Towards a Laser-driven polarized $^3$He Ion Beam Source

Full text

PoS(PSTP2015)002
Towards a Laser-driven polarized 3He Ion–Beam
Source
Ilhan Engin∗a, Markus Büscher b,c, Oliver Deppert d, Laura Di Lucchio e,
Ralf Engels a, Simon Frydrych d, Paul Gibbon f, Annika Kleinschmidt d,
Andreas Lehrach a,g,h, Markus Roth d, Friederike Schlüter b,
Katharina Strathmann b, Florian Wagner d
aInstitut für Kernphysik, Forschungszentrum Jülich GmbH, D-52425 Jülich, DE
bPeter Grünberg Institut, Forschungszentrum Jülich GmbH, D-52425 Jülich, DE
cInstitut für Laser- und Plasmaphysik, Heinrich-Heine-Universität Düsseldorf, Universitätsstr. 1,
D-40225 Düsseldorf, DE
dInstitut für Kernphysik, Technische Universität Darmstadt, Schloßgartenstr. 9, D-64289 Darmstadt, DE
eFLA Linear Accelerator Technologies, Deutsches Elektronen-Synchrotron DESY, Notkestrasse 85,
D-22607 Hamburg, DE
fInstitute for Advanced Simulation, Jülich Supercomputing Centre, Forschungszentrum Jülich GmbH,
D-52425 Jülich, DE
gJARA-FAME (Forces and Matter Experiments), FZJ and RWTH Aachen University, DE
hIII. Physikalisches Institut B, RWTH Aachen University, D-52056 Aachen, DE
e-mail list:
i.engin@fz-juelich.de —m.buescher@fz-juelich.de —o.deppert@gsi.de —
laura.di.lucchio@desy.de —r.w.engels@fz-juelich.de —
s.frydrych@gsi.de —p.gibbon@fz-juelich.de —a.kleinschmidt@gsi.de —
a.lehrach@fz-juelich.de —markus.roth@physik.tu-darmstadt.de —
f.schlueter@fz-juelich.de —k.strathmann@fz-juelich.de —f.wagner@gsi.de
ABS TR ACT: In order to investigate the polarization degree of laser-accelerated 3He ions from
a pre-polarized 3He gas–jet target, several challenges have to be overcome beforehand. One
of these includes the demonstration of the feasibility of laser-induced ion acceleration out of
gas–jet targets. In particular, the ion–emission angles as well as the ion–energy spectra have to
be determined for future polarization measurements. Such an experiment was performed at the
PHELIX Petawatt Laser Facility, GSI Darmstadt. As laser target, both 4He, and in a second step,
unpolarized 3He gas were applied.
XVIth International Workshop in Polarized Sources, Targets, and Polarimetry, PSTP2015,
14-18 September 2015,
Bochum, Germany
∗Speaker.
c
Copyright owned by the author(s) under the terms of the Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). http://pos.sissa.it/

PoS(PSTP2015)002
Towards a Laser-driven polarized 3He Ion–Beam Source Ilhan Engin
1. Introduction
Nuclear polarized 3He is of particular importance for fundamental research since the spins
of the two protons are oriented anti-parallel so that the nuclear spin is basically carried by the
unpaired neutron. That is why polarized 3He [1] can be used, for example, as an effective polarized
neutron target for studying the neutron structure by scattering with polarized electrons [2]. For
many experiments in nuclear and particle physics, such as experiments with stored particle beams,
the use of polarized 3He–ion beams would be advantageous. 3He gas can be polarized for long
times at standard conditions. However, building a spin-polarized 3He ion source for nuclear and
particle physics experiments with high degrees of polarization is extremely challenging. Until
now, only a few approaches have shown promise — but not with the desired particle currents or
an adequate beam polarization [3,4,5]. At Brookhaven National Lab’s Relativistic Heavy Ion
Collider (RHIC) attempts are now being made to develop a polarized 3He ion–beam source [6].
Conventional accelerators reach fundamental, technological, and, as one of the most important
aspects, financial limits of the achievable particle energies. Some of these limitations do not apply
to laser-induced particle acceleration. During the past 50 years the achievable laser intensities have
been increased continuously. Since the invention of chirped pulse amplification (CPA) in 1985
[7], the higher-and-higher intensities have opened new applications for laser–physics experiments.
With a high-intensity laser pulse impinging on a suitable target, a relativistic plasma is formed out
of which charged particles can be accelerated to energies of several MeV. An unsolved question in
this context is the influence of the strong laser and plasma fields on the spins of the accelerated par-
ticles. Two scenarios are possible here: either the magnetic fields of the incoming laser beam or the
induced plasma change the spin direction of the accelerated particles, or the spins are sufficiently
robust that the short laser pulse has no effect on the spin alignment of a polarized target [8].
For the second scenario, there is no data given which would allow a scientifically proven es-
timation of the behavior of nuclear spins (ωlarmor,3He =32.4MHzT−1B) in magnetic laser–plasma
fields (B∼O103−105T, temporal continuance of 102−103fs). It has to be experimentally in-
vestigated if the polarization can be conserved inside plasma during laser–acceleration processes,
which would open up the possibility of nuclear fusion with polarized fuel, in which the cross-
sections for nuclear fusion reactions theoretically can be enhanced, leading to higher energy yields
compared to the case of unpolarized fuel [9].
While the first scenario (polarization creation by laser–particle interactions) has already been
investigated with conventional foil targets by spin-dependent hadronic proton scattering off silicon
nuclei [8], for the second one (polarization conservation during laser–plasma interactions) polar-
ized 3He gas can be used as production target. The relaxation time of 3He depends on several
conditions, e.g. gas pressure or field gradients of an external magnetic holding field. During the
ionization processes, also the absence of one electron in the atomic shell leads to a rapid decrease
of the polarization degree: the interaction time τHF for the coupling of the nuclear spins to the spin
of the remaining electron is of the order of about 100 ps (GHz energy level). Thus, a full ionization
of the polarized 3He has to be accomplished within a few picoseconds. This can readily be achieved
with currently available laser intensities. We report here on preliminary results of a first measure-
ment to demonstrate the general feasibility of laser-induced ion acceleration out of helium–gas
jets at PHELIX, GSI Darmstadt. The experiments were conducted with both 4He and unpolarized
2

PoS(PSTP2015)002
Towards a Laser-driven polarized 3He Ion–Beam Source Ilhan Engin
3He gas as laser target. The main ion–emission angles as well as the ion–energy spectra could be
determined.
2. Ion Acceleration in underdense Plasmas
When a high-intensity laser pulse with laser frequency ωLis focused onto a target, an energy
transfer from the incident pulse to the atoms or molecules occurs (the so-called heating process) and
a plasma is formed. The eigenfrequency of the collective electron motion against the plasma-ion
background is called (electron) plasma frequency ωp∝√ne, with nebeing the electron number
density. The relation between plasma and laser frequencies determines whether a laser pulse is
able to propagate through plasma or not: it can propagate through the plasma until the plasma fre-
quency becomes equal to the frequency of the electromagnetic wave. This limiting case is given
for a critical plasma density nc∝ω2
L. Hence, depending on nc, and therefore on the applied laser
and target parameters, laser–plasma interactions generally can be categorized in overdense (e.g. for
solid targets, ωp>ωL) and underdense (e.g. for gaseous targets, ωp<ωL) interactions with var-
ious absorption and charged particle–acceleration mechanisms [10,11]. In underdense plasma
targets, relativistic self-focusing and channeling can occur such that the laser pulse stays focused
over several Rayleigh lengths and the energy transfer to the plasma electrons can remain for longer
interaction times [12]. With today’s laser intensities, only electrons can be accelerated directly by
the incident laser pulse, e.g. by the ponderomotive force which is proportional to the negative radial
gradient of the laser intensity, Fpon ∝−∇IL: electrons are irreversibly expelled from regions of
higher intensity while the inert massive ions remain virtually unaffected on their original position.
This ponderomotive process −Ov2
oscis symmetric, the resulting density channel is cylindrical.
Large secondary electric fields of the order of several TV/m can be generated by this charge sep-
aration. The arising Coulomb forces and the given ponderomotive forces compete against each
other until a Coulomb explosion takes place in which the remaining ions are accelerated radially
outwards from the location of highest ion density [13].
The physics of this process can be simulated with the help of a particle-in-cell code. Within
this study, EPOCH 2D [14] was used on the Jülich supercomputer JURECA. A 2D simulation
box was filled with 50,000 ×6,250 grid points distributed over an area of 2,000µm in xtimes
250µm in y. The spatial resolution was 25 grid points per µm (grid size of 40 nm). Inside this
box, a neutral 4He/3He gas jet was initialized which is then ionized by the simulated laser pulse
(which propagates in positive x-direction, centered in y). The particle–density distribution was
adapted from experimental interferometrical characterizations of the gas flow through a supersonic
de Laval nozzle (nozzle throat of 0.5 mm, backing pressure of 25 bar): 6th-order superGaussian
particle–density profile with a maximal particle density of ngas
max =0.06nc. The laser parameters
were set according to the PHELIX parameters: focus intensity IL=1.4×1019 W cm−2, wavelength
λL=1.053µm, pulse duration τL=0.8 ps, focus diameter FWHM of 25.7µm, critical density
nc=1021 cm−3, leading to a normalized vector potential of a0≈3.3.
Figure 1(a) illustrates the temporal evolution of the normalized electron (top) and He2+ion
densities (bottom) in pseudo colors for t=3.5ps (left column) and t=6.5ps (right column) after
the laser pulse entered the box at the left boundary. A channel both in ion and electron density is
generated around the laser–propagation axis. The simulation predicts strong self-focusing effects
3

PoS(PSTP2015)002
Towards a Laser-driven polarized 3He Ion–Beam Source Ilhan Engin
followed by filamentation. For larger times, the channel widens and the sharp structures smear out
due to the presence of several filaments. Approximately at the location of the gas–jet center (at
x≈900µm), the laser pulse starts to disperse in the underdense plasma regions. The sheath of the
channel for both particle species is characterized by an enhanced particle density. In Fig. 1(b),
the angular He2+ion–energy distribution can be regarded. The number of simulated particles is
indicated by pseudo colors. It becomes obvious, that in the transversal direction, i.e. around ±90◦
with respect to the laser–propagation direction (0◦), two peaks in ion energy Eare present. This
information is important for planning the corresponding laser beamtime (or to be more precise:
the setup for ion diagnostics) beforehand. Although the simulation predicts a maximal ion energy
of about 10 MeV at ±90◦, the density of the plotted pseudo colors, i.e. the ion number, has to be
considered as well: an experimentally detectable signal is not expected for the highest ion energies.
Figure 1: EPOCH-simulated data (PHELIX): (a) Temporal evolution of the normalized electron– and 4He2+ion–
number densities for t=3.5ps (left column) and t=6.5ps (right column) after the laser pulse entered the simulation
box at the left boundary. (b) Angular He2+ion–energy distribution for t=6.5ps.
3. Experimental Studies
The PHELIX parameters were set to: beam energy (after compression) of EL=40−50 J, pulse
duration of τL=0.4−1.1ps, wavelength of λL=1.053µm. The laser beam was focused using a
90◦off-axis parabolic mirror with an f-number of 6.8. Before each shot, the elliptical laser focus
was aligned to a minimal spot size (11µm×15 µm). Peak intensities of about 1.4×1019 W cm−2
could be reached in the focus. As laser target, both 4He and unpolarized 3He gas were used. A de
Laval nozzle with a nozzle throat of 0.5 mm was attached in order to shape a gas jet with adequate
density profile (sharp density ramps at the outer gas–jet regions with a near-flat-top profile in the
center). Different backing pressures were applied during the experimental beamtime: in case of
4He gas, the pressures were 30 bar and 15 bar, and in case of 3He gas a maximal backing pressure
of 8 bar was available. The maximal particle density inside the gas jet is proportional to the applied
backing pressure in front of the nozzle: the densities for these pressure regimes in the focus height
were 6×1019 cm−3≈0.06 nc, 3.25 ×1019 cm−3≈0.03 nc, and (in case of 3He) 1.67×1019 cm−3≈
4

PoS(PSTP2015)002
Towards a Laser-driven polarized 3He Ion–Beam Source Ilhan Engin
0.02nc, respectively. As main ion diagnostics, a removable radiochromic film (RCF) [15] wrap-
around detector was mounted close to the laser–target interaction region cylindrically around the
nozzle in order to measure the angular ion distribution qualitatively. The GAFCHROMIC HD-
V2 RCFs were wrapped in 5µm thick Al foil in order to protect the detectors from side-scattered
laser light. Hence, He1+,2+ions with energies >1.6MeV were able to pierce the Al shielding
and irradiate the RCF. The recorded dose on the RCF detectors was due to both ion species and
gamma radiation as background signal. In addition, three Thomson parabola spectrometers (TP)
[16,17,18] were placed at three angles relative to the laser direction (wedge-shaped capacitor with
a HV of 3 kV, B≈0.6T, covered solid angles 170 −370 nSr). The TPs were armed with Agfa
MD4.0 image plates (IP) [19,20] and for one laser shot with TASTRAK CR-39 SSNTDs [21].
Appropriate CST simulations for the given TP setup were conducted in order to gain information
about the energy–deflection dependencies of all ion species 3,4He1+,2+inside the TP fields. Thus,
experimental data for the specific deflection parameters on the IP–detector plane could directly be
related to the corresponding ion energies.
Within the PHELIX experiments, 3,4He1+,2+ions could successfully be accelerated to MeV
energies. The ion–angular distribution as well as the energy spectra for all ion species for specific
emission angles could be extracted. The results are in line with the corresponding EPOCH simula-
tions. Figure 2illustrates the scanned image of an irradiated RCF from the left side relative to the
laser–propagation direction. The gray scale values were converted in pseudo colors. In transversal
direction (±90◦) a peak in ion signal with a FWHM of ϕfwhm =23◦indicates the main ion–emission
angles at which the TPs were aligned for the following measurements: −{80◦,90◦,100◦}.
Figure 2: Irradiated RCF (left side relative to the laser–propagation direction). The FWHM of the transversal peak
around ±90◦on the focus height is ϕfwhm =23◦.
Since the IPs used in the experiment could not be calibrated with 3,4He1+,2+ions before-
hand, the obtained IP signals did not yield any credible quantitative information about the laser-
accelerated ion number in a specific solid angle. But, the ion–energy spectra, i.e. the normalized
signal intensity (per MeV and Sr, lin. scale) against the ion energy (in MeV, lin. scale), could be
extracted from the detector raw data. In order to get an impression of the real ion number, one TP
was equipped with CR-39 detectors for one laser shot. The risk of oversaturating the SSNTDs was
minimized by choosing not the main ion–emission angle for the TP measurement (here: −80◦)
and by applying a decreased helium backing pressure (here: 15 bar, i.e. ngas
max =0.03nc). Figure 3
exemplifies the laser-accelerated ion–energy spectra for one laser shot (laser energy of 43.8 J, max-
imal neutral He–gas density in a height of 500µm above the nozzle exit, i.e. the shooting height,
ngas
max =0.06nc) and for the main ion–emission angle of −90◦. Both spectra look familiar regarding
5

PoS(PSTP2015)002
Towards a Laser-driven polarized 3He Ion–Beam Source Ilhan Engin
thermal energy spectra as they are common for laser-accelerated particles: a peak in ion number
(or signal intensity) at low energies and a decrease in signal for higher energies with a saddle-like
structure in between. 4He2+ions could be accelerated to higher energies, which is an important
information regarding future polarization measurements of laser-accelerated 3He2+ions out of a
polarized 3He gas–jet target. Since the 3He1+ions are assumed to be unpolarized they will cause
a disturbing background signal. In order to suppress these signals, the lower-energetic 3He1+ions
can be filtered with Al degrader foils of adequate thickness so that a more or less pure 3He2+ion
signal can be investigated. The high-energy and low-energy cut-offs in case of 4He2+in Fig. 3
are given with 4.6 MeV (with a normalized energy uncertainty of ∆E E −1=0.032) and 0.84 MeV
(∆E E −1=0.014). The single CR-39 measurement for extracting the real ion number for one laser
shot (obtained at −80◦,i.e. in a solid angle of 356 nSr, and with a smaller gas density) in total
yielded approximately 1.95 ×1011 Sr−1He1+and He2+ions. This number will be increased by
choosing the main ion–emission angle for the polarimetry as well as by increasing the backing
pressure to 30 bar.
Figure 3: 4He1+,2+ion–energy spectra at −90◦relative to the laser–propagation direction: in blue 4He2+, in red
4He1+. Laser energy: 43.8 J. Target parameters: maximal neutral He–gas density in a height of 500 µm above the nozzle
edge ngas
max =0.06nc.
4. Conclusion
The experimental results proved the general feasibility of laser-driven 3,4He–ion acceleration
out of unpolarized gas–jet targets at PHELIX. With this sine qua non for a spin–polarization mea-
surement of laser-accelerated 3He2+ions, an appropriate layout of a polarized 3He gas–jet target
available for laser–acceleration experiments can be prepared. The optimal laser and target param-
eters could be identified for a future laser beamtime at PHELIX.
6

PoS(PSTP2015)002
Towards a Laser-driven polarized 3He Ion–Beam Source Ilhan Engin
5. Acknowledgments
The authors gratefully acknowledge the strong personal support of both R. Maier (IKP, FZ
Jülich) and T. Stöhlker (HI Jena, GSI Darmstadt). Furthermore, sincere appreciation is expressed
to the Institute for Nuclear Physics (IKP, FZJ) and to the Plasma Physics staff (PHELIX, GSI
Darmstadt) for their great assistance. Last but not least, the authors show their gratitude to the
Central Institute for Engineering, Electronics and Analytics (ZEA, FZJ) for the technical support
as well as to the Institute for Laser and Plasma Physics (ILPP, HHU Düsseldorf) staff for fruitful
discussions and for using their scanner.
References
[1] K. Krimmer et al.,Nucl. Instr. Meth., vol. 611, issue 1, 2009, doi:10.1016/j.nima.2009.09.064
[2] M. Tanaka, Nucl. Instr. Meth., vol. 402, no. 2-3, 1998, doi:10.1016/S0168-9002(97)00896-6
[3] D. O. Findley et al.,Nucl. Instr. Meth., vol. 71, no. 2, 1969, doi:10.1016/0029-554X(69)90001-9
[4] W. E. Burcham et al.,Nucl. Instr. Meth., vol. 116, no. 1, 1974, doi:10.1016/0029-554X(74)90569-2
[5] R. J. Slobodrian, Nucl. Instr. Meth., vol. 185, no. 1-3, 1981, doi:10.1016/0029-554X(81)91257-X
[6] J. Maxwell et al.,Phys. Part. Nuclei, vol. 45, no. 1, 2014, doi:10.1134/S1063779614010651
[7] D. Strickland, and G. Mourou, Opt. Commun., vol. 56, no. 3, 1985, doi:10.1016/0030-4018(85)90120-8
[8] N. Raab et al.,Phys. Plasmas, vol. 21, no. 2, 2014, doi:10.1063/1.4865096
[9] H. Paetz gen. Schieck, Few-Body Syst., vol. 54, no. 12, 2013, doi:10.1007/s00601-012-0485-0
[10] V. T. Tikhonchuk, Nucl. Instr. Meth., vol. 620, no. 1, 2010, doi:10.1016/j.nima.2010.01.051
[11] H. Daido et al.,Rep. Prog. Phys., vol. 75, no. 5, 2012, doi:10.1088/0034-4885/75/5/056401
[12] P. Monot et al.,Phys. Rev. Lett., vol. 74, 1995, doi:10.1103/PhysRevLett.74.2953
[13] G. S. Sarkisov et al.,Phys. Rev. E, vol. 59, 1999, doi:10.1103/PhysRevE.59.7042
[14] T. D. Arber et al.,Plasma Phys. Contr. F., vol. 57, no. 11, 2015, doi:10.1088/0741-3335/57/11/113001
[15] W. L. McLaughlin et al.,Radiat. Phys. Chem., vol. 10, no. 2, 1977, doi:10.1016/0146-5724(77)90017-6
[16] R. F. Schneider et al.,J. Appl. Phys., vol. 57, no. 1, 1985, doi:10.1063/1.335389
[17] D. C. Carroll et al.,Nucl. Instr. Meth., vol. 620, no. 1, 2010, doi:10.1016/j.nima.2010.01.054
[18] J. T. Morrison et al.,Rev. Sci. Instrum., vol. 82, no. 3, 2011, doi:10.1063/1.3556444
[19] I. J. Paterson et al.,Meas. Sci. Technol., vol. 19, no. 9, 2008, doi:10.1088/0957-0233/19/9/095301
[20] P. Leblans et al.,Materials, vol. 4, no. 6, 2011, doi:10.3390/ma4061034
[21] R. P. Henke et al.,Nucl. Instr. Meth., vol. 97, no. 3, 1971, doi:10.1016/0029-554X(71)90250-3
7