A major redesign of the University of Manitoba cyclotron and a study for axial injection of ions into the Princeton University AVF cyclotron
The concept of a cyclotron was visualized by Lawrence in 1930 [LAW30], who then implemented the first prototype model at Berkeley which produced a proton beam of 1 MeV [LAW32]. This successful acceleration of particles stimulated many laboratories in many countries around the world, hence more higher energy cyclotrons were built during the next two decades, culminating with the University of Birmingham cyclotron which could accelerate a proton beam up to about 40 MeV. However, the classical cyclotron, as it is usually called nowadays, imposed a limitation on the maximum obtainable energy because of two mutually contradictory requirements on the magnetic field ([LIV61], [LIV62]). Since the ion rotates a hundred turns or more to reach its maximum energy, the motion inside a cyclotron must be stable in both the radial and the axial directions. In the axial direction, the ion has to feel a force toward the median plane and this requires that the magnetic field decrease with radius ([LIV61],[LIV62]). However, if the cyclotron resonance frequency, w = qB/m, were to be constant, the magnetic field B would have to increase with radius in order to compensate for the relativistic increase in mass m with velocity. Though attempts were amde to solve this problem, eventually the best solution was devised by Thomas, which led to the appearance of the relativistic cyclotron. In his paper published in 1938 [THO38], Thomas formulated the additional focusing term by employing the idea of an azimuthally varying magnetic field (AVF). The focusing term due to the AVF component turned out to be independent of the requirement on isochronism. Therefore isochronism can be retained by increasing the average magnetic field with radius, while the axial focusing force can be obtained from the AVF component of the magnetic field. The cyclotron based on this principle is called an AVF cyclotron. Later, Kerst also discovered that further axial focusing could be obtained when spirally shaped magnet sectors were employed. This idea was suggested by the principle of edge focusing in the theory of beam optics [LAW77]. The term due to the spiral focusing was again found to be independent of the requirement on isochronism, and therefore a spiral-ridge cyclotron emerged. The advent of the idea of the relativistic cyclotron gave an impetus to build new cyclotrons in many laboratories around the world since the 1950's. Both the University of Manitoba cyclotron and the Princeton University cyclotron (whose upgrading studies form the two main constituents of this thesis) were built in the early and in the late 1960s, respectively. From then till the present time, they have been continuously devoted to fundamental subatomic physics investigations. In the two decades that have passed since these two cyclotrons were built, there has been a significant advance in cyclotron technology, progress that was in part due to the appearance of solid-state electronic devices and the emergence of powerful computers. The solid-state based instrumentation made it possible to carry out precise mapping and analysis of the cyclotron magnetic field. The powerful new generation of computers enabled one to calculate the RF electric field distribution inside a cyclotron and then to trace particles' trajectories, from the ion source to the extraction radius, under the influence of this calculated electric field and the measured magnetic field. The accuracy of these calculations is much better than what could be obtained at the birth of these first-generation AVF cyclotons. In a separate development, nuclear physics experiments became more and more sophisticated during the same time span, resulting in needs for beams of higher intensity and better quality. In response to this demand, together with the availability of enormously more advanced cyclotron technology of the 1980's, many laboratories around the world began to embark upon ambitious upgrading programs for their cyclotrons. The University of Manitoba Cyclotron laboratory was no exception. In fact, members of the machine development group at this laboratory started investigating the possibility of such an upgrading as early as 1976, when they carried out exploratory magnetic field mappings. Later, in 1982, more elaborate field mappings [DER83], based on computer-aided technology were performed with much higher precision. The result was quite encouraging; it convinced us that a substantial improvement in beam quality would be achievable by upgrading the cyclotron. This finally led to a decision to initiate an extensive and intensive improvement program for the cyclotron, a project which was started in 1984. The author's contribution to the project dates from 1982 when he analyzed the 1982 field mapping data. During the next two years, the author was engaged in design studies for a new central region of the cyclotron (based on the 1982 field mapping data). The 1984 upgrading program included, among other projects, a series of field mappings and shimmings in which some 80 field maps were taken as successively better magnetic shims were installed. The result was better than expected; an H- beam was subsequently accelerated and extracted without having to retune the cyclotron parameters from design values. The author's contribution to this project was to analyze the measured data and then to suggest the improved shape and position of the magnetic shims for the next mapping. After completion of this mapping program, the author then refined the design study for the central region based on the new magnetic field data. The Princeton University AVF Cyclotron Laboratory also turned its attention to the type of improvements described earlier. This cyclotron is noted for its single-turn extraction capability. It can accelerate high charge-state light heavy ions as well as protons and deuterons. In the past, such ions were provided by an internal PIG (Penning Ionization Gauge) [BEN69] source. Naturally, however, the research needs increasingly required higher beam intensity and energy as well as better beam quality. The development in the 1970s of an ECR (Electron Cyclotron Resonance) source [GEL79] started to have a major effect on cyclotrons. This source cannot be installed inside the cyclotron. However, its capability for producing high quality, intense beams fo high-charge state light heavy ions made it far more profitable to externally inject the beam from the ECR source into the cyclotron than to rely on the internal PIG source. Another area of interest is the study of spin-dependent nuclear interactions, an area which requires polarized beams. Such beams with reasonable intensity can only be obtained by the external injection of polarized ions into the cyclotron. Thus, the Princeton University AVF Cyclotron Laboratory decided to initiate a feasibility study for converting the cyclotron to external injection of ions. The central question was whether the single-turn extraction capability could still be retained at the same time as achieving an excellent overall beam transmission efficiency through the cyclotron. An accurate assessment of these points necessitated an extensive computer-based design study of the central region based on the beam orbit dynamics investigations inside the cyclotron, an investigation which the author has performed since July 1984. This thesis consists of two independent parts under the following headings: (1) A major redesign of the University of Manitoba cyclotron, and (2) A study for axial injection of ions into the Princeton University AVF cyclotron. The first part discusses the magnetic field mapping and the design study of the new central region of the University of Manitoba cyclotron as integral parts of the major upgrading projects. At the beginning, a brief historical background of the University of Manitoba cyclotron is presented. Then the motivation for and the importance of, the cyclotron improvement program are presented in some detail. A considerable amount of space is allocated to describe the methods and the results of the new magnetic field mappings and the shimming program performed in 1984. The need for a new central region and the design study of it, as well as the beam orbit dynamics for H- ions are then delineated. This part concludes with a presentation of the markedly improved performance of the cyclotron after the upgrade. Possible future improvements are also suggested. The second part of this thesis deals with the improvement program of the Princeton University cyclotron. This consists of a design study of the axial injection system and of the new central region. It starts with an introduction to the Princeton University cyclotron. A detailed design study of the axial injection system then follows. Finally, design studies for the new central region, based on the beam orbit dynamics investigations, are presented. A brief consideration associated with the depolarization problem of polarized ions during acceleration in the cyclotron central region is also given.