Microscopic Spin-Parity Distributions of Fission Fragments
Extracted source text · arXiv 2607.24156
\documentclass[reprint,amsmath,amssymb,aps,preprintnumbers,nofootinbib]{revtex4-2}
\usepackage{graphicx}
\usepackage{dcolumn}
\usepackage{bm}
\usepackage{braket}
\usepackage[colorlinks=true, linkcolor=blue, citecolor=blue, urlcolor=blue]{hyperref}
\usepackage{soul}
\usepackage{enumitem}
\usepackage{color}
\usepackage{xcolor}
\usepackage{array}
\usepackage{float}
\usepackage{comment}
\usepackage[normalem]{ulem}
\newcommand{\di}{i}
\newcommand{\equalcontrib}{Both authors contributed equally to this work.}
\newcommand{\axialhohfb}{{\sc AxialHOHFB }}
\newcolumntype{P}[1]{>{\centering\arraybackslash}p{#1}}
\newcommand{\modif}[1]{{\color{blue} #1}}
\newcommand{\gs}[1]{{\color{blue} #1}}
\begin{document}
\title{Microscopic Spin--Parity Distributions of Fission Fragments}
\author{Guillaume Scamps}
\thanks{\equalcontrib}
\affiliation{Universit\'e de Toulouse, CNRS/IN2P3, L2IT, Toulouse, France}
\author{Petar Marevi\'c}
\thanks{\equalcontrib}
\affiliation{University of Zagreb, Faculty of Science,
Department of Physics, HR-10000 Zagreb, Croatia}
\author{Antonio Bjel\v{c}i\'c}
\affiliation{Nuclear data and Theory Group, Nuclear and
Chemical Science Division, Lawrence Livermore National Laboratory,
California, USA 94550}
\author{Nicolas Schunck}
\affiliation{Nuclear data and Theory Group, Nuclear
and Chemical Science Division, Lawrence Livermore National Laboratory,
California, USA 94550}
\begin{abstract}
Recent microscopic studies have investigated various
features of the spin distributions of fission fragments,
but their parity distributions
remain largely unexplored.
In this Letter, we provide a complete characterization
of the spin--parity content of fission fragments within
a time-dependent Hartree–Fock–Bogoliubov framework,
performing for the first time simultaneous projections on angular momentum,
particle number, and parity.
Calculations are carried out for the thermal
neutron-induced fission of $^{239}$Pu using both
the Gogny and Skyrme energy density functionals.
We find that dynamical pair breaking during fission populates a
significant fraction of unnatural-parity states
and generates components with non-zero spin projections $K$.
The parity content is found to depend on the number
parity of the fragments, with odd-mass nuclei exhibiting
pronounced parity staggering and odd-odd nuclei favoring
negative parity.
These results show that the parity distribution of
fragments can depart significantly
from the equiprobable partition commonly assumed in
statistical de-excitation models,
with potential implications for the modeling of fragment decay.
\end{abstract}
\date{\today}
\maketitle
\textit{Introduction.} In nuclear fission, a single compound quantum system
characterized by definite angular momentum and parity evolves into two
separated fragments, each described by its own angular momentum and parity
content~\cite{krappe2012,talou2023}.
Following the renewed interest sparked by recent experimental
results~\cite{Wilson2021}, microscopic models \cite{schunck2022}
have focused on determining
the total angular momentum of the fragments, its
projection on the fission axis ($K$ quantum number),
and correlations between the
fragments~\cite{bertsch2019,bulgac2021,marevic2021,scamps2023b,Marevic2026,li2026}.
However, these approaches do not fully characterize the combined
angular momentum and parity content of the fragments,
information that is essential to model their subsequent
de-excitation and access their internal structure. Consequently, the population
of parity states in the fragments -- both positive versus negative
and natural versus unnatural -- remains unexplored within a microscopic framework.
In the absence of such predictions, state-of-the-art models of fragment decay,
based on statistical reaction theory,
typically assume a simple, equiprobable
partition between positive and
negative parities~\cite{talou2021,Kawano2023,verbeke2018,litaize2015}.
Parity is a fundamental symmetry of the nuclear wave function
under spatial inversion, taking one of two values:
$\pi = +1$ (positive parity) or $\pi = -1$ (negative parity).
This distinction, however,
does not encode any relation to the angular momentum (spin)
$J$ of the state.
A complementary classification is provided by
the notions of natural and unnatural parity,
which link parity directly to $J$.
States satisfying $\pi = (-1)^J$ (integer $J$) or $\pi = (-1)^{J-1/2}$
(half-integer $J$)
are referred to as natural parity states,
while those that do not are called unnatural parity states.
In spherical shell-model configurations,
natural parity states typically
arise from normal single-particle excitations
within a given major shell,
whereas unnatural parity states require cross-shell
or more complex multi-quasiparticle excitations.
In odd-mass nuclei, the parity of the ground
state and low-lying excitations is additionally
governed by the parity of the odd-particle
orbital. Consequently, unnatural-parity states can
emerge directly from single-particle
degrees of freedom, without
invoking collective or multi-quasiparticle
excitations~\cite{bohr1970nuclear}.
From a reaction perspective, fission can be viewed as a quantum
transition in which an initial compound state $(J_0,\pi_0)$ evolves
into a two-fragment configuration characterized by $(J_1,\pi_1)$ and $(J_2,\pi_2)$,
together with a relative orbital angular momentum $L$ carrying parity $\pi_L = (-1)^L$.
Conservation of the total angular momentum
and parity imposes the constraints
$\bm{J}_0 = \bm{J}_1 + \bm{J}_2 + \bm{L}$ and $ \pi_0 = \pi_1 \, \pi_2 \, \pi_L$.
The partitioning of these quantities
among the fragments can be understood in terms of the symmetry properties
of the fragment wave functions and their breaking during the fission dynamics.
Each broken symmetry leaves a characteristic fingerprint on the
spin--parity distributions of the emerging fragments.
For instance, a spherically-symmetric even-even fragment
would be characterized by a $J^\pi=0^+$ state.
Axial quadrupole deformations, which
break rotational symmetry, generate even-$J$ configurations.
Octupole correlations, associated with the breaking of reflection symmetry,
induce negative-parity components and allow access to odd-$J$ configurations.
Further breaking of axial symmetry causes $K$ to no longer be a good quantum number,
leading to non-zero $K$ components in the fragment spin distribution.
Such $K$-mixing can arise either
from a triaxial deformation of the fragment's
intrinsic density
or from a pair-breaking mechanism. In our approach,
as in most dynamical calculations
starting from an axial shape after saddle,
the fragments' shapes remain axially symmetric; the non-zero
$K$ components therefore originate entirely from the
dynamical breaking of pairing correlations.
While a Cooper pair formed by two time-reversed
nucleons carries $K=0 $ and positive parity,
the violent dynamics at scission and the imparted
excitation energy induce pair breaking.
This generates multi-quasiparticle configurations,
lifting the constraints of the paired vacuum
and naturally giving rise to non-zero $K$ and
unnatural-parity components in the emerging fragments,
while the total $K_{tot}$ is exactly zero.
The goal of this work is to investigate these
mechanisms microscopically, with particular emphasis on
the previously unexplored role of parity and its
connection to dynamical pair breaking. We study primary fragments
produced in the thermal neutron-induced fission of $^{239}$Pu,
using two independent, state-of-the-art microscopic models
based on the Gogny and Skyrme energy density functionals (EDFs).
Comparing their predictions
allows us to assess the model dependence of the predicted
spin--parity distributions.
\textit{Method.} Fission is simulated using
the time-dependent Hartree-Fock-Bogoliubov
(TDHFB) framework \cite{Bulgac2019b,simenel2025}
with both Gogny and Skyrme EDFs.
Gogny calculations are performed using
an in-house code
\cite{hashimoto2012,hashimoto2013,hashimoto2016,scamps2017,scamps2025}
in a hybrid basis
consisting of two-dimensional harmonic oscillator (HO)
eigenfunctions and a one-dimensional spatial mesh
along the fission axis. Skyrme calculations
are performed with the
\axialhohfb code
\cite{bjelcic2026}
in an optimized three-dimensional HO basis.
In both cases, the time evolution of the
nuclear wave function is initiated from a static
configuration beyond the saddle point
in $^{240}$Pu, modeling the thermal neutron-induced
fission of $^{239}$Pu.
The fragments are followed up to the point
of full separation, defined
either as the point where their
centers of mass are separated
by more than $30$~fm (Skyrme)
or as the point where the distance between
their skins is $6$~fm (Gogny).
Both the Gogny and the Skyrme models employ very large bases
($N_{\rm{basis}} > 5700$ states)
and have recently been successfully applied
in fission studies \cite{scamps2025,bjelcic2026}.
However, it is worth noting that
there are some important differences between
the models, including the mean-field interaction
(Gogny D1S \cite{berger1984} vs Skyrme SkM* \cite{bartel1982}),
the pairing interaction
(Gogny vs mixed pairing with a quasiparticle cutoff
\cite{dobaczewski2002}),
and the symmetry restrictions (\axialhohfb imposes
axial symmetry).
Because the two calculations
rely on independent computer
codes with different underlying assumptions and different EDFs,
their agreement provides a stringent test of the
robustness of the predicted spin–parity mechanisms.
To extract the contributions of states with well-defined parities
for a given fragment ($\pi = \pm 1$) from the global, parity-breaking TDHFB
wave function $\ket{\Psi}$, we introduce a localized parity projection operator.
Let $\hat{\Pi}_F$ be the local space reflection operator acting
exclusively on the single-particle coordinates within a spatial box defined
around the fragment $F$ (where $F \in \{H, L\}$ for the heavy or light fragment,
respectively), relative to the fragment's center of mass. The local parity
projection operator $\hat{P}_F^{\pi}$ onto a subspace with a definite fragment
parity is then defined as
\begin{equation}
\hat{P}_F^{\pi} = \frac{1}{2} \left( \hat{1} + \pi \hat{\Pi}_F \right),
\label{eq:parity_projector}
\end{equation}
where $\hat{1}$ is the identity operator. This operator satisfies the
standard idempotency and orthogonality relations, namely $(\hat{P}_F^{\pi})^2
= \hat{P}_F^{\pi}$ and $\hat{P}_F^{+} \hat{P}_F^{-} = 0$.
The probability of finding the fragment $F$ with a specific parity $\pi$ is
then obtained by evaluating the expectation value of this localized projection
operator with respect to the total wave function $\ket{\Psi}$ of the entire system.
When performed simultaneously with the local projections onto fragment particle
numbers $N$ and $Z$, the total angular momentum $J$, and its intrinsic projection $K$,
this allows us to determine the joint distribution
\begin{equation}
P_{F}(J,K,\pi,N,Z)
=
\langle \Psi |
\hat{P}^{N}_F
\hat{P}^{Z}_F
\hat{P}^{J}_{KK,F}
\hat{P}^{\pi}_F
| \Psi \rangle
\label{eq:joint_prob}
\end{equation}
for the fully separated primary fragments.
Here, $\hat{P}^{N,Z}_F$ and $\hat{P}^{J}_{K K, F}$
are, respectively, the operators projecting on good particle
numbers \cite{simenel2010,robledo2026} and angular momentum \cite{scamps2023b}
in the fragments. Together with $\hat{P}_F^\pi$, they form a set of
mutually commuting operators.
The overlaps in \eqref{eq:joint_prob} are evaluated using the
Pfaffian method \cite{robledo2009,bertsch2012,scamps2013,wimmer12}.
Note that both the angular momentum and parity projections
are performed in the fragment's frame of reference.
The distribution \eqref{eq:joint_prob} is normalized
to one in each fragment.
More details on the projection procedure
are available in the Supplemental Material \cite{SM}.
\begin{figure*}[t]
\centering
\includegraphics[width=0.99\linewidth]{Fig1.pdf}
\caption{
Panels (a)--(d): Spin--parity distributions $P(J,\pi)$ in
the light fission fragment (LFF) and
the heavy fission fragment (HFF), calculated with Gogny and Skyrme EDFs.
Red pluses (blue crosses) indicate positive (negative) parity
components. Black solid (gray dashed) lines connect states
with natural
(unnatural) parity.
Alternating gray bands highlight even-$J$ intervals to guide the eye.
Panels (e)--(f): Distribution $P(K)$ in the fragments,
calculated with Gogny and Skyrme EDFs. Black circles
correspond to the total distribution, while
green squares (purple triangles) denote natural (unnatural)
parity components.
Note that the two fragments have the same distribution
due to $K_{\rm{tot}} = 0$. Only $K \ge 0$ values are shown because
the distributions are symmetric around $K=0$.
}
\label{fig:PJK_distributions}
\end{figure*}
\textit{Results.} In Fig.~\ref{fig:PJK_distributions}(a)--(d) we
show results for the primary fragments,
which are centered around $\braket{Z_H}=52.1$
and $\braket{N_H}\!=\! 82.5$ (Gogny), and
$\braket{Z_H}=52.6$ and $\braket{N_H}=83.2 $ (Skyrme).
The spin--parity distributions are
obtained by marginalizing the full
distribution \eqref{eq:joint_prob} over $N, Z$, and $K$.
Despite differences between the models,
Gogny [panels (a) and (c)] and Skyrme [(b) and (d)]
calculations yield very similar predictions.
The spin distribution of the well-deformed ($\beta_2 \approx 0.50$)
light fragment [(a) and (b)] is well described by a
standard spin-cutoff behavior,
with an average value of $11-12\,\hbar$. Even-$J$,
positive-parity states clearly dominate, indicating only modest
reflection-symmetry breaking at scission. Concurrently,
natural-parity configurations comprise the vast majority of
populated states at lower spins.
However, in the odd-mass fragment channel, characterized by
half-integer spins, this suppression
is less pronounced and the population of unnatural-parity
states closely approaches that of natural-parity states.
In contrast, the nearly spherical ($\beta_2 \approx 0.05$) heavy fragment [(c) and (d)]
displays a significantly lower average
spin of $5-6\,\hbar$, reflecting its more compact shape at scission.
Its spin--parity landscape is more complex, exhibiting larger deviations from a simple
spin-cutoff distribution. In particular, the $0^+$ state is prominently enhanced beyond
statistical expectations. Furthermore, while natural parity generally dominates, this
trend can occasionally be inverted. For instance, in Gogny calculations,
the $7/2^+$ state is populated with a
noticeably higher probability than its natural $7/2^-$ counterpart.
Microscopically, these features can be interpreted as stemming from the
dynamical competition between preservation and breaking
of nucleon pairs.
The final fragment wave functions represent a quantum
superposition of a fully paired vacuum state,
which strictly preserves natural parity,
and various multi-quasiparticle excitations.
The non-negligible population of
unnatural-parity states, alongside local state inversions,
provides
strong microscopic
evidence of the crucial role of dynamical pair breaking during the rapid
transition to separated fragments.
To gain deeper insight into the interplay between the fragments' spin
orientation and intrinsic structure, we analyze the $K$-dependence of
the parity distributions. In Fig.~\ref{fig:PJK_distributions}(e)--(f),
we show the total $P(K)$ distribution in the fragments,
together with its decomposition
into natural- and unnatural-parity components.
Note that the fragments exhibit a non-trivial $P(K)$ distribution
despite being axially symmetric, since
$K$ is populated dynamically at scission
rather than fixed by the intrinsic shape \cite{scamps2023b}.
The two fragments share the same $P(K)$ distribution
because $K_{\rm{tot}} = 0$ and each fragment's
distribution is itself symmetric around $K = 0$.
The natural-parity states overwhelmingly dominate the $K=0$ mode,
while the unnatural-parity contribution remains strongly suppressed.
For the lowest non-zero projection ($|K|=1/2$), natural parity
still maintains a noticeable, albeit less pronounced, dominance.
However, a qualitative transition occurs as we move toward larger values
of $K$ ($|K| \ge 1$). In this regime, the natural and unnatural
parity curves converge, rendering the
two contributions essentially equiprobable.
This behavior reinforces the link between
axial symmetry breaking, spin
alignment, and pairing dynamics. While the
strongly paired $K=0$ base configurations strictly favor natural parity,
the multi-quasiparticle excitations required to build larger $|K|$ values
naturally break time-reversal pairs. Consequently, the high-$|K|$
configurations carry an equal statistical mixture of natural and unnatural
parity components, reflecting a pair-broken state.
\begin{figure*}[htbp]
\centering
\includegraphics[width=0.99\textwidth]{Fig2.pdf}
\caption{Spin--parity distributions $P(J, \pi\,|\,N, Z)$ of
several representative fragments with different neutron
and proton number parities, calculated
with the Gogny EDF.
Panels (a)--(d) show light fragments, while panels
(e)--(h) show heavy fragments.
Red pluses (blue crosses) indicate positive (negative) parity
components. Black solid (gray dashed) lines connect states
with natural
(unnatural) parity. As an example of the sensitivity
to the choice of EDF, the results for
$^{133}$Sb obtained with the Skyrme EDF are
shown in panel (g) for comparison.}
\label{fig:pu240_fragments}
\end{figure*}
Due to the particle number projection, the distribution
\eqref{eq:joint_prob} contains information
on a large number of fragments
around the most likely fragmentation. In the Gogny
calculation, for instance, independent
projections in the two fragments yield
$218$ light
fragments and $162$ heavy fragments $(N_F, Z_F)$ whose weight in the total
distribution, $w(N_F,Z_F) = \sum_{J,K,\pi} P_F(J,K,\pi,N=N_F,Z=Z_F)$
exceeds $10^{-5}$. Note, however, that since these fragments
all originate from fluctuations around a single dynamical
trajectory, they do not fully capture fragmentations
associated with substantially different
scission configurations, such as the symmetric
or highly asymmetric split.
In the next step, we focus on properties of
several representative fragments
with different neutron and proton number
parities. In Fig.~\ref{fig:pu240_fragments},
we show the spin--parity distributions $P(J,\pi\,|\,N,Z)$ of
$^{104,105}\mathrm{Mo}$ ($Z=42$) and $^{105,106}$Tc ($Z=43$) for the light
fragment [panels (a)-(d)], and $^{132,133}$Sn ($Z=50$)
and $^{133,134}$Sb ($Z=51$) for the heavy fragment
[(e)-(h)]. These distributions
are obtained by fixing the number of nucleons ($N, Z$)
to a given value and marginalizing
the distribution \eqref{eq:joint_prob} over $K$
and $\pi$. All distributions have been renormalized to one,
$\sum_{J,\pi}P(J,\pi\,|\,N,Z) = 1$.
The distributions exhibit a clear dependence on the
neutron and proton number parity.
In the even-even $^{104}$Mo (a), natural-parity states dominate,
with an enhanced population of even-spin, positive-parity states.
This behavior is consistent with a structure dominated by pairing
correlations, where pair breaking is minimal and reflection symmetry
is only weakly broken.
For odd-mass nuclides $^{105}$Mo (b) and $^{105}$Tc (c),
natural parity still dominates, but a marked staggering between
positive and negative parity states is observed. This staggering
appears to depend on whether the odd nucleon is a neutron or a proton.
In the present calculations,
odd-neutron fragments are found to favor positive-parity states,
while odd-proton fragments favor negative-parity states.
For the odd neutron around $N \simeq 62$,
this reflects the predominantly positive-parity character
of the $sdg$ ($N_{\rm{sh}} = 4$) shell.
For the odd proton around $Z \simeq 42$,
the situation is more delicate: at the large deformations
reached by the light fragment, the positive-parity
$g_{9/2}$ orbitals and the negative-parity $pf$
($N_{\rm{sh}} = 3$) orbitals lie close together near the Fermi surface,
and the calculations
favor negative-parity configurations.
Furthermore, the odd-odd nucleus $^{106}$Tc (d)
exhibits a slight dominance of negative-parity states.
This can be understood as a consequence of coupling
an odd proton and an odd neutron
with opposite intrinsic parities, which increases
the probability of producing negative-parity configurations
in the final fragment state distribution.
The situation is somewhat different for the heavy fragments,
located in the vicinity of the double shell closure.
In $^{132}$Sn (e), the $0^+$ state overwhelmingly dominates
the distribution, reflecting the nearly spherical character
of this doubly-magic nucleus. The remaining population is
concentrated in low-$J$ states, predominantly of natural parity,
with a clear preference for even-$J$ values.
Adding one proton to the doubly-magic core leads to a
spin distribution dominated by the $5/2^+$ and $7/2^+$ states
in $^{133}$Sb (g).
In a simple shell-model picture, the first proton orbitals
above the $Z=50$ shell closure are the $2d_{5/2}$ and
$1g_{7/2}$ states, both carrying positive parity~\cite{ring2004}.
Similarly, adding one neutron produces a distribution dominated
by the $7/2^-$ state in $^{133}$Sn (f),
which can be associated with the
occupation of the $2f_{7/2}$ orbital located above
the $N=82$ shell gap.
\begin{figure*}[htbp]
\centering
\includegraphics[width=0.99\textwidth]{Fig3.pdf}
\caption{
Parity distributions as a function of fragment mass number $A$,
partitioned by fragment category:
Even-Even ($Z$-even, $N$-even), Odd-Even ($Z$-odd, $N$-even),
Even-Odd ($Z$-even, $N$-odd), and Odd-Odd ($Z$-odd, $N$-odd).
For each fragment $(N,Z)$, the fraction of the yield carrying
positive (red pluses), negative (blue crosses), natural (open black circles),
and unnatural (open gray squares) parity is shown.
Only fragments with $N \in [77, 88]$, $Z \in [46, 58]$ (heavy) and
$N \in [57, 69]$, $Z \in [36, 48]$ (light) are included.
Results are obtained with the Gogny (left) and Skyrme (right) EDFs.}
\label{fig:parity_vs_mass}
\end{figure*}
These results suggest a remarkably simple interpretation
of spin distributions in this region. The heavy fragment can be
viewed as a doubly magic $^{132}$Sn core, predominantly in
its $0^+$ ground state, coupled to an additional valence nucleon
occupying a low-lying single-particle orbital. The resulting
fragment spin then directly reflects the spin
and
parity of this valence particle. Such a behavior illustrates
the strong influence of shell structure on the quantum states
populated in the fission fragments and demonstrates how simultaneous
projections on particle number, angular momentum, and parity
can serve as a proxy to access this underlying microscopic structure.
Although the Gogny and Skyrme EDFs predict the same
microscopic mechanisms governing the spin--parity distributions,
they can nonetheless
produce different single-particle structures in the fragments.
This is illustrated in Fig.~\ref{fig:pu240_fragments}(g)
for the $^{133}$Sb ($Z=51$) fragment.
In both calculations,
the $5/2^+$ state is the most strongly populated. However,
the Skyrme functional also predicts a significant population of
the $11/2^-$ state, reflecting the influence of the $1h_{11/2}$
orbital in the underlying single-particle spectrum.
To assess the global impact of the underlying
single-particle structure on the parity distributions,
Fig.~\ref{fig:parity_vs_mass} shows the probabilities of
positive- versus negative-parity, and natural- versus unnatural-parity,
configurations as a function of fragment mass.
Despite some quantitative differences,
the two EDFs provide a consistent qualitative
picture.
The largest deviation from equiprobable
partition is observed for even-even
fragments, where the probability of preserving
paired configurations is highest. For both
the heavy and the light even-even fragments,
positive and natural parity strongly dominate
at low excitation energies, in contrast to
the assumptions commonly adopted in current fission-fragment decay models.
For odd-even, even-odd, and odd-odd fragments,
the deviation from equiprobable partition
is much more pronounced in heavy fragments.
For odd-even and even-odd fragments, in particular,
the difference between natural
and unnatural parity is rather modest.
In these cases,
the parity is largely determined by the unpaired nucleon.
For the heavy proton-odd and light neutron-odd fragments,
the unpaired nucleon most frequently
carries positive parity,
whereas for the heavy neutron-odd and light proton-odd fragments
it more frequently carries negative parity.
These effects are less pronounced with the Skyrme than with the
Gogny EDF, most likely because of the differences in predicted
single-particle spectra.
In particular, the enhanced population of negative-parity orbitals,
such as the $1h_{11/2}$ shell illustrated in
Fig.~\ref{fig:pu240_fragments}(g) for $^{133}$Sb,
increases the probability of negative-parity states
for the heavy odd-proton states.
Finally, as a consequence of the unpaired neutron and proton having opposite parities,
odd-odd fragments exhibit an enhanced population of negative-parity states,
again more pronounced in the heavy fragments and
particularly in the Gogny case.
\textit{Conclusion.}
Using simultaneous projections on the fragment
quantum numbers $J$, $K$, $\pi$, $N$, and $Z$, we have provided
a fully microscopic characterization of the spin--parity
content of fission fragments
in the thermal neutron-induced
fission of $^{239}$Pu.
These predictions are consistent between two independent,
state-of-the-art models,
based on Gogny and Skyrme functionals.
The light, well-deformed fragment follows
a standard spin-cutoff behavior dominated
by collective effects, while the nearly spherical heavy
fragment is strongly influenced by shell structure.
Dynamical pair
breaking is found to drive both the population of
unnatural-parity states and the generation
of non-zero
$K$ components, with high-$|K|$ configurations
tending toward an equal mixture of natural and
unnatural parity.
Parity distributions are also sensitive to the fragment's number parity,
with odd-mass and odd-odd fragments showing
distinct parity patterns tied to
the orbitals of the unpaired nucleons.
These results show that the parity content of
fission fragments at low
energy may depart significantly from the
equiprobable partition commonly assumed
in statistical de-excitation models.
The present study is based on a single TDHFB trajectory
corresponding to the most probable fission path.
While this is sufficient to establish the qualitative mechanisms
governing the fragment spin--parity distributions,
a quantitative description of the full range of fragments
will require considering an ensemble of fission trajectories,
e.g., within the
time-dependent generator coordinate method \cite{verriere2020,marevic2023,li2023}.
Future work should also investigate the dependence of the
spin-parity distributions on excitation energy, as well as how
incorporating them into de-excitation
simulations affects predictions
of observables
such as photon multiplicities and spectra.
\textit{Acknowledgements.}
We gratefully acknowledge support from the CNRS/IN2P3
supercomputer Center (Lyon, France) for providing
computing and data-processing resources needed for this work.
This work was also granted access to the HPC resources
of IDRIS and CINES under Allocation No. 2024-AD010515531R1 made by GENCI.
The work of P.~M. was funded by the European Union’s
Horizon Europe research and innovation program under the
Marie Sk\l{}odowska-Curie Actions Grant Agreement No.~101149053.
Support for this work was partly provided through Scientific Discovery
through Advanced Computing (SciDAC) program funded by U.S. Department of
Energy, Office of Science, Advanced Scientific Computing Research and
Nuclear Physics. This work was partly performed under the auspices of
the US Department of Energy by the Lawrence Livermore National Laboratory under
Contract DE-AC52-07NA27344. Computing support for this work came from the Lawrence
Livermore National Laboratory Institutional Computing Grand Challenge program.
\bibliography{bibliography}
\end{document}