An iterative approach for the exact solution of the pairing Hamiltonian☆,☆☆
Introduction
The introduction of pairing correlation represents a great pillar in our understanding of nuclear structure [1]. Nuclear pairing is now commonly accepted as one of the dominant residual correlations in atomic nuclei. It plays a key role not only in reproducing the ground-state properties of nuclei, including binding energies, excitation spectra, and level densities, odd-even staggering effects, single-particle occupancies, electromagnetic transition rates, and moments of inertia [2], [3], [4] but also in understanding nuclear reactions [5], and particle decays [6]. The simple Bardeen-Cooper-Schrieffer (BCS) approximation of superconductors and the generalized Hartree-Fock-Bogoliubov (HFB) approach are well known in describing the pairing properties of open-shell nuclei. Unlike electrons in solids, drawbacks of applying BCS theory and HFB methods to nuclei can be pronounced since the number of valence nucleons under the influence of the pairing force is too few to be treated by such particle-number nonconservation (quasi-particle) approximations. There have been efforts trying to overcome those drawbacks by using exact diagonalization [7], [8], [9], [10], [11], [12], [13], [14], [15] and particle-number conserved seniority approaches [16], [17], [18]. Nearly exact solutions may also be obtained through quantum Monte Carlo approaches [19], [20], [21]. On the other hand, the exact solution of the standard pairing problem was firstly obtained by Richardson [22] and is now referred to as the Richardson-Gaudin method [22], [23], [24]. We deem the Richardson-Gaudin method as the most promising approach for a microscopic treatment of clustering in heavy nuclei, the description of which requires a huge model space and a proper treatment of pairing and shell effects. In contrary to the exact diagonalization, the Richardson-Gaudin method has essentially no dimension limitation. Moreover, it provides structural information for each correlated pair, which can strongly influence the decay process. This paper presented an efficient new iterative algorithm for solving the standard pairing problem within the Richardson-Gaudin method in both spherical and deformed systems.
Section snippets
Exact solution to the pairing problem
For the standard pairing model, the Hamiltonian of is given as where the sum runs over single-particle levels j with total number n, is the overall pairing interaction strength, are single-particle energies, is the number operator for valence particles in the j-th level, and () are pair creation (annihilation) operators. For a simple system with doubly-degenerate levels (i.e., Nilsson levels in deformed nuclei), one
Numerical method
The new iterative approach to the exact pairing solutions can be implemented using a simple MATHEMATICA code (see Appendix A) with the following steps.
- Step 0.
Input values for the number of levels n, the number of pairs k, pairing strength G and the initial values of , single-particle energies , dimensionality , pairing strength G.
- Step 1.
Construct symbolic polynomials and it's derivative in terms of the symbolic variables, x, ; Set .
- Step 2.
Comparison of approaches and numerical algorithms
We compare the runtimes of the Bethe-Ansatz approach (BAE), the Heine-Stieltjes polynomial approach (HS) in Refs. [29] and the present iterative approach with the single-particle energies are taken as equally spaced with MeV, for doubly-degenerate system with . Execution times are quoted for a CPU/32 GB RAM desktop computer with windows 10 and using MATHEMATICA v.11.2. The BAE approach is limited to only a few pairs. For example, for two pairs, it takes
Conclusion
In summary, a new iterative algorithm is developed for solving the standard pairing problem within the Richardson-Gaudin method. It is based on the polynomial approach, Eq. (8), and provides an efficient and robust solver for both spherical and deformed systems in super large model spaces. For that, we start with simple systems with k nucleon pairs and levels, which can be solved iteratively by including one pair and one level at each step using the Newton-Raphson algorithm with a Monte
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This work was supported by the Swedish Research Council (VR) under grant nos. 621-2012-3805, 621-2013-4323 and Natural Science Foundation of China (grant no. 12175097). Liaoning Provincial Universities Overseas Training Program (grant no. 2019GJWYB024).
References (40)
- et al.
Prog. Part. Nucl. Phys.
(2019) - et al.
Phys. Lett. B
(2001) - et al.
Nucl. Phys. A
(2015) - et al.
Phys. Lett. B
(2013) - et al.
Comput. Phys. Commun.
(2021) - et al.
Comput. Phys. Commun.
(2021) - et al.
Phys. Rev. C
(2011) J. Phys.
(1976)- et al.
Phys. Lett. B
(1998)et al.Phys. Rev. Lett.
(2002) - et al.
Phys. Rev. C
(2012)
JPS Conf. Proc.
Phys. Rev. C
Nuclear Structure, vol. 1: Single-Particle Motion
Mat.-Fys. Medd. Danske Vid. Selsk.
The Nuclear Many-Body Problem
Rev. Mod. Phys.
JPS Conf. Proc.
Phys. At. Nucl.
Phys. Rev. C
Cited by (0)
- ☆
The review of this paper was arranged by Prof. Z. Was.
- ☆☆
This paper and its associated computer program are available via the Computer Physics Communications homepage on ScienceDirect (http://www.sciencedirect.com/science/journal/00104655).