Heavy and heavy-light tensor and axial-tensor mesons in the Covariant Spectator Theory
Elmar P. Biernat Alfred Stadler · 2026
Read Heavy and heavy-light tensor and axial-tensor mesons in the Covariant Spectator Theory by Elmar P. Biernat Alfred Stadler (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-04113.txt.
Introduction
Elmar Biernat and Alfred Stadler gave the first calculation of tensor and axial-tensor mesons with total spin at least 2 in Covariant Spectator Theory. A refined quark-antiquark interaction kernel now carries the momentum dependence of the strong coupling, replacing a constant kernel term. Global least-squares fits to established heavy and heavy-light meson masses described the spectrum for spins 0, 1, 2, and 3 parities with only eight adjustable parameters. The fit quality was excellent across the full spin-parity set.
Catalog record
| CREATOR | Elmar P. Biernat Alfred Stadler |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 11006 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.04113
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Heavy and heavy-light tensor and axial-tensor mesons in the Covariant Spectator Theory Thanks: Presented at Excited QCD Workshop 2026, Granada, Spain Elmar P. Biernat Alfred Stadler Abstract We present the first calculation of tensor and axial-tensor mesons with total spin J ≥ 2 J\geq 2 within the Covariant Spectator Theory. We employ a refined quark-antiquark interaction kernel that incorporates the momentum dependence of the strong coupling, replacing the previously used constant term of the kernel. Global least-squares fits to the masses of experimentally established heavy and heavy-light meson states yield an excellent description of the mass spectrum for J P = 0 ± , 1 ± , 2 ± J^{P}=0^{\pm},1^{\pm},2^{\pm} , and 3 ± 3^{\pm} using only eight adjustable parameters. \PACS 11.10.St, 14.40.Pq, 12.39.Pn, 03.65.Ge 1 Introduction The Covariant Spectator Theory (CST) is a modern quantum-field theoretical approach for the study of few-body systems. Its simplest version is the one-channel CST, defined by the Gross equation (GE) [ 1 , 2 ] . Two- and four-channel extensions of this framework were first applied to hadronic systems by Gross, Milana, and Şavklı in the studies of pseudoscalar and vector mesons [ 3 , 4 , 5 , 6 ] . More recently, we employed the four-channel CST to calculate the dressed quark propagator [ 7 , 8 ] , the pion form factor [ 9 , 10 ] , and the \pi - \pi scattering amplitude in the chiral limit [ 11 ] . For the quark-antiquark bound state problem, we used the GE so far to compute the masses and vertex functions of heavy and heavy-light mesons with total spin J\leq 1 [ 12 , 13 , 14 , 15 ] . The present work extends these bound-state calculations by generalizing the formalism to mesons of arbitrary spin-parity J^{P} . This extension enables, for the first time, a unified fit of all currently established quark-antiquark meson states containing at least one heavy quark. It also predicts higher-spin states and guides the J^{P} assignment of experimentally observed states listed by the Particle Data Group (PDG) that remain unconfirmed or have uncertain quark content. Furthermore, whereas previous CST calculations treated the strong coupling \alpha_{\rm s} as a constant, we incorporate its momentum dependence here, yielding an improved overall description of the heavy and heavy-light meson spectrum. 2 The one-channel CST formalism The GE for a q\bar{q} meson with spin-parity J^{P} is given by \Gamma(\hat{p}_{1},p_{2})=-\int\frac{\mathrm{d}^{3}k_{1}}{(2\pi)^{3}}\frac{m_{1}}{E_{1k}}\mathcal{V}(\hat{p}_{1},\hat{k}_{1})\Lambda(\hat{k}_{1})\Gamma(\hat{k}_{1},k_{2})S(k_{2})\, (1) where \hat{p}_{1}=(E_{1p},\bm{p}) and \hat{k}_{1}=(E_{1k},\bm{k}) are the external and internal on-mass-shell four-momenta of quark 1 with constituent mass m_{1} and energies E_{1p}=\sqrt{m_{1}^{2}+\bm{p}^{2}} and E_{1k}=\sqrt{m_{1}^{2}+\bm{k}^{2}} , respectively; the external and internal four-momenta of quark 2 are p_{2}=\hat{p}_{1}-P and k_{2}=\hat{k}_{1}-P , respectively, where P=(\mu,\bm{0}) is the (on-mass-shell) four-momentum of the meson with mass \mu in the rest frame. The operators \displaystyle\Lambda(\hat{k}_{1})=\frac{m_{1}+\hat{\not{k}}_{1}}{2m_{1}}\quad\text{and}\quad S(k_{2})=\frac{m_{2}+\not{k}_{2}}{m_{2}^{2}-k_{2}^{2}-\mathrm{i}\epsilon} (2) are, respectively, the positive-energy projector of quark 1 and the dressed propagator of quark 2 with constant constituent mass m_{2} . The CST interaction kernel is given by [ 13 ] \mathcal{V}(\hat{p}_{1},\hat{k}_{1})=1_{1}\otimes 1_{2}V_{\rm L}(\hat{p}_{1},\hat{k}_{1})-\gamma^{\mu}_{1}\otimes\gamma_{\mu 2}\left[V_{\rm C}(\hat{p}_{1},\hat{k}_{1})+V_{\rm G}(\hat{p}_{1},\hat{k}_{1})\right]\,. (3) The three terms in Eq. ( 3 ) correspond to covariant generalizations of the linear confining ( V_{\rm L} ) and a constant ( V_{\rm C} ) potentials, and a one-gluon-exchange (OGE) interaction in Feynman gauge ( V_{\rm G} ). These are given by \displaystyle V_{\rm L}(\hat{p}_{1},\hat{k}_{1})=-8\pi\sigma\left[\left(\frac{1}{(\hat{p}_{1}-\hat{k}_{1})^{4}}-\frac{1}{(\lambda_{\rm L}m_{1})^{4}+(\hat{p}_{1}-\hat{k}_{1})^{4}}\right)\right. \displaystyle\left.-\frac{E_{1p}}{m_{1}}(2\pi)^{3}\delta^{3}(\bm{p}-\bm{k})\int\frac{\mathrm{d}^{3}k_{1}^{\prime}}{(2\pi)^{3}}\frac{m_{1}}{E_{1k^{\prime}}}\left(\frac{1}{(\hat{p}_{1}-\hat{k}_{1}^{\prime})^{4}}-\frac{1}{(\lambda_{\rm L}m_{1})^{4}+(\hat{p}_{1}-\hat{k}_{1}^{\prime})^{4}}\right)\right]\,, (4) V_{\rm C}(\hat{p}_{1},\hat{k}_{1})=\frac{E_{1k}}{m_{1}}(2\pi)^{3}C\delta^{3}(\bm{p}-\bm{k})\,, (5) and V_{\rm G}(\hat{p}_{1},\hat{k}_{1})=-\frac{16\pi}{3}\alpha_{\rm s}\left((\hat{p}_{1}-\hat{k}_{1})^{2}\right)\left(\frac{1}{(\hat{p}_{1}-\hat{k}_{1})^{2}}-\frac{1}{(\hat{p}_{1}-\hat{k}_{1})^{2}-(\lambda_{\rm G}m_{1})^{2}}\right)\,, (6) where the strong running coupling is \alpha_{\rm s}(q^{2})=\frac{1}{\beta_{0}\ln\left(\frac{-q^{2}}{\Lambda_{\rm QCD}}+\tau\right)}\,, (7) with \beta_{0}=\frac{33-2N_{f}}{12\pi} , and N_{f} is the number of active flavours. The IR regulator \tau is fixed by specifying the value of \alpha_{s}(q^{2}) at q^{2}=0 . The remaining parameter, \Lambda_{\rm QCD} , is determined by requiring that \alpha_{s}(q^{2}) reproduces the experimental value at q^{2}=-M_{Z}^{2} . We regularize the UV-behaviour of linear-confining and OGE kernels via Pauli-Villars subtraction, leaving the corresponding dimensionless cut-off parameters ( \lambda_{\rm L} and \lambda_{\rm G} ) and the coupling strengths ( \sigma , \alpha_{\rm s}(q^{2})|_{q^{2}=0} , and C ) as adjustable fit parameters. The spin- J meson vertex function \Gamma(\hat{p}_{1},p_{2}) can be written, depending on the parity P , as \displaystyle\Gamma(\hat{p}_{1},p_{2})=\zeta^{\mu\nu\sigma\ldots}_{m_{J}} \displaystyle\times\begin{cases}FM_{\mu\nu\sigma\ldots}+GN_{\mu\nu\sigma\ldots}+\left[HM_{\mu\nu\sigma\ldots}+IN_{\mu\nu\sigma\ldots}\right]\Lambda(-p_{2}),&P=(-1)^{J}\\ \left\{\tilde{F}M_{\mu\nu\sigma\ldots}+\tilde{G}N_{\mu\nu\sigma\ldots}+\left[\tilde{H}M_{\mu\nu\sigma\ldots}+\tilde{I}N_{\mu\nu\sigma\ldots}\right]\Lambda(p_{2})\right\}\gamma^{5},&P=(-1)^{J+1}\end{cases} (8) where \zeta^{\mu\nu\sigma\ldots\omega}_{m_{J}}=\left\{\xi_{\lambda_{1}}^{\mu}\otimes\xi_{\lambda_{2}}^{\nu}\otimes\cdots\otimes\xi_{\lambda_{J}}^{\omega}\right\}_{m_{J}}^{(J)} (9) is the rank- J spherical polarization tensor associated with angular momentum J and spin-polarization m_{J}=-J,\ldots,J . The four-vectors \xi_{\lambda_{i}}^{\mu} are the three spherical polarization vectors for massive spin-1 states, corresponding to spin polarizations \lambda_{i}=\pm 1 and 0 . The shorthand functions F,G,\ldots,\tilde{I} implicitly depend on the Lorentz invariants \hat{p}_{1}\cdot p_{2} and p_{2}^{2} , and \displaystyle M^{\mu\nu\sigma\ldots\omega} \displaystyle=\begin{cases}\frac{1}{J}\left(\gamma^{\mu}p^{\nu}p^{\sigma}\cdots p^{\omega}+p^{\mu}\gamma^{\nu}p^{\sigma}\cdots p^{\omega}+\ldots+p^{\mu}p^{\nu}p^{\sigma}\cdots\gamma^{\omega}\right),&J>0\\ 0,&J=0\end{cases} \displaystyle N^{\mu\nu\sigma\ldots\omega} \displaystyle=\begin{cases}p^{\mu}p^{\nu}p^{\sigma}\cdots p^{\omega},&J>0\\ \mathbf{1},&J=0\end{cases} (10) are rank- J Lorentz tensors. For the identification of states listed by the PDG, it is convenient to switch from the Lorentz-tensor basis to a partial-wave basis by decomposing Eq. ( 1 ) into its positive- and negative-energy channels. The resulting vertex-function matrix elements between u and v Dirac spinors can be expanded for a fixed J^{P} in terms of orbital angular momentum L and spin S eigenfunctions. The resulting eigenvalue problem is then solved for the set of possible partial waves and the corresponding energy eigenvalue \mu . 3 Results and Conclusions We have solved the GE, Eq. ( 1 ), for each meson sector from b\bar{b} to c\bar{q} (where q=u,d ), and the J^{P}=0^{\pm},1^{\pm},2^{\pm},3^{\pm} channels in each sector. The initially nine model parameters (including the constant C and the constituent quark masses m_{u},m_{s},m_{c} , and m_{b} ) were adjusted through global least-squares fits to experimentally measured states. We evaluated three parameter models, differing by the set of states included in the fit: 10 pseudoscalar states, 33 non-axial states ( J^{\mathcal{P}}=0^{\pm},1^{-},2^{+},3^{-} ), and 49 states of all channels. The parameter values for the three models are listed in Table 1 . Note that the strength C of the constant interaction is omitted from the table because the fit consistently yields C\approx 0 . In models with a fixed \alpha_{\mathrm{s}} , the constant potential effectively simulates the missing running behaviour. By explicitly incorporating the momentum dependence of \alpha_{\mathrm{s}} , the constant interaction becomes redundant, leaving us with only eight adjustable model parameters. Furthermore, implementing the running of \alpha_{\mathrm{s}} substantially improves the global fit compared to using a constant \alpha_{\mathrm{s}} . The best overall agreement is obtained with N_{f}=2 active flavours, which performs slightly better than N_{f}=3 . Table 1: Parameter values of the three fit models with N_{f}=2 . Fitted Strengths Masses [GeV] Cut-offs states \sigma [GeV 2 ] \alpha_{\rm s}(0) m_{b} m_{c} m_{s} m_{q} \lambda_{\rm L} \lambda_{\rm G} 10 0.216 0.419 4.794 1.441 0.274 0.133 1.219 1.786 33 0.179 0.507 4.852 1.508 0.343 0.185 2.812 2.266 49 0.176 0.522 4.859 1.517 0.353 0.197 2.903 2.243 The resulting mass spectra for the three models are presented in Fig. 1 . Figure 1: The results of fits to pseudoscalar (empty squares), non-axial (grey-filled diamonds), and all established (black dots) states, compared with the established (solid lines) and unconfirmed (dashed lines) experimental data, with the grey shading displaying the width. The red dotted lines indicate the open-flavour thresholds. We find that fitting only the pseudoscalar states yields fairly accurate predictions for the remaining J^{P} channels, including the higher-spin tensor states. The most significant deviations from the other two models occur primarily in the higher excited states. As expected, fitting all established states provides the