Search

International audience

In recent years, integral imaging has received increasing attention as a promising 3D imaging technique with one single exposure. In medical diagnostics, the X-ray integral imaging system potentially has a much shorter exposure time than the conventional computed tomography, reducing the radiation damage to the patient. By replacing the micro-lens array with a micro-Fresnel Zone Plate (FZP) array, the classical visible integral imaging system can be transposed to an X-ray system. However, limited by the micro-scale dimensions of FZP in the array and current manufacturing techniques, the number of zones of FZP is required to be small. This may have an important impact on the FZP imaging performance. Based on the scalar diffraction theory, this paper introduces a simulation method and numerically investigated the effect of the number of zones on the FZP imaging performance under different coherence conditions.

APOLLON laser facility was initiated by Nobel Laureat Gerard Mourou to become one of the few European 10 PW laser facilities. End of 2018, APOLLON reached 1 PW peak power. Experiments at this power are under preparation with an expected start during spring 2019 or early summer. In parallel, today the laser is under upgrade for reaching 10 PW in few years with an intermediate step at around 5 PW dedicated to the development of experiments at this high power. APOLLON will be open for external users on 2020 at 1 PW and very likely on 2021 at several PW. APOLLON has two experimental areas on which X-ray sources will be developed. The so-called "Long focal length area" (LFA) is dedicated to electron acceleration and related X-ray emission. The room may accommodate focal length up to 9m. The experiments are designed in a way to allows single or double stage electron acceleration with expected electron energies reaching several 10's of GeV. X-ray emission from betatron will be used as a diagnostic of electron acceleration processes but will be also developed independently aiming at achieving energetic and well-collimated X-ray beam. The experiment has been set in a way to allows heads-on electron-laser collision for Compton scattering experiments. The second target area called "Short Focal area", aims at achieving intensities as high as possible for electron and ion acceleration, non-linear QED tests and X-ray generation through high harmonic generation on solid. X-ray emission ranging up to 10 keV is foreseen, with very high peak power at lower energy.

Emergence of ultrafast EUV to X-ray sources, Free Electron Laser, High harmonic generation, betatron and Compton, has opened new paradigms in physical, chemical, biological and medical sciences by either producing ultrahigh intensities or for ultrafast imaging or by enabling pump-probe experiments at new timescales. Most of these experiments require an excellent or at least a properly defined wavefront (WF). A number of WF sensing techniques have been proposed in the X-Rays, including grating-based interferometry, speckle tracking, pencil beam deflectometry, or curvature sensors. Among these techniques, Hartmann WF sensing demonstrated a number of advantages, such as insensitivity to vibrations, achromaticity and very large dynamic range. Furthermore, the phase and intensity maps are directly retrieve and nearly instantaneously without the use of complex and long algorithms. Imagine Optic developed EUV to X-ray WF sensors since more than 15 years taking benefit of decades of experience in the visible range. We will show several experiments using our EUV sensors for metrology and optimization of ultrafast EUV sources. We recently developed a Hartmann sensor in the 5 - 25keV range. The device is based on a custom scintillator-to-detector optical relay system, as well as on an optimal Hartmann array geometry, providing 20µm spatial WF sampling resolution, over a 3x3 mm² pupil. We show the results of first experiments on a synchrotron beamline at 10 keV, achieving 4pm WF repeatability.

Since its creation in 1996, Imagine Optic designed and manufactured high performance Shack-Hartmann wavefront (WF) sensors for many kinds of applications such as telescope alignment, laser characterization, optics qualification or adaptive optics, and for many different fields such as space optics, microscopy, high power lasers or lithography. Since 2003, Imagine Optic is actively developing EUV to X-ray Hartmann WF sensors for applications on metrology beams emitted by synchrotrons, free-electron lasers, plasma-based soft X-ray lasers and high harmonic generation. Our most recent developments include the realization of a EUV sensor adapted to strongly convergent or divergent beams having numerical aperture as high as 0.15, as well as the production of a hard X-ray sensor working above 10 keV, providing outstanding repeatability as good as 4 pm rms. Our sensors have demonstrated their high usefulness for the metrology of EUV to X-ray optics from single flat or curved mirrors to more complex optical systems (Schwarzschild, Kirkpatrick-Baez static or based on bender technology or with activators). In terms of optics qualification is a clear advantage of actually measuring the wavefront at-wavelength. Also, we show active Kirkpatrick-Baez alignment in few minutes using our WF sensor in both manual and automatic loops at the benefit of strong improvement of the beam focalization on the sample. Recently we started developing our own compact deformable grazing incidence mirror bender. We present a review of the developed sensors, as well as experimental demonstrations of their benefits for optical metrology of various EUV and X-ray optics.

Imagine Optic (IO) is actively developing EUV to X-ray wavefront (WF) sensors since 2003 for applications on metrology of EUV to X-ray beams emitted by synchrotrons, free-electron lasers, plasma-based soft X-ray lasers and high harmonic generation. Our sensors have demonstrated their high usefulness for metrology of EUV to X-ray optics from single flat or curved mirrors to more complex optical systems (Schwarzschild, Kirkpatrick-Baez static or based on bender technology or with activators). Our most recent developments include the realization of a EUV sensor adapted to strongly convergent or divergent beams having numerical aperture as high as 0.15, as well as the production of a hard X-ray sensor working at 10 keV and higher energies, providing repeatability as good as 4 pm rms. We present a review of the developed sensors, as well as experimental demonstrations of their benefits for various metrology and WF optimization requirements.

With the progresses achieved on the development of high quality, highly coherent soft X-ray (or EUV) sources from synchrotrons, free-electrons laser, high harmonics and plasma based soft X-ray laser, as well as for the need of always better optics for imaging (EUV lithography, EUV spatial telescopes, EUV microscopes), the demand on metrology is increasing very fast. Nowadays, many synchrotrons have developed metrology beamlines but with the limit of being too expensive and too large for transposing them to university-scale laboratories or optical firms. At Laboratoire d'Optique Appliquee, we have developed a compact and versatile metrology beamline to test at-wavelength different EUV optics, from single component to full assembly and adaptive optics. The beamline is based on the use of high harmonics generated by the interaction of a 35 fs, 4 kHz, 3 mJ laser with neutral gases. The high harmonics span from 10 to 50 nm and are fully coherent, collimated and exhibit a good wavefront of about lambda/5 rms. The beamline covers a footprint of about 5*1.5 m2 while the driving laser occupies about 4 m2. Itis composed of an interaction chamber where high harmonics are generated, a spectrometer and the metrology chamber (1.5m*0.7m) . We have tested many optical components from flat or curved mirrors to toroidal mirror or Schwarzschild microscope. We will present in detail the beamline as well as results from optic metrology. The beamline is also used for calibration of wavefront sensors. This beamline is well suited for testing EUV adaptive optic in any configurations.

International audience; Ultra-intense X-ray sources have opened new avenues by creating new states of matter or probing and imaging living or inert matter. Free-electron lasers have a strong leadership by delivering pulses combining femtosecond duration and 10s of microJoules to milliJoule energy. However, these sources remain highly expensive limiting their number to a few worldwide. In parallel, laser-pumped soft X-ray lasers hold outstanding promises having demonstrated the most energetic monochromatic soft x-ray pulse and being intrinsically fully synchronized with any secondary source of the pump laser. Since the first successful demonstration of amplification of a high harmonic pulse in a plasma from gas in 2003 and from solid in 2008, we have developed an extensive numerical study. 2D hydrodynamic simulations showed that optimized Transient Collisional Excitation plasma amplifiers, may store up to 0.4 mJ in the population inversion. If carefully seeded, pulses of 80 fs and 20 mu J might be generated with table-top lasers (10J). As the energy extracted is far from the milliJoule requirements of most exciting applications, we studied the seminal experiment of Ditmire et al who seeded a plasma emitting milliJoules in the form of Amplified Spontaneous Emission (ASE).We retrieved and explained for the first time the experimental result (ASE 1,000 times stronger than amplified seed). We thus proposed and fully modeled the transposition of the so-called Chirped Pulse Amplification (CPA) in the soft X-ray range, showing that 6 mJ, 200 fs, fully coherent soft X-ray pulse is accessible with compact pump lasers.

An experiment was set up to measure the wavefront of an injection-seeded soft x-ray laser based on a solid-target plasma amplifier. The 43rd harmonic signal from a Ti:Sa laser was used to seed a molybdenum plasma amplifier operating in the λ=18.9 nm line of Ni-like Mo. A Hartmann wavefront senor with an accuracy of λ/32 rms at this wavelength was employed to measure the wavefront of both the high harmonics seed and the seeded soft x-ray lasers. A significant improvement in wavefront aberration from 0.51±0.04λ rms to 0.25±0.03λ rms was observed as a function of plasma column length. The variation of wavefront characteristic by the time delay between the injection of the seed and the peak of soft x-ray amplifier pump was studied in this paper.

In this paper we treat electron distributions produced in high intensity X-ray laser experiments and compare the behaviour of the sample assuming thermalization or following the non-equilibrium distribution at densities well below that of the solid.

Circularly polarized high order harmonics in the extreme ultraviolet range (18 - 27 nm) have been obtained by a two steps process. Harmonics were generated from a linearly polarized infrared laser (40 fs, 0.25 TW) focused into a neon filled gas cell. The harmonics have then been circularly polarized by a four-reflector phase-shifter. The polarization of the harmonics have been measured using a rotating multilayer broadband mirror set at an incidence angle of 45°. Fully circularly polarized radiation has been obtained with an efficiency of a few percents. This is significantly more efficient than currently demonstrated direct generation of elliptically polarized harmonics. This demonstration opens up new experimental capabilities based on high order harmonics, for example, in materials science for time-resolved nanometric magnetic imaging.

Seeding plasma-based soft x-ray laser (SXRL) demonstrated diffraction-limited, fully coherent in space and in time beam but with energy not exceeding 1 μJ per pulse. Quasi-steady-state (QSS) plasmas demonstrated to be able to store high amount of energy and then amplify incoherent SXRL up to several mJ. Using 1D time-dependant Bloch-Maxwell model including amplification of noise, we demonstrated that femtosecond HHG cannot be efficiently amplified in QSS plasmas. However, using Chirped Pulse Amplification concept on HHG seed allows to extract most of the stored energy, reaching up to 5 mJ in fully coherent 130 fs pulses. Original pump-probe experiments will be proposed thanks to the high laser energy available in ELI facilities.

We present in this paper theoretical and experimental investigations of temporal and spectral properties of seeded soft x-ray lasers. Bloch-Maxwell simulations of the harmonic pulse propagation in a soft x-ray laser plasma have been performed. Results show a growing wake of coherent radiation formed after the harmonic pulse.We describe the first measurement of the spectral bandwidth of a seeded soft x-ray laser. Using a varying path difference interferometer the spectral profile of a seeded OFI x-ray laser has been experimentally determined, leading to a Fourier-transform pulse duration of 5ps. The measured bandwidth is in good agreement with simulations. Finally we present the progress toward the implementation of a seeded soft x-ray laser at 18.9 nm at the new LASERIX facility.

In the race towards attosecond (as) pulses for which high order harmonics generated in rare gases are the best candidates, both the Harmonic spectral range and spectral phase have to be controlled. We present in this proceeding four mirrors numerically optimized and designed to compensate for the intrinsic Harmonic chirp recently discovered and which is responsible for a temporal broadening of the pulses. They are capable of compressing the duration down to 100 as. We present the fabrication of those aperiodic multilayers and show the measurement of reflectivity, which prooves that those multilayers are in agreement with the specifications and so let us think that they will be able to compress attosecond high harmonics trains.

Below saturation, X-ray laser output shows a reduction in pulse duration and frequency bandwidth as the gain-length product increases. Above saturation, both quantities can be expected to rebroaden. The duration of gain can be close to an order-of-magnitude longer than the output pulse duration. With gain-length products just below saturation, X-ray lasing at 13.9 nm in nickel-like silver has been measured with a pulse duration Δ t of 3 - 4 ps and an estimated frequency bandwidth Δ v of 5×10 11 Hz. Such values imply that the pulses are close to transform limited with Δ t Δ v ≈ 1.5. Measurements of x-ray laser pulse-lengths and gain duration will described in this paper.

We present the longitudinal coherence measurement of the transient inversion collisional x-ray laser for the first time. The Ni-like Pd x-ray laser at 14.68 nm is generated by the LLNL COMET laser facility and is operating in the gain-saturated regime. Interference fringes are produced using a Michelson interferometer setup in which a thin multilayer-coated membrane is used as a beam splitter. The longitudinal coherence length for the picosecond duration 4d{sup 1}S{sub 0} {yields} 4p{sup 1}P{sub 1} lasing transition is determined to be {approx}400 {micro}m (1/e HW) by adjusting the length of one interferometer arm and measuring the resultant variation in fringe visibility. This is four times improved coherence than previous measurements on quasi-steady state schemes largely as a result of the narrower line profile in the lower temperature plasma. The inferred gain-narrowed linewidth of {approx}0.29 pm is also substantially narrower than previous measurements on quasi-steady state x-ray laser schemes. This study shows that the coherence of the x-ray laser beam can be improved by changing the laser pumping conditions. The x-ray laser is operating at 4 - 5 times the transform-limited pulse.

We present a review of new progress performed in several laboratories (Laboratoire pour l'Utilisation des lasers Intenses, Rutherford Appleton Laboratory, Prague Asterix Laser System, Institute of Laser Engineering, Laboratoire d'Optique Appliquee). Concerning the realization of x-ray lasers sources, using different laser pumping techniques (600 ps, 100 ps, ns/ps, OFI) and the optimization of their optical properties, using curved and plane half-cavity mirrors. In parallel of these developments, we present the main results obtained with x-ray laser in interferometry applications. These studies concern on the one hand the Michelson interferometry with an x-ray laser emitting at 13.9 nm (recently realized at LULI), and on the other hand the Fresnel bi-mirror with an x-ray laser emitting at 21.2 nm (recently realized at PALS).© (2001) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Large amplification at 25.5 nm in neon-like iron has been demonstrated in experiments where prepulses are used. We show that the interaction between the x-ray laser beam and the amplifying medium must be taken into account in a reliable modeling of the saturation regime. Two approaches for intensity calculations are presented in this contribution. The first one combines the radiative transfer equation and the population rate equations. This approach is fully consistent, in the sense that beam amplification and population kinetics are treated simultaneously. A formalism based on a paraxial Maxwell-Bloch approach is presented. The Maxwell-Bloch calculations give the variation with length of intensity, local gain...Moreover, in the small-signal regime, it is possible to define an effective gain which is comparable to the measured gain. The second approach is based on a raytrace calculation where the saturation effect has been introduced empirically. The two codes need the electron density and the electron and ion temperatures as inputs. These quantities are given by the hydrocode EHYBRID. The two approaches give similar results.

We report in this article the experimental and numerical tools, developed at the LSAI, for a complete characterization of an x-ray laser (XRL) beam. First, a Michelson interferometer has been used to realize a Fourrier transform spectroscopy experiment. A full comprehension of the measured linewidth requires a comparison of the XRL beam amplification in the plasma to raytrace simulation. Results of transient pumping XRL simulations are presented in this article. The last section is dedicated to a description of the XUV Shack-Hartmann wavefront sensor we have developed, and to the study of the capillary discharge XRL beam.

We present new progress in the optimization and understanding of the transient collisional pumping scheme using an ultra-short sub-ps heating pulse. The effect of traveling-wave irradiation in enhancing the lasing output of the 4d-4p Ni-like Ag line is studied in detail. A new irradiation scheme using a frequency-doubled 600 ps pulse to preform a plasma is tested. Strong lasing is also obtained on a new line at 16.05 nm that we identify to a 4f-4d transition in Ni-like Ag. Finally we review our recent work in the development of applications of the 21.2 nm zinc laser for imaging or exciting matter. New experiments include the probing of a plasma by imaging Fresnel interferometry and a first attempts to demonstrate two-photon ionization in a xenon gas.

We use x-ray laser interferometry to probe defects induced by a strong electric field on niobium surface. Niobium has been chosen on account of its frequent use in superconductive cavities of particle accelerators. The x-ray laser emits bright, 50 ps-duration pulses at (lambda) equals 21.2 nm. The beam is reflected on the niobium surface under grazing incidence. The interferometer is of the wave-front division type. Interferograms are single shot recorded, which enables to probe `instantaneous' defect morphology. We observed appearance and evolution of defects between 14 MV/m and 35 MV/m. The vertical set amplitude is of 10 - 20 nm. The defect structure has been observed to shift by 500 micrometers along the metal surface under a constant 35 MV/m electric field, during the 20 minutes time interval between two laser shots.© (1998) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

This work has consisted in demonstrating that high gain can be achieved by pumping x-ray lasers (XRL) with a combination of a high intensity and short duration driving pulses (approximately 100 ps). Short pulses are very well suited for pumping collisional XRL since a high lasant ion density, electron density and temperature can be achieved simultaneously. We have successfully tested this pumping scheme on the 4d-4p (J equals 0 - 1) transition of Ni-like tin (lambda approximately 11.93 nm) and silver (lambda approximately 13.89 nm) as well as on the 3p-3s (J equals 0 - 1) Ne-like iron (lambda approximately 25.5 nm) at an intensity of approximately 2 X 1013 Wcm-2 (130 ps in duration). The driving laser (lambda equals 1.06 micrometer) was composed of three pulses (a prepulse and two main pulses). Large amplifications were demonstrated in tin and silver (respectively GL approximately 12 and GL approximately 16). Finally, the saturation of the 3p-3s (J equals 0 - 1) transition of Ne-like iron at 25.5 nm was achieved on both pumping pulses, using a prepulse of 109 Wcm-2. A gain coefficient of 15 plus or minus 3 cm-1 (GL approximately 26 plus or minus 5) on the first main pulse and 12 plus or minus cm-1 (GL approximately 23 plus or minus 2) on the second one was measured.© (1997) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Results of an experiment aimed at a detailed characterization of line plasmas created by intensities equivalent to those generated by low-level prepulses in the collisional excitation soft x-ray lasers are presented. Electron density profile in the plane perpendicular to the line plasma axis, at times 4 and 10 ns subsequent to the laser pulse, time-integrated electron temperature and time-integrated lateral extent of the plasma were obtained. These parameters were studied for zinc, copper and iron plasmas created on slab target by approximately 400 ps (FWHM) pulses of an iodine laser ((lambda) equals 1.315 micrometer) giving rise to irradiances ranging from approximately 4.8 multiplied by 10 9 to approximately 1.2 multiplied by 10 11 Wcm -2 .

The main feature of x-ray laser research at LULI is the development of a saturated laser at 212 angstrom with a relatively small pump laser of 0.4 kJ in 600 ps. The laser works with the 3p- 3s J equals O yields 1 transition of neon-like zinc, by using the double-pass of amplified radiation in the active medium. Plasma parameters (temperature, density, homogeneity), and x-ray laser emission properties (intensity, pointing angle, divergence, and coherence) have been studied. Lasing action needs the main laser pulse to be preceded by a ten-prepulse train (contrast ratio less than 103) due to the remnant oscillator. The effect of a single prepulse was investigated as a function of contrast ratio and delay between the prepulse and the main pulse.

Current successful approaches for achieving soft x-ray lasing typically require pumping laser pulses of duration approximately ns and energy approximately kJ (collisionally pumped schemes) or approximately ps pulses and powers of approximately several TW (recombination-pumped schemes). For applications, it is important to improve the efficiency of soft x-ray lasers and so reduce the required power of pumping lasers. The effect of pre- pulse on neon-like collisionally pumped lasers has been investigated using the LULI laser (Ecole Polytechnique, France). A small pre-pulse level approximately 10-3 of the main pulse energy was found to increase the J equals 0 minus 1 neon-like zinc laser output at 21 nm by an order-of-magnitude with a comparable increase in efficiency. A double pumping laser pulse on neon-like yttrium lasing output at 15 nm obtained with the VULCAN laser (Rutherford Appleton Laboratory, England) was also found to increase the x-ray lasing efficiency. With adiabatically cooled recombination lasing, it is shown that approximately 2 ps pulses are optimum for achieving the desired ionization balance for lasing output. The possibility of achieving recombination lasing at short wavelengths on lithium-like ions with longer pulse lasers has been investigated using the ASTERIX laser (Max-Planck Quantenoptik, Germany). These results are presented and interpreted to provide possible directions for improving the efficiency of x-ray lasers.

We present our recent efforts to produce X-ray lasers in the 200 angstroms range by using the moderate power drive of the LULI facility in Palaiseau. The 4 - 5 transitions of Li-like sulfur exhibit large gain-length products in recombining plasmas, and appear to be less sensitive to plasma non-uniformity than the 3 - 4 and 3 - 5 transitions previously studied. From numerical simulations this is likely due to smaller radiative and collisional excitation from 4f than from 3d levels. In collisional scheme, neon-like zinc gives analogous results to similar works on other elements for the 3p - 3s, J equals 2 yields 1 transitions, but the J equals 0 yields 1 transition shows a surprisingly large gain coefficient of 4.9 cm-1. From a detailed comparison of time-dependent intensities of the J equals 0 yields 1 and the J equals 2 yields 1 lines, we conclude that transitions from J equals 0 and from J equals 2 are not emitted in the same region of the plasma.© (1994) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.