betaFeedback

WorksPhysics and Astronomy

Parton distribution functions of the proton in a hot isospin-asymmetric nuclear medium and finite nuclei

Abstract

When a nucleon is bound in a nucleus, its quark distributions are modified, and the change in the spin-dependent distributions can differ in size from that in the spin-independent ones. We compute the unpolarized f1(x)f_1(x), helicity g1(x)g_1(x), and transversity h1(x)h_1(x) distributions of the uu and dd quarks in a proton immersed in hot and isospin-asymmetric nuclear matter. The proton is described in the diquark spectator model, with the medium effects entering through the effective quark, diquark, and proton masses computed using the chiral SU(3) quark mean field (CQMF) model. The axial and tensor charges are obtained for baryon densities up to saturation density and temperatures T=0T = 0 and 0.10.1 GeV, in both symmetric nuclear matter and pure neutron matter. The distributions are then carried to 40^{40}Ca, 56^{56}Fe, 197^{197}Au, and 208^{208}Pb at zero temperature through the local density approximation. In symmetric matter at saturation density and zero temperature, the peaks of the three uu quark distributions move toward larger xx and fall least for the unpolarized distribution and most for the transversity distribution, so the spin-dependent uu quark distributions are modified more strongly. Between vacuum and saturation density, the axial charge of the proton drops from 1.281.28 to 0.8440.844, and the tensor charge of the uu quark decreases while that of the dd quark grows in magnitude. At the center of each nucleus, the axial charge of the proton rises from 0.6990.699 in 40^{40}Ca to 0.8170.817 in 208^{208}Pb as the central baryon density decreases, which suggests that the modification at the center follows the central density rather than the mass number.

Introduction

The internal structure of a nucleon bound in a nucleus differs from that of a free nucleon. The European Muon Collaboration (EMC) measured the ratio F2N(Fe)/F2N(D)F_2^N(\mathrm{Fe})/F_2^N(D) of the structure functions per nucleon for iron and deuterium, and found the iron value to be smaller [1]. Electron scattering at SLAC confirmed the effect on steel and aluminum [2, 3], and a later measurement traced its atomic weight dependence over nine nuclei from deuterium to gold [4]. Fermi motion alone predicts a cross-section ratio σFe/σD2=1.069\sigma_{\mathrm{Fe}}/\sigma_{\mathrm{D2}} = 1.069 at a longitudinal momentum fraction x=0.65x = 0.65, where the measurement gives 0.870±0.0170.870 \pm0.017 [2]. The effect does not correlate directly with the binding energy per nucleon, which peaks near iron while the cross-section ratios continue to fall for heavier nuclei [4]. This suggests that the binding energy alone may not govern the size of the effect. Corrections applied at the nuclear level, through binding and Fermi motion, are not sufficient, and reproducing the data requires the internal quark structure of the bound nucleon itself to be modified [5]. Interpretation in a finite nucleus is complicated because even in 197Au{}^{197}\mathrm{Au}, fewer than half of the nucleons sit in the constant-density interior, and the experimental data have therefore been extrapolated to infinite nuclear matter in the local density approximation [6]. In the present work, the proton is placed in such a medium, and we compute its distributions as the baryon density, temperature, and isospin asymmetry are varied separately. The distributions are then carried to the finite nuclei 40Ca{}^{40}\mathrm{Ca}, 56Fe{}^{56}\mathrm{Fe}, 197Au{}^{197}\mathrm{Au}, and 208Pb{}^{208}\mathrm{Pb} at zero temperature through the local density approximation. For the proton and neutron densities, we use Woods-Saxon distributions with the radius and diffuseness parameters of Ref. [7], obtained from fits to Dirac-Hartree-Bogoliubov densities, as in Ref. [8].

The quark structure of the proton at leading twist is described by three distributions, the unpolarized f1(x)f_1(x), the helicity g1(x)g_1(x), and the transversity h1(x)h_1(x). The distribution f1f_1 counts the quarks of a given flavor. The other two give the difference between quarks polarized parallel and antiparallel to the proton spin, longitudinally for g1g_1 and transversely for h1h_1. In Ref. [9], the light-front quark correlation function is shown to require nine distribution functions, three each at twists 2, 3, and 4. Only f1f_1, g1g_1, and h1h_1 enter a complete quark-parton model of the nucleon at leading twist. Of the three, only h1h_1 is chiral odd. It does not contribute to inclusive deep inelastic scattering (DIS), where f1f_1 and g1g_1 are measured [9]. The distribution h1h_1 originates in the analysis of dimuon production from polarized proton-proton collisions, where the average transverse spin of the quarks enters the cross section alongside their average helicity [10]. The first extraction of h1h_1 within collinear factorization used pion-pair production in semi-inclusive DIS and in proton-proton collisions, where h1h_1 pairs with a dihadron fragmentation function in a simple product [11]. A recent global analysis of dihadron production extracted h1h_1 together with the fragmentation functions, and determined the tensor charge gTqg_T^q, its first moment, both with and without lattice results included as a prior [12]. In contrast, the helicity distributions and their uncertainties have been determined from a global analysis of polarized DIS and proton-proton data [13]. These results show how unevenly the three distributions are known in vacuum, with transversity the least constrained of the three.

In-medium nucleon distributions have been computed in several quark-level frameworks. In Ref. [14], the Nambu-Jona-Lasinio (NJL) model is used to treat the nucleon as a relativistic quark-diquark state in mean-field nuclear matter. The quenching in the valence region was found to come from the vector field, while the scalar potential, taken together with Fermi motion, produces an enhancement near x≃0.7x \simeq0.7. A further NJL calculation obtains the nucleon from the homogeneous Faddeev equation with proper-time regularization, and the constituent quark mass falls from 0.40 to 0.32 GeV and the nucleon mass from 0.94 to 0.75 GeV at saturation density [15]. Carried to finite nuclei through a shell model and a convolution, it found the spin-independent distributions quenched at large xx and enhanced at small xx, and the polarized effect roughly twice the unpolarized one [5]. The quark-meson coupling (QMC) model supplies the medium to a nucleon built in the MIT bag, where the dominant contributions are the σ\sigma-generated masses and the ω\omega vector potential, the polarized and unpolarized effects being about the same size with an axial-charge suppression of 8.9% [16]. The chiral quark-soliton model reaches nuclear-matter saturation from an attractive nucleon-nucleon interaction with vector-meson exchange, and describes the EMC effect together with the related Drell-Yan measurements [17]. In a light-front constituent quark treatment with QMC input, the quark momentum is shifted by the vector potential alongside the replacement of the masses [18]. The present calculation carries no vector potential, and the medium enters through the effective quark, diquark, and proton masses alone, which in the finite nuclei are evaluated at the local baryon density and isospin asymmetry. We therefore compute the response of the distributions to those masses. Since the valence quenching in Ref. [14] comes from the vector field, the distributions obtained here may not show the quenching seen in the EMC measurements.

All three distributions of the free proton have been obtained in the valence sector of Basis Light-Front Quantization (BLFQ), as forward limits of the corresponding generalized parton distributions (GPDs) [19]. With one dynamical gluon in the Fock space, the same approach gives the unpolarized and helicity quark distributions together with the gluon distribution [20]. That calculation has since been embedded in QMC nuclear matter, where the valence component of the proton loses probability to the quark-gluon component as the density rises [21]. That in-medium calculation covers unpolarized distributions in symmetric matter at zero temperature. In Ref. [22], the NJL quark-diquark treatment is extended to asymmetric matter with an isovector-vector ρ0\rho^{0} field, and the EMC effect is found to be largest near Z/N≃0.6Z/N \simeq0.6. We are not aware of an in-medium calculation of the proton transversity distribution. None of the in-medium calculations cited above is carried out at finite temperature.

For the mesons, in-medium valence distributions of the pion have been obtained from a light-front wave function (LFWF) carrying the QMC effective quark masses [23]. The regulator mass is kept at its vacuum value of 600 MeV, and the kaon calculation makes the same choice [24]. In the light-front quark model, the Gaussian parameter has been held density independent because its medium dependence is not well understood [25], and fixing the harmonic parameter leaves the constituent mass from the NJL model as the only medium-dependent input [26]. In Ref. [27], the NJL model with QMC input gives the pion and kaon valence distributions with light-quark momenta shifted by the vector potential and rescaled xx, subject to the baryon number and momentum sum rules in the medium [28]. All these calculations stay in symmetric nuclear matter at zero temperature.

Valence quark distributions of the pion have been obtained in isospin-asymmetric nuclear matter at finite temperature, with the effective quark masses taken from the chiral SU(3) quark mean field (CQMF) model [29]. The distributions respond more strongly to baryon density than to temperature or isospin asymmetry, and the peak of the valence distribution moves toward larger xx with increasing density. The same construction has been extended to the kaon and antikaon doublets, whose valence quark and antiquark carry unequal masses [30]. In a denser medium, the light quark carries a smaller fraction of the longitudinal momentum and the strange antiquark a larger one. In the CQMF model, quarks are confined inside the baryon and interact through the scalar fields σ\sigma and ζ\zeta together with the vector fields ω\omega and ρ\rho [31]. The broken scale invariance of QCD enters through a dilaton field χ\chi that reproduces the trace anomaly [32]. The scalar-isovector field δ\delta vanishes in symmetric matter and splits the effective masses of the uu and dd quarks when the medium is asymmetric [33]. The pion and kaon calculations thus describe the valence quark structure of the pseudoscalar mesons at finite baryon density, temperature, and isospin asymmetry.

Motivated by this, we aim to examine the influence of the medium on the three leading-twist parton distributions of the proton within the diquark spectator model [34], with the in-medium quark, diquark, and proton masses taken from the CQMF model. In vacuum, the same diquark model has been used to study the chiral-odd GPDs of the octet baryons, with the transversity distributions and tensor charges obtained in the forward limit [35]. We evaluate the distributions of the uu and dd quarks across a range of baryon density, temperature, and isospin asymmetry, and obtain from them the second Mellin moment ⟨x⟩q\langle x\rangle^q of f1f_1, together with the axial charge Δq\Delta q and the tensor charge δq\delta q as the first moments of g1g_1 and h1h_1. The same three distributions are then obtained at zero temperature for a proton at the center of each of the four nuclei, and as a function of the radial distance in 208Pb{}^{208}\mathrm{Pb}. Compared with the in-medium calculations cited above, we obtain in addition the transversity distribution and the tensor charge of the proton in the medium, and we follow all three distributions at finite temperature. Medium effects are carried by the effective masses alone, and we use the vacuum definition of the longitudinal momentum fraction, as was done for the pion and the kaon [29, 30].

The rest of this paper is organized as follows. In § 0.2, we outline the CQMF model and the in-medium masses it supplies. In § 0.3, we introduce the diquark spectator model and the leading-twist distributions, and discuss how they respond to the baryon density, temperature, and isospin asymmetry of the medium. In § 0.4, we obtain the distributions of a proton bound in finite nuclei. Finally, in § 0.5, we summarize our findings.

Chiral SU(3) quark mean field model

We obtain the in-medium quark and baryon masses from the CQMF model, in which the total effective Lagrangian density is written as [33]

Leff=Lq0+Lqm+LΣΣ+LVV+LχSB+LΔm+Lc,(1)\mathcal{L}_{\mathrm{eff}} = \mathcal{L}_{q0} + \mathcal{L}_{qm} + \mathcal{L}_{\Sigma\Sigma} + \mathcal{L}_{VV} + \mathcal{L}_{\chi\mathrm{SB}} + \mathcal{L}_{\Delta m} + \mathcal{L}_{c}, \tag*{(1)}

where Lq0=ψˉiγμ∂μψ\mathcal{L}_{q0} = \bar{\psi}i\gamma^\mu\partial_\mu\psi is the free part for massless quarks, Lqm\mathcal{L}_{qm} is the chiral SU(3) invariant quark-meson interaction, and LΣΣ\mathcal{L}_{\Sigma\Sigma} and LVV\mathcal{L}_{VV} are the self-interactions of the scalar and vector mesons. The term LχSB\mathcal{L}_{\chi\mathrm{SB}} breaks chiral symmetry explicitly, LΔm=−ΔmsψˉS1ψ\mathcal{L}_{\Delta m} = -\Delta m_s\bar{\psi}S_1\psi with S1=diag⁡(0,0,1)S_1 = \operatorname{diag}(0,0,1) shifts the mass of the strange quark, and Lc=−ψˉχcψ\mathcal{L}_c = -\bar{\psi}\chi_c\psi confines the quarks. In the mean-field approximation, the meson field operators are replaced by their classical expectation values, only the timelike components of the vector fields survive, and the pseudoscalar fields vanish by parity conservation [32]. For nuclear matter, we retain the scalar fields σ\sigma, ζ\zeta, and δ\delta, the dilaton χ\chi, and the vector fields ω\omega and ρ\rho. The couplings of the quarks to the scalar-isoscalar fields σ\sigma and ζ\zeta follow from the SU(3) flavor structure of Lqm\mathcal{L}_{qm} and are expressed through a single coupling constant gsg_s [31],

gσu=gσd=gs2,gσs=gs,gζu=gζd=gζs=0.(2)g_{\sigma}^{u}=g_{\sigma}^{d}=\frac{g_s}{\sqrt{2}}, \qquad g_{\sigma}^{s}=g_s, \qquad g_{\zeta}^{u}=g_{\zeta}^{d}=g_{\zeta}^{s}=0. \tag*{(2)}

The effective quark masses are written as [30]

mq∗=−gσqσ−gζqζ−gδqI3qδ+Δmq,(3)m_q^*=-g_{\sigma}^{q}\sigma-g_{\zeta}^{q}\zeta-g_{\delta}^{q}I_3^q\delta+\Delta m_q, \tag*{(3)}

where I3qI_3^q is the third component of isospin of the quark, with I3u=−I3d=1/2I_3^u=-I_3^d=1/2 and I3s=0I_3^s=0, and Δmu=Δmd=0\Delta m_u=\Delta m_d=0. The δ\delta field couples to the uu and dd quarks with the same strength, gδu=gδd=gs/2g_{\delta}^{u}=g_{\delta}^{d}=g_s/\sqrt{2}, and not to the ss quark gδs=0g_{\delta}^{s}=0, so the sign of its contribution to each quark mass is fixed by I3qI_3^q alone [30]. Equations (2) and (3) then give

mu∗=−gs2(σ+δ2),md∗=−gs2(σ−δ2),ms∗=−gsζ+Δms.(4)m_u^*=-\frac{g_s}{\sqrt{2}}\left(\sigma+\frac{\delta}{2}\right), \qquad m_d^*=-\frac{g_s}{\sqrt{2}}\left(\sigma-\frac{\delta}{2}\right), \qquad m_s^*=-g_s\zeta+\Delta m_s. \tag*{(4)}

The vacuum values of the nonstrange and strange scalar fields are fixed by the pion and kaon decay constants as σ0=−fπ\sigma_0=-f_\pi and ζ0=(fπ−2fK)/2\zeta_0=(f_\pi-2f_K)/\sqrt{2} [32].

The scalar meson self-interaction, including the scalar-isovector field δ\delta, is written as [32, 33]

LΣΣ=−12k0χ2(σ2+ζ2+δ2)+k1(σ2+ζ2+δ2)2+k2(σ42+δ42+3σ2δ2+ζ4)+k3χ(σ2−δ2)ζ−k4χ4−14χ4ln⁡χ4χ04+ξ3χ4ln⁡(σ2−δ2)ζσ02ζ0.(5)\begin{aligned} \mathcal{L}_{\Sigma\Sigma}={}&-\frac{1}{2}k_0\chi^2\left(\sigma^2+\zeta^2+\delta^2\right)+k_1\left(\sigma^2+\zeta^2+\delta^2\right)^2 \\ &+k_2\left(\frac{\sigma^4}{2}+\frac{\delta^4}{2}+3\sigma^2\delta^2+\zeta^4\right)+k_3\chi\left(\sigma^2-\delta^2\right)\zeta-k_4\chi^4 \\ &-\frac{1}{4}\chi^4\ln\frac{\chi^4}{\chi_0^4}+\frac{\xi}{3}\chi^4\ln\frac{\left(\sigma^2-\delta^2\right)\zeta}{\sigma_0^2\zeta_0}. \tag*{(5)} \end{aligned}

The first logarithmic term, −14χ4ln⁡(χ4/χ04)-\frac{1}{4}\chi^4\ln(\chi^4/\chi_0^4), breaks scale invariance and makes the trace of the energy-momentum tensor proportional to χ4\chi^4. Comparison with the trace anomaly of QCD then identifies the dilaton with the gluon condensate. The coefficient ξ\xi of the second logarithm is guided by the one-loop QCD β\beta function, which suggests ξ=6/33\xi=6/33 for three colors and three flavors [32]. The vector meson self-interaction is written as

LVV=12χ2χ02(mω2ω2+mρ2ρ2)+g4(ω4+6ω2ρ2+ρ4),(6)\mathcal{L}_{VV}=\frac{1}{2}\frac{\chi^2}{\chi_0^2}\left(m_\omega^2\omega^2+m_\rho^2\rho^2\right)+g_4\left(\omega^4+6\omega^2\rho^2+\rho^4\right), \tag*{(6)}

where the masses mωm_\omega and mρm_\rho are taken as 783 MeV. The explicit chiral symmetry breaking term, which gives the pseudoscalar mesons nonvanishing masses and satisfies the partially conserved axial-vector current relations for the pion and the kaon, is written as

LχSB=−χ2χ02[mπ2fπσ+(2mK2fK−mπ2fπ2)ζ].(7)\mathcal{L}_{\chi SB}=-\frac{\chi^2}{\chi_0^2}\left[m_\pi^2f_\pi\sigma+\left(\sqrt{2}m_K^2f_K-\frac{m_\pi^2f_\pi}{\sqrt{2}}\right)\zeta\right]. \tag*{(7)}

The confining term Lc\mathcal{L}_{c} in Eq. (1) is built from the scalar-vector potential [31, 33]

χc(r)=14kcr2(1+γ0)(8)\chi_{c}(r)=\frac{1}{4}k_{c}r^{2}\left(1+\gamma^{0}\right) \tag*{(8)}

and each quark in the meson mean fields obeys the Dirac equation

[−iα⋅∇+χc(r)+βmq∗]ψq=eqψq(9)\left[-i\boldsymbol{\alpha}\cdot\nabla+\chi_{c}(r)+\beta m_{q}^{*}\right]\psi_{q}=e_{q}\psi_{q} \tag*{(9)}

where α\boldsymbol{\alpha} and β\beta are the Dirac matrices and mq∗m_{q}^{*} is given by Eq. (3). For this potential, the energy eqe_{q} of a quark in the lowest (1S1S) state follows from the condition

eq=mq∗+3kc2(eq+mq∗)(10)e_{q}=m_{q}^{*}+\frac{3\sqrt{k_{c}}}{\sqrt{2\left(e_{q}+m_{q}^{*}\right)}} \tag*{(10)}

The strength of the confining potential, kc=300k_{c}=300 MeV fm−2^{-2}, is fixed by requiring the model to reproduce the saturation properties of symmetric nuclear matter.

The energy of a nucleon built from three quarks of energy eqe_{q} is written as [30, 33]

Ei=∑qnqieq+Espini(11)E_{i}=\sum_{q}n_{q}^{i}e_{q}+E_{\mathrm{spin}}^{i} \tag*{(11)}

where nqin_{q}^{i} is the number of quarks of flavor qq in the nucleon i=p,ni=p,n and EspiniE_{\mathrm{spin}}^{i} accounts for the spin correlations [36]. The independent-quark state is not translationally invariant, so EiE_{i} contains the energy of a spurious center-of-mass motion [37]. The effective nucleon mass in the medium is obtained by removing this contribution [36, 37],

Mi∗=Ei2−⟨Pc.m.2⟩i,⟨Pc.m.2⟩i=∑qnqi⟨pq2⟩(12)M_{i}^{*}=\sqrt{E_{i}^{2}-\left\langle P_{\mathrm{c.m.}}^{2}\right\rangle_{i}},\qquad\left\langle P_{\mathrm{c.m.}}^{2}\right\rangle_{i}=\sum_{q}n_{q}^{i}\left\langle p_{q}^{2}\right\rangle \tag*{(12)}

where the quarks are taken to move independently in the confining potential. For the equally mixed scalar-vector harmonic potential of Eq. (8), the mean square momentum of a quark in its 1S1S state is [30, 37]

⟨pq2⟩=(11eq+mq∗)(eq2−mq∗2)6(3eq+mq∗)(13)\left\langle p_{q}^{2}\right\rangle=\frac{\left(11e_{q}+m_{q}^{*}\right)\left(e_{q}^{2}-m_{q}^{*2}\right)}{6\left(3e_{q}+m_{q}^{*}\right)} \tag*{(13)}

Following Ref. [36], EspiniE_{\mathrm{spin}}^{i} is fixed by the free nucleon mass and is assumed not to change in the nuclear medium. We take Espinp=Espinn=−567.012E_{\mathrm{spin}}^{p}=E_{\mathrm{spin}}^{n}=-567.012 MeV, which reproduces the nucleon mass in vacuum MN=939.02M_{N}=939.02 MeV. The Δ\Delta baryons are not included as constituents of the medium. Their masses are evaluated in the same way from the quark energies in nuclear matter, with EspinΔE_{\mathrm{spin}}^{\Delta} fixed by the vacuum mass MΔ=1232M_{\Delta}=1232 MeV.

At finite temperature TT, the thermodynamic potential per unit volume of isospin-asymmetric nuclear matter is written as [30, 33]

Ω=−γNT(2π)3∑i=p,n∫d3k{ln⁡[1+e−(Ei∗(k)−μi∗)/T]+ln⁡[1+e−(Ei∗(k)+μi∗)/T]}−LM−Vvac,(14)\Omega=-\frac{\gamma_N T}{(2\pi)^3}\sum_{i=p,n}\int d^3 k\left\{\ln\left[1+e^{-(E_i^*(k)-\mu_i^*)/T}\right]+\ln\left[1+e^{-(E_i^*(k)+\mu_i^*)/T}\right]\right\}-\mathcal{L}_M-\mathcal{V}_{\mathrm{vac}}, \tag*{(14)}

where γN=2\gamma_N=2 is the spin degeneracy of the nucleons, Ei∗(k)=k2+Mi∗2E_i^*(k)=\sqrt{k^2+M_i^{*2}}, LM=LΣΣ+LVV+LχSB\mathcal{L}_M=\mathcal{L}_{\Sigma\Sigma}+\mathcal{L}_{VV}+\mathcal{L}_{\chi\mathrm{SB}}, and Vvac\mathcal{V}_{\mathrm{vac}} is subtracted to obtain zero vacuum energy. The vector fields shift the chemical potential μi\mu_i of the nucleon to the effective value [31]

μi∗=μi−gωNω−gρNI3iρ,(15)\mu_i^*=\mu_i-g_\omega^N\omega-g_\rho^N I_3^i\rho, \tag*{(15)}

with I3p=−I3n=1/2I_3^p=-I_3^n=1/2. The vector and scalar densities of the nucleon ii are

ρvi=γN(2π)3∫d3k[fi(k)−fˉi(k)],ρsi=γN(2π)3∫d3kMi∗Ei∗(k)[fi(k)+fˉi(k)],(16)\rho_v^i=\frac{\gamma_N}{(2\pi)^3}\int d^3 k\left[f_i(k)-\bar{f}_i(k)\right],\qquad\rho_s^i=\frac{\gamma_N}{(2\pi)^3}\int d^3 k\frac{M_i^*}{E_i^*(k)}\left[f_i(k)+\bar{f}_i(k)\right], \tag*{(16)}

where fi(k)={1+e[Ei∗(k)−μi∗]/T}−1f_i(k)=\left\{1+e^{[E_i^*(k)-\mu_i^*]/T}\right\}^{-1} and fˉi(k)={1+e[Ei∗(k)+μi∗]/T}−1\bar{f}_i(k)=\left\{1+e^{[E_i^*(k)+\mu_i^*]/T}\right\}^{-1} are the particle and antiparticle distribution functions. The baryon density is ρB=ρvp+ρvn\rho_B=\rho_v^p+\rho_v^n, and the isospin asymmetry of the medium is described by η=(ρvn−ρvp)/(2ρB)\eta=(\rho_v^n-\rho_v^p)/(2\rho_B), which vanishes in symmetric nuclear matter and equals 0.5 in pure neutron matter. The meson fields follow from the stationarity conditions ∂Ω/∂φ=0\partial\Omega/\partial\varphi=0, where φ\varphi stands for each of σ\sigma, ζ\zeta, δ\delta, χ\chi, ω\omega, and ρ\rho. The nucleon term of Eq. (14), which we denote by ΩN\Omega_N, depends on σ\sigma and δ\delta only through Mi∗M_i^* and on ω\omega and ρ\rho only through μi∗\mu_i^*. At fixed TT and μi\mu_i, its derivatives are

∂ΩN∂Mi∗=ρsi,∂ΩN∂μi∗=−ρvi,(17)\frac{\partial\Omega_N}{\partial M_i^*}=\rho_s^i,\qquad\frac{\partial\Omega_N}{\partial\mu_i^*}=-\rho_v^i, \tag*{(17)}

so that the σ\sigma and δ\delta equations acquire the sources ∑i=p,n(∂Mi∗/∂φ)ρsi\sum_{i=p,n}(\partial M_i^*/\partial\varphi)\rho_s^i, and the ω\omega and ρ\rho equations acquire gωN∑i=p,nρvig_\omega^N\sum_{i=p,n}\rho_v^i and gρN∑i=p,nI3iρvig_\rho^N\sum_{i=p,n}I_3^i\rho_v^i. The equations of motion for the six mean fields then read

∂Ω∂σ:k0χ2σ−4k1(σ2+ζ2+δ2)σ−2k2(σ3+3σδ2)−2k3χσζ−2ξ3χ4σσ2−δ2+χ2χ02mπ2fπ+∑i=p,n∂Mi∗∂σρsi=0,(18)\frac{\partial\Omega}{\partial\sigma}: \quad k_0\chi^2\sigma-4k_1\left(\sigma^2+\zeta^2+\delta^2\right)\sigma-2k_2\left(\sigma^3+3\sigma\delta^2\right)-2k_3\chi\sigma\zeta-\frac{2\xi}{3}\frac{\chi^4\sigma}{\sigma^2-\delta^2}+\frac{\chi^2}{\chi_0^2}m_\pi^2f_\pi+\sum_{i=p,n}\frac{\partial M_i^*}{\partial\sigma}\rho_s^i=0, \tag*{(18)}
∂Ω∂ζ:k0χ2ζ−4k1(σ2+ζ2+δ2)ζ−4k2ζ3−k3χ(σ2−δ2)−ξ3χ4ζ+χ2χ02(2mK2fK−mπ2fπ2)=0,(19)\frac{\partial\Omega}{\partial\zeta}: \quad k_0\chi^2\zeta-4k_1\left(\sigma^2+\zeta^2+\delta^2\right)\zeta-4k_2\zeta^3-k_3\chi\left(\sigma^2-\delta^2\right)-\frac{\xi}{3}\frac{\chi^4}{\zeta}+\frac{\chi^2}{\chi_0^2}\left(\sqrt{2}m_K^2f_K-\frac{m_\pi^2f_\pi}{\sqrt{2}}\right)=0, \tag*{(19)}
∂Ω∂δ:k0χ2δ−4k1(σ2+ζ2+δ2)δ−2k2(δ3+3σ2δ)+2k3χδζ+2ξ3χ4δσ2−δ2+∑i=p,n∂Mi∗∂δρsi=0,(20)\frac{\partial\Omega}{\partial\delta}: \quad k_0\chi^2\delta-4k_1\left(\sigma^2+\zeta^2+\delta^2\right)\delta-2k_2\left(\delta^3+3\sigma^2\delta\right)+2k_3\chi\delta\zeta+\frac{2\xi}{3}\frac{\chi^4\delta}{\sigma^2-\delta^2}+\sum_{i=p,n}\frac{\partial M_i^*}{\partial\delta}\rho_s^i=0, \tag*{(20)}
∂Ω∂χ:k0χ(σ2+ζ2+δ2)−k3(σ2−δ2)ζ+[1+4k4+4ln⁡χχ0−4ξ3ln⁡((σ2−δ2)ζσ02ζ0)]χ3+2χχ02[mπ2fπσ+(2mK2fK−mπ2fπ2)ζ]−χχ02(mω2ω2+mρ2ρ2)=0,(21)\frac{\partial\Omega}{\partial\chi}: \quad k_0\chi\left(\sigma^2+\zeta^2+\delta^2\right)-k_3\left(\sigma^2-\delta^2\right)\zeta+\left[1+4k_4+4\ln\frac{\chi}{\chi_0}-\frac{4\xi}{3}\ln\left(\frac{\left(\sigma^2-\delta^2\right)\zeta}{\sigma_0^2\zeta_0}\right)\right]\chi^3+\frac{2\chi}{\chi_0^2}\left[m_\pi^2f_\pi\sigma+\left(\sqrt{2}m_K^2f_K-\frac{m_\pi^2f_\pi}{\sqrt{2}}\right)\zeta\right]-\frac{\chi}{\chi_0^2}\left(m_\omega^2\omega^2+m_\rho^2\rho^2\right)=0, \tag*{(21)}
∂Ω∂ω:−χ2χ02mω2ω−4g4ω3−12g4ωρ2+gωN∑i=p,nρvi=0,(22)\frac{\partial\Omega}{\partial\omega}: \quad-\frac{\chi^2}{\chi_0^2}m_\omega^2\omega-4g_4\omega^3-12g_4\omega\rho^2+g_\omega^N\sum_{i=p,n}\rho_v^i=0, \tag*{(22)}
∂Ω∂ρ:−χ2χ02mρ2ρ−4g4ρ3−12g4ω2ρ+gρN∑i=p,nI3iρvi=0.(23)\frac{\partial\Omega}{\partial\rho}: \quad-\frac{\chi^2}{\chi_0^2}m_\rho^2\rho-4g_4\rho^3-12g_4\omega^2\rho+g_\rho^N\sum_{i=p,n}I_3^i\rho_v^i=0. \tag*{(23)}

The ζ\zeta and χ\chi equations carry no nucleon source, because the uu and dd quark masses of Eq. (4), and hence Mi∗M_i^*, contain neither ζ\zeta nor χ\chi. At given ρB\rho_B, TT, and η\eta, we solve Eqs. (18) to (23) self-consistently together with Eq. (10) for the uu and dd quarks.

The nucleon source terms in Eqs. (18) and (20) contain the derivatives of the effective nucleon masses with respect to the σ\sigma and δ\delta fields. We define the couplings of the nucleon to these two fields as

gσi=−∂Mi∗∂σ,gδi=−∂Mi∗∂δ,(24)g_\sigma^i=-\frac{\partial M_i^*}{\partial\sigma},\qquad g_\delta^i=-\frac{\partial M_i^*}{\partial\delta}, \tag*{(24)}

so that the source terms read −∑i=p,ngσiρsi-\sum_{i=p,n}g_\sigma^i\rho_s^i and −∑i=p,ngδiρsi-\sum_{i=p,n}g_\delta^i\rho_s^i, respectively. We evaluate both derivatives exactly through Eq. (12), separately for the proton and the neutron. They are recomputed together with the mean fields at every ρB\rho_B, TT, and η\eta, so that gσig_\sigma^i and gδig_\delta^i are density-dependent couplings. In the quark mean field model of Ref. [36], which uses the confining potential of Eq. (8), the in-medium σ\sigma-nucleon coupling is identified with the derivative of the effective nucleon mass with respect to σ\sigma. We extend the derivative prescription of Ref. [36] to the scalar-isovector field. We are not aware of an earlier calculation that obtains the nucleon-δ\delta coupling from −∂Mi∗/∂δ-\partial M_i^*/\partial\delta.

The chain rule through mu∗m_u^* and md∗m_d^*, worked out in Appendix § 0.3.1, gives the closed forms

gσi=gs2Mi∗∑q=u,dnqicqi,gδi=gs2Mi∗∑q=u,dnqiI3qcqi,cqi=Eirq−12sq,(25)g_\sigma^i=\frac{g_s}{\sqrt{2}M_i^*}\sum_{q=u,d}n_q^i c_q^i,\qquad g_\delta^i=\frac{g_s}{\sqrt{2}M_i^*}\sum_{q=u,d}n_q^i I_{3q}c_q^i,\qquad c_q^i=E_ir_q-\frac{1}{2}s_q, \tag*{(25)}

where rq=deq/dmq∗r_q=de_q/dm_q^* and sq=d⟨pq2⟩/dmq∗s_q=d\langle p_q^2\rangle/dm_q^* are the derivatives of the quark energy and of its mean square momentum along Eq. (10), given in Eqs. (51) and (53). Here nup=ndn=2n_u^p=n_d^n=2 and ndp=nun=1n_d^p=n_u^n=1. The coupling gδpg_\delta^p is positive and gδng_\delta^n is negative at every ρB\rho_B, TT, and η\eta considered here. Both couplings depend on ρB\rho_B, TT, and η\eta only through σ\sigma and δ\delta, which fix mq∗m_q^*, eqe_q, and Mi∗M_i^*. In isospin-asymmetric matter the δ\delta field is finite, so that mu∗≠md∗m_u^*\ne m_d^* and, in general, gσp≠gσng_\sigma^p\ne g_\sigma^n. At η=0\eta=0, gσNg_\sigma^N decreases from 8.477 in vacuum to 7.960 at ρB=ρ0\rho_B=\rho_0 and T=0T=0, and equals 8.037 at ρB=ρ0\rho_B=\rho_0 and T=0.1T=0.1 GeV. In pure neutron matter (η=0.5\eta=0.5) at ρB=ρ0\rho_B=\rho_0 and T=0T=0, we obtain gσp=7.977g_\sigma^p=7.977 and gσn=7.968g_\sigma^n=7.968, and gδp=1.347g_\delta^p=1.347 and gδn=−1.311g_\delta^n=-1.311.

The parameters of the model are listed in Table I. The decay constants fπf_\pi and fKf_K take the values of Ref. [40]. The pion and kaon masses enter Eq. (7) directly and are therefore reproduced by construction. For the dilaton vacuum value χ0=282\chi_0=282 MeV, the coefficients k0k_0 to k4k_4 are fixed by five vacuum conditions. The σ\sigma, ζ\zeta, and χ\chi equations must hold at σ0\sigma_0, ζ0\zeta_0, and χ0\chi_0, the scalar meson mass must equal the value mσ=417.5m_\sigma=417.5 MeV of Ref. [31], and the sum of the squared η\eta and η′\eta' masses must equal its physical value, mη2+mη′2=(5472+9572)m_\eta^2+m_{\eta'}^2=(547^2+957^2) MeV2^2, as in Refs. [42, 43]. The coupling gs=4.76g_s=4.76 gives the vacuum constituent quark mass mu∗=md∗=313m_u^*=m_d^*=313 MeV, as in Ref. [40], and Δms\Delta m_s is chosen to give ms∗=450m_s^*=450 MeV in vacuum. The confinement strength kck_c and the spin correlation energies are fixed as described after Eqs. (10) and (13). We then fit the nucleon-ω\omega coupling gωNg_\omega^N and the vector coupling g4g_4 to the saturation point of symmetric nuclear matter at T=0T=0. The pressure P=−ΩP=-\Omega must vanish there, and the energy per nucleon, measured from the vacuum nucleon mass, E/A=(Ω+∑i=p,nμiρiv)/ρB−MNE/A=(\Omega+\sum_{i=p,n}\mu_i\rho_i^v)/\rho_B-M_N, must equal −16.00-16.00 MeV at ρ0=0.15\rho_0=0.15 fm−3^{-3}. The resulting parameter set, obtained with the density-dependent coupling gσig_\sigma^i of Eq. (24) in the σ\sigma equation, differs from those of Refs. [30, 44]. Finally, we fit the nucleon-ρ\rho coupling gρNg_\rho^N to the symmetry energy, J=32J=32 MeV at saturation.

The saturation properties of the model are obtained at T=0T=0 with the definitions of Ref. [41]. In symmetric nuclear matter, E/AE/A has its minimum, with vanishing pressure, at the saturation density ρ0\rho_0, and its curvature there defines the incompressibility

ParameterValueTermConstraint
k0k_04.5701∂LM/∂σ∣vac=0\left.\partial\mathcal{L}_M/\partial\sigma\right|_{\mathrm{vac}}=0
k1k_12.6635∂LM/∂ζ∣vac=0\left.\partial\mathcal{L}_M/\partial\zeta\right|_{\mathrm{vac}}=0
k2k_2−9.9879-9.9879∂LM/∂χ∣vac=0\left.\partial\mathcal{L}_M/\partial\chi\right|_{\mathrm{vac}}=0
k3k_3−4.3201-4.3201LΣΣ\mathcal{L}_{\Sigma\Sigma}mσ=417.5 MeVm_\sigma=417.5\ \mathrm{MeV} [31]
k4k_4−0.1613-0.1613mη2+mη′2=(5472+9572) MeV2m_\eta^2+m_{\eta'}^2=(547^2+957^2)\ \mathrm{MeV}^2
ξ\xi0.182one-loop βQCD\beta_{\mathrm{QCD}}, ξ≃6/33\xi\simeq6/33 [32]
χ0\chi_0282 MeVK0≃250 MeVK_0\simeq250\ \mathrm{MeV}
fπf_\pi, fKf_K93, 115 MeVLχSB\mathcal{L}_{\chi\mathrm{SB}}Ref. [40]
mπm_\pi, mKm_K139, 494 MeVπ±\pi^\pm, K±K^\pm masses
gsg_s4.76Lqm\mathcal{L}_{qm}mu∗=md∗=313 MeVm_u^*=m_d^*=313\ \mathrm{MeV} in vacuum
Δms\Delta m_s−11.11-11.11 MeVLΔm\mathcal{L}_{\Delta m}ms∗=450 MeVm_s^*=450\ \mathrm{MeV} in vacuum
kck_c300 MeV fm−2^{-2}Lc\mathcal{L}_cE/AE/A and P=0P=0 at ρ0\rho_0 attainable
EspiniE_{\mathrm{spin}}^i−567.012-567.012 MeVEq. (11)MN=939.02 MeVM_N=939.02\ \mathrm{MeV} in vacuum
EspinΔE_{\mathrm{spin}}^\Delta−309.956-309.956 MeVMΔ+=1232 MeVM_{\Delta^+}=1232\ \mathrm{MeV} in vacuum
mωm_\omega, mρm_\rho783 MeVLVV\mathcal{L}_{VV}ω\omega mass, mρ=mωm_\rho=m_\omega
gωNg_\omega^N, g4g_411.8165, 65.5791Eq. (15), LVV\mathcal{L}_{VV}E/A=−16.00 MeVE/A=-16.00\ \mathrm{MeV}, P=0P=0 at ρ0\rho_0
gρNg_\rho^N9.9922Eq. (15)J=32 MeVJ=32\ \mathrm{MeV} at ρ0\rho_0

Table 1. Parameters of the CQMF model, the term of the Lagrangian or the equation in which each enters, and the constraint that fixes it. Vacuum denotes ρB=T=0\rho_B=T=0. The coefficients k0k_0 to k4k_4 are fixed together by the five conditions listed against them.

K0=9ρ02d2(E/A)dρB2∣ρB=ρ0(26)K_0=9\rho_0^2\left.\frac{d^2(E/A)}{d\rho_B^2}\right|_{\rho_B=\rho_0} \tag*{(26)}

The symmetry energy Esym(ρB)E_{\mathrm{sym}}(\rho_B) is taken as the difference between E/AE/A in pure neutron matter and in symmetric nuclear matter, and JJ and LL are its value and slope at saturation,

Esym(ρB)=EA∣η=0.5−EA∣η=0,J=Esym(ρ0),L=3ρ0dEsymdρB∣ρB=ρ0.(27)E_{\mathrm{sym}}(\rho_B)=\left.\frac{E}{A}\right|_{\eta=0.5}-\left.\frac{E}{A}\right|_{\eta=0},\qquad J=E_{\mathrm{sym}}(\rho_0),\qquad L=3\rho_0\left.\frac{dE_{\mathrm{sym}}}{d\rho_B}\right|_{\rho_B=\rho_0}. \tag*{(27)}

The values of these quantities are listed in Table II, together with the ratio MN∗/MNM_N^*/M_N of the effective to the vacuum nucleon mass at ρ0\rho_0.

ρ0\rho_0 (fm−3^{-3})E/AE/A (MeV)K0K_0 (MeV)JJ (MeV)LL (MeV)MN∗/MNM_N^*/M_N
0.150−16.00-16.0025032.0167.670.684

Table 2. Saturation properties of symmetric nuclear matter at T=0T=0 obtained with the parameters of Table I.

The medium enters the light-front calculation of § 0.3 only through the effective masses obtained in this section. These are the constituent quark masses mu∗m_u^* and md∗m_d^*, the proton mass Mp∗M_p^*, and the average mass ⟨MΔ∗⟩\langle M_\Delta^*\rangle of the four Δ\Delta charge states, each evaluated at the ρB\rho_B, TT, and η\eta of interest, and the diquark masses are built from them,

MS∗=mu∗+md∗,MV∗=34Mp∗+⟨MΔ∗⟩2,(28)M_S^*=m_u^*+m_d^*,\qquad M_V^*=\frac{3}{4}\frac{M_p^*+\langle M_\Delta^*\rangle}{2}, \tag*{(28)}

where a single axial-vector diquark mass is used for both the udud and the uuuu axial-vector diquarks. The factor 34\frac{3}{4} follows Ref. [45], where the diquark mass is three-quarters of the degenerate nucleon and Δ\Delta mass in the absence of one-gluon exchange. Here, the degenerate mass is replaced by the average of Mp∗M_p^* and ⟨MΔ∗⟩\langle M_\Delta^*\rangle, and the one-gluon-exchange shifts of the diquark masses are not included. At zero density and temperature, Eq. (28) gives MS=0.626M_S=0.626 GeV and MV=0.814M_V=0.814 GeV, the vacuum values used in § 0.3. The diquark masses carry no explicit δ\delta term, so the δ\delta field reaches them only through the splitting of mu∗m_u^* and md∗m_d^* and the resulting shifts in Mp∗M_p^* and ⟨MΔ∗⟩\langle M_\Delta^*\rangle. The proton and quark masses MpM_p and mqm_q of § 0.3 take the in-medium values Mp∗M_p^* and mq∗m_q^* obtained here.

The effective masses of the constituent quarks, the scalar and axial-vector diquarks, the proton, and the average Δ\Delta baryon in symmetric nuclear matter are plotted as functions of the baryon density in Figure 1. We begin with T=0T=0, where all five masses decrease monotonically as ρB\rho_B increases from zero to ρ0\rho_0, but by different fractions. The constituent quark mass decreases from mq∗=0.313m_q^*=0.313 GeV in vacuum to 0.1920.192 GeV at ρ0\rho_0, a drop of about 39%. The scalar diquark mass MS∗=mu∗+md∗M_S^*=m_u^*+m_d^* is twice the quark mass at η=0\eta=0 and follows it from 0.626 GeV to 0.384 GeV. The proton mass decreases by about 32%, from 0.939 GeV to 0.643 GeV, so that MN∗/MN=0.684M_N^*/M_N = 0.684, as listed in Table II. The axial-vector diquark mass decreases by about 26%, from 0.814 GeV to 0.600 GeV, less than the proton mass. This can be traced to Eq. (28), where MV∗M_V^* is three-quarters of the mean of Mp∗M_p^* and ⟨MΔ∗⟩\langle M_\Delta^* \rangle, and ⟨MΔ∗⟩\langle M_\Delta^* \rangle falls by about 22%, from 1.232 GeV to 0.958 GeV, the smallest fraction of the five masses.

Figure 1. Effective masses of (a) the constituent quarks, mq∗=mu∗=md∗m_q^* = m_u^* = m_d^*, (b) the scalar and axial-vector diquarks, MS∗=mu∗+md∗M_S^* = m_u^* + m_d^* and MV∗M_V^* of Eq. (28), and (c) the proton, Mp∗M_p^*, and the average Δ\Delta baryon mass ⟨MΔ∗⟩\langle M_\Delta^* \rangle, as functions of the baryon density ρB/ρ0\rho_B/\rho_0 in symmetric nuclear matter (η=0\eta= 0) at T=0T = 0 (solid lines) and T=0.1T = 0.1 GeV (dash-dotted lines). The shifts of these masses between η=0\eta= 0 and η=0.5\eta= 0.5 are listed in Table III.

At T=0.1T = 0.1 GeV (dashed lines in Figure 1), the masses again decrease monotonically with ρB\rho_B but lie above their T=0T = 0 values. For example, at ρB=ρ0\rho_B = \rho_0 the constituent quark mass is mq∗=0.207m_q^* = 0.207 GeV at T=0.1T = 0.1 GeV, against 0.192 GeV at T=0T = 0, which is higher by about 8%. The scalar diquark mass rises by the same fraction, to 0.414 GeV, since it remains twice the quark mass in symmetric matter. The proton mass rises by about 6%, to 0.679 GeV, and the axial-vector diquark mass by about 4%, to 0.626 GeV. As at T=0T = 0, the smaller change of MV∗M_V^* can be traced to Eq. (28), since ⟨MΔ∗⟩\langle M_\Delta^* \rangle rises by only about 3%, from 0.958 GeV to 0.990 GeV. The temperature, therefore, partly offsets the density-driven decrease of the effective masses. The masses that fall most with density, mq∗m_q^* and MS∗M_S^*, also rise most with temperature, followed by Mp∗M_p^*, MV∗M_V^*, and ⟨MΔ∗⟩\langle M_\Delta^* \rangle.

In asymmetric matter, the δ\delta field is sourced by −∑i=p,ngδiρsi-\sum_{i=p,n} g_\delta^i \rho_s^i in Eq. (20), and since gδp>0g_\delta^p > 0 and gδn<0g_\delta^n < 0, this source follows the imbalance between the proton and neutron scalar densities. It vanishes in symmetric nuclear matter and is finite at η=0.5\eta= 0.5, which corresponds to pure neutron matter. Through Eq. (4), δ\delta shifts mu∗m_u^* and md∗m_d^* by equal amounts in opposite directions and drops out of mu∗+md∗m_u^* + m_d^*, which depends on σ\sigma alone. The resulting relative changes from η=0\eta= 0 to η=0.5\eta= 0.5 are listed in Table III. At T=0T = 0, the σ\sigma field also changes with η\eta and lifts both quark masses by 1.26% at ρ0\rho_0, so mu∗m_u^* rises by 2.69% while md∗m_d^* falls by only 0.17%. At T=0.1T = 0.1 GeV, the change of σ\sigma stays within 0.11%, and the splitting is close to symmetric, with 1.25% for mu∗m_u^* and −1.03%-1.03\% for md∗m_d^* at ρ0\rho_0. The scalar diquark mass MS∗=mu∗+md∗M_S^* = m_u^* + m_d^* does not depend on δ\delta and changes by at most 1.26% at T=0T = 0 and 0.11% at T=0.1T = 0.1 GeV, while MV∗M_V^* and Mp∗M_p^* change by at most 0.81% and 1.23% at T=0T = 0 and 0.17% and 0.36% at T=0.1T = 0.1 GeV. The isospin asymmetry has the smallest effect on the masses, at most 2.69% in Table III, for mu∗m_u^* at ρ0\rho_0 and T=0T = 0, compared with the decrease of mq∗m_q^* by about 39% from vacuum to ρ0\rho_0 and its increase by about 8% at ρ0\rho_0 from T=0T = 0 to T=0.1T = 0.1 GeV.

TT (GeV)ρB\rho_BΔm∗/m∗\Delta m^*/m^* (%)
mu∗m_u^*md∗m_d^*MS∗M_S^*MV∗M_V^*Mp∗M_p^*
00.25ρ00.25\rho_00.34−0.17-0.170.080.090.14
00.5ρ00.5\rho_00.91−0.26-0.260.320.260.40
00.75ρ00.75\rho_01.70−0.27-0.270.720.520.78
0ρ0\rho_02.69−0.17-0.171.260.811.23
0.10.25ρ00.25\rho_00.17−0.25-0.25−0.04-0.040.000.03
0.10.5ρ00.5\rho_00.46−0.50-0.50−0.02-0.020.040.11
0.10.75ρ00.75\rho_00.82−0.77-0.770.030.100.22
0.1ρ0\rho_01.25−1.03-1.030.110.170.36

Table 3. Relative change of the effective quark, diquark and proton masses from symmetric (η=0\eta= 0) to asymmetric (η=0.5\eta= 0.5) nuclear matter, Δm∗/m∗=[m∗(η=0.5)−m∗(η=0)]/m∗(η=0)\Delta m^*/m^* = [m^*(\eta= 0.5) - m^*(\eta= 0)]/m^*(\eta= 0), at the listed temperatures and baryon densities.

Parton distribution functions

In this section, we investigate the implications of medium effects on the valence quark structure of the proton through the PDFs. We choose a frame where the proton carries zero transverse momentum, and its four-momentum in vacuum condition is given by

P=[P+,Mp22P+,0T](29)P = \left[P^{+}, \frac{M_p^{2}}{2P^{+}}, \boldsymbol{0}_{T}\right] \tag*{(29)}

using the generic light-front four-vector notation v=[v+,v−,vT]v = [v^{+},v^{-},\boldsymbol{v}_{T}], where MpM_p denotes the proton mass and the four-momentum of the active quark qq is defined as

k=[xP+,k2+kT22xP+,kT].(30)k = \left[xP^{+},\frac{k^{2}+\boldsymbol{k}_{T}^{2}}{2xP^{+}},\boldsymbol{k}_{T}\right]. \tag*{(30)}

Here, the longitudinal momentum fraction carried by an active quark is expressed by x=k+/P+x = k^{+}/P^{+}. For semi-inclusive deep inelastic scattering, the integrated quark-quark correlator is defined as

Φ(x,kT;S)=∫dξ−d2ξT(2π)3eik⋅ξ⟨P;S∣ψˉ(0)W[0,ξ]ψ(ξ)∣P;S⟩∣ξ+=0,(31)\Phi(x,\boldsymbol{k}_{T};S)=\left.\int\frac{d\xi^{-}d^{2}\boldsymbol{\xi}_{T}}{(2\pi)^{3}}e^{ik\cdot\xi}\langle P;S|\bar{\psi}(0)W_{[0,\xi]}\psi(\xi)|P;S\rangle\right|_{\xi^{+}=0}, \tag*{(31)}

with Γ={1,γμ,σμν,iγ5,γ5γμ}\Gamma= \{1,\gamma^{\mu},\sigma_{\mu\nu},i\gamma_{5},\gamma_{5}\gamma^{\mu}\} [46]. Wilson line WW preserves the gauge invariance that connects the points 0 and ξ\xi with an infinitely long staple-shaped path along the lightlike direction nn. In a Lorentz covariant fashion, the light-front decomposition of a vector nn can be written in two light-like vectors n+=[1,0,0T]n_{+}=[1,0,\boldsymbol{0}_{T}] and n−=[0,1,0T]n_{-}=[0,1,\boldsymbol{0}_{T}]. In the light-front gauge, A+=0A^{+}=0, the Wilson line reduces to unity. At the leading twist, the integration over kT\boldsymbol{k}_{T} leads to the collinear quark correlator, whose leading-twist parametrization involves the familiar collinear distributions f1q(x)f_{1}^{q}(x), g1q(x)g_{1}^{q}(x), and h1q(x)h_{1}^{q}(x) [47], which is defined as

Φ(x;S)=12[f1q(x)̸n++g1q(x)SLγ5̸n++h1q(x)iσμνγ5n+μSTν].(32)\Phi(x;S)=\frac{1}{2}\left[f_{1}^{q}(x)\not{n}_{+}+g_{1}^{q}(x)S_{L}\gamma_{5}\not{n}_{+}+h_{1}^{q}(x)i\sigma_{\mu\nu}\gamma_{5}n_{+}^{\mu}S_{T}^{\nu}\right]. \tag*{(32)}

The spin vector SμS^{\mu} of the proton target can be defined in terms of the longitudinal and transverse components {SL,ST}\{S_{L},S_{T}\} as

Sμ=SL(P⋅n−)n+μ−(P⋅n+)n−μMp+STμ.(33)S^{\mu}=S_{L}\frac{(P\cdot n_{-})n_{+}^{\mu}-(P\cdot n_{+})n_{-}^{\mu}}{M_{p}}+S_{T}^{\mu}. \tag*{(33)}

We evaluate this kT\boldsymbol{k}_{T} integrated quark-quark correlator in the spectator approximation, where we consider a proton as a two-body system of an active quark qq and a spectator diquark DD. Depending on the spin of spectator diquark SD={0,1}S_{D}=\{0,1\}, it is termed as scalar SS and axial-vector VV diquarks, respectively. Further, isospin categorizes the axial-vector diquarks as isoscalar (udud-like system) and isovector (uuuu-like system) [48]. So, assuming spin-isospin symmetry, the probabilistic weights among scalar-isoscalar, axial-vector-isoscalar, and axial-vector-isovector are 3:1:2 as the proton wave function [49] is written as

∣p⟩λ=12∣uS(ud)⟩λ−16∣uV(ud)⟩λ+13∣dV(uu)⟩λ.(34)|p\rangle^{\lambda}=\frac{1}{\sqrt{2}}|uS(ud)\rangle^{\lambda}-\frac{1}{\sqrt{6}}|uV(ud)\rangle^{\lambda}+\frac{1}{\sqrt{3}}|dV(uu)\rangle^{\lambda}. \tag*{(34)}

This choice leads to the relations fu=3fu(S)/2+fu(V)/2f^u = 3f^{u(S)}/2 + f^{u(V)}/2 and fd=fd(V)f^d = f^{d(V)}, where f={f1(x),g1(x),h1(x)}f = \{f_1(x), g_1(x), h_1(x)\}. On parameterization, the PDFs can be written in terms of the overlap form of respective scalar ψλqλ\psi^{\lambda}_{\lambda_q} and axial-vector ψλqλDλ\psi^{\lambda}_{\lambda_q\lambda_D} LFWFs, where λq={↑,↓}\lambda_q = \{\uparrow,\downarrow\} and λD={±1,0}\lambda_D = \{\pm1,0\} denote the helicities of an active quark and a spectator diquark, sequentially. The contribution of the scalar diquark is given by

f1q(S)(x)=∫d2kT16π3[∣ψ↑↑(x,kT)∣2+∣ψ↓↑(x,kT)∣2],(35)f_1^{q(S)}(x) = \int\frac{\mathrm{d}^2\mathbf{k}_{\mathrm{T}}}{16\pi^3} \left[|\psi_{\uparrow}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 + |\psi_{\downarrow}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2\right], \tag*{(35)}
g1q(S)(x)=∫d2kT16π3[∣ψ↑↑(x,kT)∣2−∣ψ↓↑(x,kT)∣2],(36)g_1^{q(S)}(x) = \int\frac{\mathrm{d}^2\mathbf{k}_{\mathrm{T}}}{16\pi^3} \left[|\psi_{\uparrow}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 - |\psi_{\downarrow}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2\right], \tag*{(36)}
h1q(S)(x)=∫d2kT16π3[ψ↑↑∗(x,kT)ψ↓↓(x,kT)−ψ↓↑∗(x,kT)ψ↑↓(x,kT)].(37)h_1^{q(S)}(x) = \int\frac{\mathrm{d}^2\mathbf{k}_{\mathrm{T}}}{16\pi^3} \left[\psi_{\uparrow}^{\uparrow*}(x,\mathbf{k}_{\mathrm{T}})\psi_{\downarrow}^{\downarrow}(x,\mathbf{k}_{\mathrm{T}}) - \psi_{\downarrow}^{\uparrow*}(x,\mathbf{k}_{\mathrm{T}})\psi_{\uparrow}^{\downarrow}(x,\mathbf{k}_{\mathrm{T}})\right]. \tag*{(37)}

On the parallel lines, the contribution of the axial-vector diquark is given by

f1q(V)(x)=∫d2kT16π3[∣ψ↑1↑(x,kT)∣2+∣ψ↑0↑(x,kT)∣2+∣ψ↑−1↑(x,kT)∣2+∣ψ↓1↑(x,kT)∣2+∣ψ↓0↑(x,kT)∣2+∣ψ↓−1↑(x,kT)∣2],(38)\begin{aligned} f_1^{q(V)}(x) ={}& \int\frac{\mathrm{d}^2\mathbf{k}_{\mathrm{T}}}{16\pi^3} \big[|\psi_{\uparrow1}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 + |\psi_{\uparrow0}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 + |\psi_{\uparrow-1}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 + |\psi_{\downarrow1}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 \\ &\quad+ |\psi_{\downarrow0}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 + |\psi_{\downarrow-1}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2\big], \tag*{(38)} \end{aligned}
g1q(V)(x)=∫d2kT16π3[∣ψ↑1↑(x,kT)∣2+∣ψ↑0↑(x,kT)∣2+∣ψ↑−1↑(x,kT)∣2−(∣ψ↓1↑(x,kT)∣2+∣ψ↓0↑(x,kT)∣2+∣ψ↓−1↑(x,kT)∣2)],(39)\begin{aligned} g_1^{q(V)}(x) ={}& \int\frac{\mathrm{d}^2\mathbf{k}_{\mathrm{T}}}{16\pi^3} \big[|\psi_{\uparrow1}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 + |\psi_{\uparrow0}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 + |\psi_{\uparrow-1}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 - (|\psi_{\downarrow1}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 \\ &\quad+ |\psi_{\downarrow0}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2 + |\psi_{\downarrow-1}^{\uparrow}(x,\mathbf{k}_{\mathrm{T}})|^2)\big], \tag*{(39)} \end{aligned}
h1q(V)(x)=∫d2kT16π3[ψ↑1↑∗(x,kT)ψ↓1↓(x,kT)+ψ↑0↑∗(x,kT)ψ↓0↓(x,kT)+ψ↑−1↑∗(x,kT)ψ↓−1↓(x,kT)+ψ↓1↑∗(x,kT)ψ↑1↓(x,kT)+ψ↓0↑∗(x,kT)ψ↑0↓(x,kT)+ψ↓−1↑∗(x,kT)ψ↑−1↓(x,kT)].(40)\begin{aligned} h_1^{q(V)}(x) ={}& \int\frac{\mathrm{d}^2\mathbf{k}_{\mathrm{T}}}{16\pi^3} [\psi_{\uparrow1}^{\uparrow*}(x,\mathbf{k}_{\mathrm{T}})\psi_{\downarrow1}^{\downarrow}(x,\mathbf{k}_{\mathrm{T}}) + \psi_{\uparrow0}^{\uparrow*}(x,\mathbf{k}_{\mathrm{T}})\psi_{\downarrow0}^{\downarrow}(x,\mathbf{k}_{\mathrm{T}}) + \psi_{\uparrow-1}^{\uparrow*}(x,\mathbf{k}_{\mathrm{T}})\psi_{\downarrow-1}^{\downarrow}(x,\mathbf{k}_{\mathrm{T}}) \\ &\quad+ \psi_{\downarrow1}^{\uparrow*}(x,\mathbf{k}_{\mathrm{T}})\psi_{\uparrow1}^{\downarrow}(x,\mathbf{k}_{\mathrm{T}}) + \psi_{\downarrow0}^{\uparrow*}(x,\mathbf{k}_{\mathrm{T}})\psi_{\uparrow0}^{\downarrow}(x,\mathbf{k}_{\mathrm{T}}) + \psi_{\downarrow-1}^{\uparrow*}(x,\mathbf{k}_{\mathrm{T}})\psi_{\uparrow-1}^{\downarrow}(x,\mathbf{k}_{\mathrm{T}})]. \tag*{(40)} \end{aligned}

The required LFWFs are computed by using

ψλqλ(x,k⊥)=k+(P−k)+1k2−mq2uˉ(k,λq)YSU(P,λ),(41)\psi_{\lambda_q}^{\lambda}(x,\mathbf{k}_{\perp}) = \sqrt{\frac{k^+}{(P-k)^+}}\frac{1}{k^2-m_q^2}\bar{u}(k,\lambda_q)\mathcal{Y}_S U(P,\lambda), \tag*{(41)}
ψλqλVλ(x,k⊥)=k+(P−k)+1k2−mq2uˉ(k,λq)ϵμ∗(P−k,λV)⋅YVμU(P,λ),(42)\psi_{\lambda_q\lambda_V}^{\lambda}(x,\mathbf{k}_{\perp}) = \sqrt{\frac{k^+}{(P-k)^+}}\frac{1}{k^2-m_q^2}\bar{u}(k,\lambda_q)\epsilon_{\mu}^{*}(P-k,\lambda_V)\cdot\mathcal{Y}_V^{\mu}U(P,\lambda), \tag*{(42)}

with scalar and axial-vector vertex YS=igS(k2)1\mathcal{Y}_S = ig_S(k^2)\mathbf{1} and YVμ=igV(k2)γμγ5/2\mathcal{Y}_V^{\mu} = ig_V(k^2)\gamma^{\mu}\gamma_5/\sqrt{2}, sequentially [34]. The dipolar form factor for spectator diquarks has a form

gqD(k2)=gqDk2−mq2∣k2−ΛqD2∣2.(43)g_{qD}(k^2) = g_{qD}\frac{k^2-m_q^2}{|k^2-\Lambda_{qD}^2|^2}. \tag*{(43)}

Here, the ΛqD\Lambda_{qD} and gqDg_{qD} are the cutoff parameter and coupling constant, respectively. The momentum space wave function in the spectator diquark model is defined as

ϕqD=−gqD1−xx(1−x)2(k⊥2+LqD2)2,(44)\phi_{qD} = -\frac{g_{qD}}{\sqrt{1-x}}\frac{x(1-x)^2}{(\mathbf{k}_{\perp}^2+L_{qD}^2)^2}, \tag*{(44)}

with LqD2=xMD2+(1−x)ΛqD2−x(1−x)Mp2L_{qD}^{2}=xM_{D}^{2}+(1-x)\Lambda_{qD}^{2}-x(1-x)M_{p}^{2}, where MDM_{D} is the mass of the spectator diquark, MSM_{S} for the scalar and MVM_{V} for the axial-vector diquark. Here, ΛqD\Lambda_{qD} is an appropriate cutoff to avoid the singularity and satisfy the relation MD>Mp−ΛqDM_{D}>M_{p}-\Lambda_{qD}. The explicit expressions of the LFWFs are expressed in Table IV. The vacuum numerical values of proton and quark mass are considered as Mp=0.939M_{p}=0.939 and mq=0.313m_{q}=0.313 in the units of GeV. The diquark masses are those of Eq. (28) in Section II, MS=0.626M_{S}=0.626 GeV and MV=0.814M_{V}=0.814 GeV in vacuum. The cutoff parameters ΛuS(ud)=0.5\Lambda_{uS(ud)}=0.5 GeV, ΛuV(ud)=0.6\Lambda_{uV(ud)}=0.6 GeV, and ΛdV(uu)=0.4\Lambda_{dV(uu)}=0.4 GeV are chosen in such a way that they give the value of axial-charge of proton as close as to its experimental value, i.e., 1.27 [50, 51]. The original couplings gDg_{D} are rescaled by normalization constants NqDN_{qD} [34] and the f1qf_{1}^{q} TMD is normalized by

DDλq\lambda_qλD\lambda_Dλ=⇑\lambda=\Uparrowλ=⇓\lambda=\Downarrow
SS+12+\frac{1}{2}−-mq+xMpxϕqS\frac{m_q+xM_p}{x}\phi_{qS}kx−ikyxϕqS\frac{k_x-i k_y}{x}\phi_{qS}
SS−12-\frac{1}{2}−-−kx+ikyxϕqS-\frac{k_x+i k_y}{x}\phi_{qS}mq+xMpxϕqS\frac{m_q+xM_p}{x}\phi_{qS}
VV+12+\frac{1}{2}+1+1kx−ikyx(1−x)ϕqV\frac{k_x-i k_y}{x(1-x)}\phi_{qV}00
VV−12-\frac{1}{2}+1+1mq+xMpxϕqV\frac{m_q+xM_p}{x}\phi_{qV}−kx−iky1−xϕqV-\frac{k_x-i k_y}{1-x}\phi_{qV}
VV+12+\frac{1}{2}00k⊥2−xMV2−mqMp(1−x)22x(1−x)MVϕqV\frac{\mathbf{k}_{\perp}^{2}-xM_V^{2}-m_qM_p(1-x)^{2}}{\sqrt{2x(1-x)}M_V}\phi_{qV}(mq+Mp)(kx−iky)2xMVϕqV\frac{(m_q+M_p)(k_x-i k_y)}{\sqrt{2x}M_V}\phi_{qV}
VV−12-\frac{1}{2}00(mq+Mp)(kx+iky)2xMVϕqV\frac{(m_q+M_p)(k_x+i k_y)}{\sqrt{2x}M_V}\phi_{qV}−k⊥2−xMV2−mqMp(1−x)22x(1−x)MVϕqV-\frac{\mathbf{k}_{\perp}^{2}-xM_V^{2}-m_qM_p(1-x)^{2}}{\sqrt{2x(1-x)}M_V}\phi_{qV}
VV+12+\frac{1}{2}−1-1−kx+iky1−xϕqV-\frac{k_x+i k_y}{1-x}\phi_{qV}−mq+xMp1−xϕqV-\frac{m_q+xM_p}{1-x}\phi_{qV}
VV−12-\frac{1}{2}−1-100kx+ikyx(1−x)ϕqV\frac{k_x+i k_y}{x(1-x)}\phi_{qV}

Table IV. The light-front wave functions of Ref. [34] for a proton carrying helicity λ\lambda, comprised of an active quark and a spectator diquark with helicities λq\lambda_q and λD\lambda_D, respectively, used in the present calculations with ϕqD\phi_{qD} as momentum space wave functions. MpM_p, mqm_q, and MDM_D are the masses of the proton, active quark, and spectator diquark, which take their in-medium values from § 0.2.

∫dx∫d2k⊥f1normq(x,k⊥)=1.(45)\int dx \int d^{2}\boldsymbol{k}_{\perp} f_{1\mathrm{norm}}^{q}(x,\boldsymbol{k}_{\perp})=1. \tag*{(45)}

In the current section, the formalism and LFWFs discussed above are formulated under vacuum conditions at model scale Q02=0.2Q_{0}^{2}=0.2 GeV2^{2}. In order to incorporate medium effects, the corresponding vacuum masses of the quarks, diquarks, and proton are replaced by their baryon density, temperature, and isospin asymmetry dependent in-medium values, along with satisfying the normalization condition at each fixed value of {ρB,T,η}\{\rho_{B},T,\eta\}. The cutoffs ΛqD\Lambda_{qD} keep their vacuum values.

Baryon density effects on parton distribution functions

A set of PDFs is presented as a function of longitudinal momentum fraction xx for discrete values of baryon density ρB\rho_{B} in Figure 2. The unpolarized PDFs for f1uf_{1}^{u} and f1df_{1}^{d} are shown in Figure 2(a) and Figure 2(d), respectively. With increasing baryon density, the distributions exhibit a suppression of their peak amplitudes, accompanied by a broadening of the xx-dependence and a shift of the peak toward larger xx. This behavior reflects the medium-induced redistribution of the longitudinal momentum fraction of quarks, while the valence-quark number remains conserved through the normalization of the corresponding first moment at each baryon density. The peak amplitude of u(d)u(d) quark decreases by 6.6% (12.6%), while the peak position shifts toward higher xx by 27.9% (26%). The second Mellin moment of unpolarized PDF is tabulated in the Table V for different values of baryon density. For the vacuum case, the sum of the second moment of f1uf_{1}^{u} and f1df_{1}^{d} gives 0.926, which is not exactly unity because in our spectator model, it is not possible to fulfill at the same time the momentum sum rule and the quark number sum rule [34]. The same limitation persists in the in-medium extension of the model. The results of the in-medium second Mellin moment show a gradual increase in the Mellin moment with increasing baryon density. This behavior can be attributed to the rapid decrease in the quark, diquark, and proton masses with increasing baryon density, as our calculation incorporates only the corresponding effective masses. A similar trend has been reported in Ref. [14], where the use of effective masses alone leads to a shift of the unpolarized PDF peak toward higher xx value. Our results are therefore consistent with the scalar-field only results of Ref. [14], where the distribution becomes stiffer. In that work, the quenching in the valence region is also reported to arise from the vector field, which is not included here.

Figure 2. The leading-twist proton parton distribution functions as a function of the longitudinal momentum fraction xx for different baryon densities ρB\rho_B at zero temperature and isospin asymmetry. The top (bottom) row corresponds to the uu (dd) quark distributions, while the first, second, and third columns represent the unpolarized distribution xf1(x)xf_1(x), helicity distribution xg1(x)xg_1(x), and transversity distribution xh1(x)xh_1(x), respectively.

baryon density⟨x⟩u\langle x\rangle^u⟨x⟩d\langle x\rangle^d
T=0 GeVT=0\ \mathrm{GeV}T=0.1 GeVT=0.1\ \mathrm{GeV}T=0 GeVT=0\ \mathrm{GeV}T=0.1 GeVT=0.1\ \mathrm{GeV}
η=0\eta=0η=0.5\eta=0.5η=0\eta=0η=0.5\eta=0.5η=0\eta=0η=0.5\eta=0.5η=0\eta=0η=0.5\eta=0.5
00.6880.6880.6880.6880.2380.2380.2380.238
0.25ρ00.25\rho_00.7200.7200.7150.7150.2500.2490.2480.247
0.5ρ00.5\rho_00.7500.7500.7410.7420.2620.2590.2580.256
0.75ρ00.75\rho_00.7800.7790.7660.7670.2730.2700.2680.265
ρ0\rho_00.8060.8040.7890.7900.2840.2790.2770.274

Table V. Second Mellin moment of unpolarized parton distribution function ⟨x⟩q\langle x\rangle^q for different baryon density ρB\rho_B of the hot and isospin-asymmetric nuclear matter at discrete values of temperature TT (GeV) and isospin-asymmetry η\eta for the uu and dd quark flavors of the proton.

The results of helicity distribution for uu and dd quarks are shown in the respective Figure 2(b) and Figure 2(e). The uu quark distribution shows a gradual suppression of its peak, from 1.43 in vacuum to 1.30 at saturation density, with a shift of the peak position from x=0.42x=0.42 to x=0.56x=0.56 (see Figure 2(b)). In the vacuum, the dd quark distribution becomes positive as x→1x\to1. This behavior is consistent with the increasing contribution of the nonzero orbital angular momentum components of the LCWFs at large xx [38]. With increasing baryon density, density, the positive large-xx component of the dd quark helicity distribution becomes progressively enhanced, while its negative contribution at small xx is suppressed. Consequently, the zero-cross over point shifts toward smaller xx. This behavior can be associated with the medium-induced modification of the LFWFs, where the density-dependent quark, diquark, and proton masses alter the relative contributions of the different orbital angular momentum components to the helicity distribution. First Mellin moment of the helicity distribution, i.e., axial charge, is tabulated in the Table VI, corresponding to both quarks for discrete values of the baryon density. The magnitude of the axial charge for both Δu\Delta u and Δd\Delta d is found to decrease with an increase in the baryon density, which is consistent qualitatively with the result of Ref. [5]. For the case of proton, the axial charge gA=Δu−Δdg_A = \Delta u - \Delta d drops from 1.28 value in vacuum to 0.844 at saturation density. In particular, the reduction of the axial charge suggests that the quark helicity contribution to the proton spin becomes smaller in the medium.

baryon densityΔu\Delta uΔd\Delta d
T=0 GeVT = 0\ \mathrm{GeV}T=0.1 GeVT = 0.1\ \mathrm{GeV}T=0 GeVT = 0\ \mathrm{GeV}T=0.1 GeVT = 0.1\ \mathrm{GeV}
η=0\eta= 0η=0.5\eta= 0.5η=0\eta= 0η=0.5\eta= 0.5η=0\eta= 0η=0.5\eta= 0.5η=0\eta= 0η=0.5\eta= 0.5
001.0951.0951.0951.0951.0951.0951.0951.095−0.183-0.183−0.183-0.183−0.183-0.183−0.183-0.183
0.25ρ00.25\rho_01.0501.0501.0521.0521.0561.0561.0581.058−0.126-0.126−0.127-0.127−0.136-0.136−0.135-0.135
0.5ρ00.5\rho_01.0061.0061.0121.0121.0191.0191.0231.023−0.062-0.062−0.062-0.062−0.083-0.083−0.081-0.081
0.75ρ00.75\rho_00.9640.9640.9740.9740.9840.9840.9900.9900.0080.0080.0060.006−0.026-0.026−0.024-0.024
ρ0\rho_00.9230.9230.9380.9380.9500.9500.9590.9590.0790.0790.0740.0740.0320.0320.0340.034

Table VI. Valence quark contributions of the axial charge Δq\Delta q for different baryon density ρB\rho_B of the hot and isospin-asymmetric nuclear matter at discrete values of temperature TT (GeV) and isospin-asymmetry η\eta for the uu and dd quark flavors of the proton.

The results of transversity distribution for uu and dd quark as a function of longitudinal momentum fraction xx for discrete values of the baryon density ρB\rho_B are presented in Figure 2(c) and Figure 2(f), respectively. With increasing baryon density from ρB=0\rho_B = 0 to the saturation density, the qualitative behavior of both distributions remains unchanged. The uu quark peak decreases while the dd quark peak deepens, and both move to higher xx. Quantitatively, the uu quark peak decreases from 1.43 at x=0.42x = 0.42 to 1.18 at x=0.54x = 0.54 when ρB\rho_B increases from zero to saturation density. For the dd quark distribution, the corresponding negative peak value changes from −0.20-0.20 at x=0.32x = 0.32 to −0.26-0.26 at x=0.45x = 0.45. The first Mellin moments, corresponding to the tensor charges, exhibit a decrease in the uu quark contribution, while the dd quark contribution becomes increasingly negative with increasing baryon density. The corresponding values of the tensor charge δq\delta q at discrete values of the baryon density are listed in the Table VII. In a nutshell, the decrease in the tensor charge reflects a corresponding modification of the transverse spin structure.

baryon densityδu\delta uδd\delta d
T=0 GeVT = 0\ \mathrm{GeV}T=0.1 GeVT = 0.1\ \mathrm{GeV}T=0 GeVT = 0\ \mathrm{GeV}T=0.1 GeVT = 0.1\ \mathrm{GeV}
η=0\eta= 0η=0.5\eta= 0.5η=0\eta= 0η=0.5\eta= 0.5η=0\eta= 0η=0.5\eta= 0.5η=0\eta= 0η=0.5\eta= 0.5
01.1131.1131.1131.113−0.287−0.287−0.287−0.287
0.25ρ00.25\rho_{0}1.0511.0531.0601.061−0.296−0.295−0.295−0.294
0.5ρ00.5\rho_{0}0.9880.9931.0071.010−0.312−0.309−0.306−0.305
0.75ρ00.75\rho_{0}0.9260.9350.9550.960−0.332−0.328−0.322−0.319
ρ0\rho_{0}0.8660.8810.9060.913−0.356−0.349−0.340−0.336

Table VII. Valence quark contributions of the tensor charge δq\delta q for different baryon density ρB\rho_{B} of the hot and isospin-asymmetric nuclear matter at discrete values of temperature TT (GeV) and isospin-asymmetry η\eta for the uu and dd quark flavors of the proton.

Temperature effects on parton distribution functions

To examine the interplay between baryon density and temperature TT (GeV), we present the corresponding quark distributions in Figure 3 as a function of longitudinal momentum fraction xx in symmetric nuclear matter for two different values of baryon density. The PDFs are essentially insensitive to temperature at zero baryon density, whereas finite density effects become increasingly temperature dependent at saturation density. Increasing TT (GeV) from 0 to 0.1 partly counteracts the reduction of the effective masses induced by finite baryon density, leading to a slightly enhanced peak and a displacement of the peak toward lower xx. This behavior is consistent for xf1uxf_{1}^{u}, xf1dxf_{1}^{d}, xg1uxg_{1}^{u} and xh1uxh_{1}^{u}. Quantitatively, Figure 3(a) and Figure 3(d) show that the amplitude of the distributions xf1uxf_{1}^{u} and xf1dxf_{1}^{d} shift from 1.892 at x=0.53x = 0.53 to 1.897 at x=0.51x = 0.51 and 0.581 at x=0.35x = 0.35 to 0.589 at x=0.34x = 0.34, respectively. Similarly, the peak of xg1uxg_{1}^{u} distribution increases from 1.304 at x=0.56x = 0.56 to 1.308 at x=0.54x = 0.54 for the given values of T=0T = 0 and 0.1 GeV, respectively, as shown in Figure 3(b). For the case of transversity distribution (see Figure 3(c)), the peak value shifts from 1.176 at x=0.54x = 0.54 to 1.201 at x=0.52x = 0.52 when the temperature rises from T=0T = 0 to 0.1 GeV. In contrast, the xg1dxg_{1}^{d} distribution exhibits a nonuniform response to increasing temperature at the saturation density (see Figure 3(e)). While the negative contribution in the low-xx region increases with a shift of 0.019 value and the zero-cross over point moves toward higher xx, the magnitude of the distribution decreases in the high-xx region from 0.145 to 0.124. Thus, for the case of xg1dxg_{1}^{d}, the combined effects of baryon density and temperature modify not only the magnitude and peak position but also the shape of the distribution. Figure 3(f) shows the case of xh1dxh_{1}^{d} in which the peak magnitude of the distribution is found to decrease from 0.257 to 0.243 as the temperature increases to T=0.1T = 0.1 GeV from zero. The effects observed in the PDFs are transported to their corresponding momentum fractions carried by quarks, axial and tensor charges, as summarized in Table V, Table VI, and Table VII, respectively.

Figure 3. The leading-twist proton parton distribution functions as a function of the longitudinal momentum fraction xx for fixed values of temperature T=0T = 0 and 0.10.1 (GeV) at baryon densities ρB=0\rho_B = 0 and ρB=ρ0\rho_B = \rho_0 in isospin symmetric nuclear matter, i.e., η=0\eta= 0. The top (bottom) row corresponds to the uu (dd) quark distribution, while the first, second, and third columns represent the unpolarized distribution xf1(x)xf_1(x), helicity distribution xg1(x)xg_1(x), and transversity distribution xh1(x)xh_1(x), respectively.

Isospin asymmetry effects on parton distribution functions

To assess the impact of baryon density and isospin asymmetry η\eta on the quark distributions, we present the difference of PDFs between pure neutron matter (η=0.5\eta= 0.5) and symmetric nuclear matter (η=0\eta= 0) in Figure 4 as a function of longitudinal momentum fraction xx at zero temperature and saturation density. The difference of unpolarized PDFs for uu quark exhibits a characteristic sign change, being negative at small xx, positive at intermediate xx, and again negative at large xx. The zero-cross over lies at x=0.19x = 0.19 and 0.59, which is shown in Figure 4(a). For the case of dd quark, the distribution shows a positive change in the range x<0.4x < 0.4 and then flips to negative. This indicates that η\eta enhances the intermediate-xx contribution while suppressing the small- and large-xx contribution for the case of unpolarized PDF. Unlike the unpolarized distributions, the helicity distributions exhibit a distinct flavor dependence as the uu quark contribution undergoes a reversal of sign at x=0.62x = 0.62, whereas the dd quark contribution remains negative over the whole range (see Figure 4(b)). The transversity distributions presented in Figure 4(c) display another distinct flavor dependence, with both uu and dd quark contributions exhibiting sign changes at x=0.61x = 0.61 and 0.420.42, respectively, but in opposite directions. In all three cases, the uu quark distribution shows a substantially stronger response to isospin asymmetry than the dd quark distribution. The stronger response of uu quark follows the mass shifts of § 0.2. At ρ0\rho_0 and T=0T = 0 GeV, mu∗m_u^* rises by 2.69%, while the md∗m_d^* falls by 0.17% (Table III). The values of momentum fractions carried by quarks, axial and tensor charge as a function of η\eta are presented in Table V, VI and VII, respectively, for zero and finite T=0.1T = 0.1 GeV. The comparison indicates that the modifications induced by isospin asymmetry are comparatively smaller than those arising from variations in temperature TT (GeV) and baryon density ρB\rho_B.

Figure 4. Difference between the (a) unpolarized, (b) helicity, and (c) transversity quark distributions in asymmetric (η=0.5\eta= 0.5) and symmetric (η=0\eta= 0) nuclear matter for the proton at saturation density ρB=ρ0\rho_B = \rho_0 and zero temperature.

Parton distribution functions in nuclei

In this section, we compute the parton distributions of proton bound in various nuclei, including 40Ca{}^{40}\mathrm{Ca}, 56Fe{}^{56}\mathrm{Fe}, 197Au{}^{197}\mathrm{Au}, and 208Pb{}^{208}\mathrm{Pb}. To approximate parton distributions in these nuclei, we employ a Woods-Saxon-type density distribution of nucleons, which is expressed as follows [7],

ρi(r)=ρi,01+exp⁡[r−Riai](46)\rho_i(r)=\frac{\rho_{i,0}}{1+\exp\left[\frac{r-R_i}{a_i}\right]} \tag*{(46)}

where ρi,0\rho_{i,0} (i=p,ni=p,n) is not a free parameter but is instead determined by the normalization condition, 4π∫0∞ρi(r)r2dr=Z4\pi\int_0^\infty\rho_i(r)r^2\mathrm{d}r=Z for proton and NN for neutron. The surface behavior of the density profile is governed by two factors: the diffuseness parameter aia_i and the half-density radius RiR_i, both in fm. For protons and neutrons, we take these parameters from the fits of Ref. [7] to Dirac-Hartree-Bogoliubov densities, as in Ref. [8].

Rp=1.81Z1/3−1.12,Rn=1.49N1/3−0.79,(47)R_p=1.81Z^{1/3}-1.12,\qquad R_n=1.49N^{1/3}-0.79, \tag*{(47)}

and

ap=0.47−0.00083Z,an=0.47+0.00046N.(48)a_p=0.47-0.00083Z,\qquad a_n=0.47+0.00046N. \tag*{(48)}

In order to obtain a medium impact on parton distribution within finite nuclei, we utilize the CQMF model to solve the coupled equations of motion governing the mesonic fields, which are coupled to the baryonic density of nucleons. The total baryonic density of the nucleus ρB(r)\rho_B(r) is then given by the sum over proton and neutron contributions, ρB(r)=∑i=p,nρi(r)\rho_B(r)=\sum_{i=p,n}\rho_i(r), where the nuclei are treated in the local density approximation. From these densities, we obtain the in-medium masses of quarks, diquarks, and protons confined within various nuclei, specifically 40Ca{}^{40}\mathrm{Ca}, 56Fe{}^{56}\mathrm{Fe}, 197Au{}^{197}\mathrm{Au}, and 208Pb{}^{208}\mathrm{Pb}. We now plot the total baryonic density of the nuclei, ρB(r)\rho_B(r), varying with the radial distance rr from the center of the nuclei 40Ca{}^{40}\mathrm{Ca}, 56Fe{}^{56}\mathrm{Fe}, 197Au{}^{197}\mathrm{Au}, and 208Pb{}^{208}\mathrm{Pb} in Figure 5. The total baryonic density decreases with increasing distance from the center of the nucleus. Specifically, the values of ρB\rho_B in the center of the nuclei

Figure 5. The radial profile of total baryonic density ρB(r)\rho_B(r) in fm−3\mathrm{fm}^{-3} for various nuclei, including 40Ca{}^{40}\mathrm{Ca}, 56Fe{}^{56}\mathrm{Fe}, 197Au{}^{197}\mathrm{Au}, and 208Pb{}^{208}\mathrm{Pb}, given by the Woods-Saxon distributions of the proton and neutron densities.

(r=0)(r = 0) are found to be 0.191, 0.184, 0.158, and 0.157 fm−3\mathrm{fm}^{-3} for 40Ca{}^{40}\mathrm{Ca}, 56Fe{}^{56}\mathrm{Fe}, 197Au{}^{197}\mathrm{Au}, and 208Pb{}^{208}\mathrm{Pb}, respectively. For these values of ρB\rho_{B}, the leading-twist proton distributions with respect to the longitudinal momentum fraction xx for 40Ca{}^{40}\mathrm{Ca}, 56Fe{}^{56}\mathrm{Fe}, 197Au{}^{197}\mathrm{Au}, and 208Pb{}^{208}\mathrm{Pb} nuclei are plotted in Figure 6. The results of unpolarized PDFs for uu and dd quarks for different nuclei are presented in Figure 6(a) and Figure 6(d). For light nuclei, the peak position of f1uf_{1}^{u} is centered around x≈0.55x \approx0.55, while the peak position of f1df_{1}^{d} lies near x≈0.35x \approx0.35. Going from light to heavy nuclei, both peaks shift toward lower xx, reflecting how these PDFs respond to the baryonic density of the nuclear medium. In Figure 6(b) and Figure 6(e), we plot the results of the helicity distribution of the light quarks, g1ug_{1}^{u} and g1dg_{1}^{d}, as a function of xx for all nuclei considered in this study. As the central density falls, the valence uu quark distribution develops a modest rise near its maximum. In contrast, the dd quark distribution is progressively suppressed for x>0.27x > 0.27 for heavy nuclei. Below this value of xx, a negative modification appears only for the heavy nuclei as illustrated in Figure 6(e). Similarly, the results of the transversity distribution h1(x)h_{1}(x) in Figure 6(c) and Figure 6(f) are displayed. We observe a positive contribution of h1u(x)h_{1}^{u}(x) till x<0.85x < 0.85, but its effect turns negative beyond this point. On the other hand, h1d(x)h_{1}^{d}(x) contributes negatively across the considered xx range. Further, we present these leading-twist proton parton distributions for the 208Pb{}^{208}\mathrm{Pb} nucleus at various values of radial distance r=0r = 0, 3, 6, 9 fm in Figure 7. The distributions at r=0r = 0 and 3 fm coincide, since the density stays close to its central value there (Figure 5). At r=6r = 6 fm, in the nuclear surface, the density is lower and the distributions move part of the way back toward their vacuum forms, while at r=9r = 9 fm, where the density has almost vanished, they coincide with the vacuum distributions. At the center of each nucleus, the axial charge gA=Δu−Δdg_A = \Delta u - \Delta d of the bound proton is 0.699 for 40Ca{}^{40}\mathrm{Ca}, 0.741 for 56Fe{}^{56}\mathrm{Fe}, 0.815 for 197Au{}^{197}\mathrm{Au}, and 0.817 for 208Pb{}^{208}\mathrm{Pb}, against 1.28 in vacuum, so it increases as the central density decreases. In Ref. [5], the axial charge of the bound proton tends to decrease with mass number, from 1.267 for the free proton to 1.19 in 7Li{}^{7}\mathrm{Li}, 1.16 in 11B{}^{11}\mathrm{B}, and 1.15 in 15N{}^{15}\mathrm{N} and 27Al{}^{27}\mathrm{Al}, and it reaches 1.05 in nuclear matter. The calculation of Ref. [5] describes the nucleus in a relativistic shell model with Woods-Saxon scalar and vector potentials of the same depth for every nucleus. The spin-dependent distributions there are carried mainly by the valence proton or proton hole, which feels the mean fields averaged over its single-particle state. The values obtained here are instead taken at the center of each nucleus, where the central density falls from 40Ca{}^{40}\mathrm{Ca} to 208Pb{}^{208}\mathrm{Pb}, so the opposite dependence on mass number may reflect the different quantity computed. In both calculations, the axial

Figure 6. The leading-twist proton parton distribution functions are depicted as a function of the longitudinal momentum fraction xx for various nuclei: 40Ca{}^{40}\mathrm{Ca}, 56Fe{}^{56}\mathrm{Fe}, 197Au{}^{197}\mathrm{Au}, and 208Pb{}^{208}\mathrm{Pb}. The top row illustrates the uu quark distributions, while the bottom row shows the dd quark distributions. The columns are organized to display the unpolarized distribution xf1(x)xf_1(x) in subfigures (a) and (d), the helicity distribution xg1(x)xg_1(x) in subfigures (b) and (e), and the transversity distribution xh1(x)xh_1(x) in subfigures (c) and (f).

Figure 7. The leading-twist proton parton distribution functions are depicted as a function of the longitudinal momentum fraction xx for 208Pb^{208}\mathrm{Pb} nucleus at various values of radial distance rr. The top row illustrates the uu quark distributions, while the bottom row shows the dd quark distributions. The columns are organized to display the unpolarized distribution xf1(x)xf_1(x) in subfigures (a) and (d), the helicity distribution xg1(x)xg_1(x) in subfigures (b) and (e), and the transversity distribution xh1(x)xh_1(x) in subfigures (c) and (f).

charge is lower in nuclear matter than in the free proton.

Summary

The impact of hot and isospin-asymmetric nuclear matter and finite nuclei on the proton parton distribution functions (PDFs) is investigated within the diquark spectator model. The medium effects are incorporated through the effective masses of the quarks, diquarks, and proton, which are calculated within the chiral SU(3) quark mean-field model. The resulting modifications of these effective masses arise from their interactions with the scalar fields and consequently influence the light-front wave functions and the partonic structure of the proton. The effective masses decrease with increasing baryon density. At temperature T=0.1T = 0.1 GeV, they lie above their T=0T = 0 values, while in pure neutron matter, mu∗m_u^* lies above and md∗m_d^* below its value in symmetric nuclear matter, by at most 2.7%.

The in-medium unpolarized f1qf_1^q, helicity g1qg_1^q, and transversity h1qh_1^q distributions for both quark flavors are calculated as functions of longitudinal momentum fraction xx at different baryon densities ρB\rho_B, temperatures TT (GeV), and isospin asymmetries η\eta. The impact of the medium conditions on the parton distributions follows the order: baryon density >> temperature >> isospin asymmetry. As a function of baryon density, the distributions of xf1uxf_1^u, xf1dxf_1^d, xg1uxg_1^u, and xh1uxh_1^u exhibit a decrease in their peak values, accompanied by a shift of the peaks toward higher xx values. In contrast, the distribution xg1dxg_1^d exhibits a decrease in the magnitude of its negative contribution in the low-xx region, accompanied by a shift of the crossover point toward lower xx. Conversely, the positive contribution increases in magnitude in the high-xx domain when the medium becomes denser. For the case of xh1dxh_1^d, the distribution exhibits an increase in the magnitude of its negative contribution in the whole span of xx region when the baryon density increases from zero to its saturation value. Corresponding to the unpolarized, helicity, and transversity distributions, their Mellin moments are also computed. Second Mellin moment of the f1qf_1^q increases with increasing the density of the matter, whereas the axial charge gAg_A and the uu quark tensor charge decrease and the dd quark tensor charge grows in magnitude. The axial charge of the proton falls from 1.28 in vacuum to 0.844 at saturation density. This agrees in sign with the NJL quark-diquark calculation [5], where the axial charge of the bound proton falls from 1.267 for the free proton to 1.05 in nuclear matter, while the reduction obtained here is larger. A comparison of the spin-dependent and spin-independent distributions reveals more pronounced changes in the spin-dependent distributions.

At zero baryon density, the distributions remain essentially unchanged with temperature and isospin asymmetry. However, at saturation density, the distributions vary with TT and η\eta. Moving from T=0T = 0 to 0.1 GeV, the distributions xf1uxf_1^u, xf1dxf_1^d, xg1uxg_1^u, and xh1uxh_1^u show a slight increase in their peak heights, and their peaks shift toward smaller xx. In contrast, the negative peak of xh1dxh_1^d becomes shallower, while also moving to smaller xx. For xg1dxg_1^d, the negative contribution increases in the low-xx region, whereas the positive contribution decreases in the high-xx region. To study the impact of isospin asymmetry, the difference between asymmetric (η=0.5\eta= 0.5) and symmetric (η=0\eta= 0) nuclear matter over the distributions is presented. The uu quark distributions exhibit a characteristic sign change over the range of xx for all three cases: unpolarized, helicity, and transversity. In contrast, the dd quark helicity distribution remains negative, while the other dd quark distributions exhibit opposite signs in the low- and high-xx regions. In the corresponding Mellin moments, the effect of η\eta is small, while at saturation density, raising TT to 0.1 GeV increases the axial charge from 0.844 to 0.918.

In addition, we also examined parton distributions of a proton bound in various nuclei, including 40Ca^{40}\mathrm{Ca}, 56Fe^{56}\mathrm{Fe}, 197Au^{197}\mathrm{Au}, and 208Pb^{208}\mathrm{Pb} at zero temperature by employing Woods-Saxon-type density distributions. As the central density falls from 40Ca^{40}\mathrm{Ca} to 208Pb^{208}\mathrm{Pb}, the peaks of xf1qxf_{1}^{q} move to lower xx, xg1uxg_{1}^{u} rises slightly near its peak, the positive part of xg1dxg_{1}^{d} is suppressed, and xh1dxh_{1}^{d} stays negative for all xx. At the center of each nucleus, the axial charge of the proton rises from 0.699 in 40Ca^{40}\mathrm{Ca} to 0.817 in 208Pb^{208}\mathrm{Pb} as the central baryon density decreases, which suggests that the modification at the center follows the central density rather than the mass number. In Ref. [5], the axial charge of the bound proton instead decreases with mass number, and the spin-dependent distributions are carried mainly by the valence proton or proton hole, which feels the mean fields averaged over its single-particle state. The opposite trend may therefore reflect the different quantity computed, since the values here are taken at the center of the nucleus. In the present work, we studied the in-medium quark dynamics by incorporating effective masses through scalar-field interactions. Future extensions will include vector-field interactions to provide a more comprehensive description of the nuclear medium and examine their effects on the quark dynamics and the resulting parton distribution functions.

Acknowledgements

H.D. would like to thank the Science and Engineering Research Board, Anusandhan-National Research Foundation, Government of India under the scheme SERB-POWER Fellowship (Ref No. SPF/2023/000116) for financial support.

Derivatives of the effective nucleon mass

Here we derive Eq. (25), writing φ\varphi for either of the two fields σ\sigma and δ\delta in Eq. (24), so that both couplings take the form gφi=−∂Mi∗/∂φg_{\varphi}^{i}=-\partial M_{i}^{*}/\partial\varphi. The spin correlation energy EspiniE_{\mathrm{spin}}^{i} and the confinement strength kck_{c} are held fixed, so the fields reach Mi∗M_{i}^{*} only through the quark masses, along the chain φ→mq∗→(eq,⟨pq2⟩)→(Ei,⟨Pc.m.2⟩i)→Mi∗\varphi\to m_{q}^{*} \to(e_{q},\langle p_{q}^{2}\rangle) \to(E_{i},\langle P_{\mathrm{c.m.}}^{2}\rangle_{i}) \to M_{i}^{*}. By Eqs. (10) and (13), eqe_{q} and ⟨pq2⟩\langle p_{q}^{2}\rangle depend on mq∗m_{q}^{*} alone, so the entire dependence on φ\varphi enters through the derivatives of the quark masses. From Eq. (4) these are

∂mq∗∂σ=−gs2,∂mq∗∂δ=−gs2I3q,q=u,d.(49)\frac{\partial m_q^*}{\partial\sigma}=-\frac{g_s}{\sqrt{2}},\qquad\frac{\partial m_q^*}{\partial\delta}=-\frac{g_s}{\sqrt{2}}I_3^q,\qquad q=u,d. \tag*{(49)}

Since eq−mq∗>0e_q-m_q^*>0, Eq. (10) is equivalent to (eq−mq∗)2(eq+mq∗)=9kc/2(e_q-m_q^*)^2(e_q+m_q^*)=9k_c/2. Differentiating this form with respect to mq∗m_q^* and dividing by eq−mq∗e_q-m_q^* gives

2(eq+mq∗)(rq−1)+(eq−mq∗)(rq+1)=0,rq≡deqdmq∗,(50)2\left(e_q+m_q^*\right)\left(r_q-1\right)+\left(e_q-m_q^*\right)\left(r_q+1\right)=0,\qquad r_q\equiv\frac{d e_q}{d m_q^*}, \tag*{(50)}

so that

rq=eq+3mq∗3eq+mq∗.(51)r_q=\frac{e_q+3m_q^*}{3e_q+m_q^*}. \tag*{(51)}

The quark energy then responds to the fields as ∂eq/∂φ=rq∂mq∗/∂φ\partial e_q/\partial\varphi=r_q\partial m_q^*/\partial\varphi. The mean square momentum of Eq. (13) depends on mq∗m_q^* both directly and through eqe_q. Writing it as ⟨pq2⟩=Nq/Dq\langle p_q^2\rangle=N_q/D_q, with Nq=(11eq+mq∗)(eq2−mq∗2)N_q=(11e_q+m_q^*)(e_q^2-m_q^{*2}) and Dq=6(3eq+mq∗)D_q=6(3e_q+m_q^*), and using deq/dmq∗=rqd e_q/dm_q^*=r_q, we have

dNqdmq∗=(11rq+1)(eq2−mq∗2)+2(11eq+mq∗)(rqeq−mq∗),dDqdmq∗=6(3rq+1).(52)\frac{dN_q}{dm_q^*}=(11r_q+1)(e_q^2-m_q^{*2})+2(11e_q+m_q^*)(r_q e_q-m_q^*),\qquad\frac{dD_q}{dm_q^*}=6(3r_q+1). \tag*{(52)}

The quotient rule then gives

sq≡d⟨pq2⟩dmq∗=1Dq(dNqdmq∗−⟨pq2⟩dDqdmq∗),(53)s_q\equiv\frac{d\langle p_q^2\rangle}{dm_q^*}=\frac{1}{D_q}\left(\frac{dN_q}{dm_q^*}-\langle p_q^2\rangle\frac{dD_q}{dm_q^*}\right), \tag*{(53)}

and the mean square momentum responds to the fields as ∂⟨pq2⟩/∂φ=sq∂mq∗/∂φ\partial\langle p_q^2\rangle/\partial\varphi=s_q\partial m_q^*/\partial\varphi. Since EspiniE_{\mathrm{spin}}^i is constant, Eqs. (11) and (12) give

∂Ei∂φ=∑q=u,dnqirq∂mq∗∂φ,∂⟨Pc.m.2⟩i∂φ=∑q=u,dnqisq∂mq∗∂φ.(54)\frac{\partial E_i}{\partial\varphi}=\sum_{q=u,d}n_q^i r_q\frac{\partial m_q^*}{\partial\varphi},\qquad\frac{\partial\langle P_{\mathrm{c.m.}}^2\rangle_i}{\partial\varphi}=\sum_{q=u,d}n_q^i s_q\frac{\partial m_q^*}{\partial\varphi}. \tag*{(54)}

Differentiating Mi∗2=Ei2−⟨Pc.m.2⟩iM_i^{*2}=E_i^2-\langle P_{\mathrm{c.m.}}^2\rangle_i and inserting these sums gives

∂Mi∗∂φ=1Mi∗(Ei∂Ei∂φ−12∂⟨Pc.m.2⟩i∂φ)=1Mi∗∑q=u,dnqicqi∂mq∗∂φ.(55)\frac{\partial M_i^*}{\partial\varphi}=\frac{1}{M_i^*}\left(E_i\frac{\partial E_i}{\partial\varphi}-\frac{1}{2}\frac{\partial\langle P_{\mathrm{c.m.}}^2\rangle_i}{\partial\varphi}\right)=\frac{1}{M_i^*}\sum_{q=u,d}n_q^i c_q^i\frac{\partial m_q^*}{\partial\varphi}. \tag*{(55)}

with cqi=Eirq−sq/2c_q^i=E_i r_q-s_q/2, as in Eq. (25). The energy EiE_i in cqic_q^i is the full nucleon energy of Eq. (11), including EspiniE_{\mathrm{spin}}^i. Inserting the quark mass derivatives of Eq. (49) into Eq. (55) and changing the sign gives

gσi=−1Mi∗∑q=u,dnqicqi(−gs2)=gs2Mi∗∑q=u,dnqicqi,gδi=−1Mi∗∑q=u,dnqicqi(−gs2I3q)=gs2Mi∗∑q=u,dnqiI3qcqi.(56)\begin{aligned} g_\sigma^i&=-\frac{1}{M_i^*}\sum_{q=u,d}n_q^i c_q^i\left(-\frac{g_s}{\sqrt{2}}\right)=\frac{g_s}{\sqrt{2}M_i^*}\sum_{q=u,d}n_q^i c_q^i,\\ g_\delta^i&=-\frac{1}{M_i^*}\sum_{q=u,d}n_q^i c_q^i\left(-\frac{g_s}{\sqrt{2}}I_3^q\right)=\frac{g_s}{\sqrt{2}M_i^*}\sum_{q=u,d}n_q^i I_3^q c_q^i. \tag*{(56)} \end{aligned}

which is Eq. (25).

References

References

  1. [1]J. J. Aubert et al. [European Muon Collaboration], Phys. Lett. B 123, 275-278 (1983) doi:10.1016/0370-2693(83)90437-9DOI
  2. [2]A. Bodek et al., Phys. Rev. Lett. 50, 1431-1434 (1983) doi:10.1103/PhysRevLett.50.1431DOI
  3. [3]A. Bodek et al., Phys. Rev. Lett. 51, 534-537 (1983) doi:10.1103/PhysRevLett.51.534DOI
  4. [4]R. G. Arnold et al., Phys. Rev. Lett. 52, 727-730 (1984) doi:10.1103/PhysRevLett.52.727DOI
  5. [5]I. C. Cloet, W. Bentz and A. W. Thomas, Phys. Lett. B 642, 210-217 (2006) doi:10.1016/j.physletb.2006.08.076 [arXiv:nucl-th/0605061 [nucl-th]].DOI
  6. [6]I. Sick and D. Day, Phys. Lett. B 274, 16-20 (1992) doi:10.1016/0370-2693(92)90297-HDOI
  7. [7]L. C. Chamon, B. V. Carlson, L. R. Gasques, D. Pereira, C. De Conti, M. A. G. Alvarez, M. S. Hussein, M. A. Cândido Ribeiro, E. S. Rossi, Jr. and C. P. Silva, Phys. Rev. C 66, 014610 (2002) doi:10.1103/PhysRevC.66.014610DOI
  8. [8]W. Ma, Y. G. Ma, W. He and B. Zhou, Chaos 34, 093121 (2024) doi:10.1063/5.0213717DOI
  9. [9]R. L. Jaffe and X. D. Ji, Phys. Rev. Lett. 67, 552-555 (1991) doi:10.1103/PhysRevLett.67.552DOI
  10. [10]J. P. Ralston and D. E. Soper, Nucl. Phys. B 152, 109-124 (1979) doi:10.1016/0550-3213(79)90082-8DOI
  11. [11]M. Radici and A. Bacchetta, Phys. Rev. Lett. 120, 192001 (2018) doi:10.1103/PhysRevLett.120.192001 [arXiv:1802.05212 [hep-ph]]DOI
  12. [12]C. Cocuzza, A. Metz, D. Pitonyak, A. Prokudin, N. Sato and R. Seidl, Phys. Rev. Lett. 132, 091901 (2024) doi:10.1103/PhysRevLett.132.091901DOI
  13. [13]D. de Florian, R. Sassot, M. Stratmann and W. Vogelsang, Phys. Rev. Lett. 101, 072001 (2008) doi:10.1103/PhysRevLett.101.072001DOI
  14. [14]H. Mineo, W. Bentz, N. Ishii, A. W. Thomas and K. Yazaki, Nucl. Phys. A 735, 482-514 (2004) doi:10.1016/j.nuclphysa.2004.02.011 [arXiv:nucl-th/0312097 [nucl-th]].DOI
  15. [15]I. C. Cloet, W. Bentz and A. W. Thomas, Phys. Rev. Lett. 95, 052302 (2005) doi:10.1103/PhysRevLett.95.052302 [arXiv:nucl-th/0504019 [nucl-th]].DOI
  16. [16]S. Tronchin, H. H. Matevosyan and A. W. Thomas, Phys. Lett. B 783, 247-252 (2018) doi:10.1016/j.physletb.2018.06.065 [arXiv:1806.00481 [nucl-th]].DOI
  17. [17]J. R. Smith and G. A. Miller, Phys. Rev. Lett. 91, 212301 (2003) [erratum: Phys. Rev. Lett. 98, 099902 (2007)] doi:10.1103/PhysRevLett.91.212301 [arXiv:nucl-th/0308048 [nucl-th]].DOI
  18. [18]W. R. B. de Araújo, J. P. B. C. de Melo and K. Tsushima, Nucl. Phys. A 970, 325-352 (2018) doi:10.1016/j.nuclphysa.2017.12.005 [arXiv:1707.00168 [hep-ph]].DOI
  19. [19]S. Kaur, S. Xu, C. Mondal, X. Zhao and J. P. Vary, Phys. Rev. D 109, 014015 (2024) doi:10.1103/PhysRevD.109.014015DOI
  20. [20]S. Xu, C. Mondal, X. Zhao, Y. Li and J. P. Vary, Phys. Rev. D 108, 094002 (2023) doi:10.1103/PhysRevD.108.094002DOI
  21. [21]X. Wu, S. Nair, S. Kaur, C. Mondal, J. Lan, X. Zhao, J. P. B. C. de Melo and T. Frederico, [arXiv:2605.31030 [hep-ph]].arxiv.org/abs/2605.31030
  22. [22]I. C. Cloet, W. Bentz and A. W. Thomas, Phys. Rev. Lett. 102, 252301 (2009) doi:10.1103/PhysRevLett.102.252301DOI
  23. [23]J. P. B. C. de Melo, K. Tsushima and I. Ahmed, Phys. Lett. B 766, 125-131 (2017) doi:10.1016/j.physletb.2017.01.004 [arXiv:1608.03858 [hep-ph]].
  24. [24]G. H. S. O. Yabusaki, J. P. B. C. de Melo, K. Tsushima, T. Frederico and W. de Paula, Phys. Rev. D 109, 054024 (2024) doi:10.1103/PhysRevD.109.054024 [arXiv:2310.11262 [hep-ph]].DOI
  25. [25]A. J. Arifi, P. T. P. Hutauruk and K. Tsushima, Phys. Rev. D 107, 114010 (2023) doi:10.1103/PhysRevD.107.114010 [arXiv:2302.12382 [hep-ph]].DOI
  26. [26]A. Dwibedi, S. Puhan, S. Ghosh and H. Dahiya, PTEP 2026, 063B04 (2026) doi:10.1093/ptep/ptag089 [arXiv:2512.24921 [hep-ph]].DOI
  27. [27]P. T. P. Hutauruk, J. J. Cobos-Martínez, Y. Oh and K. Tsushima, Phys. Rev. D 100, 094011 (2019) doi:10.1103/PhysRevD.100.094011 [arXiv:1908.02406 [hep-ph]].DOI
  28. [28]P. T. P. Hutauruk and S. i. Nam, Phys. Rev. D 105, 034021 (2022) doi:10.1103/PhysRevD.105.034021 [arXiv:2112.05435 [hep-ph]].DOI
  29. [29]S. Puhan, N. Kaur, A. Kumar, S. Dutt and H. Dahiya, Phys. Rev. D 110, 054042 (2024) doi:10.1103/PhysRevD.110.054042 [arXiv:2408.07334 [nucl-th]].DOI
  30. [30]D. Singh, S. Puhan, N. Kaur, M. Kaur, A. Kumar, S. Dutt and H. Dahiya, Phys. Rev. D 111, 054001 (2025) doi:10.1103/PhysRevD.111.054001 [arXiv:2410.20181 [hep-ph]].DOI
  31. [31]P. Wang, Z. Y. Zhang and Y. W. Yu, Commun. Theor. Phys. 36, 71 (2001).
  32. [32]P. Papazoglou, D. Zschiesche, S. Schramm, J. Schaffner-Bielich, H. Stöcker and W. Greiner, Phys. Rev. C 59, 411-427 (1999) doi:10.1103/PhysRevC.59.411 [arXiv:nucl-th/9806087 [nucl-th]].DOI
  33. [33]H. Singh, A. Kumar and H. Dahiya, Eur. Phys. J. Plus 134, 128 (2019) doi:10.1140/epjp/i2019-12486-4 [arXiv:1710.08328 [nucl-th]].DOI
  34. [34]A. Bacchetta, F. Conti and M. Radici, Phys. Rev. D 78, 074010 (2008) doi:10.1103/PhysRevD.78.074010 [arXiv:0807.0323 [hep-ph]].DOI
  35. [35]N. Kaur and H. Dahiya, Phys. Rev. D 112, 074024 (2025) doi:10.1103/rt3y-2b8gDOI
  36. [36]H. Shen and H. Toki, Phys. Rev. C 61, 045205 (2000).
  37. [37]N. Barik, B. K. Dash and M. Das, Phys. Rev. D 31, 1652 (1985).
  38. [38]H. Avakian, S. J. Brodsky, A. Deur and F. Yuan, Phys. Rev. Lett. 99, 082001 (2007) doi:10.1103/PhysRevLett.99.082001 [arXiv:0705.1553 [hep-ph]].DOI
  39. [40]P. Wang, D. B. Leinweber, A. W. Thomas and A. G. Williams, Nucl. Phys. A 744, 273 (2004) doi:10.1016/j.nuclphysa.2004.08.015DOI
  40. [41]R. Kumar et al. (MUSES Collaboration), Living Rev. Relativ. 27, 3 (2024) doi:10.1007/s41114-024-00049-6DOI
  41. [42]J. T. Lenaghan, D. H. Rischke and J. Schaffner-Bielich, Phys. Rev. D 62, 085008 (2000) [arXiv:nucl-th/0004006].DOI
  42. [43]S. Sakai and D. Jido, Phys. Rev. C 88, 064906 (2013) [arXiv:1309.4845].DOI
  43. [44]M. Kaur, A. Abi Jebarson, D. Singh, N. Kaur, S. Dutt, H. Dahiya and A. Kumar, Eur. Phys. J. A 61, 251 (2025) doi:10.1140/epja/s10050-025-01737-9DOI
  44. [45]F. E. Close and A. W. Thomas, Phys. Lett. B 212, 227-230 (1988).
  45. [46]S. Meissner, A. Metz and M. Schlegel, JHEP 08, 056 (2009).
  46. [47]A. Bacchetta, M. Diehl, K. Goeke, A. Metz, P. J. Mulders and M. Schlegel, JHEP 02, 093 (2007).
  47. [48]R. Jakob, P. J. Mulders and J. Rodrigues, Nucl. Phys. A 626, 937-965 (1997).
  48. [49]D. B. Lichtenberg, L. J. Tassie and P. J. Keleman, Phys. Rev. 167, 1535-1542 (1968).
  49. [50]E. Leader, A. V. Sidorov and D. B. Stamenov, Phys. Rev. D 82, 114018 (2010).
  50. [51]B. Märkisch, H. Mest, H. Saul, X. Wang, H. Abele, D. Dubbers, M. Klopf, A. Petoukhov, C. Roick and T. Soldner, et al. Phys. Rev. Lett. 122, 242501 (2019).

Paper details

vlnc:W_y5ec943r
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 1

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