DE102013204391B3 - Projection lens for imaging projection lens pattern from object plane into image plane, has field point in field plane of outgoing beam illuminating manipulator surface with sub-aperture, and manipulation system comprising manipulator - Google Patents

Abstract

The lens (PO) has multiple optical elements with optical surfaces arranged in a projection optical path between an object plane (OS) and an image plane (IS). A wave front manipulation system comprises a manipulator (MAN1) with a manipulator surface (MS1) arranged in the projection optical path, and a positioning device (DR1) is adapted for reversible change of surface shape and/or index of refraction distribution of the manipulator surface. A field point in a field plane of an outgoing beam illuminates the manipulator surface with a sub-aperture. Independent claims are also included for the following: (1) a projection exposure method (2) a projection exposure system.

Description

BACKGROUND

Technical area

The invention relates to a projection objective for imaging a pattern arranged in an object plane of the projection objective into an image plane of the projection objective by means of electromagnetic radiation having a working wavelength λ <260 nm and a projection exposure method which can be carried out with the aid of the projection objective.

State of the art

For the production of semiconductor devices and other fine-structured components, such. As photolithography masks, nowadays predominantly microlithographic projection exposure methods are used. In this case, masks (reticles) or other pattern generating means are used, which carry or form the pattern of a structure to be imaged, for. B. a line pattern of a layer (layer) of a semiconductor device. The pattern is positioned in a projection exposure apparatus between a lighting system and a projection lens in the region of the object plane of the projection lens and illuminated with an illumination radiation provided by the illumination system. The radiation changed by the pattern passes through the projection lens as projection radiation, which images the pattern onto the substrate to be exposed on a reduced scale. The surface of the substrate is arranged in the image plane of the projection lens optically conjugate to the object plane. The substrate is usually coated with a radiation-sensitive layer (resist, photoresist).

The WO 2011/120821 A1 describes a method for adapting a projection exposure apparatus for microlithography to a mask having structures of different pitches and / or different feature widths in different structural directions, wherein the wavefront aberrations induced by the mask are reduced by a manipulator of the projection exposure apparatus for microlithography.

One of the goals in the development of projection exposure equipment is to lithographically produce structures of increasingly smaller dimensions on the substrate. Smaller structures lead z. B. in semiconductor devices to higher integration densities, which generally has a favorable effect on the performance of the microstructured components produced.

The size of the structures that can be generated largely depends on the resolution capability of the projection objective used and can be increased on the one hand by reducing the wavelength of the projection radiation used for the projection and on the other hand by increasing the image-side numerical aperture NA of the projection objective used in the process.

High-resolution projection lenses today operate at wavelengths less than 260 nm in the deep ultraviolet (DUV) or extreme ultraviolet (EUV) regions.

In order to ensure a sufficient correction of aberrations (eg chromatic aberrations, field curvature) at wavelengths from the deep ultraviolet range (DUV), catadioptric projection objectives are used which include both transparent refractive optical elements with refractive power (lenses) and reflective elements with refractive power, ie curved mirrors. Typically, at least one concave mirror is included. Nowadays, with immersion lithography at NA = 1.35 and λ = 193 nm resolution can be achieved here, enabling a projection of 40 nm structures.

Integrated circuits are produced by a sequence of photolithographic structuring steps (exposures) and subsequent process steps, such as etching and doping, of the substrate. The individual exposures are usually carried out with different masks or different patterns. For the finished circuit to show the desired function, it is necessary that the individual photolithographic exposure steps are matched as well as possible, so that the fabricated structures, such as contacts, lines and the components of diodes, transistors and other electrically functional units, as close as possible approach the ideal of planned circuit layouts.

Manufacturing errors can u. a. then arise when the structures generated in successive exposure steps are not sufficiently accurate to each other, so if the overlap accuracy is not sufficient. The overlay accuracy of structures from different production steps of a photolithographic process is commonly referred to by the term "overlay". This term denotes z. For example, the registration accuracy of two consecutive lithographic planes. The overlay is an important parameter in the fabrication of integrated circuits because alignment errors of any kind can cause manufacturing defects such as short circuits or missing connections, thus limiting the operation of the circuit.

Even with multiple exposure methods, high demands are placed on the registration accuracy of successive exposures. For example, in the double patterning method (or double-exposure method), a substrate, for example a semiconductor wafer, is exposed twice in succession and the photoresist is subsequently processed. In a first exposure process z. For example, a normal structure with a suitable feature width is projected. For a second exposure process, a second mask is used which has a different mask structure. For example, periodic structures of the second mask may be shifted by half a period with respect to periodic structures of the first mask. In the general case, especially for more complex structures, the differences between the layouts of the two masks can be large. Double patterning can be used to reduce the period of periodic structures on the substrate. This can only succeed if the coverage accuracy of the successive exposures is sufficiently good, ie if the overlay errors do not exceed a critical value.

Insufficient overlay can thus significantly reduce the yield of good parts during production, thereby increasing the manufacturing costs per good part.

TASK AND SOLUTION

The invention has for its object to provide a projection lens and a projection exposure method for microlithography, which allow to perform different photolithographic processes with low overlay errors.

This object is achieved by a projection objective having the features of claim 1 and by a projection exposure method having the features of claim 13.

Advantageous developments are specified in the dependent claims. The wording of all claims is incorporated herein by reference.

The projection objective has a wavefront manipulation system for dynamically influencing the wavefront of the projection radiation running from the object plane to the image plane of the projection objective. The effect of the arranged in the projection beam path components of the wavefront manipulation system can be variably adjusted in response to control signals of a control device, whereby the wavefront of the projection radiation can be selectively changed. The optical effect of the wavefront manipulation system can, for. B. at certain, pre-defined events or situation-dependent before an exposure or even during an exposure to be changed.

The wavefront manipulation system has a first manipulator which has a first manipulator surface arranged in the projection beam path. The first manipulator includes a first adjusting device, which makes it possible to reversibly change the surface shape and / or the refractive index distribution of the first manipulator surface. As a result, the wavefront of the projection radiation, which is influenced by the first manipulator surface, can be selectively changed dynamically. This change in the optical effect is possible without exchanging the first manipulator for another manipulator.

In this case, a manipulator surface is understood to mean a plane or curved surface which (i) is arranged in the projection beam path, and (ii) in which a change in its surface shape and / or orientation with respect to the projection radiation leads to a change in the wavefront of the projection radiation. For example, each curved surface of a lens displaceable relative to the other optical components of a projection lens is a manipulator surface. Other examples are mechanically or thermally deformable surfaces of lenses or mirrors.

In a local, thermal manipulation of a lens i. d. R. also vary the refractive index of the lens locally spatially. Can one - z. Due to the thickness of the lens - assume that this variation has no component in the direction of the projection radiation, i. H. the refractive index varies only orthogonally to the direction of the projection radiation, so it makes sense to also consider a local variation of the refractive index of a lens as an effect that occurs on a manipulator surface. This applies, for example, to thin plane plates.

In contrast, listed above known displacing, deforming or thermal manipulators, for example, by global displacement of an optical element such. B. tilting, decentration and / or axis-parallel displacement, or act by global deformation on the wavefront, the first manipulator according to the claimed invention is configured such that over an optically used area of the first manipulator surface within an effective diameter D _{FP of} this first manipulator surface more Maxima and several minima of an optical path length change of the projection radiation can be generated. If N _{MAX is} the number of maxima and N _{MIN is} the number of minima of the optical path length change in the considered Direction, the effect of the first manipulator in the direction of the effective diameter can be described by means of a characteristic period P _CHAR = D _FP / ((N _MAX + N _MIN ) / 2). The multiple change of the optical path length change caused by the first manipulator over the influenced cross section of the projection radiation does not have to be strictly periodic, so that, for example, the absolute values of the maxima and / or the minima of the optical path length change and / or their lateral distances over the cross section of the influenced Projection radiation may vary. Strict periodic optical path length change, which can be described for example by a sine function, are also possible.

The first manipulator surface is arranged "in optical proximity" of a nearest field plane of the projection objective. Among other things, this "near-field arrangement" means that the first manipulator surface is arranged substantially closer to the closest field plane than to a pupil plane of the projection objective. Each beam emanating from a field point of the field plane illuminates at the first manipulator surface a subaperture with a subaperture diameter SAD, which is substantially smaller than the maximum diameter D _{FP of} the optically used region of the first manipulator surface, so that the condition SAD / D _FP <0.2 applies , In particular, even the condition SAD / D _FP <0.1 apply.

Here, the subaperture diameter SAD is to be understood as meaning the diameter of the bundle of projection light emanating from a single field point. The quotient SAD / D _FP is generally independent of the height of the considered field point.

Due to the arrangement of the first manipulator surface in optical proximity to the nearest field plane and the possibility of varying the optical effect over the beam cross section, the wavefront manipulation system is capable of selectively setting or changing the distortion of the projection objective in the image field. This means u. a., that different values of the distortion can be set specifically for different field points. A wavefront manipulation system which is able to selectively set a field-dependent distortion in dependence on control signals makes it possible to introduce a specific field-dependent distortion correction or directory change with each exposure process. As a result, it is also possible in particular to match the field-dependent distortion in a second exposure to a structure produced during a preceding first exposure in such a way that the structures produced in successive exposure steps lie on one another with high coverage accuracy. The overlay accuracy can be better by activating the manipulator than in the absence of an activated manipulator. This allows overlay errors to be kept to a tolerable level.

This influence of the field-dependent distortion should be achieved without simultaneously generating other aberrations to a disturbing extent. Particular emphasis is placed here on the elimination or minimization of wave front contributions with radially at least quadratic dependence in the pupil, d. H. of focus terms and astigmatism terms.

The projection objective preferably has an off-axis field having an aspect ratio greater than 2: 1 between a longer and a shorter side, wherein the optically used region has approximately a rectangular shape with an aspect ratio greater than 2: 1 and the first manipulator acts parallel to the longer side. In this direction, the first manipulator should be able to generate several maxima and several minima of the optical path length change of the projection radiation. The longer side can be used particularly easily to vary the path length change.

There are various ways to achieve the advantages of the novel wavefront manipulation system in practice.

In one embodiment, the wavefront manipulation system has only the first manipulator in the optical proximity of the field plane and the finite first distance is dimensioned such that the condition on the first manipulator surface upon activation of the first manipulator 0.25 · NA _M ² <SAD / P _CHAR <0.48 (1) is satisfied, where NA _{M is} the numerical aperture of the projection radiation at the first manipulator surface.

Here, therefore, only one manipulator is provided. As a result, the structural design can be relatively simple and the need for space can be low.

In order to achieve the desired effect on the wavefront with sufficient strength when using only a single manipulator without simultaneously producing unwanted residual aberrations to disturbing extent, the first manipulator surface should be located neither too close to this nearest field level nor too far away from this closest field level , A favorable first distance is given when the above condition (1) is satisfied. The parameter SAD stands for the subaperture diameter of the projection radiation the first manipulator surface. This parameter takes into account that each beam emanating from a field point of a field at the first manipulator surface illuminates a subaperture with a subaperture diameter SAD. The subaperture diameter can be understood as the footprint footprint of a single beam emerging from a field point on an optical surface. In the case of a divergent beam bundle, the subaperture diameter increases with increasing distance to the field.

From the condition (1) given above, it becomes clear that the subaperture diameter should be in a certain ratio to the characteristic period P _CHAR . The ratio SAD / P _CHAR is also referred to in this application as "normalized manipulator distance" and represented by the parameter D _NORM : = SAD / P _CHAR .

If the subaperture diameter SAD becomes too large in relation to the characteristic period so that the upper limit is exceeded, a sufficiently strong influence on the field dependence of the distortion can generally be achieved; at the same time, however, the level of normally unwanted aberrations, in particular the level of defocus and / or astigmatism, increases to an extent which can have a noticeable disturbing effect on the image. On the other hand, if the lower limit is undershot, the level of unwanted aberrations, such as astigmatism, for example, can be kept low. At the same time, however, normally the distortion can no longer be influenced to a sufficiently large extent depending on the field, so that the effect of the first manipulator remains limited to relatively small values of the distortion. In addition, unfavorable contributions to defocus bending may result if the first manipulator surface is too close to the field plane.

In some embodiments, the closest field plane is the object plane of the projection objective. It is particularly advantageous if no optical surface with refractive power is arranged between the object plane and the first manipulator surface, so that the numerical aperture NA _{M of} the projection radiation at the first manipulator surface is equal to the object-side numerical aperture NA _O. Immediately behind the pattern of the mask to be imaged, the effect of the manipulator can not be disturbed by aberrations of intervening optical elements, so that the first manipulator can be used in a particularly targeted manner. In addition, there may be no or no accessible intermediate image layer.

The manipulation of the distortion by a single manipulator can be difficult if z. B. the then required relatively small first distance to a field level due to structural conditions is difficult to implement. In addition, the maximum effect of a single manipulator is limited if demands for negligible higher-order contributions are to be met. For this reason, among others, it is provided in some embodiments that the wavefront manipulation system has, in addition to the first manipulator, a second manipulator having a second manipulator surface arranged in the projection beam path and a second actuating device for reversibly changing the surface shape and / or refractive index distribution of the second manipulator surface ,

The two manipulators can preferably be adjusted independently of each other.

Using at least two manipulators results in larger areas of suitable distances to adjacent field levels, which can facilitate the insertion of manipulators into the projection objective. In addition, two or more manipulators may be operated in concert so that their desired field-dependent distortion effects mutually reinforce each other, while the undesirable effects on other aberrations, such as aberrations, are minimized. B. defocus and / or astigmatism, can compensate each other at least partially.

Preferably, the first manipulator surface and the second manipulator surface are arranged such that the numerical aperture of the projection radiation on the first manipulator surface is equal to the numerical aperture of the projection radiation on the second manipulator surface. There should be no optical element with refractive power between the manipulator surfaces. This can u. a. can be achieved that the adjustments of the manipulators (eg, in terms of the characteristic period) can be particularly easily matched.

It is also possible that a wavefront manipulation system has more than two independently controllable, close to the field arranged manipulators whose effects can be coordinated with each other, for. B. three or four manipulators.

If the projection objective is configured in such a way that at least one real intermediate image is generated in the region of an intermediate image plane between the object plane and the image plane, then the field plane closest to the first manipulator surface can also be this intermediate image plane.

The first manipulator surface can lie in the radiation direction in front of the intermediate image plane or behind the intermediate image plane.

The numerical aperture at the location of the first manipulator surface then depends on the magnification scale with which the object is imaged into the corresponding intermediate image. With a diminishing image between object and intermediate image, the numerical aperture in the vicinity of the intermediate image is larger than the object-side numerical aperture NA _O , so that the suitable first distances D1 are smaller than in the case of an arrangement in the vicinity of the object plane. If, on the other hand, there is a magnifying image between the object plane and the intermediate image plane, the distance values suitable for mounting the first manipulator surface increase to the intermediate image plane. An arrangement in the vicinity of a real intermediate image can therefore z. B. be favorable for space reasons.

It may be particularly favorable to arrange two manipulators on different sides of the intermediate image in the vicinity of an intermediate image. In one embodiment, the first manipulator surface is arranged in front of the intermediate image plane and the second manipulator surface behind the intermediate image plane. A reverse arrangement is also possible. As a result, it is possible to achieve, for example, that the effects on the odd aberrations increase, while in the case of even aberrations, in particular defocus and astigmatism, the contributions of the two manipulators partially or completely compensate each other.

By using two manipulators on different sides of an intermediate image, the spacing requirements can be relaxed compared to using only a single manipulator. It has been found that then the first distance and the second distance should be dimensioned so that at the first manipulator surface (when activating the first manipulator) and at the second manipulator surface (when activating the second manipulator) respectively the condition 0.012 <SAD / P _CHAR <0.85 (2) is satisfied. The permissible range for the normalized manipulator distance increases in comparison to condition (1). In addition, the lower limit is no longer dependent on the numerical aperture of the projection radiation on the manipulator surface, so that this variant can be advantageous especially in high-aperture regions of the projection beam path.

It is possible to arrange two manipulators immediately following one another on the same side of a field plane at different distances therefrom. In one embodiment, the second manipulator surface is arranged immediately behind the first manipulator surface, wherein the numerical aperture of the projection radiation at the first manipulator surface is equal to the numerical aperture of the projection radiation at the second manipulator surface and the first distance is smaller than the second distance, so that the Subaperture differ at the manipulator surfaces, wherein for the first manipulator surface the condition 0.25 · NA _M ² <SAD / P _CHAR <0.8 (3A) and for the second manipulator surface the condition SAD / P _CHAR <1.5 (3B) applies. Although the first manipulator surface should not be moved closer to the field plane than when using only a single manipulator (see condition (1)). However, the usable distance range increases to larger tolerable intervals.

The invention may, for. B. in catadioptric projection lenses or dioptric projection lenses are used, possibly also in other imaging systems.

The invention also relates to a projection exposure method for exposing a radiation-sensitive substrate arranged in the region of an image surface of a projection lens to at least one image of a pattern of a mask arranged in the region of an object surface of the projection lens in which a projection objective according to the invention is used.

The invention further relates to a projection exposure apparatus for exposing a radiation-sensitive substrate arranged in the region of an image plane of a projection lens to at least one image of a pattern of a mask arranged in the region of an object plane of the projection lens, comprising: a primary radiation source for emitting primary radiation; an illumination system for receiving the primary radiation and for producing an illumination radiation directed onto the mask; and a projection objective for generating at least one image of the pattern in the region of the image surface of the projection objective, wherein the projection objective is designed according to the invention.

The projection exposure apparatus preferably has a central controller for controlling functions of the projection exposure apparatus, wherein the control apparatus is a control module for controlling the wavefront manipulation system is assigned and a manipulator or more manipulators on the control module z. B. can be controlled by means of electrical signals in coordination with other control signals during operation of the projection exposure system.

The above and other features are apparent from the claims and from the description and from the drawings, wherein the individual features are realized individually or in each case in the form of sub-combinations in an embodiment of the invention and in other fields and advantageous and can represent protectable embodiments. Embodiments of the invention are illustrated in the drawings and explained in more detail below.

BRIEF DESCRIPTION OF THE DRAWINGS

1 shows a schematic representation of a microlithography projection exposure apparatus according to an embodiment of the invention;

2 shows a schematic longitudinal section in an xz plane in the region of the mask and an immediately following first manipulator element;

3 shows a plan view of the first manipulator element in 2 parallel to the optical axis;

4 to 8th illustrate effects of a field-first positioned first manipulator element on the correction state of the wavefront in the image field of a projection lens in dependence on different influencing parameters;

9 schematically shows elements of a wavefront manipulation system, which has manipulator elements in front of and behind an intermediate image plane;

10 shows in analogy to the representations of 4 and 5 the dependence of different Zernike coefficients on the normalized manipulator distance on both sides of an intermediate image plane;

11 schematically shows elements of a wavefront manipulation system, which has two closely adjacent manipulator elements behind a field plane;

12 to 15 each show bar graphs illustrating the effects of manipulations on selected aberrations, the height of the bars indicating the contributions to each of the x-axis aberrations (represented by Zernike coefficients), respectively;

16 and 17 show schematic meridional lens sections of embodiment catadioptric projection lenses, which are equipped with field-near manipulator elements;

18 to 21 show schematically embodiments of dynamically adjustable manipulators that can be used at near-field positions within a projection lens in the context of wavefront manipulation systems.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

In 1 An example of a microlithography projection exposure apparatus WSC is shown which can be used in the manufacture of semiconductor devices and other finely-structured components and works to achieve resolutions down to fractions of a micron with deep ultraviolet (DUV) electromagnetic radiation. The primary radiation source or light source LS is an ArF excimer laser with an operating wavelength λ of approximately 193 nm. Other UV laser light sources, for example F ₂ lasers with a working wavelength of 157 nm or ArF excimer lasers with a working wavelength of 248 nm, are also used possible.

An illumination system ILL connected downstream of the light source LS produces in its exit surface ES a large, sharply delimited and substantially homogeneously illuminated illumination field, which is adapted to the telecentricity requirements of the projection objective PO arranged behind the light path. The illumination system ILL has facilities for setting different illumination modes (illumination settings) and can be switched, for example, between conventional on-axis illumination with different degree of coherence σ and off-axis illumination. The off-axis illumination modes include, for example, an annular illumination or a dipole illumination or a quadrupole illumination or another multipolar illumination. The construction of suitable lighting systems is known per se and will therefore not be explained in detail here. The patent application US 2007/0165202 A1 (corresponding to WO 2005/026843 A2 ) shows examples of lighting systems that can be used in various embodiments.

Those optical components which receive the light from the laser LS and form illumination radiation from the light which is directed onto the reticle M belong to the illumination system ILL of the projection exposure apparatus.

Behind the illumination system, a device RS for holding and manipulating the mask M (reticle) is arranged such that the pattern arranged on the reticle lies in the object plane OS of the projection objective PO, which coincides with the exit plane ES of the illumination system and here also referred to as reticle plane OS becomes. The mask is movable in this plane for scanner operation in a scan direction (y-direction) perpendicular to the optical axis OA (z-direction) by means of a scan drive.

Behind the reticle plane OS follows the projection objective PO which acts as a reduction objective and an image of the pattern arranged on the mask M on a reduced scale, for example at a scale of 1: 4 (| β | = 0.25) or 1: 5 (| β | = 0.20 ) is imaged onto a substrate W coated with a photoresist layer or photoresist layer whose photosensitive substrate surface SS lies in the region of the image plane IS of the projection objective PO.

The substrate to be exposed, which in the example is a semiconductor wafer W, is held by a device WS comprising a scanner drive to move the wafer in synchronism with the reticle M perpendicular to the optical axis OA in a scanning direction (y direction). to move. The device WS, which is also referred to as "wafer days", and the device RS, which is also referred to as "reticle days", are part of a scanner device which is controlled via a scan control device, which in the embodiment is transferred to the central control device CU the projection exposure system is integrated.

The illumination field generated by the illumination system ILL defines the effective object field OF used in the projection exposure. This is rectangular in the example case, has a parallel to the scanning direction (y-direction) measured height A * and a perpendicular thereto (in the x-direction) measured width B *> A *. The aspect ratio AR = B * / A * is generally between 2 and 10, in particular between 3 and 6. The effective object field is located at a distance in the y-direction next to the optical axis (off-axis field or off-axis field). The effective image field optically conjugate to the effective object field in the image area IS has the same shape and the same aspect ratio between height B and width A as the effective object field, but the absolute field size is reduced by the imaging scale β of the projection objective, i. H. A = | β | A * and B = | β | B *.

If the projection objective is designed and operated as an immersion objective, then during operation of the projection objective, a thin layer of immersion liquid, which is located between the exit face of the projection objective and the image plane IS, is transmitted. In immersion mode, image-side numerical apertures NA> 1 are possible. A configuration as a dry objective is also possible, here the image-side numerical aperture is limited to values NA <1. Under these conditions, which are typical of high-resolution projection objectives, projection radiation with a relatively large numerical aperture is present in the region of some or all field levels (object plane, image plane, possibly one or more intermediate image planes) of the projection objective, eg. With values greater than 0.15 or greater than 0.2 or greater than 0.3.

The projection objective or the projection exposure apparatus is equipped with a wavefront manipulation system WFM, which is configured to dynamically change the wavefront of the projection radiation passing from the object plane OS to the image plane IS in the sense that the optical action of the wavefront manipulation system is variable via control signals can be adjusted. The wavefront manipulation system of the exemplary embodiment has a first manipulator MAN1 which has a first manipulator ME1, which is arranged in the immediate vicinity of the object plane of the projection lens in the projection beam path and has a first manipulator surface MS1 arranged in the projection beam path, whose surface shape and / or refractive index distribution with help a first adjusting device DR1 can be reversibly changed.

For further explanation shows 2 a schematic longitudinal section in an xz plane in the region of the mask M and the immediately following first manipulator element ME1. The first manipulator element ME1 is a plate-shaped optical element made of a material that is transparent to the projection radiation, for example made of synthetic quartz glass (fused silica). A light entry side facing the object plane OS serves as first manipulator surface MS1, the opposite light exit surface is a plane surface.

The first adjusting device comprises a plurality of independently actuatable actuators (not shown) which engage the plate-shaped manipulator element ME1 in such a way that the surface shape of the first manipulator surface MS1 can be defined in a wave form. In this case, both the "amplitude" of the waves measured parallel to the z-direction, ie the deflection of the manipulator surface in the z-direction, as well as the distance measured in the x-direction between adjacent wave peaks, ie the wavelength or the period of the wave pattern, on different values are set. In the example, a sinusoidal waveform is set in the x-direction, the (mitlere) wavelength in the x-direction can be characterized by a characteristic period P _CHAR .

The first manipulator surface is arranged in a finite first distance D1 to the object plane OS of the projection lens, which is the closest field plane to the first manipulator surface. The first manipulator surface MS1 is visually arranged in the immediate vicinity of the object plane OS, ie in a "near-field position". This is among other things 3 recognizable. 3 shows a plan view of the first manipulator surface MS1 and the first manipulator ME1 parallel to the optical axis OA (in the z-direction). The rectangular area FP with rounded corners represents that area of the first manipulator area which is illuminated by the rays coming from the effective object field OF. This area is also called a "footprint".

The footprint or footprint of the projection radiation in this case represents in size and shape the sectional area between the projection beam and the area penetrated by the projection beam (here the first manipulator area MS1). The optical proximity to the object plane OS can be recognized by the fact that the footprint essentially has the rectangular shape of the object field OF, the corner regions being slightly rounded. In addition, the footprint is just like the object field outside the optical axis OA. While the optical region used by the projection radiation in the optical proximity of the field essentially has the shape of the illuminated field region, a substantially circular region is illuminated in the region of the Fourier-transformed pupil plane to a field plane, so that a footprint in the region of a pupil is at least approximately Has circular shape.

The area illuminated on the first manipulator surface MS1 has an effective diameter D _FP in the x direction. The 2 and 3 show schematically that the surface form of the first manipulator surface in this direction has several local maxima (represented by peaks in 2 and by "+" signs in 3 ) and several intermediate local minima (represented by troughs or "-" signs in 3 ) having. This results in a "ripple" in the x-direction or in the direction of the effective diameter.

In the case of a projection objective, a ray bundle starts from each field point of the object plane whose diameter increases with increasing distance from the object plane. The object-side numerical aperture NA _O corresponds to the sine of the opening angle α of each of the beam bundles. Each beam emanating from a field point illuminates at the first manipulator surface MS1 a circular subaperture whose diameter is designated as Subaperturdurchmesser SAD. Out 2 it is immediately apparent that the subaperture diameter SAD increases with increasing first distance D1 as well as with increasing numerical aperture on the image side. The first manipulator is arranged so close to the object plane that several subapertures without mutual overlap in the x direction next to one another fit into the illuminated area FP. Preferably, the condition SAD / D _FP <0.2 should be satisfied, in particular even the condition SAD / D _FP <0.1.

If these conditions are met, it is possible, with the aid of the dynamic first manipulator, to influence the distortion in the image field of the projection objective in a location-dependent manner such that field-dependent distortion correction becomes possible. This is achieved in that the first manipulator is able to introduce different changes in the optical path length for beams emanating from different field points. The change in the optical path length, which is also referred to as optical path length change for short, should be referred to here as parameter ΔOPL.

There are various causes for a change in the optical path length of a beam when passing through a transparent optical element made of a material with refractive index n and irradiated thickness d. If Δd is the change in the thickness d of the irradiated optical element in the z-direction, then, in the case of vertical transmission, ΔOPL = Δd · n-1. If, in the irradiated optical element, z. B. due to heating or cooling results in a refractive index change Δn, then ΔOPL = Δn · d. These causes can be alternative or cumulative. Often a cause dominates. Both effects can be used within the scope of the invention.

In the example of 2 From the first field point FP1 a first ray bundle emanates whose main ray (dashed line) running parallel to the optical axis strikes in the region of a "wave valley", ie in the region of a local minimum of the plate thickness. In a second beam shown by way of example, which starts from a second field point FP2 laterally offset from the first field point, the main beam (shown in dashed lines) strikes in the region of a wave crest, ie in the region of a local maximum of the plate thickness. The outgoing from the two field points beam thus undergo different changes in the optical path length.

The local distribution of the optical path length changes in the x direction can be varied by triggering the actuating devices of the first manipulator such that a differently deformed surface shape with a different waviness can be set, so that the field dependence of the wavefront correction can also be set. The The shape and strength of the wavefront change depend on the course of the optical path length change by the first manipulator surface MS1 within the subaperture belonging to the beam bundle. For example, a linear increase or decrease in the optical path length change over the subaperture diameter leads to a tilt of the wavefront; a quadratic profile leads to an influence on focus and astigmatism.

Based on 4 to 8th The effects of a first manipulator element positioned near the field on the correction state of the wavefront in the image field IF of the projection objective as a function of different influencing parameters will now be explained. In the field of geometric optics, Zernike polynomials are commonly used to represent wavefronts, which in turn describe the aberrations of optical systems. The individual aberrations can be described by the coefficients of the Zernike polynomials, ie the Zernike coefficients or their values (in [nm]). In the representation chosen here, the Zernike coefficients Z2 and Z3 represent the tilting of a wavefront in the x-direction or the y-direction, resulting in a distortion-like error. The Zernike coefficient Z4 describes a curvature of the wavefront, whereby a defocus error can be described. The Zernike coefficient Z5 describes a saddle-shaped deformation of the wavefront and thus the astigmatism component of a wavefront deformation. The Zernike coefficients Z7 and Z8 stand for coma, the Zernike coefficient Z9 for spherical aberration and the Zernike coefficients Z10 and Z11 for three-wave.

Based on 4 to 8th Let us first describe an example of a wavefront manipulation system which has only a single first manipulator with a first manipulator surface, which is arranged in the optical proximity of the object plane. In the 4 to 7 In this case, the sensitivities of such a manipulator are represented as a function of the axial position, ie as a function of the distance from the nearest field plane. The term "sensitivity" here refers to the effect of the manipulator on the wavefront at a defined "deflection" of the manipulator, wherein the "deflection" can be given for example by a maximum optical path length change ΔOPL = 1 nm.

In the 4 to 7 In this case, the "normalized manipulator distance" D _NORM from the closest field plane, ie the ratio SAD / P _{CHAR, is} indicated on the x-axis. A value SAD / P _CHAR = 0.0 stands for a position directly in the field level. Positive values represent positioning in the transmission direction behind this field level, negative values for positions in front of this field level. In each case the value Z _{MAX of} the respective Zernike coefficients in the form of the maximum value is plotted on the y-axis over the considered field (in [nm]). 4 shows the values for NA _M = 0.3375 while 5 shows the corresponding values for NA _M = 0.675. The numerical aperture of the projection radiation at the first manipulator surface is thus in the case of 5 twice as big as in the case of 4 ,

It can be seen that the effect on the distortion error (Z2 / 3) is an odd function in the sense that the effect on placement before the field level and the effect on placement behind the field plane have opposite signs. The sensitivity curve of Z2 / 3 has a turning point at the field level. The sensitivities for defocus (Z4) and astigmatism (Z5), on the other hand, are just functions in the sense that their course is mirror-symmetrical with respect to the field level and has a local extreme value (maximum at defocus, minimum at astigmatism) at the field level. Incidentally, the astigmatism term and the distortion term disappear in the field plane, while the defocus pattern Z4 has a local maximum directly in the field plane and only disappears outside the field plane at a certain distance. (The solid lines represent higher-order aberrations, namely coma (Z7 / 8), spherical aberration (Z9), and three-waviness (Z10 / 11), which are not considered here.) The ratios at higher numerical aperture ( 5 ) are similar, it being understood that the extent of the defocus pattern increases with increasing numerical aperture at the manipulator surface.

The first manipulator should be used in the example case to influence distortion errors field dependent. These are also described by the value Z3, which thus stands for the aberration that is to be influenced, ie. H. the "desired aberration". The other aberrations, in particular Z4 (defocus) and Z5 (astigmatism) should as far as possible not be influenced or be influenced so little, the resulting aberrations are in a regularly negligible order of magnitude.

To illustrate how the positioning of the manipulator surface on the one hand to the desired aberrations (Z2, Z3) and on the other hand on the unwanted residual aberrations (Z4, Z5, etc.), are in the 6 and 7 the the 4 and 5 corresponding values are plotted again in such a way that the sensitivities are plotted on the y-axis as normalized values based on a Z3 sensitivity of 1 nm. In this application it can be seen which "price" (at here unwanted residual aberrations) is payable to to produce a (desired) distortion correction on the order of at most 1 nm Z3.

In this normalized plot, it can be seen that at finite distance from the field plane there is a useful range (UR) of finite width in which both the defocus pattern and the astigmatism term remain below the limits considered critical for these aberration contributions. If, for example, it is assumed that the absolute value of the defocus pattern Z4 should not be greater than 0.2 nm for a maximum distortion contribution of Z2 = 1 nm, it can be seen that the first manipulator surface should not be brought closer to the field plane than to a normalized manipulator _{distance value} D _NORM = 0.03. If the first manipulator surface is brought closer to this field level than this limit value, a very strong increase in the defocus contribution results due to a singularity of the defocus pattern at this point. At greater distances towards the well usable distance range UR is limited by the contributions to astigmatism. Typically, this application assumes that astigmatism terms of less than 0.4 nm are tolerable in most cases, while the astigmatism contribution (Z5) should not be higher than this limit. The limit value Z5 = 0.4 nm includes a normalized manipulator distance D _NORM = approx. 0.48.

These limits arise 6 for a value N _AM = 0.3375. In 7 a corresponding plot for twice the numerical aperture is plotted on the first manipulator surface. A corresponding evaluation of the usable distance range UR shows that due to the greater increase of the defocus pattern in the vicinity of the field plane, a larger normalized manipulator distance to the field plane should now be maintained, whereby now the lower limit of the normalized manipulator distance is approximately D _NORM = 0.12. The upper limit caused by the astigmatism term is independent of the numerical aperture at the manipulator surface.

A more detailed analysis shows that the lower limit of the usable distance range for a given upper limit value for the defocus contribution scales to a good approximation with the square of the numerical aperture at the manipulator surface. 8th shows a diagram which shows on the x-axis the numerical aperture N _{AM of} a manipulator surface and on the y-axis the normalized manipulator distance D _NORM for a given limit value of the Defocusterms with Z4 / Z3 = 0.2. The dashed line shows in each case the results of the simulations described in the preceding figures, the solid line is given by 0.25 = NA _M ² .

Thus, if a single, close-to-field, dynamically manipulatable manipulator surface is used, the distance range of the normalized manipulator distance that can be used for a well-functioning manipulator results from the following inequality: 0.25 · NA _M ² <SAD / P _CHAR <0.48 (1)

The upper limit is determined by the ratio Z5 / Z3 = 0.4. The corresponding value SAD / P _CHAR > 0.48 is independent of NA _M. The NA-dependent lower limit is given by 0.25 NA _M ² .

Wavefront manipulation systems will now be described by way of example which have a plurality of independently operable manipulators. For clarity of illustration examples are explained in which in addition to the first manipulator exactly a second manipulator is provided, a second manipulator element arranged in the projection beam path second manipulator surface and a second adjusting device for reversibly changing the surface shape and / or refractive index distribution of the second manipulator surface having. A wavefront manipulation system may also include more than two independently adjustable manipulators, for example three or four manipulators.

In the embodiment of 9 the projection lens is configured such that between the object plane OS and the image plane IS there is an intermediate image plane IMIS, in which a real intermediate image IMI is generated. For this purpose, the projection objective has a plurality of optical elements between the object plane OS and the intermediate image plane IMIS, which together form an optical imaging system which forms a first objective part of the projection objective. Depending on the configuration of the first objective part, the intermediate image can be enlarged or reduced in size compared to the effective object field located in the object plane OS or have the same size as this one. The intermediate image is imaged directly or via at least one further intermediate image in the image plane IS, typically with a greatly reduced magnification.

The wavefront manipulation system has a first manipulator MAN1 with a first manipulator element ME1, whose first manipulator surface MS1 is arranged at a first distance D1 behind the intermediate image plane, ie between it and the image plane. A second manipulator MAN2 has a second manipulator element ME2 with a second manipulator surface MS1, which is arranged in a finite second distance D2 in front of the intermediate image plane, that is between this and the object plane. One in relation to the Intermediate image plane symmetrical arrangement with equal distances D1 = D2 is favorable because then the unwanted even aberrations can be fully compensated. The plate-shaped manipulator elements are each near the field to the intermediate image plane, so that the conditions SAD / D _FP <0.2 applies to the respective manipulator surfaces.

Each of the manipulator elements can be controlled such that in the x direction a multiple change of the effect on the change of the optical path length can be generated. The patterns for the adjustment of the optical path length changes ΔOPL are inverse to each other. This is symbolized by the symbols "+" and "-", where "+" represents a local maximum and "-" represents a local minimum of the optical path length changes ΔOPL. It can thus be achieved that the contributions to the even aberrations generated by the two manipulators, such as, for example, defocus and astigmatism, compensate each other, so that the combination of the two manipulators produces practically no contributions to defocus and astigmatism.

10 shows in analogy to the representations of the 4 and 5 the dependence of different Zernike coefficients on the normalized manipulator _distance D _NORM on both sides of the intermediate image plane at SAD / P _CHAR = 0. The values are calculated for NA _M = 0.3375.

Since the contributions to defocus and astigmatism can cancel each other out with opposite control and equal distances to the intermediate image plane, larger usable distance ranges arise in front of and behind the intermediate image plane than in the case of only a single manipulator. A certain minimum distance of the respective manipulator surfaces to the intermediate image plane is required in order to obtain a sufficiently high sensitivity of the manipulators for the aberrations to be manipulated (Z2 / 3) without unduly deforming or stressing the manipulators. Assuming a desired correction range for these aberrations in the order of magnitude of 2 nm and strives for a correction range up to 50 nm as the maximum permissible deformation per manipulator surface, the condition SAD / P _CHAR > results as a condition for the lower limit of the normalized manipulator _distance 0012. The upper limit of the usable distance range can be suitably described by the condition SAD / P _CHAR <0.85, if too large contributions for coma and tristimulus are to be avoided. Thus, preferably 0.012 <SAD / P _CHAR <0.85 (2)

There are other ways to configure a wavefront manipulation system with two independently operable manipulators in the projection beam path. 11 shows a schematic representation of an embodiment in which the wavefront manipulation system comprises two manipulators MAN1, MAN2, which are arranged directly behind each other at different distances to a front in the transmission direction field plane (here object plane OS) of the projection lens. The essentially plate-shaped first and second manipulator elements ME1, ME2 are arranged directly behind one another without the interposition of an optical element with refractive power in the divergent beam path such that the numerical aperture of the projection radiation on both manipulator surfaces (first manipulator surface MS1, second manipulator surface MS2) is the same. The first manipulator element is closer to the nearest field plane OS, so that the first distance D1 is smaller than the second distance D2.

Due to the different distances to the field points, each of which emanate beam bundles, the Subaperturdurchmesser are each smaller on the first manipulator surface than on the second manipulator surface. This leads to different sensitivities with regard to the target variable to be changed (in this case Z2 / 3 distortion) and with regard to the other unwanted aberrations. By a suitable control of the two manipulators can be achieved that the same effect on the target size, the unwanted aberrations compared to the case of only one manipulator (see 2 ) to reduce.

This results in more relaxed distance requirements for placement of the manipulator elements, wherein it is particularly possible to arrange a manipulator element further away from the field plane than in the case of only a single manipulator element.

The analyzes explained in more detail below have shown that the first manipulator closer to the field level should not lie closer to the field plane than in the case of just a single manipulator in order to achieve a good compromise between sufficiently high sensitivity and sufficiently small contributions to the unwanted aberrations for this manipulator to achieve. The normalized manipulator distance should not be less than 0.25 = NA _M ² . However, it is possible to arrange the first manipulator surface further away from the field plane than in the case of a single manipulator. With appropriate boundary conditions, an upper limit of 0.8 for the normalized manipulator distance appears feasible. The second manipulator surface can lie much further away from the nearest field plane, wherein the normalized Manipulator distance should preferably be less than 1.5.

To illustrate the effects on the aberrations show the 12 to 15 each bar graphs, wherein the height of the bars respectively represent the contributions to the individual aberrations Z _i [nm] indicated on the x-axis. The first example is in the 12 and 13 shown. The respective left bars with darker gray tone represent a reference case in which only a single manipulator is arranged in a normalized manipulator distance SAD / P _CHAR = 0.5 from the nearest field level. The respective right bars with lighter gray tone represent an exemplary situation with two directly successive manipulator elements, wherein the first manipulator element closer to the field plane has the same normalized manipulator distance (0.5) to the nearest field plane, while the further manipulator element is arranged at SAD / P _CHAR = 0.7 is. The bar heights in all diagrams are normalized to a value of 1 nm of the desired aberration Z3. This means that the manipulator is driven so that it provides a contribution of 1 nm to the tilt of the wavefront.

First, only the darker bars in the 12 and 13 Considering the situation with a single manipulator. It can be seen that a desired contribution to distortion correction on the order of 1 nm Z3 generates various unwanted aberrations, with defocus (Z4) and astigmatism (Z5) being the dominant residual aberrations. This corresponds to the situation associated with 6 has already been explained.

If, in contrast to the first manipulator, a second manipulator at SAD / P _CHAR = 0.5 is directly connected downstream at SAD / P _CHAR = 0.7, the residual aberrations can be significantly reduced, especially for defocus (Z4) and astigmatism (Z5). With the same sensitivity for the desired aberration (Z3), the contribution to the defocus is reduced from about -0.18 to about -0.03. For astigmatism (Z5) there is a reduction to about ¼ of the comparison value, namely from 0.4 to about 0.09. The combination of two directly connected in series manipulators thus results in much lower contributions to defocus and astigmatism as a single manipulator, which is arranged at a comparable distance to the nearest field level.

To estimate a practicable upper limit for the distances to the nearest field level, the 14 and 15 in a corresponding plot, the aberration contributions for a case in which the first manipulator element closer to the field plane is arranged in a normalized manipulator distance SAD / P _CHAR = 0.8 and the _remoter second manipulator element is located at SAD / P _CHAR = 1.0. The darker left bars each show the aberration contributions for the case where only the closest first manipulator element is used. The brighter bars show the corresponding results with inverse control of the two manipulator elements arranged immediately one behind the other. It can be seen that, given a desired sensitivity Z3 = 1 nm, the individual manipulator element delivers contributions in defocus (Z4) as well as in astigmatism (Z5) which are greater than the upper limit values (Z4 = 0.2 nm and Z5 assumed here as tolerable) = 0.4 nm). The combination of two manipulators can reduce these contributions so significantly that the contribution to the defocus (now approx. -0.08) and astigmatism (now approx. 0.2) are below the upper limit values assumed here, so that no disturbing aberrations of this kind are generated any longer. The odd aberrations, such as coma (Z8) and tristimulus (Z11) are just within a tolerable range of the order of approximately 0.3 nm each. This example shows that the upper bounds given here should not be exceeded if the contributions are too odd Aberrations should not be too high. Overall, it is considered favorable if the condition 0.25 · NA _M ² <SAD / P _CHAR <0.8 applies to the first manipulator _surface and the condition SAD / P _CHAR <1.5 applies to the second manipulator _surface .

In the following, various combinations of projection objectives and manipulator devices that can be used therein will be illustrated by means of more concrete embodiments.

16 shows a schematic meridional lens section of an embodiment of a catadioptric projection lens 1600 with selected beam bundles to illustrate the imaging beam path of the running through the projection lens projection radiation. The projection objective is provided as a reduction-acting imaging system for imaging a pattern of a mask arranged in its object plane OS on a reduced scale, for example on a scale of 4: 1, onto its image plane IS aligned parallel to the object plane. In this case, exactly two real intermediate images IMI1, IMI2 are generated between the object plane and the image plane. A first objective part OP1, which is constructed exclusively with transparent optical elements and is therefore purely refractive (dioptric), is designed such that the pattern of the object plane is imaged into the first intermediate image IMI1 substantially without a change in size. A second catadioptric objective part OP2 essentially forms the first intermediate image IMI1 on the second intermediate image IMI2 Resizing. A third, purely refractive objective part OP3 is designed to image the second intermediate image IMI2 into the image plane IS with great reduction.

Between the object plane and the first intermediate image, between the first and the second intermediate image and between the second intermediate image and the image plane, there are respective pupil surfaces P1, P2, P3 of the imaging system where the main beam CR of the optical image intersects the optical axis OA. In the area of the pupil surface P3 of the third objective part OP3, the aperture stop AS of the system is attached. The pupil surface P2 within the catadioptric second objective part OP2 is in the immediate vicinity of a concave mirror CM.

If the projection objective is designed and operated as an immersion objective, then during operation of the projection objective, a thin layer of immersion liquid, which is located between the exit face of the projection objective and the image plane IS, is transmitted. Immersion lenses with a comparable basic structure are z. In the international patent application WO 2004/019128 A2 shown. In immersion mode, image-side numerical apertures NA> 1 are possible. A configuration as a dry objective is also possible, here the image-side numerical aperture is limited to values NA <1.

This in 16 embodiment shown corresponds to the embodiment in the optical structure (without manipulator elements) 6 of the US 2009/0046268 A1 , The disclosure of this document regarding the basic structure of the projection lens (optical specification) is incorporated herein by reference.

The catadioptric second objective part OP2 contains the single concave mirror CM of the projection objective. Immediately in front of the concave mirror is a negative group NG with two negative lenses. In this arrangement, sometimes called Schupmann achromat, the Petzvalkorrektur, d. H. the correction of the field curvature, by the curvature of the concave mirror and the negative lenses in the vicinity, the chromatic correction achieved by the refractive power of the negative lenses in front of the concave mirror and the diaphragm position with respect to the concave mirror.

A reflective deflection device serves to separate the beam bundle extending from the object plane OS to the concave mirror CM or the corresponding partial beam path from the beam bundle or partial beam path that runs after reflection at the concave mirror between it and the image plane IS. For this purpose, the deflecting device has a plane first deflecting mirror FM1 for reflecting the radiation coming from the object plane to the concave mirror CM and a second deflecting mirror FM2 aligned at right angles to the first deflecting mirror FM1, which deflects the radiation reflected by the concave mirror in the direction of the image plane IS. Since the optical axis is folded at the deflecting mirrors, the deflecting mirrors in this application are also referred to as folding mirrors. The deflecting mirrors are tilted with respect to the optical axis OA of the projection lens about tilting axes extending perpendicular to the optical axis and parallel to a first direction (x-direction), e.g. B. by 45 °. In a design of the projection objective for the scanning operation, the first direction (x-direction) is perpendicular to the scanning direction (y-direction) and thus perpendicular to the direction of movement of the mask (reticle) and substrate (wafer). For this purpose, the deflection device is realized by a prism, the outside of which mirrored, perpendicularly aligned catheter surfaces serve as deflecting mirrors.

The intermediate images IMI1, IMI2 are in each case in the optical proximity of their closest folding mirrors FM1 and FM2, but can have an optical minimum distance to them so that any errors on the mirror surfaces are not sharply imaged in the image plane and the plane deflection mirrors (plane mirrors) FM1, FM2 are in the range of moderate radiant energy density.

The positions of the (paraxial) intermediate images define field levels of the system which are optically conjugate to the object plane or to the image plane. The deflection mirrors are thus in optical proximity to field levels of the system, which is also referred to as "close to field" in the context of this application. In this case, the first deflecting mirror is arranged in the optical proximity of a first field plane belonging to the first intermediate image IMI1 and the second deflecting mirror is arranged in optical proximity to a second field plane which is optically conjugate to the first field plane and belongs to the second intermediate image IMI2.

For quantifying the position of an optical element or an optical surface in the beam path z. B. the Subaperturverhältnis SAR can be used.

According to an illustrative definition, the subaperture ratio SAR of an optical surface of an optical element in the projection beam path is defined as the quotient between the subaperture diameter SAD and the optically free diameter DCA according to SAR: = SAD / DCA. The subaperture diameter SAD is given by the maximum diameter of a partial surface of the optical element illuminated with rays of a beam emanating from a given field point. The optically free diameter DCA is the Diameter of the smallest circle about a reference axis of the optical element, the circle enclosing that area of the surface of the optical element which is illuminated by all rays coming from object field.

SAR = 0 in a field plane (object plane or image plane or intermediate image plane). In a pupil plane, SAR = 1. "Field-near" surfaces thus have a subaperture ratio that is close to 0, while "pupil-near" surfaces have a subaperture ratio that close to 1.

The optical proximity or the optical distance of an optical surface to a reference plane (eg a field plane or a pupil plane) is described in this application by the so-called subaperture ratio SAR. The subaperture ratio SAR of an optical surface is defined as follows for the purposes of this application: SAR = signCRH (MRH / (| CRH | + | MRH |)) where MRH denotes the boundary ray height, CRH the principal ray height and the sign function sign x the sign of x, where by convention sign 0 = 1. The main beam height is understood to be the beam height of the main beam of a field point of the object field with the maximum field height in terms of magnitude. The beam height is to be understood here signed. The term "edge beam height" is understood to mean the beam height of a beam with maximum aperture starting from the point of intersection of the optical axis with the object plane. This field point does not have to contribute to the transmission of the pattern arranged in the object plane, especially in the case of off-axis image fields.

The subaperture ratio is a signed variable, which is a measure of the field or pupil near a plane in the beam path. By definition, the subaperture ratio is normalized to values between -1 and +1, where the subaperture ratio is zero at each field level, and at a pupil level, the subaperture ratio jumps from -1 to +1 or vice versa. An absolute subaperture ratio of 1 thus determines a pupil plane.

Near-field levels thus have subaperture ratios that are close to 0, while near-pupal levels have subaperture ratios that are close to 1 in terms of magnitude. The sign of the subaperture ratio indicates the position of the plane in front of or behind a reference plane.

In the projection beam path, a first manipulator element ME1, which is transparent to the projection radiation, of a wavefront manipulation system is arranged with a small finite spacing immediately behind the object plane OS. The manipulator element essentially has the shape of a rectangular plate whose longer side extends in the x-direction and whose shorter side extends in the y-direction (scan direction). The plate sits completely outside the optical axis at a distance next to the optical axis (off-axis arrangement). As can be seen in the axial plan view shown on the left, the manipulator element is dimensioned and arranged such that the footprint (footprint FP, shown in dashed lines) of the projection radiation, which is substantially rectangular in this region, lies on the transmittable region of the manipulator element (cf. 3 ). With the aid of an adjusting device, not shown, the surface shape and / or refractive index distribution of the first manipulator element formed first manipulator surface can be reversibly changed in such a way that in the x-direction over the optically used range of time several maxima and adjacent minima an optical path length change Projection radiation can be generated. The curves of the maxima and minima are shown schematically with solid lines running in the y-direction.

The distance between the effective manipulator surface and the object plane measured parallel to the optical axis is so small that a subaperture ratio SAR of less than 0.1 is present at the location of the manipulator surface and the condition SAD / D _FP <0.1 applies.

The first manipulator element ME1 is arranged between the object plane OS and the first refractive lens L1 of the projection lens, closer to the object plane than to this lens.

In projection lenses of this type, there are further positions in which, alternatively or in addition to the object-oriented manipulator element, a manipulator element can be arranged in the projection beam path. A position is geometrically located between the prism carrying the folding mirrors FM1, FM2 and the concave mirror CM, close to the folding mirrors FM1, FM2. The illustrated position of the first manipulator element ME1 'is in direct optical proximity to the first intermediate image IMI1 immediately behind it, and furthermore in direct optical proximity to the second intermediate image IMI2 immediately before this intermediate image. The subaperture ratio SAR is less than 0.2 in both cases, the condition SAD / D _FP <0.2 applies in each case. A position is preferred in which the partial beam path leading from the object plane to the concave mirror is spatially separate from the second partial beam path after reflection at the first deflection mirror FM1, which leads from the concave mirror to the image plane via the second folding mirror FM2. The first partial beam path is created on Location of the manipulator, a first footprint FP1, while in the second part of the beam path, a second footprint FP2 is present. The footprints are located diametrically opposite each other with respect to the optical axis OH, as the detail below shows on the left.

In the illustrated position, it is now possible to arrange a manipulator element so that it acts only in one of the partial beam paths, ie either only in the first partial beam path (between object plane and concave mirror) or only in the second partial beam path (between concave mirror and image plane). It is also possible to attach a manipulator element at this position, which is irradiated in different directions and acts in both partial beam paths.

Similar to the position immediately behind the object plane, it is possible to set a desired wavefront change, which alternately alternates between maximum values and minimum values in the x direction, so that the optical path length change can be assigned a characteristic period in the x direction.

In 17 is an embodiment of a projection lens 1700 which is equipped with a first manipulator element ME1 arranged in the field of projection in the projection beam path. The optical design of the projection lens (without manipulator element) corresponds to that of 13 of the patent US 7,446,952 B2 , which is incorporated herein by reference. The projection objective has a first, refractive objective part OP1, which, starting from the object in the object plane OS, generates a first intermediate image IMI1. This is imaged by means of a catoptric second objective part OP2 into a second intermediate image IMI2. The second objective part consists of two concave mirrors CM1, CM2, whose mirror surfaces face each other and are arranged close to the field (near an intermediate image, away from the pupil plane lying therebetween). A refractive third objective part OP3 images the second intermediate image into a third intermediate image IMI3, which in turn is imaged with a refractive fourth objective part OP4 with a strong reduction to the final image in the image plane IS. All lens parts have a common, straight (not folded) optical axis OA (in-line system). More details can the US 7,446,952 B2 are taken and are therefore not specified here in detail.

The intermediate field plane, in which the third intermediate image IMI3 is smaller than the object, is relatively easily accessible between the last object-side concave lens of the third objective part OP3 and the first object-side convex lens of the fourth objective part. Here, a first manipulator element ME1 can be inserted into the projection beam path without danger of collision with the lens of the basic structure. Optionally, a changing device can be provided for exchanging and exchanging manipulator elements in the projection beam path. The first manipulator element ME1 can have the rectangular shape shown in the axial schematic detail view with a long side in the x direction and a short side in the y direction and can be arranged outside the optical axis OA. The dashed line footprint FP of the projection radiation is here fully in the usable area of the first manipulator element. By means of a suitable adjusting device, it is possible to set an optical effect varying in the x-direction repeatedly between maximum values and minimum values, which corresponds to a varying optical path length change of the projection radiation. Alternatively or additionally, the projection lens may also have at other places a corresponding manipulator element, for example, directly behind the object plane OS in a field-near position. It would also be possible, in addition to the first manipulator element shown ME1, which is located in the transmission direction behind the third intermediate image, yet another manipulator element to install, which is also in immediate optical proximity to the third intermediate image in front of this (see 9 ).

Exemplary embodiments of further projection objectives with rectilinear optical axis (in-line systems) and several (exactly two) intermediate images are in the WO 2005/069055 A2 , the disclosure of which is incorporated herein by reference. These can also be equipped with one or more field-near manipulator elements of a wavefront manipulation system of the type described here. For example, in one of the projection lenses of the local 30 to 32 or 36 to 38 a field-close dynamically varying manipulator element may be arranged immediately behind the object plane between this and the first local lens in a position in which the Subaperturverhältnis SAR is less than 0.2 and / or the condition SAD / D _FP <0.2 applies.

Manipulator elements of wavefront manipulation systems of the type described herein can operate on different principles. In the 18 to 21 By way of example, manipulators are shown which, as an alternative or in combination, within a projection objective of the type shown here (eg. 16 and 17 ) or other suitable projection lenses in wavefront manipulation systems.

The manipulator 1850 in 18 has two transparent plates P1, P2, which can be installed with their plate surfaces in each case perpendicular to the optical axis OA of the projection lens in a near-field position with SAR <0.2, for example, directly behind the object plane OS. The plates may have a rectangular shape with an extent that can be several times greater in the x-direction than in the perpendicular thereto y-direction of the projection lens. The entrance side of the first panel P1 facing the object plane OS is flat. Similarly, the image plane facing exit side of the second plate P2 is flat. The facing plate surfaces each have a wave-like surface shape, wherein the wave troughs and wave crests are parallel to the y-direction and in the x-direction, the plate thickness d varies periodically. The wave pattern on the exit side of the first plate P1 and the wave pattern of the entrance side of the second plate P2 are complementary to each other with respect to the x-direction measured characteristic period and the amplitude of the thickness variation. The plates can be displaced relative to each other in the x-direction, which is achieved in the embodiment in that the first plate P1 is permanently installed and the second plate P2 by means of an adjusting device DR1 parallel to the x-direction is movable. But it can also be performed both plates movable.

Each of the plates P1, P2 is made of a material with refractive index n (eg quartz glass or calcium fluoride) and, in the case of vertical transmission, causes an optical path length difference ΔOPL = Δd · n-1, where Δd is the change in the thickness d of the irradiated optical element in the z direction.

The surface of this manipulator effective for generating a wavefront change results from the difference in the effects of the two plates. If these are moved to a neutral position, in which the "wave crests" of the first plate are exactly aligned with the "wave troughs" of the second plate (ie lie one behind the other in the z direction), the effects of the two plates compensate each other in such a way that the Overall effect of a plane plate arises. Relative displacement of the second plate relative to the first plate perpendicular to the transmission direction from the zero position results in a wavefront change that varies periodically in the x direction and alternates between maximum and minimum values, whereby the strength of the resulting wavefront change can be adjusted continuously via the displacement path (compare also EP 0 851 304 A2 ). The effective manipulator surface lies in the area of the gap formed between the plates. This produces the desired wavefront effect, which varies repeatedly across the transmission direction.

The use of a single such two-plate manipulator may be sufficient in some cases, especially in those cases where only basic deformations need to be adjusted. To increase the flexibility and variety of adjustable wavefront changes, two or more plate pairs connected in series may be provided, as indicated by the second manipulator shown in dashed lines. Several series-connected manipulators can be designed so that different directions of displacement perpendicular to the optical axis are possible, for. B. mutually perpendicular directions of displacement. As a result, more flexible manipulator functions can be realized.

The 19 and 20 show schematically sections through another embodiment of a manipulator element 1950 which is in 19 in a neutral position (zero position) and in 20 is shown in sections in a functional position, in which a multiple change between minimum values and maximum values of the optically effective thickness or an optical Weglängenänderung are set in the x direction. The manipulator element uses principles that are for other purposes and in another embodiment in the patent US 7,830,611 B2 the applicant are described. The disclosure of this document is incorporated herein by reference.

The manipulator element 1950 has a multi-layered structure. In a frame-shaped holder 1952 is a relatively thick, rigid transparent plane plate 1955 recorded whose thickness z. B. can be in the cm range. At a distance to this plate is another plane plate 1960 recorded, which is much thinner than the relatively torsion-resistant plate 1955 , The thickness can z. B. in the range of one to 2 mm. Between the plates 1955 . 1960 remains a plane-parallel gap, in the ready-to-use manipulator transparent to the projection radiation liquid 1970 is filled. The thickness of the gap is usually low, z. B. less than 1 mm, in particular less than 10 microns. The liquid reservoir between the plates is connected to a pressure device 1980 connected, via which the liquid pressure of the liquid in the intermediate space z. B. can be set to a constant value. The liquid and the transparent materials of the plates have a very similar refractive index, wherein a ratio between the refractive index of the plates and the refractive index of the liquid is preferably between 0.99 and 1.01.

The entire arrangement is in axial plan view (parallel to the z-direction) rectangular and slightly larger than the area which is irradiated at near-field arrangement of the manipulator in the area of the "footprint". On the side of the thin plate 1960 are outside the radiopaque region at regular intervals to each other on the longitudinal sides of the manipulator pairwise opposed actuators 1990 appropriate. It may, for example, be piezoelectrically actuated actuators. The actuators are designed such that they can engage with a pressing force acting on the outside of the thin plate substantially perpendicular to the plate surface of the thin plate (cf. 20 ). By suitable control of the actuators can be a wave-shaped deformation of the thin plate 1960 set with predetermined amplitude, wherein the "wave troughs" parallel to the y-direction and the measured in the x-direction characteristic period length P _CHAR the wavy deformation can be set to different values by selecting the respectively driven actuators.

If this arrangement is parallel to the z-direction, ie perpendicular to the warp-resistant thicker plate 1955 irradiated, so the overall arrangement in the acts in 19 shown neutral position as a plane-parallel plate, so that over the entire illuminated cross section, the same optical path length change ΔOPL is introduced. If an optical path length change which varies periodically in the x-direction is to be introduced, corresponding actuators are activated so that the thin plane plate at the corresponding points is activated 1960 is pressed in the direction of the liquid. In these areas, local minima of the irradiated total thickness d are formed, while local maxima arise between the activated actuators. In a first approximation, a wavelength variation which varies periodically in the x direction can thus be set, similar to those associated with the 2 and 3 was explained. By controlling different groups of actuators, it is possible to set different "wavelengths" or different characteristic periods P _CHAR in the x direction.

If necessary, modifications of this arrangement can be used in which actuators are arranged not only on the edge outside of the radiopaque region, but also within the radiopaque region (see. 10 . 11 out US 7,830,611 B2 ).

The oblique perspective schematic representation in 21 shows a further embodiment of a manipulator 2100 which may be used in the context of a wavefront manipulation system of the type described herein. At the edge of the deformable thin rectangular plan plate 2155 are groups of actuators 2190 alternately mounted on opposite sides of the plane plate. By appropriate group actuation of actuators, the plane plate can be wavy deformed, in which case the wave troughs and wave crests are parallel to the y-direction and the wave peaks or valleys in the x-direction at a distance from each other, the characteristic period P _{CHAR of} the corresponding application equivalent.

Claims (18)

Projection objective (PO) for imaging a pattern arranged in an object plane (OS) of the projection lens into an image plane (IS) of the projection objective by means of electromagnetic radiation having a working wavelength λ <260 nm, comprising: a plurality of optical elements with optical surfaces arranged in a projection beam path between the object plane (OS) and the image plane (IS) are arranged such that a pattern arranged in the object plane can be imaged by means of the optical elements into the image plane, and a wavefront manipulation system (WFM) for the dynamic influencing of the wavefront of the object plane Projection radiation extending image plane, wherein: the wavefront manipulation system comprises a first manipulator (MAN1), the one arranged in the projection beam path first manipulator surface (MS1) and a first adjusting device (DR1) for reversible change of surface shape and / or refractive index distribution of the first manipulator atorfläche has; the first manipulator is configured such that a visually used area of the first manipulator surface having an effective diameter D _FP across a number N _MAX> 1 of maxima and a number N _MIN> 1 of minima of an optical path length of the projection radiation in accordance with a characteristic period P _CHAR = D _FP / ((N _MAX + N _MIN / 2); and the first manipulator surface is arranged at a finite first distance (D1) to a nearest field plane of the projection lens in optical proximity of that field plane, such that each of one Field spot of the field plane outgoing beam at the first manipulator surface a subaperture with a Subaperturdurchmesser SAD and illuminates at the first manipulator surface the condition SAD / D _FP <0.2, wherein the Subaperturdurchmesser SAD is the diameter of a projecting radiation from a single field point of the nearest field plane bundle of projection radiation , Projection objective according to claim 1, wherein the wavefront manipulation system has only the first manipulator (MAN1) in the optical proximity of the field plane and the finite first distance (D1) is dimensioned such that the condition on the first manipulator surface upon activation of the first manipulator 0.25 · NA _M ² <SAD / P _CHAR <0.48 is satisfied, where NA _{M is} the numerical aperture of the projection radiation at the first manipulator surface. Projection objective according to claim 1 or 2, wherein the nearest field plane is the object plane (OS) of the projection objective (PO). Projection objective according to one of the preceding claims, wherein no optical surface with refractive power is arranged between the object plane (OS) and the first manipulator surface (MS1), so that the numerical aperture NA _{M of} the projection radiation at the first manipulator surface is equal to the object-side numerical aperture NA _O , Projection objective according to one of the preceding claims, wherein the wavefront manipulation system comprises in addition to the first manipulator (MAN1) a second manipulator (MAN2) having a second manipulator surface (MS2) arranged in the projection beam path and a second actuator device for reversibly changing surface shape and / or Having refractive index distribution of the second manipulator surface; the second manipulator is configured such that a number N _MAX > 1 of maxima and a number N _MIN > 1 of minima of an optical path length change of the projection radiation in accordance with a characteristic period P over an optically used region of the second manipulator surface with an effective diameter D _{FP CHAR} = D _FP / ((N _MAX + N _MIN ) / 2) can be generated; and the second manipulator surface is arranged in a finite second distance (D2) to the nearest field plane of the projection lens in the optical proximity of this field plane such that each beam emanating from a field point of the field plane illuminates a subaperture with a subaperture diameter SAD at the second manipulator surface and at the second manipulator surface the condition SAD / D _FP <0.2 applies. Projection objective according to claim 5, wherein the first manipulator surface (MS1) and the second manipulator surface (MS2) are arranged such that the numerical aperture of the projection radiation at the first manipulator surface is equal to the numerical aperture of the projection radiation at the second manipulator surface. Projection lens according to one of the preceding claims, wherein the projection lens is configured such that between the object plane (OS) and the image plane (IS) at least one real intermediate image (IMI) in the region of an intermediate image plane (IMIS) is generated, wherein the nearest field plane, the intermediate image plane is. Projection objective according to claim 5 and 7, wherein the first manipulator surface (MS1) behind the intermediate image plane (IMIS) and the second manipulator surface (MS2) in front of the intermediate image plane (IMIS) is arranged. A projection lens according to any one of claims 5 to 8, wherein the first distance (D1) is equal to the second distance (D2). Projection objective according to one of claims 5 to 9, wherein the first distance (D1) and the second distance (D2) are dimensioned such that on the first manipulator surface upon activation of the first manipulator and on the second manipulator surface upon activation of the second manipulator the condition 0.012 <SAD / P _CHAR <0.85 is satisfied. Projection objective according to one of claims 5 to 7, wherein the second manipulator surface (MAN2) is arranged directly behind the first manipulator surface (MAN1), wherein the numerical aperture of the projection radiation at the first manipulator surface is equal to the numerical aperture of the projection radiation at the second manipulator surface and the first distance (D1) is smaller than the second distance (D2), so that the Subaperturdurchmesser differ on the manipulator surfaces, wherein for the first manipulator surface the condition 0.25 · NA _M ² <SAD / P _CHAR <0.8 And for the second manipulator area the condition SAD / P _CHAR <1.5 applies. A projection lens according to any one of the preceding claims, wherein the projection objective comprises an off-optical axis (OA) effective object field having an aspect ratio greater than 2: 1 between a longer and a shorter side, the optically used portion having a rectangular shape with an aspect ratio greater than 2: 1 and the first manipulator acts parallel to the longer side in such a way that the first manipulator can generate in this direction several maxima and several minima of the optical path length change of the projection radiation. Projection exposure method for exposing a radiation-sensitive substrate to at least one image of a pattern of a mask, comprising the following steps: Providing a pattern between an illumination system and a projection objective of a projection exposure apparatus such that the pattern is arranged in the region of the object plane of the projection objective; Holding the substrate such that a radiation-sensitive surface of the substrate in the region of an object plane to the optically conjugate image plane of the projection lens is arranged; Illuminating an illumination region of the mask with an illumination radiation provided by the illumination system having an operating wavelength λ <260 nm; Projecting a part of the pattern lying in the illumination area onto an image field on the substrate with the aid of the projection lens, all the beams of the projection radiation contributing to image formation in the image field forming a projection beam path, influencing the wavefront of the projection radiation extending from the object plane to the image plane by activating a first manipulator, which has a first manipulator surface arranged in the projection beam path and a first actuating device for reversibly changing the surface shape and / or refractive index distribution of the first manipulator surface; wherein the first manipulator surface is arranged in a finite first distance D1 to a nearest field plane of the projection lens such that each beam originating from an object point of an object field illuminates a subaperture SA with a subaperture blade SAD on the first manipulator surface and the condition SAD / on the first manipulator surface D _FP <0.2 applies; and wherein the first manipulator is controlled in such a way that, over an optically used region of the first manipulator surface with an effective diameter D _FP , a number N _MAX > 1 of maxima and a number N _MIN > 1 of minima of an optical path length change of the projection radiation according to a characteristic Period P _CHAR = D _FP / ((N _MAX + N _MIN ) / 2), and where the subaperture diameter SAD is the diameter of a beam of projection radiation emanating from a single field point of the nearest field plane. A projection exposure method according to claim 13, wherein a first exposure and a second exposure are successively performed, wherein a field dependent distortion in the second exposure is adjusted by activation of the first manipulator to a structure generated during the previous first exposure such that the structures produced in successive exposures lie with higher coverage accuracy than without activation of the first manipulator. A projection exposure method according to claim 13 or 14, wherein the wavefront is influenced to set different values of distortion for different field points. A projection exposure method according to any one of claims 13 to 15, wherein a projection lens according to any one of claims 1 to 12 is used. A projection exposure apparatus for exposing a radiation-sensitive substrate arranged in the region of an image surface of a projection objective to at least one image of a pattern of a mask arranged in the region of an object surface of the projection objective, comprising: a light source (LS) for emitting ultraviolet light; an illumination system (ILL) for receiving the light of the light source and for forming illumination radiation directed to the pattern of the mask; and a projection objective (PO) for imaging the structure of the mask onto a photosensitive substrate; wherein the projection lens according to any one of claims 1 to 12 is configured. A projection exposure apparatus according to claim 17, wherein the projection exposure apparatus comprises a central controller for controlling functions of the projection exposure apparatus, wherein the control means is associated with a control module for driving the wavefront manipulation system (WFM) and one or more manipulators via the control module in coordination with other control signals the operation of the projection exposure system can be controlled.

Images