random-interaction study on linear systematics of I = 11/2 electromagnetic moments

in Cd isotope chain
Z. Z. Qin ( )1 and Y. Lei ( )2
1
School of Science, Southwest University of Science and Technology, Mianyang 621010, China
2
Key Laboratory of Neutron Physics, Institute of Nuclear Physics and Chemistry,
China Academy of Engineering Physics, Mianyang 621900, China
(Dated: July 25, 2017)
In the random-interaction ensembles, electromagnetic moments of Cd I = 11/2 isomers pre-
dominately present linear systematics as changing the neutron number, which has been reported in
realistic nuclear system. Quadrupole-like and -like pn interaction are responsible for such linear
systematics of quadrupole and magnetic moments, respectively.
arXiv:1707.07311v1 [nucl-th] 23 Jul 2017

PACS numbers:
Keywords:

I. INTERACTION II. CALCULATION FRAMEWORK

Our random-interaction calculations covers 112130 Cd,

whose single-particle orbits are limited to 0g9/2, 2s1/2 ,
The low-lying spectra and magnetic properties of dou- 1d3/2 and 0h11/2 within Z = 40 50 and N = 64
bly even nuclei are highly regulated with simple patterns. 82 shell closures. No further truncation is introduced.
For examples, they always have I = 0+ ground states, Single-neutron energies are set to be degenerated, al-
and I = 2+ second excited states with few exceptions; most as reality. The two-body interaction is randomized
their quadrupole moments of 2+ +
1 and 22 states gener- within the two-body random ensemble (TBRE) [1416].
ally present a strong correlation with similar magnitudes In other words, any two-body interaction element de-
and different signs across the whole nuclide chart [1]. On noted by VjJ1 j2 j3 j4 follows the Gaussian distribution with
the other hand, these two regularities robustly exist in an ( = 0, 2 = 1 + j1 j2 ,j3 j4 ), where j1 , j2 , j3 and j4 rep-
ensemble of nuclear models with random numbers as two- resent the four single-particle orbits, and the superscript
body interaction matrix elements [2, 3]. These two and J labels the rank.
other robust regularities in the random-interaction en-
semble demonstrates how simple regularity emerges out We generate 3 000 000 sets of two-body interaction el-
of complex nuclear system, even with interactions mostly ements, and input them into the shell-model code [17].
deviating from reality [48]. With each set of two-body interaction elements, we first
calculate corresponding low-lying spectra of even-mass
Cd. If the calculation produces I = 0+ ground states
for all the even-mass Cd isotopes, we further calculate
Recently, it was reported that the I = 11/2 elec- the I = 11/2 electromagnetic moments of odd-mass
tromagnetic moments of neutron rich Cd isotopes are Cd with the same set of elements. To simplify our de-
also simply regulated with an obvious linear systemat- scription, these quadrupole and magnetic moments are
ics as changing neutron number [9]. Several theoret- denoted by Q and , respectively. For Q calculations,
ical investigations tried to understand this linear sys- effective charges are set as e = 1.5e and e = 0.5e;
tematics based on the BCS [10], density functional the- For calculations, single-particle g factors are set as
ory [11], and schematic Shell Model [12]. However, its gs = 5.586 0.7N , gl = 1N , gs = 3.826 0.7N
still the challenge to explain how such simplicity survives and gl = 0, where the spin g factors are conventionally
out of the complex nuclear structure [13]. We believe quenched by 0.7.
that the random-interaction ensemble may provide some
clue. Furthermore, most previous random-interaction To quantitatively describe the Q and systematics, we
works focused on the robust properties of a single dou- introduce the Pearson correlation coefficient (denoted by
bly even nucleus. Its novel to apply random interaction ) [18] as a measure of linear correlation between electro-
to a nuclear-systematics study on an odd-mass isotope magnetic moment and neutron number. This coefficient
chain. Therefore, our work aims to probe and under- has a value between 1, where 1, 0 and -1 correspond to
stand the robustness of such linear systematics in the totally positive linear correlation, no linear correlation,
random-interaction ensemble. and totally negative linear correlation, respectively. For
instance, according to Table I of Ref. [9], the experimen-
tal Q values present = 0.997 linear systematics, and
values present = 0.862 smaller than Q. Thus, the Q
corresponding author: [email protected] linearity is more evident than one as observed.
2

1.0
+
increasing J 0.4 J=10 J=10
J=8 J=8
0 g.s. sampling 0.5 J=8 J=8 J=10 J=10
J=6 J=6 J=6 J=8
J=8
Sampling (I) J=6 J=6 J=6
J=4 J=4 J=6
J=8 J=4
J=6
0.5 Sampling (II) J=4 J=4
J=2 J=8
J=2 J=2 J=4
J=2 J=2 J=2 J=2
0.0 0.0 J=4
J=4 J=2

0.0
Vj j j j >

-0.5
1 2 3 4

-0.4
VdJ3d 3d0 3d 3 (a) (b) (c) (d)
J

VsJ1s 1s01s 1
-0.5
<

2 2 2 2
-1.0
2 2 2 2
parabolic rule -0.8
-1.0 J=0 J=0
J=0 J=0

-1.5

V gJ9 g 90g 9 g 9 V hJ11 h110h11 h11 V gJ9 h11 g 9 h11

-1.5
2 2 2 2 2 2 2 2
V gJ9 g 9 g 9 g 9 force V hJ11 h11 h11 h11 force
2 2 2 2
2 2 2 2 2 2 2 2

FIG. 1: (Color online) hVjJ1 j2 j3 j4 i values after the 0+ -g.s. sam- FIG. 2: Comparison between hVjJ1 j2 j3 j4 i values after the 0+ -
pling, sampling (I) and sampling (II) in an ascending order of g.s. sampling and the -force elements.
j1 j2 j3 j4 J indexes. Sampling (I) corresponds to the sampling
with > 0.9 Q systematics and 0+ ground states for all the
even-mass Cd isotopes. Sampling (II) corresponds to the sam- 0.15 (a)
pling with > 0.9 Q systematics regardless of even-mass Cd

)
0.10
ground states. The grey areas highlight the proton-neutron

P(
interaction between 0g9/2 and 0h11/2 , where the evolution 0.05
with the parabolic rule [19] is schematically illustrated with
a solid curve. The dash line illustrates the ensemble mean of 0.00
back-ground distribution
the TBRE. The error bar is determined by statistic error.
300
(b) Q systematics with 0
+
g.s.
bkg

Q systematics in TBRE
P( )/P

200

100
+
III. INTERACTION PROPERTY WITH I = 0 0
GROUND STATES
-1.0 -0.5 0.0 0.5 1.0

Our Q and calculations are based on the interac-

tions, which can provide I = 0+ ground states (denoted FIG. 3: (Color online) distributions of the Q systematics
after the 0+ -g.s. sampling and in the whole TBRE compared
by 0+ g.s.) for all the even-mass Cd isotopes. In other
with the back-ground distribution (denoted by Pbkg , see text
words, we perform a sampling in the TBRE before the for definition). The error bar is determined by statistic error.
Q and calculations. Only 1% interactions can sur-
vive such sampling, although 0+ -g.s. predominance is
still preserved for each single even-mass Cd isotope in
the TBRE [2]. We present VjJ1 j2 j3 j4 average values (de- IV. Q LINEARITY
noted by hVjJ1 j2 j3 j4 i) after this 0+ -g.s. sampling as the
background of our analysis in Fig. 1. After the 0+ -g.s. sampling, we calculate the distribu-
tion of the Q systematics as presented in Fig. 3(a), where
According to Fig. 1, all the hVjJ1 j2 j3 j4 i values with J = the predominance with P ( > 0.9) 15% emerges, cor-
0 after the 0+ -g.s. sampling are relatively more attractive responding to the almost linear Q evolution in the real-
than others, corresponding to the short-range property istic nuclear system. However, there is still considerable
of nuclear force in realistic nuclear system. To visualize distribution of P ( 0) 5%, corresponding to totally
this more straightforwardly, we compare the hVgJ9 g 9 g 9 g 9 i chaotic Q evolutions. These chaotic evolutions should be
2 2 2 2
taken as the back ground of our analysis, because the
and hVhJ11 h 11 h 11 h 11 i values with two-body interactions TBRE intrinsically and trivially maintains chaos. To il-
2 2 2 2
elements of a typical short-range interaction, i.e. the lustrate the back ground, we generate 25 000 000 sets
force, in Fig. 2. The similarity between them is obvious. of independently randomized Q values from the normal
Therefore, we conclude that even through the interaction distribution instead of from the TBRE, and present their
origin of the 0+ -g.s. predominance for a single nucleus in distribution as our back-ground distribution (denoted
the TBRE is unclear, the short-range property of nuclear by Pbkg ) in Fig. 3(a). To eliminate this interference
force is still the key to keep all the doubly even nuclei from the back-ground distribution, we present the ratio
have 0+ g.s. in both TBRE and realistic nuclear system. of P () for the Q systematics over Pbkg in Fig. 3(b),
3

13
where the predominance of the Q linearity is more obvi-
systematics in TBRE
ous with P ( > 0.9)/Pbkg 300.
+

One may argue that, the linear Q systematics out of systematics after 0 -g.s. sampling

the 0+ -g.s. sampling is trivial, because the 0+ -g.s. sam- 10

pling favors the -like interaction, and thus should en-

bkg
hance the seniority scheme, which has been proposed to

P( )/P
be the origin of the linearity of Q systematics in Ref. [9]. 7

To examine this argument, we also calculate the distri-

bution of the Q systematics in the whole TBRE without
the 0+ -g.s. sampling as shown in Fig. 3. P ( > 0.9) with 4

the 0+ -g.s. sampling is 3 times of that without this sam-

pling according to Fig. 3(a), which agrees with the claim
in Ref. [9], that the seniority scheme indeed enhances the 1

linearity of the Q systematics. However, in Fig. 3(b), the -1.0 -0.5 0.0 0.5 1.0
predominance of P ( > 0.9)/Pbkg is still obvious, even
without the 0+ -g.s. sampling. This means that the Q
linearity is robust in the whole TBRE, which can not be FIG. 4: (Color online) distributions of the mu systematics in
totally attributed to the seniority scheme here. the whole TBRE and with the 0+ -g.s. sampling, normalized
To search the origin of this Q linearity in the TBRE, with Pbkg . The dash line with P ()/Pbkg = 1 highlights a
we perform two additional samplings: totally independent variation from neutron number. The
error bar is determined by statistic error.
(I) the sampling with > 0.9 and 0+ ground states for
all the even-mass Cd isotopes; 1.0
+
0 g.s. sampling

(II) the sampling with > 0.9 regardless of even-mass increasing J

>0.9 systematics

Cd ground states. 0.5

The hVjJ1 j2 j3 j4 i values of these two samples are presented

in Fig. 1 compared with those after the 0+ -g.s. sampling. 0.0

Sampling (I) and the 0+ -g.s. sampling share the same

Vj j j j >
1 2 3 4

short-range property of like-nucleons interaction, i.e., rel-

VdJ3d 3d0 3d 3
J

atively attractive interaction elements with rank J = 0, -0.5

VsJ1s 1s01s 1
<

which can be trivial, because sampling (I) is actually 2 2 2 2

based on the 0+ -g.s. sampling. Furthermore, after sam- 2 2 2 2

-1.0
pling (I), the proton-neutron (pn) interaction elements
between 0g9/2 and 0h11/2 orbits obviously follow the
parabolic rule [19] as increasing rank J, corresponding to
V gJ9 g 90g 9 g 9 V hJ11 h110h11 h11 V gJ9 h11 g 9 h11
-1.5
2 2 2 2 2 2 2 2 2 2 2 2
the quadrupole interaction [20]. Sampling (II) also favors
a quadrupole-like pn interaction, even though the rank-
J = 0 interaction elements after this sampling present FIG. 5: (Color online) hVjJ1 j2 j3 j4 i values after the 0+ -g.s. sam-
not short-range property. Therefore, we conclude that pling and the > 0.9 sampling of systematics beyond the
the quadrupole pn interaction is responsible to induce 0+ -g.s. sampling, in an ascending order of j1 j2 j3 j4 J indexes.
the Q linearity, and the seniority scheme is a boost to The grey areas highlight characteristic of the pn interaction
this linearity in random-interaction ensemble. elements surviving the > 0.9 sampling. The dash line il-
lustrates the ensemble mean of the TBRE. The error bar is
determined by statistic error.
V. LINEARITY

In Fig. 4 we present distributions of the system- explained. However, pure seniority scheme can only pro-
atics with the 0+ -g.s. sampling and in the whole TBRE, vide a constant as agued in Ref. [9]. Thus, we need
normalized with Pbkg . Here Pbkg for systematics should to further probe other origin of the linearity beside the
be the same as that for Q. In the TBRE, P ()/Pbkg is seniority scheme. We perform a sampling for > 0.9
always close to 1, which demonstrates that the TBRE systematics based on the 0+ -g.s. sampling, and present
does not characterize the systematics. However, af- hVjJ1 j2 j3 j4 i values after such sampling in Fig. 5. < 0.9
ter the 0+ -g.s. sampling, relatively larger possibility for linearity is totally against experiments, and thus omitted
|| > 0.9 emerges (1% in the 0+ -g.s. sample), corre- here.
sponding to the the predominance of the linearity. In Fig. 5, the > 0.9 sampling and the 0+ -g.s. sam-
It seems that the linearity requires even-mass Cd pling share the short-range property of the like-nucleon
0+ ground states, i.e. the seniority scheme as we have interaction, which means that the seniority scheme is still
4

With V T =0 and V T =1 as fitting variables, we fit even-

J and odd-J hVgJ9 h 11 g 9 h 11 i values to Eqs. (1) and (2),
0.4 V gJ9 h11 g 9 h11 2 2 2 2
2 2 2 2 respectively. For even J, the best-fit V T =0 =-0.68(7); for
odd J, the best-fit V T =0 =-0.6(4) and V T =1 =0.4(2). The
0.0 fitting error for odd J is larger than that for even J. This
is because the odd-J fitting involves one more fitting vari-
able. We also note that the best-fit V T =0 values of even
-0.4 J and odd J agree with each other, which implies that
odd J: hVgJ9 h 11 g 9 h 11 i for > 0.9 systematics are describable
T=0
V = -0.6(4)
2 2 2 2

-0.8 T
with a unified pn interaction. According to the best-fit
V J:
=1
= 0.4(2) even
T=0
V T =0 and V T =1 values, to induce a linear evolution,
V = -0.68(7) the T = 0 pn interaction shall be attractive, while the
-1.2 T = 1 one is repulsive. This has also been long noted
0 2 4 6 8 10 in the realistic nuclear system, which may explain the
J linearity of experimental data.

FIG. 6: (Color online) hVgJ9/2 h11/2 g9/2 h11/2 i after the > 0.9
sampling of systematics with all the even Cd having 0+
ground states. The even-J behavior of hVgJ9/2 h11/2 g9/2 h11/2 i is
different from odd-J one, and thus they are presented sepa- VI. SUMMARY
rately. Straight lines illustrates the fitting to pn interaction
elements governed by the force. The error bar is determined
by statistic error. To summarize, the random-interaction ensemble pre-
dominantly reproduces the linear Q and systematics in
the Cd isotopes chain. The pn interaction is the key to
important for the linearity. Furthermore, the > 0.9 linearize the the Q and systematics, although the se-
sampling presents additional structure of hVgJ9 h 11 g 9 h 11 i. niority scheme is a significant boost. For the Q linearity,
2 2 2 2 the pn interaction presents quadrupole-like feature. For
We replot the detail of hVgJ9 h 11 g 9 h 11 i of the > 0.9 sam- the linearity, the -like pn interaction is required with
2 2 2 2
ple in Fig. 6. The even-J behavior of hVgJ9 h 11 g 9 h 11 i is repulsive T = 1 and attractive T = 0 components.
2 2 2 2
different from odd-J one: the even-J hVgJ9 h 11 g 9 h 11 i val- Our work also emphasizes that the short-range inter-
2 2 2 2 action between like nucleons is responsible to reproduce
ues present an obvious parabolic evolution, while those the I = 0+ ground states for all the even-mass nuclei in
with odd J seem less regulated. This odd-even differ- both TBRE and realistic nuclear system, which may pro-
ence also characterizes the pn interaction governed by vides a new viewpoint to understand the predominance
the force [20]. More specifically, the evolution of even- of I = 0 ground states in the TBRE.
J interaction between 0g9/2 and 0h11/2 orbit are only
attributed to the T = 0 force as
2
J=even 9/2 11/2 J
Vg 9 h 11 g 9 h 11
2 2 2 2 1/2 1/2 0 Acknowledgments
(1)
T =0 121
V 1+ ,
J(J + 1) This work was supported by the National Natural Sci-
ence Foundation of China under Grant Nos. 11647059,
while the odd-J interaction elements have both T = 0
11305151, the Research Fund for the Doctoral Program
and T = 1 contributions as
of the Southwest University of Science and Technology

9/2 11/2 J
2 under Grant No. 14zx7102, and the Graduate Education
VgJ=odd
9 h 11 g 9 h 11
Reform Project of the Southwest University of Science
2 2 2 2 1/2 1/2 0
(2) and Technology under Grant No. 17sxb119.
T =1 T =0 1
V +V .
J(J + 1)

[1] J. M. Allmond, Phys. Rev. C 88, 041307 (2013). [4] V. K. B. Kota, Phys. Rep. 347, 223 (2001).
[2] C. W. Johnson, G. F. Bertsch, and D. J. Dean, Phys. [5] V. Zelevinsky and A. Volya, Phys. Rep. 391, 311 (2004).
Rev. Lett. 80, 2749 (1998). [6] Y. M. Zhao, A. Arima, and N. Yoshinaga, Phys. Rep.
[3] Y. Lei, Phys. Rev. C 93, 024319 (2016). 400, 1 (2004).
5

[7] H. Weidenmueller and G. E. Mitchell, Rev. Mod. Phys. (1970).

81, 539 (2009). [15] O. Bohigas and J. Flores, Phys. Lett. B 34, 261 (1971).
[8] V. K. B. Kota, Embedded Random Matrix Ensembles in [16] S. S. M. Wong and J. B. French, Nucl. Phys. A 198, 188
Quantum Physics (Springer, Heidelberg, 2014). (1972).
[9] D. T. Yordanov, et al., Phys. Rev. Lett. 110, 192501 [17] E. Caurier and F. Nowacki, Acta Phys. Pol. 30, 705
(2013). (1999).
[10] N. B. de Takacsy, Phys. Rev. C 89, 034301 (2014). [18] K. Pearson, Proc. R. Soc. London 58, 240 (1895).
[11] P. W. Zhao, S. Q. Zhang, and J. Meng, Phys. Rev. C 89, [19] V. Paar, Nucl. Phys. A 331, 16 (1979).
011301 (2014). [20] R. F. Casten, in Nuclear Structure From A Simple Per-
[12] Y. Lei, H. Jiang, and S. Pittel, Phys. Rec. C 92, 024321 spective, edited by P. E. Hodgson (Oxford University
(2015). Press, 1990).
[13] J. Wood, Physics 6, 52 (2013).
[14] J. B. French and S. S. M. Wong, Phys. Lett. B 33, 449