The University of Manchester Research
A compact linear Paul trap cooler buncher for CRIS
DOI:
10.1016/j.nimb.2019.04.054
Document Version
Accepted author manuscript
Link to publication record in Manchester Research Explorer
Citation for published version (APA):
Ricketts, C., Cooper, B., Edwards, G., Perrett, H., Billowes, J., Binnersley, C., Cocolios, T., Flanagan, K., Garcia
Ruiz, R. F., de Groote, R. P., Gustafsson, F. P., Koszorús, Á., Neyens, G., Vernon, A., & Yang, X. F. (2019). A
compact linear Paul trap cooler buncher for CRIS. Nuclear Instruments & Methods in Physics Research. Section B:
Beam Interactions with Materials and Atoms. https://doi.org/10.1016/j.nimb.2019.04.054
Published in:
Nuclear Instruments & Methods in Physics Research. Section B: Beam Interactions with Materials and Atoms
Citing this paper
Please note that where the full-text provided on Manchester Research Explorer is the Author Accepted Manuscript
or Proof version this may differ from the final Published version. If citing, it is advised that you check and use the
publisher's definitive version.
General rights
Copyright and moral rights for the publications made accessible in the Research Explorer are retained by the
authors and/or other copyright owners and it is a condition of accessing publications that users recognise and
abide by the legal requirements associated with these rights.
Takedown policy
If you believe that this document breaches copyright please refer to the University of Manchester’s Takedown
Procedures [http://man.ac.uk/04Y6Bo] or contact uml.scholarlycommunications@manchester.ac.uk providing
relevant details, so we can investigate your claim.
Download date:10. Aug. 2022
A compact linear Paul trap cooler buncher for CRIS
C.M. Ricketts
a,∗
, B.S. Cooper
a,b
, G. Edwards
a,b
, H.A. Perrett
a,b
, J. Billowes
a
, C.L. Binnersley
a
, T.E. Cocolios
c
, K.T. Flanagan
a,b
,
R.F. Garcia Ruiz
d
, R.P. de Groote
f
, F.P. Gustafsson
c
,
´
A. Koszor
´
us
c
, G. Neyens
c,d
, A.R. Vernon
a
, X.F. Yang
e
a
School of Physics and Astronomy, The University of Manchester, Manchester, M13 9PL, United Kingdom
b
Photon Science Institute, Alan Turing Building, The University of Manchester, Manchester, M13 9PY, United Kingdom
c
KU Leuven, Instituut voor Kern- en Stralingsfysica, B-3001 Leuven, Belgium
d
EP Department, CERN, CH-1211 Geneva 23, Switzerland
e
School of Physics and State Key Laboratory of Nuclear Physics and Technology, Peking University, Beijing 100871, China
f
Department of Physics, University of Jyv¨askyl¨a, PB 35 (YFL) FIN-40351 Jyv¨askyl¨a, Finland
Abstract
A gas-filled linear Paul trap for the Collinear Resonance Ionisation Spectroscopy (CRIS) experiment at ISOLDE, CERN is currently
under development. The trap is designed to accept beam from both ISOLDE target stations and the CRIS stable ion source. The
motivation for the project along with the current design, simulations and future plans, will be outlined.
Keywords: Ion trapping, Laser spectroscopy, 3D printing
1. Introduction
Gas-filled linear Paul traps are used at a variety of radioac-
tive beam facilities to provide bunched beams with low emit-
tance to a range of experimental setups [1, 2, 3]. Bunched
beams are essential to Collinear Resonance Ionisation Spec-
troscopy (CRIS) [4, 5, 6] because the required high-power lasers
are only available with repetition rates of the order of 100 Hz,
meaning the ion beam must also be pulsed to remove duty-cycle
losses. The pulsed-time structure also reduces optical pumping
to atomic dark states, further increasing ionisation efficiency
[7]. Minimising the ion beam energy spread via in-trap buffer-
gas cooling reduces the Doppler broadening of the resonant ex-
citation steps, leading to higher precision in extracted hyperfine
structure and isotope shifts. For more information on the CRIS
technique, see the contributions from
´
A. Koszor
´
us et al. and
A.R. Vernon et al. in these proceedings [8, 9].
CRIS at ISOLDE [10] currently utilises the ISCOOL gas-
filled linear Paul trap (RFQcb). ISCOOL has been essential to
the success of the CRIS experiment, providing bunched beams
with <1 eV energy spread and <5 µs width. When fully opti-
mised the transmission efficiency of ISCOOL can reach 70%,
however for lower masses the usable efficiency decreases sig-
nificantly [11]. This article describes the development of an
independent RFQcb for the CRIS beamline, to be placed af-
ter the CRIS ion source (CRISIS) [12]. Installing a dedicated
RFQcb would decouple the CRIS beamline from the rest of
ISOLDE; ions injected into the trap from ISOLDE and from
CRISIS would have the same properties when they reach the
CRIS interaction region. This would allow for continual opti-
misation of trapping parameters and efficiency with stable iso-
topes, improving beam quality and transport efficiency for ra-
dioactive beam experiments. Frequent reference measurements,
∗
christopher.ricketts@manchester.ac.uk
performed by extracting a stable reference isotope from ISOLDE,
are required during experiments on radioactive atoms to keep
track of slow drifts in wavenumber readout. Switching masses
on the ISOLDE separator for this purpose is time consuming
(approximately 6 minutes per mass change); during recent CRIS
campaigns, between 5% and 10% of the total beamtime was
spent cycling the magnets. In addition, more than 33% of the
remaining time was spent adjusting bending and focussing elec-
tric potentials to maximise ion beam transmission through CRIS.
A dedicated trap would allow CRISIS to be used for the refer-
ence measurements, and would almost entirely eliminate the
beam transport optimisation required at the start of radioactive
beam experiments. This period is when the yield of radioac-
tive cases is at its maximum and additional data collection at
this point would have a disproportionate effect on the overall
signal significance. As the majority of background counts at
CRIS scale with the beam current, for example non-resonant
laser ionisation of isobaric contamination, the total background
observed is proportional to the interaction region atom through-
put, hence the signal significance scales with the product of the
beam intensity, measurement time and total efficiency.
2. Prototype RFQcb overview
Anticipating the available lab space, the current prototype
design uses a short trapping region, 20 cm in length, with four
cylindrical rods at 5.3 mm internal radius. Sine or square wave
RF driven at frequencies of the order of MHz will be applied
to the rods to trap the ions radially, while helium buffer gas (at
pressure in the range 10
−3
to 10
−1
mbar) will be used to reduce
the amplitude of the radial oscillation. The ions will be trapped
longitudinally using DC potentials applied to four printed cir-
cuit boards inserted between the rods. The rods and PCBs are
mounted on two 3D printed endcap pieces.
Preprint submitted to Elsevier April 4, 2019
Figure 1: The current linear RFQ design. Upper - schematic of the trap design including the buffer gas containment cylinder (dark grey) around the trapping region,
the 3D printed components (light yellow) and the PCB electrodes (dark yellow). 70% of the losses occur in the region indicated by the red box. Lower - SIMION
simulation showing electric potential contours (red) and A = 70 ion trajectories starting at the beam waist, at 0.034 mbar buffer gas pressure (blue). The ions travel
from left to right. Losses are minimal during extraction.
We have demonstrated that polylactic acid biopolymer (PLA)
used in 3D printers is vacuum compatible. Figure 3 shows
the pressure over time in a 3-way T-piece vacuum chamber
with 440 L s
−1
pumping capacity. As the figure shows, when
a 20 mm side length cube of 3D printed PLA was placed in
the vacuum chamber, the pressure decreased to 2 × 10
−7
mbar
after 5.5 hours of pumping, equal to the pressure reached by
the empty chamber within the same pumping time. This pres-
sure is compatible with the vacuum requirements of the RFQcb.
Residual gas analysis was performed at 3.9 × 10
−8
mbar with
the PLA sample present and at 6.7 × 10
−8
mbar when the cham-
ber was empty. No significant change was seen in the compo-
sition of the gas after adding the sample, indicating that out-
gassing of the PLA is not significant at these pressures. The
RGA spectra are shown in Figure 4. Flow rate calculations in-
dicate the 4 mm apertures on either side of the trapping region
will provide the required differential pumping to reach pres-
sures below 5 × 10
−6
mbar outside the trap with two stages of
differential pumping and up to 0.1 mbar internal buffer gas pres-
sure.
The entire trap is biased to a potential slightly below 30 kV
to decelerate the incoming beam from 30 keV to approximately
100 eV. At 30 keV, the Doppler broadening of the resonant exci-
tation lineshape is comparable to the natural linewidth of many
resonant transitions and a larger platform operating voltage is
not required. SIMION simulations, shown in the lower part of
Figure 1, have indicated that more than 70% of the ion beam
losses occur during the injection-deceleration phase due to the
narrow input aperture required for differential pumping and col-
lisions with residual buffer gas. To mitigate these losses, a se-
ries of PCB ring electrodes were introduced into the model to
produce a longitudinal potential structure of the form V(z) =
V
0
(1 − e
−αz
). This produces a transverse focussing effect with
a focal point at a position proportional to 1/α [13]. The re-
sults of the simulations are shown in Table 1 and an output
emittance plot for A = 70 is shown in Figure 2. These used
an input 30 keV focussed beam with 20 π mm mrad transverse
emittance and a Gaussian radial profile with 0.5 mm standard
deviation. The ions were released from the simulated trap after
2 ms. The transmission efficiency at A = 70 was 54% and the
final simulated emittance of the cooled beam at this mass was
1.5 π mm mrad.
3. Outlook
The benefits of installing an independent gas-filled linear
Paul trap at the CRIS beamline have been explained and the de-
sign outlined is under development. Simulations have indicated
the feasibility of a short trap design and have informed possi-
ble design improvements. The design offers a cost effective and
high efficiency method to trap ions for radioactive beam stud-
ies. This work also illustrates the possible uses of 3D printed
parts in vacuum, enabling cheaper and faster methods for de-
sign prototyping.
Acknowledgements
The authors would like to thank Colin Reed and Andrew
McFarlane for preparing the PLA printed component and vac-
uum testing samples. This work was supported by the ERC
2
Mass, A Pressure / mbar Transmission probability / % Acceptance / π mm mrad Output emittance / π mm mrad
12 0.012 11 10 16
70 0.034 54 20 1.5
Table 1: Simulated RFQ trapping parameters with total ion transmission probabilities, the input beam acceptance and output beam emittance after reacceleration at
optimised buffer gas pressures. A 30 keV focussed input beam with 20 π mm mrad transverse emittance and a Gaussian radial profile with 0.5 mm standard deviation
was used. The ions were released after 2 ms.
Figure 2: A plot showing the displacement along the transverse x-axis vs the
velocity angle relative to the beam axis in the xz-plane, where z is aligned along
the beam axis, of the transmitted A = 70 ions in a SIMION simulation of the
RFQcb (green points). The ellipse showing the RMS transverse emittance in x
is also included (red line).
Figure 3: Pump down curves for an empty chamber (yellow) and for the same
chamber with a 20 mm side length cube of PLA (orange). The chamber con-
taining the PLA sample reached a pressure equal to that of the empty chamber,
2 × 10
−7
mbar, after 5.5 hours of pumping, indicated by the pink dashed line.
Features in both curves below the order of 2 × 10
2
s are related to different
pumping stages turning on.
Figure 4: A plot showing the partial pressure at each mass detected by a residual
gas analyser of the empty vacuum chamber at 6.7 × 10
−8
mbar (red) and the
vacuum chamber containing the PLA test cube at 3.9 × 10
−8
mbar (black). No
significant change in the RGA spectrum is seen after adding the PLA sample,
indicating that outgassing from the sample is not significant at these pressures.
Consolidator Grant No. 648381 and the Science and Technol-
ogy Facilities Council Grants ST/P004423/1 and ST/S002316/1.
References
[1] A. Nieminen, et al. Phys. Rev. Lett., 88:094801, 2002.
[2] F. Herfurth, et al. Nuclear Instruments and Methods in Physics Research
Section A: Accelerators, Spectrometers, Detectors and Associated Equip-
ment, 469(2):254 , 2001. ISSN 0168-9002.
[3] T. Brunner, et al. Nuclear Instruments and Methods in Physics Research
Section A: Accelerators, Spectrometers, Detectors and Associated Equip-
ment, 676:32 , 2012. ISSN 0168-9002.
[4] Y. A. Kudriavtsev et al. Applied Physics B, 29(3):219, 1982. ISSN 1432-
0649.
[5] K. T. Flanagan, et al. Phys. Rev. Lett., 111:212501, 2013.
[6] C. Schulz, et al. Journal of Physics B: Atomic, Molecular and Optical
Physics, 24(22):4831, 1991.
[7] R. P. de Groote, et al. Phys. Rev. Lett., 115:132501, 2015.
[8]
´
A. Koszor
´
us, et al. In preparation.
[9] A. Vernon, et al. In preparation.
[10] M. J. G. Borge et al. Journal of Physics G: Nuclear and Particle Physics,
45(1):010301, 2018.
[11] E. Man
´
e, et al. The European Physical Journal A, 42(3):503, 2009. ISSN
1434-601X.
[12] R. F. Garcia Ruiz, et al. Phys. Rev. X, 8:041005, 2018.
[13] R. F. Garcia Ruiz et al. Collinear Laser Spectroscopy on exotic Ca iso-
topes towards new magic numbers N=32 and N=34, 2015. Presented 01
Oct 2015.
3