Study of the Ion Hose Instability in the DARHT-II Downstream Transport Region

摘要

The second axis of the DARHT flash X-ray facility at Los Alamos National Laboratory ('DARHT-II') is a multiple-pulse, 18.4 MeV, 2 kA induction electron linear accelerator. A train of short ((approx) 50 ns) pulses are converted via bremsstrahlung to X-rays, which are then used to make radiographic images at various times (nominally four) during a ''hydrotest'' experiment. The train of pulses is created by carving them out of a two microsecond long macropulse, using a fast switching element called a kicker. The unused portion of the macropulse is absorbed in a beam dump. Thus, upstream of the kicker, two microseconds of beam are transported through a vacuum system roughly sixty meters long. These conditions involve length and, specifically, time scales which are new to the transport of high-current beams. A concern under such conditions are the macroscopic interactions between the electron beam and positive ions created by impact ionization of the residual gas in the vacuum system. Over two microseconds, the ion density can develop to a hundredth or even a tenth of a percent of the beam density--small, to be sure, but large enough to have cumulative effects over such a long transport distance. Two such effects will be considered here: the ion hose instability, where transverse forces conspire to pull the electron beam farther and farther off axis, and background gas focusing, where radial forces (with respect to the beam) change the beam envelope during the course of the macropulse. The former effect can cause beam emittance growth (affecting the ability to focus the beam on the target) and eventually catastrophic beam loss; the latter can cause either serious degradation of the statically tuned final focus on the converter target, or a pinching of the beam on the surface of the main dump to the point where the heat flux causes damage. The beam transport upstream of the kicker has two distinct phases. First, the beam is created and accelerated up to 18.4 MeV over a distance of about fifty meters. Then the true downstream transport begins: the beam drifts through a matching section in preparation for the kicker, over some ten meters; the long-pulse beam then travels about four more meters from the kicker to the main dump. In the accelerator, the beam energy is obviously not constant; the transport is emittance-dominated and done through nearly continuous solenoidal focusing. In the downstream section, there are only two discrete solenoids over the entire fourteen meters and the transport is largely ballistic. Since ion hose has been studied in the accelerator and since the lack of continuous focusing is considered a concern with respect to ion hose in the downstream section, the focus of this study is only from the exit of the accelerator to the main dump.