# A Microscopic Cluster Description of .

###### Abstract

We investigate both bound and resonance states in embedded in a three--cluster continuum using a three-cluster microscopic model. The model relies on the Hyperspherical Harmonics basis to enumerate the channels describing the three-cluster discrete and continuous spectrum states. It yields the most probable distribution of the three -clusters in space, and the dominant decay modes of the three-cluster resonances.

## 1 Introduction

The nucleus is an interesting example of the so-called Borromean nuclei, as it has no bound states in any two-cluster subsystem of its three-cluster configuration. The lowest dissociation threshold (7.276 MeV above the ground state) is that of a three particles disintegration. This three-cluster configuration is thus responsible to a great extent for the formation of a few bound, and many resonance states. The next threshold is of a two-cluster nature: [1]. It opens when the excitation energy of exceeds 15.96 MeV. One therefore expects only a negligible influence of the latter channel on the bound and resonance states of in the vicinity of the threshold.

The nucleus is unique because of its excited “Hoyle state”. This state is important in the context of the nucleosynthesis of carbon in helium-burning red giant stars. It is a state with an energy of 7.65 MeV above the ground state, or 0.4 MeV above the three-cluster threshold. Its width is only 8.5 eV, indicating a long lifetime. One immediately relates this to the state in described by two particles, with an energy of 0.092 MeV above the threshold, and a width of 5.57 eV.

Many efforts have been made to reproduce the experimentally observed structure of , and to explore and understand the nature of the ground, excited and resonance states. This was e.g. done within so-called semi-microscopic models (considering structureless -particles) [2, 3, 4, 5, 6, 7, 8] and within fully microscopic models [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24].

A rather general feature of the calculations is that, with potentials which adequately reproduce the - interaction (this includes the phase shifts for , and states, and the position of the corresponding resonance states), one obtains a noticeably overbound ground state for .

To determine the energies and widths of the resonance states created by a three-cluster continuum, only a few methods can be used. One popular method for obtaining the resonance properties in many-cluster, many channel systems is the Complex Scaling Method (see reviews [25, 26] and references therein). Other methods start from a calculated form of the -matrix in a wide energy range, and determine the resonance states as the pole(s) of the -matrix. The advantage of these methods is that they provide the scattering quantities (such as phase-shifts, cross-sections, …) and the resonance properties (energies and widths), as well as the wave functions of scattering and resonance states. The latter then allow one to obtain more information about the nature of the resonance states.

is known from theory and experiment (see, e.g., [32] and [33]) to have some very narrow resonances above the three threshold. One may wonder why a system with several open channels does not decay instantly, but manifests these narrow resonance states. There are two possible answers to this question. First, a resonance state appears in one single channel of the multi-channel system. Such particular channel is usually weakly coupled to a number, or all, of the other open channels. It is well-known that this weak coupling of channels predetermines the existence of long-lived resonance states. Second, a resonance can be more or less uniformly distributed over all open channels, and the compound system needs (some) time for the resonance to be accumulated by one or a few number of open channels in order to decay into. Such a distribution over many open channels leads to very narrow resonances, as was predicted by A. Baz’ [31]. It is referred to as diffusion-like processes in scattering. This type of resonance is attributed to the effect that “the system spends most of its time wandering from one channel to another” [31].

In this paper we wish to calculate and analyze the bound and continuum structure of , and gain some insight in the nature of these states. Indeed, in some publications (e.g. [27, 28, 29, 30]) the suggestion for a dominant linear, chain-like, three-cluster structure appears for some of the resonances. We will look for confirmation of this structure. To this end, we determine the most probable configuration of the three particles both in coordinate and momentum space. We also qualify those channels on which the resonance states of preferentially decay.

The main results of this paper are obtained by applying the “Algebraic Model in a Hyperspherical Harmonics Basis” (AMHHB) [34, 35, 36] on a configuration of three -particles. In this model the three clusters are treated equally, and their relative motion descrybed by Hyperspherical Harmonics. The latter enumerate the channels of the three-cluster continuum and allow to implement the correct boundary conditions for the three-cluster exit channels. The AMHHB has been applied successfully to study resonances in nuclei with a large excess of protons or neutrons such as , , . The method provides the energies and widths of the resonances, and their total and partial widths, as well as the corresponding wave functions. The latter allow to analyze the nature of the resonance states. The results of this model are compared to those obtained in other, more or less comparable, microscopic descriptions from the literature, and to experiment.

In the next section we elaborate on the method used to calculate the spectrum of . Section three focuses on the results obtained using this method. We also present correlation functions and density functions to characterize more precisely the spatial configuration of the three particles for specific resonance states. We also compare the results to those of other microscopic calculations as well as to experiment.

## 2 The microscopic cluster model

In this section we describe the microscopic model used to determine the structure of in the present paper. As it has already been introduced and used in several publications, we will limit ourselves to the most important notations and aspects of importance to the current calculations.

### 2.1 The three-cluster AMHHB model

The three-cluster “Algebraic Model in a Hyperspherical Harmonics Basis” (AMHHB) [34, 35, 37, 36] will be applied to a single three-cluster configuration.

This model takes a Hyperspherical Harmonics basis (HH) to characterize and enumerate the different three-cluster channels. In each of these channels an oscillator basis describes the radial behavior, and is used to expand the many-particle wave function. A matrix version of the Schrödinger equation is obtained after substitution of this wave function. It solved by the Algebraic Method (also called the Modified J-Matrix method [36]) for both bound and scattering states using the correct asymptotics.

A similar approach, using the Hyperspherical Harmonics, was proposed in [39, 40] in coordinate representation, using the generator coordinate technique to solve the corresponding Schrödinger equation.

The AMHHB wave function for is written as

where is a cluster oscillator function [34]:

(2) | |||

These functions are enumerated by the number of hyperradial excitations , hyperspherical momentum and two partial orbital momenta . The vectors and form a set of Jacobi coordinates, and and are hyperspherical coordinates, related to the Jacobi vectors by:

(3) |

The notation and refers to unit length vectors. Vector corresponds to the distance between two selected particles, with an associated partial orbital angular momentum . Vector is the displacement of the third particle with respect to the center of mass of the other two, with an associated angular momentum . The three quantum numbers determine the channels of the three-cluster system in the AMHHB.

The fact that all three clusters are identical leads to some specific issues. The wave function (2.1) for is antisymmetric with respect to the permutation of any pair of nucleons. Because the three clusters are identical, this function should be symmetric with respect to the permutation of any pair of alpha particles. This imposes constraints on the allowed quantum numbers of the wave function. Due to this symmetry, for instance, the partial orbital momentum of a two-cluster subsystem can only have even values. As the parity of states is defined as , it is fully determined by the partial orbital angular momentum of the relative motion of the remaining cluster with respect to the two-cluster subsystem.

In [41] and [12] it was suggested to use a symmetrization operator to construct the proper basis states. For a discussion on the symmetry of a system with three identical clusters we refer to [42].

The symmetrical Hyperspherical Harmonics basis for a three-particle system was realized many years ago (see, e.g., [43, 44]). An explicit form of a few basis functions for small values of the total angular momentum ( and 2) can be derived. However it is extremely intricate to use for explicit calculation of matrix elements.

An alternative approach to obtain such matrix elements without an explicit realization of the basis functions consists in using the generating function technique. One can indeed construct a generating function for the overlap and hamiltonian kernels of , using the procedure explained in [34], that satisfies all required symmetry conditions, including the cluster symmetric permutation behavior. Explicit matrix elements of the operators can then be obtained by using recurrence relations. The standard approach in the AMHHB is to extract matrix elements characterized by explicit quantum numbers. These, however, do not yet correspond to the desired symmetrical Harmonics. Indeed, the states for fixed and do not belong to the desired symmetrical irreducible representation of S(3), the permutation group of the three clusters, with Young tableau [3]. They are, in fact, linear combinations of the Young tableau [3] and the non-symmetrical Young tableaus [2,1] and [111].

The antisymmetrization operator in the standard AMHHB basis has non-zero matrix elements

(4) |

for fixed oscillator shells with =. By selecting only the matrix elements with hyperradial quantumnumber and hypermomentum , one obtains relatively small matrices whose eigenfunctions with non-zero eigenvalues are of the correct symmetrical Hyperspherical type, due to the symmetry properties of the generating function. This procedure is similar to the procedure of obtaining the Pauli allowed states in three-cluster systems (for details see [45]).

This is demonstrated in Figure 1 where the diagonal matrix elements of the antisymmetrization operator between the original Hyperspherical Harmonics are displayed for total angular momentum , for all channels up to . One notices that the matrix elements

(5) |

do not tend to unity, as one could expect, but to some fixed values. Analysis shows that these asymptotic values of (5) correspond to the weights of the symmetrized Hyperspherical Harmonics with Young tableau [3], within the original Harmonic.

The eigenvalues obtained after diagonalization however, which are matrix elemts of symmetrized Harmonics, do display the correct asymptotic behavior, i.e. all tend to unity, as can be seen in Figure 2.

In Table 1 we display both the total number of (original) nonsymmetrized and of symmetrized channels for different values of the total orbital momentum. The symmetrization significantly reduces the number of channels compatible with the maximal value of Hypermomentum . Only even values of the partial orbital momentum are considered because of the symmetry rules for two-cluster subsystems.

14 | 14 | 14 | 13 | 13 | |

() | 20 | 44 | 54 | 28 | 42 |

() | 8 | 16 | 19 | 9 | 14 |

In [34] and [35] the importance and meaning of the effective charges were defined in the context of the AMHHB, and explicitly discussed for the nucleus in three-cluster configuration . The effective charge determines the asymptotic form of the three-cluster potential originating from the Coulomb interaction, which has the form

(6) |

It was shown that it is of crucial importance for implementing the correct boundary conditions for the three-cluster continuum states.

The symmetrization influences the behavior of the effective charges. In Table 2 we display the effective charges for the state of , calculated in the original, nonsymmetrized, basis of the Hyperspherical Harmonics, for . One easily verifies that they coincide with those calculated in [23].

(0,0,0) | (4,0,0) | (4,2,2) | (6,0,0) | (6,2,2) | (8,0,0) | (8,2,2) | (8,4,4) | |
---|---|---|---|---|---|---|---|---|

(0,0,0) | 28.81 | 2.47 | 3.49 | 2.74 | -2.74 | 0.87 | 0.00 | 1.04 |

(4,0,0) | 2.47 | 32.157 | -1.13 | 3.95 | -0.31 | 4.67 | 0.00 | 1.95 |

(4,2,2) | 3.49 | -1.13 | 31.35 | 1.72 | -4.30 | 0.00 | 0.66 | 0.00 |

(6,0,0) | 2.74 | 3.95 | 1.72 | 33.48 | -2.51 | 4.63 | 0.00 | 0.45 |

(6,2,2) | -2.74 | -0.31 | -4.30 | -2.51 | 34.29 | 0.00 | 0.62 | 0.00 |

(8,0,0) | 0.87 | 4.67 | 0.00 | 4.63 | 0.00 | 34.29 | 0.00 | -2.38 |

(8,2,2) | 0.00 | 0.00 | 0.66 | 0.00 | 0.62 | 0.00 | 33.08 | 0.00 |

(8,4,4) | 1.04 | 1.95 | 0.00 | 0.45 | 0.00 | -2.38 | 0.00 | 32.41 |

In Table 3 we display the effective charges in the symmetrized basis. Only four channels remain after symmetrization. In particular no channel remains, so we ommitted these also in Table 2 even though they have a non-zero contribution.

(0,1) | (4,1) | (6,1) | (8,1) | |
---|---|---|---|---|

(0,1) | 28.810 | 4.277 | 3.880 | 1.139 |

(4,1) | 4.277 | 30.556 | 5.217 | 2.301 |

(6,1) | 3.880 | 5.217 | 35.990 | 1.457 |

(8,1) | 1.139 | 2.301 | 1.457 | 31.450 |

It goes without saying that the asymptotic form of the effective three-cluster potential which originates from the nucleon-nucleon interaction [34]

(7) |

is also influenced by the symmetrization. This asymptotic component is very important for obtaining the correct values of the matrix. We do not dwell on its explicit form here, but apply a procedure similar to that for the effective charges.

### 2.2 Phases, Eigenphases and Resonances

After solving the system of linear equation of the AMHHB model, we obtain the wave functions of the continuous spectrum states, and the scattering -matrix. We consider two different representations of the -matrix.

In the first representation, the elements of the -matrix are described through the phase shifts and the inelastic parameters :

(8) |

of which one usually only analyzes the diagonal matrix elements by displaying the and quantities. In the second representation the -matrix is reduced to diagonal form, leading to the so-called eigenphases, which now represent the elastic scattering of the many-channel system in independent (uncoupled) eigenchannels:

(9) |

Here is an orthogonal matrix, connecting both representations, and is a diagonal matrix with nonzero elements

(10) |

defining the eigenphases .

The phases shifts , inelastic parameters and eigenphases then provide sufficiently detailed information about the channels that are involved in the production of resonance states. The eigenphases are used to extract the resonance positions and total widths in the traditional way

(11) |

whereas the orthogonal matrix leads to the partial decay widths of the resonance (for details see, e.g., [36]).

### 2.3 Correlation functions and density distributions.

As we pointed out, the AMHHB model allows to calculate the scattering properties, but also to obtain the obtain wave function at any energy, in particular at the resonance positions. The latter is of the utmost importance to analyze the nature of the system at these energies.

Within the AMHHB model the solution is fully expressed by the expansion coefficients and the -matrix. The expansion coefficients determine both the total three-cluster wave function of a compound system , as well as the wave function of the relative motion of three clusters (see eq. (2.1)).

The latter contains all information on the dynamic behavior of the three-cluster system for bound as well as continuum states. It is interesting to note that these coefficients are identical in both the representations of the wave function in coordinate and momentum space, because of the Fourier transform properties of the oscillator states. The wave function in momentum space has arguments that are directly related to the coordinate representation: is the momentum of relative motion of two clusters, whereas is the momentum of the third cluster with respect to the center of mass of the two-cluster subsystem.

We obtain the density distribution in coordinate space as

(12) |

and the corresponding correlation function as

(13) |

directly from the wave function of relative motion . Both the density distribution and correlation function in momentum space are obtained in the same way using the wave function of relative motion in momentum space .

In a calculation with open channels, one obtains independent wave functions describing the elastic and inelastic processes in the many-channel system. It is quite impossible to analyze all of these wave functions when many channels are open. Some principles have to be set up on how to select the most important wave functions. In [36] we formulated some criteria for selecting the dominant wave function of a resonance. We will use the same criteria in this paper to select the “resonance wave functions”.

## 3 Calculations and results

In the present calculations for we consider for the nucleon-nucleon interaction the Minnesota potential [46]. The oscillator basis is characterized by an oscillator length fm, to minimize the ground state energy of the particle using the above potential.

Parameter of the Minnesota potential is taken to be in order to reproduce the phase shifts for scattering, and the , and resonances in . The same parameters were used by Arai [17].

The two-cluster substructure is of key importance in the description of . We present resonance properties in Table 4.

AMOB | Arai [17] | |||
---|---|---|---|---|

, MeV | , keV | , MeV | , MeV | |

0.022 | 6.30 10 | 0.03 | 10 | |

2.93 | 1.51 | 2.9 | 1.4 | |

12.55 | 5.01 | 12.5 | 4.8 |

The AMOB model takes a set of oscillator functions to describe the intercluster behavior, and the Algebraic Model to obtain the phase shifts for scattering (see e.g. [47]). We include a comparison to the results of of Arai in his paper on [17], where he uses the “analytical continuation of the matrix to the complex plane” method with Complex Scaling to obtain the resonance characteristics.

These results form a first test of the consistency of the different expansion methods used, applied to the two-cluster subsystem. Although quite similar, one still notices that the resonance properties of the two-cluster system have a slight dependence on the method used.

### 3.1 The Potential and Coulomb interaction in AMHHB

In Figures 3 and 4 the diagonal matrix elements of the nucleon-nucleon and Coulomb interactions within the AMHHB model are displayed, again for channels up to . One observes that the nucleon-nucleon interaction creates a deep potential well with a long tail in the Hyperspherical coordinate. This tail reflects the asymptotic form of the potential, indicated in (7). The matrix elements of the Coulomb interaction indicate the magnitude of the Coulomb barrier, which is the main factor for generating the resonance states in .

### 3.2 Phase shifts and eigenphases

In Figure 5 we show results of the AMHHB calculations for the state in terms of the symmetrical Hyperspherical Harmonic channels through the (diagonal) phase shifts and the inelastic parameters .

The scattering parameters are obtained from a calculation with maximal Hypermomentum . One observes from Figure 5 that for small energies the channels are totally uncoupled (). A first resonance appears at MeV, and is mainly produced in the first channel with Hypermomentum , whereas a second resonance at energy MeV is dominated by Hypermomentum . The inelastic parameters for the first two channels have a pronounced minimum at the energy of the first resonance, and a shallow minimum at the second resonance energy. Also, the first resonance displays a “shadow resonance” behavior in the second channel. This is a typical behavior for resonances in a many-channel system (see, for instance, the detailed analysis of two-channel resonances in in [48]). The minimum in the inelastic parameters indicates that the compound system is being reconstructed at this energy, and transits from one channel to another.

In Figure 6 we display the corresponding eigenphase shifts for the first three eigenchannels. One observes now that both resonance states are mainly associated with the first eigenchannel, and that the second eigenchannel only contributes marginally.

### 3.3 Convergence properties

A convergence study of the energies (and widths) for bound and resonance states should indicate whether the Hilbert space is sufficiently large for stable and reliable results. The AMHHB model space is characterized by two parameters: the maximal value of Hypermomentum , and the maximal value of the Hyperradial excitation . Usually the choice is a compromise between the convergence of the results and the computational burden. A set of Hyperspherical Harmonics with for even parity states, and for odd parity states, seems sufficient and remains computationally feasible. This choice accounts for a large number of three-cluster configurations or, in other words, for a sufficient number of inherent (triangular) shapes for the three clusters. We refer to [38] for examples of most probable triangular shapes for the Hyperspherical Harmonics from to .

A first convergence test considers the , and bound states of , shown in Figure 7 as a function of .

One observes that the deeply bound states (, ) require significantly less Hyperspherical Harmonics for a converged energy than the shallow, or weakly bound, state with . At least all Hyperspherical Harmonics with are required to bind the latter state, whereas for one already obtains binding with a single Hyperspherical Harmonic with . Figure 7 further demonstrates that the above choice of amply leads to sufficient precision for the bound states.

In Table 5 we turn to the energies and widths of the and resonances obtained with increasing number of Hyperspherical Harmonics. One observes that sufficient convergence of the resonances occurs at . It is furthermore interesting to note that these resonances already appear with reasonable energy and width values when only the lowest channel ( for the , and for the state) is considered. This is a remarkable result for , as e.g. for it was impossible to generate a resonance with a single channel (see [35]).

0 | 4 | 6 | 8 | 10 | 12 | 14 | ||
---|---|---|---|---|---|---|---|---|

0.40 | 0.75 | 0.74 | 0.72 | 0.70 | 0.68 | 0.68 | ||

205.08 | 13.40 | 11.79 | 7.10 | 4.35 | 2.71 | 2.77 | ||

1.15 | 7.34 | 6.09 | 5.55 | 5.54 | 5.16 | 5.14 | ||

510.16 | 897.64 | 422.50 | 539.21 | 586.08 | 534.33 | 523.46 | ||

- | 3.28 | 2.89 | 2.83 | 2.78 | 2.74 | 2.73 | ||

- | 30.19 | 13.07 | 11.85 | 9.95 | 8.84 | 8.75 | ||

- | 3.50 | 3.27 | 3.22 | 3.17 | 3.14 | 3.11 | ||

- | 274.51 | 351.57 | 308.29 | 280.23 | 263.80 | 246.78 |

In all calculations we have considered states with hyperradial excitation up to =70, which covers a large range of intercluster distances, and reaches well into the asymptotic region.

### 3.4 Partial widths.

In Table 6 we display the energy, the total width () and the partial widths (, ) in the corresponding decay channels for the even parity resonances, and in Table 7 for the odd parity resonances.

0.68 | 2.78 | 3.17 | 5.60 | |||||

2.79 | 9.95 | 280.24 | 0.55 | |||||

2.79 | 6.11 | 13.46 | 0.23 | |||||

0 | 3.84 | 278.89 | 0.15 | |||||

0 | 10 | 10 | 0.16 |

3.52 | 0.67 | |||

0.21 | 8.34 | |||

0.206 | 8.34 | |||

0.002 | 0 | |||

10 | 0 |

One observes that in most cases only one or two channels are responsible for the decay of the resonance states. The remaining channels contribute negligibly, and the corresponding partial width does not exceed keV. Only for the resonance a significant distribution over multiple channels is apparent.

One should note that, although the resonances are created by only a few channels, the role of the other, very weakly coupled, channels is still important. This can be seen from Table 5 for the first resonance: it is indeed generated mainly by the channel with minimal Hypermomentum , but modified substantially with increasing number of Hypermomentum. The same applies to the other resonance states.

### 3.5 Correlation functions and density distributions.

In Figure 8 we show the correlation function for the ground state, and observe that this state displays a compact spatial configuration, as it is expected for such a deeply bound state. The most probable shape of the three -cluster system is an almost equilateral triangle with a distance between any two -particles of approximately 3 fm.

The correlation function for the first resonance state on the other hand, shown in Figure 9, shows a more deformed system with two particles relatively close to one another (about 3.5 fm) and the third alpha-particle further away (approximately 5 fm). So features a prolate triangle as a dominant configuration for this state.

One also observes on Figure 9 a small maximum for the correlation function corresponding to an almost linear configuration of three particles, two of them being approximately 4 fm apart, and the third 0.2 fm away from their centre of mass. However, the weight of this linear configuration is approximately 6 times less than the weight of the prolate triangular configuration. Our calculations therefore do not agree with other authors advancing a dominant linear structure [27, 28, 29, 30].

In Figure 10 we display the correlation function of the first resonance state in momentum space. One observes a huge maximum corresponding to relatively slowly moving particles. A small maximum corresponding to faster moving alpha-particles is also present.

### 3.6 Comparison to the literature

We now compare the AMHHB results to the existing literature. In Table 8 we display the AMHHB results to those of Arai [17] and Pichler et al. [9], both obtained by the Complex Scaling Method (CSM). The latter authors [9] use a somewhat different value for the parameter in the Minnesota potential, and a different oscillator length ; because of this, different results are obtained for the bound states.

Method | AMHHB | CSM-Arai | CSM-Pichler et al. | |||
---|---|---|---|---|---|---|

Reference | Present paper | [17] | [9] | |||

, MeV | , keV | , MeV | , keV | , MeV | , keV | |

0.684 | 2.71 | 0.4 | 0.64 | 14 | ||

5.156 | 534.00 | 4.7 | 1000 | 5.43 | 920 | |

2.775 | 9.95 | 2.1 | 800 | 6.39 | 1100 | |

3.170 | 280.24 | 4.9 | 900 | |||

5.603 | 7.82 | 5.1 | 2000 | |||

3.516 | 0.21 | 3.4 | 200 | 3.71 | 360 | |

0.672 | 8.34 | 0.6 | 1.16 | 25 | ||

4.348 | 2.89 | 7.1 | 5400 | 11.91 | 1690 | |

5.433 | 334.90 | 9.6 | 400 |

Comparison with the results of Arai [17] indicates that the AMHHB model leads to resonance states with higher energy and smaller widths than those obtained with the CSM. This can be attributed to the difference in the methods, and to the different Hilbert spaces. Formally the Hilbert space of basis functions used by Arai [17] is quite close to the one considered in the AMHHB. Actually, in the present calculations the partial orbital momenta and are restricted by the condition

so that, for instance, for total orbital momentum , they run from to with . Arai on the other hand, restricted himself with . In [35, 36, 37] we observed the tendency that the more Hyperspherical Harmonics (thus the more channels) are involved in the calculation, the smaller the resonance energy and width becomes. This tendency is again confirmed by the present AMHHB calculations. Thus some reduction of the width of the resonances, observed in our calculations with respect to Arai [17], can be attributed to the larger number of channels in our model.

Comparing the AMHHB results to the Complex Scaling Model calculations of Pichler et al. [9], one observes that both yield close results for the first and second resonance states.

On the whole one can conclude that there is consistency in the results for resonance properties in all three microscopic models.

### 3.7 Comparison to experiment

In Table 9 we compare the theoretical AMHHB results for to available experimental data.

Method | AMHHB | Experiment | ||

Reference | Present paper | [1] | ||

, MeV | , keV | , MeV | , keV | |

0.684 | 2.71 | |||

5.156 | 534.00 | |||

2.775 | 9.95 | |||

3.170 | 280.24 | |||

5.603 | 7.82 | |||

3.516 | 0.21 | |||

0.672 | 8.34 | |||

4.348 | 2.89 | |||

5.433 | 334.90 |

One notices that the first resonance state (the Hoyle state) appears in the current calculations as a narrow resonance with an energy of 0.684 MeV and width 2.7 keV, which is considerably wider than the experimental Hoyle state (about 8.5 keV). This contrasts with the generally observed feature of the AMHHB calculations that the calculated widths are significantly less than the corresponding experimental widths of the resonances. The discrepancies between the theoretical and experimental data have essentially two origins. The first one relates to the choice of the nucleon-nucleon interaction: it has been tuned to reproduce the phase shifts and resonance properties for alpha-alpha scattering. As a result it leads to overbound and states in , and binds the state. The second one relates to the specific choice of three-cluster model and corresponding model space, as well as to the method by which the energy and width of the resonance states are obtained.

### 3.8 Optimizing the nucleon-nucleon potential

In this paper we used a Minnesota nucleon-nucleon potential tuned to reproduce the phase shifts for scattering, as well as the resonances. This however leads to overbound and states, and a bound state. Moreover, the obtained resonance structure for the three-cluster continuum deviates from the experimentally observed one, which can also be attributed to the specific choice of semi-realistic nucleon-nucleon potential.

We therefore wish to discuss the dependence of the results to the choice of parameter on the results. To do so we use different criteria to optimize this parameter. We first determine a value to reproduce the ground state energy of , followed by an attempt to reproduce the energy and width of the Hoyle state.

In Figure 11 we display the ground state energy as a function of the parameter , compared to experiment (dashed line). One observes that the ground state is reproduced with .

One observes a monotonously decreasing linear dependence of the ground state energy on within the selected range. For the Hoyle state position and width the dependency is less trivial, as is shown in Figure 12. One however observes that

the value reproduces the position of the Hoyle state, and leads to a close match for its width too.

## 4 Conclusions

In this paper we described the nucleus with a three-cluster microscopic model.

The model correctly handles the three-cluster continuum, i.e. correctly implements the suitable boundary conditions, by using a Hyperspherical Harmonics basis. It leads to the scattering matrix in many-channel space, and the energy, total and partial widths of the resonance states and their corresponding wave functions can be obtained.

It was shown that the obtained resonances of agree well with other methods, and that the lowest resonances are generated by only a few number of weakly coupled channels, leading narrow resonance states. The partial widths determine the most probable channels for resonance decay. Correlation functions and density distributions revealed the dominant shape of the three-cluster triangle configuration for the lowest bound and resonance states of . There were no indications of a prominent linear three-cluster structure for the resonance states.

It was also shown that it is impossible to fix a unique value for the parameter of the Minnesota nucleon-nucleon potential to fit all desired physical properties for , and for the disintegrating particles. However the qualitative conclusions remained unaltered under slight adaptation of .

As a final conclusion we can state that the model is consistent with other microscopic models using the Complex Scaling methodology to determine the energy and total width of three-cluster resonance states.

## 5 Acknowledgments

Support from the Fonds voor Wetenschappelijk Onderzoek Vlaanderen (FWO), G0120-08N is gratefully acknowledged. V. S. Vasilevsky is grateful to the Department of Mathematics and Computer Science of the University of Antwerp (UA) for hospitality. This work was supported in part by the Program of Fundamental Research of the Physics and Astronomy Department of the National Academy of Sciences of Ukraine.

## References

- [1] F. Ajzenberg-Selove, “Energy levels of light nuclei = 11-12,” Nucl. Phys. A, vol. 506, pp. 1–158, Jan. 1990.
- [2] S. I. Fedotov, O. I. Kartavtsev, V. I. Kochkin, and A. V. Malykh, “3 -cluster structure of the states in and the effective - interactions,” Phys. Rev. C, vol. 70, p. 014006, July 2004.
- [3] I. Filikhin, V. M. Suslov, and B. Vlahovic, “ states of the nucleus: the Faddeev calculation in configuration space,” Journal of Physics G Nuclear Physics, vol. 31, pp. 1207–1224, Nov. 2005.
- [4] R. Álvarez-Rodrıguez, E. Garrido, A. S. Jensen, D. V. Fedorov, and H. O. U. Fynbo, “Structure of low-lying resonances,” Eur. Phys. J. A, vol. 31, pp. 303–317, Mar. 2007.
- [5] R. Álvarez-Rodríguez, A. S. Jensen, D. V. Fedorov, H. O. U. Fynbo, and E. Garrido, “Decay of low-lying resonances within a 3 cluster model,” J. Phys. Conf. Ser., vol. 111, p. 012017, May 2008.
- [6] R. Álvarez-Rodríguez, A. S. Jensen, E. Garrido, D. V. Fedorov, and H. O. U. Fynbo, “Momentum distributions of particles from decaying low-lying resonances,” Phys. Rev. C, vol. 77, p. 064305, June 2008.
- [7] M. Chernykh, H. Feldmeier, T. Neff, P. von Neumann-Cosel, and A. Richter, “Pair Decay Width of the Hoyle State and its Role for Stellar Carbon Production,” Phys. Rev. Lett., vol. 105, p. 022501, July 2010.
- [8] R. de Diego, E. Garrido, D. V. Fedorov, and A. S. Jensen, “The triple alpha reaction rate and the resonances in ,” Phys. Lett. B, vol. 695, pp. 324–328, Jan. 2011.
- [9] R. Pichler, H. Oberhummer, A. Csótó, and S. A. Moszkowski, “Three-alpha structures in ,” Nucl. Phys. A, vol. 618, pp. 55–64, Feb. 1997.
- [10] C. Kurokawa and K. Katō, “Three-alpha resonances in ,” Nuclear Physics A, vol. 738, pp. 455–458, June 2004.
- [11] Y. Fujiwara, K. Miyagawa, M. Kohno, Y. Suzuki, D. Baye, and J.-M. Sparenberg, “A consistent 3 and 2 Faddeev calculation using the 2 RGM kernel,” Nucl. Phys. A, vol. 738, pp. 495–498, June 2004.
- [12] Y. Fujiwara, Y. Suzuki, and M. Kohno, “Case of almost redundant components in 3 Faddeev equations,” Phys. Rev. C, vol. 69, p. 037002, Mar. 2004.
- [13] Y. Fujiwara, M. Kohno, and Y. Suzuki, “Solving Three-Cluster OCM Equations in the Faddeev Formalism,” Few-Body Systems, vol. 34, pp. 237–257, 2004.
- [14] Y. Fujiwara and H. Nemura, “Three-Cluster Equation Using Two-Cluster RGM Kernel,” Prog. Theor. Phys., vol. 107, pp. 745–757, Apr. 2002.
- [15] Y. Fujiwara, K. Miyagawa, M. Kohno, Y. Suzuki, D. Baye, and J.-M. Sparenberg, “Faddeev calculation of 3 and systems using resonating-group method kernels,” Phys. Rev. C, vol. 70, p. 024002, Aug. 2004.
- [16] C. Kurokawa and K. Katō, “New broad state in ,” Phys. Rev. C, vol. 71, p. 021301, Feb. 2005.
- [17] K. Arai, “Resonance states of in a microscopic cluster model,” Phys. Rev. C, vol. 74, p. 064311, Dec. 2006.
- [18] M. Theeten, H. Matsumura, M. Orabi, D. Baye, P. Descouvemont, Y. Fujiwara, and Y. Suzuki, “Three-body model of light nuclei with microscopic nonlocal interactions,” Phys. Rev. C, vol. 76, p. 054003, Nov. 2007.
- [19] C. Kurokawa and K. Katō, “Spectroscopy of within the boundary condition for three-body resonant states,” Nucl. Phys. A, vol. 792, pp. 87–101, Aug. 2007.
- [20] P. Descouvemont and D. Baye, “Microscopic theory of the reaction in a three-cluster model,” Phys. Rev. C, vol. 36, pp. 54–59, July 1987.
- [21] M. Orabi, Y. Suzuki, H. Matsumura, Y. Fujiwara, D. Baye, P. Descouvemont, and M. Theeten, “3 description of with microscopic nonlocal potentials,” J. Phys. Conf. Ser., vol. 111, p. 012045, May 2008.
- [22] Y. Suzuki, H. Matsumura, M. Orabi, Y. Fujiwara, P. Descouvemont, M. Theeten, and D. Baye, “Local versus nonlocal interactions in a 3 description of ,” Phys. Lett. B, vol. 659, pp. 160–164, Jan. 2008.
- [23] P. Descouvemont, “Three- continuum states,” J. Phys. G Nucl. Phys., vol. 37, p. 064010, June 2010.
- [24] T. Yoshida, N. Itagaki, and K. Katō, “Symplectic structure and monopole strength in ,” Phys. Rev. C, vol. 83, p. 024301, Feb. 2011.
- [25] N. Moiseyev, “Quantum theory of resonances: calculating energies, widths and cross-sections by complex scaling,” Phys. Rep., vol. 302, pp. 212–293, Sept. 1998.
- [26] Y. K. Ho, “The method of complex coordinate rotation and its applications to atomic collision processes,” Phys. Rep., vol. 99, pp. 1–68, Oct. 1983.
- [27] G. S. Anagnostatos, “Alpha-chain states in ,” Phys. Rev. C, vol. 51, pp. 152–159, Jan. 1995.
- [28] A. C. Merchant and W. D. M. Rae, “Systematics of alpha-chain states in 4N-nuclei,” Nucl. Phys. A, vol. 549, pp. 431–438, Nov. 1992.
- [29] Y. Kanada-En’yo, “The Structure of Ground and Excited States of ,” Progr. Theor. Phys., vol. 117, pp. 655–680, Apr. 2007.
- [30] T. Neff and H. Feldmeier, “Cluster structures within Fermionic Molecular Dynamics,” Nucl. Phys. A, vol. 738, pp. 357–361, June 2004.
- [31] A. I. Baz’, “Diffusion-like processes in the quantum theory of scattering,” Soviet J. Exp. Theor. Phys., vol. 43, pp. 205–211, Feb. 1976.
- [32] T. Muñoz-Britton, M. Freer, N. I. Ashwood, T. A. D. Brown, W. N. Catford, N. Curtis, S. P. Fox, B. R. Fulton, C. W. Harlin, A. M. Laird, P. Mumby-Croft, A. S. J. Murphy, P. Papka, D. L. Price, K. Vaughan, D. L. Watson, and D. C. Weisser, “Search for the excitation of the Hoyle state in using the reaction,” J. Phys. G Nucl. Phys., vol. 37, p. 105104, Oct. 2010.
- [33] M. Freer, H. Fujita, Z. Buthelezi, J. Carter, R. W. Fearick, S. V. Förtsch, R. Neveling, S. M. Perez, P. Papka, F. D. Smit, J. A. Swartz, and I. Usman, “ excitation of the Hoyle state,” Phys. Rev. C, vol. 80, p. 041303, Oct. 2009.
- [34] V. Vasilevsky, A. V. Nesterov, F. Arickx, and J. Broeckhove, “Algebraic model for scattering in three-s-cluster systems. I. Theoretical background,” Phys. Rev. C, vol. 63, p. 034606 (16 pp), Mar. 2001.
- [35] V. Vasilevsky, A. V. Nesterov, F. Arickx, and J. Broeckhove, “Algebraic model for scattering in three-s-cluster systems. II. Resonances in the three-cluster continuum of and ,” Phys. Rev. C, vol. 63, p. 034607 (7 pp), Mar. 2001.
- [36] J. Broeckhove, F. Arickx, P. Hellinckx, V. S. Vasilevsky, and A. V. Nesterov, “The resonance structure studied with a three-cluster -matrix model,” J. Phys. G Nucl. Phys., vol. 34, pp. 1955–1970, Sept. 2007.
- [37] V. S. Vasilevsky, F. Arickx, J. Broeckhove, and V. N. Romanov, “Theoretical analysis of resonance states in , and above three-cluster threshold,” Ukr. J. Phys., vol. 49, no. 11, pp. 1053–1059, 2004.
- [38] V. Vasilevsky, A. V. Nesterov, F. Arickx, and J. Broeckhove, “S factor of the and reactions using a three-cluster exit channel,” Phys. Rev. C, vol. 63, p. 064604 (8 pp), June 2001.
- [39] S. Korennov and P. Descouvemont, “A microscopic three-cluster model in the hyperspherical formalism,” Nucl. Phys. A, vol. 740, pp. 249–267, Aug. 2004.
- [40] A. Damman and P. Descouvemont, “Three-body continuum states in a microscopic cluster model,” Phys. Rev. C, vol. 80, p. 044310, Oct. 2009.
- [41] Y. Fujiwara, Y. Suzuki, K. Miyagawa, and Michio, “Redundant Components in the 3 Faddeev Equation Using the 2 RGM Kernel,” Prog. Theor. Phys., vol. 107, pp. 993–1000, May 2002.
- [42] Y. A. Lashko and G. F. Filippov, “The role of the Pauli principle in three-cluster systems composed of identical clusters,” Nucl. Phys. A, vol. 826, pp. 24–48, July 2009.
- [43] W. Zickendraht, “Construction of a complete orthogonal system for the quantum-mechanical three-body problem,” Ann. Phys., vol. 35, pp. 18–41, Oct. 1965.
- [44] J. Nyiri and Y. A. Smorodinsky, “Symmetrical basis in the three-body problem. (In russian),” Yad. Fiz., vol. 29, pp. 833–844, 1979.
- [45] A. V. Nesterov, F. Arickx, J. Broeckhove, and V. S. Vasilevsky, “Three-cluster description of properties of light neutron- and proton-rich nuclei in the framework of the algebraic version of the resonating group method,” Phys. Part. Nucl., vol. 41, pp. 716–765, Sept. 2010.
- [46] D. R. Thompson, M. LeMere, and Y. C. Tang, “Systematic investigation of scattering problems with the resonating-group method,” Nucl. Phys., vol. A286, no. 1, pp. 53–66, 1977.
- [47] A. Sytcheva, F. Arickx, J. Broeckhove, and V. S. Vasilevsky, “Monopole and quadrupole polarization effects on the -particle description of ,” Phys. Rev. C, vol. 71, p. 044322, Apr. 2005.
- [48] K. Wildermuth and Y. Tang, A unified theory of the nucleus. Braunschweig: Vieweg Verlag, 1977.