The invention relates to a method and a device for determining the one-dimensional or multidimensional structure of objects by means of high-energy radiation of short wavelength, in particular EUV and XUV or soft X-radiation. To ISO 21348 EUV is the spectral range between 121 nm and 10 nm and XUV the spectral range of 10 nm to 0.1 nm. The invention covers both ranges, so in the following the term "XUV" is used for the entire spectral range of the invention. The proposed device constitutes an apparatus for performing optical coherence tomography (OCT) for the aforementioned radiation wavelengths and is called XCT in the following.
Near-infrared (NIR) OCT is an established method for the non-destructive three-dimensional imaging of mostly biological or medically relevant samples. The resolution is in the range of a few micrometers with a maximum penetration depth of a few millimeters. The imaging in the axial direction (propagation direction of the light source used) is achieved independently of the numerical aperture of the imaging optics by taking advantage of the short coherence length of a broadband radiation source. Technically, OCT devices are realized by an interferometric design. The imaging in the lateral dimensions is done with the help of a microscope objective. A three-dimensional image is created by scanning the sample to be examined ( W. Drexler and JG Fujimoto: Optical Coherence Tomography, 2008, Springer Verlag ).
The basic idea of coherence tomography with short wavelengths is already in the patent US 7,656,538 B2
known. Due to the significantly shorter wavelength of the XCT (XUV radiation) used compared to infrared radiation, an axial resolution of a few nanometers independent of the focusing can be achieved. However, the implementation of OCT in this spectral range with an interferometer extremely challenging and technically unsatisfactory solved so far.
A first experimental setup according to US 7,656,538 B2
was tested in 2011. Typically, OCT devices in the infrared spectral range use a Michelson interferometer to measure the sample in the axial direction. However, the design of a broadband beam splitter in the XUV range with sufficient surface accuracy with the current state of the art - if at all - only with great effort and especially with losses in bandwidth or resolution feasible. The final implementation was therefore based on a so-called "common-path OCT" scheme ( AB Vakhtin, DJ Kane, WR Wood and KA Peterson: Common-path interferometer for frequency-domain optical coherence tomography, Appl. Opt. 42, 2003, 6953-6958
), in which can be dispensed with a beam splitter.
A theoretical description of how this method works and some simulations have already been published ( S. Fuchs, A. Blinne, C. Rödel, U. Zastrau, V. Hilbert, M. Wünsche, J. Bierbach, E. Frumker, E. Förster, GG Paulus: Optical coherence tomography using broad-bandwidth XUV and soft X -ray radiation, Appl. Phys. B 106, 2012, 789-795 ); see. also description too 1 , Due to the large bandwidth and the consequent low coherence length of the radiation used, the interference leads to modulations in the spectrum of the reflected light. These modulations carry the information about the deep structure of the sample. Since only depth differences can be measured with the aid of the modulation frequencies, an excellent reference depth or position is necessary for an unambiguous reconstruction of the deep structure of the sample. The deep structure can only be uniquely determined with those interferences in which the surface reflection participates. In order to distinguish these distances or the modulation frequencies in the spectrum from the distances within the sample, a highly reflective thin surface coating had to be applied to each sample. The amplitudes of the interference with the cover layer are thus generally greater than those within the sample.
Despite the highly reflective surface coating, it is not possible to clearly reconstruct the depth structure due to the interference between the layers. Accordingly, it comes to the detection of dummy structures that can not be distinguished from the real structures. The surface coating weakens this effect, but it can not completely prevent it, so that a clear structural analysis is not possible.
In addition, each sample to be examined must previously be prepared very costly with the said cost-intensive highly reflective surface coating (eg gold layer). As a result of this covering layer applied to the sample surface, the depth structure of the sample can only be determined relative to its surface. Absolute measurements of the structure of objects are therefore not possible.
The resolution in the lateral direction of the known XCT method is limited by the size of the focal point of the XUV radiation. Due to the short wavelength, the high absorption and weak dispersion of most materials in the XUV, the focusing of the radiation is technically very complicated because z. B. refractive optical elements can not be used. Instead, complex manufacturable diffractive or reflective focusing optics are used. In the said XCT method, therefore, especially toroidally shaped mirror surfaces are used for focusing. With a synchrotron undulator source, lateral structures up to about 100 microns could be resolved. This is up to four orders of magnitude above the structure size of the method which can be resolved in the axial direction, so that an improvement of the lateral resolution of XCT is absolutely necessary for a three-dimensional image with comparable resolution in all dimensions.
The samples to be examined with said device are to be introduced into a vacuum chamber. For this, the samples must be suitable for vacuum (vacuum-dried, no water, no gas inclusions) or prepared accordingly. The samples can be affected or even destroyed by the vacuum. In addition, the exchange of samples in a vacuum is complex.
In summary, it should be noted that this method is particularly by the above-mentioned highly reflective surface coating, which is applied to each sample to be examined extra and possibly not fully feasible in any application, and by the necessary investigation in a vacuum with the required vacuum-compatible preparation of the sample, which is also not or not sufficiently for all objects of investigation is possible to implement only limited practical impact.
The object of the invention is to make possible, as far as possible, a universally applicable unambiguous measurement of the object structure with high efficiency of three-dimensional imaging, in which process-related ambiguities (dummy structures) can be detected in the measurement signal and completely eliminated in the signal evaluation.
According to the invention, the object to be examined is exposed to radiation of short wavelength, in particular XUV or soft X-ray radiation, and the radiation reflected or backscattered by the object is detected by a detector and spectrally evaluated, wherein at a fixed or relatively variable distance from the surface of the object the radiation partially permeable element, in particular a film is arranged. With such a partially transparent element (foil), the XUV or soft X-ray radiation is split into a measuring beam which transmits this partially transmissive element (film) and penetrates the object, and into a reference beam reflected by the partially transmitting element (foil) is superimposed on the measured beam reflected or backscattered by the object for the purpose of common spectral evaluation. In this way, it is possible to unambiguously identify signal components in the evaluation signal, which indicate apparent structures, so that these signal components can be excluded from the structure determination.
It is advantageous if the partially transparent foil, for example a gold foil, is accommodated in a stationary or positionally displaceable foil holder and can be moved laterally or in the plane of this foil relative to the surface of the object to be examined.
In the dependent claims further embodiments of the invention are set forth.
In contrast to the aforementioned XUV device according to US 7,656,538 B2
is not applied to each object to be examined with elaborate preparation and sample preparation a highly reflective surface coating, but separated from the object to be examined, the proposed partially transparent film in the beam path of the XUV or soft X-radiation is arranged. With this measure, not only false evaluations of dummy structures can be reliably avoided, but it also eliminates the adverse high cost of surface coating the objects to be measured in comparison to the known prior art.
The partially transmissive element (foil) is such that it both sufficiently reflects and transmits the incident XUV radiation. In the case of a film, this can be achieved both by the film material itself and the film thickness as well as by the structuring of the film (for example perforation). The measurement of the reflected spectrum of the arrangement of film and object to be measured (sample) is analogous to the known XCT method. By positioning the partially transparent film to the object surface at a distance from the sample, both the depth structure and the surface condition can be reconstructed from the measured signal without ambiguous dummy structures. For this purpose, the partially transparent element (foil) is positioned at a distance greater than the maximum measurable depth (penetration depth of the radiation) of the sample. The radiation reflex of the sample surface is then visible in the reconstructed deep structure (cf. 5 ). All possible dummy structures now appear in the representation of the evaluation signal only in front of the actual sample surface and thus can be unambiguous be identified. Alternatively, two measurements can be taken with a slightly different distance between the sample and the partially transmissive element (foil) (the condition that the distance is greater than the maximum measurable depth does not necessarily have to be complied with here). Only the reconstructed depth of real structures will change depending on the distance. However, the dummy structure signal components always appear in the same place, whereby they can also be clearly identified.
If the lateral extent of the transmitting film surface is smaller than the XUV focus used, then the lateral resolution of the method can be significantly increased thereby, without changing the focus (see 4 ). Thus, lateral structure sizes smaller than the XUV focal length can be resolved.
The partially transmissive element (foil) according to the invention can advantageously be formed as a radiation passage window of a vacuum system for the XUV or soft X-ray radiation, so that although the examination radiation (as before) is to be introduced into a vacuum chamber, but not the object to be examined, which has considerable advantages for the Sample preparation and examination per se, including sample changes, brings (cf. 3 ). Thus it is no longer necessary for the samples to be vacuum-dried and vacuum-compatible, which considerably extends the application possibilities of the measuring method. For scanning the sample surface, the sample can be moved behind the radiation passage window.
The invention will be explained in more detail below with reference to embodiments shown in the drawing for determining the structure by means of XUV radiation.
1 : Schematic representation of a known XCT device in which an object with a deposited on its surface highly reflective reflection layer in a vacuum by XUV radiation is examined
2 : Schematic representation of the XCT device according to the invention, in which before the object to be examined and separated from this a highly reflective partially transparent film is arranged and movable relative to this
3 : Special embodiment of the device according to 2 in which the highly reflective partially transparent film is arranged as a radiation passage window of a vacuum chamber for the XUV radiation
4 : Schematic partial representation of the XCT device with the object and the highly reflective partially transparent film to improve the lateral resolution of the process
5 : Simulated spectrum of the known manner from a surface-coated object reflected or backscattered detected XUV radiation (left figure) and a reconstructed and actual deep structure (right figure)
Simulated spectrum of the detected XUV radiation reflected or backscattered by a structurally to be examined object with partially transparent film according to the invention upstream (left figure) and a deep structure completely reconstructed therefrom (right figure) 1
is as a comparison to the invention one of the principle of US 7,656,538 B2
known XCT device shown schematically, in which a in a vacuum chamber 1
located and evaluated in its depth structure object 2
with a very short-wave, broadband radiation 3
(XUV or soft X-rays) is irradiated. The object 2
is hung in such a way (for reasons of clarity not explicitly shown, but only by arrow 4
symbolizes) that it is in all dimensions relative to the radiation 3
is movable. The radiation 3
penetrates the object 2
one and is at different depths, which are the optical properties (refractive index) of the object 2
change, reflected or backscattered (see enlarged partial representation on the right in the picture of 1
with reflected partial radiations 3a
), so that all reflected partial beams 3a
interfere with each other.
Due to the large bandwidth and the resulting short coherence length of the radiation used, the interference leads to modulations in the spectrum of the reflected or backscattered light (reflected partial beams 3a . 3b . 3c . 3d ) of the radiation 3 , These modulations carry the information about the deep structure of the object 2 , A modulation frequency corresponds to a distance of the origin depths of two reflected or backscattered interfering partial beams 3a . 3b . 3c . 3d , This is the higher, the farther the source depths are apart. To evaluate these modulations, one of the object 2 reflected or backscattered total radiation 5 (Totality of superimposed partial beams 3a . 3b . 3c . 3d ) with a spectral resolution intensity detector 6 , For example, a spectrometer detected.
Since only depth differences can be measured with the help of said modulation frequencies, is for a clear reconstruction of the deep structure of the object 2 a excellent reference depth or position required. The depth structure can only be determined with those interferences at which the surface reflection (partial beam 3a ) is involved. These distances or the modulation frequencies in the detected spectrum with respect to the distances within the object 2 to mark is directly on the surface of the object 2 in addition a thin and for the radiation 3 semi-permeable, highly reflective surface coating 7 applied. Amplitudes of interferences with surface coating 7 are generally larger than those inside the object 2 starting from the partial beams 3b . 3c . 3d ,
The interferences of partial beams 3b . 3c . 3d from the object 2 However, they can not be completely suppressed, so that a clear reconstruction is not possible (see also 5 ). Through this highly reflective surface coating 7 , for example, made of gold, which mandatory on each object to be examined 2 and the above-described ambiguous reconstruction of the structure of the object 2 From the spectrum detected arise the aforementioned disadvantages, which are eliminated by the invention.
In 2 the device according to the invention is shown schematically. An object to be evaluated in its depth structure 8th is again in the vacuum chamber 1 and becomes with the very short-wave, broadband radiation 3 (XUV or soft X-rays) irradiated. Unlike the object 2 in 1 owns the object 8th no surface coating 7 but at a distance to the surface not specially coated for the structure determination 9 of the object 8th is (in the direction of the radiation 3 in front of the object 8th ) According to the invention a highly reflective and for the radiation 3 semi-permeable film 10 , For example, a gold foil arranged. The radiation 3 first meets the partially transparent film 10 which acts like a beam splitter, so that the radiation 3 in a partially permeable film 10 transmitting measuring beam 11 for penetrating the deep structure of the object to be evaluated 8th and in one of the partially transparent film 10 reflected reference beam 12 is split.
The partially transparent film 10 is mounted in or on a (for reasons of clarity not explicitly shown) film holder, which is designed such that the partially transparent film 10 in all dimensions relative to the radiation 3 and to the object 8th can move (symbolized by arrow display 13 ). In addition, the object 8th in all dimensions relative to the radiation 3 (Measurement beam 11 ) movable (also not shown for reasons of clarity).
The measuring beam 11 the radiation 3 penetrates through the partially transparent film 10 in the object 8th one and is at different depths, which are the optical properties (refractive index) of the object 8th change, reflected or backscattered (see enlarged partial representation on the right in the picture of 2 with reflected or backscattered partial beams 11a . 11b . 11c . 11d ), so that these partial beams with that of the partially transparent film 10 reflected reference beam 12 due to the overlay interfere for common evaluation.
One through this beam fusion of the object 8th reflected or backscattered partial beams 11a . 11b . 11c , and 11d and of the partially transparent film 10 reflected reference beam 12 resulting overlay radiation 14 will (as in 1 ) with the spectral resolution intensity detector 6 , For example, also a spectrometer detected. The detected spectrum is modulated. These modulations carry the information about the deep structure of the object 8th , Is the semi-permeable film 10 at a greater distance from the surface 9 of the object 8th removed as the through the measuring beam 11 to penetrate depth of the last structure to be evaluated in the object 8th , the entire depth structure can be unambiguously reconstructed from the modulated spectrum in the detected evaluation signal, since all signal components directly from the sub-beams 11b . 11c . 11d arise, temporally in the spectrum according to the signal component of the surface 9 of the object 8th reflected or backscattered partial beam 11a appear. All signal components caused by interference from the sub-beams 11a . 11b . 11c . 11d arise appear earlier in the detected evaluation signal and are recognized by the invention clearly as so-called. Pseudo structures and can of the determination of the depth structure of the object 8th be excluded (see also 6 ).
Alternatively or in addition to the said determination of the distance of the partially transparent film 10 to the surface 9 of the object 8th can also each a second measurement (determination of the depth structure of the object 8th ) at the same lateral position thereof with changed distance between the partially transparent film 10 as well as the surface 9 be performed. In this case, only the signal components of the actual depth structures and not the signal components, which due to the interference of the superposition radiation, change in the spectrum of the detected evaluation signal 14 indicate dummy structures (apparent depths) so that they can be clearly identified.
A preparation of the object to be determined in the deep structure, as in US 7,656,538 B2
known XCT device (see Object 2
), with the highly reflective Surface coating is not required in the present invention by the unique identification of pseudo-structures (depths of vision) as described.
3 shows a further embodiment of an XCT device according to the invention, in which the partially transparent film 10 simultaneously as a radiation passage window 15 a vacuum chamber 16 is trained. In the vacuum chamber 16 (in 3 characterized by the term 'vacuum') are now only the beam paths of the radiation 3 as well as the merging of the reflected or backscattered partial beams 11a . 11b . 11c . 11d and the reference beam 12 (see. 2 ) resulting overlay radiation 14 , The object to be evaluated in terms of its structure depth 8th is outside the vacuum chamber 16 in the area labeled 'Atmosphere'.
The partially transparent film 10 is as a radiation passage window 15 in this device immovable part of the vacuum chamber 16 while the object to be examined 8th at a short distance to the radiation passage window 15 and relative to the partially transparent film 10 , For example, on or on a three-dimensionally movable coordinate table, not shown, is preferably arranged movable, symbolized by arrow 17 , The radiation 3 to the object 8th as well as the reflected or backscattered by this partial beams each penetrate the radiation passage window 15 with the too 2 described operation. Such a device has the advantage that the object to be examined is not exposed to any vacuum caused by the XUV radiation and thus does not necessarily have to be suitable for vacuum.
4 shows a detailed view of an XCT device according to the invention with the in a film holder 18 received partially permeable film 10 and a focus 19 the incident radiation 3 , Here is the extent of the partially transparent film 10 to form a diaphragm or a pinhole smaller than the focus 19 on the semi-permeable film 10 meeting radiation 3 to increase the lateral resolution. The mode of operation of the evaluation of the detected evaluation signal from the superposition radiation 14 (see. 2 . 3 ) in itself remains unaffected.
shows as a comparison to the invention a simulated spectrum (left figure) of a known manner of one with a surface coating 7
provided sample (see object 2
) reflected or backscattered detected XUV radiation (dependence of the reflectivity of the sample R sample
of the detected photon energy). In the right picture of 5
are shown from this spectrum by a Fourier transform (FT) reconstructed and actual depth structures. The sample consists of two 5 nm thick gold layers ( 20
) through a 12 nm thick silicon layer ( 22
) and under a 120 nm thick silicon layer ( 23
) are buried. Additionally is on the object 2
a 5 nm thick gold layer ( 24
) as a surface coating 7
(according to US 7,656,538 B2
) applied. The original structure of the sample (Object 2
) is shown in the right figure in the background. The deep structure reconstructed from the spectrum is also a graph 25
shown in the right part. The structures are resolved, however, a signal peak appears in the evaluation curve at a depth of 17 nm 26
which, when compared to the actual depth structure, indicates a pseudo-structure which is the difference of the depths of both gold layers ( 20
) corresponds. Without previous or additional knowledge about the sample structure, the reconstruction of the deep structure is therefore not clear.
6 shows, however, in an adequate representation 5 a simulated spectrum (left picture) of a structurally to be examined sample (see Object 8th in 2 ) with the partially transparent film according to the invention upstream 10 reflected or backscattered detected XUV radiation and a completely reconstructed deep structure (right figure).
As before, the sample consists of the two 5 nm thick gold layers 20 . 21 passing through the 12 nm thick silicon layer 22 separated and under said 120 nm thick silicon layer 23 are buried. In addition, at a distance of 200 nm in front of the sample (in the direction of the radiation 3 in front of the surface 9 of the object 8th , see. 2 ) according to the invention the partially transparent film 10 as a 5 nm thin gold foil 27 arranged. The original structure of the sample is comparable to 5 ) in the right figure in the background (hatched area). The deep structure reconstructed from the spectrum by Fourier transformation FT is a graph 28 in the right figure of 6 shown. The structures are resolved and it appears in the graph 28 at 17 nm a signal spec 29 , which in turn indicates a said dummy structure. This signal spec 29 However, it appears in one area of the graph 28 , which lies before the (for structure evaluation eligible) curve area after the sample surface, which represents signal components of the actual sample structure, and is thus clearly identifiable as a dummy structure.
LIST OF REFERENCE NUMBERS
- 1, 16
- vacuum chamber
- 2, 8
- 3a, 3b, 3c, 3d
- reflected or backscattered partial beams of the radiation 3
- 4, 13, 17
- Arrow depiction for movable suspension
- 5, 14
- Overlay radiation
- intensity detector
- surface coating
- Surface of the object 8th
- semi-permeable film
- Measuring beam
- 11a, 11b, 11c, 11d
- reflected or backscattered partial beams of the radiation 3
- Reference beam
- Radiation passage window
- foil holder
- Focus of radiation 3
- 20, 21, 24
- gold layer
- 22, 23
- silicon layer
- 25, 28
- 26, 29
- Gold layer as a partially transparent film 10
- Fourier transformation (calculation of the depth structure)
- R sample
- Reflectivity of the sample (object 2 respectively. 8th )
QUOTES INCLUDE IN THE DESCRIPTION
This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.
Cited patent literature
- US Pat. No. 7656538 B2 [0003, 0004, 0015, 0026, 0035, 0039]
Cited non-patent literature
- ISO 21348 
- W. Drexler and JG Fujimoto: Optical Coherence Tomography, 2008, Springer Verlag 
- AB Vakhtin, DJ Kane, WR Wood and KA Peterson: Common-path interferometer for frequency-domain optical coherence tomography, Appl. Opt. 42, 2003, 6953-6958 
- S. Fuchs, A. Blinne, C. Rödel, U. Zastrau, V. Hilbert, M. Wünsche, J. Bierbach, E. Frumker, E. Förster, GG Paulus: Optical coherence tomography using broad-bandwidth XUV and soft X -ray radiation, Appl. Phys. B 106, 2012, 789-795