• Z. Zhang, R.D. Ruth, S.G. Tantawi
  SLAC, Menlo Park, California

A Bragg waveguide consisting of multiple dielectric layers with alternating index of refraction becomes an excellent option to form electron accelerating structure powered by high power laser sources. It provides confinement of a synchronous speed-of-light mode with extremely low loss. However, laser field can not be coupled into the structure collinearly with the electron beam. There are three requirements in designing input coupler for a Bragg electron accelerator: side-coupling, selective mode excitation, and high coupling efficiency. We present a side coupling scheme using a Bragg-grating-assisted input coupler to inject the laser into the waveguide. Side coupling is achieved by a second order Bragg grating with a period on the order of an optical wavelength. The phase matching condition results in resonance coupling thus providing selective mode excitation capability. The coupling efficiency is limited by profile mismatch between the outgoing beam and the incoming beam, which has normally, a Gaussian profile. We demonstrate a non-uniform distributed grating structure generating an outgoing beam with a Gaussian profile, therefore, increasing the coupling efficiency.

• C.D. Nantista, V.A. Dolgashev, R. Siemann, S.G. Tantawi, J. Weisend
  SLAC, Menlo Park, California
• I.E. Campisi
  ORNL, Oak Ridge, Tennessee

Superconducting rf cavities are increasingly used in accelerators. Gradient is a parameter of particular importance for the ILC. Much progress in gradient has been made over the past decade, overcoming problems of multipacting, field emission, and breakdown triggered by surface impurities. However, the quenching limit of the surface magnetic field for niobium remains a hard limitation on cavity fields sustainable with this technology. Further exploration of materials and preparation may offer a path to surpassing the current limit. For this purpose, we have designed a resonant test cavity. One wall of the cavity is formed by a flat sample of superconducting material; the rest of the cavity is copper or niobium. The H field on the sample wall is 74% higher than on any other surface. Multipacting is avoided by use of a mode with no surface electric field. The cavity will be resonated through a coupling iris with high-power rf at superconducting temperature until the sample wall quenches, as detected by a change in the quality factor. This experiment will allow us to measure critical magnetic fields up to well above that of niobium with minimal cost and effort.

• V.A. Dolgashev, C.D. Nantista, S.G. Tantawi
  SLAC, Menlo Park, California
• Y. Higashi, T. Higo
  KEK, Ibaraki

Operating accelerating gradient in normal conducting accelerating structures is often limited by rf breakdown. The limit depends on multiple parameters, including input rf power, rf circuit, cavity shape and material. Experimental and theoretical study of the effects of these parameters on the breakdown limit in full scale structures is difficult and costly. We use 11.4 GHz single-cell traveling wave and standing wave accelerating structures for experiments and modeling of rf breakdown behavior. These test structures are designed so that the electromagnetic fields in one cell mimic the fields in prototype multicell structures for the X-band linear collider. Fields elsewhere in the test structures are significantly lower than that of the single cell. The setup uses matched mode converters that launch the circular TM01 mode into short test structures. The test structures are connected to the mode launchers with vacuum rf flanges. This setup allows economic testing of different cell geometries, cell materials and preparation techniques with short turn-around time. Simple 2D geometry of the test structures simplifies modeling of the breakdown currents and their thermal effects.

• J. Guo, S.G. Tantawi
  SLAC, Menlo Park, California

In this paper, we present the recent results of our research on the high power ultra-fast silicon RF switches. This switch is composed of a group of PIN diodes on a high purity SOI (silicon on oxide) wafer. The wafer is inserted into a cylindrical waveguide under T·10⁰¹ mode, performing switching by injecting carriers into the bulk silicon. Our current design is using a CMOS compatible process and the fabrication is accomplished at SNF (Stanford Nanofabrication Facility). This design is able to achieve sub-100ns switching time, while the switching speed can be improved further with 3-D device structure and faster circuit. Power handling capacity of the switch is at the level of 10MW. The switch was designed for active X-band RF pulse compression systems - especially for NLC, but it is also possible to be modified for other applications and other frequencies such as L-band.

• S.H. Gold
  NRL, Washington, DC
• H. Chen, Y. Hu, Y. Lin, C. Tang
  TUB, Beijing
• W. Gai, C.-J. Jing, R. Konecny, J.G. Power
  ANL, Argonne, Illinois
• A.K. Kinkead
  ,
• C.D. Nantista, S.G. Tantawi
  SLAC, Menlo Park, California

This paper will describe a joint project by the Naval Research Laboratory (NRL) and Argonne National Laboratory (ANL), in collaboration with the Stanford Linear Accelerator Center (SLAC), to develop a dielectric-loaded accelerator (DLA) test facility powered by the high-power 11.424-GHz magnicon that was developed by NRL and Omega-P, Inc. The magnicon can presently produce 25 MW of output power in a 250-ns pulse at 10 Hz, and efforts are in progress to increase this to 50 MW.* The facility will include a 5-MeV electron injector being developed by the Accelerator Laboratory of Tsinghua University in Beijing, China. The DLA test structures are being developed by ANL, and some have undergone testing at NRL at gradients up to ~8 MV/m.** SLAC is developing a means to combine the two magnicon output arms, and to drive an injector and accelerator with separate control of the power ratio and relative phase. The installation and testing of the first dielectric-loaded test accelerator, including injector, DLA structure, and spectrometer, should take place within the next year. The initial goal is to produce a compact 20-MeV dielectric-loaded test accelerator.

*O. A. Nezhevenko et al., Proc. PAC 2003, p. 1128.**S. H. Gold et al., AIP Conf. Proc. 691, p. 282.

• S. Doebert, C. Adolphsen, G.B. Bowden, D.L. Burke, J. Chan, V.A. Dolgashev, J.C. Frisch, R.K. Jobe, R.M. Jones, R.E. Kirby, J.R. Lewandowski, Z. Li, D.J. McCormick, R.H. Miller, C.D. Nantista, J. Nelson, C. Pearson, M.C. Ross, D.C. Schultz, T.J. Smith, S.G. Tantawi, J.W. Wang
  SLAC, Menlo Park, California
• T.T. Arkan, C. Boffo, H. Carter, I.G. Gonin, T.K. Khabiboulline, S.C. Mishra, G. Romanov, N. Solyak
  Fermilab, Batavia, Illinois
• Y. Funahashi, H. Hayano, N. Higashi, Y. Higashi, T. Higo, H. Kawamata, T. Kume, Y. Morozumi, K. Takata, T. T. Takatomi, N. Toge, K. Ueno, Y. Watanabe
  KEK, Ibaraki

During the past five years, there has been an concerted effort at FNAL, KEK and SLAC to develop accelerator structures that meet the high gradient performance requirements for the Next Linear Collider (NLC) and Global Linear Collider (GLC) initiatives. The structure that resulted is a 60-cm-long, traveling-wave design with low group velocity (< 4% c) and a 150 degree phase advance per cell. It has an average iris size that produces an acceptable short-range wakefield in the linacs, and dipole mode damping and detuning that adequately suppresses the long-range wakefield. More than eight such structures have operated over 1000 hours at a 60 Hz pulse rate at the design gradient (65 MV/m) and pulse length (400 ns), and have reached breakdown rate levels below the limit for the linear collider. Moreover, the structures are robust in that the breakdown rates continue to decrease over time, and if the structures are briefly exposed to air, the rates recover to their low values within a few days. This paper presents a final summary of the results from this program, which effectively ended last August with the selection of ‘cold’ technology for a next generation linear collider.