Optimization of the SIS100 Lattice and a Dedicated Collimation System for Ionisation Losses

Transcription

1 Optimization of the SIS100 Lattice and a Dedicated Collimation System for Ionisation Losses P. Spiller, K. Blasche, B. Franczak, J. Stadlmann, and C. Omet GSI Darmstadt, D Darmstadt, Germany Abstract: A triplet lattice with six-fold symmetry was originally proposed for SIS100 and presented in the Conceptual Design Report (CDR) [1]. Meanwhile the design strategy for the SIS100 lattice had to be changed and a new lattice was presented in the Technical Status Report. Especially for the planned operation with intermediate charge state heavy ions (U 28+ ), a new lattice design concept was developed. Measurements in SIS18 and theoretical studies have shown that significant beam loss appears by the interaction of beams of these ions with the residual gas. The reason for the importance of the interaction process even at very low residual gas pressures is the large ionization cross section of heavy ions with intermediate charge states. During high intensity operation, beam loss induces a large amount of ion desorption which leads to a fast degradation of the residual gas pressure. Therefore, a new type of collimation system was proposed to control beam loss due to charge changing processes and to confine the generated desorption gas. The optimization of the efficiency of this new collimation concept was chosen as the major lattice design criteria for SIS100. A lattice structure was developed which contributes to an effective stabilization of the dynamic behaviour of the residual gas pressure. INTRODUCTION SIS100 is the main accelerator in the FAIR (facility of antiproton and ion research) project. It is a new large synchrotron designed for a maximum magnetic rigidity of Bρ= 100 Tm, i.e. comparable in size to the proton synchrotrons PS (CERN) and AGS (BNL). SIS100 will be built for the acceleration of high intensity and high energy proton and ion beams. Based on the FAIR research program the following key beam parameters were defined: a) For the radioactive beam program about uranium ions per second at energies from 400 to 2700 MeV/u in a single short bunch with pulse lengths from 50 to 100 ns, b) For the antiproton facility protons per pulse at 29 GeV every 5 seconds, c) For plasma physics research at least uranium ions in a single short bunch (50 to 100 ns) at energies from 400 to 1000 MeV/u Since the space charge limit after injection into synchrotrons scales with the ratio of A/q 2, these intensities of heavy ion beams e.g. uranium beams, can only be reached by using intermediate charge states (q=28+). The maximum beam intensity per synchrotron pulse is increased by the factor 6.8 by using the charge state 28+ instead of 73+, which is the charge state currently used in SIS18 at much lower intensities. In addition, a short synchrotron cycle time of τ 1 s is required to achieve the high average beam intensity of about uranium ions per second. Therefore, new technical features have to be developed for SIS100: (1) operation at a very low static base pressure of p= mbar, i.e. in the XHV range, (2) careful control of beam losses, especially beam losses due to charge exchange q=28 29 with residual gas molecules, by implementing a well-designed collimator system, (3) s.c. synchrotron magnet operation at a high ramp rate of 4 T/s, (4) bunch compression for high intensity proton and uranium ion beams to provide a single short bunch with a pulse length of about 50 ns for the production and storage of secondary beams. These technical features define new challenges for the synchrotron design, which go beyond the design of existing proton synchrotrons. DYNAMIC VACUUM AND COLLIMATION CONCEPT Severe beam losses driven by ionisation processes were observed in machine experiments with U 28+ beams in SIS18 (fig. 1). The strong dynamics of the residual gas pressure and the ionisation rates can be explained by the large multiplication factor for gas particles desorbed from the beam pipe. In SIS18 ionised projectiles (i.e. U 28+ U 29+ ) hit the inner beam pipe with an angle of about 35mrad in and behind the dipole magnets. Secondary particles are desorbed with a rate of η~10 4. These particles increase the local pressure, which causes growing ionization

2 and beam loss rates and may finally end up in a vacuum instability [2]. FIGURE 1: Beam intensity evolution during a vacuum instability in SIS18. The red curve shows significant losses of an intermediate charge state heavy ion beam (U 28+ ) during its interaction with a desorption driven residual gas bump. The black curves indicates the situation at lower initial intensity realized by a shorter chopper window. A significant reduction of the ionisation rate may be achieved if the desorbed gases could be prevented from interaction with the revolving beam. For this purpose a dedicated collimation system was proposed which shall localize the beam losses and confine the desorbed gases in a secondary vacuum chamber [3]. An experimental collimator (in this case as an injection septum protection) similar to the planned system of desorption collimators has been built and installed in the SIS18 section S12. First experiments have proven that this concept should be suitable to control the desorption gases [4]. In the frame of the SIS18 upgrade program each section of SIS18 will be equipped with a collimator of this type. Each collimator of the planned system consists of a wedge with an inclination of about 15, made of (or coated with) low desorption material. The surface on which the beam ions are stopped points in opposite direction to the beam axis (fig. 2). This wedge is placed in a heat-shielded secondary chamber, which is connected to the first (70 K for the heat shield) and second (20 K for the secondary chamber) stage of a cold head of a cryo pump. The large pumping power (estimated to be l/s) will even enhance the SIS12/18 main vacuum. The ionised beam particles will enter the secondary chamber through a small window, which allows only 7% of the desorbed gases to leave the secondary chamber. Finally only the gases and the part of ions which can not be caught by the collimators directly (5-13% of the ionised particles, depending on the number of installed collimators) have to be handled by the UHV/XHVsystem. This means that the effective desorption rate and thereby gas load for the UHV system will be extremely reduced. It is planned to constructed, built and test a first prototype of the SIS18 collimation system in This system, together with the planned NEG coating of SIS18 is expected to lead to the performance improvement needed for the FAIR project. A similar concept of collimator technology is being considered for SIS100. However, the collimator module will be installed inside the quadrupole cryostat. First design studies for the collimator module have been performed. In order to avoid additional heat load to the cryosystem and to avoid freezing out of rest gas on the surface, the operation temperature of the wedge will be in the range between 50K and 100K. The wedge is installed inside a dedicated cooled chamber and separates these in two UHV sections. The wedge is movable in radial direction. The cooled chamber surface with a temperature of about T 5 K is used as a cryopump for the desorbed gases. More detailed technical design of the collimation chamber based on the technical specification listed in the table 1 has been started [5]. The R&D phase is scheduled for TABLE 1: Parameters of the SIS100 cryogenic desorption collimator system. Number of collimators per 10 superperiod Wedge Length of the wedge, m Thickness, kg/m 2 Material SS Temperature of the wedge, K Weight of the wedge, kg 2.3 Coating low-η-material Alignement tolerances (x/z), mm 0.5 Radial movement, mm 20 Accuracy of radial position, mm 0.5 Open chamber aperture, mm 130x65 Heat realise from the beam, J 100 Cold components Cold chamber diameter, m Operating temperature, K 4.5 Static heat leaks at T= 4.5 K, W 3 Cooling power, W >100 Length of the collimator module, m

3 support frame wedge collimator radiation shield (~77K) manipulator primary stage (~77K) cold heat ion beam secondary chamber (~4-20K) secondary stage (~4-20 K) FIGURE 2: Proposed collimator for SIS18 with ion beam, support frame and secondary chamber/heat shield. SIS100 CHARGE SEPARATOR LATTICE The SIS100 lattice must provide the required space in each cell for the collimation system and ensure that practically no ions are lost elsewhere. This can be achieved if the lattice structure is optimized such that each lattice cell acts as a charge separator, providing a waist at the position of maximum separation. The acceptance of the synchrotron should not be affected by the collimation system. Several lattice structures have been designed and investigated. For each lattice, the maximum fraction of ions which can be caught by the collimators has been determined and compared to the total amount of lost ions. This collimation efficiency was investigated as a function of the radial distance between the wedge and the beam edge. The goal was to achieve 100% collimation efficiency over a large radial range by an optimum layout of the lattice structure. The calculations have shown that within the CDR triplet lattice, a maximum collimation efficiency of only 98 % could be achieved. Consequently several versions of doublet DF-structures were designed. It could be shown that only doublet lattices provide practically 100% of collimation efficiency up to large distances from the beam edge. In addition, a much larger acceptance (a factor of 1.74 larger than the CDR lattice) could be reached. The collimation efficiency of the CDR triplet lattice and an optimized doublet lattice is plotted in figure 3. Two different positions of the collimators in the arcs of SIS100 were considered: in front of each D-quadrupole and behind the D- quadrupole (in between the two quadrupoles). An advantage of the first option is a possible simplification of the cryostat design and, as a result, reduced investments cost and simplified maintenance. FIGURE 3: Collimation efficiency as a function of the distance to the beam edge for the CDR (triplet) and the TR (doublet) reference lattices. The collimators are situated in each regular cell with dispersive elements. The position between the quadrupoles results in a higher collimation efficiency over a large radial distance. Figure 4 shows the optimized doublet focusing structure with two dipoles in each cell.

4 -40. y[mm].. x[mm] 80. = Dispersion 0. path length [mm] FIGURE 4: SIS100 lattice structure with missing dipoles at both sides of the arc and minimised dispersion function In order to suppress the dispersion in the straight sections and to achieve a minimum dispersion function in the arcs only one dipole is foreseen in the first and last cell of an arc. The parameters of the new SIS100 lattice, which has still six-fold symmetry are listed in table 2. TABLE 2: SIS100 lattice parameters as presented in the technical report Number of superperiods 6 Number of regular cells 90 # Dipole magnets 120 Magnetic field [T] (max) 1.9 / (2.0) Length [m] 2.74 Bending radius [m] Bending angle [ ] 3 Aperture h / v [mm] # Quadrupole magnets 180 Lenght [m] 1 Gradient [T/m] / / 65 (elliptic) Max. beta functions h/v [m] 19.5 / 22.2 Max dispersion [m] 1 m In order to provide the additional space required for the 60 collimators, without reducing the total drift length in the straight sections needed for injection, extraction, Rf- and other systems, the circumference was enlarged to m (11/2 times the circumference of SIS 18). Correction elements and pickups will be combined with the quadrupoles and the collimator to a cold mass. STORAGE MODE LATTICES Some of the investigated lattice structures showed a jump rise of the collimation efficiency up to almost 100% over a wide range of radial distance of the collimator from the beam edge (fig. 3). FIGURE 5: Comparison of the storage efficiency in standard doublet lattices (red) and lattices showing storage mode properties (green). For comparison the TR triplet lattice (black) and a FODO lattice (blue) are plotted. This jump is a consequence of the capability of the lattice to store a significant fraction of the single ionized beam particles (U 29+ ). However, the lattice structure itself may differ only slightly in geometry from others. Storage capability is achieved by increasing the number of regular cells while keeping the ring circumference constant. In parallel the focusing strength in the horizontal plane is being enhanced and the tune value increased. These measures, in connection with the missing dipole concept, leads to a very low dispersion function in the arcs and the quality of collimation efficiency shown in figure 3.

5 2,5 E ,0 E N 1,5 E ,0 E N 2 0 % In je c tio n lo s s e s N 0 % In je c tio n lo s s e s N 2 0 % In je c tio n lo s s e s w ith N E G N 0 % In je c tio n lo s s e s w ith N E G N 2 0 % In je c tio n lo s s e s w ith N E G a n d C o llim a to rs N 0 % In je c tio n lo s s e s w ith N E G a n d C o llim a to rs 5,0 E ,0 E ,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0 Figure 6: Calculated SIS18 beam intensity evolution during a 4 Hz booster operation. The curves indicate the influence of initial beam losses, the collimation system and the enhancement of the UHV pumping power. t / s INTEGRATED BEAM LOSS AND DYNAMIC VACUUM SIMULATIONS For a fast prove of the collimation efficiency of different lattices, a dedicated tracking code was developed. The original purpose of the code was to calculate the loss distribution of ionized particles along the circumference. It became clear that this performance may be used as a basis for estimates of the dynamic behaviour of the residual gas pressure. On the other side more quantitative results were required on the contribution of the proposed collimation systems for stabilization of the residual gas pressure in SIS18 and SIS100. Therefore the tracking core of the program was supplemented by the most important physical processes affecting the residual gas pressure and its composition. The following mechanisms were implemented: Initial systematic beam losses (e.g. multi turn injection losses, Rf capture losses) Projectile and target ionization and capture cross sections and the resulting ionization and multiple ionization degree Collimation efficiency for each generated charge state Energy dependence of the collimation efficiency and of the cross sections Effective desorption rate of the collimation system (leakage rate) Initial residual gas composition Desorption coefficient and assumption for the composition of the desorped gases Desorption generated by target ionisation Coulomb scattering with the residual gas By means of the extended tracking code it was possible to estimate the pressure rise during a U 28+ machine cycle in SIS18. The beam survival curve and the pressure curves were calculated. It could be shown that, as expected, only a combination of different measures will be able to stabilize the SIS18 pressure and to enable the high current booster operation with intermediate charge state heavy ions (fig 6). The figure shows the importance of a) avoiding any initial systematic beam losses, b) enhancing the average pumping power by a factor of 100, and c) installing a collimation system which confines the generated desorption gases. It is planned to improve the code such that also short term mechanisms which may affect the beam life time in SIS100 can be estimated, e.g. gas desorption from one side of a cryogenic beam pipe, single transition of the generated pressure bump and sticking on the surface at the opposite side. REFERENCES [1] CDR Conceptual Design Report, GSI (2002) [2] E. Mustafin et. al., Nucl. Instr. and Meth. A 510 (2003) [3] P. Spiller, Desorptionskollimator zur Kontrolle von Desorptionsgasen, GSI internal note (2004) [4] C. Omet and P. Spiller, GSI internal note GSI- Acc-Report [5] A.D. Kovalenko et.al., JINR final report contract 2/AC, GSI internal note (2004)