Search

We are exploring the physics of vacuum rf breakdowns in high gradient metallic accelerating structures at sub-THz frequencies. We present the experimental results of rf tests performed in beam-driven travelling-wave accelerating structures. We measured the breakdown rates, the frequency of the emitted radiation, and the gradients. As the next step in this study, we are developing accelerating structures powered by an rf source. We report on the status of this project.

Silicon woodpile photonic crystal provides a three-dimensional dielectric waveguide system for high-gradient laser driven acceleration. The woodpile waveguide is periodically loaded in the longitudinal direction; therefore simple cross-sectional mode profile matching is not sufficient to launch the accelerating mode appropriately and will result in significant scattering loss. Hinted by the common nature of longitudinal periodicity between disk-loaded waveguide and woodpile waveguide, several coupler design schemes developed for multi-cell RF cavity are implemented in the woodpile accelerator design. Among them there are the travelling-wave match method based on S-matrix, the periodic VSWR method, and the TE-to-TM coupling iris design. This paper presents design procedures and simulation results using these methods. According to simulations, nearly 100% power transmission between SOI and woodpile waveguides with a travelling-wave match is achieved with a specially designed mode-launching coupler. Constructed by silicon rods extruding into the defect waveguide, the coupling iris provides necessary transition from TE mode to TM accelerating mode, also with negligible coupling loss.

We describe a comprehensive project, funded under the DARPA AXiS program, to develop an all-optical table-top X-ray FEL based on dielectric acceleration and electromagnetic undulators, yielding a compact source of coherent X-rays for medical and related applications. The compactness of this source demands that high field (>GV/m) acceleration and undulation-inducing fields be employed, thus giving rise to the project’s acronym: GV/m AcceLerator And X-ray Integrated Experiment (GALAXIE). There are numerous physics and technical hurdles to surmount in this ambitious scenario, and the integrated solutions include: a biharmonic photonic TW structure, 200 micron wavelength electromagnetic undulators, 5 μm laser development, ultra-high brighness magnetized/asymmetric emittance electron beam generation, and SASE FEL operation. We describe the overall design philosophy of the project, the innovative approaches to addressing the challenges presented by the design, and the significant progress towards realization of these approaches in the nine months since project initialization.

We present beam dynamics studies of a proposed circularly polarized RF undulator cavity to be operated at 50 MW to yield the undulator strength parameter K = 0.7. A symplectic integration scheme is used for numerical studies. The expressions for field integrals of the RF undulator are also developed. Simulation results for the electron trajectories, showing off-axis drift and bunch defocusing, for 60 and 120 MeV beam energies are presented.

Niobium (Nb) superconducting radio-frequency (SRF) cavities have become the technology of choice for many recent and future particle accelerators. Increasing their accelerating gradient beyond the theoretical limit of Nb (~50 MV/m) by coating multilayer thin film superconductors is our ultimate goal. This idea proposed by Gurevich in 2005 is based on the assumption that the lower critical magnetic field parallel to the surface (Bc1//) increases with very thin films having a thickness close to its magnetic penetration depth. We measured the magnetic field Bpen at which a large number of vortices or fluxons start to penetrate into the material using a SQUID magnetometer. While we found that the measurements for the films thinner than ~200 nm is difficult due to insufficient signal strength and alignment accuracy, we also found that even the films with a thickness a few times the penetration depth show significant increase in Bpen, e.g., ~135 mT and ~210 mT for 500 nm and 300 nm films at 4.5 K, respectively, as compared to ~46mT for bulk MgB2. RF measurements of 50.8 mm diameter MgB2(100 nm)/Al2O3(20 nm)/Nb sample using 11.4 GHz pulsed power and a TE013-mode cavity showed quenches at ~42 mT at 4 K and ~34 mT at 10 K. These results were significantly lower than the numbers predicted from the low-power DC measurement results mentioned earlier, but these were found to be thermal quenches due to a high RF surface resistance caused by the inter-diffusion of coated components.

We are investigating a standing wave accelerator structure that uses a rf feed to each individual cell. This approach minimizes rf power flow and electromagnetic energy absorbed by an rf breakdown. The objective of this work is a robust high-gradient (above 100 MV/m) Xband accelerator structure.

An X‐band RF cryogenic material testing system has been developed in the past few years. This system employs a high‐Q copper cavity with an interchangeable flat bottom working under a TE013 like mode. By measuring the cavity Qs with a network analyzer, the system can characterize the surface resistance of different samples at different temperatures. Using a 50 MW 2μs pulsed klystron, the system can measure the quenching H field for superconducting samples, up to 300–400 mT. In this paper, we will present the most recent developments of the system and testing results.

In this paper, we will present our recent results on the research of the ultra-fast high power RF switches based on silicon. We have developed a switch module at X-band which can use a silicon window as the switch. The switching is realized by generation of carriers in the bulk silicon. The carriers can be generated electrically or/and optically. The electrically controlled switches use PIN diodes to inject carrier. We have built the PIN diode switches at X-band, with

We discuss the various aspects of the ILC RF system. We suggest a new RF distribution system that does not require circulators at each cavity port. We present a new type of fundamental mode couplers that eliminates electric field lines from most of the coupler wall — hence, we expect an improved reliability. We present a new technique for testing and conditioning these couplers. Finally, we discuss the overall design of the RF system and related economic issues.

Summarized results obtained in several experiments on active RF pulse compression at X‐band using the Omega‐P/NRL magnicon are presented. In these experiments RF energy is stored in two parallel‐fed resonators and then discharged using switches activated with rapidly‐fired plasma discharge tubes. New results with both transmission and reflection switches are reported, with peak power gains in the range of 8–10. The application of developed plasma switches for an electrically controlled switching of an existing passive pulse compressor SLEDII for increasing its efficiency is discussed. The low‐power test of the plasma switch with a 35 m SLEDII delay line is described.

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 TE01 mode, performing switching by injecting carriers into the bulk silicon. Our current design use a CMOS compatible process and the device was fabricated 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.

This paper presents a progress report on 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 test accelerator in the magnicon facility at NRL. The accelerator will be powered by an experimental 11.424‐GHz magnicon amplifier that presently produces 25 MW of output power in a ∼250‐ns pulse at up to 10 Hz. The accelerator will include a 5‐MeV electron injector originally developed at the Tsinghua University in Beijing, China, and can incorporate DLA structures up to 0.5 m in length. The DLA structures are being developed by ANL, and shorter test structures fabricated from a variety of dielectric materials have undergone testing at NRL at gradients up to ∼8 MV/m. SLAC has developed components to distribute the power from the two magnicon output arms to the injector and to the DLA accelerating structure with separate control of the power ratio and relative phase. RWBruce Associates, Inc., working with NRL, has investigated means to join short ceramic sections into a continuous accelerator tube by a brazing process using an intense 83‐GHz beam. The installation and testing of the first dielectric‐loaded test accelerator, including injector, DLA test structure, and spectrometer, should take place within the next year.

For two sub‐systems with a common bi‐periodic (2D) perforated wall, conditions for selective coupling between modes are analyzed. Two models are discussed: 1) a directional coupler between a cylindrical waveguide operating in a high‐order mode and a whispering‐gallery mode of an outer coaxial waveguide and 2) a launcher into a gyrotron cavity of a high‐order rotating mode, using a coaxial perforated waveguide insert.

Experience gained during the recent operation of high power 11.424 GHz rf sources for accelerators led to new, stricter requirements for system components. One of the basic components of such a system is a mode converter that transforms the rectangular waveguide mode into the TE01 circular waveguide mode. With such a converter, it is possible to minimize the use of WR90 rectangular waveguide which was shown to be a weak part of the previous system at power levels higher than 100 MW and pulse lengths on the order of a microsecond. We used several methods to design a compact mode converter with extremely low parasitic mode conversion. These methods employ HFSS and include multi‐parameter searches, concurrent optimization with a mode‐matching code CASCADE, cascading of resulting S‐matrices, and tolerance analyses. This paper describes the design methods and presents results.

RF breakdown is one of the major factors determining performance of high power rf components and rf sources. RF breakdown limits working power and produces irreversible surface damage. The breakdown limit depends on the rf circuit, structure geometry, and rf frequency. It is also a function of the input power, pulse width, and surface electric and magnetic fields. In this paper we discuss multi‐megawatt operation of X‐band rf structures at pulse width on the order of one microsecond. These structures are used in rf systems of high gradient accelerators. Recent experiments at Stanford Linear Accelerator Center (SLAC) have explored the functional dependence of breakdown limit on input power and pulse width. The experimental data covered accelerating structures and waveguides. Another breakdown limit of accelerating structures was associated with high magnetic fields found in waveguide‐to‐structure couplers. To understand and quantify these limits we simulated 3D structures with the electrodynamics code Ansoft HFSS and the Particle‐In‐Cell code MAGIC3D. Results of these simulations together with experimental data will be discussed in this paper.

We will review the design of the dual‐mode X‐band rf system proposed for the Next Linear Collider (NLC). Recent experimental data are presented. The system is to produce 400 ns pulses with power levels up to 600 MW. A proof‐of‐principle experiment is being constructed at SLAC. Four 50 MW klystrons will power a fully dual‐moded resonance delay line pulse compression system. Both the transfer line and the delay lines are dual‐moded. The modes carried by the transfer line are controlled by the rf phases of the different klystrons. The modes in the delay lines are controlled by a set of mode converters at the input and the end of each delay line. By manipulating the modes in the transfer line, one can achieve either no pulse compression or a pulse compression ratio of 4. The total output power will be 200 MW for 1.6 microseconds or 600 MW for 400 nanoseconds.

In recent years, R&D for pulse compression and power distribution systems for the Next Linear Collider has led to the invention of many novel rf components, some of which must handle up to 600 MW of pulsed power at X‐band. These include passive waveguide components, active switch designs, and non‐reciprocal devices. Among the former is a class of multi‐moded, highly efficient rf components based on planar geometries with overmoded rectangular ports. Multi‐moding allows us, by means of input phasing, to direct power to different locations through the same waveguide. Planar symmetry allows the height to be increased to improve power handling capacity. Features that invite breakdown, such as coupling slots, irises and H‐plane septa, are avoided. This class includes hybrids, directional couplers, an eight‐port superhybrid/dual‐mode launcher, a mode‐selective extractor, mode‐preserving bends, a rectangular mode converter, and mode‐mixers. We are able to utilize such rectangular waveguide components in systems incorporating low‐loss, circular waveguide delay lines by means of specially designed tapers that efficiently transform multiple rectangular waveguide modes into their corresponding circular waveguide modes, specifically TE10 and TE20 into circular TE11 and TE01. These extremely compact tapers can replace well‐known mode converters such as the Marie type. Another component, a reflective TE01‐TE02 mode converter in circular waveguide, allows us to double the delay in reflective or resonant delay lines. Ideas for multi‐megawatt active components, such as switches, have also been pursued. Power‐handling capacity for these is increased by making them also highly overmoded. We present a design methodology for active rf magnetic components which are suitable for pulse compression systems of future X‐band linear colliders. We also present an active switch based on a PIN diode array. This component comprises an array of active elements arranged so that the electric fields are reduced and the power handling capability is increased. Novel designs allow these components to operate in the low‐loss circular waveguide TE01 mode. We describe the switching elements and circuits.

Increasing the power handling capabilities of rf components is an important issue for the design of rf accelerators and rf sources. RF breakdown is a phenomena that limit the high power performance. A major concern is the damage that can occur in rf components from breakdown. To better understand this damage, we have studied rf breakdown in a rectangular waveguide experimentally and theoretically. The breakdown process in a waveguide is both easier to measure and simulate than breakdown in a complex geometry such as an accelerating structure. We used a particle tracking code and a Particle‐In‐Cell code to model the breakdown behavior. Models developed for the waveguide were applied to the breakdown in accelerating structures. RF breakdown in traveling wave and standing wave accelerating structures was simulated. We compare the experimental data with results of the simulations for the accelerating structures.

The Delay Line Distribution System (DLDS) is an alternative to conventional pulse compression, which enhances the peak power of rf sources while matching the long pulse of those sources to the shorter filling time of accelerator structures. We present an implementation of this scheme that combines pairs of parallel delay lines of the system into single lines. The power of several sources is combined into a single waveguide delay line using a multi-mode launcher. The output mode of the launcher is determined by the phase coding of the input signals. The combined power is extracted from the delay line using mode-selective extractors, each of which extracts a single mode. Hence, the phase coding of the sources controls the output port of the combined power. The power is then fed to the local accelerator structures. We present a detailed design of such a system, including several implementation methods for the launchers, extractors, and ancillary high power rf components. The system is designed so that it can handle the 600 MW peak power required by the NLC design while maintaining high efficiency.

We present a design methodology and analysis for an SPDT switch that is capable of handling hundreds of megawatts of power at X-band. The switch is designed for application in high power RF systems in particular future linear colliders. In these systems switching needs to be fast in one direction only. We use this to our advantage to reach a design for a super high power switch. In our analysis we treat the problem from an abstract point of view. We introduce a unified analysis for the microwave circuits irrespective of the switching elements. The analysis is, then, suitable for different kinds of switching elements such as photoconductors, PIN diodes, and plasma discharge in low-pressure gases.

This paper describes the design and performance of the XL-1 klystron; a 50 MW klystron operating at a frequency of 11.424 GHz for use on the SLAC Next Linear Collider Test Accelerator (NLCTA). Problems associated with the development of high-power rf sources for NLC, and the solutions implemented on XL-1 are discussed.

X‐Band and K‐Band gyroklystrons are being evaluated for possible application to future linear colliders. So far we have examined then different two‐ and three‐cavity configurations. We have achieved a maximum peak power of 27 MW in ∼1 μs pulses at a gain of 36 dB and an efficiency exceeding 32%. The nominal parameters include a 430 kV, 150–200 A beam with an average perpendicular to parallel velocity ratio near one. In this paper, we detail our progress to data and describe our plans for future experiments that should culminate in amplifier outputs in excess of 100 MW in 1 μs pulses.

The concept and development of gyroklystron amplifiers with parameters relevant to providing rf driver tubes for TeV linear colliders is reviewed. Phase noise is shown to be acceptable for the collider application. Recent experimental results on a 3‐cavity, 9.87‐GHz gyroklystron (viz., 24 MW output power in a 1.7‐μs pulse with 36% efficiency and 33‐dB gain) are presented. These results establish the gyroklystron as a microwave amplifier with uniquely attractive capabilities. Prospects for scaling performance to still higher power, higher frequency and higher gain and efficiency are bright; advanced concepts for scaling are discussed.

The development of future linear colliders operating at ≳1 TeV center of mass energy will likely depend on the availability of microwave amplifiers which are both affordable and are capable of significantly surpassing the performance of klystrons in terms of power and frequency. One of the most promising concepts for the required amplifiers is the gyroklystron. This paper reviews the physical principles of the gyroklystron and contrasts its phase bunching behavior with that of a klystron. Experiments in a low power gyroklystron which have studied phase noise and penultimate cavity tuning are summarized. The stages in developing an X‐band gyroklystron resulting in the achievement of 29 MW, 1 μsec output pulses with 34% efficiency and 36 dB gain is recapitulated. Finally, plans are described for experiments which will extend gyroklystron performance to the ∼100 MW level at X‐band, and explore operation at a frequency of ∼20 GHz by operating the output cavity at the second harmonic of the electron cyclotron frequency.