1,721,286 research outputs found
Many-body calculations with deuteron based single-particle bases and their associated natural orbits
We use the recently introduced single-particle states obtained from localized deuteron wavefunctions
as a basis for nuclear many-body calculations. We show that energies can be
substantially lowered if the natural orbits (NOs) obtained from this basis are used. We use this
modified basis for 10B, 16O and 24Mg employing the bare NNLOopt nucleon–nucleon interaction.
The lowering of the energies increases with the mass. Although in principle NOs require a full
scale preliminary many-body calculation, we found that an approximate preliminary many-body
calculation, with a marginal increase in the computational cost, is sufficient. The use of natural
orbits based on an harmonic oscillator basis leads to a much smaller lowering of the energies for
a comparable computational cost
A new single-particle basis for nuclear many-body calculations
Predominantly, harmonic oscillator single-particle wave functions are the preferred choice for a basis in ab initio nuclear many-body calculations. These wave-functions, although very convenient in order to evaluate the matrix elements of the interaction in the laboratory frame, have too fast a fall-off at large distances. In the past, as an alternative to the harmonic oscillator, other single-particle wave functions have been proposed. In this work, we propose a new single-particle basis, directly linked to nucleon-nucleon interaction. This new basis is orthonormal and complete, has the proper asymptotic behavior at large distances and does not contain the continuum which would pose severe convergence problems in nuclear many body calculations. We consider the newly proposed NNLO-opt nucleon-nucleon interaction, without any renormalization. We show that, unlike other bases, this single-particle representation has a computational cost similar to the harmonic oscillator basis with the same space truncation and it gives lower energies for 6He and 6Li
A shell model calculation for in the full fp space
We discuss a shell model calculation for Fe-52 in the full fp space using the GXPF1A interaction. Several energy levels for the same angular momentum are obtained. The results for the energy levels and transition rates are compared with previous calculations. Using collective models an attempt is made to classify the spectrum into band
Ab initio calculations of energies of light nuclei with the Hybrid Multideterminant scheme
We use the AV8' nucleon-nucleon potential renormalized with the Lee-Suzuki prescription with the Hybrid Multideterminant scheme
to evaluate energies of some light nuclei. The Lee-Suzuki prescription is used to evaluate the the two-body matrix elements up
to 6 major oscillator shells in the lab frame. The Hybrid Multideterminant
scheme, is used to deal with the nuclear structure problem.
The results obtained for , and are compared
with the results obtained with other methods.
The results suggest reasonable convergence of the renormalization prescription
for major shells
About the sign ambiguity in the evaluation of grand canonical traces for quasi-particle statistical density operators
A simple and general prescription for evaluating unambiguously the sign of the grand-canonical trace of quasi-particle statistical density operators (the so-called sign ambiguity in taking the square root of determinants) is given. Sign ambiguities of this kind appear in the evaluation of the grand-canonical partition function projected to good quantum numbers (angular momentum, parity and particle number) in the Hartree-Fock-Bogoliubov approximation at finite temperature, since traces
are usually expressed as the square root of determinants.
A comparison is made with the numerical continuity method
Ab-initio calculation of the binding energy with the hybrid multideterminant scheme
We perform an ab-initio calculation for the binding energy of using the CD-Bonn 2000 NN potential renormalized with the
Lee-Suzuki method. The many-body approach to the problem is the Hybrid Multideterminant method. The results indicate a binding energy
of about , within a few hundreds KeV uncertainty. The center of mass diagnostics are also discussed
A calculation of the position of the quasi-beta and quasi-gamma bands for the transitional nucleus Dy-154 with Monte Carlo methods
A calculation of the excitation energy of the 0(+) states and of the 2(+) states is performed using Monte Carlo methods for the nucleus Dy-154. The Hamiltonian is assumed to be a monopole+quadrupole pairing+quadrupole with the parameters fixed by the spectroscopic Monte Carlo method so as to reproduce the experimental excitation energies of the yrast states up to J = 8 within the 50-82 and 82-126 proton and neutron major shells. The resulting Hamiltonian has been diagonalized in the J = 0 and J = 2 subspaces using the quantum Monte Carlo method. The size of the basis is fixed by comparing the yrast energies obtained with the basis -independent spectroscopic Monte Carlo method, and those obtained with the quantum Monte Carlo method. The excitation energy of the 0(2)(+) is much higher than the experimental value. The structure of the 0(2,3)(+) and of the 2(2,3)(+) eigenstates is discussed in terms of fluctuating intrinsic states and resolved in terms of the deformation variables
A model study of short range correlations with a multi determinantal method
Using a simple model of fermions interacting with a strong short range repulsive potential, we discuss how short range correlations can be described with a linear combination of Slater determinants using variational methods. The many-body wavefunction obtained in this way is used to evaluate the two-particle correlation function which shows the typical depletion at distances between particles comparable with the range of the repulsive potential
Use of multi-determinant wave functions as input for standard shell model calculations
We show that the use of wavefunctions obtained with the hybrid multi-determinant method for both the yrast states and the excited states of a given Jπ, as an input for the standard shell model diagonalization method, can greatly reduce the computational cost of large-scale shell model calculations. Few examples are given using the GXPF1A interaction in the fp region
About efficient quasi-newtonian schemes for variational calculations in nuclear structure
The Broyden-Fletcher-Goldhaber-Shanno (BFGS) quasi-newtonian scheme is known as the most efficient scheme for variational calculations
of energies. This scheme is actually a member of a one-parameter family of variational methods, known as the Broyden -family.
In some applications to light nuclei
using microscopically derived effective hamiltonians starting from accurate nucleon-nucleon potentials, we actually found other members of the same family which have better performance than the BFGS method. We also extend the Broyden -family of algorithms to a two-parameter family of rank three updates which has even better performances
- …
