Constraining Neutrinoless Double-Beta Decay Matrix Elements
from Ab Initio Nuclear Theory
A. Belley,1, 2, a) T. Miyagi,3, 4 S. R. Stroberg,5, b) and J. D. Holt1, 6
1)TRIUMF, 4004 Wesbrook Mall, Vancouver BC V6T 2A3, Canada
2)Department of Physics & Astronomy, University of British Columbia, Vancouver, British Columbia V6T 1Z1, Canada
3)Technische Universität Darmstadt, Department of Physics, 64289 Darmstadt, Germany
4)ExtreMe Matter Institute EMMI, GSI Helmholtzzentrum für Schwerionenforschung GmbH, 64291 Darmstadt, Germany
5)Physics Division, Argonne National Laboratory, Lemont, IL, 60439, USA
6)Department of Physics, McGill University, Montréal, QC H3A 2T8, Canada
a)Corresponding author: abelley@triumf.ca
b)Current address: Department of Physics and Astronomy, University of Notre Dame, Notre Dame, IN, 46556, USA
Abstract. As experimental searches for neutrinoless double-beta (0νβ β ) decay are entering a new generation, with hopes to
completely probe the inverted mass hierarchy, the need for reliable nuclear matrix elements, which govern the rate of this decay,
is stronger than ever. Since a large discrepancy in results is typically found with nuclear models [1], a large unknown still exists
on the sensitivity of these experiments to the effective neutrino mass. We consider this problem from a first-principles perspective,
using the ab initio valence-space in medium similarity renormalization group. In particular, we study correlations of the 0νβ β -
decay matrix elements in 76Ge with other observables, such as the double Gamow-Teller giant resonance, from 34 input chiral
interactions in an attempt to constrain our uncertainties and investigate the interaction dependence of the nuclear matrix element.
INTRODUCTION
The observation of neutrinoless double-beta (0νβ β ) decay, a hypothetical process in which two neutrons transform
into two protons without emitting an antineutrino, would show that lepton number is not a conserved quantity, thus
having important implications for the matter-antimatter asymmetry puzzle [2]. As the nuclear matrix element (NME)
of this decay is intrinsically related to its half-life, a reliable way of obtaining the NME of this transition with asso-
ciated theoretical uncertainty is of crucial importance in order to compare experimental limits in different isotopes,
pinpoint the mechanism responsible for lepton number violation, and extract key quantities such as the absolute mass
scale of the neutrino upon potential observation of the process [1, 3, 4]. Ab initio nuclear theory offers an unpar-
alleled tool to tackle the challenging task of computing the nuclear matrix elements, as it is based on systematically
improvable methods that allow for rigorous uncertainty quantification [5]. In these first-principle methods, the nuclear
Hamiltonian is constructed from two- (NN) and three-nucleon (3N) forces obtained from chiral effective field theory
(EFT) and the time-independent Schrödinger equation is then solved using nonperturbative many-body methods.
Chiral EFT offers a systematic expansion of nuclear and electroweak forces involved in this decay, while encoding
neglected higher-energy information in contact terms, whose low-energy constants (LECs) are fit to reproduce chosen
experimental data [6, 7]. In this work, the valence-space in-medium similarity renormalization group (VS-IMSRG) [8,
9, 10, 11] is used to approximately solve the Schrödinger equation and obtain NMEs from different starting chiral
EFT potentials. This method decouples an effective valence-space Hamiltonian from the full space via continuous
unitary transformations, therefore (in the absence of many-body truncations) preserving the eigenstates of the original
Hamiltonian, while reducing significantly the size of the space considered. The valence-space Hamiltonian is then
diagonalized using the shell-model code KSHELL [12], thereby extending the reach of ab initio calculations to that
of the traditional shell model [13]. Operators are then evolved using the same unitary transformations, allowing them
to be treated consistently with the Hamiltonian. To make the problem computationally tractable, all operators are
truncated at the normal-ordered two-body level, introducing the primary many-body approximation, IMSRG(2), in
the NMEs. We further restrict the size of the initial single-particle space following the emax truncation, which limits
the possible state to those having e=2n+l≤emax where nis the principal quantum number and lis the orbital angular
momentum. Finally, 3N forces are truncated via e1+e2+e3≤E3max to satisfy computational memory limitations.
Until recently, the E3max truncation proved to be the key bottleneck for ab initio methods to reach the heavy isotopes
of experimental relevance for 0νβ β . Fortunately, recent advances [14] have overcome this limitation, allowing for
the first ab initio NMEs for 130Te and 136Xe [15], two of the most predominant candidates for experimental searches,
joining previous calculations for 48Ca [16, 17, 18], as well as 76Ge, and 82Se [17].
arXiv:2210.05809v1 [nucl-th] 11 Oct 2022