Prog. Theor. Exp. Phys. 2012, 03C001 (23 pages)
DOI: 10.1093/ptep/pts059
Research with fast radioactive isotope beams
at RIKEN
Tohru Motobayashi1,∗ and Hiroyoshi Sakurai1,2
1
RIKEN Nishina Center, Wako, Saitama 351-0198, Japan
Department of Physics, University of Tokyo, Bunkyo, Tokyo 113-0033, Japan
∗
E-mail: [email protected]
2
Received October 11, 2012; Accepted November 6, 2012; Published December 10, 2012
...............................................................................
The RIKEN RI Beam Factory (RIBF) provides intense radioactive isotope (RI) beams with energies around 200 MeV/nucleon in a wide range of unstable nuclei. Part of the facility started
routine operation in 1987 and, in 1990, RI beams from this part became available, with energies of a few tens of MeV/nucleon and the world’s highest intensities at the time for many light
exotic nuclei. To exploit the research opportunities available with such fast RI beams, various
experimental devices have been constructed and several new methods have been developed.
A number of modern aspects of nuclear structure, such as the neutron halo and the disappearance/appearance of magic numbers, have been investigated. Several attempts to approach
astrophysical nuclear processes involving short-lived nuclei have been made. The present RIBF
facility was completed in 2006. It far exceeds other existing facilities in the world in its RI
beam production capability. A variety of research programs have been started, and several
groundbreaking results have already been obtained.
...............................................................................
1.
Introduction
Radioactive isotopes or unstable nuclei have played crucial roles in developing our knowledge on
atomic nuclei. For example, at the beginning of the last century, α particles emitted from the decay
of polonium—one of the radioactive isotopes (RIs) known at that time—were used in the Rutherford
scattering experiment, providing the first indication of the existence of the atomic nucleus. Later
on, various methods for artificial RI production were found, and nuclear physics research has been
extended accordingly.
The idea of employing RIs in the form of a beam for nuclear reaction studies was realized at
LBL in 1985 [1,2]. The projectile-fragmentation reaction with fast primary beams, in this case
800 MeV/nucleon 11 B beams, was used to produce RIs in-flight. The nature of projectile fragmentation, the large cross sections creating fragments with velocities similar to that of the beam, and
emission angles concentrating on the beam direction made this idea applicable to experiments. Using
RI beams of light unstable nuclei, interaction cross sections were measured, and enhancement of the
matter radius was observed, e.g., for the neutron-rich nucleus 11 Li, revealing its halo structure [3].
Another idea for RI beams is to use the ISOL (isotope separator on-line) technique. Radioactive
nuclei produced by nuclear reactions, such as spallation by high-energy protons, are separated and
post-accelerated by an independent accelerator. The first realization of this idea was made a little
later at the laboratory of Leuvain la Neuve in 1989. The first experiment was performed in 1990 on
the astrophysical (p, γ ) reaction with 13 N beams [4]. Radioactive 13 N nuclei were produced at rest
© The Author(s) 2012. Published by Oxford University Press on behalf of the Physical Society of Japan.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0),
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
by the (p,n) reaction on 13 C with protons accelerated to 30 MeV by a cyclotron, then ionized and
post-accelerated to about 1 MeV/nucleon by another cyclotron.
At RIKEN, the first idea, the use of projectile fragmentation, was adopted, and fast RI beams
have been produced. One of the first experimental results was published in 1991, almost at the same
time and on the same subject as the first experiment at Leuvain la Neuve [4] mentioned above.
The Coulomb dissociation method, discussed later, with fast RI beams of 14 O at 87.5 MeV/nucleon
was employed to extract the 13 N(p, γ )14 O cross section, relevant to hot-CNO cycle nuclear burning [5]. The results of these two experiments agreed with each other, demonstrating the usefulness
of experimental studies with RI beams.
These first physics cases already indicate the major directions of research with RI beams: exploring
exotic structures in nuclei with a large imbalance between proton and neutron numbers, and access
to astrophysical nuclear processes involving unstable nuclei. Indeed, several milestones in nuclear
physics along these lines were achieved with RI beams, and the studies at RIKEN have made a big
contribution.
Planning for a facility with a much higher RI beam production capability had been discussed for
many years at RIKEN. After long efforts of design and construction, a new accelerator complex
started operation in 2007. The entire facility, including the old part that had been in operation, was
named the “RI Beam Factory (RIBF)” [6,7]. RIBF is the first “new generation” RI beam facility, and
its performance in providing RI beams is already higher compared with any of the existing facilities
in the world.
The present article is intended to give an overview of the research at RIKEN, especially for experimental studies using fast RI beams. Detailed discussions on several specific subjects are given in
separate articles in this special issue of Prog. Theor. Exp. Phys.
2.
The RI Beam Factory
A heavy-ion accelerator system started routine operation at RIKEN in 1987. RI beams have been
available since 1991 with the projectile-fragment separator, RIPS [8]. The accelerators and RIPS are
shown schematically in Fig. 1 as part (on the left-hand side) of the present RIBF facility. This “old
part” of the RIBF accelerators consists of a four-sector ring cyclotron (RIKEN ring cyclotron, RRC)
coupled with two injectors, an azimuthally varying field (AVF) cyclotron for ions up to around Ca
and a variable-frequency linear accelerator (RILAC) for heavier ions. The maximum energy for light
heavy-ions such as 16 O, or fully stripped ions with N = Z , is 135 MeV/nucleon.
Vector- and tensor-polarized deuteron beams are also available. This unique capability is kept in the
new facility. A series of experiments measuring spin-dependent observables as well as cross sections
of d+p elastic scattering have been performed [9]. Clear signatures of three nucleon forces are seen,
providing an important basis for the fundamental understanding of nuclear interaction. Deuteroninduced reactions were also studied by using the high-resolution spectrometer SMART [10], which is
located in the place of fRC in Fig. 1. For example, the tensor-analyzing powers of the 12 C(d, 2 He)12 N
reaction were obtained, demonstrating its ability to identify J π for spin-flip dipole states [11].
Intense lower-energy ion beams with energies of several MeV/nucleon are also available in the
RILAC experimental hall. They are used for super-heavy element searches by heavy-ion fusion reactions. Long-running experiments have been performed for producing the super-heavy isotope 278 113
by 209 Bi(70 Zn,n) with 5 MeV/nucleon 70 Zn beams. In total, three events have been observed, in 2004,
2005, and 2012 [12–14].
2/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 1. Schematic view of the RIKEN accelerator complex. The part with a gray background is the new part
of the RIBF facility.
With beams from the AVF injector cyclotron, beams of light unstable nuclei at low energies,
typically 5 MeV/nucleon, are produced in the CRIB facility [15]. CRIB was constructed and operated by the Center for Nuclear Study (CNS), University of Tokyo. A number of experiments, mostly
for reactions of astrophysical interest, have been conducted.
In addition, the beams have also been used for various applications to nuclear chemistry, bio- and
medical science, radioisotope production, and materials science.
Fast RI beams, with energies of typically 30–100 MeV/nucleon, are produced with RIPS by projectile fragmentation of light heavy-ion primary beams obtained in the entire accelerator complex.
Thanks to the large angular and momentum acceptance and high bending power of the fragment
separator, RIPS, the RI beam intensity for many light nuclei far from the stability line had been the
world’s highest for many years. Many experiments with the RI beams have been performed, and a
variety of interesting results have been obtained1 .
The new or higher-energy part of RIBF, indicated by the colored background in Fig. 1, has three
cyclotrons: fRC (fixed-frequency ring cyclotron), IRC (intermediate stage ring cyclotron), and SRC
(superconducting ring cyclotron). By further accelerating the beam from the pre-existing accelerator, RRC, the RIBF accelerators can provide beams from protons to uranium up to an energy of
345 MeV/nucleon. A new fragment separator, BigRIPS [16], was built for RI beam production. It
is designed to accept about half of the reaction products from the in-flight fission of uranium ions,
which is expected to have the advantage of efficient production of neutron-rich medium-mass nuclei,
in addition to projectile fragmentation. The RI beam production capability is much enhanced both in
the intensity and in the mass range. The RI beam energies are in the range of 100 to 250 MeV/nucleon,
higher than in the old facility. After the first extraction of primary beams at the end of 2006, RI beam
production started in 2007. With a still very weak primary beam, the first experiment identified two
neutron-rich isotopes, 125 Pd and 126 Pd, in the products of in-flight fission of 238 U [17]. More details
of the accelerator complex are described in a separate article [7].
1
The number of publications to date is approximately 300.
3/23
PTEP 2012, 03C001
3.
T. Motobayashi and H. Sakurai
Methods for experiments with fast RI beams
As demonstrated in the first examples mentioned in Sect. 1, RI beams enable studies of secondary
reactions involving short-lived nuclei. Previously, this was a “dream” of experimentalists, except for
reactions with long-lived isotopes that could serve as a target. Therefore, the study of reactions of
astrophysical interest was limited to cases with stable or long-lived nuclei and therefore to reactions in
steady nuclear burning, such as the one in the sun. RI beams opened new opportunities for studying
explosive nucleosynthesis involving unstable nuclei. The secondary reactions are also, of course,
useful for nuclear structure studies. It should be noted that, in the incident-energy region of RIBF, the
reactions without mass transfer, such as inelastic scattering, charge exchange, nucleon removal, and
fragmentation reaction, have reasonable cross sections, and are therefore suitable for experimental
studies with limited RI-beam intensities.
Another type of application takes an RI beam itself as the object of a nuclear-property study, either
in-flight or after being stopped in a material. Isotopic selection can be made to a certain extent in a
fragment separator such as RIPS and BigRIPS. In addition, fast and firm particle identification can be
made on-line using the signals from detectors in the beam line. The new isotope search is an example
using these fast RI beam properties. The radioactive decays of the stopped RIs provide valuable
information, and delayed coincidence with particle-identified RIs is powerful for spectroscopy of
rarely produced isotopes. More generally, manipulation of stopped or decelerated exotic nuclei is an
important technique in nuclear physics study and other applications.
3.1.
Secondary reaction studies
The first study of secondary reactions was the interaction cross section measurement at LBL mentioned in Sect. 1. For more complicated reactions, new methods should be developed. For example, in
comparison with most stable-beam direct reaction experiments, the roles of the projectile and target
are interchanged: the projectile-like product is the object of interest and the target particle serves as
a probe. Together with a large beam spot size, large angular and energy spreads (typically 2–3 cm in
diameter, a few % and a few degrees, respectively) of the fragmentation-based RI beams, this kinematical situation makes the usual missing-mass measurement impractical. The energy and angle of
the scattered particle are less sensitive to the excitation energy of the residual nucleus, and the poor
beam quality prevents their precise measurement. One solution is to detect the reaction recoils for
cases with light targets such as protons. In direct reactions without mass transfer, e.g., proton inelastic scattering, the recoil energy is usually low at small center-of-mass scattering angles. This limits
the target thickness, keeping it small enough not to stop or seriously degrade the recoil particles.
As alternative methods suitable for experiments with low-intensity beams, we often adopted the following two schemes. One is to measure deexcitation γ rays to identify the excited levels in the final
nucleus. For particle-unbound states, the invariant mass is measured instead by detecting the decay
particles. In both cases, the energy resolution of the final state is not much affected by the large
energy-spread of the incident secondary beams. Therefore, relatively thick targets can be used to
compensate for the low RI-beam intensity.
In the usual fast RI-beam experiments, the beam particle is identified event-by-event, and therefore the experiment does not suffer from possible impurities of unwanted nuclei in the secondary
beam. On the other hand, the maximum tolerable counting-rate in the beam-line detectors limits
the beam intensity in this scheme, which reduces, in a few cases, the overall throughput of the
experiment.
4/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 2. Schematic view of the setup used in the first Coulomb excitation experiment in 1994.
3.1.1. Interaction cross section. In the interaction cross section studies, the attenuation method
was adopted. The number of particles in the beam and after passing through the target are counted.
This method is simple but powerful for systematic investigation of the matter radius along, e.g., an
isotopic chain in a wide range. One of the “Day One” experiments at the new RIBF facility was on
the interaction cross sections of neon isotopes [18]; this will be described later.
3.1.2. Measurement of γ rays. The measurement of deexcitation γ rays can be used to identify
particle-bound excited states populated by RI-beam induced reactions. Since the γ rays are emitted
from the residual nucleus moving at almost the same velocity (v/c ≈ 0.3–0.7) as the beam, the measured photon energy is largely Doppler-shifted; this should be corrected according to the γ emission
angle relative to the direction of the γ ray emitter, which is well approximated by the beam direction
for the direct reaction at energies relevant to RIBF.
The first experiment of this type was performed in 1994 [19], with a setup shown schematically
in Fig. 2. Low-intensity radioactive 32 Mg beams (300 cps) of 49.2 MeV/nucleon bombarded a thick
208 Pb target (350 mg/cm2 ). Deexcitation γ rays from the 2+ state in 32 Mg (moving as fast as 30% of
light speed) were detected by a stack of Na I (Tl) detectors called DALI 2 . The large Doppler shift was
corrected for by referring to the location of the Na I (Tl) crystal that detected the γ rays. Despite the
low-intensity 32 Mg beam, a clear peak for the 2+ –0+ transition was observed and the B(E2) value
could be obtained, demonstrating the usefulness of this method. Also, the extracted large B(E2) of
454 ± 78 e2 fm4 was of high significance, because it supported the idea of the disappearance of the
N = 20 shell closure in 32 Mg.
Later, a new Na I (Tl) array, called DALI2 [20], was constructed. It has a higher efficiency and a
better angular resolution, which is the key to those experiments with higher velocity beams provided
by the new RIBF facility. The DALI2 array consists of 160 Na I (Tl) scintillators with a size of 4.5 ×
8 × 16 cm2 , a full-energy-peak efficiency of 21% and an energy resolution of 8% (FWHM) for 1 MeV
photons.
These DALI and DALI2 setups have been used in many spectroscopic studies with Coulomb
excitation, proton inelastic scattering, and nucleon removal reactions. The high experimental efficiency enables one to find new excited states and/or to measure transition probabilities or deformation lengths with low-intensity beams for nuclei quite far from stability. More details of γ ray
spectroscopy at the new RIBF facility are discussed separately [21] in this special issue.
2
Detector array for low-intensity radiation. Descriptions can be found in Ref. [19].
5/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 3. Velocity diagram for p + 7 Be particle decay from a fast-moving 8 B nucleus in its unbound state.
Fig. 4. Setup of the Coulomb dissociation experiment at RIKEN. The inset shows an extended view of the
detector system for the 8 B Coulomb dissociation experiment.
3.1.3. Measurement of invariant mass. Particle-unbound states excited by fast RI-beam-induced
reactions can be studied by measuring decay particles. Figure 3 shows the velocity diagram for the
process where a fast-moving 8 B in its unbound excited state decays by emitting a proton and 7 Be,
as an example. The excitation energy is obtained from the invariant mass of the p + 7 Be system
relative to the ground-state mass of 8 B. It is equivalent to the sum of the proton separation energy
of 8 B and the relative energy of p + 7 Be. The latter is determined by measuring the velocities of
the two decaying particles and their relative angle. As discussed above, the important point is the
insensitivity of the measured relative energy to the velocity of the excited 8 B nucleus just after the
reaction.
The experimental setup for the 8 B Coulomb dissociation study [22] is shown in Fig. 4. This is the
second invariant-mass experiment at RIKEN, following the first 14 O dissociation study mentioned in
Sect. 1. Radioactive 8 B nuclei were produced by the 12 C + 9 Be interaction at 92 MeV/nucleon, then
analyzed by RIPS, and transported to a 50 mg/cm2 thick secondary target of 208 Pb. The energies and
directions of the fragments were measured to obtain the p − 7 Be relative energy event-by-event, using
6/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 5. Experimental setup for the study of neutron-unbound states in 13 Be.
a E–E plastic scintillator hodoscope. The results provided useful information on the solar-neutrino
production reaction 7 Be(p, γ )8 B, independent of several preexisting data from “direct method”
experiments, among which sizable discrepancies had been noted [22].
Neutron-unbound states can also be studied with the same method. Figure 5 shows the setup used
in the study of the neutron-rich nucleus 13 Be [23]. Beams of 14 Be with an energy of 70 MeV/nucleon
hit a liquid hydrogen target [24]. Unbound states in 13 Be were created by the one-neutron removal
reaction, and their decays were measured by detecting neutrons with neutron counters and protons
with a charged particle hodoscope after magnetic analysis. An “intruder” resonant state with p-waves
at a low energy was identified for the first time in the relative-energy spectrum, showing a trace of
the disappearance of the N = 8 shell gap [23].
In the new RIBF facility, a magnetic spectrometer SAMURAI [25] has been constructed, and the
first experimental studies were made in 2012. SAMURAI is an extension of the setup shown in Fig. 5
to applications in wider ranges of magnetic rigidity and of particle type. SAMURAI is designed to
make efficient invariant-mass measurements of both neutron- and proton-unbound states.
3.1.4. Attempts at low-energy reactions. As discussed earlier, fast beams from RIPS and BigRIPS
are suitable for studying direct reactions without mass transfer. There are other types of reactions
useful as tools for nuclear spectroscopy. Various kinds of transfer reaction and heavy-ion fusion are
examples of such reactions, but their suitable incident energies are typically 5–20 MeV/nucleon for
transfer reactions and a few MeV/nucleon for fusion, much lower than the energy ranges of the RI
beams at RIKEN.
For some specific cases, however, transfer reaction experiments turned out to be feasible at higher
energies within the range of the RIPS fragment separator. Figure 6 shows the incident-energy dependence of the transfer reactions, which is essentially determined by kinematic matching. They are
calculated for the (d,p) and (α, 3 He) reactions on 18 O in the range of 20–100 MeV/nucleon within
the framework of distorted wave born approximation, assuming single-particle final states with an
angular momentum of 3 or 4. As seen in the plots, the (d,p) cross sections (dashed curves) decrease
7/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 6. Energy dependence of the (d,p) and (α, 3 He) reaction cross section on
transitions.
18
O for = 3 and = 4
as a function of energy, whereas the ones for (α, 3 He) (solid curves) exhibit maxima at around
30 MeV/nucleon. This difference is interpreted as being due to the large negative Q values for the (α,
3 He) reaction. At lower angular momentum transfers ( = 0 and 1), the energy dependence does not
show a peak, but (α, 3 He) or (α, t) experiments should still be feasible in the 30 MeV/nucleon region.
Indeed, successful experiments have been done for 22 F [26] and 12 Be [27] beams with a liquid helium
target [24].
Another way is to decelerate the beam to a lower energy. Nuclear spectroscopy experiments by
multistep Coulomb excitation [28] and fusion reaction [29] are such cases. Deceleration was achieved
by using a thick production target together with a thick degrader in the intermediate focus of RIPS.
Due to energy-loss straggling in the production target and the degrader, the decelerated beam has a
considerably large energy spread, which should be measured event-by-event to determine the incident
energy of the reaction of interest. Application of this method to higher-energy RI beams from the new
RIBF facility with sophisticated beam optics is planned; this should extend the research opportunity
to low-energy reactions [30]. It should be noted that this method is free from the chemical properties
of the beam particle, in contrast to the usual ISOL-based RI beams.
3.2.
Studies using separated RIs
3.2.1. Identification of new isotopes. The fragment separator (RIPS or BigRIPS) analyzes RI
beams by their magnetic rigidities with help of a simultaneous energy-loss analysis. The isotopic
selection is not complete, but it indeed provides a useful means to confine the range of isotopes
8/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 7. Setup for nuclear moment studies using the β decay asymmetry.
to be transmitted. The new isotope search experiment mentioned earlier [17] uses this function of
BigRIPS. Thanks to the increase in the 238 U primary beam, another 45 new isotopes were found by
an experiment in 2008 [31]. Isotope identification was made on-line using signals from beam-line
detectors. The magnetic selection of isotopes by BigRIPS considerably reduced the counting rates
of the detectors, and enabled measurement with the maximum primary-beam intensity. The experiment that found the particle stability of 31 F was performed at RIPS with the same method [32]. More
detailed descriptions on the fragment separator and the new isotope search experiments are found in
a separate article in this special issue [33].
3.2.2. Radioactive decay studies. By stopping separated isotopes in a material, various studies
are possible by measuring their radioactive decays. The first successful example along these lines
at RIKEN uses spin-polarized RI beams [34]. With the help of the β-NMR method, data for the
magnetic dipole- or electric quadrupole-moment for many light unstable nuclei have been obtained
for their ground states. The experimental setup is shown in Fig. 7. The RI beam is degraded and
implanted in an appropriate single crystal. The nucleus of interest is finally stopped at a crystallographic site, where a certain magnet field or electric field gradient is acting. By observing the
asymmetric emission of β rays from spin-oriented nuclei, nuclear moments are precisely obtained
with the help of the nuclear magnetic resonance (NMR). Some of the studies with RIPS are discussed
later. Application of this technique at the BigRIPS separator has already been made for an isomeric
state in 43 Sc [35]. Recently a new method to produce highly spin-aligned RI beams was developed.
The two-step projectile-fragmentation reaction combined with the dispersion matching technique
highly enhances the spin-alignment of the relevant RI beams [36].
More standard and traditional techniques measure β decays, isomeric γ decays, or β-delayed
radiation, and spectroscopic information such as the lifetime and transition energy are obtained.
An example is the spectroscopic study of the neutron-rich nucleus 14 Be, in which delayed neutron
emission following β decays was measured [37]. The states populated are identified by measuring
the time-of-flight of decay neutrons. Since β and isomeric γ emissions are important decay modes
in heavier isotopes, this technique is and will be employed more in the new RIBF facility. The recent
experiments on the lifetime of r-process nuclei [38] and isomers in neutron-rich Zr isotopes [39]
are examples of such an application. Fast RI beams from BigRIPS are further delivered downstream
9/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 8. Setup for the laser spectroscopy experiment of trapped 7 Be degraded from a fast RI beam.
through the ZeroDegree spectrometer (ZDS in Fig. 1), and often implanted in an active stopper as
a stack of position-sensitive silicon detectors. Delayed emission is studied with β and/or γ detectors surrounding the stopper. To thoroughly exploit the potential of RIBF, the EURICA (EUroball
RIken Cluster Array) program was launched in 2011 [40]. Euroball cluster Ge detectors in the former
RISING setup at GSI will stay at RIBF for a few years, mainly to conduct decay experiments.
3.2.3. Slow-beam production. Slowing down fast beams to thermal energy is attractive, because
a lot of novel techniques could be applied without being limited by the chemical properties of the
isotope of interest.
Developments based on the rf ion-guide technique have been made by using light neutron-rich
beams from RIPS [41]. Precision hyperfine structure spectroscopy of 7 Be+ in an ion trap [42] is one
of the results of the development. A schematic diagram of the experimental setup is shown in Fig. 8.
A primary beam of 13 C produces a fast 7 Be beam by RIPS, which is degraded in energy and injected
into an ion-guide cell. Slowing down of 7 Be is achieved by the rf ion-guide technique; it is then
transported to an ion-trap apparatus for optical spectroscopy. Trapped 7 Be+ ions are further cooled by
laser radiation down to the μeV range, which allows laser-microwave double resonance spectroscopy
for ground state hyperfine splittings with a precision of 10−7 . The development continues toward
applications with the more extensive RI beams from BigRIPS, under the name of SLOWRI (see
Fig. 1). Various experiments, including precision mass measurements with the multi-reflection timeof-flight MR-TOF technique [43–45], collinear laser spectroscopy to determine root-mean-square
charge radii of unstable nuclei by measuring their isotope shifts [46,47], and hyperfine structure
spectroscopy to extract the radius of a valence nucleon through the Bohr–Weisskopf effect [48,49],
are planned.
4.
New phenomena in unstable nuclei
Since the use of fast RI beams started in the mid-1980s, various new phenomena have been found,
which have deepened our understanding of nuclear structure and nuclear interaction. A few of the
most important subjects studied at RIKEN are presented in this section.
4.1.
Halo and p–n “decoupling”
In nuclei in and close to the stability valley, constituent protons and neutrons behave similarly, under
the influence of the strong proton–neutron attractive interaction. This leads to almost the same radii
for protons and neutrons, and also similar proton- and neutron-transition amplitudes, with the double
ratio (Mn /Mp )/(N /Z ) being almost equal to unity. Here Mn and Mp are respectively the neutron- and
proton-transition matrix elements, with Mp being related to B(E2) by B(E2) = e2 Mp2 . The interest
is in whether any deviation from the above established nature occurs in neutron- or proton-rich nuclei
with a large proton–neutron imbalance. The first example is the neutron halo observed in 11 Li, as
mentioned in Sect. 1.
10/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
4.1.1. Neutron halo and skin. Following the pioneering work at LBL on 11 Li mentioned earlier [1,
2], studies of interaction cross section were performed in RI-beam facilities such as RIKEN and GSI.
Neutron halo and neutron skin structures were also found to occur in some light neutron-rich nuclei
other than 11 Li [50], where neutrons have extended spatial distributions outside of the core where
neutrons and protons are equally distributed. This may be an indication that in certain neutron-rich
regions some neutrons become decoupled from protons despite the strong p–n interaction.
A method to extract the spatial distribution of valence neutrons is to measure interaction cross
sections at different incident energies. The energy dependence of the nucleon–nucleon interaction is
the basis of the method [51]. Comparison with theoretical predictions gives useful information on
the halo wave function.
The nucleus 11 Li has been studied by various different methods. The Coulomb dissociation method
was applied to measure the dipole strength for excitation to an unbound continuum state, which was
used to extract the spatial correlation of the two valence neutrons in the latest study [52].
Attempts to find new neutron halo nuclei have been made at RIPS. The Coulomb dissociation
method was also applied to 11 B and 19 C and their halo nature brought about by an s-wave valence
nucleon was established [53,54]. The halo structure was also found in 22 C by an interaction cross
section experiment at RIPS [55].
The most recent interaction cross section study is on neon isotopes performed at the new RIBF
facility [18]. The results are shown in Fig. 9. Systematic deviation from the trend expected from data
for stable nuclei is clearly seen, and further enhancement is seen in a few isotopes, particularly in
31 Ne, in which the most prominent increase is observed. The former is interpreted as an effect of
deformation, whereas the latter is attributed to the neutron halo.
This halo structure in 31 Ne, the heaviest halo nucleus so far, was also found in an independent
experiment with inclusive nuclear and Coulomb dissociation at 230 MeV/nucleon [56]. Their analysis leads to a possible p-wave neutron contribution to the halo structure, in contrast to the usual halo
nucleus with halo neutron(s) in an s-orbital.
4.1.2. Proton and neutron contributions to quadrupole collectivity. Another aspect of the deviation from the nature of stable nuclei can be found in the behavior of quadrupole collectivity. The
static electric quadrupole moments of neutron-rich boron isotopes were measured by the β-NMR
technique. The isospin dependence of the effective charges of protons and neutrons was observed.
In particular, the neutron effective charge was found to take a small value of about 0.1 for 15 B and
17 B [57]. It should be noted that the observed isospin dependence is compatible with the prediction
by Bohr and Mottelson [58].
This result may be related to the hindered E2 transition in 16 C and its neighbors. The measured
mean lifetime of the 2+ state, 17.7 ± 1.6(sta) ± 4.6(syst) ps [59]3 was particularly long, as seen in
Fig. 10. The figure plots the E(2+ ) and B(E2) data for even–even carbon isotopes. Deviation from
the general trend, E(2+ ) × B(E2) ≈ constant, is clearly seen at A = 16 and A = 18, even for the
shorter lifetime or the larger B(E2) value of 16 C measured at LBL [63].
Together with the large hindrance of the Coulomb excitation amplitude [61], the small B(E2)
values point to a picture in which neutrons contribute almost solely to the 2+ state excitation. Another
3
The value of 75 ± 23 ps, published previously [60], was revised to 34 ± 14(sta) ± 7(syst) ps by taking into
account experimental bias [60], which could not be completely removed in the analysis in Ref. [60]. In view
of the better statistical accuracy of the setup with a high detection efficiency, the quoted lifetime is taken from
the new experiment [59].
11/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 9. Interaction cross sections for neon isotopes measured at 240 MeV/nucleon with carbon targets (solid
circles). Data for stable nuclei mostly taken at around 1 GeV/nucleon are indicated by the open circles. A factor
is multiplied by each original value to take into account the incident energy dependence of the cross sections
obtained by Glauber model calculations. The solid curve represent the mass-number dependence proportional
to A2/3 expected for stable nuclei with constant matter density.
Fig. 10. E(2+ ) and B(E2) of carbon isotopes. The open circle, red solid circle, and blue asterisk represent
the data for 16 C from Refs. [60], [61], and [59], respectively. The solid line for B(E2) is obtained by simple
systematics with constant E(2+ ) × B(E2).
measurement of the proton inelastic scattering, which is more sensitive to neutrons than to protons in
the intermediate energy region, supports the above picture [62]. This suggests decoupling of neutron
and proton motions, and is consistent with the small neutron effective charge in the neighboring
nuclei 15 B and 17 B [57].
12/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
The theoretical predictions of the AMD (asymmetric molecular dynamics) framework [64], a
microscopic shell model [65], or a shell model assuming 14 C to be inert [63] reproduce the B(E2)
hindrance and support the picture of neutron-dominant excitation in 16 C.
4.2.
Disappearance and appearance of magic numbers
The disappearance of the N = 20 magic number was pointed out in the 1970s, based on the mass
anomalies in neutron-rich sodium isotopes [66] and the low excitation energy of 885 keV for the first
2+ state in 32 Mg [67]. The result of the first Coulomb excitation experiment at RIKEN described in
Sect. 3.1.2 is consistent with the picture of the disappearance. The phenomenon is interpreted by a
considerable amount of intruder configurations in which a few (mostly two) neutrons are in higher pf
orbitals, and the region of nuclei with this behavior is called the “island of inversion” [68]. Since then,
many neutron-rich nuclei around 32 Mg have been studied. The appearance of a new magic number
at N = 16 [69] may be associated with this disappearance. Similar phenomena, disappearance or
weakening of shell closures, were found for the N = 8 and N = 28 regions, and will be studied for
larger magic numbers; some of them are related to r-process nucleosynthesis as well.
4.2.1. Nuclei around the island of inversion. Following the 32 Mg Coulomb excitation experiment,
further studies were performed at RIPS and BigRIPS.
The most proton-deficient even–even nucleus with N = 20, 30 Ne, was studied by a (p,p’) experiment employing a liquid hydrogen target [24]. The location of the first 2+ state, 791 ± 26 keV [71]4 ,
indicates that the 30 Ne nucleus is also in the island of inversion.
For the more neutron-rich nucleus 34 Mg, the measured Coulomb excitation cross section leads to
a larger B(E2 : 0+ → 2+ ) value of 600 e2 fm4 , corresponding to the quadrupole deformation β2 ≈
0.6 [72]. This supports the picture of well developed deformation suggested by the ratio (R4/2 ) 3.2,
close to the rigid-rotor limit of 3.3, between the energies of the 2+ (660 keV) and 4+ states (2120 keV)
observed in the γ -ray spectroscopy study with the two-step fragmentation scheme [73]. This is in
contrast to the result of the recent (p,p’) study on 32 Mg. The R4/2 value of 2.6 is in between the
vibrational limit (2.0) and the rotational limit (3.3). This indicates that 32 Mg is not of rigid axial
deformation.
The B(E2 : 0+ → 2+ ) value for 28 Ne, a neighboring even–even nucleus of 30 Ne, was obtained by
a Coulomb excitation study [74]. As shown in Fig. 11, the result deviates from the systematic trend
between the excitation energy (E(2+ )) and the transition probability (B(E2)): E(2+ ) is lower than the
shell model prediction within the sd-shell space, suggesting contributions of intruder configurations,
whereas B(E2) agrees with the prediction, pointing to a normal configuration. Similar behavior was
seen in an experiment at Michigan State University for 34 Si [76], in which the N = 20 nucleus with
two more protons was compared with 32 Mg. This suggests that the nuclei 28 Ne and 34 Si are on the
boundary of the island of inversion. In contrast, the behavior of both 26 Ne and 36 S agrees with the
predictions, indicating that they are “normal” nuclei without intruder configurations. On the other
hand, 32 Mg and 30 Ne are inside the island of inversion. B(E2) is larger and E(2+ ) is smaller than
the prediction but their product is close to unity, which follow the general trend.
Attempts to explore the island on its more neutron-rich side have been started at the new RIBF
facility. The first result is on 32 Ne [70]. The low 2+ energy of 0.722 MeV indicates that this nucleus
is also in the island.
4
An energy value with higher precision, 801 ± 7 keV, was obtained in a recent BigRIPS experiment [70].
13/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 11. The ratios of E(2+ ) (a) and B(E2) (b) to the shell model results E(2+ )sd and B(E2)sd , respectively,
calculated within the sd-shell model space [75,76]. The red circles represent the data taken at RIKEN.
31Si
32Si
33Si
34Si
35Si
36Si
30Al
31Al
32Al
33Al
34Al
35Al
29Mg 30Mg 31Mg 32Mg 33Mg
34Mg
28Na
29Na 30Na
31Na 32Na
33Na
27Ne 28Ne 29Ne
30Ne 31Ne
32Ne
26F
27F
28F
29F
30F
31F
Fig. 12. Nuclear chart around the island of inversion. The nuclei known to be in the island based on experimental data are shown in blue. The ones on the boundary are in light blue. The other nuclei, shown in gray, are
not experimentally confirmed in view of the island of inversion.
Such mapping of the island of inversion becomes more complete by data for odd or odd–odd
nuclei. This is possible by measuring the magnetic and electric quadrupole moments by the β-NMR
method described in Sect. 3.2.2. A series of experiments, including the one at RIPS [77], have been
performed with a group at RIKEN. The study continues in the new RIBF facility with the new method
mentioned in Sect. 3.2.2.
Figure 12 summarizes the current knowledge of the island of inversion, based on the experimental level energy and transition probability for even–even nuclei, the electric quadrupole and and/or
magnetic moment for odd and odd–odd nuclei, and other information such as the binding energy
and level scheme. Information from the mass anomalies mentioned earlier for sodium isotopes [66]
is also included. The figure exhibits the recent progress in the mapping and necessity of extending
14/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
the study, especially toward nuclei farther from stability, most of which are out of the chart in the
figure.
4.2.2. Appearance of the new magic number N = 16. Based on the systematic trend of the binding energy, the appearance of the new magic number N = 16 in place of N = 20 in some neutron-rich
nuclei has been pointed out [69]. There are several works exhibiting evidence for the N = 16 magicity in 27 Na and more proton-deficient nuclei. The size of the N = 16 shell gap was obtained for 23 O
by observing its resonance states excited by the (d,p) reaction [78]. The invariant mass method was
employed in that study.
Recently, the location of the unbound 2+ state in 24 O was measured at RIPS with a similar setup
to that shown in Fig. 5 [79]. The highest excitation energy of 4.65 ± 0.14 MeV and the smallest
deformation parameter of 0.15 ± 0.04 among the neighboring oxygen isotopes indicate N = 16 shell
closure.
Another study on the shell closure is on 28 S with the proton number Z = 16 instead of the neutron
number [80]. Figure 13 plots the B(E2 : 0+ → 2+ ) (top) and E(2+ ) (middle) values for even–
even N = 12 isotones and Z = 16 isotopes. A special feature was observed in the double ratio
(Mn /Mp )/(N /Z ) (bottom), which exhibited the largest value for 28 S. The large double ratio indicates neutron-dominant excitation or hindrance of the proton contribution. This may be interpreted
as being due to the Z = 16 shell closure, analogously to the N = 16 magic number. The behavior
of B(E2) and E(2+ )—disagreement of the former and agreement of the latter with the shell model
prediction by the USDB [81] effective interaction (dotted lines)—is similar to the case of the analog
nucleus 28 Ne. Therefore, the appearance of the N (Z ) = 16 magic number in 28 Ne(28 S) could be
related to its transitional character, discussed in Sect. 4.2.1.
4.2.3. Disappearance of the N = 8 magic number. A series of experiments with the neutron-rich
nucleus 12 Be with the neutron magic number N = 8 have been performed at RIPS. The first 1− state
was found at 2.7 MeV excitation energy. The spin assignment was made by comparing the γ ray
yields obtained in 12 Be + 208 Pb and 12 Be + 12 C inelastic scattering [82]. An isomeric 0+ state at
2.24 MeV was found in the 12 Be secondary beam [83]. The locations of these states are much lower
than those in neighboring N = 8 nuclei, indicating disappearance of the N = 8 shell closure in 12 Be.
4.2.4. Rise and fall of the shell closure. Evolution of the shell closure attracts a great deal of attention. It should be governed by change in the relative location of orbitals caused by the low binding
energy [69] and/or the spin–isospin dependent part of the residual nucleon–nucleon interaction [84].
Some recent work at the new RIBF facility is on the nucleus 42 Si with Z = 14 and N = 28, both of
which are for the subshell closures [85]. Thanks to the high RI-beam intensities at BigRIPS, γ − γ
coincidence measurements could be made and the 4+ state, as well as the first excited 2+ state, was
identified for the first time. The obtained R4/2 ratio of 3.1 suggests well developed deformation,
despite the expectation of the “doubly subshell-closed” nature. The result finally confirms one of the
earlier works, indicating a low-energy 2+ state [86], in conflict with another work, claiming a doubly
closed shell nature for 42 Si [87,88].
It is also known that disappearance or weakening of shell closures affects the behavior of nuclear
reaction networks and hence the abundance predictions of, e.g., r-process nucleosynthesis. Weakening could hinder the roles of classical waiting-point nuclei in explosive nuclear burning processes.
Along these lines, several spectroscopy experiments are proposed or ongoing for N = 50 and N = 82
neutron-rich nuclei, which play crucial roles in r-process nucleosynthesis.
15/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 13. B(E2 : 0+ → 2+ ) (top) and E(2+ ) (middle) for even–even N = 12 isotones and Z = 16 isotopes.
The ratio of neutron- and proton-transition matrix elements divided by the N /Z ratio (bottom) exhibits the
largest value for 28 S. The filled circles and open triangles show, respectively, the present results and known
data. The open squares represent the double ratios obtained by the combinations of B(E2) and the result of the
(p,p’) reaction.
5. Nuclear processes of astrophysical interest
5.1. Unstable nuclei in astrophysics
Certain short-lived nuclei play important roles in explosive nuclear burning. A nuclear reaction on
the relevant nucleus may become faster than its β decay, depending on the temperature and density
of the astronomical environment. Non-explosive situations, such as hydrogen burning in the sun,
sometimes involve unstable nuclei. These reactions are, however, difficult to study in laboratory
experiments, because the short-lived nucleus is hard to use as a target. The use of RI beams is a
solution. Direct measurements can be performed by a low-energy RI beam of the nucleus of interest.
Alternatively, indirect methods, such as the ANC method [89,90] with transfer reactions, the “Trojan
Horse” method [91,92], and Coulomb dissociation, are sometimes employed. The first and third
methods are used for radiative capture reactions, whereas the second one is for transfer reactions.
For applications to experiments with RI beams, measurements should be quite efficient, as discussed earlier. Reactions with large cross sections are preferable. The Coulomb dissociation method
is most suited for fast RI beams.
On the other hand, the explosive burning scenario sometimes satisfies the condition for statistical
equilibrium between the inverse processes, especially in burning at high temperatures and high densities, as in the r-process. In that case, determination of reaction Q values or the nuclear masses in
16/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 14. Schematic picture of the Coulomb dissociation. The lower panel shows the electric force felt by the
projectile as a function of time and its Fourier transform.
the initial and final channels is crucial, instead of individual reaction cross sections, as a first approximation. The lifetime and delayed-neutron emission probability are also important quantities to be
experimentally determined.
5.2.
Coulomb dissociation
For a fast-beam induced Coulomb dissociation experiment, the residual nucleus A of the reaction
B(x, γ )A bombards a high-Z target as Pb, and is Coulomb-excited to its unbound state, which decays
to the B+x channel. The cross sections of the Coulomb dissociation and the relevant capture reactions
are related, since the dissociation process is regarded as absorption of a virtual photon, A(γ , x)B,
the inverse of the radiative capture of interest. Figure 14 shows an intuitive picture for the Coulomb
dissociation process. When a projectile nucleus (8 B in this case) passes the vicinity of a high-Z
nucleus, the projectile feels a rapid change in the electric field. This field change (shown in the bottom
left of Fig. 14) can be decomposed by electromagnetic waves with different frequencies, and hence
is equivalent to exposure to photons with different energies. The photon spectrum is obtained by the
Fourier transform of the impulse. With the fast beams at RIPS or BigRIPS, the maximum energy,
which is related to the projectile velocity, can cover the range of excitation energies of particle A in
most of the cases of astrophysical interest.
The idea of applying this Coulomb dissociation method to astrophysical radiative capture was first
proposed by Baur, Bertulani, and Rebel [93], and some review articles are available [22,94,95].
In more detail, the cross sections of the radiative capture and the photo-absorption are connected
by the detailed balance relation
σA+γ →B+x =
(2 jB + 1)(2 jx + 1) k 2
σB+x→A+γ .
2(2 jA + 1)
kγ2
(1)
The spins of A, B, and x are denoted by j, respectively. The symbols k and kγ denote the wave
numbers in the B + x and A + γ channels, respectively. Due to the large phase space volume for the
B+x channel, the photo-absorption cross section is enhanced. For example, the wave-number ratio
k 2 /kγ2 in Eq. 1 is calculated to be about 1000 for the case of the 7 Be(p, γ )8 B reaction.
17/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
The energy-dependent Coulomb dissociation cross section dσCD /d E γ and that of photoabsorption can be related by the equivalent photon number n γ as
dσCD
1
=
n γ σphoto .
d Eγ
Eγ
(2)
n γ can be very large, and the Coulomb dissociation cross section is greatly enhanced. For example, the n γ value is evaluated to be about 1000 for the 8 B Coulomb dissociation at around
50 MeV/nucleon.
In Coulomb dissociation experiments, the invariant mass of the B+x system is measured, as discussed in Sect. 3.1.3. The relative-energy resolution in the case of the setup shown in Fig. 4 is about
200 keV at a relative energy of 1 MeV with an angular resolution of 0.6◦ and an energy resolution of
2%. These values are reasonable for experiments with light RI beams.
5.3.
Studies with the Coulomb dissociation method
In addition to the examples quoted earlier (14 O and 8 B dissociation), more Coulomb dissociation
studies have been performed. An example is on the 22 Mg(p, γ )23 Al reaction [96], which might play
a role in the rp-process relevant in thermonuclear runaway, expected to occur in certain novae. From
the experimental relative-energy spectrum, the radiative width of the first resonance state in 23 Al
was obtained, and provided the first experimental support for the recent evaluation of the reaction
rate [97]. It should be noted that in direct capture measurement data with equivalent statistics can be
obtained only by the 22 Mg beam intensity of the order of 1012 s−1 , which is impossible with current
technology. Another rp-process reaction, 26 Si(p, γ )27 P, was recently studied by a 54 MeV/nucleon
beam of 27 P [98].
Detector developments are ongoing for a near-future experiment of astrophysical Coulomb
dissociation with SAMURAI at the new RIBF facility.
5.4.
Properties of nuclei in the pathway of explosive nucleosynthesis
As discussed at the beginning of the present section, to measure the properties of nuclei in explosive processes as the r- and rp-processes is of crucial importance. The first lifetime measurement
at BigRIPS in this context determined the lifetime for 38 isotopes near or in the expected r-process
path. Among them, the data for 18 nuclei are new. The lifetime of some isotopes disagree with the
theoretical ones that are used in current network calculations for nucleosynthesis [38,40].
As discussed in Sect. 5.1, the nuclear mass is the crucial property, because it determines the reaction
Q value, and hence controls the nucleosynthesis. The SLOWRI project with the MR-TOF device
and construction of the rare-RI ring (see Fig. 1) are in progress. The former is for precision mass
measurements, whereas the latter is designed to measure with reasonable resolution the masses of
rarely produced isotopes, such as those in the r-process path.
6.
RIBF at present and in the future
The capability of RI beam production at present and in the future is indicated in Fig. 15. The isotopes
found at RIKEN so far are indicated by red squares, including the 47 new ones created at RIBF.
When the intensity reaches the goal of 1 pµA for all kinds of primary beams, nuclides in the regions
indicated in purple-red and light blue will be produced at rates higher than one particle per day. As
seen in the figure, most of the expected path of r-process nucleosynthesis is covered. In the near
future, the intensities of most of the primary beams will reach the goal, whereas that of uranium
18/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Fig. 15. Nuclear chart potentially covered by RIBF. The nuclei shown in purple-red (by the in-flight fission and
projectile fragmentation of uranium beams) and light blue (by the fragmentation of other beams) indicate those
with a production rate higher than 1 particle per day, expected with a primary-beam intensity of 1 particle µA.
will be about 100 pnA. The RIKEN RI Beam Factory is the only “new-generation” facility that is
working, before completion of the facilities planned or under construction in Asia (HIAF at IMP
China, the IBS facility in Korea), the United States (FRIB), and Europe (GSI FAIR, Spiral2).
The new RIBF facility has started operation and several experiments there have already produced
several highly significant results, as briefly presented in this article. Its RI beam production capability
is already superior to those in existing facilities for many neutron-rich isotopes. To exploit fully the
research opportunity, construction of several experimental installations has been planned.
In addition to SAMURAI, the spectrometer SHARAQ [99], seen in Fig. 1, is in operation. For
high-resolution missing-mass measurements, a dispersion-matched RI beam line is coupled. The
first experimental result, isovector spin-monopole excitation studied by the charge-exchange (t, 3 He)
reaction, has been obtained [100].
The idea of SCRIT for scattering experiments with storage electrons and RI ions trapped by electron beams has been experimentally confirmed [101]. Construction of the electron–RI scattering
equipment with SCRIT is ongoing [102] (see Fig. 1). The major purpose is the most straightforward
determination of the charge distribution of exotic nuclei.
Construction has begun on an isochronous ring based on a novel idea of trapping rarely produced
fast RIs [103] (see Fig. 1). The primary purpose of this “rare RI ring” is to directly measure the masses
of very neutron-rich nuclei, for which no other method can be applied, due to the low production yield.
The mass measurement is expected to reach nuclei in the r-process path. It should be noted that the
mass or the reaction Q value is one of the most crucial quantities that govern the reaction network
for explosive nucleosynthesis.
The SLOWRI project utilizes slow RI ions produced by stopping fast ions in gas, as mentioned
in Sect. 3.2.3. Another project for slow RI ions is planned by a group from KEK. Using the laserionization technique, heavy neutron-rich nuclei, created by the multi-nucleon transfer reaction and
stopped in gas, are to be extracted. This equipment, called KISS (KEK isotope separation system), is
19/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
designed to study the decay properties of neutron-rich nuclei around the N = 126 classical waitingpoint in the r-process [104]. Research and development (R&D) studies are in progress.
Another R&D project is the construction of a beam line delivering ions from IRC to RIPS for
nuclear moment studies and materials science (see Fig. 1).
In the coming years, the conduct of various experiments with the world’s strongest RI beams is a
major focus of RIBF. Improvement of the accelerator performance and construction of the new equipment mentioned above might enhance the research activity. Future possibilities of extending RI-beam
production by modification and construction of accelerators and equipment are under consideration.
7.
Summary
At the RIKEN RI Beam Factory, fast RI beams have been used for nuclear structure and nuclear
astrophysics studies for more than twenty years. The new RIBF facility has the potential to drastically
enlarge the region of isotopes that can be approached experimentally. One of the motivations for
extending the number of isotopes under control might be to put the nuclear system into large proton–
neutron asymmetry, expecting the discovery of new characteristic properties in nuclear interaction
that are difficult to see in stable nuclei.
The difference in the roles played by protons and neutrons in the ground-state structure and in
the excitation process is one of the subjects discussed in this article. The most striking phenomenon
in this line is the neutron halo. Spatially extended neutron distributions were found in several light
neutron-rich nuclei, such as 11 Be, 19 C, and 22 C, as well as in 11 Li, the first nucleus in which the neutron halo structure was found. By the accumulation of experimental results over the last twenty years,
it has become well known that weakly bound neutrons with low angular momenta, typically in the sstate, are responsible for the halo phenomenon, and are not strongly coupled with the remaining part
of the nucleus. Moreover, two-neutron halo nuclei like 11 Li provide a testing ground to understand
the two-neutron correlation in low-density situations. The Coulomb dissociation study on 11 Li [52]
is a good example. Searching for halo structures in heavier nuclei is one of the major subjects at
the new RIBF facility, and a candidate has already been found in 31 Ne using the interaction cross
section [18] and inclusive Coulomb dissociation [56].
Another interesting nuclear property is the behavior of the shell structure or the appearance and
disappearance of magic numbers. The disappearance of the N = 8 and N = 20 magic numbers
was established by various experiments with fast RI beams. The existence of the “island of inversion” around 32 Mg was confirmed by measurements of nuclear moments with spin-polarized RI
beams [77] and of cross sections for inelastic scattering, including Coulomb excitation performed
in inverse kinematics [5]. Through continuous efforts with various ideas, mapping of the island of
inversion is considerably advanced. The appearance of the new magic number N = 16 is now established for several neutron-rich nuclei. Such a “rise and fall” of the shell closure is an indication of
nuclear-structure change according to the characteristic properties of nuclear interaction that appear
in unstable nuclei. In the shell model theory, the roles of the tensor interaction become more visible
in nuclei with a proton–neutron imbalance [84]. A search for similar phenomena in different regions
of magic numbers is of interest. The behavior of the N = 28 magic number was studied with nuclei
around 42 Si at the new RIBF facility [85]. The study will be extended to heavier nuclei at RIBF, such
as N = 50, 82, 126, where possible weakening of the shell closure is expected to affect explosive
nucleosynthesis, including the r-process.
Mapping the deformation in the extended nuclear chart should be useful for understanding the
atomic nucleus, and should also be one of the major programs at the new RIBF facility.
20/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
Another opportunity created by the intense RI beams is to access astrophysical nuclear processes
involving short-lived nuclei. Experimental studies on reactions involving short-lived nuclei are made
possible by RI beams. The 13 N(p, γ )14 O reaction in the hot-CNO cycle was the first successful example of experiments with RI beams. The two experiments on direct capture with a low-energy beam [4]
and Coulomb dissociation with a fast beam [5] agreed with each other and thus demonstrated the
usefulness of the RI beam. Another well studied reaction is the radiative proton capture of 7 Be, a reaction crucial for solar neutrino production. As well as the two methods mentioned above, an indirect
method to measure the asymptotic normalization coefficient (ANC) was also employed, and experimental results for the astrophysical S-factor mostly converge [105]. Determination of cross sections
of astrophysical interest has been made for many reactions in explosive nucleosynthesis, such as
the hot-CNO cycle and the rp-process. However, there remain a lot of crucial reactions untouched
due to various difficulties, including low cross sections. Some of these are the targets of research at
new-generation RI beam facilities, including RIBF.
The r-process is believed to be responsible for synthesizing about half of the elements heavier than
iron. However, extensive experimental studies on it have not been made, because most of the nuclei
involved are very neutron-rich and are hardly accessible. The capability of RIBF may reach most
of the r-process nuclei, and systematic studies on them may become possible. The first attempt was
successfully made by the β-decay lifetime measurement [38].
Various important results obtained in RI-beam-based experiments at RIKEN and other institutes
have triggered and will trigger worldwide challenges in nuclear physics and nuclear astrophysics
research. With its newly developed regions of nuclei, the RIKEN RIBF plays and will play pivotal
roles in this research domain.
Acknowledgment
We are indebted to all the people who have developed research with fast RI beams in the two decades at RIKEN.
Special thanks are due to S. Takeuchi, M. Takechi, H. Ueno, M. Wada, and T. Nakamura, for providing us with
detailed information on their work together with drawings.
References
[1] I. Tanihata et al., Phys. Lett. B 160, 380 (1985).
[2] J. R. Alonso, B. Feinberg, J. G. Kalnins, G. F. Krebs, M. A. McMahan, and I. Tanihata, in The First
International Conference on Radioactive Nuclear Beams, eds. W. D. Myers, J. M. Nitschke, and E. B.
Norman (World Scientific, Berkeley, CA, 1990), p. 112.
[3] I. Tanihata, H. Hamagaki, O. Hashimoto, Y. Shida, N. Yoshikawa, K. Sugimoto, O. Yamakawa,
T. Kobayashi, and N. Takahashi, Phys. Rev. Lett. 55, 2676 (1985).
[4] P. Decrock et al., Phys. Rev. Lett. 67, 808 (1991).
[5] T. Motobayashi et al., Phys. Lett. B 264, 259 (1991).
[6] Y. Yano, Nucl. Instrum. Meth. B 261, 1009 (2007).
[7] H. Okuno, N. Fukunishi, and O. Kamigaito, Prog. Theor. Exp. Phys. 2012, 03C002 (2012).
[8] T. Kubo, M. Ishihara, N. Inabe, H. Kumagai, I. Tanihata, K. Yoshida, T. Nakamura, H. Okuno,
S. Shimoura, and K. Asahi, Nucl. Instrum. Methods Phys. Res., Sect. B 70, 309 (1992).
[9] K. Sekiguchi et al., Phys. Rev. C 83, 061001 (2011), and references therein.
[10] H. Okamura et al., Nucl. Instrum. Methods Phys. Res., Sect. A 406, 78 (1998).
[11] H. Okamura et al., Phys. Lett. B 345, 1 (1995).
[12] K. Morita et al., J. Phys. Soc. Jpn. 73, 2593 (2004).
[13] K. Morita et al., J. Phys. Soc. Jpn. 76, 045001 (2007).
[14] K. Morita et al., J. Phys. Soc. Jpn. 81, 103201 (2012).
[15] Y. Yanagisawa et al., Nucl. Instrum. Methods Phys. Res., Sect. A 539, 74 (2005).
[16] T. Kubo, Nucl. Instrum. Methods Phys. Res., Sect. B 204, 97 (2003).
21/23
PTEP 2012, 03C001
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
T. Motobayashi and H. Sakurai
T. Ohnishi et al., J. Phys. Soc. Jpn. 77, 083201 (2008).
M. Takechi et al., Phys. Lett. B 707, 357 (2012).
T. Motobayashi et al., Phys. Lett. B 346, 9 (1995).
S. Takeuchi, T. Motobayashi, H. Murakami, K. Demichi, and H. Hasegawa, RIKEN Accel. Prog. Rep.
36, 148 (2003).
P. Doornenbal, Prog. Theor. Exp. Phys. 2012, 03C004 (2012).
T. Motobayashi, Nucl. Phys. A 693, 258 (2001).
Y. Kondo et al., Phys. Lett. B 690, 245 (2010).
H. Ryuto, M. Kunibu, T. Minemura, T. Motobayashi, K. Sagara, S. Shimoura, M. Tamaki,
Y. Yanagisawa, and Y. Yano, Nucl. Instrum. Methods Phys. Res., Sect. A 555, 1 (2005).
T. Kobayashi, in preparation for publication in Prog. Theor. Exp. Phys., 2012.
S. Michimasa et al., Phys. Lett. B 638, 146 (2006).
S. Ota et al., Phys. Lett. B 666, 311 (2008).
T. Morikawa et al., Phys. Lett. B 350, 169 (1995).
M. Niikura et al., Eur. Phys. J. A 42, 471 (2009).
E. Ideguchi, Prog. Theor. Exp. Phys. 2012, 03C005 (2012).
T. Ohnishi et al., J. Phys. Soc. Jpn. 79, 073201 (2010).
M. Notani et al., Phys. Lett. B 542, 49 (2002).
T. Kubo et al., Prog. Theor. Exp. Phys. 2012, 03C003 (2012).
K. Asahi et al., Phys. Lett. B 251, 488 (1990).
R. Chevrier et al., Phys. Rev. Lett. 108, 162501 (2012).
Y. Ichikawa et al., Nat. Phys., 8, 918–922 (2012).
N. Aoi et al., Phys. Rev. C 66, 014301 (2002).
S. Nishimura et al., Phys. Rev. Lett. 106, 052502 (2011).
T. Sumikama et al., Phys. Rev. Lett. 106, 202501 (2011).
S. Nishimura, Prog. Theor. Exp. Phys. 2012, 03C006 (2012).
M. Wada, Nucl. Instrum. Methods Phys. Res., Sect. A 532, 40 (2004).
K. Okada et al., Phys. Rev. Lett. 101, 212502 (2008).
H. Wollnik and M. Przewloka, Int. J. Mass Spectrom. 96, 267 (1990).
Y. Ishida, M. Wada, and H. Wollnik, Nucl. Instrum. Methods Phys. Res., Sect. B 241, 983 (2005).
P. Schury, K. Okada, S. Shchepunov, T. Sonoda, A. Takamine, M. Wada, H. Wollnik, and Y. Yamazaki,
Eur. Phys. J. A 42, 343 (2009).
E. W. Otten, in Treatise on Heavy-Ion Science, ed. D. A. Bromley (Plenum, New York, 1989), Vol. 8,
p. 517.
H.-J. Kluge and W. Nörtershauser, Spectrochim. Acta B 58, 1031 (2003).
A. Bohr and V. F. Weisskopf, Phys. Rev. 77, 94 (1950).
J. S. Grossman, L. A. Orozco, M. R. Pearson, J. E. Simsarian, G. D. Sprouse, and W. Z. Zhao, Phys.
Rev. Lett. 83, 935 (1999).
T. Suzuki et al., Phys. Rev. Lett. 75, 3241 (1995).
See, e.g., I. Tanihata, J. Phys. G: Nucl. Part. Phys. 22, 157 (1996).
T. Nakamura et al., Phys. Rev. Lett. 96, 252502 (2006).
N. Fukuda et al., Phys. Rev. C 70, 054606 (2004).
T. Nakamura et al., Phys. Rev. Lett. 83, 1112 (1999).
K. Tanaka et al., Phys. Rev. Lett. 104, 062701 (2010).
T. Nakamura et al., Phys. Rev. Lett. 103, 262501 (2009).
H. Ogawa et al., Phys. Rev. C 67, 064308 (2003).
A. Bohr and B. Mottelson, Nuclear Structure, Vol. II. (W. A. Benjamin, New York, 1975).
H. J. Ong et al., Phys. Rev. C 78, 014308 (2008).
N. Imai et al., Phys. Rev. Lett. 92, 062501 (2004).
Z. Elekes et al., Phys. Lett. B 586, 34 (2004).
H. J. Ong et al., Phys. Rev. C 73, 024610 (2006).
M. Wiedeking et al., Phys. Rev. Lett. 100, 152501 (2008).
Y. Kanada-En’yo and H. Horiuchi, Phys. Rev. C 55, 2860 (1997).
S. Fujii, T. Mizusaki, T. Otsuka, T. Sebe, and A. Arima, Phys. Lett. B 650, 9 (2007).
C. Thibault, R. Klapisch, C. Rigaud, A. M. Poskanzer, R. Prieels, L. Lessard, and W. Reisdorf, Phys.
Rev. C 12, 644 (1975).
22/23
PTEP 2012, 03C001
T. Motobayashi and H. Sakurai
[67] C. Détraz, D. Guillemaud, G. Huber, R. Klapisch, M. Langevin, F. Naulin, C. Thibault, L. C. Carraz,
and F. Touchard, Phys. Rev. C 19, 164 (1979).
[68] E. K. Warburton, J. A. Becker, and B. A. Brown, Phys. Rev. C 41, 1147 (1990).
[69] A. Ozawa, T. Kobayashi, T. Suzuki, K. Yoshida, and I. Tanihata, Phys. Rev. Lett. 84, 5493 (2000).
[70] P. Doornenbal et al., Phys. Rev. Lett. 103, 032501 (2009).
[71] Y. Yanagisawa et al., Phys. Lett. B 566, 84 (2003).
[72] H. Iwasaki et al., Phys. Lett. B 522, 227 (2001).
[73] K. Yoneda et al., Phys. Lett. B 499, 233 (2001).
[74] H. Iwasaki et al., Phys. Lett. B 620, 118 (2005).
[75] E. Caurier, F. Nowacki, A. Poves, and J. Retamosa, Phys. Rev. C 58, 2033 (1998).
[76] R. W. Ibbotson et al., Phys. Rev. Lett. 80, 2081 (1998).
[77] H. Ueno et al., Phys. Lett. B 615, 186 (2005).
[78] Z. Elekes et al., Phys. Rev. Lett. 98, 102502 (2007).
[79] K. Tshoo et al., Phys. Rev. Lett. 109, 022501 (2012).
[80] Y. Togano et al., Phys. Rev. Lett. 108, 222501 (2012).
[81] B. A. Brown and W. A. Richter, Phys. Rev. C 74, 034315 (2006).
[82] H. Iwasaki et al., Phys. Lett. B 491, 8 (2000).
[83] S. Shimoura et al., Phys. Lett. B 560, 31 (2003).
[84] T. Otsuka, R. Fujimoto, Y. Utsuno, B. A. Brown, M. Honma, and T. Mizusaki, Phys. Rev. Lett.
87, 082502 (2001).
[85] S. Takeuchi et al., Phys. Rev. Lett. 109, 182501 (2012).
[86] B. D Bastin et al., Phys. Rev. Lett. 99, 022503 (2007).
[87] J. Fridmann et al., Nature 435, 922 (2005).
[88] J. Fridmann et al., Phys. Rev. C 74, 034313 (2006).
[89] A. Azhari, V. Burjan, F. Carstoiu, C. A. Gagliardi, V. Kroha, A. M. Mukhamedzhanov, F. M. Nunes,
X. Tang, L. Trache and R. E. Tribble, Phys. Rev. C 63, 055803 (2001).
[90] H. M. Xu, C. A. Gagliardi, R. E. Tribble, A. M. Mukhamedzhanov and N. K. Timofeyuk, Phys. Rev.
Lett. 73, 2027 (1994).
[91] G. Baur, Phys. Lett. B 178, 135 (1986).
[92] C. Spitaleri, M. Aliotta, S. Cherubini, M. Lattuada, D. Miljani, S. Romano, N. Soi, M. Zadro, and
R. A. Zappalà, Phys. Rev. C 60, 055802 (1999).
[93] G. Baur, C. A. Bertulani, and H. Rebel, Nucl. Phys. A 458, 188 (1986).
[94] G. Baur and H. Rebel, J. Phys. G. 20, 1 (1994).
[95] G. Baur and H. Rebel, Annu. Rev. Nucl. Part. Sci. 46, 321 (1996).
[96] T. Gomi et al., Nucl. Phys. A 734, E77 (2004).
[97] J. A. Caggiano et al., Phys. Rev. C 64, 025802 (2001).
[98] Y. Togano et al., Phys. Rev. C 84, 035808 (2011).
[99] T. Uesaka, S. Shimoura, H. Sakai, G. P. A. Berg, K. Nakanishi, Y. Sasamoto, A. Saito, S. Michimasa,
T. Kawabata, and T. Kubo, Nucl. Instrum. Methods Phys. Res., Sect. B 266, 4218 (2008).
[100] K. Miki et al., Phys. Rev. Lett. 108, 262503 (2012).
[101] M. Wakasugi et al., Phys. Rev. Lett. 100, 164801 (2008).
[102] T. Suda et al., Prog. Theor. Exp. Phys. 2012, 03C008 (2012).
[103] Y. Yamaguchi et al., Nucl. Instrum. Methods Phys. Res., Sect. B 266, 4575 (2008).
[104] S. C. Jeong, N. Imai, H. Ishiyama, Y. Hirayama, H. Miyatake, and Y. X. Watanabe, KISS: KEK
isotope separation system for β-decay spectroscopy - Toward the funn understanding of r-process
nucleosynthesis, KEK Report 2010-2, (KEK, Tsukuba, Japan, 2010).
[105] E. G. Adelberger et al., Rev. Mod. Phys. 83, 195 (2011).
23/23