NPS, Monterey, California
Stanford University, Stanford, California
The new experimental facilities for the Naval Postgraduate School Beam Physics Lab are at the 95% completion level for exterior construction, and work has begun on the internal lab spaces. A general timeline for the commencement of first experiments is presented, along with an overview of the experimental path forward. The NPS-BPL is rated for considerably higher average powers (40 kW) than most university accelerator facilities, which presents unique challenges in both the physical and administrative realms. Design considerations, radiation approval processes and other “lessons learned” in a non-U.S. Department of Energy government facility are discussed.
NASU/IRE, Kharkov
Using a time-dependent approach the analysis and optimization of a planar FEL-amplifier with an axial magnetic field and an irregular waveguide is performed. By applying methods of nonlinear dynamics three-dimensional equations of motion and the excitation equation are partly integrated in an analytical way. As a result, a self-consistent reduced model of the FEL is built in special phase space. The reduced model is the generalization of the Colson-Bonifacio model and takes into account the electrons’ intricate dynamics and intramode scattering. The reduced model and concepts of evolutionary computation are used to find optimal waveguide profiles. The numerical simulation of the original non-simplified model is performed to check the effectiveness of found optimal profiles. The FEL parameters are chosen to be close to the parameters of the experiment*, in which a sheet electron beam with the moderate thickness interacts with the TE01 mode of a rectangular waveguide. The results strongly indicate that one can improve the efficiency by a factor of five or six if the FEL operates in the magnetoresonance regime and if the irregular waveguide with the optimized profile is used.
*S. Cheng et al. IEEE Trans. Plasma Sci. 1996, vol. 24, p. 750
SLAC, Menlo Park, California
The success of LCLS [1] generates strong motivation and solid technical basis to extend its capabilities. The upgrade will extend x-rays wavelength range down to 0.06 nm. A new soft x‐ray adjustable‐gap undulator line will produce FEL with wavelengths up to 6 nm. To allow full electron beam rate and independent electron beam parameters in each line, a new injector and pair of bunch compressors will be added to the second kilometer of SLAC linac. The electron from this linac part will bypass the LCLS accelerator into the soft x‐ray undulators which can provide two FEL pulses with variable delay and photon energy and may be configured for narrow bandwidth pulse via self‐seeding. External seeding with the echo‐enabled harmonic generation can improve temporal coherence. The new bypass line can add multiple electron bunches within each RF pulse. LCLS‐II will provide polarization control and can incorporate the low‐charge, few‐femtosecond pulse duration operating mode. A THz radiation source will be included to provide x‐ray/THz pump‐probe capabilities. The schemes and parameters are based on measurements and experience at LCLS.
1. P. Emma et al., Nature Photonics (accepted, 2010).
Mesa+, Enschede
ELETTRA, Basovizza
UT-MESA+ OS, Enschede
FERMI@Elettra is a free electron laser user facility currently under construction at Sincrotrone Trieste S.C.p.A. Its goals are to produce high-brightness, ultra-short pulses with wavelengths ranging from 100 - 20 nm (FEL1) and 40 - 4 nm (FEL2) and deliver these pulses to a wide range of user experiments. Currently, FERMI uses the HGHG technique to improve both the stability and the longitudinal and spectral coherence of the output of the laser. Direct seeding of FEL1 using a High Harmonic (HH) source is also foreseen and allows a direct comparison between the two seeding methods. For an HH source, we will use neutral atoms in a hollow waveguide in combination with coherent control of the drive laser pulse to provide wavelength tuning as well as selective enhancement of the harmonic orders. For direct seeding of FEL 2 we propose HH generation from ions in a modulated plasma waveguide. The ions allow generation of shorter wavelengths, while the modulated plasma waveguide provides a long interaction length as well as quasi-phase matching for boosting the output energy of the source. In this paper, we will present the HH source for FEL1 as well as a concept for HH seeding of FEL2.
ANL, Argonne
X-ray free electron lasers (FELs) are well suited to pursue a long-standing goal of studying matter in a transient state that is far from equilibrium. This state often determines the functions of materials and, thus, holds a key to understanding how to control them. The natural time scale for most of the dynamic processes involving atoms is of the order of 100 femtoseconds, and existing x-ray FELs have already surpassed this mark. The natural time scale for dynamic processes driven by electrons is of the order of 100 attoseconds, and this is the next Rubicon for FELs. In this talk I will review the state of the art in generation of femtosecond x-ray pulses using FELs and will discuss a number of new ideas en route to sub-femtosecond x-ray pulses.
JLAB, Newport News, Virginia
A sophisticated research device such as an advanced photo-cathode injector for a high energy accelerator-based X-ray light source requires drive lasers with a flat-top shape both in time and space in order to generate high-quality short electron beam bunches. There are a number of different ways to spatially shape laser beams, but the practical methods for temporal shaping, in particular in the picosecond or femtosecond regime, are quite limited. One simple way to shape laser pulses is pulse stacking by birefringent crystals. This method has been adopted for several applications. While the method itself has the great advantage of simplicity, the overall performance depends on many factors. In this paper, we will present both analysis and a recent experimental study about important pulse shaping characteristics that, to our knowledge, have not been adequately explored before. Evaluation on the pros and cons of the method and how to improve the overall performance will be discussed.
RIKEN/SPring-8, Hyogo
JASRI/SPring-8, Hyogo-ken
An 8-GeV C-band (5712 MHz) accelerator is employed as a main accelerator for XFEL/SPring-8. Since a C-band accelerating structure generates a high accelerating gradient of higher than 35 MV/m, the total length of the accelerator fits within 400 m, including the injector and three bunch compressors. We use 64 C-band rf units, which consist of 128 accelerating structures, 64 rf pulse compressors and waveguide components, 64 klystrons and modulators, etc. Mass-production of the C-band rf components has been done by several Japanese manufacturers. The components reliability has been improved during the production, and all the components finally have excellent quality. The production quality was also confirmed by a high power rf test. We achieved the accelerating gradient of 40 MV/m without any problem. Since XFEL realizes high bunch compression with precise control of the energy chirp, the rf should be quite stable. We developed a high precision high voltage charger combined with a low-noise klystron modulator. The pulse-to-pulse stability of the PFN voltage was less than 0.01%. Installation of the components started in August 2009 and was now almost completed on schedule.
LBNL, Berkeley, California
SLAC, Menlo Park, California
The scientific potential of femtosecond x-ray pulses at linac-driven FELs such as the LCLS is tremendous. Time-resolved pump-probe experiments require a measure of the relative arrival time of each x-ray pulse with respect to the experimental pump laser. To achieve this, precise synchronization is required between the arrival time diagnostic and the laser which are often separated by hundreds of meters. For seeded FELs, synchronization is necessary between the seed and pump laser. We describe an optical timing system based on stabilized fiber links which has been developed for the LCLS. Preliminary results show stability of the timing distribution at the sub-20 fsec level. We present details of the results measured during LCLS operation for the first pump-probe experiment in October 2009 and the present user run starting in April 2010. We conclude with a discussion of potential for development.
DESY, Hamburg
The Andrzej Soltan Institute for Nuclear Studies, Centre Swierk, Swierk/Otwock
The Free electron LASer at Hamburg (FLASH) is a linear accelerator of 330m length. It provides laser pulses with pulse duration between 10 and hundreds fs in the soft X-ray wavelength range below 5nm produced in SASE process from electron bunches with an energy up to 1.2 GeV. FLASH works in pulse mode with repetition rate of 10 Hz where up to 800 bunches at a bunch spacing of 1 us are accelerated in one macro-pulse. The electron beam time structure is well suited for fast intra-train feedbacks using beam based measurements incorporated to the Low Level Radio Frequency (LLRF) control system of the accelerator structures to further improve the bunch compressions, bunch arrival and bunch energy stability directly impacting the quality of the FEL photon beam. In this paper, we present the beam based signal pre-processing, the implementation into LLRF system, the mandatory exception handling for robust operation and the imbedding of the real-time ~ 2us latency fast intra-train feedback with feedbacks for the removal of slow and repetitive errors. First results of the achieved intra-train bunch arrival and peak current stability will be presented together with observed limitations.
RIKEN/SPring-8, Hyogo
JASRI/SPring-8, Hyogo-ken
A pump-probe experiment at XFEL/SPring-8 is one of the most prominent parts to extract the future of a coherent short-pulse X-ray laser. A commercial Ti:Sapphire mode-locked laser is presently used as a pump laser, while a probe laser is the XFEL. However, the time jitter of the commercial mode locked laser, as which is caused by the noise of an electrical mode-locking circuit, is around several hundred femto-seconds. This jitter value is not sufficient for a temporal resolution requirement of our pump-probe experiment with a laser pulse width of several ten femto-seconds. To improve this time jitter, the method, using a 770 nm Ti:Sapphire laser amplifiers directly triggered by a 1540 nm master optical oscillator as a time reference signal source for an XFEL accelerator, was devised. This method could eliminate the noise caused by the electrical mode-locking circuit. The basic principle of the method was proved by a preliminary experiment with laser pulse manipulation employing an E/O crystal shutter with a several ten ps response. This presentation describes a basic idea of this pumpprobe method, a preliminary experiment set-up to check its feasibility, and experiment results.
DESY, Hamburg
Warsaw University of Technology, Institute of Electronic Systems, Warsaw
Bunch arrival-time monitors measure the arrival-time of each bunch in the electron bunch train at several locations at FLASH. The temporal reference for the monitors is provided by the optical synchronization system which distributes laser pulses with a repetition rate of 216 MHz and a length of around 200 fs FWHM. The pulses are delivered to the monitors with an arrival-time stability of about 10 fs. The bunch arrival-time is encoded as an amplitude modulation of a laser pulse from the optical synchronization system. These laser pulse amplitudes need to be sampled and processed together with additional input parameters. Because the arrival-time information is used in a feedback loop to adjust the accelerator fields, the signal processing, calibration and transmission of the bunch arrival-time information via a low-latency, high-speed link to an accelerator RF control station is needed. The most challenging problems of the signal processing are the synchronisation of several clock domains, regeneration and conversion of optical laser pulses, on-line calibration, and exception handling.
Monash University, Faculty of Science, Victoria
ASCo, Clayton, Victoria
From a detection perspective, short wavelength phase information is lost when event sizes exceed radiation wavelengths, making conventional holography impossible above a material-dependent quantum energy limit. Despite this, and prior to the invention of lasers or holography, Bragg's X-ray microscope* opened the door to optical computation in short-wavelength studies using spatially coherent visible light, including phase retrieval methods. This optical approach lost ground to semiconductor detection and digital computing in the 1960s. Since then, visible optics such as spatial light modulators, array detectors and femtosecond lasers have become widely available, routinely allowing versatile and computer-interfaced imposition of optical phase, detection, and molecular coherent control in pump-probe studies. Today, FELs begin to offer opportunities for atomic resolution and ultrafast studies. Thus we investigate an overlooked aspect of Bragg's X-ray microscope: the short-wavelength to visible-wavelength, incoherent to coherent conversion that is a necessary prerequisite for coherent optical computations. Some potential approaches, techniques and opportunities are outlined.
* W.L. Bragg, A new type of 'X-Ray microscope', Nature 143, 678 (1939)
Kyoto IAE, Kyoto
The bulk high temperature superconductor staggered array undulator (Bulk HTSC SAU) has potential to generate strong periodic magnetic field in short period and to control K value without a mechanical gap control structure.* However, availability of the bulk HTSC magnets having matched performance of critical current density is a problem to be solved. In this study, we have numerically and experimentally estimated influence of variation of critical density upon field error. It was numerically found that the field error was naturally compressed, because the difference in critical current density was compensated by natural variation of the region where the supercurrent flows. In the conference, the experimental results of the field error compression and principle of the compression will be discussed.
* R. Kinjo, et al., “BULK HIGH-TC SUPERCONDUCTOR STAGGERED ARRAY UNDULATOR”, Proceedings of FEL 2008, 473 (2009).
MSL, Stockholm
DESY, Hamburg
European XFEL GmbH, Hamburg
The Manne Siegbahn Laboratory at Stockholm University is currently involved in two separate projects at the European XFEL. The first concerns the fiducialization and characterization of the quadrupole magnets in the undulator sections. A recently upgraded rotating coil system measures the magnetic centre stability during magnet excitation, magnet gradient and field error components. In connection, a coordinate measuring machine is used to fiducialize the quadrupole magnetic centre to better than 0.050 mm. The second project concerns high precision measurements of the undulator temperature. The SASE radiation intensity depends strongly on the undulator period and the magnetic field strength, which are both sensitive to temperature. Instead of keeping the temperature within 0.1 degrees along the undulator tunnel, a temperature compensation scheme can be applied. Here, a change in temperature initiates adjustment of the undulator gap to compensate for changes in magnetic field. A system for undulator segment temperature measurement, with resolution of 0.03 degrees, necessary for the compensation scheme, is presented together with a brief overview of the upgraded rotating coil system.
Kyoto IAE, Kyoto
The bulk high temperature superconductor staggered array undulator (Bulk HTSC SAU) has several advantages: such as strong magnetic field, potential of short period undulator, K value variability without gap control. In addition to these advantages, the Bulk HTSC SAU can be used near the electron beam because the undulator is expected to show good performance at 20 30 K. In the conference, we will report the expected performance of the undulator at low temperature through magnetic measurement by using a superconducting quantum interference device (SQUID) magnetometer. Also we will report the results of the first operation at 4 77 K of new prototype undulator consisting of a helium cooling system and a 2 T superconducting solenoid.
ELETTRA, Basovizza
Technische Universität Berlin, Berlin
Physikalisches Institut Albert-Ludwig, Freiburg
Università di Roma "La Sapienza", Roma
The Low Density Matter beamline (LDM) at FERMI@Elettra is scheduled to begin operation in early 2011 as a large collaborative project for experiments on neutral matter beams, and later on trapped species and mass selected ions. FERMI@Elettra is a seeded source comprising two Free Electron Lasers(FELs) that will generate short pulses (25200fs) of VUV (FEL1:12-60eV) and XUV/soft-X-rays (FEL2:60-300eV; third harmonic: up to 900eV) with close-to-transform-limited transverse and longitudinal coherence, and full polarization control. It includes a synchronized broadly-tunable user laser for pump-probe experiments. LDM modular design seeks to exploit these unique features with a flexible choice of target system and detection method. It will supply intense beams of neutral atoms, closed-shell molecules, radicals, and pure/doped clusters (the latter ranging from ultracold helium nanodroplets, to atomic and molecular van der Waals clusters, especially water, to clusters of refractory materials such as metals and their oxides). These can be combined with a set of detectors, working in tandem when possible, for photoelectron/photoion spectroscopy, fluorescence emission, and photon scattering.