Program for Plasma-Based Concepts for Future High Energy Accelerators

Abstract

Program for Plasma-Based Concepts for Future High Energy AcceleratorsT. Katsouleas, PI; P. Muggli, Co-PI USC Progress ReportThe progress made under this program in the period since November 15, 2002 is reflected in this report. The main activities for this period were to conduct the first run of the E-164 high-gradient wakefield experiment at SLAC, to prepare for run 2 and to continue our collaborative effort with CERN to model electron cloud interactions in circular accelerators. Each of these is described briefly below. Also attached to this report are papers that were prepared or appeared during this period.Highlights of the accomplishments during this contract period:1. E-164 run I. The first run of the E-164 experiment took place from march 17 to April 28 2003. For this run electron bunch length was shortened from ≈700 µm to ≈100 µm in order to bring the expected gradient from a few hundred MeV/m to the GeV/m level. To optimize the accelerating gradient the plasma density had to be increased by the bunch length reduction factor squared, or a factor of about 50 to n e ≈6x10 15 cm -3. The pre-ionized plasma is obtained by laser photo-ionization. This high density could be reached only over a distance of ≈30 cm with the available laser beam energy and fluence. However, experimental results showed that, for most of the beam parameters (radius ≈25 µm at the plasma entrance, >1.3x10 10 electrons per bunch), the beam radial field were sufficiently large to field ionize the ambient neutral lithium vapor in the plasma source. Field ionization is a threshold process which, when the threshold is exceeded, leads to full ionization of the neutral vapor density n 0 (n e =n 0 ), therefore leading to a plasma density much larger that the one expected for laser ionization (≈10% of n 0 ). Images of the beam dispersed in energy showed that the energy loss by the beam in the field-ionized plasma was extremely large, exceeding 1.5 GeV over the 30 cm plasma length, therefore corresponding to decelerating gradients well exceeding 4 GeV/m. Ionizing the vapor to a low density prior to the beam arrival led to more of the beam charge losing energy, but did not increase the maximum energy loss. Clear thresholds for the ionization process were observed as a function of the number of particles in the bunch, bunch radius, and bunch length. The electron bunch enters the plasma with a correlated energy spread resulting from the strong longitudinal linac wakefields. The full width of energy chirp is of the order of ≈3% of the incoming energy (or ≈1 GeV), with the head of the bunch at the higher energies. In previous long bunch experiments the energy spectrum was dispersed in time to observe energy gains in the back of the bunch, with magnitudes smaller than the incoming chirp. With 100 µm long bunches (1/3 ps), time resolution is not available and only energy gains larger than the incoming energy chirp can be observed on time integrated energy spectra. No clear energy gain was observed during the experiment. Numerical simulations indicate t...

Review of Beam Driven Plasma Wakefield Accelerators

“…12 Finally at the ATF, a user group from USC has proposed to resonantly excite a wakefield using electrons prebunched on a 10 µm scale by an IFEL prebuncher. 13 By sending a train of such microbunched pulses into a 10 19 cm -3 plasma, they hope to excite a sizable wakefield. A few percent detuning between the drive and the wakefield frequency is sufficient for the later bunches to extract energy from the wake built-up by the earlier bunches.…”