Patent Application
20140038315
This application claims priority to and the benefit of European patent application No. 12179065.3, filed Aug. 2, 2012, and European patent application No. 12199128.1, filed Dec. 21, 2012, the contents of which are herein incorporated by reference in their entirety.
The present invention relates to an apparatus and a method for measuring the dimensions of 1-dimensional and 0-dimensional nanostructures in real-time during epitaxial growth.
Optical real-time monitoring and related real-time analysis of thin-film structures during epitaxial growth are state-of-the-art techniques. Epitaxy methods such as MOCVD (metal-organic vapour-phase epitaxy) and MBE (molecular beam epitaxy) are used both in industry and in academic research. Usually there are appropriate optical view-ports available in the MOCVD or MBE growth chambers and optical real-time monitoring of the epitaxial growth process can be performed by reflection, transmission or ellipsometry (more general: optical in-situ measurements). Even if the thin films grown have a total thickness of only a few nanometers (so called ‘quantum wells’ or 2D nano-structures), today's optical in-situ metrology tools are sufficiently sensitive to follow in real-time the formation of such 2D nano-structures. The related real-time analysis of in-situ measurements on growing 2D nano-structures is also straight-forward, because the mathematical equations describing the optical properties of thin-film stacks are valid also for extremely thin layers (total thicknesses of only a few nanometers). For real-time growth analysis these thin-film optical equations (e.g., for normal incidence reflection) are numerically calculated and by comparison between the measured and the calculated reflection the current status of film thickness and/or film composition can be determined.
The conventional optical thin-film analysis described above works well for 2D nano-structures—but usually completely fails for ‘nano-wires’ (1D nano-structures, see
It is therefore an object of the present invention to provide an apparatus and a method for measuring the dimensions of 1-dimensional and 0-dimensional nanostructures in real-time during epitaxial growth, which overcomes the deficiencies of the prior art.
According to an aspect of the present invention, a method for measuring the dimensions (of nanostructures) of a sample is disclosed.
According to a preferred embodiment of the invention, characteristic dimensions of the sample (nanostructures) can be measured in real-time.
According to another preferred embodiment of the invention, dimensions of the sample (nanostructures on a substrate) can be measured during epitaxial growth.
According to an aspect of the present invention, a method for determining the structural data of a sample which comprises nano-structures on a substrate is disclosed, the method comprising: irradiating optical radiation onto the sample, wherein an area of the sample onto which the optical radiation is irradiated ranges between 0.0004 mm2 to 4 mm2, determining an optical reflectance signal of the sample from the reflected optical radiation, and determining structural data of the area of the sample onto which the optical radiation is irradiated, by either comparing the determined reflectance signal with reference signals of a data base and assigning three-dimensional structural data from the data base to the sample or by calculating structural data from the determined optical reflectance signal.
Even though it is possible to use monochromatic optical radiation, it is preferred that polychromatic optical radiation is used. Accordingly, a method for determining the structural data of a sample which comprises nano-structures on a substrate is disclosed, the method comprising: irradiating polychromatic optical radiation onto the sample, wherein an area of the sample onto which the polychromatic optical radiation is irradiated ranges between 0.0004 mm2 to 4 mm2, determining an optical reflectance spectrum of the sample from the reflected polychromatic optical radiation, and determining structural data of the area of the sample onto which the polychromatic optical radiation is irradiated, by either comparing the determined reflectance spectrum with reference spectra of a data base and assigning structural data from the data base to the sample or by calculating structural data from the determined optical reflectance signal spectrum.
Preferably, the assignment of the structural data from the data base to the sample is made based on the comparison result.
Preferably, the data base contains a plurality of pairs of pre-defined three-dimensional structures and pre-calculated optical reflectance spectra.
In case that the characteristic dimensions of the sample (three-dimensional structural data, nanostructures) are determined by assigning such data from a data base to the sample, the characteristic dimensions of the sample can advantageously be determined (measured) in real-time.
The polychromatic optical radiation which is irradiated onto the sample preferably comprises substantially parallel propagating rays (radiation), more preferably consists of substantially parallel propagating rays, still more preferably consists of parallel propagating rays. Preferably, substantially parallel radiation is radiation having a divergence angle of less than 10°, more preferably less than 5°, more preferably less than 2° and still more preferably less than 0.5°.
Preferably, collimated polychromatic optical radiation is used. Preferably, the polychromatic optical radiation is irradiated perpendicularly onto the sample. Preferably a cross-section of the irradiated polychromatic optical radiation comprises a circular shape or rectangular shape.
Preferably, a ratio of the minimum energy density and the maximum energy density of the polychromatic optical radiation ranges between 0.1 and 1, more preferably between 0.3 and 1, more preferably between 0.5 and 1, more preferably between 0.7 and 1, more preferably between 0.9 and 1. Still more preferably, a ratio of the minimum energy density and the maximum energy density of the polychromatic optical radiation is 1, i.e. the energy density is uniform over the whole cross-section of the irradiated polychromatic optical radiation.
Preferably, the polychromatic optical radiation being irradiated onto the sample comprises a wavelength spectrum from the range between 200 nm and 2000 nm, preferably from the range between 400 nm and 900 nm. Preferably, the wavelength spectrum is continuous (with respect to intensity) in the mentioned ranges.
Preferably, a ratio of an intensity of a wavelength from said wavelength spectrum which has the lowest intensity and an intensity of a wavelength from said wavelength spectrum which has the highest intensity ranges between 0.1 and 1, more preferably between 0.3 and 1, more preferably between 0.5 and 1, more preferably between 0.7 and 1 and still more preferably between 0.9 and 1.
Preferably, an area of the sample onto which the polychromatic optical radiation is irradiated ranges between 0.0025 mm2 to 1 mm2, preferably between 0.0025 mm2 to 0.25 mm2 and more preferably between 0.005 mm2 to 0.025 mm2.
Preferably, the data base contains a plurality of pre-defined three-dimensional structures, each pre-defined structure being assigned with a pre-calculated optical reflectance spectrum, wherein each pre-calculated optical reflectance spectrum is calculated by including interference effects due to superposition of neighbouring wave-fronts reflected from the nano-structure and wave-fronts reflected from the substrate between the nano-structures.
Preferably, the measured optical reflectance spectrum of the sample is compared with a plurality of pre-calculated optical reflectance spectra from the data base, and the structural data of the best-fitting pre-calculated optical reflectance spectrum is assigned to the sample.
Preferably, the step of calculating the three-dimensional structural data of the sample includes a scattering matrix method.
Preferably, at least one from the group selected from pre-known parameters of the sample, optical properties of the material of the nano-structures, optical properties of the material of the substrate, optical properties of the material of an ambient medium and the irradiated polychromatic optical radiation are used as input parameters for pre-calculating the structural data of the sample.
Preferably, the sample is a structure of nano-wires arranged on a semiconductor substrate. Preferably, the sample comprises a (substantially) regular pattern of nano-wires. Preferably, a ratio of a pitch L between adjacent nano-wires and a median wavelength λM of the irradiated radiation ranges between 0.01 and 100, more preferably between 0.1 and 10, more preferably between 0.3 and 3, more preferably between 0.5 and 2, more preferably between 0.7 and 1.4, more preferably between 0.8 and 1.25 and still more preferably between 0.9 and 1.1. Preferably, a ratio of a (maximum) diameter D of the cross-section of the nano-wires and a median wavelength λM of the irradiated radiation ranges between 0.01 and 100, more preferably between 0.1 and 10, more preferably between 0.3 and 3, more preferably between 0.5 and 2, more preferably between 0.7 and 1.4, more preferably between 0.8 and 1.25 and still more preferably between 0.9 and 1.1. The pitch of the nano-wires means a distance in which the regularly arranged nano-wires are repeated. A median wavelength λM is the wavelength which divides the total irradiated energy into two parts (over the time of irradiation) of wavelengths lower than median wavelength λM and wavelengths greater than median wavelength λM, the two parts having the same total amount of energy.
The time period in which the polychromatic optical radiation is irradiated onto the sample (for each irradiation spot=an area of the sample) preferably ranges between 0.1 s and 10 s, more preferably between 0.01 s and 1 s and still more preferably between 0.0001 s and 0.01 s.
Preferably, the polychromatic optical radiation is irradiated onto a plurality of different areas of the sample, an optical reflectance spectrum is determined for each of the plurality of areas and structural data each of the plurality of areas of the sample are determined from the respective reflectance spectra. More preferably, the polychromatic optical radiation is continuously scanned over the sample, and structural data of the sample are continuously determined for the areas over which the polychromatic optical radiation is continuously scanned, and from the obtained structural data of the plurality of measuring spots, the overall structural data of the sample are determined. That is, that the spot size is limited and due to scanning the measuring spot, the structural data of a plurality of partial areas of the sample can be determined. From these structural data of the plurality of partial areas, the structural data for (the whole or) a larger are of the sample can be determined. At least, it is possible to determine the structural data for the whole area of the sample for which structural data shall be obtained. In this case, scanning over sample can be made such that neighbouring scanning areas are either adjacent or partially overlapping. The spatial resolution of the structural data becomes more precise if the overlap of neighbouring scanning areas becomes larger. That is, the overlap of neighbouring scanning areas is preferably larger than 1%, more preferably larger than 5%, more preferably larger than 10%, more preferably larger than 20% and still more preferably larger than 40%.
Preferably, an area covered by the plurality of areas covers at least 10% of the total area of the sample. More preferably, an area covered by the plurality of areas covers at least 30%, more preferably at least 50%, more preferably at least 70% and still more preferably at least 90% of the total area of the sample.
Preferably, pre-known parameters of the sample structure and the optical properties of the sample material, the substrate material and of the ambient medium are used as input parameters.
Preferably a calculation of structural data and/or a calculation of the reference spectra (for pre-defined structural data) is performed by using a scattering matrix method as e.g. known from Anttu, N. & Xu, H. Q., “Scattering matrix method for optical excitation of surface plasmons in metal films with periodic arrays of sub-wavelength holes”, Physical Review B 83, 165431 (2011).
Preferably 3D-modelling of the light-sample interaction is performed by using versions of the Fourier Modal Method or, alternatively, formulations of the Rigorous Coupled Wave Approximation (RCWA).
For real-time analysis of the sample structure (e.g. an epitaxial nano-structure growth) it is preferred that the calculated optical response (reflectances) of the sample (e.g. nano-structure on a substrate) is stored before the measurement of the reflectances in a data base for all structural and compositional parameters relevant for the sample (e.g. a specific nano-structure growth process). During the measurement of the reflectances (e.g. during the growth process), from this pre-calculated multi-dimensional data-base of optical response data carefully optimized fast data-mining algorithms will find the best-fitting optical spectra (in case of spectroscopic in-situ monitoring) or the best fitting single-/multi-wavelength optical transients (in case of single-/multi-wavelength optical in-situ monitoring) for real-time determination of the characteristic sample parameters (e.g. nano-structure parameters).
For samples with structure dimensions of approximately the same order of magnitude as the transversal coherence length of the probing light, a semi-classical approach as described in Strittmatter, A. & , ReiBmann, L. & Trepk, T. & Pohl, U. W. & Bimberg, D. & Zettler, J.-T., “Optimization of GaN MOVPE growth on patterned Si substrates using spectroscopic in situ reflectance”, Journal of Crystal Growth 272, 76 (2004) can be taken. Here, spectra of virtual planar samples corresponding to the bottom and the top of the nano-structure are taken from the database. Their complex reflection coefficients are subsequently added, assuming a layer of air (or other suitable material) in between whose thickness can be varied to fit the aspect ratio of the nanowire structure. For a certain class of nano-dimensional structures and for an optimised polarisation direction of the incident light, this approach is faster than selecting the best-fitting spectrum from the pre-calculated database. Preferably, samples with structure dimensions of approximately the same order of magnitude as the transversal coherence length of the probing light are understood as samples where a ratio of a transversal coherence length of the probing light and a pitch L between adjacent nano-structures ranges between 0.1 and 10, more preferably between 0.3 and 3, more preferably between 0.5 and 2, more preferably between 0.7 and 1.4 and still more preferably between 0.9 and 1.1.
Preferably, the sample is a nano-wire (NW) structure on the surface of a semiconductor wafer. Preferably, the sample comprises a nano-wire (NW) structure having a regular pattern.
Preferably, the lateral dimension D of the sample, the central wavelength X of the light used for measuring the optical reflectance and/or the pitch L of the regular pattern of the sample range between 100 and 2000 nm, more preferably between 200 and 1500 nm and still more preferably between 400 and 1000 nm.
According to an aspect of the present invention, a method for controlling fabrication process parameters for the sample during epitaxial growth is disclosed, wherein structural data of at least a part of the sample are determined according to the method of the present invention, and fabrication process parameters of the sample are controlled according to the determined structural data. Such fabrication process parameters may be the pressure within a chamber in which the sample is grown, the temperature of the sample and the like.
According to an aspect of the present invention, an apparatus for determining structural data of a sample, comprising: means for irradiating polychromatic optical radiation onto the sample, wherein an area of the sample onto which the polychromatic optical radiation is irradiated ranges between 0.0004 mm2 to 4 mm2, means for determining an optical reflectance spectrum of the sample from the reflected polychromatic optical radiation, and means for determining structural data of the area of the sample onto which the polychromatic optical radiation is irradiated, by either assigning three-dimensional structural data from a data base to the sample or by calculating three-dimensional structural data from the determined optical reflectance spectrum.
Preferably, the apparatus further comprises means for repeatedly scanning the polychromatic optical radiation over the sample.
According to an aspect of the present invention, an apparatus for controlling fabrication process parameters for the sample during epitaxial growth is disclosed, comprising an apparatus for determining structural data of the sample according to the method of the present invention, and further comprising means adapted to control fabrication process parameters for the sample according to the determined structural data of the sample.
According to an aspect of the present invention, a method is disclosed comprising: arranging the nano-structures on a substrate, both together forming the sample and measuring the optical reflectance of a plurality of different measuring positions of the sample (which is arranged on a substrate) with a spatial resolution ranging between 0.02 mm×0.02 mm and 2 mm×2 mm. More preferably, the spatial resolution ranges between 0.05 mm×0.05 mm and 0.5 mm×0.5 mm. Preferably, the optical reflectance of the sample is measured by scanning a focused white-light beam or a single-wavelength focused laser-beam over the sample. Preferably, the measuring positions are arranged uniformly over the sample; however, the invention is not restricted hereto.
Preferably, the optical reflectance is measured either in an appropriate spectral range (preferably between 200 nm and 2000 nm) for spectroscopic analysis of the sample's optical response or with a sufficiently large time-resolution in case of single wavelength or multiple-wavelength growth monitoring. This is useful for detecting sufficiently clear and structured interference patterns of the sample structure in the optical in-situ measurement.
The method further comprises calculating a 3D-model from the measured optical reflectances of the plurality of different measuring positions of the sample, wherein 3D-modelling is performed by incorporating both, the interference effects of light reflected from the front and back interfaces of the sample structure and the interference effects due to superposition of neighbouring wave-fronts reflected from the sample top area and wave-fronts reflected from the substrate area between the samples.
Preferably, pre-known parameters of the sample structure and the optical properties of the sample material, the substrate material and of the ambient medium are used as input parameters.
Preferably 3D-modelling of the light-sample interaction is performed by using a scattering matrix method as e.g. known from Anttu, N. & Xu, H. Q., “Scattering matrix method for optical excitation of surface plasmons in metal films with periodic arrays of sub-wavelength holes”, Physical Review B 83, 165431 (2011).
Preferably 3D-modelling of the light-sample interaction is performed by using versions of the Fourier Modal Method or, alternatively, formulations of the Rigorous Coupled Wave Approximation (RCWA).
For real-time analysis of the sample structure (e.g. during epitaxial nano-structure growth) it is preferred that the calculated optical response (reflectances) of the sample (e.g. nano-structure on a substrate) is stored before the measurement of the reflectances in a data base for all structural and compositional parameters relevant for the sample (e.g. a specific nano-structure growth process). During the measurement of the reflectances (e.g. during the growth process), from this pre-calculated multi-dimensional data-base of optical response data carefully optimized fast data-mining algorithms will find the best-fitting optical spectra (in case of spectroscopic in-situ monitoring) or the best fitting single-/multi-wavelength optical transients (in case of single-/multi-wavelength optical in-situ monitoring) for real-time determination of the characteristic sample parameters (e.g. nano-structure parameters).
For samples with structure dimensions of approximately the same order of magnitude as the transversal coherence length of the probing light, a semi-classical approach as described in Strittmatter, A. & , ReiBmann, L. & Trepk, T. & Pohl, U. W. & Bimberg, D. & Zettler, J.-T., “Optimization of GaN MOVPE growth on patterned Si substrates using spectroscopic in situ reflectance”, Journal of Crystal Growth 272, 76 (2004) can be taken. Here, spectra of virtual planar samples corresponding to the bottom and the top of the nano-structure are taken from the database. Their complex reflection coefficients are subsequently added, assuming a layer of air (or other suitable material) in between whose thickness can be varied to fit the aspect ratio of the nanowire structure. For a certain class of nano-dimensional structures and for an optimised polarisation direction of the incident light, this approach is faster than selecting the best-fitting spectrum from the pre-calculated database.
Preferably, the sample is a nano-wire (NW) structure on the surface of a semiconductor wafer. Preferably, the sample comprises a nano-wire (NW) structure having a regular pattern.
In case of single-/multi-wavelength optical in-situ monitoring, the light used for measuring the optical reflectance of a plurality of different measuring positions of the sample, preferably is monochromatic.
In case of spectroscopic optical in-situ monitoring, the white light used for measuring the optical reflectance of a plurality of different measuring positions of the sample, preferably is in the spectroscopic range between 200 nm<λ<2000 nm (λ is the central wavelength of the spectroscopic range).
Preferably, the pitch L of the regular pattern of the sample and the central wavelength λ of the light used for measuring the optical reflectance satisfy the condition 0.1<L/λ<10, more preferably 0.25<L/λ<4 and still more preferably 0.5<L/λ<2. Preferably, the (maximum) lateral dimension D of the regular pattern of the sample and the wavelength λ of the light used for measuring the optical reflectance satisfy the condition 0.1<D/λ<10, more preferably 0.25<D/λ<4 and still more preferably 0.5<D/λ<2.
Preferably, the lateral dimension D of the sample, the central wavelength λ of the light used for measuring the optical reflectance and/or the pitch L of the regular pattern of the sample range between 100 and 2000 nm, more preferably between 200 and 1500 nm and still more preferably between 400 and 1000 nm.
According to another aspect of the present invention, an apparatus for measuring the dimensions of nano-structures on a substrate (of a sample) is disclosed.
The apparatus comprises means for measuring the optical reflectance of a plurality of different measuring positions on a sample with a spatial resolution ranging between 0.02 mm×0.02 mm and 2 mm×2 mm.
The apparatus further comprises means for calculating a 3D-model from the measured optical reflectances of the plurality of different measuring positions of the sample, wherein 3D-modelling is performed incorporating both, the interference effects of light reflected from the front and back interfaces of the nano-structure and the interference effects due to superposition of neighbouring wave-fronts reflected from the nano-structure area and wave-fronts reflected from the substrate area between the nano-structures.
Preferably, the apparatus further comprises a data base for storing structural and compositional parameters relevant for the sample.
Preferably, the apparatus further comprises means for comparing measured reflectances with pre-calculated reflectances stored in the data base. Preferably, the apparatus further comprises means for assigning the measured reflectances to best-fitting pre-calculated reflectances stored in the data base. The structure used for pre-calculating the assigned best-fitting pre-calculated reflectances is then used as the structure of the sample.
For samples with structure dimensions of approximately the same order of magnitude as the transversal coherence length of the probing light, a semi-classical approach as described in Strittmatter, A. &, ReiBmann, L. & Trepk, T. & Pohl, U. W. & Bimberg, D. & Zettler, J.-T., “Optimization of GaN MOVPE growth on patterned Si substrates using spectroscopic in situ reflectance”, Journal of Crystal Growth 272, 76 (2004) can be taken. Here, spectra of virtual planar samples corresponding to the bottom and the top of the nano-structure are taken from the database. Their complex reflection coefficients are subsequently added, assuming a layer of air (or other suitable material) in between whose thickness can be varied to fit the aspect ratio of the nanowire structure. For a certain class of nano-dimensional structures and for an optimised polarisation direction of the incident light, this approach is faster than selecting the best-fitting spectrum from the pre-calculated database.
In the following the invention will be described in further detail. The examples given are adapted to describe the invention, but not to limit the invention in any case.
The present invention discloses a method and an apparatus for performing optical in-situ measurements and related real-time analysis of critical structural parameters of ‘nano-wires’ and ‘quantum dots’ during their formation in epitaxial processes.
More specifically, the present invention is of special importance for the optical analysis of such 1D and 0D nanostructures having critical dimensions (diameter D and nano-wire length L, see
Only in the lower-limit case, where L<<λ and D<<λ, or in the upper limit case, where L<<λ and D<<λ, the conventional optical thin-film analysis can be applied successfully. In the lower-limit case so called effective medium models can be used for transforming the structure into a quasi-homogeneous film with an effective refractive index and an effective extinction coefficient. Certain scaling parameters in the specific effective medium model can be fitted to the nano-structure.
In the upper-limit case a straight-forward weighted superposition of the conventionally calculated reflectance response of the NW top area and of the substrate area between the NWs can be applied.
In the case of L˜λ and/or D˜λ, which is of significant practical importance, the conventional thin-film optical modelling methods completely fail and the following new method has to be applied:
In
Such NW arrays are currently intensively investigated as potential nano-scaled building blocks for enhancing the performance of semiconductor devices in the areas of photonics and optoelectronics. Possible applications are for example solar cells, light emitting diodes, lasers and photo-detectors. In all these applications both the NW diameter and length play important roles in the optical response of the devices. Using the method described here, one is able to determine in real-time and simultaneously both the NW diameter and length with accuracy far better than the diffraction limit and well within 10% of the values obtained from SEM measurements.
The method and apparatus is applicable for an even wider range of nano-structures and other periodically arranged 3-dimentional objects of nano-meter dimension.
The angle of incidence between the incoming light and the substrate below the nano-structure in not restricted to 0° (normal incidence), i.e., the method works well also for non-normal (oblique) incidence.
The different areas of the nano-structure contributing to the reflection of plane waves are not restricted to only two (as described above)—more complex nano-structures having more than two contributing nano-sized sub-areas can be analysed with the same method, requiring only slightly more complex models for the numerical solution of Maxwell's equation.
The different areas of the nano-structures contributing to the reflection of plane waves can even be tilted with respect to each other. Such nano-structures (e.g., having tilted facets with a certain tilt-angle) can be analysed, e.g., by multi-angle-of-incidence reflectance and basically the same numerical calculation method, requiring more complex models for the numerical solution of Maxwell's equation, respectively.
The method can be expanded from in-situ reflectance and transmittance analysis also to in-situ ellipsometry: the phase-shifting effects of the 1D- and 0D-nanostructures are included in the scattering matrix method.
The sample 10 comprising the substrate 12 and the plurality of regularly arranged nano-wires 14 is arranged within a process chamber 24 which is adapted to perform epitaxial growth of the sample 10. Preferably, the nano-wires 14 are arranged in a regular matrix but the invention is not limited thereto.
The process chamber 24 comprises means to control the temperature of the sample 10 by direct or indirect heating, the chamber pressure—through a combination of pressure controllers and process pumps—, the concentration and mixture of precursor material, e.g. by using mass flux controllers, and the rotation speed of the sample 10 for homogenisation of the growth.
The InP NW samples 14 shown in
The process chamber 24 comprises one window 26. A light source 20 produces white light which has a continuous spectrum within the visible wavelength range (380 to 760 nm). The light source 20 is preferably a heat radiator such as a tungsten lamp. The incident polychromatic optical radiation 16 is perpendicularly irradiated through the window 26 onto the sample 10 which is located within the process chamber 24. Before reaching the window 26, the incident polychromatic optical radiation 16 passes a partial mirror 34. After the incident polychromatic optical radiation 16 reaches the substrate 12 and the nano-wires 14, the optical radiation is reflected in dependence of the structural data of the substrate 12 and the plurality of nano-wires 14 which has been illuminated by the light source 20.
The diameter of the incident polychromatic optical radiation 16 (or equivalently the area of the sample which is irradiated) is chosen such that it is on the one hand side large enough to integrate over a sufficiently large number of nano-structures 14 and that it is on the other hand small enough to avoid suppression of the interference patterns due to local variations of the characteristic nano-wire parameters L and D within the detection spot (
Due to the chosen dimensions of the detection spot (measuring spot or area of the sample which is irradiated in relation to the spatial dimensions of the nano structures), not just interference effects of light reflected from the front and back interfaces of the nano-wire structure but also the interference effects due to superposition of neighbouring wave-fronts reflected from the nano-wire top area and wave-fronts reflected from the substrate area between the nano-wires contribute to the reflection spectrum.
Then, the reflected optical radiation 18 passes through the window 26 of the process chamber 24, is diverted by the mirrors 34 and 36, and is focussed by the lens 28 onto the detector 30, which is part of the detection means 22 which detects a spectrum of the reflected optical radiation 18. In the embodiment of the present invention, the light source 20, the detection means 22, the control means 32, the mirrors 34 and 36 and the lens 28 are integrated in a mutual housing of the scanning unit 40. However, the invention is not limited thereto, and it is alternatively possible to arrange the light source 20, the detection means 22, the control means 32, the mirrors 34 and 36 and the lens 28 separately from each other. The scanning unit 40 and/or the sample 10 itself may be moved along a direction 38 which is perpendicular to the incident optical radiation 16. In case of scanning the scanning unit 40 laterally over the sample 10, the window 26 has to be adapted such that incident optical radiation 16 and reflected optical radiation 18 can pass into and from the chamber 24 along the whole scanning path.
Then, the spectrum, which is obtained from the detector 30, is input into the control means 32 which then determines structural 3D data (i.e. the dimensions of the wires 14 as well as their pitch L) of the area of the sample 10 onto which the optical radiation 16 is irradiated by comparing the obtained reflectance spectrum with a plurality of pre-calculated reference spectra of a data base which is integrated in the control means 32. In the present embodiment, the measured optical reflectance spectrum of the sample 10 is compared with a plurality of pre-calculated optical reflectance spectra as e.g. shown in
| Number | Date | Country | Kind |
|---|---|---|---|
| 12 179 065.3 | Aug 2012 | EP | regional |
| 12 199 125.1 | Dec 2012 | EP | regional |
