photondiag2017 workshop
Diffraction gratings metrology and ray-tracing results for an XUV Raman spectrometer at FLASH
aDESY, Notkestrasse 85, Hamburg 22607, Germany, bHelmholz-Zentrum Berlin für Materialien und Energie, Albert-Einstein-Strasse 15, Berlin 12489, Germany, cUniversity of Hamburg, Notkestrasse 85, Hamburg 22607, Germany, and dCenter for Free-Elektron Laser Science, Notkestrasse 85, Hamburg 22607, Germany
*Correspondence e-mail: siarhei.dziarzhytski@desy.de
The extreme-ultraviolet double-stage imaging Raman spectrometer is a permanent experimental endstation at the plane-grating monochromator beamline branch PG1 at FLASH at DESY in Hamburg, Germany. This unique instrument covers the photon energy range from 20 to 200 eV with high energy resolution of about 2 to 20 meV (design values) featuring an efficient elastic line suppression as well as effective stray light rejection. Such a design enables studies of low-energy excitations like, for example, phonons in solids close to the vicinity of the elastic line. The Raman spectrometer effectively operates with four reflective off-axial parabolic mirrors and two plane-grating units. The optics quality and their precise alignment are crucial to guarantee best performance of the instrument. Here, results on a comprehensive investigation of the quality of the spectrometer diffraction gratings are presented. The gratings have been characterized by ex situ metrology at the BESSY-II Optics Laboratory, employing slope measuring deflectometry and interferometry as well as atomic force microscopy studies. The efficiency of these key optical elements has been measured at the at-wavelength metrology laboratory using the reflectometer at the BESSY-II Optics beamline. Also, the metrology results are discussed with respect to the expected of the instrument by including them in ray-tracing studies of the instrument.
Keywords: X-ray optics; XUV Raman spectrometer; metrology for synchrotron optics; NOM; reflectometry; ray tracing.
1. Introduction
The plane-grating monochromator beamline PG1 (Gerasimova et al., 2011; Dziarzhytski et al., 2016) at the soft X-ray/extreme-ultraviolet (XUV) free-electron laser FLASH in Hamburg (Ackermann et al., 2007; Tiedtke et al., 2009) is permanently equipped with a unique high-resolution XUV double-stage Raman spectrometer, dedicated to (resonant) inelastic soft X-ray scattering (IXS) experiments in the spectral region from 20 to 200 eV (Rübhausen et al., 2004; Rusydi et al., 2014). The optical design of the spectrometer is based on a confocal additive coupling of two high-resolution monochromators (SP1 and SP2) mediated by a middle slit (MS) (see Fig. 1).
Each monochromator is equipped with two off-axis parabolic mirrors and a plane grating.1 The spectrometer has no entrance slit and disperses along the vertical direction, thus the vertical size of the focal spot produced by the PG1 beamline Kirkpatrick–Baez (KB) refocusing optics on the sample (Dziarzhytski et al., 2016; Siewert et al., 2010) together with the resolution of the primary monochromator PG1 defines the resolution of the first spectrometer stage to a large extent. The spectrometer has a designed spectral resolution of 2 to 20 meV.
Such spectral resolution puts high demands on all optical elements in terms of figure and surface quality. In general, a slope error of the plane grating leads to a reduced spectral resolution. However, slope errors of the order of 0.05 arcsec r.m.s. are nowadays achievable. Such a small slope error does not affect the θ = 7° is the incidence angle of the mirror (Rübhausen et al., 2004). Off-axis parabolic mirrors of the spectrometer have a sagittal slope error below 1 arcsec. Such values of the optical quality parameters were chosen during the design phase of the spectrometer and pursuit in fabricating to minimize unwanted specular deflections of the rays from their ideal path resulting in reduction of the spectrometer resolution. Also, these values represent the technical limits of parabola production at that time and have been chosen in the closed discussion with the manufacture. Extremely precise metrology instruments are mandatory to characterize high-quality optical elements of the beamline and spectrometer operating in the soft X-ray/VUV spectral range. The Nanometer Optical Component Measuring Machine (NOM) and the atomic force microscope (AFM) at the BESSY-II Optics laboratory (Siewert et al., 2014) were used in combination with at-wavelength metrology at the BESSY-II Optics beamline (Schäfers et al., 2016; Sokolov et al., 2016) of the Helmholtz Zentrum Berlin to characterize the diffraction gratings of the Raman spectrometer. The gratings were produced by Carl Zeiss Optronics GmbH and first tested in 2008. Since then they have been partly used in operation/commissioning but also kept in storage under air pressure for a considerable amount of time. In order to exclude possible performance issues due to coating delamination or other unwanted degradation effects the coating quality and the overall grating efficiency have been thoroughly re-characterized. Parabolic mirrors have not been re-measured in the present work as they were kept in the spectrometer since their production and characterization at Carl Zeiss Optronics GmbH in 2006.
significantly. Of crucial importance are the slope errors of the parabolic mirrors. The reflection of the mirrors is perpendicular to the dispersion plane and the resolution is proportional to the sagittal slope error multiplied by the `forgiveness factor' , whereThe obtained results from metrology demonstrated some efficiency degradation and deviations from the optics specifications, as will be discussed in the following. These findings were implemented into ray-tracing package SHADOW (Cerrina & Sanches del Rio, 2010) to quantify their influence on the performance of the XUV Raman spectrometer (§3).
2. Gratings metrology
The spectrometer gratings are plane and blazed to yield maximum efficiency in first order. They are mechanically ruled, ion etched and coated with diamond-like carbon (DLC). The substrate material is Zerodur and the coating thickness is 45 nm. All gratings have been characterized ex situ. The at-wavelength efficiency of the gratings G1-3 and G2-32 were also measured with the reflectometer at the BESSY-II Optics beamline.
2.1. Ex situ metrology
Spectrometer gratings have been characterized ex situ by means of the BESSY-NOM (Siewert et al., 2004), regarding topography in terms of slope, and curvature in terms of the substrate meridional radius. The sagittal slope error was not measured due to the forgiveness factor assumption for the application of the grating. AFM measurements were made to characterize the groove profile of the gratings regarding blaze profile, groove density and micro-roughness on the grooves. The instrument applied is a NaniteAFM (SPM S200) by Nanosurf. While the NOM measurements allow the spatial frequency range from 1.2 mm up to aperture length to be verified (Siewert et al., 2016), the AFM gives a view on the nano-topology of the grating with a spatial resolution in the range from ≤10 nm up to a few µm, depending on the tip radius and the field of view applied for such measurements (Breil et al., 2002). The spatial frequency range covered by the slope error has an impact on the effects of classical aberration. The higher spatial frequency range as measured by means of the AFM has an impact on the efficiency (e.g. losing because of scattering) and spectral purity provided by the grating. Fig. 2 shows the results of the slope measurements for the gratings G1-3 and G2-3. Fig. 3 shows the state of the groove profile and micro-roughness on the grooves as measured by using an AFM.
Table 1 shows the results of the measurements in detail. The measurements reveal a compliance with the specification for most of the parameters like the slope error, radius of curvature, groove density and blaze angle. The AFM measurements have shown high values for the micro-roughness on the grooves of 1–6 nm r.m.s. for grating G1-3. This is probably because of aging effects during the years of storage. The micro-roughness on grating G2-3 is 0.80–1.61 nm r.m.s. slightly better compared with grating G1-3. However, the micro-roughness is out of specification for both gratings.
|
2.2. In situ metrology
Diffraction efficiencies of the Raman spectrometer gratings G1-3 and G2-3 have been measured with the reflectometer at the BESSY-II Optics beamline. A standard beam focus size ax (along the grating grooves) × ay (across the grooves) of 0.2 mm × 0.36 mm was used, which results in a footprint size of ax × ay/sin(θ) (where θ is the grazing incidence angle) on the grating. Since the measured area is rather small compared with the total grating working aperture, measurements at different points on gratings have been carried out in order to test a larger grating area (see Fig. 4).
Fig. 5 presents dispersion scans at fixed photon energy of 136 eV at different positions on both gratings under investigation. For both gratings only a small difference of the low-level background signal was observed. The area between the zeroth- and first-order peaks does not reveal any abnormal structure like a ghost peak or peak shape distortion, which could be related to grating structure defects. In general, both gratings tested exhibit an efficiency variation of only ∼0.1% across the measured points.
However, the measured efficiency energy dependence does not match very well with the efficiency calculations based on grating parameters obtained from the ex situ metrologies carried out after the manufacturing and from the present work (see Fig. 6).
The deviations between the efficiency values measured and calculated are on average in the limits of 10–15% for grating G1-3 and 2–5% for grating G2-3. In the case of grating G2-3 one can see a deviation in the shape of the measured curve compared with the calculated one. This points to possible structural deviation in the grating profile, which is not described by the model used. Overall, in spite of the found efficiency deviations, the investigated gratings show an acceptable performance. The design and the measurements results for gratings G1-3 and G2-3 are compiled in Table 1.
3. Ray tracing
The spectrometer is designed for high-resolution
experiments in the XUV spectral region, thus high-quality optical elements are mandatory to meet the designed performance specifications. Generally, figure and slope error imperfections manifest themselves in specular deflections of the rays from their ideal path, which thus results in focal spot broadening and a possible reduction of the spectrometer resolution. Furthermore, surface roughness (random irregularities in microscopic scale) can lead to a blurring of the image and a loss of contrast at the focus due to wide-angle scattering of the photon rays. The effects of such imperfections on the spectrometer performance are analyzed here.The measured r.m.s. slope error values were used in the pre-processor `WAVINESS' to simulate maps of the slope errors of the gratings. We also applied measured slope error profiles to the grating surfaces in the ray tracing and compared results of both approaches. The source for the ray tracing has two spectral lines in the vicinity of the blazed energy of the gratings and a rectangular shape with spatial dimensions of 5 µm × 20 µm (V × H) and uniform divergence of 37 mrad × 82 mrad (V × H). Such parameters reflect a realistic experimental PG1 focal size formed by the KB optics on the sample of the Raman spectrometer (Dziarzhytski et al., 2016).
Fig. 7 demonstrates a spectrometer resolution of 5.6 meV at a photon energy of 90 eV in the ideal case when no slope errors are taken into account.
First, slope errors of 0.056 and 0.08 arcsec r.m.s. in the meridional direction for gratings G1-3 and G2-3, respectively, and 0.1 arcsec r.m.s. in the sagittal directions for both gratings measured by the NOM instrument were used in the pre-processor `WAVINESS' to create maps of slope errors for the gratings. These maps were applied to the surface of the gratings to calculate the focal spot size and also to estimate the resolution of the spectrometer (see Fig. 8).
As one can see from a comparison of Figs. 7 and 8 the resolution of the spectrometer of 5.6 meV is not affected by the application of the calculated slope errors to the gratings. The vertical focus size is about 35 µm full width at half-maximum (FWHM) in both the ideal case and when slope errors are applied. Furthermore, the three-dimensional maps of the surface error generated out of the one-dimensional profiles measured by the NOM instrument were introduced into the simulations using the `PRESURFACE' pre-processor. Examples of the surface spline for gratings G1-3 and G2-3 are shown in Fig. 9. The slope error in the X direction (along the grooves) was generated for a slope error of 0.1 arcsec r.m.s. for both gratings G1-3 and G2-3. The ray-tracing results with measured surfaces for gratings G1-3 and G2-3 are shown in Fig. 10.
The resolution of the spectrometer is also not reduced due to the applied measured slope error profiles. The vertical focal size is roughly 36 µm FWHM versus 35 µm in the ideal case. The horizontal size of the focal spot is not affected and remains about 43 µm FWHM. In general, the effect of the slope error of the diffraction gratings on the spectrometer resolution is much weaker compared with that from the focusing elements, namely the four (M1–M4) off-axial parabolic mirrors used in the spectrometer. The mirrors' parameters are summarized in Table 2.3
|
The spectrometer resolution reduces from 5.6 meV to 10 meV at 90 eV photon energy if the slope errors of the off axial parabolas are taken into account.
The effect of the measured micro-roughness was also estimated. The power spectral density function (PSD) was created by means of the JNTPSCALC tool in SHADOW. A Gaussian correlation function was chosen with correlation length of 2 cm−1 for grating G1-3 and 5 cm−1 for grating G2-3. RMS values of the surface roughness of 60 and 16 Å in the X and Y directions were taken for gratings G1-3 and G2-3, respectively. The results are shown in Fig. 11.
The measured micro-roughness is out of specification for both gratings which leads to a blurring effect of the image and a loss of contrast at the focus as well as less efficient suppression of unwanted scattered light. However, the signal-to-background ratio in this case is still high enough to use these gratings in the spectrometer.
4. Conclusion
Our optics metrology and ray tracing have clearly demonstrated that, although the spectrometer diffraction gratings had experienced some degradation in efficiency and roughness, no strong negative effect on the slope error and micro-roughness due to possible suspected coating delamination and other processes has been revealed. The gratings provide reasonable efficiency and can be further used for the high-resolution RIXS experiments at the XUV Raman spectrometer at the PG1 beamline at FLASH.
Footnotes
1In fact, each monochromator stage has a set of four gratings to cover the full spectral range; however, only one grating is used in each stage at a time.
2Grating GX-Y corresponds to grating Y (Y = 1–4) at the monochromator stage X (X = 1, 2).
3Mirrors were measured by Carl Zeiss Optronics GmbH in 2006. Data are taken from the production and control report.
References
Ackermann, W. et al., (2007). Nat. Photon. 1, 336–342. Google Scholar
Breil, R., Fries, T., Garnaes, J., Haycocks, J., Hüser, D., Joergensen, J., Kautek, W., Koenders, L., Kofod, N., Koops, K. R., Korntner, R., Lindner, B., Mirandé, W., Neubauer, A., Peltonen, J., Picotto, G. B., Pisani, M., Rothe, H., Sahre, M., Stedman, M. & Wilkening, G. (2002). Precis. Eng. 26, 296–305. CrossRef Google Scholar
Cerrina, F. & Sanches del Rio, M. (2010). Handbook of Optics, 3rd ed., ch. 35. New York: McGraw Hill. Google Scholar
Dziarzhytski, S., Gerasimova, N., Goderich, R., Mey, T., Reininger, R., Rübhausen, M., Siewert, F., Weigelt, H. & Brenner, G. (2016). J. Synchrotron Rad. 23, 123–131. Web of Science CrossRef IUCr Journals Google Scholar
Gerasimova, N., Dziarzhytski, S. & Feldhaus, J. (2011). J. Mod. Opt. 58, 1480–1485. Web of Science CrossRef Google Scholar
Rübhausen, M., Schulz, B., Burth, K., Bäckström, J., Kunze, J., Reininger, R., Nordgren, J., Söderström, J., Rubensson, J.-E., Börjesson, L., Abbamonte, P., Cooper, S. L., Martins, M., Föhlisch, A., Wurth, W., Feldhaus, J. & Schneider, J. (2004). Technical Design Report. BMBF Project No. 05KS4GU2. DESY, Hamburg, Germany. Google Scholar
Rusydi, A., Goos, A., Binder, S., Eich, A., Botril, K., Abbamonte, P., Yu, X., Breese, M. B. H., Eisaki, H., Fujimaki, Y., Uchida, S., Guerassimova, N., Treusch, R., Feldhaus, J., Reininger, R., Klein, M. V. & Rübhausen, M. (2014). Phys. Rev. Lett. 113, 067001. Web of Science CrossRef PubMed Google Scholar
Schäfers, F. (1996). BESSY Technischer Bericht TB 202/96. BESSY, Berlin, Germany. Google Scholar
Schäfers, F., Bischoff, P., Eggenstein, F., Erko, A., Gaupp, A., Künstner, S., Mast, M., Schmidt, J.-S., Senf, F., Siewert, F., Sokolov, A. & Zeschke, T. (2016). J. Synchrotron Rad. 23, 67–77. Web of Science CrossRef IUCr Journals Google Scholar
Schäfers, F. & Krumrey, M. (1996). BESSY Technischer Bericht TB 201/96. BESSY, Berlin, Germany. Google Scholar
Siewert, F., Buchheim, J., Zeschke, T., Störmer, M., Falkenberg, G. & Sankari, R. (2014). J. Synchrotron Rad. 21, 968–975. Web of Science CrossRef CAS IUCr Journals Google Scholar
Siewert, F., Noll, T., Schlegel, T., Zeschke, T. & Lammert, H. (2004). AIP Conf. Proc. 705, 847–850. CrossRef Google Scholar
Siewert, F., Reininger, R., Rübhausen, M., Garrett, R., Gentle, I., Nugent, K. & Wilkins, S. (2010). AIP Conf. Proc. 1234, 752–755. CrossRef CAS Google Scholar
Siewert, F., Zeschke, T., Arnold, T., Paetzelt, H. & Yashchuk, V. V. (2016). Rev. Sci. Instrum. 87, 051907. Web of Science CrossRef PubMed Google Scholar
Sokolov, A., Bischoff, P., Eggenstein, F., Erko, A., Gaupp, A., Künstner, S., Mast, M., Schmidt, J.-S., Senf, F., Siewert, F., Zeschke, T. & Schäfers, F. (2016). Rev. Sci. Instrum. 87, 052005. Web of Science CrossRef PubMed Google Scholar
Tiedtke, K., Azima, A., von Bargen, N., Bittner, L., Bonfigt, S., Düsterer, S., Faatz, B., Frühling, U., Gensch, M., Gerth, C., Guerassimova, N., Hahn, U., Hans, T., Hesse, M., Honkavaar, K., Jastrow, U., Juranic, P., Kapitzki, S., Keitel, B., Kracht, T., Kuhlmann, M., Li, W. B., Martins, M., Núñez, T., Plönjes, E., Redlin, H., Saldin, E. L., Schneidmiller, E. A., Schneider, J. R., Schreiber, S., Stojanovic, N., Tavella, F., Toleikis, S., Treusch, R., Weigelt, H., Wellhöfer, M., Wabnitz, H., Yurkov, M. V. & Feldhaus, J. (2009). New J. Phys. 11, 023029. Web of Science CrossRef Google Scholar
This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.