DE102011077234A1 - Extreme UV mirror arrangement for use as e.g. pupil facet mirror arranged in region of pupil plane of e.g. illumination system, has multilayer arrangement including active layer arranged between entrance surface and substrate - Google Patents

Abstract

The arrangement (100) has mirror elements (110-112) connected to each other for jointly forming a mirror surface (115) of the arrangement. A multilayer arrangement (130) is applied on a substrate (120), and includes reflective effect with respect to radiation from an extreme UV range (EUV). The multilayer arrangement includes an active layer (140) arranged between a radiation entrance surface and the substrate, where thickness (z) of the layer is altered by an action of an electric field. A provision is made of an electrode arrangement for generating the field acting on the layer. The active layer is a pulsed laser deposition layer. The mirror surface is a flat mirror surface, plane mirror surface or curved mirror surface such as convex mirror surface, concave mirror surface and cylindrical mirror surface. The substrate is a metal substrate, silicon substrate and a glass substrate. An independent claim is also included for a method for operating an optical system.

Description

BACKGROUND

Technical area

The invention relates to an EUV mirror arrangement according to the preamble of claim 1, to an optical system with an EUV mirror arrangement according to the preamble of claim 16 and to a method for operating an optical system according to the preamble of claim 21. A preferred field of application is the EUV microlithography. Other applications include EUV microscopy and EUV mask metrology.

Description of the Prior Art

For the production of semiconductor components and other finely structured components, predominantly microlithographic projection exposure methods are used today. 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 (bearing) 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 surface 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 coated with a radiation-sensitive layer.

The pattern is illuminated by means of an illumination system which shapes from the radiation of a primary radiation source an illumination radiation directed onto the pattern, which is characterized by specific illumination parameters and impinges on the pattern within an illumination field of defined shape and size. Within the illumination field there should be a given local intensity distribution, which should normally be as uniform as possible.

As a rule, different illumination modes (so-called illumination settings) are used, which can be characterized by different local intensity distributions of the illumination radiation in a pupil surface of the illumination system, depending on the type of structures to be imaged. In this way, a specific illumination angle distribution or a specific distribution of the incident intensity in the angular space can be specified in the illumination field.

In order to be able to produce ever finer structures, different approaches are pursued. For example, the resolution of a projection lens can be increased by increasing the image-side numerical aperture (NA) of the projection lens. Another approach is to work with shorter wavelengths of electromagnetic radiation.

Attempting to improve the resolution by increasing the numerical aperture may present problems in that as the numerical aperture increases, the depth of focus (DOF) achievable decreases. This is disadvantageous because, for example, for reasons of achievable flatness of the substrates to be patterned and mechanical tolerances, a depth of field of the order of at least 0.1 nm is desirable.

For this reason, among other things, optical systems have been developed which operate at moderate numerical apertures and which achieve the increase in resolving power essentially by the short wavelength of the extreme ultraviolet (EUV) electromagnetic radiation used, in particular with working wavelengths in the range between 5 nm and 30 nm. In the case of EUV lithography with working wavelengths around 13.5 nm, for example, at image-side numerical apertures of NA = 0.3, theoretically a resolution of the order of 0.03 μm can be achieved at typical depths of field in the order of about 0.15 μm.

Radiation from the extreme ultraviolet range can not be focused or guided by refractive optical elements because the short wavelengths are absorbed by the known optical materials transparent at higher wavelengths. Therefore, mirror systems are used for EUV lithography. A reflecting mirror (EUV mirror) which is reflective for EUV radiation typically has a substrate on which one is reflective of extreme ultra violet (EUV) radiation effective multi-layer arrangement (multilayer) is applied, which has many pairs of layers with alternately low-refractive and high-refractive layer material. Layer pairs for EUV mirrors are often built up with the layer material combinations molybdenum / silicon (Mo / Si) or ruthenium / silicon (Ru / Si).

In order to ensure the best possible uniformity of the lithographic image, the aim is generally to produce as uniform as possible an intensity distribution in the illumination field illuminated by the illumination system. Furthermore, the aim is normally to approximate as closely as possible to the desired spatial intensity distribution the desired spatial intensity distribution of the illumination beams in the pupil surface of the illumination system or to minimize deviations from a desired spatial intensity distribution as much as possible. These requirements must not only be satisfied by the lithographic optical system at the time of its delivery, but must be maintained throughout the life of the optical system without significant change. While in the former case, any deviations are largely due to design residuals and manufacturing defects, changes over the lifetime are often caused mainly by aging.

In optical systems for lithography with ultraviolet light from the deep or very deep ultraviolet range (DUV or VUV), possibly resulting non-uniformities can generally be compensated by controllable mechanical compensators (cf., for example, US Pat. US 2008/113281 A1 or US 7,545,585 B2 ).

In optical systems for EUV microlithography such compensators are significantly more difficult to realize, among other things for geometrical reasons. For example, there is often no optically conjugate intermediate plane of the object plane to the object plane of the projection objective in which the field homogeneity can be easily corrected. The WO 2010/049020 A1 discloses possibilities for correcting the illumination intensity distribution and the illumination angle distribution in the illumination field of an EUV illumination system. Other correction devices are in US 2003/0063266 A1 . EP 1 349 009 A2 . US 2008/0165925 A1 or WO 2009/135576 A1 disclosed.

TASK AND SOLUTION

It is an object of the invention to provide an EUV mirror assembly and a thus equipped optical system, the z. B. in a microlithography projection exposure system can be used to ensure a high fidelity and stability of the illumination intensity in the field and pupil with respect to a given distribution and thus the lithographic imaging quality throughout the lifetime of the projection exposure system.

To achieve this object, the invention provides an EUV mirror assembly having the features of claim 1. Furthermore, an optical system with an EUV mirror arrangement having the features of claim 16 and a method for operating such an optical system with the features of claim 21 are provided.

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

The EUV mirror assembly has a plurality of juxtaposed mirror elements that together form a mirror surface of the mirror assembly. The element mirror surface of a mirror element forms a fraction of the entire mirror surface. Mirror elements can, for example, be arranged in rows and columns essentially in a surface-filling manner or completely in a surface-filling manner or at a mutual distance next to one another. The mirror elements can be separate from one another and, if appropriate, separated by intermediate spaces, can be mirror elements which can be mounted on a carrier structure.

It is also possible for the mirror elements to have a common substrate and for the multilayer layer arrangement to have continuous layers over the entire usable area. In this case, the electrode arrangement may have one or more structured electrodes in order to be able to independently apply an electric field of predeterminable strength to the regions of an active layer assigned to the individual mirror elements.

A multilayer laminate has a plurality of pairs of layers, each comprising a layer of relatively high refractive layer material and a layer of (relatively) low refractive layer material. Such pairs of layers are also referred to as "bilayer" or "bilayer". A layer arrangement with many pairs of layers acts in the manner of a "Distributed Bragg Reflector". The layer arrangement simulates a crystal whose lattice planes leading to Bragg reflection are formed by the layers of the material with the lower real part of the refractive index. The optimum period thickness of the layer pairs is determined for a given wavelength and for a given angle of incidence (range) by the Bragg equation and is usually between 1 nm and 10 nm.

In addition to the two layers of relatively high-refractive or relatively low-refractive-index material, a layer pair may also have one or more further layers, for example an intermediate barrier layer for reducing interdiffusion between adjacent layers.

The multilayer layer arrangement of a mirror element has at least one active layer which is arranged between a radiation entrance surface and the substrate and which consists of a piezoelectrically active layer material. Due to this material property of the active layer material, the layer thickness of the active layer can be changed by applying an electrical voltage. For each active layer, an electrode arrangement is provided for generating an electric field acting on the active layer. As a result, the active layers of the mirror elements can be activated independently of one another as required and thus changed with respect to their layer thicknesses. As a consequence, the reflection properties of the EUV mirror arrangement can be influenced locally differently over the mirror surface.

In this case, the inverse piezoelectric effect is exploited, in which the active layer material reversibly deforms under the action of an electric field. The crystalline active layer material undergoes no phase transformation, but there is only a shift of positive and negative charge centers within the crystal structure of the electrically non-conductive active layer material instead.

An electrode of the electrode assembly may be in physical contact with the active layer. It is also possible to arrange one or more electrodes at a distance from the active layer to be influenced, as long as the electric field can penetrate the material-filled or material-free interspace up to the active layer. Thus, one or more layers of the layer arrangement may also lie between an electrode and the active layer. In particular, to generate the electric field, a voltage can be applied between an outer, substrate-distant layer of a layer arrangement and an inner, substrate-near layer of a layer arrangement, wherein in each case many pairs of layers are located between the electrode layers and the active layer.

The lateral resolution (spatial resolution) of the influence is dependent on the lateral dimensions of the element mirror surfaces of the individual mirror elements. Lateral dimensions can vary depending on the application z. B. in the range of one or more millimeters or centimeters. Smaller lateral dimensions, eg. B. between 1 micron and 900 microns, are also possible. In the mirror surface of the mirror arrangement more than 10 or more than 100 or more than 1000 independently controllable mirror elements can be provided. It may also be sufficient to provide less than 10, for example, only two or three or four separately controllable mirror elements. This can be z. B. be useful for adjustment purposes or calibration purposes.

At least one active layer is integrated into the multi-layer layer arrangement, the layer thickness of which can be selectively changed by electrical activation of the associated electrode arrangement.

There are different possibilities for arranging the at least one active layer with respect to the layer pairs of the multilayer layer arrangement.

In some embodiments, the multilayer stack has a first layer group having a first number N1 of layer pairs disposed between the radiation entrance surface and the active layer and a second layer group having a second number N2 of layer pairs disposed between the active layer and the substrate, the numbers N1 and N2 of layer pairs of the first layer group and the second layer group are selected such that, for at least one angle of incidence of the radiation impinging on the radiation entrance surface, the first layer group transmits a portion of the incoming radiation through the active layer to the second layer group and reflects through the multilayer stack Radiation includes a first portion reflected by the first layer group and a second portion reflected by the second layer group.

In general, the first and the second layer group each have multiple pairs of layers, for. B. each 10 or more, or 15 or more pairs of layers.

In this case, both the substrate-distant first layer group and the substrate-near second layer group contribute to the overall reflectivity of a mirror element. By the intermediate active layer, the distance of the layer groups (measured perpendicular to the layer surface) can be changed by applying an external voltage. The layer structure of the first layer group is preferably chosen such that, for the considered angle of incidence or angle of incidence range, a constructive interference of the radiation components (partial waves) reflected at the individual boundary surfaces within the first layer group occurs. The same applies preferably also to the layers of the second layer group. The intermediate active layer introduces an optical path length difference or a phase shift between the radiation components reflected at the first group of layers and the radiation components reflected at the second layer group. By applying an external voltage, the amount of phase shift can be varied steplessly.

If, for example, the introduced phase shift in the absence of an electric field is substantially one wavelength or an integral multiple of the wavelength of the electromagnetic radiation, then the first component and the second component of the reflected radiation interfere constructively with each other, so that the total reflectivity of the mirror element is in the range for the Incidence angle range applicable maximum reflectivity can be. If, on the other hand, the layer thickness of the active layer is adjusted so that the phase shift between the first and the second component is in the region of half a wavelength or in the region of three half wavelengths, etc., a destructive interference takes place between the first component and the second component , so that the total reflectivity resulting from the first portion and the second portion is lower than the maximum reflectivity possible with the layer groups.

For example, if the change in the optical path length in single pass through the active layer is one quarter of the operating wavelength, and the active layer is placed at a suitable depth such that the first and second portions are substantially equal in intensity, the reflection may be substantially complete be suppressed. Between these extremes (maximum reflectivity of a mirror element and complete suppression of the reflection of a mirror element), numerous variants result, which are explained in more detail in connection with the exemplary embodiments.

The integrated between the first and the second layer group active layer acts like an integrated Fabry-Perot interferometer (etalon) with electrically adjustable distance between its reflective interfaces.

In many cases it is not necessary or required to vary the reflectivity of a mirror element between maximum reflection and complete reflection suppression. Often it is sufficient if the reflectance of a mirror element is only varied by a maximum of 20% or a maximum of 10%. In some embodiments, in the absence of an electric field, the active layer has a layer thickness that is selected such that a reflectivity of the multilayer layer arrangement can be changed by applying an electric field by a maximum of 20%, in particular a maximum of 10%, for a reference incidence angle of the incident radiation is.

Preferably, exactly one active layer is provided between two adjacent layer groups each having a plurality of layer pairs. This can u. a. the Fehlbeschichtungsrisiko due to manufacturing tolerances are kept small. In addition, this results in only a small complexity between the transmitted and the absorbed radiation components. However, a multi-layer arrangement may also have more than one active layer arranged between two adjacent layer groups with a plurality of layer pairs, which serves for the controllable phase shift between the reflected radiation components of these layer groups. For example, two or three such active layers may be provided, between which then also layer groups with several layer pairs lie.

When selecting active layer materials for such an integrated active layer, it should be noted that, on the one hand, the layer material only has relatively low absorption for the radiation to be transmitted to the second layer group and, on the other hand, enables a sufficiently strong "lift" of the layer thickness for controlling the phase shift. In some embodiments, the active layer material consists essentially of barium titanate (BaTiO ₃ ).

In general, layer materials which have a relatively low absorption (low extinction coefficient or imaginary part of the complex refractive index) in the selected EUV wavelength range and at the same time show a relatively strong piezoelectric effect are preferred for the piezoelectrically active layer. to be able to produce sufficiently strong layer thickness changes. The piezoelectric-active layer material may be a material having a perovskite structure, which exhibits a relatively strong piezoelectric effect. In particular, the piezoelectrically active layer material may be selected from the group: Ba (Sr, Zr) TiO ₃ , Bi (Al, Fe) O ₃ , (Bi, Ga) O ₃ , (Bi, Sc) O ₃ , CdS, (Li , Na, K) (Nb, Ta) O _3, Pb (Cd, Co, Fe, In, Mg, Ni, Sc, Yb, Zn, Zr) (Nb, W, Ta, Ti) O _3, ZnO, ZnS or contain at least one material of this group in combination with at least one other material. Here, the notation (A, B) indicates that in a certain lattice position of the crystal structure, an element or ion of type A or an element or ion of type B may be present.

In other embodiments, the multilayer stack includes a plurality of active layers of a piezoelectric active layer material, wherein the active layers are each disposed alternately with layers of a non-piezoelectric active layer material. In this case, the layers arranged between the active layers preferably consist of an electrically conductive layer material, so that these layers can simultaneously serve as electrode layers for the active layers arranged in each case between them. The active layer material may be either the relatively high refractive or the relatively low refractive layer material as compared to the non-active layer material. An active layer material with a relatively high absorption can be used advantageously as an absorber layer.

In this embodiment, by applying an electric field to the active layers, an optionally stepless variation of the layer period within the multilayer layer arrangement can be produced. In this case, the layer period denotes the distance between the delimiting outer boundary surfaces of a layer pair measured perpendicular to the layer surface. Since for a given operating wavelength and a given angle of incidence only certain layer periods lead to a full constructive interference and thereby to maximum reflectance, the reflectivity of the multilayer layer arrangement of the mirror element at the operating wavelength can be varied steplessly by the variation of the layer period. Furthermore, there is an effect on the phase of the reflected radiation, so that a spatially resolving wavefront interference is possible.

The detuning or change of the layer period can also be used to adjust the reflectivity to a possibly deviating from the desired central wavelength, so that z. B. a compensation of variations of the source spectrum or the spectral transmission of the overall optical system can be performed. Alternatively or additionally, an adaptation to intentionally or unintentionally changed angles of incidence on the mirror is possible.

In multi-layer arrangements having a plurality of active layers, it is particularly important to ensure that the active layer material has low absorptions for the radiation used. In this context, it has proven to be advantageous if the active layer material consists predominantly or exclusively of a ceramic material of the type (Li, Na, K) (Nb, Ti) O ₃ . Such materials are for. In EP 2 050 726 A2 described. These materials may also be beneficial for health reasons because they contain no lead (Pb).

In particular, the active layer material may contain or consist of a material from the group of potassium niobate (KNbO ₃ ), lithium niobate (LiNbO ₃ ), PbNb ₂ O ₆ and sodium potassium niobate (Na _0.9 K _0.1 NbO ₃ ), or a combination thereof Materials. Among other things, these materials are characterized by particularly low absorption in the EUV range.

It is also possible to design the EUV mirror arrangement such that a spatially resolving phase correction of the wave front of the incident radiation is possible substantially without influencing the spatial distribution of the reflectivity. Such embodiments can be used in particular as a mirror in an EUV projection objective. In some such embodiments, the multilayer stack has a third layer group having a third number N3 of layer pairs disposed between the radiation entrance surface and the active layer, wherein the third number N3 is selected such that for at least one angle of incidence the radiation impinging on the radiation entrance surface is third Layer group substantially completely reflects or absorbs the incoming radiation before reaching the active layer. It can z. B. at least 20 or at least 30 or at least 40 pairs of layers may be provided. Typically, there are less than 70 or fewer than 60 pairs of layers.

The reflectivity (or reflectance) of the mirror element is determined in this case almost exclusively by the layer structure of the third layer group. This can be done with the help of active Layer can be raised or lowered relative to the substrate by applying an electrical voltage without tilting perpendicular to the layer surface as a whole.

The layer thicknesses of the individual layers of the layer pairs and possibly also of the active layer are generally of the order of a few nanometers. In order to keep the influence of boundary surface roughness on the optical effect of the mirror elements as low as possible, it is provided in preferred embodiments to apply the active layer and / or possible electrode layers by means of laser beam evaporation (Pulsed Laser Deposition, PLD), so that the active layer and / or or an electrode layer is present as a PLD layer. With the aid of laser beam evaporation, it is possible to produce very thin layers with low surface roughness. If necessary, it is also possible to produce single-crystalline piezoelectric layer materials with high piezoelectric coefficients, the surface of which can then be used without polishing as a contact surface for further layers.

Preferably, at least the layer to which the active layer is applied is produced as a crystalline (non-amorphous) layer, in particular by means of laser beam evaporation. This facilitates crystal growth of the active layer. In favorable cases, this can grow epitaxially to an underlying crystalline layer. In some embodiments, most or all of the layers lying between the substrate and the active layer are crystalline.

In some embodiments, the electrode assembly of a mirror element has a first electrode layer and a second electrode layer, and the active layer is disposed between these electrode layers. As a result, it can be achieved that the electric field penetrates the active layer essentially perpendicular to the layer surface, as a result of which the layer thickness changes can be generated particularly effectively. An electrode layer may consist of a metallic layer material or of a semimetal, such as silicon. Electrode layers of silicon may be arranged, for example, in alternation with active layers of a piezoelectrically active layer material.

In some cases, it has proven to be advantageous if an electrode layer consists of an electrically conductive ceramic material, for example of SrRuO ₃ or aluminum nitride (AlN). The use of electrically conductive ceramic materials as electrode material in combination with ceramic active layer materials allows to keep the lattice mismatch at the interfaces between the electrode layer and active layer material low, whereby layer stresses in the region of the interfaces and thus the risk of delamination kept low and thus the life of the Layer arrangement can be improved.

The invention also relates to an optical system with at least one EUV mirror arrangement. In particular, the optical system may be an illumination system of a microlithography projection exposure apparatus. The EUV mirror arrangement can be arranged in the beam path of the illumination system between a light source and an illumination field to be illuminated in or near a field plane which is optically conjugate to the plane of the illumination field. The EUV mirror arrangement in this case can serve as a field facet mirror. Alternatively or additionally, an EUV mirror arrangement may be arranged in the region of a pupil plane of the illumination system, that is to say in the region of a plane which is Fourier-transformed to the plane of the illumination field. In this case, the EUV mirror assembly may serve as a pupil facet mirror. The optical system may also be a projection objective of a microlithography projection exposure apparatus.

In a method for operating an optical system with at least one such EUV mirror arrangement, the local reflectivity distribution over the mirror surface of the EUV mirror arrangement can be varied depending on the location by selective activation of individual or all active layers. If an EUV mirror arrangement is arranged in the region of a field plane of the optical system, this can influence the illumination intensity distribution in this field plane and in optically conjugate field planes. With an arrangement in the region of a pupil plane, the illumination intensity distribution in the illumination field can be changed as a function of the angle by locally changing the reflectivities.

These and other features will become apparent from the claims but also from the description and drawings, wherein the individual features each alone or more in the form of sub-combinations in an embodiment of the invention and in other fields be realized and advantageous and protectable Can represent versions. Embodiments are illustrated in the drawings and are explained in more detail below.

BRIEF DESCRIPTION OF THE DRAWINGS

1 shows a schematic, oblique perspective view of an embodiment of an EUV mirror assembly in partial section;

2 shows a diagram for explaining the influence of the layer thickness of an active layer on the reflectivity R and the transmittance T of the entire layer arrangement;

3 shows a diagram for the influence of the number of layer pairs in the substrate-distant first layer group on the reflectivity curve with increasing layer thickness of the active layer;

4A shows a diagram of the reflectivity curve in the region of a maximum reflectivity with increasing number of layer pairs in the substrate-second second layer group with a constant number of layer pairs in the first layer group;

4B shows a diagram of the Reflektivitätsverlauf in the range of a first Reflektivitätsmaximums;

5 shows a diagram of the Reflektivitätsverlaufs in the region of a reflection maximum, in which an adjustment range .DELTA.R of the reflectivity of about 2.5% at a layer thickness variation .DELTA.z of 0.127 nm can be achieved;

6 shows a diagram with reflectivity and transmission curve of a multi-layer layer arrangements, which can be used as a beam splitter with variably adjustable transmission;

7 schematically shows a part of an EUV layer arrangement with an activated and a non-activated mirror element;

8th shows a schematic, oblique perspective view of another embodiment of an EUV mirror assembly in partial section;

9 shows a schematic, oblique perspective view of another embodiment of an EUV mirror assembly in partial section;

10 shows in 10A . 10B and 10C various embodiments of structured layer electrodes; and

11 shows optical components of an EUV microlithography projection exposure apparatus with embodiments of EUV mirror assemblies used as field facet mirrors and pupil facet mirrors, respectively.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

In 1 is a schematic, oblique perspective view of an embodiment of an EUV mirror assembly 100 shown in partial section. The mirror assembly has a plurality of juxtaposed mirror elements 110 . 111 . 112 , which in the example case each have a rectangular cross-section. Each mirror element may be referred to as a single mirror and has a rectangular element mirror surface, wherein the element mirror surfaces connect to each other largely seamlessly or with spaces next to each other and together a mirror surface 115 form the mirror assembly. The mirror surface can be flat overall (plane mirror) or curved (eg convex mirror, concave mirror, cylindrical mirror, etc.).

The structure of a mirror element is based on the mirror element 110 explained in more detail. Each mirror element has a substrate 120 , which may for example consist of metal, silicon, a glass, a ceramic material, a glass ceramic or a composite material. On a substrate surface smoothly machined with high precision is a multi-layer coating arrangement which is reflective for radiation from the extreme ultraviolet region 130 applied by suitable coating technique. To produce some or all individual layers z. As the magnetron sputtering, electron beam sputtering or ion beam sputtering can be used. If a crystalline layer structure is desired, z. B. also be coated with laser beam evaporation (PLD).

The multi-layer layer arrangement has a multiplicity of layer pairs (bilayers) 135 , each of which has alternately applied layers of a higher real part refractive index (also called "spacer") layer material and a relatively lower real part refractive index (also called "absorber") layer material. In the example case are relatively thin layers 136 with molybdenum (Mo) as an absorber material alternately with relatively thicker layers 134 applied with silicon (Si) as a spacer material. A layer pair may also contain at least one further layer, in particular an intermediate barrier layer, the z. B. from C, B ₄ C, Si _x N _y , SiC or a composition with one of these materials can and should prevent interdiffusion at the interface. As a result, permanently sharply defined interfaces can be ensured even under radiation exposure.

The layer pairs are grouped into two groups of layers. A substrate-distant, near-surface first layer group 131 has a first number N1 of layer pairs 135 , Between the first layer group and the substrate-remote radiation entrance surface, in the example case, there is still a cap layer 137 applied to protect the underlying layers. The cover layer may, for. Example of ruthenium, rhodium, gold, palladium, Si _x N _y or SiC or contain one of these materials. The free surface of the cover layer forms the beam entry surface.

A substrate-near second layer group 132 has a second number N2 of layer pairs 135 , This second layer group can be applied directly to the substrate surface, but it can also be provided a single- or multi-layer intermediate layer, which acts for example as a smoothing layer. The layers of the near-second second layer group are preferably produced as crystalline layers, in particular by means of Pulsed Laser Deposition (PLD).

Between the first shift group 131 and the second layer group 132 is a single active layer 140 arranged from a piezoelectrically active layer material. The layer thickness z of the active layer can be changed by applying an electric field to the active layer material. For this purpose, a first electrode layer is in direct contact with the active layer between it and the first layer group 142 and a second electrode layer between the active layer and the second layer group 143 arranged. The electrode layers which are in surface contact with the active layer material consist of an electrically conductive layer material and are by electrically conductive connections with a switchable or controllable voltage source 145 connected. The layer thickness z can be varied continuously as a function of the voltage generated by the voltage source between a minimum value z _min (in the absence of an electric field) and a maximum value z _max .

For each of the mirror elements, a corresponding electrode arrangement is provided. The electrode arrangements can be controlled independently of each other, so that for each mirror element, its active layer can be changed independently of the active layers of other mirror elements by applying electrical voltage with respect to its layer thickness.

Both the first shift group 131 , as well as the second layer group 132 is designed with respect to the layer thicknesses of the spacer and absorber layers so that they have a reflective effect for the angle of incidence range with which the mirror arrangement is to be operated. The first number N1 of layer pairs of the near-surface first layer group 131 is chosen so that it reflects only a portion of the radiation entering at the mirror surface by Bragg reflection and another portion of the incoming radiation through the active layer 140 through to the second layer group 132 is transmitted. The second number N2 of layer pairs of the second layer group is selected such that this fraction, which is transmitted to the second layer group, is reflected almost completely (and possibly partially absorbed) by the second layer group.

The one from the second shift group 132 reflected portion is reflected back through the active layer and through the first layer group. Accordingly, the total radiation reflected by the layer arrangement contains a first portion reflected by the first group of layers and a second portion reflected by the second group of layers. The resulting overall reflectivity of the multi-layer arrangement (ie the ratio between reflected and incident intensity, represented by the reflectance or reflectivity R) is determined by interference between those of the first layer group 131 reflected and that of the second layer group 132 reflected partial waves determined. The nature and extent of the interference can be changed separately for each mirror element by changing the layer thickness of the active layer, so that within predetermined setting limits the weighting between shares of destructive interference and constructive interference can be changed. The amount of interference is thereby by the optical angular distance difference (phase difference) between the of the lower second layer group 132 reflected partial waves and that of the upper first layer group 131 reflected partial waves determined. This basic principle will be explained in more detail below with reference to calculated exemplary embodiments.

In the calculated embodiment, barium titanate (BaTiO ₃ ) becomes an active layer material for the active layer 140 used. The first electrode 142 and the second electrode 143 are each formed by layers of SrRuO ₃ . This electrically conductive ceramic exhibits relatively low lattice mismatch when in contact with barium titanate. Alternatively, for example, aluminum nitride (AlN) or another electrically conductive material, such as a metallic material may be used. The layer pairs 135 consist, as mentioned, of molybdenum as absorber material and silicon as a spacer material.

At least the substrate-side lower electrode layer 143 is available as crystalline layers, it can be generated in particular by pulsed laser deposition (PLD). The crystal surface then serves as a support for the growth of the active layer, which is also a PLD-applied crystalline layer. Due to the low lattice mismatch, an epitaxial growth of a monocrystalline active layer on a monocrystalline lower electrode layer may be possible.

Depending on the layer thickness of the active layer and the coating process used to produce it, its surface may be relatively rough. To improve the growth conditions for subsequent layers, a smoothing layer, e.g. B. of amorphous silicon, are applied, the surface of which can then be smoothly polished by means of ion beam, before the next layer is applied. This smoothing layer can serve as an electrode layer.

In order to demonstrate the effect of the layer thickness z of the active material, the following examples are presented, which were calculated for an operating wavelength λ = 13.5 nm of the EUV radiation and vertical radiation incidence (angle of incidence AOI = 0 °). The term "angle of incidence" here refers to the angle between the direction of incidence of a ray and the surface normal of the mirror at the point of impact of the ray on the mirror. The starting structure consists of 10 Mo / Si layer pairs in the first layer group 131 (ie, N1 = 10) and also 10 Mo / Si layer pairs in the second layer group 132 (ie N2 = 10). With a layer thickness of the Mo layers of 2.76 nm and 4.14 nm for the Si layers, a periodicity length d = 6.9 nm results for the stack of layer pairs. The layer thickness of the electrode layers 142 . 143 are each 2.76 nm. The layer thickness z of the active layer is variable.

To explain the influence of the layer thickness on the reflectivity R and the transmittance T of the entire layer arrangement shows 2 a corresponding diagram in which the reflectivity R and the transmittance T are shown as a function of the layer thickness z [nm] for the above example. Depending on the layer thickness z and the resulting phase difference between the radiation fractions reflected by the layer groups, z increases periodically with increasing layer thickness, so that maxima (peaks) and minima (valleys) arise in the curves, the maxima of the transmission T naturally occurring in the curve Range of minima of the reflectance R are.

The layer thickness z can now be determined with regard to different targets. In the first region R1 delimited by horizontal dashed lines, regions with maximum reflectivity R are at layer thicknesses around 5 nm and 12 nm, respectively. In these regions, due to the small slope of the reflectivity curve for a given layer thickness variation Δz, there is only a relatively small adjustment range (tuning range) for the variation of the total reflectivity. The underlying second region R2 covers the regions of relatively high slope of the reflectivity curves on the left and right of the reflectivity maxima. For a given layer thickness variation .DELTA.z of the active layer, particularly large adjustment ranges .DELTA.R for the total reflectivity result here (cf. 5 However, the absolute value of the reflectivity is slightly lower than in the areas of maximum reflectivity (region R1). The third region R3 marks the areas of minimum reflectivity and correspondingly maximum transmission of the layer arrangement. If appropriate layer thicknesses of the active layer are selected, the layer arrangement can also be used as a beam splitter layer with an adjustable ratio between reflectivity and transmission.

Based on 3 Now, the influence of the number of layer pairs in the substrate-distant first layer group 131 explained on the Reflektivitätsverlauf with increasing layer thickness z. 3 shows the region of the first reflectivity maximum (at layer thicknesses z around 5 nm), the second number N2 = 10 remaining constant and the first number N1 of the layer pairs in the near-surface first layer group varying between N2 = 10 and N2 = 25. It can be seen that the maximum reflectivity in the region of the first reflectivity maximum increases with increasing N2 from about 0.6 to about 0.72 and that the cuts in the Become flat region of the adjacent Reflektivitätsminima, so that the reflectivity over the layer thickness at N2 = 25 only between R = 0.6 and R = 0.72 varies.

Based on 4A It is demonstrated how the reflectivity curve in the region of the reflectivity maximum with increasing number of layer pairs in the substrate-second second layer group 132 with constant number N1 = 10 the layer pairs in the first layer group behaves. With increasing number of layer pairs in the substrate-second second layer group, the maximum reflectivity increases from about 0.6 to about 0.7 in the region of the reflectivity maximum, while the decrease in reflectivity in the region of the adjacent reflectivity minima (at about z = 1.5 nm and z = 8.4 nm ) increases.

It can be seen that there are many pairs of first and second layer groups (corresponding to first numbers N1 and second numbers N2), which give a high reflectivity in the vicinity of the absolute maximum reflectivity. Based on these curves, pairings of first and second numbers N1 and N2 can be selected which, with relatively high overall reflectivity, allow a large adjustment range with varying layer thickness z of the active layer.

It should be noted that the maximum reflectivity can not be arbitrarily increased by a larger number of layer pairs. Rather, z. As with Mo / Si layer pairs experience shows a saturation at about 50 pairs of layers. In the example calculations, the maximum number of layer pairs (N1 + N2) was limited to 48, since higher layer numbers hardly bring any significant changes in the overall behavior.

The tuning range (tuning range) is determined primarily by the elasticity and the yield strength of the piezoelectrically active layer material. When the yield strength (σ _y ) is exceeded, an irreversible deformation of the layer material begins. The yield strength is associated with the elasticity (described by the modulus of elasticity E, also called Young's modulus) of the material and the dimensional change or deformation of the material for which the strain ε (strain) serves as a normalized measure. The relationship between the dimensional change (elongation) of the layer thickness (ε _max = Δz / z) possible without plastic deformation of the material, the yield strength and the modulus of elasticity is given by e _max = σ _y / E. Where z is the initial layer thickness and Δz is the layer thickness change. The yield strength for piezoelectric materials is typically between 1% and 5% and for BaTiO _{3 is} about 4.8% ( RFCook, CJ Fairbanks, BR Lawn and Y.-W. May "Crack Resistance by Interfacial Bridging: Its Role in Determining Strength Characteristics," J. Mater Res., 2, 345-356 (1987) ).

Furthermore, the layer thickness of the active layer before the layer expansion with z _min and the layer thickness change with Δz are described. Depending on which side of a reflectivity maximum the optimization process is considered, the thickness of the piezoelectric material at the minimum reflectivity of the adjustment range and the maximum reflectivity of the adjustment range is given by z _max = z _min + Δz. Using this information and the yield strength of BaTiO ₃ (σ _y = Δz / z _min ), z _min , z _max and Δz can be calculated.

If, for example, the desired maximum reflectance is set to R _max = 72% and the number of layer pairs (N1 or N2) is limited to 48, the values given in Table 1 are obtained for the first five reflection maxima (peak 1 to peak 5). N1 = N ₁ and N2 = N ₂ .

The values z _min , z _max and Δz are given in Tables 1, 2 and 3 respectively in the unit 10 ^-10 m and 0.1 nm (corresponding to the usual, but no longer generally permissible unit of length Å (angstrom)).

A suitable solution in the region of the first reflection maximum (with layer thicknesses z of approximately 5 nm) is in 4B where R _max = 72%, N1 = 16 and N2 = 16. Table 1 N ₁ N ₂ R _{max Min} .DELTA.R z _max z _min Az Peak 1 16 16 72.0455 71.9528 0.0927 45.50 43.4160 2.0840 Peak 2 16 36 72.0044 71.5439 0.4605 115.25 109.9714 5.2786 Peak 3 19 25 72.0210 71.0087 1.0123 185.75 177.2424 8.5076 Peak 4 21 48 72.4302 70.8417 1.5886 255.50 243.7977 11.7023 Peak 5 22 48 72.2566 69.4827 2.7739 325.25 310.3531 14.8969

If a layer structure is to be optimized with regard to a maximum adjustment range of the reflectivity at a relatively high degree of reflection, it is preferable in the region of the second region R2 in FIG 2 worked. In this area, one can reach large adjustment ranges with relatively small dimensional changes (changes in layer thickness) of the active layer, whereby the maximum reflectance is not completely achievable there. If, for example, the minimum reflectivity is limited to 65%, it is possible, for example, to obtain a particularly large adjustment range in the range of the first reflection maximum with N1 = 18 and N2 = 28. The results of an optimization in the area of the first reflection maxima are summarized in Table 2. Table 2 N ₁ N ₂ R _{max Min} .DELTA.R z _max z _min Az Peak 1 18 28 67.5713 65.0722 7.4992 27.7720 26.5 1.2720 Peak 2 28 48 71.0973 65.1657 5.9317 94.3200 90 4.3200 Peak 3 25 48 72.0965 65.1191 6.9774 169.2520 161.5 7.7520 Peak 4 24 29 72.1154 5.0037 7.1117 242.6120 231.5 11.1120 Peak 5 24 32 72.3050 65.0045 7.3005 314.9240 300.5 14.4240

Among other things, it can be seen that larger absolute layer thicknesses of the active layer (corresponding to the second, third, fourth etc. reflectivity maximum) allow a larger layer thickness deviation Δz and thus a larger adjustment range ΔR of the reflectivity. A suitable compromise with regard to absorption by the active layer can be chosen.

5 shows the Reflektivitätsverlauf in the range of the first reflection maximum for N1 = 18 and N2 = 28. From Table 2 it can be seen that in this range a setting range .DELTA.R of the reflectivity of about 2.5% at a layer thickness variation .DELTA.z of 0.127 nm can be achieved.

• (1) 10 < N1 < 30
• (2) 15 < N2 < 50
• (3) 30 < (N1 + N2) < 70 und N1 > 10 und N2 > 10
• (4) N1 ≤ N2
• (5) z ≥ 2 nm
• (6) z ≤ 35 nm
• (7) Δz ≥ 0.1 nm
• (8) 0.15 nm ≤ Δz ≤ 2 nm

In one embodiment, in particular, at least one of the following conditions may apply:

• (1) 10 <N1 <30
• (2) 15 <N2 <50
• (3) 30 <(N1 + N2) <70 and N1> 10 and N2> 10
• (4) N1 ≤ N2
• (5) z ≥ 2 nm
• (6) z ≤ 35 nm
• (7) Δz ≥ 0.1 nm
• (8) 0.15 nm ≦ Δz ≦ 2 nm

Finally, be on the basis of 6 explained that multilayer layer arrangements of the type described can also be used as a physical beam splitter with variably adjustable transmission. The third region R3 off 1 with relatively large values of transmission is particularly suitable for this application. 6 shows by way of example the range around the second local transmission maximum (corresponding to the first reflectivity minimum) at layer thicknesses z 1.2 nm. Again, the calculation was carried out for incidence angle AOI = 0 °. Corresponding results would also result for the 45 ° angle of incidence better suited for use. For the purposes of the calculation, the minimum transmission was set to 30% and the transmittance varied above this level in the considered range. Table 3 gives exemplary values for N1 = 8 and N2 = 8. In this case, Tzmin and Rzmin are the transmission and the reflectivity at the minimum layer thickness zmin and Tzmax and Rzmax the corresponding values at the maximum layer thickness zmax. Table 3 Tzmin 30.3426 Rzmin 28.4563 zmax 94.5 Tzmax 36.1263 rzmax 17.2831 zmin 90.1718 .DELTA.T 5.7837 .DELTA.T 11.1732 Az 4.3282

Based on 7 schematically the operation of an EUV mirror assembly 700 shown, which has a plurality of mirror elements whose multilayer layer arrangement each have an integrated piezoelectrically active layer 140 between a near-surface first layer group 131 and a substrate-proximate second layer group 132 exhibit. The periodicity of the successive layer pairs should in each case be adapted to the used angle of incidence AOI such that maximum reflectivity results both within the first layer group and within the second layer group. Furthermore, let the layer thickness z be the active layer 140 in the absence of an electric field, such that completely constructive interference results between the second component A2 originating from the second group of layers and the first component A1 originating from the first group of layers. In the example shown, this leads to a total reflectivity in the vicinity of a reflectivity maximum in the mirror element shown on the left (see diagram above). The resulting intensity of the reflected radiation is represented by the arrow length of the rays emerging from the mirror surface.

If now a location-dependent variation of the reflectivity over the entire mirror surface is to be set, differently high electrical voltages can be applied to the active layers of the individual mirror elements, so that different layer thicknesses of the active layers occur within the individual mirror elements. In the example case, an increase of the layer thickness by Δz results from the application of an electrical voltage to the active layer of the right-hand mirror element. In contrast to the case of complete constructive interference (left), this leads to a phase shift between the partial waves originating from the first group of layers (component A1) and the partial waves (component A2) originating from the second group of layers in such a way that partially destructive interference occurs. As a result, the reflected total intensity in the example falls by ΔR (see diagram above), which is illustrated by the relatively shorter arrows of the outgoing beams.

It should be noted that each mirror element may have a different operating point or a different nominal layer thickness of the active layer. Layer elements can also have the same layer thicknesses in groups, which differ between two or more groups.

In 8th is schematically an EUV mirror arrangement 800 according to another embodiment shown in oblique perspective and partial section. This too has many mirror elements 810 . 811 . 812 arranged in rows and columns next to each other so that their individual element mirror surfaces as a whole form the total mirror surface of the mirror assembly. The layer structure of the mirror element 810 is explained in more detail by way of example. On a substrate 820 is a multilayer sandwich 830 applied by suitable coating techniques. The multilayer sandwich has a strictly periodic construction over most of its thickness with a plurality of pairs of layers 835 wherein each layer pair is a relatively thin layer 836 of a layer material with a relatively low real part of the refractive index and a thicker layer 840 of a layer of a layer material having a relatively higher real part of the refractive index.

The thicker layers 840 Each consists of the same, piezoelectrically active layer material and thus form an active layer whose layer thickness is changed by the action of an electric field. The thinner layers 836 each consist of an electrically conductive material. Each adjacent layers 836 close a single active layer 840 and serve as electrode layers for the intermediate active layer so as to apply an electric voltage between the adjacent layers 836 to generate an electric field that is the intermediate active layer 840 penetrates perpendicular to the layer surface. The electrode layers 836 are alternately connected to respective poles of a switchable DC voltage source 845 connected with variably adjustable voltage.

The through the sequence of electrically non-conductive active layers 840 and electrically conductive electrode layers 836 formed periodic arrangement has a periodicity P, which in the example, the layer thickness of a layer pair 835 equivalent. The periodicity length can be determined by creating a electrical voltage to the electrode layers are changed continuously, since the layer thickness z of the active layers changes in dependence on the applied voltage.

By varying the layer period P in response to the application of an electrical voltage, the reflectivity of the mirror element concerned at the operating wavelength can be influenced. Coupled by the Bragg equation, reflectivity changes can also be compensated by slightly detuned working wavelengths and / or angles of incidence. In addition, since the absolute thickness of the entire layer stack, ie the distance between the substrate and the element mirror surface of the individual mirror element, increases with an increase in the layer thicknesses of the active layers, there is also an influence on the wavefront of the radiation impinging on the mirror surface, since, for example, when the element mirror surface is raised, the optical path of the radiation reflected by the element mirror surface is shortened overall. Thereby, a phase shift may be introduced relative to adjacent non-activated or to some extent raised element mirror surfaces. This embodiment can thus be used simultaneously for spatially resolving the influence of the wavefront and for spatially resolving the reflectivity.

Since, in this embodiment, relatively large thicknesses of the active layer material are irradiated overall, a material with a low absorption coefficient (imaginary part of the complex refractive index) should be used as the active layer material.

The number of layer pairs 835 may be between 10 and 70, for example.

In the embodiment of an EUV mirror arrangement 900 in 9 are also mirror elements 910 . 911 . 912 arranged in rows and columns substantially side by side so that their rectangular element mirror surfaces as a whole form the mirror surface of the mirror assembly. The layer structure of the multi-layer arrangement reflecting the EUV radiation 930 is designed so that the EUV mirror arrangement can be used as a whole as a spatially resolving effective device for wave front correction, without that arise in the different modes of operation different spatial distributions of reflectivity. For this purpose, the multi-layer arrangement initially has a third layer group starting from the mirror surface 933 consisting of a third number N3 of similar layer pairs 935 consists. Each layer pair (bilayer) has a thinner absorber layer 936 made of molybdenum and a thicker spacer layer 934 made of silicon. The number N3 of the layer pairs 835 is chosen so that the periodic stack reflects the entire radiation entering from the radiation entrance surface (or to a smaller extent also absorbed). For this example, between 40 and 50 pairs of layers 935 be provided. The period of the layer pairs is chosen as a function of the occurring angle of incidence range and the operating wavelength such that maximum or almost maximum reflectivity occurs according to the Bragg equation.

Between the third shift group 933 and the substrate 920 there is an active layer 940 from a piezoelectric active layer material. At the top and the bottom of the active layer are respectively electrode layers 942 respectively. 943 arranged, wherein the substrate near the electrode layer 943 directly on the substrate 920 or, in other embodiments, may be disposed on an intermediate layer therebetween. The electrode layers 942 . 943 are connected to a switchable DC voltage source 945 connected, with the DC voltage of predeterminable height between the electrode layers can be applied if necessary, so that the piezoactive layer 940 is penetrated by an electric field and changes its layer thickness z as a function of the applied voltage.

The active layer 940 and the adjacent electrode layers 943 . 942 are each applied by means of laser beam evaporation (Pulsed Laser Deposition, PLD). With appropriate process control, the layers can be applied as monocrystalline layers, so that the surface of the upper electrode layer 942 , are then applied to the Mo / Si-layer pairs, has such a low roughness that a repolishing can be omitted.

In connection with the 10A to 10C some variants for the use of structured electrodes in the construction of mirror arrays are explained. 10A shows in an oblique perspective a schematic view of three rectangular cross-section mirror elements 1010 . 1011 . 1012 a mirror arrangement 1000 , which can be used, for example, as a field facet mirror in the region of a field plane of the illumination system of an EUV projection exposure apparatus (cf. 11 ). The layer structure of the mirror element 1010 is shown in detail. Similar to the arrangement of 1 is on a substrate 1020 a multilayer sandwich 1030 applied from many individual layers. A single active layer 1040 of piezoelectric active crystalline layer material is between a lying in the vicinity of the beam entrance surface first layer group 1031 and a substrate-proximate second layer group 1032 arranged. Both layer groups each consist of several (eg between 10 and 30) layer pairs of suitable layer period and in each case have a reflective effect on the penetrating EUV radiation. About the electrically variable layer thickness z of the active layer 1040 may be a certain and with regard to their extent continuously adjustable phase shift between that of the second layer group 1032 reflected second portion and that of the first layer group 1032 be reflected reflected first proportion.

The electrode arrangement for driving the active layer has a substrate-side second electrode layer which extends over the entire cross-section of the mirror element 1043 connected to one pole of an adjustable DC voltage source 1045 connected. The on the opposite surface of the active layer 1040 arranged first electrode layer 1042 is designed as a structured layer electrode and in a plurality of juxtaposed, electrically mutually insulated electrode segments 1042A . 1042B divided. Each of the electrode segments covers only a fraction of the total cross-sectional area of the mirror element, e.g. Less than 50% or less than 40% or less than 30% or less than 20% or less than 10%. As a rule, the area of a single electrode segment is at least 1% or at least 5% of the total area of the structured layer electrode. Between adjacent electrode segments is in each case a narrow insulating section 1044 made of an electrically non-conductive layer material. The insulating regions each extend obliquely to the long edge (x-direction) and obliquely to the short edge (y-direction) of the mirror element. Other orientations are possible.

The structured electrode 1042 can be produced for example by means of microlithographic methods.

Each of the electrode segments is connected to the other pole of the DC voltage source via a separate electrical line 1045 connected and independent of the other electrode segments to a suitable potential relative to the continuous second electrode layer 1043 be placed. As a rule, there is a separate switchable or continuously adjustable DC voltage source for each electrode segment.

With the aid of the structured electrode it is possible to determine the layer thickness of the active layer 1040 To change location-dependent, to produce a running in the x direction layer thickness profile. Corresponding to the layer thickness varying in the x-direction, a phase shift of the reflected radiation components that varies spatially in the x-direction then sets in, so that the reflectivity R of this individual mirror element can be adjusted and varied as a function of location in the x-direction as a result. The schematic diagram shown above the mirror arrangement shows that a higher overall reflectivity R is set in the region of the electrode segments shown on the left than in the region of the opposite narrow side, with a transitional region therebetween.

In this mirror arrangement 1000 It is thus possible, not only the reflectivity level of individual mirror elements 1010 . 1011 . 1012 each individually to control, but also within each individual mirror element to set a desired course with varying spatial reflectivity R. Thus, a single mirror element again forms an EUV mirror arrangement with two or more mirror elements which can be individually adjusted with regard to their reflectivity, the shape and size of these mirror elements being determined by the shape and size of the electrode segments 1042A . 1042B are determined.

The 10B and 10C schematically show other structuring geometries of structured layer electrodes, which can be used in embodiments of individual mirror elements or mirror elements of a mirror array equipped with a plurality of mirror elements. The overall circular structured electrode 1050 can be used in a circular cross-section mirror element in combination with a likewise circular, but not divided into segments counter electrode. In the example, the structured electrode has twelve individually controllable electrode segments 1050A . 1050B identical shape and size, each covering an angular range of about 30 °. In each case at the free outer edge of an electrode segment, a contact point KP for connecting an electrical connection to a voltage source is provided. Such a structured electrode arrangement may, for example, be provided on a single mirror of a projection objective in order to set a radially symmetric nonuniform distribution of the reflectivity and / or the reflected phase on the mirror surface, the reflection behavior varying in the azimuthal direction (circumferential direction) and targeting this variation can be adjusted. For example, a local reflectivity distribution may be adjusted with a two-fold or a three-fold or a four-fold or a six-fold azimuth symmetry.

The structured electrode 1060 in 10C has a plurality of mutually electrically isolated electrode segments 1060A . 1060B , which cover the circular surface of the mirror element with a small mutual distance. Instead of the checkered arrangement of square electrode segments, other polygonal shapes, for example triangles or hexagons, may also be provided, for example. The counter-electrode, not shown, on the other side of the active layer is continuous, ie not segmented. Between the respective square electrode segments extending in different directions insulating regions 1064 arranged from eleltrisch insulating material. Those electrode segments which adjoin the circular outer edge of the electrode arrangement can be contacted directly from the outside via corresponding first contact points KP1. The internal electrode segments without connection to the outside of the patterned electrode layer are contacted via narrow lines which are electrically insulated on both sides within the insulating regions 1064 extend to the respective electrode segments to be contacted and to second contact points KP2. This makes it possible to set each electrode segment separately and independently of other electrode segments to a specific potential with respect to the counter electrode (not shown) and thereby adjust the layer thickness of the active layer in the associated layer region.

EUV mirror arrangements which enable a spatially resolving adjustment of the reflectivity profile over the total mirror surface of the mirror arrangement and / or over the surface of a single mirror element with the aid of piezoelectrically active layers can be used for various tasks. In the following, uses within the illumination system for an EUV microlithography projection exposure apparatus are presented.

11 shows optical components of an EUV microlithography projection exposure machine 1100 for exposing one in the area of a picture surface 1160 a projection lens 1130 arranged radiation-sensitive substrate having at least one image of one in the region of an object surface 1120 of the projection lens arranged pattern of a reflective pattern generating device or mask.

The plant is using the radiation of a primary radiation source 1114 operated. A lighting system 1110 serves to receive the radiation from the primary radiation source and to form illumination radiation directed at the pattern. The projection lens 1130 serves to image the structure of the pattern onto a photosensitive substrate.

The primary radiation source 1114 may be, inter alia, a laser plasma source or a gas discharge source or a synchrotron-based radiation source. Such radiation sources generate radiation 520 in the EUV range, in particular with wavelengths between 5 nm and 30 nm. In order for the illumination system and the projection objective to be able to work in this wavelength range, they are constructed with components reflective to EUV radiation.

The of the radiation source 1114 outgoing radiation 1120 is done by means of a collector 1115 collected and into the lighting system 1110 directed. The radiation passes through a Zwischenfokusebene 1122 in which devices for the separation of unwanted radiation components can be provided. The lighting system comprises a mixing unit 1112 , a telescope optics 1116 and a field-forming mirror 1118 , The illumination system shapes the radiation and illuminates an illumination field in the object plane 1150 of the projection lens 1130 or near it. The shape and size of the illumination field determine the shape and size of the effectively used object field in the object plane 1150 ,

In the object plane 1150 When operating the system a reflective reticle or another reflective pattern generating device is arranged. The projection lens here has six mirrors and forms the pattern of the pattern generating device in the image plane, in which a substrate to be exposed, z. B. a semiconductor wafer is arranged.

The mixing unit 1112 consists essentially of two faceted mirrors 1170 . 1180 , The first facet mirror 1170 is in a plane 1172 of the lighting system arranged to the object plane 1150 is optically conjugated. It is therefore also called a field facet mirror. The second facet mirror 1180 is in a pupil plane 1182 arranged the illumination system, which is optically conjugate to a pupil plane of the projection lens. It is therefore also referred to as a pupil facet mirror.

With the help of the pupil facet mirror 1180 and the downstream in the beam path imaging optical assembly, the telescope optics 1116 and the grazing incidence field-forming mirror 1118 includes the individual specular facets (individual mirrors) of the first facet mirror 1170 in the object field 1152 displayed.

About the facets of the field facet mirror 1170 on the one hand and the pupil facet mirror 1180 on the other hand, the radiation beam coming from the radiation source is divided into a plurality of illumination channels, wherein each illumination channel is assigned exactly one facet pair with a field facet and a pupil facet. The following components guide the radiation of all illumination channels to the object field 1152 ,

The spatial (local) illumination intensity distribution at the field facet mirror determines the local illumination intensity distribution in the object field. The spatial (local) illumination intensity distribution at the pupil facet mirror 1180 determines the illumination angle intensity distribution in the object field.

EUV projection exposure systems with a similar basic structure are z. B. from the WO 2009/100856 A1 or WO 2010/049020 A1 the disclosure of which is incorporated herein by reference.

In the embodiment shown, the channel-dependent transmissions and thus the energetic illumination angle distribution can be influenced via the reflectivities of the individual field and pupil facets. The spatial illumination intensity distribution in the object field can be influenced by location-dependent variation of the field facet reflectivities.

Each of the facet mirrors 1170 . 1180 is an EUV mirror assembly having a plurality of individual mirror elements. Their reflecting front surfaces are referred to as element mirror surfaces and form the facets (mirror surfaces) of the facet mirror.

The field facet mirror 1170 and the pupil facet mirror 1180 are by type of in 1 shown EUV mirror assembly 100 built up. Each of the mirror elements thus has a multi-layer arrangement in which a single piezoelectric active layer lying between two electrode layers is arranged between a second layer group close to the substrate and a first layer group close to the surface. The electrode layers are electrically connected to a control device 1190 which is configured to selectively energize the electrode pairs belonging to the individual active layers as needed to vary the layer thickness. This can be done separately for each mirror element and independently of the other mirror elements, so that different local distributions of the layer thickness variation and thus different local reflectivity distributions at the facet mirror concerned can be set.

The ability to precisely control the local reflectivity of the facet mirrors can be used to control the illumination intensity distribution across the pupil of the illumination system and in the illumination field. If the reflectivity of an associated field facet and / or the reflectivity of an associated pupil facet is changed by electrical actuation in one or more of the illumination channels, then the illumination intensity in this illumination channel can be selectively changed within a certain adjustment range. Since this is possible for several and optionally all illumination channels independently of other illumination channels, a controllable manipulator for the illumination intensity distribution over the pupil can be realized in order to provide exactly a desired intensity distribution in the illumination field as a function of the illumination angle.

The field facet mirror can also have mirror elements which are constructed with structured electrodes and thus enable a location-dependent adjustment of the reflectivity of each individual mirror element (cf. 10 ). Since it is possible in this embodiment, the reflectivity of the field facet mirror 1170 It is possible to control the local illumination intensity distribution in the optically conjugated illumination field within the individual facets 1152 location-dependent approach to the desired illumination intensity distribution. In this way, a desired value of field uniformity can be set precisely.

In the image field, the images of the individual field facets are superimposed. Their long side runs parallel to the x-direction (cross-scan direction), while the short side runs parallel to the y-direction, which corresponds to the scanning direction in scanner systems. Due to the inclination of the insulating sections 1044 ( 10A ) becomes ensures that their projection into the image plane neither parallel nor perpendicular to the scan direction, but obliquely to it. During the scanning process, the image field has an integrated effect in the y-direction over the scanning process so that any artifacts of the insulating sections in the image field are practically invisible.

Thus, a fully programmable "gray filter" is realized for intensity distributions in the pupil and field of the illumination system. The lateral resolution of the controllable gray filter is determined by the lateral extent of the separately controllable mirror elements or of their number across the illuminated cross section.

In the embodiment, the illumination intensity distribution in the illumination field becomes 1152 through a field uniformity sensor 1153 and the illumination intensity distribution in the pupil area via a pupil intensity sensor 1183 supervised. Deviating from the schematic representation, these sensors can be located in the region of the image plane of the projection lens. These are to the controller 1190 connected, which controls based on the sensor signals, the voltage between the electrode layers of the individual piezoelectrically active layers and thus the local reflectivity distribution of the facet mirror. This control circuit permanently ensures a high degree of precision of the important illumination parameters in the pupil and field of the illumination system.

Using the exemplary embodiments, some possible uses of EUV mirror arrangements in the illumination system of a microlithography projection exposure apparatus have been explained. Alternatively or additionally, it may also be provided, at least one mirror of the telescope optics 1116 the illumination system and / or at least one mirror of the projection lens 1130 according to one embodiment of an EUV mirror arrangement.

The total mirror surface of an EUV mirror arrangement can be flat in the manner of a plane mirror. It is also possible to design an EUV mirror arrangement in a convexly or concavely curved mirror surface. In the examples, the individual mirror surfaces are each plane surfaces. However, this is not mandatory. Individual or all individual mirror surfaces of the mirror elements can also be curved convexly or concavely.

In the exemplary embodiments, the relative orientations of the individual mirror surfaces of the EUV mirror arrangement are respectively defined, wherein the electrically induced layer thickness change of the active layer merely leads to an elevation or lowering of this mirror surface with respect to the substrate. In addition, it is also possible that individual or all mirror elements of an EUV mirror arrangement can be tilted relative to one another by means of independent actuators in order to change the illumination angle distribution of the reflected radiation in a targeted manner (cf. WO 2009/100856 A1 ).

The shape of the individual mirror elements can be adapted to the desired application. If, for example, an EUV mirror arrangement is to be used as a field facet mirror, the individual mirror surfaces can be rectangular with a predeterminable aspect ratio or curved in an arc shape. In the case of EUV mirror arrangements which are to be used as a pupil facet mirror, in addition to polygonal cross sections, round cross sections of the individual mirror elements may also be useful.

The layer structure of the multi-layer layer arrangement in the region of the successive layer pairs can be adapted to the desired application. If high maximum reflectivities are required for a relatively small angle of incidence range, then a full-period sequence of layer pairs can be advantageous. If, on the other hand, a broadband in the angular space and / or a spectral broadband are desired, it is also possible to combine pairs of layers with different periods (cf. DE 101 55 711 B4 or WO 2010/118928 A1 , In principle, in order to reduce the dependence of the reflectivity on the angle of incidence, the layer arrangement can also be constructed as in the US Pat US 7,382,527 B2 disclosed. In particular, different material pairings may be provided in the case of the layer pairs of a multilayer layer arrangement.

The described layer structures of the multi-layer layer arrangements can in principle also be provided in the case of mirrors having only a single mirror element. As a result, z. B. by means of electrically induced change in the layer period of a layer arrangement to adapt to a slightly different central wavelength and / or adaptation to changing angles of incidence (angle of incidence). In addition, the global intensity or dose can be adjusted.

When using structured layer electrodes (see eg 10 ) is a location-dependent control of the reflectivity and / or phase of the incident on a single mirror element radiation possible.

Embodiments of the invention can be used not only in optical systems for projection microlithography. For example, a use in the field of X-ray microscopy is possible, in particular in the field of EUV mask metrology. For example, one or more mirror arrays can be used in an Aerial Image Monitoring System (AIMS) or in an Actinic Patterned Mask Inspection System (APMI) or in an Actinic Blank Inspection System (ABI). Lenses for EUV AIMS systems are z. In international publications WO 2011/012267 A1 and WO 2011/012266 A1 shown. Applications in the EUV system measuring technology, eg. B. in an actinic system interferometer, are also conceivable. Furthermore, applications in the field of EUV astronomy and for optics in synchrotron systems or FEL beam lines are conceivable (FEL = free electron laser).

The embodiments were configured for a central wavelength of 13.5 nm. Other embodiments may be optimized for other wavelengths (ranges), for example, for a center wavelength of about 6.8 nm. In particular, in this case, other layer materials may be used for the alternating layers of the layer pairs, e.g. As the combination La / B ₄ C. With shorter central wavelengths, it may be useful to increase the numbers of layer pairs in layer groups compared to the examples described above.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 2008/113281 A1 [0010]
• US 7545585 B2 [0010]
• WO 2010/049020 A1 [0011, 0120]
• US 2003/0063266 A1 [0011]
• EP 1349009 A2 [0011]
• US 2008/0165925 A1 [0011]
• WO 2009/135576 A1 [0011]
• EP 2050726 A2 [0038]
• WO 2009/100856 A1 [0120, 0131]
• DE 10155711 B4 [0133]
• WO 2010/118928 A1 [0133]
• US 7382527 B2 [0133]
• WO 2011/012267 A1 [0136]
• WO 2011/012266 A1 [0136]

Cited non-patent literature

• RFCook, CJ Fairbanks, BR Lawn and Y.-W. May "Crack Resistance by Interfacial Bridging: Its Role in Determining Strength Characteristics," J. Mater Res., 2, 345-356 (1987) [0080]

Claims (22)

EUV mirror arrangement having a multiplicity of juxtaposed mirror elements which together form a mirror surface of the mirror arrangement, each mirror element comprising a substrate and a multilayer layer arrangement (FIG. 2) applied to the substrate for radiation from the extreme ultraviolet region (EUV) ( 130 . 830 . 930 ) having a plurality of layer pairs ( 135 . 835 . 935 ) comprising alternating layers of a high-refracting layer material and a low-refractive layer material, characterized in that the multilayer-layer arrangement comprises an active layer consisting of a piezoelectrically active layer material arranged between a radiation entrance surface and the substrate (US Pat. 140 . 840 . 940 ), whose layer thickness is variable by the action of an electric field; and that for each active layer an electrode arrangement is provided for generating an electric field acting on the active layer. An EUV mirror assembly according to claim 1, wherein the multilayer stack ( 130 ) one between the radiation entrance surface and the active layer ( 140 ) arranged first layer group ( 131 ) with a first number N1 of layer pairs and one between the active layer ( 140 ) and the substrate ( 120 ) arranged second layer group ( 132 ) having a second number N 2 of layer pairs, wherein the numbers N 1 and N 2 of layer pairs of the first layer group and the second layer group are selected such that for at least one angle of incidence of the radiation incident on the radiation surface incidence the first layer group ( 131 ) a portion of the incoming radiation through the active layer ( 140 ) to the second layer group ( 132 ) and the radiation reflected by the multilayer stack includes a first portion (A1) reflected by the first group of layers and a second portion (A2) reflected by the second group of layers. EUV mirror assembly according to claim 2, wherein the active layer in the absence of an electric field has a layer thickness which is selected such that for a reference incidence angle of the incident radiation, a reflectivity of the multilayer film arrangement by applying an electric field by a maximum of 20% changeable is. An EUV mirror device according to claim 2 or 3, wherein the piezoelectric active layer material consists essentially of barium titanate (BaTiO ₃ ). An EUV mirror device according to claim 2, 3 or 4, wherein at least one of the following conditions holds: (1) 10 <N1 <30 (2) 15 <N2 <50 (3) 30 <(N1 + N2) <70 and N1> 10 and N2> 10 (4) N1 ≤ N2 (5) z ≥ 2 nm (6) z ≤ 35 nm (7) Δz ≥ 0.1 nm (8) 0.15 nm ≤ Δz ≤ 2 nm, where z is the layer thickness of the active layer ( 140 ) and Δz is a layer thickness change produced by the action of the electric field. An EUV mirror assembly according to claim 1, wherein the multilayer stack ( 830 ) a plurality of active layers ( 840 ) of a piezoelectrically active layer material, wherein the active layers in each case alternately with non-active layers ( 836 ) are arranged. EUV mirror assembly according to claim 6, wherein the active layer material predominantly of a ceramic material of the type (Li, Na, K) (Nb, Ti) O ₃ , wherein the active layer material is preferably selected from the group potassium niobate (KNbO3), lithium niobate (LiNbO3), PbNb2O6 and sodium potassium niobate (Na _0.9 K _0.1 NbO ₃ ). An EUV mirror assembly according to claim 1, wherein the multilayer stack ( 930 ) one between the radiation entrance surface and the active layer ( 940 ) arranged third layer group ( 933 ) having a third number N3 of layer pairs, wherein the third number N3 is selected such that for at least one angle of incidence of the radiation incident on the radiation entrance surface, the third group of layers reflects or absorbs the incoming radiation before reaching the active layer. An EUV mirror assembly according to claim 8, wherein the third number N3 is between 10 and 70. EUV mirror assembly according to one of the preceding claims, wherein the electrode assembly comprises a first electrode layer ( 142 . 836 . 942 . 1042 ) and a second electrode layer ( 143 . 836 . 943 . 1043 ) and the active layer ( 140 . 840 . 940 . 1040 ) is disposed between the electrode layers. EUV mirror arrangement according to one of the preceding claims, wherein the electrode arrangement for driving the active layer ( 1040 ) an electrode layer ( 1042 ), which is designed as a structured layer electrode and in a plurality of juxtaposed, electrically mutually insulated electrode segments ( 1042A . 1042B ), wherein each of the electrode segments only a fraction of the total cross-sectional area of the mirror element ( 1010 ) covers. EUV mirror assembly according to one of the preceding claims, wherein the active layer is a PLD layer applied by laser beam evaporation (PLD). EUV mirror arrangement according to one of claims 10, 11 or 12, wherein at least one electrode layer is a PLD layer applied by laser beam evaporation (pulsed laser deposition, PLD). EUV mirror assembly according to one of claims 10 to 13, wherein an electrode layer consists of an electrically conductive ceramic material, in particular SrRuO ₃ or AlN. EUV mirror arrangement according to one of the preceding claims, wherein the piezoelectrically active layer material is selected from the group consisting of: Ba (Sr, Zr) TiO _3, Bi (Al, Fe) O ₃ (Bi, Ga) O ₃ (Bi, Sc ) O ₃ , CdS, (Li, Na, K) (Nb, Ta) O ₃ , Pb (Cd, Co, Fe, In, Mg, Ni, Sc, Yb, Zn, Zr) (Nb, W, Ta, Ti) O ₃ , ZnO, ZnS or at least one material of this group in combination with at least one other material. Optical system with at least one EUV mirror arrangement according to one of the preceding claims. An optical system according to claim 16, wherein the optical system is a lighting system ( 1110 ) or a projection lens ( 1130 ) of a microlithography projection exposure apparatus ( 1100 ). Optical system according to claim 17, wherein the EUV mirror arrangement ( 1000 . 1170 ) in a beam path of the illumination system ( 1110 ) between a light source ( 1114 ) and a lighting field to be illuminated ( 1152 ) in or near a field level ( 1172 ) which is optically conjugate to a plane ( 1120 ) of the illumination field is located. An optical system according to claim 18, wherein in the case of several or all mirror elements ( 1010 . 1011 . 1012 ) of the mirror arrangement ( 1000 ) the layer thickness of the active layer ( 1040 ) Can be changed depending on location, that the reflectivity R of a mirror element is location-dependent adjustable and variable. An optical system according to any one of claims 16 to 19, wherein the EUV mirror assembly ( 1180 ) in the region of a pupil plane ( 1182 ) of the illumination system is Fourier-transformed to the plane ( 1120 ) of the illumination field is located. A method of operating an optical system having at least one EUV mirror assembly according to any of claims 1 to 15, wherein a local reflectivity distribution across the mirror surface of the EUV mirror assembly is varied depending on location by selectively driving one or all of the active layers of the mirror elements. The method of claim 21, wherein a detuning of the layer period is generated by changing the layer thickness of the active layer, in particular for adapting the reflectivity to a deviating from a nominal center wavelength of the radiation and / or to a modified angle of incidence distribution at the mirror assembly.

Images