[en] The conceptual design of a superconducting (linac) booster (named ALPI PROJECT) for the 17 MV XTU-TANDEM of Laboratori Nazionali di Legnaro has been recently accepted by the National Institute of Nuclear Physics as one of the leading projects to be funded in the next five year plan. Money for resonator and cryostat prototypes is already available and the building is going to be funded next January. The project aims at a machine capable of accelerating all the stable isotopes up to Uranium at energies above the Coulomb barrier of very possible ion-ion interaction with beam quality comparable to that of d.c. accelerators. At LNL the advantage of coupling the linac postaccelerator to the 17 MV XTU Tandem is taken which is able to produce even the very heavy beams with reliable intensity and velocities β ≥ 0.04 which can be matched by superconducting resonators feasible with the present available technology. As accelerating structures in the ALPI project straight line quarter wave resonators (QWR) have been chosen on the basis of their intrinsic mechanical stability and broad velocity acceptance (two gap resonator) particularly important for a national facility like ALPI which is expected to produce as many different beams as possible. Lead has been chosen as superconductor on the basis of the following considerations: (i) lead technology being much more applied for QWR resonators than the Nb one can be easier and faster introduced in a Nuclear Physics Laboratory without any experience in the field; (ii) the performances of SUNYLAC have demonstrated that their initial goal of reaching accelerating gradient of 3 MV/m is feasible; (iii) the difficulty in fabricating the OFHC copper base of the resonators (number of EB welds, joints) is relatively modest if compared with the solutions involving Nb as superconductor. 7 references, 3 figures

[en] The Stony Brook Nuclear Structure Laboratory (NSL) has operated a superconducting heavy-ion booster linac since April 1983. The 40 copper split-loop resonators were developed and fabricated at Cal-Tech and plated with lead at Stony Brook. These original lead surfaces have given stable performance for the last 4 years, at an average accelerating gradient of about 2.5 MV/m in the high-β section. The low-β resonators however have never run reliably on-line much better than 2.0 MV/m, due to excessive vibration of their rather soft loop arms in the working accelerator environment. For the last 2-3 years the efforts of the Stony Brook accelerator development group have been focused on (1) a retrofit of the low-beta section of the linac with new QWRs and (2) the further development of plated superconducting surfaces. In particular a Sn/Pb alloy has been shown to give resonator performance at least comparable to that obtained with pure Pb but with a greatly simplified plating technique, as discussed below. Recently a possible heavy-ion injector based on superconducting RF quadrupole (RFQ) structures has also been studied. 13 references, 3 figures, 1 table

[en] The RF Superconductivity activities at Canberra have been inspired by the Second Workshop on RF Superconductivity. The main areas of effort have been in: (1) vacuum chemistry kindled by Tuong's report on ozone treatment of surface contaminants, (2) electron emission enhanced by surface contaminant interfaces by Lathan, and (3) magnetron sputtering of niobium onto copper by Benvenuti. Before reporting the details of these projects, a progress report on the booster project is obligatory. 4 references, 1 figure

[en] A superconducting heavy ion linac is being proposed for the JAERI-tandem booster. For the accelerating structure of the tandem booster which ought to accelerate heavy ions of wide range of mass numbers, quarter wave resonator (QWR)s are suitable because of their wide ion-velocity acceptance. Ions of hydrogen to bismuth from the JAERI tandem can be accelerated by β = 0.1 QWRs. The excellent result of a niobium QWR at Argonne National Laboratory was a motive for the development of niobium QWRs. Further considerations on the design were required, because the Argonne's QWR did not have beam ports nor frequency tuners. As a result of considerations on these points, it has been decided to have an oval cylinder for the outer conductor. The prototype resonator has been built and tested. The fabrication techniques of explosive bonding, electron beam welding and heat treatment were found to be available in domestic companies in 1984. After obtaining niobium and niobium-clad-copper materials in 1985, the prototype resonator was built in 1985-86. Electro-polishing was done in their laboratory. Tests at 4.2 K have been repeated several times in combination of treatments of the niobium surface. The work is proceeding to the construction of a buncher and a prototype linac unit which are composed of superconducting QWRs. 4 references, 4 figures, 2 tables

[en] Daresbury Laboratory houses a 20 MV tandem which has been in operation for over 4 years. Various proposals have been made to upgrade the facility by the addition of a post accelerator. One of these involved the addition of a large superconducting heavy ion linac capable of accelerating all masses up to the Coulomb barrier (1). Because of financial restrictions it was not possible to proceed with this scheme, so that when the resonators and cryostats which were to have been installed on the Oxford University folded tandem became available it was decided to install these as a post accelerator on the Daresbury machine. These modules will add about 5 MeV/charge to the ion energy and this will extend the range of nuclear physics experiments possible by a useful factor. The addition of further linac modules is an obvious possibility and this has been kept in mind in the design and specification of the layout and the helium refrigeration system. 3 references, 2 figures

[en] Higher order mode (HOM) couplers are integral parts of a superconducting accelerator cavity. The damping which the couplers must provide is dictated by the frequency and shunt impedance of the cavity modes as well as by the stability requirements of the accelerator incorporating the cavities. Cornell's 5-cell 1500 MHz elliptical cavity was designed for use in a 50 x 50 GeV electron-positron storage ring with a total beam current of 3.5 mA (CESR-II). HOM couplers for the Cornell cavity were designed and evaluated with this machine in mind. The development of these couplers is described in this paper. 8 references, 8 figures

[en] URMEL computation and beadpull measurements showed that a 4-cell, 500 MHz HERA cavity has five parasitic mode families with high R/Q which must be damped by HOM couplers. To reduce induced power in cavity, Q/sub ext/ of these resonances should be low due to the fact that maximum design current of the HERA electron ring is Ib = 58 mA. During the last two years couplers based on two inductive stubs were developed and tested in SC version. The two stubs construction of HOM couplers was chosen as a relatively simple cryogenic solution for the cooling of all inner components of the coupler. In November 1987 a test of the complete accelerating module consisting of two 4-cell cavities, two fundamental mode couplers and six HOM couplers will be carried out. Each 4-cell cavity is equipped with a set of three HOM couplers: one mainly for reducing Q of TM₀₁₁ family (coupler TM) and two for loading the most dangerous dipole modes and TM₀₁₂ family (couplers TE). 7 references, 6 figures, 1 table

[en] Two techniques are usually employed to prepare the so-called δ-NbN phase having good superconducting properties: reactive sputtering in a nitrogen-argon atmosphere, and thermal diffusion of nitrogen into niobium followed by a rapid quench cooling. Sputtered NbN films have granular type microstructure, yielding high B/sub C2/ - and J/sub C/ - fluxoid properties (4), and are well-adapted for the protection of surfaces with simple forms. However for complicated geometries it is rather difficult to obtain homogeneous NbN films by this technique. Thermal diffusion technique has been used for the first time about forty years ago (5), but has been rarely employed since then, due to the difficulty for obtaining the δ - NbN phase at temperature below 1300⁰C. For their application to RF cavities, the NbN obtained by thermal diffusion is intended not to be of the granular type structure, having thus lower RF residual losses R/sub res/ and higher RF breakdown fields B/sub crit/. The results reported here are related to NbN obtained by the thermal diffusion technique on bulk niobium at temperatures well below 1300⁰C. 14 references, 3 figures

[en] As early as the fall of 1977 it was decided that the future research needs of their nuclear structure laboratory required an increase in energy capability to at least 8 MeV per nucleon for the lighter ions, and that these needs could be met by the installation of a 17 MV tandem Van de Graaff accelerator. The chief problem with this proposal was the high cost. It became apparent that a far less expensive option was to construct a linear accelerator to boost the energy from their existing 9 MV tandem. The options open to them among linac boosters were well represented by the room temperature linac at Heidelberg and the superconducting Stony Brook and Argonne systems. By the Spring of 1979 it had been decided that both capital cost and electric power requirements favored a superconducting system. As regards the two superconducting resonator technologies - the Argonne niobium-copper or the Caltech-Stony Brook lead plated copper - the Argonne resonators, though more expensive to construct, had the advantages of more boost per resonator, greater durability of the superconducting surface and less stringent beam bunching requirements. In 1980 pilot funding from the State of Florida enabled the construction of a building addition to house the linac and a new target area, and the setting up of a small, three resonator, test booster. Major funding by the NSF for the laboratory upgrade started in 1984. With these funds they purchased their present helium liquefaction and transfer system and constructed three large cryostats, each housing four Argonne beta = 0.105 resonators and two superconducting solenoids. The last large cryostat was completed and installed on-line early this year and the linac was dedicated on March 20. Nuclear physics experiments using the whole linac began in early June. 4 references, 6 figures, 1 table

[en] At the Accelerator Laboratory of the Universities of Munich a booster for the existing MP-tandem is under construction. The Tritron project was finally funded in January this year. The Tritron is a cyclotron with separated orbits with both the magnets and cavities superconducting. It will increase the ion energies by a factor of 4.9, so a ¹²C⁶⁺-beam will have a maximum energy of 21 MeV/u e.g. The cryostat has a diameter of 3.6 m. The beam is injected at a radius r₁ = 66 cm and extracted after almost 20 turns at r₂ = 145 cm. The turn separation is δR = 4 cm. The bending is made by 12 flat magnet sectors, each consisting of 20 magnetic channels of window frame type. In each channel the magnetic field (∼ 1.4 T) is adjustable individually, so that the beam always can be guided along the spiral orbit. The gradient is alternating from one sector to the next resulting in strong focusing in both transversal directions. In each second intermediate gap a wedge-shaped accelerating cavity of the reentrant type is installed for the 20 parallel beams. The radial cavity length is 1.2 m, the outside width 0.7 m. The radial length of the accelerating lips is 0.9 m, the gap width at injection radius 6 cm, at extraction radius 13 cm. The rf-frequency is ∼ 170 MHz, the maximum voltage at injection U(r₁) ≅ 270 kV, at extraction U(r₂) ≅ 540 kV per cavity. If the bunchers cross the cavity at a rf-phase with increasing accelerating voltage, they are focused longitudinally, resulting in ∼ 0.2 synchrotron oscillations per turn. The longitudinal focusing causes the rf-phase to change automatically corresponding to the necessary energy gain and to the radial characteristic of the voltage amplitude along the gap. 2 references, 6 figures, 1 table