This paper was converted on www.awesomepapers.org from LaTeX by an anonymous user.
Want to know more? Visit the Converter page.

Generalized coherent states satisfying the Pauli principle in a nuclear cluster model

Takayuki Myo takayuki.myo@oit.ac.jp General Education, Faculty of Engineering, Osaka Institute of Technology, Osaka, Osaka 535-8585, Japan Research Center for Nuclear Physics (RCNP), Osaka University, Ibaraki, Osaka 567-0047, Japan    Kiyoshi Katō kato@nucl.sci.hokudai.ac.jp Nuclear Reaction Data Centre, Faculty of Science, Hokkaido University, Sapporo 060-0810, Japan
Abstract

We propose a new basis state, which satisfies the Pauli principle in the nuclear cluster model. The basis state is defined as the generalized coherent state of the harmonic oscillator wave function using a pair of the creation operators and is orthogonal to the Pauli-forbidden states having smaller quanta. In the coherent basis state, the range parameter is changeable and controls the radial dilation. This property is utilized for the precise description of the relative motion between nuclear clusters. We show the reliability of this framework for the 2α2\alpha system of 8Be in the semi-microscopic orthogonality condition model. We obtain the resonances and non-resonant continuum states of 2α2\alpha with complex scaling. The resonance solutions and the phase shifts of the α\alpha-α\alpha scattering agree with those using the conventional projection operator method to remove the Pauli-forbidden states. We further discuss the extension of the present framework to the multi-α\alpha cluster systems using the SU(3) wave functions.

pacs:
21.60.Gx, 27.20.+n 

I Introduction

Nuclear clustering is a fundamental aspect of nuclei [1, 2, 3], such as the spatial formation of α\alpha clusters in nuclei. The 8Be nucleus is a typical cluster system decaying into two α\alpha particles. In 12C, the 02+0^{+}_{2} state is known as the Hoyle state having a three α\alpha-structure located near the three α\alpha threshold energy.

In nuclear cluster models, the resonating group method (RGM) [4, 5] is a microscopic approach starting from the degrees of freedom of nucleons and used to solve the relative motions between clusters in nuclei. The orthogonality condition model (OCM) [6] is a semi-microscopic approach, in which the local potential is often used as the intercluster potential to fit the experimental data of the cluster systems. This is the advantage of OCM to reproduce the threshold energies of every cluster emission in nuclei.

The Pauli principle is an essential statistics of nuclei and this property is fully treated in RGM. The Pauli-forbidden states are defined as the zero-eigenvalue states of the RGM norm kernel. In OCM, the Pauli-forbidden states are removed from the space of relative motion between clusters, and only the Pauli-allowed states are treated and obtained dynamically.

When the nuclear clusters are described with the harmonic oscillator (HO) shell model wave functions, the Pauli-forbidden states are also expressed by using the HO states for the relative wave function between clusters. Technically there are several methods to remove the Pauli-forbidden states in relative motion in OCM. One is the Gram-Schmidt orthonormalization method [7]. When the relative motion is precisely solved by using the linear combination of the appropriate basis functions, Kukulin’s projection operator method works to push the Pauli-forbidden states in every relative motion to the irrelevant energy region [8]. In this method, the pseudo potential with the projection operator form to the Pauli-forbidden states is added to the Hamiltonian and the orthogonal solutions can be obtained as physical states. This method sometimes makes difficulty increasing the number of clusters in multicluster systems such as 4α4\alpha and 5α5\alpha, because precise projections are necessary for every cluster-pair to eliminate the Pauli-forbidden states in all the relative motions, which causes the numerical efforts with many basis states. In this situation, one needs an efficient method to treat the Pauli-allowed states in the description of multicluster systems based on OCM.

In this paper, we propose a new scheme to treat the Pauli-allowed states in OCM; all the basis states in relative motion are automatically orthogonal to the Pauli-forbidden states and it is not necessary to use the projection operator in the Hamiltonian and the wave function. We formulate this method in the generalized coherent states [9] of the HO basis states for relative motion between clusters using the raising operator 𝒂^𝒂^\hat{\mbox{\boldmath$a$}}^{\dagger}\cdot\hat{\mbox{\boldmath$a$}}^{\dagger} [10, 11]. This operator increases the quanta of every HO state and can be utilized to define the Pauli-allowed states above the Pauli-forbidden states. In this method, we can describe the resonances in cluster-cluster scattering using the complex scaling [12].

In this paper, we formulate the new method and confirm its reliability by calculating the 2α2\alpha system of 8Be, in which we use the complex-scaled solutions of the resonant and nonresonant continuum states. The present work becomes the foundation to investigate the multicluster systems in the OCM approach.

In Sec. II, we provide the formulation of the generalized coherent state with the HO basis states and its application to the nuclear cluster systems. In Sec. III, we discuss the resonances and scattering of the 2α\alpha system of 8Be. In Sec. IV, we discuss the extension to the multi-α\alpha cluster systems using the SU(3) wave functions. In Sec. V, we summarize this work. In the Appendix, we give the mathematical derivation of the generalized coherent states with the HO basis states.

II Theoretical methods

II.1 Generalized coherent states

We begin with the harmonic oscillator (HO) basis state ϕnm(𝒓,ν)\phi_{n\ell m}(\mbox{\boldmath$r$},\nu) with a range ν=1/b2\nu=1/b^{2} and a principal quantum number N=2n+N=2n+\ell, where nn represents the number of nodes in the radial wave function and \ell is an orbital angular momentum. Using the operators of the creation and annihilation of a quanta NN, the HO basis state can be written as [13, 14]

ϕnm(𝒓,ν)=An(𝒂^𝒂^)n𝒴m(𝒂^)ϕ0(𝒓,ν),An=(1)n4π(2n+2+1)!!(2n)!!,ϕ0(𝒓,ν)=(νπ)3/4e12νr2,\begin{split}\phi_{n\ell m}(\mbox{\boldmath$r$},\nu)&=A_{n\ell}\,\left(\hat{\mbox{\boldmath$a$}}^{\dagger}\cdot\hat{\mbox{\boldmath$a$}}^{\dagger}\right)^{n}{\cal Y}_{\ell m}(\hat{\mbox{\boldmath$a$}}^{\dagger})\,\phi_{0}(\mbox{\boldmath$r$},\nu),\\ A_{n\ell}&=(-1)^{n}\sqrt{\frac{4\pi}{(2n+2\ell+1)!!\,(2n)!!}},\\ \phi_{0}(\mbox{\boldmath$r$},\nu)&=\left(\frac{\nu}{\pi}\right)^{3/4}e^{-\frac{1}{2}\nu r^{2}},\end{split} (1)

where 𝒴m(𝒓)=rYm(𝒓^){\cal Y}_{\ell m}(\mbox{\boldmath$r$})=r^{\ell}Y_{\ell m}(\hat{\mbox{\boldmath$r$}}) is a solid spherical harmonics and ϕ0\phi_{0} is a vacuum with N=0N=0. This HO basis state can be used to represent the single-nucleon wave function in nuclei and also the relative wave function between nuclear clusters.

In this study, we introduce the following scalar operators D^\hat{D}^{\dagger} (raising) and D^\hat{D} (lowering) following Ref. [10] as

D^=𝒂^𝒂^,D^=𝒂^𝒂^.\begin{split}\hat{D}^{\dagger}&=\hat{\mbox{\boldmath$a$}}^{\dagger}\cdot\hat{\mbox{\boldmath$a$}}^{\dagger},\qquad\hat{D}=\hat{\mbox{\boldmath$a$}}\cdot\hat{\mbox{\boldmath$a$}}.\end{split} (2)

These operators belong to the symplectic Sp(3,𝑹R) Lie algebra of the coherent state of the collective motion and change the quanta of the wave function by two for the radial part without changing the angular momentum. Using the raising operator D^\hat{D}^{\dagger}, we introduce the following new basis state ϕnmβ(𝒓,ν)\phi_{n\ell m}^{\beta}(\mbox{\boldmath$r$},\nu), in which D^\hat{D}^{\dagger} is coherently multiplied by the HO basis state with the weight of the real parameter β\beta in the exponential form as

ϕnmβ(𝒓,ν)=exp(12βD^)ϕnm(𝒓,ν)=1(1+β)N+3/2exp(β2(1+β)νr2)×ϕnm(𝒓,ν1+β).\begin{split}\phi_{n\ell m}^{\beta}(\mbox{\boldmath$r$},\nu)&=\exp\left(\frac{1}{2}\beta\hat{D}^{\dagger}\right)\,\phi_{n\ell m}(\mbox{\boldmath$r$},\nu)\\ &=\frac{1}{\sqrt{(1+\beta)^{N+3/2}}}\,\exp\left(\frac{\beta}{2(1+\beta)}\nu r^{2}\right)\,\\ &\times\phi_{n\ell m}(\mbox{\boldmath$r$},\frac{\nu}{1+\beta}).\end{split} (3)

The derivation of this equation is given in Appendix A. This new basis state is a kind of generalized coherent state [9] in terms of D^\hat{D}^{\dagger} and can be represented by the HO basis state with the same quanta NN and the different range parameter ν/(1+β)\nu/(1+\beta) and multiplying the Gaussian function with the coordinate rr. This equation plays an essential role in this study. It is noted that the basis state has the following exponential dependence;

ϕnmβ(𝒓,ν)exp(1β2(1+β)νr2).\begin{split}\phi_{n\ell m}^{\beta}(\mbox{\boldmath$r$},\nu)&\propto\exp\left(-\frac{1-\beta}{2(1+\beta)}\,\nu r^{2}\right).\end{split} (4)

This form gives a condition of |β|<1|\beta|<1 to satisfy the asymptotically damping behavior, which is imposed throughout this study. When n=0n=0, the basis state ϕ0mβ(𝒓,ν)\phi_{0\ell m}^{\beta}(\mbox{\boldmath$r$},\nu) becomes the nodeless Gaussian function multiplying rr^{\ell}, which is often used in the Gaussian expansion technique [15, 16, 17].

From the property of the raising operator D^\hat{D}^{\dagger}, the function ϕnmβ(𝒓,ν)\phi_{n\ell m}^{\beta}(\mbox{\boldmath$r$},\nu) includes only the quanta larger than or equal to NN of the HO basis states with the range ν\nu. Hence the following orthogonal condition is satisfied;

ϕnm(ν)|ϕnmβ(ν)=0forn<n(N<N).\langle\phi_{n^{\prime}\ell m}(\nu)|\phi_{n\ell m}^{\beta}(\nu)\rangle=0\qquad\mbox{for}~{}n^{\prime}<n~{}(N^{\prime}<N). (5)

This property is useful to construct the HO basis states with a quanta NN and any values of β\beta, which are orthogonal to the HO states with a lower quanta NN^{\prime}. If one regards the HO basis states with the lower quanta NN^{\prime} as the occupied states in the nucleus, namely the Pauli-forbidden states, the generalized coherent basis states ϕnmβ(𝒓,ν)\phi_{n\ell m}^{\beta}(\mbox{\boldmath$r$},\nu) can be the unoccupied states in the nucleus automatically, and represents the Pauli-allowed states. A specific case of this formulation is introduced in the shell model, in which the HO particle state with a free range parameter is taken to be orthogonal to the HO hole states by adjusting the polynomial in the HO particle state [18, 19].

The parameter β\beta controls the spatial range of the generalized coherent basis states. When β\beta is close to unity, the basis state has a long tail and is suitable to describe a weak-binding state of nuclei such as a halo structure and the low-energy scattering solution in the nuclear reaction. When β\beta is close to 1-1, the basis state becomes a short-range and is suitable to describe the short-range and tensor correlations of nucleons with high momenta in nuclei [18]. From these properties, the parameter β\beta plays a role on the radial dilation of the coherent basis state, and then we call β\beta “dilation parameter” hereafter.

In the cluster model, the present coherent basis state is useful to describe relative motion between clusters with the orthogonality condition from the Pauli principle for the following two reasons:

  1. (i)

    When the cluster wave functions are the HO shell-model ones, the Pauli-forbidden states in relative motion become the HO states with a specific quanta NPFN_{\rm PF}. Hence, the coherent basis states with a relative oscillator quanta NN and β\beta become the Pauli-allowed states that are orthogonal to the Pauli-forbidden states with the condition of NPF<NN_{\rm PF}<N [20].

  2. (ii)

    The relative motion between clusters is solved precisely and the relative wave function is optimized by superposing the coherent basis states ϕnmβ(𝒓,ν)\phi_{n\ell m}^{\beta}(\mbox{\boldmath$r$},\nu) with various dilation parameters β\beta, each of which shows a different spatial distribution.

In the multicluster system, we can prepare the cluster wave function using the coherent basis states in relative motion between every cluster-pair. In this paper, we consider the two-cluster case with clusters C1 and C2 and one intercluster motion with the coordinate 𝒓r in the single channel. We express the total nuclear wave function Ψ\Psi, in which the relative wave function Φrel(𝒓)\Phi_{\rm rel}(\mbox{\boldmath$r$}) is in the linear combination form of the coherent basis states {ϕnmβi(𝒓,νrel)}\{\phi^{\beta_{i}}_{n\ell m}(\mbox{\boldmath$r$},\nu_{\rm rel})\} with the range parameter νrel\nu_{\rm rel}, the set of {βi}\{\beta_{i}\} with i=1,,Nbasei=1,\cdots,N_{\rm base}, and the condition of N=2n+N=2n+\ell for Pauli-allowed states, N>NPFN>N_{\rm PF};

Ψ=𝒜{ϕC1ϕC2Φrel(𝒓)},Φrel(𝒓)=i=1NbaseCiϕnmβi(𝒓,νrel),\begin{split}\Psi&={\cal A}\{\phi_{{\rm C}_{1}}\phi_{{\rm C}_{2}}\Phi_{\rm rel}(\mbox{\boldmath$r$})\},\\ \Phi_{\rm rel}(\mbox{\boldmath$r$})&=\sum_{i=1}^{N_{\rm base}}C_{i}\,\phi^{\beta_{i}}_{n\ell m}(\mbox{\boldmath$r$},\nu_{\rm rel}),\end{split} (6)

where 𝒜{\cal A} is the antisymmetrizer of nucleons between different clusters and ϕC\phi_{{\rm C}} is the internal wave function of the cluster C\rm C. Hereafter we omit the notation of the quantum numbers nn,\ell, and mm in the basis states for simplicity. It is possible to add the basis states with different nn to Φrel(𝒓)\Phi_{\rm rel}(\mbox{\boldmath$r$}) as well as βi\beta_{i}.

In the present study, we adopt the orthogonality condition model (OCM). The eigenvalue problem of the Hamiltonian HH for relative motion is given to obtain the relative energy EE between clusters:

H=Trel+VC1C2,HΦrel(𝒓)=EΦrel(𝒓),j=1Nbase(HijENij)Cj=0,Hij=ϕβi(νrel)|H|ϕβj(νrel),Nij=ϕβi(νrel)|ϕβj(νrel),\begin{split}H&=T_{\rm rel}+V_{{\rm C}_{1}{\rm C}_{2}},\\ H\Phi_{\rm rel}(\mbox{\boldmath$r$})&=E\Phi_{\rm rel}(\mbox{\boldmath$r$}),\\ \sum_{j=1}^{N_{\rm base}}(H_{ij}&-EN_{ij})\,C_{j}=0,\\ H_{ij}&=\langle\phi^{\beta_{i}}(\nu_{\rm rel})|H|\phi^{\beta_{j}}(\nu_{\rm rel})\rangle,\\ N_{ij}&=\langle\phi^{\beta_{i}}(\nu_{\rm rel})|\phi^{\beta_{j}}(\nu_{\rm rel})\rangle,\end{split} (7)

where TrelT_{\rm rel} and VC1C2V_{{\rm C}_{1}{\rm C}_{2}} are the kinetic energy and the potential of relative motion between clusters, respectively. The matrix elements of HijH_{ij} and NijN_{ij} are those of the Hamiltonian and norm with the individual β\beta-values, respectively. In this paper, we call the present framework “coherent basis method”.

In the coherent basis method, the matrix elements can be calculated analytically, and we use the formulas using the HO basis states with the independent range parameters in the bra and ket states [7], characterized by βi\beta_{i} and βj\beta_{j} in Eq. (7). For kinetic energy, we give the formula in Appendix B.

II.2 α\alpha-α\alpha system

We demonstrate the present new scheme in the α\alpha-α\alpha cluster system of 8Be. The α\alpha cluster is represented by the (0s)4(0s)^{4} configuration of the HO basis state where the range parameter ν\nu of the single-nucleon state is taken as 0.535 fm-2, which corresponds to the length of b=1.3672b=1.3672 fm, to reproduce the charge radius of the α\alpha particle. We prepare the coherent basis states for the relative wave function of 2α2\alpha, the range parameter of which is νrel=2ν=\nu_{\rm rel}=2\nu= 1.070 fm-2 corresponding to the length of brel=0.9667b_{\rm rel}=0.9667 fm. We employ the folding potential between α\alpha-α\alpha with the nucleon-nucleon interaction and the Coulomb interaction using the α\alpha cluster wave function. We adopt the Schmid-Wildermuth effective nucleon-nucleon interaction [21], which is often used in the previous studies of the multi-α\alpha cluster systems [22, 23, 24, 25]. The form of the α\alpha-α\alpha folding potential Vαα(r)V_{\alpha\alpha}(r) is given with nuclear (N) and Coulomb (C) parts as

Vαα(r)=VααN(r)+VααC(r),VααN(r)=2XDV0a3/2eaμr2,XD=8W+4B4H2M,a=2ν2ν+3μVααC(r)=4e2erf(cr)r,c=2ν3,\begin{split}V_{\alpha\alpha}(r)&=V^{\rm N}_{\alpha\alpha}(r)+V^{\rm C}_{\alpha\alpha}(r),\\ V^{\rm N}_{\alpha\alpha}(r)&=2\,X_{D}\,V_{0}\,a^{3/2}e^{-a\mu r^{2}},\\ X_{D}&=8W+4B-4H-2M,\quad a=\frac{2\nu}{2\nu+3\mu}\\ V^{\rm C}_{\alpha\alpha}(r)&=4\,e^{2}\,\frac{{\rm erf}(cr)}{r},\qquad c=\sqrt{\frac{2\nu}{3}},\end{split} (8)

where r=|𝒓|r=|\mbox{\boldmath$r$}|, V0=72.98V_{0}=-72.98 MeV, μ=0.46\mu=0.46 fm-2, W=M=0.4075W=M=0.4075, and B=H=0.0925B=H=0.0925.

The lowest shell-model configuration of the 2α2\alpha system is (0s)4(0p)4(0s)^{4}(0p)^{4} in the HO basis state with a total quanta of 4. Hence the Pauli-forbidden HO states, ωF(𝒓,νrel)\omega_{\rm F}(\mbox{\boldmath$r$},\nu_{\rm rel}), are defined by the condition of quanta NPF<4N_{\rm PF}<4 in the relative motion with the range νrel\nu_{\rm rel}. In the coherent basis state, we impose this condition of the Pauli-allowed states and set N=4N=4 for the 0+0^{+}, 2+2^{+}, and 4+4^{+} states and N=6N=6 for the 6+6^{+} state in the present study. For the 4+4^{+} and 6+6^{+} states, there is no Pauli-forbidden state.

We take various dilation parameters βi\beta_{i} in Eq. (6) to optimize the radial wave function. In the present study, we choose the set of βi\beta_{i} in the form of the geometric progression of the length parameters bib_{i} of the HO basis state [15, 16] according to Eq. (4) as

1βi1+βiνrel=1bi2=1(b0γi1)2.\begin{split}\frac{1-\beta_{i}}{1+\beta_{i}}\,\nu_{\rm rel}&=\frac{1}{b_{i}^{2}}=\frac{1}{(b_{0}\gamma^{i-1})^{2}}.\end{split} (9)

We set b0=0.2b_{0}=0.2 fm, γ=1.2\gamma=1.2, and Nbase=30N_{\rm base}=30 in the present calculation, which are transformed to βi\beta_{i} in the coherent basis states.

To show the reliability of the coherent basis method, we compare the obtained results with those of the conventional projection operator method (PO) [8]. In PO, one usually adds the pseudo potential of the projection operators with a positive prefactor λ\lambda to the original Hamiltonian given as;

Hλ=H+λf|ωF,fωF,f|.\begin{split}H_{\lambda}&=H+\lambda\sum_{f}|\omega_{{\rm F},f}\rangle\langle\omega_{{\rm F},f}|.\end{split} (10)

One uses a large value of λ\lambda to make the solutions orthogonal to the Pauli-forbidden states {ωF,f}\{\omega_{{\rm F},f}\} with the index ff and we take λ=106\lambda=10^{6} MeV in this study [17]. The number of the Pauli-forbidden states is determined from the condition of of quanta as NPF=2nf+f<4N_{\rm PF}=2n_{f}+\ell_{f}<4. For the basis states of the relative motion in PO, we adopt the nodeless HO basis functions with n=0n=0, which are often used in the OCM calculation, as

Φrel(𝒓)=i=1NbaseC¯iϕ0m(𝒓,bi),ϕ0m(𝒓,bi)=N(bi)e1/2(r/bi)2𝒴m(𝒓),\begin{split}\Phi_{\rm rel}(\mbox{\boldmath$r$})&=\sum_{i=1}^{N_{\rm base}}\bar{C}_{i}\,\phi_{0\ell m}(\mbox{\boldmath$r$},b_{i}),\\ \phi_{0\ell m}(\mbox{\boldmath$r$},b_{i})&=N_{\ell}(b_{i})\,e^{-1/2(r/b_{i})^{2}}\,{\cal Y}_{\ell m}(\mbox{\boldmath$r$}),\end{split} (11)

where N(b)N_{\ell}(b) is a normalization factor of the basis state. The choice of the length parameters {bi}\{b_{i}\} is the same as those of the coherent basis states in Eq. (9), which is suitable for comparing the solutions.

II.3 Complex scaling

We describe the resonances and the scattering of the α\alpha-α\alpha system using the complex scaling [12, 17, 26, 27, 28] in both the coherent basis method and the projection operator method. In the complex scaling, the relative coordinate 𝒓r, the relative momentum 𝒑p in the Hamiltonian HH, and the relative wave function Φrel(𝒓)\Phi_{\rm rel}(\mbox{\boldmath$r$}) are transformed using a scaling angle θ\theta with an operator U(θ)U(\theta) as

U(θ):𝒓𝒓eiθ,𝒑𝒑eiθ,U(\theta)~{}:~{}\mbox{\boldmath$r$}\to\mbox{\boldmath$r$}\,e^{i\theta},\qquad\mbox{\boldmath$p$}\to\mbox{\boldmath$p$}\,e^{-i\theta}, (12)

where θ\theta is a real positive number. The complex-scaled Hamiltonian HθH^{\theta}, the complex-scaled relative wave function Φrelθ\Phi_{\rm rel}^{\theta}, and the corresponding energy EθE^{\theta} are given as

Hθ=U(θ)HU1(θ),Φrelθ(𝒓)=U(θ)Φrel(𝒓)HθΦrelθ(𝒓)=EθΦrelθ(𝒓).\begin{split}H^{\theta}&=U(\theta)HU^{-1}(\theta),\\ \Phi_{\rm rel}^{\theta}(\mbox{\boldmath$r$})&=U(\theta)\Phi_{\rm rel}(\mbox{\boldmath$r$})\\ H^{\theta}\Phi_{\rm rel}^{\theta}(\mbox{\boldmath$r$})&=E^{\theta}\Phi_{\rm rel}^{\theta}(\mbox{\boldmath$r$}).\end{split} (13)

After solving the last equation, EθE^{\theta} are obtained for bound, resonant, and continuum states on the complex energy plane according to the ABC theorem [29]. The energies of the continuum states start from the α\alpha+α\alpha threshold energy and are obtained along the line rotated down by 2θ2\theta from the real energy axis. The energies of the bound and resonant states are independent of θ\theta. The resonance has a complex energy ER=EriΓ/2E_{\rm R}=E_{r}-i\Gamma/2 with a resonance energy ErE_{r} and a decay width Γ\Gamma. The asymptotic behavior of the resonance wave function becomes a damping form if 2θ>|arg(ER)|2\theta>|\arg(E_{\rm R})| [29]. In calculations with a finite number of the basis states, the resonances are identified from the stationary property of ERE_{\rm R} with respect to θ\theta on the complex energy plane [17, 26, 27], and the continuum states are discretized with the complex energies. The wave function Φ~relθ(𝒓)\tilde{\Phi}_{\rm rel}^{\theta}(\mbox{\boldmath$r$}) is the biorthogonal state of Φrelθ(𝒓)\Phi_{\rm rel}^{\theta}(\mbox{\boldmath$r$}) [30], and used for the bra state in the complex-scaled matrix elements. One does not take the complex conjugate of the radial part of the bra state in the matrix elements [26, 27].

The Pauli forbidden state ωF(𝒓,νrel)\omega_{\rm F}(\mbox{\boldmath$r$},\nu_{\rm rel}) is also transformed in the complex scaling as

U(θ)ωF(𝒓,νrel)=ωFθ(𝒓,νrel)=e3iθ/2ωF(𝒓eiθ,νrel)=ωF(𝒓,νrele2iθ).\begin{split}U(\theta)\omega_{\rm F}(\mbox{\boldmath$r$},\nu_{\rm rel})&=\omega_{\rm F}^{\theta}(\mbox{\boldmath$r$},\nu_{\rm rel})=e^{3i\theta/2}\omega_{\rm F}(\mbox{\boldmath$r$}e^{i\theta},\nu_{\rm rel})\\ &=\omega_{\rm F}(\mbox{\boldmath$r$},\nu_{\rm rel}e^{2i\theta}).\end{split} (14)

In the last equation, we use the explicit form of the HO basis state, and the range parameter νrel\nu_{\rm rel} is transformed instead of 𝒓r.

In the projection operator method, the Hamiltonian HλH_{\lambda} in Eq. (10) is transformed as Hλθ=U(θ)HλU1(θ)H_{\lambda}^{\theta}=U(\theta)H_{\lambda}U^{-1}(\theta), in which the Pauli-forbidden states are transformed in the pseudo potential [17]. In analogy with Eq. (11), the complex-scaled wave function is given as

Φrelθ(𝒓)=i=1NbaseC¯iθϕ0m(𝒓,bi),\begin{split}\Phi^{\theta}_{\rm rel}(\mbox{\boldmath$r$})&=\sum_{i=1}^{N_{\rm base}}\bar{C}^{\theta}_{i}\,\phi_{0\ell m}(\mbox{\boldmath$r$},b_{i}),\end{split} (15)

where the θ\theta dependence is included in {C¯iθ}\{\bar{C}^{\theta}_{i}\}. This expansion is often used in the conventional OCM calculation with complex scaling [17].

In the coherent basis method, the coherent basis state with a dilation parameter β\beta in Eq. (6) is transformed because the basis state should be orthogonal to the complex-scaled Pauli-forbidden states as

ω~Fθ(νrel)|ϕβ,θ(νrel)=ω~F(νrele2iθ)|ϕβ(νrele2iθ)=0.\langle\tilde{\omega}_{\rm F}^{\theta}(\nu_{\rm rel})|\phi^{\beta,\theta}(\nu_{\rm rel})\rangle=\langle\tilde{\omega}_{\rm F}(\nu_{\rm rel}e^{2i\theta})|\phi^{\beta}(\nu_{\rm rel}e^{2i\theta})\rangle=0. (16)

Hence the relative wave function Φrelθ(𝒓)\Phi_{\rm rel}^{\theta}(\mbox{\boldmath$r$}) is expanded in terms of the complex-scaled coherent basis states {ϕβi,θ(𝒓,νrel)}\{\phi^{\beta_{i},\theta}(\mbox{\boldmath$r$},\nu_{\rm rel})\} with the index ii for βi\beta_{i} as

Φrelθ(𝒓)=i=1NbaseCiθϕβi,θ(𝒓,νrel),\begin{split}\Phi_{\rm rel}^{\theta}(\mbox{\boldmath$r$})&=\sum_{i=1}^{N_{\rm base}}C_{i}^{\theta}\phi^{\beta_{i},\theta}(\mbox{\boldmath$r$},\nu_{\rm rel}),\end{split} (17)

One solves the following eigenvalue problem of the complex-scaled Hamiltonian matrix and obtains EθE^{\theta} and {Ciθ}\{C_{i}^{\theta}\} for each eigenstate;

i=1Nbase(HijθEθNijθ)Cjθ=0.\sum_{i=1}^{N_{\rm base}}\left(H_{ij}^{\theta}-E^{\theta}N_{ij}^{\theta}\right)C_{j}^{\theta}=0. (18)

Technically, the matrix elements with the complex-scaled coherent basis states are calculated in the following procedure;

Hijθ=ϕ~βi,θ(νrel)|Hθ|ϕβj,θ(νrel)=p,qNpϕ~βi,θ(νrel)|ϕpϕ~p|Hθ|ϕqϕ~q|ϕβj,θ(νrel)=p,qNpDp,iθHpqθDq,jθ,\begin{split}H_{ij}^{\theta}&=\langle\tilde{\phi}^{\beta_{i},\theta}(\nu_{\rm rel})|H^{\theta}|\phi^{\beta_{j},\theta}(\nu_{\rm rel})\rangle\\ &=\sum_{p,q}^{N_{p}}\langle\tilde{\phi}^{\beta_{i},\theta}(\nu_{\rm rel})|\phi_{p}\rangle\,\langle\tilde{\phi}_{p}|H^{\theta}|\phi_{q}\rangle\,\langle\tilde{\phi}_{q}|\phi^{\beta_{j},\theta}(\nu_{\rm rel})\rangle\\ &=\sum_{p,q}^{N_{p}}D_{p,i}^{\theta}H_{pq}^{\theta}D_{q,j}^{\theta},\end{split} (19)
Hpqθ=ϕ~p|Hθ|ϕq=ϕ~pθ|H|ϕqθ,Dp,iθ=ϕ~βi,θ(νrel)|ϕp=ϕβi(νrel)|ϕpθ,Nijθ=ϕ~βi,θ(νrel)|ϕβj,θ(νrel)=pNpDp,iθDp,jθ.\begin{split}H_{pq}^{\theta}&=\langle\tilde{\phi}_{p}|H^{\theta}|\phi_{q}\rangle=\langle\tilde{\phi}_{p}^{-\theta}|H|\phi_{q}^{-\theta}\rangle,\\ D_{p,i}^{\theta}&=\langle\tilde{\phi}^{\beta_{i},\theta}(\nu_{\rm rel})|\phi_{p}\rangle=\langle\phi^{\beta_{i}}(\nu_{\rm rel})|\phi_{p}^{-\theta}\rangle,\\ N_{ij}^{\theta}&=\langle\tilde{\phi}^{\beta_{i},\theta}(\nu_{\rm rel})|\phi^{\beta_{j},\theta}(\nu_{\rm rel})\rangle=\sum_{p}^{N_{p}}D_{p,i}^{\theta}D_{p,j}^{\theta}.\end{split} (20)

Here we insert the the completeness relation consisting of the states with a finite number NpN_{p} ; 1=p=1Np|ϕpϕ~p|1=\sum_{p=1}^{N_{p}}|\phi_{p}\rangle\,\langle\tilde{\phi}_{p}|. In this study, we construct the completeness relation in terms of the nodeless HO basis function with θ=0\theta=0, which are the same as those used in the projection operator method. We use the same set of {bi}\{b_{i}\}.

ϕp(𝒓)=i=1NbaseCi,pϕ0m(𝒓,bi),ϕ~p|ϕq=δpq.\begin{split}\phi_{p}(\mbox{\boldmath$r$})&=\sum_{i=1}^{N_{\rm base}}C_{i,p}\,\phi_{0\ell m}(\mbox{\boldmath$r$},b_{i}),\\ \langle\tilde{\phi}_{p}|\phi_{q}\rangle&=\delta_{pq}.\\ \end{split} (21)

We diagonalize the norm matrix of ϕ0m(𝒓,bi)\phi_{0\ell m}(\mbox{\boldmath$r$},b_{i}) and construct the orthonormalized basis states {ϕp}\{\phi_{p}\} in the linear combination of ϕ0m(𝒓,bi)\phi_{0\ell m}(\mbox{\boldmath$r$},b_{i}) with the coefficients Ci,pC_{i,p}, which nicely describe the completeness relation in the present calculation. The states {ϕp}\{\phi_{p}\} involve the Pauli-forbidden states, which are removed by diagonalizing the norm matrix with the elements of NijθN_{ij}^{\theta} in Eq. (20), because of the overlap with the coherent basis state in {Dp,iθ}\{D_{p,i}^{\theta}\}. In the eigenvalue problem in Eq. (18), when one diagonalizes the norm matrix, the eigenstates of the Pauli-forbidden states show the zero-energy eigenvalue, which are removed from the basis states before diagonalizing the Hamiltonian matrix. It is noted that this procedure is used for the calculation of the unbound states with the complex scaling only, and is not necessary for the bound-state calculation with θ=0\theta=0.

II.4 Level density

In the complex scaling, the solutions {Φnθ,Φ~nθ}\{\Phi_{n}^{\theta},\tilde{\Phi}_{n}^{\theta}\} construct the completeness relation [30, 31] given as

1=n|ΦnθΦ~nθ|,1=\sum_{n}|\Phi_{n}^{\theta}\rangle\langle\tilde{\Phi}_{n}^{\theta}|, (22)

where nn is the state index. Using the energy eigenvalues {Enθ}\{E_{n}^{\theta}\}, the complex-scaled Green’s function 𝒢θ(E){\cal G}^{\theta}(E) is expressed as

𝒢θ(E)=1EHθ=n|ΦnθΦ~nθ|EEnθ.{\cal G}^{\theta}(E)=\frac{1}{E-H^{\theta}}=\sum_{n}\frac{|\Phi^{\theta}_{n}\rangle\langle\tilde{\Phi}^{\theta}_{n}|}{E-E_{n}^{\theta}}. (23)

We calculate the level density ρ(E)=nδ(EEn)\rho(E)=\sum_{n}\delta(E-E_{n}) with complex scaling [32, 33, 34]. The complex-scaled level density ρθ(E)\rho^{\theta}(E) is given with 𝒢θ(E){\cal G}^{\theta}(E) as

ρθ(E)=1πIm{Tr𝒢θ(E)}=1πnIm(1EEnθ).\rho^{\theta}(E)=-\frac{1}{\pi}{\rm Im}\left\{{\rm Tr}\,{\cal G}^{\theta}(E)\right\}=-\frac{1}{\pi}\sum_{n}{\rm Im}\left(\frac{1}{E-E_{n}^{\theta}}\right). (24)

We also consider the asymptotic Hamiltonian H0θH_{0}^{\theta} with the energy eigenvalues {E0,nθ}\{E_{0,n}^{\theta}\}, and define the asymptotic level density ρ0θ(E)\rho_{0}^{\theta}(E) as

ρ0θ(E)=1πnIm(1EE0,nθ).\rho_{0}^{\theta}(E)=-\frac{1}{\pi}\ \sum_{n}{\rm Im}\Biggl{(}\frac{1}{E-E_{0,n}^{\theta}}\Biggr{)}. (25)

One defines the continuum level density Δ(E)=ρθ(E)ρ0θ(E)\Delta(E)=\rho^{\theta}(E)-\rho_{0}^{\theta}(E), which is related to the scattering matrix S(E)S(E) [35]:

Δ(E)=12πImddEln{detS(E)}.\Delta(E)=\frac{1}{2\pi}{\rm Im}\frac{{\rm d}}{{\rm d}E}{\rm ln}\bigl{\{}{\rm det}\,S(E)\bigr{\}}. (26)

In the single channel, Δ(E)\Delta(E) becomes the derivative of the phase shift δ(E)\delta(E) and the phase shift is obtained as

δ(E)=πEΔ(E)dE.\begin{split}\delta(E)&=\pi\int_{-\infty}^{E}\Delta(E^{\prime}){\rm d}E^{\prime}.\end{split} (27)

We define the asymptotic Hamiltonian H0H_{0} for the 2α2\alpha system as [36]

H0=Trel+4e2r.H_{0}=T_{\rm rel}+\frac{4e^{2}}{r}. (28)

We omit the nuclear interaction, and replace the Coulomb interaction with the point type. In the asymptotic wave function of 2α2\alpha, one omits the antisymmetrization between the nucleons in the different α\alpha clusters [36, 37]. This means no Pauli-forbidden state in the relative motion between 2α2\alpha and then we set N=N=\ell with n=0n=0 in the coherent basis method. We also omit the projection operator in H0H_{0} in the projection operator method.

III Results

III.1 α\alpha-α\alpha system

In this study, we treat the 2α2\alpha system of 8Be and discuss the α\alpha-α\alpha resonances. First, we compare the diagonal energies of the basis states in the coherent basis method (CH) and the projection operator method (PO) as functions of the HO length parameter bb in the Gaussians using Eq. (9). In two methods, the treatments of Pauli-forbidden states are different, and affect the diagonal energies. We show the results of the 0+0^{+} and 2+2^{+} states in Fig. 1 (top) in a logarithmic scale. We also show the results as functions of the dilation parameter β\beta used in the coherent basis states in Fig. 1 (bottom). These figures are useful to understand the treatment of the Pauli principle in the coherent basis method, which leads to the low-energy states in the large value of bb, namely a large α\alpha-α\alpha distance, and also in the values of β\beta close to unity.

In the projection operator method, the basis states in Eq. (11) can involve the Pauli-forbidden states, and then the pseudo potential with the strength of λ=106\lambda=10^{6} MeV makes the states have high energies. The HO length of Pauli-forbidden states is brel=b_{\rm rel}=0.9667 fm and the maximum energies appear at this length for two spin states. For the 0+0^{+} state, there are two Pauli-forbidden states with n=0n=0 and 11 and then the repulsive effect is distributed in a wider range of bb than the results of the 2+2^{+} state, which includes one Pauli-forbidden state with n=0n=0. In the projection operator method, the superposition of the basis states makes the Pauli-allowed states with low energies. The comparison of the two methods explains the reasonable treatment of the Pauli-allowed states in the coherent basis method.

Refer to caption
Refer to caption
Figure 1: Diagonal energies of the relative motion in 8Be for 0+0^{+} (red) and 2+2^{+} (blue) states as functions of the HO length parameter bb (top) and dilation parameter β\beta (bottom) in the coherent basis method (CH) and the projection operator method (PO) without complex scaling.

Next, we solve the eigenvalue problem of the Hamiltonian matrix. For 0+0^{+} state, there are two Pauli-forbidden states and in the projection operator method, two states are obtained to have the high energies close to λ\lambda. On the other hand in the coherent basis method, the basis states do not involve the Pauli-forbidden states, and all eigenstates are obtained as the Pauli-allowed states.

Before the calculation of resonances, we discuss the reliability of the present coherent basis method for the bound state. For this purpose, we artificially strengthen the α\alpha-α\alpha nuclear potential VααN(r)V^{\rm N}_{\alpha\alpha}(r) to make the 0+0^{+} and 2+2^{+} states of 8Be bound. We introduce the enhancement factor δ\delta in VααN(r)V^{\rm N}_{\alpha\alpha}(r) as

VααN(r)VααN(r)(1+δ).V^{\rm N}_{\alpha\alpha}(r)\to V^{\rm N}_{\alpha\alpha}(r)(1+\delta). (29)

We compare the resulting energies with those obtained in the projection operator method.

In Table 1, we show the energies of 8Be (0+0^{+} and 2+2^{+}) measured from the α\alpha+α\alpha threshold energy by changing δ\delta. It is found that the two methods give the same energies of the 0+0^{+} and 2+2^{+} states from weak to strong bindings with various values of δ\delta. These results indicate the reliability of the present coherent basis method.

Table 1: Energies of 8Be (0+0^{+} and 2+2^{+}) measured from the α\alpha+α\alpha threshold energy in MeV, calculated in two methods; coherent basis method (CH) and projection operator method (PO). The parameter δ\delta is the enhancement factor of the α\alpha-α\alpha nuclear potential.
 
0+0^{+} 0+0^{+} 2+2^{+} 2+2^{+}
 δ\delta CH PO CH PO
 
0.050.05 0.072-0.072 0.072-0.072
0.100.10 0.593-0.593 0.593-0.593
0.150.15 1.256-1.256 1.256-1.256
0.200.20 2.065-2.065 2.065-2.065
0.250.25 3.026-3.026 3.026-3.026
0.300.30 4.147-4.147 4.147-4.147 0.954-0.954 0.954-0.954
0.350.35 5.430-5.430 5.430-5.430 2.175-2.175 2.175-2.175
0.400.40 6.880-6.880 6.880-6.880 3.553-3.553 3.553-3.553
 

Next, we keep δ=0\delta=0 in the α\alpha-α\alpha nuclear interaction and describe the unbound states of 8Be in the complex scaling. We solve the complex-scaled eigenvalue problem in Eq. (18) for 2α\alpha of 8Be (0+0^{+}, 2+2^{+}, 4+4^{+}, and 6+6^{+}). In Fig. 2, we show the energy eigenvalues {Enθ}\{E_{n}^{\theta}\} of four spin states on the complex energy plane. The scaling angle θ\theta is optimized in each spin state from the stationary condition of the energy eigenvalues of resonances on the complex energy plane with respect to θ\theta. This condition gives θ=16,18,20\theta=16^{\circ},18^{\circ},20^{\circ}, and 2525^{\circ} for 0+,2+,4+0^{+},2^{+},4^{+}, and 6+6^{+}, respectively. We show two kinds of solutions obtained in the coherent basis method (CH) and projection operator method (PO) in the 0+0^{+} and 2+2^{+} states. For 4+4^{+} and 6+6^{+}, the results obtained in the coherent basis method are shown. The continuum states are discretized along a straight line and we obtain one resonance in each state deviating from the line of the continuum states. In Fig. 2, the discretized continuum states also agree with each other by using the same range parameters in the relative wave function of 2α2\alpha.

In Table 2, we list the resonance energies and decay widths of four resonances of 8Be obtained in the coherent basis method in comparison with the projection operator method. We also include the experimental data. It is found that resonance energies and decay widths of two states of 8Be agree with each other in the two methods. These results mean the reliability of the coherent basis method to describe resonances with complex scaling.

Table 2: Resonance parameters of 8Be measured from the α\alpha+α\alpha threshold energy in MeV, in the coherent basis method (CH) and the projection operator method (PO). The experimental values (Exp.) are in the square brackets [38, 39].
 
 J±J^{\pm} energy decay width
 
CH 0.294~{}~{}~{}0.294 0.0140.014
0+0^{+} PO 0.296~{}~{}~{}0.296 0.0150.015
Exp. [0.0918] [5.57(25)×1065.57(25)\times 10^{-6}]
CH 3.01~{}~{}~{}3.01 1.65~{}~{}1.65
2+2^{+} PO 3.00~{}~{}~{}3.00 1.67~{}~{}1.67
Exp. [3.12(1)] [1.513(15)]
4+4^{+} CH 12.13~{}12.13 5.19~{}~{}5.19
Exp. [11.44(15)] [\approx 3.5]
6+6^{+} CH 30.49~{}30.49 37.88~{}37.88
Exp. [\approx 28] [\approx 20]
 
Refer to caption
Refer to caption
Figure 2: Energy eigenvalues of 8Be (top: 0+0^{+} and 2+2^{+}, bottom: 4+4^{+} and 6+6^{+}) for the coherent basis method (CH, solid symbols) and the projection operator method (PO, open symbols) on the complex energy plane, measured from the α\alpha+α\alpha threshold energy. Scaling angle θ\theta is taken as 16 (0+0^{+}), 18 (2+2^{+}), 20 (4+4^{+}), and 25 (6+6^{+}). The eigenvalues deviated from the line of discretized continuum states are resonances.

III.2 Phase shifts

We calculate the eigenstates of the asymptotic Hamiltonian H0θH_{0}^{\theta} of 2α2\alpha using Eq. (28) to obtain the continuum level densities and phase shifts in the coherent basis method. We employ the same set of the dilation parameters {βi}\{\beta_{i}\} as used in the calculation with the full Hamiltonian HθH^{\theta} and set the same scaling angle θ\theta for each state.

Using the energy eigenvalues {Enθ}\{E_{n}^{\theta}\} and {E0,nθ}\{E_{0,n}^{\theta}\} of 2α\alpha, we calculate the continuum level density, Δ(E)\Delta(E), and evaluate the phase shift of the α\alphaα\alpha scattering by integrating Δ(E)\Delta(E) in Eq. (27). In Fig. 3, we show the phase shifts of the four states obtained, where we put the arrows at the resonance energies of four states shown in Table 2.

The resulting phase shifts with dashed or dotted lines are obtained in the coherent basis method, and they agree with the gray lines obtained in the projection operator method for the 0+0^{+} and 2+2^{+} states. In each state, the energy at the maximum derivative of the phase shift is close to the resonance energy shown in the arrow. From these results, one can apply the present coherent basis method to the scattering problem between various nuclear clusters with complex scaling. One does not need the projection operator to eliminate the Pauli-forbidden states between clusters, which are automatically removed in the coherent basis method.

Refer to caption
Refer to caption
Figure 3: Phase shifts of the α\alphaα\alpha scattering (0+0^{+}, 2+2^{+}, 4+4^{+}, and 6+6^{+}) in the center-of-mass frame. The lines using dashed or dotted ones are the results in the coherent basis method and the gray solid lines for 0+0^{+}, and 2+2^{+} are the ones in the projection operator method. The upper arrows from the bottom indicate the resonance energies of 0+0^{+}, 2+2^{+}, 4+4^{+}, and 6+6^{+} in Table 2.

IV Discussion

We discuss the application of the present coherent basis method to the multicluster system beyond the two-cluster case. We shall consider the 3α3\alpha system for 12C with two Jacobi coordinates of the α\alpha-α\alpha and 2α2\alpha-α\alpha systems. We adopt the SU(3) representation for 12C with the coherent HO basis states [11, 40], which is defined as

ΦQ,(λ,μ),JKMβ=exp(12βD^)ΦQ,(λ,μ),JKM,D^=i=12D^i,\begin{split}\Phi_{Q,(\lambda,\mu),JKM}^{\beta}&=\exp\left(\frac{1}{2}\beta\hat{D}^{\dagger}\right)\,\Phi_{Q,(\lambda,\mu),JKM},\\ \hat{D}^{\dagger}&=\sum_{i=1}^{2}\hat{D}_{i}^{\dagger},\end{split} (30)

where i=1(2)i=1(2) is for the α\alpha-α\alpha (2α2\alpha-α\alpha) system with a quanta Ni=2ni+iN_{i}=2n_{i}+\ell_{i}. The total quanta of the basis state is given as Q=N1+N2Q=N_{1}+N_{2} with the quanta of each Jacobi coordinate under the irreducible SU(3) representation of (λ,μ)(\lambda,\mu) in the total spin JJ with the KK-quantum number. The total raising operator D^\hat{D}^{\dagger} is a summation of those for each Jacobi coordinate with the single dilation parameter β\beta in the exponent. Using Eq. (30), the basis state for the 3α3\alpha system is expressed as the product of the relative wave functions with the coherent basis states ΦNiiγiβ\Phi_{N_{i}\,\ell_{i}\,\gamma_{i}}^{\beta} as

ΦQ,(λ,μ)JKMβ=N1,N2CN1N2(λ,μ)×1,2(N1,0)1,(N2,0)2||(λ,μ)JK×[e12βD^1ΦN11γ1,e12βD^2ΦN22γ2]JM=N1,N2CN1N2(λ,μ)×1,2(N1,0)1,(N2,0)2||(λ,μ)JK×[ΦN11γ1β,ΦN22γ2β]JM,\begin{split}\Phi^{\beta}_{Q,(\lambda,\mu)JKM}&=\sum_{N_{1},N_{2}}C_{N_{1}N_{2}(\lambda,\mu)}\\ &\times\sum_{\ell_{1},\ell_{2}}\langle(N_{1},0)\ell_{1},(N_{2},0)\ell_{2}||(\lambda,\mu)JK\rangle\,\\ &\times\left[e^{\frac{1}{2}\beta\hat{D}^{\dagger}_{1}}\,\Phi_{N_{1}\,\ell_{1}\,\gamma_{1}},e^{\frac{1}{2}\beta\hat{D}^{\dagger}_{2}}\,\Phi_{N_{2}\,\ell_{2}\,\gamma_{2}}\right]_{JM}\\ &=\sum_{N_{1},N_{2}}C_{N_{1}N_{2}(\lambda,\mu)}\\ &\times\sum_{\ell_{1},\ell_{2}}\langle(N_{1},0)\ell_{1},(N_{2},0)\ell_{2}||(\lambda,\mu)JK\rangle\,\\ &\times\left[\Phi_{N_{1}\,\ell_{1}\,\gamma_{1}}^{\beta},\Phi_{N_{2}\,\ell_{2}\,\gamma_{2}}^{\beta}\right]_{JM},\end{split} (31)

where γ1=2ν\gamma_{1}=2\nu for α\alpha-α\alpha and γ2=8ν/3\gamma_{2}=8\nu/3 for 2α2\alpha-α\alpha are the HO range parameters in each relative motion, and ||\langle\cdots||\cdots\rangle is a SU(3) Clebsch Gordan coefficient. The specific coefficient CN1N2(λ,μ)C_{N_{1}N_{2}(\lambda,\mu)} is determined from the quanta in each relative motion and the (λ,μ)(\lambda,\mu) representation. The total variational wave function is a superposition of the above basis states with various values of the quanta QQ, N1N_{1}, N2N_{2} with (λ,μ)(\lambda,\mu) and the dilation parameter β\beta. It is noted that the common β\beta is used in the two relative motions in the single basis state. This condition comes to keep the symmetry of the identical α\alpha clusters, which fixes the ratio of the range parameters of the coherent basis states for the two Jacobi coordinates to γ1/γ2\gamma_{1}/\gamma_{2}.

We show the case of Q=8Q=8, (λ,μ)=(0,4)(\lambda,\mu)=(0,4), J=0J=0 and K=0K=0 for 12C, which uniquely gives N1=4N_{1}=4 and N2=4N_{2}=4, as

Φ8,(0,4)β=1=2=0,2,4(4,0)1,(4,0)2||(0,4)00×[Φ41γ1β,Φ42γ2β]00,\begin{split}\Phi^{\beta}_{8,(0,4)}&=\sum_{\ell_{1}=\ell_{2}=0,2,4}\langle(4,0)\ell_{1},(4,0)\ell_{2}||(0,4)00\rangle\\ &\times\left[\Phi_{4\,\ell_{1}\,\gamma_{1}}^{\beta},\Phi_{4\,\ell_{2}\,\gamma_{2}}^{\beta}\right]_{00},\end{split} (32)
(4,0)0,(4,0)0||(0,4)00=815,(4,0)2,(4,0)2||(0,4)00=435,(4,0)4,(4,0)4||(0,4)00=35.\begin{split}\langle(4,0)0,(4,0)0||(0,4)00\rangle&=\frac{8}{15},\\ \langle(4,0)2,(4,0)2||(0,4)00\rangle&=\frac{4}{3\sqrt{5}},\\ \langle(4,0)4,(4,0)4||(0,4)00\rangle&=\frac{3}{5}.\end{split} (33)

We also define the basis states for the linear-chain states of 12C, in which the lowest total-quanta is Q=12Q=12 with (λ,μ)=(12,0)(\lambda,\mu)=(12,0). In this configuration, the sets of the quanta (N1,N2)(N_{1},N_{2}) are given as (4, 8), (6, 6), (8, 4), and (10, 2). In a similar way, extending the 3α3\alpha case, the heavier multi-α\alpha cluster states can be constructed systematically in the SU(3) representation with the coherent basis states. We plan to investigate the 3α\alpha structure in 12C in the present framework in the future.

V Summary

We presented a new scheme to construct the Pauli-allowed states in nuclei with the harmonic oscillator (HO) basis states. We introduced a generalized coherent state of the HO basis state in terms of the raising operator 𝒂^𝒂^\hat{\mbox{\boldmath$a$}}^{\dagger}\cdot\hat{\mbox{\boldmath$a$}}^{\dagger} in the exponential form. This basis state results in the HO basis state with the same quanta, but with the changeable range parameters, namely, the radial dilation character. This property is important and controlled by one parameter, which we call the dilation parameter. This coherent basis state is automatically orthogonal to the lower quanta state and represents the short-range and long-range properties of the particle motion from the dilation property of the basis state. In this study, we utilized this property to treat the Pauli-allowed states appearing in relative motion of nuclear cluster systems. We also extend this framework to treat the resonances and the cluster-cluster scattering in the complex scaling.

We show the application to the 2α2\alpha system of 8Be in the orthogonality condition model. We compare the results in the coherent basis method with the conventional projection operator method, in which the projection operator is imposed in the Hamiltonian to obtain the Pauli-allowed states. It is confirmed that the present coherent basis method gives reasonable solutions of resonance energies, decay widths, and the phase shifts of the α\alpha-α\alpha scattering, which agree with those obtained in the projection operator method. These results indicate the reliability of the coherent basis method.

We further discuss the extension of the present method to the multicluster systems and explain the basic framework of the 3α3\alpha system of 12C. We adopt the SU(3) representation of the HO basis states in the relative motions with the Jacobi coordinates and introduce the coherent basis states in each relative motion with a common dilation parameter. It would be interesting to apply this framework to investigate the multi-α\alpha cluster states of nuclei.

Acknowledgments

This work was supported by JSPS KAKENHI Grants No. JP20K03962 and No. JP22K03643.

Appendix A Generalized coherent state

We formulate the generalized coherent state [9] of the harmonic oscillator (HO) basis state using the raising operator D^=𝒂^𝒂^\hat{D}^{\dagger}=\hat{\mbox{\boldmath$a$}}^{\dagger}\cdot\hat{\mbox{\boldmath$a$}}^{\dagger} [10]. The HO basis state ϕnm(𝒓,ν)\phi_{n\ell m}(\mbox{\boldmath$r$},\nu) with a range ν=1/b2\nu=1/b^{2} is usually defined using the associated Laguerre polynomials Ln(+1/2)(νr2)L_{n}^{(\ell+1/2)}(\nu r^{2}) as follows

ϕnm(𝒓,ν)=Nn(ν)e12νr2Ln(+1/2)(νr2)𝒴m(𝒓),Nn(ν)=ν+3/2 2+2(2n)!!π(2n+2+1)!!,\begin{split}\phi_{n\ell m}(\mbox{\boldmath$r$},\nu)&=N_{n\ell}(\nu)\,e^{-\frac{1}{2}\nu r^{2}}L_{n}^{(\ell+1/2)}(\nu r^{2})\,{\cal Y}_{\ell m}(\mbox{\boldmath$r$}),\\ N_{n\,\ell}(\nu)&=\sqrt{\frac{\nu^{\ell+3/2}\,2^{\ell+2}\,(2n)!!}{\sqrt{\pi}\,(2n+2\ell+1)!!}},\end{split} (34)

where nn represents the number of nodes in the radial wave function, and N=2n+N=2n+\ell is a principal quantum number. First, we start from the generating function for the associated Laguerre polynomials with α=+1/2\alpha=\ell+1/2 as

eνr2t/(1t)(1t)α+1=m=0Lm(α)(νr2)tm,\begin{split}\frac{e^{-\nu r^{2}\,t/(1-t)}}{(1-t)^{\alpha+1}}&=\sum_{m=0}^{\infty}L_{m}^{(\alpha)}(\nu r^{2})\,t^{m},\end{split} (35)

where |t|<1|t|<1. We introduce the following nn-th derivative of the generating function 𝒮n{\cal S}_{n} with its expansion;

𝒮n=1n!dndtn{eνr2t/(1t)(1t)α+1}=1n!m=0Lm(α)(νr2)dntmdtn=k=0Ln+k(α)(νr2)(n+k)!k!n!tk,\begin{split}{\cal S}_{n}&=\frac{1}{n!}\frac{d^{n}}{dt^{n}}\left\{\frac{e^{-\nu r^{2}\cdot t/(1-t)}}{(1-t)^{\alpha+1}}\right\}\\ &=\frac{1}{n!}\sum_{m=0}^{\infty}L_{m}^{(\alpha)}(\nu r^{2})\,\frac{d^{n}t^{m}}{dt^{n}}\\ &=\sum_{k=0}^{\infty}L_{n+k}^{(\alpha)}(\nu r^{2})\,\frac{(n+k)!}{k!\,n!}\,t^{k},\end{split} (36)

where m=n+km=n+k. It is also proven that 𝒮n{\cal S}_{n} is proportional to the associated Laguerre polynomials with the order of nn and the argument of νr2/(1t)\nu r^{2}/(1-t) as;

𝒮n=eνr2t/(1t)(1t)n+α+1Ln(α)(νr21t).\begin{split}{\cal S}_{n}&=\frac{e^{-\nu r^{2}\,t/(1-t)}}{(1-t)^{n+\alpha+1}}\,L_{n}^{(\alpha)}\left(\frac{\nu r^{2}}{1-t}\right).\end{split} (37)

This formula can be confirmed in the mathematical induction using the relation of 𝒮n+1=(d𝒮n/dt)/(n+1){\cal S}_{n+1}=(d{\cal S}_{n}/dt)\,/(n+1) and the properties of the associated Laguerre polynomials. From two expressions of 𝒮n{\cal S}_{n} in Eqs. (36) and (37), we obtain the following relation

Ln(α)(νr21t)=(1t)n+α+1exp(t1tνr2)×k=0(n+k)!k!n!tkLn+k(α)(νr2).\begin{split}L_{n}^{(\alpha)}\left(\frac{\nu r^{2}}{1-t}\right)&=(1-t)^{n+\alpha+1}\exp\left(\frac{t}{1-t}\nu r^{2}\right)\\ &\times\sum_{k=0}^{\infty}\frac{(n+k)!}{k!\,n!}t^{k}\,L_{n+k}^{(\alpha)}\left(\nu r^{2}\right).\end{split} (38)

This formula means that the associated Laguerre polynomial with the argument of νr2/(1t)\nu r^{2}/(1-t) and the order nn is expanded by tkt^{k} in terms of those with the argument of νr2\nu r^{2} and the order of n+kn+k. This property is applicable to the HO basis states to connect the HO basis states with different range parameters. Hereafter we define β=t\beta=-t for the dilation parameter in the coherent basis state and use this relation in the HO basis state with the range ν/(1+β)\nu/(1+\beta).

Next, we discuss the generalized coherent state with the dilation parameter β\beta, which can be expanded in the HO basis states using Eq. (1), the quanta of which is larger than or equal to NN, because of the raising operator D^=𝒂^𝒂^\hat{D}^{\dagger}=\hat{\mbox{\boldmath$a$}}^{\dagger}\cdot\hat{\mbox{\boldmath$a$}}^{\dagger} as

ϕnmβ(𝒓,ν)=exp(12βD^)ϕnm(𝒓,ν)=k=0βk2kk!An(𝒂^𝒂^)n+k𝒴m(𝒂^)ϕ0(𝒓,ν)=k=0βk2kk!AnAn+kϕn+km(𝒓,ν).\begin{split}\phi^{\beta}_{n\ell m}(\mbox{\boldmath$r$},\nu)&=\exp\left(\frac{1}{2}\beta\hat{D}^{\dagger}\right)\phi_{n\ell m}(\mbox{\boldmath$r$},\nu)\\ &=\sum_{k=0}^{\infty}\frac{\beta^{k}}{2^{k}k!}\,A_{n\ell}\left(\hat{\mbox{\boldmath$a$}}^{\dagger}\cdot\hat{\mbox{\boldmath$a$}}^{\dagger}\right)^{n+k}{\cal Y}_{\ell m}(\hat{\mbox{\boldmath$a$}}^{\dagger})\,\phi_{0}(\mbox{\boldmath$r$},\nu)\\ &=\sum_{k=0}^{\infty}\frac{\beta^{k}}{2^{k}k!}\,\frac{A_{n\,\ell}}{A_{n+k\,\ell}}\,\phi_{n+k\,\ell m}(\mbox{\boldmath$r$},\nu).\end{split} (39)

We get the following relation for the ratio of the coefficients An/An+kA_{n\,\ell}/A_{n+k\,\ell} as

AnAn+k=(1)k(2n+2k+2+2)!(n++1)!(n+k)!(n+k++1)!(2n+2+2)!n!.\begin{split}\frac{A_{n\,\ell}}{A_{n+k\,\ell}}&=(-1)^{k}\sqrt{\frac{(2n+2k+2\ell+2)!\,(n+\ell+1)!\,(n+k)!}{(n+k+\ell+1)!\,(2n+2\ell+2)!\,n!}}.\end{split} (40)

On the other hand, we define the following function φnmβ\varphi^{\beta}_{n\ell m} with a normalization constant CC and the HO basis state with the range ν/(1+β)\nu/(1+\beta),

φnmβ(𝒓,ν)=Cexp(β2(1+β)νr2)ϕnm(𝒓,ν1+β).\begin{split}\varphi^{\beta}_{n\ell m}(\mbox{\boldmath$r$},\nu)&=C\,\exp\left(\frac{\beta}{2(1+\beta)}\nu r^{2}\right)\,\phi_{n\ell m}(\mbox{\boldmath$r$},\frac{\nu}{1+\beta}).\end{split} (41)

Using Eq. (38) with β=t\beta=-t,

φnmβ(𝒓,ν)=Cexp(ν(1β)2(1+β)r2)Nn(ν1+β)𝒴m(𝒓)×(1+β)n++3/2exp(β1+βνr2)×k=0(n+k)!k!n!(β)kLn+k(+1/2)(νr2)=C(1+β)n++3/2k=0(n+k)!n!(β)kk!×Nn(ν1+β)Nn+k(ν)ϕn+km(𝒓,ν).\begin{split}\varphi^{\beta}_{n\ell m}(\mbox{\boldmath$r$},\nu)&=C\,\exp\left(-\frac{\nu(1-\beta)}{2(1+\beta)}r^{2}\right)\,N_{n\ell}\left(\frac{\nu}{1+\beta}\right)\,{\cal Y}_{\ell m}(\mbox{\boldmath$r$})\\ &\times(1+\beta)^{n+\ell+3/2}\exp\left(-\frac{\beta}{1+\beta}\nu r^{2}\right)\\ &\times\sum_{k=0}^{\infty}\frac{(n+k)!}{k!\,n!}(-\beta)^{k}\,L_{n+k}^{(\ell+1/2)}\left(\nu r^{2}\right)\\ &=C\,(1+\beta)^{n+\ell+3/2}\sum_{k=0}^{\infty}\frac{(n+k)!}{n!}\,\frac{(-\beta)^{k}}{k!}\,\\ &\times\frac{N_{n\ell}\left(\frac{\nu}{1+\beta}\right)}{N_{n+k\,\ell}(\nu)}\,\phi_{n+k\,\ell m}(\mbox{\boldmath$r$},\nu).\end{split} (42)

Here,

Nn(ν1+β)Nn+k(ν)=(2n)!!(2n+2k+2+1)!!(1+β)+3/2(2n+2k)!!(2n+2+1)!!=1(1+β)+3/2n!(1)k2k(n+k)!AnAn+k.\begin{split}\frac{N_{n\ell}\left(\frac{\nu}{1+\beta}\right)}{N_{n+k\,\ell}(\nu)}&=\sqrt{\frac{(2n)!!\,(2n+2k+2\ell+1)!!}{(1+\beta)^{\ell+3/2}\,(2n+2k)!!\,(2n+2\ell+1)!!}}\\ &=\frac{1}{\sqrt{(1+\beta)^{\ell+3/2}}}\,\frac{n!\,(-1)^{k}}{2^{k}\,(n+k)!}\,\frac{A_{n\,\ell}}{A_{n+k\,\ell}}.\end{split} (43)

Hence we can rewrite φnmβ\varphi^{\beta}_{n\ell m} using Eq. (39) as

φnmβ(𝒓,ν)=C(1+β)2n++3/2k=0βk2kk!AnAn+k×ϕn+km(𝒓,ν)=C(1+β)N¯exp(12βD^)ϕnm(𝒓,ν)=C(1+β)N¯ϕnmβ(𝒓,ν),\begin{split}\varphi^{\beta}_{n\ell m}(\mbox{\boldmath$r$},\nu)&=C\,\sqrt{(1+\beta)^{2n+\ell+3/2}}\,\sum_{k=0}^{\infty}\frac{\beta^{k}}{2^{k}\,k!}\,\frac{A_{n\,\ell}}{A_{n+k\,\ell}}\\ &\times\phi_{n+k\,\ell m}(\mbox{\boldmath$r$},\nu)\\ &=C\,\sqrt{(1+\beta)^{\bar{N}}}\,\exp\left(\frac{1}{2}\beta\hat{D}^{\dagger}\right)\,\phi_{n\ell m}(\mbox{\boldmath$r$},\nu)\\ &=C\,\sqrt{(1+\beta)^{\bar{N}}}\,\phi^{\beta}_{n\ell m}(\mbox{\boldmath$r$},\nu),\\ \end{split} (44)

where N¯=2n++3/2=N+3/2\bar{N}=2n+\ell+3/2=N+3/2. Imposing the relation of φnmβ=ϕnmβ\varphi^{\beta}_{n\ell m}=\phi^{\beta}_{n\ell m}, we can determine CC as

C=1(1+β)N¯.\begin{split}C&=\frac{1}{\sqrt{(1+\beta)^{\bar{N}}}}.\end{split} (45)

Finally, we define the generalized coherent state of the HO basis state.

ϕnmβ(𝒓,ν)=exp(12βD^)ϕnm(𝒓,ν)=1(1+β)N¯exp(β2(1+β)νr2)×ϕnm(𝒓,ν1+β).\begin{split}\phi^{\beta}_{n\ell m}(\mbox{\boldmath$r$},\nu)&=\exp\left(\frac{1}{2}\beta\hat{D}^{\dagger}\right)\,\phi_{n\ell m}(\mbox{\boldmath$r$},\nu)\\ &=\frac{1}{\sqrt{(1+\beta)^{\bar{N}}}}\,\exp\left(\frac{\beta}{2(1+\beta)}\nu r^{2}\right)\\ &\times\phi_{n\ell m}(\mbox{\boldmath$r$},\frac{\nu}{1+\beta}).\end{split} (46)

It is noted that the above coherent basis state is not normalized, and one can normalize it in the calculation of the norm matrix element.

Appendix B Kinetic energy

We give the formula of the matrix element of the kinetic energy TT with a reduced mass μ\mu in the generalized coherent basis states with the independent values of β\beta and nn in the bra and ket states.

ϕnβ(ν)|T|ϕnβ(ν)=2νβ2μfnnββ(ν),fnnββ(ν)=(1β)2AnGnn+1ββ(ν)+(1β)(1+β)BnGnnββ(ν)+(1+β)2CnGnn1ββ(ν),\begin{split}\langle\phi_{n^{\prime}\ell}^{\beta^{\prime}}(\nu)|T|\phi_{n\ell}^{\beta}(\nu)\rangle&=\frac{\hbar^{2}\nu_{\beta}}{2\mu}\,f_{n^{\prime}n\,\ell}^{\beta^{\prime}\beta}(\nu),\\ f_{n^{\prime}n\,\ell}^{\beta^{\prime}\beta}(\nu)&=\left(1-\beta\right)^{2}\,A_{n}\,G_{n^{\prime}n+1\,\ell}^{\beta^{\prime}\beta}(\nu)\\ &+\left(1-\beta\right)\left(1+\beta\right)\,B_{n}\,G_{n^{\prime}n\,\ell}^{\beta^{\prime}\beta}(\nu)\\ &+\left(1+\beta\right)^{2}\,C_{n}\,G_{n^{\prime}n-1\,\ell}^{\beta^{\prime}\beta}(\nu),\end{split} (47)

where

Gnnββ(ν)=ϕn(νβ)|e(γβ+γβ)r2|ϕn(νβ)(1+β)N¯(1+β)N¯,An=(n+1)(n++32),Bn=2n++32,Cn=n(n++12),νβ=ν1+β,γβ=β1+βν2.\begin{split}G_{n^{\prime}n\,\ell}^{\beta^{\prime}\beta}(\nu)&=\frac{\langle\phi_{n^{\prime}\ell}(\nu_{\beta^{\prime}})|e^{\left(\gamma_{\beta^{\prime}}+\gamma_{\beta}\right)r^{2}}|\phi_{n\ell}(\nu_{\beta})\rangle}{\sqrt{(1+\beta^{\prime})^{\bar{N}^{\prime}}(1+\beta)^{\bar{N}}}},\\ A_{n}&=\sqrt{(n+1)(n+\ell+\frac{3}{2})},\\ B_{n}&=2n+\ell+\frac{3}{2},\quad C_{n}=\sqrt{n(n+\ell+\frac{1}{2})},\\ \nu_{\beta}&=\frac{\nu}{1+\beta},\qquad\gamma_{\beta}=\frac{\beta}{1+\beta}\,\frac{\nu}{2}.\\ \end{split} (48)

References

References

  • [1] K. Ikeda, N. Takigawa, and H. Horiuchi, Prog. Theor. Phys. Suppl. E68, 464 (1968).
  • [2] H. Horiuchi, K. Ikeda, and K. Katō, Prog. Theor. Phys. Suppl. 192, 1 (2012).
  • [3] M. Freer, H. Horiuchi, Y. Kanada-En’yo, D. Lee, and Ulf-G. Meißner, Rev. Mod. Phys. 90, 035004 (2018).
  • [4] J. A. Wheeler, Phys. Rev. 52, 1083 (1937), and Phys. Rev. 52, 1107 (1937).
  • [5] H. Horiuchi, Prog. Theor. Phys. Suppl. 62, 90 (1977).
  • [6] S. Saito, Prog. Theor. Phys. 40, 893 (1968), Prog. Theor. Phys. 41, 705 (1969), Prog. Theor. Phys. Suppl. 62, 11 (1977).
  • [7] A. T. Kruppa and K. Katō, Prog. Theor. Phys. 84, 1145 (1990).
  • [8] V. I. Kukulin, and V. N. Pomenertsev, Ann. Phys. (NY) 111, 330 (1978).
  • [9] A. Perelomov, Generalized Coherent States and their Applications (Springer-Verlag, Berlin, 1986).
  • [10] D. J. Rowe, Rep. Prog. Phys. 48, 1419 (1985).
  • [11] T. Yoshida, N. Itagaki, and K. Katō, Phys. Rev. C 83, 024301 (2011).
  • [12] T. Myo, Y. Kikuchi, H. Masui, and K. Katō, Prog. Part. Nucl. Phys. 79, 1 (2014).
  • [13] V. Bergmann and M. Moshinsky, Nucl. Phys. 18, 697 (1960).
  • [14] M. Moshinsky, Y. F. Smirnov, The Harmonic Oscillator in Modern Physics, Contemporary Concepts in Physics, Vol.9 (Harwood Academic, Reading, 1996).
  • [15] M. Kamimura, Phys. Rev. C 38, 621 (1988).
  • [16] H. Kameyama, M. Kamimura and Y. Fukushima, Phys. Rev. C 40, 974 (1989).
  • [17] S. Aoyama, T. Myo, K. Katō, and K. Ikeda, Prog. Theor. Phys. 116, 1 (2006).
  • [18] T. Myo, S. Sugimoto, K. Katō, H. Toki and K. Ikeda, Prog. Theor. Phys. 117, 257 (2007).
  • [19] T. Myo, A. Umeya, H. Toki, and K. Ikeda, Phys. Rev. C 84, 034315 (2011).
  • [20] K. Katō, V. S. Vasilevsky and N. Zh. Takibayev, Presentation at the Workshop ”Nuclear Physics, Nuclear Astrophysics and Cosmic Rays” in Almaty, Kazakhstan, 2019.
  • [21] E. W. Schmid, and K. Wildermuth, Nucl. Phys. 26, 463 (1961).
  • [22] K. Fukatsu, K. Katō, Prog. Theor. Phys. 87, 151 (1992).
  • [23] C. Kurokawa, K. Katō, Phys. Rev. C 71, 021301(R) (2005).
  • [24] C. Kurokawa, K. Katō, Nucl. Phys. A 792, 87 (2007).
  • [25] Y. Funaki, T. Yamada, H. Horiuchi, G. Röpke, P. Schuck, and A. Tohsaki, Phys. Rev. Lett. 101, 082502 (2008).
  • [26] Y. K. Ho, Phys. Rep. 99, 1 (1983).
  • [27] N. Moiseyev, Phys. Rep. 302, 212 (1998).
  • [28] T. Myo and K. Katō, Prog. Theor. Exp. Phys. 2020, 12A101 (2020).
  • [29] J. Aguilar and J.M. Combes, Commun. Math. Phys. 22, 269 (1971), E. Balslev and J.M. Combes, Commun. Math. Phys. 22, 280 (1971).
  • [30] T. Berggren,  Nucl. Phys. A109, 265 (1968).
  • [31] T. Myo, A. Ohnishi, and K. Katō, Prog. Theor. Phys. 99, 801 (1998).
  • [32] R. Suzuki, T. Myo, and K. Katō, Prog. Theor. Phys. 113, 1273 (2005).
  • [33] M. Odsuren, K. Katō, M. Aikawa, and T. Myo, Phys. Rev. C 89, 034322 (2014).
  • [34] M. Odsuren, T. Myo, Y. Kikuchi, M. Teshigawara, and K. Katō, Phys. Rev. C 104, 014325 (2021).
  • [35] R. D. Levine, Quantum Mechanics of Molecular Rate Processes (Clarendon Press, Oxford, 1969), Chap. 2.5.
  • [36] M. Kamimura, Prog. Theor. Phys. Suppl. 62, 236 (1977).
  • [37] T. Myo and H. Takemoto, Phys. Rev. C 107, 064308 (2023).
  • [38] D. R. Tilley, J. H. Kelley, J. L. Godwin, D. J. Millener, J. E. Purcell, C. G. Sheu, and H. R. Weller, Nucl. Phys. A 745, 155 (2004).
  • [39] https://www.nndc.bnl.gov/nudat3/ .
  • [40] K. Katō, K. Fukatsu, H. Tanaka, Prog. Theor. Phys. 80, 663 (1988).