Production of electric power by using a beam of heavy ions to ignite an ine
rtially-confined fusion target requires the focusing of high-power beams on
to a small spot several meters distant from the final lens system. Beams wi
th the necessary intensity generally behave like warm nonneutral bounded pl
asmas where beam kinetic temperatures are sufficiently high that a cold-pla
sma description can be inadequate for describing the collective space-charg
e modes. In view of the complexity of the self-consistent nonlinear dynamic
s, analytic study has largely been limited to the singular Kapchinskij-Vlad
imirskij (K-V) distribution. Numerical simulations, primarily using the WAR
P [D. P. Grote, A. Friedman, I. Haber, W. Fawley, and J. L. Vay, Nucl. Inst
rum. Methods Phys. Res. A 415, 428 (1998)] particle-in-cell (PIC)/accelerat
or code, have been employed to identify the degree to which the analytic re
sults, especially the predictions of unstable modes, are applicable to real
istic beam distributions. During extensive benchmarking of the code against
experiments at Lawrence Berkeley National Laboratory, Lawrence Livermore N
ational Laboratory, and the University of Maryland, a particular feature of
the beam which has been seen in both experiment and simulation is the laun
ching, in the source region, of collective warm-plasma oscillations similar
to those predicted on the basis of the K-V analysis. (C) 1999 American Ins
titute of Physics. [S1070-664X(99)96605-3].