LFTC-25-10/104 In-medium mass shifts of $B_{c}^{(*)},B_{s}^{(*)}$ and $D_{s}^{(*)}$ mesons

Abstract

We present our predictions for the Lorentz scalar mass shifts of two-flavored heavy mesons, $B_{c}^{(*)},B_{s}^{(*)}$ and $D_{s}^{(*)}$ in symmetric nuclear matter. The in-medium mass shifts are estimated by evaluating the lowest order one-loop self-energies of the mesons based on a flavor-SU(5) effective Lagrangian approach. In-medium properties necessary for the estimates are calculated by the quark-meson coupling (QMC) model. The enhanced self-energies of the mesons in symmetric nuclear matter relative to those in free space, yield the negative mass shifts of these mesons.

1 Introduction

Studies of hadron interactions with nuclear medium, particularly for the mesons composed of two-flavored heavy quarks, can shed light on the roles of gluons in quantum chromodynamics (QCD). Since such mesons do not share light quarks with the nucleons (nuclear medium), the interactions with the nuclear medium are expected primarily by gluons at the lowest order. Here, our focus is on the mesons $B_{c}^{(*)},B_{s}^{(*)}$ and $D_{s}^{(*)}$, and their mass shifts in symmetric nuclear matter based on Ref. Zeminiani:2023gqc . (See, e.g., Refs. Tsushima:1998ru ; Krein:2010vp ; Krein:2017usp ; Cobos-Martinez:2020ynh ; Zeminiani:2024dyo ; Cobos-Martinez:2025iqg for the studies made on the mass shifts of quarkonia and heavy-meson-nucleus bound states.)

In Fig. 1, we show the self-energy graphs included for evaluating the mass-shifts of $B_{c}^{(*)},B_{s}^{(*)}$ and $D_{s}^{(*)}$ mesons. (Note that, for the self-energies of the vector mesons $B_{c}^{*},B_{s}^{*}$ and $D_{s}^{*}$, we include the ”lowest order graphs” excluding the two-vector-meson excited intermediate states, based on the discussions given in Ref. Zeminiani:2020aho .)

Figure 1: Meson self-energy diagrams for mesons $B_{c}$ [(a) and (a’)], $B_{s}$ [(b) and (b’)], $D_{s}$ [(c) and (c’)], $B_{c}^{*}$ [(d)], $B_{s}^{*}$ [(e)], and $D_{s}^{*}$ [(f)], included in the present study.

To evaluate the in-medium self-energy graphs depicted in Fig. 1, we use the in-medium masses of the intermediate state excited mesons, $B^{(*)},D^{(*)}$ and $K^{(*)}$, calculated by the quark-meson coupling (QMC) model. Thus, we will briefly explain next the QMC model Guichon:1987jp ; Saito:2005rv .

2 Quark-meson coupling (QMC) model

The Lorentz-scalar and Lorentz-vector mean field potentials in symmetric nuclear matter for the $B^{(*)},D^{(*)}$ and $K^{(*)}$ mesons enter to evaluate the self-energy graphs shown in Fig. 1, we use those calculated by the QMC model Guichon:1987jp ; Saito:2005rv . In the QMC model, the binding of nucleons in nuclei (nuclear medium) is achieved by the self-consistent exchange of the scalar-isoscalar-$\sigma$, vector-isoscalar-$\omega$ and vector-isovector-$\rho$ meson fields directly coupled to the light quarks $u$ and $d$, that are confined and relativistically moving inside the nucleon (described by the MIT bag model). The meson mean fields are generated by those confined light quarks in the nucleons, and the (light-quark)-(meson field) coupling constants in any hadrons should be the same as those in nucleon, after they are calibrated once properly. Then, one can naturally expect that the hadrons with light quarks, when they are immersed in nuclear medium, are expected to change their properties in the QMC model.

In the rest frame of nuclear matter, the Dirac equations for the quarks and antiquarks (no Coulomb force) in the QMC model, assuming SU(2) symmetry for the light quarks ($q=u,d$ and $m_{q}\equiv m_{u}=m_{d}$) are given by Tsushima:1997df ,

$$\hskip-34.44434pt\left[i\gamma\cdot\partial_{x}-\left(m_{q}-V^{q}_{\sigma}\right)\mp\gamma^{0}\left(V^{q}_{\omega}+\frac{1}{2}V^{q}_{\rho}\right)\right]\begin{pmatrix}\psi_{u}\left(x\right)\\ \psi_{\overline{u}}\left(x\right)\end{pmatrix}=0,$$ (1)

$$\hskip-34.44434pt\left[i\gamma\cdot\partial_{x}-\left(m_{q}-V^{q}_{\sigma}\right)\mp\gamma^{0}\left(V^{q}_{\omega}-\frac{1}{2}V^{q}_{\rho}\right)\right]\begin{pmatrix}\psi_{d}\left(x\right)\\ \psi_{\overline{d}}\left(x\right)\end{pmatrix}=0,$$ (2)

$$\hskip-34.44434pt\left[i\gamma\cdot\partial_{x}-m_{Q}\right]\psi_{Q,\overline{Q}}\left(x\right)=0,$$ (3)

where $Q=s,c$, or $b$ and the mean-field potentials for the light quark ($q$) in nuclear matter are defined by $V^{q}_{\sigma}\equiv g^{q}_{\sigma}\sigma$, $V^{q}_{\omega}\equiv g^{q}_{\omega}\omega=g^{q}_{\omega}\,\delta^{\mu,0}\omega^{\mu}$, $V^{q}_{\rho}\equiv g^{q}_{\rho}b=g^{q}_{\rho}\,\delta^{i,3}\delta^{\mu,0}\rho^{i,\mu}$, with $\sigma$, $\omega$ and $b$ being the meson mean fields, and $g^{q}_{\sigma}$, $g^{q}_{\omega}$ and $g^{q}_{\rho}$ the corresponding quark-meson coupling constants. Hereafter, we consider symmetric nuclear matter in Hartree approximation, and set $V^{q}_{\rho}=0$ in Eqs. (1) and (2).

The eigenenergies for the quarks and antiquarks in a hadron $h\,(=\hskip-4.30554ptB,B^{*},D,D^{*},K,\rm or\,K^{*})$ in units of the in-medium bag radius of hadron $h$, $1/R^{*}_{h}$ are given by (recall that $V^{q}_{\rho}=0$ in symmetric nuclear matter below):

$$\begin{pmatrix}\epsilon_{u}\\ \epsilon_{\overline{u}}\end{pmatrix}=\Omega^{*}_{q}\pm R^{*}_{h}\left(V^{q}_{\omega}+\frac{1}{2}V^{q}_{\rho}\right),$$ (4)

$$\begin{pmatrix}\epsilon_{d}\\ \epsilon_{\overline{d}}\end{pmatrix}=\Omega^{*}_{q}\pm R^{*}_{h}\left(V^{q}_{\omega}-\frac{1}{2}V^{q}_{\rho}\right),$$ (5)

$$\epsilon_{s,c,b}=\epsilon_{\overline{s},\overline{c},\overline{b}}=\Omega^{*}_{s,c,b}.$$ (6)

The in-medium mass of the hadron $h$, $m^{*}_{h}$, for a given nuclear density is calculated by

$$\hskip-8.61108ptm^{*}_{h}=\sum_{j=q,\overline{q},Q,\overline{Q}}\frac{n_{j}\Omega^{*}_{j}-Z_{h}}{R^{*}_{h}}+\frac{4}{3}\pi R^{*3}_{h}B_{p},\hskip 8.61108pt\left.\frac{dm^{*}_{h}}{dR_{h}}\right|_{R_{h}=R^{*}_{h}}=0,$$ (7)

with $\Omega^{*}_{q}=\Omega^{*}_{\overline{q}}=\left[x^{2}_{q}+\left(R^{*}_{h}m^{*}_{q}\right)^{2}\right]^{\frac{1}{2}}$, where $m^{*}_{q}=m_{q}-V^{q}_{\sigma}$ and $\Omega^{*}_{Q}=\Omega^{*}_{\overline{Q}}=\left[x^{2}_{Q}+\left(R^{*}_{h}m_{Q}\right)^{2}\right]^{\frac{1}{2}}$ when $h$ contains a light quark (when $h$ contains no light quarks, $R_{h}^{*}\to R_{h}$), and $x_{q,Q}$ are the lowest mode bag eigenfrequencies. Note that, when $h$ contains a light quark, $x_{Q}$ is also modified slightly from the free space value due to the in-medium modification of $R^{*}_{h}$, and thus $\Omega_{Q}$ is also modified as $\Omega^{*}_{Q}$, from that in free space. $B_{p}$ is the bag constant, and $n_{q,Q}$ ($n_{\overline{q},\overline{Q}}$) are the lowest mode quark (antiquark) numbers for the quark flavors $q$ and $Q$ in the hadron $h$, and the $Z_{h}$ parametrizes the sum of the center-of-mass and gluon fluctuation effects, which is assumed to be independent of density Guichon:1995ue .

The current quark mass values used are $(m_{q},m_{s},m_{c},m_{b})$ = (5, 250, 1270, 4200) MeV. (See Ref. Tsushima:2020gun for the other values used, $(m_{q},m_{s},m_{c},m_{b})$ = (5, 93, 1270, 4180) MeV.) Note that, in phenomenological quark models, the quark mass values are not necessarily related with the values in quantum chromodynamics (QCD). The free space nucleon bag radius is chosen to be $R_{N}$ = 0.8 fm, and the light quark-meson coupling constants, $g^{q}_{\sigma}$, $g^{q}_{\omega}$ and $g^{q}_{\rho}$, are determined by the fit to the symmetric saturation energy (-15.7 MeV) at the saturation density ($\rho_{0}=0.15$ fm^-3), and the bulk symmetry energy (35 MeV), and the explicit parameter values obtained are given in Refs. Guichon:1987jp ; Saito:2005rv ; Tsushima:2020gun .

In Table 1 we summarize the free space meson mass values (input) taken from Particle Data Group (PDG) ParticleDataGroup:2022pth except for that of the $m_{B^{*}_{c}}$ which is not determined experimentally, and for this, we use the average value ofo $m_{B^{*}_{c}}$ from Table III in Ref. Martin-Gonzalez:2022qwd . We also list the in-medium mass values for some mesons appearing in Fig. 1, at $\rho_{0},2\rho_{0}$ and $3\rho_{0}$ ($\rho_{0}=0.15$ fm^-3) calculated by the QMC model. The entries with the bold face are the focus of the present study.

Table 1: Free space meson mass values (2nd column, input) from Particle Data Group (PDG) ParticleDataGroup:2022pth , and in-medium masses for some mesons, at densities $\rho_{0}$, $2\rho_{0}$ and $3\rho_{0}$ with $\rho_{0}$ = 0.15 fm^-3 calculated by the QMC model with $(m_{u,d}$,$m_{s}$,$m_{c}$,$m_{b})$ = (5,250,1270,4200) MeVTsushima:2020gun . (For $m_{B^{*}_{c}}$ we use the averaged value from Table III in Ref. Martin-Gonzalez:2022qwd .) The entries with bold face are those for the two-flavored heavy mesons, that we focus on the present study. (All units are in MeV.)

	$\rho_{B}=0$	$\rho_{B}=\rho_{0}$	$\rho_{B}=2\rho_{0}$	$\rho_{B}=3\rho_{0}$
$m_{K}$	493.7	430.5	393.6	369.0
$m_{K^{*}}$	893.9	831.9	797.2	775.0
$m_{D}$	1867.2	1805.2	1770.6	1748.4
$m_{D^{*}}$	2008.6	1946.9	1912.9	1891.2
$m_{B}$	5279.3	5218.2	5185.1	5164.4
$m_{B^{*}}$	5324.7	5263.7	5230.7	5210.2
$\bf m_{B_{c}}$	6274.5
$\bf m_{B^{*}_{c}}$	6333.0
$\bf m_{B^{0}_{s}}$	5366.9
$\bf m_{B^{*}_{s}}(\equiv m_{B^{*0}_{s}})$	5415.4
$\bf m_{D_{s}}$	1968.4
$\bf m_{D^{*}_{s}}$	2112.2

In the QMC model the density dependent negative mass shifts of the $B$, $B^{*}$, $D$, $D^{*}$, $K$ and $K^{*}$ mesons are obtained Tsushima:2020gun ; Zeminiani:2020aho ; Zeminiani:2023gqc . The model predicts the mass shifts ($\Delta m$ $\equiv$ $m^{*}-m$) of these mesons in symmetric nuclear matter at $\rho_{0}=0.15$ fm^-3: $(\Delta m_{B},\Delta m_{B^{*}},\Delta m_{D},\Delta m_{D^{*}},\Delta m_{K},\Delta m_{K^{*}})$ = (-61.13, -61.05, -61.97, -61.66, -63.20, -61.97) MeV. We use the density dependent in-medium masses of these mesons for estimating the $B_{c}$, $B^{*}_{c}$, $B_{s}$, $B^{*}_{s}$, $D_{s}$ and $D^{*}_{s}$ meson self-energies in symmetric nuclear matter, and since the vector potentials cancel out in the intermediate excited state mesons, we do not need them.

Figure 2: $B$ and $B^{*}$ (top), $D$ and $D^{*}$ (middle) and $K$ and $K^{*}$ (middle) meson Lorentz-scalar effective masses in symmetric nuclear matter versus baryon density ($\rho_{B}/\rho_{0}$), calculated by the QMC model.

3 Effective Lagrangian approach for two-flavored heavy meson mass shift

In the following, we mainly focus on the $B_{c}$ and $B_{c}^{*}$ mesons. The $B_{c}$ [$B^{*}_{c}$] in-medium mass shift arises from the $B^{*}D+BD^{*}$ [$BD$] loop contribution to the self-energy, relative to those in free space (see figures (a) and (a’) [(d)] in Fig. 1).

The self-energies are calculated based on a flavor-SU(5) symmetric effective Lagrangian densities Lin:2000ke . (Hereafter we will simply call ”Lagrangian”.) The free Lagrangian for pseudoscalar and vector mesons (denoted respectively by $P$ and $V$ of $5\times 5$ matrix representations to be shown later) is given by Lin:2000ke ,

$${\cal L}_{0}=Tr\left(\partial_{\mu}P^{\dagger}\partial^{\mu}P\right)-\frac{1}{2}Tr\left(F_{\mu\nu}^{\dagger}F^{\mu\nu}\right),$$ (8)

with $F_{\mu\nu}=\partial_{\mu}V_{\nu}-\partial_{\nu}V_{\mu}$, where $P$ and $V$ (suppressing the Lorentz indices for $V$) are, respectively, the $5\times 5$ pseudoscalar and vector meson matrices in SU(5) and given below (for a clearer identification of the matrix elements, see, e.g., Refs. Lin:2000ke ; Zeminiani:2020aho ; Zeminiani:2023gqc ):

$$\hskip-60.27759ptP=\frac{1}{\sqrt{2}}\begin{pmatrix}\frac{\pi^{0}}{\sqrt{2}}+\frac{\eta}{\sqrt{6}}+\frac{\eta_{c}}{\sqrt{12}}+\frac{\eta_{b}}{\sqrt{20}}&\pi^{+}&K^{+}&\overline{D}^{0}&B^{+}\\ \pi^{-}&\frac{-\pi^{0}}{\sqrt{2}}+\frac{\eta}{\sqrt{6}}+\frac{\eta_{c}}{\sqrt{12}}+\frac{\eta_{b}}{\sqrt{20}}&K^{0}&D^{-}&B^{0}\\ K^{-}&\overline{K}^{0}&\frac{-2\eta}{\sqrt{6}}+\frac{\eta_{c}}{\sqrt{12}}+\frac{\eta_{b}}{\sqrt{20}}&D_{s}^{-}&B_{s}^{0}\\ D^{0}&D^{+}&D_{s}^{+}&\frac{-3\eta_{c}}{\sqrt{12}}+\frac{\eta_{b}}{\sqrt{20}}&B_{c}^{+}\\ B^{-}&\overline{B^{0}}&\overline{B_{s}^{0}}&B_{c}^{-}&\frac{-2\eta_{b}}{\sqrt{5}}\\ \end{pmatrix},$$

$$\hskip-60.27759ptV=\frac{1}{\sqrt{2}}\begin{pmatrix}\frac{\rho^{0}}{\sqrt{2}}+\frac{\omega}{\sqrt{6}}+\frac{J/\Psi}{\sqrt{12}}+\frac{\Upsilon}{\sqrt{20}}&\rho^{+}&K^{*+}&\overline{D}^{*0}&B^{*+}\\ \rho^{-}&\frac{-\rho^{0}}{\sqrt{2}}+\frac{\omega}{\sqrt{6}}+\frac{J/\Psi}{\sqrt{12}}+\frac{\Upsilon}{\sqrt{20}}&K^{*0}&D^{*-}&B^{*0}\\ K^{*-}&\overline{K}^{*0}&\frac{-2\omega}{\sqrt{6}}+\frac{J/\Psi}{\sqrt{12}}+\frac{\Upsilon}{\sqrt{20}}&D_{s}^{*-}&B_{s}^{*0}\\ D^{*0}&D^{*+}&D_{s}^{*+}&\frac{-3J/\Psi}{\sqrt{12}}+\frac{\Upsilon}{\sqrt{20}}&B_{c}^{*+}\\ B^{*-}&\overline{B^{*0}}&\overline{B_{s}^{*0}}&B_{c}^{*-}&\frac{-2\Upsilon}{\sqrt{5}}\\ \end{pmatrix}.$$

By the following minimal substitutions,

$$\partial_{\mu}P\rightarrow\partial_{\mu}P-\frac{ig}{2}\left[V_{\mu}\text{, P}\right],$$ (9)

$$F_{\mu\nu}\rightarrow\partial_{\mu}V_{\nu}-\partial_{\nu}V_{\mu}-\frac{ig}{2}\left[V_{\mu},\penalty 10000\ V_{\nu}\right],$$ (10)

one obtains the effective interaction Lagrangians Lin:2000ke . For the $B_{c}B^{*}D$, $B_{c}BD^{*}$ and $B^{*}_{c}BD$ interactions the relevant Lagrangians are Zeminiani:2023gqc ; Lodhi:2007zz :

$$\hskip-25.83325pt\mathcal{L}_{B_{c}B^{*}D} = ig_{B_{c}B^{*}D}[(\partial_{\mu}B^{-}_{c}){D}-B^{-}_{c}(\partial_{\mu}{D})]B^{*\mu}+h.c.,$$

$$\hskip-25.83325pt\mathcal{L}_{B_{c}BD^{*}} = ig_{B_{c}BD^{*}}[(\partial_{\mu}B^{+}_{c})\overline{B}-B_{c}^{+}(\partial_{\mu}\overline{B})]\overline{D^{*}}^{\mu}+h.c.,$$

$$\hskip-25.83325pt\mathcal{L}_{B^{*}_{c}BD} = -ig_{B^{*}_{c}BD}B^{*+{\mu}}_{c}[\overline{B}(\partial_{\mu}\overline{D})-(\partial_{\mu}\overline{B})\overline{D}]+h.c.,$$ (11)

where the following conventions are used with the superscript $T$ standing for the ”transposition” operation:

$$B =\begin{pmatrix}B^{+}\\ B^{0}\end{pmatrix}, \hskip-8.61108pt\overline{B}=\begin{pmatrix}B^{-}\\ \overline{B}^{0}\end{pmatrix}^{T}, \hskip 4.30554ptB^{*}=\begin{pmatrix}B^{*+}\\ B^{*0}\end{pmatrix}, \hskip-8.61108pt\overline{B^{*}} =\begin{pmatrix}B^{*-}\\ \overline{B}^{*0}\end{pmatrix}^{T},$$

$$\overline{D} =\begin{pmatrix}\overline{D}^{0}\\ D^{-}\end{pmatrix}, \hskip-8.61108ptD=\begin{pmatrix}D^{0}\\ D^{+}\end{pmatrix}^{T}, \hskip 4.30554pt\overline{D}^{*}=\begin{pmatrix}\overline{D}^{*0}\\ D^{*-}\end{pmatrix}, \hskip-8.61108ptD^{*} =\begin{pmatrix}D^{*0}\\ D^{*+}\end{pmatrix}^{T}.$$

The universal coupling constant $g$ appearing in the flavor SU(5) Lagrangian has the following relation, and also the value is fixed as Zeminiani:2023gqc , $g_{\Upsilon BB}=\frac{5g}{4\sqrt{10}}\approx 13.2$, using the $\Upsilon$ decay data $\Gamma(\Upsilon\to e^{+}e^{-})$ and the vector meson dominance (VMD) model Lin:2000ke ; Zeminiani:2020aho . The coupling constant values appearing in Eq. (11) are fixed by the flavor SU(5) symmetry:

$$g_{B_{c}B^{*}D} = \frac{2}{\sqrt{5}}g_{\Upsilon BB},$$ (12)

$$g_{B_{c}B^{*}D} = g_{B_{c}BD^{*}}=g_{B^{*}_{c}BD}=\frac{g}{2\sqrt{2}}\approx 11.9.$$ (13)

The in-medium $B_{c}$ mass shift, $\Delta m_{B_{c}}$ (Lorentz-scalar), is computed by the difference of the in-medium $m^{*}_{B_{c}}$ and the free space $m_{B_{c}}$ masses as,

$$\Delta m_{B_{c}}=m^{*}_{B_{c}}-m_{B_{c}},$$ (14)

where, the free space mass $m_{B_{c}}$ (input) is used to determine the bare mass $m^{0}_{B_{c}}$ with the $B_{c}$ self-energy $\Sigma_{B_{c}}$ by

$$m^{2}_{B_{c}}=\left(m^{0}_{B_{c}}\right)^{2}-{\displaystyle|}\Sigma_{B_{c}}(k^{2}=m^{2}_{B_{c}}){\displaystyle|}.$$ (15)

Note that, the total self-energy $\Sigma_{B_{c}}$ is calculated by the ($B^{*}D$ + $BD^{*}$) meson loop contribution in free space, ignoring any possible $B_{c}$ meson width as well as those of the other mesons (or imaginary part) in the self-energy. The in-medium $B_{c}$ mass $m^{*2}_{B_{c}}$ is similarly calculated using the same bare mass value $m^{0}_{B_{c}}$ determined in free space by reproducing the observed mass, using the density dependent in-medium masses of the ($B,B^{*},D,D^{*}$) mesons ($m^{*}_{B},m^{*}_{B^{*}},m^{*}_{D},m^{*}_{D^{*}}$), namely,

$$\hskip-8.61108ptm^{2}_{B_{c}} = \left[m^{0}_{B_{c}}(B^{*}D+BD^{*})\right]^{2}$$ (16)

$$-\left|\Sigma_{B_{c}}(B^{*}D)+\Sigma_{B_{c}}(BD^{*})\right|(k^{2}=m^{2}_{B_{c}}),$$

$$\hskip-8.61108ptm^{*2}_{B_{c}} = \left[m^{0}_{B_{c}}(B^{*}D+BD^{*})\right]^{2}$$ (17)

$$-\left|\Sigma^{*}_{B_{c}}(B^{*}D)+\Sigma^{*}_{B_{c}}(BD^{*})\right|(k^{*2}=m^{*2}_{B_{c}}).$$

We note that, when the self-energy graphs contain different contributions as the present case as $\Sigma_{B_{c}}({\rm total})=\Sigma(B^{*}D)$ + $\Sigma(BD^{*})$ in free space, $m^{0}$ depends on both $\Sigma(B^{*}D)$ and $\Sigma(BD^{*})$ in reproducing the physical mass $m_{B_{c}}$. Thus, one must be careful when discussing the $B_{c}$ in-medium mass and mass shift of each loop contribution $\Sigma(B^{*}D)$ and $\Sigma(BD^{*})$, since $m^{0}(B^{*}D+BD^{*})\neq m^{0}(B^{*}D)\neq m^{0}(BD^{*})$, and $m^{0}(B^{*}D+BD^{*})\neq m^{0}(B^{*}D)+m^{0}(BD^{*})$. The dominant loop contribution should be known by the decomposition of the total contribution in the self-energy, $\Sigma^{(*)}_{B_{c}}(B^{*}D+BD^{*})=\Sigma^{(*)}_{B_{c}}(B^{*}D)+\Sigma^{(*)}_{B_{c}}(BD^{*})$.

We focus on the $B^{*}D$ loop contribution now, without that of the $BD^{*}$ loop. The $B_{c}$ self-energy is calculated by

$$\Sigma^{B^{*}D}_{B_{c}}(m^{*}_{B_{c}})=\frac{-4g^{2}_{B_{c}B^{*}D}}{\pi^{2}}\int d|\textbf{k}|\,|\textbf{k}|^{2}I_{B_{c}}^{B^{*}D}(|\textbf{k}|)F_{B_{c}B^{*}D}(\textbf{k}^{2}),$$ (18)

where $I_{B_{c}}^{B^{*}D}(|\textbf{k}|)$ is expressed as below, after the Cauchy integral with respect to $k^{0}$ complex plane by shifting $k^{0}$ variable for the vector potential:

$$\hskip-43.05542ptI_{B_{c}}^{B^{*}D}(|\textbf{k}|)=$$

$$\hskip-34.44434pt\left.\frac{m^{*2}_{B_{c}}\left(-1+k^{2}_{0}/m^{*2}_{B^{*}}\right)}{(k_{0}-\omega^{*}_{B^{*}})(k_{0}-m^{*}_{B_{c}}+\omega^{*}_{D})(k_{0}-m^{*}_{B_{c}}-\omega^{*}_{D})}\right|_{k_{0}=-\omega^{*}_{B^{*}}}$$

$$\left.\hskip-34.44434pt+\frac{m^{*2}_{B_{c}}\left(-1+k^{2}_{0}/m^{*2}_{B^{*}}\right)}{(k_{0}+\omega^{*}_{B^{*}})(k_{0}-\omega^{*}_{B^{*}})(k_{0}-m^{*}_{B_{c}}-\omega^{*}_{D})}\right|_{k_{0}=m^{*}_{B_{c}}-\omega^{*}_{D}}.$$ (19)

In Eq. (18), $F_{B_{c}B^{*}D}$ is the product of vertex form factors to regularize the divergence in the loop integral, $F_{B_{c}B^{*}D}(\textbf{k}^{2})$ $\equiv$ $u_{B_{c}B^{*}}(\textbf{k}^{2})\times u_{B_{c}D}(\textbf{k}^{2})$, and $u_{B_{c}B^{*}}(\textbf{k}^{2})$ and $u_{B_{c}D}(\textbf{k}^{2})$ are respectively given by $u_{B_{c}B^{*}}(\textbf{k}^{2})$ = $\left(\frac{\Lambda^{2}_{B^{*}}+m^{2}_{B_{c}}}{\Lambda^{2}_{B^{*}}+4\omega^{2}_{B^{*}}(\textbf{k}^{2})}\right)^{2}$ and $u_{B_{c}D}(\textbf{k}^{2})$ = $\left(\frac{\Lambda^{2}_{D}+m^{2}_{B_{c}}}{\Lambda^{2}_{D}+4\omega^{2}_{D}(\textbf{k}^{2})}\right)^{2}$ with $\Lambda_{B^{*}}$ and $\Lambda_{D}$ being the corresponding cutoff masses associated with the $B^{*}$ and $D$ mesons, respectively. We use the common value $\Lambda\equiv\Lambda_{B^{*}}=\Lambda_{D}$. A similar calculation can be performed for the $BD^{*}$ loop contribution by replacing $(B^{*},D)\to(B,D^{*})$ in Eqs. (18), and (19) (including the form factors).

Furthermore, the $B_{c}^{*}$ meson self-energy can be calculated straightforwardly. (See Ref. Zeminiani:2023gqc for details.)

The choice of cutoff mass value has nonegligible impacts on the calculated self-energy results. In the present study, we use the common cutoff mass value $\Lambda\equiv\Lambda_{B,B^{*},D,D^{*},K,K^{*}}$ for all the form factors appearing in the self-energy integrals, and vary the $\Lambda$ value. The $\Lambda$ value may be associated with the energies to probe the internal structure (finite sizes) of the mesons. In the previous study Zeminiani:2020aho it was observed that when the value of the cutoff mass becomes close to the masses of the mesons appearing in the self-energies, such large cutoff mass value and those values larger than that value did not make sense to serve as ”proper” form factor. This is because the Compton wavelengths of the corresponding cutoff mass values reach near and/or smaller than those of the meson sizes, and they do not represent the effect of the ”meson’s finite size” properly. Therefore, we need to constrain the $\Lambda$ value in such a way that the form factors reflect properly the finite sizes of the mesons. Based on the heavy quark and heavy meson symmetry in QCD, it may be justified that we use the same range of values for $\Lambda$ as it was practiced for the cases of quarkonia Zeminiani:2020aho . Namely, we use the values, $\Lambda$ = 2000, 3000, 4000, 5000 and 6000 MeV.

In Table 2 we give the cutoff mass ($\Lambda$) value dependence of the bare mass $m^{0}$ in Eq. (15) for different cases of the self-energy calculations of $B_{c}$ meson, namely, only the $B^{*}D$ loop, only the $BD^{*}$ loop, and $B^{*}D+BD^{*}$ loops.

Table 2: Cutoff $\Lambda$ value dependence of bare mass $m^{0}$ in Eq. (15) for the cases of only the $B^{*}D$ loop, only the $BD^{*}$ loop, and $B^{*}D+BD^{*}$ loops. (All numbers are in MeV).

$\Lambda$	$m^{0}(B^{*}D)$	$m^{0}(BD^{*})$	$m^{0}(B^{*}D+BD^{*})$
2000	7906.1	9925.4	11029.6
3000	8032.6	10313.9	11468.7
4000	8249.0	10913.0	12156.0
5000	8561.5	11732.3	13098.8
6000	8968.6	12766.6	14284.6

One can notice from the results in Table 2 as follows. First, the differences in the $m^{0}$ values for the different self-energy graphs with the same cutoff $\Lambda$ value (each low) as already mentioned. Second, the bare mass $m^{0}$ value increases as the cutoff mass $\Lambda$ value increases (each column), for all the individual cases of the self-energy contributions, $B^{*}D,BD^{*}$ and $B^{*}D+BD^{*}$ loops.

Similarly, for the other two-flavored heavy mesons $B_{s},B_{s}^{*},D_{s}$ and $D_{s}^{*}$, with the same five $\Lambda$ values and the corresponding form factors, we can estimate their in-medium mass shifts using the following interaction Lagrangians Zeminiani:2023gqc :

$$\hskip-25.83325pt\mathcal{L}_{B_{s}B^{*}K} = ig_{B_{s}B^{*}K}[(\partial_{\mu}\overline{B^{0}_{s}})\overline{K}-\overline{B^{0}_{s}}(\partial_{\mu}\overline{K})]{B}^{*\mu}+h.c.,$$

$$\hskip-25.83325pt\mathcal{L}_{B_{s}BK^{*}} = ig_{B_{s}BK^{*}}[(\partial_{\mu}{B^{0}_{s}})\overline{B}-{B^{0}_{s}}(\partial_{\mu}\overline{B})]{K}^{*\mu}+h.c.,$$

$$\hskip-25.83325pt\mathcal{L}_{B^{*}_{s}BK} = -ig_{B^{*}_{s}BK}{B_{s}^{*}}^{0\mu}[\overline{B}(\partial_{\mu}K)-(\partial_{\mu}\overline{B})K]+h.c.,$$

$$\hskip-25.83325pt\mathcal{L}_{D_{s}D^{*}K} = ig_{D_{s}D^{*}K}[(\partial_{\mu}D_{s}^{+})\overline{K}-D_{s}^{+}(\partial_{\mu}\overline{K})]\overline{D^{*}}^{\mu}+h.c.,$$

$$\hskip-25.83325pt\mathcal{L}_{D_{s}DK^{*}} = ig_{D_{s}DK^{*}}[(\partial_{\mu}D^{-}_{s})D-D_{s}^{-}(\partial_{\mu}D)]{K}^{*{\mu}}+h.c.,$$

$$\hskip-25.83325pt\mathcal{L}_{D^{*}_{s}DK} = -ig_{D^{*}_{s}DK}{D_{s}^{*-}}^{\mu}[D(\partial_{\mu}K)-(\partial_{\mu}D)K]+h.c.$$ (20)

The coupling constant values associated with each vertex in the above, are obtained from the universal SU(5) coupling $g$ value as,

$$\hskip-34.44434ptg_{B_{s}KB^{*}}= g_{B_{s}BK^{*}}=g_{B^{*}_{s}BK}=g_{D_{s}KD^{*}}=g_{D_{s}DK^{*}}=g_{D^{*}_{s}DK}$$

$$= \frac{g}{2\sqrt{2}}\approx 11.9.$$ (21)

4 Mass shift results

We are now in a position to present our estimates (predictions) of the in-medium mass shifts for the two-flavored heavy mesons, ($B_{c},B_{c}^{*}$), ($B_{s},B_{s}^{*}$), and ($D_{s},D_{s}^{*}$) mesons Zeminiani:2023gqc respectively in Figs. 3, 4 and 5.

Figure 3: The total ($B^{*}D$ + $BD^{*}$) loop contribution for the in-medium $B_{c}$ mass shift (upper panel) and that of the $BD$ loop for $B_{c}^{*}$ (lower panel), versus baryon density ($\rho_{B}/\rho_{0}$) for five different cutoff mass values $\Lambda$.

Figure 4: The total ($B^{*}K$+$BK^{*}$) loop contribution for the in-medium $B_{s}^{0}$ mass shit (upper panel) and that of the $BK$ loop for the $B_{s}^{*}\equiv B_{s}^{0*}$ (lower panel), versus baryon density ($\rho_{B}/\rho_{0}$) for five different cutoff mass values $\Lambda$.

Figure 5: The total ($D^{*}K$ + $DK^{*}$) loop contribution for the in-medium $D_{s}$ mass shift (upper panel) and that of the $DK$ loop for the $D_{s}^{*}$ (lower panel), versus baryon density for five different cutoff mass values $\Lambda$.

Among the all results given in Figs. 3, 4 and 5, the cutoff mass value dependence of $\Delta m_{B_{s}^{*}}$ (the lower panel of Fig. 4) shows ”anomalous” behavior. Namely, $\Lambda=2000$ MeV (thick dotted black line) gives the most negative mass shift, despite all the other cases give the smallest negative mass shift for each case with this value. Furthermore, the mass-shift tendency (order) with the cutoff mass values are also different from the ”typical one”, e.g., that shown in the lower panel of Fig. 5 – as the cutoff mass value becomes larger, the more negative mass shift is obtained. This may be attributed that the mass difference in the intermediate state excited mesons of $B$ and $K$, both in free space as well as in medium. That is, the mass difference of them is the largest (see Table 1), and the role of $4\omega_{K}{\textbf{k}^{2})}$ relative to $m_{B_{c}^{*}}^{2}$ in the form factor $u_{B_{c}^{*}K}$ gives larger effects (in particular $\Lambda=2000$ MeV) than that of $4\omega_{K}{\textbf{k}^{2})}$ relative to $m_{D_{s}^{*}}^{2}$ in the form factor $u_{D_{s}^{*}K}$. The former tends to give larger value than that of the latter, and thus larger self-energy enhancement to yield larger negative mass shift. But this analysis may not be sufficiently quantitative.

In addition to the density dependence of the mass shifts shown in Figs. 3, 4 and 5, we summarize below the mass shift ranges obtained at normal nuclear matter density $\rho_{0}=0.15$ fm^-3 with the five different cutoff mass values in the form factors:
(1) for $B_{c}$, $\Delta m_{B_{c}}$ = [-101.1, -90.4] MeV,
(2) for $B^{*}_{c}$, $\Delta m_{B^{*}_{c}}$ = [-19.7, -14.5] MeV,
(3) for $B_{s}$, $\Delta m_{B_{s}}$ = [-178.8, -133.0] MeV,
(4) for $B^{*}_{s}$, $\Delta m_{B^{*}_{s}}$ = [-20.5, -16.0] MeV,
(5) for $D_{s}$, $\Delta m_{D_{s}}$ = [-119.0, -78.0] MeV,
(6) for $D^{*}_{s}$, $\Delta m_{D^{*}_{s}}$ = [-20.2, -52.1] MeV.

5 Summary and Conclusions

We have presented our predictions for the in-medium mass shifts of two-flavored heavy mesons, $B_{c},B_{c}^{*},B_{s},B_{s}^{*},D_{s}$ and $D_{s}^{*}$ in symmetric nuclear matter. The estimates are made by evaluating the lowest order one-loop self-energy graphs at the hadronic level, using a flavor-SU(5) effective Lagrangian approach, with the in-medium inputs calculated by the quark-meson coupling (QMC) model. We have not included any possible meson widths nor imaginary parts of the potentials. Our results show the negative mass shifts for all the mesons studied, $B_{c},B_{c}^{*},B_{s},B_{s}^{*},D_{s}$ and $D_{s}^{*}$. In the evaluation, we have used some specific form factors to regularize the divergent loop integrals with five values of the cutoff mass. This part may be improved or has merit to study further in the future.

The next, interesting and important step, is to study possibilities of meson-nucleus bound states for these mesons including the Coulomb potentials and/or the in-medium widths. Concerning this, we would like to comment that, some initial results for the $B_{c}$-${}^{12}C$ system single-particle energies are already available in Ref. Zeminiani:2024dyo .

Acknowledgments:   G.N.Z and K.T. thank the OMEG (Origin of Matter and Evolution of Galaxies) Institute at Soongsil University (Seoul, South Korea) for the supports and the collaboration works. G.N.Z and S.L.P.G.B. were supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Brazil. K.T. was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Brazil, Processes No. 304199/2022-2, and Fundação de Amparo Pesquisa do Estado de São Paulo (FAPESP), Brazil, Process No. 2019/00763-0 and No. 2023/07313-6. This work was in the projects of Instituto Nacional de Ciência e Tecnologia - Nuclear Physics and Applications (INCT-FNA), Brazil, Process No. 464898/2014-5.