Design of a Post Accelerator for the Rare Isotope Accelerator Facility*

The proposed Rare Isotope Accelerator (RIA) Facility includes a post-accelerator for rare isotopes (RIB linac) which must produce high-quality beams of radioactive ions over the full mass range, including uranium, at energies above the coulomb barrier, and with high transmission and efficiency. The latter requires the RIB linac to accept at injection ions in the 1+ charge state. A concept for such a post accelerator suitable for ions up to mass 132 has been previously presented [1-3]. This paper presents a modified concept which extends the mass range to uranium. The RIB linac will utilize existing superconducting heavy-ion linac technology for all but a small portion of the accelerator system. The exceptional piece, a very-low-charge-state injector section needed for just the first few MV of the RIB accelerator, consists of a pre-buncher followed by several sections of cw, normallyconducting RFQ. Two stages of charge stripping are provided: helium gas stripping at energies of a few keV/u, and additional foil stripping at ~600 keV/u for the heavier ions. In extending the mass range to uranium, however, for best efficiency the helium gas stripping must be performed at different energies for different mass ions. We present numerical simulations of beam dynamics for a design for the complete RIB linac which provides for several stripping options and uses cost-effective solenoid focussing elements in the drift-tube linac. 1 LINAC DESIGN CONCEPT The RIB accelerator system must have the following properties: x Efficiently accept and accelerate ions over the full mass range, including uranium. x Provide an output beam at any energy up to 5-10 MeV/nucleon. x Maintain a longitudinal emittance of a 0.5 S keV/u˜nsec over the full range of energy and mass. An important technical challenge is to design a linac for low-charge-state ions which can provide low longitudinal emittance (i.e. good time and energy resolution). The ability of the linac to maintain small longitudinal emittance, will be critical in enabling experiments to use time-of-flight techniques while simultaneously having available good energy resolution and adequate beam intensity. The technology developed for existing superconducting heavy-ion linacs, characterized by excellent performance and a very high degree of modularity, provides a basis for all but a small portion of such a RIB accelerator system. The exceptional piece is a very low charge state injector section. The most efficient generation of beams of rare isotopes requires singly charged ions at initial injection. Very low charge state ions can most efficiently be bunched and accelerated by using several sections of cw, normally-conducting RFQ for the first few MV of the RIB accelerator . The design goal for the RIB linac is to accelerate heavy ions in the mass range from 6 to 240, starting with ions at charge state 1+. As discussed in references [3,4], the RIB linac provides for charge stripping at two stages: nonequilibrium gas stripping at energies of ~10 keV/u, and an additional foil stripping at ~600 keV/u for the heavier ions. In operation, radioactive nuclei will be either collected or ionised at charge state 1+ and extracted at a voltage of 100 kV. A state-of-the-art high-acceptance isobar separator with mass resolution m/'m|20000 [5] will provide final separation of the desired species. For further acceleration, the velocities of different ions must be matched into the normally-conducting injector RFQ. This is accomplished by mounting the first two sections of RFQ on a variable voltage platform. For best efficiency over the full mass range, helium gas stripping must be performed at different energies for different mass ions. The linac following this stripping can accelerate ions of charge to mass ratio 1/66 and above. By stripping at 7 keV/u some 55% of an incident Sn beam, for example, can be stripped into charge state 2+ and further accelerated. For the heavier ions, higher charge states are required, for which the best stripping efficiency is achieved at higher energy. Figure 1 shows the efficiency of helium stripping [4] as a function of beam energy with charge number as a parameter. The yield of heavy ions is shown at charge state q which is required for the following accelerator section. To cover the full mass range, we propose to provide options for He gas stripping at either of two energies: at 7 keV/u for the lighter ions, and for ions of Z>54, at 20 keV/u. Providing two different options for helium gas stripping ensures a high efficiency of operation over the full mass range, including mass 240. In the proposed RIB linac ions with masses 66 and below do not require this helium gas stripping. The RIB linac consists of the following main sections (see Fig. 2): x An injector with three sections of normallyconducting RFQ. x A superconducting linac which will accelerate ions of q/m > 1/66 to 600 keV/u or more. ________________________________________ * Work supported by the U. S. Department of Energy under contract W-31-109-ENG-38. † On leave from the Institute of Theoretical and Experimental Physics,