JP5225433B2 - Illumination optical system and exposure apparatus - Google Patents

Description

  The present invention generally relates to an illumination optical system, and more particularly to control of an incident angle distribution or orientation characteristic (also referred to as “effective light source” or “intra-σ distribution”) on a surface to be illuminated. The illumination optical system of the present invention is suitable for an exposure apparatus for microlithography in the production of fine patterns such as semiconductor elements, liquid crystal elements, and magnetic materials.

  Projection of transferring circuit patterns by projecting circuit patterns drawn on a reticle (mask) onto a wafer or the like by a projection optical system when manufacturing fine semiconductor elements such as semiconductor memories and logic circuits using photolithography technology An exposure apparatus has been conventionally used. Due to the recent demand for smaller and thinner electronic devices, there is an increasing demand for miniaturization of semiconductor elements mounted on electronic devices. As one means for achieving high resolution, it is known to increase the numerical aperture (NA) of the projection optical system (high NA).

  Further, in order to perform high-quality exposure, it is necessary to form an optimum effective light source according to the pattern formed on the surface to be illuminated (reticle). The effective light source distribution is realized, for example, by adjusting the intensity distribution near the exit surface of the fly-eye lens to a desired shape (normal illumination condition, annular illumination condition, quadrupole illumination condition). Also, means for changing the NA of the projection optical system, the coherence factor σ (NA of the illumination optical system / NA of the projection optical system), and the effective light source to optimally set each of the processes having various characteristics. Is required.

  In recent years, the optical path of illumination optical systems has become longer due to complicated functions such as the formation of various effective light sources. For this reason, it is difficult to linearly arrange the illumination optical system, and it is necessary to bend the optical path using a folding mirror (deflection mirror) in order to reduce the size of the exposure apparatus. As one means for monitoring the amount of exposure light, it is also known to arrange a half mirror in the illumination optical system and monitor the transmitted / reflected light.

  Considering the case where exposure light having a certain light beam width enters a certain range of the folding mirror or half mirror, the incident angle of the light beam may vary depending on the location within that range. Conventionally, the variation in the transmittance / reflectance of the mirror with respect to the incident angle of the light beam has been suppressed to an allowable range by the design technology, but it has become impossible to suppress the allowable range as the NA increases due to the demand for miniaturization. . In addition, for light having a wavelength of 250 nm or less, the film material constituting the mirror is limited, and the degree of freedom in design has also been limited.

  As a result, a desired effective light source distribution can be formed without a mirror, but the effective light source distribution does not become a desired distribution due to the transmittance and reflectance characteristics of the mirror. This is because exposure is performed with a coherence factor σ that is different from the coherence factor σ optimally set to transfer the minimum line width of a certain pattern, and the set resolution line width (especially the minimum line width) is obtained. Invite the problem of not being able to. In addition, the “HV difference” in which the line width transferred to the wafer is different between the horizontal direction and the vertical direction of the pattern occurs, resulting in a problem that the yield decreases.

  Further, the contrast of the interference fringes formed on the photosensitive agent by the line and space (L & S) pattern is lowered when the light is P-polarized with respect to the L & S diffracted light, and becomes particularly noticeable as the NA increases. For this reason, polarized illumination using S-polarized light (that is, light in which the vibration direction of the electric field vector of light is parallel to the wafer surface and perpendicular to the light traveling direction) has been studied. However, since the transmittance / reflectance of the mirror differs between S-polarized light and P-polarized light, an HV difference similarly occurs.

For example, Patent Documents 1 to 3 are known as conventional techniques for solving the non-uniformity of the transmittance distribution in the illumination optical system.
JP 2002-09377000 A JP 2003-243276 A JP 2002-75843 A

  Patent Document 1 proposes correcting the transmittance by adjusting the relative angle of two filters having a discrete transmittance distribution, but setting the relative angle actually measures the effective light source distribution. Since it is not known until it is done, it takes time to make adjustments, and there is a problem that the correction accuracy is poor because the transmittance distribution is discrete. On the other hand, Patent Document 2 aims at correcting the non-uniformity of the transmittance depending on the incident position of the lens, but does not consider the non-uniformity of the reflectance / transmittance due to the incident angle of the mirror.

  Accordingly, an object of the present invention is to provide an illumination optical system, an exposure apparatus, and a device manufacturing method that can obtain a desired effective light source relatively easily and quickly.

The illumination optical system according to one aspect of the present invention is an illumination optical system for illuminating an illumination target surface, and a mirror, wherein the relationship between the irradiated surface and substantially the Fourier transform with a light beam from a light source having a filter member disposed at a position, the transmittance distribution of the filter member is changed in a first direction, Ri second substantially constant der in a direction perpendicular to the first direction, the In the first direction of the filter member, light having different incident angles to the mirror is incident, and in the second direction of the filter member, light having a substantially constant incident angle to the mirror is incident. the transmittance distribution in a first direction of the filter member, characterized that you have changed so as to reduce the intensity difference of the light reflected by the mirror according to the incident angle characteristics of the mirror.

  Further objects and other features of the present invention will become apparent from the preferred embodiments described below with reference to the accompanying drawings.

  According to the present invention, it is possible to provide an illumination optical system, an exposure apparatus, and a device manufacturing method capable of obtaining a desired effective light source relatively easily and quickly.

1 is a schematic block diagram of an exposure apparatus as one aspect of the present invention. It is a schematic block diagram of the light beam shape conversion means of the exposure apparatus shown in FIG. It is a schematic plan view for demonstrating operation | movement of the light beam change means in the exposure apparatus shown in FIG. It is a schematic plan view for demonstrating another operation | movement of the light beam change means shown in FIG. It is a schematic plan view for demonstrating another operation | movement of the light beam change means shown in FIG. It is a graph for demonstrating the nonuniformity of the transmittance distribution by the mirror in the exposure apparatus shown in FIG. It is a schematic plan view for demonstrating the nonuniformity of the transmittance distribution by the mirror in the exposure apparatus shown in FIG. It is a schematic plan view of the filter member of the exposure apparatus shown in FIG. It is a flowchart which shows an example of the illumination method in the exposure apparatus shown in FIG. It is the schematic for demonstrating operation | movement of the (sigma) shape correction mechanism of the exposure apparatus shown in FIG. FIG. 11 is a schematic plan view of the σ shape correction mechanism shown in FIG. 10. It is a flowchart for demonstrating manufacture of devices (semiconductor chips, such as IC and LSI, LCD, CCD, etc.). 13 is a detailed flowchart of the wafer process in Step 4 shown in FIG. 12. It is the figure which showed the example of the polarization state of the light beam in effective light source distribution. It is a graph for demonstrating the nonuniformity of the transmittance | permeability distribution of the polarization state by the mirror in the exposure apparatus shown in FIG. It is the figure which showed the example of the effective light source correction | amendment by the phase plate structure in the case of generating a different polarization state in a pupil in the case of polarized illumination, and a filter member and a stop. FIG. 2 is an enlarged configuration diagram of the illumination optical system shown in FIG. 1 when performing polarized illumination and when performing non-polarized illumination. It is a figure explaining the method to set the transmittance | permeability distribution of the filter in tangential polarization illumination. It is a figure explaining the method to set the transmittance | permeability distribution of the filter in cross pole polarization illumination.

  The exposure apparatus 1 will be described below as one aspect of the present invention with reference to the accompanying drawings. Here, FIG. 1 is a schematic block diagram of the exposure apparatus 1. As shown in FIG. 1, the exposure apparatus 1 includes an illumination apparatus 100, a reticle 200, a projection optical system 300, a plate 400, a plate stage 450, and a control unit 500. The exposure apparatus 1 is a projection exposure apparatus that exposes a plate 400 with a circuit pattern formed on a reticle 200 by, for example, a step-and-scan method or a step-and-repeat method.

  The illumination device 100 illuminates a reticle 200 on which a transfer circuit pattern is formed, and includes a light source unit 102 and an illumination optical system 110.

For the light source unit 102, for example, an ArF excimer laser with a wavelength of about 193 nm, a KrF excimer laser with a wavelength of about 248 nm, or the like can be used as the light source, but the type of the light source is not limited to the excimer laser. A 153 nm F ₂ laser may be used, and the number of light sources is not limited. Moreover, when a laser is used for the light source unit 102, it is preferable to use an incoherent optical system that incoherently converts a coherent laser beam. The light source that can be used for the light source unit 102 is not limited to the laser, and one or a plurality of lamps such as a mercury lamp and a xenon lamp can be used. The exposure apparatus 1 of the present embodiment is particularly effective when using light having a wavelength of 250 nm or less, as will be described later.

  The illumination optical system 110 is an optical system that illuminates the reticle 200, and includes a lens, a mirror, an optical integrator, a stop, and the like. The illumination optical system 110 according to this embodiment includes a light beam shape conversion unit 120, a light beam changing unit 130, an imaging optical system 140, bending mirrors 112, 150, 151, a filter member 154, and a fly-eye lens 156. , Variable aperture 158, condensing optical system 160, half mirror 152, detection unit 170, masking blade 172, and imaging optical system 180.

  The bending mirror 112 guides the light emitted from the light source unit 110 to the light beam shape conversion unit 120.

  The light beam shape conversion means 120 changes the light from the light source unit 110 into a light beam having a desired shape distribution on a predetermined surface (A surface) as necessary, such as a circular shape, an annular shape, or a multipolar shape. That is, the A surface is a surface that forms the basic shape of the effective light source. The distribution on the A surface is a basic shape, the shape is changed by a light beam changing means 130 described later, the size is changed by an imaging optical system 140 with variable magnification, the restriction is made by a diaphragm member (for example, 158) arranged at each position, etc. Thus, a desired effective light source shape is formed on the irradiated surface.

  The light beam shape conversion means 120 is composed of at least one of a fly-eye lens, an optical pipe using internal reflection, a diffractive optical element, etc., or a plurality of optical integrators, relay optical systems, condensing optical systems, mirrors, etc. that combine them. Is done. The light beam shape converting means 120 of the present embodiment includes optical systems 121, 123, 126, optical integrators 122, 124, and diffractive optical elements 125a, 125b. Here, FIG. 2 is a schematic configuration diagram of the light beam shape conversion means 120.

  The optical system 121 is configured by a cylindrical lens or the like, and changes an incident light beam to a light beam having a desired size such as a substantially circular or square shape. In this embodiment, the optical integrator 122 is composed of a fly-eye lens in which microlenses are two-dimensionally arranged or have an equivalent effect, and the incident surface of the optical integrator 124 is made uniform through the optical system 123. Illuminate. The light distribution shape on the incident surface of the optical integrator 124 is determined by the numerical aperture of the microlens constituting the optical integrator 122 and the focal length of the optical system 123, and is constant regardless of the distribution of the light beam incident on the optical integrator 122.

  In this embodiment, the optical integrator 124 is composed of a fly-eye lens in which microlenses are two-dimensionally arranged or have an equivalent effect, and uniformly illuminates the incident surface of the diffractive optical element 125a or 125b. The light beam emitted from each region of the optical integrator 124 becomes a light beam having substantially the same NA (light beam spread angle) in all regions.

  The diffractive optical elements 125a and 125b are provided in the vicinity of the exit surface of the optical integrator 124 so as to be switched by a driving device (not shown). The number of diffractive optical elements is not limited to two, and the drive device is, for example, a device that can arrange any one diffractive optical element on the optical path, such as a turret. The diffractive optical element is an element that diverges incident light into a desired angular distribution, and the emitted light angular distribution is projected onto a rear focal plane (so-called Fourier transform plane) of the optical system 126. The A plane corresponds to this Fourier transform plane. The Fourier transform relationship refers to the relationship between the object plane and the pupil plane or between the pupil plane and the image plane.

  Each of the optical integrators 122 and 124 has a configuration in which the incident side focal point (front focal point) position substantially coincides with the incident surface, and even when the angle of the incident light beam fluctuates, the light flux emitted from each of the optical integrators 122 and 124. The angle characteristics of do not change. By adopting the configuration as described above, a light beam having a constant angular characteristic is always incident on the diffractive optical element. For example, even if the light flux from the light source fluctuates, the A surface always has a constant light. It can be a distribution. The optical integrators 122 and 124 include an optical pipe using internal reflection, a diffraction grating, a fly-eye mirror in which an optical element constituting the multi-light source forming means is a mirror having a reflecting action, or a plurality of combinations thereof. An optical integrator or other uniformizing means may be used.

  The optical integrator 124 and the diffractive optical element are arranged at a predetermined distance from each other, and are set such that the light beams from adjacent microlenses constituting the optical integrator 124 partially overlap each other. Further, by illuminating the optical integrator 124 with a light beam that passes through the optical integrator 122 and the optical system 123, it is possible to irradiate the diffractive optical element with light on average and prevent light concentration. For example, when an ArF laser is used as a light source and the diffractive optical element is made of quartz, the diffractive optical element is damaged if the irradiation energy density on a part thereof is large. Such a configuration is taken as a preventive measure.

One of the reasons for illuminating the optical integrator 124 with a light beam that passes through the optical integrator 122 and the optical system 123 is that the light distribution characteristics of the incident light beam on the A-plane can be made constant. Even if the optical integrator 122 and the optical system 123 are not provided, a constant illuminance distribution can be obtained on the A plane. However, if the light distribution incident on the integrator 124 changes, the light distribution characteristic (incident angle distribution) of the incident light beam on the A plane changes. Change. This means that the angle distribution of the incident light on the reticle 200 surface changes slightly depending on the subsequent optical system and its aberration. In other words, even if there are fluctuations or machine differences in the light incident from the light source, this configuration is such that the A surface has a constant controlled light distribution and incident angle distribution unless the diffractive optical element is switched. It has been. In addition, although the light beam shape conversion means 120 of this embodiment is a structure of a double integrator, you may use the structure of a triple integrator.

  The pattern that can be formed on the A surface is a convolution of the Fourier pattern of the diffractive optical element (pattern formed on the Fourier transform plane when light is incident vertically at NA = 0) and the angular distribution of the light incident on the diffractive optical element. Result. Therefore, in order to bring the distribution on the A-plane closer to the desired distribution, it is desirable to make the NA of the light emitted from the optical integrator 124 as small as possible. For this reason, it is desirable to preserve the NA × diameter of the light beam emitted from the optical system 121 as much as possible and transmit it to the diffractive optical element. By switching the diffractive optical element, for example, a circular, annular or multipolar distribution can be formed on the A plane. Depending on the design of the diffractive optical element, for example, the intensity of each pole region in a multipole such as a quadrupole can be made different.

  Returning to FIG. 1, in the vicinity of the A surface, a conical optical element 132, a conical optical element 134 whose interval can be changed, a parallel flat plate (not shown), and an appropriately shaped diaphragm member (for example, an annular aperture diaphragm or quadrupole) (Aperture stop, circular stop, etc.) The light beam that has been changed to the basic shape by the light beam shape conversion means 120, such as a quadrangular pyramid type optical element, a roof type optical element, or an expansion / reduction beam expander for changing the magnification is further changed. A light beam changing means 130 for switching is arranged on the optical axis so as to be switchable. The light flux changing means 130 can be withdrawn from the optical path, or a plurality of them can be simultaneously arranged on the optical axis.

  The conical optical element 132 is a conical optical element having a concave conical shape on the entrance surface and a convex conical shape on the exit surface. For example, when the basic shape on the A surface is circular, the conical optical element 132 is arranged on the optical axis. An annular light beam is formed.

  The conical optical element 134 includes a conical optical element 134a having a concave incident surface and a flat exit surface, and a conical optical element 134b having a flat incident surface and a convex exit surface. When the basic shape on the A-plane is circular, if the conical optical element 134 is arranged on the optical axis, an annular light beam is formed, and the distance between the optical element 134a and the optical element 134b is changed. The ring-shaped luminous flux (ring zone ratio) and size can be changed. By adopting such a configuration for the conical optical element 134, it becomes possible to efficiently form an annular light beam in a smaller space. In addition, when the basic shape on the A surface is, for example, a quadrupole shape or a double pole shape, the inner diameter and the outer diameter can be changed.

  For example, by appropriately selecting the diffractive optical element shown in FIG. 2 and forming a circular distribution (FIG. 3A) on the A surface, the distance between the optical element 134a and the optical element 134b is changed. As shown in FIGS. 3 (b) and 3 (c), the shape (ring zone ratio) and size of the ring zone can be changed. By selecting another diffractive optical element in FIG. 2, when the distribution of the annular zone (FIG. 4 (a)) is formed on the A plane, the distance between the optical element 134a and the optical element 134b is changed. As shown in FIG. 4 (b) and FIG. 4 (c), the shape (annulus ratio) and size of the annular zone can be changed. By selecting another diffractive optical element in FIG. 2, when the quadrupole (FIG. 5 (a)) distribution is formed on the A plane, the distance between the optical element 134a and the optical element 134b is changed. As shown in FIGS. 5B and 5C, the ratio and size of the quadrupole can be changed.

  As will be described later, since the imaging optical system 140 is a variable magnification optical system, by changing the magnification and the aperture diameter of the stop 158, not only the ring zone ratio formed by the light beam changing means 130 but also a larger ring An effective light source having a desired size of band ratio (for example, 2/3 ring zone, 3/4 ring zone, etc.) can be formed.

  When the conical optical element 134 is used, the fly-eye lens 156 needs to satisfy the limit of the incident angle of light. If the angle of the incident light beam exceeds a certain angle, the light beam not only becomes unnecessary light, but also the shape of the effective light source collapses or uneven illuminance occurs. Therefore, the distance between the optical elements 134a and 134b and the magnification of the imaging optical system 140 are set within a range that does not exceed the limit of the incident angle to the fly-eye lens 156, and the aperture diameter of the diaphragm 158 is changed to change the outer diameter. It is effective to change the ring zone ratio by reducing. This is applicable not only to the annular illumination but also to quadrupole or dipole illumination.

The angles of the conical surfaces of the optical elements 134a and 134b are substantially the same angle. By using the same angle, it is possible to suppress an increase in the angle of the emitted light beam from the light beam shape changing means 122, and to minimize the flux of the light beam in the subsequent optical system. When the rear optical system has an angular margin, it is not always necessary to set the same angle. For example, the angle may be changed in order to reduce the zone width.

  Like the conical optical element 134, a variable-distance quadrupole conversion element in which the incident surface of the incident optical element is a concave quadrangular pyramid surface and the exit surface of the optical element on the exit side is a convex quadrangular pyramid surface, Similarly, it is also possible to apply a triangular roof type dipole conversion element. The shape formed by the light beam shape conversion unit 120 may be maintained without using the light beam conversion unit 130. Thus, by combining the optical elements of the light beam shape conversion means 120 and the light beam changing means 130, light beams having various shape distributions can be formed as real images or virtual images near the A plane.

  The magnification of the light beam formed on the surface A or changed to a desired shape by the light beam changing means 130 is changed in magnification by the imaging optical system 140 with variable magnification, and passes through a filter member 154 to be described later. Projected onto the entrance surface of the lens 156. The imaging optical system 140 includes lenses 142, 144, and 146 in this embodiment, but the number of lenses is not limited.

  On the incident surface of the fly-eye lens 156, when the light amount distribution on the predetermined surface A forms an image without aberration, the outline of the light intensity distribution becomes clear. In this case, unevenness in illuminance or non-uniformity of the effective light source on the screen occurs on the plate 400 that is the exposed surface. Therefore, it is desirable that the image formation relationship between the predetermined surface A and the incident surface of the fly-eye lens 156 is an image with some aberration (including defocus). However, this is not the case when there are a large number of lenses (microlenses) constituting the fly-eye lens 156 and the influence on the illuminance unevenness or the like is small.

  The fly-eye lens 156 forms a plurality of light source images (secondary light sources) in the vicinity of its exit surface by the incident light flux, and uniformly illuminates the reticle 200 surface. Near the surface (B surface) on which a plurality of light source images are formed, an aperture 158 with a variable diameter (including switching) is arranged. Note that the surface on which a plurality of light source images are formed (the rear condensing point surface of the microlens constituting the fly-eye lens) has a relatively high energy density of the luminous flux, and is thus slightly defocused with respect to that surface. A diaphragm 158 is disposed at (a position within a range of −several mm to + several mm). However, when the diaphragm 158 can withstand the high energy density, the diaphragm 158 may be arranged so as to coincide with a surface on which a plurality of light source images are formed.

  The aperture stop 158 and the aperture stop 310 are disposed at optically nearly conjugate positions. On the exit surface side of the stop 158, the image at the position of the aperture stop 310 in the shape of the multiple light source formed by the fly-eye lens 156 and the stop 158 is the shape of the illumination light (effective light source) at each point on the plate 400 surface. Shape).

  Of the light beams from the plurality of light source images, the light beam that is not limited by the stop 158 efficiently illuminates the surface on which the masking blade 172 is disposed by the condensing optical system 160. The masking blade 172 is disposed at a position optically conjugate with the surface on which the reticle 200 is disposed by the imaging optical system 180, and determines an illuminated area on the surface of the reticle 200. The condensing optical system 160 includes lenses 162 and 164 in the present embodiment, and the imaging optical system 180 includes lenses 182 and 184 in the present embodiment, but the number of these lenses is not limited.

  A half mirror 152 is disposed between the lenses 162 and 164 of the condensing optical system 160. The half mirror 152 divides the incident light beam into reflected light and transmitted light, one of which is indirectly monitored by the detector 170 with the exposure amount incident on the reticle 200 as one of the illumination light and the other incident on the reticle 200. It has a configuration. The arrangement of the half mirror 152 and the detector 170 is not limited to that shown in FIG. 1, and may be arranged in the optical path between the light source unit 102 and the masking blade 172. In the vicinity of the reticle 200, a detector 190 for measuring the effective light source distribution is provided that can be inserted and removed between the reticle 200 and the projection optical system 300.

  In this embodiment, the folding mirrors 150 and 151 and the half mirror 152 are arranged so that the optical path from the diffractive optical element shown in FIG. 2 to the surface of the plate 400 is within one plane as shown in FIG. ing. As a result, as will be described later, the configuration of the filter member 154 can be simplified.

  Hereinafter, the filter member 154 and the σ shape correction mechanism 128 will be described.

  First, the non-uniformity of the transmittance distribution of the illumination optical system 110 provided by the mirrors 150 to 152 will be described with reference to FIG. In FIG. 1, regarding the principal ray a and the rays b and c having a predetermined NA when viewed in a cross section parallel to the paper surface, when there are no folding mirrors 150 and 151 and half mirror 152, the rays b and c are optical axes. They are symmetrical, and there is no difference in transmittance between them in the optical path from the A plane to the reticle 200 plane. Even if these mirrors are present, if the light rays b and c are parallel in the region where each mirror is present, there is no difference in the transmittance of the light rays b and c.

  However, in practice, it is difficult to make the angles incident on the mirrors 150 and 151 and the half mirror 152 exactly the same due to design restrictions (optimization of space and aberration). There is no problem as long as a coating that does not exhibit angular characteristics in the transmittance (reflectance) of these optical elements is provided, but if the wavelength is 250 nm or less, particularly ArF (193 nm) or less, the coating material is also limited. Therefore, coating without angular characteristics is difficult. Therefore, a difference occurs in the transmittance of the rays a to c from the A surface to the reticle 200 surface.

  FIG. 6 is a graph showing an example of the transmittance. FIG. 7 shows an example of two-dimensional transmittance in the σ distribution. The density in FIG. 7 indicates that the transmittance distribution is non-uniform. Since the incident angles of light rays d and e positioned in the direction perpendicular to the paper surface on the A surface are almost the same as those of the light beam a, the transmittance distribution in σ is almost constant in the X direction as shown in FIG. And the distribution changes in the Y direction. Therefore, even if an XY symmetric distribution is formed on the A plane, the σ distribution on the reticle 200 plane is a distribution whose intensity in the Y direction is weaker than the X direction. When exposure is performed using this σ distribution, the imaging performance varies between the X-direction pattern and the Y-direction pattern, and a line width difference occurs between the two-direction patterns. Further, in the case of a catadioptric system using a mirror for the projection optical system after the reticle, even if the σ distribution on the reticle 200 surface is uniform, the σ distribution on the wafer 400 surface becomes non-uniform by the projection system mirror, and the X direction The imaging performance changes depending on the pattern and the pattern in the Y direction, and a line width difference occurs between the patterns in the two directions.

  That is, since the transmittance distribution to the pupil 310 is not uniform, the effective light source distribution does not become uniform. In the present embodiment, as a first means for solving such a problem, a filter member 154 that makes the transmittance distribution uniform is arranged. That is, the filter member 154 has a transmittance distribution that cancels the non-uniformity of the transmittance distribution caused by the illumination system mirror and the projection system mirror.

  In this embodiment, the filter member 154 is disposed immediately before the fly-eye lens 156. However, the filter member 154 is not limited to this position. The filter member 154 is substantially in a Fourier transform relationship with the reticle 200 surface that is the irradiated surface. (Or a position conjugate with the pupil 310 plane). In the present embodiment, the transmittance distribution to the pupil 310 is made uniform, the pupil 310 is in a Fourier transform relationship with the reticle 200, and the exit surface of the fly-eye lens 156 is conjugate with the pupil 310. The filter member 154 can be disposed in the vicinity of the entrance surface or exit surface of the fly-eye lens 156, or a surface optically conjugate with these surfaces. The reason why the filter member 154 may be disposed on the entrance surface of the fly-eye lens 156 is that the fly-eye lens 156 has a Fourier transform relationship between the entrance surface and the exit surface in each lens element. This is because the entire eye lens 156 substantially maintains an effective light source on the entrance surface and the exit surface (that is, before and after subdivision). In this sense, it is sufficient that the filter member 154 is provided on the surface (A surface) that determines the basic shape of the effective light source, or a position optically conjugate with or near the surface. "Position that is in a Fourier transform relationship" is a concept that includes a surface (A surface) that determines the basic shape of the effective light source, or a position optically conjugate with it or its vicinity.

  FIG. 8A and FIG. 8B show filter members 154a and 154b having different transmittance distributions. The filter member 154a has a transmittance that is symmetrical with respect to the X axis, and the filter member 154b has a slight offset in the Y direction. The density in FIG. 8 indicates the transmittance distribution, which is a distribution in which the reflectance and transmittance differences of the light rays a to c by the mirrors are canceled.

  With the filter member 154, the transmittance distribution in the pupil 310 can be made to be a substantially uniform distribution or a rotationally symmetric distribution. This distribution may be uniquely determined according to the design values of the illumination optical system 110 (mirror reflection characteristics and transmission characteristics). Alternatively, the detection unit 190 may measure the σ distribution on the reticle 200 surface, and the control unit 500 may selectively arrange the plurality of filter members 154. Alternatively, a larger filter member 154 may be arranged, and the position may be determined by determining the optimum position by shifting it in the Y direction for each device / each illumination state.

  A method for determining the transmittance distribution of the filter member 154 will be described with reference to FIG. Here, FIG. 9 is a flowchart for setting the transmittance distribution of the filter member 154.

  First, the illumination optical system 110 is designed so as to form a uniform transmittance distribution on the pupil 310 without the mirrors 150 to 152 (step 1002). At that time, it is preferable that the optical axis of the optical path from the light source unit 110 to the reticle 200 surface as the illumination target surface is maintained in one plane. This is because the configuration of the filter member 154 can be simplified like a density change filter in only one direction.

  Next, the non-uniformity of the transmittance distribution caused by the insertion of the projection system mirrors and illumination system mirrors 150 to 152 is obtained by simulation (step 1004). The non-uniformity of the transmittance distribution can be obtained by comparing the transmittance distribution before and after the insertion of the mirrors 150 to 152. If necessary, obtain transmission non-uniformity based on other factors.

  Next, the transmittance distribution that corrects the non-uniformity of the transmittance distribution caused by the projection system mirrors and the illumination system mirrors 150 to 152 measured in step 1004 is set in the filter member 154 (step 1006).

  For example, in FIG. 7, if the transmittance distributions at positions corresponding to the light rays a to e are 95%, 90%, 90%, 95%, and 95%, respectively, the light rays a to e are canceled. Transmittance distributions are set in the filter member 154 so that the transmittance distributions at corresponding positions are relatively 90%, 95%, 95%, 90%, and 90%, respectively. Such a transmittance distribution applies to the filter member 154a.

  On the other hand, for example, in FIG. 7, if the transmittance distributions at positions corresponding to the light rays a to e are 95%, 90%, 93%, 95%, and 95%, respectively, first, as described above, the light rays a to The transmittance distributions at positions corresponding to e are relatively 90%, 95%, about 92%, 90%, and 90%, respectively, and the transmittance distributions for the rays a to e are uniform. The transmittance distribution is determined by offsetting it in the Y direction by a predetermined direction so as to be set to the filter member 154. Such a transmittance distribution applies to the filter member 154b. In the above, an example of correction at three points in the Y direction has been shown. However, in actuality, it is possible to measure or interpolate more subdivided points and set a filter member having a distribution (cancelled) in the distribution. desirable.

  The filter members 154a and 154b have substantially the same distribution in the X direction and continuously change in the Y direction. This is because, as shown in FIG. 1, the mirrors 150 to 152 are arranged so that the optical paths of the A plane and the reticle plane on which the basic shape of the effective light source is determined are within one plane. As described above, since the incident angles of the light beams d and e positioned in the direction perpendicular to the paper surface to the mirrors are almost the same as those of the light beam a, the correction can be made with such a filter distribution. Manufacture is easy because of the filter with density change in one direction. For example, when this filter is manufactured by vapor deposition of a metal film or the like, such a filter is deposited on the substrate by performing vapor deposition while controlling the positions of two shielding plates disposed between the substrate and the vapor deposition source during vapor deposition. A filter having a distribution in the direction can be easily and inexpensively manufactured.

  When the optical path from the A surface to the reticle 200 surface cannot be accommodated in the plane due to the arrangement of the illumination optical system 110, two filter members 154 having a distribution in one direction are arranged with the distribution direction rotated by 90 °. May be. Of course, even when the optical path from the A surface to the reticle 200 surface can be accommodated in a plane, two filter members arranged by rotating the distribution direction by 90 ° may be disposed. The two sheets may be shifted independently in the respective distribution directions and their positions may be adjusted. If necessary, two or more filter members 154 may be provided. At least two or more filter members 154 may have the same transmittance distribution or different transmittance distributions.

  Next, the filter member 154 having the transmittance distribution set in step 1006 is arranged on the illumination optical system 110 (on the optical path thereof) (step 1008). As described above, the arrangement position is a position substantially in a Fourier transform relationship with the surface to be illuminated. As a result, the transmittance distribution of the filter member 154 has a transmittance distribution set in advance so as to correct the non-uniformity of the transmittance distribution of the illumination optical system 110 and the projection optical system 300 caused by the mirror.

  Note that the control unit 500 may switch the filter member 154 each time the illumination state is changed. Changing the illumination state typically means changing the σ distribution on the reticle 200 surface. However, in the present embodiment, even if the distribution itself is not changed, the illumination state is also changed. Include in changes. When the NA of the projection optical system increases as in a recent exposure apparatus, the imaging characteristics change depending on the polarization state of light incident on the wafer surface. Therefore, when a finer pattern is formed by exposure with high NA, illumination with a controlled polarization state has been proposed in order to improve image quality. For example, in the case of a line and space extending in the X direction, it is preferable that the polarization direction of the illumination light that illuminates the pattern is also specialized in the X direction. That is, depending on the pattern, it is conceivable to perform optimal exposure by switching the polarization state.

  In general, the mirror described above has a difference in characteristics with respect to polarized light in addition to the angle characteristics as shown in FIG. Therefore, even with the same distribution shape on the A plane, if the polarization state is different (especially in the case of illumination in which the polarization is positively controlled as described above), naturally the transmittance distribution to the pupil 310 of the projection lens will be different. A plurality of filter members 154 are provided in a switching means such as a turret, and the detection unit 190 measures the σ distribution on the reticle 200 surface, and the control unit 500 selects from among the plurality of filter members 154 each time the illumination state is changed. An appropriate filter member 154 may be selected and placed on the optical path.

  In this case, the illumination optical system 100 can use a configuration as shown in FIG. Here, FIG. 17 is an enlarged configuration diagram of the illumination optical system shown in FIG. 1 when performing polarized illumination and when performing non-polarized illumination. If the light source unit 102 is a laser, the linearly polarized light of the laser can be used as it is. In addition, X-polarized light must be incident on the illumination optical system uniformly regardless of the difference in the polarization direction of the emitted laser light depending on the laser installation state and the configuration of the laser routing optical system. Therefore, in the case where the polarization direction of the emitted laser light and the reflection by the deflecting mirrors 103, 104, and 112 result in Y deflection incidence, it is desirable to arrange the λ / 2 phase plate 111 so that X polarization incidence occurs.

  The phase canceling plate (or phase adjusting plate) 113 is for converting linearly polarized light into random polarized light and is inserted into the optical path during non-polarized illumination, and retracts from the optical path during polarized illumination. The mirrors 150 and 151 are broadband high reflection film mirrors (BBHR) mirrors, and the BBHR film suppresses the phase difference between the S-polarized light and the P-polarized light generated in the film with respect to a wide-band incident angle. Designed to be

  Reference numeral 154 denotes an ND filter, and 155 denotes a λ / 2 phase plate. In this embodiment, a plurality of types of ND filters 154 and a plurality of types of λ / 2 phase plates 155 are provided, which are paired. FIGS. 18A and 19A show different types of ND filter 154 and λ / 2 phase plate 155 pairs. FIG. 18A is a schematic plan view showing a pair of an ND filter 154a and a λ / 2 phase plate 155a used for tangential polarization illumination, and FIG. 19A is used for cross-pole polarization illumination. It is a schematic plan view showing a pair of an ND filter 154b and a λ / 2 phase plate 155b. Reference number 154 summarizes 154a and 154b, and reference number 155 summarizes 155a and 155b.

  In the polarization illumination of this embodiment, the λ / 2 phase plate 155 sets a predetermined polarization state for a plurality of regions of the effective light source in this way. The ND filter 154 includes a filter member corresponding to each region, and each filter member is set in advance so as to correct the non-uniformity of the transmittance distribution due to the difference in the polarization state of light caused by the mirror.

  FIGS. 18B and 19B correspond to FIG. 6 and show transmittance distributions depending on the incident angles of the ND filters 154a and 154b (the vertical axis of the graph represents the transmittance). An example of the transmittance distribution set in the ND filter 154a is shown in FIG. For example, as shown in FIG. 18 (a), although the same polarization states A and E are set, the region a and e of the ND filter 154a is 96% open (see FIG. 18C). The reason why the transmittance distribution of 100% is set is that the angle dependency of the transmittance shown in FIG. 18B is considered. In this embodiment, the transmittance of one filter is set to be constant. In the case of polarized illumination, modified illumination is usually used like dipole illumination, and in modified illumination, a limited area such as a dipole is used. This is because the transmittance difference due to the incident angle in a region actually used in one region (in the filter) is small. However, it is optional whether or not to consider the angle-dependent characteristics in one filter.

  Thus, when different polarization states are generated in different regions of the pupil using the phase plate 155, the filter member 154 suitable for the polarization state of each region is provided with the influence of the mirror showing different transmittance for each region. Can be corrected. In FIG. 1, for convenience of drawing, the control unit 500 is not connected to the filter member 154, but the control unit 500 can control the filter member 154 via switching means (not shown). The filter member 154 and the phase plate 155 are exchanged in pairs. As described above, step 1008 is not limited to the initial arrangement of the filter member 154.

  The detection unit 190 may be configured to be able to measure for each polarization component (for example, an X-direction polarization component and a Y-direction polarization component) using, for example, two sensors. Of course, the position of the effective light source detector is not limited to this position, and may be arranged on the wafer stage, for example. Further, it may be arranged at the position of the light quantity detector 170. However, the detection at this position does not reflect the reflection / transmission characteristics of the half mirror 152 and the reflection characteristics of the mirror 151, and therefore, these characteristics are taken into consideration and re-calculated before being effective on the irradiated surface. It is necessary to calculate the light source distribution.

  The filter member 154 is not limited to an optical filter in which the optical density distribution changes. For example, like the filter member 154c in FIG. 8C, the mechanical light shielding portion 155 (the pitch and the light shielding width are different in the Y direction). May be used). In FIG. 8C, the black line is the light shielding portion 155, and the white area is the light transmitting portion. However, if a mechanical light-shielding member is disposed immediately before the fly-eye lens 156, the structure appears on the incident surface of the fly-eye lens 156, and the influence thereof varies depending on the irradiation position of the illuminance distribution and σ distribution on the illuminated surface. It will appear. For this reason, it is desirable that the fine structure be disposed a predetermined amount away from the incident surface of the fly-eye lens 156 so that the fine structure does not appear on the incident surface of the fly-eye lens 156.

  The mechanical filter 154 is particularly effective for an optical system in which a refracting member cannot be used (only a mirror) like an exposure apparatus using EUV as a light source. The configuration of the EUV exposure apparatus is completely different from that described above, and is basically a reflection optical system using a mirror from the light source to the irradiation surface. Since many mirrors are used, asymmetry of the effective light source tends to occur. Therefore, asymmetry can be corrected by mounting one or more mechanical filters such as 154 near the pupil plane of the illumination surface.

  Although the description has been made with the configuration in which the diffractive optical element is included in the light beam shape conversion unit 120, the gist of the present embodiment is that the projection system mirrors and the illumination system mirrors 150 to 152 bring about, or other factors The diffractive optical element is not necessarily an essential configuration. As shown in FIG. 1, it is apparent that the present embodiment can be applied to any optical system that forms a predetermined light shape distribution controlled on the A plane.

  It is also possible to keep away from the ideal σ distribution. For example, the reticle pattern itself has a direction difference in line width, a pattern direction difference due to the aberration factor of the projection lens, or the exposure apparatus 1 has a line width exposed in the scanning direction and a non-scanning direction perpendicular thereto. If there is a difference, a filter that corrects the direction difference of the actually exposed pattern may be selected. As a second means for solving the problem that the effective light source distribution does not become uniform because the transmittance distribution up to the pupil 310 is not uniform, the integrated intensity on the illuminated surface is made uniform by changing the effective light source. It is. For this reason, in this embodiment, a σ shape correction mechanism 128 is provided. The σ shape correction mechanism 128 may be applied simultaneously with the filter member 154, or both may be applied independently.

  In the preferred embodiment, in order to shorten the adjustment time and facilitate the adjustment, the transmittance of the fixed σ distribution that does not depend on the illumination state such as the projection system mirror and the illumination system mirrors 150 to 152 is corrected by the filter member 154 as much as possible. The σ shape correction mechanism 128 may perform fine correction for each illumination state. The σ shape correction mechanism 128 can bring the σ distribution on the illumination surface closer to the ideal. In particular, when performing illumination in which the polarization state is controlled, asymmetry of the σ distribution is likely to occur due to the mirror, half mirror, and antireflection film. Can be adjusted to the state.

  It is also possible to keep away from the ideal σ distribution. For example, the reticle pattern itself has a direction difference in line width, a pattern direction difference due to the aberration factor of the projection lens, or the exposure apparatus 1 has a line width exposed in the scanning direction and a non-scanning direction perpendicular thereto. If a difference has occurred, the σ shape correction mechanism 128 can correct the difference in the direction of the line width of the pattern that is finally exposed by changing the symmetry of the σ distribution.

  Hereinafter, the operation of the σ shape correction mechanism 128 will be described with reference to FIG. Here, FIG. 10 shows the operation of the σ shape correction mechanism 128 when the light beam shape converting means 120 forms a uniform light intensity distribution on the A surface and the light intensity distribution on the A surface (cross-sectional view cut on the paper surface). FIG. Each reference number in the figure corresponds to FIG. A distribution having a predetermined shape is formed on the A surface by the light beam emitted from the light beam shape conversion means 120. By forming two or more optical integrators (haenome lens, internal reflection type integrator, diffractive optical element, etc., and combinations thereof) in the light beam shape conversion means 120, a desired shape whose distribution is controlled on the A plane Not only the distribution but also the angle characteristic of the incident light beam is controlled. Of course, even if the light flux from the light source fluctuates, both distribution and angle characteristics are maintained constant. The σ shape correction mechanism 128 includes light shielding portions 129a and 129b.

  In FIG. 10A, the light shielding portions 129a and 129b of the σ shape correction mechanism 128 are arranged in the vicinity of the A plane, and are set in a state in which the light beam from the light beam shape conversion unit 120 is not limited. Therefore, the light intensity distribution on the A plane remains uniform.

  FIG. 10B shows a state in which the light shielding portion 129a is inserted into the light beam in the state of FIG. Since the light shielding portion 129a shields the light flux, the distribution on the A surface is partially lost. Thereafter, the distribution on the A plane is converted into another shape by the light beam changing means 130 as required. However, the incident surface of the fly-eye lens 156, that is, the σ distribution depends on the distribution shape on the A plane. Partly missing. Thus, the σ distribution can be changed by changing the state of the σ shape correction mechanism 128. For example, similarly to the light shielding part 129a, the σ distribution (size) in two orthogonal directions can be changed by inserting the light shielding part 129b into the optical path. The movement of the light shielding unit 129a is performed by the control unit 500 based on the output of the detection unit 190, for example. The control unit 500 determines whether the distribution of the effective light source deviates from a predetermined distribution, and changes the effective light source by the diaphragm 128 based on the determination result. For example, when the transmittance is non-uniform, the movement of the light-shielding portion 129a is controlled so that the integrated intensity in the plurality of directions from the optical axis of the effective light source distribution on the reticle 200 surface is substantially uniform.

  FIG. 10C shows a case where the light shielding portions 129a and 129b of the σ shape correction mechanism 128 are arranged away from the A plane. In this case, the distribution on the A plane is not a partially missing shape as shown in FIG. 10B, but a distribution with a reduced strength while maintaining the outer shape. In the final σ distribution, such a configuration is preferable when it is desired to reduce a part of the strength without changing the outer shape. The movement of the light shielding unit 129a is performed by the control unit 500 based on the output of the detection unit 190, as in FIG. In this way, by adjusting the light shielding portion of the σ shape correction mechanism 128, a part of the intensity can be changed with a simple configuration. The position of the σ shape correction mechanism 128 from the A surface may be changed as necessary as described above, or may be fixed somewhere. Further, the position at which the σ shape correction mechanism 128 is disposed is not limited to the vicinity of the A surface, but may be, for example, the vicinity of the incident surface of the fly-eye lens 156.

  FIG. 11 is a schematic plan view showing a configuration example of the σ shape correction mechanism 128. FIG. 11A shows a σ shape correction mechanism 128a configured by four light shielding portions 129c that can be independently driven. For example, the control unit 500 may acquire the output from the measurement unit 190 of FIG. 1 and move the position of each light shielding unit 129c so that the XY difference of the effective light source shape is minimized based on the information. Alternatively, the position of each light shielding portion 129c may be changed depending on the exposure result so that the pattern direction difference after exposure is minimized.

  When the telecentricity (perpendicularity of light rays) on the reticle 200 surface is changed by driving the σ shape correction mechanism 128a, for example, some lenses of the light beam shape conversion means 120 (some lenses of the optical system 126). ) Or a part of the lens of the imaging optical system 140 capable of zooming may be adjusted by decentering from the optical axis. Of course, telecentric adjustment can also be performed by the σ shape correction mechanism 128a itself. The σ shape correction mechanism 128a can reduce the eccentricity and asymmetry of the σ distribution.

  FIG. 11B shows a σ shape correction mechanism 128b using eight light-shielding portions 129d that can be independently driven. The σ shape correction mechanism 128b can provide multidirectional correction. FIG. 11C shows a σ shape correction mechanism 128c in which the light shielding edges of the four light shielding portions 129e are formed in a curved shape. Usually, the σ distribution is based on a circular shape or an annular shape, and conforms to these shapes.

  In the polarized illumination, the stop 128 can be used instead of the filter member 154 or together with the filter member. In this case, the diaphragm of the σ shape correction mechanism 128b or the σ shape correction mechanism 128d shown in FIG. 16A can be used for the filter member 154a and the phase plate 155a shown in FIG. Further, the diaphragm of the σ shape correction mechanism 128e shown in FIG. 16B can be used for the filter member 154b and the phase plate 155b shown in FIG. 18B. The functions of the σ shape correction mechanisms 128d and 128e are the same as those of the σ shape correction mechanisms 128a to 128c.

  When the effective light source is adjusted by the σ shape correction mechanisms 128a to 128e, the effective coherence factor σ in the direction in which the light shielding portion is moved to the center decreases. That is, in order to reduce the direction difference, the effective coherence factor σ is aligned to the smaller side. In this case, it is conceivable that the magnitude of σ becomes small on average. In order to correct this, it is possible to adjust the average size of σ by increasing the magnification of the imaging optical system 140 capable of zooming.

  Further, by using the σ shape correction function, it is possible not only to finely correct the shape but also to actively change the effective light source. For example, from this state of annular illumination or quadrupole illumination, a dipole illumination state can be created using this mechanism.

  The reticle 200 is made of, for example, quartz, on which a circuit pattern (or image) to be transferred is formed, and is supported and driven by a reticle stage (not shown). Diffracted light emitted from the reticle 200 passes through the projection optical system 300 and is projected onto the plate 400. The reticle 200 and the plate 400 are optically conjugate. Since the exposure apparatus 1 of the present embodiment is a step-and-scan system (also called “stepper”), the pattern of the reticle 200 is scanned on the plate 400 by scanning the reticle 200 and the plate 400 at the speed ratio of the reduction ratio. Transfer on top. In the case of a step-and-repeat type exposure apparatus (also referred to as “stepper”), exposure is performed with the reticle 200 and the plate 400 being stationary.

  The projection optical system 300 is an optical system that projects light reflecting the pattern on the reticle 200 onto the plate 400. The projection optical system 300 includes an aperture stop 310 and can be set to an arbitrary numerical aperture (NA). The aperture stop 310 has a variable aperture diameter that defines the NA of the imaging light beam on the plate 400, and the aperture diameter is changed to change the NA as necessary. In the present embodiment, the coherence factor σ can be said to be the ratio of the image size at the position of the aperture stop 310 of the plurality of light sources formed by the fly-eye lens 156 and the aperture diameter of the aperture stop 310.

  The B surface (the surface on which a plurality of multiple light sources are formed) and the aperture stop 310 having a variable diameter are disposed at optically substantially conjugate positions, and the distribution on the B surface is substantially σ on the plate 400 surface. Distribution or effective light source. When the diaphragm 158 is mounted on the B surface, the distribution that is not limited by the diaphragm 158 is the σ distribution. Further, if there is no stop 158 on the B surface and the fly-eye lens 156 is sufficiently fine (several tens of rows in one direction), the light beam shape converting means 120, the light beam changing means 130, and the imaging optical system The distribution on the entrance surface of the fly-eye lens 156 formed by the combination of 140 becomes a substantial σ distribution.

  The position of the diaphragm 158 is not limited to the vicinity of the B surface. For example, the diaphragm 158 may be selectively inserted into the optical path by the switching means such as the turret together with the light flux changing means 130 on the A surface, or may be disposed immediately before the fly-eye lens 156, or a plurality of these May be arranged at the same time. For example, a diaphragm having a mechanism that can change only the aperture angle such as a quadrupole without limiting the radiation direction (direction in which the size is limited) is selectively disposed at the position of the light beam changing unit 130, and the size is set. It is also possible to arrange the iris diaphragm to be limited immediately before the fly-eye lens 156, selectively arrange the fixed diaphragm on the B surface, and create a desired σ distribution by this combination. In this way, by arranging the apertures that share the functions at a plurality of positions and changing / switching the apertures, it becomes possible to cope with more various σ conditions.

  In the present embodiment, the projection optical system 300 is an optical system having a plurality of lens elements 320 and 322, but an optical system (catadioptric optical system) having a plurality of lens elements and at least one concave mirror, a plurality of optical elements. An optical system having a lens element and at least one diffractive optical element such as a kinoform, an all-mirror optical system, or the like can be used. When correction of the chromatic aberration of the projection optical system 300 is necessary, a plurality of lens elements made of glass materials having different dispersion values (Abbe values) are used, or dispersion of the diffractive optical element in a direction opposite to the lens element occurs. Or configure as follows. In such a catadioptric optical system, even if the non-uniformity of the transmittance distribution generated in the illumination system 110 is corrected, the non-uniformity of the transmittance distribution at the pupil 310 of the projection lens is caused by the concave mirror in the projection system. Occur. For this reason, it is necessary to arrange a filter and a diaphragm that cancel the non-uniformity of the transmittance distribution generated by both the illumination system mirror and the projection system mirror. In another embodiment, the final surface of the projection optics 300 and the plate 400 are filled with liquid (such as pure). In such a so-called immersion type projection exposure apparatus, the effect of polarization control is particularly great due to high NA.

  The plate 400 is a wafer in this embodiment, but widely includes a liquid crystal substrate and other objects to be processed. Photoresist is applied to the plate 400.

  The plate stage 450 supports the plate 400. Since any configuration known in the art can be applied to the plate stage 450, a detailed description of the structure and operation is omitted here. For example, the plate stage 450 can move the plate 400 in the XY directions using a linear motor. The reticle 200 and the plate 400 are synchronously scanned, for example, and the positions of the plate stage 450 and the reticle stage (not shown) are monitored by, for example, a laser interferometer, and both are driven at a constant speed ratio. The plate stage 450 is provided on a stage surface plate supported on a floor or the like via a damper, for example. The reticle stage and the projection optical system 300 (not shown) are mounted on a base frame placed on the floor or the like, for example. Are provided on a lens barrel surface plate (not shown) supported via a damper.

  In the exposure, the light beam emitted from the light source unit 110 illuminates the reticle 200 by the illumination optical system 120, for example, Koehler illumination. Light that passes through the reticle 200 and reflects the reticle pattern is imaged on the wafer 400 by the projection optical system 300. Since the exposure apparatus 1 forms a desired effective light source by the filter member 154 and / or the σ shape correction mechanism 128, a device (semiconductor element, LCD element, imaging element (CCD, etc.), high-resolution and excellent performance, A thin film magnetic head).

  Although the combined use of the filter member 154 and the σ shape correction mechanism 128 has been briefly described, a specific example will be described below.

  First, selection of the filter member 154 will be described. As described above, it is desirable that the filter member 154 is mounted so as to correct fixed asymmetry (asymmetry generated by the mirror) independent of the effective light source distribution. The transmittance distribution of the filter 154 to be mounted is determined by one of the following methods.

  The first method calculates the transmittance distribution within the σ distribution by using the design values of the mirrors arranged from the A surface to the non-irradiated surface and the results of measuring the reflection (or transmission) characteristics after fabrication, Select a distribution filter that cancels the distribution.

  The second method irradiates a fly's eye incident surface with a distribution in which the effective σ region is made as uniform as possible on the A plane, and measures the actual effective light source distribution on the irradiated surface. Compare the effective light source distribution (design value effective light source distribution) on the irradiated surface that is expected when there is no mirror and the actual effective light source distribution, and the actual effective light source distribution is the design value effective light source distribution. Select a filter with a distribution that is nearly identical.

  In the third method, a pattern in two directions (X direction and Y direction) is actually exposed in a typical illumination mode (annular illumination with a large outer σ, etc.). Select a filter whose directional patterns are approximately the same size.

  The filter member 154 is selected by the determination method as described above. For example, in the case of an illumination optical system that controls the polarization state, the filter may be selected for each polarization state. In the case of an illumination system that controls the polarization state, the polarization state is roughly divided into five as shown in FIGS. 14 (a) to 14 (e). Here, FIG. 14 shows the polarization state of the light beam in the σ distribution. FIG. 14A shows ordinary non-polarized illumination (or circularly polarized illumination). FIG. 14B shows Y-direction linearly polarized illumination. FIG. 14C shows X-direction linearly polarized illumination. FIG. 14 (d) shows the illumination polarized in the tangential direction. FIG. 14 (e) shows illumination polarized in the radial direction.

  One filter may be prepared for each polarization state, arranged in a turret, and switched according to the polarization state to be set. Of course, for example, only a filter optimized for a non-polarization state may be attached regardless of the polarization state.

  Further, as shown in FIG. 16, a configuration in which different polarization states are generated in different regions of the pupil using a plurality of phase plates is also conceivable. In this case, although the transmittance varies depending on the region due to the influence of the mirror or the like, it is desirable to provide a filter suitable for the polarization state of each region in each region to correct the transmittance distribution of the entire pupil. Also, since the transmittance distribution changes every time the polarization state is changed, at the same time as the polarization state changes, that is, when the phase plate changes, the filter is exchanged or combined with one suitable for each polarization state. The structure which does is preferable. For example, if several kinds of polarization states used for exposure are known, and several kinds of phase plates that form the polarization states are prepared in advance such as a turret, the polarization state generated when the phase plate is used By investigating the optimum filter and installing the filter on the turret that moves in synchronization with the phase plate turret, the phase plate turret is moved to change the polarization state, and at the same time, a filter suitable for the polarization state is obtained. A changed configuration is desirable.

  Next, the setting of the σ shape correction mechanism 128 will be described. The σ shape correction mechanism is set at each position according to a change in the illumination state. The setting method is determined by one of the following.

  In the first method, the effective light source distribution is measured by the detector 190 or the like without limiting the light beam by the σ shape correction mechanism 128. The measured effective light source distribution is divided into a plurality of (for example, four) regions, and the light quantity ratio of the plurality of portions is measured. The respective light shielding portions of the σ shape correction mechanism 128 are driven so that the light quantity ratio becomes a desired value.

In the second method, exposure of a plurality of patterns extending in a plurality (for example, two directions) is performed without limiting the light beam by the σ shape correction mechanism 128. The respective light shielding portions of the σ shape correction mechanism 128 are driven so that the difference between the line widths of the respective patterns becomes a desired difference.
The above describes the case where two means are used in combination. Of course, even if used alone, there is a considerable effect.

  In the case of an illumination optical system that controls the polarization state, the light shielding unit may be driven for each polarization state. In the case of an illumination system that controls the polarization state, the polarization state is as shown in FIGS. 14A to 14E, and the light shielding unit may be driven for each polarization state. Of course, the light shielding unit may be driven to a position optimized for the non-polarized state, for example, regardless of the polarization state.

Further, as shown in FIG. 16, a configuration in which different polarization states are generated in different regions of the pupil using a plurality of phase plates is also conceivable. In this case, the transmittance varies depending on the region due to the influence of the mirror, etc., but the shading unit is driven to a position suitable for the polarization state of each region, and the effective light source contour is not corrected or the contour is not changed. The intensity distribution may be corrected.
In addition, since the transmittance distribution changes every time the polarization state is changed, it is preferable that the stop is made into a stop shape suitable for each polarization state at the same time as the polarization state changes, that is, the phase plate changes. .

  For example, if several kinds of polarization states used for exposure are known, and several kinds of phase plates that form the polarization states are prepared in advance such as a turret, the polarization state generated when the phase plate is used Investigate the optimal aperture shape (width and position of the light shielding plate) and aperture configuration (number of light shielding plates), install the aperture on the turret that moves in synchronization with the phase plate turret, and move the phase plate turret It is desirable that the diaphragm suitable for the polarization state be changed simultaneously with changing the polarization state. In addition, when the polarization state changes slightly, it is desirable to drive the light shielding unit each time.

  Next, with reference to FIGS. 12 and 13, an embodiment of a device manufacturing method using the above-described exposure apparatus 1 will be described. FIG. 12 is a flowchart for explaining how to fabricate devices (ie, semiconductor chips such as IC and LSI, LCDs, CCDs, and the like). Here, the manufacture of a semiconductor chip will be described as an example. In step 1 (circuit design), a device circuit is designed. In step 2 (mask production), a mask on which the designed circuit pattern is formed is produced. In step 3 (wafer manufacture), a wafer is manufactured using a material such as silicon. Step 4 (wafer process) is called a pre-process, and an actual circuit is formed on the wafer by the lithography technique of the present invention using the mask and the wafer. Step 5 (assembly) is called a post-process, and is a process for forming a semiconductor chip using the wafer created in step 4. The assembly process (dicing, bonding), packaging process (chip encapsulation), and the like are performed. Including. In step 6 (inspection), inspections such as an operation confirmation test and a durability test of the semiconductor device created in step 5 are performed. Through these steps, a semiconductor device is completed and shipped (step 7).

  FIG. 13 is a detailed flowchart of the wafer process in Step 4. In step 11 (oxidation), the surface of the wafer is oxidized. In step 12 (CVD), an insulating film is formed on the surface of the wafer. In step 13 (electrode formation), an electrode is formed on the wafer by vapor deposition or the like. Step 14 (ion implantation) implants ions into the wafer. In step 15 (resist process), a photosensitive agent is applied to the wafer. Step 16 (exposure) uses the exposure apparatus 1 to expose a circuit pattern on the mask onto the wafer. In step 17 (development), the exposed wafer is developed. In step 18 (etching), portions other than the developed resist image are removed. In step 19 (resist stripping), the resist that has become unnecessary after the etching is removed. By repeatedly performing these steps, multiple circuit patterns are formed on the wafer. According to this device manufacturing method, it is possible to manufacture a higher quality device than before. Thus, the device manufacturing method using the exposure apparatus 1 and the resulting device also constitute one aspect of the present invention.

  The preferred embodiments of the present invention have been described above, but the present invention is not limited to these embodiments, and various modifications and changes can be made within the scope of the gist.

  The exposure apparatus 1 of the present embodiment employs a mirror arrangement in which the optical path from the surface (A surface) forming the basic shape of the effective light source to the reticle 200 surface is contained in one plane. In addition, the exposure apparatus 1 has a filter member 154 that corrects its mirror characteristics in the vicinity of the effective light source formation surface, makes the σ distribution symmetrical with respect to the optical axis with a simple structure, and the direction of the exposure pattern of the effective light source factor. The difference is suppressed. In the exposure apparatus 1, a σ shape correction mechanism 128 having a plurality of light-shielding portions that can be independently driven is arranged in the vicinity of the A surface so that the σ shape can be changed. In addition, it is possible to reduce the direction difference of the exposure pattern caused by factors other than the illumination optical system. Further, the exposure apparatus 1 is provided with various light-shielding members (fixed diaphragm, variable aperture angle diaphragm, iris diaphragm, σ shape correction mechanism, etc.) that determine the illumination state at a plurality of locations and share functions. It makes it possible to create an illumination state.

DESCRIPTION OF SYMBOLS 1 Exposure apparatus 110 Illumination optical system 120 Light beam shape conversion means 128 (sigma) shape correction mechanism 150,151 Bending mirror 152 Half mirror 154 Filter member 156 Effective light source formation means (fly eye lens)
158 Aperture 200 Reticle 300 Projection optical system 310 Aperture stop 400 Plate

Claims (9)

With a light beam from the light source to a illumination optical system for illuminating an illumination target surface,
Mirror,
A filter member disposed at a position substantially in a Fourier transform relationship with the irradiated surface;
The transmittance distribution of the filter member is changed in a first direction, Ri second substantially constant der in a direction perpendicular to the first direction,
In the first direction of the filter member, light incident at different angles to the mirror is incident,
In the second direction of the filter member, light having an angle of incidence on the mirror is substantially constant,
First transmittance distribution in the direction of the illumination optical system characterized that you have changed so as to reduce the intensity difference of the light reflected by the mirror according to the incident angle characteristics of the mirror of the filter member. A plurality of the filter members;
A light beam shape converting means for forming an effective light source corresponding to an incident angle distribution of a light beam incident on the irradiated surface,
The filter member more than the illumination optical system according to claim 1, characterized in Rukoto switched in accordance with the effective light source. A plurality of the filter members;
A light beam shape converting means for forming an effective light source corresponding to an incident angle distribution of a light beam incident on the irradiated surface;
And a polarization setting means for setting a polarization state of the light beam that put the plurality of regions of the effective light source,
The filter member more than, the effective light source and an illumination optical system according to claim 1, characterized in that switched in response to at least one of said polarization states. The illumination optical system according to any one of claims 1 to 3, further comprising a plurality of the filter members having different transmittance distributions in the first direction. The illumination optical system according to claim 1, wherein the mirror is a deflection mirror. The illumination optical system according to claim 1, wherein the mirror is a half mirror. The illumination optical system according to claim 1, comprising a plurality of the mirrors. The illumination optical system according to any one of claims 1 to 7 , for illuminating a reticle arranged on an irradiated surface using a light beam from a light source;
An exposure apparatus comprising: a projection optical system that projects a reticle pattern onto an object to be exposed. Exposing an object to be exposed using the exposure apparatus according to claim 8 ;
And developing the exposed object to be exposed.

Images