WO2016001092A1

WO2016001092A1 - Optical element, optical manipulator, projection lens and projection exposure apparatus

Info

Publication number: WO2016001092A1
Application number: PCT/EP2015/064558
Authority: WO
Inventors: Eric Eva
Original assignee: Carl Zeiss Smt Gmbh
Priority date: 2014-07-01
Filing date: 2015-06-26
Publication date: 2016-01-07
Also published as: TW201606334A; DE102014212711A1

Abstract

The invention relates to an optical element (OE) made of fused silica, which has an inner region (FP) that is continuous in a thickness direction (z), which inner region is surrounded by a circumferential edge region (BA) that is continuous in the thickness direction (z), wherein a hydrogen content ( [H₂] ) in the circumferential edge region (BA) is less than a hydrogen content ( [H₂] ) in the inner region (FP). The invention also relates to an optical manipulator with such an optical element (OE), a projection lens with at least one such optical manipulator and a projection exposure apparatus with such a projection lens.

Description

Optical element, optical manipulator, projection lens and projection exposure apparatus

Reference to related applications

This application claims the priority of the German patent application DE 10 2014 212 711.4, dated 01.07.2014, the entire disclosure of which is made to be the content of this application by reference.

Background of the invention The invention relates to an optical element made of fused silica. The invention also relates to an optical manipulator having such an - in particular plate-shaped - optical element, a microlithographic projection lens with at least one such optical manipulator and a microlithographic projection exposure apparatus with such a projection lens.

The use of optical manipulators for wavefront correction or for correcting aberrations in microlithographic projection lenses is known. By way of example, US 2008/0239503 Al describes a projection lens of a microlithographic projection exposure apparatus which has an optical manipulator for reducing non- rotationally symmetric aberrations. In one exemplary embodiment, the manipulator has a first optical element, e.g. a planar plate, and a second optical element, with an interspace into which a liquid has been introduced being formed therebetween. At least one actuator is coupled to the first optical element in such a way that operating the actuator brings about a non-rotationally symmetric deformation of the first optical element. It is possible to attach at least two actuators along the circumference of the first optical element, which actuators introduce mechanical bending moments into the optical element in order to reversibly change the surface form thereof .

In order to bring about a sufficiently large change in the surface form (bending) of the optical element for the purposes of correcting wavefront aberrations, the thickness of the optical element to be manipulated should not be selected to be too large: as the thickness of the optical element increases, there is also an increase in the mechanical tensions that are required for the same deflection and fatigue fractures appear earlier. The optical element needs to have a sufficient mechanical durability in order to achieve the envisaged service life in the case of the continuous stresses by the action of the actuators over a large number of cycles. In addition to the question of the mechanical service life, there is also the problem of damage to the fused silica by the laser light in the UV wavelength range.

Optical elements made of fused silica are produced by mechanical processing of a blank made of synthetic fused silica. The synthetic fused silica of the blank is generally produced by combusting an organic or inorganic precursor substance, which contains silicon, in H₂ and 0₂, possibly under addition of a fuel gas such as e.g. natural gas. Si0₂ particles are already formed in the flame and deposited on a target. In fused silica production, a distinction is made between the so-called direct process, in which a very hot flame is directed onto a hot target such that the particles vitrify directly, and the so-called soot process, in which a porous body is deposited at lower temperatures and sintered to make a solid glass body.

These days, soot glasses, which are dried, sintered and subject to stress-relief annealing to OH contents of typically less than approximately 200 ppm, are generally used in projection exposure apparatuses that are operated at a wavelength of approximately 193 nm. In order to saturate E' F-centres in the fused silica, which would strongly absorb laser radiation at a wavelength of 193 nm, the fused silica is loaded with hydrogen after annealing, to be precise typically at "cold" temperatures of at most 600°C, in order to avoid the formation of silane (SiH) . By contrast, directly deposited fused silica already obtains high concentrations of hydrogen and silane directly during deposition.

The currently envisaged pulse energy density in microlithographic projection exposure apparatuses lies in the range of approximately 0.2 mj/cm², but it is not possible to preclude that energy densities of up to 1 mJ/cm² or more may occur in future. The required total number of laser pulses will be of the order of 300 to 500 billion. In order to ensure that the fused silica material survives these requirements without significant induced absorption, induced wavefront aberration and formation of micro-channels , a low- defects fused silica which is loaded with hydrogen is required.

Object of the invention

An object of the invention lies in the provision of an optical element which is suitable for use in an optical manipulator of a microlithographic projection exposure apparatus .

Subject matter of the invention In accordance with a first aspect, this object is achieved by an optical element made of fused silica, which has an inner region that is continuous in a thickness direction, which inner region is surrounded by a circumferential edge region that is continuous in the thickness direction, wherein a hydrogen content in the circumferential edge region is less than a hydrogen content in the inner region.

The material of the optical element is artificially produced fused silica, which is transparent to the used wavelengths relevant here, which lie in the UV wavelength range at less than approximately 250 nm. In the installed state of the optical element in an optical arrangement, e.g. in a projection lens, the inner region forms the optically used region, i.e. the inner region is that region of the optical element through which used radiation passes in a directed manner. The optically used inner region is surrounded by a ring-shaped circumferential edge region, in which e.g. actuators of an actuation device of an optical manipulator can act on the optical element in order to bend the optical element as desired and thereby achieve a wavefront correction.

The inner region, through which the radiation passes, and the circumferential edge region extend through the optical element in the thickness direction, i.e. the two regions divide the volume of the optical element formed between the two surfaces of the optical element lying opposite one another in the thickness direction into an outer, ring-shaped or frame-shaped region and an inner region, which differ in terms of hydrogen content. Thus, the optical element has a gradient in the hydrogen content in the lateral direction (i.e. perpendicular to the thickness direction) . To the extent that the optical element has rotational symmetry, the thickness direction is that direction along which the axis of symmetry extends. Otherwise, for example in the case of a plate-shaped optical element with e.g. a rectangular basic form, the thickness direction corresponds to the direction along which the optical element has the minimum extent.

The hydrogen content in the inner region is selected to be greater than in the circumferential edge region since the inner region, as described above, forms the optically used region which is exposed to laser radiation in the UV wavelength range. As described further above, the hydrogen serves to saturate the E' F-centres in the fused silica and thus to prevent laser- induced absorption in the fused silica. In the case of high energy densities of 0.2 mJ/cm² or more, a high hydrogen content is needed in the fused silica for this purpose. However, the hydrogen introduced into the fused silica material already diffuses at room temperature and so depletion zones with a thickness of a few 10 μτα form in surface-near volume regions of the optical element over time, e.g. over several years. As a result of the typically low thickness of the optical element, there is the additional problem that there is a significant hydrogen depletion, at least in the surface-near volume regions adjacent to the surfaces of the optical element, as a result of heating processes, for example by the application of optical antireflection coatings at an increased temperature.

The hydrogen depletion can lead to the connection between the optical element and the actuators, which engage on the optical element with the contact faces thereof, being weakened over time as a result of the emerging hydrogen. By way of example, such a connection can be a soldered connection, a welded connection or an adhesively bonded connection. In order to prevent a weakening of the connection to the actuators, the hydrogen content in the circumferential edge region is (significantly) lower than the hydrogen content in the optically used inner region, i.e. the optical element has a lateral gradient in the hydrogen content . Within the meaning of this application, the term hydrogen or the notation H (or OH) is understood to mean all isotopes of hydrogen (protium, deuterium, tritium) should nothing else be specified. Deuterium (D₂) in particular can likewise be used for loading with hydrogen in addition to protium (H₂) . Typically, loading is only carried out with one of the hydrogen isotopes, i.e. with either H₂ or D₂, but loading with mixtures of the two (or optionally with tritium T₂) is likewise possible .

The optical element is preferably a plate-shaped optical element, i.e. a plane parallel plate, which simplifies the use of the optical element in a manipulator. Alternatively, the optical element can also deviate from a plate-shaped geometry and can, for example, have an inner, conically extending region which forms the optically used region and which is surrounded by a ring-shapedly circumferential, e.g. collar- shaped edge region, which is not optically used.

The optical element typically has a small thickness of e.g. less than 5 mm, in particular of less than 3 mm, in order to enable sufficient bending and hence sufficient deformation for the wavefront manipulation. As described above, there is an additional problem as a result of the low thickness of the optical element that heating processes after the loading with hydrogen and the storage at room temperature lead to significant hydrogen depletion. The loading profile, to be selected, of the optical element with hydrogen should anticipate these diffusion processes during the production and use of the optical element.

In one embodiment, the minimum hydrogen content in the inner region is more than 1x10 16 molecules/cm³ , preferably more than 5x10 16 molecules/cm³. Since hydrogen diffuses inward through the two e.g. plate- shaped surfaces of the optical element lying opposite one another, the minimum hydrogen content typically occurs at half the thickness of the (plate- shaped) optical element, provided that the same amount of material is removed from both sides after loading. The strength or the profile of the drop in hydrogen content from the surfaces or from the surface-near volume regions to the centre can be adjusted to a certain extent, depending on the requirements on the optical element .

In one embodiment, the optical element has a first (planar) surface and a second (planar) surface that is opposite to the first surface in the thickness direction, wherein the hydrogen content in the inner region, at least up to a distance of 0.25 mm from the first and second surface, is at least lxlO¹⁶ molecules/cm³ , preferably at least 5xl0¹⁶ molecules/cm³, particularly preferably at least 2xl0¹⁷ molecules/cm³. As described further above, a high hydrogen content is required due to the high energy densities to which the optical element is exposed in the optically used inner region. However, a hydrogen content that is too high should be avoided since this can lead to a laser- induced change in the optical properties.

Since hydrogen, during loading, diffuses into the fused silica through the e.g. planar, plate-shaped surfaces, said fused silica typically has a higher hydrogen content in the surface-near volume region than in the volume region situated further inside. The distribution of the hydrogen content in the thickness direction of the optical element therefore typically has a maximum in a surface-near volume region and decreases with increasing distance from the surfaces of the optical element. There can be a depletion as a result of the outward diffusion of hydrogen, in particular in the surface-near volume region, for example if antireflection coatings are applied to the surfaces at an increased temperature or if the fused silica is stored at room temperature for a long period of time. The loading profile, to be selected, in the thickness direction of the optical element should anticipate the diffusion processes that occur after loading. It is possible to set a suitable distribution of the hydrogen content in the optical element by suitably selecting the parameters when loading with hydrogen.

Since the energy density remains approximately constant in the thickness direction, the critical variable is normally the hydrogen content at the centre of the optical element. An increase towards the edge, i.e. towards the mutually opposing surfaces, generally occurs automatically since hydrogen enters the optical element by the surfaces (see above) . The increase towards the edge can be deliberately increased by a second short-term loading at a higher hydrogen partial pressure if losses due to heating processes or outward diffusion at room temperature are intended to be compensated for . In a further embodiment, the maximum hydrogen content in the circumferential edge region is no more than 5xl0¹⁵ molecules/cm³, preferably no more than lxlO¹⁵ molecules/cm³. As described above, it is advantageous if the hydrogen content in the circumferential edge region is not too high so as to avoid possible damage of the connection to the actuators . The maximum occurring hydrogen content in the circumferential edge region depends on the manner in which the inward diffusion of hydrogen into the edge region is prevented. In general, the hydrogen content is higher in the case of masking- off the edge region than in the case of pre-contouring . In order to avoid, or at least reduce, the inward diffusion of hydrogen into the circumferential edge region during the loading, the circumferential edge region can be covered or masked-off, for example by virtue of frame-shaped bodies being placed onto the surfaces of the optical element, the geometry of which frame-shaped bodies corresponds to that of the circumferential edge region. The material of the frame- shaped bodies likewise can be fused silica or a different material, in which the diffusion of hydrogen progresses more slowly than in fused silica.

A further option for reducing the inward diffusion of hydrogen into the circumferential edge region is represented by pre-contouring the optical element. During pre-contouring, a blank made of fused silica, from which the optical element is cut or manufactured, is produced. The blank has protective beads or bulges for protecting the circumferential edge region. The protective beads or bulges are removed after loading such that the circumferential edge region has a lower hydrogen content in relation to the optically used region since the hydrogen can diffuse (almost) directly into the inner, optically used region.

In general, the subsequently optically used inner region is milled prior to loading with hydrogen to an excess thickness of only e.g. approximately 0.2 mm to 1 mm per surface or per side. This subsequently renders it possible to remove a glass region, which was possibly contaminated during loading, without damage. Moreover, the removal of the beads (and the excess thickness in the inner region) can be carried out with a lower precision. Moreover, highly precise machining of an optical surface, which lies within a circumferential bead, is extremely difficult or extremely laborious. In a further embodiment, the ratio between the minimum hydrogen content in the inner region and a mean hydrogen content in the circumferential edge region is greater than three. The mean hydrogen content is understood to mean the hydrogen content averaged over the volume of the circumferential edge region. It is understood that the ratio between the minimum hydrogen content in the inner region and the minimum hydrogen content in the edge region is also greater than three.

In a further embodiment, the fused silica has an OH content of less than 60 ppm by weight, preferably of less than 1 ppm by weight. Such a so-called "dry" fused silica with a low OH content is well-suited for lithography and is typically produced, sintered and subject to stress -relief annealing in a soot process, before it is loaded with hydrogen.

In addition to a low OH content, fused silica which is used for microlithography should also have a low number of defects, i.e. the content of ODC and peroxy-centres should not be too great; this can be achieved, for example, by controlled oxidative healing of defect centres. Moreover, the fictive temperature T_f of the fused silica should not be too high and, for example, be less than 1100 °C or optionally less than 1000 °C. The fictive temperature T_f is a parameter which characterizes the specific network structure of the fused silica and can be determined by e.g. IR spectroscopy or Raman scattering, as described in e.g.

US 7, 928, 026 B2.

In a further embodiment, the optical element, when irradiated by pulsed laser radiation at a wavelength of 193 nm, has a change dK/dF in the absorption coefficient K as a function of the energy density F of less than 5xl0^~4 cmxpulse/mJ, preferably of less than lxlO^"4 cmxpulse/mJ, particularly preferably of less than 5xlCT⁵ cmxpulse/mJ, in particular of less than lxlO^"5 cmxpulse/mJ .

One requirement for fused silica for a lithography system at an operating wavelength of approximately 193 nm, in particular for immersion lithography, lies in only admitting a low silane content since fused silica with a high silane content has strong polarization- induced birefringence and compaction under irradiation with higher energy densities and rarefaction under irradiation with lower energy densities, as well as a reducing transmission during the irradiation. The silane content of fused silica can only be determined comparatively imprecisely by means of Raman spectroscopy: the detection limit lies at approximately 1 to 5xl0¹⁵ molecules/cm³, which typically constitutes a value that is too high for lithography applications. US 7,928,026 B2 describes an indirect method, in which the gradient dK/dF of the change of transmission K as a function of the energy density F is used as a measure for the silane content. A low silane content, and therefore a small change in the gradient dK/dF, can be achieved by virtue of the fused silica being loaded with hydrogen at "cold" temperatures of at most 600 °C, normally at approximately 400°C to 500°C. Loading at lower temperatures of no more than 300°C is also possible in order to generate the smallest possible gradient dK/dF or a silane content that is as low as possible. The fused silica material also has a small (reversible) gradient dK/dF in the circumferential edge region which has not, or only slightly, been loaded with hydrogen. However, in the edge region, the fused silica material has high (irreversible) ageing dK/dN (N: number of cycles) , which makes measuring the gradient dK/dF more difficult.

A further aspect of the invention relates to an optical manipulator, comprising an optical element as described above, as well as an actuation device for reversibly changing the surface form of the optical element, wherein the actuation device has a plurality of actuators for mechanically acting on the optical element at a plurality of contact faces, which are formed at the circumferential edge region of the optical element.

At the contact faces, the actuators are connected to the optical element in a secure manner or in a manner secured against shearing, i.e. in a manner that enables the introduction of tensile stress or compressive stress into the optical element. The actuators typically work in the manner of levers in order, by way of tilting the levers, to transmit both tensile stress and compressive stress onto the material of the optical element and bend the latter. In the circumferential edge region which surrounds the optically used inner region, the actuators, more precisely the contact faces of the actuators, engage on the optical element.

In one embodiment, at least one actuator has an actuatable foot, at which the contact face is formed for connecting the actuator with the surface of the optical element. The foot, which is securely connected to the surface of the optical element, forms a lever, on the side of which facing away from the contact face an actuating component, for example in the form of a piezo-element, can engage in order to apply a lateral force onto the foot. Additionally or alternatively, the foot can also be moved in a direction perpendicular to the surface of the optical element in order to generate a compressive force or deflection of the optical element in the region of the contact face. The optical element is tensioned and bent by adjusting a foot or lever, or, in general, a plurality of feet or levers, as a result of which higher orders of the wavefront can be influenced in a targeted manner. In one development, the contact face of the actuatable foot is adhesively bonded, welded or soldered onto a surface of the optical element. The foot serving as a lever can be connected at the contact face to the typically planar surface of the optical element with the aid of joining means, for example by means of an adhesive. A circumferential groove can serve to divert tensions away from the adhesive. To the extent that the generally organic adhesive does not have a sufficient shearing strength or if adhesives may not be used for reasons of UV degradation or outgassing, the foot can also be soldered or welded onto a metallization layer. Alternatively, the feet or levers can also be screwed.

In order to distribute tensions introduced into the optical element during the actuation in an improved spatial manner, it is possible to apply an adhesive layer, which is fibre-reinforced, onto the surface of the optical element. The fibre-reinforced adhesive layer in this case typically extends beyond the diameter of the foot or beyond the contact face to the outside. Smearing of the tensions in the optical element can also be achieved by the selection of the diameter and the thickness profile of the employed solder, contributing to an increase in the service life of the optical element.

If the lever or the foot at which the contact face is formed is made of aluminium, steel or a high-strength ceramic and if a connection that is sufficiently secured against shearing can be obtained with the glass of the optical element, this results in a composite material with increased strength since most of the aforementioned materials have a higher fatigue strength than the glass material of the optical element. However, the joining means, i.e. the adhesive or the metallization layer and/or the solder, can be weakened by the emerging hydrogen, which is why the circumferential edge region of the optical element, at which the actuators engage, should have a (significantly) lower hydrogen content than the inner, optically used region.

A further aspect relates to a microlithographic projection lens comprising at least one optical manipulator, embodied as described above, for the wavefront correction of the projection lens. In contrast to conventional optical manipulators, the wavefront correction can be implemented by the actuation of the optical element in real time during the exposure process by virtue of the optical element being tensioned or bent. In this manner, it is possible to influence individual orders of the wavefront, i.e. individual Zernike coefficients, in a targeted manner. The projection lens is typically embodied for used radiation in the UV wavelength range. The optical element is introduced into the beam path of the projection lens as an additional component. Here, for example, the optical element can be arranged between the object plane with the mask to be imaged and the first lens element of the projection lens, or at a different location. It is understood that it is possible to arrange the optical element at a different location in the beam path of the projection lens.

A further aspect relates to a microlithographic projection exposure apparatus comprising a projection lens embodied as described above. Such a projection exposure apparatus is typically provided for scanning operation, i.e. the object field, in which the structure to be imaged is arranged, is scanned during the exposure process. The optical manipulator serves for the wavefront manipulation or for correcting aberrations in real time during the scanning process and should have a service life of several years. Over the course of this service life, there should be no weakening of the mechanical connections between the contact faces of the actuators and the optical element, since, otherwise, it may no longer be possible, or no longer be possible in the desired manner, to bend the optical element for wavefront correction purposes .

Further features and advantages of the invention emerge from the following description of exemplary embodiments of the invention, on the basis of the figures in the drawing, which show details essential to the invention, and from the claims. In a variant of the invention, the individual features can be implemented either individually on their own or a number thereof can be implemented together in any combination.

Drawing

Exemplary embodiments are depicted in the schemat drawing and explained in the following description. detail : gure 1 shows a schematic illustration of an exampl of a microlithographic projection exposure apparatus , gures 2a, b show schematic illustrations

exemplary embodiment of an optical manipulator with a plate-shaped optical element and an actuation device in a side view and a top view,

Figures 3a-d show schematic illustrations of a plurality of options for fastening a foot of an actuator of the actuation device to the surface of the optical element of Figures 2a, b, Figures 4a, b show schematic illustrations of two options for loading an optical element of the optical manipulator with hydrogen, and Figures 5a, b show schematic illustrations of the plate-shaped optical element (in Figure 5a) and (in Figure 5b) the distribution of the hydrogen content of the plate- shaped optical element in the thickness direction along two sections in accordance with Figure 5a.

In the following description of the drawings, identical reference signs are used for equivalent or functionally equivalent components.

Figure 1 shows an example of a microlithographic projection exposure apparatus WSC, which is applicable in the production of semiconductor components and other finely structured components and which operates with light or electromagnetic radiation from the deep ultraviolet (DUV) range for obtaining resolutions down to fractions of micrometres. An ArF excimer laser with a used wavelength λ of approximately 193 nm serves as primary radiation source or light source LS . Other UV laser light sources, e.g. an F₂ laser with the working wavelength of 157 nm or an ArF excimer laser with a working wavelength of 248 nm are likewise possible.

In the emergence surface ES thereof, an illumination system ILL disposed downstream of the light source LS generates a large, sharply delimited and substantially homogeneously illuminated illumination field, which is adapted to the telecentricity requirements of the projection lens PO arranged therebehind in the light path. The illumination system ILL has devices for setting different illumination modes (illumination settings) and can, for example, be switched between conventional on-axis illumination with a different degree of coherence σ and off-axis illumination. The off-axis illumination modes for example comprise an annular illumination or a dipole illumination or a quadrupole illumination or a different type of multi- pole illumination. The setup of suitable illumination systems is known per se and is therefore not explained in any more detail here. The patent application US 2007/0165202 Al (corresponding to WO 2005/026843 A2) shows examples for illumination systems which can be used within the scope of various embodiments.

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

A device RS for holding and manipulating the mask M (reticle) is arranged behind the illumination system in such a way that the pattern arranged at the reticle lies in the object plane OS of the projection lens PO which coincides with the emergence plane ES of the illumination system and is also referred to here as reticle plane OS. In this plane, the mask is movable in a scanning direction (y-direction) perpendicular to the optical axis OA (z-direction) with the aid of a scanner drive for the purposes of a scanning operation.

The projection lens PO follows behind the reticle plane OS, which projection lens acts as a reduction lens and images an image of the pattern arranged at the mask M with a reduced scale, for example with the scale of 1:4 (|β| = 0.25) or 1:5 (|β| = 0.20), on a substrate W covered by a photoresist layer, the light-sensitive substrate surface SS of which substrate lies in the region of the image plane IS of the projection lens PO.

The substrate to be exposed which, in the exemplary case, is a semiconductor wafer W is held by a device WS, which, comprises a scanner drive in order to move the wafer synchronously with the reticle M perpendicular to the optical axis OA in a scanning direction (y-direction) . The device WS, which is also referred to as "wafer stage", and the device RS, which is also referred to as "reticle stage", are components of a scanner device, which is controlled by means of a scan control device, which is integrated in the central control device CU of the projection exposure apparatus in this embodiment.

The illumination field generated by the illumination system ILL defines the effective object field OF used during the projection exposure. In the exemplary case, the latter is rectangular, has a measured height A parallel to the scanning direction (y-direction) and a measured width B>A perpendicular thereto (in the x- direction) . The aspect ratio AR = B/A generally lies between 2 and 10, in particular between 3 and 6. The effective object field lies at a distance in the y- direction next to the optical axis (off-axis field) . The effective image field, optically conjugate to the effective object field, in the image plane 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 lens, i.e. A* = | β | A and B* = Ιβ|Β.

If the projection lens is designed and operated as an immersion lens, a thin layer of an immersion liquid is transilluminated during the operation of the projection lens, which immersion liquid is situated between the emergence surface of the projection lens and the image plane IS. During immersion operation, image-side numerical apertures NA > 1 are possible. A configuration as a dry lens is also possible; in this case, the image- side numerical aperture is restricted to values NA < 1. Under these conditions, which are typical for highly resolving projection lenses, projection radiation with a relatively large numerical aperture, for example with values greater than 0.15 or greater than 0.2 or greater than 0.3, is present in the region of some or all field planes (object plane, image plane, possibly one or more intermediate image planes) of the projection lens.

The projection lens PO shown in Figure 1 is a catadioptric projection lens, which has a first, purely refractive lens part, a second, catadioptric lens part with a concave mirror CM and a third, purely refractive lens part. The second lens part has a deflection device, which is embodied in the style of a prism and which has a first planar deflection mirror FM1 for reflecting projection radiation coming from the object plane OS to the concave mirror CM and a second deflection mirror FM2 , aligned at a right angle to the first deflection mirror FM1, for deflecting the projection radiation reflected by the concave mirror CM in the direction of the image plane IS. Immersion lenses with a comparable basic design are shown e.g. in the international patent application WO 2004/019128 A2. During immersion operation, image-side numerical apertures NA > 1 are possible. A configuration as a dry lens is also possible; in this case, the image-side numerical aperture is restricted to values NA < 1.

The projection lens or the projection exposure apparatus is equipped with an optical wavefront manipulation system WFM, which is configured to dynamically modify the wavefront of the projection radiation propagating from the object plane OS to the image plane IS, the optical effect of the wavefront manipulation system WFM being able to be set variably by way of control signals. The wavefront manipulation system WFM of the exemplary embodiment has an optical manipulator MAN with an optical element OE which is arranged in the direct vicinity of the object plane OS of the projection lens PO in the projection beam path and the surface form SF of which can be changed reversibly with the aid of an actuation device DR.

For further explanation purposes, Figure 2a shows a schematic longitudinal section through the optical manipulator MAN in an xz -plane. The optical manipulator MAN has a plate -shaped optical element OE made of a material transparent to the projection radiation, for example synthetic fused silica. In the shown example, the optical element OE is embodied as a plane plate and has a thickness of less than 5 mm, typically of approximately 2 mm to 3 mm. The surface form SF of the optical element OE, which is plane without the effect of the actuation device, is modified by the effect of the actuation device DR into the wave- shaped surface form SF shown in Figure 2a. In order to generate the wave-shaped surface form SF, the actuation device DR comprises a plurality of actuators AK that are actuatable independently of one another and depicted in a simplified manner in the form of circles in Figure 2b. The actuators AK engage on the optical element OE in a circumferential, frame-shaped edge region BA outside an optically used region FP shown in Figure 2b. The outer boundary of the optically used region FP is depicted by a dashed line in Figure 2a. The optically used region FP is illuminated by the beams coming from the effective object field OF. The optically used region FP is also referred to as "footprint". Since the optical element OE is arranged in the vicinity of the object field OF, the optical region FP used by the projection radiation substantially has the rectangular form of the illuminated object field OF, with the corner regions being slightly rounded-off . The actuators AK are arranged outside of the optically used region FP in order to prevent the projection radiation from being scattered or reflected at the actuators AK. The actuators AK engage at the plate- shaped optical element OE in such a manner that the latter can be bent and the surface form SF can be brought into a defined wave form. Here, both the "amplitude" of the waves measured parallel to the z- direction, i.e. the deflection of the optical element OE in the z -direction, and the distance between adjacent wave crests measured in the x-direction, i.e. the wavelength or period of the wave pattern, can be set to have different values. In the exemplary case, a sinusoidal profile is set in the x-direction.

Figure 2b schematically shows that the surface form of the optical element OE in this direction has a plurality of local maxima (represented by wave crests in Figure 2a and by "+" signs in Figure 2b) and a plurality of local minima (represented by wave troughs in Figure 2a and signs in Figure 2b) lying therebetween. It is understood that the arrangement of the actuators AK shown in Figure 2b is merely exemplary and that the actuators AK can also be arranged differently around the optically used region FP in order to bend the optical element OE in the desired manner .

For the purposes of acting on the optical element OE, the actuators AK each have a foot F, which is depicted in Figures 3a-d. At the lower end thereof, the foot F has a contact face K, which serves for a (secure) connection with a first surface 01 of the optical element OE facing the foot F. An actuation element of the actuator AK, which can e.g. be embodied as a piezo- element, engages on the foot F on the end thereof facing away from the surface 01 of the optical element OE. What can be seen from Figures 3a-d is that the foot F acts as a lever since the force acting on the foot F in an upper portion of the foot F has a force component in the lateral direction. In the example depicted in Figure 3a, the foot F is adhesively bonded to the surface 01 of the optical element OE, i.e. an adhesive layer 10 has been introduced between the contact face K and the surface 01, which adhesive layer in the shown example is applied to a protective layer 11 for protecting the adhesive layer 10 from stray radiation in the UV wavelength range from the optical element OE. The protective layer 11 consists of a material that is impermeable to radiation in the VUV wavelength range.

If the force visualized by an arrow in Figure 3a acts on the foot F, a tensile stress is transmitted from the foot F onto the material of the optical element OE, which tensile stress assumes a maximum value in an impact region IA which in the shown example partly overlaps with the region at the surface 01 of the optical element OE covered by the contact face K or by the adhesive layer 10. The impact region IA is formed on the opposite side of the foot F to the effective direction of the lateral force. Since the effective direction of the force of the actuator AK engaging at the foot F can typically be varied virtually as desired in the xy-plane, the impact region IA, in which a maximum tensile stress occurs in the optical element OE, typically extends in a ring-shaped manner around the contact face K.

In order to divert the tensile stress away from the adhesive layer 10, the foot F or the contact face K thereof is surrounded by an annulus-shaped groove N in the optical element OE shown in Figure 3b, which groove is milled into the surface 01 of the optical element OE. In this case, the impact region IA is formed within the groove N, i.e. the maximum tensile stresses during the deflection of the foot F that serves as a lever occur within the groove N. Figure 3c shows a further option for fastening the foot F to the surface 01 of the optical element OE, in which a bore through which a screw 12 is inserted has been introduced into the optical element OE, which screw is screwed into a female thread formed in the foot F. A shim 13, 14 has respectively been introduced between the foot F and the first surface 01 of the optical element OE and between the head of the screw 12 and the second surface 02 of the optical element OE. In the foot F depicted in Figure 3d, a metallization layer 15 has been applied to the surface 01 of the optical element OE in order to solder or weld the foot F thereon by means of a solder 16. It is understood that, as an alternative to the examples shown in Figures 3a-d, the actuators AK or the feet F thereof can also engage on both mutually opposite surfaces 01, 02 of the optical element OE.

In all cases shown in Figures 3a-d, a connection that is secured against shearing is established between the foot F and the optical element OE in order to enable the bending of the optical element OE. The central control device CU of the projection exposure apparatus WSC acts continuously on the actuation device DR of the optical manipulator MAN during the exposure operation of the projection exposure apparatus WSC in order to correct wavefront errors in real time.

Since the fused silica material of the optical element OE is transilluminated by the projection radiation within the optically used region FP, it is necessary, after the production in a soot process, the annealing and the drying to an OH content of less than 60 ppm by weight, preferably less than 1 ppm by weight, to load the optical element OE with hydrogen. The hydrogen loading serves to saturate E' F-centres in the fused silica material of the optical element OE, which would otherwise absorb laser radiation at a wavelength of 193 nm.

Hydrogen loading of the optical element OE is implemented in an oven, in which a hydrogen- containing atmosphere (H₂ content of more than 5%) is prevalent. The oven is heated to a constant temperature T of more than 250°C and of typically less than 500°C so that the hydrogen H₂ can diffuse into the optical element OE from the surroundings, as depicted in Figures 4a, b. The duration of the loading with hydrogen H₂ depends on the desired hydrogen content [H₂] in the volume of the optical element OE and on the selected temperature T during the loading with hydrogen H₂. Loading typically takes several weeks or months; loading at low temperatures of e.g. approximately 300°C may possibly also take longer. What is advantageous for the loading duration in the present case is that the plate- shaped optical element OE has a small thickness of typically no more than about 5 mm so as to enable the bending.

Since the circumferential edge region BA does not transmit the projection radiation and consequently there is no damage from a decrease in the transmission in the circumferential edge region BA, which is caused by e.g. stray light from the optically used region FP, loading of the optical element OE with hydrogen H₂ is merely required within the optically used region FP. It would be possible to assume that loading of the circumferential edge region BA with hydrogen is non- damaging, and so the whole optical element OE can be loaded uniformly with hydrogen. However, this is not the case since loading the circumferential edge region BA with hydrogen H₂ can be damaging, namely for the following reason: the hydrogen H stored during loading already diffuses outward at room temperature when the optical element OE is stored for a relatively long period of time or used in the projection lens PO for a relatively long period of time of e.g. a number of years. The outwardly diffusing hydrogen H₂ can weaken the connection between the contact faces K of the actuators AK and the surface 01 (and 02) of the optical element OE, particularly if the connection is brought about by way of the adhesive layer 10 shown in Figure 3b, but possibly also in the case where the connection is implemented by way of a metallization layer 15 and a solder 16.

In order to avoid, or at least reduce, an inward diffusion of hydrogen H₂ into the circumferential edge region BA during the loading, in the example shown in Figure 4a, a circumferential frame 20a, 20b is placed onto the optical element OE in the circumferential edge region BA on both the first surface 01 and the second surface 02 , which frame covers the edge region BA in a planar manner. Figure 4a additionally shows a sealing ring between the circumferential frames 20a, 20b and the respective surfaces 01, 02 , which sealing ring is intended to prevent the ingress of hydrogen into the region between the frames 20a, 20b and the circumferential edge region BA. In place of the sealing ring, it is also possible to introduce sealing substances between the circumferential frames 20a, 20b and the optical element OE. Alternatively, the circumferential frames 20a, 20b can also be placed directly onto the surfaces 01, 02. The circumferential frames 20a, 20b can be formed from fused silica or a different material in which the inward diffusion of hydrogen progresses more slowly than in fused silica. A further option for reducing the inward diffusion of hydrogen H₂ into the circumferential edge region BA lies in the pre-contouring of a blank R made of fused silica, from which the optical element OE is subsequently cut. Such a pre-contoured blank R is depicted in Figure 4b. In the circumferential edge region BA, the blank R has bulges in the form of circumferential protective beads 20a, 20b, which take up the inwardly diffusing hydrogen H₂ and thereby protect the circumferential edge region BA lying therebelow from the inward diffusion of hydrogen H₂ during loading. The optically used region FP also has an excess thickness (which is, however, significantly lower) of typically approximately 0.2 mm to approximately 1.0 mm in order to prevent inward diffusion of hydrogen H₂. After loading, the protective beads 20a, 20b and the small excess thickness at the optically used region FP are removed such that the plate- shaped optical element OE formed by the removal has a uniform thickness and the circumferential edge region BA of the optical element OE has a lower hydrogen content [H₂] in relation to the optically used region FP. Removing the excess thickness and the protective beads 20a, 20b is followed by fine machining and polishing of the optical element OE .

Figure 5a shows the plate-shaped optical element OE of Figure 4b after removing the protective beads 20a, 20b. Since loading with hydrogen H₂ is implemented at a temperature T of no more than 500°C, in particular of no more than approximately 300°C, the fused silica material only has a low silane content. A measure for the silane content is represented by the change dK/dF of the absorption coefficient K of the fused silica as a function of the energy density F. As a result of the "cold" loading, the optical element OE has a change dK/dF of less than 5xl0^~4 cmxpulse/mJ, preferably of less than lxlO^"4 cmxpulse/mJ, in particular of less than 5x10⁵ cmxpulse/mJ, optionally even of less than lxlO^"5 cmxpulse/mJ .

The distribution of the hydrogen content [H₂] in the optical element OE of Figure 5a along the thickness direction z of the optical element OE is shown in Figure 5b for two sections E, F and F' within the circumferential edge region BA (section E) and within the optically used region (section F, F' ) · In the shown example, the plate- shaped optical element OE has a thickness D of 2.5 mm. The hydrogen content [H₂] (in molecules/cm³) along the section F has a substantially parabolic, symmetrical distribution about the longitudinal central plane or at the half thickness D/2 of the plate-shaped optical element OE (at z = 1.25 mm) . It is likewise possible to identify in Figure 5b that the distribution of the hydrogen content [H₂] along the section F has a maximum at the two surfaces 01, 02 (i.e. at z = 0 mm and z = 2.5 mm) , with the maximum hydrogen content [H₂] in the shown example lying at lxlO¹⁸ molecules/cm³.

Section F shows the distribution of the hydrogen content [H₂] immediately after removing the protective beads 20a, 20b. Section F' shows the distribution of the hydrogen content [H₂] at the same location but at a later time, namely after performing a heating process on the optical element OE, during which the latter is heated, for example in order to apply a coating, e.g. a metallization layer, or any other functional coating, for example one or more antireflection layers. As a result of the heating process, there is a reduction in the hydrogen content [H₂] in the surface-near volume regions, i.e. there was a depletion of hydrogen H₂ in the surface-near volume regions near the surfaces 01, 02. Nevertheless, in the thickness direction Z, the optical element OE has a hydrogen content [H₂] of at least 2xl0¹⁷ molecules/cm³ in the inner region FP, at least up to a distance di of 0.25 mm from the first surface 01 (at z = 0 mm) and from the second surface 02 (at z = 2.5 mm) , as can be seen in Figure 5b. It is essential that the hydrogen content [H₂] does not fall, or does not fall significantly, below the minimum hydrogen content [H₂] in the centre of the optical element OE, even in the vicinity of the surfaces 01, 02, after many years of operation as a result of depletion of hydrogen H₂.

The hydrogen content [H₂] reduces significantly in the volume region lying between the two surface-near volume regions (i.e. between z = 0.25 mm and z = 2.25 mm), namely to a minimum hydrogen content [H₂] which lies at 5xl0¹⁶ molecules/cm³ in the shown example. The value of the minimum hydrogen content [H₂] depends, inter alia, on the glass type, the expected energy density and the specified number of laser pulses or the specified service life of the optical element OE. The hydrogen content [H₂] , shown in Figure 5b, in the region of the second section E, which lies in the circumferential edge region BA, is much smaller compared thereto and lies at no more than 5xl0¹⁵ molecules/cm³ or, if the protective beads 20a, 20b shown in Figure 4b are used, at no more than approximately lxlO¹⁵ molecules/cm³. As a result of masking-off or pre-contouring, a lower hydrogen content [H₂] is present in the circumferential edge region BA at each thickness (each z-value) and in each section in the thickness direction Z than in the inner, optically used region FP. Typically, the ratio between the minimum hydrogen content [H₂] in the optically used region FP and a mean (or minimum) hydrogen content [H₂] in the circumferential edge region BA is greater than three. It is understood that the optically used region FP is optionally not completely illuminated by the projection radiation such that possibly no projection radiation passes through the optical element OE in a tightly delimited outer edge of the optically used region FP.

What is essential is that there is no significant outward diffusion of hydrogen within the circumferential edge region BA, in which the actuators AK engage at the optical element OB with the contact faces K thereof. This renders it possible to ensure that the connection between the contact faces K of the actuator AK and the optical element OE is not weakened, such that a permanent connection secured against shearing is maintained between the surface OE and the contact faces K and the desired bending of the optical element OE by the actuators AK can still be implemented with the desired accuracy after a period of several years after the production thereof.

Claims

Patent Claims

Optical element (OE) made of fused silica,

characterized

in that that the optical element (OE) has an inner region (FP) that is continuous in a thickness direction (z) , which inner region is surrounded by a circumferential edge region (BA) that is continuous in the thickness direction (z) , wherein a hydrogen content ( [H₂] ) in the circumferential edge region (BA) is less than a hydrogen content ( [H₂] ) in the inner region (FP) .

Optical element according to Claim 1, which has a thickness (D) of less than 5 mm.

Optical element according to Claim 1 or 2 , which is a plate-shaped optical element (OE) .

Optical element according to one of the preceding claims, in which the minimum hydrogen content ( [H₂] ) in the inner region (FP) is more than lxl0¹⁶ molecules/cm³, preferably more than 5xl0¹⁶ molecules/cm³.

Optical element according to one of the preceding claims, which has a first surface (01) and a second surface (02) that is opposite to the first surface in the thickness direction (z) , wherein the hydrogen content ( [H₂] ) in the inner region (FP) , at least up to a distance (di) of 0.25 mm from the first and second surface (01, 02) , is at least lxlO¹⁶ molecules/cm³, preferably at least 5xl0¹⁶ molecules/cm³, particularly preferably at least 2xl0¹⁷ molecules/cm³.

6. Optical element according to one of the preceding claims, in which the maximum hydrogen content ( [H₂] ) in the circumferential edge region (BA) is no more than 5xl0¹⁵ molecules/cm³, preferably no more than IxlO¹⁵ molecules/cm³.

Optical element according to one of the preceding claims, in which the ratio between the minimum hydrogen content ( [H₂] ) in the inner region (FP) and a mean hydrogen content ( [H₂] ) in the circumferential edge region (BA) is greater than three .

Optical element according to one of the preceding claims, which has an OH content of less than 60 ppm by weight, preferably of less than 1 ppm by weight .

Optical element according to one of the preceding claims, which, when irradiated by pulsed laser radiation at a wavelength of 193 nm, has a change dK/dF in the absorption coefficient K as a function of the energy density F of less than 5xl0^"4 cmxpulse/mJ, preferably of less than IxlO^"4 cmxpulse/mJ, particularly preferably of less than 5xl0^~5 cmxpulse/mJ, in particular of less than IxlO^"5 cmxpulse/mJ .

Optical manipulator (MAN) , comprising:

an optical element (OE) according to one of the preceding claims, as well as an actuation device (DR) for reversibly changing the surface form (SF) of the optical element (OE) , wherein the actuation device (DR) has a plurality of actuators (AK) for mechanically acting on the optical element (OE) at a plurality of contact faces (K) , which are formed at the circumferential edge region (BA) of the optical element (OE) . Optical manipulator according to Claim 10, in which at least one actuator (AK) has an actuatable foot (F) , at which a contact face (K) is formed for connecting the actuator (AK) with the surface (Ol) of the optical element (OE) .

Optical manipulator according to Claim 11, in which the contact face (K) of the actuatable foot (F) is adhesively bonded, welded or soldered onto a surface (01) of the optical element (OE) .

Microlithographic projection lens (PO) , comprising :

at least one optical manipulator (MAN) according to one of Claims 10 to 12 for the wavefront correction of the projection lens (PO) .

Microlithographic projection exposure apparatus ( SC) , comprising a projection lens (PO) according to Claim 13.