ION CATCHER SYSTEM FOR THE STABILISATION OF THE DYNAMICPRESSURE IN SIS18∗

C. Omet† , H. Kollmus, H. Reich-Sprenger, P. SpillerGSI Darmstadt, Planckstr. 1, 64291 Darmstadt, Germany

Abstract

In synchrotrons operated with intermediate charge stateheavy ion beams, intensity dependent beam losses havebeen observed. The origin of these losses is the changeof charge state of the beam ions at collisions with residualgas atoms. The resulting m/q deviation from the referencebeam ion leads to modified trajectories in dispersive ele-ments, which finally results in beam loss. At the impact onthe beam pipe, gas molecules are released by ion stimulateddesorption which increase the vacuum pressure locally. Inturn, this pressure rise will enhance the charge change- andparticle loss process and finally cause significant beam losswithin a very short time. In order to suppress and controlthe gas desorption process, a dedicated ion catcher systemincorporating NEG coated surfaces and low-desorption ratematerials has been developed and two prototypes were in-stalled in SIS18. The design of the scraper and measuredeffect on the dynamic residual gas pressure are presented.

INTRODUCTION

The new accelerator facility FAIR at GSI relies on theexisting SIS18 synchrotron as injector. Currently most ofthe experiments are performed with highly charged U 73+-beams. Higher intensities will be available by avoidingfurther stripping in front of SIS18 and accelerating lowcharged ions, which also raises the space charge limit.Therefore, U28+ has been selected as the reference ion forthe FAIR project.

During machine experiments in SIS18, extremely fastand intensity dependent particle losses in connection witha very strong dynamic behavior of the residual gas pres-sure have been observed. The intensities were far be-low the space charge limit of the machine. The rea-son for the observed losses is a charge change of thebeam particles, which finally degrades the vacuum by ionstimulated desorption. A quantitative treatment of theselosses has been developed and implemented into the code”STRAHLSIM” [1].

In order to reduce the charge change induced beamlosses, an upgrade of the SIS18 UHV system and the in-stallation of a dedicated desorption catcher system has beenplanned. In the following, the most recent experimental re-sults obtained with the catcher prototypes are summarized.

∗Work supported by EU, contract No. 515876† c.omet@gsi.de

PROJECTILE IONISATION

Heavy ions change their charge state at collisions withresidual gas atoms. Cross sections for electron capturecan be described with Schlachter’s formula [2]. A big un-certainty for simulations are still the projectile ionisationcross sections, especially at high energies. Detailed cal-culations for the electron loss cross sections of U28+ havebeen conducted recently by means of the relativistic LOSS-R code [3] and at the same time were measured experimen-tally in the ESR at GSI [4]. The LOSS-R calculations showthat the cross sections slowly decrease with energy up to 2GeV/u, as shown in figure 1.

For multiple ionisation, cross sections are mainly de-termined by heavy residual gas constituents, e.g. Argon.Therefore, in order to achieve sufficiently large beam lifetimes, these residual gas components have to be removedfrom the vacuum system.

1E-24

1E-23

1E-22

1E-21

1E-20

1E-19

1 10 100 1.000 10.000 100.000

cros

s se

ctio

n / m

²/at

om

beam energy / MeV/u

HCNOAr

e--capture

e--loss

Figure 1: Calculated projectile charge change cross sec-tions for collisions of U28+-ions with various residual gascomponents.

ION IMPACT INDUCED DESORPTION

Particles with a trajectory modified by projectile ionisa-tion will most probably hit the vacuum tube under grazingincidence. Thereby, depending on the projectile energy,molecules attached loosely to the wall’s surface are des-orbed at rates of η � ≈ 5× 103 . . . 1× 106 molecules/ion.If a beam ion hits the surface of the beam pipe or anyother component under perpendicular angle of incidence,the desorption rate has been measured to be about η⊥ ≈1 × 102 . . . 3 × 103 molecules/ion for energies of a few

Proceedings of EPAC08, Genoa, Italy MOPC099

04 Hadron Accelerators T19 Collimation and Targetry

295

MeV/u. The desorbed gases consist to a large extent of COand H2.

The physics behind the desorption process itself is mean-while quite good understood. Detailed investigations havebeen performed by the GSI UHV group using an ERDA(elastic recoil detection analysis) technique [5]. It could beshown that the lowest measured desorption rate of η⊥ ≈25 . . . 80 molecules/ion are reached with a pure Gold sur-face. The measured desorption rate depends on thermaltreatment and history [6]. The measurements have beenconducted at the HLI (high charge state injector) in GSIunder perpendicular angle of incidence with a dE/dx nor-malized for 11.4 MeV/u U28+-ions.

CATCHER SYSTEM FOR IONISATIONBEAM LOSS

SIS18 is currently being upgraded for the operation withhighest beam intensities required for the FAIR project. Forthis purpose, the dynamics of the residual gas pressure hasbeen studied with STRAHLSIM. It has been shown that be-side reducing the injection losses and increasing the ramprate, a large distributed pumping speed and an effectivecatcher system are required for the minimization of beamlosses.

Injection losses produce a very high pressure bump inthe range of several orders of magnitude during a few μs.This pressure bump has to be removed by the UHV systemas fast as possible. For this purpose, NEG coating of thedipole and quadrupole chambers is foreseen to increase thedistributed pumping speed by a factor of 100. For the con-trol of the charge change generated beam losses of U 28+

and to confine the desorbed gases, a dedicated catcher sys-tem will be installed in each section of SIS18.

As the static beam lifetime can always be enhanced by alower residual gas base pressure, the dynamic pressure canbe stabilized by the stronger pumping and the dedicated ioncatcher system. Catchers can be designed such that theycontrol most of the ionized particles and confine the pro-duced desorption gases. It is essential to prevent the des-orbed gases from reaching the beam axis and to minimizethe charge change rate. Due to the SIS18 triplet lattice, thecatcher system could be built such that it does not reducethe acceptance of the machine.

Two prototypes of the ion catcher system have beendeveloped and installed during the winter shutdown2007/2008 in two sections of SIS18 (S02 and S03). Asshown in figure 2, the catcher consists of two transversallymovable copper beam absorbers, coated with a few 100 nmGold and an underlying diffusion barrier made of Nickel.To prevent the desorption gases from reaching the beamaxis, the absorber is enclosed by a secondary chamber. Toachieve a high pumping speed near by the absorbers, bothprimary and secondary catcher chambers have been coatedwith TiZrV-NEG at GSI.

During machine experiments with U28+-beams inFebruary 2008, the qualitative behavior of the projectile

Figure 2: Horizontal cut through the installed SIS18 ioncatcher prototype. Yellow: beam, red: secondary chamber,brown: beam absorbers.

0,0

0,2

0,4

0,6

0,8

1,0

1,2

0

1E+09

2E+09

3E+09

4E+09

5E+09

6E+09

7E+09

8E+09

9E+09

0 50 100 150 200

Ion

curr

ent o

n ab

sorb

er /

a.u.

num

ber

of U

28+

part

icle

s

beam energy / MeV/u

Number of particles

I_S02 (E-Loss)

I_S03 (E-Loss)

I_S02 (EC)

I_S03 (EC)

Figure 3: Measured ion currents on absorbers as a functionof energy. Dotted: beam loss by electron capture, solid:beam loss by electron loss.

ionisation cross sections vs. energy have been confirmedby measuring the ion current onto the beam absorbers. Forthese measurements, the absorbers have been positionedleft and right of the beam axis. Thereby it was possibleto measure directly the ion current of U27+ on the outerside and U29+ on the inner side of the ring. As can be seenfrom figure 3 and figure 1, the measured ion currents forboth ionisation and electron loss have a similar dependenceon energy as the cross sections calculated by the relativisticLOSS-R code. The measured ion currents have amplitudesproportional to the amount of ionisation, i.e. proportionalto the residual gas pressure in the section before the catch-ers. This observation is the first direct prove that projec-tile ionisation is indeed the driving process of the observedstrong beam loss.

The measured and gas specific corrected pressures inthe catcher chamber together with the average pressure ofSIS18 are shown in figure 4. Under beam loading, the aver-age pressure in SIS18 raises due to ion stimulated desorp-

MOPC099 Proceedings of EPAC08, Genoa, Italy

04 Hadron Accelerators

296

T19 Collimation and Targetry

160

140

120

100

80

60

40

20

0

20

1E-10

1E-09

15:20:10 15:48:58 16:17:46 16:46:34

Abs

orbe

r po

sitio

n / m

m

pres

sure

/ m

bar

time

p Primaryp Secondaryp SIS avg.Absorber pos.

WedgeWedge

Block Block Block

Wedge

Figure 4: Measured residual gas pressures in the catcherchamber and the whole SIS18.

tion from 3.3× 10−10 mbar to about 1× 10−9 mbar. Thepressure rise in the catcher chamber itself decreases signif-icantly when the absorber block is moved to its designatedposition.

From the measured pressure rise, the effective, for thebeam visible desorption rate of the wedge shaped absorberwas calculated to be η � ≈ 278 mol/ion and for the blockshaped absorber η⊥ ≈ 55 mol/ion. These values are quiteclose to the measured desorption rates at the ERDA teststand. Therefore, the low desorption rate Gold coatingworks exactly as intended only for perpendicular angle ofincidence. After approval of the catcher prototypes, a se-ries of 10 catchers will be installed together with 6 NEGcoated quadrupole chambers in the shutdown end of 2008.Since two of the 12 SIS18 sections are blocked by acceler-ator installations, the efficiency of the overall system willbe reduced by a few percent.

U28+ beam intensities

The successive optimizations during the ongoing SIS18machine development program have pushed the number ofextracted U28+-particles in comparison to 2001. The high-est measured number of extracted particles is now about8 × 109, as can be seen in figure 5. As the upgrade is stillprogressing, the U28+-experiments have been conductedwith only half of the dipole magnet chambers replaced byNEG coated chambers (which is about 7 % of the ring’s cir-cumference) and with two prototype ion catchers installed.The maximum transmission was achieved at a ramp rate ofB = 7 T/s.

Major improvements have been reached by a new closedorbit correction scheme [7] and 10 s long pumping breaksbetween the machine cycles. The fact that the last measureimproves the number of extracted particles by more than10-13 % indicates that the observed losses are strongly de-pendent on pressure and not driven by other high-currenteffects such as resonances or longitudinal or transversal in-stabilities.

0.0E+00

2.0E+09

4.0E+09

6.0E+09

8.0E+09

1.0E+10

1.2E+10

1.4E+10

1.6E+10

1.8E+10

2.0E+10

0 0.2 0.4 0.6

num

ber

of U

28+

part

icle

s

t / s

24.10.2006

22.03.2007

12.06.2007

28.08.2007

14.05.2008

Figure 5: Measured (solid lines) and calculated (dottedlines) time dependent number of U28+-particles in SIS18.

CONCLUSIONS

Both the UHV upgrade with the NEG coated chambersand the dedicated catcher prototypes show the intended sta-bilization of the dynamic vacuum in SIS18. Furthermore, itcould be shown that the main upgrade measures have to beaccompanied by a precise machine setting to avoid pressurebumps by systematic beam losses.

REFERENCES

[1] C Omet, P Spiller, J Stadlmann, and D H H Hoffmann.Charge change-induced beam losses under dynamic vacuumconditions in ring accelerators. New Journal of Physics,8(11):284, 2006.

[2] A. S. Schlachter, J. W. Stearns, W. G. Graham, K. H. Berkner,R. V. Pyle, and J. A. Tanis. Electron capture for fast highlycharged ions in gas targets: An empirical scaling rule. Phys.Rev. A, 6(10.1103/PhysRevA.27.3372):3372–3374, 1983.

[3] V.P. Shevelko et.al. New relativistic LOSS code. to be pub-lished, 2008.

[4] G. Weber. Untersuchung der Umladungsverluste undStrahllebensdauern gespeicherter U28+ Ionen. Diplomar-beit, Ruprecht-Karls-Universitat Heidelberg, 2006.

[5] M. Bender, H. Kollmus, and W. Assmann. Understandingof Ion Induced Desorption Using the ERDA Technique. InProceedings of the EPAC06, 2006.

[6] Markus Bender. Untersuchung der Mechanismen Schweri-oneninduzierter Desorption an Beschleunigerrelevanten Ma-terialien. Doktorarbeit, Johann Wolfgang Goethe-UniversitatFrankfurt, 12 2007.

[7] A. Parfenova, G. Franchetti, B. Franzcak, M. Kirk, andC. Omet. SIS18 closed orbit correction using local bumpmethod. Internal note, GSI, 12 2006.

Proceedings of EPAC08, Genoa, Italy MOPC099

04 Hadron Accelerators T19 Collimation and Targetry

297