betaFeedback

WorksPhysics and Astronomy

Hybrid stars and QCD phase transition with a NJL-like model

Abstract

In this paper, we introduce a self-consistent mean field approximation to study the QCD phase transition and the structure of hybrid stars within the framework of NJL model. In our practice, a phenomenological parameter αα is introduced, which reflects the weights of "direct" channel and "exchange" channel under a finite chemical potential. The mass-radius relation is obtained by solving the Tolman-Oppenheimer-Volkoff equation using a crossover equation of state (EOS). We calculate the density distribution in a two solar-mass hybrid star to show the effects of different parameters. We also calculate the tidal Love number k2k_2 and the deformability (Λ)(Λ). It is found that the stiffness of the EOS increases with αα, which allows us to obtain a hybrid star with a maximum mass of 2.40 solar-mass through our model. The observation of over 2.06 solar-mass neutron stars may indicates that the chiral transition may be a crossover on the whole T−μT-μ plane.

Introduction

Neutron stars are the densest celestial bodies in our universe except for black holes. Their extreme environment provides us a natural laboratory for cold and condensed matter. Especially, in recent years, the detection of gravitational waves has given us different valuable information about compact stars. During the inspiral and merger of binary stars, the tidal deformability can be measured by the LIGO and VIRGO network directly [1]. The Love number k2k_2 is used to measure the distortion of a neutron star. It can be calculated through the exterior solutions that are related to the tidal deformability (Λ\Lambda) [2, 3, 4]. On the other hand, electromagnetic observations on the cooling rate{1}, gamma-ray bursts [6], thermal X-rays{2} and radio bursts [8] from neutron stars give us diverse data to study neutron stars. These astronomical observations can help us to constrain theoretical models and improve our understanding of neutron stars.

The inner structure of neutron stars is still not well known. The possibility that the neutron star is composed of a quark matter inner core, nucleon matter outer core and crust, i.e., being a hybrid star, is widely discussed. The key to study hybrid stars is the equation of state (EOS) of the strongly interacting matter. For quark matter, lattice quantum chromodynamics (LQCD) is widely accepted as the ab-initio method dealing with QCD nonperturbatively. Unfortunately, present lattice QCD calculations at finite chemical potential are plagued with the so called “sign problem” [9]. Thus, to study QCD systems at finite chemical potential, it is necessary to employ some QCD effective models, such as the Nambu-Jona-Lasinio (NJL) model [10, 11, 12, 13, 14]. The NJL model displays important features of QCD, i.e., the dynamical chiral symmetry breaking and restoring. The chiral restoring phase transition is usually considered to happen simultaneously (or almost simultaneously) with the deconfinement phase transition [15, 16, 17, 18].

Thus, a problem naturally arises: which type of phase transition should it be, a first order phase transition or a crossover? Another related issue is that in the case of zero chemical potential and finite temperature, the result of LQCD and other methods shows there is a crossover[15, 16, 17, 18]. Thus, at zero temperature and finite chemical potential the type of phase transition determines whether there exists a critical end point (CEP) in the QCD phase diagram. Experimentally, the second stage beam energy scanning plan of Relativistic Heavy Ion Collider (RHIC) is still to try to find the possible CEP [19, 20, 21]. Theoretically, different models give different perspectives and answers. In this paper, we will study this issue by using a self-consistent mean field approximation in the framework of NJL model [22]. People generally predict that the hadronic matter will completely transit to quark matter when the density increases to 4–8 times the saturation density of nucleon matter in the hadron physical pictures [23, 24, 25, 26], which means that the corresponding quark chemical potential is at least larger than 430 MeV in NJL model. However, the strong interaction phase transition predicted by the standard NJL model usually occurs near 330 MeV [27, 28, 29, 30], which indicates that there is an contradiction between the prediction of the standard NJL model and the corresponding prediction of the hadron physical pictures.

In order to overcome this contradiction, the authors of Ref. [22] proposed a mean field approximation method under the framework of NJL model. In this method, a phenomenological parameter α\alpha is introduced to reflect the weights of the “exchange” channel and “direct” channel, because the vector interaction term generated by “exchange” channel plays a very important role in the case of finite chemical potential. The relative results will influence the CEP and hybrid stars. α\alpha is a theoretical parameter. The finite density experiments could help to constrain α\alpha. For example, the experiments of RHIC can provide useful clues on the issue of whether CEP exists or not, which itself can help constrain the value of α\alpha. Anyway, note that such a constraint will still be too rough to determine α\alpha, since there may be other factors influencing the CEP. Thus we need various experiments and observations, such as the astronomical observations of neutron stars, to further constrain the parameter. Another point we focus on is the uncertain region between the nucleon matter and the quark matter. The uncertainty forces us to adopt some nonphysical methods to connect the nucleon part and the quark part, such as the Gibbs construction [31], or other mathematical operations [24, 32, 33]. However, these methods unavoidably lead to a phase transition themselves, thus they are not self-consistent in the practice. The EOS will be studied under a crossover-type model with a self-consistent connection method in this paper.

This paper is organized as follows. In § 1.1, we introduce the three flavor NJL model and its Fierz transformation. The three-flavor QCD phase diagram is calculated and the existence of CEP in our model is discussed. In Section III, the thermodynamical relations will be solved to obtain the EOS for quark matter. We use a soft hadronic EOS (APR) [34] and a crossover connection method to get the EOS of hybrid stars. We will show how hybrid stars are changed in this framework. Finally, we give a summary in Section IV.

Mean Field Approximation and Phase Transition

The standard 2+1 flavor NJL Lagrangian contains four-fermion and six-fermion interactions written as

\mathcal{L}=\bar{\psi}(i\slashed{\partial}-m)\psi+G\sum_{a=0}^{8}\left[(\bar{\psi}\lambda^{a}\psi)^{2}+(\bar{\psi}i\gamma_{5}\lambda^{a}\psi)^{2}\right]-K\left[\det\bar{\psi}(1+\gamma_{5})\psi+\det\bar{\psi}(1-\gamma_{5})\psi\right]+\mu\bar{\psi}\gamma^{0}\psi \tag*{(1)}

where mm is the current quark mass, GG and KK denote the coupling constants, which will be calibrated to reproduce the physical pion meson mass, kaon meson mass and their decay constant. The corresponding mean field approximation is

\mathcal{L}_{1}\approx\bar{\psi}\left[i\slashed{\partial}-\gamma^{0}\mu-(m+A)\right]\psi, \tag*{(2)}

with

A=−4G(σuσdσs)+2K(σdσsσuσsσuσd)A=-4G\begin{pmatrix}\sigma_{u}\\ \sigma_{d}\\ \sigma_{s}\end{pmatrix}+2K\begin{pmatrix}\sigma_{d}\sigma_{s}\\ \sigma_{u}\sigma_{s}\\ \sigma_{u}\sigma_{d}\end{pmatrix}

and

σi=⟨ψˉiψi⟩=−∫d4p(2π)4Tr⁡[s(p)i]i=u,d,s,\sigma_{i}=\langle\bar{\psi}_{i}\psi_{i}\rangle=-\int\frac{d^{4}p}{(2\pi)^{4}}\operatorname{Tr}[s(p)_{i}]\qquad i=u,d,s,

where σ\sigma is the quark condensation, Tr denotes a trace over color and spinor indices.

As shown in Eq. (1), the standard NJL model Lagrangian contains only the interactions of scalar and pseudoscalar channels. It is insufficient to handle the interaction of vector channels, such as in the case of finite chemical potential. To get a self-consistent result in the sense of mean field approximation, the contribution of Fierz transformation Lagrangian must be taken into account [35]. The authors of Ref. [11] demonstrated that a Fierz transformation of six-fermion interaction can be defined as an operation that leaves the interaction invariant under all possible permutations of the quark spinors ψ\psi occurring in it, thus the six-fermion term dose not change after Fierz transformation. So in this paper we only need to consider the Fierz transformation of four-fermion term, which can be written as [11]:

F[∑a=08[(ψˉλaψ)2+(ψˉiγ5λaψ)2]]=−G2∑c=08[(ψˉγμλcψ)2−(ψˉγμγ5λcψ)2].(3)\mathcal{F}\left[\sum_{a=0}^{8}\left[(\bar{\psi}\lambda^{a}\psi)^{2}+(\bar{\psi}i\gamma_{5}\lambda^{a}\psi)^{2}\right]\right] =-\frac{G}{2}\sum_{c=0}^{8}\left[(\bar{\psi}\gamma^{\mu}\lambda^{c}\psi)^{2}-(\bar{\psi}\gamma^{\mu}\gamma_{5}\lambda^{c}\psi)^{2}\right]. \tag*{(3)}

In this study, we only consider the scalar and vector channel contribution, this is because other terms in our modeling make no contribution at the level of mean field approximation. The mean field approximation of the Fierz transformation of the original NJL Lagrangian is

\mathcal{L}_{2}\approx\bar{\psi}[i\slashed{\partial}-\gamma^{0}(\mu+B)-(m+A)]\psi. \tag*{(4)}

Here B=−2G3(nu+nd+ns)IB=-\frac{2G}{3}(n_{u}+n_{d}+n_{s})I, II is the identity matrix in flavor space and nn denotes the quark number density. L1\mathcal{L}_{1} contains only the Hartree term, while L2\mathcal{L}_{2} contains only the Fock term. We can rewrite Eq. (1) by taking the linear combination of L1\mathcal{L}_{1} and L2\mathcal{L}_{2} [22]:

LR=(1−α)L1+αL2.(5)\mathcal{L}_{R}=(1-\alpha)\mathcal{L}_{1}+\alpha\mathcal{L}_{2}. \tag*{(5)}

To illustrate the characteristics of this approach, let us recall the mean field approximation utilized in previous studies within the framework of NJL model. In the original mean field approximation, the Fierz transformation is not adopted, which corresponds to the case of α=0\alpha=0 in our method [36]. But as pointed out in Refs. [35, 37], this approximation is theoretically not self-consistent at the level of mean field approximation.

As shown in Refs. [35, 37], to get a self-consistent result in the sense of mean field approximation, the contribution of the Fierz transformed Lagrangian (“exchange channel”) must be taken into account. Later, when the finite chemical potentials (in this situation, the vector-isoscalar channel interactions are very important) are involved, people usually put a vector-isoscalar channel interaction by hand to the original Lagrangian, so they must phenomenologically introduce a coupling parameter to reflect the strength of the explicit vector-isoscalar channel interaction [38]. This will cause a serious problem, that is, the vector-isoscalar channel interaction is introduced phenomenologically, and the Lagrangian at this time is no longer the original Lagrangian. What is more serious is that the vector-isoscalar interaction introduced by hand is not theoretically self-consistent in the framework of the mean field approximation, because the Fierz transformed Lagrangian is not considered.

At the same time, if people want to discuss the problems related to the axialvector chemical potential (under these circumstances, the axialvector channel interaction plays an extremely important role), then people must artificially introduce the axial-vector channel interaction. So, the introduction of a term by hand is quite arbitrary and will make the method lose its reliability [39]. In order to overcome the arbitrariness brought by the above method, the self-consistent mean field approximation method is necessary, which can release all the interaction channels hidden in the original Lagrangian by the Fierz transformation [40]. Because the Fierz transformation is an equivalent transformation, the new Lagrangian LR=(1−α)L+αLF\mathcal{L}_{R}=(1-\alpha)\mathcal{L}+\alpha\mathcal{L}_{F} used in this paper is equivalent to the original NJL model, which is another advantage of our method.

Actually, there are many studies that have considered the contribution of an explicit Fierz transformed term and have chosen the weight of vector-isoscalar channel factor α\alpha as 0.5 [11, 27]. However, there is no physical basis supporting that the “direct” channel and the “exchange” channel have the same weight. In principle, the value of α\alpha could be constrained or hinted from related experiments and neutron star observations. On the other hand, in the commonly used NJL model, μ0\mu_{0} (when the chemical potential μ\mu is smaller than the critical value μ0\mu_{0}, the quark number density becomes zero.) is very close to μc\mu_{c} (the critical chemical potential of chiral restoration). It means that a vacuum phase transition occurs shortly after the baryon is excited from the vacuum, which is physically unreasonable [24, 41]. This long-standing problem can be well solved by adopting our self-consistent mean field approximation.

Based on the new Lagrangian of Eq. (5), the three-flavor quark gap equation is then given by

Mi=mi−4Gσi+2Kσjσk,(6)M_i=m_i-4G\sigma_i+2K\sigma_j\sigma_k, \tag*{(6)}
μ~i=μi−23αG′(nu+nd+ns),(7)\tilde{\mu}_i=\mu_i-\frac{2}{3}\alpha G^{\prime}(n_u+n_d+n_s), \tag*{(7)}

where G′=G1−αG^{\prime}=\frac{G}{1-\alpha}, and μ~i\tilde{\mu}_i is the effective chemical potential and MiM_i is the constituent quark mass. Similarly, μi\mu_i is the chemical potential and mim_i is the quark mass. Because NJL model is non-renormalizable, in this study, we employ three momentum cutoff scheme to regulate the divergence. The quark condensation and number density in three momentum cutoff scheme is given by:

σi=−NcMiπ2∫0Λp2Ep,i[1−(eEp,i−μ~iT+1)−1−(eEp,i+μ~iT+1)−1]dp,(8)\sigma_i=-\frac{N_cM_i}{\pi^2}\int_0^\Lambda\frac{p^2}{E_{p,i}}\left[1-\left(e^{\frac{E_{p,i}-\tilde{\mu}_i}{T}}+1\right)^{-1}-\left(e^{\frac{E_{p,i}+\tilde{\mu}_i}{T}}+1\right)^{-1}\right]dp, \tag*{(8)}
ni=Ncπ2∫0Λp2[1eEp,i−μ~iT+1−1eEp,i+μ~iT+1]dp.(9)n_i=\frac{N_c}{\pi^2}\int_0^\Lambda p^2\left[\frac{1}{e^{\frac{E_{p,i}-\tilde{\mu}_i}{T}}+1}-\frac{1}{e^{\frac{E_{p,i}+\tilde{\mu}_i}{T}}+1}\right]dp. \tag*{(9)}

The parameters of NJL we adopted are shown in Table I, which are consistent with the experimental results of π\pi, KK and η\eta meson masses and decay constants [42].

mum_u [MeVMeV]msm_s [MeVMeV]ΛUV\Lambda_{UV} [MeVMeV]GΛUV2G\Lambda_{UV}^{2}KΛUV5K\Lambda_{UV}^{5}
5.5135.7630.11.7819.29

Table I. Parameters adopted in our calculations

We can use the method of finite temperature field theory

[43] to write the grand canonical potential density as:

Ω=−NcTπ2∑i=u,d,s∫p2[Ep,iT+ln⁡(1+e−Ep,i+μ~iT)+ln⁡(1+e−Ep,i−μ~iT)]dp+2G(σu+σd+σs)−αG3(nu+nd+ns)2−4Kσuσdσs+Ω0.(10)\Omega=-\frac{N_cT}{\pi^2}\sum_{i=u,d,s}\int p^2\left[\frac{E_{p,i}}{T}+\ln\left(1+e^{-\frac{E_{p,i}+\tilde{\mu}_i}{T}}\right)+\ln\left(1+e^{-\frac{E_{p,i}-\tilde{\mu}_i}{T}}\right)\right]dp+2G(\sigma_u+\sigma_d+\sigma_s)-\frac{\alpha G}{3}(n_u+n_d+n_s)^2-4K\sigma_u\sigma_d\sigma_s+\Omega_0. \tag*{(10)}

From this, the pressure p=−Ωp=-\Omega and the energy density ϵ=−p+∑iμini\epsilon=-p+\sum_i\mu_i n_i can be easily obtained. In addition, the baryon number and electric charge are conserved for neutron stars [44]. Therefore, we need to include the following equilibrium-condition:

{23nu−13nd−13ns−ne=0,μe=μμ=μd−μu,μd=μs.\begin{cases} \frac{2}{3}n_u-\frac{1}{3}n_d-\frac{1}{3}n_s-n_e=0,\\ \mu_e=\mu_\mu=\mu_d-\mu_u,\\ \mu_d=\mu_s. \end{cases}

The NJL model with zero bare quark mass at high energy scale, keeps the chiral symmetry. With the decrease of energy, the quarks and anti-quarks condensate together to give quarks dynamical mass, thus the chiral symmetry is broken at low energy scale. To illustrate this phase transition more clearly, we can use the chiral susceptibility to reveal the order of phase transition and the corresponding physical condition[45, 46]. The chiral susceptibility is defined as :

χm=−∂σu∂mu.(11)\chi_m=-\frac{\partial\sigma_u}{\partial m_u}. \tag*{(11)}

We show the results of chiral susceptibility for different α\alpha in Figure 1. It can be seen that when α\alpha is small, there is a singular point on the susceptibility curve. It corresponds to a first order phase transition occurring at a particular critical chemical potential. On the other hand, when α\alpha is large enough (α>0.47\alpha>0.47), the curve becomes relatively smooth. It means that the phase transition changes to a crossover, and the chemical potential corresponding to the peak point is regarded as “pseudo-critical chemical potential”. Figure 1 also shows that the critical (pseudo-critical) chemical potential gradually becomes larger with the increase of the parameter α\alpha. We can see this more clearly in Figure 2. The above results indicate that our quark model can be used to describe different kinds of phase transitions. By evaluating the parameter α\alpha properly, it can present a good approximation for various hadron models.

Figure 1. The chiral susceptibility as a function of baryon chemical potential at zero temperature, for different α\alpha.

Critical pseudo-critical chemical potential as a function of $\alpha$

Figure 2. The critical (pseudo-critical) chemical potential as a function of α\alpha at zero temperature. When α<0.47\alpha< 0.47, it is a first order phase transition. But when α>0.47\alpha> 0.47, the phase transition is a crossover.

Similarly, using the chiral susceptibility we can plot the QCD phase diagram for different α\alpha. The result is shown in Figure 3. In particular, we are interested in the critical end point (CEP) which denotes a critical point connecting two different types of transitions. It is one of the most important features of QCD phase diagram. However, whether there is a CEP in the phase diagram or not is still highly debated in the literature and it depends on the chosen physical environment [47]. Our model indicates that there is CEP for a small α\alpha but no CEP exists for a large α\alpha.

QCD phase diagram on the $T-\mu$ plane for different $\alpha$

Figure 3. The QCD phase diagram on the T−μT-\mu plane for different α\alpha for the conditions of neutron stars.

More restrictions on α\alpha and the type of phase transition could be hinted by the experiments of RHIC. Note that the above calculations about the QCD phase diagram are in the conditions of neutron stars. For the experiments of RHIC, the conditions are quite different. An important difference is that the ratio of electric charge over baryon is special and is approximately 0.4, i.e., nq=0.4nBn_q=0.4n_B. Second, the strangeness neutrality requires the density of strange quark (nsn_s) to be zero. Additionally, the experiments of RHIC are not in β\beta equilibrium, but are connected with an ambiguous chemical potential condition. However, comparing with the special electric charge condition, such a chemical potential condition will have less influence on the QCD phase diagram. To compare with the RHIC experiments, we have calculated the QCD phase diagram under the above experimental conditions (but still with β\beta equilibrium). The results are shown in Figure 4.

QCD phase diagram on the $T-\mu$ plane under the condition of $n_q=0.4n_B$ and $\beta$ equilibrium

Figure 4. The QCD phase diagram on the T−μT-\mu plane for different α\alpha, under the condition of nq=0.4nBn_q=0.4n_B and β\beta equilibrium. When α>0.58\alpha> 0.58, there is no CEP in our model.

In these cases, our model predicts that there will be no CEP when α>0.58\alpha> 0.58 in the phase diagram.

Hybrid Stars with a Crossover EOS

The EOS plays a critical role in calculating the mass-radius relation and tidal deformability (Λ\Lambda) of neutron stars. Using Eq. (10), we can get the quark section of EOS. The hybrid star also includes a hadron section, but the NJL model is difficult to describe the hadron state at low densities. Thus we use the APR EOS for hadronic matter, which is a soft EOS [34]. It has been adopted by many authors to describe hybrid stars at low densities[48, 49]. For the whole EOS, we use a simple mathematical method to link the hadron and quark section:

p=s(μ)pH(μ)+(1−s(μ))pQ(μ),(12)p=s(\mu)p_H(\mu)+(1-s(\mu))p_Q(\mu), \tag*{(12)}
n=dpdμ=s(μ)nH(μ)+(1−s(μ))nQ(μ)+dsdμ(pH−pQ).(13)n=\frac{dp}{d\mu}=s(\mu)n_H(\mu)+(1-s(\mu))n_Q(\mu)+\frac{ds}{d\mu}(p_H-p_Q). \tag*{(13)}

Here pHp_H and pQp_Q is the pressure of hadron and quark section, pp is the total pressure. The case is similar for the density nn. The function ss is defined as:

s(μ)=e−μ−μ0μ1.(14)s(\mu)=e^{-\frac{\mu-\mu_0}{\mu_1}}. \tag*{(14)}

Here the parameter μ0=923 MeV\mu_0=923\ \mathrm{MeV} corresponds to the chemical potential that nucleon matter begins to appear in the ARP EOS model. The parameter μ1=1200 MeV\mu_1=1200\ \mathrm{MeV} we assume here can make the contribution of hadron matter reduce and the contribution of quark matter increase smoothly when the density changes from 2n02n_0 to 8n08n_0, which means it is a crossover between quark and hadron matter. So, a reasonable requirement is that the EOS of quark matter also has a crossover, which means α>0.47\alpha>0.47 here. The whole EOS is shown in Figure 5, from which it can be seen that a larger α\alpha will lead to a harder EOS.

The crossover equation of state for different alpha

Figure 5. The crossover equation of state for different α\alpha.

The square of sound velocity for different alpha

Figure 6. The square of sound velocity for different α\alpha.

To study the stiffness of the system in detail, we calculate the square of sound velocity

v2=dpdε,(15)v^2=\frac{dp}{d\varepsilon}, \tag*{(15)}

which can reflect the stiffness of the system. The results are plot in Figure 5. From this figure, we can see that a larger α\alpha leads to a larger sound velocity, which demonstrates that the stiffness of the EOS will increase with α\alpha.

We then study the mass-radius relation by solving the Tolman-Oppenheimer-Volkoff equations (in the natural unit system)

dp(r)dr=−(ε+p)(M+4πr3p)r(r−2M),(16)\frac{dp(r)}{dr}=-\frac{(\varepsilon+p)(M+4\pi r^3p)}{r(r-2M)}, \tag*{(16)}
dM(r)dr=4πr2ε.(17)\frac{dM(r)}{dr}=4\pi r^2\varepsilon. \tag*{(17)}

In addition, we use the small crust approximation[50] to describe the crust of our hybrid stars. To show the influence of α\alpha on the neutron star, the mass-radius relation of hybrid stars for different values of α\alpha is presented in Figure 7. Astronomical observations show that the masses of PSR J0384+0432 and PSR J1614-2230 are 2.01±0.04 M⊙2.01\pm0.04\ M_\odot and 1.928±0.017 M⊙1.928\pm0.017\ M_\odot respectively. Our result indicates that when α=0.5\alpha=0.5, the maximum strange matter star is 2.01 M⊙2.01\ M_\odot, which is consistent with observations. However, α=0.5\alpha=0.5 is difficult to meet the requirement of PSR J0740+6620 (2.14±0.10 M⊙2.14\pm0.10\ M_\odot). Thus we may need an even larger α\alpha, such as in the range of 0.6∼0.80.6\sim0.8. As shown in Figure 7, the maximum mass of hybrid stars increases with the parameter α\alpha. If a larger mass of neutron star is observed, then a larger α\alpha will be preferred.

Figure 7. Mass-Radii relations of hybrid stars for different α\alpha. The maximum mass increases with α\alpha. Mass are in unites of M⊙M_\odot.

Another astronomical phenomenon that can be engaged to limit the range of α\alpha is the cooling rate of neutron stars. The direct Urca process provides the fastest neutrino emission in nucleon matter and quark matter. It is the main cooling mechanism in neutron stars [51]. The neutrino emission rate is density-dependent. A high density will lead to a high emission rate, which makes the cooling of neutron stars faster. This could give us more restrictions on the EOS. We plot in Figure 8 the density distribution inside a two solar mass hybrid star. We can see that a small α\alpha results in a high density, which will lead to a quicker cooling rate. A more detailed comparison is beyond the scope of this study and will be studied later.

Density distribution in a $2M_\odot$ hybrid star

Figure 8. The density distribution in a 2M⊙2M_\odot hybrid star. The color depth denotes the density value.

During the inspiral and merger of two stars, the ratio of each star’s induced mass quadrupole moment to the tidal field is defined as the tidal deformability (Λ\Lambda), and the Love number

α\alphaMmax⁡(M⊙)M_{\max}(M_\odot)R(km)R(\mathrm{km})R1.4(km)R_{1.4}(\mathrm{km})Λ(1.4M⊙)\Lambda(1.4M_\odot)
0.52.0110.8111.72312
0.62.0610.8411.80332
0.72.1710.9211.87348
0.82.2211.0311.93363
0.92.4011.2811.98377

Table 2. Properties of hybrid stars in our framework

k2k_2 describes the distortion of the surface of a star. The relation between the tidal deformability (Λ\Lambda) and the l=2l=2 dimensionless tidal Love number k2k_2 is (in the natural unit system)

k2=32Λ(MR)5.(18)k_2 = \frac{3}{2}\Lambda\left(\frac{M}{R}\right)^5. \tag*{(18)}

According to Ref. [52], k2k_2 can be calculated as

k2=8C55(1−2C)2[2+2C(y−1)−y]×{2C[6−3y+3C(5y−8)]+4C3[13−11y+C(3y−2)+2C2(1+y)]+3(1−2C)2[2−y+2C(y−1)]ln⁡(1−2C)}−1,(19)\begin{aligned} k_2={}&\frac{8C^5}{5}(1-2C)^2[2+2C(y-1)-y] \\ &\times\{2C[6-3y+3C(5y-8)] \\ &\quad+4C^3[13-11y+C(3y-2)+2C^2(1+y)] \\ &\quad+3(1-2C)^2[2-y+2C(y-1)]\ln(1-2C)\}^{-1}, \tag*{(19)} \end{aligned}

where C=M/RC=M/R is the compactness of the star, and

y=Rβ(R)H(R)−4πR3ε0M(20)y=\frac{R\beta(R)}{H(R)}-\frac{4\pi R^3\varepsilon_0}{M} \tag*{(20)}

is related to the metric variable HH and surface energy density ε0\varepsilon_0. The metric variable HH depends on the EOS and can be obtained by integrating two differential equations

dH(r)dr=β,(21)\frac{dH(r)}{dr}=\beta, \tag*{(21)}
dβ(r)dr=2(1−2Mr)−1H{−2π[5ε+9p+f(ε+p)]+3r2+2(1−2Mr)−1(Mr2+4πrp)2}+2βr(1−2Mr)−1[−1+Mr+2πr2(ε−p)],(22)\begin{aligned} \frac{d\beta(r)}{dr}={}&2\left(1-\frac{2M}{r}\right)^{-1}H\left\{-2\pi[5\varepsilon+9p+f(\varepsilon+p)]\right.\\ &\left.+\frac{3}{r^2}+2\left(1-\frac{2M}{r}\right)^{-1}\left(\frac{M}{r^2}+4\pi rp\right)^2\right\}\\ &+\frac{2\beta}{r}\left(1-\frac{2M}{r}\right)^{-1}\left[-1+\frac{M}{r}+2\pi r^2(\varepsilon-p)\right], \tag*{(22)} \end{aligned}

where ff is defined as

f=dεdp.(23)f=\frac{d\varepsilon}{dp}. \tag*{(23)}

We can integrate Eqs. (22) and (23) from the center via expansions H(r)=a0r2H(r)=a_0r^2 and β(r)=2a0r\beta(r)=2a_0r for r≪Rr\ll R with constant a0a_0. Since a0a_0 can be reduced in the expression of the Love number k2k_2, here we take a0=1a_0=1,

We have calculated the main parameters of hybrid stars for different α\alpha. The results are shown in Table II. For a 1.4M⊙1.4M_\odot low-spin star, the tidal deformability (Λ\Lambda) has been constrained in a range of (200, 800) through gravitational wave observations [53]. It can be seen that the results of our hybrid stars are consistent with the observational requirement.

Summary and Discussion

In this paper, we adopt a self-consistent mean field approximation to discuss the three-flavor NJL model. A free parameter α\alpha is introduced to denote the weight of different interaction channels. Using this method, we study the phase transition properties by calculating chiral susceptibility. It is found that the (pseudo) critical chemical potential increases with α\alpha. We plot the QCD phase diagram under the condition applicable for neutron stars as well as the condition for heavy ion collision experiments. It is found that there is no CEP when α>0.47\alpha>0.47 for neutron stars. But for the heavy ion collision experiments, there will be no CEP when α>0.58\alpha>0.58. We discuss how the stiffness of EOS changes with α\alpha by examining the sound velocity. It is found that the phase transition of quark matter to hadron matter is a crossover when α>0.47\alpha>0.47 in neutron stars, thus we need a crossover connection method to describe the hybrid star. The corresponding EOS can also support a more massive star since it gets harder with a larger α\alpha. The density-radius diagram of a 2M⊙2M_\odot hybrid star is plot to show the effects of different parameters. Finally we calculate the tidal deformability and show that our EOS can meet the requirements of gravitational wave observations.

To be specific, the EOS with α=0.5\alpha=0.5 yields a maximum compact star of 2.01M⊙2.01M_\odot, which can match the masses of PSR J1614-2230 (1.928±0.017M⊙1.928\pm0.017M_\odot) [54] and PSR J0348+0432 (2.01±0.04M⊙2.01\pm0.04M_\odot) [55]. However, α=0.5\alpha=0.5 is difficult to satisfy PSR J0740+6620 (2.14±0.10M⊙2.14\pm0.10M_\odot) [56] and PSR J2215+5135 (2.27−0.15+0.17M⊙2.27^{+0.17}_{-0.15}M_\odot) [57]. A quark model with a larger α\alpha may provide a harder EOS to support a compact star with a larger mass. A more mild phase transition, which means a larger α\alpha in this paper, makes quarks have a larger energy at the same density due to interactions with each other. As a result, α>0.7\alpha>0.7 is preferred to satisfy PSR J0740+6620 (2.14±0.10M⊙2.14\pm0.10M_\odot)… in our framework. More astronomical observations will help to further constrain the value of α\alpha. We calculate the density-radius distribution because the density of matter may affect the cooling rate of neutron stars. Measuring the temperature of neutron stars may also help to constrain α\alpha.

Acknowledgements

We thank the anonymous referee for helpful comments and suggestions. This work is supported by National SKA Program of China No. 2020SKA0120300, by the National Natural Science Foundation of China (Grant Nos. 11873030, 12041306, U1938201), by the National Key R&D Program of China (2021YFA0718500), and by the science research grants from the China Manned Space Project with NO. CMS-CSST-2021-B11.

References

  1. [1]B. P. Abbott, R. Abbott, T. Abbott, F. Acernese, K. Ackley, C. Adams, T. Adams, P. Addesso, R. Adhikari, V. Adya, et al., Physical Review Letters 119, 161101 (2017).
  2. [2]T. Damour, M. Soffel, and C. Xu, Physical Review D 45, 1017 (1992).
  3. [3]É. É. Flanagan and T. Hinderer, Physical Review D 77, 021502(R) (2008).
  4. [4]T. Hinderer, The Astrophysical Journal 677, 1216 (2008).
  5. [6]E. Berger, Ann. Rev. Astron. Astrophys. 52, 43 (2014), arXiv:1311.2603 [astro-ph.HE].DOI
  6. [8]J. Geng, B. Li, and Y. Huang, The Innovation 2, 100152 (2021).
  7. [9]S. Bethke, Progress in Particle and Nuclear Physics 58, 351 (2007).
  8. [10]U. Vogl and W. Weise, Progress in Particle and Nuclear Physics 27, 195 (1991).
  9. [11]S. Klevansky, Reviews of Modern Physics 64, 649 (1992).
  10. [12]T. Hatsuda and T. Kunihiro, Physics Reports 247, 221 (1994).
  11. [13]W. Fan, X. Luo, and H.-S. Zong, International Journal of Modern Physics A 32, 1750061 (2017).
  12. [14]C.-M. Li, P.-L. Yin, and H.-S. Zong, Physical Review D 99, 076006 (2019).
  13. [15]O. Philipsen, Prog. Part. Nucl. Phys. 70, 55 (2013), arXiv:1207.5999 [hep-lat].DOI
  14. [16]Z. Fodor and S. D. Katz, (2009), arXiv:0908.3341 [hep-ph].arxiv.org/abs/0908.3341
  15. [17]S. Gupta, K. Huebner, and O. Kaczmarek, Phys. Rev. D 77, 034503 (2008), arXiv:0711.2251 [hep-lat].DOI
  16. [18]S. Borsanyi, Z. Fodor, C. Hoelbling, S. D. Katz, S. Krieg, and K. K. Szabo, Phys. Lett. B 730, 99 (2014), arXiv:1309.5258 [hep-lat].DOI
  17. [19]M. Aggarwal, Z. Ahammed, A. Alakhverdyants, I. Alekseev, J. Alford, B. Anderson, D. Arkhipkin, G. Averichev, J. Balewski, L. Barnby, et al., Physical Review Letters 105, 022302 (2010).
  18. [20]L. Adamczyk, J. Adkins, G. Agakishiev, M. Aggarwal, Z. Ahammed, I. Alekseev, J. Alford, C. Anson, A. Aparin, D. Arkhipkin, et al., Physical Review Letters 112, 032302 (2014).
  19. [21]L. Adamczyk, J. Adams, J. K. Adkins, G. Agakishiev, M. Aggarwal, Z. Ahammed, N. Ajitanand, I. Alekseev, D. Anderson, R. Aoyama, et al., Physics Letters B 785, 551 (2018).
  20. [22]F. Wang, Y. Cao, and H. Zong, Chinese Physics C 43, 084102 (2019).
  21. [23]G. Baym and S. Chin, Physics Letters B 62, 241 (1976).
  22. [24]G. Baym, T. Hatsuda, T. Kojo, P. D. Powell, Y. Song, and T. Takatsuka, Reports on Progress in Physics 81, 056902 (2018).
  23. [25]C.-M. Li, Y. Yan, J.-J. Geng, Y.-F. Huang, and H.-S. Zong, Physical Review D 98, 083013 (2018).
  24. [26]Z. Bai and Y.-X. Liu, AIP Conference Proceedings 2127, 020030 (2019).
  25. [27]A. Masayuki and Y. Koichi, Nuclear Physics A 504, 668 (1989).
  26. [28]J. O. Andersen and M. Strickland, Physical Review D 66, 105001 (2002).
  27. [29]M. Fortin, C. Providência, A. R. Raduta, F. Gulminelli, J. L. Zdunik, P. Haensel, and M. Bejger, Phys. Rev. C 94, 035804 (2016).
  28. [30]C.-M. Li, J.-L. Zhang, Y. Yan, Y.-F. Huang, and H.-S. Zong, Physical Review D 97, 103013 (2018).
  29. [31]H. Chen, M. Baldo, G. F. Burgio, and H. J. Schulze, Phys. Rev. D 84, 105023 (2011), arXiv:1107.2497 [nucl-th].
  30. [32]K. Masuda, T. Hatsuda, and T. Takatsuka, PTEP 2013, 073D01 (2013), arXiv:1212.6803 [nucl-th].DOI
  31. [33]J. I. Kapusta and T. Welle, Phys. Rev. C 104, L012801 (2021).
  32. [34]A. Akmal, V. R. Pandharipande, and D. G. Ravenhall, Phys. Rev. C 58, 1804 (1998), arXiv:nucl-th/9804027.DOI
  33. [35]T. Hatsuda and T. Kunihiro, Progress of Theoretical Physics 74, 765 (1985).
  34. [36]Y. Nambu and G. Jona-Lasinio, Phys. Rev. 122, 345 (1961).
  35. [37]T. Kunihiro and T. Hatsuda, Prog. Theor. Phys. 71, 1332 (1984).
  36. [38]M. Buballa, Physics Reports 407, 205 (2005).
  37. [39]L.-K. Yang, X. Luo, and H.-S. Zong, Phys. Rev. D 100, 094012 (2019), arXiv:1910.13185 [nucl-th].DOI
  38. [40]L.-Q. Su, C. Shi, Y.-H. xia, and H. Zong, “Color superconductivity with self-consistent njl-type model,” (2020), arXiv:2008.07678 [nucl-th].DOI
  39. [41]T. Zhao, W. Zheng, F. Wang, C.-M. Li, Y. Yan, Y.-F. Huang, and H.-S. Zong, Physical Review D 100, 043018 (2019).
  40. [42]R. C. Pereira, P. Costa, and C. Providência, Physical Review D 94, 094001 (2016).
  41. [43]J. Kapusta and C. Gale, Finite-Temperature Field Theory: Principles and Applications, Cambridge Monographs on Mathematical Physics (Cambridge University Press, 2006).
  42. [44]N. K. Glendenning, Compact stars: Nuclear physics, particle physics and general relativity (Springer Science & Business Media, 2012).
  43. [45]Y. Aoki, Z. Fodor, S. Katz, and K. Szabó, Physics Letters B 643, 46 (2006).
  44. [46]K. Fukushima and T. Hatsuda, Rept. Prog. Phys. 74, 014001 (2011), arXiv:1005.4814 [hep-ph].DOI
  45. [47]P. Costa, M. Ferreira, H. Hansen, D. P. Menezes, and C. Providência, Phys. Rev. D 89, 056013 (2014), arXiv:1307.7894 [hep-ph].DOI
  46. [48]S. Y. Lau, P. T. Leung, and L. M. Lin, Phys. Rev. D 95, 101302 (2017), arXiv:1705.01710 [astro-ph.HE].DOI
  47. [49]M. Marczinkowski, D. Blaschke, K. Redlich, and C. Sasaki, Phys. Rev. D 98, 103021 (2018), arXiv:1805.06886 [nucl-th].
  48. [50]J. L. Zdunik, M. Fortin, and P. Haensel, Astronomy and Astrophysics 599, A119 (2016).
  49. [51]D. G. Yakovlev and C. J. Pethick, Ann. Rev. Astron. Astrophys. 42, 169 (2004), arXiv:astro-ph/0402143.DOI
  50. [52]T. Damour and A. Nagar, Physical Review D 80, 084035 (2009).
  51. [53]A. Bauswein, N.-U. F. Bastian, D. Blaschke, K. Chatzioannou, J. A. Clark, T. Fischer, H.-T. Janka, O. Just, M. Oertel, and N. Stergioulas, AIP Conference Proceedings 2127, 020013 (2019).
  52. [54]E. Fonseca et al., Astrophys. J. 832, 167 (2016), arXiv:1603.00545 [astro-ph.HE].DOI
  53. [55]J. Antoniadis et al., Science 340, 6131 (2013), arXiv:1304.6875 [astro-ph.HE].DOI
  54. [56]H. T. Cromartie et al. (NANOGrav), Nature Astron. 4, 72 (2019), arXiv:1904.06759 [astro-ph.HE].DOI
  55. [57]M. Linares, T. Shahbaz, and J. Casares, Astrophys. J. 859, 54 (2018), arXiv:1805.08799 [astro-ph.HE].

Paper details

vlnc:W_b9x9gd47
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