WO2020244937A1

WO2020244937A1 - Measuring apparatus for interferometrically determining a shape of an optical surface of a test object

Info

Publication number: WO2020244937A1
Application number: PCT/EP2020/064294
Authority: WO
Inventors: Ulrich Löring
Original assignee: Carl Zeiss Smt Gmbh
Priority date: 2019-06-03
Filing date: 2020-05-22
Publication date: 2020-12-10
Also published as: DE102019208029A1

Abstract

A measuring apparatus (10) for interferometrically determining a shape of an optical surface (12) of a test object (14) comprises a diffractive optical module (24) for generating a test wave (26), which is directed at the optical surface and which has at least one wavefront partially adapted to a target shape (12-2) of the optical surface, and a reference wave (28), a reflective reference element, which is arranged in the beam path of the reference wave and which is displaceably mounted for the purpose of being arranged at different reference positions in the direction of the beam path of the reference wave, a holding device (15), which is displaceably mounted in the direction of the beam path of the test wave (26) and which is provided to arrange the test object at different test positions (17), a detection device (32) for detecting interferograms generated at the different test positions of the test object, said interferograms each being formed by a superposition of the test wave following an interaction with the optical surface and the reference wave following a reflection at the reference element. Here, a plurality of interferograms are respectively assigned to at least one of the test positions, the reference element being arranged at different reference positions when said interferograms are generated. Furthermore, the measuring apparatus comprises an evaluation device (38), which is configured to determine a deviation of an actual shape (12-1) of the optical surface (12) from the target shape (12-2) by a respective ascertainment of a deviation result for each of the test positions by means of evaluating the interferograms generated at the different test positions and averaging the deviation results, the deviation result being ascertained on the basis of the assigned interferograms for the at least one test position with a plurality of assigned interferograms.

Description

Measuring device for the interferometric determination of a shape of an optical surface of a test object

The present application claims the priority of German patent application 10 2019 208 029.4 of June 3, 2019. The entire disclosure of this patent application is incorporated into the present application by reference. Background of the invention

The invention relates to a measuring device and a method for interferometric determination of a shape of an optical surface of a test object. A highly precise interferometric measurement of a surface shape of a test object, for example an optical element for a projection objective of a microlithographic exposure system, is often carried out by means of a measuring device comprising a diffractive optical module as so-called zero optics or compensation optics. The wavefront of a measurement wave is adapted to a target shape of the surface by the diffractive optical module in such a way that it strikes the surface perpendicularly at every location and is reflected back by this when the target shape is present. Deviations from the nominal shape can be determined by superimposing the reflected test wave with a reference wave likewise generated by the diffractive optical module. The diffractive optical module can comprise a diffractive element in the form of a computer-generated hologram (CGH).

However, scattered radiation that occurs in the measuring device often falsifies the measurement result. This scattered radiation is often referred to as “false light” and in particular includes measurement radiation diffusely reflected in the measuring device, caused for example due to lens reflections, mirror inaccuracies, light leakage and soiling in the measuring device as well as reflection and scattering on imperfectly blackened mechanical components / surfaces. The scattered radiation interferes with the useful light and causes falsifications of the measurement result due to the accompanying phase change of the useful light, which conventionally often does not differ from the measurement results of actual surface deviations. In any case, scattered radiation reduces the measurement accuracy.

Underlying task

It is an object of the invention to provide a measuring device and a method with which the aforementioned problems are solved, in particular the measurement accuracy is increased and in particular measurement errors which can be attributed to scattered radiation are suppressed.

Solution according to the invention

The aforementioned object can be achieved according to the invention, for example, with the measuring device described below for interferometric determination of a shape of an optical surface of a test object. This measuring device comprises a diffractive optical module for generating a test wave which is directed onto the optical surface and which has at least a wave front which is partially adapted to a nominal shape of the optical surface, as well as a reference wave. Furthermore, the measuring device comprises a reflective reference element arranged in the beam path of the reference wave, which is mounted displaceably in the direction of the beam path of the reference wave for arrangement at different reference positions, a holding device which is mounted displaceably in the direction of the beam path of the test shaft for arranging the test object at different test positions, and a detection device for capturing interferograms generated at the different test positions of the test object, each of which is based on a superimposition of the test wave Interaction with the optical surface and the reference wave are formed after reflection on the reference element. In this case, at least one of the test positions is assigned a plurality of interferograms in each case, the generation of which the reference element is arranged at different reference positions. Furthermore, the measuring device comprises an evaluation device which is configured to determine a deviation of an actual shape of the optical surface from the nominal shape by determining a deviation result for each of the test positions by evaluating the interferograms generated at the different test positions and averaging the deviation results, for the at least one test position with a plurality of assigned interferograms, the determination of the deviation result takes place on the basis of the assigned interferograms.

The interferogram is generated in particular by superimposing the test wave with the reference wave in a detection plane after the test wave has interacted with the optical surface of the test object and renewed diffraction on the diffractive optical module and back reflection of the reference wave on the reference element and renewed diffraction on the diffractive optical element. The shape of the optical surface is determined by calculating a deviation of the optical surface from its nominal shape. The diffractive optical module can consist of a diffractive optical element; alternatively, it can additionally comprise a further optical element, in particular also a further diffractive optical element.

By arranging the test object at the different test positions and recording the interferograms generated at the different test positions, it is possible to suppress "static" measurement errors, i.e. measurement errors that are due to phase-stable scattered radiation during the measurement process, in the measurement result and thus increase the measurement accuracy of the measurement device. The scattered radiation is also referred to as "false light" in this document. The reference element is mounted displaceably for arrangement at different reference positions. The reference element is mounted in such a way that it can be displaced in the direction of the beam path of the reference wave. According to one embodiment variant, the reference element is mounted displaceably over an area of at least 1 times, in particular at least 2 times or at least 3 times the wavelength of the test shaft. The displaceability relates to an optical displacement, ie a displacement occurring in the wavefront effect. Mechanically, the reference element may only have to be shifted by half the value.

According to one embodiment, the evaluation device is configured to evaluate the interferograms in such a way that errors in the evaluation result generated by scattered radiation are minimized.

According to a further embodiment, adjacent test positions are at most a fraction of a wavelength of the test shaft spaced apart from one another. In particular, the adjacent test positions are spaced apart from one another by exactly a fraction of the wavelength of the test wave.

The evaluation device is configured to determine the deviation of the actual shape of the optical surface from the target shape by determining a deviation result for each of the test positions by evaluating the interferograms generated at the different test positions and averaging the deviation results. By averaging the deviation results, a portion of the end result that can be attributed to scattered radiation can be reduced or eliminated. The averaging can be the formation of a simple mean value in which the deviation results are included in the mean value without weighting. Alternatively, a weighted average can also be used. As a further alternative, a mathematical adaptation (fit) of the course of the error expected for different test positions can also take place by means of scattered radiation in order to separate it as far as possible. According to one embodiment, the reference element is mounted in such a way that it can be displaced in steps of less than a fifth, in particular less than a tenth or less than a twentieth of the wavelength of the test shaft. Here, too, the displaceability relates to an optical displacement, that is to say which occurs in the wavefront effect.

According to one embodiment, the respective deviation result is determined for each of the test positions using interferograms, the generation of which the reference element is arranged at different reference positions.

According to a further embodiment, the reference element for arrangement at different reference positions is mounted displaceably in the direction of the beam path of the reference wave and the measuring device is configured to arrange the test object one after the other at the different test positions and at the same time to change the reference position of the reference element.

According to a further embodiment, the measuring device is configured to move the reference element simultaneously when the test object is moved between test positions in such a way that the effect of the displacement of the test object and the effect of the displacement of the reference element on the phase difference between the test wave and the reference wave in the resulting interferogram have the same sign. This means that the phase difference generated in the interferogram is generated by simultaneous phase shifting of the test object and reference element.

According to a further embodiment, the evaluation device is configured to use the interferograms as a basis for at least one measurement point on the optical surface, which varies from interferogram to interferogram as a function of the respective interferogram to determine lying phase difference between the test wave and the reference wave and to calculate a portion to be traced back to scattered radiation from the determined intensity profile. This calculated portion of the scattered radiation is, in particular, a phase difference variable scattered radiation portion. By "calculating" the portion attributable to scattered radiation is to be understood as first determining the portion and then partially or completely mathematically eliminating it from the intensity curve. In particular, with respect to a plurality of measurement points on the optical surface, an intensity profile that varies from interferogram to interferogram is determined as a function of the phase difference, and one for each of the measurement points

Part of the scattered radiation attributable to the corresponding intensity curve is calculated.

According to a further embodiment, the evaluation device is configured to perform a Fourier decomposition of the intensity profile determined when calculating out the portion attributable to scattered radiation.

According to a further embodiment, the measuring device is configured to determine the shape of a surface of an EUV mirror for microlithography.

According to a further aspect of the invention, the measuring device described below for interferometric determination of a shape of an optical surface of a test object is also provided according to the invention. This measuring device comprises a diffractive optical module for generating a test wave directed onto the optical surface, which test wave has at least a wavefront partially adapted to a nominal shape of the optical surface, and a reference wave directed onto a reflective reference element. Furthermore, the measuring device comprises a holding device, mounted displaceably in the direction of the beam path of the test shaft, for arranging the test object at different test positions, as well as a detection device for detecting those generated at the different test positions of the test object Interferog rams, which are each formed by superimposing the test wave after interaction with the optical surface and the reference wave after reflection on the reference element. Furthermore, the measuring device comprises an evaluation device which is configured to determine a deviation of an actual shape of the optical surface from the nominal shape by evaluating the interferograms generated at the different test positions. The reference element is mounted displaceably in the direction of the beam path of the reference shaft for arrangement at different reference positions and the measuring device is configured to arrange the test object one after the other at the different test positions and at the same time to change the reference position of the reference element.

According to one embodiment, the measuring device is configured to move the reference element simultaneously when the test object is moved between test positions in such a way that the effect of the displacement of the test object and the effect of the displacement of the reference element on the phase difference between the test wave and the reference wave resulting interferogram have the same sign. This means that the phase difference generated in the interferogram is generated by simultaneous phase shifting of the test object and reference element.

According to a further embodiment, the evaluation device is configured to determine from the interferograms with respect to at least one measurement point on the optical surface an intensity profile that varies from interferogram to interferogram depending on a phase difference between the test wave and the reference wave on which the respective interferogram is based, and from the determined intensity profile to subtract a portion attributable to scattered radiation. This calculated portion of the scattered radiation is, in particular, a phase difference variable scattered radiation portion. By “calculating” the portion that can be traced back to scattered radiation is to be understood to first determine the portion and then to partially or completely eliminate it from the intensity curve. Especially an intensity profile that varies from interferogram to interferogram is determined as a function of the phase difference with respect to several measuring points on the optical surface, and one for each of the measuring points

The features specified in relation to the above-mentioned embodiments, exemplary embodiments or design variants, etc. of the measuring device according to the invention listed at the beginning can be transferred accordingly to the measuring device of the further aspect of the invention.

According to the invention, a method for the interferometric determination of a shape of an optical surface of a test object is also provided. The method comprises the steps of: generating a test wave directed onto the optical surface, which has at least a wavefront that is partially adapted to a nominal shape of the optical surface, and a reference wave directed onto a reflective reference element by means of a diffractive optical module, arranging the reference element at different reference positions in Direction of the beam path of the reference wave, arranging the test object in the direction of the beam path of the test wave one after the other at different test positions, recording of interferograms generated at the different test positions of the test object, each of which occurs by superimposing the test wave after interaction with the optical surface and the reference wave after reflection the reference element are formed, with at least one of the test positions being assigned a plurality of interferograms, when the reference element is generated at different Refe renzpositionen is arranged, and determining a deviation of an actual shape of the optical surface from the desired shape by respective determination a deviation result for each of the test positions by evaluating the interferograms generated at the different test positions and averaging the deviation results, the deviation result being determined using the assigned interferograms for the at least one test position with several assigned interferograms.

According to a further embodiment of the method according to the invention, the test object is arranged one after the other at the different test positions and, at the same time, a reference position of the reference element is changed in the direction of the beam path of the reference wave. According to one embodiment variant, an intensity profile varying from interferogram to interferogram is determined from the interferograms with respect to at least one measuring point on the optical surface, depending on a phase difference between the test wave and reference wave on which the respective interferogram is based, and a portion attributable to scattered radiation is calculated from the determined intensity profile.

According to a further embodiment of the method according to the invention, a Fourier decomposition of the intensity profile determined is carried out in the case of calculating out the portion attributable to scattered radiation.

According to the further aspect of the invention, a method for interferometric determination of a shape of an optical surface of a test object is also provided according to the invention. The method comprises the steps of: generating a test wave directed onto the optical surface, which has at least one wavefront partially adapted to a nominal shape of the optical surface, and a reference wave directed onto a reflective reference element by means of a diffractive optical module, arranging the test object in the direction of the Beam path of the test wave one after the other at different test positions, acquisition of interferograms generated at the different test positions of the test object, each of which is achieved by superimposing the Test wave are formed after interaction with the optical surface and the reference wave after reflection on the reference element, as well as determination of a deviation of an actual shape of the optical surface from the target shape by evaluating the interferograms generated at the different test positions. The test object is arranged one after the other at the different test positions and at the same time a reference position of the reference element is changed in the direction of the beam path of the reference wave.

According to one embodiment, an intensity profile that varies from interferogram to interferogram is determined from the interferograms with respect to at least one measuring point on the optical surface, depending on a phase difference between test wave and reference wave on which the respective interferogram is based, and from the determined intensity profile

Scattered radiation deducted part. According to a further embodiment, a Fourier decomposition of the intensity profile determined is carried out when calculating out the portion attributable to scattered radiation.

The features specified with regard to the above-mentioned embodiments, exemplary embodiments or design variants, etc. of the method according to the invention listed at the beginning can be transferred accordingly to the method of the further aspect of the invention.

The features specified with regard to the above-mentioned embodiments, exemplary embodiments or design variants, etc. of the measuring device according to the invention can be transferred accordingly to the method according to the invention. These and other features of the embodiments according to the invention are explained in the description of the figures and the claims. The individual features can be realized either separately or in combination as embodiments of the invention. Furthermore, they can describe advantageous embodiments that can be independently protected and whose protection may only be claimed during or after the application is pending.

Brief description of the drawings

The above and further advantageous features of the invention are illustrated in the following detailed description of exemplary embodiments according to the invention with reference to the attached schematic drawings. It shows:

1 shows an illustration of a measuring device according to the invention for the interferometric determination of a shape of an optical surface of a test object from recorded interferograms,

2 shows the course of an intensity value from interferogram to interferogram at one point of the interferograms as a function of a wavefront shift for a nominal shape compared to an actual shape of the optical surface,

3 shows a diagram to illustrate the vectors of the field strength of a test wave and of false light,

4 shows intensity profiles analogous to FIG. 2 for the nominal shape of the optical surface without false light and with false light of a first phase difference,

5 shows the diagram according to FIG. 3 for different test positions of the test object, FIG. 6 shows the intensity curves according to FIG. 4 for the nominal shape of the optical surface without false light and with false light of a further phase difference, 7 shows the course of a phase error Aw as a function of the phase difference between the field vectors of the false light and the test wave,

8 shows the intensity curves according to FIG. 4 for the nominal shape of the optical surface without false light and with false light when the test object is arranged at different test positions,

FIG. 9 shows the profile of the phase error Aw according to FIG. 7 for the different test positions,

10 shows the course of the curves from FIG. 9, averaged according to a first embodiment according to the invention,

11 shows a maximum shift <Aw> max of the phase (maximum phase error) as a function of the number of test positions, as well as

FIG. 12 shows the intensity profile recorded according to a second embodiment according to the invention according to FIG. 8 for the nominal shape of the optical surface without and with false light, as well as intensity profiles determined during the evaluation according to the invention.

Detailed description of exemplary embodiments according to the invention

In the exemplary embodiments or embodiments or design variants described below, elements that are functionally or structurally similar to one another are provided as far as possible with the same or similar reference symbols. Therefore, in order to understand the features of the individual elements of a specific exemplary embodiment, reference should be made to the description of other exemplary embodiments or the general description of the invention. To make the description easier, a Cartesian xyz coordinate system is indicated in the drawing, from which the respective positional relationship of the components shown in the figures results. In Fig. 1, the y-direction runs perpendicular to the plane of the drawing into this, the x-direction to the right and the z-direction upwards.

In FIG. 1, an exemplary embodiment of a measuring device 10 for the interferometric determination of the shape of an optical surface 12 of a test object 14 is illustrated. The optical surface 12 has an actual shape 12-1 (TF) which deviates from a nominal shape 12-2 (SF). This is illustrated by way of example in the left section of the test object 14: Here the actual shape TF has a spherical deviation from the nominal shape SF, at the point 29 (Mj) arranged there this is a deviation in the direction of the surface normals by Ah.

With the measuring device 10, in particular a deviation of the actual shape 12-1 of the surface 12 from its desired shape 12-2 can be determined. As a test object 14, for example, a mirror of a projection lens for EUV microlithography with a non-spherical surface for reflecting EUV radiation with a wavelength of less than 100 nm, in particular a wavelength of about 13.5 nm or about 6.8 nm. The non-spherical nominal shape 12-2 of the surface of the mirror can have a free-form surface with a deviation from each rotationally symmetrical asphere of more than 5 μm and a deviation from each sphere of at least 1 mm.

The measuring device 10 contains a holding device 15 for arranging the test object 14 at different test positions 17 as well as a measuring radiation source 16 for providing a sufficiently coherent measuring radiation 19 as an input wave 20. In this embodiment, the measuring radiation source 16 comprises an optical waveguide with an exit surface 18. The optical waveguide is on a radiation generator not shown in FIG. 1, for example in the form of a Lasers, connected. A helium-neon laser with a wavelength of approximately 633 nm, for example, can be provided for this purpose. The measuring radiation 19 can, however, also have a different wavelength in the visible or invisible wavelength range of electromagnetic radiation. The measuring radiation source 16 with the optical waveguide is only one example of a radiation source that can be used for the measuring device. In alternative embodiments, instead of the optical waveguide, an optical arrangement with lens elements, mirror elements or the like to provide a suitable input shaft 20 can be provided.

The measuring device 10 also contains a beam splitter 22, a diffractive optical module 24 and a reference element 30. After passing through the beam splitter 22 undisturbed, the input shaft 20 meets the diffractive optical element 24 to generate a test wave 26 and a reference wave 28 from the input wave 20.

The diffractive optical module 24 can comprise one or more diffractive optical elements. In the present case, it has an optical element in the form of a complex coded CGH and contains diffractive structures which form two diffractive structure patterns that are superimposed on one another in a plane. These two diffractive structure patterns can e.g. be formed by a first structure pattern in the form of a basic lattice and a second diffractive structure pattern in the form of a superlattice. One of the diffractive structure patterns is configured to generate the test wave 26 with a wave front that is at least partially adapted to the nominal shape 12 - 2 of the optical surface 12. The other diffractive structure pattern generates the reference wave 28 with an even wavefront.

In alternative exemplary embodiments, a simply coded CGH with a diffractive structure or another optical grating can also be used instead of the complex coded CGH. The test shaft 26 can for example in a first diffraction order and the reference wave 28 can be generated in the zeroth or any other diffraction order on the diffractive structure. The reference element 30 is configured as a reflective optical element and is used to reflect back the reference wave 28. Since the reference wave 28 is a plane wave in the present case, the reference element 30 is configured here as a plane mirror. In another embodiment, the reference wave 28 can have a spherical wavefront and the reflective optical element can be designed as a spherical mirror.

Furthermore, the measuring device 10 contains a detection device 32. This includes the already mentioned beam splitter 22 for leading out the combination of the reflected test wave 26 and the reflected reference wave 28 from the beam path of the input shaft 20. Furthermore, the detection device includes a two-dimensionally resolving detector 34 and a Focussing lens 36 located in front of it for detecting an interferogram generated by superimposing test wave 26 with reference wave 28. The interferograms recorded by the detector are evaluated by an evaluation device 38, as described in more detail below.

In a conventional measuring device of the type shown in Fig. 1, the test object 14 is arranged in a fixed test position, i. not displaceable between different test positions 17 (1 to Nmax) as in the measuring device 10. The functional principle of such a conventional measuring device will now be explained below with reference to the measuring device 10 illustrated in FIG. 1, in which the holding device 15 for the test object 14 is arranged in a test position 1.

The reference element 30 is mounted to be displaceable in the direction of the beam path of the reference shaft 28 for arrangement at different reference positions 31 (1 to Mmax). During the measurement, the reference element 30 is successively arranged at the different reference positions 31 (1 to Mmax) and the Interferograms imaged on the detector 34 at the different reference positions 31 are recorded. Depending on the reference position, there is a shift w of the wavefront of the reference wave 28 (the greater the index of the reference position 1 to Mmax, the greater the wavefront shift w).

FIG. 2 shows an example of an evaluation of the interferograms recorded at the different reference positions 31 for the two cases illustrated in FIG. 1 in which the optical surface 12 has the actual shape 12-1 (TF) or the desired shape 12-2 (SF ) having. In the diagram shown in FIG. 2, the intensity at the point of the respective interferogram which corresponds to the measurement point 29 (Mj) on the surface 12 is plotted on the y-axis as a function of the wavefront shift w. In FIG. 1, the individual beams 26i and 28i of the test wave 26 and the reference wave 28 are superimposed at the point of the corresponding interference pattern corresponding to the point 29 (Mj).

As already mentioned above, the adjustment of the wavefront shift w takes place by shifting the reference element 30. This can be expressed as follows:

where l is the wavelength of the measurement radiation 19 and F is the phase difference between the phase q> p of the returning test wave 26 upon reflection on the nominal shape SF of the optical surface 12 and the phase qm of the reference wave 28 (f = <pp - cpR). Since the test position 17 of the test object 14 is not changed in the present case, F corresponds to the phase qm of the reference wave 28.

Fig. 2 now illustrates the intensity profile at the point of the interferograms corresponding to the measuring point Mj, on the one hand with Ig _es (S) for the case in which the optical surface 12 has the nominal shape 12-2 (SF), and on the other hand with Ig _it

for the case in which the optical surface 12 has the actual shape 12-1, which is lowered by Ah at the measuring point Mj. The reduction by Ah leads in the diagram of FIG. 2 to the fact that the intensity profile

I _g ^Q _it

compared to the intensity curve Ig _es

shifted to the left by Aw = 2 Ah. By measuring the intensity profile I (TF) _ges at different measurement points 29 and compared with the determined by simulation or Kalibiermessung I _ges

the distribution Ah (x, y) can thus be determined in principle.

According to the invention, however, it has now been recognized that coherent scattered radiation present in the measuring device 10 can produce effects similar to the deviation Ah in the aforementioned intensity profiles, as explained below with reference to FIGS. 3 and 4. This has the consequence that the measurement results for the deviation between the actual shape 12-1 from the desired shape 12-2 of the optical surface are often falsified by such scattered radiation effects. The above-mentioned scattered radiation is also referred to below as "false light" and in particular comprises measurement radiation 19 that is diffusely reflected in the measuring device 10, caused for example by lens reflections, reflections and scattering on imperfectly blackened mechanical components / surfaces, e.g. by light outside the target beam cone of the measuring radiation 19, mirror inaccuracies, light leakage and contamination in the measuring device 10, as well as other interfering light.

In the diagram of FIG. 3, the real part (Re (E)) and imaginary part (lm (E)) of the vector of the field strength E of the test wave 26 (Ep) in connection with a field strength vector of false light (EFL) are shown by way of example. The phase of the field strength vector EP of the test wave 26 is denoted by f ° r and the phase of the field strength vector EFL of the false light is denoted by cpFL. The phase <PFL of the false light can assume any value between 0 and 2p, so that the vector EFL can be anywhere within of the circle drawn in with a broken line in FIG. 3, ie the tip of the vector EFL can point to any point on the circle assuming a constant vector length. The phase difference Df ° between the phase <PFL of the false light and the phase f ° r of the test wave 26 can thus vary between 0 and 2p.

In FIG. 3, the situation for false light is shown by way of example, for which Df ° = tt / 2 and thus the phase fr + FL of the resulting overall vector EP + FL

(Sum of the vectors Ep and EFL) deviates almost maximally from the phase for the test wave 26. The phase angle between the vectors Ep and EP + FL is denoted by DF. The above-mentioned phase difference F is changed by DF.

In addition to the intensity profile Ig _es (SF) from FIG. 2, FIG. 4 shows the intensity _profile Ig _gS

at the point corresponding to the measuring point Mj as a function of the wavefront shift w, which results when the optical surface 12 has the desired shape 12-2 (SF) and furthermore false light with the phase illustrated in FIG. 3 (Df ° = tt / 2 ) is present. The intensity curve Ige _S (SF

is shifted to the left analogously to the intensity curve Ig _es (TF) illustrated in FIG. 2, which results without the presence of false light for the actual shape TF which deviates from the nominal shape SF by Ah. The shift represents an error in the wavefront of test wave 26 and is referred to as Aw or as deviation result 40. False light with Df ° = tt / 2 thus generates a signature which cannot be distinguished from the signature generated by the deviation Ah. This means that if there is an intensity curve shifted to the left with respect to Ig _es (5F), it is not conventionally possible to determine which part of the shift is due to a deviation Ah of the actual shape TS from the target shape SF and which part is due to the presence of false light. In the extreme case, the entire shift can be caused by false light. The field vectors illustrated in Figure 3 are described mathematically below:

Ep = Ap

where the following applies to the amplitude Ap: A _P = -JT _p and Ip is the intensity of the test wave 26. Alternatively, E _P , Ap or Ip can also describe the corresponding variables of the reference shaft 28.

where A

Ipi and IFL is the intensity of the false light.

The following applies: Ap> A _FL > 0 (3)

Ep + FL— E _P + E _FL - A _{P + FL} e ^{i (pp + FL} (4)

As mentioned above, the phase FR + FL of the resulting total vector EP + FL deviates almost at a maximum from the phase f ° r of the test wave 26 when Df ° =

TT 12. The same applies to Df ° = -p / 2. In this case: or ^A FL si h Df (9)

^A NL ~ l-sin ² Df

For DF «1 applies: or. (10)

The error Aw shown in FIG. 4 in the wavefront of the test wave 26 is calculated as follows:

The following applies to the defects Ah in the optical surface 12 illustrated in FIG. 2:

For l = 532 nm and Aw = 1 pm (Ah ~ 0.5 pm) then follows in the case: Dfo = ± p / 2:

with fully coherent lighting.

Assuming 16% for the diffraction efficiency (amplitude) of the CGH and 4% for the reflectivity of the test object 14 (uncoated), only «0.1% of the total light Iges goes into the test wave as“ useful light ”(16% · 16 % 4% = 0.1%):

The following table 1 shows simulation results for the resulting phase error Awmax (in pm) as a function of different false light levels I FL / IP.

Tab. 1 simulation results

While the interference between the test wave 26 (Ep) and the false light (EFL) was considered in the above mathematical formalism, the influence of the reference wave 28 is also taken into account below. The field strength vector of the reference wave 28 is denoted by ER, the amplitude of the reference wave 28 is denoted by AR and the intensity of the reference wave 28 is denoted by IR. The intensity

undisturbed interferogram, which is present without the presence of false light, can be described as follows:

f - f _R - (p _R (18)

If the false light is now also taken into account, the intensity l _ge s of the interferogram then resulting from 3-beam interference can be described as follows:

where Epi = A _FL e ^{i (f> FL} and A _FL =

(19)

If the sum of the field strengths Ep and EFL is now combined as field strength EP + FL, the mathematical consideration can be traced back to a 2-beam interference:

EP + FL is analogous to the above when considering the interference between

Ep and EFL used formula shown:

Ep + E _FL - A _{P + FL} r ΐfr + D f

Ep + FL -

(23)

as

Df ° = f _R i ^~ fr ⁰ . (25)

With phase shifting of the reference (test object fix, yr = const), Df and thus Df and Ap + _{FL are} constant for the phase-stable false light under consideration (<p _Fi = const).

For the total intensity l _ge s (<P) of the interferogram disturbed by stray light then results: IgesW = h + WLGV) + 2 h ¹ P + FL (AF ⁵ ) 'cos (0 + Df)

(26)

If f is varied by phase shifting the reference (test object fixed, y _r = const ₎ , Df ° and thus Df and A _{NL + FL are} constant for the phase-stable false light under consideration (yr _i = const). The overall intensity

() of the interferogram disturbed by stray light has the same shape as the intensity Ig _es (0) of the undisturbed interferogram:

Iges (O) differs from Ig _{es (} 0) only in contrast, offset and phase shift DF. The phase shift DF can be incorrectly interpreted as a surface feature, such as a surface depression Ah according to FIG. 1, of the test object, as already explained above with reference to FIG. If one sets lp = 1 and IR = 0.5 as an example, the simulation results already listed in Tab. 1 result for Aw _max . As already mentioned above, the almost maximum phase error Awmax results for Dfo = p / 2. This situation is shown in the diagram already explained above. If, for example, I FL = 1.4% is set for the intensity IFL of the false light, the result for the phase error in FIG. 4 is Aw = 10 nm. FIG. 5, analogously to FIG. 3, illustrates the field vectors Ep, EFL and EP + FL for the situation in which Df ° = p / 4. In this case, the intensity profile / es (-S ^' ) shown in FIG. 6 results at the point corresponding to the measuring point Mj as a function of the wavefront shift w. This intensity curve is compared to Ig _es

Shifted to the left analogously to FIG. 4, but by a smaller amount (Aw = 6.541 nm for IFL = 1.4%). Furthermore, / _gS (5F) is shifted slightly upwards and has an increased amplitude.

The diagram shown in FIG. 7 illustrates the phase error Aw resulting for IFL = 1.4% continuously as a function of Acpo for angles between 0 and 2TT. Stray light can thus 26 perform a function of the phase difference ACPO between the field vector EFL of false light and the field vector Ep of the test wave to a commonly indistinguishable from the surface deviation Ah phase error Aw with a value between -AW _ma x and + Aw _ax.

Embodiments according to the invention are described below with which the influence of false light on the measurement result of the surface deviation Ah can be suppressed or significantly attenuated. For this purpose, the test object 14 is arranged at different test positions 17 (N = 1 to Nmax) during the measurement. At each of the test positions 17, the above-described recording of the interferograms imaged at the different reference positions 31 on the detector 34 and determination of the resulting intensity profile Iges as a function of the wavefront shift w takes place.

FIG. 8 illustrates the intensity _curves / _gS (N = 1), Ig ^ _s N = 2) resulting for the point Mi for three test positions 17 (Nmax = 3), as well as

/ e _S (A / = 3), for the case in which the measured optical surface 14 the

Has nominal shape 12-2 (SF) and false light with the phase difference Dfo = tt / 4 to the field vector EP of the test shaft 26 is present. As illustrated in Fig. 5, has the field vector Ep at the test position N = 1 has a phase f _R = f _R. At the test position N = 2, the phase fr is shifted by 2p / 3 (f _R = f _r + -). The

Phase CPFL of the false light remains unchanged. In FIG. 5, this situation is shown for the sake of a simplified illustration using the analog situation in which the phase CPFL of the false light is shifted by 2p / 3 and the phase for the test wave 26 remains the same. As a result, the phase difference Df between the field vector EFL of the false light and the field vector Ep of the test wave 26 shifts by 2TT / 3. At the test position N = 3, the phase <pp is shifted by 4p / 3 (f _R = f _R + 2 · -), which is again shown in FIG. 5 using the analogous situation in which the phase CPFL of the false light is changed 4TT / 3 is shifted.

The intensity _profile Ig _gS illustrated in FIG. 8

which corresponds to the intensity profile Ig _PS (SF) shown in FIG. 6, is, as mentioned above, for IFL = 1.4% around Aw = 6541 pm compared to Ig _es

shifted to the left. For the intensity curve

takes place opposite

⁼ 1) also a shift to the left, namely by Aw = 2936 pm. For the intensity curve Ig _PS

however, takes place opposite

a shift to the right, namely by Aw = -9377 pm.

If the displacements Aw determined for N = 1 to 3 are now averaged, the result for the example illustrated in FIG. 8 is an averaged displacement of <Aw> = 33 pm.

As already mentioned, the displacements Aw illustrated in FIG. 8 result for Dfo = p / 4. 9 shows the displacements D w that result in an analogous manner for other values (0 to 2 p) of the phase difference Dfo between Ep and EFL for the individual test positions 17 with N = 1 to 3. Fig. 10 shows the derived therefrom Average shift <Dnn> as a function of the phase difference Dfo. It should be noted here that this oscillates back and forth between +/- <Aw> max. In the present case <Aw> max = 47 pm.

In the case where Nmax = 5, i.e. test object 14 on five different test

Positions 17 with yr = yr, yr + 3

2p. ₀ 2p

- as well as y _r + 4 -

5 5

is arranged, results in <Aw> max = 0.4 pm. In general it can be observed that <Aw> _ma x decreases continuously for increasing Nmax.

Fig. 11 shows this behavior for different levels of false light. Curve 1 is based on the false light level of 1.4% used as a basis in the preceding illustrations, for which the following applies when measuring in the basic test position (N = 1): <Aw> max = 10070pm. Curves 2 and 3 are based on lower false light levels, for which the following applies when measuring in the basic test position: <Aw> max =

1000 pm or <Aw> max = 10 pm.

According to a first embodiment of the invention, during the interferometric shape measurement of the optical surface 12 of the test object 14, phase error distributions Aw (x, y) resulting from 1 to Nmax depending on the location (x, y) of the surface 12 are determined for each of the test positions 17. In order to determine each of the individual phase error distributions A \ N (x, y), the reference element 30 can, as described above, be arranged one after the other at the different reference positions 31. This variant of the phase measurement is based on the configuration of the measuring device 10 shown in FIG. 1, which is designed as a so-called “common path interferometer”. Alternatively, the phase error distributions Aw (x, y) can also be measured without shifting the reference element 30 in an interferometer which differs from the measuring device 10 according to FIG. 1 only in that the Reference element 30 is tilted somewhat, so that the beam path of the returning reference wave 28 deviates somewhat from the beam path of the returning test shaft 26. After determining the phase error distributions Aw (x, y) representing a deviation result 40 for the different test positions 17, these are averaged; this is preferably a simple mean, i.e. all phase error distributions Aw (x, y) are included in the mean without weighting . Alternatively, the phase error distributions Aw (x, y) can also be weighted before averaging.

The averaged phase error distribution serves as a measurement result for the deviation of the actual shape 12-1 of the optical surface 12 from the nominal shape 12-2.

False light effects are suppressed in the averaged phase error distribution. As illustrated above, the degree of suppression of the false light effects is dependent on the number of test positions 17 in the measurement.

In order to mathematically illustrate this effect of strong suppression by averaging, the limiting case of very small false light levels (- «1) is considered below. Based on equation (25)

lv

with Df = f _R i - f _r as well as Ap = _^ JT _p and A _FL = jl one finds in the borderline case very small false light levels - «1, ie in the Limes lim -

-> 0, lp h

that The total wavefront error D w (pp) in the test item position f _r (sum of the wavefront error D w ° of the actual surface error (TF) h and l

the error - Df Dy) of the false light contribution),

(31) lFL

varies in this limit - «1 thus sinusoidally with the test item position y _R b

In the case of the previously chosen special discretization of the test item positions yr = (p _P with

then applies in this borderline case (as a special property of the sine function):

The phase errors induced by the false light are therefore completely and exactly averaged to zero. The mean wavefront error

describes in this limit case the exact wavefront error DW ° of the actual surface error (TF) h. This exact means of removing the false light interference ^! FL

to zero applies only in the Limes very low false light levels - «1

0. Jen¬

Ip

Aside from this limit, the phase error due to false light according to equation (29) no longer runs exactly sinusoidally with the position of the test object and the averaging is no longer exactly zero. But the course of the phase error is still approximately sinusoidal (given a sufficiently small false light). As explained above, the averaging still supplies very small values, the closer to zero the greater Nmax is.

The choice of the special equidistant discretization (expression 32) of the test item positions and averaging of the resulting phase error distribution is in principle the simplest method, but in general practice it is not necessarily the most favorable.

If one considers the phase positions f _{r of} the test object point by point, ie effectively for each location (x, y) of the surface 12, the special equidistant discretization f _r for a generally curved surface 12 at one location

2p

(x, y) no longer take place exactly with the globally defined phase distances,

Nmax but possibly only for a selected reference point. The setting accuracy of the test object (z-shift, tilt, ...)

2n

wisely or globally) the exactly defined phase distances - no more

Nmax

adhered to. The systematic suppression of the false light effects by simple averaging is thereby disturbed and possibly only reduced to a significantly less efficient statistical suppression. Therefore, in this context, instead of simple averaging, it is better to use an evaluation method in which the special discrete setting (32) of the test object position (global or for a reference point on the mirror surface) is retained (roughly), but deviations

gen can allow. Instead one fits e.g. the verb expected for - «1

run

the wavefront deviations (for the effective point-by-point test object positions known from the mirror geometry and from the phase reconstruction) to the wavefront deviations measured in different test object positions point-by-point at each location (x, y). The adaptation determines the three point by point

(Fit-) parameters ZI W °, and yri- With Zlw ° one obtains the (un-

disturbed) wavefront error of the actual surface error and man ge

winnt with - - and yri additional information about the disturbing false light.

b

Hi

Instead of the approximate course (for

1)

b

In equation (35), the exact expression in equation (29) can also be used for the greatest possible accuracy to determine the three parameters Zlw °, and yri

to determine.

According to a second embodiment according to the invention, the displacement of the reference element 30 and the test object 14 takes place simultaneously, in fact stepwise in opposite directions, for example by the same phase difference. Referring to the measuring device 10 of FIG. 1, this might mean that Nmax and Mmax have the same value, and wherein the first interferometer, the test object 14 grams recording at the position N = N _ma x and the reference element 30 at the position M = 1, or vice versa, are arranged. The test object 14 is then moved to the position Nmax-1 and the reference element 30 to the position 2 and so on. The distance between the positions is dimensioned so that the phase shift of the test wave 26 + F / 2 (ie fr = f ° r + F / 2) caused by the displacement of the test object 14 and the phase shift of the caused by the displacement of the reference element 30 Reference wave 28 is -F / 2 (ie q> R = cp ° R - F / 2). This results in the phase difference between test wave 26 and reference wave 28: fr - cpR = (cp ° p -cp ° R) + f = f ₀ + f, where Fo = (cp ° p -f ° r). The effect of the displacement of the test object 14 on the phase difference epp - q> R is + F / 2 with the effect of the displacement of the reference element 30 on the phase difference <pp - q> R (namely - (- F / 2) = + F / 2) identical. In particular, the two effects should have at least the same sign.

If the same value is to be set for F in the successive measurements as in the above-described operation of the measuring device 10, in which the reference element 30 is shifted step by step for a constant position of the test piece 14, then in the present embodiment the reference element 30 must be in each step can only be shifted by half the shift amount of the above-described operation. As already explained above, the reference element only contributes F / 2 to the phase shift, while the remaining phase shift by F / 2 is brought about by shifting the test object 14.

With reference to the intensity l _ges listed in expression (20) in conjunction with expressions (16), (17) and (19) of the interferogram then resulting from 3-beam interference, half of this is on the reference element 30 and the test object 14 split phase shifts of F described by: The intensity Ig _es (0) of the undisturbed interferogram according to (18), which is present without the presence of false light, is now as follows:

With (36) one finds for l _ge s according to expression (20):

where Df = arctan2 x, y) (4-quadrant-inverse tangent)

With

Iges () can be written as follows:

as

according to (40) corresponds to the sum of the intensity Ig _es (0) des

undisturbed interferogram and I FL. IFL represents a small modification of the offset through a constant stray light intensity and can therefore be neglected. Ig _es f (0) thus varies with the shift frequency 1 (ie with F).

The first factor

of / y (0) is constant with variation of F, while the second factor

ges, ₂ varies with half the shift frequency V2 (ie with F / 2).

Due to the different frequency dependencies (F or F / 2), the component I f (0) abse- w ges, ₂ can be _pared by means of Fourier decomposition of I _qes (0) according to (38), which means the remaining intensity Ig _es (0) of the undisturbed phase shift interferogram can be determined. 12 shows the result of this procedure using the intensity profile Iges (SF), which is at the point corresponding to the measuring point Mj as a function of the wavefront shift w for the case in which there is false light with Df ° = p / 4 and the optical Surface 12 has the desired shape 12-2, results. Furthermore, FIG. 12 shows the portion I g _es F which can be separated off by means of Fourier decomposition of I _ge s (SF) and the remaining portion I g _es y which, if the constant offset I FL of the intensity is neglected

(0) ^ens P ^richt · The phase error Aw generated by the false light can now be read off as an offset in the x coordinate direction between l _ge s (SF) and Ig _es y.

For other phase deviations Acp ° of the false light, if the component I g is separated off accordingly, _it also results in F as the remaining component

Ig _es y the intensity distribution / ^ _es (0) that would result in the absence of stray light.

The division of the phase shifting made above simultaneously on

0 0 0

DUT and reference f _R = f + _r - p and _R p = _R -, with 1: ^_1/2

is exemplary. Other divisions are also possible for which the following generally applies:

and p _R = (p _R - (1-a) F, where a 6 [0, 1].

When carrying out the above-described measuring method according to the second embodiment according to the invention, the intensity profile 42 determined when measuring the optical surface 12 with an actual shape 12-1 deviating by Ah (x, y) from the nominal shape 12-2

in the case of false light with unknown phase offsets Df °, the portion 44 (I y) attributable to scattered radiation is separated off by means of Fourier decomposition.

9 ^es ' ^~ 2

By comparing the remaining portion l _total y with the intensity distribution determined by simulation or calibration measurement for the target shape 12-2

Then the distribution of the deviation Ah (x, y) is determined. The above description of exemplary embodiments, embodiments or design variants is to be understood as exemplary. The disclosure thus made enables the person skilled in the art, on the one hand, to understand the present invention and the advantages associated therewith, and, on the other hand, also includes obvious changes and modifications of the structures and methods described in the understanding of the person skilled in the art. Therefore, it is intended to cover all such changes and modifications insofar as they come within the scope of the invention as defined in the appended claims, and equivalents of the scope of protection of the claims.

List of reference symbols

10 measuring device

12 optical surface

12-1 actual shape

12-2 Target form

14 T est object

15 holding device

16 Measurement radiation source

17 test positions

18 exit area

19 Measurement radiation

20 input shaft

22 beam splitters

24 diffractive optical module

26 test shaft

26i single beam of the test wave

28 reference wave

28i single beam of the reference wave

29 measuring point

30 reference element

31 reference positions

32 acquisition device

34 detector

36 focusing lens

38 Evaluation device

40 Deviation result Aw

42 determined intensity curve Ig _es (0)

44 Part due to scattered radiation /

Claims

Expectations

1. Measuring device for the interferometric determination of a shape of an optical surface of a test object with:

- A diffractive optical module for generating a test wave directed onto the optical surface, which has at least a wave front that is partially adapted to a nominal shape of the optical surface, and a reference wave,

- A reflective reference element arranged in the beam path of the reference wave, which is mounted displaceably in the direction of the beam path of the reference wave for arrangement at different reference positions,

- a holding device mounted displaceably in the direction of the beam path of the test shaft for arranging the test object at different test positions,

A detection device for detecting interferograms generated at the different test positions of the test object, which are each formed by superimposing the test wave after interaction with the optical surface and the reference wave after reflection on the reference element, with at least one of the test positions each being assigned several interferograms , during the generation of which the reference element is arranged at different reference positions, and

- An evaluation device, which is configured to determine a deviation of an actual shape of the optical surface from the nominal shape by determining a deviation result for each of the test positions by evaluating the interferograms generated at the different test positions and averaging the deviation results, with at least one Test position with several assigned interferograms, the determination of the deviation result is carried out using the assigned interferograms.

2. Measuring device according to claim 1, wherein the evaluation device is configured to evaluate the interferograms in such a way that errors in the evaluation result generated by scattered radiation are minimized.

3. Measuring device according to claim 1 or 2,

wherein adjacent test positions are at most a fraction of a wavelength of the test wave spaced apart.

4. Measuring device according to one of the preceding claims,

wherein the reference element is mounted displaceably for arrangement at different reference positions.

5. Measuring device according to one of claims 1 to 4,

wherein the reference element is mounted displaceably in the direction of the beam path of the reference shaft for arrangement at different reference positions and the measuring device is configured to arrange the test object in succession at the different test positions and at the same time to change the reference position of the reference element.

6. Measuring device according to claim 5,

wherein the measuring device is configured to move the reference element at the same time when the test object is moved between test positions such that the effect of the displacement of the test object and the effect of the displacement of the reference element on the phase difference between the test wave and the reference wave in the resulting interferogram have the same sign .

7. Measuring device according to claim 5 or 6,

wherein the evaluation device is configured to use the interferograms as a function of at least one measuring point on the optical surface to use an intensity profile that varies from interferogram to interferogram to determine a phase difference between the test wave and the reference wave on which the respective interferogram is based, and to calculate a portion attributable to scattered radiation from the determined intensity profile.

8. Measuring device according to claim 7,

wherein the evaluation device is configured to perform a Fourier decomposition of the intensity profile determined when calculating the portion attributable to scattered radiation.

9. Measuring device according to one of the preceding claims,

which is configured to determine the shape of a surface of an EUV mirror for microlithography.

10. A method for the interferometric determination of a shape of an optical surface of a test object, with the steps:

- Generating a test wave directed onto the optical surface, which has at least a wavefront which is partially adapted to a nominal shape of the optical surface, and a reference wave directed onto a reflective reference element by means of a diffractive optical module,

- Arranging the reference element at different reference positions in the direction of the beam path of the reference wave,

- Arranging the test object in the direction of the beam path of the test wave one after the other at different test positions,

Acquisition of interferograms generated at the different test positions of the test object, which are each formed by superimposing the test wave after interaction with the optical surface and the reference wave after reflection on the reference element, with at least one of the test positions being assigned several interferograms when they are generated the reference element is arranged at different reference positions, and - Determination of a deviation of an actual shape of the optical surface from the nominal shape by respective determination of a deviation result for each of the test positions by evaluating the interferograms generated at the different test positions and averaging the deviation results, with several assigned interferograms for the at least one test position the determination of the deviation result takes place on the basis of the assigned interferograms.

11. The method according to claim 10,

in which the test object is arranged one after the other at the different test positions and at the same time a reference position of the reference element is changed in the direction of the beam path of the reference wave.

12. The method according to claim 11,

in which from the interferograms with respect to at least one measuring point on the optical surface, an intensity profile that varies from interferogram to interferogram is determined depending on a phase difference between test wave and reference wave on which the respective interferogram is based, and a proportion attributable to scattered radiation is calculated from the intensity profile determined.

13. The method according to claim 12,

in which when calculating the portion attributable to scattered radiation, a Fourier decomposition of the intensity profile determined is carried out.