Ingen bildrubrik

1
Cooling of Highly-Charged Ions in a Penning Trap
SMILEtrap (Stockholm-Mainz Ion LEvitation trap)
Team T. Fritjoff, Nagy Szilard (PhD student),
R. Schuch, Atomic Physics Department, Stockholm
University I. Bergström,Manne Siegbahn
Laboratory, Stockholm U. K. Blaum,
GSI/CERN

• Penning trap (SMILE) connected to an
  electron-beam ion source (EBIS) with isotope
  injector for mass measurements
• Exploit highly charged ions for mass
  measurements, because ?cq/mB
• An accuracy in 10-10 range for atomic masses is
  reached by ion selection without cooling.
• Do we really need (want) cooling? How far down
  can one get?
• Prospects of evaporative cooling and adiabatic
  cooling
• Electron cooling

2
Ion Budget
Precision Trap After injection aperture 150
ions Captured 50 ions After energy selection
1-4 ions (2V- 50 mV)

• PreTrap
• After magnet and retardation 106 ions
• 1/250 beam captured 4000
• pretrap length/beam length
• 2 V deep trap / 7 eV energy spread

EBIS delivers e.g. 108 Geq ions 25 on Ge22
2.5x107 ions
3
Mass Measurements using SMILETRAP

• Full computerized control (LabVIEW)
• Easy, comfortable to handle

4
High Resolution Mass Spectrometry with Heavy
High-q Ions

• Result from 208Pb and 198Hg runs

5
Projects in SMILEtrap collaboration with
relevance for HITRAP Goal reach mass uncertainty
of lt 0.1 ppb for very highly charged heavy
ions 0.1 ppb is in reach for light ions Behind
this success were the following measures 1.
Stabilisation of the trap temperature 2.
Stabilisation of the helium pressure 3.
Stabilisation of the frequency synthesizer 4. The
use of short measurement cycles (2 min) 5. The
use of H2 as a intermediary mass reference 6.
The fast exchange (a few s) between the mass of
ref. ion and the

ion of interest. 7. The application of Ramsey
excitation 8. Exactly calculated binding energies
(Lindroth and Indelicato) And now a table which
summarizes our results
6
Isotope q Atomic mass (u)
Uncertainty(ppb) 1H 1 1.00727646666
(90) 0.90 3H 1 3.016 049
278 4(29) 0.96 3He 2 3.016 029 323
5(28) 0.93 4He 2 4.002 603 256 8(13)
0.33 20Ne 9, 10 19.992 440 186(14)
0.70 22Ne 9, 10 21.991 385 115(19)
0.86 28Si 12, 13, 14 27.976 926 531
(14) 0.50 36Ar 13, 14 35.967 545
105(15) 0.42 76Ge 22, 23 75.921 402
758(96) 1.3 76Se 24, 25 75.919
213 795(81) 1.1 86Kr 26 85.910 610
729(110) 1.3 133Cs 36, 37 132.905
45159(41) 3.1 198Hg 52 197.966 768
4(6) 3.0 204Hg 52 203.973 494 2(6)
2.9
7
Systematic Errors
8
How could we check?

• Examples

Agreement between charge states for SMILEtrap
values!
9

• Is cooling absolutely necessary?
• Yes, for the resonant circuit pick-up
  detection(Seattle, MIT, Harvard groups) they use
  ion and ref. ion and need them in the same
  position.

We still have to show for mass measurements how
far we can go by selection rather than cooling.
Essential also for HITRAP, but other
experiments may need cooling (g-factor, cold
ions,)
10
Evaporative cooling
pre-trap
precision trap
11
Adiabatic cooling
is based on invariant of action integral with
adiabatic change of fields in this case 1/B
dB/dt ltlt ?c angular momentum Lrmvi is
invariant therefore vi has to decrease when r
increases as B decreases
t
Example (Li et al. Z.Phys.D22, 375,91) for dense
U92 cloud z1cm, B 3 T ? 0.08 Tesla T
300K ? 10 K density 106 ? 4.7 103 cm-3
12
Electron (positron) cooling
cooling times tcool k /q2 memi ve2/ne
Lc tcool 8.9 s (kTe)3/2 A/neq2 kT(eV), ne
(107cm-3)
warm ions
cold e-
Electron-ion recombination (radiative for bare
ions) ?RR 5 1013 Z2/(kTe)1/2 cm3/s
Ratio tcool/tRR 3 10-3 A kTe(eV)
Resistive cooling times tcool k mi /q2 R
13
Present Plans in SMILEtrap

• testing limits in accuracy with present set-up
• 204-208Pbq, Thq, Caq , He,2, 3H masses
• Development of Ramsey method
• for standard procedure in data taking and
  analysis
• Q-value measurements
• Input for neutrino mass determination
• Installation of new 1.2 T pre-trap magnet
• Test of evaporative cooling
• Test of e- cooling