[en] This publication is the collection of the paper presented at the title meeting. The 28 of the presented papers are indexed individually. (J.P.N.)

[en] RF pulse compression is a method of exchanging the long low-power rf pulse by a short pulse of higher peak power. This is essential for the Future Linear Collider, where the required peak power of about 100 MW per meter of accelerator can not be produced with conventional rf sources. The most advanced methods of rf pulse compression are described. Theories are briefly presented, with their special features. Various schemes are illustrated with their specific designs and compared for particular collider configurations. (author)

[en] In this article, we have attempted to review the current status of knowledge on the subject of field emission and rf breakdown in high-gradient room-temperature accelerator structures. The reader will have noticed that this field of research is very dynamic, that continuous progress is being made, and that some basic and practical questions still remain to be answered. A brief summary is presented below. 1. The basic phenomenon of electron field emission as a source of dark current and as a trigger of rf breakdown in accelerator structures is well-established. What is still missing is a deeper understanding of what exactly makes up the value of the enhancement factor β and how it is to be correlated with, and ultimately predicted from an observable surface condition. When many similar surface emitters are present, is the measured β the result of the sum of these emitters or is it always dominated by one of them? 2. In the fairly well-established chain of events that leads to rf breakdown, is the sudden burst of gas and ion formation the ultimate enhancer of the local microscopic field which triggers a breakdown event? 3. Why does the surface field at which breakdown occurs depend on rf frequency when the classical Fowler-Nordheim expression for field emission is independent of frequency? Is this empirical frequency dependence real, or is it perhaps influenced by the fact that the rf pulse lengths used in experiments are generally shorter at higher frequency? On the other hand, in those experiments where the pulse length was varied at a given rf frequency, why is it that pulse length, admittedly over a small available range, did not have much of an effect on the breakdown threshold? Or perhaps, is the frequency dependence just due to the fact that higher frequency structures are smaller and therefore contain fewer defects and impurities? 4. Finally, from a practical point of view, what must be done to limit dark current and push the threshold of breakdown to the highest possible level? It is clear that every step in the fabrication, cleaning and installation of an accelerator structure can affect these two parameters: i.e., the raw material, machining practices, brazing, cleaning methods, environmental conditions during assembly and installation, in situ baking, the number of impurities per cm² left on the surface before the structure is pumped down, and the ultimate rf processing schedule. What is not clear is which of these steps are more crucial than others. Indeed, building an accelerator structure is a complicated and costly process, and eliminating unnecessary steps is very important, particularly for machines that may require thousands of accelerator sections. Some very interesting research still lies ahead. (author)

[en] This lecture has two parts: waveguides and cavities. Basic topics are discussed which can serve as bases for the following lectures. Many of the results obtained in the first part concerning waveguides are applied in the second part in the discussion on cavity properties, since a cavity can be considered as a part of a waveguide, a cavity resonant mode being a superposition of two counter-travelling waves in the waveguide. In deriving most of the mathematical formulas, complex number representation - that is, phasor forms - are used. For the final results, however, real number representations are also provided as much as possible as an aid to a more intuitive understanding. (author)

[en] The transient response of periodic structures is analyzed numerically by a coupled-resonator model. The traveling and the standing waves are approximated by superposition of the cell-mode (eigenmode in an isolated boundary) in each cell. The wave propagation is modeled by the signal transmission along the equivalent-circuit chain through the mutual inductance. A computer simulation code was developed to solve the time-dependent circuit equations using the difference equation. The code was used to study the transient beam-loading in a disk-loaded accelerating structure, and its compensation method was tested. This method has also been used to analyze the rf-pulse compressor as well as the traveling-wave output-circuit in a high-power klystron inside a particle-in-cell klystron simulator. This lecture describes the basic concept of the coupled-resonator model, application to the disk-loaded structure, and some details of the numerical method. (author)

[en] Superconducting RF cavities excel in applications requiring continuous waves or long pulse voltages. Since power losses in the walls of the cavity increase as the square of the accelerating voltage, copper cavities become uneconomical as demand for high continuous wave voltage grows with particle energy. For these reasons, RF superconductivity has become an important technology for high energy and high luminosity accelerators. The state of art in performance of sheet metal niobium cavities is best represented by the statistics of more than 300 5-cell, 1.5-GHz cavities built for CEBAF. Key aspects responsible for the outstanding performance of the CEBAF cavities set are the anti-multipactor, elliptical cell shape, good fabrication and welding techniques, high thermal conductivity niobium, and clean surface preparation. On average, field emission starts at the electric field of 8.7 MV/m, but there is a large spread, even though the cavities received nominally the same surface treatment and assembly procedures. In some cavities, field emission was detected as low as 3 MV/m. In others, it was found to be as high as 19 MV/m. As we will discuss, the reason for the large spread in the gradients is the large spread in emitter characteristics and the random occurrence of emitters on the surface. One important phenomenon that limits the achievable RF magnetic field is thermal breakdown of superconductivity, originating at sub-millimeter-size regions of high RF loss, called defects. Simulation reveal that if the defect is a normal conducting region of 200 mm radius, it will break down at 5 MV/m. Producing high gradients and high Q in superconducting cavities demands excellent control of material properties and surface cleanliness. The spread in gradients that arises from the random occurrence of defects and emitters must be reduced. It will be important to improve installation procedures to preserve the excellent gradients now obtained in laboratory test in vertical cryostats. (Y. Tanaka)

[en] This paper treats Thermionic emission, Cathode as an e^- emitter, Space-charge limited effect and 3/2 power law, Perveance, Beam spread due to space charge, Pierce guns, Magnetically immersed guns, Method of gun design including simulations, and Examples, mainly treating E3786, which attendees will operate above 1 MW-CW in a practical exercise course at KEK. (author). 74 refs

[en] The mechanism of a ferrite-loaded rf cavity is explained from the point of view of its operation. Then, an analysis of the automatic cavity-tuning system is presented using the transfer function; and a systematic analysis of a beam-feedback system using transfer functions is also presented. (author)

[en] The relationship between advanced accelerator research and future directions for particle physics is discussed. Comments are made about accelerator research trends in hadron colliders, muon colliders, and e⁺e^- linear colliders. (author)

[en] Some properties of the waves propagating in a constant cross-section tube are recalled, in particular the phase and group velocities of the guided wave. The dispersion diagram of a periodically-loaded waveguide is then derived by considering it either as a smooth tube perturbed with obstacles or as a chain of coupled resonators. Some examples of travelling-wave structures are presented and, finally, the interaction of the structure with beam is analyzed. This includes transit-time-factor effects as well as beam loading. (author)