Linac Coherent Light Source

Update On The Linac Coherent Light Source (LCLS)

R. Tatchyn and H. Winick

This is an update on recent developments on the Linac Coherent Light Source (LCLS) project. The LCLS aims at using 15-20 GeV electrons from the last kilometer of the 3-kilometer SLAC linac to drive a short wavelength free-electron laser (FEL) using self-amplified spontaneous emission (SASE). The SASE approach opens up lasing possibilities at very short wavelength since it does not employ an optical cavity, which is difficult or impossible to make at short wavelengths. The last kilometer of the SLAC linac becomes available for this project when the PEP-II B-Factory, which uses only the first two kilometers of SLAC for injection, becomes operational in the 1998-99 time frame.

The possibility that the unique SLAC linac could be used to produce such a coherent x-ray beam in the 1-2 Å wavelength range has attracted the interest and participation of scientists from LBL, UCLA, the University of Milan and the University of Rochester. In addition, we are collaborating with groups at BNL and DESY which also have plans to develop SASE-based FELs using linacs at their laboratories.

The BNL project aims at the wavelength range down to about 1000 Å using an existing 210 MeV linac. There has been much collaboration with BNL, particularly in developing an advanced photocathode rf gun, essential to all SASE-based projects. This is described in more detail below.

The DESY project aims at the wavelength range down to about 60 Å using the 0.5-1 GeV TESLA Test Facility prototype superconducting linac now in construction at DESY as part of their linear collider R&D. Eventually, DESY plans an LCLS in the Å range using the much higher energy (~250 GeV) linac they propose for the final TESLA linear collider project.

Here is a brief update on the status of the various activities on LCLS R&D:

  1. Photocathode RF Gun : The collaboration among BNL, UCLA, and SLAC to produce an advanced photocathode rf electron gun is making good progress. The goal is to develop a gun that would produce one nanocoulomb of electrons at about 5 MeV in a 10 ps FWHM pulse with a normalized transverse emittance of 1 p millimeter-milliradian, which is required to reach the 1-2 Angström range in the LCLS with a 15-20 GeV electron beam. [Normalized emittance is the actual geometrical emittance times gamma, where gamma = E/(mc 2). In a linac, the emittance decreases linearly as the energy increases. Thus the required final emittance, close to the diffraction limited emittance given by l/4 p, can be reached at electron energies of 15-20 GeV if there is little or no emittance degradation during the acceleration and compression of the electrons from the gun.] Encouraging results have been reported at BNL, where an earlier version of this gun has reached a normalized emittance of about 1 pi millimeter-radian at low charge.
    Components of the photocathode RF sun before brazing. This gun is the result of a collaboration between Brookhaven National Laboratory, UCLA and SLAC. It is designed to reach a normalized mittance (which is defined as the actual emittance times the electron energy in rest mass units). Also shown are Dennis Palmer, a Stanford graduate student working on this gun as a PhD project with Professor Roger Miller of SLAC and Herman Winick.
    After several months of effort on microwave measurements, the detailed design of the new 1.6 cell S-band gun was completed at BNL in early April 1995. As of this writing (August 17, 1995) machining of the first gun has been completed at UCLA. The parts have been delivered to SLAC where cold tests ( i.e., tests at low RF power) and final machining will be done, after which the assembly will be brazed at SLAC and cold tested again. The gun will then be sent to BNL for initial tests at full RF power at the Accelerator Test Facility (ATF), using the low energy beam line which has been simulated and is now being designed. Integrated and slice emittance as a function of position will be measured, along with emittance growth due to field asymmetries which the gun has been designed to eliminate. Beam loading will also be studied. These studies will allow future guns to be unitized without the need for fully characterizing each gun.

    Following this, three more guns will be fabricated, one of which will come to SLAC for characterization using the gun test stand (see below). Most of this work is being carried out by Dennis Palmer, a Stanford graduate student working with Roger Miller, along with several researchers at BNL (Xi Jie Wang, Ken Bachelor, Ilan Ben-Zvi, Marty Woodle).

  2. Gun Test Stand: A gun test stand is under construction in the shielded vault of the SSRL 120 MeV injector linac. It is based on a 3-meter-long section of S-band linear accelerator which can accelerate the 5 MeV electrons from the gun to 30-40 MeV, which is required to demonstrate full emittance compensation. Following this is a diagnostic section in which the peak current, emittance and energy spread will be measured. Most of this work is being carried out by Michael Hernandez, a graduate student of Helmut Wiedemann, along with Jim Weaver and others at SSRL.

    The laser to drive the gun will be constructed with the help of a group from the University of Rochester, led by Adrian Melissinos and David Meyerhofer. Use will be made of an existing University of Rochester laser system that is part of their experiment at the SLAC Final Focus Test Beam (FFTB). A clean room enclosure to house the amplifier and frequency multiplier section of this laser will be constructed adjacent to the gun test stand in Building 140, the booster shelter. Alan Fisher at SLAC is providing much valuable advice on the laser system.

  3. 1.5 Å LCLS Undulator: This insertion device will be required to provide very strong distributed focusing (50 T/m-75 T/m) and a field quality of 0.1-0.3%. Whereas field qualities in this range have been attained in recently developed undulators at 3rd generation synchrotron storage rings, the focusing strength represents a new level of undulator specification and design.

    A second major consideration is the undulator length. Depending on the field strength that can be attained at a minimal period, LCLS simulations, performed by Ming Xie at LBL, have shown that the undulator length can vary from about 60 m (for a 1.2 T field and 3 cm period transverse undulator) to about 30 m (for a 1.7 T field and 2.2 cm period helical undulator). Given this broad range of parameters and field requirements, the sub-centimeter gap size at which the LCLS undulator can be operated has allowed the consideration of a number of technologies with which this device could be implemented, and these technologies constitute the main areas of the LCLS undulator R&D program.

    Specific examples under study include: 1) a canted/wedged pole hybrid/permanent magnet (PM) device introduced and under development by Ross Schlueter at LBL; 2) a standard hybrid/PM device with independently added strong-focusing lattices, two types of which are being developed, respectively, by Roman Tatchyn at SSRL and Alexander Varfolomeev at the Kurchatov Institute in Moscow; 3) a pulsed-copper structure developed by Roger Warren, previously at LANL; and 4) a bifilar-helical superconducting (SC) device being designed by Shlomo Caspi at LBL. The present emphasis is on the SC structure, as it features the shortest attainable length (30 m), at a potentially minimal cost. Studies of the likelihood of quenching in this device in SLAC's linac environment are presently underway using SLAC's EGS4 particle-scattering code under the direction of Ralph Nelson (developer of the code), with assistance from Roman Tatchyn and Ted Cremer, a physicist at SSRL.

  4. X-Ray Optics : There are three general areas of application for x-ray optics and instrumentation at the LCLS: 1) deflection and transportation of the photon pulses to experiments or beam lines; 2) diagnostic characterization of the spontaneous and coherent components of the LCLS output; and 3) extension of the LCLS intensity, coherence, temporal, spectral-angular, and polarization parameters beyond those of the unprocessed LCLS beam for selected experimental applications. In all three cases the extreme shortness (down to 160 fs) and high peak power (up to 100 GW) of the LCLS x-ray pulses strongly influence optical design and expected performance. The LCLS x-ray optics R&D program is presently concentrating on the theoretical study of short-pulse effects in relation to: 1) peak power damage effects in candidate optical materials and 2) the performance and design of instrumentation based on the principle of interference. Based on these investigations, the initial design of a grazing-incidence take-off mirror system has been completed, and designs of diagnostic instrumentation (a Michelson interferometer) and optical elements for experimental applications are underway in collaboration with John Arthur, SSRL and Gerd Materlik at DESY. The possibility of performing proof-of-principle experiments on 3rd-generation synchrotron storage rings ( e.g., APS, ESRF, and PETRA), both on peak power-damage effects and on the development of coherence techniques, is being explored.

  5. SASE Demonstration Experiments : Because there is little experimental data on the SASE process, experiments at longer wavelength to study the basic physics of the process and compare the results with simulation codes are underway at SSRL and also at BNL and UCLA. A group led by Helmut Wiedemann has observed exponential growth at 50 micron wavelength at the SUNSHINE facility, a small linac equipped with a thermionic rf gun on the Stanford Campus. In collaboration with Rodolfo Bonifacio from the University of Milan, Max Cornacchia, Bob Hettel, and Heinz-Dieter Nuhn are studying the use of the SSRL injector linac for such demonstration experiments in the few micron wavelength range.

    Most of the above described R&D activity is being carried out with the part-time help of several staff at SSRL and in the SLAC Technical Division. To augment these efforts and to add efforts on other aspects of the project which are not being worked on at present, such as the design of electron bunch length compressors, an opening has been posted for an accelerator physicist to work full time on LCLS R&D.