A 2-GeV, 1-MW PULSED PROTON SOURCE FOR A SPALLATION SOURCE* Y. Cho, Y.-C. Chae, E. Crosbie, H. Friedsam, D. Horan, R. Kustom, E. Lessner, W. McDowell, D. McGhee, H. Moe, R. Nielsen, G. Norek, K. Peterson, K. Thompson, M. White Argonne National Laboratory, Argonne, IL 60439 USA ABSTRACT A design study of a 1-MW pulsed proton source based on a 2-GeV rapid cycling synchrotron (RCS) has been completed. The RCS operates at a 30-Hz repetition rate. A 400-MeV H - injector linac allows transverse phase- space painting during injection. The linac beam is chopped near the ion source so that the phase-space of the incoming beam fits into the waiting synchrotron bucket. Chopping in this way minimizes potential beam loss during the rf capture process. An rf system provides a peak voltage of about 180 kV with a frequency swing of 1.1 to 1.5 MHz. The rf voltage programming was developed using particle tracking that included space- charge effects, in order to eliminate possible beam losses. The design takes into account re-use of existing buildings and infrastructure of the former 12-GeV Zero Gradient Synchrotron. 1 INTRODUCTION A feasibility study for upgrading the Intense Pulsed Neutron Source (IPNS) at Argonne National Laboratory has been completed [1]. The 1-MW spallation source is based on a proton synchrotron that accelerates the 400- MeV linac beam to 2 GeV and delivers 1.04 x 10 14 protons per pulse at a repetition rate of 30 Hz. The choice of 30 Hz as the repetition rate was based on preferences expressed by the neutron community. Full power can be delivered to a 30-Hz target station or one out of three pulses to a 10-Hz station and the remaining two pulses to the high-frequency station. The synchrotron system and associated research facilities are housed in the former 12- GeV Zero Gradient Synchrotron (ZGS) area and occupy about 50,000 m 2 . The 190-m, 2-GeV RCS is housed in the ZGS tunnel and the two neutron-producing targets, each serving 18 neutron beamlines, are placed in former experimental area buildings. Enclosures for the linac and low energy transport line (LET) are the only new conventional facility construction. 2 LATTICE The lattice design provides: a) a large transition energy so that the lattice has a relatively large slip factor, η γ γ = - - - 2 2 t , b) enough straight-section length for a radio-frequency cavity system that could have a total * Work supported by the U.S. Department of Energy, Office of Basic Energy Sciences under Contract W-31-109-ENG-38. length of 20-30 m, and c) dispersion-free straight sections for implementation of charge-exchange injection. Figure 1 shows 1/2 of a superperiod with reflective symmetry at both ends. Each cell of the FODO structure has a phase advance of ~ 90 o in both transverse planes. The normal cells, dispersion-suppressor cell and the straight-section cells are evident in the figure. The dispersion-suppressor cell is made by removing a dipole from a 90 o phase advance cell. The vertical phase advance is slightly less than 90 o but the horizontal phase advance is maintained at 90 o , so that the missing dipole scheme effectively supresses the dispersion function. An advantage of this arrangement is that the horizontal tune is about one unit higher than the vertical. Taking alignment and construction imperfections into account, tracking studies showed that the RCS has a dynamic aperture larger than the physical aperture of the ring vacuum chamber [2]. Table 1 is a summary of the main RCS parameters. η x β x β y Distance (m) B B B B QD QD QD QD SF SF QF QF QF QF SD SD S1 Figure 1: Lattice functions for 1/2 superperiod. 3 INJECTION The injection energy was determined by the incoherent space-charge limit of the lattice and the defined acceptance of the synchrotron. The injected beam stack has an emittance of 375 π mm mr in both transverse planes. With a bunching factor of 0.4 and an allowed tune shift due to space-charge of 0.15, an injection energy of 400 MeV is sufficient to provide a time-averaged current of 0.5 mA with a repetition rate of 30 Hz. The 400-MeV H - ion injector linac design for this feasibility study was performed by the industrial firm