Fast, Efficient Calculations of the Two-Body Matrix Elements of the Transition Operators for Neutrinoless Double Beta Decay

TL;DR

This paper introduces a new, faster algorithm for calculating two-body matrix elements in neutrinoless double beta decay, enabling larger model space computations with significantly reduced processing time.

Contribution

The authors developed an analytical approach to radial matrix element calculation, drastically reducing computational time for shell model NMEs in double beta decay.

Findings

Achieved a 30-fold reduction in computation time.

Results agree well with existing literature.

Enabled calculations in larger model spaces (9-10 shells).

Abstract

To extract information about the neutrino properties from the study of neutrinoless double-beta (0\nu\beta\beta) decay one needs a precise computation of the nuclear matrix elements (NMEs) associated with this process. Approaches based on the Shell Model (ShM) are among the nuclear structure methods used for their computation. ShM better incorporates the nucleon correlations, but have to face the problem of the large model spaces and computational resources. The goal is to develop a new, fast algorithm and the associated computing code for efficient calculation of the two-body matrix elements (TBMEs) of the 0\nu\beta{\beta} decay transition operator, which are necessary to calculate the NMEs. This would allow us to extend the ShM calculations for double-beta decays to larger model spaces, of about 9-10 major harmonic oscillator shells. The improvement of our code consists in a faster calculation of the radial matrix elements. Their computation normally requires the numerical evaluation of two-dimensional integrals: one over the coordinate space and the other over the momentum space. By rearranging the expressions of the radial matrix elements, the integration over the coordinate space can be performed analytically, thus the computation reduces to sum up a small number of integrals over momentum. Our results for the NMEs are in a good agreement with similar results from literature, while we find a significant reduction of the computation time for TBMEs, by a factor of about 30, as compared with our previous code that uses two-dimensional integrals.