Plasma Possibilities in the NLC*

Basic idea and preliminary analysis are presented of a possibility to compensate the energy spread and to collimate beam in the next generation of linear colliders (NLC) using tunneling ionization of a gas by the beam. Contributed to ICFA Workshop, Nonlinear and Collective Phenomena in Beam Physics, Arcidosso, Italy September 2-September 6, 1996 *Work supported by Department of Energy contract DE-AC03-76SFO0515. t ‘Permanent address: Stanford Linear Accelerator Center, Stanford, University, Stanford, CA 94309, USA Substantial progress inplasma based methods of acceleration opens newoptionforoptimization of the design of future colliders. In this paper, we discuss two possible uses of plasma in the Next Linear Collider (NLC). First, we discuss the use of a short wavelength plasma accelerator to reduce the correlated energy spread along the bunch at the end of the NLC linac. Second, we consider plasma generated by the bunch in a gas chamber by tunneling ionization as the mean to simplify the beam collimation. The optimal BNS energy spread 613/13 of a bunch is of the order of 1% rms in the NLC [1] and is induced by the longitudinal wakefields. A subst ant ial component of this correlated energy variation is linear along the bunch, see Fig. 1. Presently, it is removed by shifting the rf phase to 30° off crest along the last quarter of the linac. This both reduces the effectiveness of the BNS damping and decreases the net acceleration. Instead, this energy spread could be compensated if the bunch with the total length /B = 401, where typical rms length o ~ 100 – 150p, passing through a section with preliminary ionized plasma with the density of the order of ng = 1015 cm-3 excites plasma wave with the wave length of the order of 21B. The amplitude of the accelerating field in such a wave would increase from the head of the bunch to the tail producing desirable energy compensation. Another section, with higher plasma density and shorter plasma wave length, may be used to compensate remaining nonlinear variation of the energy spread. With a gradient of 1 GeV/m, a plasma length 1~ of five meters is needed to achieve the energy compensation. This length is restricted by the. Coulomb scattering. NLC collimation is designed with the restriction AN <104 on the fraction of the electrons AN scattered to the angle @ > ka’, k = 35 for the vertical plane. This sets the limit 1~< 10-6k2 yzo; 27rr~ngZ2~f” (1) Taking the NLC parameters: cry = lp, beta function PV = 35 m, and ~ = 1.0 x 106, and the density n~ = 1015 cm-s we get 1~ < 1.6(k2/Z2) cm. Therefore, the light gases (Z & 1) are preferable. In this case, for the design value of k = 35, the length 1~ would be limited to 1~<20.3 m which should not present a problem. Experiments with the goal to demonstrate accelerating gradients of the order of 1 GeV/m in a meter long channel of preliminary ionized plasma, are proposed or in progress today [2] and, if succeed, would make such scheme feasible. Another problem which may have a solution based on progress achieved in our understanding of beam-plasma interaction is the beam collimation. In the present design, a substantial part of the total length of the machine is dedicated to collimation of the halo particles, which is necessary to reduce the background in the detector. In this paper, we study collimation based on strongly nonlinear focusing produced by plasma generated in neutral gas by the beam itself. Collimation based on nonlinear optics was studied before [3]. There are two primary limitations with this approach: first, it requires very strong nonlinear magnets and, second, the alignment tolerances on the magnets is severe (fractions of a micron). a strong As it which is A plasma generated by tunneling ionization has the advantage that it can produce nonlinear field which is self-aligning; the field is centered at the beam location. is shown in the following, the tunneling ionization produces focusing on the beam almost linear for the core particles and is strongly nonlinear for the halo parti--