betaFeedback

WorksPhysics and Astronomy

Relations between three-particle interactions in nuclear matter to observable quantities

Abstract

In the first part of this paper, we use the framework of the Fermi liquid theory to derive model-independent relations between the slope parameters of the symmetry energy and of the incompressibility in nuclear matter to three-particle interaction parameters. Based on these relations, we present simple estimates and compare with the empirical information. In the second part, we discuss the general structure of the three-particle scattering amplitude in nuclear matter, and use methods similar to the Bethe-Brueckner-Goldstone theory to show how three-particle cluster diagrams emerge naturally in the Fermi liquid theory.

Introduction

Among the basic physical quantities which play a critical role in shaping the structure of nuclei and neutron stars are the symmetry energy (asa_s) [1, 2] and the incompressibility (KK) [3, 4] of nuclear systems. In nuclei, the density dependence of the symmetry energy and its associated pressure determine the neutron skin and the excitation modes of isovector character, in particular the electric dipole oscillations [5]. In neutron stars, the symmetry pressure determines the radius of the star, which is known to be strongly correlated to the neutron skin thickness of heavy nuclei [6]. In heavy ion collisions, the same physical quantity determines the isospin distribution among the reaction products [7]. On the other hand, the density dependence of the incompressibility and its associated pressure determine the nuclear excitation modes of isoscalar character, in particular the monopole oscillations [8], and play a critical role to determine the maximum mass of a neutron star [9]. In heavy ion collisions, the same physical quantity has critical impact on the collective flow of the produced particles [10].

The dependencies of the symmetry energy and the incompressibility on the baryon density (ρ\rho) are usually expressed in terms of the slope parameters L≡3ρdasdρL \equiv3\rho\frac{\mathrm{d}a_s}{\mathrm{d}\rho} and J≡3ρdKdρ−12KJ \equiv3\rho\frac{\mathrm{d}K}{\mathrm{d}\rho} - 12K, where the latter one is usually called the skewness [4]. Because the quantities asa_s and KK in turn are defined in terms of the second derivatives of the energy density w.r.t. ρ(3)=ρ(p)−ρ(n)\rho^{(3)} = \rho^{(p)} - \rho^{(n)} and ρ=ρ(p)+ρ(n)\rho= \rho^{(p)} + \rho^{(n)}, respectively, it follows that LL and JJ are related to the third derivatives of the energy density1 and, therefore, by extension of Landau’s hypothesis [11, 12], to the ℓ=0\ell= 0 moments of the three-nucleon in-medium forward scattering amplitude [13, 14]. The first purpose of this article is to present these relations in an exact (model independent) form. Because of the long history of three-particle interactions studies in nuclear matter [15, 16], it is desirable to know such model independent relations based on first principles. We will present semi-quantitative discussions on these relations in comparison to empirical information. The second purpose of our work is to discuss the physical content of the in-medium three-particle amplitude, the ℓ=0\ell= 0 moments of which enter into the model independent relations mentioned above. We will discuss how far the structure of the three-particle amplitude can be specified by using only its definition, and illustrate how further assumptions, similar to the ones used in the Bethe-Brueckner-Goldstone (BBG) theory [17], can be used to derive more detailed expressions.

For these two purposes we will use the general field theoretic formulation of the Fermi liquid theory due to Landau and Migdal [18, 19]. Two important merits of this approach are that it exploits general symmetries like gauge invariance and Galilei invariance to derive relations between interaction parameters (Landau-Migdal parameters) and observable quantities which are in principle exact, and that it naturally leads to the shell structure in finite systems. Nowadays, several more merits are well known: The theory can be derived from the renormalization group [20], it has a close relationship to the more popular Skyrme approach [21], it can be extended to relativistic field theory [22, 23], and it has deep impacts on nuclear electromagnetic properties [24, 25]. Last but not least, the theory played an important role to promote the exchange of people, ideas, and understanding between very different societies of the world.

The layout of the paper is as follows: In § 1.1 we discuss our relations between the quantities LL and KK mentioned above and the three-nucleon ss-wave forward scattering amplitudes. Semi-quantitative estimates will be presented in comparison to the empirical information. In Sec. III we will discuss the physical content of the three-particle scattering amplitude, and make connection to the BBG theory. In Sec. IV we summarize our results.

Basic relations and semi-quantitative discussions

In this section we summarize the main points in the derivation of the model independent relations between the slope parameters of the symmetry energy and of the incompressibility to the three-nucleon ss-wave interaction parameters, and present semi-quantitative estimates in comparison to empirical values. More detailed discussions can be found in Refs. [13, 14].

Model independent relations

We start from the standard relationships [26] for the second derivatives of the energy density (EE) w.r.t. the isoscalar and isovector densities defined in the previous section:2

∂2E∂ρ2=π22pFM∗+f0≡K9ρ,(1)\frac{\partial^2 E}{\partial\rho^2} = \frac{\pi^2}{2p_F M^*} + f_0 \equiv\frac{K}{9\rho}, \tag*{(1)}
∂2E∂ρ(3)2=π22pFM∗+f0′≡2asρ.(2)\frac{\partial^2 E}{\partial\rho^{(3)2}} = \frac{\pi^2}{2p_F M^*} + f'_0 \equiv\frac{2a_s}{\rho}. \tag*{(2)}

Here M∗≡M∗(k=pF)M^* \equiv M^*(k=p_F) is the Landau effective mass in the isospin symmetric limit (pFp_F denotes the Fermi momentum in this limit), which is defined via the quasiparticle velocity as usual [26] by ∂εk/(∂k)≡k/M∗(k)\partial\varepsilon_k/(\partial k) \equiv k/M^*(k), where εk\varepsilon_k is the quasiparticle energy. The isoscalar and isovector combinations f0f_0 and f0′f'_0 are defined by [19]

f0=12(f0(pp)+f0(pn)),f0′=12(f0(pp)−f0(pn)),(3)f_0 = \frac{1}{2}\left(f_0^{(pp)}+f_0^{(pn)}\right), \qquad f'_0 = \frac{1}{2}\left(f_0^{(pp)}-f_0^{(pn)}\right), \tag*{(3)}

where f0(τ1τ2)f_0^{(\tau_1\tau_2)} are the ℓ=0\ell=0 moments of the two-particle forward scattering amplitudes

12ℓ+1fℓ(τ1τ2)(k1,k2)=∫dΩ24π(k^1⋅k^2)ℓf(τ1τ2)(k1,k2),(4)\frac{1}{2\ell+1}f_\ell^{(\tau_1\tau_2)}(k_1,k_2)=\int\frac{\mathrm{d}\Omega_2}{4\pi}\left(\hat{\mathbf{k}}_1\cdot\hat{\mathbf{k}}_2\right)^\ell f^{(\tau_1\tau_2)}(k_1,k_2), \tag*{(4)}

at k1=k2=pFk_1=k_2=p_F. Here the label τ\tau distinguishes between protons (τ=p\tau=p) and neutrons (τ=n\tau=n).

As we explained in the previous section, in order to obtain the slope parameters of the incompressibility and of the symmetry energy, we have to take the derivatives of Eqs. (1) and (2) w.r.t. ρ\rho. In the following we summarize the basic relations which are used in the calculation:

  • The variations of the quasiparticle energies ε(τ)(k)\varepsilon^{(\tau)}(k) w.r.t. the background densities are generally given by

δε(τ1)(k)δρ(τ2)=f0(τ1τ2)(k,pF(τ2)),(5)\frac{\delta\varepsilon^{(\tau_1)}(k)}{\delta\rho^{(\tau_2)}}=f_0^{(\tau_1\tau_2)}(k,p_F^{(\tau_2)}), \tag*{(5)}

where pF(τ)p_F^{(\tau)} are the Fermi momenta for protons and neutrons. By taking the derivative of Eq. (5) w.r.t. the momentum kk and setting k=pFk=p_F, we obtain in the isospin symmetric limit

∂M∗∂ρ=−M∗22pF∂f0∂pF+π22pF2∂M∗∂pF,(6)\frac{\partial M^*}{\partial\rho}=-\frac{M^{*2}}{2p_F}\frac{\partial f_0}{\partial p_F}+\frac{\pi^2}{2p_F^2}\frac{\partial M^*}{\partial p_F}, \tag*{(6)}
∂ΔM∗∂ρ(3)=−M∗2pF∂f0′∂pF+π2pF2∂ΔM∗∂pF,(7)\frac{\partial\Delta M^*}{\partial\rho^{(3)}}=-\frac{M^{*2}}{p_F}\frac{\partial f'_0}{\partial p_F}+\frac{\pi^2}{p_F^2}\frac{\partial\Delta M^*}{\partial p_F}, \tag*{(7)}

where ΔM∗≡M∗(p)−M∗(n)\Delta M^* \equiv M^*(p)-M^*(n). We note that the above relations include both the variations of the background densities, as expressed by the first terms on the r.h.s. originating from Eq. (5), and of the external momenta of the particles. We defined [13]

∂f0∂pF≡[(∂∂k1+∂∂k2)f0(k1,k2)]k1=k2=pF,\frac{\partial f_0}{\partial p_F}\equiv\left[\left(\frac{\partial}{\partial k_1}+\frac{\partial}{\partial k_2}\right)f_0(k_1,k_2)\right]_{k_1=k_2=p_F},

and similarly for f0′f'_0, and ∂M∗/∂pF≡(∂M∗(k)/∂k)k=pF\partial M^*/\partial p_F\equiv(\partial M^*(k)/\partial k)_{k=p_F}.

  • In analogy to Eq. (5), the variations of the ℓ=0\ell=0 two-particle forward scattering amplitudes for external momenta k1,k2k_1,k_2 w.r.t. the background densities are given by the ℓ=0\ell=0 moments of the three-particle forward scattering amplitudes, denoted by h0h_0 (h0′h'_0) for the isoscalar (isovector) case, where the third particle is on the Fermi surface (k3=pFk_3=p_F). If all three particles are taken on the Fermi surface, we obtain [14]

∂f0∂ρ=h0+π22pF2∂f0∂pF,∂f0′∂ρ=h0′+π22pF2∂f0′∂pF.(8)\frac{\partial f_0}{\partial\rho}=h_0+\frac{\pi^2}{2p_F^2}\frac{\partial f_0}{\partial p_F}, \qquad\frac{\partial f'_0}{\partial\rho}=h'_0+\frac{\pi^2}{2p_F^2}\frac{\partial f'_0}{\partial p_F}. \tag*{(8)}

The first terms on the r.h.s. of the above relations come from the variations of the background densities, and the second terms from the variations of the external particle momenta. The isoscalar and isovector combinations h0h_0 and h0′h'_0 are defined by

h0=14(h0(ppp)+3h0(ppn)),h_0=\frac{1}{4}\left(h_0^{(ppp)}+3h_0^{(ppn)}\right),
h0′=14(h0(ppp)−h0(ppn)).(9)h'_0=\frac{1}{4}\left(h_0^{(ppp)}-h_0^{(ppn)}\right). \tag*{(9)}

They are the ℓ=0\ell=0 moments of the three-particle forward scattering amplitudes

12ℓ+1hℓ(τ1τ2τ3)(k1,k2,k3)=∫dΩ24π∫dΩ34π(k^1⋅k^2)ℓh(τ1τ2τ3)(k1,k2,k3),(10)\frac{1}{2\ell+1}h_\ell^{(\tau_1\tau_2\tau_3)}(k_1,k_2,k_3)=\int\frac{\mathrm{d}\Omega_2}{4\pi}\int\frac{\mathrm{d}\Omega_3}{4\pi}\left(\hat{\mathbf{k}}_1\cdot\hat{\mathbf{k}}_2\right)^\ell h^{(\tau_1\tau_2\tau_3)}(k_1,k_2,k_3), \tag*{(10)}

at k1=k2=k3=pFk_1=k_2=k_3=p_F. The three-particle amplitudes on the r.h.s. of Eq. (10) are generally defined as the variations of the two-particle amplitudes w.r.t. the quasiparticle distribution functions:3

h(τ1τ2τ3)(k1,k2,k3)=δf(τ1τ2)(k1,k2)δnk3(τ3).(11)h^{(\tau_1\tau_2\tau_3)}(k_1,k_2,k_3)=\frac{\delta f^{(\tau_1\tau_2)}(k_1,k_2)}{\delta n_{k_3}^{(\tau_3)}}. \tag*{(11)}
  • Relations which follow from Galilei invariance play a very important role in the Fermi liquid theory. The best known relation follows from the requirement that under a Galilei transformation the single particle energies transform in the same way as a Hamiltonian in classical mechanics [27]. In the isospin symmetric limit, this gives the relation

kM∗(k)+2pF23π2f1(k,pF)=kM,(12)\frac{k}{M^*(k)}+\frac{2p_F^2}{3\pi^2}f_1(k,p_F)=\frac{k}{M}, \tag*{(12)}

where MM is the free nucleon mass. For k=pFk=p_F this becomes the familiar Landau effective mass relation [26]. A less well known relation follows from the requirement that the that under a Galilei transformation the two-particle scattering amplitudes are invariant. At the Fermi surface and in the isospin symmetric limit, this gives [13, 14]

pF∂∂pF(f0+13f1)+43f1+4pF33π2h1=0(13)p_F\frac{\partial}{\partial p_F}\left(f_0+\frac{1}{3}f_1\right)+\frac{4}{3}f_1+\frac{4p_F^3}{3\pi^2}h_1=0 \tag*{(13)}

and a similar relation with primes attached to all amplitudes. We can convert Eq. (13) to a more useful form by using Eq. (12) and its derivative w.r.t. kk at the Fermi surface, to obtain

pF∂f0∂pF−3π2pFM−M∗MM∗−π2M∗2∂M∗∂pF−4pF33π2h1(14)p_F\frac{\partial f_0}{\partial p_F}-\frac{3\pi^2}{p_F}\frac{M-M^*}{MM^*}-\frac{\pi^2}{M^{*2}}\frac{\partial M^*}{\partial p_F}-\frac{4p_F^3}{3\pi^2}h_1 \tag*{(14)}

This relation is used to eliminate the derivative ∂f0/∂pF\partial f_0/\partial p_F from the final expressions.

By using the above prerequisites, the derivative of Eq. (1) is calculated by using Eqs. (6), (8), (12), and (14). The derivative of Eq. (2) is obtained by using the same formulas, and in addition Eq. (7) to eliminate the derivative ∂f0′/(∂pF)\partial f'_0/(\partial p_F) in favor of the dimensionless parameter

μ≡ρ∂∂ρ(3)(ΔM∗M)(15)\mu\equiv\rho\frac{\partial}{\partial\rho^{(3)}}\left(\frac{\Delta M^*}{M}\right) \tag*{(15)}

Defining the dimensionless three-particle interaction parameters by [13]4

Hℓ=4pF4M∗3π4hℓ,Hℓ′=4pF4M∗3π4hℓ′(16)H_\ell=\frac{4p_F^4M^*}{3\pi^4}h_\ell,\qquad H'_\ell=\frac{4p_F^4M^*}{3\pi^4}h'_\ell \tag*{(16)}

the final results can be expressed as follows:

J=−9K+9pF2M×[(−3+83MM∗)−43MpFM∗2∂M∗∂pF+MM∗(H0−H1)](17)J=-9K+\frac{9p_F^2}{M}\times\left[\left(-3+\frac{8}{3}\frac{M}{M^*}\right)-\frac{4}{3}\frac{Mp_F}{M^{*2}}\frac{\partial M^*}{\partial p_F}+\frac{M}{M^*}(H_0-H_1)\right] \tag*{(17)}
L=3as−pF22M×[(1−23MM∗)+μ(MM∗)2−MM∗(H0′−13H1)](18)L=3a_s-\frac{p_F^2}{2M}\times\left[\left(1-\frac{2}{3}\frac{M}{M^*}\right)+\mu\left(\frac{M}{M^*}\right)^2-\frac{M}{M^*}\left(H'_0-\frac{1}{3}H_1\right)\right] \tag*{(18)}

One should note that Eqs. (17) and (18) are exact (model independent) relations.

Qualitative discussions

As we discussed in § 1, there have been many recent analyses on the empirical values of JJ and LL by using data on heavy nuclei, astrophysical observations, and heavy ion collisions. For definiteness we refer here to the values given in Figure 4 of Ref. [28] for JJ, and in Fig. 20 of Ref. [4] for LL, at normal nuclear matter density (ρ=0.155 fm−3\rho=0.155\ \mathrm{fm}^{-3}):

Figure 4. Contribution of third order in the ladder tt-matrix to the three-particle amplitude h~\tilde{h}. The diagram (a) is the second (next-to-leading) term in the Faddeev series. λijk\lambda_{ijk} is unity for even permutations of (123) and zero otherwise, and an independent sum over (ijk)(ijk) and (lmn)(lmn) is implied. The diagrams (b) and (c) arise from the interaction between the three given particles and the particles in the Fermi sea, and do not appear in the usual Faddeev series. εijk\varepsilon_{ijk} is the usual antisymmetric tensor, and a sum over i,j,k=1,2,3i,j,k=1,2,3 is implied. The two-particle tt-matrix in the ladder approximation is denoted by a black dot.

J=−255±245 MeV,L=59±16 MeV(19)J=-255\pm245\ \mathrm{MeV},\qquad L=59\pm16\ \mathrm{MeV} \tag*{(19)}

We mention that these values are consistent with most of the other analyses mentioned in § 1. Concerning the nucleon effective mass, most of the investigations during the last decades [4, 29] are consistent with the range M∗/M=0.85±0.15M^*/M=0.85\pm0.15 at the Fermi surface, and indicate that M∗(k)M^*(k) has a pronounced peak very close to the Fermi surface[30, 31], i.e., ∂M∗/∂pF≃0\partial M^*/\partial p_F\simeq0 or slightly positive.

Consider first Eq. (17). Using the canonical value of the incompressibility at normal nuclear matter density [4], K=250 MeVK=250\ \mathrm{MeV}, the lower limit of JJ (J>−500 MeVJ>-500\ \mathrm{MeV}) indicates that the three-particle term MM∗(H0−H1)\frac{M}{M^*}(H_0-H_1) is positive and larger than unity. Considering next Eq. (18), the canonical value of the symmetry energy at normal nuclear matter density is as=32 MeVa_s=32\ \mathrm{MeV}, and the empirical range of the parameter μ\mu defined by Eq. (15) reported in Ref. [4] is 0.27±0.250.27\pm0.25. If μ\mu is taken to be close to the central value of this range (μ≃0.27\mu\simeq0.27), the sum of the first two terms in […][\ldots] in the expression (18) for LL is ≃0.6\simeq0.6, almost independent of M∗M^*, within the range of M∗M^* given above.5 The empirical slope parameter LL in Eq. (19) then implies that the three-particle term MM∗(H0′−13H1)\frac{M}{M^*}(H'_0-\frac{1}{3}H_1) is negative, with a magnitude smaller than unity.

If we further assume that the ℓ=1\ell=1 moment of the three-particle parameter plays only a subordinate role, that is, H1≪H0H_1\ll H_0 and H1≪3H0′H_1\ll3H'_0, — and the simple calculations explained below support this assumption — we can expect that the isoscalar three-particle parameter H0H_0 is positive and larger than unity, while the isovector one H0′H'_0 is negative and smaller than unity in magnitude. The uncertainties in the values of KK and asa_s do not change this qualitative conclusion [13, 14].

In order to see how these expectations compare with simple semi-quantitative calculations, we consider the leading term in the Faddeev series for the three-particle amplitude hh, shown in Figure 1a. In this diagram, tt denotes the two-particle tt-matrix, which is the off-forward generalization of the amplitude ff of the Fermi liquid theory. We call the contribution of the leading term the two-particle correlation (2pc) piece. Its expression reads

First two terms in the Faddeev series for the three-particle scattering amplitude h

Figure 1. First two terms in the Faddeev series for the three-particle scattering amplitude hh. Lines with arrows denote nucleons, and tt denotes the two-particle tt-matrix.

h(2pc)(k1,k2,k3)=∑4δk1+k2,k3+k4P∣⟨12∣t^∣34⟩a∣2ε3+ε4−ε1−ε2+(1↔3)+(2↔3).(20)h^{(2\mathrm{pc})}(\mathbf{k}_1,\mathbf{k}_2,\mathbf{k}_3)=\sum_4\delta_{\mathbf{k}_1+\mathbf{k}_2,\mathbf{k}_3+\mathbf{k}_4}P\frac{\left|\langle12|\hat{t}|34\rangle_a\right|^2}{\varepsilon_3+\varepsilon_4-\varepsilon_1-\varepsilon_2}+(1\leftrightarrow3)+(2\leftrightarrow3). \tag*{(20)}

In this schematic notation, 1 ∼\sim 4 represent the momenta k1∼k4\mathbf{k}_1\sim\mathbf{k}_4 as well as the associated spin and isospin components, though an average over the spin components of 1, 2, 3 is assumed implicitly. The sum represents momentum integration and summation over spin and isospin components of 4, the δ\delta symbol represents a momentum conserving δ\delta-function, PP denotes the principal value, ⟨12∣t^∣34⟩a\langle12|\hat{t}|34\rangle_a is the antisymmetrized two-particle tt-matrix element, and εi\varepsilon_i are the quasiparticle energies. An example for the three-particle correlation (3pc) term of hh, which is the next-to-leading term in the Faddeev series, is shown in Figure 1b. We will derive its form in the next section.

In order to get a rough estimate of h(2pc)h^{(2\mathrm{pc})}, we assume that the matrix elements ⟨12∣t^∣34⟩a\langle12|\hat{t}|34\rangle_a can be replaced by the ℓ=0\ell=0 part of an effective interaction of the Landau-Migdal type [19], which corresponds to an approximation of the scattering matrix by the scattering length decomposed into the various spin-isospin channels:

a⟨34∣t^∣12⟩=f0(δ31⋅δ42)+f0′(τ31⋅τ42)+g0(σ31⋅σ42)+g0′(σ31⋅σ42)(τ31⋅τ42),(21)a\langle34|\hat{t}|12\rangle=f_0(\delta_{31}\cdot\delta_{42})+f'_0(\boldsymbol{\tau}_{31}\cdot\boldsymbol{\tau}_{42})+g_0(\boldsymbol{\sigma}_{31}\cdot\boldsymbol{\sigma}_{42})+g'_0(\boldsymbol{\sigma}_{31}\cdot\boldsymbol{\sigma}_{42})(\boldsymbol{\tau}_{31}\cdot\boldsymbol{\tau}_{42}), \tag*{(21)}

where the notation indicates that the spin and isospin operators are defined to act in the particle-hole channel. With the further assumption that the quasiparticle energies can be approximated as εi=ki22M∗\varepsilon_i=\frac{k_i^2}{2M^*}, where M∗M^* is in the range given above, the angular averages of Eq. (10) can be carried out analytically [13], and show that the contribution of H1H_1 in Eqs. (17) and (18) is indeed small. In this approximation, the dimensionless isoscalar and isovector three-particle ℓ=0\ell=0 amplitudes take the very simple forms

H0(2pc)=ln⁡24(F02+3F0′2+3G02+9G0′2),(22)H_0^{(2\mathrm{pc})}=\frac{\ln2}{4}\left(F_0^2+3F_0'^2+3G_0^2+9G_0'^2\right), \tag*{(22)}
H0′(2pc)=ln⁡24(13F02+43F0F0′−13F0′2+G02+4G0G0′−G0′2),(23)H_0'^{(2\mathrm{pc})}=\frac{\ln2}{4}\left(\frac{1}{3}F_0^2+\frac{4}{3}F_0F_0'-\frac{1}{3}F_0'^2+G_0^2+4G_0G_0'-G_0'^2\right), \tag*{(23)}

where the dimensionless Landau-Migdal parameters are defined as F0=(2pFM∗/π2)f0F_0=(2p_FM^*/\pi^2)f_0 and similar for the other cases. For illustrative purposes, we show in Tab. I the results for three sets of the extended Skyrme interaction [33], which are all consistent with the assumptions made at the beginning of this subsection. We see that the 2pc contribution to H0H_0 is positive definite, which is consistent with our expectations, but tends to be smaller than unity, which gives us a hint that 3pc terms like shown in Figure 1b may be necessary to explain the empirical values of the skewness. The 2pc contribution to H0′H'_0, on the other hand, is negative – mainly because of the terms −13F0′2-\frac{1}{3}F_0'^2 and −G0′2-G_0'^2 in Eq. (23) – and smaller in magnitude than unity, which is consistent with our expectations. This result thus provides no hint that the 3pc term is necessary to explain the empirical values of the slope parameter of the symmetry energy.

Set123
M∗/MM^*/M0.70.80.9
μ\mu0.2330.2290.360
H0(2pc)H_0^{(2\mathrm{pc})}0.5280.7621.095
H0′(2pc)H_0'^{(2\mathrm{pc})}−0.063−0.101−0.135

Table 1. Values of various physical quantities entering Eqs. (17) and (18) at nuclear matter saturation density. Sets 1–3 correspond to the results for the extended Skyrme interactions [33] eMSL07, eMSL08, eMSL09, respectively. The values for μ\mu, defined by Eq. (15), given in this Table refer to the approximate expression given in footnote 5. H0(2pc)H_0^{(2\mathrm{pc})} refers to Eq. (22), and H0′(2pc)H_0'^{(2\mathrm{pc})} refers to Eq. (23).

Physical content of the three-particle amplitude

As we have seen in the previous section, the ℓ=0\ell=0, ℓ=1\ell=1 moments of the three-particle amplitude, Eq. (11), which we express here schematically as6^6 h(1,2,3)=δf(1,2)/(δn3)h(1,2,3)=\delta f(1,2)/(\delta n_3), are related to observable quantities. It is the aim of this section to gain some insight into the physics contained in this quantity, and in methods to calculate it by using certain approximations.

The two-particle amplitude has the schematic form [26] f(1,2)=Z1Z2t(1,2)f(1,2)=Z_1Z_2t(1,2), where the ZZ-factors are the residues of the single particle propagators at the quasiparticle poles, and t(1,2)t(1,2) is the on-shell value of the two-particle tt-matrix in the forward limit, which satisfies the Bethe-Salpeter (BS) equation in the particle-hole channel with the two-body kernel given by K(2)(1,2)=iδΣ(1)/(δS(2))K^{(2)}(1,2)=i\delta\Sigma(1)/(\delta S(2)) [34], where Σ\Sigma is the skeleton part of the self energy and SS is the single particle propagator. The variation δ/(δn3)\delta/(\delta n_3) of the ZZ-factors and of the energy variables in t(1,2)t(1,2) produces retardation terms which will be denoted by h(prod)h^{(\mathrm{prod})}, and have the form of symmetrized products of functions depending only on pairs of variables. For the derivation we refer to Ref. [14], and just quote the result for the case when all particles are on the Fermi surface:

h(prod)(1,2,3)=12f(1,2)((ZΣ′′′)+∂∂εF)(f(1,3)+f(2,3))+(1↔3)+(2↔3),h^{(\mathrm{prod})}(1,2,3)=\frac{1}{2}f(1,2)\left((Z\Sigma''')+\frac{\partial}{\partial\varepsilon_F}\right)(f(1,3)+f(2,3))+(1\leftrightarrow3)+(2\leftrightarrow3),

where ZZ is the values of the ZZ-factor at the Fermi surface, Σ′′\Sigma'' is the second derivative of Σ\Sigma w.r.t. the energy variable at the Fermi surface, and

∂f(1,2)∂εF≡Z2[(∂∂ε1+∂∂ε2)t(1,2)]εi=εF.\frac{\partial f(1,2)}{\partial\varepsilon_F}\equiv Z^2\left[\left(\frac{\partial}{\partial\varepsilon_1}+\frac{\partial}{\partial\varepsilon_2}\right)t(1,2)\right]_{\varepsilon_i=\varepsilon_F}.

The variation δ/(δn3)\delta/(\delta n_3) of the off-shell two-particle tt-matrix gives a contribution h~(1,2,3)=Z1Z2δt(1,2)/(δn3)\tilde{h}(1,2,3)=Z_1Z_2\delta t(1,2)/(\delta n_3), taken on the energy shells of particles 1 and 2. This can be calculated by straight forward variation of the BS equation, which has the schematic form t=K(2)−iK(2)SStt=K^{(2)}-iK^{(2)}SSt, and one obtains the structure shown in Figure 2. The particle-hole irreducible three-body kernel which appears there is given by K(3)(1,2,3)=iδ2Σ(1)/(δS(2)δS(3))K^{(3)}(1,2,3)=i\delta^2\Sigma(1)/(\delta S(2)\delta S(3)) [35]. Typically, one finds that the loop diagrams shown in Figure 2 represent “direct terms”, with the corresponding exchange terms contained in K(3)K^{(3)}. The total three-particle amplitude can then be expressed as h(1,2,3)=h(prod)(1,2,3)+h~(1,2,3)h(1, 2, 3) = h^{(\mathrm{prod})}(1, 2, 3) + \tilde{h}(1, 2, 3).

Figure 2. Graphical representation of the three-particle amplitude h~(1,2,3)\tilde{h}(1, 2, 3). The three-particle kernel K(3)K^{(3)}, defined in the main text, is irreducible in the particle-hole channel, and contains terms of second, third and higher orders in the two-particle tt-matrix (denoted as tt). The products of pole parts of propagators with the same momenta (Spole2(k)S_{\mathrm{pole}}^{2}(k) or Spole3(k)S_{\mathrm{pole}}^{3}(k)) in these diagrams are defined as Sp2(k)+Sh2(k)S_{p}^{2}(k)+S_{h}^{2}(k) or Sp3(k)+Sh3(k)S_{p}^{3}(k)+S_{h}^{3}(k), without products of particle and hole parts.

In order to calculate the two-body and three-body kernels K(2)K^{(2)} and K(3)K^{(3)} from their definitions, one needs to model the functional dependence of the self energy Σ\Sigma on the single-particle propagator SS. One model which has been widely used in the literature since the works of Brueckner, Day, Bethe and others [17, 36, 37], is to express Σ\Sigma in terms of the two-particle tt-matrix calculated in the ladder approximation to the BS equation, which is essentially the same as Brueckner’s G-matrix. We have recently developed a scheme [14] to expand Σ\Sigma in powers of the ladder tt matrix, and to carry out the required functional derivatives to get the kernels K(2)K^{(2)} and K(3)K^{(3)}. The self energy up to the third power of the ladder tt-matrix is shown graphically in Figure 3.

Figure 3. Skeleton diagrams for the self energy to first order (a) and to third order (b) in the ladder tt-matrix. Note that there is no second order contribution. The two-particle tt-matrix in the ladder approximation is denoted by a black dot.

Figure 3a is the standard result in Brueckner approximation, and Figure 3b describes the rescattering of the hole line in Figure 3a, and the RPA-type correlations of third order in the ladder tt-

matrix. By calculating K(3)K^{(3)} from these diagrams and inserting the result into Figure 2, one obtains the term h(2pc)h^{(2\mathrm{pc})} of Figure 1a or Eq. (20) as the second order term, and the contributions shown in Figure 4 as the third order terms.

The diagram Figure 4a is the next-to-leading term of the in-medium Faddeev series, with nn and kk being the spectators of the first and last interactions. The term with (ijk)=(lmn)(ijk)=(lmn) comes from Figure 2b, with the three-particle kernel K(3)K^{(3)} of second order in the ladder tt-matrix shown by Figure 1a. The other terms with (ijk)≠(lmn)(ijk)\ne(lmn) come from the third order contributions to the three-particle kernel K(3)K^{(3)}. The diagrams in Figure 4b and Figure 4c arise from the interaction of the three given particles on the Fermi surface with the particles in the Fermi sea, i.e., they are of four-body nature. These diagrams do not appear in the usual Faddeev series, because there it is assumed from the outset that the first and last interactions occur among the three given particles, and not between one of them and a background particle. The contribution where two intermediate lines in Figure 4b and Figure 4c have the same momenta come again from Figure 2b, and the contribution where all three intermediate lines have the same momenta comes from Figure 2c. Those terms where all three intermediate lines in Figure 4b and Figure 4c have different momenta come from the third order contributions to the three-particle kernel K(3)K^{(3)}.

Summary

The motivation for the first part of our work was the rapidly expanding interest in the slope parameters of the symmetry energy and the incompressibility of nuclear matter, two physical quantities which have decisive impact on the structure of nuclei and neutron stars. By using the Fermi liquid theory of Landau and Migdal, we derived two model independent relations for these observables, Eqs. (17) and (18). The crucial new point about these relations is that they connect three-body interaction parameters, where the isoscalar ss-wave parameter H0H_0 and its isovector counterpart H0′H'_0 play the dominant roles, to observables of nuclear matter. By using empirical information about several other quantities in those relations, we concluded that H0H_0 must be positive and larger than unity, and H0′H'_0 must be negative and smaller in magnitude than unity. We made a semi-quantitative estimate of these two parameters by using the leading term of the Faddeev series and simple two-body contact interactions. We found that H0H_0 is indeed positive but tends to be smaller than unity, which gives us a hint that higher order terms in the Faddeev series, in particular the three-body cluster of Figure 1b, give an important contribution to the skewness of nuclear matter. The results for H0′H'_0 were in the range of expectations (negative and smaller in magnitude than unity), which gives no clear hint about the role of three-body cluster for the slope parameter of the symmetry energy.

The motivation for the second part of our work was to get more understanding about the physics contained in the three-particle amplitudes in nuclear matter. By using only its definition in the Fermi liquid theory, we expressed it as the sum of retardation terms (h(prod)h^{(\mathrm{prod})}), which come from the energy dependence of the self energy and/or the two-particle tt-matrix, and the part h~\tilde{h} of Figure 2, which is characterized by the particle-hole irreducible three-particle kernel K(3)K^{(3)}. In order to get more detailed expressions, we followed the method of the Bethe-Brueckner-Goldstone (BBG) theory and used the ladder approximation to the two-body Bethe-Salpeter equation for the tt-matrix as a building block. By expanding all quantities up to the third power of the ladder tt-matrix, we found that the Fermi liquid theory naturally leads not only to the first few terms in the Faddeev series (see Figure 1), but also to medium modifications of four-body nature, induced by the interaction of the three particles on the Fermi surface with the particles in the Fermi sea. It would be very interesting to see how large these terms are in comparison to the usual Faddeev terms, and which kind of medium induced three-body processes appear in higher orders.

References

Acknowledgments

The authors would like to dedicate their work to Professor Akito Arima, the great physicist, teacher, and friend. Shortly after one of us (W.B.) joined the Nuclear Theory Group at the University of Tokyo more than 40 years ago, Prof. Arima suggested him to study the connection between the work of the Tokyo group on nuclear electromagnetic properties, which started with the famous Arima-Horie Effect for magnetic moments in 1954, and the Fermi liquid theory. With Prof. Arima’s help, it was enlightening to see how many similar physical ideas can be expressed in a very different and independent language.

I.C. would like to contribute a photo, which was taken in front of the Physics Building of Argonne National Laboratory in summer 1963, a few years after Prof. Arima’s stay as a researcher at Argonne. We can imagine how comfortable and happy he felt among so many active people and friends. With his experiences in Argonne, he started the shell-model tradition in Japan, being still developed to date.

Finally, the authors would like to express their thanks to the Editors and Publishers of this Memorial Book for their kind cooperation.

I.C. was supported by the U.S. Department of Energy, Office of Science, Office of Nuclear Physics, contract no. DE-AC02-06CH11357.

Group photograph in front of the Physics Building

Figure 5. Photo taken in summer 1963 in front of the Physics Building at Argonne National Laboratory. From left to right: A. W. Martin, A. Saperstein, K. C. Wali, H. J. Lipkin, M. Peshkin, W. D. McGlinn, F. Coester, J. Monahan, K. Hiida, A. M. Green, G. E. Brown, A. Arima, K. Tanaka, D. Kurath, M. A. Melvin, N. Rosenzweig, K. Eckstein, S. Tani, M. N. Hack, B. Kursunoglu, T. Sebe, R. A. Ferrell, D. R. Inglis, B. M. Udaonkar, S. Fallieros, S. P. Pandya. Reprinted with kind permission of Argonne National Laboratory.

References

  1. [1]M. Baldo and G. F. Burgio, The nuclear symmetry energy, Prog. Part. Nucl. Phys. 91, 203–258 (2016).DOI
  2. [2]C. J. Horowitz and J. Piekarewicz, Neutron star structure and the neutron radius of Pb-208, Phys. Rev. Lett. 86, 5647 (2001).
  3. [3]J. P. Blaizot, Nuclear Compressibilities, Phys. Rept. 64, 171–248 (1980).DOI
  4. [4]B.-A. Li, B.-J. Cai, L.-W. Chen, and J. Xu, Nucleon Effective Masses in Neutron-Rich Matter, Prog. Part. Nucl. Phys. 99, 29–119 (2018).DOI
  5. [5]G. Colo, U. Garg, and H. Sagawa, Symmetry energy from the nuclear collective motion: constraints from dipole, quadrupole, monopole and spin-dipole resonances, Eur. Phys. J. A50, 26 (2014).DOI
  6. [6]J. Piekarewicz and F. J. Fattoyev, Neutron rich matter in heaven [7] V. Baran, M. Collonna, V. Greco, and M. Di Toro, Reaction dynamics with exotic beams, Phys. Rept. 410, 335–466 (2005).
  7. [8]J. M. Pearson, The incompressibility of nuclear matter and the breathing mode, Phys. Lett. B271, 12–16 (1991).DOI
  8. [9]A. W. Steiner, J. M. Lattimer, and E. F. Brown, The Neutron Star Mass-Radius Relation and the Equation of State of Dense Matter, Astrophys. J. 765, L5 (2013).DOI
  9. [10]P. Danielewicz, R. Lacey, and W. G. Lynch, Determination of the equation of state of dense matter, Science. 298, 1592–1596 (2002).DOI
  10. [11]L. D. Landau, The Theory of a Fermi Liquid, Sov. Phys. JETP. 3, 920 (1956).
  11. [12]L. D. Landau, Oscillations in a Fermi Liquid, Sov. Phys. JETP. 5, 101 (1957).
  12. [13]W. Bentz and I. C. Cloët, Skewness of nuclear matter and three-particle correlations, Phys. Rev. C100(1), 014303 (2019).DOI
  13. [14]W. Bentz and I. C. Cloët, Slope parameter of the symmetry energy and the structure of three-particle interactions in nuclear matter, Phys. Rev. 105(1) 014320 (2022).arxiv.org/abs/2004.11605
  14. [15]H. A. Bethe, Three-Body Correlations in Nuclear Matter, Phys. Rev. 138, B804–B822 (1965).
  15. [16]B. D. Day, Three-body correlations in nuclear matter, Phys. Rev. C24, 1203–1271 (1981).
  16. [17]H. A. Bethe, Theory of nuclear matter, Ann. Rev. Nucl. Part. Sci. 21, 93–244 (1971).
  17. [18]L. D. Landau, On the Theory of the Fermi Liquid, Sov. Phys. JETP. 8, 70 (1959).
  18. [19]A. B. Migdal, Theory of finite Fermi systems and applications to atomic nuclei, New York, Wiley (1967).DOI
  19. [20]R. Shankar, Renormalization group approach to interacting fermions, Rev. Mod. Phys. 66, 129–192 (1994).DOI
  20. [21]J. Speth, K. Rewald, F. Grümmer, P. G. Reinhard, N. Lyutorovich, and V. Tsel yaev, Landau–Migdal vs. Skyrme, Nucl. Phys. A928, 17–29 (2014).
  21. [22]G. Baym and S. A. Chin, Landau Theory of Relativistic Fermi Liquids, Nucl. Phys. A262, 527–538 (1976).DOI
  22. [23]W. Bentz, A. Arima, H. Hyuga, K. Shimizu, and K. Yazaki, Ward identity in the many body system and magnetic moments, Nucl. Phys. A 436, 593–620 (1985).DOI
  23. [24]A. Arima, K. Shimizu, W. Bentz, and H. Hyuga, Nuclear Magnetic Properties and Gamow-Teller Transitions, Adv. Nucl. Phys. 18, 1–106 (1987).
  24. [25]W. Bentz and A. Arima, The Relation between the photonuclear E1 sum rule and the effective orbital g factor, Nucl. Phys. A736, 93–125 (2004).DOI
  25. [26]J. Negele and H. Orland, Quantum Many-particle Systems. West-view Press (1998).
  26. [27]G. Baym and C. Pethick, Landau Fermi-Liquid Theory: Concepts and Applications. WILEY-VCH Verlag (2004).
  27. [29]C. Mahaux, P. F. Bortignon, R. A. Broglia, and C. H. Dasso, Dynamics of the shell model, Phys. Rept. 120, 1–274 (1985).DOI
  28. [30]J. P. Blaizot and B. L. Friman, On the nucleon effective mass in nuclear matter, Nucl. Phys. A372, 69–89 (1981).DOI
  29. [31]E. N. E. van Dalen, C. Fuchs, and A. Faessler, Momentum, density, and isospin dependence of the symmetric and asymmetric nuclear matter properties, Phys. Rev. C72, 065803 (2005).DOI
  30. [32]S. Goriely, N. Chamel, and J. M. Pearson, Further explorations of Skyrme-Hartree-Fock-Bogoliubov mass formulas. XII: Stiffness and stability of neutron-star matter, Phys. Rev. C82, 035804 (2010).DOI
  31. [33]Z. Zhang and L.-W. Chen, Extended skyrme interactions for nuclear matter, finite nuclei, and neutron stars, Phys. Rev. C 94, 064326 (Dec, 2016).DOI
  32. [34]G. Baym and L. P. Kadanoff, Conservation Laws and Correlation Functions, Phys. Rev. 124, 287–299 (1961).DOI
  33. [35]J. Speth, Transition probabilities and static moments of excited states in even-even nuclei, Zeitschrift für Physik A Hadrons and nuclei. 239(3), 249–265 (1970).DOI
  34. [36]K. A. Brueckner and C. A. Levinson, Approximate Reduction of the Many-Body Problem for Strongly Interacting Particles to a Problem of Self-Consistent Fields, Phys. Rev. 97, 1344–1352 (1955).DOI
  35. [37]B. D. Day, Elements of the Brueckner-Goldstone Theory of Nuclear Matter, Rev. Mod. Phys. 39, 719–744 (1967).DOI

Paper details

vlnc:W_a5vbf98t
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents