PHYSICAL REVIEW LETTERS 134, 062502 (2025)

Deformation and Collectivity in Doubly Magic 208Pb

J. Henderson ,1,* J. Heery ,1 M. Rocchini ,2,3 M. Siciliano ,4 N. Sensharma ,4 A. D. Ayangeakaa ,5,6

R. V. F. Janssens ,5,6 T. M. Kowalewski ,5,6 Abhishek ,1 P. D. Stevenson ,1 E. Yüksel ,1 B. A. Brown ,7,8
T. R. Rodriguez ,9,10,11 L. M. Robledo ,10,11,12 C. Y. Wu ,13 S. Kisyov ,13,† C. Müller-Gatermann ,4 V. Bildstein ,3
L. Canete,1 C. M. Campbell ,14 S. Carmichael ,15 M. P. Carpenter,4 W. N. Catford ,1 P. Copp,4,‡ C. Cousins,1
M. Devlin ,16 D. T. Doherty,1 P. E. Garrett ,3 U. Garg ,15 L. P. Gaffney ,17 K. Hadynska-Klek ,18 D. J. Hartley ,19
S. F. Hicks ,20,21 H. Jayatissa ,4,‡ S. R. Johnson ,5,6 D. Kalaydjieva ,22 F. Kondev ,4 D. Lascar ,23 T. Lauritsen,4
G. Lotay,1 N. Marchini ,2 M. Matejska-Minda ,24 S. Nandi,4 A. Nannini ,2 C. O’Shea ,1 S. Pascu ,1,25 C. J. Paxman ,1
A. Perkoff,26 E. E. Peters,20 Zs. Podolyák ,1 A. Radich ,3 R. Rathod,15 B. J. Reed ,1,§ P. H. Regan ,1,27 W. Reviol,4
E. Rubino,7,∥ R. Russell ,1 D. Seweryniak,4 J. R. Vanhoy,19 G. L. Wilson ,4,28,¶
K. Wrzosek-Lipska ,18 S. W. Yates,20 and I. Zanon 29
1
School of Mathematics and Physics, University of Surrey, Guildford GU2 7XH, United Kingdom
2
Istituto Nazionale di Fisica Nucleare, Sezione di Firenze, I-50019 Sesto Fiorentino (Firenze), Italy
3
Department of Physics, University of Guelph, Guelph, Ontario N1G2W1, Canada
4
Argonne National Laboratory, Lemont, Illinois 60439, USA
5
University of North Carolina, Chapel Hill, North Carolina 27599, USA
6
Triangle Universities Nuclear Laboratory, Duke University, Durham, North Carolina 27708, USA
7
Facility for Rare Isotope Beams, Michigan State University, East Lansing, Michigan 48824, USA
8
Department of Physics & Astronomy, Michigan State University, East Lansing, Michigan 48824, USA
9
Departamento de Estructura de la Materia, Física Térmica y Electrónica, Universidad Complutense de Madrid,
E-28040 Madrid, Spain
10
Departamento de Física Teórica, Universidad Autónoma de Madrid, E-28049 Madrid, Spain
11
Centro de Investigación Avanzada en Física Fundamental-CIAFF-UAM, E-28049 Madrid, Spain
12
Center for Computational Simulation, Universidad Politécnica de Madrid, Campus de Montegancedo,
Bohadilla del Monte, E-28660 Madrid, Spain
13
Lawrence Livermore National Laboratory, Livermore, California 94550, USA
14
Nuclear Science Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA
15
Department of Physics and Astronomy and The Joint Institute for Nuclear Astrophysics, University of Notre Dame,
30 Notre Dame, Indiana 46556, USA
16
Los Alamos National Laboratory, Los Alamos, New Mexico 87545, USA
17
Department of Physics, University of Liverpool, Liverpool, United Kingdom
18
Heavy Ion Laboratory, University of Warsaw, PL-02-093 Warsaw, Poland
19
Department of Physics, U.S. Naval Academy, Annapolis, Maryland 21402, USA
20
Departments of Chemistry and Physics & Astronomy, University of Kentucky, Lexington, Kentucky 40506, USA
21
Department of Physics, University of Dallas, Irving, Texas 75062, USA
22
IRFU, CEA Saclay, Université Paris-Saclay, France
23
Department of Physics and Astronomy, Northwestern University, Evanston, Illinois 60208, USA
24
Institute of Nuclear Physics Polish Academy of Sciences, Kraków, Poland
25
National Institute for Physics and Nuclear Engineering, Bucharest-Magurele, R-77125, Romania
26
U.S. Naval Academy, Annapolis, Maryland 21402, USA
27
National Physical Laboratory, Teddington, Middlesex TW11 0LW, United Kingdom
28
Department of Physics and Astronomy, Louisiana State University, Baton Rouge, Louisiana 70803, USA
29
Laboratori Nazionali di Legnaro, Laboratori Nazionali di Legnaro, Legnaro, Italy
(Received 5 July 2024; revised 3 October 2024; accepted 18 December 2024; published 14 February 2025)

*
Contact author: [email protected]
†
Present address: Nuclear Science Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA.
‡
Present address: Los Alamos National Laboratory, Los Alamos, New Mexico 87545, USA.
§
Present address: TRIUMF, Vancouver V6T 2A3, British Columbia, Canada.
∥
Present address: Lawrence Livermore National Laboratory, Livermore, California 94550, USA.
¶
Present address: UK Atomic Energy Authority, Culham Campus, Abingdon OX14 3DB, United Kingdom.

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further
distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI.

0031-9007=25=134(6)=062502(8) 062502-1 Published by the American Physical Society

 PHYSICAL REVIEW LETTERS 134, 062502 (2025)

Lead-208 is the heaviest known doubly magic nucleus and its structure is therefore of special interest.
Despite this magicity, which acts to provide a strong restorative force toward sphericity, it is known to
exhibit both strong octupole correlations and some of the strongest quadrupole collectivity observed in
doubly magic systems. In this Letter, we employ state-of-the-art experimental equipment to conclusively
demonstrate, through four Coulomb-excitation measurements, the presence of a large, negative,
spectroscopic quadrupole moment for both the vibrational octupole 3−1 and quadrupole 2þ 1 state, indicative
of a preference for prolate deformation of the states. The observed quadrupole moment is discussed in the
context of the expected splitting of the 3− ⊗ 3− two-phonon states, due to the coupling of the quadrupole
and octupole motion. These results are compared with theoretical values from three different methods,
which are unable to reproduce both the sign and magnitude of this deformation. Thus, in spite of its well-
studied nature, 208Pb remains a puzzle for our understanding of nuclear structure.

DOI: 10.1103/PhysRevLett.134.062502

Since the identification of the nuclear spin-orbit inter- 208

Pb, which lies at the frontier of state-of-the-art ab initio
action, and its role in the creation of so-called magic nuclear theory [15].
numbers [1,2], doubly magic nuclei have acted as corner- The nuclear magic numbers act to stabilize spherical
stones of nuclear physics. These isotopes act to delimit the nuclear configurations, whereas collective properties
nuclear landscape, allowing it to be subdivided into emerge naturally from the deviation of the nucleus from
relatively manageable valence spaces. Of these key iso- sphericity. Any distinction is, however, somewhat con-
topes, 208Pb is the heaviest known example, lying at the trived, as evidenced by the existence of shape coexistence
intersection of the proton Z ¼ 82 and neutron N ¼ 126 in doubly magic systems (see, e.g., Ref. [16]). The nucleus
magic numbers that dictate the nuclear structure of the mass is not a rigid object, with a broad (“soft”) minimum in its
region (see, e.g., Refs. [3,4]). Studying this nucleus is thus potential energy surface with respect to deformation,
key to understanding not only the structure of doubly magic allowing for its collective wave function to have some
nuclei, but also the entire region of the nuclear landscape distribution in deformation parameter space [17].
around mass number A ¼ 200 with broad implications. For Collective observables typically relate to the square of
example, the aforementioned N ¼ 126 magic number plays the nuclear collective wave function and any softness in
a critical role in the formation of the third abundance peak deformation space introduces a collective behavior, even if
of the astrophysical r process—a series of rapid neutron the nucleus itself exhibits a spherical minimum in its
captures leading to the formation of approximately half of potential energy. Indeed, these oscillations about sphericity
the elements heavier than iron in our Galaxy. Specifically, have led to recent assertions that the very meaning of a
the intersection of the r process capture path with the nuclear shape for doubly magic nuclei is to be treated with
N ¼ 126 shell closure results in an accumulation of caution [17].
material in the A ∼ 195 region [5], producing a broad peak Arguably the most famous example of a vibrational
in the observed abundance pattern of many stars in the excitation in nuclear physics is the 3−1 state in 208Pb [18].
Milky Way. Given that this is one of the key signatures of This state is unusual in many senses, not least as a rare
“main” r-process nucleosynthesis, understanding the prop- example of a first-excited state in an even-even nucleus that
erties of nuclei in this region is of paramount importance is not Jπ ¼ 2þ . The presence of a 3− state at low excitation
for our understanding of the production of heavy elements. energy in an even-even nucleus is indicative of a relatively
Beyond the role the N ¼ 126 shell closure plays in strong degree of octupole collectivity, further borne out by
r-process nucleosynthesis, 208Pb is also key to the deter- experiment, with a BðE3; 3−1 → 0þ 1 Þ transition strength of
mination of the symmetry energy, J, and its slope, L, in the 33.8(6) W.u. [19]. Nonetheless, 208Pb remains a doubly
nuclear equation of state, which governs the properties of magic system, resistant to any permanent deformation in its
nuclear matter under extreme conditions, such as those ground state, and this excitation is thus thought to be
found in neutron stars. The neutron-skin thickness of 208Pb vibrational. As a one-phonon octupole vibration, this state
has been experimentally investigated using parity-violating is expected to be accompanied by a quartet of 3− ⊗ 3− two-
electron scattering [6,7], hadronic [8,9], electroweak [10], phonon partners, with Jπ ¼ 0þ ; 2þ ; 4þ ; 6þ , lying at
and gravitational wave [11] probes. Because of the strong approximately twice its energy. The location and splitting
correlation between this nuclear property and L, it has been of this quartet is, however, strongly influenced by the
possible to place constraints on the nuclear equation of state coupling of the octupole and quadrupole motion in the
around saturation densities [12,13] as well as on the system [20–22]. To date, this quartet of two-phonon states
maximum neutron star mass radius [14]. Key to this, has eluded conclusive identification [23], with the best
however, is a detailed knowledge of the structure of evidence presented for the 0þ member [24,25], raising

062502-2
 PHYSICAL REVIEW LETTERS 134, 062502 (2025)

questions about the potential strength of this quadrupole

coupling.
Our understanding of the couplings of the quadrupole
and octupole motions, and the quadrupole behavior of
the nucleus more generally, is aided by the study of
three experimental observables: the transition strength
between the 2þ þ
1 excited state and the 01 ground state
þ þ
(BðE2; 21 → 01 Þ), and the spectroscopic quadrupole
moments (Qs ) of the 3−1 and 2þ 1 states. Of these, the
BðE2; 2þ 1 → 0 þ
1 Þ value in 208
Pb has been well measured,
and is found to be 8.3(5) W.u., which is considerably
smaller than expected for a nucleus exhibiting permanent
deformation (i.e., a rotor), but is also the largest value
observed for any doubly magic system [26]. The Qs ð3−1 Þ
value has also been measured [22,27–29] previously, albeit
with some concern raised for the early results [27] regard- FIG. 1. Total, angle-integrated γ-ray spectra, Doppler-corrected
ing the contribution of nuclear interference terms. A careful for 208Pb reaction kinematics, showing the 2614-keV 3−1 → 0þ 1
analysis, excluding such terms, was performed in transition [top] and the 4085-keV 2þ þ
1 → 01 [bottom] transition
Refs. [22,29], and demonstrated a large, negative quadru- for beams of 166Er, 150Nd, 130Te, and 70Ge.
pole moment, Qs ð3−1 Þ ¼ −0.34ð15Þ e b, inconsistent with,
although with a large uncertainty, theoretical descriptions
available at the time. In the same work, a measurement of a combined analysis enhances sensitivity through the use of
Qs ð2þ beams of differing Z and with differing angular coverage in
1 Þ ¼ −0.7ð3Þ e b was also made, with very large
uncertainties. In this Letter, we present new results using the center-of-mass frame. Both of these variables influence
a state-of-the-art experimental setup, making use of heavy- the strength of multistep Coulomb excitation and the
ion induced Coulomb excitation, demonstrating conclu- reorientation effect, improving sensitivity to observables
sively a significant role for quadrupole deformation in associated with those effects, such as spectroscopic quadru-
208
Pb. We discuss these results within the context of state- pole moments.
of-the-art nuclear theory, finding that the observed deforma- Gamma rays detected in GRETINA were Doppler-cor-
tion cannot be reproduced. Finally, we present the anticipated rected for 208Pb recoils on the basis of two-body reaction
scale of the splitting of the 3− ⊗ 3− two-phonon states due to kinematics deduced from the measured scattering angle in
the coupling of octupole and quadrupole motion. CHICO2. Figure 1 shows two portions of the resultant
Excited states in 208Pb were populated in heavy-ion Doppler-corrected γ-ray spectrum with clear peaks corre-
induced Coulomb excitation by beams of 70Ge (297 MeV), sponding to the 3−1 → 0þ þ þ
1 and 21 → 01 deexciting transi-
130
Te (527 MeV), 150Nd (640 MeV), and 166Er (749 MeV), tions at 2614 and 4085 keV, respectively. A time-random
subtraction was performed to remove contributions to the
that impinged upon 99.86-% enriched, 0.45-mg=cm2 thick
208 2614-keV peak due to natural background. This subtraction
Pb targets. The Pb was deposited upon a 40-μg=cm2
was validated by observing the complete removal of the
carbon backing, and the target was oriented with the
backing facing the upstream direction. For the scattering 1461-keV γ-ray from the decay of 40K from the spectra. In
angles used in the analysis the beam energies used order to enhance sensitivity to the spectroscopic quadrupole
correspond to distances of closest approach of more than moment, the data were subdivided by the angle measured
6 fm, sufficient to almost completely suppress any nuclear in CHICO2, allowing for the angular distribution of the
component to the excitation process [30]. Deexciting γ-rays reaction cross section to be assessed. Eight angular ranges
were detected in the GRETINA array of high-purity were used for the data acquired with 166Er and 150Nd beams,
germanium detectors [31], consisting of 12 modules at and four angular ranges each for the 130Te and 70Ge data. The
the time of the experiments. The relative efficiency of the size, location, and number of angular ranges was partly
GRETINA array was verified experimentally for energies dictated by the reaction kinematics and the statistics of the
up to 4806 keV using a source of 66Ga alongside standard 2þ þ
1 → 01 transition but also so as to avoid any discontinuities
γ-ray calibration sources. Scattered beamlike and targetlike in the angular distribution, or any regions where the shape
nuclei were detected simultaneously using the CHICO2 of the partner nucleus (i.e., 166Er, 150Nd, 70Ge, and 130Te)
[32] array of 20 parallel-plate avalanche counters, with the excitation angular distribution deviated significantly from
beam and target nuclei distinguished based on their relative expectations. This allowed the use of a single, common
time of flight and scattering angle. This combined setup normalization parameter for each beam for all angular ranges.
provides exceptional efficiency, angular resolution, and Gamma-ray yields for each angular range (see
high-rate capabilities. Compared to a single-beam analysis, Supplemental Material) were analyzed using the GOSIA

062502-3
 PHYSICAL REVIEW LETTERS 134, 062502 (2025)

TABLE I. Transition and diagonal E2 and E3 matrix elements

determined from an analysis combining all data obtained in the
present work. Where indicated, the literature BðE2; 2þ þ
1 → 01 Þ
value was included to further constrain the minimization. Note
that, in all cases, the literature BðE3; 3−1 → 0þ
1 Þ strength was
included as a constraint.

J πi → J πf hJ πi jE2jJ πf i (e b) BðE2Þ↓ (W.u.) Note

0þ þ
1 → 21 0.489(77) 6.5(21)

J πi hJ πi jE2jJ πi i (e b) Qs ðJ πi Þ (e b) Note
3−1 −0.81ð14Þ −0.63ð11Þ
3−1 −0.73ð11Þ −0.57ð9Þ BðE2Þ incl.
2þ1 −0.31ð33Þ −0.24ð25Þ
2þ1 −0.55ð9Þ −0.42ð7Þ BðE2Þ incl.

Table I both with and without the inclusion of the literature

BðE2; 2þ þ
1 → 01 Þ value in the fitting routine.
FIG. 2. Mean values (points) and confidence regions corre-
sponding to 1σ (solid lines) and 2σ (dotted lines) for the
The results in Table I are presented alongside the reduced
h0þ − − − transition strengths [BðEλÞ] in Weisskopf units (W.u.),
1 jE3j31 i and h31 jE2j31 i matrix elements (top), and the
h0þ jE2j2 þ
i h2 þ
jE2j2 þ indicating the expected strength of a single-particle tran-
1 1 and 1 1 i matrix elements (bottom). All data
were constrained by the known BðE3; 3−1 → 0þ sition. BðEλÞ values far in excess of 1 W.u. are indicative of
1 Þ ¼ 33.8ð6Þ W:u:
[19] strength. Limits are shown for the case where BðE2; 2þ a collective enhancement due to deformation. The results
1 →
0þ Þ ¼ 8.3ð5Þ W:u: [40,41] was (red) and was not (black) therefore paint a picture of a system with relatively strong
1
included as a constraint in the minimization. Also shown are octupole deformation, coupled to a modest quadrupole
bands corresponding to the 1σ limits on literature values for the deformation, with a large spectroscopic quadrupole moment
h3−1 jE2j3−1 i, h0þ þ þ þ
1 jE2j21 i, and h21 jE2j21 i matrix elements. approaching the limit of an axial rotor [20] in the case of the
Shown by the dashed line is the axial-rotor limit for the 2þ1 state (Fig. 2). In order to better understand these results,
h2þ þ þ
1 jE2j21 i matrix element as a function of h01 jE2j21 i.
þ
we compare them to new predictions from three theoretical
models: the nuclear shell model, a Gogny density functional
semiclassical Coulomb-excitation code [33], with minimi- theory coupled to a symmetry-conserving configuration-
zation handled through an external program [34]. Data mixing method (SCCM), the self-consistent Hartree Fock
from each beam were analyzed independently to confirm ðHFÞ þ random-phase approximation (RPA) method, and
consistency before a simultaneous analysis of all the data the time-dependent Hartree-Fock (TDHF) method, using
was performed. A summary of the modern Coulomb- Skyrme interactions in both cases. We also compare the
excitation analysis procedures used in this work can be results to predictions from a previous theoretical study using
found in Ref. [35]. Four parameters were varied in fitting the a generator coordinate method with a relativistic energy
experimental data: h0þ − − − þ þ density functional [42]. Details of the new calculations
1 jE3j31 i, h31 jE2j31 i, h01 jE2j21 i and
h21 jE2j21 i matrix elements. The literature BðE3; 31 → 0þ
þ þ − performed in this work are included in the Supplemental
1Þ
Material [36].
value [18,22] and its uncertainty was included in the χ 2
Shell-model calculations were performed using the
minimisation as an additional constraint, allowing for the
method outlined in Ref. [43]. Particle-hole excitations
determination of absolute E2 and E3 matrix elements. Only
are only permitted at the 1p − 1h and 2p − 2h level.
the 3− → 0þ þ þ
1 and 21 → 01 transitions were observed in the
The excitation energy of the 2þ 1 state in
208Pb in these
present work, however a comprehensive level scheme (see
Supplemental Material [36]) was included in the GOSIA calculations is overpredicted by approximately 700 keV
simulation, including the E1 transition coupling the two and that of the 3−1 state by approximately 400 keV. The E3
states of interest. The resulting confidence intervals are given values are calculated with effective charges of ep ¼ 1.6 and
in black in Fig. 2, indicating the strong correlations between en ¼ 0.6 and E2 values with ep ¼ 1.5 and en ¼ 0.98 [44].
the transitional h0þ þ þ þ
1 jE2j21 i and diagonal h21 jE2j21 i matrix
These shell-model calculations yield BðE3; 3−1 → 0þ 1Þ¼
þ þ
elements. To better constrain h21 jE2j21 i, the BðE2; 2þ
þ þ
1 →
36 W:u: and BðE2; 21 → 01 Þ ¼ 3.8 W:u:, with Qs ð3−1 Þ ¼
0þ1 Þ strength was included as an additional datum [40,41], −0.12 e b and Qs ð2þ 1 Þ ¼ 0.08 e b. While the experimental
resulting in the more tightly constrained confidence intervals BðE3; 3−1 → 0þ 1 Þ value is well reproduced, all electric-
shown in red in Fig. 2. The resulting values are presented in quadrupole observables are significantly underpredicted.

062502-4
 PHYSICAL REVIEW LETTERS 134, 062502 (2025)

Of particular interest, the magnitudes of the computed

Qs ð3−1 Þ and Qs ð2þ 1 Þ values are not only considerably
smaller than the experimental ones, but also have opposing
signs, whereas the experimental values are both negative
and of comparable magnitude.
Symmetry-conserving configuration-mixing calculations
were performed using the method outlined in Ref. [45]. In
general, excitation energies are poorly reproduced, with
Eð3−1 Þ ¼ 5.6 MeV [experiment: Eð3−1 Þ ¼ 2.6 MeV] and
Eð2þ þ
1 Þ ¼ 9.2 MeV [experiment: Eð21 Þ ¼ 4.1 MeV]. This
is thought to be due to an as-yet unexplained enhancement
in the octupole correlations in the ground state at defor-
FIG. 3. Spectroscopic quadrupole moments (Qs ) and electric
mation β2 ¼ 0. Transition strengths are overpredicted with
transition strengths [BðEλÞ] for the 2þ −
1 and 31 states studied in the
respect to experiment, with BðE3; 3−1 → 0þ 1 Þ ¼ 41.3 W:u: present work. Shown for comparison are the HF þ RPA calcu-
and BðE2; 2þ 1 → 0 þ
1 Þ ¼ 14.3 W:u: Spectroscopic quadru- lations using the Skyrme-type SkM* and SKX functionals,
pole moments, on the other hand, are underpredicted, with calculations using the SkM* and SKX functionals, symmetry-
Qs ð3−1 Þ ¼ −0.104 e b and Qs ð2þ 1 Þ ¼ −0.117 e b. We note
conserving configuration mixing (SCCM) calculations, and shell-
that similar calculations were performed in Ref. [42]. The model calculations, as well as the results using the generator
coordinate method (GCM) based on a covariant energy density
calculations presented here differ in the choice of inter-
functional by Yao and Hagino in Ref. [42]. BðEλÞ uncertainties
action and in the fact that we do not perform a phenom- are smaller than the symbol size.
enological quenching of excitation energy. The calculations
of Ref. [42] yield BðE2; 2þ þ
1 → 01 Þ ¼ 5.5 W:u: and
BðE3; 3−1 → 0þ 1 Þ ¼ 21.9 W:u:, but do not report values A comparison of all experimental and theoretical data for
for Qs ð2þ 1 Þ or Qs ð3−1 Þ. Using one dimensional generator the states of interest is presented in Fig. 3. Interestingly,
coordinate method calculations with intrinsic wave func- both SCCM and shell-model results predict similar behav-
tions (i.e., no projection) one can also estimate the ior for the 3−1 state, with a modest, negative Qs ð3− Þ value
excitation energies and transition strengths for the 2þ and a BðE3; 3−1 → 0þ 1 Þ strength of comparable magnitude
and 3− states, using β2 and β3 , respectively, as coordinates. to that observed in experiment. In this work we find an
The 2þ state appears at 7.03 MeV with BðE2; 2þ þ experimental Qs ð3− Þ moment that is considerably larger in
1 → 01 Þ ¼
− absolute terms than either calculated value. This indicates a
4.13 W:u. On the other hand [46] the 3 level appears at
4.01 MeV with BðE3; 3−1 → 0þ stronger quadrupole coupling to the octupole deformed
1 Þ ¼ 7.7 W:u. The lower
value of the excitation energies, as compared to the projected state than either model predicts. This would point to a more
calculation, confirms the unexpectedly large stretching of extended collective wave function than observed in Fig. 4(c)
with respect to the quadrupole degree of freedom, and with a
excitation energies arising from the angular momentum
projection and exacerbated in the spherical system. The
low transition strengths in the intrinsic calculation are due to
the unsuitable use of the rotational formula to connect
multipole moments and transition strengths in a spherical
system [47]. In the projected calculation, transition strengths
are computed exactly and there is no need for approximation.
Finally, we calculate excitation spectra from Skyrme
density functionals in RPA and TDHF calculations, which
produce consistent results. Using the SKX interaction [48],
Eð2þ þ þ
1 Þ ¼ 4.08 MeV with BðE2; 21 → 01 Þ ¼ 6.26 W:u.
þ
The SkM* interaction [49] yields Eð21 Þ ¼ 4.79 MeV with
BðE2; 2þ þ
1 → 01 Þ ¼ 8.82 W:u. The RPA can also be used to
predict the properties of the 3−1 state. Using the SKX
interaction, Eð3−1 Þ ¼ 2.619 MeV with BðE3; 3−1 → 0þ 1Þ¼
30.45 W:u:, while the SkM* interaction gives Eð3−1 Þ ¼
FIG. 4. Total energy surfaces for the angular-momentum
3.301 MeV with BðE3; 3−1 → 0þ 1 Þ ¼ 33.78 W:u. The projected (a) 0þ , (b) 2þ , and (c) 3− states in 208Pb calculated
results obtained using the SKX interaction are in a better with the SCCM method described in the text. The collective wave
agreement with the experimental data in terms of the functions for the 0þ þ −
1 , 21 , and 31 states are shown in (d), (e), and
excitation energies and transition probabilities. (f), respectively.

062502-5
 PHYSICAL REVIEW LETTERS 134, 062502 (2025)

more prolate-dominant form. The BðE3; 3−1 → 0þ 1 Þ strength, In conclusion, we present a state-of-the-art Coulomb-
on the other hand, is approximately consistent with experi- excitation study of the heaviest known doubly magic
ment in its magnitude. nucleus, 208Pb. Thanks to the experimental sensitivity, we
The clearest discrepancy between theories and experi- are able to conclusively demonstrate the large, negative
ment is found in the behavior of the 2þ 1 state. Both shell- spectroscopic quadrupole moments of both the 3−1 and 2þ 1
model and SCCM predict a relatively small absolute states, as well as their consistent magnitudes. These results
Qs ð2þ Þ value, whereas the experiment indicates the pres- are indicative of a time-averaged prolate deformation for the
ence of a large, negative moment. The shell model system. As with previous attempts to calculate the Qs ð3−1 Þ
significantly underpredicts the BðE2; 2þ þ
1 → 01 Þ value as [50–52] moment, sophisticated contemporary nuclear
well. In general, the failure of the shell model to reproduce models fail to reproduce both the sign and magnitude of
E2 properties can be understood from the limited mp − mh the quadrupole moments for both the 2þ −
1 and 31 states. Thus,
excitations included in the calculation, which are well even as a cornerstone of the nuclear landscape, 208Pb remains
known to be essential to reproduce quadrupole effects. The a puzzle for nuclear structure theories.
inability of the SCCM values to reproduce the relative
BðE2; 2þ þ þ
1 → 01 Þ and Qs ð2 Þ behavior is more surprising. Acknowledgments—The authors thank the beam-delivery
While the overprediction of the BðE2; 2þ þ
1 → 01 Þ strength team at the ATLAS facility of Argonne National Laboratory
can be understood from deficiencies in the model, for providing the beams. Work at the University of Surrey
described earlier, the relative behavior of the two observ- was supported under UKRI Future Leaders Fellowship Grant
ables might be expected to be more reliable. In Fig. 4(e), the No. MR/T022264/1. Work at the University of Surrey was
collective wave function for the 2þ 1 state can be seen to be supported under the Science and Technologies Facilities
an admixture of two components, with both oblate (β2 < 0) Council (STFC) under Grant No. ST/V001108/1. Work at
and prolate (β2 > 0) contributions. The present experimen- the University of Liverpool was supported by STFC. Work
tal work implies that the prolate contribution to this state, was supported by the NSF under Grants No. PHY-2208137,
contributing negatively to the quadrupole moment, is No. PHY-2011890, No. PHY-1913028, and No. PHY-
stronger than predicted in the SCCM calculations and 2110365. Work of Argonne National Laboratory was sup-
the amplitude of the oblate component is too strong. The ported by the U.S. Department of Energy, Office of Science,
sign of β2 was then monitored in the Skyrme TDHF Office of Nuclear Physics, under Contract No. DE-AC02-
calculations, which reproduce both excitation energies 06CH11357. Work at Lawrence Livermore National
and transition strengths. While Qs values cannot be Laboratory was performed under the auspices of the U.S.
extracted from the model, the time average of the defor- Department of Energy under Contract No. DE-AC52-
mation was found to be zero, with no preference for prolate 07NA27344. Work at UNC is sponsored by the U.S.
or oblate deformation. Thus, within the scope of the model, Department of Energy under Grants No. DE-FG02-
there was no indication of the stronger prolate contribution 97ER41041 (UNC) and No. DE-FG02-97ER41033
to the 2þ1 state posited above. Qualitatively, however, this (TUNL). Los Alamos National Laboratory is operated by
prolate dominance appears to be reproduced in the collec- Triad National Security, LLC, for the National Nuclear
tive wave functions calculated in Ref. [42], with a pre- Security Administration of the U.S. Department of Energy
dominantly prolate configuration for the 2þ 1 state and a (Contract No. 89233218CNA000001). Work was supported
more extended prolate wave function for the 3−1 level. under U.S. Department of Energy under Contract No. DE-
Importantly, the presently measured Qs ð3−1 Þ moment is SC0021243. This material is based upon work supported by
expected [20] to have implications for the splitting between the U.S. Department of Energy, Office of Science, Office of
the energies of the states in the two octupole-phonon Nuclear Physics, under Contract No. DEAC02-05CH11231
quartet in 208Pb. Using the approximate relation for the (LBNL). GRETINA was funded by the U.S. DOE, Office of
splitting of the two-phonon states presented in Ref. [20] Science, Office of Nuclear Physics, and operated by the ANL
suggests a splitting of between 800 and 1500 keV between and LBNL contract numbers above. This research uses
the Jπ ¼ 0þ , 2þ , 4þ , and 6þ states due to the coupling of resources of ANL’s ATLAS facility, which is a U.S.
quadrupole and octupole modes. As previously mentioned, Department of Energy Office of Science User facility.
a candidate has been identified for the 0þ member of this Work was supported by the National Science Centre,
quartet, lying at 5241 keV. Within the model presented in Poland under Grant No. 2023/50/E/ST2/00621.
Ref. [20], the splitting required to reproduce this excitation
energy would imply no coupling between octupole and
quadrupole modes. This observed 0þ state viewed along-
side the present result thus presents something of a puzzle. [1] M. G. Mayer, On closed shells in nuclei. II, Phys. Rev. 75,
Either the model presented in Ref. [20] is not appropriate, 1969 (1949).
or there exists some phenomenon providing a near-exact [2] O. Haxel, J. H. D. Jensen, and H. E. Suess, On the “magic
cancellation of the implied splitting. numbers” in nuclear structure, Phys. Rev. 75, 1766 (1949).

062502-6
 PHYSICAL REVIEW LETTERS 134, 062502 (2025)

[3] E. Caurier, G. Martínez-Pinedo, F. Nowacki, A. Poves, and [20] A. Bohr and B. R. Mottelson, Nuclear Structure (Benjamin,
A. P. Zuker, The shell model as a unified view of nuclear New York, 1975), Vol. 2, pp. 45, 569–570.
structure, Rev. Mod. Phys. 77, 427 (2005). [21] J. Blomqvist, The 3− × 3− two-phonon quartet and the
[4] T. Otsuka, A. Gade, O. Sorlin, T. Suzuki, and Y. Utsuno, proton pairing vibration in 208Pb, Phys. Lett. 33B, 541
Evolution of shell structure in exotic nuclei, Rev. Mod. (1970).
Phys. 92, 015002 (2020). [22] R. Spear, W. Vermeer, M. Esat, J. Kuehner, A. Baxter, and S.
[5] J. J. Cowan, C. Sneden, J. E. Lawler, A. Aprahamian, M. Hinds, An improved determination of the quadrupole mo-
Wiescher, K. Langanke, G. Martínez-Pinedo, and F.-K. ment of the first excited state of 208Pb, Phys. Lett. 128B, 29
Thielemann, Origin of the heaviest elements: The rapid (1983).
neutron-capture process, Rev. Mod. Phys. 93, 015002 [23] K. Vetter et al., Fragmentation of the two-phonon octupole
(2021). vibrational states in 208Pb, Phys. Rev. C 58, R2631 (1998).
[6] S. Abrahamyan et al. (PREX Collaboration), Measurement [24] M. Yeh, P. E. Garrett, C. A. McGrath, S. W. Yates, and T.
of the neutron radius of 208Pb through parity violation in Belgya, Two-phonon octupole excitation in 208pb, Phys.
electron scattering, Phys. Rev. Lett. 108, 112502 (2012). Rev. Lett. 76, 1208 (1996).
[7] D. Adhikari et al. (PREX Collaboration), Accurate deter- [25] M. Yeh, M. Kadi, P. E. Garrett, C. A. McGrath, S. W. Yates,
mination of the neutron skin thickness of 208Pb through and T. Belgya, Candidates for two-phonon octupole ex-
parity-violation in electron scattering, Phys. Rev. Lett. 126, citations in 208Pb, Phys. Rev. C 57, R2085 (1998).
172502 (2021). [26] B. Pritychenko, M. Birch, B. Singh, and M. Horoi, Tables of
[8] A. Trzcińska, J. Jastrzębski, P. Lubiński, F. J. Hartmann, R. E2 transition probabilities from the first 2þ states in even–
Schmidt, T. von Egidy, and B. Kłos, Neutron density even nuclei, At. Data Nucl. Data Tables 107, 1 (2016).
distributions deduced from antiprotonic atoms, Phys. Rev. [27] A. R. Barnett and W. R. Phillips, Coulomb excitation and
Lett. 87, 082501 (2001). reorientation of the octupole state in 208Pb, Phys. Rev. 186,
[9] J. Zenihiro, H. Sakaguchi, T. Murakami, M. Yosoi, Y. 1205 (1969).
Yasuda, S. Terashima, Y. Iwao, H. Takeda, M. Itoh, H. P. [28] A. M. R. Joye, A. M. Baxter, M. P. Fewell, D. C. Kean, and
Yoshida, and M. Uchida, Neutron density distributions of R. H. Spear, Quadrupole moment of the first 3− state in
204;206;208 208
Pb deduced via proton elastic scattering at Pb, Phys. Rev. Lett. 38, 807 (1977).
Ep ¼ 295 mev, Phys. Rev. C 82, 044611 (2010). [29] W. J. Vermeer, M. T. Esat, J. A. Kuehner, R. H. Spear, A. M.
[10] C. M. Tarbert et al. (Crystal Ball at MAMI and A2 Baxter, and S. Hinds, Coulomb excitation of the 2.615 MeV
Collaborations), Neutron skin of 208Pb from coherent pion (3þ ) and 4.086 MeV (2þ ) states of 208Pb, Aust. J. Phys. 37,
photoproduction, Phys. Rev. Lett. 112, 242502 (2014). 123 (1984).
[11] F. J. Fattoyev, J. Piekarewicz, and C. J. Horowitz, Neutron [30] P. Lesser, D. Cline, P. Goode, and R. Horoshko, Static electric
skins and neutron stars in the multimessenger era, Phys. quadrupole moments of the first excited states of 56Fe and the
Rev. Lett. 120, 172702 (2018). even titanium nuclei, Nucl. Phys. A190, 597 (1972).
[12] P.-G. Reinhard, X. Roca-Maza, and W. Nazarewicz, Com- [31] S. Paschalis et al., The performance of the gamma-ray
bined theoretical analysis of the parity-violating asymmetry energy tracking in-beam nuclear array GRETINA, Nucl.
for 48Ca and 208Pb, Phys. Rev. Lett. 129, 232501 (2022). Instrum. Methods Phys. Res., Sect. A 709, 44 (2013).
[13] E. Yüksel and N. Paar, Implications of parity-violating [32] C. Wu, D. Cline, A. Hayes, R. Flight, A. Melchionna, C.
electron scattering experiments on 48Ca (CREX) and 208Pb Zhou, I. Lee, D. Swan, R. Fox, and J. Anderson, CHICO2, a
(PREX-II) for nuclear energy density functionals, Phys. two-dimensional pixelated parallel-plate avalanche counter,
Lett. B 836, 137622 (2023). Nucl. Instrum. Methods Phys. Res., Sect. A 814, 6 (2016).
[14] B. T. Reed, F. J. Fattoyev, C. J. Horowitz, and J. [33] T. Czosnyka, D. Cline, and C. Y. Wu, Bull. Am. Phys. Soc.
Piekarewicz, Implications of PREX-2 on the equation of 28, 745 (1983).
state of neutron-rich matter, Phys. Rev. Lett. 126, 172503 [34] J. Henderson, https://github.com/jhenderson88/GOSIA
(2021). Fitter.
[15] B. Hu, W. Jiang, T. Miyagi, Z. Sun, A. Ekström, C. Forssén, [35] M. Zielińska, L. P. Gaffney, K. Wrzosek-Lipska, E.
G. Hagen, J. D. Holt, T. Papenbrock, S. R. Stroberg, and I. Clément, T. Grahn, N. Kesteloot, P. Napiorkowski, J.
Vernon, Ab initio predictions link the neutron skin of 208Pb Pakarinen, P. Van Duppen, and N. Warr, Analysis methods
to nuclear forces, Nat. Phys. 18, 1196 (2022). of safe coulomb-excitation experiments with radioactive ion
[16] K. Heyde and J. L. Wood, Shape coexistence in atomic beams using the GOSIA code, Eur. Phys. J. A 52, 99 (2016).
nuclei, Rev. Mod. Phys. 83, 1467 (2011). [36] See Supplemental Material at http://link.aps.org/
[17] A. Poves, F. Nowacki, and Y. Alhassid, Limits on assigning supplemental/10.1103/PhysRevLett.134.062502 for gamma-
a shape to a nucleus, Phys. Rev. C 101, 054307 (2020). ray yields, level-schemes used in the present work and
[18] D. Goutte, J. B. Bellicard, J. M. Cavedon, B. Frois, M. Huet, details of the theoretical calculations, which includes
P. Leconte, P. X. Ho, S. Platchkov, J. Heisenberg, J. Refs. [37–39].
Lichtenstadt, C. N. Papanicolas, and I. Sick, Determination [37] J. Berger, M. Girod, and D. Gogny, Time-dependent
of the transition charge density of the octupole vibration in quantum collective dynamics applied to nuclear fission,
208Pb, Phys. Rev. Lett. 45, 1618 (1980). Comput. Phys. Commun. 63, 365 (1991).
[19] NNDC et al., Evaluated Nuclear Structure Data File [38] L. M. Robledo, T. R. Rodríguez, and R. R. Rodríguez-
(ENSDF), 10.18139/nndc.ensdf/1845010. Guzmán, Mean field and beyond description of nuclear

062502-7
 PHYSICAL REVIEW LETTERS 134, 062502 (2025)

structure with the Gogny force: A review, J. Phys. G 46, J.-M. Régis, G. De Gregorio, and A. Gargano, Lifetimes
013001 (2019). and structures of low-lying negative-parity states of 209Po,
[39] G. Colò, L. Cao, N. Van Giai, and L. Capelli, Self-consistent Phys. Rev. C 103, 044309 (2021).
RPA calculations with Skyrme-type interactions: The [45] R. N. Bernard, L. M. Robledo, and T. R. Rodríguez, Octu-
Skyrme_rpa program, Comput. Phys. Commun. 184, 142 pole correlations in the 144Ba nucleus described with
(2013). symmetry-conserving configuration-mixing calculations,
[40] J. Heisenberg, J. Lichtenstadt, C. N. Papanicolas, and J. S. Phys. Rev. C 93, 061302(R) (2016).
McCarthy, Excitation of low lying natural parity levels in [46] L. M. Robledo and G. F. Bertsch, Global systematics of
208 octupole excitations in even-even nuclei, Phys. Rev. C 84,
Pb by inelastic electron scattering, Phys. Rev. C 25, 2292
(1982). 054302 (2011).
[41] J. Enders, P. von Brentano, J. Eberth, A. Fitzler, C. Fransen, [47] L. M. Robledo and G. F. Bertsch, Electromagnetic transition
R.-D. Herzberg, H. Kaiser, L. Käubler, P. von Neumann strengths in soft deformed nuclei, Phys. Rev. C 86, 054306
Cosel, N. Pietralla, V. Ponomarev, A. Richter, R. (2012).
Schwengner, and I. Wiedenhöver, Nuclear resonance fluo- [48] B. A. Brown, New Skyrme interaction for normal and exotic
rescence experiments on 204;206;207;208 Pb up to 6.75 MeV, nuclei, Phys. Rev. C 58, 220 (1998).
Nucl. Phys. A724, 243 (2003). [49] J. Bartel, P. Quentin, M. Brack, C. Guet, and H.-B.
[42] J. M. Yao and K. Hagino, Anharmonicity of multi– Håkansson, Towards a better parametrisation of Skyrme-
octupole-phonon excitations in 208Pb: Analysis with multi- like effective forces: A critical study of the SkM force, Nucl.
reference covariant density functional theory and sub Phys. A386, 79 (1982).
barrier fusion of 16O þ 208Pb, Phys. Rev. C 94, 011303(R) [50] P. Ring and J. Speth, Nuclear structure calculations with a
(2016). density-dependent force in 208Pb, Nucl. Phys. A235, 315
[43] B. A. Brown, Double-octupole states in 208Pb, Phys. Rev. (1974).
Lett. 85, 5300 (2000). [51] B. Sørensen, Static quadrupole moment of the collective
[44] V. Karayonchev, M. Stoyanova, G. Rainovski, J. Jolie, A. octupole vibration in 208Pb, Phys. Lett. 35B, 101 (1971).
Blazhev, M. Djongolov, A. Esmaylzadeh, C. Fransen, [52] I. Hamamoto, Nuclear octupole vibrations in the isotopes of
K. Gladnishki, L. Knafla, D. Kocheva, L. Kornwebel, Pb, Nucl. Phys. A155, 362 (1970).

062502-8