Test of the Rex-rfq and Status of the Front Part of the Rex-isolde Linac*

For REX-ISOLDE (Radioactive beam EXperiments at ISOLDE / CERN) [1], a test beamline is built up at the Garching Accelerator Lab to perform He-experiments with the RFQ [2], the matching (rebunching) section between RFQ and IH-DT-linac, the IH-structure [3] and several electrostatic lenses of the REX-ISOLDE-mass separator [4]. In a first step, the beamline is conceived for tests with the RFQ. This paper presents the parameters and the status of the REX-RFQ, the experimental setup and the particle dynamics simulations with the COSY infinity code for beam injection and beam analysis. Furthermore it shows the design and status of the mass separator, the IHstructure and the buncher section. 1 TEST OF THE REX-RFQ The REX-ISOLDE Linac consists basically of a 4-RodRFQ and an Interdigital-H Drift-Tube-accelerator, followed by three seven-gap resonators for energy variation from 0.8 MeV/u to 2.2 MeV/u [5]. 5 keV/u 300 keV/u 1.1-1.2 MeV/u 0.8-2.2 MeV/u RFQ buncher IH-structure 7-gap resonators 0 1m highly charged ions (A/q<4.5) to the target and the MINIBALL Fig. 1: The REX-ISOLDE Linac 1.1. The RFQ The REX-RFQ will accelerate the radioactive ions with an A/q 3 4.5 from 5 keV/u to 300 keV/u. The according electrode-voltage is 28 42 kV, which will require an rfpower of 30 to 40 kW. The accelerator is similar to the GSI HLI-RFQ and to the Heidelberg High current injector RFQ. It has a total length of 3 m (18 stems) and operates at a resonant frequency of 101.28 MHz (duty-cycle: 10%). Low-level measurements of the rf-parameters were in good agreement with the design values. The measurements showed: f=101.18 MHz, Q=3900 and Rp=170 kΩm. The frequency can be shifted with two mounted piston tuners in the range ±200 kHz. The flatness was tuned from ±20% to ±1%. The distribution of the built in tuning plates has been calculated with the code MAFIA. Fig. 2a shows the voltage distribution of a MAFIA-model with the tuning plate heigths from experimental data and unmodulated electrodes together with the measured voltage distribution of the REX-RFQ without tuning plates. The diagram illustrates the compensation of the variable capacitive load (changes in modulation and aperture) in each RFQ-cell by an appropriate variation of its inductance (different heigth of the tuning plates). Figure 2b shows the tuned flatness measured with three different pertubation capacitors. The bars on the bottom of the diagram indicate the heigth of the Tuning-plates.