Tensor Force and Nuclear Mean Field

a b c
Takaharu Otsuka , and Daisuke Abea
a
Department of Physics, University of Tokyo,
Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan
b
Center for Nuclear Study, University of Tokyo,
Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan
c
RIKEN,Hirosawa, Wako-shi, Saitama 351-0198, Japan

The effects of the tensor force, the most prominent spin-isospin interaction, will be
discussed from the viewpoints of the spherical single-particle energies. This point will
be started with basic properties of the tensor force, and a new mean-field model will be
introduced. It will be shown that single-particle properties, including (sub-)magic gaps
and spin-orbit splittings, are varied rather strongly by the tensor force. The effects are
not weaker than the effects of the diffuse surface due to neutron skin. An example is
presented for the so-called island of inversion, for which the boundary is being modified
significantly.

1. Introduction
The spin-isospin interaction has a special importance for the COMEX conferences as
well as their preceding conferences, largely because this interaction is closely connected
to spin-isospin excitations, including the Gamov-Teller (GT) resonances. Recently, it has
been realized that the spin-isospin interaction, especially the tensor interaction, is also
of particular and direct importance to single-particle properties, or mean-field properties,
of exotic nuclei. Like other many-body quantal systems, e.g., electrons in hydrogen-like
atoms, the mean field governs the structure of nuclei [1]. The deformation, spin-isospin
excitation, and even existence of nuclei depend on single-particle properties. Recently
it has been shown that the tensor force has a specific, robust and systematic effect on
the single-particle energies of nuclei, even breaking or creating magic numbers in some
cases [2]. We shall overview the present situation of this subject, and also shall explore
perspectives.

2. Tensor force and pions

The nuclear binding is basically due to the exchange of mesons. Even with QCD, this
picture remains unchanged. While the QCD may give us more natural explanations of
short-distance behaviours, for a longer distance between nucleons, the meson exchange
picture is still valid, and its basis may be given by Chiral Perturbation idea of QCD [4].

1
2 T. Otsuka, and D. Abe

Yukawa predicted 70 years ago that the binding of atomic nuclei is due to exchange of
mesons [3]. We now know that the nuclear binding comes indeed mostly from exchange
of more than one meson. This situation can be seen in the importance of the σ meson,
which is nothing but a pair of pions coupled to spin 0. Besides the effects of multiple
meson exchanges, there should be the effect of single pion exchange. It is well-known
that the dominant effect of the single-pion exchange is nothing but the tensor force. We
have presented [2] that a systematic and significant effect is produced from the one-pion
exchange on single-particle energies with many examples over the nuclear chart. The
effect can be explaind in a very simple and intuitive way.

100
GT2
50 π
π+ρ
AV8’
VTE [MeV]

-50

-100

-150
0 1 2 3
r [fm]

Figure 1. Triplet-Even potential of the tensor force for various potential models.

We here explain that the same tensor force can be expressed in two ways, in order to
answer to some questions. The tensor force is expressed usually as,

VT = (~τ1 · ~τ2 ) S12 V (r) (1)

where ~τ1,2 denotes the isospin operators of nucleons 1 and 2, ( · ) means a scalar product,
and V is a function of the distance, r, between two nucleons. The function S12 is defined
as

S12 = 3 (s~1 · ~r/r) (s~2 · ~r/r) − (s~1 · s~2 ), (2)

Tensor Force and Nuclear Mean Field 3

where ~s1,2 is the spin of nucleons 1 and 2, and ~r denotes relative coordinate from nucleon
1 to 2 with r being its absolute value. This equation is rewritten as
S12 = 3 ( [s~1 × s~2 ](2) · [~r × ~r](2) /r 2 ), (3)
q
= 24π/5 ( [s~1 × s~2 ](2) · Y (2) ) (4)

where [ × ](K) means the coupling of two operators in the brackets to an angular momen-
tum (or rank) K, and the spherical harmonics is introduced:
q
[~r × ~r](2) /r 2 = 8π/15 Y (2) . (5)
Thus, the tensor force can be expressed in two different ways. Equation (2) may be more
familiar, while eq. (4) is equivalent to it and is used in [2].
Figure 1 shows the so-called Triplet-Even Potential defined from V (r) as
VT E (r) = − 3 V (r). (6)
In Fig. 1, the tensor part of the one-pion exchange potential (π), the one-pion + one-ρ-
meson exchange potential (π + ρ), AV8’ potential [5], and GT2 potential is shown. The
tensor part of the AV8’ potential is similar to taht of the M3Y potenial [6]. The GT2
potential will be discussed later. One notices similarities among the potentials except
for the π potential which is shown only for the sake of comparison. Differences of these
potentials at short distances do not matter in the present study, because the relative
motion between two nucleons cannot be in the S state in both bra and ket states (See [2]
for details).
We have investigated the tensor monopole effects using the π + ρ potential in the shell-
model based studies [2]. Since the parameters of this potential has been well established
by Osterfeld [7], we took the values in [7] without further adjustment. There could be
minor higher-order-medium and/or core-polization effects, but they are up to about 10
%. At moment, we focus on global studies, leaving such fine details for later works.
Since the tensor monopole effects have been discussed extensively in [2], it is not ap-
propriate to repeat the arguments here. We instead mention that more examples of the
monopole tensor effects have been seen recently. One of them is for the monopole inter-
action between proton 1d3/2 and neutron 1f7/2,5/2 orbits [8]. In fact, the change of the
spin-orbit splitting has been measured for the first time in this experiment.
The tensor force should be responsible for the change of the magic number around
N=20. Figure 2 shows effective single-particle energies (ESPE’s) and their changes. The
original figure was presented in [9], where the change of the shell structure had been dis-
cussed for the first time as a systematic effect of the effective nucleon-nucleon interaction,
particularly, its spin-isospin part. It was pointed out in [9] that the monopole interaction
between spin-flip proton-neutron partners is crucial. Up to this point, the argument is
still correct. In the same paper, we vaguely attributed the origin of this interaction to
the σστ τ central interaction. However, it turns out that the σστ τ central interaction is
not the right origin : For instance, it is not strong enough, and if we make it stronger
artificially the whole scheme breaks down. We have shown in [2] that the major origin
of the shell evolution (change of relative energies of single-particle states as a function
of N or Z) is the tensor force and we can explain the shell evolution without destroying
4 T. Otsuka, and D. Abe

Figure 2. Schematic picture of magicity change due to the spin-isospin interaction. Neu-
tron ESPE’s for (a) 30 Si and (b) 24 O, relative to 1s1/2 . The dotted line connecting (a) and
(b) is drawn to indicate the change of the 0d3/2 level. (c) The major interaction producing
the basic change between (a) and (b). (d) The process relevant to the interaction in (c)
is dominantly due to the tensor force.

existing descriptions of nuclei. Thus, Fig. 2 differs in the part (d) from the corresponding
one in [9].
The N=20 shell gap can be small for nuclei with Z ∼ 10 as we shall show later.
This trend has been confirmed by many experiments published recently [10–14], with
theoretical explanations obtained with the SDPF-M interaction [15,12].
The GXPF1 and GXPF1A interactions should have appropriate tensor components
[16,17].

3. Mean-field model with tensor force

The existing Skyrme [18,19] and Gogny [20–22] calculations do not include the tensor
force. We have developed a mean-field model with the tensor force. Details are given in
[23].
The tensor force is expressed as [2],

VT = (~τ1 · ~τ2 ) ( [s~1 s~2 ](2) · Y (2) )fG (r), (7)

Here, fG (r) is a function of the relative distance, r. Similarly to the central part of the
interaction, we take a Gaussian for fG (r) too, for simplicity, with the range 1.2 fm, i.e.,
the longer range of the central part. The overall strength of fG (r) is then determined
Tensor Force and Nuclear Mean Field 5

A possible mechanism of ls splitting change

exotic nucleus
stable nucleus wi th n e ut r on sk i n

proton
n eut ron

dρ/dr

l s splitting smaller
(Scale-type l s q uen c h i n g)

Figure 3. Conventional picture of smaller spin-orbit splitting in neutron-rich nuclei.

so that its volume integral reproduces that of the AV8’ [5], resulting in fG (r=0)=3156
(MeV). The Triplet-Even potentials in Fig. 1 includes this potential.
One of the major motivations of this work is to compare the effect of neutron skin [25]
to the effect of the tensor force. Figure 3 shows the conventional and rather naive picture
of the change of the spin-orbit splitting due to more diffuse surface in neutron-rich nuclei.
This effect can be a good subject of future experiments as indicated in RIA Physics White
Paper [26]. Although the reduction of the splitting may occur very near driplines [27], the
reality of the mechanism shown in Fig. 3 should be examined. This mechanism is reffered
to as the scale-type ls quenching for the sake of clarification.
In [23], we suggested that another type of the quenching of the spin-orbit splitting is
the right picture for most (or all) of exotic nuclei accesible at present. Their effects are
smaller than the tensor force effect, and in fact the shell evolution cannot be studied
without the tensor force. The sysmtematic change of proton single-particle levels of Sb
isotopes as shown by Schiffer et al. [28] can be explained with the tensor force, but have
not been explained in many attmepts by existing mean-field models. Figure 4 presents
such a comparison.

4. Summary
In summary, we gave an overview on the evolutions of the nuclear shells. The tensor-
force effects are of the same order of magnitude as the changes of spin-orbit splitting due
6 T. Otsuka, and D. Abe

5
GT2
D1S
SLy4
EXP

Energy [MeV]

0
64 70 82 90 94 104
Neutron Number

Figure 4. Evolution of 1h11/2 - 1g7/2 energy gap. The difference from the value of N=64
is plotted for experimental data [28] and calculated results with GT2, D1S and SLy4
interactions.

to the neutron skin, and thereby both have to be considered simultaneously. Even if the
neutron skin effect is absent or very small, notable tensor effects on SPE’s are suggested for
examples such as exotic N=20 isotones with Z ∼10, exotic Ni isotopes with N =40∼50,
and a long isotopic chain of Sb. Note that many of interesting physics cases are found
well inside the dripline, but quite far away from the β-stability line. The predictions seem
to provide us with a paradigm to be tested and developed by future RI-beam physics. We
stress that changes in SPE’s affect essentially all features of the nuclear structure, e.g.,
deformation, and also that many of the nuclei involved participate to the astrophysical
r-process most likely. We point out that a large fraction of the tensor force comes from
one π exchange, and the present GT2 model includes this mechanism. So, the question
is “Are pions so visible in exotic nuclei ?”.
The ls quenching has two types. The surface type has been expected [26] but does
not appear in nuclei studied in this work. The situation could be different if the nucleus
sits on or very near the dripline [27]. The position type, which has not been reported in
publications, can be found in many cases.
This work was supported in part by a Grant-in-Aid for Specially Promoted Research
(13002001) from the MEXT and by the JSPS Core-to-Core Program. This work has
been a part of the RIKEN-CNS joint research project on large-scale nuclear-structure
calculations.

REFERENCES
1. P. Ring and P. Schuck, The Nuclear Many-Body Problem, (Springer, Berlin, 1980).
2. T. Otsuka, T. Suzuki, R. Fujimoto, H. Grawe and Y. Akaishi, Phys. Rev. Lett. 95,
232502 (2005).
3. H. Yukawa, Proc. Phys. Math. Soc. Japan 17 48 (1935).
Tensor Force and Nuclear Mean Field 7

4. S. Weinberg, Phys. Lett. B 251, 288 (1990).

5. B.S. Pudliner et al., Phys. Rev. C56, 1720 (1997).
6. G. Bertsch, J. Borysowicz, H. McManus and W.G. Love, Nucl. Phys. A284, 399
(1977).
7. F. Osterfeld, Rev. Mod. Phys. 64, 491(1992).
8. O. Sorlin et al., Phys. Rev. Lett. 97, 92301 (2006).
9. T. Otsuka et al., Phys. Rev. Lett. 87, 082502 (2001).
10. E. Becheva, et al., Phys. Rev. Lett. 96, 012501 (2006).
11. M. Belleguic et al., Phys. Rev. C72, 054316 (2005).
12. Y. Utsuno et al., Phys. Rev. C70, 044307 (2004).
13. V. Tripathi, et al., Phys. Rev. Lett. 94, 162501 (2005).
14. G. Neyens, et al., Phys. Rev. Lett. 94, 022501 (2005).
15. Y. Utsuno et al., Phys. Rev. C60, 054315 (1999).
16. M. Honma et al., Phys. Rev. C65, 061301 (2002); Phys. Rev. C68, 034316 (2003).
17. M. Honma et al., Euro. Phys. J. A25, Suppl. 1, 499 (2005).
18. T.H.R. Skyrme, Nucl. Phys. 9, 615 (1959).
19. D. Vautherin and D.M. Brink, Phys. Rev. C5, 626 (1972).
20. J. Decharge, M. Girod and D. Gogny, Phys. Lett. 55B, 361 (1975).
21. J. Decharge and D. Gogny, Phys. Rev. C21, 1568 (1980).
22. J.F. Berger, M. Girod and D. Gogny, Nucl. Phys. A428, 23c (1984); Comput. Phys.
Commun. 63, 365 (1991),
23. T. Otsuka, T. Matsuo and D. Abe, Phys. Rev. Lett. accepted.
24. Fl. Stancu, D.M. Brink and H. Flocard, Phys. Lett. 68B, 108 (1977).
25. N. Fukunishi, T. Otsuka and I. Tanihata, Phys. Rev. C48, 1648 (1993).
26. RIA Physics White Paper.
27. J. Dobaczewski et al., Phys. Rev. Lett. 72, 981 (1994).
28. J.P. Schiffer et al., Phys. Rev. Lett. 92, 162501 (2004).