- Email: [email protected]
Development of a compact polarization analysis apparatus for plasma soft X-ray laser
TSF-33145; No of Pages 4 Thin Solid Films xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Thin Solid Films journal homepage: www.elsevier.com/locate/tsf
Development of a compact polarization analysis apparatus for plasma soft X-ray laser Takashi Imazono, Masato Koike Quantum Beam Science Directorate, Japan Atomic Energy Agency (JAEA), 8-1-7, Umemidai, Kizugawa, Kyoto 619-0215, Japan
a r t i c l e
i n f o
Available online xxxx Keywords: Soft X-ray laser Multilayer Polarizer Ellipsometer
a b s t r a c t Laser-driven plasma soft X-ray laser, XRL, at a wavelength of 13.9 nm is generated from nickel-like silver plasmas. The polarization state at an end station is considered to be vertically linearly polarized due to the reflections at some Mo/Si multilayer mirrors installed in the XRL beamline, but the detail has not been verified experimentally. To evaluate and control the polarization state, a compact polarization analysis apparatus to adapt for the XRL end station is developed. Two Mo/Si multilayer mirrors are fabricated and the polarization properties are evaluated by using synchrotron radiation, SR. As a preliminary test of the apparatus, the reflectivity of the multilayer is measured by the XRL and it shows good agreement with that by SR. © 2014 Elsevier B.V. All rights reserved.
1. Introduction A laser-driven plasma soft X-ray laser, XRL, system has been constructed and operated at Kansai Photon Science Institute, KPSI, of JAEA [1]. The XRL is generated from nickel-like silver plasmas created by 10 J and several-ps-duration Nd:glass laser pulse and has a 7-ps-duration pulse at a wavelength of 13.9 nm with high spectral purity of b10−4. The XRL is helpful for speckle measurements and surface observations in the nano-scale [2–4]. The characteristics of XRL depend on the gradient of plasma density and the electron temperature of plasmas generated by Nd:glass laser pulses [5]. Therefore, it is important to obtain the information on the plasmas, which is associated with the polarization state. Hence, the polarization of plasma should be characterized by polarization analysis. It is of importance in light source development. Also, it is considered that the polarization state at an XRL end station changes due to the reflections from some Mo/Si multilayer mirrors at an angle of incidence of ~ 45° installed between the XRL source point and the end station for the sake of focusing and spatial management, because a Mo/Si multilayer mirror designed at around 45° acts as a good polarizer. The pseudo-Brewster angle is ~45°, resulting from the fact that the refractive index is almost unity in a soft X-ray region. However, the detail of the polarization state has not been evaluated since the construction of the XRL beamline. The information on the polarization state is helpful for calibrating experimental data. Therefore, it is an important subject to characterize the polarization state in polarization dependent measurements. We have developed a soft X-ray polarimeter and ellipsometer for complete polarization analysis, SXPE, for use in a synchrotron radiation, E-mail address: [email protected] (T. Imazono).
SR, facility [6–8]. The SXPE can be equipped with a pre-polarizer, P, and analyzer, A, and perform polarization measurements based on rotatinganalyzer ellipsometry by means of nine axes, i.e., the azimuth and incidence angles and height of P, the azimuth and incidence angles, and height of A, the arm angle connected to a rotating-analyzer unit, the detector arm angle, and a variable slit in just front of a detector in high vacuum. It is currently operated at a soft X-ray SR beamline, BL-11 [9], of the SR Center, Ritsumeikan University, Shiga, Japan [10]. Unfortunately, since it is somewhat large to install at an XRL end station of KPSI, a more compact polarization analysis apparatus than SXPE is required to evaluate and control the polarization state of the XRL. In this paper, we describe the design of the compact polarization analysis apparatus to adapt for an XRL end station along with the polarization characterization of Mo/Si multilayer mirrors using SR. Furthermore, a reflection measurement of the Mo/Si multilayer mirror using XRL is performed as a preliminary test of the apparatus showing good agreement with that measured by SR. 2. Design of a compact polarization analysis apparatus The SXPE at the BL-11 beamline consists of nine motorized stages. It is somewhat large to install at the XRL end station of KPSI, therefore a more compact polarization analysis apparatus than SXPE was designed. Fig. 1 shows a schematic diagram of polarization measurements based on rotating-analyzer ellipsometry using two polarization components, i.e., a pre-polarizer, P, and an analyzer, A (a), and a projection drawing of the main body of the polarization analysis apparatus (b). This apparatus allows us to perform polarization measurements, i.e., single-reflection and transmission measurements, double-reflection and transmission
http://dx.doi.org/10.1016/j.tsf.2014.02.011 0040-6090/© 2014 Elsevier B.V. All rights reserved.
Please cite this article as: T. Imazono, M. Koike, Thin Solid Films (2014), http://dx.doi.org/10.1016/j.tsf.2014.02.011
2
T. Imazono, M. Koike / Thin Solid Films xxx (2014) xxx–xxx
axis, L1. The reflected beam by P is irradiated to A at ω on the arm rotated by ψ, and then the intensity of the reflected beam from A is measured by a detector, D, on the detector arm rotated by θ while changing η along a rotation axis L4. The angles of φ and ψ, and ω and θ rotate along axes, L2 and L3, respectively. Also L4 is superposed on L1 when ψ = 0. The SXPE was equipped with the translation axes of polarization components and a variable slit in just front of a detector which were used for the adjustment of the angles of incidence in the vacuum. However these functions were not introduced in the instrument developed in this study because of miniaturization. Regardless of this, using 2-mm diameter pinholes denoted by PH1–PH4 in Fig. 1(b), the angles of incidence and height positions of P and A can be aligned at relatively high precision in an atmosphere. It is noted that the pinholes are removed during measurements. Also the size of a polarization component for P and A is limited up to 25 mm square × 5 mm thickness, which is about 1/2 times smaller than that of SXPE. Consequently, the instrument assembly shown in Fig. 1(b) was able to be installed in a vacuum chamber with a dimension of 400 × 400 × 400 mm3 at the XRL end station, so it was named cSXPE after a compact SXPE. In addition, a special silicon photodiode (AXUV100Si/Zi, IRD Inc., CA, U.S.A.) having a filter deposited directly on the surface for the reduction of stray light was employed as the detector, which was the same with the one used in SXPE.
(a)
(b)
3. Design and fabrication of Mo/Si multilayer polarizers
Table 1 Symbol, function, operation range, and resolution per pulse of each axis in cSXPE. Symbol
Function
Operation range
Resolution per pulse
χ φ ψ η ω θ
Azimuth angle of P Incidence angle of P Arm angle Azimuth angle of A Incidence angle of A Detector arm angle
−5–+100° −10–+90° −5–+120° 0–+360° −10–+90° −5–+125°
0.00125° 0.0003° 0.001° 0.00125° 0.0003° 0.0003°
1.0
1
Z Rs 10-1 0.5
Rp
10-2
10-3 0
15
30
45
60
75
Polarizance, Z
measurements, reflection–transmission and transmission–reflection measurements by means of six axes in a high vacuum: azimuth angle of P, χ; incidence angle of P, φ; azimuth angle of A, η; incidence angle of A, ω; arm angle, ψ; detector arm angle, θ. All axes can be controlled individually and the specifications are listed in Table 1. In the case of double-reflection measurement shown in Fig. 1(a), an XRL beam is incident to P at φ along the optical axis identical to χ indicated by a rotation
Reflectivity,Rs and Rp
Fig. 1. Schematic diagram of polarization measurements based on rotating-analyzer ellipsometry using a pre-polarizer, P, and an analyzer, A (a), and a projection drawing of the main body of the polarization analysis apparatus (b).
A reflection-type Mo/Si multilayer polarizer for 13.9 nm was designed by means of a layer-by-layer method [11]. Fig. 2 shows the calculated reflectivities for s-polarization, Rs, and p-polarization, Rp, and the polarizance, Z, defined as (Rs − Rp) / (Rs + Rp), of the Mo/Si multilayer as a function of the incident angle, assuming that the periodic length, D, the ratio, Γ, of a Mo layer thickness to D, the number of layers, N, and the topmost layer are 10.2 nm, 0.44, 47, and Mo, respectively. The value of Rs is 75% at the pseudo-Brewster angle of about 45° and two orders of magnitude higher than Rp. Because of Z N 99%, the multilayer is expected to work as the polarizer with high performance at 13.9 nm. In order to be used as a pre-polarizer, P, and analyzer, A, in the cSXPE, the designed Mo/Si multilayer mirror described above was fabricated on commercially available Si(100) wafers of 1 mm thickness with a root-mean-squire roughness of b 0.3 nm at ambient temperature by ion beam sputtering method. The base pressure was lower than 1 × 10− 5 Pa and the Ar gas pressure was 15 mPa during deposition. The periodic structures of the fabricated multilayers were examined by X-ray diffraction with Cu-Kα radiation. Fig. 3 shows small-angle X-ray diffraction profiles of two Mo/Si multilayer mirrors for P and A. For the comparison, the calculation curve is also shown. From the Bragg peak positions, D and Γ were determined to
0.0 90
Angle of incidence (deg) Fig. 2. Calculated Rs, Rp, and Z of the Mo/Si multilayer as a function of the incident angle.
Please cite this article as: T. Imazono, M. Koike, Thin Solid Films (2014), http://dx.doi.org/10.1016/j.tsf.2014.02.011
T. Imazono, M. Koike / Thin Solid Films xxx (2014) xxx–xxx
1
Measured (Polarizer) Measured (Analyzer) Calculated
(a) 1
Reflectivity
Reflectivity
1
1
0
1
2
3
4
5
6
10 -2
10 -3 35
The polarization properties of the two Mo/Si multilayer mirrors were evaluated by means of SXPE at the SR beamline BL-11. The Mo/Si multilayer mirrors were equipped in the SXPE as the pre-polarizer, P, and analyzer, A. The grazing incidence Monk-Gillieson monochromator with a laminar-type varied-line-spacing holographic plane grating, G1, of 300 lines/mm and a spherical mirror, M5, of the deviation angle of 172° was employed [9]. The scanning wavelength was set to 13.9 nm due to the XRL radiation at KPSI. The resolving power was estimated to be a few hundred when entrance and exit slit widths were set to 200 μm and 220 μm, respectively. A thin Si film of ~0.5 μm thickness between the exit slit and a refocusing mirror, M7, was used as a bandpass filter. The beam size at P in SXPE was estimated to be ~2 mm squared by zero-th order light from the monochromator. The BL-11 is a bending magnet beamline, in which it is known that the state of polarization strongly depends on the vertical acceptance angle. It has been confirmed to provide a horizontally linearly polarized light due to the small acceptance angle of the beamline [7]. The degree of linear polarization was determined together with the polarization characteristics of the Mo/Si multilayer mirrors in this polarization measurement. As a preliminary experiment with cSXPE installed at an end station of the XRL beamline of KPSI, a reflection measurement of the Mo/Si multilayer polarizer, which has been already characterized by SR, was carried out using XRLs generated by the single-target scheme. The XRL illuminates a concave mirror with a radius of curvature of 2000 mm at a normal incidence of ~3° and then three plane mirrors at ~45° deposited with Mo/Si multilayers successively before reaching the cSXPE chamber located at the end station. The state of polarization of XRL is considered to be vertically linearly polarized because of the reflections at three plane mirrors used at near the pseudo-Brewster angle. A beam size was estimated to be ~4 mm at D. 5. Results and discussion Fig. 4(a) shows measured reflectivities for s- and p-polarization components (Rs and Rp) of Mo/Si multilayer polarizer and analyzer as a function of the incidence angle. The values of Rs are 54.5% and 51.5% for the polarizer and analyzer, respectively, at the angle of incidence of 42.75°. The deviation of the reflectivity from the theoretical value of 75% is considered to result from the roughness of the multilayer. If the probe light is completely linearly polarized, Z is easily derived from the definition, but the degree of linear polarization, PL, will be revealed to be smaller than unity in the experiment described below. The azimuth angle dependencies of the intensities of the reflection lights from P, A, and A after reflected by P are shown in Fig. 4(b). The angles
45
50
(b)
55
P
Intensity (arb. unit)
4. Experiments
40
Angle of incidence (deg)
A
P&A Measured Calculated 0
90
180
270
360
Azimuth angle (deg) Fig. 4. Measured reflectivities of Mo/Si multilayer polarizer, P, and analyzer, A, as a function of the incident angle (a) and the azimuth angle dependencies of the intensities of the reflection lights from P, A, and A after reflected by P (b).
of incidence of both P and A were set at 42.75°, at which the maximum polarizance was recorded. The curve-fitting analysis results in Z = 99.3% for P, Z = 99.5% for A, PL = 93%, and δ = 0.23°, where the azimuth angle is measured counterclockwise from the horizontal plane by an observer facing the light. Fig. 5 shows the reflectivity for vertical polarization component, which corresponds to s-polarization, of the Mo/Si polarizer measured
0.8 Measured by SR Measured by XRL 0.6
Reflectivity
be 10.280 nm and 0.45 for P, and 10.328 nm and 0.46 for A, respectively, and they were verified to agree with the design values.
Polarizer Analyzer Rs Rs Rp Rp
10 -1
Diffraction angle (deg) Fig. 3. Small-angle X-ray diffraction profiles of the Mo/Si multilayer polarizer and analyzer.
3
0.4
0.2
0.0 35
40
45
50
55
Angle of incidence (deg) Fig. 5. Comparison of reflectivities for s-polarization component of Mo/Si multilayer polarizer between measured by XRL and SR.
Please cite this article as: T. Imazono, M. Koike, Thin Solid Films (2014), http://dx.doi.org/10.1016/j.tsf.2014.02.011
4
T. Imazono, M. Koike / Thin Solid Films xxx (2014) xxx–xxx
by using cSXPE installed at the XRL end station. It is noted that the reflection curve is in good agreement with the s-polarization reflectivity measured by the SR. Therefore, it indicates that the developed cSXPE instrument acts as an apparatus for polarization analysis and allows us to perform polarization measurements using XRL light sources. The degree of polarization of the XRL using cSXPE will be described in another paper.
6. Conclusion To evaluate and control the polarization state of the XRL generated from nickel-like silver plasmas excited by Nd:glass laser at KPSI, we developed a compact polarization analysis apparatus, cSXPE, based on rotating analyzer ellipsometry to adapt to the XRL end station. The cSXPE has six motion axes to control two polarizing elements and the detector. The polarization property of a Mo/Si multilayer polarizer and analyzer fabricated by ion beam sputtering method was characterized by using a polarization analysis apparatus, SXPE, at the soft X-ray SR beamline. It was clarified that the Mo/Si multilayer mirrors act as an analyzer with high performance at 13.9 nm as well as they clearly determined the degree of linear polarization of the SR light. In addition, in a preliminary experiment using cSXPE at the XRL end station, the reflectivity for s-polarization of the Mo/Si multilayer polarizer was found to be in good agreement with the result obtained by using the SR. Although the detail of this experiment using XRL would be described in another article, the experimental results show that the cSXPE is a useful apparatus for polarization analysis and control of XRL light sources.
Acknowledgments This work was partly supported by JSPS KAKENHI Grant Number 23760040. We thank our colleagues of KPSI for their experimental support. References [1] M. Nishikino, M. Tanaka, K. Nagashima, M. Kishimoto, M. Kado, T. Kawachi, K. Sukegawa, Y. Ochi, N. Hasegawa, Y. Kato, Phys. Rev. A 68 (2003) 061802. [2] R.Z. Tai, K. Namikawa, A. Sawada, M. Kishimoto, M. Tanaka, P. Lu, K. Nagashima, H. Maruyama, M. Ando, Phys. Rev. Lett. 93 (2004) 087601. [3] S. Namba, N. Hasegawa, M. Nishikino, T. Kawachi, M. Kishimoto, K. Sukegawa, M. Tanaka, Y. Ochi, K. Takiyama, K. Nagashima, Phys. Rev. Lett. 99 (2007) 043004. [4] T. Suemoto, K. Terakawa, Y. Ochi, T. Tomita, M. Yamamoto, N. Hasegawa, M. Deki, Y. Minami, T. Kawachi, Opt. Express 13 (2010) 14114. [5] T. Kawachi, K. Murai, G. Yuan, S. Ninomiya, R. Kodama, H. Daido, Y. Kato, T. Fujimoto, Phys. Rev. Lett. 75 (1995) 3826. [6] T. Imazono, K. Sano, Y. Suzuki, T. Kawachi, M. Koike, Rev. Sci. Instrum. 80 (2009) 085109. [7] T. Imazono, K. Sano, Y. Suzuki, T. Kawachi, M. Koike, in: R. Garrett, I. Gentle, K. Nugent, S. Wilkins (Eds.), 10th International Conference on Synchrotron Radiation Instrumentation, Melbourne, Australia, September 27–October2, 2009, AIP Conference Proceedings 1234, 2010, p. 347. [8] T. Imazono, K. Sano, M. Koike, in: S.V. Bulanov, A. Yokoyama, Y.I. Malakhov, Y. Watanabe (Eds.), Laser-Driven Relativistic Plasmas Applied to Science, Energy, Industry, and Medicine: The 3rd International Symposium, Kyoto, Japan, 30 May–2 June, 2011, AIP Conference Proceedings 1465, 2012, p. 28. [9] M. Koike, K. Sano, O. Yoda, Y. Harada, M. Ishino, N. Moriya, H. Sasai, H. Takenaka, E. Gullikson, S. Mrowka, M. Jinno, Y. Ueno, J.H. Underwood, T. Namioka, Rev. Sci. Instrum. 73 (2002) 1541. [10] H. Iwasaki, Y. Nakayama, K. Ozutsumi, Y. Yamamoto, Y. Tokunaga, H. Saisho, T. Matsubara, S. Ikeda, J. Synchrotron Radiat. 5 (1998) 1162. [11] M. Yamamoto, T. Namioka, Appl. Opt. 31 (1992) 1622.
Please cite this article as: T. Imazono, M. Koike, Thin Solid Films (2014), http://dx.doi.org/10.1016/j.tsf.2014.02.011
Contents lists available at ScienceDirect
Thin Solid Films journal homepage: www.elsevier.com/locate/tsf
Development of a compact polarization analysis apparatus for plasma soft X-ray laser Takashi Imazono, Masato Koike Quantum Beam Science Directorate, Japan Atomic Energy Agency (JAEA), 8-1-7, Umemidai, Kizugawa, Kyoto 619-0215, Japan
a r t i c l e
i n f o
Available online xxxx Keywords: Soft X-ray laser Multilayer Polarizer Ellipsometer
a b s t r a c t Laser-driven plasma soft X-ray laser, XRL, at a wavelength of 13.9 nm is generated from nickel-like silver plasmas. The polarization state at an end station is considered to be vertically linearly polarized due to the reflections at some Mo/Si multilayer mirrors installed in the XRL beamline, but the detail has not been verified experimentally. To evaluate and control the polarization state, a compact polarization analysis apparatus to adapt for the XRL end station is developed. Two Mo/Si multilayer mirrors are fabricated and the polarization properties are evaluated by using synchrotron radiation, SR. As a preliminary test of the apparatus, the reflectivity of the multilayer is measured by the XRL and it shows good agreement with that by SR. © 2014 Elsevier B.V. All rights reserved.
1. Introduction A laser-driven plasma soft X-ray laser, XRL, system has been constructed and operated at Kansai Photon Science Institute, KPSI, of JAEA [1]. The XRL is generated from nickel-like silver plasmas created by 10 J and several-ps-duration Nd:glass laser pulse and has a 7-ps-duration pulse at a wavelength of 13.9 nm with high spectral purity of b10−4. The XRL is helpful for speckle measurements and surface observations in the nano-scale [2–4]. The characteristics of XRL depend on the gradient of plasma density and the electron temperature of plasmas generated by Nd:glass laser pulses [5]. Therefore, it is important to obtain the information on the plasmas, which is associated with the polarization state. Hence, the polarization of plasma should be characterized by polarization analysis. It is of importance in light source development. Also, it is considered that the polarization state at an XRL end station changes due to the reflections from some Mo/Si multilayer mirrors at an angle of incidence of ~ 45° installed between the XRL source point and the end station for the sake of focusing and spatial management, because a Mo/Si multilayer mirror designed at around 45° acts as a good polarizer. The pseudo-Brewster angle is ~45°, resulting from the fact that the refractive index is almost unity in a soft X-ray region. However, the detail of the polarization state has not been evaluated since the construction of the XRL beamline. The information on the polarization state is helpful for calibrating experimental data. Therefore, it is an important subject to characterize the polarization state in polarization dependent measurements. We have developed a soft X-ray polarimeter and ellipsometer for complete polarization analysis, SXPE, for use in a synchrotron radiation, E-mail address: [email protected] (T. Imazono).
SR, facility [6–8]. The SXPE can be equipped with a pre-polarizer, P, and analyzer, A, and perform polarization measurements based on rotatinganalyzer ellipsometry by means of nine axes, i.e., the azimuth and incidence angles and height of P, the azimuth and incidence angles, and height of A, the arm angle connected to a rotating-analyzer unit, the detector arm angle, and a variable slit in just front of a detector in high vacuum. It is currently operated at a soft X-ray SR beamline, BL-11 [9], of the SR Center, Ritsumeikan University, Shiga, Japan [10]. Unfortunately, since it is somewhat large to install at an XRL end station of KPSI, a more compact polarization analysis apparatus than SXPE is required to evaluate and control the polarization state of the XRL. In this paper, we describe the design of the compact polarization analysis apparatus to adapt for an XRL end station along with the polarization characterization of Mo/Si multilayer mirrors using SR. Furthermore, a reflection measurement of the Mo/Si multilayer mirror using XRL is performed as a preliminary test of the apparatus showing good agreement with that measured by SR. 2. Design of a compact polarization analysis apparatus The SXPE at the BL-11 beamline consists of nine motorized stages. It is somewhat large to install at the XRL end station of KPSI, therefore a more compact polarization analysis apparatus than SXPE was designed. Fig. 1 shows a schematic diagram of polarization measurements based on rotating-analyzer ellipsometry using two polarization components, i.e., a pre-polarizer, P, and an analyzer, A (a), and a projection drawing of the main body of the polarization analysis apparatus (b). This apparatus allows us to perform polarization measurements, i.e., single-reflection and transmission measurements, double-reflection and transmission
http://dx.doi.org/10.1016/j.tsf.2014.02.011 0040-6090/© 2014 Elsevier B.V. All rights reserved.
Please cite this article as: T. Imazono, M. Koike, Thin Solid Films (2014), http://dx.doi.org/10.1016/j.tsf.2014.02.011
2
T. Imazono, M. Koike / Thin Solid Films xxx (2014) xxx–xxx
axis, L1. The reflected beam by P is irradiated to A at ω on the arm rotated by ψ, and then the intensity of the reflected beam from A is measured by a detector, D, on the detector arm rotated by θ while changing η along a rotation axis L4. The angles of φ and ψ, and ω and θ rotate along axes, L2 and L3, respectively. Also L4 is superposed on L1 when ψ = 0. The SXPE was equipped with the translation axes of polarization components and a variable slit in just front of a detector which were used for the adjustment of the angles of incidence in the vacuum. However these functions were not introduced in the instrument developed in this study because of miniaturization. Regardless of this, using 2-mm diameter pinholes denoted by PH1–PH4 in Fig. 1(b), the angles of incidence and height positions of P and A can be aligned at relatively high precision in an atmosphere. It is noted that the pinholes are removed during measurements. Also the size of a polarization component for P and A is limited up to 25 mm square × 5 mm thickness, which is about 1/2 times smaller than that of SXPE. Consequently, the instrument assembly shown in Fig. 1(b) was able to be installed in a vacuum chamber with a dimension of 400 × 400 × 400 mm3 at the XRL end station, so it was named cSXPE after a compact SXPE. In addition, a special silicon photodiode (AXUV100Si/Zi, IRD Inc., CA, U.S.A.) having a filter deposited directly on the surface for the reduction of stray light was employed as the detector, which was the same with the one used in SXPE.
(a)
(b)
3. Design and fabrication of Mo/Si multilayer polarizers
Table 1 Symbol, function, operation range, and resolution per pulse of each axis in cSXPE. Symbol
Function
Operation range
Resolution per pulse
χ φ ψ η ω θ
Azimuth angle of P Incidence angle of P Arm angle Azimuth angle of A Incidence angle of A Detector arm angle
−5–+100° −10–+90° −5–+120° 0–+360° −10–+90° −5–+125°
0.00125° 0.0003° 0.001° 0.00125° 0.0003° 0.0003°
1.0
1
Z Rs 10-1 0.5
Rp
10-2
10-3 0
15
30
45
60
75
Polarizance, Z
measurements, reflection–transmission and transmission–reflection measurements by means of six axes in a high vacuum: azimuth angle of P, χ; incidence angle of P, φ; azimuth angle of A, η; incidence angle of A, ω; arm angle, ψ; detector arm angle, θ. All axes can be controlled individually and the specifications are listed in Table 1. In the case of double-reflection measurement shown in Fig. 1(a), an XRL beam is incident to P at φ along the optical axis identical to χ indicated by a rotation
Reflectivity,Rs and Rp
Fig. 1. Schematic diagram of polarization measurements based on rotating-analyzer ellipsometry using a pre-polarizer, P, and an analyzer, A (a), and a projection drawing of the main body of the polarization analysis apparatus (b).
A reflection-type Mo/Si multilayer polarizer for 13.9 nm was designed by means of a layer-by-layer method [11]. Fig. 2 shows the calculated reflectivities for s-polarization, Rs, and p-polarization, Rp, and the polarizance, Z, defined as (Rs − Rp) / (Rs + Rp), of the Mo/Si multilayer as a function of the incident angle, assuming that the periodic length, D, the ratio, Γ, of a Mo layer thickness to D, the number of layers, N, and the topmost layer are 10.2 nm, 0.44, 47, and Mo, respectively. The value of Rs is 75% at the pseudo-Brewster angle of about 45° and two orders of magnitude higher than Rp. Because of Z N 99%, the multilayer is expected to work as the polarizer with high performance at 13.9 nm. In order to be used as a pre-polarizer, P, and analyzer, A, in the cSXPE, the designed Mo/Si multilayer mirror described above was fabricated on commercially available Si(100) wafers of 1 mm thickness with a root-mean-squire roughness of b 0.3 nm at ambient temperature by ion beam sputtering method. The base pressure was lower than 1 × 10− 5 Pa and the Ar gas pressure was 15 mPa during deposition. The periodic structures of the fabricated multilayers were examined by X-ray diffraction with Cu-Kα radiation. Fig. 3 shows small-angle X-ray diffraction profiles of two Mo/Si multilayer mirrors for P and A. For the comparison, the calculation curve is also shown. From the Bragg peak positions, D and Γ were determined to
0.0 90
Angle of incidence (deg) Fig. 2. Calculated Rs, Rp, and Z of the Mo/Si multilayer as a function of the incident angle.
Please cite this article as: T. Imazono, M. Koike, Thin Solid Films (2014), http://dx.doi.org/10.1016/j.tsf.2014.02.011
T. Imazono, M. Koike / Thin Solid Films xxx (2014) xxx–xxx
1
Measured (Polarizer) Measured (Analyzer) Calculated
(a) 1
Reflectivity
Reflectivity
1
1
0
1
2
3
4
5
6
10 -2
10 -3 35
The polarization properties of the two Mo/Si multilayer mirrors were evaluated by means of SXPE at the SR beamline BL-11. The Mo/Si multilayer mirrors were equipped in the SXPE as the pre-polarizer, P, and analyzer, A. The grazing incidence Monk-Gillieson monochromator with a laminar-type varied-line-spacing holographic plane grating, G1, of 300 lines/mm and a spherical mirror, M5, of the deviation angle of 172° was employed [9]. The scanning wavelength was set to 13.9 nm due to the XRL radiation at KPSI. The resolving power was estimated to be a few hundred when entrance and exit slit widths were set to 200 μm and 220 μm, respectively. A thin Si film of ~0.5 μm thickness between the exit slit and a refocusing mirror, M7, was used as a bandpass filter. The beam size at P in SXPE was estimated to be ~2 mm squared by zero-th order light from the monochromator. The BL-11 is a bending magnet beamline, in which it is known that the state of polarization strongly depends on the vertical acceptance angle. It has been confirmed to provide a horizontally linearly polarized light due to the small acceptance angle of the beamline [7]. The degree of linear polarization was determined together with the polarization characteristics of the Mo/Si multilayer mirrors in this polarization measurement. As a preliminary experiment with cSXPE installed at an end station of the XRL beamline of KPSI, a reflection measurement of the Mo/Si multilayer polarizer, which has been already characterized by SR, was carried out using XRLs generated by the single-target scheme. The XRL illuminates a concave mirror with a radius of curvature of 2000 mm at a normal incidence of ~3° and then three plane mirrors at ~45° deposited with Mo/Si multilayers successively before reaching the cSXPE chamber located at the end station. The state of polarization of XRL is considered to be vertically linearly polarized because of the reflections at three plane mirrors used at near the pseudo-Brewster angle. A beam size was estimated to be ~4 mm at D. 5. Results and discussion Fig. 4(a) shows measured reflectivities for s- and p-polarization components (Rs and Rp) of Mo/Si multilayer polarizer and analyzer as a function of the incidence angle. The values of Rs are 54.5% and 51.5% for the polarizer and analyzer, respectively, at the angle of incidence of 42.75°. The deviation of the reflectivity from the theoretical value of 75% is considered to result from the roughness of the multilayer. If the probe light is completely linearly polarized, Z is easily derived from the definition, but the degree of linear polarization, PL, will be revealed to be smaller than unity in the experiment described below. The azimuth angle dependencies of the intensities of the reflection lights from P, A, and A after reflected by P are shown in Fig. 4(b). The angles
45
50
(b)
55
P
Intensity (arb. unit)
4. Experiments
40
Angle of incidence (deg)
A
P&A Measured Calculated 0
90
180
270
360
Azimuth angle (deg) Fig. 4. Measured reflectivities of Mo/Si multilayer polarizer, P, and analyzer, A, as a function of the incident angle (a) and the azimuth angle dependencies of the intensities of the reflection lights from P, A, and A after reflected by P (b).
of incidence of both P and A were set at 42.75°, at which the maximum polarizance was recorded. The curve-fitting analysis results in Z = 99.3% for P, Z = 99.5% for A, PL = 93%, and δ = 0.23°, where the azimuth angle is measured counterclockwise from the horizontal plane by an observer facing the light. Fig. 5 shows the reflectivity for vertical polarization component, which corresponds to s-polarization, of the Mo/Si polarizer measured
0.8 Measured by SR Measured by XRL 0.6
Reflectivity
be 10.280 nm and 0.45 for P, and 10.328 nm and 0.46 for A, respectively, and they were verified to agree with the design values.
Polarizer Analyzer Rs Rs Rp Rp
10 -1
Diffraction angle (deg) Fig. 3. Small-angle X-ray diffraction profiles of the Mo/Si multilayer polarizer and analyzer.
3
0.4
0.2
0.0 35
40
45
50
55
Angle of incidence (deg) Fig. 5. Comparison of reflectivities for s-polarization component of Mo/Si multilayer polarizer between measured by XRL and SR.
Please cite this article as: T. Imazono, M. Koike, Thin Solid Films (2014), http://dx.doi.org/10.1016/j.tsf.2014.02.011
4
T. Imazono, M. Koike / Thin Solid Films xxx (2014) xxx–xxx
by using cSXPE installed at the XRL end station. It is noted that the reflection curve is in good agreement with the s-polarization reflectivity measured by the SR. Therefore, it indicates that the developed cSXPE instrument acts as an apparatus for polarization analysis and allows us to perform polarization measurements using XRL light sources. The degree of polarization of the XRL using cSXPE will be described in another paper.
6. Conclusion To evaluate and control the polarization state of the XRL generated from nickel-like silver plasmas excited by Nd:glass laser at KPSI, we developed a compact polarization analysis apparatus, cSXPE, based on rotating analyzer ellipsometry to adapt to the XRL end station. The cSXPE has six motion axes to control two polarizing elements and the detector. The polarization property of a Mo/Si multilayer polarizer and analyzer fabricated by ion beam sputtering method was characterized by using a polarization analysis apparatus, SXPE, at the soft X-ray SR beamline. It was clarified that the Mo/Si multilayer mirrors act as an analyzer with high performance at 13.9 nm as well as they clearly determined the degree of linear polarization of the SR light. In addition, in a preliminary experiment using cSXPE at the XRL end station, the reflectivity for s-polarization of the Mo/Si multilayer polarizer was found to be in good agreement with the result obtained by using the SR. Although the detail of this experiment using XRL would be described in another article, the experimental results show that the cSXPE is a useful apparatus for polarization analysis and control of XRL light sources.
Acknowledgments This work was partly supported by JSPS KAKENHI Grant Number 23760040. We thank our colleagues of KPSI for their experimental support. References [1] M. Nishikino, M. Tanaka, K. Nagashima, M. Kishimoto, M. Kado, T. Kawachi, K. Sukegawa, Y. Ochi, N. Hasegawa, Y. Kato, Phys. Rev. A 68 (2003) 061802. [2] R.Z. Tai, K. Namikawa, A. Sawada, M. Kishimoto, M. Tanaka, P. Lu, K. Nagashima, H. Maruyama, M. Ando, Phys. Rev. Lett. 93 (2004) 087601. [3] S. Namba, N. Hasegawa, M. Nishikino, T. Kawachi, M. Kishimoto, K. Sukegawa, M. Tanaka, Y. Ochi, K. Takiyama, K. Nagashima, Phys. Rev. Lett. 99 (2007) 043004. [4] T. Suemoto, K. Terakawa, Y. Ochi, T. Tomita, M. Yamamoto, N. Hasegawa, M. Deki, Y. Minami, T. Kawachi, Opt. Express 13 (2010) 14114. [5] T. Kawachi, K. Murai, G. Yuan, S. Ninomiya, R. Kodama, H. Daido, Y. Kato, T. Fujimoto, Phys. Rev. Lett. 75 (1995) 3826. [6] T. Imazono, K. Sano, Y. Suzuki, T. Kawachi, M. Koike, Rev. Sci. Instrum. 80 (2009) 085109. [7] T. Imazono, K. Sano, Y. Suzuki, T. Kawachi, M. Koike, in: R. Garrett, I. Gentle, K. Nugent, S. Wilkins (Eds.), 10th International Conference on Synchrotron Radiation Instrumentation, Melbourne, Australia, September 27–October2, 2009, AIP Conference Proceedings 1234, 2010, p. 347. [8] T. Imazono, K. Sano, M. Koike, in: S.V. Bulanov, A. Yokoyama, Y.I. Malakhov, Y. Watanabe (Eds.), Laser-Driven Relativistic Plasmas Applied to Science, Energy, Industry, and Medicine: The 3rd International Symposium, Kyoto, Japan, 30 May–2 June, 2011, AIP Conference Proceedings 1465, 2012, p. 28. [9] M. Koike, K. Sano, O. Yoda, Y. Harada, M. Ishino, N. Moriya, H. Sasai, H. Takenaka, E. Gullikson, S. Mrowka, M. Jinno, Y. Ueno, J.H. Underwood, T. Namioka, Rev. Sci. Instrum. 73 (2002) 1541. [10] H. Iwasaki, Y. Nakayama, K. Ozutsumi, Y. Yamamoto, Y. Tokunaga, H. Saisho, T. Matsubara, S. Ikeda, J. Synchrotron Radiat. 5 (1998) 1162. [11] M. Yamamoto, T. Namioka, Appl. Opt. 31 (1992) 1622.
Please cite this article as: T. Imazono, M. Koike, Thin Solid Films (2014), http://dx.doi.org/10.1016/j.tsf.2014.02.011