Method for producing mirror of microlithographic projection exposure apparatus

Abstract

The invention relates to a method for producing a mirror (26) of a microlithographic projection exposure apparatus (1), a first mirror part (27) and a second mirror part (28) are provided, which are in contact in the region of a first connection surface (29) of the first mirror part (27) and a second connection surface (30) of the second mirror part (28). In order to form a permanent connection between the first mirror part (27) and the second mirror part (28), the first mirror part (27) and the second mirror part (28) are heated to a holding temperature of at least 400 DEG C and are held at this holding temperature for a holding time. After the holding time has elapsed, the first mirror part (27) and the second mirror part (28) are cooled to a first cooling temperature at a first cooling rate of at most 100 K / h.

Description

Field of the Invention

[0001] The present invention relates to a method for manufacturing a mirror of a microlithographic projection exposure apparatus. The present invention also relates to a mirror of a microlithographic projection exposure apparatus, an illumination optical unit, a projection optical unit, and a microlithographic projection exposure apparatus. Background Art

[0002] Microlithographic projection exposure apparatuses are particularly used for the production of semiconductors and generally have an illumination optical unit and a projection optical unit. The illumination optical unit generates a desired light distribution from the light of a light source for the illumination of a reticle, which is also commonly referred to as a mask. In this case, light should be understood in the general sense of electromagnetic radiation, i.e., there is no limitation to a specific wavelength. Thus, hereinafter, the terms "light" and "radiation" are used as synonyms, i.e., a light source may also be referred to as a radiation source, a light distribution may also be referred to as a radiation distribution, and so on. Using the projection optical unit, the reticle is imaged onto a photosensitive material, such as a wafer or another substrate, which is particularly made of a semiconductor material. In this way, the photosensitive material is exposed in a structured manner in the pattern predefined by the reticle. Since the reticle has tiny structural elements that are intended to be transferred onto the substrate with high accuracy, it is required that the illumination optical unit precisely and reproducibly generates the desired light distribution, and that the projection optical unit precisely and reproducibly performs the imaging.

[0003] In addition to other optical elements, the illumination optical unit and the projection optical unit may have at least one mirror in the optical path, which deflects light in a predetermined manner by reflection on its optical surface. The specific manner in which the light deflection occurs depends on the formation of the optical surface, in particular its shape. For example, the optical surface may be formed as a metal layer or a series of layers with alternating refractive indices. Regardless of the formation of the optical surface, perfect reflection cannot be achieved, so that part of the light is always absorbed and causes the mirror to heat up. The heating up, in turn, is accompanied by thermal expansion and thus causes deformation of the mirror, in particular of the optical surface, which may ultimately lead to a change in the optical properties of the mirror.

[0004] In order to minimize the influence of the heating of the mirror on its optical properties as much as possible, the mirror can be made of a material with an extremely low coefficient of thermal expansion. Such a material can have a zero-crossing temperature at which the thermal expansion is particularly low, ideally even zero, in the vicinity of this zero-crossing temperature. Therefore, when using such a material, it is generally advisable to try to operate the mirror at or near its zero-crossing temperature. This, in turn, has the effect that the zero-crossing temperature should correspond exactly, or should be slightly higher than, the temperature assumed by the mirror under the expected operating conditions. The advantage of a zero-crossing temperature slightly higher than the expected operating temperature of the mirror is that local heating of the mirror caused by the maximum value of the light intensity will only result in a very small expansion of the mirror.

[0005] It is known from US 10427974 B2 that the coefficient of thermal expansion of doped quartz glass and its increase with respect to temperature can be influenced by heat treatment.

[0006] For example, if the mirror is intended to have cooling channels, or if the mirror is intended to be larger than the blank available for manufacturing the mirror, it may be advantageous, or even necessary, to form the mirror from more than one component.

[0007] A multi-component mirror with cooling channels is known from DE102020208648.6, which is not a prior publication.

[0008] One problem with a mirror formed from more than one component is how to connect the individual components of the mirror to meet the high requirements of a lithographic optical component. Summary of the Invention

[0009] The present invention is based on the object of forming a multi-component mirror of a microlithographic projection exposure apparatus that meets very high optical requirements and has great dimensional stability under the expected operating conditions. In particular, the mirror undergoes very little thermally induced deformation under the expected operating conditions.

[0010] This object is achieved by the combination of features of claim 1.

[0011] In the case of a method according to the invention for manufacturing a mirror of a microlithographic projection exposure apparatus, a first mirror component and a second mirror component are provided, and the first mirror component and the second mirror component come into contact in the region of a first connecting surface of the first mirror component and a second connecting surface of the second mirror component. In order to form a permanent connection between the first mirror component and the second mirror component, the first mirror component and the second mirror component are heated to a holding temperature of at least 400 °C and held at this holding temperature for a holding time. After the holding time, the first mirror component and the second mirror component are cooled to a first cooling temperature at a first cooling rate of at most 100 K / h.

[0012] The method according to the invention has the advantage of allowing a very durable connection to be formed between the mirror components. The process of forming such a connection is also referred to hereinafter as bonding. Another advantage of the method according to the invention is that the multi-component mirror of the microlithographic projection exposure apparatus manufactured therefrom meets very high optical requirements and can be formed such that it has great dimensional stability under the intended operating conditions. In particular, the mirror can be formed such that it undergoes very small thermally induced deformations under the intended operating conditions. This can be achieved by a mirror formed at the zero-crossing temperature, where the zero-crossing temperature is averaged over the relevant volume and has a very small deviation from a predetermined value and / or the zero-crossing temperature has only a very small spatial variation within this relevant volume and is thus very uniform. Of particular relevance here are the regions of the mirror volume that are arranged at a very small distance from its optical surface such that they can deform the mirror to an unacceptable extent.

[0013] An important finding of the present invention is that the heat treatment for connecting the mirror components has an impact on the zero-crossing temperature of the mirror, and this should be taken into account when designing the heat treatment in order to be able to manufacture a mirror of particularly high quality. In other words, by bonding, the zero-crossing temperature of the material used to manufacture the mirror components, which has been strongly set to a desired value, can be changed. For this reason, in the method according to the invention, for example, a relatively low cooling rate is selected, so that the heat treatment process can be monitored precisely at any time and as much as possible throughout the entire volume of the mirror components.

[0014] For example, the advantage of a low cooling rate is that when the cooling rate changes, the temperature difference in different regions of the volume of the mirror components is relatively small. This in turn results in a relatively small non-uniformity of the zero-crossing temperature of the mirror. Another advantage is that the temperature error of the oven in which the mirror components are held at a certain temperature or the temperature error of the mirror components themselves does not have such a strong impact on the uniformity of the zero-crossing temperature. A low cooling rate generally has a greater impact on the absolute value of the zero-crossing temperature than a high cooling rate. However, this has been taken into account when manufacturing the material of the mirror components.

[0015] The heat treatment can be carried out in an air or inert gas environment. Similarly, the heat treatment can also be carried out under vacuum conditions.

[0016] The first mirror part and the second mirror part can be provided in the form of an intermediate product which is produced by wringing the first connecting surface of the first mirror part and the second connecting surface of the second mirror part. This makes it easier to perform heat treatment because the mirror parts do not have to be handled separately and there is no need to laboriously position the mirror parts correctly relative to each other during the heat treatment process. In addition, if wringing has been successfully performed previously, the likelihood of successful joining is high because during the joining process, the connecting surfaces of the mirror parts must be very closely adjacent to each other and the connecting surfaces must meet joining prerequisite conditions similar to those of wringing. Furthermore, it is possible to investigate whether there are possible defects in the intermediate product before heat treatment. For example, a defect may be an inclusion between the connecting surfaces. If a defect is found, the intermediate product may be rejected or otherwise processed. Thus, complex heat treatment of defective intermediate products can be avoided.

[0017] During the wringing process, the first mirror part and the second mirror part can be joined together such that a first partial region of the first connecting surface and a partial region of the second connecting surface come into contact and form a common contact surface, and the size of the contact surface is increased by continuing to join the first mirror part and the second mirror part in the rolling direction.

[0018] The transition region between the common contact surface on which the second mirror part has been wrung onto the first mirror part and the non-wrung regions of the first and second connecting surfaces is also referred to as the wringing front. The effect of continuing to join the first mirror part and the second mirror part together is that the wringing front advances in a monitored manner in the rolling direction. The advantage of this continuous wringing is that the relative alignment of the two mirror parts with respect to each other can be very precisely monitored throughout the wringing operation and unwanted spontaneous wringing can be avoided. In addition, at any time during the wringing operation, there is a fluid connection between the region of the wringing front and the surrounding region, so it is ensured that gas can escape from this region at any time and gas inclusions between the connecting surfaces can be largely prevented.

[0019] The first mirror part and the second mirror part can be joined such that contact with the first mirror part first occurs in a region adjacent to the second surface of the outer contour of the second mirror part. Similarly, the first mirror part and the second mirror part can also be joined together such that contact with the first mirror part first occurs in the central region of the second connecting surface. This enables the way in which the wringing operation is performed to be adapted to the corresponding situation, in particular the formation of the second mirror part.

[0020] On the side of the second mirror part facing away from the second connecting surface, a region for forming an optical surface can be provided.

[0021] The first connecting surface of the first mirror element and / or the second connecting surface of the second mirror element may have a curvature. A curved first connecting surface and / or second connecting surface is advantageous, especially when it is envisaged to form a curved optical surface on the second mirror element, because in this way a constant thickness is possible, for example, and thus a uniform deformability of the second mirror element is possible. In addition, a constant distance can be easily established between the cooling channels and the optical surface. The curvature of the first connecting surface and / or the second connecting surface may correspond to the curvature of the intended optical surface of the mirror.

[0022] It is also possible that the first mirror element has a first connecting surface which is formed in a planar manner or has a large mean radius of curvature in at least one direction, in particular a mean radius of curvature greater than 10 m. Then, a thin second mirror element can be used, in particular a second mirror element with a thickness of 0.5 mm to 10 mm, which has a second connecting surface formed in a planar manner. If the second mirror element has cooling channels, the thickness is determined from the bottom of the cooling channels. The advantage of such a thin second mirror element is that its second connecting surface can be deformed so as to adapt to the shape of the first connecting surface of the first mirror element. This means that there is no need to laboriously manufacture a second mirror element with a curved second connecting surface.

[0023] For a wavelength λ of 633 nm, the deviation of the shape of the first connecting surface of the first mirror element and / or the second connecting surface of the second mirror element from the predetermined shape may be less than λ / 2, preferably less than λ / 10. The first connecting surface of the first mirror element and / or the second connecting surface of the second mirror element may have a roughness of less than 5 nm, preferably less than 1 nm. In this case, the term "roughness" refers to the randomly distributed deviation from a smooth surface. An accurately formed and smooth connecting surface makes twisting and joining easier, reduces the risk of gas inclusions and increases the load-carrying capacity of the connection. Before twisting or joining, one of the two connecting surfaces can be adapted to the actual shape of the other of the two connecting surfaces. In this way, the accuracy can be further improved.

[0024] It is also possible that the curvature of the first connecting surface is formed convex and the curvature of the second connecting surface is formed concave, or the curvature of the first connecting surface is formed concave and the curvature of the second connecting surface is formed convex, and when the first mirror element and the second mirror element are joined together, the absolute value of the mean radius of curvature of the convex curvature is less than the absolute value of the mean radius of curvature of the concave curvature. This has the advantage that local continuous twisting can be ensured, because the distance between the two connecting surfaces increases as the distance from the twisting front increases. The risk of unwanted twisting at a certain distance from the twisting front is reduced. In addition, the risk of gas inclusions is also reduced.

[0025] Before heating to the holding temperature, the intermediate product can be stored at a storage temperature of at least 80 °C and at most 400 °C for a storage time of at least 2 hours, preferably at least 12 hours, particularly preferably at least 24 hours. Preferably, the storage temperature is at least 160 °C, particularly preferably at least 250 °C. Additionally, the storage temperature is preferably at most 350 °C. The advantage of storing at these temperatures is that excess moisture diffuses out from the connection surface area of the mirror component, and the strength is increased. In particular, anhydrous materials are beneficial for the formation of covalent bonds during the subsequent heat treatment at the holding temperature. If a relatively low value in the range of, for example, 400 °C to 550 °C is selected as the holding temperature, then storage at a low storage temperature is carried out for a longer time, for example, storing for one to two weeks at a holding temperature of at most 250 °C.

[0026] In particular, the first mirror component and the second mirror component are heated to a holding temperature of at least 500 °C, preferably at least 600 °C, particularly preferably at least 700 °C. With the assistance of a high holding temperature, a particularly durable connection of the mirror components can be achieved because a particularly large number of covalent bonds are formed.

[0027] The first mirror component and the second mirror component can be heated to a holding temperature of at most 900 °C. In addition to the above advantages, a high holding temperature is also accompanied by the disadvantage that the heat treatment has a greater impact on the zero-crossing temperature. By the above maximum value of the holding temperature, a good compromise can be achieved.

[0028] The first mirror component and the second mirror component can be heated to the holding temperature at a heating rate of at most 150 K / h, preferably at most 100 K / h. The advantage of slow heating is that less stress is formed between the inner and outer regions of the mirror component, and there is more time for the outward diffusion of water.

[0029] The holding time can be at least 8 hours, preferably at least 12 hours, particularly preferably at least 20 hours. The advantage of a long holding time is that the connection between the mirror components has greater tensile strength because more covalent bonds are formed. For efficient manufacturing, the holding time is preferably at most 72 hours, particularly preferably at most 48 hours.

[0030] The first cooling temperature can be 700 °C or lower. Specifically, the first cooling temperature can be 600 °C or lower. The first cooling rate is preferably 50 K / h or lower. Specifically, the first cooling rate is between 1 K / h and 20 K / h. The advantage of a lower cooling rate is that the first mirror component and the second mirror component are cooled relatively uniformly throughout the volume. Higher spatial temperature gradients are avoided, and the temperature change over time in the entire volume is similar. This in turn leads to a relatively uniform distribution of the zero-crossing temperature in the mirror.

[0031] After cooling to the first cooling temperature at the first cooling rate, the first mirror member and the second mirror member may be cooled to a second cooling temperature at a second cooling rate greater than the first cooling rate, and the second cooling temperature is lower than the first cooling temperature. The second cooling temperature may be 500 °C or lower, preferably 400 °C or lower. The second cooling rate may be 150 K / h or less. In particular, the second cooling rate falls between 5 K / h and 100 K / h. After cooling to the second cooling temperature, the first mirror member and the second mirror member may be further cooled in the direction of room temperature. The further cooling may be carried out at a third cooling rate which at least initially has a greater value than the second cooling rate. Using a greater cooling rate at a low cooling temperature can significantly accelerate the heat treatment speed without significantly losing the quality of the mirror because at a low cooling temperature, the influence of the cooling curve on the zero-crossing temperature of the mirror is significantly lower than at a high cooling temperature.

[0032] The first mirror member and the second mirror member may be cooled such that the sliding average of the cooling rate formed within a temperature interval of 20 K has the same relative change over time throughout the volume of the first mirror member and / or the second mirror member within the temperature range from 900 °C to 700 °C, except for a deviation of at most 2 K / h. This means that the cooling can be carried out in such a way that, after taking into account the corresponding possible time differences, the change over time of this sliding average at all positions of the first mirror member and / or the second mirror member differs by at most 2 K / h. Preferably, the deviation of the sliding average within the temperature range from 900 °C to 700 °C is at most 1 K / h. In addition, it may be stipulated that the deviation of the sliding average within the temperature range from 700 °C to 500 °C is at most 4 K / h, preferably at most 2 K / h. With the assistance of this stipulation, a consistently high quality of the manufactured mirror can be achieved. In this case, by having a slightly larger tolerance at a low temperature, the heat treatment can in turn be accelerated without any major adverse effects. In addition, the first mirror member and the second mirror member may be cooled such that the non-uniformity of the zero-crossing temperature of the mirror due to heat treatment is at most + / −1 K, preferably at most + / −0.5 K. These tolerances should in particular be maintained within the volume region of the mirror that laterally overlaps with the optical surface of the mirror and extends in depth from the optical surface to the cooling channel (i.e., from the optical surface to the starting point of the cooling channel). Similarly, these tolerances may also be maintained within the volume region of the mirror that laterally overlaps with the optical surface of the mirror and extends in depth from the optical surface to 1 cm below the cooling channel.

[0033] The first cooling rate and / or the second cooling rate can be selected according to the material composition of the first mirror component and / or the second mirror component. Specifically, the first cooling rate and / or the second cooling rate are selected according to the OH content and / or the titanium content of the first mirror component and / or the second mirror component. Since the influence of heat treatment on the zero-crossing temperature of the mirror depends on the material composition of the mirror component, the mirror performance, in particular the achievable accuracy of its zero-crossing temperature, can be improved by taking the material composition into account when designing the heat treatment.

[0034] The first cooling rate and / or the second cooling rate can be selected according to the desired average zero-crossing temperature of the mirror. In this way, a mirror with an average zero-crossing temperature with a very small deviation from a predetermined value can be manufactured. Specifically, the first cooling rate and / or the second cooling rate can be selected according to the desired average zero-crossing temperature within a volume region of the mirror that laterally overlaps the optical surface of the mirror and extends in depth from the optical surface to the cooling channel or 1 cm below the cooling channel.

[0035] To determine the first cooling rate and / or the second cooling rate, a material sample of the material used to manufacture the first mirror component and / or the second mirror component can be analyzed. It can be specifically specified that the material sample undergoes all the heat treatments of the material of the first mirror component and / or the second mirror component in the same manner. In this way, the first cooling rate and / or the second cooling rate can be adjusted very precisely to suit the material composition of the first mirror component and / or the second mirror component, without any risk of damaging the mirror component, and thus a very high accuracy and a very low variance can be achieved for the average zero-crossing temperature of the manufactured mirror.

[0036] The present invention also relates to a mirror of a microlithographic projection exposure apparatus, which has a first mirror component, a second mirror component, and an optical surface for light reflection. The first mirror component and the second mirror component are permanently connected to each other in the region of the first connection surface of the first mirror component and the second connection surface of the second mirror component by heat treatment. The mirror has an average zero-crossing temperature within a volume region that laterally overlaps the optical surface and extends in depth from the optical surface to 1 cm, and the maximum deviation of the average zero-crossing temperature from a predetermined value is + / - 1 K. In this volume region, the predetermined value may decrease as the distance from the optical surface increases.

[0037] Despite the influence of heat treatment on the zero-crossing temperature, such an accurate average zero-crossing temperature can still be achieved by connecting the first mirror component and the second mirror component to each other by means of the method according to the present invention. Preferably, the average zero-crossing temperature deviates from the predetermined value by at most + / -0.5 K, and particularly preferably at most + / -0.2 K.

[0038] Such a high accuracy would hardly be possible if the influence of the heat treatment taken into account in the method according to the invention, by means of which the mirror components are connected to one another, on the zero-crossing temperature of the mirror was not taken into account.

[0039] The first connection surface of the first reflector component and / or the second connection surface of the second reflector component may have a curvature. In this case, the curvature of the first connection surface of the first reflector component may not be formed rotationally symmetrically, so it has different average curvature radii, and / or the curvature of the second connection surface of the second reflector component may not be formed rotationally symmetrically, so it has different average curvature radii.

[0040] The cooling channel may be formed in the first reflector component and / or in the second reflector component. In this case, the reflector may have an average zero-crossing temperature in a volume region which overlaps laterally with its optical surface and extends in depth from the optical surface to the cooling channel (i.e. from the optical surface to the start of the cooling channel) which deviates from a predetermined value by a maximum of + / -1 K, preferably by a maximum of + / -0.5 K, particularly preferably by a maximum of + / -0.2 K. If the reflector has a cooling channel, this condition may replace the above-mentioned condition, which is independent of any cooling channel involving a volume region from the optical surface to a depth of 1 cm, so that in the case of a reflector with a cooling channel, the volume region may extend to a depth of less than 1 cm. Similarly, the reflector may also have an average zero-crossing temperature in a volume region which overlaps laterally with its optical surface and extends in depth from the optical surface to 1 cm below the cooling channel, which deviates from a predetermined value by a maximum of + / -1 K, preferably by a maximum of + / -0.5 K, particularly preferably by a maximum of + / -0.2 K. In the volume region mentioned, the predetermined value may in each case decrease with increasing distance from the optical surface. The cooling channel may be formed to be open toward the first connection surface of the first reflector component and / or toward the second connection surface of the second reflector component.

[0041] In each case, the cooling channel can be configured such that its greatest longitudinal extension lies transversely in the region for the optical surface. This allows effective cooling of regions which are particularly susceptible to temperature-induced deformations of the optical surface.

[0042] In addition, the cooling channel can extend parallel to the optical surface. This can make the cooling of the optical surface particularly uniform. Specifically, the cooling channel can reach a distance of 1 to 6 mm from the optical surface. The shape deviation between the second connecting surface of the second reflector component and the optical surface can be a maximum of 3 mm, preferably no more than 20 μm.

[0043] In an improvement, the net distance between adjacent cooling channels does not exceed 15 mm, preferably does not exceed 5 mm. The small distance between the cooling channels enables effective and uniform cooling, and can contribute to the deformability of the second mirror component during torsion, especially when the rolling direction is transverse to the cooling channels of the second mirror component. Specifically, the values of the width of the cooling channels, the net distance between adjacent cooling channels, and the distance between the cooling channels and the optical surface can be substantially the same.

[0044] The lateral dimension of the cooling channels can be 0.2 to 10 mm. The depth of the cooling channels (i.e., the lateral dimension substantially perpendicular to the first connection surface of the first mirror component or the second connection surface of the second mirror component) can be greater than the width of the cooling channels (i.e., the lateral dimension perpendicular to the depth). Specifically, the depth of the cooling channels can be more than twice the width of the cooling channels. Therefore, in the case of the same flow cross-section, the risk of deformation of the mirror optical surface caused by the fluid pressure in the cooling channels is smaller because the proportion of the surface area of the cavity formed by the cooling channels parallel to the optical surface region is smaller than that of wide cooling channels with a small depth.

[0045] The first mirror component can have a greater thickness than the second mirror component. This enables effective cooling of the optical surface while achieving good mechanical stability of the mirror.

[0046] The invention also relates to an illumination optical unit having a mirror formed according to the invention and / or manufactured by a method according to the invention.

[0047] The invention also relates to a projection optical unit having a mirror formed according to the invention and / or manufactured by a method according to the invention.

[0048] Finally, the invention relates to a microlithography projection exposure apparatus having an illumination optical unit according to the invention and / or a projection optical unit according to the invention. Description of the Drawings

[0049] The invention will be explained in more detail below based on exemplary specific embodiments represented in the drawings, wherein:

[0050] Figure 1 An exemplary specific embodiment of a projection exposure apparatus for EUV projection lithography is schematically shown in a meridional section;

[0051] Figure 2 A schematic diagram of an exemplary specific embodiment of a projection exposure apparatus for DUV projection lithography is shown;

[0052] Figure 3 An exemplary element of a mirror manufactured by the method according to the invention is shown in a schematic cross-section;

[0053] Figure 4 Shown in a further schematic cross - section Figure 3 a first exemplary embodiment of the mirror shown;

[0054] Figure 5 shows a flow chart for explaining the procedure when connecting the lower part and the upper part according to the present invention; and

[0055] Figure 6 shows a snapshot during the process of twisting the upper part to the lower part in a schematic cross - section. Detailed Description of the Invention

[0056] Figure 1 Schematically shows an exemplary embodiment of a projection exposure apparatus 1 for EUV projection lithography in a meridional cross - section.

[0057] Hereinafter, first by referring to Figure 1 examples to describe the basic components of the microlithography projection exposure apparatus 1. The description of the basic structure of the projection exposure apparatus 1 and its components should not be construed as limiting.

[0058] In addition to the light source or radiation source 3, a specific embodiment of the illumination system 2 of the projection exposure apparatus 1 further has an illumination optical unit 4 for illuminating the object field 5 in the object plane 6. In an alternative specific embodiment, the radiation source 3 can also be provided as a module independent of the rest of the illumination system. In this case, the illumination system does not include the radiation source 3.

[0059] Exposes a mask blank 7 disposed in the object field 5. The mask blank 7 is held by a mask blank holder 8. The mask blank holder 8 can be displaced, particularly in the scanning direction, by a mask blank displacement driver 9.

[0060] For the sake of explanation, a Cartesian xyz coordinate system is shown in Figure 1 . The x - direction is perpendicular to the drawing plane. The y - direction travels horizontally, and the z - direction travels vertically. The scanning direction travels along the Figure 1 y - direction in. The z - direction is perpendicular to the object plane 6.

[0061] The projection exposure apparatus 1 includes a projection optical unit 10. The projection optical unit 10 is used to image the object field 5 onto an image field 11 in an image plane 12. The image plane 12 is parallel to the object plane 6. Alternatively, the angle between the object plane 6 and the image plane 12 may also be different from 0°.

[0062] The results show that the structures on the mask blank 7 are imaged onto the photosensitive layer of the wafer 13 or some other substrate, and the wafer 13 is arranged in the region of the image field 11 in the image plane 12. The wafer 13 is held by a wafer holder 14. The wafer holder 14 can be displaced by a wafer displacement drive 15, in particular in the y direction. The displacement of the mask blank 7 is effected on the one hand by a mask blank displacement drive 9 and the displacement of the wafer 13 is effected on the other hand by the wafer displacement drive 15, and these two displacements can be effected in a synchronized manner with each other.

[0063] The radiation source 3 is an EUV radiation source. The radiation source 3 emits illumination radiation 16, which is also referred to hereinafter as use radiation or illumination light. In the exemplary embodiment shown, the illumination radiation 16 has a wavelength in the EUV range, in particular in the range between 5 nm and 30 nm. The radiation source 3 can be a plasma source, such as an LPP source (laser-produced plasma) or a GDPP source (gas-discharge-produced plasma). It can also be a synchrotron-based radiation source. Similarly, the radiation source 3 can be a free electron laser (FEL).

[0064] The illumination radiation 16 emitted from the radiation source 3 is focused by a condenser 17. The condenser 17 can be a condenser having one or more ellipsoidal and / or hyperboloidal reflecting surfaces. The illumination radiation 16 can be incident on at least one reflecting surface of the condenser 17 at a grazing incidence (GI) (i.e., the angle of incidence is greater than 45°) or at a normal incidence (NI) (i.e., the angle of incidence is less than 45°). The condenser 17 can be structured and / or coated, on the one hand for optimizing its reflectivity for the illumination radiation 16 and on the other hand for suppressing stray light.

[0065] Downstream of the condenser 17, the illumination radiation 16 propagates through an intermediate focus in an intermediate focal plane 18. The intermediate focal plane 18 can represent the separation between the radiation source module (having the radiation source 3 and the condenser 17) and the illumination optical unit 4.

[0066] The illumination optical unit 4 includes a deflecting mirror 19 and a first faceted mirror 20 arranged downstream of it in the beam path. The deflecting mirror 19 can be a planar deflecting mirror or a mirror having a beam influencing effect that goes beyond a pure deflecting effect. Alternatively or additionally, the deflecting mirror 19 can be implemented as a spectral filter that separates the use light wavelength of the illumination radiation 16 from stray light having a wavelength deviated therefrom. If the first faceted mirror 20 is arranged in the plane of the illumination optical unit 4 that is optically conjugate to the field plane and the object plane 6, it is also referred to as a field faceted mirror. The first faceted mirror 20 includes a plurality of individual first facets 21, which are also referred to hereinafter as field facets. Figure 1 Only some of these facets 21 are shown by way of example.

[0067] The first facet 21 can be implemented as a macro facet, particularly as a rectangular facet or a facet with an arcuate or partially circular outer contour. The first facet 2 can be implemented as a planar facet or alternatively as a convexly or concavely curved facet.

[0068] As is known, for example, from DE 10 2008 009 600 A1, the first facet 21 itself can also in each case consist of a plurality of individual mirrors, particularly a plurality of micromirrors. The first facet mirror 20 can in particular be formed as a microelectromechanical system (MEMS system). For further details, reference is made to DE 10 2008 009 600 A1.

[0069] Between the condenser 17 and the deflecting mirror 19, the illumination radiation 16 propagates horizontally, i.e., in the y direction.

[0070] In the beam path of the illumination optical unit 4, the second facet mirror 22 is arranged downstream of the first facet mirror 20. If the second facet mirror 22 is arranged in the pupil plane of the illumination optical unit 4, it is also referred to as a pupil facet mirror. The second facet mirror 22 can also be arranged at a certain distance from the pupil plane of the illumination optical unit 4. In this case, the combination of the first facet mirror 20 and the second facet mirror 22 is also referred to as a specular reflector. Specular reflectors are disclosed in US 2006 / 0132747 A1, EP 1 614 008 B1, and US 6,573,978.

[0071] The second facet mirror 22 comprises a plurality of second facets 23. In the case of a pupil facet mirror, the second facets 23 are also referred to as pupil facets.

[0072] The second facets 23 can likewise be macro facets, which can, for example, have a circular, rectangular, or hexagonal outer periphery, or facets consisting of micromirrors. In this regard, reference is also made to DE 10 2008 009 600 A1.

[0073] The second facets 23 can have a planar or alternatively convexly or concavely curved reflective surface.

[0074] Thus, the illumination optical unit 4 forms a two-facet system. This basic principle is also referred to as a honeycomb condenser (fly's eye integrator).

[0075] It may be advantageous not to arrange the second facet mirror 22 completely in a plane that is optically conjugate to the pupil plane of the projection optical unit 10. In particular, the second pupil facet mirror 22 can be arranged inclined with respect to the pupil plane of the projection optical unit 10, as described, for example, in DE 10 2017 220 586 A1.

[0076] With the assistance of the second partial mirror 22, the individual first partials 21 are imaged into the object field 5. The second partial mirror 22 is the last beam-forming mirror, or in fact the last mirror in the beam path before the object field 5 for the illumination radiation 16.

[0077] In a further specific embodiment of the illumination optical unit 4 (not shown), a transmission optical unit that particularly helps to image the first partial 21 into the object field 5 can be arranged in the beam path between the second partial mirror 22 and the object field 5. The transmission optical unit can have exactly one mirror, or can have two or more mirrors, which are arranged one after another in the beam path of the illumination optical unit 4. The transmission optical unit can particularly include one or two normal-incidence mirrors (NI mirrors) and / or one or two grazing-incidence mirrors (GI mirrors).

[0078] In Figure 1 In the illustrated specific embodiment, the illumination optical unit 4 has exactly three mirrors downstream of the collector 17, specifically the deflection mirror 19, the field partial mirror 20, and the second partial mirror 22.

[0079] In a further specific embodiment of the illumination optical unit 4, the deflection mirror 19 is also not required, so the illumination optical unit 4 can have exactly two mirrors downstream of the collector 17, specifically the first partial mirror 20 and the second partial mirror 22.

[0080] Imaging the first partial 21 into the object plane 6 by means of the second partial 23 or using the second partial 23 and the transmission optical unit is generally only an approximate imaging.

[0081] The projection optical unit 10 includes a plurality of mirrors Mi, which are successively numbered according to their arrangement in the beam path of the projection exposure apparatus 1.

[0082] In Figure 1 In the illustrated example, the projection optical unit 10 includes six mirrors M1 to M6. Alternatives with four, eight, ten, twelve, or any other number of mirrors Mi are equally possible. The projection optical unit 10 is a double-masking optical unit. The penultimate mirror M5 and the last mirror M6 each have a through-hole, through which the radiation that contributes to the exposure is transmitted from the mask reticle 7 to the wafer 13 during the exposure of the wafer 13. The projection optical unit 10 has an image-side numerical aperture greater than 0.5, which can also be greater than 0.6, and can be, for example, 0.7 or 0.75.

[0083] The reflective surface of the mirror Mi can be implemented as a freeform surface without a rotational axis of symmetry. Alternatively, the reflective surface of the mirror Mi can be designed as an aspherical surface with exactly one rotational axis of symmetry. Similar to the mirrors of the illumination optical unit 4, the mirror Mi can have a highly reflective coating for the illumination radiation 16. These coatings can be designed as multilayer coatings, in particular alternating layers of molybdenum and silicon.

[0084] The projection optical unit 10 has a large object-image offset in the y-direction between the y-coordinate of the center of the object field 5 and the y-coordinate of the center of the image field 11. This object-image offset in the y-direction can be approximately the same size as the z-distance between the object plane 6 and the image plane 12.

[0085] In particular, the projection optical unit 10 can have an anamorphic form. Specifically, it has different imaging ratios βx, βy in the x- and y-directions. The two imaging ratios βx, βy of the projection optical unit 10 are preferably (βx, βy) = (+ / -0.25, + / - 0.125). A positive imaging ratio β means imaging without image inversion. A negative sign of the imaging ratio β indicates imaging with image inversion.

[0086] Therefore, the projection optical unit 10 results in a size reduction with a ratio of 4:1 in the x-direction (i.e., in the direction perpendicular to the scanning direction).

[0087] The projection optical unit 10 results in a size reduction by a factor of 8:1 in the y-direction (i.e., in the scanning direction).

[0088] Other imaging ratios are also possible. Imaging ratios with the same sign and the same absolute value in the x- and y-directions are also possible, for example, with an absolute value of 0.125 or 0.25.

[0089] In the beam path between the object field 5 and the image field 11, the number of intermediate image planes in the x- and y-directions can be the same or (depending on the specific embodiment of the projection optical unit 10) can be different. US 2018 / 0074303 A1 discloses an example of a projection optical unit 10 with different numbers of such intermediate images in the x- and y-directions.

[0090] In each case, one of the second facets 23 is assigned to exactly one of the first facets 21 to respectively form illumination channels for illuminating the object field 5. This can generate illumination particularly according to the Köhler principle. With the aid of the first facet 21, the far field is decomposed into a plurality of object fields 5. The first facet 21 produces a plurality of images of the intermediate foci on the second facets 23 assigned to it respectively.

[0091] Via the assigned second facet 23, the first facet 21 is imaged onto the mask master 7 in each case in a mutually superimposed manner for illuminating the object field 5. The illumination of the object field 5 is particularly as uniform as possible. The uniformity error is preferably less than 2%. By superimposing different illumination channels, field uniformity can be achieved.

[0092] The illumination of the entrance pupil of the projection optical unit 10 can be geometrically defined by the configuration of the second facet 23. The intensity distribution in the entrance pupil of the projection optical unit 10 can be set by selecting the illumination channels guiding the light, in particular a subset of the second facet 23. This intensity distribution is also referred to as illumination setting or illumination pupil filling.

[0093] Similarly good pupil uniformity in the defined illumination segment area of the illumination pupil of the illumination optical unit 4 can be achieved by redistributing the illumination channels.

[0094] Further aspects and details of the illumination of the object field 5 and in particular of the entrance pupil of the projection optical unit 10 will be described below.

[0095] In particular, the projection optical unit 10 can have a homocentric entrance pupil. The latter can be accessible. It can also be inaccessible.

[0096] The entrance pupil of the projection optical unit 10 generally cannot be precisely illuminated using the second facet mirror 22. When imaging the projection optical unit 10, it images the center of the second facet mirror 22 eccentrically onto the wafer 13, and the aperture rays generally do not intersect at a single point. However, a region can be found where the distance between the pairwise determined aperture rays becomes minimal. This region represents the entrance pupil or the region in real space conjugate thereto. In particular, this region has a finite curvature.

[0097] The projection optical unit 10 may have different entrance pupil positions for the tangential beam path and the sagittal beam path. In this case, imaging elements (in particular the optical components of the transmission optical unit) should be arranged between the second facet mirror 22 and the mask master 7. With the aid of this optical element, the different positions of the tangential entrance pupil and the sagittal entrance pupil can be taken into account.

[0098] In Figure 1 In the configuration of the components of the illumination optical unit 4 shown, the second facet mirror 22 is arranged in the region conjugate to the entrance pupil of the projection optical unit 10. The first facet mirror 20 is arranged such that it is inclined with respect to the object plane 6. The first facet mirror 20 is arranged such that it is inclined with respect to the configuration plane defined by the deflection mirror 19.

[0099] The first facet mirror 20 is arranged such that it is inclined with respect to the configuration plane defined by the second facet mirror 22.

[0100] Figure 2 Shows a schematic diagram of an exemplary specific embodiment of a projection exposure apparatus 1 for DUV projection lithography. DUV stands here for "deep ultraviolet". In particular, the projection exposure apparatus 1 can be designed to operate at a wavelength of 193 nm.

[0101] The projection exposure apparatus 1 has an illumination optical unit 4 and a projection optical unit 10. The internal structure of the illumination optical unit 4 and the internal structure of the projection optical unit 10 are not shown in detail here, which can each contain, for example, optical components, sensors, manipulators, etc. In the case of the projection optical unit 10, the mirror M is shown as representative of its optical components. The mirror M can be cooled by means of a cooling medium, which is provided by a cooling device 24. The cooling medium is a fluid, for example water. Supplementally or alternatively, the illumination optical unit 4 can have a cooled mirror M and an associated cooling device 24. The projection optical unit 10 and / or the illumination optical unit 4 can also have a plurality of cooled mirrors M and cooling devices 24. In the case of the illumination optical unit 4 and in the case of the projection optical unit 10, lenses and further mirrors (cooled or uncooled) can be present, for example, as additional optical components.

[0102] Similarly, in Figure 1 at least one cooling device 24 can also be provided in the exemplary specific embodiment of the projection exposure apparatus 1 shown, which can be connected, for example, to the mirror M3.

[0103] The radiation required for the operation of the projection exposure apparatus 1 is generated by a radiation source 3. The radiation source 3 can in particular be an excimer laser, for example an argon fluoride laser, which generates illumination radiation 16 with a wavelength of 193 nm.

[0104] A mask holder 8 is provided between the illumination optical unit 4 and the projection optical unit 10, on which a mask blank 7 (also called a mask) is fixed. The mask holder 8 has a mask blank displacement drive 9. Viewed from the radiation direction, the wafer holder 14 is arranged downstream of the projection optical unit 10. The wafer holder holds a wafer 13 or some other substrate and has a wafer displacement drive 15.

[0105] Figure 2 A control device 25 is also shown, which is connected to the illumination optical unit 4, the projection optical unit 10, the cooling device 24, the radiation source 3, the mask holder or the mask blank displacement drive 9, and the wafer holder 14 or the wafer displacement drive 15. Similarly, Figure 1 the projection exposure apparatus of

[0106] The projection exposure apparatus 1 is used to image the mask reticle 7 onto the wafer 13 with high precision. For this purpose, the mask reticle 7 is illuminated with the assistance of the illumination optical unit 4, and the illuminated mask reticle 7 is imaged onto the wafer 13 with the assistance of the projection optical unit 10. Specifically, the following procedure is adopted:

[0107] The illumination optical unit 4 converts the illumination radiation 16 generated by the radiation source 3 in a precisely defined manner through its optical components and guides it onto the mask reticle 7. According to a specific embodiment, the illumination optical unit 4 can be configured such that it illuminates the entire mask reticle 7 or only a partial area of the mask reticle 7. The illumination optical unit 4 is capable of illuminating the mask reticle 7 such that almost the same illumination conditions exist at each illumination point on the mask reticle 7. In particular, the intensity and angular distribution of the incident illumination radiation 16 are almost the same for each illumination point on the mask reticle 7.

[0108] The illumination optical unit 4 is capable of selectively illuminating the mask reticle 7 with illumination radiation 16 having a variety of different angular distributions. These angular distributions of the illumination radiation 16 are also referred to as illumination settings. The required illumination setting is usually selected according to the structural elements formed on the mask reticle 7. For example, relatively frequently used are dipole or quadrupole illumination settings, in which case the illumination radiation 16 is incident on each illumination point of the mask reticle 7 from two different directions or from four different directions, respectively. Depending on the form of the illumination optical unit 4, different illumination settings can be generated, for example, by a combination of different diffractive optical elements and a zoom axicon optical unit, or by a mirror array (which in each case has a plurality of small mirrors that are arranged adjacent to each other and can be adjusted individually with respect to their angular positions).

[0109] The mask reticle 7 can be formed as, for example, a glass plate that is transparent to the illumination radiation 16 provided by the illumination optical unit 4 and is provided with opaque structures (for example, in the form of a chromium coating).

[0110] The projection exposure apparatus 1 can be configured such that the entire mask reticle 7 is simultaneously illuminated by the illumination optical unit 4 and is completely imaged onto the wafer 13 in a single exposure step through the projection optical unit 10.

[0111] Alternatively, the projection exposure apparatus 1 may also be configured such that only a partial area of the mask blank 7 is illuminated by the illumination optical unit 4 at the same time, and the mask blank displacement driver 9 is activated by the control device 25 such that during the exposure of the wafer 13, the mask blank 7 moves relative to the illumination optical unit 4, so that the illuminated partial area as a whole moves on the mask blank 7. By appropriately adjusting the activation of the wafer displacement driver 15 to move the wafer 13 synchronously, taking into account the imaging characteristics of the projection optical unit 10 as well, the respectively illuminated partial areas on the mask blank 7 are imaged onto the partial areas provided for it on the wafer 13. This movement of the mask blank 7 and the wafer 13 is also referred to as scanning.

[0112] In two specific embodiments of the projection exposure apparatus 1, in order to be able to transfer the latent image generated by the exposure of the wafer 13 into a physical structure, a photosensitive layer is applied to the wafer 13. The image of the mask blank 7 is formed in this photosensitive layer by exposure, and a permanent structure can be generated from it on the wafer 13 by means of subsequent chemical processes.

[0113] The mask blank 7 is usually imaged not only once but multiple times adjacent to each other on the wafer 13. For this purpose, after each imaging of the mask blank 7 onto the wafer 13, the wafer holder 14 is laterally moved in a manner corresponding to the size of the image of the mask blank 7 on the wafer 13. In each case, the imaging of the mask blank 7 can be carried out either as a whole or sequentially by scanning. The chemical treatment of the wafer 13 only begins when the required number of imagings of the mask blank 7 has been carried out on the wafer 13.

[0114] Figure 3 The cross-sectional schematic diagram shows an exemplary specific embodiment of the mirror 26 manufactured by the method of the present invention. Figure 4 The same exemplary specific embodiment of the mirror 26 is shown in a further cross-sectional schematic diagram, where the cross-section is taken along Figure 3 the cross-section line A-A depicted in. In order to be able to represent as many details of the mirror 26 as possible, the cross-section line A-A has a lateral offset when passing through the mirror 26. In Figure 3 the cross-section is taken along Figure 4 the cross-section line B-B shown in. In Figure 3 and Figure 4 the representation of the mirror 26 and its component parts is not to scale and is highly abstract in order to present the present invention as clearly as possible. This also applies to all other figures.

[0115] In Figure 1 and Figure 2In the case of one of the projection exposure apparatuses 1 shown, a mirror 26 can be used and it has a lower part 27 and an upper part 28, each of which is made of fused silica, titanium-doped fused silica or glass-ceramics in each case. The terms "lower part" and "upper part" are chosen because the lower part 27 is usually much thicker than the upper part 28, so it is as if it carries the upper part 28. However, these terms have nothing to do with the orientation of the mirror 26 in the installed state relative to the direction of gravity. During operation of the projection exposure apparatus 1, the upper part 28 can be arranged above, below or beside the lower part 27 relative to the direction of gravity, or assume some other relative position relative to it. The lower part 27 is also referred to as the first mirror part 27, and the upper part 28 is also referred to as the second mirror part 28.

[0116] The lower part 27 and the upper part 28 are rigidly connected to each other in the region of the connecting surface 29 of the lower part 27 and the connecting surface 30 of the upper part 28 by the method of the present invention described in more detail below. In the exemplary embodiment shown, the connecting surface 29 of the lower part 27 is formed in a concave-curved manner. The curvature can be spherical, aspherical or formed according to a free-form surface. The connecting surface 30 of the upper part 28 is curved in a manner complementary to the connecting surface 29 of the lower part 27, and thus has a convex curvature, which can be spherical, aspherical or formed according to a free-form surface. Thus, the connecting surface 30 of the upper part 28 and the connecting surface 29 of the lower part 27 can be joined in close contact with each other. As an alternative to the curved formation, the connecting surface 29 of the lower part 27 and the connecting surface 30 of the upper part 28 can also be formed in a planar manner.

[0117] On the side facing away from its connecting surface 29, as shown at the bottom of Figure 3 , the lower part 27 is formed in a planar manner. On the side facing away from its connecting surface 30, as shown at the top of Figure 3 , the upper part 28 is formed in a concave-curved manner and has a reflective optical surface 31 with the same curvature. The curvature can be formed as spherical, aspherical or according to a free-form surface, and particularly corresponds to the curvature of the connecting surface 30 of the upper part 28. Even though the connecting surfaces 29 and 30, the surface of the upper part 28 facing away from the connecting surface 30 and the optical surface 31 are each formed in a rotationally symmetric manner, they have slightly different curvatures in the cross-sectional regions of Figure 3 and Figure 4 respectively, because these cross-sectional regions are formed at different distances from the centers of the said surfaces. However, compared with the absolute value of the radius of curvature, the differences in distance are very small in each case, so different curvatures are not perceptible in the representations of Figure 3 and Figure 4 .

[0118] As an alternative to this, it is also possible for the optical surface 31 to be formed in a planar manner. This is particularly the case when the connecting surface 29 of the lower part 27 and the connecting surface 30 of the upper part 28 are formed in a planar manner.

[0119] The optical surface 31 is embodied as a coating applied to the upper part 28. The formation of the coating depends on at which wavelength the optical surface 31 is intended to produce its reflection effect. In the case of a mirror M having the required reflection in the DUV range, i.e., Figure 2 in the case of the mirror M, the coating can be formed as an aluminum layer. On the other hand, if reflection is intended in the EUV range, for example, Figure 1 in the case of the mirror M3, the coating can in particular be formed from alternately successive layers of silicon and molybdenum and may contain one or more additional layers of different composition, for example, as a protective layer.

[0120] The upper part 28 also has a number of elongated cooling channels 32, which are parallel to each other and parallel to the optical surface 31 and extend transversely in the region of the optical surface 31 and may slightly exceed it. Thus, in the case of the exemplary embodiment shown, the cooling channels 32 are formed in a curved manner. The cooling channels 32 reach a distance of approximately 1 to 6 millimeters from the optical surface 31 and are formed to open towards the connecting surface 30 of the upper part 28. Thus, the cooling channels 32 have elongated cooling channel openings 33 in the region of the connecting surface 30 of the upper part 28. The lateral dimension of the cooling channels 32 can be approximately 0.2 to 10 millimeters, and the depth of the cooling channels 32 (i.e., the dimension substantially perpendicular to the connecting surface 30 of the upper part 28) is generally much greater than the width (i.e., the dimension substantially parallel to the connecting surface 30 of the upper part 28). For example, the depth of the cooling channels 32 can be more than twice the width of the cooling channels 32. The net distance between adjacent cooling channels 32 generally does not exceed 15 millimeters, preferably not more than 5 millimeters. The values of the width of the cooling channels 32, the net distance between adjacent cooling channels 32, and the distance between the cooling channels 32 and the optical surface 31 can be approximately the same.

[0121] In the case where the connecting surface 30 of the upper part 28 is formed to be non-rotationally symmetric, it may have different average radii of curvature for different azimuth angles, or in other words, with respect to different directions. In this case, the region of the connecting surface 30 of the upper part 28 in which the absolute value of the average radius of curvature has a maximum can extend in a direction transverse to (in particular, perpendicular to) the longitudinal extent of the cooling channel openings 33. The mutually adjacent regions of the connecting surface 29 of the lower part 27 and the connecting surface 30 of the upper part 28 may have slightly different average radii of curvature in the unconnected state, and the average radius of curvature of the convex surface region generally has a smaller absolute value compared to the average radius of curvature of the corresponding region of the concave surface.

[0122] The lower part 27 has a plurality of distributor channels 34 and collector channels 35 which extend from the connection surface 29 of the lower part 27 in a direction away from the connection surface 29 of the lower part 27 towards the inside of the lower part 27. Inside the lower part 27, the distributor channels 34 lead to a fluid distributor 36 and the collector channels 35 lead to a fluid collector 37. The distributor channels 34 and the collector channels 35 widen in a stepped manner towards the connection surface 29 of the lower part 27 in each case and overlap with opposite ends of two cooling channel openings 33 respectively. In other words, one distributor channel 34 overlaps with one end of two cooling channel openings 33 respectively, and one collector channel 35 overlaps with the other two ends of the same two cooling channel openings 33 respectively. Accordingly, one distributor channel 34 is fluidly connected to the fluid distributor 36 at one end and to two cooling channels 32 at the other end. Similarly, one collector channel 35 is fluidly connected to the fluid collector 37 at one end and to two cooling channels 32 at the other end.

[0123] Accordingly, a fluid (such as water) can be supplied to the cooling channels 32 and to the distributor channels 34 via the fluid distributor 36, each distributor channel 34 supplying two cooling channels 32 respectively, and the fluid distributor 36 supplying fluid to all distributor channels 34. The fluid flows through the cooling channels 32 and then flows into the fluid collector 37 via the collector channels 35, and the fluid can be removed via the fluid collector 37. In this case, the fluid of every two cooling channels 32 flows into one collector channel 35 respectively, and the fluid of all collector channels 35 flows into the fluid collector 37. The supply of fluid to the fluid distributor 36 and the removal of fluid from the fluid collector 37 can be carried out by means of the cooling device 24, and for this purpose, the cooling device 24 can be connected to the fluid distributor 36 and the fluid collector 37.

[0124] Adjusting the temperature of the supplied fluid to be lower than the temperature of the upper part 28 can achieve the effect that the fluid extracts heat from the upper part 28 when flowing through the cooling channels 32. This heat extraction is especially for compensating the heat input generated due to the radiation incident on the optical surface 31 during the operation of the lithography system. Since the optical surface 31 does not completely reflect the incident radiation, part of the radiation is absorbed by the optical surface 31, and depending on the formation of the optical surface 31, part of the radiation is also absorbed by the upper part 28 and converted into heat. Since the optical surface 31 and the upper part 28 have a certain thermal conductivity, part of this heat is guided to the cooling channels 32 and can be absorbed and carried away by the fluid there. In this way, compared with the uncooled mirror 26, the temperature rise of the mirror 26 caused by the radiation can be limited, and the deformation of the optical surface 31 caused by the thermal expansion effect can be reduced. Accordingly, the imaging error caused by the deformation is also reduced.

[0125] The formation and configuration of the cooling channels 32, the distributor channels 34, the collector channels 35, the fluid distributor 36, and the fluid collector 37 can also be modified in various ways. For example, the cooling channels 32 can be configured in the lower part 27 and formed to open towards the connecting surface 29 of the lower part 27, so that the cooling channels 32 have elongated cooling channel openings 33 in the region of the connecting surface 29 of the lower part 27. It is also possible to do without the distributor channels 34 and the collector channels 35, and the fluid distributor 36 and the fluid collector 37 can be directly connected to the cooling channels 32, etc.

[0126] To manufacture the mirror 26, the lower part 27 and the upper part 28 are manufactured as separate components and connected to each other. The optical surface 31 is generally only formed after connecting the lower part 27 and 28 in order to achieve the highest possible accuracy, especially in terms of its shape, and to avoid damaging the optical surface 31 during the connection process. Materials with a very low coefficient of thermal expansion are used to manufacture the lower part 27 and the upper part 28. Suitable materials are, for example, fused silica, fused silica doped with titanium oxide, or special glass ceramics. The lower part 27 and the upper part 28 can be cut out from common material blanks. However, if the lower part 27 and the upper part 28 do not differ much in terms of material parameters, it is also possible to use different material blanks for the lower part 27 and the upper part 28.

[0127] For example, fused silica mixed with titanium oxide (whose average zero-crossing temperature over the volume of the lower part 27 is between 22 °C and 25 °C) can be used to manufacture the lower part 27. During the operation of the lithography system, taking into account the cooling of the mirror 26, the zero-crossing temperature has been adjusted here to match the expected temperature of the lower part 27. Depending on the method of manufacturing the material of the lower part 27, the increase in the coefficient of thermal expansion at 22 °C is preferably less than about 1.35 ppb / K 2 . At least in the region of the finished mirror 26 that laterally overlaps with the optical surface 31, the zero-crossing temperature has a uniformity better than + / - 5 K, which is better in the region of the connecting surface 29 of the lower part 27 than on the side facing away from it.

[0128] The lower part 27 is cut out from the material blank and then processed by grinding and polishing. The connecting surface 29 of the lower part 27 can here be specifically formed in a form that roughly corresponds to the expected shape of the optical surface 31 (which will later be formed on the upper part 28). The shape deviation between the connecting surface 29 of the lower part 27 and the optical surface 31 is at most 3 mm, preferably not exceeding 200 µm. If the connecting surface 29 of the lower part 27 has a circular peripheral boundary, its diameter is generally between 20 cm and 100 cm, with a sagittal up to 3% of the diameter. Any desired function can be superimposed on the spherical basic shape of the connecting surface 29 of the lower part 27, so that a free-form surface can be obtained.

[0129] For a wavelength λ of 633 nm used for measurement, the polishing of the connecting surface 29 of the lower part 27 can be carried out in such a way that the deviation from a predetermined shape (pattern) is less than λ / 2, preferably less than λ / 10. Furthermore, in this case, a roughness of less than 5 nm, preferably less than 1 nm, can be achieved. In this context, the term "roughness" encompasses the random distribution deviation from a smooth surface. Shape or pattern deviations are considered to include long-wave and short-wave deviations from the predetermined shape, as well as more punctiform, non-randomly distributed deviations. The long-wave form deviations may also be slightly larger than the values of λ / 2 and λ / 10 mentioned above. However, for the short-wave deviations, the values mentioned will be kept as constant as possible. For example, deviations with a spatial frequency of several centimeters do not have such a destructive effect. Deviations with a spatial frequency below 100 µm are generally more destructive.

[0130] In order to prevent the residues of the grinding and polishing agents from drying, the lower part 27 is rinsed directly after grinding and polishing. Furthermore, mechanical wet cleaning can be carried out. The surface of the lower part 27 can be additionally cleaned with solvents, organic and inorganic cleaning agents, or by plasma treatment, etc.

[0131] When processing the lower part 27, it is necessary to ensure that damage in depth is avoided, because a small amount of water, other processing agents (such as polishing agents) or contaminants may remain in the microcracks, and they may evaporate and cause local blistering when the lower part 27 is heated. In order to avoid or eliminate possible damage in depth, dry and / or wet etching steps or other appropriate optical processing steps can be used. Furthermore, a graded grinding step can be used, or non-contact removal can be carried out, for example, by using oxygen plasma etching, followed by a polishing step. For example, a short etching with hydrofluoric acid does not result in local etching pits, which proves that there is no damage in depth. The hydrofluoric acid bath is suitable as a cleaning step for removing hydrocarbons at the same time.

[0132] After grinding and polishing or sanding, if applicable, the distributor channels 34 and the collector channels 35, as well as the fluid distributor 36 and the fluid collector 37, are formed in the lower part 27. Depending on the formation of the lower part 27, the cooling channels 32 can also be formed in the lower part 27. The distributor channels 34 and the collector channels 35, as well as the fluid distributor 36 and the fluid collector 37, generally have a lateral dimension, i.e., a dimension transverse to the fluid flow direction, which is much larger than the corresponding dimension of the cooling channels 32.

[0133] The corresponding channels can be formed by milling, grinding, drilling, laser ablation, ultrasonic ablation, or etching, or a combination of these methods. As a final method step, etching with hydrofluoric acid can be provided in each case to avoid subsequent crack propagation and to form a surface with only slightly roughness in the channels to facilitate flow.

[0134] In order to protect the machined, in particular polished, surface of the lower part 27 from cracking at the edges (e.g., at the periphery of the cooling channel opening 33) and from similar damage, a protective resist can be applied to the surface of the lower part 27 before the channels are formed. Despite this preventive measure, cracking at the edges can still occur, which can be locally smoothed out or eliminated by, for example, a chamfer extending around the cooling channel opening 33. However, due to the notch effect, this chamfer is disadvantageous in terms of flow and strength and should be avoided as much as possible.

[0135] As an alternative to the above process, the channels can first be formed in the lower part 27, and then the surface of the lower part 27 can be machined only by grinding, sanding, and / or polishing. However, this results in a certain rounding at the edges of the cooling channel opening 33 or other openings because, for example, the polishing disc has a limited stiffness and protrudes into the cooling channel opening 33 or other openings. This rounding at the edges can be compensated for by inserting or gluing a placeholder made of the same material as the lower part 27 into the cooling channel opening 33 or other openings. The connecting surface 29 of the placeholder with the lower part 27 is ground off and polished, and the width of the gap caused by the cooling channel opening 33 or other openings is reduced. The remaining gap between the surrounding material of the lower part 27 and the placeholder can also be grouted with a putty filled with glass powder or ceramic powder, so that there is at most a slight bulge during the polishing process and thus very little rounding at the edges. Alternatively, local reworking can be carried out by robotic polishing or ion beam polishing to minimize the rounding at the edges. After machining the surface of the lower part 27, the placeholder and the grouting material (if applicable) are removed again.

[0136] In order to avoid or at least reduce the rounding at the edges, ablation or etching methods can be used as an alternative to mechanical surface machining. If, in these methods, a resist mask is applied to the areas that are not to be machined, underetching may occur in the peripheral area of the resist mask. The adverse effects of this underetching can be compensated for by appropriate mechanical, laser ablation, or etching reworking.

[0137] The upper part 28 can be made of a material having the same or similar properties as those described for the lower part 27. The material properties of the upper part 28 can deviate from those of the lower part 27 only within a predetermined limit. For example, very different thermal expansion behaviors of the lower part 27 and the upper part 28 would be a problem because the lower part 27 and the upper part 28 are rigidly connected to each other.

[0138] After cutting out the upper part 28 from the material blank, the surface of the upper part 28 (in particular the connecting surface 30) can be machined in a manner similar to that described for the lower part 27, and similar requirements for shape deviation and roughness must be met. If the upper part 28 is relatively thin, the requirements for shape deviation can be slightly relaxed.

[0139] According to a variant of a specific embodiment, the upper part 28 may have cooling channels 32 and / or other channels, such as a distributor channel 34 and a collector channel 35. All these channels can be manufactured in a manner similar to that described for the lower part 27, and similar measures can be used to reduce rounding at the edges, etc.

[0140] The thickness of the upper part 28 is generally between 10 and 40 millimeters, but may also have a smaller thickness under certain preconditions. Depending on the thickness, it may be necessary to bond or twist the upper part 28 to a glass substrate as a carrier material for processing to achieve sufficient stiffness for precise processing.

[0141] Tools with the same or very similar nominal curvature as the tools used for processing the connection surface 29 of the lower part 27 can be used to process the connection surface 30 of the upper part 28 by grinding, polishing, and / or buffing. Optionally, further steps can be provided thereafter, in which the actual shapes of the connection surface 29 of the lower part 27 and the connection surface 30 of the upper part 28 are measured, and the connection surface 30 of the upper part 28 is locally adapted to the connection surface 29 of the lower part 27 by robotic polishing, plasma etching, ion beam ablation, or compaction of high-energy radiation.

[0142] In the case of a specific embodiment of the mirror 26, where the connection surface 29 of the lower part 27 is substantially planar or has a very small curvature, so that the absolute value of the average radius of curvature of at least one azimuth is greater than 10 m, it is possible to use a relatively thin upper part 28 with a thickness of 0.5 to 10 millimeters, and the connection surface 30 of the upper part 28 is formed in a planar manner. In this case, the upper part 28 may deform when connected to the lower part 27, so that the connection surface 30 of the upper part 28 can adapt to the shape of the connection surface 29 of the lower part 27. For the processing required before connection, the upper part 28 can be bonded or twisted to a sufficiently rigid carrier material.

[0143] As an alternative to using the upper part 28 cut out from a material blank, a thin upper part 28 of titanium-doped silica glass can also be produced by outflow (e.g., on molten metal (float glass)), pouring on a long-separated edge, pressing and sintering glass soot, or indirectly through a gray body. The advantage of these methods is that the upper part 28 is made of a sheet several millimeters thick, where good uniformity of material parameters can be achieved, and machining can be omitted. In order to enable the upper part 28 made of such a sheet to have a thermal expansion behavior similar to that of the lower part 27, the OH content of the sheet should deviate from the corresponding content of the lower part 27 by a maximum of 5%, and the titanium content of the sheet should deviate from the corresponding content of the lower part 27 by a maximum of 0.05%. For very thin sheets with a thickness of less than 2 mm for manufacturing the upper part 28, these requirements can also be slightly relaxed.

[0144] When the lower part 27 and the upper part 28 are already completed, they can be connected to each other in the region of their connection surfaces 29, 30. The connection is carried out in two steps: in the first step, the upper part 28 is twisted onto the lower part 27, that is, the upper part 28 is pulled so close to the lower part 27 that the upper part 28 is attached to the lower part 27 due to the action of van der Waals forces between the atoms of the lower part 27 and the atoms of the upper part 28. The first step can be carried out at room temperature and an intermediate product comprising the lower part 27 and the upper part 28 is produced. In the second step, the connection between the upper part 28 and the lower part 27 is strengthened by a joining process, whereby covalent bonds are formed between the atoms of the lower part 27 and the atoms of the upper part 28, so that the upper part 28 and the lower part 27 are very firmly and permanently connected to each other. The procedure according to the invention in the joining process and the preparatory work taken for this will be explained in more detail below.

[0145] Figure 5 A flow chart is shown which explains the procedure according to the invention when connecting the lower part 27 and the upper part 28. The lower part 27 and the upper part 28 can be produced in the above-described manner in procedures carried out substantially in parallel, or completely independently of each other in procedures separated in space and / or time.

[0146] In order to be able to form a permanent connection, in step S1, the lower part 27 and the upper part 28 are cleaned, activated and dried, in particular in the region of their connection surfaces 29, 30. For example, the connection surface 29 of the lower part 27 and the connection surface 30 of the upper part 28 can be cleaned, ablated and activated simultaneously by etching (for example with fluoric acid), by sputtering, or by plasma treatment (in particular using hydrogen or oxygen plasma).

[0147] Step S1 is followed by step S2, in which the upper part 28 is twisted onto the lower part 27. In order to be twisted onto the lower part 27, the connection surface 30 of the upper part 28 is so close to the connection surface 29 of the lower part 27 that a strong attraction is generated due to the van der Waals force between the atoms close to the surface of the upper part 28 and the atoms close to the surface of the lower part 27, and finally the upper part 28 and the lower part 27 are connected to each other. Since the van der Waals force only acts over a very short distance, it is necessary to make the connection surface 30 of the upper part 28 and the connection surface 29 of the lower part 27 very close. This is only possible in a relatively large surface area if the connection surface 30 of the upper part 28 and the connection surface 29 of the lower part 27 are very smooth, have almost the same shape, and are substantially free of contaminants. It is also very important to make the connection surface 30 of the upper part 28 and the connection surface 29 of the lower part 27 close enough to prevent air or other gases from being trapped between the connection surfaces 30, 29 as much as possible. These gas inclusions may also have adverse effects during the subsequent use of the reflector 28. Figure 6 A possible twisting procedure is shown in .

[0148] Figure 6 A snapshot is shown in a schematic cross-section during the twisting of the upper part 28 onto the lower part 27, which cross-section is taken along Figure 4 the cross-section line C-C shown, and is thus perpendicular to the longitudinal extent of the cooling channel 32.

[0149] In the case of the exemplary embodiment shown, the connecting surface 29 of the lower part 27 and the connecting surface 30 of the upper part 28 have at least approximately complementary curvatures. As can be seen from Figure 6 it, the connecting surface 30 of the upper part 28 does not align parallel to the connecting surface 29 of the lower part 27 over the entire surface area, but rather the connecting surfaces 29, 30 form an angle with each other when placed together, and initially only intersect in the region of their peripheral surfaces shown on the Figure 6 left side. In Figure 6 it, the angle is magnified to see the effect. When they are placed together, initially there is only contact between the upper part 28 and the lower part 27 in a small area of the connecting surfaces 29, 30, and thus only a relatively small contact surface 38 is initially formed between the upper part 28 and the lower part 27. The van der Waals forces exert their attractive effect within the contact surface 38, and thus the twisting of the connecting surface 30 of the upper part 28 onto the connecting surface 29 of the lower part 27 occurs in the region of the contact surface 38.

[0150] Next, the upper part 28 continues to pivot further towards the lower part 27, and thus gets closer and closer to the lower part 27, and thus the size of the contact surface 38 increases. Accordingly, the size of the surface area on which the upper part 28 twists onto the lower part 27 also increases. The linearly formed transition zone between the twisted region and the non-twisted region of the connecting surfaces 29, 30 is hereinafter referred to as the twist front. In the Figure 6 representation, the twist front extends perpendicularly with respect to the drawing plane and is generally parallel to the longitudinal extent of the cooling channel 32. This applies at least to the region where the twist front passes through the drawing plane. The twist front can be formed as a straight line or slightly curved, and as the upper part 28 gradually approaches the lower part 27, the twist front will move from left to right, i.e., the process of the upper part 28 approaching the lower part 27 will proceed sequentially from left to right. The progressive manner is also hereinafter referred to as "rolling", and thus in the Figure 6 representation, the upper part 28 rolls onto the lower part 27 from left to right. The direction of the rolling movement is shown by an arrow in Figure 6 and is hereinafter referred to as the rolling direction 39.

[0151] The continuous twisting with a linearly advancing and stable twist front has the following effect: gas is continuously removed from the intermediate space between the connecting surface 30 of the upper part 28 and the connecting surface 29 of the lower part 27, and thus the risk of gas inclusions is greatly reduced. When the connecting surface 30 of the upper part 28 has twisted onto the connecting surface 29 of the lower part 27 over the entire surface area, the twisting operation ends.

[0152] In the case of an exemplary embodiment in which the connecting surface 30 of the upper part 28 and the connecting surface 29 of the lower part 27 are not formed in a rotationally symmetric manner, the connecting surface 30 of the upper part 28 preferably rolls onto the connecting surface 29 of the lower part 27 in a direction parallel to the direction in which the connecting surfaces 29, 30 have the smallest absolute value of the mean curvature radius.

[0153] By virtue of the connecting surface 30 of the upper part 28 and the connecting surface 29 of the lower part 27 not having exactly the same curvature, the escape of gas from the intermediate space between the connecting surface 30 of the upper part 28 and the connecting surface 29 of the lower part 27 can be made easier to some extent.

[0154] From Figure 6 It can be seen that the rolling direction 39 is transverse to, in particular perpendicular to, the longitudinal extent of the cooling channel opening 33 and the cooling channel 32. Since the thickness of the upper part 28 is significantly reduced in the region of the cooling channel 32, these regions act in a manner similar to a joint and make it easier for the upper part 28 to bend in these regions. As a result, only a relatively small force is required to temporarily hold the upper part 28 at a sufficient distance from the lower part 27 outside the contact surface 38 to prevent twisting from occurring there and to bring the connecting surface 30 of the upper part 28 continuously closer to the connecting surface 29 of the lower part 27. Furthermore, in the region of the contact surface 38, a better engagement with the connecting surface 30 of the upper part 28 in close contact with the connecting surface 29 of the lower part 27 is achieved, and thus a very reliable twist can be achieved.

[0155] Similarly, the upper part 28 can also be brought to the lower part 27 such that the rolling direction 39 is parallel to the longitudinal extent of the cooling channel opening 33 and the cooling channel 32. In the case of this exemplary embodiment, since gas can escape along the cooling channel opening 33, particularly reliable exhaust of the intermediate space between the upper part 28 and the lower part 27 is ensured. Compared with the above exemplary embodiment, the deformation ability in the rolling direction is reduced to some extent.

[0156] Twisting the upper part onto the lower part in step S2 has the effect of producing an intermediate product comprising the lower part 27 and the upper part 28, wherein the lower part 27 and the upper part 28 are rigidly connected to each other in the region of their connecting surfaces 29, 30. The intermediate product must meet various quality requirements, and only when it meets these quality requirements can it enter the heat treatment stage described in detail below. Otherwise, the intermediate product will be reprocessed or rejected. The quality requirements relate, for example, to the occurrence of inclusions between the connecting surfaces 29 and 30.

[0157] In step S3 following step S2, the intermediate product is stored at a storage temperature for a storage time. The storage time can be, for example, 24 hours. The storage temperature is higher than room temperature and can be, for example, 80 °C or preferably between 160 °C and 400 °C. Specifically, the storage temperature can be between 250 °C and 350 °C. A heating ramp can also be combined with the storage at a constant storage temperature. Storing at an elevated temperature has the effect that excess moisture diffuses out of the regions of the connecting surfaces 29, 30, in particular via the cooling channel openings 33. Depending on the pretreatment of the lower part 27 and the upper part 28 before twisting and the time interval between the manufacture and the further use of the intermediate product, step S3 may not be required either, especially when the net distance between adjacent cooling channels 32 is small.

[0158] Following step S3 is step S4, in which a heat treatment is started to form a very durable connection between the lower part 27 and the upper part 28. The connection formed by the heat treatment is also referred to as a joint hereinafter. In step S4, the intermediate product is placed in a joining furnace which ensures a temperature uniformity of, for example, better than 2 K. An air or inert gas atmosphere can be provided in the joining furnace. Similarly, the joining furnace can also be evacuated and the heat treatment can be carried out under vacuum conditions. In the joining furnace, the intermediate product is heated to a holding temperature at a heating rate. The heating rate has a value lower than 150 K / h, preferably a value lower than 100 K / h. The holding temperature lies in the range from 400 °C to 900 °C, preferably a temperature higher than 500 °C, in particular higher than 600 °C, and very preferably higher than 700 °C.

[0159] Next is step S5, according to which the intermediate product is held at the holding temperature for a holding time. The holding time can be at least 8 hours, especially at least 12 hours or at least 20 hours.

[0160] After step S5 comes step S6, in which the intermediate product is cooled so that its temperature decreases from the holding temperature. In particular, the cooling can be carried out in multiple stages, with the temperature of the intermediate product decreasing at different cooling rates. For example, the decrease in the temperature of the intermediate product can first occur at a cooling rate of less than 100 K / h or less than 50 K / h, especially between 1 K / h and 20 K / h, until a temperature of 700 °C or 600 °C is reached. At temperatures below 700 °C or 600 °C, the temperature decrease can occur more rapidly, for example at a cooling rate of less than 150 K / h, which is especially in the range of 5 K / h to 100 K / h or in the range of 20 K / h to 100 K / h. When a temperature such as 500 °C or 400 °C is reached, the limitation of the cooling rate can be completely abandoned, and then the cooling can be carried out at any cooling rate until room temperature is reached. For example, in each case, the cooling can be achieved by reducing the temperature of the gas around the intermediate product and / or the temperature of the surrounding area to a value lower than the temperature of the intermediate product. Step S6 completes the flow chart.

[0161] When setting the parameters of the heat treatment, especially the holding temperature and the cooling rate, it should be noted that they will have an impact on the zero-crossing temperature and the increase in the coefficient of thermal expansion of the mirror 26 manufactured therewith. Although these material properties have usually been set to desired values during the process of manufacturing the materials for the lower part 27 and the upper part 28, they can be changed again by heat treatment. Therefore, if a lower tolerance for the zero-crossing temperature and the increase in the coefficient of thermal expansion of the mirror 26 is required, it is not sufficient to only use materials that meet these manufacturing requirements. In addition, the heat treatment of the intermediate product needs to be precisely planned, taking into account the potential impact of the above parameters. When doing so, it should be noted that the impact of the heat treatment on the zero-crossing temperature and the increase in the coefficient of thermal expansion depends on the material composition of the intermediate product, especially its OH and titanium content. Therefore, in order to achieve a low tolerance, the OH and titanium content should be considered. If the content is not known accurately enough, it can be measured in the method according to the invention to determine the OH content and the titanium content. In addition, measurements can be carried out before the heat treatment to determine the actual zero-crossing temperature and the actual increase in the coefficient of thermal expansion of the intermediate product. Since some of these measurements are not non-destructive, they are not carried out on the intermediate product itself, but on the attached test piece, that is, on a piece of material used to manufacture the intermediate product. The attached test piece undergoes the same heat treatment as the intermediate product, and if the lower part 27 and the upper part 28 have undergone heat treatment before the intermediate product is formed, this is also carried out on the attached test piece. If the lower part 27 and the upper part 28 are made of different blanks, attached test pieces can be provided separately for the lower part 27 and the upper part 28.

[0162] Subsequently, based on known or measured material properties, a heat treatment curve is determined for each intermediate product or each group of intermediate products with very similar material properties, i.e., the expected variation of the temperature of the intermediate product over time. Of particular importance here is the cooling curve, especially the cooling rate used.

[0163] In this way, a mirror 26 can be manufactured with a maximum deviation of the average zero-crossing temperature from the desired zero-crossing temperature of + / - 1 K. Maximum deviations of + / - 0.5 K or even + / - 0.2 K are also possible. These values are in each case related to a zero-crossing temperature that is averaged over the volume region of the mirror 26 that laterally overlaps the optical surface 31 and extends from the optical surface 31 to 1 cm in depth. Similarly, this volume region can also extend from the optical surface 31 to the cooling channel 32 or to 1 cm below the cooling channel 32 in depth. In Figure 3 the representation, the range to the cooling channel 32 refers to the range to the upper edge of the cooling channel 32, and the range to 1 cm below the cooling channel 32 refers to the range to 1 cm below the lower edge of the cooling channel 32. In this volume region, the predetermined value can in each case decrease as the distance from the optical surface 31 increases.

[0164] Another difficulty in heat treatment is to ensure that the entire volume of the intermediate product is treated according to the expected heat treatment curve. For example, the peripheral region of the intermediate product reacts faster to the cooling acting from the outside and thus cools faster than the core region. This problem is alleviated to some extent because it is not necessary to achieve approximately the same temperature throughout the volume of the intermediate product at each time point of cooling. Indeed, the predetermined cooling curve, i.e., the decrease in temperature over time, should be maintained as accurately as possible at each location of the intermediate product. However, time differences between the individual locations are allowed. Thus, the cooling is controlled in such a way that the cooling curve is maintained locally in each case, and for this purpose, the cooling rate is adjusted step by step during cooling, if appropriate several times and over a period of several hours, to achieve this goal. For example, the adjustment can last for a period of one to six hours.

[0165] In particular, it should be ensured that the relative change over time of the sliding average of the cooling rate formed within a temperature interval of 20 K is the same throughout the entire volume of the intermediate product within the temperature range from 900 °C to 700 °C, except for a maximum deviation of 2 K / h. This means that the cooling should be carried out in such a way that, after taking into account the corresponding potential time differences, the change over time of this sliding average at all positions of the intermediate product differs by at most 2 K / h. Preferably, the deviation of the sliding average within the temperature range from 900 °C to 700 °C is at most 1 K / h. In addition, it can be stipulated that the deviation of the sliding average within the temperature range from 700 °C to 500 °C is at most 4 K / h, preferably at most 2 K / h. By these measures, ultimately, it should be achieved that within the volume region of the mirror 26 that laterally overlaps with the optical surface 31 and extends 1 cm in depth from the optical surface 31, the non-uniformity of the zero-crossing temperature caused by the heat treatment is at most + / - 1 K, preferably at most + / - 0.5 K. Similarly, this volume region can also extend to the cooling channel 32 or 1 cm below the cooling channel 32.

[0166] The purpose of the heat treatment is to achieve a tensile strength of 50 MPa or higher for the connection between the lower part 27 and the upper part 28.

[0167] In the modified example, the lower part 27 and the upper part 28 are pressed against each other during the heat treatment.

[0168] The above joining procedure can also be achieved in the same way without twisting. In this case, the procedure is similarly the same, where each step does not have to be carried out with the intermediate product but with the lower part 27 and the upper part 28, which are not connected to each other by twisting.

[0169] After the joining procedure, the upper part 28 can be machined to a thickness of 2 to 10 mm above the cooling channel 32.

[0170] After the machining step of reducing the thickness of the upper part 28, or instead of this machining step, the upper part 28 is reworked so that on the side facing away from the connection surface 30 and on which the optical surface 31 will be formed, it has a predetermined shape with high precision and conforms to a predetermined roughness standard. Then, the optical surface 31 is formed there. This can be carried out, for example, by applying an aluminum layer or alternately applying a molybdenum layer and a silicon layer.

[0171] Reference numerals

[0172] 1 Projection exposure apparatus

[0173] 2 Illumination system

[0174] 3 Radiation source

[0175] 4 Illumination optical unit

[0176] 5 Object field

[0177] 6 Object plane

[0178] 7 Mask master

[0179] 8 Mask master holder

[0180] 9 Mask master displacement driver

[0181] 10 Projection optical unit

[0182] 11 Image field

[0183] 12 Image plane

[0184] 13 Wafer

[0185] 14 Wafer holder

[0186] 15 Wafer displacement driver

[0187] 16 Illumination radiation

[0188] 17 Condenser

[0189] 18 Intermediate focal plane

[0190] 19 Deflection mirror

[0191] 20 First split surface mirror

[0192] 21 First split surface

[0193] 22 Second split surface mirror

[0194] 23 Second split surface

[0195] 24 Cooling device

[0196] 25 Control device

[0197] 26 Mirror

[0198] 27 Lower part

[0199] 28 Upper part

[0200] 29 Connection surface

[0201] 30 Connection surface

[0202] 31 Optical surface

[0203] 32 Cooling channel

[0204] 33 Cooling channel opening

[0205] 34 Distributor channel

[0206] 35 Collector Channel

[0207] 36 Fluid Distributor

[0208] 37 Fluid Collector

[0209] 38 Contact Surface

[0210] 39 Rolling Direction

[0211] Mirror M

[0212] Mirror M1

[0213] Mirror M2

[0214] Mirror M3

[0215] Mirror M4

[0216] Mirror M5

[0217] Mirror M6

Claims

1. A method for manufacturing a mirror (26) for a microlithographic projection exposure apparatus (1), comprising: - providing a first mirror part (27) and a second mirror part (28), which first mirror part (27) and second mirror part (28) are in contact in the region of a first connection surface (29) of the first mirror part (27) and a second connection surface (30) of the second mirror part (28); - in order to form a permanent connection between the first mirror part (27) and the second mirror part (28), the first mirror part (27) and the second mirror part (28) are heated to a holding temperature of at least 400 °C and held at the holding temperature for a holding time; and - after the holding time has elapsed, the first mirror part (27) and the second mirror part (28) are cooled to a first cooling temperature at a first cooling rate of at most 100 K / h.

2. The method according to claim 1, wherein the first mirror part (27) and the second mirror part (28) are provided in the form of an intermediate product which is manufactured by twisting the first connection surface (29) of the first mirror part (27) and the second connection surface (30) of the second mirror part (28).

3. The method according to any one of the preceding claims, wherein the intermediate product is stored at a storage temperature of at least 80 °C and at most 400 °C for a storage time of at least 2 hours before being heated to the holding temperature.

4. The method according to any one of the preceding claims, wherein the first mirror part (27) and the second mirror part (28) are heated to a holding temperature of at most 900 °C.

5. The method according to any one of the preceding claims, wherein the first mirror part (27) and the second mirror part (28) are heated to the holding temperature at a heating rate of at most 150 K / h.

6. The method according to any one of the preceding claims, wherein the holding time is at least 8 hours.

7. The method according to any one of the preceding claims, wherein the first cooling temperature is 700 °C or below.

8. The method according to any one of the preceding claims, wherein after cooling to the first cooling temperature at the first cooling rate, the first mirror part (27) and the second mirror part (28) are cooled to a second cooling temperature at a second cooling rate, wherein the second cooling rate is greater than the first cooling rate and the second cooling temperature is lower than the first cooling temperature.

9. The method according to any one of the preceding claims, wherein throughout the entire volume of the first mirror part (27) and / or the second mirror part (28), the first mirror part (27) and the second mirror part (28) are cooled such that the sliding average of the cooling rate formed within a temperature interval of 20 K has the same relative change over time within the temperature range from 900 °C to 700 °C, except for a deviation of at most 2 K / h.

10. The method according to any one of the preceding claims, wherein the selection of the first cooling rate and / or the second cooling rate depends on the material composition of the first mirror member (27) and / or the second mirror member (28).

11. The method according to any one of the preceding claims, wherein the selection of the first cooling rate and / or the second cooling rate depends on the desired average zero-crossing temperature of the mirror (26).

12. The method according to any one of the preceding claims, wherein, in order to determine the first cooling rate and / or the second cooling rate, a material sample of the material for manufacturing the first mirror member (27) and / or the second mirror member (28) is analyzed.

13. A mirror for a microlithographic projection exposure apparatus (1), comprising: - a first mirror member (27); - a second mirror member (28); - an optical surface (31) for reflecting light; wherein - the first mirror member (27) and the second mirror member (28) are permanently connected to each other by heat treatment in the region of a first connection surface (29) of the first mirror member (27) and a second connection surface (30) of the second mirror member (28), and - the mirror (26) has an average zero-crossing temperature in a volume region that laterally overlaps the optical surface (31) and extends from the optical surface (31) to 1 cm in depth, and the deviation of the average zero-crossing temperature from a predetermined value is at most + / - 1 K.

14. An illumination optical unit having a mirror (26), wherein the mirror is manufactured by the method according to any one of claims 1 to 12 and / or formed as claimed in claim 13.

15. A projection optical unit having a mirror (26), wherein the mirror is manufactured by the method according to any one of claims 1 to 12 and / or formed as claimed in claim 13.

16. A microlithographic projection exposure apparatus having the illumination optical unit (4) according to claim 14 and / or the projection optical unit (10) according to claim 15.

Images