Computational aspects of nuclear coupled-cluster theory

Abstract

Abstract. Coupled-cluster (CC) theory represents an important theoretical tool that we use to solve the quantum many-body problem. CC theory also lends itself to computation in a parallel computing environment. In this paper, we present selected results from ab initio studies of stable and weakly bound nuclei utilizing computational techniques that we employ to solve CC theory. We also outline several perspectives for future research directions in this area.

Coupled-cluster computations of atomic nuclei

“…Rather, one uses iterative procedures to obtain the solution. In nuclear physics, Dean et al (2008) and Baran et al (2008) explored different Krylov sub-space methods to increase the convergence rate of iterative methods. They found that the Broyden method (Broyden, 1965) or DIIS (Direct Inversion in the Iterative Subspace) (Pulay, 1980) achieve convergence of the CCSD equations already with 20-30 iterations.…”

“…The inset shows the decay widths due to neutron emission. Adapted from (Dean et al, 2008). Bacca et al (2009) studied the cutoff-dependence for low-momentum N N interactions based on chiral EFT potentials in neutron-rich isotopes of helium and performed benchmarks between the hyperspherical harmonics method and the coupledcluster method in the CCSD and Λ−CCSD(T) approximations for 4 He.…”

Computational Science and Innovation

“…A similar story can be told for the NCSM technique [26] where applied mathematicians worked with domain scientists to achieve scale up to the largest computer systems available. Coupled-cluster techniques are also benefiting from similar interactions [27].…”