incident direction of the illumination light to be irradiated to the first object to a plurality of specific substantially discrete directions. Thus, the imaging characteristics of the space pattern can be improved by performing illumination with dipole illumination or quadrupole illumination, and the like, and by arranging lines at a fine pitch in a predetermined direction.

The second projection exposure apparatus according to the present invention is a projection exposure apparatus including an illumination optical system (ILS) for irradiating a first object (R) with illumination light from a light source (1), and a projection optical system (25) for projecting a pattern image on the first object onto a second object (W), the light source generating illumination light substantially in a single polarization state, the illumination optical system including a plurality of birefringent members (12, 13) arranged along a proceeding direction of the illumination light, and a direction of a phase-advancing axis of at least one birefringent member among the plurality of birefringent members being different from directions of phase-advancing axes of the other birefringent members, and passing at least a part of an illumination region within a specific annular region (36) of the specific annular region centered on an optical axis of the illumination optical system in a pupil plane or a vicinity thereof in the illumination optical system The light is polarized in a polarized state mainly composed of linearly polarized light in which the circumferential direction of the specific zone region is the polarized direction.

In the case of the above-described invention, for example, by setting the distribution of the plurality of birefringent members to a predetermined distribution, the polarization state of at least a part of the illumination light passing through the specific annular zone of the illumination light emitted from the light source is in a state in which there is almost no loss of light amount, and the state is a state (predetermined polarization state) in which the main component is linearly polarized light with the circumferential direction of the specific annular zone as the polarization direction.

In this case, the illumination light applied to the first object may be substantially limited to the light beam distributed in the specific zone region by the light beam limiting members (9a, 9 b). The light beam restriction member may further restrict the light beam to a plurality of substantially discrete regions within a specific loop region. In such cases, the amount of illumination light is hardly reducedAnd annular illumination, dipolar illumination or quadrupole illumination and the like can be realized.

An example of the light flux restriction member is a diffractive optical element disposed between the light source (1) and the plurality of birefringent members (12, 13). By using the diffractive optical element, the light quantity loss can be further reduced.

In addition, as an example, at least one of the plurality of birefringent members is an inhomogeneous wavelength plate (12, 13) in which the position of the member changes nonlinearly, and which is an inter-polarization phase difference that is a phase difference given between a linearly polarized component parallel to the phase-in axis and a linearly polarized component parallel to the phase-out axis in transmitted light. Thus, the polarization state of the illumination light passing through the plurality of birefringent members can be controlled to a predetermined state with high accuracy.

In this case, the non-uniform wavelength plate may include a first non-uniform wavelength plate (12) for giving a phase difference with respect to polarization having a quadratic rotation symmetry about an optical axis of the illumination optical system to the specific illumination light or the illumination light distributed in the specific ring zone region.

The non-uniform wavelength plate may further include a second non-uniform wavelength plate (13) for giving a phase difference with respect to the polarized light having a primary rotational symmetry about the optical axis of the illumination optical system, with respect to the specific illumination light or the illumination light distributed in the specific ring zone region.

In addition, as an example, the first and second non-uniform wavelength plates are separated from each other by 45 ° in the direction of the phase advance axis with the optical axis of the illumination optical system as the rotation center. Thus, the polarization state of the illumination light passing through the two non-uniform wavelength plates can be easily controlled.

Further, an optical integrator (14) may be further included, the optical integrator being disposed between the plurality of birefringent members and the first object. Therefore, the uniformity of the illumination distribution on the first object can be improved.

Further, a zoom (zoom) optical system, a group of variable-interval conical prisms (41, 42), or a group of polygonal prisms may be further included between the birefringent members and the optical integrator. The size and position of the special zone (or the zone or a substantially discrete partial zone thereof) can be controlled by using the zoom optical system according to the pattern pitch on the first object, thereby improving the imaging characteristics when exposing various pitch patterns.

The optical integrator is exemplified by fly-eye lenses (fly-eye lenses).

The illumination device may further include a polarization control member (4) disposed between the light source and the plurality of birefringent members to change the polarization state of the illumination light from the light source. Thus, the polarization state of the illumination light from the light source can be converted into a polarization state suitable for the plurality of birefringent members without light loss.

Further, the illumination optical system may have a rotating mechanism for rotating a part or all of the plurality of birefringent members around the optical axis of the illumination optical system, so that the illumination light from the light source can be supplied to the birefringent members in a polarization state suitable for the birefringent members.

Further, the illumination optical system may include a birefringence exchange mechanism having a plurality of sets of the plurality of birefringence members, and the plurality of sets of the plurality of birefringence members may be disposed in the illumination optical system so as to be exchangeable. This makes it possible to correspond to various patterns of the transfer target.

The third projection exposure apparatus according to the present invention is a projection exposure apparatus including an illumination optical system (ILS) that irradiates a first object (R) with illumination light from a light source (1), and a projection optical system (25) that projects a pattern image on the first object onto a second object (W), the light source generating illumination light substantially in a single polarization state, the illumination optical system having diffractive optical elements (9a, 9b) and birefringent members (12, 13) arranged along a proceeding direction of the illumination light.

The second illumination optical device according to the present invention is an illumination optical device that irradiates illumination light from a light source (1) to a first object (R), and has a diffractive optical element and a birefringent member arranged along the proceeding direction of the illumination light.

According to the present invention, the first object can be efficiently illuminated in a predetermined polarization state, for example, in which S-polarization is the main component, by using the birefringent member. Further, by limiting the light distribution of the illumination light incident on the birefringent member using the diffractive optical element to a ring shape, the loss of the light amount can be made extremely small.

In the present invention, as an example, the diffractive optical element substantially limits the illumination light applied to the first object to the specific illumination light applied to the first object in a specific incident angle range, and the birefringent member causes the specific illumination light to be in a polarized state having S-polarized light as a main component.

The diffractive optical element further restricts the incident direction of the illumination light to be irradiated on the first object to a plurality of directions having a specific substantial dispersion.

The diffractive optical element substantially restricts the illumination light to a light flux within a specific annular zone distributed in a predetermined annular zone around an optical axis of the illumination optical system in a pupil plane of the illumination optical system, and the birefringent member causes the light flux to be in a polarized state in which linearly polarized light having a circumferential direction as a polarization direction is a main component.

The diffractive optical element may restrict the light flux to a plurality of substantially discrete regions within the specific annular region.

Next, according to the exposure method of the present invention, the projection exposure apparatus of the present invention is used to expose the photoreceptor (W) as the second object with the pattern image of the mask (R) of the first object. The present invention can illuminate the first object by ring illumination, dipole illumination, quadrupole illumination, or the like, and can make the polarization state of the illumination light incident on the first object substantially S-polarized as a main component. Therefore, a pattern formed at a fine pitch in a predetermined direction on the mask can be transferred with good image forming characteristics with almost no light loss.

The method for manufacturing a device according to the present invention is a method for manufacturing a device including a photolithography (lithography) process in which a pattern is transferred to a photoreceptor by using the exposure method of the present invention. According to the present invention, a pattern can be transferred with high throughput and high image formation characteristics.

According to the present invention, the relationship of the polarization state of the illumination light is controlled by using the plurality of birefringent members, and the loss of the amount of light when the first object (mask) is illuminated with the illumination light of a predetermined polarization state can be reduced.

Further, by further using the light flux restriction member, the polarization state of the illumination light passing through at least a part of the specific annular region can be set to a state mainly composed of linearly polarized light parallel to the circumferential direction of the specific annular region, with the amount of illumination light hardly decreased, when the first object is illuminated by annular illumination, dipole illumination, quadrupole illumination, or the like.

In this case, the image formation characteristic can be improved when exposing a pattern in which line patterns having a longitudinal direction along the direction of linearly polarized light on the first object are arranged at fine intervals. Therefore, it is possible to provide an illumination optical apparatus, a projection exposure apparatus, and an exposure method which can improve the image forming characteristics without reducing the throughput.

In order to make the aforementioned and other objects, features and advantages of the invention comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

Fig. 1 is a sectional view of a projection exposure apparatus according to an embodiment of the present invention, with a schematic configuration cut away.

Fig. 2A is a view of the birefringent member 12 in fig. 1, taken toward the + Y direction, and fig. 2B is a sectional view taken along line AA' of fig. 2A.

Fig. 3A is a view of the birefringent member 13 in fig. 1, viewed in the + Y direction, and fig. 3B is a cross-sectional view taken along line BB' of fig. 3A.

Fig. 4A is a diagram showing an example of a relationship between the inter-polarization phase difference Δ P1 and the position X in the first birefringent member 12, fig. 4B is a diagram showing an example of a relationship between the inter-polarization phase difference Δ P2 and the position XZ in the second birefringent member 13, and fig. 4C is a diagram showing an example of a polarization state of illumination light emitted from the second birefringent member 13.

Fig. 5 is a diagram showing an example of the polarization state of the illumination light emitted from the first birefringent member 12.

Fig. 6A is a diagram showing another example of the relationship between the inter-polarization phase difference Δ P1 and the position X in the first birefringent member 12, fig. 6B is a diagram showing another example of the relationship between the inter-polarization phase difference Δ P2 and the position XZ in the second birefringent member 13, and fig. 6C is a diagram showing another example of the polarization state of the illumination light emitted from the birefringent member 13.

Fig. 7A is a plan view showing an example of the fine periodic pattern PX formed on the grating R of fig. 1, fig. 7B is a diagram showing a diffraction light distribution pattern formed in the pupil plane 26 of the projection optical system when the pattern of fig. 7A is illuminated under a predetermined condition, and fig. 7C is a diagram showing a condition of the annular band illumination illuminating the pattern PX of fig. 7A.

Fig. 8A is a perspective view showing a relationship between the pupil plane 15 and the grating R of the illumination optical system ILS in fig. 1 in a simplified manner, fig. 8B is a view of a part of fig. 8A as viewed in the + Y direction, and fig. 8C is a view of a part of fig. 8A as viewed in the-X direction.

Fig. 9 is a view showing a plurality of conical prisms which can be arranged between the birefringent members 12 and 13 and the fly-eye lens 14 in fig. 1, in order to change the radius of the specific zone region, in an example of the embodiment of the present invention.

Fig. 10 is a diagram showing an example of a polarization control optical system that can be disposed at the position of the polarization control member 4 in fig. 1.

Fig. 11 is a diagram showing an example of a photolithography process for manufacturing a semiconductor device using the projection exposure apparatus according to the embodiment of the present invention.

R: grating

W: wafer

ILS: illumination optical system

AX 2: optical axis of illumination system

nf: phase advancing shaft

ns: phase delay shaft

1: exposure light source

4: polarization control member

9a, 9 b: diffractive optical element

12: a first birefringent member

13: a second birefringent member

14: fly's eye lens

25: projection optical system

36: field of specific endless belts

41. 42: conical prism

Detailed Description

[ embodiment ] A method for producing a semiconductor device

Hereinafter, an example of a preferred embodiment of the present invention will be described with reference to the drawings. The present inventors applied to the case where the exposure is performed by a scanning exposure type projection exposure apparatus (scanning stepper) constituted by a step and scan system.

Fig. 1 is a cross-sectional view showing a part of a schematic configuration of the projection exposure apparatus of the present embodiment, and in fig. 1, the projection exposure apparatus of the present embodiment includes an illumination optical system ILS and a projection optical system 25. The former illumination optical system ILS includes a plurality of optical members (described later in detail) arranged along an optical axis (illumination system optical axis) AX1, AX2, AX3 from an exposure light source 1 (light source) to a focusing lens (condenser lens)20, and illuminates an irradiation field of view on a pattern surface (grating surface) of a grating R of a mask with uniform illuminance distribution using an exposure beam (beam) from the exposure light source 1 as an exposure illumination light (exposure light) IL. The latter projection optical system 25 projects an image in which the pattern in the illumination field of the grating R is reduced at a projection magnification M (M is a reduction magnification of 1/4, 1/5, for example) under the illumination light onto an exposure field in a shot (shot) field on a substrate (substrate) to be exposed or a resist (photoresist) coated wafer W as a photosensitive body. The grating R and the wafer W can be considered as a first object and a second object, respectively. The wafer W is a disk-shaped substrate having a diameter of about 200 to 300mm, such as a semiconductor (silicon or the like) or a silicon-on-insulator (soi) (silicon on insulator). The projection optical system 25 of this example is, for example, a refractive optical system, and a catadioptric system may be used.

In fig. 1, the projection optical system 25, the grating R, and the wafer W are described with the optical axis AX4 parallel to the projection optical system 25 being taken as the Z axis, the scanning direction (direction parallel to the paper surface of fig. 1) along the grating R and the wafer W at the time of scanning exposure being taken as the Y axis, and the non-scanning direction (direction perpendicular to the paper surface of fig. 1) being taken as the X axis in the plane (XY plane) perpendicular to the Z axis. In this case, the illumination field of the grating R is an elongated region in the X direction in the non-scanning direction, and the exposure field on the wafer W is an elongated region conjugate to the illumination field. The optical axis AX4 of the projection optical system 25 coincides with the illumination system optical axis AX3 on the grating R.

First, a pattern-formed grating R to be exposed and transferred is held by suction on a grating stage (grating) 21, and the grating stage 21 is moved in the Y direction at a constant speed on a grating base (grating base)22 and is moved finely in the X axis direction, the Y axis direction, and the rotation direction around the Z axis so as to correct a synchronization error, thereby scanning the grating R. The position and the rotation angle of the grating machine 21 in the X direction and the Y direction are measured by a movable mirror 23 and a laser interferometer (laser interferometer)24 provided on the upper surface thereof. Based on the measured value and control information from the main control system 34, the reticle stage driving system 32 controls the position and speed of the reticle stage 21 by a driving mechanism (not shown) such as a linear motor. A grating alignment microscope (not shown) for grating alignment (correction) is disposed above the peripheral portion of the grating R.

On the other hand, the wafer is held by suction by a wafer holder (not shown) on the wafer stage 27, and the wafer stage 27 is placed on the wafer base 30 so as to be movable in the Y direction at a constant speed and movable in the X direction and the Y direction in a stepwise manner. A Z leveling mechanism is also incorporated in the wafer stage 27 to align the surface of the wafer W with the image plane of the projection optical system 25 based on the measurement value of a not-shown focus sensor (not shown). The position and rotation angle of the wafer stage 27 in the X-direction and Y-direction are measured by a movable mirror 28 and a laser interferometer 29 provided thereon. Based on the measured value and control information from the main control system 34, the wafer stage driving system 33 controls the position and speed of the wafer stage 27 by a driving mechanism (not shown) such as a linear motor. In the vicinity of the projection optical system 25, a positioning sensor (Alignment sensor)31 of an off-axis (off-axis) system, for example, a Field Image Alignment (Field Image Alignment) system, is disposed to detect the position of a mark for Alignment on the wafer W for wafer positioning.

In the projection exposure apparatus according to the present embodiment, before exposure, the reticle R is positioned by the reticle positioning microscope, and the wafer W is positioned by detecting the position of the alignment mark formed on the wafer W by the previous exposure process together with the circuit pattern by the positioning sensor 31. Then, the reticle stage 21 and the wafer stage 27 are driven in a state where the illumination field on the reticle R is illuminated with the illumination light IL, and the reticle R and one imaging region on the wafer W are scanned in the Y direction in synchronization with each other, and the wafer stage 27 is driven to move the wafer W in the X direction and the Y direction in steps by stopping the light emission of the illumination light 1L, and the above two operations are repeated. The scanning speed ratio of the reticle stage 21 and the wafer stage 27 during the synchronous scanning is equal to the projection magnification M of the projection optical system 25 in order to maintain the image formation relationship between the reticle R passing through the projection optical system 25 and the wafer W. With these operations, the pattern image of the transfer grating R is exposed to the entire image-taking area on the wafer W by the step-and-scan method.

Next, the configuration of the illumination optical system ILS of the present invention will be described in detail. In FIG. 1, an ArF (argon fluorine) excimer laser (wavelength 193nm) is used as an exposure light source 1 for this example.In addition, other KrF (KrF) excimer laser (wavelength 248nm) and F may be used as the exposure light source 1₂(molecular fluorine) laser (wavelength 157nm), or kr₂And (krypton molecule) laser (wavelength 146 nm). These laser light sources (including the exposure light source 1) are narrow-banded laser light or wavelength-selective laser light, and the illumination light IL emitted from the exposure light source 1 is in a polarized state mainly composed of linearly polarized light due to the above-mentioned narrowing or wavelength selection. In fig. 1, the illumination light IL just emitted from the exposure light source 1 will be described with linearly polarized light as a main component, the polarization direction (electric field direction) of which coincides with the X direction in fig. 1.

Illumination light IL emitted from the exposure light source 1 is incident on a polarization control means 4 (described later in detail) of a polarization control means through relay lenses (relay lenses) 2 and 3 along an illumination system optical axis AX 1. Illumination light IL emitted from the polarization control mechanism 4 passes through zoom (zoom) Optical systems (5, 6) constituted by a combination of a concave lens 5 and a convex lens 6, is reflected by a mirror 7 for Optical path bending, and is incident on a Diffractive Optical Element (DOE) 9a along an Optical axis AX2 of the illumination system, the Diffractive Optical Element 9a is constituted by a phase type diffraction lattice, and the incident illumination light IL is diffracted to advance in a predetermined direction.

As will be described later, the diffraction angle and direction of each diffracted light from the diffractive optical element 9a of the beam restricting member correspond to the position of the illumination light IL on the pupil plane 15 of the illumination optical system ILS and the incident angle and direction of the illumination light IL on the grating R. A plurality of diffraction optical elements 9a and another diffraction optical element 9b having a different diffraction function are arranged on the tower-shaped member 8. However, for example, the component 8 is driven by the exchange mechanism 10 under the control of the main control system 34, and the diffraction optical system at an arbitrary position on the component 8 is configured so that the incident angle range and direction of the illumination light to the grating R (or the position of the illumination light to the pupil surface 15) can be set within a desired range in accordance with the pattern of the grating R by filling the position on the optical axis AX2 of the illumination optical system. The incident angle range is finely adjustable in a complementary manner by moving the concave lens 5 and the convex lens 6 constituting the zoom optical systems (5, 6) in the directions of the optical axis AX1 of the illumination optical system.

The illumination light (diffracted light) IL emitted from the diffractive optical system 9a passes through the relay lens 11 along the illumination system optical axis AX2 to be sequentially incident on the first birefringent member 12 and the second birefringent member 13 of the plurality of birefringent members of the present invention. Details of such birefringent members will be described later. In the present embodiment, a fly-eye lens (fly-eye lens)14 of an optical integrator (illumination uniformizing means) is disposed after the birefringent means 13. The illumination light IL emitted from the fly-eye lens 14 passes through the field stop 17 and the focusing lens 18 to the mirror 19 for optical path (optical pass) bending, and the illumination light IL reflected thereby passes through the relay lens 20 along the illumination system optical axis AX3 to illuminate the grating R. The pattern on the grating R thus illuminated is projected and transferred onto the wafer W by the projection optical system 25 as described above.

Further, the field diaphragm 17 may be of a scanning type as needed, and scanning may be performed in synchronization with scanning of the reticle stage 21 and the wafer stage 27. In this case, the field diaphragm may be divided into a fixed field diaphragm and a movable field diaphragm.

In this configuration, the exit side surface of the fly-eye lens 14 is positioned near the pupil surface 15 of the illumination optical system ILS. The pupil plane 15 functions as an optical Fourier (Fourier) conversion plane with respect to the pattern plane (grating plane) of the grating R by optical members (the relay lens 16, the field stop 17, the focusing lenses 18 and 20, and the mirror 19) in the illumination optical system ILS from the pupil plane 15 to the grating R. That is, the illumination light exiting one point of the exit pupil plane 15 is irradiated to the grating R at a predetermined incident angle and incident direction as approximately parallel light beams. The incident angle and the incident direction are determined according to the position of the light beam on the pupil plane 15.

Further, the optical path bending mirrors 7 and 19 are not necessarily required in terms of optical performance, and are disposed at appropriate positions in the illumination optical system ILS for the purpose of space saving, in such a manner that the overall height (height in the z direction) of the exposure apparatus increases when the illumination optical system is disposed on a straight line. The optical axis AX1 of the illumination system is reflected by the mirror 7 to coincide with the optical axis AX2 of the illumination system, and the optical axis AX2 of the illumination system is reflected by the mirror 19 to coincide with the optical axis AX3 of the illumination system.

A first embodiment of the first and second birefringent members 12 and 13 in fig. 1 will be described below with reference to fig. 2 to 5.

The first birefringent member 12 is a disc-shaped member made of a birefringent material such as uniaxial crystal, and the optical axis thereof is in the in-plane direction (parallel to a plane perpendicular to the optical axis AX2 of the illumination optical system), whereas the size (diameter) of the first birefringent member 12 in the plane thereof is larger than the beam diameter of the illumination light IL at the position where the birefringent member 12 is disposed.

Fig. 2A is a view of the birefringent member 12 of fig. 1 as viewed along the optical axis AX2 of the illumination system in the + Y direction, and as shown in fig. 2A, the phase advance axis nf in the axial direction in which the refractive index is lowest for linearly polarized light having a polarization direction parallel thereto in the birefringent member 12 is rotated by 45 ° from each coordinate (x-axis and z-axis) in the xz coordinate direction of the same coordinate axis as that of fig. 1. The slow axis ns in the axial direction in which the refractive index is highest for linearly polarized light having a polarization direction parallel thereto is a direction rotated by 45 ° from both the x axis and the z axis, as well as being orthogonal to the phase advancing axis nf.

The first birefringent member 12 is not uniform in a plane parallel to the paper surface of fig. 2A, and changes according to the X coordinate (position in the X direction). Fig. 2B is a cross-sectional view of the birefringent member 12 taken along line AA' of fig. 2A, and as shown in fig. 2B, the birefringent member 12 is thinner at the center (illumination system optical axis) and thicker at the periphery in the X direction. On the other hand, the first birefringent member 12 has the same shape in the z direction in fig. 2A, and the birefringent member 12 has a shape as a negative cylindrical lens (cylinder lens) as a whole.

In general, a light beam transmitted through such a birefringent member has an optical path difference (i.e., an inter-polarization phase difference) between a linearly polarized light component having a polarization direction (i.e., "the vibration direction of the electric field of light", the same applies hereinafter) aligned with the phase advance axis nf and a linearly polarized light component having a polarization direction aligned with the phase delay axis ns. On the other hand, the refractive index of the birefringent member is high for linearly polarized light parallel to the retardation axis ns, so that the traveling speed of the linearly polarized light is delayed, and an optical path difference (phase difference between polarized lights) is generated between the two polarized lights. Therefore, the first birefringent member 12 functions as a first inhomogeneous wavelength plate which imparts a phase difference between different polarizations to the transmitted light according to the location.

However, when the optical path length difference generated by the birefringent member 12 is an integral multiple of the wavelength by optimizing the first birefringent member 12, the phases of the two light beams are substantially indistinguishable and no optical path length difference is formed. In this example, T1 at the center of the birefringent member 12 is set as described above. Hereinafter, as shown in fig. 2B, the origin of the X axis (X ═ 0) is set as the center of the birefringent member 12 (the optical axis of the irradiation system).

On the other hand, the shape of the birefringent member 12 is set so that the inter-polarization phase difference of 0.5 (unit is the wavelength of the illumination light) is generated only at a position ± 1 in the X direction from the center of the first birefringent member 12 (1 is a reference length, inside the outer diameter of the first birefringent member 12). With such a shape, the position X in the X direction of the TA of the birefringent member 12 is expressed as the following function in this example.

TA＝T1+α×(1.7×X⁴-0.7×X²)......(1)

Here, α is a proportionality coefficient, and the value of α is different from T1 in the central portion, for example, by the difference in refractive index between the phase-in axis and the phase-out axis of the birefringent material used.

When crystal of uniaxial crystal is used as the birefringent material constituting the first birefringent member 12, the refractive index of crystal is 1.6638 for ordinary rays and 1.6774 for extraordinary rays of ArF excimer laser light having a wavelength of 193 nm. The phase-in axis is the deflection direction of the normal light, and the phase-out axis is the polarization direction of the abnormal light.

The wavelengths of ordinary and extraordinary rays in the crystal are 116.001nm and 115.056nm, respectively, in the relationship of dividing the wavelength (193nm) in vacuum by the refractive indices, and a path length difference of 0.945nm is formed between the two light fluxes for each wavelength division in the crystal. Therefore, when the 122.7(═ 166.001/0.945) wavelength division is performed, a one-wavelength-division optical path difference is formed between the two light beams. When the optical path difference is exactly one wavelength or an integer wavelength, the two light beams are substantially equivalent to the light beam with no optical path difference. The 122.7 wavelength division crystal is equivalent to 14239nm calculated from 122.7X 193/1.6638, i.e., 14.239 μm. Similarly, if the ordinary ray and the extraordinary ray have an optical path difference of half a wavelength, the crystal should be 7.12 μm which is half the above-mentioned wavelength.

Thus, when the first birefringent member 12 of the first inhomogeneous wavelength plate is formed of crystal, T1 at the center of expression (1) is set to an integral multiple of 14.239 μm, and the thickness of each of the reference positions (X is 1) near the periphery may be increased by 7.12 μm, even if the proportionality coefficient is set to 7.12 μm.

At this time, the inter-polarization phase difference Δ P1 formed by the first birefringent member 12 is expressed as a function of the position X in the X direction as follows.

△P1＝0.5×(1.7×X⁴-0.7×X²)............(2)

In addition, in the first birefringent member 12, if the distance between the incident surface 12a and the output surface 12b satisfies the positional relationship with the X direction that forms the phase difference, the shapes of the incident surface 12a and the output surface 12b may be arbitrary. However, in the processing of surface formation, processing is easy when any one surface is a flat surface, and it is actually desirable to make the emission surface 12B a flat surface as shown in fig. 2B. In this case, the TA value of the incident surface 12a when the TA value of the emission surface 12b is 0 is the TA as obtained by expression (1). Of course, the incident surface 12a may be a flat surface.

Fig. 4A is a diagram showing a relationship between the inter-polarization phase difference Δ P1 (unit is the wavelength of the illumination light) expressed by expression (2) and the position X. Fig. 5 is a view showing the polarization state of the illumination light emitted from the first birefringent member 12 of this example, and in fig. 5, the polarization states of the illumination light distributed at each position on the XZ coordinate are shown by line segments, circles, and ellipses each having a position as a center. The origin (X is 0, Z is 0) of the X-axis and the Z-axis in fig. 5 is set at the center of the birefringent member 12, and the scale in the Z-direction in the X-direction is set so that the positions of X ± and Z ± 1 (the positions of the reference length from the origin X is 0 and Z is 0 in common) are located at four corners in fig. 5.

At the position identified by each XZ coordinate in fig. 5, the illumination light at the position indicated by the line segment is in a polarized state in which the linear polarization is the main component, and the direction of the line segment indicates the polarization direction. The illumination light at the position indicated by the ellipse is in a polarized state in which the principal component is elliptically polarized light, and the longitudinal direction of the ellipse is a direction indicating that the linearly polarized light component contained in the elliptically polarized light is the largest. The illumination light at the position indicated by the circle is in a polarized state having circular polarized light as a main component.

As shown in fig. 4A, the first birefringent member 12 functions as a so-called 1/2 wavelength plate at a position ± 1 away from the center in the X direction. Here, the illumination light IL emitted from the exposure light source of fig. 1 is mainly composed of linearly polarized light polarized in the X direction as described above, and the 1/2 wavelength plate is rotated by 45 ° with respect to the X direction of the polarization direction of the incident light (of the illumination light) by the phase advance axis nf and the phase delay axis ns. Therefore, as shown in fig. 5, the state of deflection of the illumination light transmitted through the first birefringent member 12 in the vicinity of the position separated by ± 1 (reference length) from the center in the X direction is converted to a polarization state having linearly polarized light in the Z direction as a main component by the action of the 1/2 wavelength plate.

In addition, as shown in fig. 4A, the inter-polarization phase difference Δ P1 is 0.25 with respect to the illumination light transmitted through the first birefringent member 12 at a position separated from the center in the X direction by ± 0.6, and the first birefringent member 12 functions as a so-called 1/4 wavelength plate. Therefore, the illumination light transmitted through this portion is converted into a polarized state having circularly polarized light as a main component.

On the other hand, with respect to the light flux passing through the center in the X direction, no optical path difference relationship is generated between linearly polarized light in the phase advance axis nf direction and the phase delay axis ns direction, and the polarization state of the transmitted light is not changed. Therefore, the light beam incident on the birefringent member 12 at the center with respect to the X direction is emitted from the birefringent member 12 with the linearly polarized state of the X direction maintained as it is as a main component. However, the light beams transmitted through positions other than the positions X ═ 0, ± 0.6, and ± 1 are polarized light states in which elliptically polarized light having different shapes as main components is transmitted through the first birefringent member 12. This biased state is shown in fig. 5.

In fig. 1, illumination light IL having different polarization states depending on the field where the illumination light IL passes through the first birefringent member 12 is incident on the second birefringent member 13. The second birefringent member 13 is also a disk-shaped member made of a birefringent material.

Fig. 3A is a view of the second birefringent member 13 of fig. 1 as viewed along the optical axis AX2 of the illumination system in the + Y direction, and when the first birefringent member 12 is different from the first birefringent member 12, as shown in fig. 3A, the phase advance axis nf of the second birefringent member 13 is set parallel to the Z axis of the XZ coordinate of the same coordinate axis as that of fig. 1, and the phase delay axis ns is set parallel to the X axis. The size (diameter) of the second birefringent member 13 in the in-plane direction is also larger than the beam diameter of the illumination light IL arranged on the second birefringent member 13.

Similarly, the second birefringent member 13 is changed in position in the direction of the function Z of the XZ coordinate in fig. 3A, i.e., in the direction of the BB' line in fig. 3A (hereinafter, this is referred to as "XZ direction"). Fig. 3B is a cross-sectional view of the second birefringent member 13 taken along line BB' of fig. 3A. As shown in fig. 3B, the birefringent member 13 has a thin left end (near B) and a thick right end (near B'). On the one hand, the direction of the second birefringent members 13 in the direction orthogonal to the XZ direction is the same. Therefore, the second birefringent member 13 also functions as a second inhomogeneous wavelength plate which imparts a phase difference between different polarizations depending on the location.

In this example, TB of the second birefringent member 13 is represented by the following function with respect to the position XZ in the XZ direction, and as shown in fig. 3B, the origin (XZ ═ 0) in the XZ direction is set as the center (optical axis of the illumination system) of the birefringent member 13, and the center thereof is set as T2.

TB＝T2+β×(2.5×XZ⁵-1.5×XZ³).....(3)

Here, β is a proportionality constant, and as with T2 in the center, the value of β differs depending on the refractive index difference between the phase advance axis and the phase delay axis of the birefringent material used, and T2 in the center is set so that the inter-polarization phase difference Δ P2 of the second birefringent member 13 is 0.25 (unit is the wavelength of the illumination light), that is, is set in the center to function as a 1/4 wavelength plate.

In addition, in the birefringent member 13, the positions separated by +1 (reference length) and-1 in the XZ direction are set so that the inter-polarization phase difference Δ P2 becomes +0.75 and-0.25, respectively. This means that a difference in phase difference between polarized lights of +0.5 and-0.5 is formed from the center.

That is, in the second birefringent member 13 of this example, the inter-polarization phase difference Δ P2 is set so as to be expressed by the following equation.

△P2＝0.25+0.5×(2.5×XZ⁵-1.5×XZ³)....(4)

In addition, when the second birefringent member 13 is also formed of crystal in the same manner as in the above-described example, T2 at the center may be 14.239 μm times (integer +1/4), and the proportionality constant β may be 7.12 μm. Fig. 4B is a diagram showing the relationship between the inter-polarization phase difference Δ P2 and the position XZ in expression (4).

In fig. 1, illumination light having different polarization states depending on the field where the first birefringent member 12 is transmitted is changed again in the polarization state depending on the field by the second birefringent member 13. Fig. 4C shows the polarization state of the illumination light IL emitted from the second birefringent member 13.

The method of fig. 4C is similar to the method of fig. 5, and the polarization state of the illumination light distributed at each position on the XZ coordinate in fig. 4C is indicated by a line segment (linearly polarized light) centered on each position or an ellipse (elliptically polarized light). In fig. 4C, the origin of the X-axis and the Z-axis (X is 0 and Z is 0) is also set at the center of the birefringent member 13.

As shown in fig. 1, in the present embodiment, the first birefringent member 12 and the second birefringent member 13 are disposed immediately in front of the fly-eye lens 14, and the exit-side surface of the fly-eye lens 14 is disposed in the vicinity of the pupil surface 15 in the illumination optical system ILS. Therefore, the first birefringent element 12 and the second birefringent element 13 are arranged at positions substantially equivalent to the pupil plane 15 of the illumination optical system ILS.

Therefore, the illumination light IL transmitted through the first birefringent element 12 and the second birefringent element 13 is incident on the grating R at an incident angle and an incident direction determined according to the positions thereof. That is, in fig. 4C, the light beam distributed at the origin (the position where X is 0 and Y is 0) is incident on the grating R perpendicularly, and the light beam distributed at a position away from the origin by a predetermined distance is incident on the grating R obliquely at an incident angle approximately proportional to the distance. The incident direction is a direction equal to the azimuth angle of the point from the origin.

The outer circle C1 and the inner circle C2 shown in fig. 4C and 5 are boundaries of the distribution of illumination light for constituting a predetermined illumination zone with respect to the grating R. The radii of the circles C1 and C2 are based on the reference length used for determining the shapes (distributions) of the first birefringent member 12 and the second birefringent member 13, and the radius of the outer circle C1 is 1.15 and the radius of the inner circle C2 is 0.85. That is, the annulus ratio (radius of the inner circle/radius of the outer circle) for annulus illumination is assumed to be 0.74. This is assumed to be what is generally used as "3/4 girdle illumination (inner radius: outer radius: 3: 4)", and it is needless to say that the conditions applied to the girdle illumination of the present invention are not limited thereto.

As is clear from fig. 4C, the illumination light emitted from the second birefringent member 13 is in a polarized state having linearly polarized light as a main component, in the specific annular region 36 of the annular region surrounded by the outer circle C1 and the inner circle C2, with the circumferential direction of the specific region 36 being the polarized direction.

When comparing fig. 4C and 5, the polarization states of the illumination light on the X axis and the Y axis are approximately equal. However, the polarization state at positions separated from each axis by about 45 ° around the origin (upper right, upper left, lower left, and lower right positions in fig. 4C and 5) is roughly circularly polarized in fig. 5, and linearly polarized in the circumferential direction converted to the specific zone region in fig. 4C. This is caused by the action of the second birefringent member 13, and the second birefringent member 13 functions as an 1/4 wavelength plate in the upper left and lower right regions in fig. 4C, and functions as a-1/4 wavelength plate and its equivalent 3/4 wavelength plate in each of the lower left and upper right regions.

In the actual exposure apparatus, the actual radius of the outer circle C1 of the specific annular zone region 36 is determined by the Number of Apertures (NA) on the grating R side of the projection optical system 25 in fig. 1 and the focal length of the optical system constituted by the relay lens 16 and the focusing lenses 18 and 19 in the illumination optical system ILS, and is determined by the value of the set coherence factor (illumination σ), and the radius of the inner circle C2 is determined by the more re-set annular zone ratio. However, it is needless to say that the polarization direction of the illumination light distributed in the specific annular region 36 is determined so as to match the circumferential direction of the annular region at each position under the condition of the annular illumination, and the shapes of the first birefringent member 12 and the second birefringent member 13 are determined.

Here, the shape of the first birefringent member 12 and the second birefringent member 13 is determined so that the shape is enlarged or reduced in proportion in the XZ plane, and the amount of unevenness does not change in the Y direction (the proceeding direction of light).

As described above, in the first embodiment of the first and second birefringent members 12 and 13, the polarization direction of the illumination light distributed in the specific annular zone can be made to coincide with the circumferential direction of the annular zone at each position by the first and second uneven wavelength plates that do not attenuate the light beam, in a state where there is no loss of the light amount of the illumination light beam. In this case, the illumination light applied to the grating R through the specific annular zone region 36, that is, the specific illumination light applied to the grating R in the specific incident angle range is light in a polarized state in which S-polarized light whose polarization direction is perpendicular to the incident surface is a main component. Accordingly, depending on the periodicity of the pattern to be transferred, the contrast (contrast), resolution, depth of focus, and the like of the transferred image may be improved (described in detail later).

Next, a second embodiment of the first and second birefringent members 12 and 13 in the illumination optical system ILS of fig. 1 will be described with reference to fig. 6A to 6C.

In this example, the configuration of the first birefringent member 12 and the second birefringent member 13 is basically the same as that of the first embodiment. That is, the first birefringent member 12 has a phase advancing axis direction and a shape as shown in fig. 2A and 2B, and the second birefringent member 13 has a phase advancing axis direction and a shape as shown in fig. 3A and 3B. However, in this example, the function of the birefringent members 12 and 13 is modified.

Fig. 6A corresponds to fig. 4A, and the second embodiment shows the characteristic of the X-direction position of the inter-polarization phase difference Δ P1 formed for the first birefringent member 12. The inter-polarization phase difference Δ P1 in fig. 6A is a function including a trigonometric function with respect to the γ position X as follows.

△P1＝0.265{1-cos(π×X²)}....(5)

Such an inter-polarization phase difference Δ P1 can be obtained by expressing the position of the TA of the first birefringent member 12 in the X direction by the following function.

TA＝T1+γ×{1-cos(π×X²)}....(6)

Here, γ is a proportionality coefficient. In the same manner as in the first embodiment, when the first birefringent member 12 is formed of crystal, the center T1 may be set to an integral multiple of 14.239 μm, and the proportionality coefficient γ may be set to 3.77 μm. 3.77 μm is 14.239 μm of crystal for giving a retardation between polarized lights of 1 wavelength, and is a value 0.265 times the coefficient of the above formula (5).

Fig. 6B shows characteristics of the position in the XZ direction of the inter-polarization phase difference Δ P2 formed in the second birefringent member 13 according to the second embodiment. The inter-polarization phase difference Δ P2 in fig. 6B can be expressed as a function including a trigonometric function with respect to the position XZ as follows.

△P2＝0.25+0.5×sin(0.5×π×XZ³)....(7)

Such an inter-polarization phase difference Δ P2 can be expressed by the following function with respect to the position XZ in the XZ direction with respect to the TB of the second birefringent member 13.

TB＝T2+δ×sin(0.5×π×XZ³)....(8)

Here, δ is a proportionality coefficient. When crystal is used as the second birefringent member 13, T2 at the center may be set to be 14.239 μm (integer +1/4) times larger than the center, and the proportionality coefficient may be set to 7.12 μm.

In this example, the first birefringent member 12 and the second birefringent member 13 function as first and second inhomogeneous wavelength plates in which the phase difference between the polarized lights that give the transmitted light is different for each location. However, the linearly polarized light polarized in the X direction incident on the first birefringent member 12 is converted into a polarization distribution as shown in fig. 6C, and is emitted from the second birefringent member 13.

As apparent from comparison between fig. 6C and fig. 4C, one of the first birefringent member 12 and the second birefringent member 13 of the second embodiment can make the polarization state of the illumination light distributed in the specific annular region 36 surrounded by the outer circle C1 and the inner circle C2 closer to the linear polarization parallel to the circumferential direction than that of the first embodiment. In this regard, the first birefringent member 12 and the second birefringent member 13 of the second embodiment adopt a relationship of a shape (i.e., a surface shape) determined by a higher-order function of a trigonometric function, and can perform a high-precision polarization control.

However, the first birefringent member 12 and the second birefringent member 13 shown in the first embodiment have a relationship of a high-order function of up to five times, and have an advantage that they are slightly inferior in polarization control characteristics, and thus can be easily processed and can be manufactured at low cost.

In order to further reduce the manufacturing cost of the first and second birefringent members 12 and 13, for example, the surface shape of the first birefringent member 12 may be a cylindrical (circular) surface (a surface having a circular cross section in the X direction), and the surface shape of the second birefringent member 13 may be a tapered (taper) surface (inclined surface). The polarization control characteristic in this case is slightly inferior to that of the first embodiment, and a sufficient effect can be obtained for the application of the projection exposure apparatus, and a high-performance exposure apparatus can be realized while the above-mentioned low manufacturing cost is achieved.

The surface shape of the second birefringent member 13 is made to be a tapered surface in this manner, and it is intended that the inter-polarization phase difference of the light beam transmitted through the second birefringent member 13 can be determined linearly (linear function) according to the in-plane position of the second birefringent member 13.

However, the shapes of the first birefringent member 12 and the second birefringent member 13 in fig. 1 are not limited to those shown in the first and second embodiments, and any shape may be used as long as the polarization state in the specific band region of the transmitted light can be matched to the shape in the circumferential direction of each portion.

For example, the shapes of the first birefringent member 12 and the second birefringent member 13 may be stepped shapes that change in shape stepwise at a predetermined position, instead of the shapes represented by the continuous and differential continuous functions described above. The step-like shape is formed by etching instead of mechanical or mechanochemical polishing.

In order to realize such a polarization state, for example, in the case of illumination light composed of a single polarization state in which the polarization state of the light flux incident on the first birefringent member 12 is linearly polarized as a main component, the first birefringent member 12 is preferably configured to impart a phase difference between polarizations having quadratic rotation symmetry (symmetry of rotation) properties about the optical axis AX2 of the illumination system. It is needless to say that the first and second embodiments include a plate having an even function in the X direction and a certain non-uniform wavelength in the Y direction.

The second birefringent member 13 is preferably an inhomogeneous wavelength plate capable of imparting a phase difference between polarized lights having primary rotational symmetry about the optical axis AX2 of the illumination system. The first order rotational symmetry is a distribution of the phase difference between the polarized lights, and is approximately symmetric with respect to one of two axes orthogonal to the optical axis AX2 of the illumination system, and approximately anti-symmetric with respect to the other axis. The antisymmetric function is generally a function in which the absolute value is inverted for the inversion of the coordinate axis, and the sign is equal, and is a function obtained by adding a constant offset (offset) to a general antisymmetric function. It goes without saying that, as shown in the first and second embodiments described above, the plate includes a plate having a certain non-uniform wavelength in the direction orthogonal to the XZ direction, which is determined by an odd function with an additional offset.

In the present embodiment, it is important to set the illumination light distributed in the specific annular zone to a predetermined polarization state, and the shapes of the first birefringent member 12 and the second birefringent member 13 are not particularly problematic in places corresponding to the specific annular zone, not to mention that the shapes do not satisfy the above conditions.

The number of the first birefringent members 12 and the second birefringent members 13 or the phase advance axis direction is not limited to those of the first and second embodiments. That is, three or more birefringent members may be arranged in series along the advancing direction of the illumination light (along the illumination system optical axis AX2), and the rotational relationship about the optical axis AX2 in the advancing axis direction is not limited to 45 °. In the case where three or more birefringent members are arranged in series in the proceeding direction of the illumination light, it is preferable that at least one of the birefringent members has a phase advancing axis direction different from that of the other birefringent members in at least a part of the specific annular zone region, preferably substantially in the entire circumferential region, in order to linearly polarize the polarization state of the illumination light substantially parallel to the circumferential direction.

Similarly, the material of the birefringent members 12 and 13 is not limited to the crystal, and other birefringent materials may be used, and the material may be formed by Intrinsic Birefringence (fluorite). Further, a material such as synthetic quartz that is not birefringent in nature may be used as the birefringent members 12 and 13, which has birefringence by applying stress or the like.

Furthermore, as the birefringent members 12 and 13, those obtained by laminating a material having birefringence on a transmissive substrate having no birefringence can be used. In this case, the above description does not necessarily mean a material having birefringence. The lamination is a method of forming a thin film having birefringence by using a means of vapor deposition on a transmissive substrate, in addition to mechanical bonding such as adhesion or pressure bonding. As described above, although the shapes of the first birefringent member 12 and the second birefringent member 13, etc. shown in the first and second embodiments described above are changed depending on the magnitude of birefringence of the material used, it goes without saying that the shape determining method described above can be applied to determine the shapes even when a material other than crystal is used.

Here, in the above-described annular illumination, the advantage of the illumination light distributed in the annular region in the polarization state thereof in the circumferential direction of the annular region is briefly described with reference to fig. 7A, 7B, and 7C and fig. 8A, 8B, and 8C.

Fig. 7A shows an example of a fine periodic pattern PX formed on the grating R of fig. 1. The periodic pattern PX is a pattern having periodicity in the X direction in the same XYZ coordinates as in fig. 1, and the pitch PT thereof is 140nm in consideration of the projection magnification of the projection optical system 25 in fig. 1 converted into a scale value on the wafer W. Fig. 7B shows the distribution of diffracted light (difffiltered light) formed in the pupil plane 26 (see fig. 1) of the projection optical system 25 having an aperture Number (NA) of 0.90 on the wafer side, when the pattern is illuminated with illumination light having a wavelength of 193nm and an annulus illumination having a coherence coefficient (illumination σ) of 0.9 and an annulus ratio of 0.74.

Fig. 7C is a diagram showing conditions of the annular band illumination for illuminating the pattern PX, and the pattern PX is illuminated with the illumination light of the annular band region ILO satisfying the above conditions of the annular band illumination at the pupil plane 15 of the illumination optical system ILS in fig. 1. The 0 th order diffracted light DO in fig. 7B from the periodic pattern PX is entirely distributed in the pupil plane 26 and reaches the wafer W through the projection optical system 25, and the first order diffracted lights D1R and D1L may be transmitted only partially through the pupil plane 26 and the projection optical system 25. The image of the pattern PI of the grating R is formed as interference fringes (moire fringe) formed on the wafer W by the 0 th order diffracted light DO and the first order diffracted lights D1R and D1L, and is limited to the couple (pair) of the 0 th order diffracted light by the illumination light emitted from the same position on the pupil plane 15 of the illumination optical system ILS.

The first-order diffracted light D1L positioned at the left end of the pupil plane 26 in fig. 7B is coupled to the right end of the 0-order diffracted light DO, and is illumination light illuminated from the right end region ILR in the annular band region ILO in fig. 7C. On the other hand, the first-order diffracted light D1R located at the right end of the pupil plane 26 in fig. 7B is coupled to the left end of the 0-order diffracted light DO, and is illumination light irradiated from the left end region ILL in the annular band region ILO in fig. 7C.

That is, when the pattern PX having a fine pitch in the X direction is exposed, the light flux contributing to the image formation of the pattern PX in the illumination light emitted from the annular zone ILO on the pupil surface 15 of the illumination optical system ILS is limited to the partial region ILR and the partial region ILL, and the illumination light emitted from the other region in the annular zone ILO is illumination light not contributing to the image formation of the pattern PX.

However, when a pattern PX having periodicity in the X direction and a longitudinal direction in the Y direction is exposed, for example, when illumination is performed with linearly polarized light having a polarization direction in the Y direction on a grating R, it is reported in non-patent document 1 (High NA Lithographic imaging at brewers's angle), SPIE vol 4691, pp1 to 24(2001), and the like, which can improve contrast of a projection image.

Therefore, when the illumination light distributed in the partial region and the partial region ILL of fig. 7C is linearly polarized light in the PR direction and the PL direction (corresponding to the Y direction on the grating R in consideration of the action of the mirror surface 19 in fig. 1) parallel to the Z direction in fig. 7C, the contrast of the projection image effective to the pattern PX is improved, and further, the resolution and the depth of focus are improved.

Next, when the grating pattern is rotated by 90 ° from the pattern PX of fig. 7A and a periodic pattern having a fine pitch in the Y direction is formed, the diffraction light distribution shown in fig. 7B is also rotated by 90 °. As a result, the partial region through which the illumination light contributing to the image formation of the periodic pattern passes is also arranged at the position (i.e., the upper end and the lower end in fig. 7C) where the partial region ILR and the partial region ILL shown in fig. 7C are rotated by 90 °, and the good polarization state is linearly polarized light whose polarization direction coincides with the X direction. As described above, when exposing a grating R including a pattern PX having a fine periodicity in the X-axis direction and a pattern PY having a fine periodicity in the Y-axis direction, it is effective to use illumination light having a polarization state as shown in fig. 8A to 8C.

Fig. 8A is a perspective view schematically showing the relationship between the pupil plane 15 and the grating R of the illumination optical system ILS in fig. 1, and the relay lens 16, the focusing lenses 18 and 20, and the like in fig. 1 are omitted. As described above, in order to improve the image formation performance of the periodic pattern PX in the X direction, the illumination light distributed in the endless belt ILO in fig. 8A is preferably linearly polarized light in the Y direction (direction passing through the paper surface in fig. 8A) at the ends ILL, ILR in the X direction, and is preferably linearly polarized light in the X direction at the ends ILU, ILD in the Y direction, in order to improve the image formation performance of the periodic pattern PY in the Y direction. That is, it is preferable to use linearly polarized light whose polarization direction approximately coincides with the circumferential direction of the annular zone ILO.

Further, in the case where the grating R is a pattern including not only the X direction and the Y direction but also the intermediate direction (45 ° and 135 ° directions), it is preferable to use linear polarization whose polarization direction completely coincides with the circumferential direction of the annular zone region in consideration of the directivity of such a pattern.

However, the above-described polarization state does not necessarily achieve an effective polarization state for a pattern orthogonal to a directivity pattern suitable for the polarization state of each portion in the annular zone ILO. For example, the illumination light polarized in the X direction from the partial region ILU is a bad polarization direction for the image formation of the pattern PX having periodicity in the X direction and the longitudinal direction in the Y direction. However, it is clear from fig. 7C that fig. 7C shows a light source contributing to an image formation of a pattern having a fine pitch in the X direction, and the partial region ILU corresponding to the upper end of the annular band region ILO in fig. 7C is a relationship of a light source not contributing to an image formation of a pattern having a fine pitch in the X direction at all, and the image formation characteristics are not deteriorated regardless of the polarization state of the partial region ILU.

As shown in fig. 8A, the linearly polarized light that is substantially aligned with the circumferential direction of the annular zone ILO on the pupil plane 15 of the illumination optical system ILS is incident on the grating R as S-polarized light, which is orthogonal to the incident plane (including the normal line of the object and the plane of the light beam) on which the light beam is incident. That is, as shown in fig. 8B, the illumination light ILL1 of the partial region ILL composed of linearly polarized light in the direction coincident with the circumferential direction of the annular region ILO enters the grating R with the polarization direction EF1 as S-polarized light perpendicular to the entrance surface (the paper surface of fig. 8B). As shown in fig. 8C, the illumination light ILD1 on the same partial area ILD is incident on the grating R with the polarization direction EF2 also being S-polarized perpendicular to the incident surface (the paper surface in fig. 8C).

As a matter of course, the illumination light from the partial regions ILL and ILD and the partial regions ILR and ILU that are symmetrically positioned with respect to the optical axis AX41 of the illumination optical system has a relationship in which the respective illumination light has a polarization direction that coincides with the circumferential direction of the annular region ILO in the partial regions ILR and ILU, and is similarly S-polarized light from symmetry and incident on the grating R. Regarding the general property of the annular illumination, the incidence angle of the illumination light distributed in the annular region ILO on the grating R is within a predetermined angle range centered on the angle Φ from the optical axis AX41 (i.e., the perpendicular to the grating R) of the illumination optical system. The light beam irradiated to the grating R at the incident angle is hereinafter referred to as "specific illumination light". The angle phi and the angle range are preferably determined according to the wavelength of the illumination light, the pitch of the pattern to be transferred of the grating R, and the like.

However, the first and second birefringent members 12 and 13 are configured such that the polarization state of illumination light distributed in a specific annular region between a predetermined outer radius (outer circle C1) and an inner radius (inner circle C2) determined from the shape inherent to the members is converted to a polarization state in which linearly polarized light parallel to the circumferential direction of the specific annular region is a main component, and the radii (C2 and C1) thereof are not easily changed.

Therefore, when it is necessary to change the desired zone area in accordance with the pitch of the pattern to be transferred onto the grating R or the like as described above, it is preferable to add a plurality of conical prisms 41 and 42 of a zoom type between the first and second birefringent members 12 and 13 and the optical integrator such as the fly-eye lens 14 in fig. 1 so that the radius of the specific zone area described above can be changed as shown in fig. 9. In fig. 9, the plurality of zoom type conical prisms are arranged along the illumination system optical axis AX2 with the interval DO variable between the concave conical prism 41 having the concave conical surface 41b and the convex conical prism 42 having the convex conical surface 42 a.

In this case, the illumination light transmitted through the first and second birefringent members 12 and 13 and distributed in a specific annular region centered on the average radius RI is enlarged to a radius R0 at the pupil plane 15 of the illumination optical system on the incident surface and the exit surface of the fly-eye lens 14 by the zoom type conical prisms 41 and 42. The radius R0 is enlarged by enlarging the interval DD between the two conical prisms 41, 42, and is reduced by reducing the interval DD.

Thus, a specific annular region in which illumination light composed of linearly polarized light parallel to the circumferential direction is distributed can be formed at any radius on the pupil surface 15 of the illumination optical system, and the illumination condition for illuminating the annular region can be changed in accordance with the pattern of the grating R to be transferred.

Further, instead of the zoom prisms 41 and 42, a zoom optical system may be used.

Although the above embodiment has been described with the illumination light amount distribution formed on the pupil plane 15 of the illumination optical system ILS in fig. 1 as the zone area, that is, with the premise of being applied to the zone illumination, the illumination conditions achievable by the projection exposure apparatus in fig. 1 are not necessarily limited to the zone illumination. That is, the birefringent members 12 and 13 in fig. 1 and the variable focal length type conical prisms 41 and 42 in fig. 9 are in a relationship in which the polarization state of the illumination light distributed in the specific annular region in the pupil plane 15 of the illumination optical system is set to the desired polarization state described above, and in the case where the distribution of the illumination light is limited to a more specific partial region in the specific annular region, that is, in the case where the distribution is limited to the partial regions ILL and ILR in fig. 7C, for example, it is needless to say that the illumination light whose main component is linearly polarized in the polarization direction distributed in the circumferential direction of the partial region can be used.

In this way, when the illumination light is focused only in a more specific area within the specific annular area, the diffractive optical element 9a in fig. 1 may be replaced to concentrate the diffracted light (illumination light) generated from the other diffractive optical element in a more specific discrete area within the specific annular area on the first birefringent member 12 and the second birefringent member 13. Although the positions where the illumination light is concentrated are, for example, two positions of the partial regions ILL and ILR in fig. 7C, the positions are not limited to these, and may be concentrated at any position in the specific annular region, and the number thereof may be four. The selection may be determined according to the pattern shape of the exposure target on the grating R.

In addition, in order to concentrate the illumination light in a more specific area within the specific area, instead of the above-described zoom type conical prisms 41 and 42, a group of optical members in which a convex polygonal prism such as a pyramid type and a concave polygonal prism are combined in the same manner with variable intervals may be used.

Further, the illumination light distributed outside these specific regions is not suitable for the pattern exposure of the exposure target, and it is preferable in some cases to substantially set the light amount distribution to 0. On the other hand, due to manufacturing errors of the diffractive optical element 9a and the like, diffracted light (hereinafter referred to as "error light") is generated from the diffractive optical element 9a and the like in a desired direction, and there is a possibility that the illumination light is distributed in a region other than the above-described partial region. For example, a diaphragm is provided on the incident surface side or the exit surface side of the fly-eye lens 14 in fig. 1 to shield the error light. Thus, the above-described distributions of the amount of illumination light in the plurality of specific regions are completely discrete. However, there are also patterns other than the pattern of the exposure target on the grating R, and the error light has a relationship effective in forming an image of the pattern other than the exposure target, and there are also cases where it is not always necessary to set the illumination light amount distribution other than the specific region to 0.

However, focusing on the incidence of the illumination light on the grating R, the illumination light amount distribution on the pupil plane 15 is limited to a more specific area within a specific area, and the incidence direction is also limited from the above-described plurality of substantially discrete directions by adding a limitation to the incidence angle range of the illumination by the annular band. Of course, when the present invention is applied to the zone illumination, the error light distributed outside the specific zone region may be configured to be blocked by providing a diaphragm on the incident surface side or the emission surface side of the fly light lens 14.

In the above embodiment, the fly-eye lens 14 is used as the optical integrator, but an internal reflection type integrator (e.g., glass rod) may be used as the optical integrator. In this case, the exit surface of the glass rod is not on the pupil surface 15 of the illumination optical system, but is disposed on a conjugate surface with the grating R.

In the above embodiment, although the laser light source of the exposure light source 1 emits linearly polarized light polarized in the X direction, depending on the form of the laser light source, the laser light source may emit linearly polarized light polarized in the Z direction in fig. 1 or a light beam in another polarized state. When the exposure light source 1 in fig. 1 emits light linearly polarized in the Y direction, that is, light linearly polarized in the Z direction at the position of the birefringent members 12 and 13, the birefringent members 12 and 13 shown in the first and second embodiments described above are rotated by 90 ° about the optical axis a × 2 of the illumination system as a rotation center, and thereby illumination light in a polarized state substantially the same as the polarized state shown in fig. 4C and 6C can be obtained (specifically, illumination light in a state in which the states shown in both figures are rotated by 90 °).

Alternatively, the linearly polarized light in the Y direction emitted from the exposure light 1 may be converted into linearly polarized light in the X direction by the polarization control means 4 (polarization control means) in fig. 1. Such a polarization control element 4 can be easily realized by a so-called 1/2 wavelength plate. In addition, when the exposure light source 1 emits circularly polarized light or elliptically polarized light, the polarization control member 4 can be similarly changed to linearly polarized light in the desired Z direction by using an 1/2 wavelength plate or a 1/4 wavelength plate.

However, the polarization control member 4 is not a member that can convert a light beam in an arbitrary polarization state irradiated from the exposure light source 1 into polarization in the Z direction without light loss. Therefore, the exposure light source 1 needs to generate a light beam having a single polarization state (a light beam that can be converted into linearly polarized light by a wavelength plate or the like without loss of light quantity) such as linearly polarized light, circularly polarized light, elliptically polarized light, or the like. However, in the case where the intensity of the light beam other than the single polarization state is not so large as to be a certain degree with respect to the entire intensity of the illumination light, the influence of the light beam other than the single polarization state on the image characteristics is slightly concerned, and the light beam irradiated from the exposure light source 1 may include the light beam other than the single polarization state as long as it is a certain degree (for example, 20% or less of the total light amount).

In consideration of the state of use of the projection exposure apparatus according to the above embodiment, it is not preferable to set the polarization state of the illumination light to be linearly polarized such that the illumination light distributed in the specific annular region is always substantially parallel to the circumferential direction of the annular region, or to set the specific illumination light to be incident on the grating R in S-polarized light. That is, it is preferable not only for the annular illumination but also for the normal illumination (illumination having a circular illumination light amount distribution on the pupil plane 15 of the illumination optical system), and in this case, it is preferable not to use the illumination light in the polarized state of the above embodiment.

In order to cope with such a usage state, the polarization control member 4 in fig. 1 may be an element or an optical system that can change the polarization state of a light beam emitted from a light source such as a laser beam into any polarization state as needed. This can be achieved by, for example, two polarized beam splitters (beam splitters) 4b, 4c, etc., as shown in fig. 10.

Fig. 10 shows a polarization control optical system provided at a position of the polarization control member 4 in fig. 1, and in fig. 10, for example, an illumination light beam ILO (corresponding to the illumination light IL in fig. 1) composed of linearly polarized light is incident on a rotating wavelength plate 4a composed of an 1/2 wavelength plate or a 1/4 wavelength plate. The illumination light IL1 converted into linearly polarized light or circularly polarized light in a direction inclined at 45 ° on the paper surface of fig. 10 is split into a light beam IL2 composed of P-polarized light components and an IL3 composed of S-polarized light components by the first polarized beam splitter 4b with respect to the splitting plane, one light beam IL2 is directed upward in fig. 10 through the prism 4b, and the other light beam IL3 is reflected rightward in fig. 10.

Although the straight light beam IL2 enters the polarized beam splitter 4C next, the light beam IL2 travels straight in the polarized beam splitter 4C due to its polarization characteristic, and the light beam IL4 advances upward in fig. 10. On the other hand, the reflected light beam IL3 is reflected by the mirrors 4d and 4e and enters the polarization beam splitter 4C, where the re-reflected light beam IL3 joins the straight light beam IL4 again. At this time, if the distances between the polarization beam splitters 4b and 4c and the mirror surfaces 4d and 4e are each DL, a 2 × DL optical path length difference is formed between the two merged light fluxes IL3 and IL 4. However, when the optical path length difference 2 × DL is set to be longer than the coherence (coherent) length of the illumination light beam, the coherence between the two light beams is lost, and the combined light beams can be polarized substantially arbitrarily.

When such a polarization control optical system is incorporated in the illumination optical system ILS in fig. 1, the illumination light IL transmitted therethrough is always polarized arbitrarily, and cannot be regarded as a failure to realize the above-described embodiment. However, in the optical system shown in fig. 10, the polarization state of the illumination light IL1 transmitted through the rotating wavelength plate 4a can be entirely changed to the linearly polarized light transmitted through the first beam splitter 4b by the rotation of the rotating wavelength plate 4a, and the above-described problem is not generated in principle. However, since some light quantity loss is unavoidable due to absorption by the polarized beam splitters 4b and 4c and reflection loss by the mirror surfaces 4d and 4e, a mechanism for avoiding the beam splitters 4b and 4c and the rotary wavelength plate 4a from the optical path of the illumination optical system may be provided when it is not necessary to polarize the illumination light arbitrarily.

However, even without using such a polarized beam splitter, the following simple method can provide an effect approximately similar to that of any polarized illumination. This is achieved by converting the polarized state of the illumination light IL incident on the first birefringent member 12 of fig. 1 to a direction separated by 45 ° from the X direction and the Y direction in fig. 1, and by converting the illumination light distributed in the specific annular zone into approximately circular polarized light. Therefore, even when the projection exposure apparatus of the present embodiment is used for applications where circular polarization can be regarded as being similar to arbitrary polarization, that is, when the required imaging performance is used for relatively inexact applications, the polarization control member 4 in fig. 1 is configured as an 1/2 wavelength plate, for example, and when the polarization state of the illumination light incident on the first birefringent member 12 is tilted by 45 ° from the X axis and the Y axis as described above, the same effect as that of arbitrary polarization illumination can be obtained. In addition, the polarization control member 4 is similarly constituted by, for example, an 1/4 wavelength plate, and when the polarization state of the illumination light incident on the first birefringent member 12 is circularly polarized, the same effect as that of arbitrary polarized illumination can be obtained.

Alternatively, the same effect as that of any polarized light illumination can be obtained by rotating the first birefringent member 12 and the second birefringent member 13 in fig. 1 by 45 °, for example, by a mechanism that can rotate the two birefringent members 12 and 13 in the direction of the linearly polarized light of the illumination light collectively around the illumination system optical axis AX2 as the optical axis of the illumination optical system ILS.

However, in the above-described normal illumination, it is sometimes preferable to set the polarization state to linearly polarized light in a predetermined direction. In the projection exposure apparatus of the above-described embodiment, when the illumination condition is to be met, the birefringent members such as the first birefringent member 12 and the second birefringent member 13 in fig. 1 may be configured to be individually rotatable about the illumination system optical axis AX2 in a lump, and the rotation direction of each birefringent member may be set so that the phase advance axis (or the phase delay axis) of each birefringent member is parallel to the linearly polarized direction of the illumination light. In this case, the illumination light is not subjected to the conversion of the polarization state at all even when the birefringent member is operated, and is emitted as if it is straight light when it is incident.

In addition, when setting the linearly polarized state in a predetermined one direction, the first birefringent member 12 and the second birefringent member 13 may be accommodated by being totally outside the optical path of the illumination optical system. That is, the exchange mechanism may be provided so as to correspond to the linearly polarized state in a predetermined one direction by collectively exchanging the birefringent members and the like. Further, when the exchanging mechanism is to be provided, a plurality of sets of birefringent member groups in the exchanging mechanism may be set, and these may be disposed at positions on the optical axis AX2 of illumination so as to be exchangeable. In this case, it is needless to say that each birefringent member group has a characteristic of converting illumination light into linearly polarized light in a circumferential direction in a specific annular zone region having different outer and inner radii.

However, it is preferable to use illumination light linearly polarized in a predetermined direction as described above, and for example, in the case of exposing a spatial frequency modulation type phase shift grating having a pattern in a complete pattern direction. In this case, however, the coherence coefficient (σ value) of the illumination light is preferably about 0.4 or less in order to further improve the resolution and the depth of focus of the pattern to be transferred by exposure.

Here, as to the action of the birefringent members (the first birefringent member 12 and the second birefringent member 13) according to the present invention, as will be described with reference to fig. 4C and 6C, as shown in the respective figures, the polarization state of the illumination light distributed inside a circle (not shown) having a radius of about half the radius of the outer radius C1 in the specific zone region is hardly affected by the optical axis (X ═ 0, Y ═ 0) of the illumination optical system, which is the first embodiment (fig. 4C) and the second embodiment (fig. 6C) of the first birefringent member 12 and the second birefringent member 13 in common.

When the radius of the outer radius C1 corresponds to, for example, 0.9 as the illumination σ (σ value), the first birefringent member 12 and the second birefringent member 13 emit the incident linearly polarized light in the X direction in the range of the illumination light beam in which the illumination σ is 0.45, in a state of polarization approximately as it is. When the first birefringent member 12 receives the linearly polarized light in the Z direction (Z-polarized light), the polarization state of the illumination light beam having the illumination σ of about 0.45 out of the light beams emitted from the second birefringent member 13 can be Z-polarized light.

Therefore, when the birefringent members (the first birefringent member 12 and the second birefringent member 13) according to the first and second embodiments are used, it is not necessary to avoid this outside the optical path of the illumination optical system, and the polarization direction of the incident light to the birefringent members is switched by the polarization control member 4 or the like, whereby it is possible to realize illumination light suitable for the illumination to the spatial frequency modulation type phase shift grating described above, which is an illumination light beam with an illumination σ of about 0.4 or less, and which is polarized in the X direction or the Z direction (polarized light in the X direction or the Y direction in each of the gratings R in fig. 1).

In this case, of course, if the illumination σ is limited to about 0.4, it is preferable to use the diffractive optical element 9a in which the directional characteristic of the generated diffracted light has an angular distribution that can correspond to the directional characteristic. Therefore, the lighting beam B section in various practical polarization states can be formed without the general exchange mechanism, and the advantages of the invention are also achieved.

Next, an example of a process for manufacturing a semiconductor device using the projection exposure apparatus according to the above-described embodiment will be described with reference to fig. 11.

Fig. 11 shows an example of a manufacturing process of a semiconductor device, and in fig. 11, first, a wafer W is manufactured from a silicon semiconductor or the like. Thereafter, a photoresist (photoresist) is applied to the wafer W (step S10), and next, in step S12, a grating (assumed to be R1) is mounted on the grating stage of the projection exposure apparatus of the above-described embodiment (fig. 1), the wafer W is mounted on the wafer stage, and the pattern (indicated by symbol a) of the grating R1 is transferred (exposed) to all the shot regions SE on the wafer W by a scanning exposure method. At this time, double exposure is performed as needed. The wafer W is, for example, a wafer (12 inch wafer) having a diameter of 300mm, and an example of the size of the imaging region SE is a rectangular region having a width in the non-scanning direction of 25mm and a width in the scanning direction of 33 mm. Next, in step S14, a predetermined pattern is formed in each imaging region SE of the wafer W by performing development, etching, ion implantation, and the like.

Next, at step S16, a photoresist is applied to the wafer W, and then at step S18, a grating (assumed to be R2) is mounted on the grating stage of the projection exposure apparatus of the above-described embodiment (fig. 1), the wafer W is mounted on the wafer stage, and the pattern (indicated by symbol B) of the grating R2 is transferred (exposed) to each shot region SE on the wafer W by a scanning exposure method. However, in step S20, a predetermined pattern is formed in each imaging region of the wafer W by performing development, etching, ion implantation, and the like of the wafer W.

The above exposure process to pattern formation process (step S16 to step S20) are repeated as many times as necessary to manufacture a desired semiconductor device. However, the semiconductor device SP, which is a product, is manufactured by performing a dicing (dicing) process (step S22) and a bonding (bonding) process, an assembling (packing) process, and the like (step S24) for cutting each chip CP on the wafer W one by one.

In the device manufacturing method according to the present example, in the case of performing exposure by the projection exposure apparatus according to the above-described embodiment, the grating can be illuminated in a predetermined polarization state in the exposure process in a state in which the utilization efficiency of the illumination light (exposure light beam) is improved. Therefore, the resolution of a fine pitch periodic pattern or the like is improved, and a high-performance semiconductor integrated circuit can be manufactured at a high throughput (throughput) at a low cost with higher integration.

The projection exposure apparatus according to the above-described embodiment can be manufactured by assembling the illumination optical system and the projection optical system, which are formed by a plurality of lenses, into the exposure apparatus body and optically adjusting them, and by assembling the reticle stage and the wafer stage, which are formed by a plurality of mechanical members, into the exposure apparatus body and connecting wires and pipes, and further by performing comprehensive adjustment (electrical adjustment, operation confirmation, and the like). In addition, the projection exposure apparatus is preferably manufactured in a clean room (clean room) in which the temperature, the cleanliness, and the like are controlled.

The present invention is applicable not only to a scanning exposure type projection exposure apparatus but also to a general exposure type projection exposure apparatus such as an exposure apparatus. The magnification of the projection optical system used may be not only a reduction magnification but also an equal magnification or an expansion magnification. Further, the present invention is also applicable to a liquid immersion type exposure apparatus as disclosed in, for example, International publication (WO) No. 99/49504.

The application of the projection exposure apparatus of the present invention is not limited to an exposure apparatus for manufacturing a semiconductor device, and an exposure process (exposure apparatus) in the case of manufacturing a mask pattern for manufacturing various devices such as an exposure apparatus for a display device such as a liquid crystal display device formed on an angle glass substrate or a plasma display, an imaging device (such as a CCD), a micromachine (micro machine), a thin film magnetic head, and a DNA wafer (including a mask of an X-ray mask, a reticle, and the like) can be widely applied to the exposure apparatus and the exposure process.

Incidentally, the illumination optical systems 2 to 20 included in the projection exposure apparatus shown in the above-described embodiment are illumination optical apparatuses that can also be applied to the first object such as the illumination grating R.

The present invention is not limited to the above-described embodiments, and it is needless to say that various configurations can be adopted without departing from the scope of the present invention. All disclosures of the specification, claims scope, drawings and abstract contained in japanese patent application laid-open No. 2003-367-963, filed on 28/10/2003, are hereby incorporated by reference in their entirety.

According to the device manufacturing method of the present invention, the efficiency of utilization of the exposure light beam (illumination light) can be improved, and the predetermined pattern can be formed with high accuracy. Therefore, various elements such as a semiconductor integrated circuit can be manufactured with high accuracy and high throughput.

Although the present invention has been described with reference to the preferred embodiments, it should be understood that various changes and modifications can be made therein by those skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (15)

1. An illumination optical apparatus for irradiating a pattern of a mask with illumination light from a light source, comprising:

a polarization control means for allowing the illumination light to enter, a diffraction optical element for diffracting the illumination light emitted from the polarization control means, and a birefringent member for allowing the illumination light emitted from the diffraction optical element to enter,

wherein the polarization control means switches the polarization direction of the illumination light incident on the birefringent member,

the diffractive optical element limits the incident angle range of the illumination light irradiated to the pattern of the mask to a specific range,

the birefringent member changes the polarization direction by the polarization control means, and changes the polarization state of the illumination light incident on the pattern of the mask while being limited to the specific range by the diffractive optical element to a polarization state mainly containing S-polarized light.

2. The illumination optical apparatus according to claim 1, wherein the polarization control means changes a polarization direction of the incident linearly polarized illumination light.

3. The illumination optical device according to claim 1, wherein the polarization control means includes an 1/2 wavelength plate and a 1/4 wavelength plate.

4. The illumination optical apparatus according to claim 1, wherein the diffractive optical element further limits an incident direction of the illumination light irradiated on the pattern of the mask to a specific plurality of discrete directions.

5. The illumination optical apparatus according to claim 1, wherein the birefringent member is a non-uniform wavelength plate.

6. The illumination optical apparatus according to claim 5, wherein the shape of the birefringent member is a shape represented by a continuous and differentially continuous function.

7. The illumination optical device according to claim 5, wherein the shape of the birefringent member is a stepwise shape which changes stepwise in shape at a predetermined position.

8. The illumination optical apparatus according to claim 5, wherein a plurality of the birefringent members are arranged along a traveling direction of the illumination light.

9. The illumination optical device according to any one of claims 1 to 8, comprising an optical integrator disposed between the birefringent member and the pattern of the mask.

10. A projection exposure apparatus, characterized in that it comprises:

an illumination optical system which is the optical illumination device according to any one of claims 1 to 8 and irradiates illumination light from a light source to a pattern of a mask; and

and a projection optical system for projecting an image of the pattern of the mask onto a substrate.

11. A projection exposure apparatus, characterized in that it comprises:

an illumination optical system which is the optical illumination device according to claim 9 and irradiates illumination light from a light source to a pattern of a mask; and

and a projection optical system for projecting an image of the pattern of the mask onto a substrate.

12. An exposure method, characterized in that the projection exposure apparatus according to claim 10 is used to expose a photosensitive body as the substrate with an image of the pattern of the mask.

13. An exposure method, characterized in that the projection exposure apparatus according to claim 11 is used to expose a photosensitive body as the substrate with an image of the pattern of the mask.

14. A method of manufacturing a device, comprising a photolithography process, characterized in that:

in the above-mentioned photolithography process, a pattern is transferred to a photoreceptor by using the exposure method according to claim 12.

15. A method of manufacturing a device, comprising a photolithography process, characterized in that:

in the above-mentioned photolithography process, a pattern is transferred to a photoreceptor by using the exposure method according to claim 13.