N. Tokuda

Construction and beam tests of a 25.5 MHz split coaxial RFQ for radioactive nuclei

Abstract A 25.5 MHz split coaxial RFQ with modulated vanes has been constructed at INS. The RFQ will accelerate radioactive nuclei with a charge-to-mass ratio greater than 1 30 from 2 to 172 keV/u. The cavity, 0.9 m in inner diameter and 8.6 m in length, comprises four unit cavities, each of which comprises three module cavities further. The modulated vanes have been made by using both of a three-dimensional cutting technique and a two-dimensional one, the former for the first unit cavity, and the latter for the other ones. The transverse radius of curvature of the vane-tip is variable along the beam axis in a low-energy part, and constant in a high-energy part (the boundary at 1.1 m down from the RFQ entrance). The vanes have been aligned with an error less than ± 40 μm before installation in the unit cavities. After completion of the four unit cavities, they have been aligned with an error less than ± 50 μm on the floor of an accelerator room. Low-power tests after cavity-tuning show that the longitudinal voltage flatness and the azimuthal field balance are better than ± 1%, the resonant frequency is 25.46 MHz, and the unloaded Q-value is 5800, corresponding to a resonant resistance of 22 kΩ. We have achieved so far the goal intervane voltage of 109 kV at a duty factor of 15%. Through acceleration tests with N+ ions, we have obtained the following results: the output-beam emittances are well in the design ellipses of 0.06π cm mrad normalized; the data of transmission efficiency versus intervane voltage agree well with PARMTEQ results and the transmission at the nominal voltage is measured to be 90%.

Design and construction of heavy ion RFQ linac with effective acceleration structure

Abstract At the Tokyo Institute of Technology, generating a plasma with a heavy ion beam has been studied for basic research on inertial fusion and a heavy-ion pumped laser. These experiments require a high intensity and high brightness beam. An 80 MHz heavy-ion RFQ linac is being constructed to meet the requirements. This linac accelerates particles with a charge to mass ratio ( q A ) greater than 1 16 from 5 keV/amu up to 214 keV/amu. Two-dimensional (2D) machining is applied for cutting of the RFQ vane-tips. We simulated the beam optics including effects of higher order harmonies in the intervane potential (H. Deitinghoff, A. Schempp, H. Klein and O. Pan, Particle Acc. 37–38 (1992) 47; Chen Chia-Erh, Fang Jia-Xun, Li Weigo, Pan Oujia, Lu Yuanrong, Li Deshan, Wang Lishan, Yu Maolin, H. Deitinghoff, A. Schempp and H. Klein, Proc. European Particle Acc. Conf. (1992) p. 1328) for optimization of the vane parameters. In order to increase the acceleration efficiency, synchronous phase was gradually raised from −30° to −20° in the accelerator section of the cavity. The expected beam transmission is 91.8% for a beam current of 0 mA and 68.4% for 10 mA.

Test Accumulation Ring for Numatron Project - TARN -

A Test Accumulation Ring for NUMATRON Project, TARN, is now under construction at INS, University of Tokyo. Heavy ions from the SF Cyclotron such as N5+ with an energy of 8.5 MeV per nucleon, are planned to be injected and stacked in the ring by a combination of multiturn injection and RF stacking method Expected intensity of the stacked beam, e.g. N5+, is 2 × 1010 particles, and a survival rate of 90 % is anticipated at the pressure of 1 × 10-10 torr during a stacking time of 1 sec. In this paper, the present status of the ring is described as well as the performances of the major subsystems.

Characteristics of Lattice and Magnet System of TARN II

TARN II is a ring which is designed to be operated both as a synchrotron and a cooler ring for ions. Its mean radius is ~12.4 m and is to be able to accelerate protons up to 1300 MeV and ions with charge to mass ratio of 1/2 up to 450 MeV/u. The ring consists of 24 dipole and 18 quadrupole magnets, which compose the lattice with sixfold symmetry for synchrotron acceleration (Synchrotron Mode) and another one with threefold symmetry to realize doubly achromatic sections for beam cooling (Cooler Ring Mode). These modes can be transferred between each other keeping the operating point at the position of (¿H, ¿V)~(1.75, 1.25). The acceptance for beam cooling experiment at TARN II is expected to be improved from 70¿ to 400¿ mm·mrad by application of pre-cooling of horizontal betatron amplitude by stochastic method with Synchrotron Mode before moving to Cooler Ring Mode. Main magnets of the ring with AC characteristics have already been fabricated. Dipole magnets with H-type are found to have realized required good field region of ±100 mm from the result of static field measurement.

Split Coaxial RFQ Linac with Modulated Vane

A split coaxial RFQ structure with modulated vanes is under development for the acceleration of very heavy ions, like uranium ions. To fix the vanes tightly to the tank and to attain accurate setting of them, a new structure is proposed: the vanes are supported by stems at several points. As a preparatory study, we have fabricated a 1/4 scaled model of a uranium machine. This model is 2 m in length, and 0.4 m in diameter. Measurements on rf characteristics have been carried out with good results: field balance among the quadrants and field flatness along the beam axis are within ±2.5%

Acceleration of high current heavy ions using a variable energy radio‐frequency quadrupole linac and measurement of the input beam emittance

A high current ion beam injection system and a variable energy radio‐frequency quadrupole (RFQ) accelerator with an external LC resonance circuit are tested. A four‐rod RFQ electrode of 2.3 m length is newly designed to obtain a milliampere class MeV ion beam. To increase accelerated beam current, injected beam emittance into the RFQ is measured and compared with the designed RFQ acceptance. Injected beam emittance tends to increase with the beam current increase. The beam acceleration tests are carried out with a cw rf power supply of 100 kW (10–30 MHz, continuously variable). Results show that the time‐averaged beam currents of N+ and Ar2+ are 1.1 mA (0.4 MeV) and 1.03 mA (1.0 MeV), respectively, and that the peak current of these beams is about 5 mA. As representative ion species for fabricating semiconductor devices, P+ and P2+ are accelerated, and the time‐averaged beam current of 0.48 mA (0.56 MeV) and 0.28 mA (0.81 MeV) are obtained, respectively. This high current MeV ion implanter has great poten...

Development of an SCRFQ heavy-ion linac for RI beams

Abstract A split coaxial RFQ (SCRFQ) is being developed to accelerate RI beams from 1 to 170 keV/u as part of the Japanese Hadron Project (JHP). Our SCRFQ is equipped with modulated vanes. On the basis of the studies on a cold model and a proton accelerating one, a 25.5 MHz prototype for the JHP SCRFQ has been constructed. The prototype, consisting of three module cavities, is 2.1 m in length and 0.9 m in inner diameter, and accelerates ions with a charge-to-mass ratio (q/A) greater than 1 30 from 1 to 45.4 keV/u. The unloaded Q-value of the cavity is 6400, corresponding to about 84% of the calculated value, and the field imbalance between vanes is within ±0.6%. The designed intervane voltage of 109 kV for ions with q/A = 1 30 is achieved with a 70 kW peak power. By using ions of three species, N2+, N+ and Ne+, acceleration tests are conducted. The transmission efficiency attained with a N+ beam is better than 80% at normalized intervane voltages higher than 1.2.

Ion beam acceleration by an RFQ with an LC resonance circuit

Abstract For the development of a high current MeV implanter, ion beam acceleration using a variable energy RFQ system has been studied. The RFQ system consisted of an LC resonance circuit and conventional RFQ electrodes of 1.3 m in length. The output energy could be continuously varied by changing the resonance frequency of the circuit. The electrodes were designed to accelerate nitrogen ion beams (N + ) from 10 keV to 270 keV in order to investigate fundamental acceleration characteristics of the system. The designed intervane voltage was 26 kV. By improving the LC resonance circuit to generate a high voltage of radio frequency, a shunt impedance of over 70 kΩ was obtained, which was sufficient for MeV-range acceleration. Experimental results showed that N + beams were accelerated to 265 keV and Ar 2+ beams to 740 keV. Energy was varied by changing the frequency, in agreement with the calculated value. It was concluded that the RFQ system driven by the LC circuit should be very useful for MeV ion implantation in semiconductor device fabrication.

Beam test results of the INS RFQ/IH linac

A linac complex for radioactive beams has been constructed at INS, which comprises a 25.5-MHz split coaxial RFQ (SCRFQ) with modulated vanes and a 51-MHz interdigital-H (IH) linac . The SCRFQ accelerates ions with a charge-to-mass ratio (q/A) greater than 1/30 from 2 to 172 keV/u. The beam from the SCRFQ is charge-stripped by a carbon-foil, and is transported to the IH linac through two magnetic-quadrupole doublets and a 25.5-MHz rebuncher cavity. The IH linac accelerates ions with a q/A greater than 1/10, and the output energy is variable in the range of 0.17 through 1.05 MeV/u. Beam tests of the linac complex performed with N ions show that the output beam energy and transmission efficiency agree well with predictions.

Beam Stacking Experiments at TARN

TARN is a low energy, around ten MeV/nucleon, ion storage ring aimed at beam accumulation in both the longitudinal and transverse phase spaces from the injector cyclotron. Up to now, proton, H/sub 2//sup +/, ..cap alpha.. and /sup 3/He/sup +/ beams have been stored and the close agreements are shown between the theoretical and experimental results on the beam dynamics. Twenty turn beams are injected in the betatron phase space area of 87 ..pi..mm X mrad by the multiturn injection method and 15 pulses are RF stacked in the momentum space ..delta..p/p of 2.2%. The overall stacking number is then attained at around 300 turns. RF stacking is performed with the repetition rate of 30 Hz and it requires about 0.5 seconds to fill the whole phase spaces with the beam. The e-folding life time of 7 MeV protons is 400 s at the vacuum pressure of 1 X 10/sup -10/ Torr, whereas that of /sup 3/He/sup +/ of the /sup 4/He gas target is 60 ms at 1 x 10/sup -8/ Torr. Collision strengths of charge exchange atomic reactions have been measured.