betaFeedback

WorksPhysics and Astronomy

Odderon as Regge oddball spin-3 in pppp and ppˉp\bar p elastic scattering

Abstract

In this work, we propose that the odderon is a Regge odd-glueball tensor spin-3. To demonstrate our proposal, we study the pppp and ppˉp\bar p elastic scattering by including contributions of the spin-3 odderon and spin-2 pomeron exchange in the processes. The phenomenological effective Lagrangian approach is used to calculate the pppp and ppˉp\bar p elastic scattering amplitudes at the tree level. In addition, a Donnachie-Landschoff ansatz of the odderon and pomeron propagators has been used in this work. We fit the theoretical results with the various experimental data of the pppp and ppˉp\bar p scattering at the TeV scale to determine the model parameters in the present work. By using the model parameters, the Chew-Frautschi plot of the tensor odderon Regge trajectory is evaluated. As a result, the odderon spin-3 mass is predicted to be 3.2 GeV. In addition, the total cross-section of our model is compatible with the results from TOTEM and its extrapolation from D0 collaboration. Moreover, the total cross-section also satisfies the Friossart bound at the Regge limit.

Introduction

Quantum ChromoDynamics (QCD) is a modern theory of strong interaction based on non-abalien color SU(3) quantum gauge field theory describing the interaction of quarks and gluons. QCD is highly successful in explaining hadronic structures and interactions at high energies (momentum exchange) where the strong coupling is small and perturbative quantum field theory is applied. However, at a low energy regime, QCD is a strongly coupled theory that we can not use the standard perturbative theory. On the other hand, hadron-hadron scattering in high center of mass energy (s\sqrt{s}) but low momentum exchange (tt) known as soft-high energy regime or a Regge limit of s→∞s \to\infty and s≫ts \gg t, the perturbative QCD is also inapplicable. Before the birth of QCD, Regge theory is invented to describe the hadron-hadron collisions by using analytical properties of the scattering amplitudes [1] including a consideration of the complex angular momentum [2]. In the Regge theory, the amplitudes of the hadronic processes are scaled as sJs^J where JJ is the spin of the exchange particles called Reggeons with fixed relevant quantum numbers [3]. One can write down the spin as a linear function of tt as J=α(t)J=\alpha(t) in the complex angular momentum plane and the linear function α(t)=α0+α′t\alpha(t)=\alpha_0+\alpha' t is called a Regge trajectory. In addition, the poles (Regge poles) correspond to the families of the exchange particles with increasing spins along the trajectories. As a result, the amplitudes of Regge theory are represented in terms of a sum over all possible exchange particles lying on the Regge trajectory. The cross-sections of various hadronic processes in the soft-high energy scattering limit are successfully described by the Regge theory [4, 5, 6, 7].

According to the experimental data of hadron-hadron scattering in the Regge limit, the total cross-sections slowly grew up with the increase of ss whereas the Regge trajectories of all known mesons are not sufficient to explain the experimental data. Then, a so-called pomeron was introduced to address this problem [8] [9]. The pomeron is a Reggeon carrying all even charge transformations, vacuum quantum number, with the intercept of Regge trajectory α0≈1\alpha_0 \approx1. This yields the slow growth of the total cross-section at large ss [10, 11]. Various approaches are trying to extract the information of the pomeron trajectory. The typical values of the parameters of the pomeron trajectory from [12–14] are α0≈1.06−1.08\alpha_0 \approx1.06 - 1.08 and α′≈0.025\alpha' \approx0.025 GeV−2^{-2} [12–14]. The pomeron is generally considered as a bound state of the gluons (glueball). On the other hand, the odd charge-conjugation counterpart of the pomeron called odderon has been proposed by Ref.[15]. Similar to the pomeron, the odderon is considered as the glueball with the odd number of gluon compositions. The odderon might cause the different observables between pppp and ppˉp\bar{p} due to its charge-conjugation property which is compatible with the experiment. However, the nature and properties of the pomeron and odderon are still unclear so far. A number of approaches have been used to calculate the properties (mass, spin, Regge trajectory et etc.) of pomeron and odderon as glueballs [16–47]. A study of pomeron and odderon exchanges in pppp and ppˉp\bar{p} elastic scatterings in the soft-high energy regime has been extensively investigated in various frameworks for instances, phenomenological approaches [48–60], QCD inspired models [61–69], holographic QCD or AdS/CFT correspondence [70, 71, 72, 73, 74, 75, 76, 77, 78, 79].

Recently, however, TOTEM and D0 collaborations have confirmed the existence of the odderon by comparing the experimental data between pppp (extrapolated from previous several data) and ppˉp\bar{p} at 1.96 TeV [80, 81]. This reveals the contributions of the odderon in tt-channel elastic scattering. After the TOTEM and D0 collaborations claimed the discovery of the odderon, several works have done to investigate the properties and scattering processes of the odderon [82, 83, 84, 85, 86, 87, 88, 89, 90, 91].

Based on the discovery of the odderon and the relevant literature on the field theoretical framework in Ref. [48], we propose the odderon as a spin-3 tensor odd-glueball within the standard field theoretical framework. We study its consequences in pppp and ppˉp\bar{p} elastic scatterings. According to a constituent quark model, the odderon is composed of three-gluon, and the lightest trajectory of the odderon is the spin-3, not the spin-1. This is because the odderon begins with a JPC=3−−J^{PC}=3^{--} three-gluon L=0L=0 state with a maximum spin of 3. Similarly, the two-gluon bound state or pomeron starts with the s-wave, and the spin and PCPC quantum number are assigned as JPC=2++J^{PC}=2^{++} glueball. Furthermore, a combined lattice QCD calculation and field theoretical Coulomb gauge QCD model confirmed that the odderon can be the oddball starting its Regge trajectory with JPC=3−−J^{PC}=3^{--} [46]. Various theoretical approaches have also shown that the pomeron is likely to be the spin-2 tensor glueball instead of the scalar one [70, 73, 74, 75, 76] [48, 49, 52, 53, 63, 68]. Then the lowest tensor odderon in its Regge trajectory is spin-3 as explained previously. However, the slope, intercept, and mass of the odderon are not well understood.

In this work, we investigate the elastic scattering of pppp and ppˉp\bar{p} with the contributions of the odderon spin-3. The effective Lagrangian of the spin-3 odderon with protons is constructed by using a similar approach in Ref.[48]. Especially, a so-called Donnachie-Landschoff ansatz is used to represent the odderon and pomeron propagators. The contributions of the spin-2 pomeron exchange are also included in the calculation where the effective Lagrangian of pomeron and protons is taken from Ref. [48]. Then the differential cross-sections of the pppp and ppˉp\bar{p} elastic scattering are calculated. After a careful statistical analysis, we fit the parameters of our model with several relevant experimental data at the TeV scale. All Feynman rules of our model such as vertices, propagators and etc., can be computed directly from the effective Lagrangians in the conventional method of perturbative QFT. The aim of the present work is to make a clear and systematic calculation in order to obtain the amplitudes. The analysis is made with the intention of compatibility with other field theoretical models.

The present work is organized as follows, in the section II, we set up our model for pppp and ppˉp\bar{p} scattering with pomeron spin-2 and odderon spin-3 exchanges. The amplitudes are also computed as well. The observables will be calculated and free parameters of our model will be fitted with the relevant experimental data at the TeV scales in the section III. In section IV, we close this work by giving discussions and conclusions.

Formalisms: Model set up and scattering amplitudes

Effective Lagrangians of the pp and p scattering in spin-2 pomeron and spin-3 odderon exchange picture

In this section, we will set up the effective Lagrangians of the pppp and ppˉp\bar{p} scattering. It is well-known that the pomeron exchange plays a major role in elastic proton-proton scattering at high energy but low momentum exchange regimes. This work we assume the pomeron as spin-2 tensor particle and the Lagrangian is given by [48, 49]

LP=−igPPμνGμν′μ′ν′ψˉγμ′∂↔ν′ψ,(1)\mathcal{L}_{\mathrm{P}}=-i g_{\mathrm{P}} \mathrm{P}^{\mu\nu} G_{\mu\nu}^{\prime\mu'\nu'}\bar{\psi}\gamma_{\mu'}\overleftrightarrow{\partial}_{\nu'}\psi, \tag*{(1)}

where PμνP^{\mu\nu} is symmetric spin-2 tensor field, ψ\psi is Dirac proton field, ψˉ∂↔μψ=(∂μψˉ)ψ−ψˉ∂μψ\bar{\psi}\overleftrightarrow{\partial}_{\mu}\psi=(\partial_{\mu}\bar{\psi})\psi-\bar{\psi}\partial_{\mu}\psi and the coupling gP=3×1.87 GeV−1g_{\mathrm{P}}=3\times1.87\ \mathrm{GeV}^{-1} as used in Ref. [48]. The totally symmetric tensor Gμν′μ′ν′G_{\mu\nu}^{\prime\mu'\nu'} is defined by

Gμν′μ′ν′=12!(gμμ′gνν′+gμν′gνμ′).(2)G_{\mu\nu}^{\prime\mu'\nu'}=\frac{1}{2!}\left(g_{\mu}^{\mu'}g_{\nu}^{\nu'}+g_{\mu}^{\nu'}g_{\nu}^{\mu'}\right). \tag*{(2)}

The corresponding vertex function of the pomeron-proton-proton coupling is given by

iΓμνP(q,q′)=−igPGμν′μ′ν′γμ′(qν′′+qν′),(3)i\Gamma_{\mu\nu}^{\mathrm{P}}(q,q')=-i g_{\mathrm{P}}G_{\mu\nu}^{\prime\mu'\nu'}\gamma_{\mu'}(q'_{\nu'}+q_{\nu'}), \tag*{(3)}

where qq and q′q' are incoming and outgoing proton/anti-proton momenta respectively.

For the spin-3 odderon exchange interaction, the effective Lagrangian reads,

LO=−gOM~0OμνρGμνρμ′ν′ρ′ψˉγμ′∂↔ν′∂↔ρ′ψ,(4)\mathcal{L}_{O}=-\frac{g_{O}}{\widetilde{M}_{0}}\mathcal{O}^{\mu\nu\rho}G_{\mu\nu\rho}^{\mu'\nu'\rho'}\bar{\psi}\gamma_{\mu'}\overleftrightarrow{\partial}_{\nu'}\overleftrightarrow{\partial}_{\rho'}\psi, \tag*{(4)}

where Oμνρ\mathcal{O}^{\mu\nu\rho} is the spin-3 tensor field that totally symmetric under interchanging of the Lorentz indices, i.e., Oμνρ=Oνρμ=Oρμν=Oμρν=Oνμρ=Oρνμ\mathcal{O}^{\mu\nu\rho}=\mathcal{O}^{\nu\rho\mu}=\mathcal{O}^{\rho\mu\nu}=\mathcal{O}^{\mu\rho\nu}=\mathcal{O}^{\nu\mu\rho}=\mathcal{O}^{\rho\nu\mu}. In order to obtain the effective Lagrangian in Eq. (4), in addition, we have followed the construction of the higher spin field coupling to the nucleons in [48] by adding the twist-2 operator as shown in Appendix B of [48] and all detail discussions therein. Moreover, the coupling gOg_{O} is a free parameter in this work and it carries the mass dimension the same as introduced for gPg_{P}. The M~0\widetilde{M}_{0} is the mass parameter with M~0=1\widetilde{M}_{0}=1 GeV. In the latter, we will see that this parameter is absent in the scattering amplitude and it is introduced for proper mass dimension. While the totally symmetric tensor Gμνρμ′ν′ρ′G_{\mu\nu\rho}^{\mu'\nu'\rho'} is used to ensure that the lower indices (μ,ν,ρ)(\mu,\nu,\rho) of the vertex functions, ΓμνρO\Gamma_{\mu\nu\rho}^{O} is totally symmetric and it is defined by

Gμνρμ′ν′ρ′=13!(gμμ′gνν′gρρ′+gνμ′gρν′gμρ′+gρμ′gμν′gνρ′+gνμ′gμρ′gρν′+gμμ′gρν′gνρ′+gρμ′gνν′gμρ′).(5)G_{\mu\nu\rho}^{\mu'\nu'\rho'}=\frac{1}{3!}\left(g_{\mu}^{\mu'}g_{\nu}^{\nu'}g_{\rho}^{\rho'}+g_{\nu}^{\mu'}g_{\rho}^{\nu'}g_{\mu}^{\rho'}+g_{\rho}^{\mu'}g_{\mu}^{\nu'}g_{\nu}^{\rho'}+g_{\nu}^{\mu'}g_{\mu}^{\rho'}g_{\rho}^{\nu'}+g_{\mu}^{\mu'}g_{\rho}^{\nu'}g_{\nu}^{\rho'}+g_{\rho}^{\mu'}g_{\nu}^{\nu'}g_{\mu}^{\rho'}\right). \tag*{(5)}

According to the Lagrangian in Eqs. (4), we can write the Feynman rules for the vertex function for LO\mathcal{L}_{O} as

iΓμνρO(q,q′)=−igOM~0Gμνρμ′ν′ρ′γμ′(qν′′+qν′)(qρ′′+qρ′),(6)i\Gamma_{\mu\nu\rho}^{O}(q,q')=-i\frac{g_{O}}{\widetilde{M}_{0}}G_{\mu\nu\rho}^{\mu'\nu'\rho'}\gamma_{\mu'}(q'_{\nu'}+q_{\nu'})(q'_{\rho'}+q_{\rho'}), \tag*{(6)}

where qq and q′q' represent the incoming and outgoing momenta of the proton/anti-proton of the vertex functions.

Scattering amplitudes of the pp and p elastic scattering

Next step, we will calculate the amplitudes for the elastic pppp and ppˉp\bar{p} scattering processes under the external momentum specifications as p(q1)p(q2)→p(q3)p(q4)p(q_{1})p(q_{2})\to p(q_{3})p(q_{4}) and pˉ(q1)p(q2)→pˉ(q3)p(q4)\bar{p}(q_{1})p(q_{2})\to\bar{p}(q_{3})p(q_{4}) for pppp and ppˉp\bar{p} elastic scattering processes, respectively. In addition, we assume that the elastic pppp and ppˉp\bar{p} scatterings are mainly dominated by the pomeron and odderon exchanges since other mesons and Reggeons exchange contributions are negligibly small in these processes at TeV scale.

By using the standard method in QFT [92], the elastic pppp scattering amplitude of the pomeron exchange is given by

iMPpp=⟨p(q3)p(q4)∣:Texp⁡(i∫LP(x) d4x):∣p(q1)p(q2)⟩=−igP2Gμ1ν1μ1′ν1′uˉ(q3)γμ1′(q3,ν1′+q1,ν1′)u(q1)iΔPμ1ν1;μ2ν2(s,t)×Gμ2ν2μ2′ν2′uˉ(q4)γμ2′(q4,ν2′+q2,ν2′)u(q2)FP(t)2,(7)\begin{aligned} i\mathcal{M}_{P}^{pp}&=\langle p(q_{3})p(q_{4})|:\mathcal{T}\exp\left(i\int\mathcal{L}_{P}(x)\,\mathrm{d}^{4}x\right):|p(q_{1})p(q_{2})\rangle\\ &=-ig_{P}^{2}G_{\mu_{1}\nu_{1}}^{\mu_{1}'\nu_{1}'}\bar{u}(q_{3})\gamma_{\mu_{1}'}(q_{3,\nu_{1}'}+q_{1,\nu_{1}'})u(q_{1})i\Delta_{P}^{\mu_{1}\nu_{1};\mu_{2}\nu_{2}}(s,t)\\ &\quad\times G_{\mu_{2}\nu_{2}}^{\mu_{2}'\nu_{2}'}\bar{u}(q_{4})\gamma_{\mu_{2}'}(q_{4,\nu_{2}'}+q_{2,\nu_{2}'})u(q_{2})\mathcal{F}_{P}(t)^{2}, \tag*{(7)} \end{aligned}

and the elastic ppˉp\bar{p} scattering amplitude of the pomeron exchange is given by

iMPppˉ=⟨pˉ(q3)p(q4)∣:Texp⁡(i∫LP(x) d4x):∣pˉ(q1)p(q2)⟩=−igP2Gμ1ν1μ1′ν1′vˉ(q1)γμ1′(q1,ν1′+q3,ν1′)v(q3)iΔPμ1ν1;μ2ν2(s,t)×Gμ2ν2μ2′ν2′uˉ(q4)γμ2′(q4,ν2′+q2,ν2′)u(q2)FP(t)2.(8)\begin{aligned} i\mathcal{M}_{P}^{p\bar{p}}&=\langle\bar{p}(q_{3})p(q_{4})|:\mathcal{T}\exp\left(i\int\mathcal{L}_{P}(x)\,\mathrm{d}^{4}x\right):|\bar{p}(q_{1})p(q_{2})\rangle\\ &=-ig_{P}^{2}G_{\mu_{1}\nu_{1}}^{\mu_{1}'\nu_{1}'}\bar{v}(q_{1})\gamma_{\mu_{1}'}(q_{1,\nu_{1}'}+q_{3,\nu_{1}'})v(q_{3})i\Delta_{P}^{\mu_{1}\nu_{1};\mu_{2}\nu_{2}}(s,t)\\ &\quad\times G_{\mu_{2}\nu_{2}}^{\mu_{2}'\nu_{2}'}\bar{u}(q_{4})\gamma_{\mu_{2}'}(q_{4,\nu_{2}'}+q_{2,\nu_{2}'})u(q_{2})\mathcal{F}_{P}(t)^{2}. \tag*{(8)} \end{aligned}

For the propagator of the spin-2 pomeron ΔPμν;ρσ\Delta_{P}^{\mu\nu;\rho\sigma}, we employ from [48] and it takes the following form

iΔPμν,ρσ(s,t)=14s[gμρgνσ+gμσgνρ−12gμνgρσ](−iαP′s)αP(t)−1,(9)i\Delta_{P}^{\mu\nu,\rho\sigma}(s,t)=\frac{1}{4s}\left[g^{\mu\rho}g^{\nu\sigma}+g^{\mu\sigma}g^{\nu\rho}-\frac{1}{2}g^{\mu\nu}g^{\rho\sigma}\right](-i\alpha'_{P}s)^{\alpha_{P}(t)-1}, \tag*{(9)}

with the conventional linear pomeron trajectory [48, 49]

αP(t)=1+ϵP+αP′t(10)\alpha_{\mathrm{P}}(t)=1+\epsilon_{\mathrm{P}}+\alpha'_{\mathrm{P}}t \tag*{(10)}
ϵP=0.0808,andαP′=0.25 GeV−2(11)\epsilon_{\mathrm{P}=0.0808,\qquad\text{and}\qquad\alpha'_{\mathrm{P}}=0.25\ \mathrm{GeV}^{-2}} \tag*{(11)}

where the factor 1+ϵP1+\epsilon_{\mathrm{P}} and αP′\alpha'_{\mathrm{P}} represent the vertical interception and slope of the pomeron trajectory, respectively. This formulation of the spin-2 pomeron propagator is an alternative approach to studying soft high energy hadronic collisions. This formalism can be reproduced in several experimental data. In addition, the pomeron propagator with the Donnachie-Landschoff ansatz in Eq. (9) is proposed by Heidelberg group [48]. Finally, the pomeron-pp/ppˉpp/p\bar{p} coupling form factor reads,

FP(t)=(1−t4mp2μpμN)(1−t4mp2)−1(1−tmD2)−2,μN=e2mp,μpμN=2.7928,mD2=0.71 GeV2(12)\mathcal{F}_{\mathrm{P}}(t)=\left(1-\frac{t}{4m_{\mathrm{p}}^2}\frac{\mu_{\mathrm{p}}}{\mu_{\mathrm{N}}}\right)\left(1-\frac{t}{4m_{\mathrm{p}}^2}\right)^{-1}\left(1-\frac{t}{m_{\mathrm{D}}^2}\right)^{-2}, \qquad \mu_{\mathrm{N}}=\frac{e}{2m_{\mathrm{p}}},\qquad \frac{\mu_{\mathrm{p}}}{\mu_{\mathrm{N}}}=2.7928,\qquad m_{\mathrm{D}}^2=0.71\ \mathrm{GeV}^2 \tag*{(12)}

This form factor is the standard Dirac form factor of proton and it is widely used to study the pppp and ppˉp\bar{p} scattering see more details discussions and its consequences of the form factor in Eq. (12) in chapter two of Ref.[93].

We turn to consider the elastic pppp and ppˉp\bar{p} scattering amplitudes for the spin-3 odderon exchange contribution. Having used the same manner, the pppp scattering amplitude is calculated and one finds,

iMOpp=⟨p(q3)p(q4)∣:Texp⁡(i∫LO(x) d4x):∣p(q1)p(q2)⟩=−gO2M02Gμνρμ1ν1ρ1 uˉ(q3)γμ1(q3,ν1+q1,ν1)(q3,ρ1+q1,ρ1)u(q1)iΔOμνρ;μ′ν′ρ′(s,t)×Gμ′ν′ρ′μ2ν2ρ2 uˉ(q4)γμ2(q4,ν2+q2,ν2)(q4,ρ2+q2,ρ2)u(q2)FO(t)2(13)\begin{aligned} i\mathcal{M}_{\mathrm{O}}^{pp} &=\langle p(q_{3})p(q_{4})|:T\exp\left(i\int\mathcal{L}_{\mathrm{O}}(x)\,\mathrm{d}^{4}x\right):|p(q_{1})p(q_{2})\rangle\\ &=-\frac{g_{\mathrm{O}}^{2}}{M_{0}^{2}}\mathcal{G}^{\mu_{1}\nu_{1}\rho_{1}}_{\mu\nu\rho}\,\bar{u}(q_{3})\gamma_{\mu_{1}}(q_{3,\nu_{1}}+q_{1,\nu_{1}})(q_{3,\rho_{1}}+q_{1,\rho_{1}})u(q_{1})i\Delta_{\mathrm{O}}^{\mu\nu\rho;\mu'\nu'\rho'}(s,t)\\ &\qquad\times\mathcal{G}^{\mu_{2}\nu_{2}\rho_{2}}_{\mu'\nu'\rho'}\,\bar{u}(q_{4})\gamma_{\mu_{2}}(q_{4,\nu_{2}}+q_{2,\nu_{2}})(q_{4,\rho_{2}}+q_{2,\rho_{2}})u(q_{2})\mathcal{F}_{\mathrm{O}}(t)^2 \tag*{(13)} \end{aligned}

while the amplitude of the ppˉp\bar{p} scattering with the odderon exchange reads

iMOppˉ=⟨pˉ(q3)p(q4)∣:Texp⁡(i∫LO(x) d4x):∣pˉ(q1)p(q2)⟩=−gO2M02Gμνρμ1ν1ρ1 vˉ(q1)γμ1(q1,ν1+q3,ν1)(q1,ρ1+q3,ρ1)v(q3)iΔOμνρ;μ′ν′ρ′(s,t)×Gμ′ν′ρ′μ2ν2ρ2 uˉ(q4)γμ2(q4,ν2+q2,ν2)(q4,ρ2+q2,ρ2)u(q2)FO(t)2(14)\begin{aligned} i\mathcal{M}_{\mathrm{O}}^{p\bar{p}} &=\langle\bar{p}(q_{3})p(q_{4})|:T\exp\left(i\int\mathcal{L}_{\mathrm{O}}(x)\,\mathrm{d}^{4}x\right):|\bar{p}(q_{1})p(q_{2})\rangle\\ &=-\frac{g_{\mathrm{O}}^{2}}{M_{0}^{2}}\mathcal{G}^{\mu_{1}\nu_{1}\rho_{1}}_{\mu\nu\rho}\,\bar{v}(q_{1})\gamma_{\mu_{1}}(q_{1,\nu_{1}}+q_{3,\nu_{1}})(q_{1,\rho_{1}}+q_{3,\rho_{1}})v(q_{3})i\Delta_{\mathrm{O}}^{\mu\nu\rho;\mu'\nu'\rho'}(s,t)\\ &\qquad\times\mathcal{G}^{\mu_{2}\nu_{2}\rho_{2}}_{\mu'\nu'\rho'}\,\bar{u}(q_{4})\gamma_{\mu_{2}}(q_{4,\nu_{2}}+q_{2,\nu_{2}})(q_{4,\rho_{2}}+q_{2,\rho_{2}})u(q_{2})\mathcal{F}_{\mathrm{O}}(t)^2 \tag*{(14)} \end{aligned}

In addition, we have modified the proton form-factor in Eq. (12) for the odderon coupling to pppp by adding new three free parameters, AA, BB and CC as,

FO(t)=(1−At4mp2μpμN)(1−Bt4mp2)−1(1−CtmD2)−2(15)\mathcal{F}_{\mathrm{O}}(t)=\left(1-\frac{At}{4m_{\mathrm{p}}^2}\frac{\mu_{\mathrm{p}}}{\mu_{\mathrm{N}}}\right)\left(1-\frac{Bt}{4m_{\mathrm{p}}^2}\right)^{-1}\left(1-\frac{Ct}{m_{\mathrm{D}}^2}\right)^{-2} \tag*{(15)}

The ΔOμ1ν1ρ1;μ2ν2ρ2(s,t)\Delta_{\mathrm{O}}^{\mu_{1}\nu_{1}\rho_{1};\mu_{2}\nu_{2}\rho_{2}}(s,t) term is the propagator of the spin-3 particle in momentum space. By analogy to the spin-2 tensor pomeron proposed by [48], we introduce the spin-3 tensor odderon propagator with the Donnachie-Lanschoff parametrization and it reads,

iΔOμνλ,ρστ(s,t)=−iM~026s3[∑Cgμρ(gνσgλτ+gντgλσ)−12∑Cgμνgλρgστ](−iαO′s)αO(t)−1(16)i\Delta_{\mathrm{O}}^{\mu\nu\lambda,\rho\sigma\tau}(s,t)=-i\frac{\widetilde{M}_{0}^{2}}{6s^{3}}\left[\sum_{C}g^{\mu\rho}\left(g^{\nu\sigma}g^{\lambda\tau}+g^{\nu\tau}g^{\lambda\sigma}\right)-\frac{1}{2}\sum_{C}g^{\mu\nu}g^{\lambda\rho}g^{\sigma\tau}\right]\left(-i\alpha'_{\mathrm{O}}s\right)^{\alpha_{\mathrm{O}}(t)-1} \tag*{(16)}
αO(t)=1+ϵO+αO′t(17)\alpha_{\mathrm{O}}(t)=1+\epsilon_{\mathrm{O}}+\alpha'_{\mathrm{O}}t \tag*{(17)}

where ∑C\sum_{C} stands for the sum over all distinct combinations of the Lorentz indices (μνλ)(\mu\nu\lambda) and (ρστ)(\rho\sigma\tau), for instance,

∑Cgμνgλρgστ=gμνgλρgστ+gμνgλσgρτ+gμνgλτgρσ+gμλgνρgστ+gμλgνσgρτ+gμλgντgρσ+gνλgμρgστ+gνλgμσgρτ+gνλgμτgρσ(18)\sum_{C}g_{\mu\nu}g_{\lambda\rho}g_{\sigma\tau} =g_{\mu\nu}g_{\lambda\rho}g_{\sigma\tau}+g_{\mu\nu}g_{\lambda\sigma}g_{\rho\tau}+g_{\mu\nu}g_{\lambda\tau}g_{\rho\sigma}+g_{\mu\lambda}g_{\nu\rho}g_{\sigma\tau}+g_{\mu\lambda}g_{\nu\sigma}g_{\rho\tau}+g_{\mu\lambda}g_{\nu\tau}g_{\rho\sigma} +g_{\nu\lambda}g_{\mu\rho}g_{\sigma\tau}+g_{\nu\lambda}g_{\mu\sigma}g_{\rho\tau}+g_{\nu\lambda}g_{\mu\tau}g_{\rho\sigma} \tag*{(18)}

The mass parameter M0M_0 is a free parameter in this work and it is introduced in order to correct the mass dimension of the odderon propagator. In addition, we consider the parameters ϵO\epsilon_O and αO′\alpha'_O as free parameters. However, the condition ϵO<ϵP\epsilon_O < \epsilon_P is imposed due to the experimental data fact that the total cross-sections of both pppp and ppˉp\bar{p} are identical at very high energies. In the other words, the pomeron exchange contributions for elastic pppp and ppˉp\bar{p} scattering at the Regge limit always dominate over the odderon ones. Moreover, the tensor structure of the spin-3 odderon propagator has been constructed in Ref. [94] (see more detailed derivations and discussions). We close this section by giving the definitions of the four-momentum conservation, the on-shell mass of the particles, and the Mandelstam variables as

q1+q2=q3+q4,q12=q22=q32=q42=mp2,s+t+u=4mp2,q_1+q_2=q_3+q_4,\qquad q_1^2=q_2^2=q_3^2=q_4^2=m_p^2,\qquad s+t+u=4m_p^2,
s=(q1+q2)2=(q3+q4)2,t=(q3−q1)2=(q4−q2)2,u=(q4−q1)2=(q3−q2)2.(19)s=(q_1+q_2)^2=(q_3+q_4)^2,\qquad t=(q_3-q_1)^2=(q_4-q_2)^2,\qquad u=(q_4-q_1)^2=(q_3-q_2)^2. \tag*{(19)}

In the next section, we will provide analytical expressions of the differential cross-section for the pppp and ppˉp\bar{p} elastic scattering with pomeron and odderon exchanges and fit the model parameters with the experimental data of the pppp and ppˉp\bar{p} elastic scatterings at TeV scale.

Results and Discussions

Differential cross section formulae

In this subsection, we will provide the analytical formulae of the differential cross section with respect to the tt variable and then the model parameters in the present work will be determined by fitting with all available experimental data of the pppp and ppˉp\bar{p} elastic scatterings at TeV level. The differential cross-section of the pppp and ppˉp\bar{p} scattering is given by [49],

dσpp/ppˉdt=116πs(s−mp2)14∑spin∣Mpp/ppˉ∣2.(20)\frac{d\sigma^{pp/p\bar{p}}}{dt}=\frac{1}{16\pi s\left(s-m_p^2\right)}\frac{1}{4}\sum_{\mathrm{spin}}\left|\mathcal{M}^{pp/p\bar{p}}\right|^2. \tag*{(20)}

First of all, let us briefly discuss the definitions of the scattering amplitudes of pppp and ppˉp\bar{p} in the pomeron and odderon exchange picture. Considering the amplitude Mab(s,t)\mathcal{M}^{ab}(s,t) of an elastic scattering for the a+b→a+ba+b\to a+b process in ss channel. While the corresponding elastic scattering by crossing to the uu-channel as a+bˉ→a+bˉa+\bar{b}\to a+\bar{b} with the amplitude Mabˉ(u,t)\mathcal{M}^{a\bar{b}}(u,t). According to the crossing symmetry of the scattering amplitudes, they are symmetric under interchange between the Mandelstam variables, ss and uu as

Mabˉ(s,t,u)=Mab(u,t,s).(21)\mathcal{M}^{a\bar{b}}(s,t,u)=\mathcal{M}^{ab}(u,t,s). \tag*{(21)}

Moreover, the amplitude M±\mathcal{M}_{\pm} are defined from Mab/aˉb\mathcal{M}^{ab/\bar{a}b} as

M±(s,t)=12(Mab(s,t)±Mabˉ(s,t)).(22)\mathcal{M}_{\pm}(s,t)=\frac{1}{2}\left(\mathcal{M}^{ab}(s,t)\pm\mathcal{M}^{a\bar{b}}(s,t)\right). \tag*{(22)}

Interchange s→us\to u, one finds that the amplitude M+\mathcal{M}_{+} is invariant under the crossing symmetry whereas the amplitude M−\mathcal{M}_{-} changes the relative sign. We therefore call M+\mathcal{M}_{+} and M−\mathcal{M}_{-} as even and odd under the crossing symmetry. As a result, one observes that the M+\mathcal{M}_{+} and M−\mathcal{M}_{-} correspond to the even and odd under charge conjugation, respectively. Since the interchange s→us\to u is equivalent to charge conjugation transformation (C) i.e., changing particle-particle scattering to particle-antiparticle scattering. We therefore identify M+\mathcal{M}_{+} and M−\mathcal{M}_{-} as pomeron (MP\mathcal{M}_{P} with C=+1C=+1) and odderon (MO\mathcal{M}_{O} with C=−1C=-1) exchange amplitudes, respectively.

As discussed above, we can define the total amplitude of the elastic pppp and ppˉp\bar{p} processes with pomeron and odderon exchange diagrams at the tree level as follow

Mpp=MPpp−MOpp,(23)\mathcal{M}^{pp}=\mathcal{M}^{pp}_{P}-\mathcal{M}^{pp}_{O}, \tag*{(23)}
Mppˉ=MPppˉ+MOppˉ.(24)\mathcal{M}^{p\bar{p}}=\mathcal{M}^{p\bar{p}}_{P}+\mathcal{M}^{p\bar{p}}_{O}. \tag*{(24)}

The absolute square amplitudes of the Mpp\mathcal{M}^{pp} and Mppˉ\mathcal{M}^{p\bar{p}} with an average sum over unpolarized initial spin states of the incoming particles are given by

∑spin∣Mpp∣2=∑spin(∣MPpp∣2−∣MPppMOpp∗∣−∣MPpp∗MOpp∣+∣MOpp∣2).(25)\sum_{\mathrm{spin}}\left|\mathcal{M}^{pp}\right|^2=\sum_{\mathrm{spin}}\left(\left|\mathcal{M}^{pp}_{P}\right|^2-\left|\mathcal{M}^{pp}_{P}\mathcal{M}^{pp*}_{O}\right|-\left|\mathcal{M}^{pp*}_{P}\mathcal{M}^{pp}_{O}\right|+\left|\mathcal{M}^{pp}_{O}\right|^2\right). \tag*{(25)}
∑spin∣Mppˉ∣2=∑spin(∣MPppˉ∣2+∣MPppˉMOppˉ∗∣+∣MPppˉMOppˉ∗∣+∣MOppˉ∣2).(26)\sum_{\mathrm{spin}} \left|\mathcal{M}^{p\bar p}\right|^2 = \sum_{\mathrm{spin}}\left(\left|\mathcal{M}^{p\bar p}_{P}\right|^2 +\left|\mathcal{M}^{p\bar p}_{P}\mathcal{M}^{p\bar p *}_{O}\right| +\left|\mathcal{M}^{p\bar p}_{P}\mathcal{M}^{p\bar p *}_{O}\right| +\left|\mathcal{M}^{p\bar p}_{O}\right|^2\right). \tag*{(26)}

The explicit forms of the absolute square amplitudes with an average sum over initial states can be calculated in the following forms,

\begin{aligned} \sum_{\mathrm{spin}}\left|\mathcal{M}^{pp}_{P}\right|^2 ={}&\operatorname{Tr}\left[ (\slashed{q}_3+m_p)\gamma_{\mu_1}(q_3,\nu'_1+q_1,\nu'_1) (\slashed{q}_1+m_p)\gamma_{\mu_3}(q_3,\nu'_3+q_1,\nu'_3) \right.\\ &\left.\quad\times (\slashed{q}_4+m_p)\gamma_{\mu'_2}(q_4,\nu'_2+q_2,\nu'_2) (\slashed{q}_2+m_p)\gamma_{\mu'_4}(q_4,\nu'_4+q_2,\nu'_4) \right]\\ &\quad\times g_P^4 g_{\mu_1\nu_1}^{\mu'_1\nu'_1}g_{\mu_3\nu_3}^{\mu'_3\nu'_3} g_{\mu_2\nu_2}^{\mu'_2\nu'_2}g_{\mu_4\nu_4}^{\mu'_4\nu'_4} \Delta_P^{\mu_1\nu_1;\mu_2\nu_2}(s,t) \Delta_P^{\mu_3\nu_3;\mu_4\nu_4}(s,t)\mathcal{F}_P(t)^4\\ \approx{}& \frac{16g_P^4\left[1-\frac{t}{4m_p^2}\frac{\mu_P}{\mu_N}\right]^4} {\left[1-\frac{t}{m_p^2}\right]^4\left[1-\frac{t}{m_D^2}\right]^8} s^2(\alpha'_P s)^{2\epsilon_P+2\alpha'_P t}, \tag*{(27)} \end{aligned}
\begin{aligned} \sum_{\mathrm{spin}}\left|\mathcal{M}^{pp}_{P}\right|^2 ={}&\operatorname{Tr}\left[ (\slashed{q}_3+m_p)\gamma_{\mu_1}(q_3,\nu'_1+q_1,\nu'_1)(q_3,\rho_1+q_1,\rho_1) (\slashed{q}_1+m_p)\gamma_{\mu_3}(q_3,\bar\nu_3+q_1,\bar\nu_3)(q_3,\bar\rho_3+q_1,\bar\rho_3) \right.\\ &\left.\quad\times (\slashed{q}_4+m_p)\gamma_{\mu_2}(q_4,\nu_2+q_2,\nu_2)(q_4,\rho_2+q_2,\rho_2) (\slashed{q}_2+m_p)\gamma_{\mu_4}(q_4,\bar\nu_4+q_2,\bar\nu_4)(q_4,\bar\rho_4+q_2,\bar\rho_4) \right]\\ &\quad\times\frac{g_O^4}{M_0^4} G_{\mu\nu}^{\mu_1\nu_1}G_{\bar\mu\bar\nu}^{\bar\nu_3\bar\rho_3} G_{\mu'\nu'}^{\mu_2\nu_2}G_{\bar\mu'\bar\nu'}^{\bar\mu_4\bar\nu_4} \Delta_O^{\mu\nu;\mu'\nu'}(s,t) \Delta_O^{\bar\mu\bar\nu;\bar\mu'\bar\nu'}(s,t)\mathcal{F}_O(t)^4\\ \approx{}& \frac{64g_O^4\left[1-\frac{A t}{4m_p^2}\frac{\mu_P}{\mu_N}\right]^4} {9\left[1-\frac{B t}{m_p^2}\right]^4\left[1-\frac{C t}{m_D^2}\right]^8} s^2(\alpha'_O s)^{2\epsilon_O+2\alpha'_O t}, \tag*{(28)} \end{aligned}
\begin{aligned} \sum_{\mathrm{spin}}\left|\mathcal{M}^{pp}_{O}\mathcal{M}^{pp *}_{O}\right| ={}&\operatorname{Tr}\left[ (\slashed{q}_3+m_p)\gamma_{\mu'_1}(q_3,\nu'_1+q_1,\nu'_1) (\slashed{q}_1+m_p)\gamma_{\bar\mu_3}(q_3,\bar\nu_3+q_1,\bar\nu_3)(q_3,\bar\rho_3+q_1,\bar\rho_3) \right.\\ &\left.\quad\times (\slashed{q}_4+m_p)\gamma_{\mu'_2}(q_4,\nu'_2+q_2,\nu'_2) (\slashed{q}_2+m_p)\gamma_{\bar\mu_4}(q_4,\bar\nu_4+q_2,\bar\nu_4)(q_4,\bar\rho_4+q_2,\bar\rho_4) \right]\\ &\quad\times\frac{g_P^2g_O^2}{M_0^2} G_{\mu'_1\nu'_1}^{\mu'_1\nu'_1} G_{\bar\mu_3\bar\nu_3}^{\bar\mu_3\bar\nu_3} G_{\mu'_2\nu'_2}^{\mu'_2\nu'_2} G_{\bar\mu_4\bar\nu_4}^{\bar\mu_4\bar\nu_4} \Delta_P^{\mu_1\nu_1;\mu_2\nu_2}(s,t) \Delta_O^{\bar\mu\bar\nu;\bar\mu'\bar\nu'}(s,t) \mathcal{F}_P(t)^2\mathcal{F}_O(t)^2\\ \approx{}& \frac{32g_P^2g_O^2\left[1-\frac{t}{4m_p^2}\frac{\mu_P}{\mu_N}\right]^2 \left[1-\frac{A t}{4m_p^2}\frac{\mu_P}{\mu_N}\right]^2} {9\left[1-\frac{t}{m_p^2}\right]^2 \left[1-\frac{t}{m_D^2}\right]^4 \left[1-\frac{B t}{m_p^2}\right]^2 \left[1-\frac{C t}{m_D^2}\right]^4} s^2(-i\alpha'_P s)^{\epsilon_P+\alpha'_P t} (i\alpha'_O s)^{\epsilon_O+\alpha'_O t}, \tag*{(29)} \end{aligned}
\begin{aligned} \sum_{\mathrm{spin}}\left|\mathcal{M}^{pp}_{P}\right|^2 ={}&\operatorname{Tr}\left[ (\slashed{q}_1-m_p)\gamma_{\mu'_1}(q_3,\nu'_1+q_1,\nu'_1) (\slashed{q}_3-m_p)\gamma_{\mu'_3}(q_3,\nu'_3+q_1,\nu'_3) \right.\\ &\left.\quad\times (\slashed{q}_4+m_p)\gamma_{\mu'_2}(q_4,\nu'_2+q_2,\nu'_2) (\slashed{q}_2+m_p)\gamma_{\mu'_4}(q_4,\nu'_4+q_2,\nu'_4) \right]\\ &\quad\times g_P^4g_{\mu_1\nu_1}^{\mu'_1\nu'_1}g_{\mu_3\nu_3}^{\mu'_3\nu'_3} g_{\mu_2\nu_2}^{\mu'_2\nu'_2}g_{\mu_4\nu_4}^{\mu'_4\nu'_4} \Delta_P^{\mu_1\nu_1;\mu_2\nu_2}(s,t) \Delta_P^{\mu_3\nu_3;\mu_4\nu_4}(s,t)\mathcal{F}_P(t)^4\\ \approx{}& \frac{16g_P^4\left[1-\frac{t}{4m_p^2}\frac{\mu_P}{\mu_N}\right]^4} {\left[1-\frac{t}{m_p^2}\right]^4\left[1-\frac{t}{m_D^2}\right]^8} s^2(\alpha'_P s)^{2\epsilon_P+2\alpha'_P t}, \tag*{(30)} \end{aligned}
\begin{aligned} \sum_{\mathrm{spin}}\left|\mathcal{M}^{pp}_{O}\right|^2 ={}&\operatorname{Tr}\left[ (\slashed{q}_1-m_p)\gamma_{\mu_1}(q_3,\nu'_1+q_1,\nu'_1)(q_3,\rho_1+q_1,\rho_1) (\slashed{q}_3-m_p)\gamma_{\mu_3}(q_3,\bar\nu_3+q_1,\bar\nu_3)(q_3,\bar\rho_3+q_1,\bar\rho_3) \right.\\ &\left.\quad\times (\slashed{q}_4+m_p)\gamma_{\mu_2}(q_4,\nu_2+q_2,\nu_2)(q_4,\rho_2+q_2,\rho_2) (\slashed{q}_2+m_p)\gamma_{\mu_4}(q_4,\bar\nu_4+q_2,\bar\nu_4)(q_4,\bar\rho_4+q_2,\bar\rho_4) \right]\\ &\quad\times\frac{g_O^4}{M_0^4} G_{\mu\nu}^{\mu_1\nu_1}G_{\bar\mu\bar\nu}^{\bar\mu_3\bar\nu_3} G_{\mu'\nu'}^{\mu_2\nu_2}G_{\bar\mu'\bar\nu'}^{\bar\mu_4\bar\nu_4} \Delta_O^{\mu\nu;\mu'\nu'}(s,t)\Delta_O^{\bar\mu\bar\nu;\bar\mu'\bar\nu'}(s,t)\mathcal{F}_O(t)^4 \tag*{(31)} \end{aligned}
≈64gO4[1−At4mp2μpμN]49[1−Btmp2]4[1−CtmD2]8s2(αO′s)2ϵO+2αO′t,(32)\approx\frac{64g_{\mathrm{O}}^{4}\left[1-\frac{At}{4m_{p}^{2}}\frac{\mu_{p}}{\mu_{N}}\right]^{4}}{9\left[1-\frac{Bt}{m_{p}^{2}}\right]^{4}\left[1-\frac{Ct}{m_{D}^{2}}\right]^{8}}s^{2}\left(\alpha'_{\mathrm{O}}s\right)^{2\epsilon_{\mathrm{O}}+2\alpha'_{\mathrm{O}}t}, \tag*{(32)}
\begin{aligned} \sum_{\mathrm{spin}}\left|\mathcal{M}_{\mathrm{P}}^{p\bar p}\mathcal{M}_{\mathrm{O}}^{p\bar p*}\right|^{2} ={}&\operatorname{Tr}\left[(\slashed{q}_{1}-m_{p})\gamma_{\mu_{1}'}(q_{3,\nu_{1}'}+q_{1,\nu_{1}'}) (q_{3}-m_{p})\gamma_{\bar{\mu}_{3}}(q_{3,\bar{\nu}_{3}}+q_{1,\bar{\nu}_{3}})(q_{3,\bar{\rho}_{3}}+q_{1,\bar{\rho}_{3}})\right.\\ &\left.\times(q_{4}+m_{p})\gamma_{\mu_{2}'}(q_{4,\nu_{2}'}+q_{2,\nu_{2}'})(q_{2}+m_{p})\gamma_{\bar{\mu}_{4}}(q_{4,\bar{\nu}_{4}}+q_{2,\bar{\nu}_{4}})(q_{4,\bar{\rho}_{4}}+q_{2,\bar{\rho}_{4}})\right]\\ &\times\frac{g_{\mathrm{P}}^{2}g_{\mathrm{O}}^{2}}{M_{0}^{2}}G^{\mu_{1}'\nu_{1}'}_{\mu\bar{\mu}}G^{\bar{\mu}_{3}\bar{\nu}_{3}\bar{\rho}_{3}}_{\mu_{1}'\bar{\nu}_{1}'\bar{\rho}_{1}'}G^{\mu_{2}'\nu_{2}'}_{\mu\bar{\mu}}G^{\bar{\mu}_{4}\bar{\nu}_{4}\bar{\rho}_{4}}_{\mu_{2}'\bar{\nu}_{2}'\bar{\rho}_{2}'}\Delta_{\mathrm{P}}^{\mu_{1}\nu_{1};\mu_{2}\nu_{2}}(s,t)\Delta_{\mathrm{O}}^{\bar{\mu}\bar{\nu}\bar{\rho};\bar{\mu}'\bar{\nu}'\bar{\rho}'}(s,t)\mathcal{F}_{\mathrm{P}}(t)^{2}\mathcal{F}_{\mathrm{O}}(t)^{2}\\ \approx{}&\frac{32g_{\mathrm{P}}^{2}g_{\mathrm{O}}^{2}\left[1-\frac{t}{4m_{p}^{2}}\frac{\mu_{p}}{\mu_{N}}\right]^{2}\left[1-\frac{At}{4m_{p}^{2}}\frac{\mu_{p}}{\mu_{N}}\right]^{2}}{9\left[1-\frac{t}{m_{p}^{2}}\right]^{2}\left[1-\frac{t}{m_{D}^{2}}\right]^{4}\left[1-\frac{Bt}{m_{p}^{2}}\right]^{2}\left[1-\frac{Ct}{m_{D}^{2}}\right]^{4}}s^{2}\left(-i\alpha'_{\mathrm{P}}s\right)^{\epsilon_{\mathrm{P}}+\alpha'_{\mathrm{P}}t}\left(i\alpha'_{\mathrm{O}}s\right)^{\epsilon_{\mathrm{O}}+\alpha'_{\mathrm{O}}t}, \tag*{(33)} \end{aligned}

where the Regge limit, s≫ts \gg t, mp2m_{p}^{2} has been applied for the approximations to obtain the final results. Here we have used the following normalizations and sum over spin of the spinors as

\bar{u}_{r}(p,m_{p})u_{s}(p,m_{p})=2m_{p}\delta_{rs},\qquad\sum_{r}u_{r}(p,m_{p})\bar{u}_{r}(p,m_{p})=\slashed{p}+m_{p},
\bar{v}_{r}(p,m_{p})v_{s}(p,m_{p})=-2m_{p}\delta_{rs},\qquad\sum_{r}v_{r}(p,m_{p})\bar{v}_{r}(p,m_{p})=\slashed{p}-m_{p}, \tag*{(34)}

where rr and ss are spin indices of the spinors. As results, we note that ∑spin∣MXpp∣2=∑spin∣MXppˉ∣2\sum_{\mathrm{spin}}\left|\mathcal{M}_{\mathrm{X}}^{pp}\right|^{2}=\sum_{\mathrm{spin}}\left|\mathcal{M}_{\mathrm{X}}^{p\bar{p}}\right|^{2} with X=P,OX=\mathrm{P},\mathrm{O}.

Next, we will present the scalar amplitudes of the pomeron and odderon exchanges as AP\mathcal{A}_{\mathrm{P}} and AO\mathcal{A}_{\mathrm{O}} respectively. Having use the results in Eqs. (27) and (28), they are written by,

AP(s,t)=4gP2[1−t4mp2μpμN]2[1−tmp2]2[1−tmD2]4s(−iαP′s)ϵP+αP′t,(35)\mathcal{A}_{\mathrm{P}}(s,t)=\frac{4g_{\mathrm{P}}^{2}\left[1-\frac{t}{4m_{p}^{2}}\frac{\mu_{p}}{\mu_{N}}\right]^{2}}{\left[1-\frac{t}{m_{p}^{2}}\right]^{2}\left[1-\frac{t}{m_{D}^{2}}\right]^{4}}s\left(-i\alpha'_{\mathrm{P}}s\right)^{\epsilon_{\mathrm{P}}+\alpha'_{\mathrm{P}}t}, \tag*{(35)}
AO(s,t)=8gO2[1−At4mp2μpμN]23[1−Btmp2]2[1−CtmD2]4s(−iαO′s)ϵO+αO′t,(36)\mathcal{A}_{\mathrm{O}}(s,t)=\frac{8g_{\mathrm{O}}^{2}\left[1-\frac{At}{4m_{p}^{2}}\frac{\mu_{p}}{\mu_{N}}\right]^{2}}{3\left[1-\frac{Bt}{m_{p}^{2}}\right]^{2}\left[1-\frac{Ct}{m_{D}^{2}}\right]^{4}}s\left(-i\alpha'_{\mathrm{O}}s\right)^{\epsilon_{\mathrm{O}}+\alpha'_{\mathrm{O}}t}, \tag*{(36)}

where the amplitude of the pppp scattering is given by App(s,t)=AP(s,t)−AO(s,t)\mathcal{A}^{pp}(s,t)=\mathcal{A}_{\mathrm{P}}(s,t)-\mathcal{A}_{\mathrm{O}}(s,t). According to the optical theorem, one can write the total cross-section formula as,

σtotpp=1sIm⁡App(s,0)=1sIm⁡[AP(s,0)−AO(s,0)]=Im⁡[4gP2(−iαP′s)ϵP−83gO2(−iαO′s)ϵO].(37)\sigma_{\mathrm{tot}}^{pp}=\frac{1}{s}\operatorname{Im}\mathcal{A}^{pp}(s,0)=\frac{1}{s}\operatorname{Im}\left[\mathcal{A}_{\mathrm{P}}(s,0)-\mathcal{A}_{\mathrm{O}}(s,0)\right]=\operatorname{Im}\left[4g_{\mathrm{P}}^{2}\left(-i\alpha'_{\mathrm{P}}s\right)^{\epsilon_{\mathrm{P}}}-\frac{8}{3}g_{\mathrm{O}}^{2}\left(-i\alpha'_{\mathrm{O}}s\right)^{\epsilon_{\mathrm{O}}}\right]. \tag*{(37)}

In the following subsection, we will use the total cross-section in Eq. (37) to compare with the data from the TOTEM collaboration at the TeV scale after the model parameters are chosen.

Parameters fitting and discussion

In this section, we perform curve fitting of the model parameters with experimental data. As mentioned in the section II, we have six free parameters i.e., the odderon-pppp coupling constant gOg_{\mathrm{O}}, ϵO\epsilon_{\mathrm{O}}, αO′\alpha'_{\mathrm{O}} and the modified odderon-pppp form-factor parameters AA, BB and CC. The observed value for differential cross sections of pppp and ppˉp\bar{p} scattering, dσobs/dtd\sigma^{\mathrm{obs}}/dt, come from various experiments with the center of mass energy ranging from 1.8 TeV [95] for ppˉp\bar{p}, 1.96 TeV [81] for pppp and [80] for ppˉp\bar{p}, 2.76 TeV [96] with 13σ13\sigma and 4.3σ4.3\sigma, 7 TeV [97] and 13 TeV [98] for pppp scattering. Since we are interested in the small −t-t limit, we only use the observed data with the linear relation between the differential cross-section and the momentum exchange because the effect of the pomeron and odderon is highly manifested in the linear regime of the differential cross-section. We define χ2\chi^2 function as

χ2(αi)=∑j=1N(dσdt(αi)jmodel−dσdtjobsdσdtjobs)2(38)\chi^2(\alpha_i) = \sum_{j=1}^{N} \left(\frac{\frac{d\sigma}{dt}(\alpha_i)_{j}^{\mathrm{model}}-\frac{d\sigma}{dt}_{j}^{\mathrm{obs}}}{\frac{d\sigma}{dt}_{j}^{\mathrm{obs}}}\right)^2 \tag*{(38)}

where αi\alpha_i are six parameters (gOg_{\mathrm{O}}, ϵO\epsilon_{\mathrm{O}}, αO′\alpha'_{\mathrm{O}}, AA, BB, CC) and j=1,…,Nj = 1,\ldots,N is the index of the data points associated with the momentum exchange, −t-t. In order to obtain the best fit parameters, we minimize χ2\chi^2 functions using iminuit[99, 100]. The results are shown in the table I. Note that the errors are calculated using the Hessian matrix where more details will be provided in Appendix § 3.1.

The central values of gOg_{\mathrm{O}}, αO′\alpha'_{\mathrm{O}}, ϵO\epsilon_{\mathrm{O}}, AA, BB and CC are consistent among various data sets. The odderon mass which can be calculated from its Regge trajectory at the pole, t=mO2t=m_{\mathrm{O}}^2 with αO(mO2)=J=3\alpha_{\mathrm{O}}(m_{\mathrm{O}}^2)=J=3 as

mO(J=3)=J−1−ϵOαO′∣J=3,(39)m_{\mathrm{O}}(J=3)=\left.\sqrt{\frac{J-1-\epsilon_{\mathrm{O}}}{\alpha'_{\mathrm{O}}}}\right|_{J=3}, \tag*{(39)}

are consistent due to its dependency on ϵO\epsilon_{\mathrm{O}} and αO′\alpha'_{\mathrm{O}}.

The quality of parameter fitting can be determined using the minimized χ2\chi^2 per degree of freedom. Among the available data sets, the best parameters with sufficient statistics come from the pppp-TOTEM 13 TeV data. However, using this particular data alone leads to an overfitting problem, i.e., these parameters lead to unsatisfying fits with ppˉp\bar{p} data sets. We, therefore, take a more global analysis using the combined χ2\chi^2 function of all available data sets. We then use the parameter fitting from the combined data set as the representation of our model. The model differential cross sections comparing with experimental data are shown in Figure 1. We have provided the statistical error analysis in detail in Appendix § 3.1.

Image not extracted

Figure 1. The best fit plots of the differential cross-section of the pppp (blue plots) and pˉp\bar{p}p (green plots) vs the model results with the parameters from the combined data set fit (see the last column in Table.I).

By using the Eq. (39) with the best-fit parameters, ϵO\epsilon_{\mathrm{O}} and αO′\alpha'_{\mathrm{O}} from the combined data set, one can determine the masses of the odderons with J=3J=3, 55, 77 as shown below,

mO(JPC=3−−)=3.201±0.609 GeV,mO(JPC=5−−)=4.563±0.868 GeV,mO(JPC=7−−)=5.603±1.066 GeV.(40)\begin{aligned} m_{\mathrm{O}}(J^{PC}=3^{--}) &= 3.201 \pm0.609\ \mathrm{GeV},\\ m_{\mathrm{O}}(J^{PC}=5^{--}) &= 4.563 \pm0.868\ \mathrm{GeV},\\ m_{\mathrm{O}}(J^{PC}=7^{--}) &= 5.603 \pm1.066\ \mathrm{GeV}. \tag*{(40)} \end{aligned}

We note that the lowest mass (pole position) of the tensor odderon with spin-3 is around 3 GeV. In addition, the Chew-Frautschi plot of the odderon Regge trajectory is depicted in Figure 2. The odderon mass results in the present work are consistent with Ref. [52]. In that work, the authors considered the odderons as the oddballs in the double pole Regge model with spin-3, 5, and 7. Then, the masses of the odderon are extracted from the experimental data. According to the literature review, we found that the theoretical estimation such as SU(3) lattice QCD for isotropic and anisotropic cases [24, 25, 45], Wilson loop approach [30], vacuum correlation method in QCD [28, 29], QCD sum rules [82, 83] [21], relativistic many body framework [46] give the ranges of odderon masses as 3.5–4.5 GeV for spin-3, 5.0–5.5 GeV for spin-5, and 6.0–6.5 GeV for spin-7. We note that the theoretical model estimations in the literature of the odderon masses are a bit heavier than the mass estimations from the data in this work by about 0.3 GeV. However, the spin-3 odderon mass from the double pole Regge model gives mODP=3.001m_{\mathrm{O}}^{DP}=3.001 GeV [52] which is lighter than our work. From the results in Table I, the odderon trajectory slope, αO′=0.189 GeV−2\alpha'_{\mathrm{O}}=0.189\ \mathrm{GeV}^{-2}, and the pomeron slope, αP′=0.25 GeV−2\alpha'_{\mathrm{P}}=0.25\ \mathrm{GeV}^{-2}, coming from Donnachie-Landshoff fit [13, 14] are compatible with approximation αO′≈αP′\alpha'_{\mathrm{O}}\approx\alpha'_{\mathrm{P}}.

Image not extracted

Figure 2. The Chew-Frautschi plot of the odderon Regge trajectory is depicted by using the αO′=0.189 GeV−2\alpha'_{\mathrm{O}} = 0.189\ \mathrm{GeV}^{-2} and ϵO=0.062\epsilon_{\mathrm{O}} = 0.062 from combined data set fitting in Table I. The masses of the odderon with spin-3, 5, and 7 including the error of the fitting parameter estimations with red dots are given by Eq. (40).

On the other hand, we obtain ϵO=0.0620≪1\epsilon_{\mathrm{O}}=0.0620\ll1. As a result, the best-fit value of αO′\alpha'_{\mathrm{O}} and ϵO\epsilon_{\mathrm{O}} in this work correspond to the assumptions in Ref. [48] that αO′≈αP′\alpha'_{\mathrm{O}}\approx\alpha'_{\mathrm{P}} and ϵO≤ϵP (=0.0808)\epsilon_{\mathrm{O}}\leq\epsilon_{\mathrm{P}\ }(=0.0808) which use for fixing the parameters αO′\alpha'_{\mathrm{O}} and ϵO\epsilon_{\mathrm{O}} due to the lack of data used to constrain at that moment.

We close this section by considering the total cross-section of the pppp in our model. The relevant parameters from the combined data set are substituted to the total cross-section formula in Eq. (37). Then, we plot the total cross section as a function of c.m. energy (s\sqrt{s}) as shown in Figure 3 and we found that our model of the odderon as Regge oddball spin-3 is compatible with the TOTEM data for pppp scattering at the TeV regime. In particular, the extrapolation of TOTEM data for the pppp total cross-section at 1.96 TeV is also laid within the error band of our model. Using the best fit parameters ϵP=0.0808\epsilon_{\mathrm{P}}=0.0808 and ϵO=0.0620\epsilon_{\mathrm{O}}=0.0620, in addition, the total cross-section of the present work in Eq. (37) has been checked numerically and it also corresponds to a Froissart bound, i.e., σtotpp≤(ln⁡s)2\sigma_{\mathrm{tot}}^{pp}\leq(\ln s)^2 at s→∞s\to\infty limit.

Image not extracted

Figure 3. The total cross-section of the pppp scattering with pomeron spin-2 and odderon spin-3 exchanges in Eq. (37) is plotted by using the averaged fitted parameters from the combined data set in Table I. As a result, our model is compatible with the TOTEM pppp results at the TeV region. Moreover, the model also agrees with the extrapolation of the TOTEM data at 1.96 TeV for the pppp scattering. All data points are laid on the model prediction band.

Conclusions

In this work, we considered the odderon as the Regge oddball spin-3. The existence of odderon can be observed in a study of the difference between pppp and ppˉp\bar{p} elastic scattering. We, therefore, investigate pppp and ppˉp\bar{p} scattering at the Regge limit by including the pomeron and odderon exchanges in the present work. The effective Lagrangians of the processes are constructed and the standard perturbative QFT method is used to calculate the relevant observables in this work. The pomeron and odderon are identified as the Regge tensor glueballs and oddballs with spin-2 and -3, respectively. We have employed the Donnachie-Landshoff ansatz for the pomeron propagator and the Regge trajectory with the electromagnetic type of the pomeron-pppp form-factor. Furthermore, we also modified the electromagnetic type of the odderon-pppp by introducing the additional 3 free parameters. There are six free parameters of the model in this work (gOg_{\mathrm{O}}, ϵO\epsilon_{\mathrm{O}}, αO′\alpha'_{\mathrm{O}}, AA, BB, CC). Having performed a careful statistical analysis, all free parameters have been fixed by fitting with all combined data of the pppp and ppˉp\bar{p} differential cross-sections at the TeV regime see results in Table I. After fixing the free parameters in the present work, the masses of the odderon spin-3 and their excited states for spin-5 and -7 are estimated from its Regge trajectory by using the best fit of the combined data set. Considering the best-fit results in Table I, the odderon Regge trajectory parameters are found to be αO′=0.189±0.038\alpha'_{\mathrm{O}} = 0.189 \pm0.038 and ϵO=0.062±0.011\epsilon_{\mathrm{O}} = 0.062 \pm0.011. These results are compatible with the assumptions in Ref. [48] that used to estimate those two parameters as αO′≈α′\alpha'_{\mathrm{O}} \approx\alpha' and ϵO≤ϵP\epsilon_{\mathrm{O}} \leq\epsilon_{\mathrm{P}}. As a result, we found that the tensor odderon masses are heavier than the phenomenological approach by using the double pole Regge model extracted from the experimental data [52]. On the other hand, the odderon masses in this work are lighter than other model theoretical calculations in the literature for all odderons along their trajectory by about 0.3 GeV. Having used the best-fit parameters, the total cross-sections also agree with the TOTEM data in the TeV regime and its extrapolation from D0 of the pppp scattering at 1.96 TeV. The odderon spin-3 contribution also provided the amplitude in Regge limit that satisfied the Froissart bound. According to our findings in this work, we can conclude that the odderon could be the tensor Regge oddball spin-3 particle. Further studies to confirm our conclusion are needed to investigate other scattering processes such as polarized proton-proton scattering, meson-baryon scattering, and photoproduction. We plan to do this in the forthcoming future work.

Descriptionχ2\chi^2d.o.fgOg_{\mathrm{O}}αO′\alpha'_{\mathrm{O}}ϵO\epsilon_{\mathrm{O}}
1.80 TeV ppˉp\bar{p} [95]0.076499.346(±\pm0.563)0.188(±\pm0.040)0.055(±\pm0.009)
1.96 TeV ppˉp\bar{p} [80], pppp [81]0.5161714.927(±\pm1.754)0.158(±\pm0.036)0.064(±\pm0.018)
2.76 TeV pppp [96] (13σ\sigma)0.0361225.114(±\pm3.574)0.199(±\pm0.042)0.070(±\pm0.020)
7.00 TeV pppp [97]0.048839.888(±\pm0.468)0.200(±\pm0.047)0.060(±\pm0.006)
13.0 TeV pppp [98]0.0091509.725(±\pm0.346)0.200(±\pm0.024)0.063(±\pm0.004)
-AVERAGES-13.800(±\pm1.112)0.189(±\pm0.038)0.062(±\pm0.011)
DescriptionABCmOm_{\mathrm{O}}
1.80 TeV ppˉp\bar{p} [95]-1.340(±\pm0.193)0.242(±\pm0.262)0.436(±\pm0.117)3.213(±\pm0.611)
1.96 TeV ppˉp\bar{p} [80], pppp [81]-0.855(±\pm0.195)1.448(±\pm0.394)1.044(±\pm0.151)3.497(±\pm0.884)
2.76 TeV pppp [96] (13σ\sigma)-1.430(±\pm0.167)7.730(±\pm1.374)0.064(±\pm0.113)3.112(±\pm0.773)
7.00 TeV pppp [97]-2.388(±\pm0.265)-0.060(±\pm0.312)0.469(±\pm0.144)3.116(±\pm0.519)
13.0 TeV pppp [98]-2.342(±\pm0.145)-0.087(±\pm0.175)0.455(±\pm0.082)3.113(±\pm0.291)
Averages-1.671(±\pm0.221)1.855(±\pm-2.121)0.493(±\pm0.264)3.201(±\pm0.609)

Table 1. Summary of best fit parameters for each data set.

Acknowledgments

DS is supported by the Fundamental Fund of Khon Kaen University and DS has received funding support from the National Science, Research and Innovation Fund. The Mindanao State University - Iligan Institute of Technology is also acknowledged through its research and extension support extended to J.B. Magallanes for his travel to Khon Kaen University, Thailand. PS, CP, and DS are financially supported by the National Astronomical Research Institute of Thailand (NARIT). CP and DS are supported by Thailand NSRF via PMU-B [grant number B05F650021]. CP is also supported by Fundamental Fund 2565 of Khon Kaen University and Research Grant for New Scholar, Office of the Permanent Secretary, Ministry of Higher Education, Science, Research and Innovation under contract no. RGNS 64-043.

The authors acknowledge the National Science and Technology Development Agency, National e-Science Infrastructure Consortium, Chulalongkorn University and the Chulalongkorn Academic Advancement into Its 2nd Century Project, NSRF via the Program Management Unit for Human Resources & Institutional Development, Research and Innovation [grant numbers B05F650021, B37G660013] (Thailand) for providing computing infrastructure that has contributed to the research results reported within this paper. URL:www.e-science.in.th.

Error analysis

We provide the detail of the error estimation method in this section. The errors for parameter fitting in this analysis are handling outside the iminuit package due to inaccuracy of the results. The errors shown in Table I are therefore improved by the following. Consider the Taylor expansion around the minimum of the χ2\chi^2 function,

χ2(αi)≈χ2(αimin⁡)+12(αi−αimin⁡)2∂χ2∂αi∣αi=αimin⁡+O((αi−αimin⁡)3).(41)\chi^2(\alpha_i) \approx\chi^2(\alpha_i^{\min}) + \frac{1}{2}(\alpha_i-\alpha_i^{\min})^2 \left.\frac{\partial\chi^2}{\partial\alpha_i}\right|_{\alpha_i=\alpha_i^{\min}} + O\left((\alpha_i-\alpha_i^{\min})^3\right). \tag*{(41)}

We can approximate the error of parameter estimation, σi\sigma_i, using the width of parabolic function defined as

1σi2=12∂χ2∂αi∣αi=αimin⁡,(42)\frac{1}{\sigma_i^2} = \frac{1}{2}\left.\frac{\partial\chi^2}{\partial\alpha_i}\right|_{\alpha_i=\alpha_i^{\min}}, \tag*{(42)}

which is also the diagonal component of the Hessian matrix. The second derivative of the χ2\chi^2 function is obtained via the finite difference method

∂χ2∂αi∣αi=αimin⁡≈χ2(αimin⁡+Δαi)+χ2(αimin⁡−Δαi)−2χ2(αimin⁡)Δαi2,(43)\left.\frac{\partial\chi^2}{\partial\alpha_i}\right|_{\alpha_i=\alpha_i^{\min}} \approx\frac{\chi^2(\alpha_i^{\min}+\Delta\alpha_i)+\chi^2(\alpha_i^{\min}-\Delta\alpha_i)-2\chi^2(\alpha_i^{\min})}{\Delta\alpha_i^2}, \tag*{(43)}

where Δαi\Delta\alpha_i is chosen to be sufficiently small compared to the value of the error. The validity of the approximation is then confirmed by the comparison between the parabolic functions and the real χ2\chi^2 function shown in Figure 4. One can see that both functions agree very well within the range of error estimations.

Image not extracted

Figure 4. The comparison plots between the parabolic function using χ2=χmin⁡+(αi−αimin⁡)2σ2\chi^2 = \chi^{\min} + \frac{(\alpha_i-\alpha_i^{\min})^2}{\sigma^2} (orange line) and the real χ2\chi^2 function (blue line). The combined data set of 1.96 GeV (pppp, ppˉp\bar{p}) is used to obtain the parameter fit in these plots. Each panel represents the variation of the χ2\chi^2 function in each particular direction of the parameter space.

References

  1. [1]R. J. Eden, P. V. Landshoff, D. I. Olive, and J. C. Polkinghorne, The analytic S-matrix (Cambridge Univ. Press, Cambridge, 1966).DOI
  2. [2]V. N. Gribov, The theory of complex angular momenta: Gribov lectures on theoretical physics, Cambridge Monographs on Mathematical Physics (Cambridge University Press, 2007).
  3. [3]P. D. B. Collins, An Introduction to Regge Theory and High-Energy Physics, Cambridge Monographs on Mathematical Physics (Cambridge Univ. Press, Cambridge, UK, 2009).
  4. [4]R. J. Eden, Rept. Prog. Phys. 34, 995 (1971).
  5. [5]P. D. B. Collins, Phys. Rept. 1, 103 (1971).
  6. [6]C. B. Chiu, Ann. Rev. Nucl. Part. Sci. 22, 255 (1972).
  7. [7]A. C. Irving and R. P. Worden, Phys. Rept. 34, 117 (1977).
  8. [8]G. F. Chew and S. C. Frautschi, Phys. Rev. Lett. 7, 394 (1961).
  9. [70]S. K. Domokos, J. A. Harvey, and N. Mann, Phys. Rev. D 80, 126015 (2009), 0907.1084.
  10. [71]S. K. Domokos, J. A. Harvey, and N. Mann, Phys. Rev. D 82, 106007 (2010), 1008.2963.
  11. [72]E. Avsar, Y. Hatta, and T. Matsuo, JHEP 03, 037 (2010), 0912.3806.
  12. [73]Z. Hu, B. Maddock, and N. Mann, JHEP 08, 093 (2018), 1710.02463.
  13. [74]W. Xie, A. Watanabe, and M. Huang, JHEP 10, 053 (2019), 1901.09564.
  14. [75]P. Burikham and D. Samart, Eur. Phys. J. C 79, 452 (2019), 1902.05706.
  15. [76]Z. Liu, W. Xie, and A. Watanabe (2022), 2210.11246.
  16. [77]Z. Liu, W. Xie, F. Sun, S. Li, and A. Watanabe, Phys. Rev. D 106, 054025 (2022), 2202.08013.
  17. [78]A. Ballon-Bayona, R. Carcassés Quevedo, and M. S. Costa, JHEP 08, 085 (2017), 1704.08280.
  18. [79]I. Iatrakis, A. Ramamurti, and E. Shuryak, Phys. Rev. D 94, 045005 (2016), 1602.05014.
  19. [80]V. M. Abazov et al. (D0), Phys. Rev. D 86, 012009 (2012), 1206.0687.
  20. [81]V. M. Abazov et al. (TOTEM, D0), Phys. Rev. Lett. 127, 062003 (2021), 2012.03981.
  21. [82]H.-X. Chen, W. Chen, and S.-L. Zhu, Phys. Rev. D 104, 094050 (2021), 2107.05271.
  22. [83]H.-X. Chen, W. Chen, and S.-L. Zhu, Phys. Rev. D 103, L091503 (2021), 2103.17201.
  23. [84]L. Zhang, C. Chen, Y. Chen, and M. Huang, Phys. Rev. D 105, 026020 (2022), 2106.10748.
  24. [85]N. Bence, A. Lengyel, Z. Tarics, E. Martynov, and G. Tersimonov, Eur. Phys. J. A 57, 265 (2021).
  25. [86]E. F. Capossoli, J. P. M. Graça, and H. Boschi-Filho, Phys. Rev. D 105, 026026 (2022), 2110.12498.
  26. [87]P. Lebiedowicz, SciPost Phys. Proc. 6, 010 (2022), 2212.03720.
  27. [88]C. Baldenegro, C. Royon, and A. M. Stasto, Phys. Lett. B 830, 137141 (2022), 2204.08328.
  28. [89]Z.-F. Cui, D. Binosi, C. D. Roberts, S. M. Schmidt, and D. N. Triantafyllopoulos (2022), 2205.15438.
  29. [90]P. Lebiedowicz, O. Nachtmann, and A. Szczurek, Phys. Rev. D 106, 034023 (2022), 2206.03411.
  30. [91]C. Bonanno, M. D’Elia, B. Lucini, and D. Vadacchino, in 39th International Symposium on Lattice Field Theory (2022), 2210.07622.
  31. [92]M. E. Peskin and D. V. Schroeder, An Introduction to quantum field theory (Addison-Wesley, Reading, USA, 1995).
  32. [93]F. Close, S. Donnachie, and G. Shaw, eds., Electromagnetic interactions and hadronic structure, vol. 25 (Cambridge University Press, 2009).
  33. [94]F. A. Berends and J. C. J. M. van Reisen, Nucl. Phys. B 164, 286 (1980).
  34. [95]C. Avila et al. (E811), Phys. Lett. B 445, 419 (1999).
  35. [96]G. Antchev et al. (TOTEM), Eur. Phys. J. C 80, 91 (2020), 1812.08610.
  36. [97]G. Antchev et al. (TOTEM), EPL 101, 21002 (2013).
  37. [98]G. Antchev et al. (TOTEM), Eur. Phys. J. C 79, 103 (2019), 1712.06153.
  38. [99]H. Dembinski and P. O. et al. (2020), URL https://doi.org/10.5281/zenodo.3949207.
  39. [100]F. James and M. Roos, Comput. Phys. Commun. 10, 343 (1975).

Paper details

vlnc:W_48f2fuse
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents