\u003ctitle\u003eSource development and novel applications of x-ray lasers for probing materials\u003c/title\u003e

Source Development and Novel Applications of X-ray Lasers for Probing Materials

T. Kawachi*a, M. Kishimotoa, M. Kadoa, Y. Ochia, N. Hasegawaa, M. Tanakaa, M. Nishikinoa,

M. Ishinoa, T. Imazonoa, T. Ohbaa, T. Kaihoria, M. Koikea, K. Namikawab, T. Suemotoc, K. Terakawac, T. Tomitad, N. Sarukurae, H. Nishimurae, A. Y. Faenova, S. Bulanova and H. Daidoa

aQuantum Beam Science Directorate, Japan Atomic Energy Agency (JAEA), 8-1 Umemidai, Kizugawa, Kyoto, 619-0216, Japan

bDept. of Physics, Tokyo Gakugei University, 4-1-1, Koganeishi, Tokyo, 184-8501, Japan, cInstitute of Solid State Physics (ISSP), University of Tokyo, 5-1-5, Kashinoha,

Kashiwa, Chiba, 277-8581, Japan, dDept. of Ecosystem Engineering, The university of Tokushima, Tokushima, 770-8506, Japan,

eInstitute of Laser Engineering (ILE), 2-6, Yamadaoka, Suita, Osaka, 565-0871, Japan

ABSTRACT

This paper gives an overview of recent progress of laser-driven plasma x-ray lasers in Japan Atomic Energy Agency (JAEA). Fully spatial coherent plasma x-ray laser (XRL) at 13.9 nm with 0.1 Hz repetition rate has been developed using new driver laser system TOPAZ, and the succeeding optimization of the pumping condition has realized more efficient generation of the coherent x-ray pulse. The 0.1 Hz XRL is now routinely used in the wide variety of the application experiments: The highlights of these applications are the study of fluctuation in the atomic structure of ferroelectric substances under the phase transition using the double XRL probe beam technique and the construction of new x-ray laser interferometer to observe nano-scale dynamics of materials.

Keywords: x-ray lasers, laser produced plasma, x-ray interferometer, x-ray speckle

1. INTRODUCTION The Advent of transient collisional excitation (TCE) laser makes it possible for us to realize small size coherent soft x-ray lasers [1-5]. In Japan Atomic Energy Agency (JAEA), we have firstly demonstrated fully spatial coherent x-ray laser beam at the wavelength of 13.9 nm by the method of double target geometry, in which the first gain medium works as the soft x-ray oscillator and the second gain medium works as soft x-ray amplifier [6, 7]. Successive optimization of the pumping condition such as the pumping intensity, traveling wave, temporal separation of the pre- and main-pulses allows us to obtain the high quality, intense x-ray laser beam with the typical parameters of beam divergence of better than 1 mrad, 1 µJ output energy and more than 109 photons in the coherent volume, where the coherent volume is determined by the spatial coherent length and temporal coherent length [8]. In a view point of the applications of the soft x-ray lasers, one of the most serious limitations so far is the repetition-rate. Typical shot interval of CPA Nd:glass driver is ~ 10 min and prohibits many potential users from conducting application experiments. The demonstration of the substantial amplification using GRIP (grazing incidence pumping) scheme induces breakthrough in this problem. The spatially coherent x-ray laser of the output energy of several tens nJ has been obtained under the repetition-rate of 5 Hz by the combination of GRIP and higher-order harmonics light as the x-ray seeder [9]. However, while this high average power soft x-ray laser is powerful tool for the soft x-ray imaging (nano-fabrication) by the multiple-shot exposure, there is strong interest and requirement from material science for the soft x-ray lasers with larger output energy or larger coherent photons per pulse: For an example, in the material science, nano-meter scale deformation under the phase transition of substances is essentially unrepeatable phenomena, and

Invited Paper

Soft X-Ray Lasers and Applications VIII, edited by James Dunn, Gregory J. Tallents, Proc. of SPIE Vol. 7451, 745107 · © 2009 SPIE · CCC code: 0277-786X/09/$18 · doi: 10.1117/12.825662

Proc. of SPIE Vol. 7451 745107-1

single-shot soft x-ray laser probe is indispensable to realize the correlation function of such the phenomena in space and time. In 2004. we have started to develop new driver laser system dedicated for the x-ray lasers. The new driver system, TOPAZ, is based upon CPA Nd:glass laser, and the oscillator and the CPA system are the same with our previous driver system [10]. The highlight of the new system is the employment of zigzag slab-type amplifier chain, which allows us to operate this system with 0.1 Hz repetition-rate due to the noticeable cooling efficiency of the zigzag slab amplifiers. In 2007, the proto-type of TOPAZ with one output beam has constructed, and intense 13.9 nm laser has been demonstrated [11]. The construction of the second beam line has been finished in 2008, and quite recently fully spatial coherent 13.9 nm laser has been obtained by TOPAZ. Now it is routinely used for the generation of the x-ray lasers and the application researches with adding minor revisions to improve the performance. Application experiments using the fully spatial coherent 13.9 nm covers the research fields of material science [12-14] and atomic and molecular physics [15] so far, and in order to extend the research area to the study of nano-scale dunamics of substances, we have started to construct new x-ray laser interferometer.

2. IMPROVEMENT OF THE PERFORMANCE OF THE TCE LASERS

2.1. TOPAZ; New Driver Laser System with 0.1 Hz repetition-rate New driver laser system, TOPAZ stands for Two OPtical Amplifiers by Zigzag slab. Figure 1 shows the schematic diagram of the optical components of TOPAZ. TOPAZ consists of the oscillator, pulse stretcher, OPCPA preamplifier, prepulse generator, zigzag slab Nd:glass power amplifiers, pulse compressor, and optics for producing a line focus on the target. The oscillator is a mode-locked Ti:sapphire laser pumped by 10-W diode-pumped solid state laser. The centre wavelength is 1053 nm and the spectral bandwidth is 20 nm (FWHM, full width at the half maximum). The frequency of the oscillator is 80 MHz with the typical power of 300 mW (~ 4 nJ / pulse). The 4-path pulse stretcher, which consists of a diffraction grating with 1740 grooves/mm (Jobin Yvon), a spherical mirror with focal length of 1500 mm, flat mirror, and steering mirrors, generates frequency chip of 250 ps/nm with the spectral bandwidth of 8 nm. Typical output power is 100 mW (~ 1 nJ/pulse). The stretched pulse is amplified by the 1st amplifier based on the optical parametric chirped pulse amplification (OPCPA). The seed light and the 532 nm YAG pumping laser with intensity of ~100 MW/cm2 are incident on BBO crystals coaxially. Totally four BBO crystal are used, and more than 106 gain is obtained. The output energy is typically ~ 10 mJ with keeping the bandwidth of 8 nm. The contrast ratio of the amplified laser pulse to the background is better then 105. After the OPCPA, the laser pulse is divided into two beams. Each laser beam goes through the pre-pulse generator, which includes Michelson-type double pulse generator together with an additional stretcher. The details of this system is described in [10]. After the double pulse generator, we have two double-pulse laser beams, and the delay time between the two laser beam and the pulse separation between the pre-pulse and main pulse in each beam are variable independently. The output beams, whose diameters are around 25 mm, are cut by the serrated apertures with 10 x 10 mm2 to obtain virtually flat-top square-shaped beams, which are amplified by the following amplifier chains. The main power amplifier chain consists of two Nd:glass zigzag slab amplifiers. The laser glass used in the amplifier chain is a silica-phosphate with 1.0 wt.% Nd concentration. The size of the 1st amplifier is 15 x 17 mm2 cross-section and 99 mm length, and the second is 110 x 17 mm2 cross-section and 205 mm length. Both the incident faces are tapered by 1 degree and are covered by antireflection coating. The slabs are installed in the stain-less steel housings with water-cooling system.

Proc. of SPIE Vol. 7451 745107-2

The first zigzag slab amplifier is installed in a cavity with the cavity length of 6 m. The schematic is shown in Fig. 2. The image of the laser pulse at the serrated aperture (A) is transferred to one of the end mirror of the cavity (B), and this image transferred again to the other end mirror (C) by use of a pair of lens with the focal length of 1500 mm. The laser light with the beam size of 10 x 10 mm2 is incident to the first zigzag slab with the angle of incidence of 28 degrees with respect to surface normal, reflects two times by the side wall of the slab and exits from the opposite side. The laser glass is pumped by 8 xenon flash lamps (4 lamps on each side) from the side faces. After 12-pass amplification (6 round trips) in the cavity, the laser light is amplified up to ~ 1 J and is extracted by means of polarization control using a pockels cell and a polarizer. The output laser beam is expanded into the rectangular shape of 90 x 10 mm2 by use of vertical and horizontal cylindrical lens pair and is amplified by the following second zigzag slab amplifier. The second zigzag slab amplifier is pumped by 24 xenon flash lamps (12 lamps on each side). The stored energy density is estimated to be 0.3 J/cm2 from the measured small signal gain. The angle of incidence is the same with that of the first zigzag slab. The reflection inside of the slab is 4 times. We can obtain the output energy of more than 10 J by 3 pass-amplification. The rectangular beam is expanded into the square shape with 100 x 100 mm2 in order to avoid the damage of the gratings in the pulse compressor. The pulse compressor is based upon two large-size gratings with the diameter of 420 mm (Jobin Yvon) aligned to be parallel each other with the distance of 970 mm. The details of the pulse compressor is described in [10]. The throughput in the energy is around 60%, and the typical duration is a few ps which is suitable

for the transient collisional excitation scheme. The compressed light is linearly focused by the combination of the off-axis parabolic and off axis spherical mirrors at the target position.

Nd:Glass Zigzag Slab Amplifier

Tape Target

Oscillator

Pulse Stretcher

OPCPA or Regen.

Pulse Compressor

Line Focusing system

Prepulse Generation

TO

PAZ

80 MHz,~ 4 nJ

80 MHz,~1 nJ / 2 ns

10 Hz,~ mJ / 2 ns

0.1 Hz, ~ 15 J / 2 ns

X-ray laser @0.1Hz

0.1 Hz, ~ 10 J / ~ ps

Nd:Glass Zigzag Slab Amplifier

Tape Target

Oscillator

Pulse Stretcher

OPCPA or Regen.

Pulse Compressor

Line Focusing system

Prepulse Generation

TO

PAZ

80 MHz,~ 4 nJ

80 MHz,~1 nJ / 2 ns

10 Hz,~ mJ / 2 ns

0.1 Hz, ~ 15 J / 2 ns

X-ray laser @0.1Hz

0.1 Hz, ~ 10 J / ~ ps

Fig.1. Schmatic diagram of new driver laser, TOPAZ

Fig 2. Scematic diagram of the main amplifier chain. A: serrated aperture with 10 x 10 mm2. The image of the laser beam at the position of A is transferred by the image relay telescope to B, C (the end mirrors of the cavity of the 1st zigzag slab amplifier.), D and E. The beam size is 10 x 10 mm2 at the position from A through D and is 10 x 90 mm2 at E.

Proc. of SPIE Vol. 7451 745107-3

The silver target is supplied by brand-new tape target system. 50 µm or 30 µm-thick, 15 mm-width silver tape is set at the position of the focus, and two DC motors coupled with planetary gear mechanism pull the silver tape in the opposite direction each other to keep the flatness of the tape. The details of the mechanism of the tape target system will published in elsewhere. We have assured that the parameters of the 13.9 nm x-ray laser, the output energy, the beam divergence and the pointing stability, are almost comparable with the case of the silver target deposited on the glass plate, and no other disadvantages can been found. By installing this tape target system, more than 103 continuous shots become possible without breaking the vacuum of the target chamber. By use of TOPAZ laser, we have already demonstrated nickel-like silver laser at a wavelength of 13.9 nm. The two targets were set with the separation of 200 mm, and each target was irradiated by the double-pulse laser beam from TOPAZ. In this experiment, the pulse separation between the pre-pulse and the main pulse was 1.1 ns, and the energy ratio was 1 : 3.3. The duration of both the pre-pulse and the main pulse was 5 ps, and the total pumping energy was 6-7 J. The x-ray laser beam was monitored by use of a 45° incidence Mo/Si multi-layer mirror coupled with a back-illuminated charge coupled device (CCD). A 2 µm-thick zirconium filter, which had the transmittance of 0.05 % for the 13.9nm line, was put in front of the Mo/Si mirror to cut the visible-UV light. Figure 3 shows the footprints of the x-ray laser from the 1st target (left), the 2nd target (middle) and the double targets (right). The beam divergence of the x-ray lasers from single target was 10-20 mrad in the vertical direction, whereas by combining the 1st and the 2nd targets, quite directional beam was obtained. The output energy reached ~ µJ, and the performance of the x-ray laser beam pumped by TOPAZ was enough to start the application experiments. It is noted that the position of the strong portion in the footprint of the double targets x-ray laser is different from that for the 2nd target (see middle of Fig. 3.). This indicates that the intense part of the x-ray laser from the 2nd target is bent by the refraction effect due to the density gradient in the plasma, and the seed x-ray from the 1st target is amplified in the region with calm density gradient but small gain coefficient. This implies that we can stand further improvement to increase the output energy of the double target x-ray laser by optimizing the spatial density and gain profile of the second plasma.

2.2. Improvement of performance of X-ray laser One of the most interesting applications of soft x-ray lasers is single-shot imaging of transient nano-scale phenomena by the methods of x-ray laser hologram, diffraction imaging and x-ray laser interferogram, etc. In order to obtain clear image of the nano-structure by these imaging techniques, more than 1010 spatially coherent photons may be required. In the transient collisional excitation lasers, since the output intensity under the gain-saturation regime is determined by the depopulation rate of the lasing levels of the gain medium ions and is 1010 -1011 Wcm-2, the generation of larger gain region under the same pumping energy is one of the key issues to improve the performance of the x-ray laser.

Fig.3. Footprints of the x-ray lasers from the 1sr target (left), the 2nd target (middle) and the double targets (right).

Proc. of SPIE Vol. 7451 745107-4

One possibility is to optimize the shape of targets to enhance the heating efficiency of the plasma. V-groove target is one of the candidates, by which the free expansion of the plasma and the radiation cooling may be suppressed leading to confine the pumping energy into localized plasma. We prepare a v-groove target with the depth of 200 µm and width of aperture is 150 µm and observe whether the heating efficiency of the plasma is improved or not. Iron v-groove target was irradiated by double pulses of CPA Nd:glass laser with pumping energy of 10 J. Figure 4 shows spatial distribution of x-ray emission in several KeV range taken by an x-ray pinhole camera for the flat target case; (a) and the v-groove target case; (b). By using the v-groove target, the x-ray emission area became 7 times larger compared with the case of flat target, and the energy conversion efficiency into the x-ray emission enhanced more than one-order of magnitude. It should be noted that in the case of v-groove target, the best result was obtained under the defocus condition, i.e., the focusing position was 800-900 mm before the bottom of the groove. With the information obtained from this preliminary test, we irradiated v-groove silver target with length of 6 mm. Figure 5 shows the far-field pattern of the 13.9 nm laser beam from flat target; (a), and v-groove target; (b). The beam divergence was improved from 10 x 20 mrad2 to 5 x 5 mrad2, although the output energy decreased only by a factor of 2. The present result may be due to that the beam propagation direction is limited by the shape of the groove and that large gain region with calm density gradient is generated by use of v-groove target. Since the beam divergence of the laser under the diffraction-limit becomes smaller as the source size becomes larger, the present result has potential to improve not only the output energy but the beam divergence under the fully spatial coherent operation using double target geometry.

Fig.5. Far-field patterns of the 13.9 nm laser with flat target; (a) and v-groove target; (b). The sight area corresponds to 20 x 20 mrad2.

(a) (b)

150 µm 150 µm

(a) (b)

150 µm 150 µm

Fig. 4. X-ray pinhole image of the plasma with the flat target; (a) and the v-groove target; (b). The width and the depth of the v-groove is 150 µm and 200 µm, respectively. The dashed line

SN_2008021401 (flat target) SN_2008021409 (v-groove)

(a) (b)

Proc. of SPIE Vol. 7451 745107-5

3. APPLICATIONS OF THE SOFT X-RAY LASERS In this section, we describe the application experiments using the fully spatial coherent 13.9 nm laser under the collaborations with universities, research institutes, and other research sections of JAEA. 3.1 Overview of the application of x-ray laser in JAEA In the application in material science, a couple of years ago, we have firstly observed pico-second snap-shot of domain-structure of ferro-electric substrates, BaTiO3, by use of x-ray laser speckle technique [12, 13]. The following double probe speckle measurement revealed the temporal-correlation of "fluctuation" in the phase-transition of this material. Fig. 6 shows typical correlation speckle signal. In this measurement, the 13.9 nm laser was divided into two beam by Michelson-type double pulse generator involving free-stand Mo/Si multi-layer beam splitter. With certain delay time, two x-ray laser pulses probed the surface of BaTiO3, and the two speckle patterns were recorded by x-ray streak camera. The speckle signal by use of the double XRL probes were taken for various time delay, and the temporal correlation function of the domain structure of BaTiO3 was determined. The single-shot x-ray speckle measurements shows us that XRLs are the powerful tool to study non-periodic ultra-fast phenomena such as the domain fluctuation under the phase transition. Our next goal is to extend the method of the measurement to optical pump & x-ray laser probe and to apply this method to the research area such as photo-induced phase transition and laser ablation process.

The study of optical property of materials is another interesting application. In the research fields relevant to next generation lithography, the development of efficient and fast imaging scintillator devices with sufficient size is one of the key issues for lithographic applications. Zinc oxide (ZnO), a wide gap semiconductor, is one of the promising materials for the scintillators, and recent progress in the fabrication technique enables us to obtain large-size homogeneous crystal at a low cost. Since the property of the fluorescence spectrum of ZnO for the EUV light pumping, e.g., the wavelength and the lifetime, has not been well-known, therefore we tried to characterize the fluorescence by the method of soft x-ray laser induced fluorescence (X-LIF) spectroscopy. Obtained fluorescence had a peak at around 380 nm. It was sufficiently intense and the lifetime was short enough (t~ 10 ns), furthermore these optical properties were virtually the same with the case pumped by 351 nm UV laser. This implied that ZnO crystal was suitable for the fast-scintillator device for the UV-EUV region [14]. In atom and molecules physics, interaction between Xe cluster and intense soft x-ray pulse has been studies by the collaboration with Hiroshima university. The 13.9 nm laser pulse with sub micro joules and 7 ps-duration irradiated the Xe cluster target, and the production rate of several ionic stages of Xe ions were measured by the method of time of flight. Our result showed the production rate of Xe3+ ions dominated that of Xe2+, which contradicted to the result

Fig.6. Typical soft x-ray speckle signal from BaTiO3 taken by x-ray streak camera. The x-ray laser pulse is divided into two pulses, separated by 120 ps in this measurement.

Proc. of SPIE Vol. 7451 745107-6

obtained in synchrotron radiation source (SR). This was due to that XRL photon flux is larger by 6 orders of magnitude than that of the SR. Under such the condition, more than 10% of atoms in the cluster were inner-shell-ionized, and this together with the following auto-ionization process formed virtually solid state density plasma before the Coulomb explosion. Our quantitative estimation indicated that substantial ionization level lowering of Xe ions in high density plasma enhanced the production channel of Xe3+ ions. This study is closely connected to the physics of strongly coupled plasma or warm dense matter. [15]. We also demonstrated single-shot Fourier transform holography. The 13.9 nm XRL beam was focused by the Fresnel zone plate with a 50 nm -thickness Au zone fabricated on a 0.75×0.75 mm2 silicon nitride (Si3N4) membrane with a thickness of 100 nm.. The diameter of the FZP was 0.434 mm, and the total zone number and outermost zone width were 1700 and 64 nm, respectively. The focal length was 2 mm for 13.9 nm. The focal spot size was 66 nm, and the focused beam was used as the reference beam. A test grid pattern with 2 microns period was put in the focal plane and was placed several tens microns from the focal spot to avoid the illumination by the reference beam. The 0th order transmitted XRL light illuminated the test patterns, and the wavefront was distorted. This distorted wave and the reference beam made hologram, and it was recorded by x-ray CCD at the distance of 0.23 m from the test pattern. Figure 3 shows the test patterns; (a), raw-data of single-shot hologram; (b), and the reconstructed intensity image; (c). In Figure 3(c), the vertical and horizontal 1 µm line-and-space pattern can be virtually resolved. In this experiment, the output energy of the 13.9 nm laser was 0.1 µJ. It was estimated that with 1 µJ output energy of the XRL, we could obtain clear single-shot holography. 3.2 Observation of low-threshold ablation of substances by use of pico-second duration of 13.9 nm laser Laser ablation has many technological applications in material processing and nano-structure fabrication. The laser-induce damage of the materials has been intensively studied and its dependence upon the pulse width, the photon energy, and the fluence has come to be understood. For the relatively long duration optical laser pulses with > 20 ps, it is shown that the damage of dielectrics is mainly originated from the heating of conduction band electrons and transferring of electron energy to the lattice. Damage occurs if the deposited energy is sufficient to melt the dielectric material. For the short enough pulses, the laser energy is absorbed by the electrons much faster than it is transferred to the lattice, resulting in drastically decrease in the ablation threshold energy. The same situation can occur in extreme ultraviolet or soft x-ray laser. In the recent studies, a well-defined ablation threshold at 0.06 for CaF2 and 0.11 J /cm2 for LiF has been found for pulse energy 0.3 mJ of the Ne-like Ar x-ray laser at the wavelength of 46.9 nm with the duration of 1.7 ns, and much lower threshold energy can be expected in the case of transient collisional excitation laser with several picoseconds duration. The 13.9 nm nickel-like silver laser from a single target was focused on a 2 mm-thick LiF crystal with the diameter of 20 mm. The output energy of the 13.9 nm laser was 1 µJ, and the vertical and horizontal beam divergence was 12 x 5 mrad2, respectively. The focusing optics was a normal incidence Mo/Si multi-layer spherical mirror with the curvature of 1050 mm, which was placed at a distance of 2715 mm from the XRL output and used at the incident angle of 2°. A 0.2 µm thick Zr filter was settled or removed in front of the x-ray mirror at 800 mm from the XRL exit in order to reduce the scattered optical radiation and the thermal x-ray emissions from the laser-produced plasma. The LiF crystal was moved after each shot along the XRL propagation direction and also perpendicular to it, in order to record the beam patterns at different focusing distances on the fresh LiF crystal surface. The total energy of the XRL beam on the LiF crystal was 170 nJ in each shot. The luminescence of stable color centers, CCs, formed by the x-ray laser radiation, was used to measure the intensity distribution in the XRL laser focal spot. The CCs in LiF crystal were observed by using a confocal fluorescence laser microscope after irradiation of the LiF crystal with the x-ray laser [16, 17]. Figure 7 shows the optical microscope image of the LiF crystal irradiated by the 13.9 nm laser (middle) together with the AFM traces of the ablation area (upper and lower). The ablation area is indicated by dotted line. From visible microscope image and AFM trace in Fig 7, we can clearly see the surface alteration of LiF crystal. In the AFM trace 1 and 2 in Fig 7 show that the ablation depths vary between 30 and 55 nm, which are almost consistent with the absorption depth of 28 nm for 13.9 nm radiation. This implies that the ablation process in the present photon energy region (= 90 eV) is mainly determined by the absorption by the electrons in the localized area. In the present experiment, the beam energy of the x-ray laser to generate the surface alteration is only 10.2 mJ/cm2, which is 3400, 300, and 10 times smaller compared with previously measured thresholds for nanosecond and femtosecond Ti:sapphire lasers, and for nanosecond 46.9 nm soft x-ray laser, respectively. To explain such a strong

Proc. of SPIE Vol. 7451 745107-7

reduction in LiF crystal ablation threshold in the case of using picoseconds x-ray laser, an ablation mechanism for dielectrics is connected with a 100% laser absorption in a thin surface layer and formation of negative pressure zone, followed by thermomechanical fragmentation, under a sufficiently strong tensile stress. As a rule, this mechanism works for metals and semiconductors irradiated by ultrashort visible laser pulses [18].

3.3 Development of soft x-ray laser interferometer toward single shot imaging of nano-scale dynamics In order to obtain single-shot image of the nano-scale structure and its dynamics, new x-ray laser interferometer has been developed. The interferometer was designed with a group of institute of solid state physics (ISSP) and Tokushima university together with good experiences obtained through the collaborations with PALS group using single Lloyd's mirror [19] and with Université de Paris Sud using bi-mirror [20]. Schematic of the x-ray interferometer is shown in Fig. 8. The fully spatial coherent 13.9 nm laser beam, with the output energy of 1-2 µJ and the duration of 7 ps, passing through the 0.1 µm-thick Zr filter was steered by a planar Mo/Si multi-layer mirror to the pair of a Pt coated grazing incidence mirror and the sample. The Pt mirror was set to be parallel to the sample, and both of them were fixed in a rotation stage to adjust the incident angle of the x-ray laser. In the present case, the sample was Pt hump with the height of 10 nm fabricated on the sample by use of 1951 USAF test pattern as the mask. The angle of the incidence of the x-ray laser to the sample was set to be θ = 70° with respect to surface normal. The image of the illuminated area on the sample was transferred to the CCD position by a Mo/Si spherical mirror with the focal length of 250 mm. The distance from the

ablation areaablation area

Fig.7 Optical microscope image of the LiF crystal irradiated by the 13.9 nm laser (middle) together with the AFM traces of the ablation area (upper and lower). The ablation area is indicated by dotted line.

Proc. of SPIE Vol. 7451 745107-8

sample to the spherical mirror and that from the spherical mirror to the CCD was about 260 mm and 4800 mm, respectively, which led to the magnification factor of 17.

Interference pattern was obtained by putting a double Llyod's mirror between the spherical mirror and CCD. The angle of the incidence for the double Lloyd's mirror was 87° with respect to surface normal. The double Lloyd's mirror consisted of two Pt coated 400 µm-thick Si wafers, which were fixed with a relative angle of η = 0.006° (= 0.1 mrad) in a hausing. One of the wafers covered the portion of the x-ray laser beam including the distorted wavefront by the hump, and the other was used as the reference. The overlap of the two wavefronts made interference pattern on the CCD (See the dark area in Fig. 8). The separation of the neighboring two fringes was ~ 130 µm, which was determined by the relative angle of the two wavefronts and the wavelength as λ/sin(η/2), where λ is the wavelength of the x-ray laser. The CCD has 1000 x 1000 pixels with the pixel size of 24 x 24 µm2. Figures 9 shows typical interference pattern of the sample. The position of the hump was indicated by open square area in Fig. 9, and apparent fringe shift was obtained. In the present case, the phase shift of 2π corresponded to the height of (λ/2)/cosθ ~ 20 nm, and the height of the hump derived from the fringe shift was 12-13 nm, which was almost consistent with the expectation.

Fig. 8 schematic diagram of the x-ray inteferometer using double Llyod's mirror.

Fig. 9. Typical interferogram taken by double Lloyd's interferometer.

The open square indicates the position of one of the humps with the height of 10nm fabricated on the sample.

Proc. of SPIE Vol. 7451 745107-9

The present preliminary experiment implies the new x-ray laser interferometer works well as a soft x-ray probe system with sufficient high resolution in the depth. Our next goal is to construct a optical pumping system and combine it with the x-ray laser interferometer to observe non-periodic ultra-fast phenomena such as photo-induced phase transition and laser ablation process.

ACKNOWLEDGEMENT

This work is partly supported by auspices of MEXT (Japanese Ministry of Education, Culture, Sports, Science and Technology) project on "Mono-energetic quantum Beam Science with PWlaser" and Japan Grant-in-Aid for Scientific Fundation, Kiban (B), No. 21360364.

REFERENCES

[1] Luther B. M., Wang, Y. Larotonda, M. A., Alessi, D., Berrill, M., Marconi, M. C., Rocca, J. J., Shlyaptsev, V. N., "Saturated high-repetition rate 18.9-nm tabletop laser in nickellike molybdenum", Optics Letters, 30, 165-167 (2005).

[2] Sebban. S., Mocek. T., Ros. D., Upcraft. L., Balcou P., Haroutunian R., Grillon. G., Rus. B., Klisnick. A., Carillon. A., Jamelot. G., Valentin. C., Rousse. A., Rousseau. J. P., Notebaert. L., Pittman. M., and Hulin. D., "Demonstration of a Ni-like Kr optical-field ionization collisional soft x-ray laser at 32.8 nm", Phys. Rev. Lett. 89, 253901 (2002).

[3] Kalachnikov. M., Nickles. P. V., Schnüer. M., Sandner. W., Shlyatsev. V. N., Danson. C., Neely. D., Wolfrum. E., Zhang. J., Behjat. A., Demir. A., Tallents. G. J., Warwick. P. J., and Lewis C. L. S., "Saturated operation of a transient collisional x-ray laser", Rev. A 57, 4778-4783 (1998).

[4] Kawachi T., Kado M., Tanaka M., Sasaki A., Hasegawa N., Kilpio A. V., Namba S., Nagashima K., Lu. P., Takahashi K., Tang H., Tai R., Kishimoto. M., Koike. M., Daido. H, and Kato Y., "Gain saturation of nickel-like silver and tin x-ray lasers by use of a tabletop pumping laser system", Phys. Rev. A 66, 033815 (2002).

[5] Dunn J, Osterheld. A. L., Nilsen. J., Hunter J. R. and Shlyaptsev. V. N., "Gain saturation regime for laser-driven tabletop, transient Ni-like ion x-ray laser", Phys. Rev. Lett. 84, 4834-4837 (2000).

[6] Tanaka. M., Nishikino. M., Kawachi. T., Hasegawa. N., Kado. M., Kishimoto. M., Nagashima. K., Kato. Y., "X-ray laser beam with diffraction-limited divergence generated with two gain media", Optics Letters 28, 1680-1682 (2003).

[7] Nishikino M., Tanaka. M., Nagashima. K., Kishimoto. M., Kado. M., Kawachi. T., Sukegawa K., Ochi. Yoshihiro., Hasegawa., N., Kato. Y., "Demonstration of a soft x-ray laser at 13.9 nm with full spatial coherence", Phys. Rev.A 68, 061802(R) (2003).

[8] Nishikino M., Hasegawa. M., Kawachi. T., Yamatani. H., Sukegawa. K., and Nagashima. K., "Characterization of a high-brilliance soft x-ray laser at 13.9 nm by use of an oscillator-amplifier configuration", Applied Optics, 47, 1129-1134 (2008).

[9] Wang Y, Granados E., Larotonda. M. A., Berrill. M., Luther. B. M., Patel. D., Menoni. C. S., and Rocca. J. J., "High-brightness injection-seeded soft x-ray-laser amplifier using a solid target.", Phys. Rev. Lett. 97, 123901 (2006).

[10] Kawachi. T., Kado. M., Tanaka. M., Hasegawa. N., Nagashima. K., Sukegawa. K., Lu. P., Takahashi. K., Namba. S., Koike. M., Nagashima. A., Kato. Y., "Development of a pumping laser system for x-ray laser research", Applied Optics 42, 2198-2205 (2003).

[11] Ochi. Y., Hasegawa. N., Kawachi. T., and Nagashima. K., "Development of a chirped pulse amplification laser with zigzag slab Nd:glass amplifiers dedicated to x-ray laser research", Applied Optics, 46, 1500-1506 (2007).

[12] Tai. R. Z., Namikawa. Kishimoto. M., Tanaka. M, Sukegawa. K., Hasegawa. N., Kawachi. T., Kado. M., Lu. P., Nagashima. K., Daido. H., Maruyama. H., Sawada. A., Ando. M., and Kato. Y., "Picosecond snapshot of the speckle from ferroelectric BaTiO3 by means of x-ray lasers", Phys. Rev. Lett. 89, 257602 (2003).

Proc. of SPIE Vol. 7451 745107-10

[13] Tai. R. Z., Namikawa. K., Sawada. A., Kishimoto. M., Tanaka. M., Lu. P., Nagashima. K., Maruyama. H., and Ando. M., “Picosecond view of microscopic-scale polarization clusters in paraelectric BaTiO3,” Phys. Rev. Lett. 93, 087601 (2004).

[14] Tanaka M., Nishikino. M., Yamatani. H., Furukawa. Y. Murakami. H., Saito S., Sarukura. N., Kagamitani. Y., Enhentraut, D., Fukuda. T., "Hydrothermal method grown large-size zinc oxide single crystal as fast scintillator for future extreme ultraviolet lithography", Applied Phyics Letters, 91, 231117 (2007).

[15] Namba. S., Hasegawa. N., Nishikino. M., Kawachi. T., Kishimoto. M., Sukegawa. K., Tanaka. M., Ochi. Y., Takiyama. K., and Nagashima. K., "Enhancement of double Auger decay probability in xenon clusters irradiated with a soft x-ray laser pulse,.", Physical Review Letters, 99, 043004 (2007).

[16] Faenov. A. Ya., Inogamov. N. A., Zhakhocskii. V. V., Khokhlov. V. A., Nishihara. K., Kato. Y., Tanaka. M., Pikuz. T. A., Kishimoto. M., Ishino. M., Nishikino. M., Nakamura. T., Fukuda. Y., Bulanov. S. V., and Kawachi. T., "Low-threshold ablation of dielectrics irradiated by picosecond soft x-ray laser pulses", Applied Physics Letters, 94, 231107 (2009).

[17] Faenov. A. Ya., Kato. Y., Tanaka. M., Pikuz. T. A., Kishimoto. M., Ishino. M., Nishikino. M., Fukuda. Y., Bulanov. S. V., and Kawachi. T., Optics Letters 34, 941-943 (2009).

[18] Anisimov. S. I., Inogamov. N. A., Petrov. Yu. V., Khokhlov. V. A., Zhakhocskii. V. V., Agranat. M. B., Ashitkov. S. I., and Komanov. P. S., "Interaction of short laser pulses with metals at moderate intensities", Applied Physics A, Material Science & processing, 92, 939-943 (2008).

[19] Rus. B., Daido. H., Tang. H., Nishiuchi. M., Kishimoto. M., Tanaka. M., Kawachi. T., Hasegawa. N., Nagashima. K., Arisawa. T., and Kato. Y., "Picosecond Fourier holography using a Lloyd's mirror and an x-ray laser at 13.9 nm", Proc. of x-ray lasers 2002, eds. by Rocca. J. J., Dunn. J, and Suckewer. S., AIP-CP 641, 522-527 (2002).

[20] Tang. H., Guilbaud. O., Jamelot. G., Ros. D., Klisnick. A., Joyeux. D., Phalippou. D., kado. M., Nishikino. M., Kishuimoto. M., Sukegawa. K., Ishino. M., Nagashima. K., and daido. H., "Diagnostics of laser-induced plasma with soft x-ray (13.9 nm) bi-mirror interference microscopy", Applied Physics B: Laser and Optics, 78, 975-977 (2004).

Proc. of SPIE Vol. 7451 745107-11