2782
OPTICS LETTERS / Vol. 39, No. 9 / May 1, 2014
Tunable two-color hard x-ray multilayer Bragg mirrors S. Roling,1,* S. Braun,2 P. Gawlitza,2 M. Wöstmann,1 E. Ziegler,3 and H. Zacharias1 1
Physikalisches Institut, Westfälische Wilhelms-Universität Münster, Wilhelm-Klemm Straße 10, 48149 Münster, Germany 2
Fraunhofer IWS, Winterbergstraße 28, 01277 Dresden, Germany 3
ESRF, 6 Rue Jules Horowitz, 38000 Grenoble, France *Corresponding author: s_roli02@uni‑muenster.de
Received March 5, 2014; accepted March 28, 2014; posted April 8, 2014 (Doc. ID 207680); published April 30, 2014 A tunable two-color multilayer Bragg coating capable of simultaneously reflecting the fundamental and the third harmonic of an x-ray free-electron laser at the same angle and with high reflectance R > 0.70 is presented. The novel coating will enable two-color x-ray pump/x-ray probe experiments. This mirror consists of a Si substrate that is coated with two different types of multilayer systems, Mo∕B4 C layers with a periodicity of d 3.2 nm directly on the substrate and Ni∕B4 C layers with a periodicity of d 11.85 nm on top. Fundamental radiation with photon energies between 3 and 9 keV is reflected by a Ni∕B4 C multilayer system while the third harmonic (9 keV < hν < 27 keV) passes this system and is reflected by the Mo∕B4 C multilayers. The principle has successfully been proven at the beamline BM05 at ESRF. © 2014 Optical Society of America OCIS codes: (340.0340) X-ray optics; (310.4165) Multilayer design; (230.4170) Multilayers; (120.5700) Reflection. http://dx.doi.org/10.1364/OL.39.002782
Hard x-ray free-electron lasers (XFELs) provide ultrashort and ultrabright light pulses that enable new classes of x-ray experiments. This is a great challenge for optical instrumentation. In addition to the already operating LCLS at the Stanford Linear Accelerator Center [1] and SACLA in Japan [2], the European XFEL is now under construction in Hamburg [3]. For the SASE 2 beamline at the European XFEL a hard x-ray split-and-delay unit (SDU) for photon energies between hν 5 and hν 20 keV is being developed that enables hard x-ray pump/probe experiments with subpicosecond time resolution [4]. The SDU is based on a geometrical wavefront beam splitter (BS) and multilayer Bragg coatings that permit larger grazing angles and therefore longer pulse delays for a given size. The setup of the optical pathway is schematically shown in Fig. 1. The XFEL beam enters the SDU from the left side and is reflected by the first mirror (S1) downward in the direction of the BS. The lower green part of the beam is reflected into the upper delay arm while the upper orange part passes the sharp edge in the direction of the lower delay arm. The mirrors of both delay arms (U1, U2, D1, D2) are moved along the split beam direction in order to introduce a temporal delay between the split pulses. After the hard x-ray pulse has passed through the lower delay line it is reflected by the recombination mirror (RC) in the direction of the last mirror (S8). The pulse on the upper beam path passes the sharp edge of the recombination mirror unaffected. In this point symmetric mirror scheme the recombination mirror RC acts as the counterpart of the BS. The last mirror (S8) reflects both beams into their original direction. Besides the fundamental photon energy FEL also emit odd order harmonics [5]. The SDU is designed to be utilized for one- and two-color pump/probe experiments. In the latter case the fundamental XFEL radiation passes along one branch of the SDU and the third harmonic via the other one. However, both the fundamental and the harmonic pulses have to be reflected by the mirrors S1 and S8, which poses a severe problem for conventional multilayer Bragg coatings. Neglecting dispersion 0146-9592/14/092782-04$15.00/0
corrections, a multilayer mirror reflects the kth harmonic of the fundamental wavelength λ into the nth reflection order, according to 2d sin θ
n λ k
(1)
for a fixed angle θ and a given periodicity d. In the x-ray region the reflectivity of all mirror materials rapidly decreases with increasing photon energy for a fixed incidence angle θ and with increasing grazing angle for a fixed wavelength λ [6]. Furthermore, due to interface roughness and absorption within the layers the reflectance of a conventional Bragg multilayer mirror strongly decreases for higher reflection orders at correspondingly larger angles θ. Similarly, the reflectance decreases for shorter wavelengths of the higher harmonics for a given incidence angle θ. An additional suppression of higher reflection orders n may occur for certain ratios Γ d0 ∕d, the ratio of the thickness d’ of the bottom layer to the total thickness d of the bilayer. For higher harmonics k the presence of absorption edges between the fundamental and harmonic photon energies may also lead to a reduced reflectance. As a consequence, there exists no realistic way to design an x-ray mirror consisting of a single multilayer system that shows a good reflectance for the fundamental photon energy as well as for the third harmonic at the same incidence angle θ. Besides these fundamental constraints the reflectance of different materials used for the multilayers changes with the photon energy range due to the presence of absorption edges. A new kind of two-color multilayer Bragg system has been designed, coated onto a silicon substrate,
Fig. 1.
Optical layout of the SDU for the European XFEL.
© 2014 Optical Society of America
May 1, 2014 / Vol. 39, No. 9 / OPTICS LETTERS
and experimentally evaluated to overcome these obstacles. The coating consists of two different multilayer systems on top of each other. The top system transmits the high-energy hard x-ray photons while it reflects the low-energy ones. The bottom system then reflects the high-energy x rays. The top system has to be designed such that a high reflectivity is achieved for the lowenergy x rays while simultaneously a low absorption for the high-energy photons is provided. It turns out that such configurations can generally be found when the hard x-ray energies are sufficiently far apart, like the fundamental and third harmonic of a FEL. In the present application x-ray energies of 5–6.7 keV and 15–19.5 keV shall be reflected. Simulations based on the CXRO database [7] show that this can ideally be achieved by using Ni∕B4 C multilayers for the top and Mo∕B4 C layers for the bottom system. Additionally, multilayers with B4 C spacer layers are known for their high thermal stability, which is advantageous for the application in intense x-ray beams. First the Mo∕B4 C system with n 120 bilayers consisting of a multilayer period of d 3.2 nm and a Γ ratio of Γ 0.3 is coated onto an ultraflat silicon substrate that shows a roughness r of less than r < 0.2 nm (r.m.s). On top of this multilayer system only n 4 bilayers of Ni∕B4 C are applied with a period of d 11.85 nm and Γ 0.4. In the present case the attenuation of the high-energy x rays is predominantly given by the distance L2·
d·γ ; sin θ
2783
The two-color multilayer Bragg mirror was deposited onto a 380 × 25 × 40 mm (LxWxH) T-shaped silicon substrate showing a roughness r of less than r < 0.2 nm (r.m.s). For all materials (Mo, Ni, B4 C) we used 1000 eV Kr ions for sputtering and obtained deposition rates of about 0.1 nm∕s and 0.03 nm∕s for the metals and B4 C, respectively. The period thicknesses and Γ ratios of the two multilayer stacks were adjusted by a time controlled, well-defined movement of the mirror across a 100 mm wide aperture. In order to evaluate the performance of the novel twocolor multilayer coating reflectivity measurements have been performed at beamline BM05 at the ESRF (Grenoble). The multilayer mirror was mounted onto a six-axis goniometer and the grazing angle θ was scanned while the detector was set at an angle of 2θ. In order to calculate the reflectivity R I∕I 0 the intensity I 0 of the x-ray beam without being reflected by the mirror was measured. This was done for the fundamental photon energies at hν 5 and 6 keV and the corresponding third harmonic energies at hν 15 and hν 18 keV, respectively. The results are shown in Fig. 2. The upper graph of Fig. 2 shows the angle scan of the fundamental radiation; the lower one shows the corresponding scan for the harmonics. It is obvious that the Bragg angles of the fundamental radiation at hν 5 and hν 6 keV perfectly match with the Bragg angles of the harmonics at θ 0.756° and θ 0.635° for hν 15 and 18 keV,
(2)
which the radiation has to pass through Ni. The attenuation by the top B4 C layers can be neglected. With an adequate number of periods for the top multilayer its reflectance for the fundamental wavelength and the transmission for the harmonic can be easily adjusted. For a photon energy of hν 15 keV and the corresponding glancing angle of θ 0.76° a total transmission through the Ni∕B4 C system of T 15 0.84 is calculated [7]. A similar calculation for a photon energy of hν 18 keV yields a total transmission of T 18 0.88. This transmission increases for harder x rays, being in the range of T ∼ 0.97 for the fifth harmonic at hν 30 keV. For the deposition of the two-color multilayer Bragg mirror we used a commercial dual ion beam sputter deposition (IBSD) machine (Roth & Rau, IonSys 1600) [8]. As main parts, the machine contains two linear electron cyclotron resonance (ECR) ion sources with beam dimensions of 400 mm × 100 mm, one for sputtering and the other one for film growth assistance. The ion sources can be operated at energies between 50 and 2000 eV enabling the deposition of smooth and dense films. Substrates are moved linearly across differently shaped slits to fulfill film thickness precision and homogeneity constraints, both typically on the order of 99.9%. Multilayer stacks are produced layer-by-layer by alternate sputtering of the materials needed. During the process up to six different targets can be used. Due to the low process pressure (∼10−4 mbar) and the rather small divergence of the depositing particle flux the IBSD technique allows the deposition of films with sharp edge boundaries.
Fig. 2. θ − 2θ angle scan of the two-color multilayer coating. For the harmonic radiation the reflectance calculated without passing the Ni∕B4 C system is depicted as dashed line. Theoretical calculations for the corresponding photon energies [7] are depicted as black lines.
2784
OPTICS LETTERS / Vol. 39, No. 9 / May 1, 2014
respectively. The reflectance for the fundamental radiation at hν 6 keV is measured to RF 0.78. The corresponding values for the third harmonics are RH 0.75 and RH 0.70 at hν 18 and hν 15 keV, respectively. Without passing the Ni∕B4 C stack the reflectivities of the lower stack are deduced to R15 RH ∕T 15 0.83 at hν 15 keV and R18 RH ∕T 18 0.85 at hν 18 keV, respectively, in accordance with theoretical expectations [7]. The 5 keV data displayed in Fig. 2 (orange line) had to be corrected for a contribution from the harmonic radiation, because in the present experimental setup the x-ray beam passes through 1.8 m air that attenuates the 5 keV intensity to a level comparable to that of the 15 keV radiation still present behind a Si(111) double crystal monochromator. For the measurement of the 6 keV radiation such corrections are not necessary, because the harmonic radiation at 18 keV does not significantly contribute to the signal. The angular widths ΔθH of the reflection curves of the harmonic radiation are determined to be ΔθH 228 μrad and ΔθH 193 μrad (FWHM) at hν 15 and hν 18 keV, respectively. The angular widths ΔθF of the reflection of the fundamental radiation are twelve times broader, with ΔθF 2.8 mrad and ΔθF 2.3 mrad at hν 5 and hν 6 keV, respectively. For the fundamental wavelength only the reflections from four bilayers constructively interfere when the Bragg condition is met, while for the harmonics the reflections from 120 bilayers contribute to the Bragg peak. By analogy with multiple slit interference, where the peak width scales with 1∕N and N are the number of slits, this leads to significantly narrower peaks for the reflection of the harmonic radiation. However, for a multilayer reflection the widths of the Bragg peaks do not scale strictly with 1∕N because the intensity reflected by one layer decreases for deeper layers. Thus their contribution to the Bragg peak also decreases. Another important aspect is the energetic width of the reflection. The relative width of the hard x-ray SASE pulses is typically on the order of ΔE∕E ∼ 0.1–0.5% [1,9], which corresponds to ΔE 18 eV–90 eV at hν 18 keV. The spectral widths of the multilayer reflections need to be broader for two reasons. First, a loss of pulse energy has to be avoided, and second the SDU is foreseen to be used for a measurement of the temporal coherence properties of the FEL [9,10]. For the latter, constraining the spectrum of the FEL radiation by the multilayers would yield overestimated values for the coherence time due to the Wiener–Khintchin theorem. The energetic width has been evaluated by aligning the mirror at the optimal angle θH 0.63° found in the previous measurement and then scanning the photon energy transmitted through the monochromator. The reflectance observed for photon energies in the vicinity of hν ∼ 18 keV is shown in Fig. 3. At a fixed angle of incidence of θH 0.63° the spectral width is measured to ΔE 280 eV, which is about 3–16 times broader than the spectrum that is expected for the XFEL pulses. According to the condition ΔE Δθ ≈ E tan θ
3
Fig. 3. Scan of the photon energy for a fixed angle of incidence of θH 0.63°.
that can be derived from the Bragg equation for multilayer mirrors [11] the measurement of the relative spectral width of ΔE∕E 0.016 appears to be in good agreement with the measurement of the relative angular width of Δθ∕ tanθ 0.017. For the experiments at the European XFEL conservation of the quality of the x-ray pulses is crucial. For instance, the influence of the x-ray optics on the spatial profile of the beam has to be minimized, which requires a homogeneous reflectance along the whole surface of the mirror. A scan along the mirror has been performed at hν 18 keV. The beam was confined by means of a 50 μm × 50 μm aperture yielding a footprint on the mirror of 4.6 mm × 50 μm. The reflectance at different positions of the mirror is shown in Fig. 4. The measurement shows a perfectly homogenous coating between l 75 mm and l 190 mm. For l 50 mm the reflectivity drops by 6%, which is still tolerable. The drop can be explained by a slightly varying multilayer period d which has been verified by Cu-Kα reflectometry. An important issue for all optics exposed to the intense x-ray beam of the European XFEL is the risk of single shot damage. The damage thresholds found by theoretical calculations [12] as well as experimental results [13] for different types of multilayer coatings are higher by more than two orders of magnitude than in the present case with expected fluences below 0.3 mJ∕cm2 .
Fig. 4. Reflectance as a function of the position along the mirror surface; hν 18 keV.
May 1, 2014 / Vol. 39, No. 9 / OPTICS LETTERS
In conclusion, the principle of the novel two-color hard x-ray multilayer coating has successfully been proven. With reflectivities higher than R 70% in all cases the new mirror will allow two-color hard x-ray pump/probe experiments with the SDU. Based on the two multilayer stacks chosen the full spectral range from ∼3 keV to greater than 27 keV can be covered with reflectivities higher than R 50%. Only at the Ni 1 s and Mo 1 s ionization thresholds at 8.33 and 20 keV, respectively, a decrease of the reflectance has to be encountered. Further optimization should allow the reflection of the 5th harmonic of the FEL together with the fundamental radiation.
4. 5. 6. 7. 8.
Funding by the Bundesministerium für Bildung und Forschung (BMBF-Projekt: 05K10PM2) is acknowledged.
9.
References 1. P. Emma, R. Akre, J. Arthur, R. Bionta, C. Bostedt, J. Bozek, A. Brachmann, P. Bucksbaum, R. Coffee, F.-J. Decker, Y. Ding, D. Dowell, S. Edstrom, A. Fisher, J. Frisch, S. Gilevich, J. Hastings, G. Hays, Ph. Hering, Z. Huang, R. Iverson, H. Loos, M. Messerschmidt, A. Miahnahri, S. Moeller, H.-D. Nuhn, G. Pile, D. Ratner, J. Rzepiela, D. Schultz, T. Smith, P. Stefan, H. Tompkins, J. Turner, J. Welch, W. White, J. Wu, G. Yocky, and J. Galayda, Nat. Photonics 4, 641 (2010). 2. D. Pile, Nat. Photonics 5, 456 (2011). 3. M. Altarelli, R. Brinkmann, M. Chergui, W. Decking, B. Dobson, S. Düsterer, G. Grübel, W. Graeff, H. Graafsma, J. Hajdu, J. Marangos, J. Pflüger, H. Redlin, D. Riley,
10. 11.
12. 13.
2785
I. Robinson, J. Rossbach, A. Schwarz, K. Tiedtke, T. Tschentscher, I. Vartaniants, H. Wabnitz, H. Weise, R. Wichmann, K. Witte, A. Wolf, M. Wulff, and M. Yurkov, “The European XFEL,” Technical Design Report (DESY Hamburg, 2006). S. Roling, L. Samoylova, B. Siemer, H. Sinn, F. Siewert, F. Wahlert, M. Wöstmann, and H. Zacharias, Proc. SPIE 8504, 850407 (2012). E. L. Saldin, E. A. Schneidmiller, and M. V. Yurkov, Phys. Rev. ST Accel. Beams 9, 030702 (2006). E. Spiller, Adv. X-Ray Anal. 42, 297 (1998). B. L. Henke, E. M. Gullikson, and J. C. Davis, At. Data Nucl. Data Tables 54, 181 (1993). P. Gawlitza, S. Braun, S. Lipfert, and A. Leson, Proc. SPIE 6317, 63170G (2006). G. Geloni, E. Saldin, L. Samoylova, E. Schneidmiller, H. Sinn, Th. Tschentscher, and M. Yurkov, New J. Phys. 12, 035021 (2010). S. Roling, B. Siemer, M. Wöstmann, H. Zacharias, R. Mitzner, A. Singer, K. Tiedtke, and I. A. Vartanyants, Phys. Rev. ST Accel. Beams 14, 080701 (2011). A. Kazimirov, D.-M. Smilgies, Q. Shen, X. Xiao, Q. Hao, E. Fontes, D. H. Bilderback, S. M. Gruner, Y. Platonov, and V. V. Martynov, J. Synchrotron Radiat. 13, 204 (2006). R. A. Loch, R. Sobierajski, E. Louis, J. Bosgra, and F. Bijkerk, Opt. Express 20, 28200 (2012). E. Louis, A. R. Khorsand, R. Sobierajski, E. D. van Hattum, M. Jurek, D. Klinger, J. B. Pelka, L. Juha, J. Chalupsky, J. Cihelka, V. Hajkova, U. Jastrow, S. Toleikis, H. Wabnitz, K. Tiedtke, J. Gaudin, E. M. Gullikson, and F. Bijkerk, Proc. SPIE 7361, 73610I (2013).
OPTICS LETTERS / Vol. 39, No. 9 / May 1, 2014
Tunable two-color hard x-ray multilayer Bragg mirrors S. Roling,1,* S. Braun,2 P. Gawlitza,2 M. Wöstmann,1 E. Ziegler,3 and H. Zacharias1 1
Physikalisches Institut, Westfälische Wilhelms-Universität Münster, Wilhelm-Klemm Straße 10, 48149 Münster, Germany 2
Fraunhofer IWS, Winterbergstraße 28, 01277 Dresden, Germany 3
ESRF, 6 Rue Jules Horowitz, 38000 Grenoble, France *Corresponding author: s_roli02@uni‑muenster.de
Received March 5, 2014; accepted March 28, 2014; posted April 8, 2014 (Doc. ID 207680); published April 30, 2014 A tunable two-color multilayer Bragg coating capable of simultaneously reflecting the fundamental and the third harmonic of an x-ray free-electron laser at the same angle and with high reflectance R > 0.70 is presented. The novel coating will enable two-color x-ray pump/x-ray probe experiments. This mirror consists of a Si substrate that is coated with two different types of multilayer systems, Mo∕B4 C layers with a periodicity of d 3.2 nm directly on the substrate and Ni∕B4 C layers with a periodicity of d 11.85 nm on top. Fundamental radiation with photon energies between 3 and 9 keV is reflected by a Ni∕B4 C multilayer system while the third harmonic (9 keV < hν < 27 keV) passes this system and is reflected by the Mo∕B4 C multilayers. The principle has successfully been proven at the beamline BM05 at ESRF. © 2014 Optical Society of America OCIS codes: (340.0340) X-ray optics; (310.4165) Multilayer design; (230.4170) Multilayers; (120.5700) Reflection. http://dx.doi.org/10.1364/OL.39.002782
Hard x-ray free-electron lasers (XFELs) provide ultrashort and ultrabright light pulses that enable new classes of x-ray experiments. This is a great challenge for optical instrumentation. In addition to the already operating LCLS at the Stanford Linear Accelerator Center [1] and SACLA in Japan [2], the European XFEL is now under construction in Hamburg [3]. For the SASE 2 beamline at the European XFEL a hard x-ray split-and-delay unit (SDU) for photon energies between hν 5 and hν 20 keV is being developed that enables hard x-ray pump/probe experiments with subpicosecond time resolution [4]. The SDU is based on a geometrical wavefront beam splitter (BS) and multilayer Bragg coatings that permit larger grazing angles and therefore longer pulse delays for a given size. The setup of the optical pathway is schematically shown in Fig. 1. The XFEL beam enters the SDU from the left side and is reflected by the first mirror (S1) downward in the direction of the BS. The lower green part of the beam is reflected into the upper delay arm while the upper orange part passes the sharp edge in the direction of the lower delay arm. The mirrors of both delay arms (U1, U2, D1, D2) are moved along the split beam direction in order to introduce a temporal delay between the split pulses. After the hard x-ray pulse has passed through the lower delay line it is reflected by the recombination mirror (RC) in the direction of the last mirror (S8). The pulse on the upper beam path passes the sharp edge of the recombination mirror unaffected. In this point symmetric mirror scheme the recombination mirror RC acts as the counterpart of the BS. The last mirror (S8) reflects both beams into their original direction. Besides the fundamental photon energy FEL also emit odd order harmonics [5]. The SDU is designed to be utilized for one- and two-color pump/probe experiments. In the latter case the fundamental XFEL radiation passes along one branch of the SDU and the third harmonic via the other one. However, both the fundamental and the harmonic pulses have to be reflected by the mirrors S1 and S8, which poses a severe problem for conventional multilayer Bragg coatings. Neglecting dispersion 0146-9592/14/092782-04$15.00/0
corrections, a multilayer mirror reflects the kth harmonic of the fundamental wavelength λ into the nth reflection order, according to 2d sin θ
n λ k
(1)
for a fixed angle θ and a given periodicity d. In the x-ray region the reflectivity of all mirror materials rapidly decreases with increasing photon energy for a fixed incidence angle θ and with increasing grazing angle for a fixed wavelength λ [6]. Furthermore, due to interface roughness and absorption within the layers the reflectance of a conventional Bragg multilayer mirror strongly decreases for higher reflection orders at correspondingly larger angles θ. Similarly, the reflectance decreases for shorter wavelengths of the higher harmonics for a given incidence angle θ. An additional suppression of higher reflection orders n may occur for certain ratios Γ d0 ∕d, the ratio of the thickness d’ of the bottom layer to the total thickness d of the bilayer. For higher harmonics k the presence of absorption edges between the fundamental and harmonic photon energies may also lead to a reduced reflectance. As a consequence, there exists no realistic way to design an x-ray mirror consisting of a single multilayer system that shows a good reflectance for the fundamental photon energy as well as for the third harmonic at the same incidence angle θ. Besides these fundamental constraints the reflectance of different materials used for the multilayers changes with the photon energy range due to the presence of absorption edges. A new kind of two-color multilayer Bragg system has been designed, coated onto a silicon substrate,
Fig. 1.
Optical layout of the SDU for the European XFEL.
© 2014 Optical Society of America
May 1, 2014 / Vol. 39, No. 9 / OPTICS LETTERS
and experimentally evaluated to overcome these obstacles. The coating consists of two different multilayer systems on top of each other. The top system transmits the high-energy hard x-ray photons while it reflects the low-energy ones. The bottom system then reflects the high-energy x rays. The top system has to be designed such that a high reflectivity is achieved for the lowenergy x rays while simultaneously a low absorption for the high-energy photons is provided. It turns out that such configurations can generally be found when the hard x-ray energies are sufficiently far apart, like the fundamental and third harmonic of a FEL. In the present application x-ray energies of 5–6.7 keV and 15–19.5 keV shall be reflected. Simulations based on the CXRO database [7] show that this can ideally be achieved by using Ni∕B4 C multilayers for the top and Mo∕B4 C layers for the bottom system. Additionally, multilayers with B4 C spacer layers are known for their high thermal stability, which is advantageous for the application in intense x-ray beams. First the Mo∕B4 C system with n 120 bilayers consisting of a multilayer period of d 3.2 nm and a Γ ratio of Γ 0.3 is coated onto an ultraflat silicon substrate that shows a roughness r of less than r < 0.2 nm (r.m.s). On top of this multilayer system only n 4 bilayers of Ni∕B4 C are applied with a period of d 11.85 nm and Γ 0.4. In the present case the attenuation of the high-energy x rays is predominantly given by the distance L2·
d·γ ; sin θ
2783
The two-color multilayer Bragg mirror was deposited onto a 380 × 25 × 40 mm (LxWxH) T-shaped silicon substrate showing a roughness r of less than r < 0.2 nm (r.m.s). For all materials (Mo, Ni, B4 C) we used 1000 eV Kr ions for sputtering and obtained deposition rates of about 0.1 nm∕s and 0.03 nm∕s for the metals and B4 C, respectively. The period thicknesses and Γ ratios of the two multilayer stacks were adjusted by a time controlled, well-defined movement of the mirror across a 100 mm wide aperture. In order to evaluate the performance of the novel twocolor multilayer coating reflectivity measurements have been performed at beamline BM05 at the ESRF (Grenoble). The multilayer mirror was mounted onto a six-axis goniometer and the grazing angle θ was scanned while the detector was set at an angle of 2θ. In order to calculate the reflectivity R I∕I 0 the intensity I 0 of the x-ray beam without being reflected by the mirror was measured. This was done for the fundamental photon energies at hν 5 and 6 keV and the corresponding third harmonic energies at hν 15 and hν 18 keV, respectively. The results are shown in Fig. 2. The upper graph of Fig. 2 shows the angle scan of the fundamental radiation; the lower one shows the corresponding scan for the harmonics. It is obvious that the Bragg angles of the fundamental radiation at hν 5 and hν 6 keV perfectly match with the Bragg angles of the harmonics at θ 0.756° and θ 0.635° for hν 15 and 18 keV,
(2)
which the radiation has to pass through Ni. The attenuation by the top B4 C layers can be neglected. With an adequate number of periods for the top multilayer its reflectance for the fundamental wavelength and the transmission for the harmonic can be easily adjusted. For a photon energy of hν 15 keV and the corresponding glancing angle of θ 0.76° a total transmission through the Ni∕B4 C system of T 15 0.84 is calculated [7]. A similar calculation for a photon energy of hν 18 keV yields a total transmission of T 18 0.88. This transmission increases for harder x rays, being in the range of T ∼ 0.97 for the fifth harmonic at hν 30 keV. For the deposition of the two-color multilayer Bragg mirror we used a commercial dual ion beam sputter deposition (IBSD) machine (Roth & Rau, IonSys 1600) [8]. As main parts, the machine contains two linear electron cyclotron resonance (ECR) ion sources with beam dimensions of 400 mm × 100 mm, one for sputtering and the other one for film growth assistance. The ion sources can be operated at energies between 50 and 2000 eV enabling the deposition of smooth and dense films. Substrates are moved linearly across differently shaped slits to fulfill film thickness precision and homogeneity constraints, both typically on the order of 99.9%. Multilayer stacks are produced layer-by-layer by alternate sputtering of the materials needed. During the process up to six different targets can be used. Due to the low process pressure (∼10−4 mbar) and the rather small divergence of the depositing particle flux the IBSD technique allows the deposition of films with sharp edge boundaries.
Fig. 2. θ − 2θ angle scan of the two-color multilayer coating. For the harmonic radiation the reflectance calculated without passing the Ni∕B4 C system is depicted as dashed line. Theoretical calculations for the corresponding photon energies [7] are depicted as black lines.
2784
OPTICS LETTERS / Vol. 39, No. 9 / May 1, 2014
respectively. The reflectance for the fundamental radiation at hν 6 keV is measured to RF 0.78. The corresponding values for the third harmonics are RH 0.75 and RH 0.70 at hν 18 and hν 15 keV, respectively. Without passing the Ni∕B4 C stack the reflectivities of the lower stack are deduced to R15 RH ∕T 15 0.83 at hν 15 keV and R18 RH ∕T 18 0.85 at hν 18 keV, respectively, in accordance with theoretical expectations [7]. The 5 keV data displayed in Fig. 2 (orange line) had to be corrected for a contribution from the harmonic radiation, because in the present experimental setup the x-ray beam passes through 1.8 m air that attenuates the 5 keV intensity to a level comparable to that of the 15 keV radiation still present behind a Si(111) double crystal monochromator. For the measurement of the 6 keV radiation such corrections are not necessary, because the harmonic radiation at 18 keV does not significantly contribute to the signal. The angular widths ΔθH of the reflection curves of the harmonic radiation are determined to be ΔθH 228 μrad and ΔθH 193 μrad (FWHM) at hν 15 and hν 18 keV, respectively. The angular widths ΔθF of the reflection of the fundamental radiation are twelve times broader, with ΔθF 2.8 mrad and ΔθF 2.3 mrad at hν 5 and hν 6 keV, respectively. For the fundamental wavelength only the reflections from four bilayers constructively interfere when the Bragg condition is met, while for the harmonics the reflections from 120 bilayers contribute to the Bragg peak. By analogy with multiple slit interference, where the peak width scales with 1∕N and N are the number of slits, this leads to significantly narrower peaks for the reflection of the harmonic radiation. However, for a multilayer reflection the widths of the Bragg peaks do not scale strictly with 1∕N because the intensity reflected by one layer decreases for deeper layers. Thus their contribution to the Bragg peak also decreases. Another important aspect is the energetic width of the reflection. The relative width of the hard x-ray SASE pulses is typically on the order of ΔE∕E ∼ 0.1–0.5% [1,9], which corresponds to ΔE 18 eV–90 eV at hν 18 keV. The spectral widths of the multilayer reflections need to be broader for two reasons. First, a loss of pulse energy has to be avoided, and second the SDU is foreseen to be used for a measurement of the temporal coherence properties of the FEL [9,10]. For the latter, constraining the spectrum of the FEL radiation by the multilayers would yield overestimated values for the coherence time due to the Wiener–Khintchin theorem. The energetic width has been evaluated by aligning the mirror at the optimal angle θH 0.63° found in the previous measurement and then scanning the photon energy transmitted through the monochromator. The reflectance observed for photon energies in the vicinity of hν ∼ 18 keV is shown in Fig. 3. At a fixed angle of incidence of θH 0.63° the spectral width is measured to ΔE 280 eV, which is about 3–16 times broader than the spectrum that is expected for the XFEL pulses. According to the condition ΔE Δθ ≈ E tan θ
3
Fig. 3. Scan of the photon energy for a fixed angle of incidence of θH 0.63°.
that can be derived from the Bragg equation for multilayer mirrors [11] the measurement of the relative spectral width of ΔE∕E 0.016 appears to be in good agreement with the measurement of the relative angular width of Δθ∕ tanθ 0.017. For the experiments at the European XFEL conservation of the quality of the x-ray pulses is crucial. For instance, the influence of the x-ray optics on the spatial profile of the beam has to be minimized, which requires a homogeneous reflectance along the whole surface of the mirror. A scan along the mirror has been performed at hν 18 keV. The beam was confined by means of a 50 μm × 50 μm aperture yielding a footprint on the mirror of 4.6 mm × 50 μm. The reflectance at different positions of the mirror is shown in Fig. 4. The measurement shows a perfectly homogenous coating between l 75 mm and l 190 mm. For l 50 mm the reflectivity drops by 6%, which is still tolerable. The drop can be explained by a slightly varying multilayer period d which has been verified by Cu-Kα reflectometry. An important issue for all optics exposed to the intense x-ray beam of the European XFEL is the risk of single shot damage. The damage thresholds found by theoretical calculations [12] as well as experimental results [13] for different types of multilayer coatings are higher by more than two orders of magnitude than in the present case with expected fluences below 0.3 mJ∕cm2 .
Fig. 4. Reflectance as a function of the position along the mirror surface; hν 18 keV.
May 1, 2014 / Vol. 39, No. 9 / OPTICS LETTERS
In conclusion, the principle of the novel two-color hard x-ray multilayer coating has successfully been proven. With reflectivities higher than R 70% in all cases the new mirror will allow two-color hard x-ray pump/probe experiments with the SDU. Based on the two multilayer stacks chosen the full spectral range from ∼3 keV to greater than 27 keV can be covered with reflectivities higher than R 50%. Only at the Ni 1 s and Mo 1 s ionization thresholds at 8.33 and 20 keV, respectively, a decrease of the reflectance has to be encountered. Further optimization should allow the reflection of the 5th harmonic of the FEL together with the fundamental radiation.
4. 5. 6. 7. 8.
Funding by the Bundesministerium für Bildung und Forschung (BMBF-Projekt: 05K10PM2) is acknowledged.
9.
References 1. P. Emma, R. Akre, J. Arthur, R. Bionta, C. Bostedt, J. Bozek, A. Brachmann, P. Bucksbaum, R. Coffee, F.-J. Decker, Y. Ding, D. Dowell, S. Edstrom, A. Fisher, J. Frisch, S. Gilevich, J. Hastings, G. Hays, Ph. Hering, Z. Huang, R. Iverson, H. Loos, M. Messerschmidt, A. Miahnahri, S. Moeller, H.-D. Nuhn, G. Pile, D. Ratner, J. Rzepiela, D. Schultz, T. Smith, P. Stefan, H. Tompkins, J. Turner, J. Welch, W. White, J. Wu, G. Yocky, and J. Galayda, Nat. Photonics 4, 641 (2010). 2. D. Pile, Nat. Photonics 5, 456 (2011). 3. M. Altarelli, R. Brinkmann, M. Chergui, W. Decking, B. Dobson, S. Düsterer, G. Grübel, W. Graeff, H. Graafsma, J. Hajdu, J. Marangos, J. Pflüger, H. Redlin, D. Riley,
10. 11.
12. 13.
2785
I. Robinson, J. Rossbach, A. Schwarz, K. Tiedtke, T. Tschentscher, I. Vartaniants, H. Wabnitz, H. Weise, R. Wichmann, K. Witte, A. Wolf, M. Wulff, and M. Yurkov, “The European XFEL,” Technical Design Report (DESY Hamburg, 2006). S. Roling, L. Samoylova, B. Siemer, H. Sinn, F. Siewert, F. Wahlert, M. Wöstmann, and H. Zacharias, Proc. SPIE 8504, 850407 (2012). E. L. Saldin, E. A. Schneidmiller, and M. V. Yurkov, Phys. Rev. ST Accel. Beams 9, 030702 (2006). E. Spiller, Adv. X-Ray Anal. 42, 297 (1998). B. L. Henke, E. M. Gullikson, and J. C. Davis, At. Data Nucl. Data Tables 54, 181 (1993). P. Gawlitza, S. Braun, S. Lipfert, and A. Leson, Proc. SPIE 6317, 63170G (2006). G. Geloni, E. Saldin, L. Samoylova, E. Schneidmiller, H. Sinn, Th. Tschentscher, and M. Yurkov, New J. Phys. 12, 035021 (2010). S. Roling, B. Siemer, M. Wöstmann, H. Zacharias, R. Mitzner, A. Singer, K. Tiedtke, and I. A. Vartanyants, Phys. Rev. ST Accel. Beams 14, 080701 (2011). A. Kazimirov, D.-M. Smilgies, Q. Shen, X. Xiao, Q. Hao, E. Fontes, D. H. Bilderback, S. M. Gruner, Y. Platonov, and V. V. Martynov, J. Synchrotron Radiat. 13, 204 (2006). R. A. Loch, R. Sobierajski, E. Louis, J. Bosgra, and F. Bijkerk, Opt. Express 20, 28200 (2012). E. Louis, A. R. Khorsand, R. Sobierajski, E. D. van Hattum, M. Jurek, D. Klinger, J. B. Pelka, L. Juha, J. Chalupsky, J. Cihelka, V. Hajkova, U. Jastrow, S. Toleikis, H. Wabnitz, K. Tiedtke, J. Gaudin, E. M. Gullikson, and F. Bijkerk, Proc. SPIE 7361, 73610I (2013).
