CN118355330A

CN118355330A - Method for producing local thickness variations of a coating, mirror and EUV lithography system

Info

Publication number: CN118355330A
Application number: CN202280080016.0A
Authority: CN
Inventors: M·凯斯; M·诺亚
Original assignee: Carl Zeiss SMT GmbH
Current assignee: Carl Zeiss SMT GmbH
Priority date: 2021-12-02
Filing date: 2022-11-23
Publication date: 2024-07-16

Abstract

The invention relates to a method for producing a local thickness variation (delta (x, y)) of a coating (26) for reflecting radiation, in particular for reflecting EUV radiation, which EUV radiation is applied to a substrate (25) of a mirror (M4), the method comprising: local thickness variations (delta (x, y)) are produced by introducing local energy inputs (E (x, y)) into the coating (26) that densify or expand the coating (26) for producing local target thicknesses (D _S (x, y)) of the coating (26) corresponding to local target reflectivities (R _S (x, y)) of the mirror (M4). The method further comprises changing the surface shape of the substrate (25) by introducing a local energy input (E (x, y)) into the substrate (25), wherein a local thickness variation (delta (x, y)) of the coating (26) is produced before and/or after changing the surface shape of the substrate (25). The invention also relates to a mirror (M4) and to an EUV lithography system having at least one such mirror (M4).

Description

Method for producing local thickness variations of a coating, mirror and EUV lithography system

Citation of related application

The present application claims priority from german patent application DE102021213679.6 filed on 12/2021, the entire disclosure of which is incorporated herein by reference.

Disclosure of Invention

The invention relates to a method for producing a local thickness variation of a coating for reflecting radiation, in particular for reflecting EUV radiation, which is applied to a substrate of a mirror. The invention also relates to a mirror comprising: a substrate and a coating applied to the substrate for reflecting radiation, in particular for reflecting EUV radiation, and also to an EUV lithography system having at least one such mirror.

Background

The local thickness of the coating for reflecting radiation (e.g. for reflecting EUV radiation) may deviate from the local target thickness or the design layer thickness profile. The deviation of the local thickness from the target thickness may be due to, for example, coating errors when applying the coating to the substrate. When operating a mirror in an optical device (e.g. in a projection system of an EUV lithographic apparatus), the deviation may adversely affect its optical properties, such as its imaging properties.

This is especially the case if the coating is a multi-layer coating having a specified number of subsystems (e.g. in layer pairs), each subsystem having the same thickness, wherein the thickness of the respective subsystem affects the reflectivity of the coating, especially the wavelength at which the coating has maximum reflectivity. Thus, a deviation of the (local) total thickness of the coating from the (local) target thickness results in a corresponding deviation of the thickness (period length or period thickness) of the specific subsystem affecting the reflectivity or spectral response of the coating.

US 6,635,391 B2 and US 7,049,033 B2 propose the production of masks (reticles) for EUV lithography by directly modulating the complex-valued reflectivity of a multilayer coating. To achieve this, locally positioned energy sources are used, for example in the form of focused electron or ion beams, which directly write a pattern into the reflective multilayer coating. This exploits the fact that: due to the energy input, locally limited interdiffusion occurs between the layers of the multilayer coating, which leads to densification and thus to shrinkage of the cycle length of the multilayer coating. The degree of densification is determined by the energy dose introduced. In this way, an adjustable variation of the phase and amplitude of the reflected field is intended to be produced, which makes it superfluous to apply a structured absorbing layer on the mask.

US 7,022,435 B2 describes a similar process for a phase shift mask for EUV lithography. There, the thickness variation of the multilayer coating by localized heating is used to produce a direct modulation of the complex-valued reflectivity of the multilayer coating in order to produce the phase shift characteristics of the mask.

US 6,821,682 B1 and EP 1336130 A2 describe a method for repairing local defects in a multilayer coating of a reticle for EUV lithography. The method alters the thickness of the coating near the localized defect by introducing energy into the coating. The thickness of the coating is locally adjusted to correct for the disturbance of the reflected field. For example, repair of the defect may include flattening the protrusions or widening the sides of the depressions.

US 6,844,272 B2 describes correcting local defects in the shape of an optical surface by irreversibly (i.e. permanently) changing the local density of the surface or of a layer near the surface, which results in a change in the height of the optical surface. The change in local density results in irreversible local expansion or irreversible local contraction of the substrate or a coating applied to the substrate. The change in height may be caused, for example, by interdiffusion or by chemical reactions between adjacent layers of the coating, which result from local energy input. The energy input may be applied, for example, with an ion beam, an electron beam, or a laser beam.

Disclosure of Invention

Object of the Invention

It is an object of the invention to provide a method, a mirror and an EUV lithography system for producing a local thickness variation of a coating, which has improved optical properties.

Object of the Invention

This object is achieved by a method of the type mentioned in the introductory portion, which comprises: local thickness variations are created by introducing local energy inputs into the coating that densify or expand the coating to create a local target thickness of the coating that corresponds to the local target reflectivity of the mirror.

US 6,844,272 B2 describes that correction of local defects in the shape of an optical surface can be achieved by a change in the local density of the material of the coating applied to the substrate, which results in local expansion or local contraction of the coating. However, correcting the surface shape of the optical surface by shrinkage or expansion results in an undesirable change in the local reflectivity of the coating in the multilayer coating, wherein the reflection is affected by interference, because the period thickness or period length of the subsystem of the reflective coating (e.g., in the form of pairs of adjacent layers having different refractive indices) varies in the process.

However, in the method according to the invention, the local thickness variation is used to set a local target thickness, which corresponds to the thickness design of the coating, and thus to set the desired local target reflectivity of the mirror. In this way, the local reflectivity characteristics of the mirror can be selectively set. In particular, the spectral response, i.e. the wavelength at which the mirror has its maximum value of reflectivity, can also be adapted to the wavelength used to be reflected.

Due to the energy input, heat is locally introduced into the coating and, if appropriate, into the substrate, in order to locally increase the temperature of the coating. The local increase in temperature results in an irreversible local expansion (increase in thickness) or an irreversible local contraction (decrease in thickness) of the coating. Whether a coating shrinks or expands under energy input depends on the type of coating material. In particular, it is important that new materials having a higher or lower density than the material prior to energy input are formed in the interdiffusion or chemical reaction between the materials of adjacent layers caused by energy input.

It should be understood that the methods described herein may be applied only to coatings that have irreversible densification or irreversible decompression (swelling) upon energy input or thermal reaction. This is often the case for multilayer coatings with multiple layers of different materials, in particular multiple subsystems with layer pairs of the same thickness, as are used for reflecting EUV radiation at normal incidence.

In one variation, the method comprises: the coating is applied to the substrate, wherein the coating is applied at a local thickness greater than a local target thickness of the coating if the local energy input causes the coating to compact, or wherein the coating is applied at a local thickness less than the local target thickness of the coating if the local energy input causes the coating to expand. For coatings, it is known or predetermined by experiments how it behaves at energy input or with increasing temperature (densification or expansion). If the coating is densified, a thicker coating is applied in a targeted manner than is required for the targeted thickness design. If the coating expands with increasing temperature, a thinner coating than specified by the target thickness design is applied in a targeted manner.

For example, the oversized or layer thickness excess may be between about 0.05% and 0.5% of the local target thickness of the coating. In the case of coatings which react to energy input by shrinking, deviations from the local target thickness may also be corrected due to the oversized dimensions when the coating is applied, wherein the local actual thickness of the coating is greater than the local target thickness. For coatings that react to energy input by expansion, this applies similarly, i.e. without excess, only deviations from the local target thickness can be compensated, wherein the local thickness of the coating is smaller than the local target thickness.

When a multilayer coating is applied, the multilayer coating has a plurality of subsystems, for example in the form of layer pairs, each layer pair having the same thickness and defining a period thickness or period length, the thickness of the respective layer pair also being scaled when a coating having a greater thickness or having a lesser thickness is applied: for example, when the thickness of the coating is increased by 0.5%, the period thickness of the applied subsystem or layer pair is increased by 0.5% each. However, the number of layer pairs applied is generally unchanged, since if there is a sufficient number of layer pairs, for example more than 50 layer pairs, a change in the number of layer pairs has no or only little effect on the reflectivity of the coating.

In principle, the method described further above can be performed for producing a local thickness variation of the coating at any time in the process chain for producing the mirror, which process chain is located after the application of the coating in terms of time. Desirably, however, the temperature in the downstream steps in the process chain is lower than the temperature resulting from the energy input in the methods described herein.

In another variation, the method includes: the surface shape of the substrate is changed by introducing a local energy input into the substrate, wherein a change in the local thickness of the coating is produced before and/or after changing the surface shape of the substrate. The surface shape of the substrate is typically changed after the coating is applied to the substrate. In this case, the local energy input is typically introduced by a treatment beam that penetrates the coating and causes an irreversible change in the local substrate density at the underlying volumetric region of the substrate, typically resulting in local densification of the substrate material. The associated local density variations of the substrate cause deformation of the mirror surface and make it possible to adapt the surface shape of the substrate to the target surface shape.

When changing the surface shape, the local energy input into the substrate may not affect or only very slightly affect the local thickness of the coating. In this case, the method for producing a local thickness variation or for producing a local target thickness of the coating, which is described further above, can be carried out before the step of changing the surface shape, without any correction. In case a strong local energy input into the substrate is generated when changing the surface shape to increase the throughput time, a relatively large temperature rise in the coating is also a consequence, wherein when an energy input is introduced into the substrate, temperature peaks may occur depending on the dose distribution, which changes the local thickness of the coating and thus the local reflectivity of the mirror in a manner depending on the dose distribution. This problem can be solved in a number of ways, as described in further detail below.

In one variation, the method comprises: the local actual thickness of the coating is determined by measuring the local actual reflectivity of the coating. The reflectivity of the coating with respect to the radiation to be reflected (e.g. with respect to EUV radiation) can be measured in a spatially resolved manner in order to record the actual reflectivity profile of the mirror. The local (actual) reflectivity of the mirror or the coating depends in a known manner on the local (actual) thickness of the coating, since the (wavelength dependent) reflectivity and thus the reflectivity at the wavelength used depend on the period length or period thickness of the coating or the individual layer pairs of the coating (see above).

In the measurement of the actual reflectance distribution described above, the region where the local actual thickness of the coating layer differs from the local target thickness can be identified by comparison with the target reflectance distribution of the coating layer. In these areas, a local thickness variation is created to create a local target reflectivity of the mirror.

In a further development of this variant, the change in the local thickness corresponds to a difference between the local target thickness and the local actual thickness. In this case, the local thickness variation is selected such that the target thickness of the coating is set based on the measured actual thickness. As further described above, local thickness variations to achieve local target thicknesses of the coating may be induced at any time in the process chain after the coating has been applied. For example, the local thickness variation may be caused before the above-described step of changing the surface shape of the substrate. This is particularly useful if the local thickness of the coating is not changed by changing the surface shape of the substrate or is only slightly changed by changing the surface shape of the substrate.

In the case where changing the surface shape of the substrate causes a significant change in the coating thickness, the local thickness change may be caused only after changing the surface shape of the substrate. In this case, the local actual thickness of the coating layer may be measured after changing the surface shape of the substrate. In this case, a local thickness variation may be generated, which corresponds to a difference between the measured local actual thickness and the local target thickness, and the coating error and the error in the coating thickness due to the variation in the surface shape of the substrate are corrected.

In a refinement, an additional local thickness variation of the coating layer that is expected when changing the surface shape of the substrate is determined, and when a local thickness variation is generated, the additional local thickness variation that is expected due to the variation of the surface shape of the substrate is compensated. In this case, the local temperature distribution and the associated expected additional variation of the local thickness of the coating caused by the variation of the surface shape are determined based on the known dose distribution generated when the surface shape of the substrate is changed. When a local thickness change is produced, which may occur before or after the change in surface shape, the expected additional change in local thickness that occurs when the surface shape is changed is compensated for. In this case, the local actual thickness of the coating can be determined by measuring the local actual reflectivity of the coating before changing the surface shape. In this case, the change in the local thickness likewise corresponds to the difference between the local target thickness and the local actual thickness, wherein the local target thickness takes into account the expected additional change in the local thickness when changing the surface shape.

In another variant, the coating and/or the substrate is irradiated with a treatment beam for locally introducing an energy input. For the local introduction of energy input, it has proven to be advantageous if the treatment beam is directed at the surface of the coating or substrate. The treatment beam may scan the surface of the coating or substrate. Typically, the energy of the treatment beam used to produce the local thickness variation is selected such that its effective range is slightly greater than the thickness of the coating, i.e., the effective range is just below the surface of the substrate, so that the coating can be irradiated throughout its thickness.

The introduction of energy input into the coating to produce local thickness variations and the introduction of energy input into the substrate to change the surface shape of the substrate may be performed by means of the same treatment beam (e.g. by means of an electron beam). In both cases, the treatment beam or electron beam has different acceleration voltages and thus different energies. The acceleration voltage or energy affects the effective penetration depth of the electron beam into the coating or substrate: the surface shape of the substrate typically varies at an accelerating voltage higher than the variation of the local thickness of the coating (e.g. in the range of 5kV-20 kV).

In a further development of this variant, the treatment beam is selected from the group comprising: ion beam, electron beam, and laser beam. In general, the local energy input may be achieved by irradiation with an electron beam, a particle beam (e.g. an ion beam) or generally by irradiation with light or radiation. Generally, the use of an electron beam may more efficiently couple heat into a coating or substrate than the use of light (e.g., a laser beam). The thermal power coupled into the coating may additionally be specified by selecting the current when generating the electron beam.

In order to compensate for the larger deviation from the local target thickness of the coating, a higher temperature is required than if the small deviation from the local target thickness were to be compensated for. In principle, the energy dose introduced is important for local thickness variations or for the resulting densification or expansion. The energy dose introduced depends on the temperature to which the coating is heated at a given location and how long the coating remains at that temperature.

Thus, the residence time of the treatment beam at a given location of the coating affects the localized densification or expansion of the coating at the given location. The dwell time of the treatment beam resulting in the desired densification or expansion also depends on the intensity of the treatment beam, which can likewise be set, for example, via electron current if desired (see above).

In another variation, the method includes: the dwell time of the treatment beam at a given location of the irradiated coating is determined, which causes a local thickness variation of the coating. The local dwell time of the treatment beam at a given location of the coating determines the degree of variation in local thickness. Longer residence times generally result in larger changes in local thickness unless saturation occurs. The dwell time of the process beam can be determined using a calibration curve that directly relates the change in local thickness to the dwell time (at a given intensity and energy of the process beam).

In a further development of this variant, the dwell time of the treatment beam is determined on the basis of a specified relationship between dwell time and temperature at a given location of the coating and on the basis of a specified relationship between dwell time at a given location of the coating at a specified temperature and local thickness variation of the coating.

As further described above, not only the temperature reached at a given location of the coating, but also the hold time and/or residence time are important for the local thickness variation. In the variant described here, two calibration curves are used to determine the residence time, wherein a first bar defines a specified relationship between residence time and temperature at a given location of the coating, and wherein a second bar defines a specified relationship between residence time at a given temperature and local thickness variation. It should be appreciated that the residence time may alternatively be determined directly by a specified relationship to the local thickness variation (see above).

In another variation, the method includes: the coating and/or the substrate is cooled during introduction of the localized energy input. To avoid large area heating of the mirror, it may be necessary to cool the coating or the coated surface simultaneously. Such cooling may be achieved, for example, by means of an air flow directed over the surface of the coating or substrate.

Another aspect of the invention relates to a mirror of the type mentioned in the introductory part, wherein the coating has a local target thickness corresponding to the local target reflectivity of the optical element, wherein the local target thickness is manufactured in accordance with the method for producing local thickness variations as claimed in any of the preceding claims or in accordance with the method for producing local thickness variations as claimed in any of the preceding claims. Such a mirror has high optical properties because it has a desired local target reflectivity or spectral response.

In one embodiment, the material of the substrate is selected from the group comprising: silicate glass, in particular quartz glass doped with titanium dioxide, or glass ceramic, wherein the material in particular has a coefficient of thermal expansion of less than 100ppb/K at 22 ℃. The material of the substrate is preferably a material that allows the shape of the surface to be changed by densification or, if appropriate, expansion during irradiation with a treatment beam, for example an electron beam.

Such substrate materials are titania-doped quartz glass having silicate glass contents typically greater than 90%. Such silicate glasses are commercially available under the trade name of Corning incAlternatively, glass ceramics can be used, wherein the ratio of crystalline phase to glass phase is set in such a way that the coefficients of thermal expansion of the different phases exactly cancel each other out, as a result of which these substrate materials are likewise characterized by an extremely low thermal expansion (less than 100ppm/K at 22 ℃) and are therefore particularly suitable for substrates for EUV mirrors. Such glass ceramics are for example sold under the trade name Schott AGOr Ohara Inc. under the trade nameProviding.

In another embodiment, the coating forms a multi-layer coating having a plurality of alternating layers of a material having a low refractive index and a material having a high refractive index. Such a multilayer coating forms an interference layer system for reflecting radiation, for example for reflecting EUV radiation. The material of the alternating layers depends on the wavelength used for the operation of the mirror. In the case of a wavelength of about 13.5nm used, the layers are typically silicon and molybdenum, which are applied alternately or in pairs to the surface of the substrate. Other material combinations such as molybdenum and beryllium, ruthenium and beryllium, or lanthanum and B ₄ C are also possible, depending on the wavelength used. As further described above, in the case of such a multilayer coating, the spectral response or reflectivity of the mirror depends on the periodic thickness of the corresponding layer pairs of two different materials.

Another aspect of the invention relates to an EUV lithography system comprising: at least one mirror, which is designed as described further above. The EUV lithography system may be an EUV lithography apparatus for exposing a wafer, or may be some other optical arrangement using EUV radiation, such as an EUV inspection system, for example for inspecting a mask, wafer, etc. used in EUV lithography. The mirror may be arranged in particular in a projection system of an EUV lithographic apparatus to image a pattern on a mask onto a wafer.

Other features and advantages of the invention will be apparent from the following description of exemplary embodiments of the invention, with reference to the accompanying drawings, which illustrate the necessary details of the invention, and from the claims. In variations of the invention, the individual features may each be implemented individually or together in any combination.

Drawings

An exemplary example is shown in the schematic diagram and explained in the following description. In the drawings:

Figure 1 schematically shows a meridional cross-section of a projection exposure apparatus for EUV projection lithography,

Figure 2a shows a schematic view of a multilayer coating applied to a substrate with an excess of thickness,

Figure 2b shows the irradiation of a coating applied to a substrate for producing a local thickness variation,

Figure 2c shows the irradiation of a coated substrate for changing the surface shape of the substrate,

Figure 3 shows a schematic view of a local thickness variation for producing a target thickness profile of a coating,

Fig. 4a, b show schematic diagrams of calibration curves for determining the local residence time of an electron beam, which is used to produce the desired change in local thickness during irradiation of the coating shown in fig. 2b,

Fig. 5a, b show schematic diagrams of an additional variation of the local thickness of the coating caused by irradiating the substrate and a local thickness variation for generating a target thickness profile of the coating, the target thickness profile taking into account the additional variation of the local thickness,

Fig. 6a, b show schematic diagrams of additional variations in the local thickness of the coating layer expected as a result of irradiating the substrate and local thickness variations for generating a target thickness profile of the coating layer, which target thickness profile takes into account the expected variations in the local thickness.

Detailed Description

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

The basic components of an optical arrangement for EUV lithography in the form of a microlithographic projection exposure apparatus 1 (EUV lithography apparatus) are described below by way of example with reference to fig. 1. The description of the basic structure of the projection exposure apparatus 1 and its components should not be regarded as limiting.

An embodiment of the illumination system 2 of the projection exposure apparatus 1 has, in addition to the light source or radiation source 3, an illumination optical unit 4 for illuminating an object field 5 in an object plane 6. In alternative embodiments, the light source 3 may also be provided in the form of a module separate from the rest of the lighting system. In this case the lighting system does not comprise a light source 3.

The reticle 7 arranged in the object field 5 is illuminated. The reticle 7 is held by a reticle holder 8. The reticle holder 8 is displaceable by a reticle displacement drive 9, in particular in the scanning direction.

The embodiment of the illumination system 2 of the projection exposure apparatus 1 has, in addition to the light source or radiation source 3, an illumination optical unit 4 for illuminating an object field 5 in an object plane 6. In alternative embodiments, the light source 3 may also be provided as a module separate from the rest of the lighting system. In this case the lighting system does not comprise a light source 3.

By way of illustration, fig. 1 shows a cartesian xyz coordinate system. The x-direction extends perpendicularly into the plane of the drawing. The y-direction extends horizontally and the z-direction extends vertically. The scanning direction extends in the y-direction in fig. 1. The z-direction extends perpendicular to the object plane 6.

The projection exposure apparatus 1 comprises a projection system 10. The projection system 10 is used for imaging the object field 5 into an image field 11 in an image plane 12. The structures on the reticle 7 are imaged onto a photosensitive layer of a wafer 13 arranged in the region of an image field 11 in an image plane 12. The wafer 13 is held by a wafer holder 14. The wafer holder 14 is displaceable, in particular in the y-direction, by a wafer displacement drive 15. The displacement of the reticle 7 firstly by the reticle displacement drive 9 and secondly by the wafer 13 of the wafer displacement drive 15 can be synchronized with one another.

The radiation source 3 is an EUV radiation source. The radiation source 3 emits in particular EUV radiation 16, which is also referred to hereinafter as the radiation used, illumination radiation or illumination light. In particular, the radiation used has a wavelength in the range between 5nm and 30 nm. The radiation source 3 may be a plasma source, such as an LPP source (laser generated plasma) or a GDPP source (gas discharge generated plasma). It may also be a synchrotron-based radiation source. The radiation source 3 may be a Free Electron Laser (FEL).

The illumination radiation 16 emitted from the radiation source 3 is focused by a collector mirror 17. The collector 17 may be a collector having one or more elliptical and/or hyperbolic reflecting surfaces. The illumination radiation 16 may be incident on at least one reflective surface of the collector mirror 17 at Grazing Incidence (GI) (i.e., at an angle of incidence greater than 45 °) or Normal Incidence (NI) (i.e., at an angle of incidence less than 45 °). The collector mirror 17 can be constructed and/or coated firstly for optimizing its reflectivity to the radiation used and secondly for suppressing extraneous light.

Downstream of the collector mirror 17, the illumination radiation 16 propagates through an intermediate focus in an intermediate focus plane 18. The intermediate focal plane 18 may constitute a separation between the radiation source module comprising the radiation source 3 and the collector mirror 17 and the illumination optical unit 4.

The illumination optical unit 4 comprises a deflecting mirror 19 and a first facet mirror 20 arranged downstream thereof in the beam path. The deflection mirror 19 may be a planar deflection mirror or, alternatively, a mirror with a beam influencing effect exceeding the pure deflection effect. Alternatively or additionally, the deflection mirror 19 may be in the form of a spectral filter which separates the used light wavelength of the illumination radiation 16 from extraneous light of a different wavelength. The first facet mirror 20 comprises a plurality of individual first facets 21, which are also referred to as field facets in the following. Fig. 1 depicts by way of example only some of the facets 21. In the beam path of the illumination optical unit 4, a second facet mirror 22 is arranged downstream of the first facet mirror 20. The second facet mirror 22 includes a plurality of second facets 23.

The illumination optical unit 4 thus forms a bipartite system. This basic principle is also known as fly's eye integrator. By means of the second facet mirror 22, the respective first facet 21 is imaged into the object field 5. The second facet mirror 22 is the last beam-shaping mirror or indeed the last mirror for the illumination radiation 16 in the beam path upstream of the object field 5.

The projection system 10 comprises a plurality of mirrors Mi, which are numbered consecutively according to their arrangement in the beam path of the projection exposure apparatus 1.

In the example shown in FIG. 1, projection system 10 includes six mirrors M1 through M6. Alternatives with four, eight, ten, twelve or any other number of mirrors Mi are equally possible. The penultimate mirror M5 and the last mirror M6 each have a through opening for the illumination radiation 16. The projection system 10 is a dual-shielding optical unit. The projection optical unit 10 has an image-side numerical aperture of more than 0.4 or 0.5 and also more than 0.6 and can, for example, be 0.7 or 0.75.

Just as the mirrors of the illumination optical unit 4, the mirrors Mi may have a coating which is highly reflective for the illumination radiation 16 (EUV radiation).

Fig. 2a-c show by way of example a fourth mirror M4 of the projection system 10 of the EUV lithographic apparatus 1 of fig. 1. Mirror M4 has a substrate 25 and a multilayer coating 26 with high reflectivity for EUV radiation 16. In the example shown, the substrate 25 is made of titanium dioxide dopedIs formed of quartz glass of (a). The substrate 25 may also be formed of another material having a low coefficient of thermal expansion, which should typically be less than 100ppb/K at 22 ℃.

A reflective multilayer coating 26 having a plurality of alternating layers 27a, 27b of low refractive index material and high refractive index material is applied to the substrate 25. In the example shown, in which the wavelength used for the EUV lithography apparatus 1 is about 13.5nm, the layers 27a, 27b are silicon and molybdenum, which are applied in pairs one above the other to the surface 25a of the substrate 25. The representation of the additional functional layers of the coating 26 is omitted in fig. 2 a-c.

The coating 26 is applied to the substrate 25 by conventional methods for depositing thin layers 27a, 27b, such as by physical or chemical vapor deposition. For the sake of simplicity, it is assumed that the surface 25a of the substrate 25 is planar and that the coating 26 applied to the surface 25a of the substrate 25 should have a constant uniform target thickness D _S (x, y) across the surface 25 a.

For operation of the mirror M4 in the projection system 10 of the EUV lithographic apparatus 1, a target reflectivity R _S (x, y) is specified which generally depends on the position (x, y) on the surface 25a of the substrate 25. In the example shown, it is assumed that the target reflectivity R _S (x, y) is constant or uniform over the surface 25a and that the target reflectivity R _S (x, y) is achieved when the coating 26 has a likewise uniform specified target thickness D _S (x, y).

In general, for the wavelengths used, the local reflectivity R (x, y) of the mirror M4 or of the coating 26 depends on the thickness d _P (x, y) of the respective layer pair 27a, 27b, which defines the period length or period thickness of the coating 26: if the thickness d _P (x, y) of the respective layer pair 27a, 27b is reduced or increased, the spectral response changes, i.e. the maximum of the spectral reflectivity R (x, y) of the mirror M4 is shifted to other wavelengths, so that the reflectivity R (x, y) at the wavelength used changes. If it is assumed that the coating 26 has a number N (e.g., n=50 to 90) of identical layer pairs 27a, 27b, the following applies to the local target thickness D _S(x,y)：D(x,y)＝N×d_P (x, y) of the coating 26, where D _P (x, y) represents the thickness of the individual layer pairs 27a, 27 b.

If the coating 26 is applied to the substrate 26 at a thickness that deviates from the local target thickness D _S (x, y), for example because of coating errors occurring during deposition of the layers 27a, 27b, the local actual thickness D _I (x, y) of the coating 26 may deviate from the local target thickness D _S (x, y) of the coating 26, assuming in the example shown that the local target thickness D _S (x, y) of the coating 26 is uniform, as shown in fig. 2 b.

In the example shown in fig. 2b, in order to produce a uniform specified target thickness D _S (x, y) of the coating 26 from the coating 26 applied with the actual thickness D _I (x, y), a local energy input E (x, y) is introduced into the coating 26. The local energy input E (x, y) causes densification in the material of the layers 27a, 27b of the coating 26 described herein, i.e., when the energy input E (x, y) is introduced into the coating 26, the thickness of the coating 26 is reduced, i.e., a local thickness change delta (x, y) of the coating 26 is produced.

Since the introduction of the local energy input E (x, y) can only cause densification but cannot cause expansion of the coating 26, it is necessary to produce a local target thickness D _S (x, y) of the coating 26 by the local thickness variation Δ (x, y) described further above in the context described herein to apply a coating 26 having a local thickness D _B (x, y) that is greater than the local target thickness D _S (x, y) of the coating 26 (see fig. 2 a). Due to this thickness excess, the coating 26, whose local actual thickness D _I (x, y) is between the target thickness D _S (x, y) and the (larger) thickness D _B (x, y) applied during coating, can be densified to a uniform target thickness D _S (x, y) as shown in the example (see fig. 2 b) due to the local energy input E (x, y).

The oversized or excessive thickness when applying the coating 26 is not created by increasing the number of layer pairs 27a, b but by scaling the local thickness d _P (x, y) of the individual layer pairs 27a, b: for example, the following may be applicable to the local thickness D _B(x,y)：D_B(x,y)＝(1+f)×D_S(x,y)＝(1+f)×N×d_P (x, y) of the applied coating 26, where f is the oversized (based on the local target thickness D _S (x, y)) when the coating 26 is applied. If the local thickness D _B (x, y) is selected to be 0.5% greater than the local target thickness D _S (x, y), for example, when the coating 26 is applied, the following applies: f=0.005.

The coating 26 heats up substantially uniformly in the thickness direction (Z direction) when the energy input E (x, y) is introduced, such that the local actual thickness D _I (x, y) of the coating 26 scales as a whole when the energy input E (x, y) is introduced. In the case where the coating 26 has a local actual thickness D _I (x, y) at a specified location (x, y) of the surface 25a of the substrate 25, this local actual thickness D _I (x, y) corresponds to the target thickness D _S (x, y) (i.e., D _I(x,y)＝(1+k)D_S (x, y)) scaled by (1+k), where the following applies to k (0.ltoreq.k.ltoreq.f), in order to produce the target thickness D _S (x, y), a local thickness change Δ (x, y) at this location (x, y) must be caused, which causes a scaling of 1/(1+k).

Depending on the type of material of the coating 26, the energy input E (x, y) may cause expansion of the coating 26 rather than densification of the coating 26. In this case, the local thickness D _B (x, y) of the coating 26 applied during coating is selected to be less than the target thickness D _S (x, y) of the coating 26. In order to scale the coating 26 in terms of expansion caused by the energy input E (x, y), the same applies as described above in connection with densification of the coating 26.

To determine the local thickness variation delta (x, y) that produces the target thickness D _S (x, y) of the coating 26, the local actual reflectivity R _I (x, y) of the mirror M4 shown in fig. 2b is measured. For this purpose, the mirror M4 can be introduced into a suitable measuring device which allows the actual reflectivity R _I (x, y) of the mirror M4 to be measured at any position (x, y) in the EUV wavelength range or at the wavelength used in the wavelength range. Determining the actual reflectivity R _I (x, y) over a wavelength range that includes the maximum reflectivity and full width at half maximum of the wavelength dependent reflectivity curve is generally sufficient to determine the local actual thickness (x, y) of the coating 26 based on the relationship of the period lengths d _P (x, y) of the pair of layers 27a, b described above.

Based on the local actual thickness D _I (x, y) and the local target thickness D _S (x, y) of the coating specified by the mirror design, it is possible to determine the local thickness variation Δ (x, y) = |d _I(x,y)-D_S (x, y) | that must be produced by the local energy input E (x, y) in order to produce the local target thickness D _S (x, y). Fig. 3 shows an example of the specification of the local thickness variation Δ (x, y) determined in the above-described manner. The local thickness variation D _S (x, y) shown in fig. 3 is shown for a square surface 25a (-10 < x <10; 10< y < 10) of the M4 mirror, but it is understood that the mirror M4 or the substrate 25 may also have a surface 25a, which surface 25a has another geometry, for example a substantially circular geometry.

In order to produce the desired localized densification delta (x, y) of the coating 26 by introducing an energy input E (x, y), it is necessary to locally heat the coating 26 at a given location (x, y) and continue to heat the coating 26 until the desired densification is achieved. To introduce a local energy input E (x, y) in the example shown in fig. 2b, an electron gun 29 is used, which generates an electron beam 28, and the coating 26 is irradiated with the electron beam 28. The electron gun 29 may be laterally displaced or deflected to direct the electron beam 28 at any location (x, y) on the surface 25 a. The energy of the electron beam 28 is selected such that the effective penetration depth of the electron beam 28 is located directly below the surface 25a of the substrate 25. In this way, the energy input E (x, y) is introduced substantially uniformly by the electron beam 28 at a given location (x, y) of the coating 26 in the thickness direction (Z direction).

The intensity of the electron beam 28 may be set by the current applied to the electron gun 29. However, in general, the current supplied to the electron gun 29 remains constant during irradiation, and the energy dose or partial compaction introduced at a given location (x, y) is set via the dwell time t _D of the electron beam 28 at the given location (x, y).

Fig. 4a shows a calibration curve, wherein the (local) densification as a function of the period of time (holding time T _H in s) for holding the respective temperatures T ₁ to T ₁₀ is plotted on the ordinate, which can be observed at different local temperatures T ₁ to T ₁₀, which local temperatures T ₁ to T ₁₀ are assumed to be constant at a given location (x, y). The local temperature T ₁ corresponds to the lowest temperature in the example shown in fig. 4a, and the local temperature T ₁₀ corresponds to the highest temperature in the example shown in fig. 4 a.

Fig. 4b shows the relationship between the time profile of the temperature T at a given location (x, y) of the surface 25a or coating 26 and the residence time T _D of the electron beam 28 at that location (x, y). As shown in fig. 4b, at a given intensity of the electron beam 28, a constant temperature T is reached after a relatively short dwell time T _D. Thus, the residence time t _D of the electron beam 28 at a given location (x, y), which causes the desired change in the local thickness Δ (x, y) of the coating 26, can be determined by the calibration curves shown in fig. 4a, 4 b.

The correction of the local thickness D _I (x, y) of the coating 26 shown in fig. 2b for producing the local target thickness D _S (x, y) can in principle be carried out at any time in the production process chain of the mirror M4 after the application of the coating 26. It should be noted, however, that in subsequent process steps no higher temperatures occur in the coating 26 than in the case of irradiation described in connection with fig. 2 b.

Fig. 2c shows such a step in the process chain, wherein the mirror M4 as in fig. 2b is irradiated with an electron beam 28 in order to cause a local energy input E (x, y). In contrast to the irradiation depicted in fig. 2b, the energy of the electron beam 28 is chosen to be larger such that it penetrates the substrate 25 below the surface 25a and substantially heats the substrate 25 without heating the reflective coating 26. Due to the illumination shown in fig. 2c, the surface shape of the surface 25a of the substrate 25 is changed to produce the planar target surface shape of the substrate 25 shown in fig. 2 a. In case a high current or high intensity electron beam 28 is used in the irradiation shown in fig. 2c to increase the throughput time, local temperature peaks may occur during irradiation of the coated substrate 25, which changes the local thickness of the coating 26 (in the example shown decreases), as shown in fig. 2c, for example, at the location (x, y) where the electron beam 28 is incident on the surface 25a of the substrate 25.

In case the change in the surface shape of the substrate 25 affects the local actual thickness D _I (x, y) of the coating 26, an additional (undesired) change Δ _S of the local thickness occurs. Fig. 5a shows an example of such additional variation Δ _S (x, y) of the local thickness, resulting from measurements of the local thickness D _I (x, y) or the reflectivity R _I (x, y) of the coating 26 before and after the irradiation shown in fig. 2 c. To compensate for the additional local thickness variation +Δ _S (x, y) caused by the illumination shown in fig. 2b, the local thickness variation Δ (x, y) shown in fig. 3 needs to be adapted appropriately: fig. 5b shows the local thickness variation delta' (x, y) required for this purpose, which consists of the thickness variations shown in fig. 3 and 5a, as follows: delta' (x, y) =delta (x, y) -delta _S (x, y).

For the generation of the target thickness D _S (x, y), knowledge of the above relationship is generally not required, as long as the actual thickness D _I (x, y) of the coating 26 is measured after the step of changing the surface shape of the substrate 25 shown in fig. 2 c. In this case, the step of producing the local target thickness D _S (x, y) of the coating 26 shown in fig. 2b is performed after changing the surface shape of the substrate 25 in the manner described in connection with fig. 2 b.

As an alternative to measuring the actual thickness D _I (x, y) of the coating 26 after changing the surface shape of the substrate 25, it is also possible to measure the actual thickness D _I (x, y) of the coating 26 before changing the surface shape of the substrate 25 and calculate or estimate the additional local thickness variation a _S,E (x, y) of the coating 26 that is expected when changing the surface shape of the substrate 25, as shown in fig. 2 c. This is possible because the local energy input E (x, y) generated when changing the surface shape of the substrate 25 and the local temperature T (x, y) generated at a given location (x, y) of the coating 26 are known in advance, for example because this has been determined or calculated experimentally.

Fig. 6a shows an example of an expected additional thickness variation Δ _S,E (x, y) due to a variation in the surface shape of the substrate 25 shown in fig. 2 c. As described in connection with fig. 5a, 5b, the resulting local thickness variation delta' "(x, y) shown in fig. 6b required to achieve the local target thickness D _S (x, y) of the coating 26 is made up of the local thickness variation delta (x, y) shown in fig. 3 and the additional local thickness variation delta _S,E (x, y) shown in fig. 6a as follows:

Δ“(x,y)＝Δ(x,y)-Δ_S,E(x,y)。

In the event of no or only negligible coating errors, the local thickness variation delta (x, y) of the coating 26 may be used to compensate only for the local thickness variation delta _S (x, y) or delta _S,E (x, y) that occurs when the surface shape of the substrate 25 is changed. In this case: delta' (x, y) = -delta _S (x, y) or delta "(x, y) = -delta _S,E (x, y).

When introducing the energy input E (x, y) into the coating 26 as shown in fig. 2b and the energy input E (x, y) into the substrate 25 as shown in fig. 2c, it may be advantageous to counteract the large area heating of the mirror M4 by cooling. This can be achieved, for example, by a gas flow 30 indicated by arrows in fig. 2b and 2c, the gas flow 30 cooling the mirror M4 when the energy input E (x, y) is introduced. However, the cooling of the mirror M4 may also be achieved in other ways, for example by bringing the substrate 25 into contact with suitable cooling means.

In the above manner, coating errors and errors due to unwanted variations in the layer thickness of the coating 26 during variations in the surface shape of the substrate 25 can be compensated for such that the coating 26 reaches its local target thickness D _S (x, y) and corresponding local target reflectivity R _S (x, y).

Claims

1. Method for producing a local thickness variation (Δ (x, y), Δ″ (x, y)) of a coating (26) for reflecting radiation, in particular for reflecting EUV radiation (16), the coating (26) being applied to a substrate (25) of a mirror (M4), the method comprising:

-generating the local thickness variation (Δ (x, y); Δ″ (x, y)) by introducing a local energy input (E (x, y)) into the coating (26), the local energy input (E (x, y)) densifying or expanding the coating (26) to generate a local target thickness (D _S (x, y)) of the coating (26) corresponding to a local target reflectivity (R (x, y)) of the mirror (M4), the method further comprising:

The surface shape of the substrate (25) is changed by introducing a local energy input (E (x, y)) into the substrate (25), wherein a local thickness variation (delta (x, y)) of the coating (26) is produced before and/or after changing the surface shape of the substrate (25).

2. The method of claim 1, further comprising:

Applying the coating (26) to the substrate (25), wherein

-Applying the coating (26) with a local thickness (D _B (x, y)) that is larger than a local target thickness (D _S (x, y)) of the coating (26) if the local energy input (E (x, y)) results in densification of the coating (26), or-wherein the coating (26) is applied with a local thickness (D _B (x, y)) that is smaller than a local target thickness (D _S (x, y)) of the coating (26) if the local energy input (E (x, y)) results in expansion of the coating (26).

3. The method of claim 1 or 2, further comprising:

The local actual thickness (D _I (x, y)) of the coating (26) is determined by measuring the local actual reflectivity (R _I (x, y)) of the coating (26).

4. A method according to claim 3, wherein the local thickness variation (Δ (x, y)) corresponds to a difference between the local target thickness (D _S (x, y)) and the local actual thickness (D _I (x, y)).

5. The method according to any of the preceding claims, wherein an additional local thickness variation (Δ _S,E (x, y)) of the coating (26) that is expected when changing the surface shape of the substrate (25) is determined, and wherein the additional local thickness variation (Δ _S,E (x, y)) that is expected due to the change in the surface shape of the substrate (25) is compensated when generating the local thickness variation (Δ "(x, y)).

6. The method according to any of the preceding claims, wherein the coating (26) and/or the substrate (25) is irradiated with a treatment beam (28) to locally introduce the energy input (E (x, y)).

7. The method of claim 6, wherein the processing beam is selected from the group consisting of: an ion beam, an electron beam (28), and a laser beam.

8. The method of claim 6 or 7, further comprising:

A dwell time (t _D) of the treatment beam (28) at a given location (x, y) of the coating (26) is determined, the dwell time (t _D) resulting in a local thickness variation (Δ (x, y); Δ' (x, y)) of the coating (26).

9. The method of claim 8, wherein the dwell time (T _D) of the processing beam (28) is determined based on a specified relationship between the dwell time (T _D) and a temperature (T) at the given location (x, y) of the coating (26) and based on a specified relationship between a hold time (T _H) at a specified temperature (T ₁,…,T₁₀) at the given location (x, y) of the coating (26) and the local thickness variation (Δ (x, y)) of the coating (26).

10. The method of any of the preceding claims, further comprising:

-cooling the coating (26) and/or the substrate (25) during introduction of the local energy input (E (x, y)).

11. A mirror (M4), comprising:

a substrate (25), and

A coating (26), the coating (26) being applied to the substrate (25) and being for reflecting radiation, in particular EUV radiation (16),

It is characterized in that

The coating (26) has a local target thickness (D _S (x, y)) corresponding to a local target reflectivity (R _S (x, y)) of the mirror (M4), wherein the local target thickness (D _S (x, y)) is produced according to the method for producing the local thickness variation (Δ (x, y), Δ″ (x, y)) as set forth in any one of the preceding claims.

12. Mirror according to claim 11, wherein the material of the substrate (25) is selected from the group comprising: silicate glass, in particular quartz glass doped with titanium dioxide, or glass ceramic, wherein the material in particular has a coefficient of thermal expansion of less than 100ppb/K at 22 ℃.

13. Mirror according to claim 11 or 12, wherein the coating (26) forms a multilayer coating having a plurality of alternating layers (27 a, 27 b) of a material having a low refractive index and a material having a high refractive index.

14. An EUV lithography system (1), comprising:

At least one mirror (M1 to M6) according to any one of claims 11 to 13.