Collimation of High Intensity Ion Beams*

Intense ion beams with small phase space occupation (high brilliance) are mandatory to keep beam losses low in high current injector accelerators like those planned for the Facility for Antiproton and Ion Research (FAIR). The low energy beam transport from the ion source towards the linac has to keep the emittance growth low and has to support the optimization of the ion source tune. The Frankfurt Neutron Source at Stern-Gerlach-Zentrum (FRANZ) is currently under construction and might be a possible test stand for a collimation channel. An intense beam of protons (2 MeV, up to 150 mA) will be used for neutron production using the Li(p, n)Be reaction for studies of the astrophysical s-process. A collimation channel, which can be adjusted to allow the transport of beams with a certain beam emittance, is an ideal tool to optimize the ion source tune in terms of beam brightness. Therefore a collimation channel in the Low Energy Beam Transport section (LEBT) will be used. Through defined apertures and transversal phase space rotation using focusing solenoids the outer part of the beam will be cut. Simulations and design done so far will be presented. COLLIMATION PRINCIPLE The brilliance of an ion beam is defined as B = I εxx′ ·εyy′ with I being the ion beam current and ε being the emittance in the xx′ and yy′ sub phase spaces. A collimation channel is an ideal tool to increase brilliance in the LEBT. The Collimation Channel comprises two main functionalities: • An online diagnostic for the ion source brilliance tuning • To dump particles outside the acceptance of the accelerator in a controlled way before entering the cavities The ion beam is extracted from the source and in an ideal case matched to the collimation channel. The channel is built upon three identical cells consisting of an aperture plate, a drift, a focussing element, another drift and ends with the aperture plate at the beginning of the following cell. A collimation channel needs to be designed according to the parameters of the ion source (maximum beam radius and angle, the particle distribution and the maximum ion ∗Work supported by HIC for FAIR within the LOEWE initiative of the state of Hesse, Germany † [email protected] beam current) as well as the acceptance of the accelerator behind the channel. Figure 1 shows the principle of phase space rotation and cutting of the particle distribution in the 2-dimensional sub phase space. (a) 1st aperture (b) 2nd aperture (c) 3rd aperture Figure 1: Phase space rotation and cutting. The beam particles are within the emittance ellipse (grey). (a) outer particles are cut, then phase space rotation, (b) beginning of second cell: cutting and rotation. (c) exit of channel, last aperture with cut. Scheme includes only two cells with 30◦ rotation per cell. The idea is to cut the outer part of the beam in coordinate space (round aperture plates), which is equal to a vertical cut in phase space. This is shown in Fig. 1 (a). The transmitted particles are drifting towards the solenoid and experiencing a transverse phase advance through the magnetic field (and the drift behind it). At the end of the cell another cutting takes place by the aperture at the beginning of the proceeding cell. Since a total phase advance of 90 degrees is required, a phase advance of approximately 30◦ per cell is needed in a 3 cell setup. The more cells are used, the more precise phase space manipulation can be done. By setting calculated values for magnetic fields as well as apertures, only particles in a defined phase space volume (emittance) are transmitted through the entire channel. All other particles are stoped at the apertures within the channel. If the beam current is measured by using a Faraday cup downstream the channel, it is possible to increase the brilliance of the beam by just tuning the source to the maximum cup current without changing the phase advance or the aperture diameters. Such a channel has been built up at National Superconducting Cyclotron Laboratory in East Lansing, MI, USA. First experimental studies of the collimation principle have been completed [1]. Since a collimation channel is a powerful tool for ion source tuning, GSI is aiming for this principle with respect to the new FAIR facility. Proceedings of IPAC2011, San Sebastián, Spain WEPC177 04 Hadron Accelerators T19 Collimation 2403 C op yr ig ht c ○ 20 11 by IP A C ’1 1/ E PS -A G — cc C re at iv e C om m on sA tt ri bu tio n 3. 0 (C C B Y 3. 0) SIMULATIONS FOR THE FRANZ LEBT The idea for a collimation channel demonstrator was the integration of such a device into the FRANZ facility at IAP in Frankfurt. FRANZ is the Frankfurt Neutron Source at Stern-Gerlach-Zentrum which is under construction. FRANZ is a neutron-production facility for astrophysical experiments [2]. The ion source produces a 150 mA proton beam which is accelerated to about 2 MeV and will bombarde a Li-target in order to produce 5 × 10 n/s. A schematic of the LEBT of the facility is shown in Figure 2. FRANZ would be an ideal test stand for the collimation channel, because one would not only have to deal with the function of the channel itself, but also the collimation of such a high-current beam and the problems of dumping a lot of power on the surface of the apertures. The simulations were initially done using COSY Infinity [3], but it had to be changed to Lintra [4], since space-charge effects have to be taken into account for this high current proton beam. The volume-type ion source of FRANZ is fed with gaseous hydrogen. The extracted ion species are H, H 2 and H 3 . A p + beam current of about 150 mA will be achieved, but there will be also the heavier beam fractions like H 2 and H + 3 which might have a total beam current up to 30-40 mA. All simulations were carried out with an extraction radius at the source of 6 mm and a maximum divergence angle of 55 mrad, which is a reasonable assumption. The main fraction of the heavier particles should be dumped before entering the chopper system (see Fig. 2). Since the beam dynamics for the FRANZ LEBT are nailed down almost completely, there are only three positions, where collimation apertures could be integrated into the beam line and therefore only 2 cells would be possible as it is shown in Fig. 2. Figure 2: Possible collimation channel layout at FRANZ (black bars represent possible positions for collimation apertures). Another critical point is the aperture of the chopper system (see Fig. 2 and 3), which is very narrow with a smallest aperture radius of 18 mm. Therefore variability of the matching solenoid 1 as well as solenoid 2 were checked in order to verify the possibility of collimation at the planned positions. The magnetic fields were varied in steps of 5 mT. Out of 945 combinations of solenoid 1 and 2 only ten combinations show a transmission of more than 98% and six combinations of more than 99% of protons through the chopper system. There is only one single combination where 100% protons are transmitted through the chopper (before looking to collimation at all). This is achieved with maximum field in solenoid 1 of 275 mT and 415 mT in solenoid 2. The envelopes of the beams are shown in Fig. 3. The boundary limits of variation of the solenoid strength is therefore restricted to a very narrow interval. Figure 3: Envelopes of p (dotted red, 150 mA), H 2 (green, 20 mA) and H 3 (magenta, 20 mA). In the nominal case 100% of p are transmitted through the LEBT. All other/unwanted fractions are stopped as early as possible. Most of the heavy H 2 and H + 3 particles are lost in front of and inside the first short solenoid. Introduction of a cooled aperture plate at position z≈1200 mm could be used in order to dump the power of these beam fractions in a controlled way. The remaining 0.6% of H 2 and H 3 are dumped between the entrance of the choppper (z=1600 mm) and the last solenoid (z=3300 mm). Protons are transported without losses towards the radio-frequency quadrupole accelerator (RFQ). The loss plot shows, that with this ideal setting it is possible to get 100% of the p to the RFQ and also get rid of all H 2 and H + 3 particles (see Fig. 4) far before being injected into the cavity. As Fig. 2 shows, apertures can only be introduced into the beamline at three distinct positions in front of solenoid 2, inbetween solenoid 3 and 4 as well as right behind solenoid 4, because the RFQ will be at the focal point of the proton beam which is at an approximate z-position of 3700 mm (see Fig. 3). Calculations have shown, that in the nominal case for the FRANZ LEBT the phase advance in the first cell would be approximately 255 degrees and 117 degrees in the second cell which corresponds to xy-space rotations of 133◦ and 84◦. WEPC177 Proceedings of IPAC2011, San Sebastián, Spain 2404 C op yr ig ht c ○ 20 11 by IP A C ’1 1/ E PS -A G — cc C re at iv e C om m on sA tt ri bu tio n 3. 0 (C C B Y 3. 0) 04 Hadron Accelerators