JP5600128B2 - Catadioptric projection objective - Google Patents

Description

  The present invention relates to a catadioptric projection objective for imaging a pattern arranged on an object plane on an image plane.

  This type of projection objective is used in projection exposure systems, particularly in wafer scanners and wafer steppers used in the manufacture of semiconductor devices and other types of microdevices, and hereinafter generally referred to as “mask” or “reticle” photo It is useful for projecting a mask or reticle pattern onto an object with a photosensitive coating on a reduced scale with very high resolution.

  For the purpose of producing finer structures, it is necessary to increase the image side edge numerical aperture (NA) of the relevant projection objective and to use ultraviolet light having a shorter wavelength, preferably less than about 260 nm. It has been.

  However, there are few fully transmissive materials, especially synthetic quartz glass and crystalline fluorides that can be used to manufacture optical elements required in that wavelength region. Since the Abbe numbers of these materials available are fairly close to each other, it is difficult to provide a purely refractive system that is sufficiently well color corrected (corrected for chromatic aberration).

  In order to expose a planar substrate such as a semiconductor wafer in lithography, a planar (planar) image is essential. However, in general, the image plane of the optical system is curved, and the degree of curvature is determined by Petzval sum. Petzval sum correction is becoming more important in that it requires the projection of a large object field onto a planar surface with increased resolution.

  One approach to obtaining a planar image surface and good color correction is the use of a catadioptric system, which combines both a refractive element such as a lens and a reflective element such as a mirror. , Preferably having at least one concave mirror. A lens with positive and negative refractive power contributes to the total refractive power in the optical system, whereas surface curvature and chromatic aberration are opposite to each other, and a concave mirror has positive power like a positive power lens However, it has the opposite effect on the surface curvature without contributing to chromatic aberration.

  In addition, the high cost of related materials that are large enough to produce large lenses and the limited availability of calcium fluoride crystals are problematic. Thus, a means is desired that can reduce the number and size of lenses and at the same time maintain the accuracy of imaging and contribute to further improvements.

  Catadioptric projection objectives having at least two concave mirrors have been proposed, providing the system with good color correction and a reasonable lens requirement. US Pat. No. 6,600,608 B1 discloses a first pure refractive objective part for imaging a pattern arranged in the object plane of a projection objective on a first intermediate image and a first intermediate image as a second intermediate. A second objective part for imaging the image and a third objective part for imaging the second intermediate image directly, i.e. without further intermediate images, on the image plane. The second objective part is a catadioptric objective part having a first concave mirror having a central hole and a second concave mirror having a central hole, and these concave mirrors are opposed to each other and have an inter-mirror space or catadioptoscope. A mirror surface defining a lick cavity; The first intermediate image is formed in the center hole of the concave mirror next to the object plane, and the second intermediate image is formed in the center hole of the concave mirror next to the object plane. The objective system has axial symmetry and provides good color correction in the axial and lateral directions. However, the reflecting surfaces of these concave mirrors are interrupted by their holes, which makes the system pupil unclear.

  The EP 1 069 448 B1 patent discloses another catadioptric projection objective having two concave mirrors facing each other. These concave mirrors are part of a first catadioptric objective part that images an object on an intermediate image located in the vicinity of the concave mirror. This is only an intermediate image formed on the image plane by the second purely refractive objective part. The object is positioned outside the intermirror space defined by these mirrors facing each other, similar to the image of the catadioptric imaging system. A similar system having two concave mirrors, a common straight optical axis, and one intermediate image formed by a catadioptric imaging system and positioned near one of these concave mirrors is disclosed in No. 208551 and US patent application US 2002/00241 A1.

  European patent application EP 1 336 887 (corresponding to US 2004/0130806 A1) shows a catadioptric projection objective with one common straight optical axis, in that order, to generate an intermediate image. One catadioptric objective part, a second catadioptric objective part for generating a second intermediate image from the first intermediate image, and a refractive third objective part forming an image from the second intermediate image . Each catadioptric system has two concave mirrors facing each other. The intermediate image is outside the intermirror space defined by the concave mirror. The concave mirror is positioned optically closer to the pupil plane closer to the pupil plane than the intermediate image of the projection objective.

  International patent application WO 2004/107011 A1 discloses catadioptric projection objectives, which have one common straight optical axis and two or more intermediate images, where NA = 1, Suitable for immersion lithography with numerical apertures up to 2. At least one concave mirror is positioned optically closer to the pupil plane closer to that of the pupil plane than to the intermediate image of the projection objective.

  T. T. et al. Matsuyama, T .; Ishiyama, Y. et al. "Nikon Projection Lens Update" by Ohmura, SPIE 5377.65 (2004) Proceedings, Optical Microlithography XVII, B. W. Smith provides an example of a catadioptric projection lens configuration, which is a combination of a conventional refractive optical DUV system and a six-mirror EUV reflective optical system inserted between the DUV lens groups. The first intermediate image is formed behind the third mirror of the reflective optical (pure reflection) group upstream of the convex mirror. A second intermediate image is formed by a purely refractive (reflecting optics) second objective part. The third objective part is purely refractive and characterizes a negative refractive power at the waist of the smallest beam diameter in the third objective part for Petzval sum correction.

  Japanese Patent Application Laid-Open No. 2003-114387 and International Patent Application WO 01/55767 A describe a common straight optical axis, a first catadioptric projection objective that forms a first intermediate image, and an intermediate image as a system. A catadioptric objective part having a second catadioptric objective part for imaging on the image plane. Concave and convex mirrors are used in combination.

  D. Dejager, “Camera viewfinder using tilted concave mirror upright elements”, SPIE. Vol. 237 (1980) p. No. 292-298 discloses a camera viewfinder consisting of two concave mirrors as elements of a 1: 1 telescope upright relay system. The system is configured to image an infinite object into a real image, which is upright and can be viewed through an eyepiece. The separate optical axes of the upstream and downstream refractive parts of the reflective optical relay system are shifted in parallel to each other. In order to construct a system with concave mirrors facing each other, the mirrors must be tilted. The authors conclude that this type of physically feasible system has low image quality.

  International Patent Applications WO 92/05462 and WO 94/06047 and OSA / SPIE Proceedings (1994) p. 389 et seq. “Innovative Wide Field Binocular Configuration” is specifically configured as an inline system with a single unfolded optical axis. Binoculars and other inspection devices are disclosed. Some embodiments include a first concave mirror having an object-side mirror surface disposed on one side of the optical axis, and a second concave mirror having a mirror surface facing the first mirror and disposed on the opposite side of the optical axis. The surface curvature of these concave mirrors defines an intermirror space. The front refractive group forms a first intermediate image near the first mirror, and the second intermediate image is formed outside the space formed by the two opposing mirrors. A narrow field of view in the horizontal direction, which is larger than the vertical direction, is arranged offset from the optical axis of the optical axis. The object side refraction group has a collimated input and the image side refraction group has a collimated output, forming a non-telecentric entrance and exit pupil. The pupil shape is semicircular, unlike the pupil plane of a lithographic projection lens, which must be circular and centered on the optical axis.

  PCT application WO 01/044682 A1 discloses a catadioptric UV imaging system for wafer inspection with one concave mirror configured as a Mangin mirror.

  Catadioptric projection objective (so-called RCR system) having a single concave mirror and having a catadioptric imaging subsystem disposed between the entrance and exit refractive imaging subsystems Is disclosed, for example, in US application serial number 60 / 573,533, filed May 17, 2004 by the applicant. Other examples of R-C-R systems are shown in US 2003/0011755, WO 03/036361 or US 2002/0197946.

  By the applicant on Jan. 14, 2005 (US provisional application 60 / 536,248 filed Jan. 14, 2004; U.S. 60 / 587,504 filed Jul. 14, 2004; Based on 60 / 617,674 filed October 13, 2004; 60 / 591,775 filed July 27, 2004: and 60 / 612,823 filed September 24, 2004. The US patent application for “catadioptric projection objectives” filed is catadioptric suitable for immersion lithography with a very high NA, NA> 1 and maximum NA = 1,2. A projection objective is disclosed. The projection objective includes a first objective part for forming a pattern provided on the object plane into a first intermediate image, and a second objective part for forming the first intermediate image into a second intermediate image. And a third objective part for directly forming the second intermediate image on the image plane. The second objective part has a first concave mirror having a first continuous mirror surface, a second concave mirror having a second continuous mirror surface, and a concave mirror surface that faces each other and defines an inter-mirror space. All concave mirrors are positioned optically away from the pupil plane. The system has the potential for very high numerical aperture with moderate lens consumption. The entire disclosure of this document and its priority documents is included in this application by reference.

US 6,600,608 B1 EP 1 069 448 B1 JP 2002-208551 A US 2002/00241 A1 EP 1 336 887 US 2004/0130806 A1 WO 2004/107011 A1 JP 2003-114387 A WO 01/55767 A WO 92/05462 WO 94/06047 WO 01/044682 A1 US application serial number 60 / 573,533 US 2003/0011755 WO 03/036361 US 2002/0197946 US Provisional Application 60 / 536,248 U. S. 60 / 587,504 U. S. 60/617, 674 U. S. 60 / 591,775 U. S. 60 / 612,823

T. T. et al. Matsuyama, T .; Ishiyama, Y. et al. "Nikon Projection Lens Update" by Ohmura, SPIE 5377.65 (2004) Preliminary Proceedings, Optical Microlithography XVII, B. W. Provided by Smith D. Dejager, “Camera viewfinder using tilted concave mirror upright elements”, SPIE. Vol. 237 (1980) p. 292-298 OSA / SPIE Proceedings (1994), page 389ff “Innovative Wide Field Binoculars”

  One of the objects of the present invention is suitable for use in the vacuum ultraviolet (VUV) region, which has the possibility of very high image-side numerical aperture that can reach values that allow immersion lithography with numerical aperture NA> 1 Is to provide a catadioptric projection objective. Another object is to provide a catadioptric projection objective that can be constructed with a relatively small amount of optical elements. A further object is to provide a compact high aperture catadioptric projection objective with reasonable dimensions.

As a solution to these and other objects, the present invention, according to one clear description, describes a catadidy for imaging a pattern provided on the object plane of the projection objective onto the image plane of the projection objective. An optical projection objective,
A first refractive objective part for imaging a pattern provided on the object plane into a first intermediate image;
A second objective part having at least one concave mirror for imaging the first intermediate image into a second intermediate image;
Having a third refractive objective part for imaging the second intermediate image on the image plane;
The projection objective has a maximum lens diameter D _max , a maximum image field height Y ′ and an image-side numerical aperture NA, and COMP1 = D _max / (Y ′ · NA ² )
Next condition COMP1 <10
Provides a catadioptric projection objective.

1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention. 1 is a longitudinal sectional view of a first embodiment of a projection objective according to the invention.

In general, the size of the projection objective increases dramatically as the image-side numerical aperture NA is increased. Empirically, it has been found that the maximum lens diameter D _max tends to increase more than the first order increase in NA according to D _{max to} NA ^k (K> 1). The value k = 2 is the approximation used for this application. Furthermore, it has been found that the maximum lens diameter D _max increases in proportion to the size of the image field (represented by the image field height Y ′). A first order dependency is assumed for this application. Based on these matters, the first compactness parameter COMP1 is defined as follows.

COMP1 = D _max / (Y ′ · NA ² )
It is clear that for a given value of image field height and numerical aperture, the first compactness parameter COMP1 should be as small as possible if a compact configuration is desired.

Considering the total material consumption required to provide the projection objective, the absolute number of lenses N _L is also a problem. Typically, a system with a small number of lenses is preferred over a system with a large number of lenses. Therefore, the second compactness parameter COMP2 is defined as follows.

COMP2 = COMP1 · N _L
A small value COMP2 indicates a compact optical system.

Furthermore, the projection objective according to the invention has at least three objective parts for imaging the entrance-side field plane onto the optically conjugate exit-side field plane, the imaging objective part being connected by an intermediate image. Has been. Typically, the number of lenses and the material required to construct the projection objective increases as the number N _OP of the imaging objective part of the optical system increases. It is desirable to keep the average N _L / N _OP of the number of lenses per objective part as small as possible. Therefore, the third compactness parameter COMP3 is defined as follows.

COMP3 = COMP1 · N _L / N _OP
Projection objectives with low optical material consumption are also characterized by a small value of COMP3.

  The value COMP1 <10 means a very compact configuration. Even values for COMP1 <9.6 have been obtained in some examples. In some embodiments, the numerical aperture is greater than 1, 2 (ie, NA> 1, 2), but low compactness is obtained. Several embodiments with NA = 1,3 or NA = 1,35 are possible and ultra-high resolution immersion lithography is possible.

  In some embodiments, a low value of the second compactness parameter can be obtained. In some embodiments, COMP2 <260 and / or COMP2 <240 are obtained. Embodiments with COMP2 <220 are possible.

  Alternatively or additionally, a low value of the third compactness parameter COMP3 is possible. In some embodiments, COMP3 <80 and low values of COMP3 <70 are also possible.

  In a preferred embodiment, a first concave mirror having a first continuous mirror surface and at least one second concave mirror having a second continuous mirror surface are arranged in the second objective part, the pupil plane being the object plane and the first Formed between one intermediate image, between the first and second intermediate images, and between the second intermediate image and the image plane, all concave mirrors being optically spaced from the pupil plane. ing.

  In these embodiments, a circular pupil centered around the optical axis is provided in the centered optical system. Two or more concave mirrors of the system part are provided to contribute to the formation of the second intermediate image, and the area of use of the concave mirror deviates considerably from axially symmetric illumination. In the preferred embodiment, exactly two concave mirrors are provided, which are sufficient to obtain excellent imaging quality and a very high numerical aperture. A system with one common unfolded (straight) optical axis can be provided, which facilitates manufacture, adjustment and integration into a photolithography projection system. A planar folding mirror is not required. However, one or more planar folding mirrors can be used to obtain a more compact configuration.

All concave mirrors are placed “optically away” from the pupil plane, which means they are placed outside the optical vicinity of the pupil plane. They may be placed optically closer to the field plane than the pupil plane. A preferred position optically away from the pupil plane (ie outside the optical vicinity of the pupil plane) may be characterized by a ray height ratio H = h _C / h _M > 1, where h _C is The principal ray height, h _M is the peripheral ray height of the imaging process. The marginal ray height h _M is the marginal ray height going from the internal field point (closest to the optical axis) to the edge of the aperture stop, and the chief ray height h _C is the outermost field (farthest from the optical axis). From the point, it can be set to the height of the principal ray that travels in parallel or at a small angle with respect to the optical axis and intersects the optical axis at the pupil plane position where the aperture stop is positioned. In other words, all the concave mirrors are at positions where the principal ray height exceeds the peripheral ray height.

  A position “optically away” from the pupil plane is a position where the cross-sectional shape of the light beam deviates significantly from the circular shape found at or in the immediate vicinity of the pupil plane. As used herein, the term “light beam” refers to the bundle of all rays traveling from the object plane to the image plane. The mirror position optically away from the pupil plane may be defined as a position where the beam diameters of the light beams deviate from each other by more than 50% or 100% in directions perpendicular to each other perpendicular to the propagation direction of the light beam. . In other words, the illuminated area of the concave mirror may have a shape that deviates significantly from a circle and is similar to a high aspect ratio rectangle corresponding to the preferred field shape in a lithographic projection objective for a wafer scanner. You may have. Therefore, a small concave mirror having a compact rectangle or shape close to a rectangle in one direction that is much smaller than the other may be used. A high aperture light beam can therefore be guided through the system without vignetting at the edge of the mirror.

  Throughout the specification, the term “object part” means an imaging subsystem of a projection objective, which is a subsystem that is optically conjugate of an object in the subsystem object plane to the subsystem object plane. Can be formed on the image plane. The object imaged by the sub-system (ie the objective part) may be an object surface object or an intermediate image of the projection objective.

  When the terms “upstream” or “downstream” are used herein, these terms mean the relative position along the optical path of the light beam traveling from the object plane of the objective system to the image plane. To do. Therefore, the upstream position of the second intermediate image is an optically position between the object plane and the second intermediate image.

  The term “intermediate image” means a “paraxial intermediate image” that is generally formed by a complete optical system and located on a surface optically conjugate with the object plane. Therefore, wherever the location or position of the intermediate image is referred to, it means the position in the axial direction of this surface that is optically conjugate with the object surface.

According to another aspect of the invention, a catadioptric projection objective for imaging a pattern provided on the object plane of the projection objective on the image plane of the projection objective,
A first refractive objective part for forming a pattern provided on the object plane into a first intermediate image;
A second objective part for imaging the first intermediate image into a second intermediate image;
A third refractive objective part for imaging the second intermediate image on the image plane;
A first concave mirror having a first continuous mirror surface and a second concave mirror having a second continuous mirror surface are disposed upstream of the second intermediate image;
A pupil plane is formed between the object plane and the first intermediate image, between the first and second intermediate images, and between the second intermediate image and the image plane;
All concave mirrors are placed optically away from the pupil plane,
The first objective part has N1 _AS aspheric lenses of the first number;
The third objective part has N3 _AS aspherical lenses of the second number;
Aspheric lens ratio ASR = N1 _AS / N3 _AS is smaller than 1,
The image-side numerical aperture NA is larger than 1 and 2.

From a manufacturing point of view, it is desirable to have a lens with only a spherical lens surface, but it is clear that a certain number of aspheric lenses are required to obtain sufficient correction of image aberrations. Configuration in which the third objective part has a number aspherical lens than the first objective part is a number N _AS of aspheric lenses in the projection objective, the number of aspheric lenses to the manufacture of aspherical lenses are produced It has been found that because of the large, there is a possibility of obtaining a good correction situation without increasing beyond the allowable limit, which is a critical problem.

In some embodiments, the first objective portion has only a spherical lens such that N1 _AS = 0. An all spherical refractive objective is particularly easy to manufacture. The all spherical first objective part may be combined with a third objective part having one or more aspheric lenses, for example one or two or three or four or five lenses. The condition 1 ≦ N3 _AS ≦ 7 is preferably satisfied.

Preferably, the first objective part has no more than four aspheric lenses, ie N1 _AS ≦ 4.

The first objective part can often be composed of a small number of lenses, which can optimize the lens material consumption of the first objective part and in particular the compact dimensions in the axial direction. I understood. In some embodiments, the first objective part has no more than five lenses, and the number N1 _L of lenses in the first objective part satisfies the condition N1 _L ≦ 5. Embodiments with N1 _L = 4 are possible. Of course, N1 _L = 5 seems to be preferable in many cases.

  In some embodiments, the first objective part has only a positive lens, so that the formation of the first intermediate image can be achieved with a small maximum lens diameter in the first objective part. In other embodiments, at least one negative lens may be advantageous, particularly to improve the correction in the first objective part. Exactly one negative lens is often preferred for that purpose. The negative lens may have a concave lens surface on the image side, and may be provided between the pupil plane of the first objective system portion and the first intermediate image.

  An aspheric surface provided in an optical element such as an essentially flat surface of a lens, mirror and / or plate, a prism, etc. is used to improve both the correction status of the optical system and both the overall dimensions and material consumption. I know you get. In some embodiments, the projection objective includes a first aspheric surface and at least one “double aspheric surface” that includes a second aspheric surface proximate to the first aspheric surface, so that the propagating beam is It is possible to pass through two aspherical surfaces without passing through an intermediate spherical surface or a planar surface. Double aspheric surfaces have been found to be very powerful correction means in some embodiments.

  The double aspheric surface may take the structure of both aspheric lenses having an aspheric entrance surface and an aspheric exit surface. In some preferred embodiments, the double aspheric surface is formed by opposing adjacent aspheric surfaces of two successive lenses. Thereby, an “air space” defined by an aspheric surface on both the incident and exit sides can be obtained. The “air space” may be filled with another gas of air having a refractive index n of about 1. If a double aspherical aspheric surface is distributed over the opposing lens surfaces of a subsequent lens, the aspherical surfaces can be positioned very close together if desired. The optical distance measured along the optical axis between the first and second aspheric surfaces of the double aspheric surface is therefore the thinner of the successive lenses forming the double aspheric surface (optical axis). Less than the thickness). A composite radiation distribution of refractive power can thus be obtained at a position defined by a narrow region in the axial direction along the optical axis.

  In some embodiments, the third objective portion has at least one double aspheric surface. Preferably, this double aspheric surface is optically positioned between the second intermediate image and the pupil plane of the third objective part, thereby preferably affecting the ray angle in the region of the generally diverging beam. . A second double aspheric surface may be provided in the objective part.

  Alternatively or in combination, the first objective part may have at least one double aspheric surface. When a double aspheric surface is provided in the first objective part, it is advantageous when the double aspherical surface of the first objective part is positioned optically close to or in the pupil plane of the first objective part. It turns out that.

  As pointed out earlier, avoiding a large number of aspheric surfaces in the lens would contribute to facilitating the production of the projection objective. Under certain conditions, the corrective action of a single aspheric surface can be approximated by one or more spheres where a large angle of incidence ray occurs on that surface. In some embodiments, the first objective portion has at least one lens having a lens surface that includes an incident angle of light rays passing through the lens surface that is greater than 60 °. Preferably, the plane may be optically close to the pupil plane. The angle of incidence (incident angle) in this case is defined by the normal of the lens surface at the point where the light beam enters the lens surface and the angle surrounded by the light beam. This type of high incident angle surface may be used to reduce the number of aspheric surfaces.

  The foregoing and other characteristics may be found not only in the claims but also in the specification and drawings, and individual features may be used alone or in combination as in the embodiments of the present invention. And may be used in other fields and may represent advantageous and patentable embodiments individually.

  In the following description of the preferred embodiments of the present invention, the term “optical axis” refers to a straight line or a series of straight line portions that pass through the center of curvature of the optical element concerned. The optical axis is bent by a bending mirror (deflecting mirror) or other reflecting surface. In the example presented here, the object concerned is a mask (reticle) carrying an integrated circuit pattern or some other pattern, for example a lattice pattern. In the example represented here, the image of the object is as a substrate covered with a layer of photoresist, although other types of substrates such as liquid crystal display components or substrates for optical gratings are also possible. Projected onto the wafer.

  Where tables are provided for disclosing specifications of configurations shown in the drawings, the tables are indicated by the same numerals as the drawings. Corresponding elements in the drawings are designated with similar or identical reference numerals for ease of understanding. When a lens is indicated, L3-2 means the second lens of the third objective part (when viewed in the light propagation direction).

  FIG. 1 shows a first embodiment of a catadioptric projection lens 100 according to the present invention configured for an approximately 193 nm ultraviolet operating wavelength. This is configured to project an image of a pattern on a reticle arranged on a planar object plane OS (object plane) onto a planar image plane IS (image plane) with a reduction scale of, for example, 4: 1. Thus, exactly two real intermediate images IMI1 and IMI2 are generated. The first refractive objective part OP1 is configured to form an object plane pattern on the first intermediate image IMI1 on an enlarged scale. The second reflection optical (pure reflection) objective part OP2 forms an image of IMI2 at a magnification close to 1: 1. The third refractive objective part OP3 forms the second intermediate image IMI2 on the image plane IS with a strong reduction ratio. The second objective part OP2 includes a first concave mirror CM1 having a concave mirror surface facing the object side, and a second concave mirror CM2 having a concave mirror surface facing the image side. Both of these mirror surfaces are continuous or complete, i.e. they have no holes or holes. These opposed mirror surfaces define a catadioptric cavity, also called an inter-mirror space, surrounded by curved surfaces defined by concave mirrors. The intermediate images IMI1 and IMI2 are sufficiently separated from the mirror surface and are both located inside the catadioptric cavity.

  Each mirror surface of the concave mirror extends beyond the edge of the physical mirror surface and defines a “curved surface” or “curved surface” that is a mathematical surface having a mirror surface. The first and second concave mirrors are portions of a rotationally symmetric curved surface having a common rotational symmetry axis.

  The objective system 100 is rotationally symmetric and has one straight optical axis AX that is common to all refractive and reflective optical components. There is no folding mirror. Concave mirrors have a small diameter, allowing them to be placed close together and very close to the intermediate image between them. Both of these concave mirrors are configured and illuminated as off-axis portions of axially symmetric surfaces. The light beam passes by the edges of these concave mirrors facing the optical axis without vignetting.

When the projection objective 100 is used with a high refractive index immersion liquid such as pure water, the image-side numerical aperture NA = 1 between the exit surface of the objective system closest to the image plane IS and the image plane IS. , 2 and λ = 193 nm immersion objective system. The refractive first objective part OP1 has only a spherical lens. The concave mirrors CM1 and CM2 are both aspherical mirrors. The third objective system part OP3 has one aspherical surface (incident surface of the lens L3-9) near the position of the pupil plane P3 of the objective system part (a place where the imaging principal ray CR intersects the optical axis AX). And the second aspherical surface on the exit side of the second lens L3-12 from the final position immediately upstream of the final image-side planoconvex lens L3-13. The final lens contacts the immersion liquid during operation of the projection objective and is also referred to herein as the “immersion lens”. Although the projection objective is not fully corrected for all aberrations, it can be seen that imaging is possible with a small number of aspheric lenses (N _AS = 2) all placed in the third lens part. .

  FIG. 2 shows a second embodiment of an objective system 200 having only one aspheric lens L3-4, a first objective part OP1 and a third objective part OP3 that are all spherical. The aperture stop AS is provided in the region of the pupil plane PS3 of the third objective part. In this case, a well-corrected position is not necessary for the aperture stop of the first objective system, which consists of only four lenses, all of which are positive spherical lenses. This provides a very simple and compact configuration of the first objective part.

  When used with a high refractive index immersion liquid, such as pure water, the projection objective 200 is a liquid of λ = 193 nm having an image-side numerical aperture NA = 1, 20 between the exit surface and the image plane of the objective system. It is configured as an immersion lens. Specifications for this configuration are summarized in Table 2. The leftmost column shows the number of refraction, reflection, or other specified surface, the second column shows the radius r [mm] of the surface, and the third column shows the “thickness” of the optical element. Is the parameter d called the distance d [mm] between the surface and the next surface, the fourth column indicates the material used to manufacture the optical element, and the fifth column is used for the manufacture The refractive index of the finished material is shown. The sixth column shows the optically available effective half diameter [mm] of the optical element. The radius r = 0 in the table means a planar surface (having an infinite radius).

  In this particular embodiment, the three surfaces (surfaces 9, 10, 18) are aspheric. Table 2A shows the relevant data for these aspheric surfaces, from which the sagittal or rising height p (h) of those surface features can be calculated using the following equation as a function of height h.

p (h) = [((1 / r) h ² ) / (1 + SQRT (1− (1 + K) (1 / r) ² h ² ))] + C1 · h ⁴ + C2 · h ⁶ +. . . . ,
Here, the reciprocal of the radius (1 / r) is the curvature of the surface at the surface vertex, and h is the distance from the optical axis to that point. The sagittal or rising height p (h) thus represents the distance from the vertex of the surface measured along the z direction, ie along the optical axis, to the point. The constants K, C1, C2, etc. are shown in Table 2A.

In this embodiment, it should be noted that many aberrations are highly corrected by only a small number of lenses (N _L = 13) and one aspheric lens (L3-4) in addition to the aspherical concave mirrors CM1 and CM2. Deserve. In particular, all third-order and fifth-order aberrations are zero. The change in telecentricity is corrected over the field of view. Higher order (seventh or higher order) distortion is corrected over the field of view. The pupil aberration on the image side is corrected so that the image-side numerical apertures NA = 1 and 2 are constant over the field of view. Two real rays are corrected on the axis, and four aperture rays are corrected at the intermediate field point. Higher-order (seventh-order and higher) astigmatism is corrected at the field edge and intermediate field point. This correction situation is obtained with an objective system in which the lens diameter of the third objective part, which is 218 mm in diameter at the largest lens (operating as a focal lens group) is considerably small. The first lens L3-1 of the third objective part portion has a relatively large geometric distance with respect to the vertex of the geometrically closest mirror (first concave mirror CM1), and its axial mirror-lens distance. The MLD is 90 mm. This is about 7.5% of the axial distance between the object plane OS and the image plane IS, and this object-image distance is also referred to as “track length”. A large geometric distance MLD between the image-side first concave mirror CM1 and the first lens of the third objective part contributes to a small lens diameter in the third objective part.

  The final lens L3-9 (immersion lens) on the image side has a spherical incident surface with a short radius (50 mm), whereby a small incident angle is obtained on that surface.

  This configuration can be optimized for residual aberrations where it is clear that higher order Petzval curvature and higher order sagittal off-axis spherical aberration are dominant. Adding one lens to the first objective part and / or providing one or more additional aspheric surfaces contributes to reducing residual aberrations. An example of a further development of the configuration of FIG. 2 is shown in FIG. 3, in which an additional negative lens L1-4 configured as a meniscus lens with a concave surface on the image side is connected to its pupil plane P1 and a first intermediate Between the images is added to the first objective part. This correction makes it possible to correct the aforementioned residual aberration. This example illustrates, among other things, that the basic configuration allows a high degree of flexibility to correct imaging errors with an overall simple configuration of a small number of lenses and a small number of aspheric lenses. Yes.

  A fourth embodiment of the projection objective 400 is shown in FIG. 4 and its specifications are given in Tables 4 and 4A. Similar to the embodiment of FIGS. 2 and 3, only one aspheric lens, ie a positive meniscus lens L3-4 having an aspheric exit surface, is the objective part of the third objective part OP3 in this system. Is located in the region of the largest beam diameter upstream of the aperture stop AS that is optically close to the pupil plane P3. The first objective part OP1 is all spherical and has only one negative lens L1-4 in the sequence PPPPPNP, where “P” means a positive lens, “N” means a negative lens. From a construction point of view, a large axial distance between the image-side concave mirror CM1 and the first lens L3-1 of the third lens part OP3 is evident, this distance MLD being greater than 10% of the track length.

  FIG. 5 shows a variant of the system of FIGS. 3 and 4, in which a slight modification mainly in the first objective part OP1 is exploited for improved correction. The resulting configuration of FIG. 5 has two field points, which are corrected for both astigmatism and Petzval curvature, so that the field zone without astigmatism is in focus. .

  In a more preferred configuration of the present invention, the distortion, astigmatism, Petzval curvature and telecentricity change with respect to the field of view are the first objective part OP1 and the aspherical mirror (acting as a relay system for forming the first intermediate image IMI1). In addition, with similar configurations of several spherical lenses, all can be corrected to very high orders.

  Obviously, the two aspherical mirrors CM1 and CM2 are important for obtaining good correction with a small number of aspherical lenses. Two aspheric mirrors generally allow a configuration in which two principal ray aberrations such as distortion and telecentricity changes are corrected to very high orders. If the deformations of the two aspherical mirrors are set correctly, these two aberrations can be accurately corrected by these mirrors. One notable point is that, in addition, astigmatism and Petzval curvature can be highly corrected in the first spherical objective part OP1.

  It is clear that there are at least three features that can contribute to the positive characteristics of the type of configuration for aberration correction, alone or in combination. One aspect is that the concave mirrors may preferably be very different in terms of radius compared to other embodiments where the concave mirrors CM1, CM2 are identical or nearly identical. Furthermore, it is clear that many coma aberrations in the intermediate images IMI1 and / or IMI2 facilitate correction with a small number of aspheric lenses. It is also clear that a significantly large air space (mirror-lens distance MLD) between the apex of the image-side concave mirror CM1 and the first lens of the third objective part contributes to advantageous properties.

  It is also clear that the object side and the image side and / or the projection objective, i.e. the first objective part OP1 and the third objective part OP3, can be configured independently. In particular, the third objective system (focus lens group) can be configured for aperture aberration without paying much attention to field aberrations, and is relatively simple with respect to the configuration. Can be configured to compensate for field aberrations, which compensation could be obtained without an aspheric lens or with only a few aspheric lenses, for example with only one aspheric lens.

  The previous example shows that a configuration with a fairly simple first objective part of only four or five lenses, where all lenses may be spherical, is available. Such a fairly simple relay lens group can correct for field aberrations to very high orders. The aperture aberration is preferably corrected in the third objective part, which may have a fairly simple construction with only a few aspheric surfaces, and the number of aspheric lenses in the third objective part is Preferably, it is larger than the number of aspheric lenses in the first objective part.

FIGS. 6 and 7 show closely related embodiments 600, 700 where the number of aspheric lenses in the refractive first and third objective parts is increased compared to the previous embodiment. The specifications for objective 700 are given in Tables 7 and 7A. Improvements regarding aperture aberration are obtained. In particular, seven aspheric lenses are used, where N1 _AS = 2 and N3 _AS = 5, and the aspheric lens ratio ASR = 0.4. The configuration of FIG. 6 having two aspheric lenses L1-2 and L1-5 in the first objective part OP1 has a wavefront aberration of 5 millimeter waves over the field of view.

  One double aspheric surface DA is optically provided between the second intermediate image IMI2 and the pupil plane P3 of the objective system part in the region of the beam diameter that increases remarkably in the third objective system part OP3. The double aspherical surface is formed by the aspherical exit surface of the positive lens L3-6 and the aspherical incident surface of the positive meniscus lens L3-7 that immediately follows the double aspherical surface. The axial distance between the two aspheric surfaces is smaller than the thickness of the thinner lens L3-7 adjacent to the double aspheric surface, and the aspheric surfaces are close to each other. Thereby, a combined radiation distribution of refractive power is obtained in a specific region of the beam, thus contributing strongly to image correction.

  Figures 8 and 9 show examples 800, 900 that are very similar in construction. The specifications of the projection objective 900 are given in Tables 9 and 9A. An image-side numerical aperture NA = 1, 2 and a wavefront error of about 6 millimeters can be obtained with only six aspheric lenses, where there is one aspheric lens (an image-side concave surface placed near the pupil plane P1). Are provided in the first objective part, and the remaining five aspherical lenses are dispersed in the third objective part OP3. These aspherical lenses include a biconcave negative meniscus L3-2, a double aspherical surface DA formed by opposing the aspherical surfaces of lens pairs L3-5 and L3-6, and a biconvex close to the aperture stop AP. A positive lens L3-9 and a positive lens L3-10 between the aperture stop and the image plane IS are included. As shown in the light distribution of FIG. 8, the front pupil of the relay part formed by the first objective part OP1 is corrected fairly well. A focal region (low coma intermediate image) with little or no coma is evident for both intermediate images IMI1 and IMI2. The low coma aberration intermediate image IMI2 has a significant waist W between the thick positive lens L3-1 on the incident side and the region of the maximum beam diameter slightly upstream of the first lens L3-1 and the aperture stop AS (ie, the beam). It is refocused on the image plane by a refractive objective part OP3 having a constricted area of diameter). The double aspheric surface DA is provided in the diverging beam between the waist and the aperture stop AS.

The projection objective 1000 of FIG. 10 may be considered in some respects as a variation of the embodiment shown in FIGS. The specifications are given in Tables 10 and 10A. The third objective part OP3 has five aspheric lenses and has a relatively similar configuration and arrangement as compared to the examples 800 and 900, but the first objective part OP1 has no aspheric lenses, N _AS = 5. It should be noted that an all-spherical doublet composed of a negative meniscus lens L1-7 having an object-side concave surface immediately downstream of the positive lens and an image-side biconvex positive lens L1-6 is positioned near the pupil plane P1. ing. A high incident angle of light rays emerging from the positive lens L1-6 and incident on the subsequent negative lens L1-7 is given to this region, and the incident angle includes an angle greater than 60 °. The optical effect of the aspherical lens L1-5 of Examples 800 and 900 having difficulty in producing an aspherical surface with a sharp tilt change uses a fairly high incident angle (range 60 ° -65 °). Thus, it is clear that a similar region near the pupil plane of the first objective part OP1 can be at least partially simulated. Because the spherical surfaces of the doublet lenses L1-6, L1-7 are easier to manufacture, sometimes the ease of manufacturing can be improved by replacing the aspherical lens with one or more lens surfaces that produce a high incidence angle. .

  When considering replacing an aspherical surface with multiple spherical surfaces, it is clear that the most important thing is the spherical curvature of the base. Spherical doublets could therefore replace aspheric surfaces-even very high order aspheric surfaces.

  The correction status of the variation of Example 1000 is between 4, 5 and 6 millimeter waves across its field of view. This suggests that this correction can all be obtained with the first objective part OP1 of the sphere, and the complete configuration does not require many aspheric lenses to obtain good performance. (Here, only 5 lens aspheric surfaces).

The specifications of the catadioptric projection objective 1100 are given in Tables 11 and 11A. The example is a good example showing that a “dual aspheric surface” with two fairly strong aspheric lens surfaces that are very close to each other may be a very powerful component. Here, the double aspheric surface DA formed by the lenses L3-4 and L3-5 is a beam that increases in the third objective part OP3, for example, as in the embodiments shown in FIGS. In the area of diameter. In addition, the second double aspheric surface DA formed by the opposing surfaces of the lenses L1-5 and L1-6 exists optically close to the front pupil, that is, the pupil plane P1 of the first objective part OP1. To do. In this example, N _AS = 8, N1 _AS = 3 and N3 _AS = 5. The correction is a maximum lens diameter of 220 mm near the aperture stop AS in the third objective part, with NA = 1, 2 and only about 2.5 millimeter waves. This illustrates the potential for this configuration to obtain good optical performance with a small number of aspheric lenses.

FIG. 12 shows a λ = 193 nm immersion objective with NA = 1,3 (rather than NA = 1,2 in the previous example). The maximum lens diameter (which is immediately upstream of the aperture stop AS in the third objective part OP3) is 270 mm. There are exactly 8 aspheric lenses with N1 _AS = 2 and N3 _AS = 6. These are formed by lenses L3-4 and L3-5 and include one double aspheric surface DA located in the area of the diverging beam between the waist W of the third objective part and the aperture stop AS. The field radius is 66 mm. The correction is about 4 to 6 millimeters over its field of view. The configuration assumes a strictly telecentric input. From the composition of the rays in the intermediate image, it is clear that there is a large amount of coma in the intermediate image.

A further embodiment of a catadioptric immersion objective 1300 with NA = 1, 3 is shown in FIG. The specifications are given in Tables 13 and 13A. N _AS = 10, N1 _AS = 4 and N3 _AS = 6. The image-side numerical aperture NA = 1, 3 corresponds to that of the objective system 1200, but the maximum lens diameter is only 250 mm (not 270 mm in Example 1200). Still, the wavefront error is only 4 millimeters over that field of view. The configuration is also characterized by a lack of real aperture, requiring telecentric input. A large amount of coma is also present in the intermediate image.

  The catadioptric immersion objective 1400 in FIG. 14 (specifications are given in Tables 14 and 14A) is another one of the high NA catadioptric immersion objective with NA = 1,3 and a relatively small maximum lens diameter. In one example, the maximum lens diameter is only 250 mm. Four of the 11 aspherical lenses are in the first objective part OP1 and the remaining 7 aspherical lenses are dispersed in the third objective part. Compared to the previous configuration, the aspheric lenses are easier to manufacture because all aspheric lens surfaces are deformed from a sphere with less than 10 mm, and a local aspheric radius greater than 150 mm for each aspheric surface. Improved by following the requirement of having Three double aspheric surfaces are provided. One double aspheric surface DA formed by the lenses L1-6 and L1-7 positioned on the pupil plane P1 in the first objective part OP1 is configured to have a fairly high incident angle, It is clear that this has the same effect as a short local aspheric radius (which is more difficult to manufacture). The third objective part OP3 has two double aspheric surfaces DA, that is, one two formed by the opposing surfaces of the negative lenses L3-1 and L3-2 in the region of the smallest lens diameter in the third objective part. The lenses L3-3 and L3-4 are opposed to each other in a region where the maximum aspherical diameter increases between the aperture stop AS positioned between the region of maximum beam diameter and the image plane IS and the second intermediate image IMI2. There is a double aspheric surface formed by a surface followed by one double aspheric surface. A large amount of coma is present in the intermediate images IMI1, IMI2 as in the embodiment of FIGS.

  The catadioptric immersion objective 1500 (specifications Tables 15 and 15A) in FIG. 15 is a variation of the embodiment shown in FIG. The dimensions and types of lenses present in the objective system 1500 are essentially the same. There is a difference in that an additional biconvex positive lens L3-1 is introduced immediately after the second intermediate image IMI2, thereby providing a positive refractive power on the entrance side of the third objective part OP3. Good performance at NA = 1,3 is obtained.

  The basic configuration is NA = 1,3 and has the possibility of higher image side numerical aperture. The catadioptric immersion objective 1600 in FIG. 16 (specifications are Tables 16 and 16A) is based on the configuration of FIG. 15, but optimized to obtain NA = 1,35. As in this example, there are 10 lenses (including 4 aspheric lenses) in the first objective part and 12 lenses (including 7 aspheric lenses) in the third objective part. The basic types of these lenses are the same, but the lens thickness, surface radius and lens position are slightly different. As the numerical aperture increases, it is advantageous to place the aperture stop AS in the region of maximum beam diameter (biconvex lens L3-8) and the region of strong focus of the beam between the image plane IS in the third objective system OP3. It is clear. Here, only three positive lenses are placed between the aperture stop and the image plane.

  The higher the desired numerical aperture, the more advantageous it is to place the second objective part OP2 closer to the image plane geometry. For convenience, the second objective part OP2, which preferably consists of only two aspheric concave mirrors CM1, CM2, will also be referred to below as “mirror group”. For the purpose of explaining this feature, a first optical axis length OAL1 is defined between the vertex of the concave mirror CM2 geometrically closest to the object plane and the object plane OS, and a third optical axis length OAL3 is defined as an image. It is defined between the vertex of the concave mirror (CM1) geometrically closest to the plane and the image plane (see FIG. 16). Based on this definition, a mirror group position parameter MG = OAL1 / OAL3 is determined, and when this value increases, the mirror group is positioned further on the image side of the projection objective. In Table 17, the values of OAL1, OAL3, and MG summarize all the examples described herein. Based on these data, it is clear that the mirror group position parameter MG> 0, 7 is desirable for the purpose of obtaining a high image-side numerical aperture. Preferably, MG ≧ 0.8. More preferably, MG ≧ 0.9.

Each projection objective described here has a high NA image side end where projection radiation emerges from the projection objective at the exit plane ES, which is a planar substrate disposed on the image plane IS. A plane is preferred to allow a uniform distance between the surface and the exit surface. The lens that forms the exit surface ES closest to the image plane is referred to herein as the “final lens” LL. Preferably, the final lens is a plano-convex positive lens having a curved entrance surface ENS and a planar exit surface, which is curved spherically in most embodiments. For the purpose of obtaining a high NA, it has been found advantageous to configure the final lens so that the large refractive power provided by the curved entrance surface ENS is located as close as possible to the image plane. Furthermore, it is clear that a large curvature of the entrance surface ENS of the final lens LL, that is, a small radius of curvature is desirable. T _LL is the thickness of the final lens on the optical axis (that is, the axial distance between the entrance surface ENS and the exit surface ES measured along the optical axis), and R _LL is the object-side apex radius of the final lens ( That is, the radius of the incident surface ENS), and DIA _LL is the optical free diameter of the incident surface of the final lens, the parameters LL1 = T _LL / R _LL and LL2 = DIA _LL / R _LL are preferably within a certain range. Should enter. In particular, it has been found to be advantageous if the conditions 0, 46 ≦ LL1 ≦ 0,69 apply to LL1. Since the parameter LL1 is 1 for a hemispherical lens whose center of curvature of the entrance surface coincides with the exit surface, the condition regarding LL1 is that the center of curvature of the curved entrance surface exists outside the final lens, particularly beyond the image plane. This indicates that a non-hemispheric final lens is preferred.

Alternately or additionally, the condition 1,15 ≦ LL2 ≦ 1,67 should preferably apply to LL2. The respective values for LL1 and LL2 are shown in Table 18. If at least one of the above conditions is true, a large positive refractive power provided by the curved entrance surface of the final lens is provided near the image surface, so that a large image-side numerical aperture NA, in particular NA> 1,1. Alternatively, NA> 1,2 and NA = 1,3 or NA = 1,35 can be obtained.

  Regarding the correction situation of the intermediate images IMI1, IMI2, in some embodiments both intermediate images are essentially focused (ie many aberrations are highly corrected) and in other embodiments considerable aberrations, in particular It was found that coma aberration occurs (compare FIGS. 11-16). A considerable coma aberration of the second intermediate image IMI2 may be advantageous with respect to the overall correction of the objective system. Since the catadioptric second objective part consisting of the concave mirrors CM1 and CM2 is effective for imaging the first intermediate image in a manner essentially symmetrical to the second intermediate image, it is usually the second catadioptric second object. Only a few frames are introduced by the objective part. Therefore, the correction statuses for the frames of both intermediate images tend to be similar. In some embodiments, it is apparent that a significant amount of coma at least for the second intermediate image contributes significantly to the overall correction. The following explanation is noteworthy in this regard.

  Correction of the sine condition of the entire objective system is difficult, especially for an objective system with a very high image side NA. Correction of the sine condition can be facilitated by the coma of the intermediate image. When considering imaging from a high NA image plane to a low NA object plane (ie in the opposite direction compared to the intended use of the projection objective in lithography), the third objective part (incident by radiation) is Provide an intermediate image having a certain correction situation. Assuming that the spherical aberration of this image is corrected, if the sine condition of this image is corrected, the intermediate image is essentially free of coma. In contrast, if the sine condition is not corrected, the intermediate image has a significant amount of coma. If the intermediate image has a significant amount of coma, correction of the sine condition in the third objective part is facilitated.

  Now consider imaging in the intended direction (towards the high NA end) on the image plane of the second intermediate image. If the second intermediate image has a good correction situation, in particular without coma, the overall correction of the sine condition is influenced by the third objective part that forms the second intermediate image on the image plane. You will have to take it. In contrast, if a certain amount of coma is present in the second intermediate image, the third objective part can be configured in a more relaxed manner. This is because the correction of the sine condition is performed at least by the objective part optically upstream of the third objective part, that is, the refractive relay system OP1 forming the first intermediate image and the catadioptric second objective part OP2. It can be partially affected. Therefore, the configuration in which the correction of the coma is distributed between the first refractive objective part OP1 and the third objective part OP3 has an objective in which each of the refractive objective parts is corrected independently of the coma aberration. Obviously, it can be advantageous when compared to the system.

  As previously mentioned, the present invention allows the construction of a high NA projection objective suitable for immersion lithography with compact dimensions and NA> 1.

Table 19 shows the values required to calculate the compactness parameters COMP1, COMP2, COMP3 and a summary of the values of these parameters for each system in the specification table (Table of Tables). Numbers (corresponding to the same numbered figures) are given in column 1 of Table 19). Further, the values of N1 _AS , N3 _AS , and ASR are shown.

  Table 19 shows that preferred embodiments in accordance with the present invention generally comply with at least one condition given above, indicating that the design rules described in this specification provide a compact configuration with reasonable material consumption. Show. In addition, specific values characterizing the aspheric lens number and distribution are shown.

  In the following, further characteristic configurations of the projection objective according to the invention are summarized, where one or more features may be present in embodiments of the invention. The parameters summarized in Tables 20 and 21 are used to clarify these characteristics.

  In some embodiments, the chief ray of the imaging process takes a characteristic path. For illustrative purposes, three consecutive runs from the outermost field point (farthest from the optical axis AX) proceeding essentially parallel to the optical axis, each in one of the imaging objective parts OP1, OP2, OP3 The principal ray CR that intersects the optical axis at the pupil plane positions P1, P2, and P3 is drawn with a thick line in FIG. 16 for easy understanding. The angle enclosed between the optical axis AX and the principal ray CR at each position along the principal ray is hereinafter referred to as “principal ray angle” CRA. The chief ray CR diverges at the position of the first intermediate image IMI1 (the chief ray height increases in the light propagation direction) and converges at the position of the second intermediate image IMI2. It is clear that a strongly converging chief ray is advantageous for obtaining a high image side NA and sufficient correction.

In the region between the two concave mirrors CM1, CM2, the chief ray traverses the optical axis with a high chief ray angle CRA (M), which angle is preferably between 58 ° and 75 °, preferably 60 ° and 70 °. It is between °. (See Table 20)
Imaging objective part OP1, OP2, with respect to the magnification provided by OP3, if is preferably magnification beta ₃ of the third objective part OP3 is imaged on the image plane with a high reduction ratio of the second intermediate image IMI2 0, 11 It is clear that it is advantageous if it is in the range ≦ β ₃ ≦ 0,17. For the purpose of obtaining a desired overall reduction ratio (eg 1: 4 or 1: 5), the second objective part OP2 can contribute to the overall reduction by having an enlargement ratio β ₂ <1. Preferably, the mirror group forming the second objective part OP2 may be configured to have a gentle reduction effect characterized by 0,85 ≦ β ₂ ≦ 1. If the second objective part contributes to some extent to the overall reduction, the third objective part responsible for the main part of the reduction can be configured in a more relaxed manner.

  The refractive power provided by the first two or three lenses on the incident side of the third objective part OP3 immediately after the second concave mirror CM2 (characterized by the focal length f) By configuring the force to be negative, it is possible to contribute to good performance. 2, 4, 14, 15, 16 the incident group is formed by the first two lenses of the third objective part, giving an incident group focal length f3 (L1 ... 2). 7, 9, 10, 11, 12, and 13, the incident group is formed by three successive lenses, thereby giving the focal length f3 (L1... 3) of the incident group. Values are given in Table 20.

On the other hand, there are not many negative lenses that should exist in the third objective part following the second concave mirror, and the number N3 _{NL of} negative lenses is 3 or less in all the examples (parameter K7a = YES in Table 21). 2, 4, 14, 15 and 16 is smaller than 3 (parameter K7 = YES in Table 21).

Furthermore, it is clear that it is advantageous if the optical free diameter DIA ₃₁ of the first lens L3-1 of the third objective part OP3 is much smaller than the diameter DIA _{AS of the} aperture stop. Preferably, the diameter ratio DR = DIA ₃₁ / DIA _AS should be less than 0.9. More preferably, the upper value of 0,8, more preferably the upper value of 0,7 should not be exceeded. The value of the diameter ratio DR is given in Table 21.

  Furthermore, if all of the lenses after the second concave mirror (ie the lenses of the third objective part) exceed 50%, the optical freedom is smaller than the diameter of the second intermediate image IMI2 following the second concave mirror CM2. If so, this condition is satisfied for all examples, as indicated by parameter K10 in Table 21.

  Also, all the lenses of the first refractive projection objective part OP1 should preferably be smaller than the paraxial dimension of the first intermediate image. If this condition is satisfied, the parameter K9 in Table 20 is satisfied.

  For the purpose of providing strong positive power to obtain strong beam convergence at the high NA image edge, at least one of the 8 and 9 consecutive lenses upstream of the image plane has positive power Is preferable. This is illustrated by parameter K11 in Table 21, which is “YES” if the condition is met and “NO” if the condition is not met.

  In this context, it may be advantageous to obtain a high NA if the position of the aperture stop AS is in the region of the convergent beam between the position of the maximum beam diameter in the third objective part OP3 and the image plane. It is worth mentioning that it is obvious. This characteristic is illustrated by the ratio AS-IS / TT shown in Table 20, where AS-IS is the geometric distance between the position of the aperture stop AS and the image plane IS, and TT is the objective " "Track length", ie the geometric distance between the object plane and the image plane. The ratio AS-IS / TT can range between 0,09 and 0,18 (see Table 20).

  This feature is particularly noticeable in the embodiments of FIGS.

  Further characteristic configurations are evident from the coma beam path. Here, “coma beam” refers to a beam that exits from the object field point farthest from the optical axis and passes through the aperture stop at the edge of the aperture. The coma beam therefore contributes to determining which lens diameter must be used. The angle surrounded by the coma beam and the optical axis is hereinafter referred to as “coma beam angle CBA”. The angle of the beam after refraction at the final lens of the first objective part (upstream of the first intermediate image IMI1) is called CBA1 and is just after refraction at the first lens of the image side third objective part OP3. The angle of the upstream coma beam is called CBA3. These angle values are given in Table 21. It is clear that values of less than 5 ° for both coma beam angles can be advantageous (Table 21).

As described above, the chief ray intersects the optical axis at the pupil planes P1, P2, P3 of the connected objective parts OP1, OP2, OP3. Since the pupil planes in the first and third objective parts are available for providing an aperture stop, these positions are also called aperture positions. The beam diameter DIA _AS at the aperture stop and the beam diameter DIA _{P1 at} the pupil plane P1 of the first objective system part conjugate to the position of the aperture stop should be within a certain range. The ratio DIA _AS / DIA _P1 should be greater than 1. Preferably, the condition DIA _AS / DIA _P1 > 2 should be met. (See Table 21)
It should be understood that all of the above systems may be complete systems for forming a real image (eg, on a wafer) from a real object. However, the system may be used as a partial system of a larger system. For example, the “object” for the above system may be an image formed by an imaging system (relay system) upstream of the object plane. Similarly, an image formed by the above system may be used as an object for a system (relay system) downstream of the image plane.

  The description of the preferred embodiment above is given as an example. From the disclosure given, those skilled in the art will not only understand the present invention and its attendant advantages, but will also find various changes and modifications to the disclosed structures and methods. Therefore, all changes and modifications and equivalents thereof are covered within the spirit and scope of the present invention as defined by the appended claims.

  The contents of all the claims are hereby incorporated by reference.

[Table 2]

[Table 2A]

[Table 4]

[Table 4A]

[Table 7]

[Table 7A]

[Table 9]

[Table 9A]

[Table 10]

[Table 10A]

[Table 11]

[Table 11A]

[Table 12]

[Table 12A]

[Table 13]

[Table 13A]

[Table 14]

[Table 14A]

[Table 15]

[Table 15A]

[Table 16]

[Table 16A]

[Table 17]

[Table 18]

[Table 19]

[Table 20]

[Table 21]

Claims (7)

A catadioptric projection objective for imaging a pattern provided on the object plane of the projection objective on the image plane of the projection objective,
A first refractive objective part for forming a pattern provided on the object plane into a first intermediate image;
A second objective part for imaging the first intermediate image into a second intermediate image;
A third refractive objective part for imaging the second intermediate image on the image plane;
The second objective part has a first concave mirror having a first continuous mirror surface and a second concave mirror having a second continuous mirror surface;
A pupil plane is formed between the object plane and the first intermediate image, between the first and second intermediate images, and between the second intermediate image and the image plane;
All concave mirrors are arranged at positions where the principal ray height traveling from the outermost field point farthest from the optical axis exceeds the peripheral ray height traveling from the inner field point closest to the optical axis ,
The first objective part has N1 _AS aspherical lenses of a first number;
The third objective part has N3 _AS aspherical lenses of a second number;
Aspheric lens ratio ASR = N1 _AS / N3 _AS is smaller than 1,
Catadioptric projection objective with an image-side numerical aperture NA greater than 1.2. A catadioptric projection objective for imaging a pattern provided on the object plane of the projection objective on the image plane of the projection objective,
A first refractive objective part for forming a pattern provided on the object plane into a first intermediate image;
A second objective part for imaging the first intermediate image into a second intermediate image;
A third refractive objective part for imaging the second intermediate image on the image plane;
The second objective part has a first concave mirror having a first continuous mirror surface and a second concave mirror having a second continuous mirror surface;
A pupil plane is formed between the object plane and the first intermediate image, between the first and second intermediate images, and between the second intermediate image and the image plane;
All concave mirrors are arranged at positions where the principal ray height traveling from the outermost field point farthest from the optical axis exceeds the peripheral ray height traveling from the inner field point closest to the optical axis ,
The first objective part has N1 _AS aspherical lenses of a first number;
The third objective portion has a second number of N3 _AS aspheric lenses;
Catadioptric projection objective with aspheric lens ratio ASR = N1 _AS / N3 _AS less than 0.5. Projection objective according to claim 2 , wherein the condition NA ≧ 1.2 applies to the image-side numerical aperture NA. A catadioptric projection objective for imaging a pattern provided on the object plane of the projection objective on the image plane of the projection objective,
A first refractive objective part for forming a pattern provided on the object plane into a first intermediate image;
A second objective part for imaging the first intermediate image into a second intermediate image;
A third refractive objective part for imaging the second intermediate image on the image plane;
The second objective part has a first concave mirror having a first continuous mirror surface and a second concave mirror having a second continuous mirror surface;
A pupil plane is formed between the object plane and the first intermediate image, between the first and second intermediate images, and between the second intermediate image and the image plane;
All concave mirrors are arranged at positions where the principal ray height traveling from the outermost field point farthest from the optical axis exceeds the peripheral ray height traveling from the inner field point closest to the optical axis ,
Catadioptric projection objective, wherein the first objective part has a first number N1 _AS aspheric lenses and the condition N1 _AS <3 applies. 5. Projection objective according to claim 4 , wherein the third objective part has a second number of N3 _AS aspheric lenses and the aspheric lens ratio ASR = N1 _AS / N3 _AS is less than one. . A projection exposure system for use in microlithography having an illumination system and a catadioptric projection objective, wherein the projection objective is configured as claimed in any one of claims 1-5. Projection exposure system. A method for manufacturing a semiconductor device or other type of microdevice,
Preparing a mask having a predetermined pattern;
Illuminating the mask with ultraviolet light having a predetermined wavelength;
Projecting an image of the pattern onto a photosensitive substrate disposed in the vicinity of the image plane of the projection objective using the catadioptric projection objective according to any one of claims 1 to 5 ; Having a method.

Images