Leadingorder nucleon selfenergy in relativistic chiral effective field theory
Abstract
We apply thermal field theory methods to compute microscopically the nucleon selfenergy arising from onepion exchange in isospinsymmetric nuclear matter and neutron matter. A selfconsistent numerical scheme is introduced and its convergence is demonstrated. The repulsive contribution from the Fock exchange diagram to the energy per nucleon in symmetric nuclear matter is obtained.
I Introduction
The equation of state of bulk nuclear matter has attracted considerable attention over time, as it has a substantial impact on the properties of neutron stars as well as finite nuclei. The equation of state is determined by two, generally related, ingredients: the force between nucleons and the manybody approximation used to compute the thermodynamic properties of nuclear matter. Nuclear interactions are accurately modeled in terms of potentials arising from meson exchange: on the one hand models have been used which are based on phaseshift equivalent oneboson exchange Wiringa et al. (1995); ?; ?; ? and, on the other hand, more recently, (multi)pionexchange potentials based on chiral power counting Weinberg (1990); ?; ?; ?; ?. The diversity of abinitio manybody approaches includes variational Monte Carlo methods Akmal et al. (1998); ?; ?; ? and propagator methods with (medium) renormalized soft interactions as, e.g., in nonrelativistic and covariant Bruecknertype theories Day and Wiringa (1985); ?; ?; ?; ?. Perturbative calculations of nuclear matter properties were revived with the advent of soft chiral potentials Hebeler et al. (2011); ?.
In this work we apply thermal field theory (TFT) methods to compute the leadingorder nucleon selfenergy in isospinsymmetric nuclear and neutron matter. The method has been applied in the past in the context of QED and QCD plasmas to compute the quasiparticle energies of electrons and quarks, respectively, in a thermal medium Blaizot and Ollitrault (1993); ?. In the lowenergy regime of interest to us, the relevant degrees of freedom are nucleons and pions. Chiral symmetry is an (approximate) symmetry of the strong interaction. Because chiral symmetry is spontaneously broken in nature, pions emerge as (pseudo) Goldstone bosons of the theory. The description of the strong interaction in terms of chiral Lagrangians admits certain approximation schemes in terms of power counting of small quantities Weinberg (1979); Kaplan et al. (1996), which in principle allow for a systematic orderbyorder improvement of a given calculation. Below, we will combine relativistic TFT methods and the chiral Lagrangian description of nuclear interactions to address the computation of the nucleon selfenergy.
Approaches similar to ours were previously developed in Refs. Fraga et al. (2004); Lutz et al. (2000); Kaiser et al. (2002a); ?; ?; Lacour et al. (2011); ?. Fraga et al. Fraga et al. (2004) derived analytical expressions for the zerotemperature selfenergy of a nucleon due to pion exchange in dilute nuclear matter to leading order in the chiral expansion. Lutz et al. Lutz et al. (2000) and Kaiser et al. Kaiser et al. (2002a); ?; ? (hereafter KFW) used chiral Lagrangians and expansions in small Fermi momentum to construct equations of state in the heavybaryon limit. The saturation in Ref. Lutz et al. (2000) arises due to correlations induced by the onepionexchange interaction. In Ref. Kaiser et al. (2002a); ?; ? twopion exchange produces nuclear binding at the threeloop level with a suitably adjusted momentum cutoff. An effective field theory of nuclear matter with nucleons and pions, which allows for both local as well as pionmediated multinucleon interactions, was developed in Ref. Lacour et al. (2011); ? in the heavybaryon limit and the main trends for the energy density of symmetric nuclear and neutron matter were already reproduced at nexttoleading order in their powercounting scheme.
In applying TFT to nuclear matter in a relativistic setting, we have to maintain the covariance of the theory by using fully relativistic thermal propagators for nucleons and pions. Keeping the Lorentz symmetries intact is of advantage, in particular, for computing the scattering and radiation amplitudes with full propagators and renormalized vertices. Specifically the computation of particular electroweak processes in nuclear and neutron matter (or transitions in nuclei), which are driven by currents with particular Lorentz symmetry are conveniently carried out in the Dirac basis.
The second goal of our work is to maintain selfconsistency among the propagators and the selfenergies of the theory, which means that the iterations are performed until the SchwingerDyson equation for the nucleons is fulfilled. If a firm perturbative expansion (with powercounting rules) exists, a selfconsistent approach generates higherorder terms in every iteration, and therefore is not really necessary. Nevertheless, using the leadingorder term in the chiral expansion of the Lagrangian, we will check the impact of these higherorder terms in a selfconsistent solution.
This paper is structured as follows. In Sec. II we discuss the chiral Lagrangian. Section III uses TFT to compute the pion contribution to the nucleon selfenergy. Our numerical method and results for the selfenergy are presented in Sec. IV. Our conclusions are collected in Sec. V. We use natural units . Fourvectors are denoted with capital letters, for instance .
Ii Lagrangians
Lowenergy nuclear dynamics can be constructed on the basis of the pion and nucleon degrees of freedom starting from a chiral Lagrangian. The interaction Lagrangian between nucleons and pions is constructed such as to reflect the spontaneous chiral symmetry breaking of strong interactions at low energies. Since the interactions of Goldstone bosons must vanish at zeromomentum transfer and in the chiral limit (i.e., the pion mass ), a lowenergy expansion in powers (the socalled chiral dimension) of the ratio of the momentum or the pion mass over ( times) the pion decay constant can be performed. Consequently, the Lagrangian can be written as
(1) 
where the superscript labels the order of the chiral dimension. The terms in the expansion (1) are constructed by introducing the following matrix in flavor space
(2) 
where is vector of Pauli matrices in isospin space, is the isotriplet of pions, and the pion decay constant. The leadingorder term is given by Gasser and Leutwyler (1988)
(3) 
where is the nucleon field, , is the nucleon mass, and is the axialvector coupling. The physical value of is determined from neutron beta decay and is given by . is the covariant derivative,
(4) 
where is the socalled chiral connection which couples an even number of pions to the nucleon and is defined as
(5)  
with . The Lagrangian (3) also includes the axialvector current which couples an odd number of pions to the nucleon
(6) 
Keeping only the lowestorder term in the chiral Lagrangian (3) the pionnucleon Lagrangian reads Machleidt and Entem (2011)
(7) 
where we have neglected the WeinbergTomozawa contribution arising from the chiral connection (5). The chiral onepion interaction term in Eq. (7) takes the light dynamical degrees of freedom, i.e., pions, explicitly into account. The complete interaction Lagrangian includes to lowest order an additional fourfermion contact term. Thus, the Lagrangian of the system can be written as
(8) 
where the first term is the Lagrangian of the noninteracting system and the last term corresponds to the fourfermion interaction. Figure 1 shows the two distinct leadingorder contributions to the selfenergy of a nucleon originating from the treelevel vertices in the Lagrangian (8).
Iii Pion contribution to the selfenergy
iii.1 Leadingorder contribution
In this section we evaluate the onepion contribution to the nucleon selfenergy, corresponding to the left diagram in Fig. 1, within the imaginarytime formalism. The free covariant propagator of the nucleons in energymomentum space is given by
(9) 
where the zeroth component of the fourmomentum takes discrete values, , and is the temperature. are projectors onto positive and negative energy states,
(10) 
where is the dispersion relation for noninteracting nucleons, is their mass, and . The free pion propagator is given by
(11) 
where the zeroth component of the fourmomentum takes discrete values , , and is the dispersion relation for noninteracting pions, with being the free pion mass. In terms of the free propagators (9) and (11) the onepion exchange contribution to the nucleon selfenergy reads
(12)  
where . We substitute the propagators (9) and (11) into Eq. (12) and carry out the summation over the fermionic Matsubara frequency . In general, this sum generates physically distinct processes involving all possible combinations of bosons and fermions and their antiparticles. The result of the summation can be arranged according to these underlying processes, but it is more convenient to separate the selfenergy into the “vacuum” and “thermal” parts , where the vacuum part is given by
(13)  
and the thermal part is given by
(14)  
where and are the Fermi/Bose distribution functions. The retarded selfenergy is obtained by analytical continuation, i.e., . We have verified that in the case of a Yukawa interaction Eq. (14) transforms to the wellknown expression for the selfenergy of a fermion in finitetemperature quantum field theory Blaizot and Ollitrault (1993); Kapusta (1989).
At sufficiently low temperature the occupation number of antiparticles is so small that we can neglect their contribution in Eq. (14). Furthermore, we assume that there is no macroscopic occupation of pionic modes in nuclear matter at any temperature and density of interest; therefore, we also drop the contributions proportional to the bosonic occupation numbers. The remaining contribution arises from the pole at and we arrive at
(15) 
where we have defined the quantities and , with components
(16)  
(17) 
Later on we will enforce selfconsistency in evaluating the selfenergy. This requires a Lorentz decomposition of the selfenergy, which in the most general case is given by
(18) 
The requirements of parity conservation, translational and rotational invariance, as well as timereversal invariance, reduce this most general decomposition to the following form
(19) 
Equation (15) can now be projected onto its Lorentz components by multiplying it with , and , and taking the trace over the matrices. Keeping only the thermal part of the selfenergy (and dropping the index on the selfenergies), this leads us to the following decomposition coefficients
(20)  
(21)  
(22)  
where is a unit vector. Equations (20)–(22) are our final result for the chiral onepionexchange contribution to the nucleon selfenergy. It is evident that the other terms in Eq. (14), which could become important at higher temperatures and densities, can be evaluated in a completely analogous way.
iii.2 Further approximations
For numerical computations the factor in Eq. (15) can be simplified. We start by rewriting the expression
(23)  
For the densities and temperatures of interest, nucleons are constrained to the vicinity of their Fermi surface, therefore the momentum of the nucleon can be expressed as . Here, is the nucleon Fermi momentum, is a unit vector, and is the residual momentum, with . Furthermore, the relativity parameter is small as well; numerically, we have
(24) 
where is the density of the systems, is the nuclear saturation density. Then, the nucleon energy is and the chemical potential . This implies that is small compared to . Therefore, we can replace .
With these approximations we obtain
(25)  
where we used the fact that .
Therefore, the factor in the integrand of the selfenergy (15) reads now
(26) 
Separating the contributions from particles and antiparticles in the boson propagator, with these approximations the selfenergy reads
(27)  
where
(28) 
is the reduced selfenergy. Therefore, the coefficients of the Lorentz decomposition of the selfenergy can be expressed in terms of ,
(29) 
The real part of the selfenergy can now be computed with the help of the Dirac identity. We obtain by explicitly evaluating the Cauchy principal value of Eq. (28)
where .
iii.3 Enforcing selfconsistency
The selfconsistency of the numerical computation of the selfenergy is achieved by replacing the free nucleon propagator in Eq. (12) by the full nucleon propagator, determined by the SchwingerDyson equation
where the free nucleon propagator is given by and we have used Eq. (27) for the nucleon selfenergy. The roots of determine the excitation spectrum of the system,
(32) 
where
(33)  
(34)  
(35)  
(36) 
We achieve selfconsistency for the selfenergy by replacing the free quantities , , , and in the integrand (but not in the integration measure) of Eq. (III.2) by the corresponding renormalized , , , and . In practice, we start by computing (III.2) with the free quantities, which define the renormalized quantities (33)–(36) to first order in the iteration process. We repeat the previous step until convergence is reached. Approximations made in the numerical part of this work are purely technical and concern only the computation of traces in the selfenergy diagram. It is evident from Eqs. (33)–(36) that the pole structure and, therefore, the quasiparticle spectra as well as the Lorentz structure of the selfenergies remains fully relativistic. During the iteration procedure the quantities in Eqs. (33)–(36) are corrected in each iteration differently according to the Lorentz structure of the selfenergy and propagators. Thus the Dirac structure of the selfenergies and propagators is maintained.
Iv Results
We will not attempt to evaluate the contributions to the energy of nuclear matter from diagrams other than the onepion exchange discussed in the previous section. For that purpose we define the energy per nucleon, excluding its rest mass , via the formula
(37) 
where is the kinetic energy, is the potential energy and is the particle number. In terms of the renormalized quantities (34)–(36) the kinetic energy reads
(38) 
where is the isospin degeneracy factor ( in isospinsymmetric nuclear matter and in neutron matter). The potential energy is given by
(39)  
As we consider only spinunpolarized matter, the spin summation has been carried out in Eqs. (38) and (39).
iv.1 Symmetric nuclear matter
The Fock diagram evaluated in the previous section represents chiral onepion exchange between two nucleons. Chiral power counting requires to leading order an additional twobody contact term. The corresponding selfenergy diagrams are shown by the left and the right diagrams in Fig. 1. In passing we note that if in the nuclear medium the contact is density independent then its contribution is given simply by where is the density and is the twobody contact.
The reduced selfenergy due to the chiral onepion exchange in isospinsymmetric nuclear matter is shown in Fig. 2 at and . As one observes, the selfconsistency procedure converges rapidly. In order to achieve a relative accuracy one needs about 10 iterations, the exact number of iteration depending on the density. For small densities the number of required iterations is small, while for large densities the number of iterations needed to achieve convergence is larger. While the selfenergy does not tend to zero for large external momenta, the result for the energy of the system is convergent due to an additional integration over a Fermi distribution function. Our numerical implementation of the selfconsistency shows that the perturbative expansion is not perfect and iterations are necessary (see Fig. 2). We see that chiral power counting at nonzero density and temperature is not reliable anymore, since several new scales (temperature and chemical potential) appear.
In Fig. 3 we show the reduced selfenergy at zero temperature in comparison to its form at MeV. The change with temperature can be seen to be rather moderate. In the nonzero temperature case the selfenergy is larger for low momenta than at zero temperature, while for higher momenta the variation of temperature does not affect the result very much. This is due to the fact that we deal with comparatively low temperatures, therefore the temperature is a relevant scale only at low momenta.
In Fig. 4 we show the Lorentz components of the real part of the full selfenergy at fm and . It can be seen that the vector selfenergy is substantially smaller than the other contributions to the selfenergy, as expected.
Our numerical calculations of the Fock contribution to the selfenergy of nucleons can be validated in certain limiting cases. We verified that our numerical result for the real part of the reduced selfenergy is in good agreement with the analytical result quoted in Ref. Fraga et al. (2004) at zero temperature. A further test is the comparison of the energies per particle of nuclear matter with those of KFW, where the energy per particle was computed directly without a reference to the selfenergy. Their result for the zerotemperature onepionexchange Fock diagram, which is an expansion in the small relativity parameter , reads Kaiser et al. (2002a); ?; ?
(40) 
where the coefficients of this expansion and are functions of the ratio alone and can be calculated analytically (34). In Fig. 5 we show the contributions of the onepion exchange to the energy of the matter from various Lorentz components along with the full result which is the sum of the three components. Our full result is in good agreement with the analytical expression by KFW, Eq. (40), also shown in Fig. 5.
iv.2 Pure neutron matter
The reduced selfenergy of neutrons in pure neutron matter is shown in Fig. 6. Compared to isospinsymmetric nuclear matter, the selfenergy in neutron matter is smaller. This is due to the fact that in pure neutron matter onepion exchange involves only the meson. At nonzero temperature, as in the case of isospinsymmetric nuclear matter, we observe that, for low momenta, the contribution of the selfenergy is larger, but the difference among the two cases tends to disappear with increasing momentum.
V Conclusions
In this work we have combined the methods of TFT and chiral Lagrangians to compute the selfenergy of nucleons to leading order in the chiral expansion. In doing so, we have maintained the covariance of the pion and nucleon propagators and we have imposed selfconsistency by solving a SchwingerDyson equation for the nucleon selfenergy. Our approach has also been applied to pure neutron matter, with similar results.
Clearly, to obtain a consistent phenomenology of both isospinsymmetric nuclear and pure neutron matter, one needs to introduce contact interactions which account for the shortrange twobody and threebody interactions. Furthermore, the importance of secondorder pion exchange was already stressed by Lutz et al. Lutz et al. (2000) and KFW. In this respect further steps might be undertaken to resolve the relativistic dynamics of pions by including higherorder terms in the chiral expansion and incorporating isobar excitations.
Methodologically, our approach differs from similar works since we address the nucleon selfenergy within a chiral effective thermal field theory by keeping a relativistic framework, and at the same time we impose selfconsistency by solving a SchwingerDyson equation.
The inmedium electromagnetic and weak interactions of nucleons can be computed in a relativistically covariant manner starting from the selfconsistent propagators derived above.
Acknowledgments
This work was partially supported by the HGSHIRe graduate program at Frankfurt University (G. C.). We thank E. S. Fraga, R. D. Pisarski, and J. SchaffnerBielich for discussions, S. Gandolfi for correspondence and the anonymous referee for pointing out a sign error in an earlier version. Furthermore, we are indebted to Bengt Friman for his continuous interest in this work and useful suggestions.
References
 R. B. Wiringa, V. G. J. Stoks, and R. Schiavilla, Phys. Rev. C 51, 38 (1995).
 V. Stoks, R. Klomp, C. Terheggen, and J. de Swart, Phys.Rev. C49, 2950 (1994).
 R. Machleidt, Phys.Rev. C63, 024001 (2001).
 M. Lacombe, B. Loiseau, J. Richard, R. Vinh Mau, J. Cote, et al., Phys.Rev. C21, 861 (1980).
 S. Weinberg, Phys.Lett. B251, 288 (1990).
 S. Weinberg, Nucl.Phys. B363, 3 (1991).
 C. Ordóñez, L. Ray, and U. van Kolck, Phys. Rev. C 53, 2086 (1996).
 R. Machleidt and D. R. Entem, Physics Report 503, 1 (2011).
 E. Epelbaum, H.W. Hammer, and U.G. Meißner, Reviews of Modern Physics 81, 1773 (2009).
 A. Akmal, V. R. Pandharipande, and D. G. Ravenhall, Phys. Rev. C 58, 1804 (1998).
 O. Benhar, V. R. Pandharipande, and S. C. Pieper, Reviews of Modern Physics 65, 817 (1993).
 S. Gandolfi, F. Pederiva, S. Fantoni, and K. E. Schmidt, Physical Review Letters 98, 102503 (2007).
 S. Gandolfi, A. Y. Illarionov, S. Fantoni, J. C. Miller, F. Pederiva, and K. E. Schmidt, Mon. Not. RAS: Letters 404, L35 (2010).
 B. D. Day and R. B. Wiringa, Phys. Rev. C 32, 1057 (1985).
 D. Gambacurta, L. Li, G. Colò, U. Lombardo, N. Van Giai, and W. Zuo, Phys. Rev. C 84, 024301 (2011).
 H.J. Schulze and T. Rijken, Phys. Rev. C 84, 035801 (2011).
 A. Sedrakian, Prog.Part.Nucl.Phys. 58, 168 (2007).
 F. Sammarruca, ArXiv eprints: 1111.0695 (2011).
 K. Hebeler, S. K. Bogner, R. J. Furnstahl, A. Nogga, and A. Schwenk, Phys. Rev. C 83, 031301 (2011).
 K. Hebeler and A. Schwenk, Phys. Rev. C 82, 014314 (2010).
 J.P. Blaizot and J.Y. Ollitrault, Phys. Rev. D 48, 1390 (1993).
 J.P. Blaizot, E. Iancu, and A. Rebhan, Phys. Rev. D 63, 065003 (2001).
 S. Weinberg, Physica A 96, 327 (1979).
 D. B. Kaplan, M. J. Savage, and M. B. Wise, Nuclear Physics B 478, 629 (1996).
 E. S. Fraga, Y. Hatta, R. D. Pisarski, and J. SchaffnerBielich, Phys. Rev. C 69, 035211 (2004).
 M. Lutz, B. Friman, and C. Appel, Physics Letters B 474, 7 (2000).
 N. Kaiser, S. Fritsch, and W. Weise, Nuclear Physics A 697, 255 (2002a).
 N. Kaiser, S. Fritsch, and W. Weise, Nuclear Physics A 700, 343 (2002b).
 W. Weise, Progress in Particle and Nuclear Physics 67, 299 (2012).
 A. Lacour, J. A. Oller, and U.G. Meißner, Annals of Physics 326, 241 (2011).
 J. A. Oller, A. Lacour, and U.G. Meißner, Journal of Physics G Nuclear Physics 37, 015106 (2010).
 J. Gasser and H. Leutwyler, Nuclear Physics B 307, 763 (1988).
 J. I. Kapusta, Finitetemperature field theory., Cambridge University Press, Cambridge (UK), 1989, 229 p., ISBN 0521351553 (1989).

The explicit expressions of these functions
are given by Kaiser et al. (2002a); ?; ?