Title: Notes on the Injection of EBIS Ions into Booster

Ions from EBIS are injected into Booster after acceleration by an RFQ and a Linac. The velocity of the ions at Booster injection is c{beta} where c is the velocity of light and (1) {beta} = 0.0655. The kinetic energy is (2) W = mc{sup 2}({gamma}-1) where m is the ion mass and (3) {gamma} = (1-{beta}{sup 2}){sup -1/2}. Putting in numbers one gets a kinetic energy of approximately 2 MeV per nucleon for each ion. The revolution period at injection is 10.276 {micro}s. The ions in the EBIS trap are delivered in a beam pulse that ranges from 10more » to 40 {micro}s in length. This amounts to 1 to 4 turns around the machine. The transverse emittance (un-normalized) of EBIS beams just prior to injection into Booster is 11{pi} mm milliradians in both planes. This is an order of magnitude larger than the nominal 1{pi} mm milliradians for Tandem beams. Injection proceeds by means of an electrostatic inflector in the C3 straight section and four programmable injection dipoles in the C1, C3, C7, and D1 straights. These devices have been in use for many years for the injection of ions from Tandem as described in [1] and [2]. The inflector brings the incoming beam to the edge of the Booster acceptance and the dipoles produce a closed orbit bump that initially places the closed orbit near the septum at the in ector exit. During injection the orbit bump must be collapsed at a rate that keeps the injected beam from hitting the septum while continuing to allow beam to be injected into the machine acceptance. The process is discussed in [2] and [3]. There it is assumed that the injected beam moves with the closed orbit as the bump collapses. In the present report this is shown to be a valid approximation if the bump collapses sufficiently slowly. It is also shown that by judiciously choosing the horizontal tune and the initial distance of the closed orbit from the septum one can inject up to 4 turns of EBIS beams without loss on the septum. The reason for wanting to inject over a period of 4 rather than fewer turns is that this allows the beam to be distributed over a larger area of the Booster acceptance, thereby reducing the space charge force on the beam particles.« less

During the commissioning of EBIS beams in Booster in November 2010 and in April, May and June 2011, it was found that the transverse emittances of the EBIS beams just upstream of Booster were much larger than expected. Beam emittances of 11{pi} mm milliradians had been expected, but numbers 3 to 4 times larger were measured. Here and throughout this note the beam emittance, {pi}{epsilon}{sub 0}, is taken to be the area of the smallest ellipse that contains 95% of the beam. We call this smallest ellipse the beam ellipse. If the beam distribution is gaussian, the rms emittance ofmore » the distribution is very nearly one sixth the area of the beam ellipse. The normalized rms emittance is the rms emittance times the relativistic factor {beta}{gamma} = 0.06564. This amounts to 0.12{pi} mm milliradians for the 11{pi} mm milliradian beam ellipse. In [1] we modeled the injection and turn-by-turn evolution of an 11{pi} mm milliradian beam ellipse in the horizontal plane in Booster. It was shown that with the present injection system, up to 4 turns of this beam could be injected and stored in Booster without loss. In the present note we extend this analysis to the injection of larger emittance beams. We consider only the emittance in the horizontal plane. Emittance in the vertical plane and the effects of dispersion are treated in [2].« less

.

Electron beam cooling is examined as an option to reduce momentum of gold ions exiting the EBIS LINAC before injection into the booster. Electron beam parameters are based on experimental data (obtained at BNL) of electron beams extracted from a plasma cathode. Many issues, regarding a low energy high current electron beam that is needed for electron beam cooling to reduce momentum of gold ions exiting the EBIS LINAC before injection into the booster, were examined. Computations and some experimental data indicate that none of these issues is a show stopper. Preliminary calculations indicate that single pass cooling is feasible;more » momentum spread can be reduced by more than an order of magnitude in about one meter. Hence, this option cooling deserves further more serious considerations.« less