betaFeedback

WorksPhysics and Astronomy

Electromagnetic radii of the nucleon in soft-wall holographic QCD

Abstract

We revisit the electromagnetic form factors of the proton and neutron in the original-minimal soft-wall holographic QCD, which has only two parameters, i.e., the mass scale κκ and the twist parameter of the nucleon ττ. We first fix τ=3τ=3 by the hard scattering rule, and extract κ=0.402 GeVκ=0.402~\text{GeV} from the world data (including the Mainz A1 data) of the Sachs magnetic form factor of the proton GMpG_M^p. We then predict among others, the charge radius of the proton to be rp=0.831 ± 0.008  fm\bf{r_{p}=0.831\,\pm\,0.008\,\,\text{fm}}, in perfect agreement with the recent charge radius of the proton measured by the PRad collaboration at Jefferson Lab, and in agreement with the muonic hydrogen experiments. Our prediction for the proton elastic form factor ratio μpGEp/GMpμ_pG_E^p/G_M^p is also in very good agreement with the recent high precision Jefferson Lab recoil polarization experiment E08-007 for Q2=0.3 − 0.6 GeV2Q^2=0.3\,-\,0.6~\text{GeV}^2, and with the recent high precision Mainz A1 experiment for Q2<0.13 GeV2Q^2< 0.13~\text{GeV}^2.

Introduction

The nucleon is a composite hadron with quarks and gluons constituents, the stuff at the origin of all hadronic matter. The quarks are charged and their distribution inside a nucleon is captured by the electric and magnetic charge radii, which measure the charge and current distributions respectively. Precision electron scattering and spectroscopic measurements show a stubborn 4% discrepancy, the so-called proton radius puzzle [1, 2, 3, 4]. This is a surprising state of affairs given the fundamental nature of the proton.

The proton structure is the quintessential QCD problem, currently addressed using ab-initio lattice simulations [5]. Empirically, the mass of the proton is known with great accuracy, but its fundamental charge radius is not, as shown in Figure 1. The high precision elastic e-p scattering measurement of 0.879 fm by Mainz A1 collaboration [6] while consistent with the value of 0.877 fm from the previous hydrogen spectroscopy [7], is larger than the reported value of 0.841 fm from muonic hydrogen spectroscopy [8], and the values of 0.833 fm and 0.848 fm from the recent hydrogen spectroscopy measurements [9, 10]. Moreover, the newest e-p measurement by the PRad collaboration at Jefferson Lab has reported a small charge radius of 0.831 fm [11]. A smaller charge radius is also supported by various re-analysis of the Mainz A1 e-p scattering data [12, 13, 14, 15, 16, 17] with few exceptions [18, 19]. In addition, the recent combined re-analysis of the Mainz A1 and PRad data has also favored a small charge radius [20, 21, 22, 23, 24].

Image not extracted

Figure 1. World data for the proton charge radius. The black dot (with its theoretical error bar) is our prediction within the original-minimal soft-wall holographic QCD [25] with τ=3\tau= 3 and κ=0.402\kappa= 0.402 GeV, see Table I. The dark-yellow dot is the prediction of Abidin and Carlson [26] (which is also within the original-minimal soft-wall holographic QCD [25] with τ=3\tau= 3 and κ=0.350\kappa= 0.350 GeV), see also Table I. The other dots are the world experimental data for the proton charge radius Pohl 2010 (μ\mu H spect.) [27], Antognini 2013 (μ\mu H spect.) [8], Beyer 2017 (H spect.) [28], CODATA 2018 [29], Bezginov 2019 (H spect.) [9], PRad 2019 (ep scatt.) [11], Grinin 2020 (H spect.) [10], Bernauer 2010 (ep scatt.) [6], Zhan 2011 (ep scatt.) [30], CODATA 2014 [31], and Fleurbaey 2018 (H spect.) [7].

All in all, tension seems to exist between e-p scattering measurements of the proton charge radius, and atomic measurements using muonic hydrogen as illustrated in Figure 1. Although ideas beyond the standard model have been suggested to fix the discrepancy [32, 33], none so far has been empirically conclusive.

The proton structure emerges from a subtle interplay between the sources composing the nucleon and the primordial glue in the vacuum [34]. Most of the proton mass arises from the primordial and topological glue at the origin of the spontaneous breaking of chiral symmetry, but its moderately large size makes it still susceptible to the effects of confinement. This last feature is also shared by the vector mesons, but not the light and more compact pseudoscalar Goldstone bosons. This last observation is further supported by the straight character of the nucleon and rho Regge trajectories, and their remarkable parallelism.

We will address the proton and neutron size problem non-perturbatively in the context of holographic QCD, which among other embodies Regge physics, and provides a field theoretical realization for the dual resonance framework [35] postulated decades ago outside the realm of QCD. It is the most economical way of enforcing basic symmetries, including crossing symmetry and unitarity, the key tenets in dispersive analyses. The approach, originates from a conjecture that observables in strongly coupled and conformal gauge theories in the limit of a large number of colors and strong gauge coupling, can be determined from classical fields interacting through gravity, in an anti-de-Sitter space in higher dimensions [36, 37]. The conjecture has been extended since to non-conformal gauge theories [25, 38].

The holographic description of the electromagnetic form factors in the context of the original-minimal soft-wall holographic QCD [25] was initially addressed in [26] (with just two parameters, the mass scale κ\kappa and twist of proton τ\tau) with rather large electromagnetic radii for κ=0.350\kappa= 0.350 GeV (fixed by simultaneously fitting the holographic mass of the rho meson and proton to their experimental values, see Table I), and τ=3\tau= 3 (fixed by the hard scattering rule). Here we show how to overcome this major shortcoming by fixing κ=0.402\kappa= 0.402 GeV (fixed by fitting the holographic Sachs magnetic form factor of the proton to the world data (including the Mainz A1 data) of the Sachs magnetic form factor of proton), see Table I, and Figure 2. We also fix κ=mρ/2=0.388\kappa= m_{\rho}/2 = 0.388 GeV, and τ=mN2/mρ2+1=2.465\tau= m_{N}^{2}/m_{\rho}^{2} + 1 = 2.465 at low energy, assuming that

Soft-wall AdS/QCDκ\kappaτ\taumρ2=4κ2m_\rho^2=4\kappa^2mN2=4κ2(τ−1)m_N^2=4\kappa^2(\tau-1)⟨rCp⟩2\sqrt{\langle r_C^p\rangle^2}
This Work0.388 GeV0.388\ \mathrm{GeV}2.4652.4650.7752 GeV20.775^2\ \mathrm{GeV}^2 (exp.)0.9382 GeV20.938^2\ \mathrm{GeV}^2 (exp.)0.839 fm0.839\ \mathrm{fm} (pr.)
This Work0.402 GeV0.402\ \mathrm{GeV}3.0003.0000.8042 GeV20.804^2\ \mathrm{GeV}^2 (pr.)1.1372 GeV21.137^2\ \mathrm{GeV}^2 (pr.)0.831 fm0.831\ \mathrm{fm} (pr.)
Abidin-Carlson [26]0.350 GeV0.350\ \mathrm{GeV}3.0003.0000.7002 GeV20.700^2\ \mathrm{GeV}^2 (pr.)0.9902 GeV20.990^2\ \mathrm{GeV}^2 (pr.)0.960 fm0.960\ \mathrm{fm} (pr.)

Table I. Comparison of the values of the holographic parameters (κ\kappa and τ\tau), and the corresponding predictions for the charge radius of the proton ⟨rCp⟩2\sqrt{\langle r_C^p\rangle^2} in our work (see (IV.14-IV.15)) within the original-minimal soft-wall holographic QCD [25], and the work of Abidin and Carlson [26] (which is also within the original-minimal soft-wall holographic QCD [25]). Note that pr. and exp. are shorthand for the predicted and experimental values, respectively.

Image not extracted

Figure 2. Sachs magnetic form factor of the proton at high and low momentum transfer. The black-solid curve is our soft-wall holographic QCD result with τ=3\tau= 3 (fixed by the hard counting rule) and κ=0.402\kappa= 0.402 GeV (extracted from the best fit to the world data (including Mainz A1 data) of Sachs magnetic form factor of proton). The green-solid curve is our soft-wall holographic QCD result with τ=mN2/mρ2+1=2.465\tau= m_N^2/m_\rho^2 + 1 = 2.465 and 4κ2=mρ2=0.77524\kappa^2 = m_\rho^2 = 0.775^2 GeV2^2. The dark-yellow-solid curve is the soft-wall holographic QCD result for τ=3\tau= 3 and κ=0.350\kappa= 0.350 GeV [26]. The solid-red curve is the Arrington fit to world data (without Mainz A1 data) [41] (brown data points). The magenta data points are the Mainz A1 data [42]. Throughout, we have used the proton magnetic moment μp=GMP(0)=F1P(0)+F2P(0)=2.793\mu_p = G_M^P(0) = F_1^P(0) + F_2^P(0) = 2.793. to PRad data [11] (orange and gray data points). The solid-red curve is the Arrington fit to world data (without PRad and Mainz A1 data) [41] (brown data points). The magenta data points are the Mainz A1 data [42]. Note that our holographic prediction (solid-black curve) completely overlaps with the PRad fit to the PRad data (solid-blue curve) in Figure 4.c for Q2<0.006Q^2 < 0.006.

the nucleon anomalous dimension or the twist τ\tau, runs with the energy scale to asymptote its hard scaling value τ=3\tau= 3 at high energy [39], see Table I. The difference between these two choices of the holographic parameters κ\kappa and τ\tau, will help us estimate our theoretical uncertainty in determining the electromagnetic radii of the nucleons.

The organization of the paper is as follows: in section II, we write down the holographic Sachs electric and magnetic form factor of the nucleon which are derived in detail, within the original-minimal holographic QCD [25, 26]. In section III, we show the three possible ways of fixing the only two parameters κ\kappa and τ\tau of the original-minimal holographic QCD, and summarize the corresponding predictions for the charge radius of the proton in Table I. Finally, in section IV, we compute the charge and magnetic radii of both the proton and neutron, and compare the holographic predictions, within the theoretical uncertainties, to the experimental values. We also summarize our results in Table II. Our conclusions are in V. More details can be found in the Appendices.

Nucleon radiiExperimental valuesSoft-wall AdS/QCD [This work]
⟨rMp⟩2\sqrt{\langle r_M^p\rangle^2}0.822 fm (e-p world average)0.803 ±\pm 0.012 fm (extracted)
⟨rCp⟩2\sqrt{\langle r_C^p\rangle^2}0.831 fm (PRad 2019)0.831 ±\pm 0.008 fm (predicted)
⟨rCp⟩2\sqrt{\langle r_C^p\rangle^2}0.841 fm (μp\mu p Lamb shift)0.831 ±\pm 0.008 fm (predicted)
⟨rMn⟩2\sqrt{\langle r_M^n\rangle^2}0.864 fm (PDG world average)0.817 ±\pm 0.007 fm (predicted)
⟨rCn⟩2\langle r_C^n\rangle^2-0.111 fm2^2 (world average)-0.094 ±\pm 0.048 fm2^2 (predicted)

Table II. Comparison of the soft-wall holographic QCD electromagnetic radii of the proton and neutron with experiment. See text for the references of the quoted experimental values.

Nucleon electric and magnetic Sachs form factors

We define the standard electric and magnetic Sachs form factors of the proton and neutron as

GEP,N(Q)=F1P,N(Q)−Q24mN2F2P,N(Q),(1)G_{E}^{P,N}(Q) = F_{1}^{P,N}(Q) - \frac{Q^{2}}{4m_{N}^{2}}F_{2}^{P,N}(Q), \tag*{(1)}
GMP,N(Q)=F1P,N(Q)+F2P,N(Q),(2)G_{M}^{P,N}(Q) = F_{1}^{P,N}(Q) + F_{2}^{P,N}(Q), \tag*{(2)}

where the Dirac F1(Q)F_{1}(Q) and Pauli F2(Q)F_{2}(Q) form factors of proton (P) and neutron (N) in the original-minimal soft-wall holographic QCD [25] are given by (see the Appendix for their detailed derivation, see also [26, 40])

F1P(Q)=(a2+τ)×B(τ,a+1)+ηP×(τ−1)×a(a(τ−1)−1)(a+τ)(a+τ+1)×B(τ−1,a+1),(3)F_{1}^{P}(Q) = \left(\frac{a}{2}+\tau\right)\times B(\tau,a+1)+\eta_{P}\times(\tau-1)\times\frac{a(a(\tau-1)-1)}{(a+\tau)(a+\tau+1)}\times B(\tau-1,a+1), \tag*{(3)}
F2P(Q)=ηP×4(τ−1)×τ×B(τ,a+1),(4)F_{2}^{P}(Q) = \eta_{P}\times4(\tau-1)\times\tau\times B(\tau,a+1), \tag*{(4)}
F1N(Q)=ηN×(τ−1)×a(a(τ−1)−1)(a+τ+1)(a+τ)×B(τ−1,a+1),(5)F_{1}^{N}(Q) = \eta_{N}\times(\tau-1)\times\frac{a(a(\tau-1)-1)}{(a+\tau+1)(a+\tau)}\times B(\tau-1,a+1), \tag*{(5)}
F2N(Q)=ηN×4(τ−1)×τ×B(τ,a+1),(6)F_{2}^{N}(Q) = \eta_{N}\times4(\tau-1)\times\tau\times B(\tau,a+1), \tag*{(6)}

Here we have defined a=Q2/4κ2a = Q^{2}/4\kappa^{2} with Q2=−q2≤0Q^{2} = -q^{2} \le0, and denoted by κ\kappa the mass scale, τ\tau the twist parameter of the nucleon, and B(x,y)=Γ(x)Γ(y)/Γ(x+y)B(x,y) = \Gamma(x)\Gamma(y)/\Gamma(x+y) the Euler beta function. The additional coefficients ηP,N=2g52(η3±η0)/2\eta_{P,N} = 2g_{5}^{2}(\eta_{3}\pm\eta_{0})/2 in Eqs. (II.3–II.6) follow from the extra couplings η0,3\eta_{0,3}, for the singlet and triplet contributions of the bulk Pauli contribution [26].

The Pauli parameter of the proton ηP\eta_{P} can be determined by matching the value of F2P(0)F_{2}^{P}(0) with the experimental data which is 1.793, i.e.

ηP=1.793C3(0)=1.7934(τ−1).(7)\eta_{P} = \frac{1.793}{C_{3}(0)} = \frac{1.793}{4(\tau-1)}. \tag*{(7)}

Similarly, the Pauli parameter of the neutron ηN\eta_{N} can be determined by matching the value of F2N(0)F_{2}^{N}(0) with the experimental data which is −1.913-1.913, i.e.

ηN=−1.913C3(0)=−1.9134(τ−1).(8)\eta_N = \frac{-1.913}{C_3(0)} = \frac{-1.913}{4(\tau-1)} . \tag*{(8)}

Fixing the mass scale κ\kappa and twist τ\tau, in the original-minimal soft-wall holographic QCD

In this work, we first fix the twist parameter of the proton τ=3\tau=3, to match the hard counting rule in the large Q2Q^2 regime (see Appendix § 9 for more details). We then fix the mass scale κ=0.402 GeV\kappa=0.402\ \mathrm{GeV} by fitting the holographic Sachs magnetic form factor GMp(Q)G_M^p(Q) ((II.2)), to the world experimental data (including the Mainz A1 data) on the Sachs magnetic form factor, see Table I, and the solid-black curves in Figure 2-Figure 4.

Image not extracted

Figure 4. Sachs electric form factor of the proton at high and low momentum transfer. The black-solid curve is our soft-wall holographic QCD result with τ=3\tau= 3 and κ=0.402\kappa= 0.402 GeV. The green-solid curve is our soft-wall holographic QCD result with τ=mN2/mρ2+1=2.465\tau= m_N^2/m_\rho^2 + 1 = 2.465 and 4κ2=mρ2=0.77524\kappa^2 = m_\rho^2 = 0.775^2 GeV2^2. The dark-yellow-solid curve is the soft-wall holographic QCD result for τ=3\tau= 3 and κ=0.350\kappa= 0.350 GeV [26]. The solid-blue curve is the PRad fit to PRad data [11] (orange and gray data points). The solid-red curve is the Arrington fit to world data (without PRad and Mainz A1 data) [41] (brown data points). The magenta data points are the Mainz A1 data [42]. Note that in (c) our holographic prediction (solid-black curve) completely overlaps with the PRad fit to PRad data (solid-blue curve).

In this work, we also fix κ\kappa, as was done in the original-minimal soft-wall holographic QCD [25], by the mass of the ρ\rho meson using the relation 4κ2=mρ24\kappa^2=m_\rho^2. Then, we fix τ\tau by the mass of the nucleon using the relation mN2=4κ2(τ−1)m_N^2=4\kappa^2(\tau-1). Therefore, for mρ=0.775 GeVm_\rho=0.775\ \mathrm{GeV} and mN=0.938m_N=0.938, we find τ=mN2/mρ2+1=2.465\tau=m_N^2/m_\rho^2+1=2.465 and κ=mρ/2=0.388 GeV\kappa=m_\rho/2=0.388\ \mathrm{GeV}, see Table I, and the solid-green curves in Figure 2-Figure 4.

Our choice is in contrast to the earlier work of Abidin and Carlson [26], within the original-minimal soft-wall holographic QCD [25], where they first fixed τ=3\tau=3 to match the hard counting rule in the large Q2Q^2 regime, and then fixed κ=0.350 GeV\kappa=0.350\ \mathrm{GeV} (by simultaneously fitting to the experimental mass of the rho meson and proton) which gives a smaller mρ=2κ=0.700 GeVm_\rho=2\kappa=0.700\ \mathrm{GeV}, and a larger mass of the nucleon mN=2κτ−1=0.990 GeVm_N=2\kappa\sqrt{\tau-1}=0.990\ \mathrm{GeV}, see Table I, and the dark-yellow-solid curves in Figure 2-Figure 4.

In Figure 2, we show the magnetic Sachs form factor of the proton. The black-solid curve is our soft-wall holographic QCD result with τ=3\tau=3 and κ=0.402 GeV\kappa=0.402\ \mathrm{GeV}. The green-solid curve is our soft-wall holographic QCD result with τ=mN2/mρ2+1=2.465\tau=m_N^2/m_\rho^2+1=2.465 and 4κ2=mρ2=0.7752 GeV24\kappa^2=m_\rho^2=0.775^2\ \mathrm{GeV}^2. The dark-yellow-solid curve is the soft-wall holographic QCD result for τ=3\tau=3 and κ=0.350 GeV\kappa=0.350\ \mathrm{GeV} [26]. The solid-red curve is the Arrington fit to world data (without the Mainz A1 data) [41] (brown data points). The magenta data points are the Mainz A1 data [42]. Throughout, we have used the proton magnetic moment μp=GMp(0)=F1p(0)+F2p(0)=2.793\mu_p=G_M^p(0)=F_1^p(0)+F_2^p(0)=2.793.

In Figure 3, we show the ratio of the Sachs electric and magnetic form factor of the proton at high and low momentum transfer. The black-solid curve is our soft-wall holographic QCD result with τ=3\tau=3 (fixed by the hard counting rule) and κ=0.402 GeV\kappa=0.402\ \mathrm{GeV} (extracted from the best fit to the world data (including Mainz A1 data) of Sachs magnetic form factor of proton). The green-solid curve is our soft-wall holographic QCD result with τ=mN2/mρ2+1=2.465\tau=m_N^2/m_\rho^2+1=2.465 and 4κ2=mρ2=0.7752 GeV24\kappa^2=m_\rho^2=0.775^2\ \mathrm{GeV}^2. The dark-yellow-solid curve is the soft-wall holographic QCD result for τ=3\tau=3 and κ=0.350 GeV\kappa=0.350\ \mathrm{GeV} [26]. The solid-red curve is the Arrington fit [41] to the world data (brown data points) (note that the world data (brown data points) is without the Mainz A1 [42] and the recent JLab recoil polarization experimental data [30, 43]). The magenta data points are the Mainz A1 data [42]. The two cyan data points are from the recent high precision JLab recoil polarization experiment E03-104 [43]. The dark-blue data points are from the recent high precision JLab recoil polarization experiment E08-007 [30].

Image not extracted

Figure 3. The ratio of the Sachs electric and magnetic form factor of the proton at high and low momentum transfer. The black-solid curve is our soft-wall holographic QCD result with τ=3\tau= 3 (fixed by the hard counting rule) and κ=0.402 GeV\kappa= 0.402\,\mathrm{GeV} (extracted from the best fit to the world data (including Mainz A1 data) of Sachs magnetic form factor of proton). The green-solid curve is our soft-wall holographic QCD result with τ=mN2/mρ2+1=2.465\tau= m_N^2/m_\rho^2 + 1 = 2.465 and 4κ2=mρ2=0.7752 GeV24\kappa^2 = m_\rho^2 = 0.775^2\,\mathrm{GeV}^2. The dark-yellow-solid curve is the soft-wall holographic QCD result for τ=3\tau= 3 and κ=0.350 GeV\kappa= 0.350\,\mathrm{GeV} [26]. The solid-red curve is the Arrington fit [41] to the world data (brown data points) (note that the world data (brown data points) is without the Mainz A1 [42] and the recent JLab recoil polarization experimental data [30, 43]). The magenta data points are the Mainz A1 data [42]. The two cyan data points are from the recent high precision JLab recoil polarization experiment E03-104 [43]. The dark-blue data points are from the recent high precision JLab recoil polarization experiment E08-007 [30]. Throughout, we have used the proton magnetic moment μp=GMp(0)=F1p(0)+F2p(0)=2.793\mu_p = G_M^p(0) = F_1^p(0) + F_2^p(0) = 2.793.

In Figure 4, we show the electric Sachs form factor of the proton. The black-solid curve is our soft-wall holographic QCD result with τ=3\tau=3 and κ=0.402 GeV\kappa=0.402\ \mathrm{GeV} (fixed using the world data of magnetic Sachs form factor of proton). The green-solid curve is our soft-wall holographic QCD result with τ=mN2/mρ2+1=2.465\tau=m_N^2/m_\rho^2+1=2.465 and 4κ2=mρ2=0.7752 GeV24\kappa^2=m_\rho^2=0.775^2\ \mathrm{GeV}^2. The dark-yellow-solid curve is the soft-wall holographic QCD result for τ=3\tau=3 and κ=0.350 GeV\kappa=0.350\ \mathrm{GeV} [26]. The solid-blue curve is the PRad fit

Electromagnetic radii

Proton

The magnetic radius of the proton is defined as

⟨rM2⟩p=−6(dln⁡GMP(Q)dQ2)0ℏ2c2(9)\langle r_M^2\rangle_p = -6\left(\frac{d\ln G_M^P(Q)}{dQ^2}\right)_0 \hbar^2 c^2 \tag*{(9)}

with ℏc=0.197\hbar c = 0.197 GeV fm. Using (II.2) in (IV.9), we find

⟨rM2⟩p=(0.803 fm)2(10)\langle r_M^2\rangle_p = (0.803\ \mathrm{fm})^2 \tag*{(10)}

(extracted from world data), with κ=0.402\kappa=0.402 GeV (fixed using the world data (including the Mainz A1 data) of the magnetic Sachs form factor of the proton, see Figure 2), and τ=3\tau= 3 (fixed by the hard scattering rule). Alternatively, we have

⟨rM2⟩p=(0.791 fm)2(11)\langle r_M^2\rangle_p = (0.791\ \mathrm{fm})^2 \tag*{(11)}

for κ=mρ/2=0.388\kappa= m_\rho/2 = 0.388 GeV, and τ=mN2/mρ2+1=2.465\tau= m_N^2/m_\rho^2 + 1 = 2.465. Combining (IV.10) and (IV.11), we find our final magnetic radius of the proton

⟨rM2⟩p=(0.803±0.012 fm)2(12)\langle r_M^2\rangle_p = (0.803 \pm0.012\ \mathrm{fm})^2 \tag*{(12)}

(extracted from world data),

in good agreement with the average of e-p scattering Mainz A1 data (with the average of the most flexible Spline and Polynomial fits) [42], and world data (with Arrington and Sick [44] world updated value [44] or equivalently with the extraction of Zhan, et al. [30] [30]), respectively, which is ⟨rM2⟩p1/2=(0.777 fm+0.867 fm)/2=0.822 fm\langle r_M^2\rangle_p^{1/2} = (0.777\,\mathrm{fm} + 0.867\,\mathrm{fm})/2 = 0.822\,\mathrm{fm}.

The charge radius of the proton is given by

⟨rC2⟩p=−6(dln⁡GEp(Q)dQ2)0ℏ2c2.(13)\langle r_C^2\rangle_p = -6\left(\frac{d\ln G_E^p(Q)}{dQ^2}\right)_0\hbar^2c^2. \tag*{(13)}

Using (II.1) in (IV.13), we find

⟨rC2⟩p=(0.831 fm)2,(14)\langle r_C^2\rangle_p = (0.831\,\mathrm{fm})^2, \tag*{(14)}

for κ=0.402 GeV\kappa=0.402\,\mathrm{GeV}, and τ=3\tau= 3, see Table I-Table II. We also find

⟨rC2⟩p=(0.839 fm)2,(15)\langle r_C^2\rangle_p = (0.839\,\mathrm{fm})^2, \tag*{(15)}

for κ=0.388 GeV\kappa=0.388\,\mathrm{GeV}, and τ=2.465\tau= 2.465, see Table I. Combining (IV.14) and (IV.15), we find our final charge radius of the proton

⟨rC2⟩p=(0.831±0.008 fm)2 (prediction),(16)\langle r_C^2\rangle_p = (0.831 \pm0.008\,\mathrm{fm})^2\ (\textbf{prediction}), \tag*{(16)}

see Figure 1.

B Neutron

The magnetic radius of the neutron is defined as

⟨rM2⟩n=−6(dln⁡GMN(Q)dQ2)0ℏ2c2(17)\left\langle r_M^2\right\rangle_n=-6\left(\frac{d\ln G_M^N(Q)}{dQ^2}\right)_0\hbar^2c^2 \tag*{(17)}

Using (II.2) in (IV.9), we find

⟨rM2⟩n=(0.817 fm)2,(18)\left\langle r_M^2\right\rangle_n=(0.817\ \mathrm{fm})^2, \tag*{(18)}

with κ=0.402\kappa=0.402 GeV, and τ=3\tau=3. We also find

⟨rM2⟩n=(0.810 fm)2,(19)\left\langle r_M^2\right\rangle_n=(0.810\ \mathrm{fm})^2, \tag*{(19)}

for κ=0.388\kappa=0.388 GeV, and τ=2.465\tau=2.465. Combining (IV.18) and (IV.19), we find our final magnetic radius of the neutron

⟨rM2⟩n=(0.817±0.007 fm)2 (prediction),(20)\left\langle r_M^2\right\rangle_n=(0.817\pm0.007\ \mathrm{fm})^2\ \mathbf{(prediction)}, \tag*{(20)}

in good agreement with the PDG world average [45]

⟨rM2⟩n1/2=0.864 fm.\left\langle r_M^2\right\rangle_n^{1/2}=0.864\ \mathrm{fm}.

The charge radius of the neutron is given by

⟨rC2⟩n=−6(dGEN(Q)dQ2)0ℏ2c2.(21)\left\langle r_C^2\right\rangle_n=-6\left(\frac{dG_E^N(Q)}{dQ^2}\right)_0\hbar^2c^2. \tag*{(21)}

Using (II.1) in (IV.21), we find

⟨rC2⟩n=(−0.094 fm)2,(22)\left\langle r_C^2\right\rangle_n=(-0.094\ \mathrm{fm})^2, \tag*{(22)}

for κ=0.402\kappa=0.402 GeV, and τ=3\tau=3. We also find

⟨rC2⟩n=(−0.142 fm)2,(23)\left\langle r_C^2\right\rangle_n=(-0.142\ \mathrm{fm})^2, \tag*{(23)}

for κ=0.388\kappa=0.388 GeV, and τ=2.465\tau=2.465. Combining (IV.22) and (IV.23), we find our final charge radius of the neutron

⟨rC2⟩n=(−0.094±0.048 fm)2 (prediction).(24)\left\langle r_C^2\right\rangle_n=(-0.094\pm0.048\ \mathrm{fm})^2\ \mathbf{(prediction)}. \tag*{(24)}

in good agreement with the average of MAMI, JLab/Hall-A, and CLAS experiments (with H. Atac, et al. (2021) fit [46]), and the world data (with Kelly [47] fit [47]), which is ⟨rE2⟩n=(−0.110 fm2−0.112 fm2)/2=−0.111 fm2\langle r_E^2\rangle_n = (-0.110\ \mathrm{fm}^2 - 0.112\ \mathrm{fm}^2)/2 = -0.111\ \mathrm{fm}^2.

We have summarized our result for the electromagnetic radii of the proton and neutron in Table II.

Conclusions

By fixing the twist parameter of the nucleon τ=3\tau= 3, and κ=0.402 GeV\kappa= 0.402\ \mathrm{GeV}, we have found a much improved description of the world data (including the Mainz A1 [6], Jefferson Lab recoil polarization experiment E08-007 [30], and PRad data [11]), of both the electric and magnetic Sachs form factors (including their ratio) of the proton. The agreement with the data range from low to high momentum transfers, all within the original-minimal soft-wall holographic QCD [25, 26], see Figure 2 - Figure 4. For example, from our electric Sachs form factor of the proton (II.1), we have found the charge radius of the proton to be rp=0.831±0.008 fmr_p = 0.831 \pm0.008\ \mathrm{fm} in perfect agreement with the recent charge radius of the proton measured by the PRad collaboration at Jefferson Lab [11]. Our theoretical error in the charge radius of the proton, stems from our alternative choice of τ=mN2/mρ2+1=2.465\tau= m_N^2/m_\rho^2 + 1 = 2.465, and κ=mρ/2=0.388 GeV\kappa= m_\rho/2 = 0.388\ \mathrm{GeV}, which gives the charge radius of proton to be rp=0.839 fmr_p = 0.839\ \mathrm{fm}, see Table I.

To our knowledge, our holographic framework is the only theoretical work that produces a perfect match to the recent PRad collaboration measurement of the charge radius of the proton. We have also computed the charge and magnetic radii of the neutron, as well as the magnetic radius of the proton, with a good agreement to the experimental values, see Table II. This is not surprising given the non-perturbative nature of the problem, and the phenomenological success of the original soft-wall holographic QCD construction [25, 26, 40, 48, 49, 50, 51, 52, 53, 54, 55], with only three parameters: the mass scale κ\kappa, the twist parameter of the nucleon τ\tau, and the ’t Hooft coupling constant λ\lambda (also needed for high energy t-channel diffractive processes).

Holography recovers the old dual description with all its phenomenological successes. As we noted earlier, it is the most economical way of enforcing QCD symmetries, duality and crossing symmetries, all usually sought by dispersive analyses, and yet within a well defined and minimal organisational principle using field theory. The present analyses and results are an illustration of that.

However, holography provides much more. Indeed, the bulk holographic description in higher dimension allows for the assessment of any n-point function on the boundary, using field theoretical methods through dual Witten diagrams, in the double limit of large NcN_c and strong gauge coupling. It is the QCD string made user friendly. It also provides for novel physics for processes at low parton-x [53], and for scattering on dense systems as dual to black holes [56, 57].

Acknowledgments

K.M. is supported by the U.S. Department of Energy, Office of Science, Office of Nuclear Physics, contract no. DE-AC02-06CH11357, and an LDRD initiative at Argonne National Laboratory under Project No. 2020-0020. I.Z. is supported by the Office of Science, U.S. Department of Energy under Contract No. DE-FG-88ER40388.

A Soft-wall holographic QCD

A simple way to capture AdS/CFT duality in the non-conformal limit is to model it using a slice of AdS5_5 with various bulk fields with assigned anomalous dimensions and pertinent boundary values, in the so-called bottom-up approach which we will follow here using the conventions in our recent work in DIS scattering [52, 53]. We consider AdS5_5 with a soft wall with a background metric gMN=(ημν,−1)R2/z2g_{MN}=(\eta_{\mu\nu},-1)R^2/z^2 with the flat metric ημν=(1,−1,−1,−1)\eta_{\mu\nu}=(1,-1,-1,-1) at the boundary. Confinement will be described by a harmonic background dilaton ϕ=κ2z2\phi=\kappa^2z^2.

Bulk vector mesons

The vector mesons fields LL, RR are described by the bulk effective action [52, 58, 59]

SM=−14g52∫d5xe−ϕ(z)ggMPgNQTr⁡(FMNLFPQL+FMNRFPQR)+∫d5x(ω5L(A)−ω5R(A))(25)S_M=-\frac{1}{4g_5^2}\int d^5x e^{-\phi(z)}\sqrt{g}g^{MP}g^{NQ}\operatorname{Tr}\left(F^L_{MN}F^L_{PQ}+F^R_{MN}F^R_{PQ}\right)+\int d^5x\left(\omega_5^L(\mathcal{A})-\omega_5^R(\mathcal{A})\right) \tag*{(25)}

with the Chern-Simons contribution

ω5(A)=Nc24π2∫d5xTr⁡(AF2+12A3F−110A5)(26)\omega_5(\mathcal{A})=\frac{N_c}{24\pi^2}\int d^5x\operatorname{Tr}\left(\mathcal{A}\mathcal{F}^2+\frac{1}{2}\mathcal{A}^3\mathcal{F}-\frac{1}{10}\mathcal{A}^5\right) \tag*{(26)}

Here F=dA−iA2\mathcal{F}=d\mathcal{A}-i\mathcal{A}^2 and A=AaTa\mathcal{A}=\mathcal{A}^aT^a with T0=3212T^0=\frac{3}{2}\mathbf{1}_2 and Ti=12τiT^i=\frac{1}{2}\tau^i, with the form notation subsumed. Also the vector fields are given by V=(R+L)/2V=(R+L)/2 and the axial-vector fields are given by A=(R−L)/2A=(R-L)/2. The coupling g5g_5 in (A.1) is fixed by the brane embeddings in bulk, or phenomenologically as 1/g52≡Nc/(12π2)1/g_5^2\equiv N_c/(12\pi^2) [60].

The flavor gauge fields solve

□Vμ+zeκ2z2∂z(e−κ2z21z∂zVμ)=0□Vz−∂z(∂μVμ)=0.(27)\Box V^\mu+ze^{\kappa^2z^2}\partial_z\left(e^{-\kappa^2z^2}\frac{1}{z}\partial_zV^\mu\right)=0 \qquad\Box V_z-\partial_z\left(\partial_\mu V^\mu\right)=0 . \tag*{(27)}

subject to the gauge condition

∂μVμ+zeκ2z2∂z(e−κ2z21zVz)=0,(28)\partial_\mu V^\mu+ze^{\kappa^2z^2}\partial_z\left(e^{-\kappa^2z^2}\frac{1}{z}V_z\right)=0 , \tag*{(28)}

with the boundary condition Vμ(z,y)∣z→0=ϵμ(q)e−iq⋅yV_\mu(z,y)|_{z\to0}=\epsilon_\mu(q)e^{-iq\cdot y}. The non-normalizable solutions are

Vμ(z,y)=ϵμ(q)e−iq⋅yΓ(1−q24κ2)κ2z2U(1−q24κ2;2;κ2z2)V_\mu(z,y)=\epsilon_\mu(q)e^{-iq\cdot y}\Gamma\left(1-\frac{q^2}{4\kappa^2}\right)\kappa^2z^2U\left(1-\frac{q^2}{4\kappa^2};2;\kappa^2z^2\right)
Vz(z,y)=i2ϵ(q)⋅q e−iq⋅yΓ(1−q24κ2)zU(1−q24κ2;1;κ2z2),(29)V_z(z,y)=\frac{i}{2}\epsilon(q)\cdot q\,e^{-iq\cdot y}\Gamma\left(1-\frac{q^2}{4\kappa^2}\right)zU\left(1-\frac{q^2}{4\kappa^2};1;\kappa^2z^2\right) , \tag*{(29)}

with U(a;b;w)U(a;b;w) the confluent hypergeometric functions of the second kind. The normalizable solutions (vector wavefunctions) are also given by [48]

ϕn(z)=cn×κ2z2Ln1(κ2z2),(30)\phi_n(z)=c_n\times\kappa^2z^2L_n^1(\kappa^2z^2) , \tag*{(30)}

with cn=2/(n+1)c_n=\sqrt{2/(n+1)} which is determined from the normalization condition (for the soft-wall model with background dilaton ϕ=κ2z2\phi=\kappa^2z^2)

∫dzge−ϕ(gxx)2ϕn(z)ϕm(z)=δnm,(31)\int dz\sqrt{g}e^{-\phi}(g^{xx})^2\phi_n(z)\phi_m(z)=\delta_{nm} , \tag*{(31)}

Here Ln(α)(ξ)L_n^{(\alpha)}(\xi) are the generalized Laguerre polynomials. Therefore, we have

Fn=1g5(−e−ϕ1z′∂z′ϕn(z′))z′=ϵ=−2g5cn(n+1)κ2,(32)F_n=\frac{1}{g_5}\left(-e^{-\phi}\frac{1}{z'}\partial_{z'}\phi_n(z')\right)_{z'=\epsilon}=-\frac{2}{g_5}c_n(n+1)\kappa^2 , \tag*{(32)}

with ϕn(z→0)≈cnκ2z2(n+1)\phi_n(z \to0) \approx c_n\kappa^2 z^2(n+1). Then, defining the decay constant as

fn=−Fn/mn,f_n=-F_n/m_n,

we have

ϕn(z)=fnmn×2g5×κ2z2Ln1(κ2z2),(33)\phi_n(z)=\frac{f_n}{m_n}\times2g_5\times\kappa^2z^2L_n^1(\kappa^2z^2), \tag*{(33)}

as required by vector meson dominance (VMD). We can also write the non-normalizable solution as sum over the normalizable solutions as [48]

V(q,z)=∑n−g5Fnϕn(z)q2−mn2=g52∑nfn2q2−mn2×2ξLn1(ξ),(34)V(q,z)=\sum_n\frac{-g_5F_n\phi_n(z)}{q^2-m_n^2}=g_5^2\sum_n\frac{f_n^2}{q^2-m_n^2}\times2\xi L_n^1(\xi), \tag*{(34)}

with ξ=κ2z2\xi=\kappa^2z^2.

Bulk Dirac fermions

The bulk Dirac fermion action in a sliced of AdS5_5 is

SF=12g52∫d5x e−ϕ(z)g(LF1+LF2)+12g52∫d4x−g(4)(LUV1+LUV2),(35)S_F=\frac{1}{2g_5^2}\int d^5x\,e^{-\phi(z)}\sqrt{g}\left(\mathcal{L}_{F1}+\mathcal{L}_{F2}\right)+\frac{1}{2g_5^2}\int d^4x\sqrt{-g^{(4)}}\left(\mathcal{L}_{UV1}+\mathcal{L}_{UV2}\right), \tag*{(35)}

The Dirac and Pauli contributions to LF1,2\mathcal{L}_{F1,2} are respectively

LDirac1,2=(i2Ψˉ1,2eANΓA(D→NL,R−D←NL,R)Ψ1,2−(±M+V(z))Ψˉ1,2Ψ1,2),\mathcal{L}_{\mathrm{Dirac}1,2}=\left(\frac{i}{2}\bar{\Psi}_{1,2}e_A^N\Gamma^A\left(\overrightarrow{D}_N^{L,R}-\overleftarrow{D}_N^{L,R}\right)\Psi_{1,2}-\left(\pm M+V(z)\right)\bar{\Psi}_{1,2}\Psi_{1,2}\right),
LPauli1,2=±2g52×η Ψˉ1,2eAMeBNσABFMNL,RΨ1,2,(36)\mathcal{L}_{\mathrm{Pauli}1,2}=\pm2g_5^2\times\eta\,\bar{\Psi}_{1,2}e_A^M e_B^N\sigma^{AB}F_{MN}^{L,R}\Psi_{1,2}, \tag*{(36)}

with V(z)=κ2z2V(z)=\kappa^2z^2, eAN=zδANe_A^N=z\delta_A^N, σAB=i2[ΓA,ΓB]\sigma^{AB}=\frac{i}{2}[\Gamma^A,\Gamma^B], and ωμzν=−ωμνz=1zημν\omega_{\mu z\nu}=-\omega_{\mu\nu z}=\frac{1}{z}\eta_{\mu\nu}. The Dirac gamma matrices ΓA=(γμ,−iγ5)\Gamma^A=(\gamma^\mu,-i\gamma^5) are chosen in the chiral representation. They satisfy the flat anti-commutation relation {ΓA,ΓB}=2ηAB\{\Gamma^A,\Gamma^B\}=2\eta^{AB}. The left and right covariant derivatives are defined as

D→NX=L,R=∂→N+18ωNAB[ΓA,ΓB]−iXNaTa≡D→N−iXNaTa\overrightarrow{D}_N^{X=L,R}=\overrightarrow{\partial}_N+\frac{1}{8}\omega_{NAB}[\Gamma^A,\Gamma^B]-iX_N^aT^a\equiv\overrightarrow{D}_N-iX_N^aT^a
D←NX=L,R=∂←N+18ωNAB[ΓA,ΓB]+iXNaTa≡D←N+iXNaTa(37)\overleftarrow{D}_N^{X=L,R}=\overleftarrow{\partial}_N+\frac{1}{8}\omega_{NAB}[\Gamma^A,\Gamma^B]+iX_N^aT^a\equiv\overleftarrow{D}_N+iX_N^aT^a \tag*{(37)}

The nucleon doublet refers to

Ψ1,2≡(Ψp1,2Ψn1,2).(38)\Psi_{1,2}\equiv \begin{pmatrix} \Psi_{p1,2}\\ \Psi_{n1,2} \end{pmatrix}. \tag*{(38)}

The nucleon fields in bulk form an iso-doublet p,np,n with 1,21,2 referring to their boundary chirality 1,2=±=R,L1,2=\pm=R,L [61]. They are dual to the boundary sources Ψp1,2↔Op,±\Psi_{p1,2}\leftrightarrow\mathcal{O}_{p,\pm} and Ψn1,2↔On,±\Psi_{n1,2}\leftrightarrow\mathcal{O}_{n,\pm} with anomalous dimensions ±M=±(Δ−2)=±(τ−3/2)\pm M=\pm(\Delta-2)=\pm(\tau-3/2).

The equation of motions for the bulk Dirac chiral doublet is

(ieANΓADNL,R−i2(∂Nϕ)eANΓA−(±M+V(z)))Ψ1,2=0,(39)\left(i e_A^N\Gamma^A D_N^{L,R}-\frac{i}{2}(\partial_N\phi)e_A^N\Gamma^A-\left(\pm M+V(z)\right)\right)\Psi_{1,2}=0, \tag*{(39)}

The normalizable solution to (A.15) are

Ψ1(p,z)=ψR(z)ΨR0(p)+ψL(z)ΨL0(p)Ψ2(p,z)=ψR(z)ΨL0(p)+ψL(z)ΨR0(p)(40)\begin{aligned} \Psi_1(p,z) &= \psi_R(z)\Psi_R^0(p)+\psi_L(z)\Psi_L^0(p)\\ \Psi_2(p,z) &= \psi_R(z)\Psi_L^0(p)+\psi_L(z)\Psi_R^0(p) \tag*{(40)} \end{aligned}

with the normalized bulk wave functions

ψR(z)=n~Rξτ−32Ln(τ−2)(ξ)=n~Rκτ−2ξ52ξτ−22Ln(τ−2)(ξ),\psi_R(z)=\tilde n_R\xi^{\tau-\frac{3}{2}}L_n^{(\tau-2)}(\xi)=\frac{\tilde n_R}{\kappa^{\tau-2}}\xi^{\frac{5}{2}}\xi^{\frac{\tau-2}{2}}L_n^{(\tau-2)}(\xi),
ψL(z)=n~Lξτ−1Ln(τ−1)(ξ)=n~Lκτ−1ξ52ξτ−22Ln(τ−1)(ξ),(41)\psi_L(z)=\tilde n_L\xi^{\tau-1}L_n^{(\tau-1)}(\xi)=\frac{\tilde n_L}{\kappa^{\tau-1}}\xi^{\frac{5}{2}}\xi^{\frac{\tau-2}{2}}L_n^{(\tau-1)}(\xi), \tag*{(41)}

Here ξ=κ2z2\xi=\kappa^2z^2, Ln(α)(ξ)L_n^{(\alpha)}(\xi), n~R=n~Lκ−1τ−1\tilde n_R=\tilde n_L\kappa^{-1}\sqrt{\tau-1} are the generalized Laguerre, and n~L=κτ2/Γ(τ)\tilde n_L=\kappa^\tau\sqrt{2/\Gamma(\tau)}. The free Weyl spinors ΨR/L0(p)=P∓u(p)\Psi^0_{R/L}(p)=P_{\mp}u(p) and ΨˉR/L0(p)=uˉ(p)P∓\bar{\Psi}^0_{R/L}(p)=\bar{u}(p)P_{\mp}, and the free boundary spinors satisfy

uˉ(p)u(p)=2mN2mN×uˉ(p′)γμu(p)=uˉ(p′)[(p′+p)μ+iσμν(p′−p)ν]u(p).(42)\bar u(p)u(p)=2m_N \qquad 2m_N\times\bar u(p')\gamma^\mu u(p)=\bar u(p')\left[(p'+p)^\mu+i\sigma^{\mu\nu}(p'-p)_\nu\right]u(p). \tag*{(42)}

The fermionic spectrum Reggeizes mn2=4κ2(n+τ−1)m_n^2=4\kappa^2(n+\tau-1). The assignments 1 = + and 2 = − at the boundary are commensurate with the substitutions ψR,L↔∓ψL,R\psi_{R,L}\leftrightarrow\mp\psi_{L,R} by parity.

Using the Dirac 1-form currents

JLaN=∂LDirac1∂LNa=Ψˉ1eANΓaTaΨ1,J_L^{aN}=\frac{\partial\mathcal{L}_{\mathrm{Dirac}1}}{\partial L_N^a}=\bar{\Psi}_1e_A^N\Gamma^aT^a\Psi_1,
JRaN=∂LDirac2∂RNa=Ψˉ2eANΓaTaΨ2,(43)J_R^{aN}=\frac{\partial\mathcal{L}_{\mathrm{Dirac}2}}{\partial R_N^a}=\bar{\Psi}_2e_A^N\Gamma^aT^a\Psi_2, \tag*{(43)}

and Pauli 2-form currents

JLaMN=∂LPauli1∂LMNa=+2g52×ηaΨˉ1eAMeBNσABTaΨ1,J_L^{aMN}=\frac{\partial\mathcal{L}_{\mathrm{Pauli}1}}{\partial L_{MN}^a}=+2g_5^2\times\eta^a\bar{\Psi}_1e_A^Me_B^N\sigma^{AB}T^a\Psi_1,
JRaMN=∂LPauli2∂RMNa=−2g52×ηaΨˉ2eAMeBNσABTaΨ2.(44)J_R^{aMN}=\frac{\partial\mathcal{L}_{\mathrm{Pauli}2}}{\partial R_{MN}^a}=-2g_5^2\times\eta^a\bar{\Psi}_2e_A^Me_B^N\sigma^{AB}T^a\Psi_2. \tag*{(44)}

we can rewrite (A.12) with the explicit isoscalar (a=0a=0) and isovector (a=3a=3) contributions

LF1+LF2⊃i2Ψˉ1eANΓA(D→N−D←N)Ψ1−(M+V(z))Ψˉ1Ψ1+i2Ψˉ2eANΓA(D→N−D←N)Ψ2−(−M+V(z))Ψˉ2Ψ2+VN0JV0N+AN0JA0N+VN3JV3N+AN3JA3N+VMN0JV0MN+AMN0JA0MN+VMN3JV3MN+AMN3JA3MN(45)\begin{aligned} \mathcal{L}_{F1}+\mathcal{L}_{F2}\supset{}& \frac{i}{2}\bar{\Psi}_1e_A^N\Gamma^A(\overrightarrow{D}_N-\overleftarrow{D}_N)\Psi_1-(M+V(z))\bar{\Psi}_1\Psi_1\\ &+\frac{i}{2}\bar{\Psi}_2e_A^N\Gamma^A(\overrightarrow{D}_N-\overleftarrow{D}_N)\Psi_2-(-M+V(z))\bar{\Psi}_2\Psi_2\\ &+V_N^0J_V^{0N}+A_N^0J_A^{0N}+V_N^3J_V^{3N}+A_N^3J_A^{3N}\\ &+V_{MN}^0J_V^{0MN}+A_{MN}^0J_A^{0MN}+V_{MN}^3J_V^{3MN}+A_{MN}^3J_A^{3MN} \tag*{(45)} \end{aligned}

with JV,AaN=JLaN±JRaNJ_{V,A}^{aN}=J_L^{aN}\pm J_R^{aN} and JV,AaMN=JLaMN∓JRaMNJ_{V,A}^{aMN}=J_L^{aMN}\mp J_R^{aMN}.

Electromagnetic form factors

The electromagnetic form factors for the nucleon in the context of the soft wall model have been originally addressed in [26], although with a different analysis at low momentum transfer than the one we will present and which will fix the numerical shortcomings they encountered. Holographic QCD maps gauge invariant operators at the boundary with supergravity fields in bulk with pertinent anomalous dimensions. The bulk string theory is in general difficult to solve, but in the double limit of a large number of colors and strong gauge coupling λ=g2Nc\lambda=g^2N_c, it reduces to a weakly coupled supergravity in the classical limit in AdS5AdS_5 geometry, with a weak string coupling gstring=g2/4πg_{\mathrm{string}}=g^2/4\pi. The nn-point functions at the boundary of AdS5AdS_5 follow from variation of the on-shell supergravity action in bulk with respect to the boundary values. The results are tree-level Feynman graphs with fixed end-points on the boundary also known as Witten diagrams [62] (and references therein).

The electromagnetic form factor for the nucleon can be extracted from the normalized boundary-to-bulk Witten diagram for the 3-point function, with pertinent LSZ reduction

WVaμ(q)=lim⁡p22,p12→mN2(p22−mN2)(p12−mN2)⟨ON(−p2)JVaμ(q)ON(p1)⟩FN(p2)FN(p1),(46)W_{V}^{a\mu}(q)=\lim_{p_{2}^{2},p_{1}^{2}\to m_{N}^{2}}(p_{2}^{2}-m_{N}^{2})(p_{1}^{2}-m_{N}^{2})\frac{\langle O_{N}(-p_{2})J_{V}^{a\mu}(q)O_{N}(p_{1})\rangle}{F_{N}(p_{2})F_{N}(p_{1})}, \tag*{(46)}

for the isoscalar a=0a=0 and the isovector a=3a=3 current JVaμJ_{V}^{a\mu}. The nucleon source constant is FN(p)=⟨0∣ON(0)∣N(p)⟩F_{N}(p)=\langle0|O_{N}(0)|N(p)\rangle. The electromagnetic current form factors follow from the identification

⟨N(p2)∣JEMμ(0)∣N(p1)⟩=uˉ(p2)(F1(Q)γμ+F2(Q)iσμνqν2mN)u(p1)=13WV0μ(Q2)+WV3μ(Q2)=WEMμ(Q2)(47)\langle N(p_{2})|J_{\mathrm{EM}}^{\mu}(0)|N(p_{1})\rangle=\bar{u}(p_{2})\left(F_{1}(Q)\gamma^{\mu}+F_{2}(Q)\frac{i\sigma^{\mu\nu}q_{\nu}}{2m_{N}}\right)u(p_{1})=\frac{1}{3}W_{V}^{0\mu}(Q^{2})+W_{V}^{3\mu}(Q^{2})=W_{\mathrm{EM}}^{\mu}(Q^{2}) \tag*{(47)}

with q2=(p2−p1)2=−Q2q^{2}=(p_{2}-p_{1})^{2}=-Q^{2} (space-like). Here N(p)N(p) refers to the U(2) proton-neutron doublet.

The vector currents JVaμJ_{V}^{a\mu} at the boundary are sourced by the dual bulk vector fields Vμa(Q,z→0)V_{\mu}^{a}(Q,z\to0). The Dirac and Pauli contributions to the direct part of the electromagnetic current can be extracted from the pertinent bulk Dirac and Pauli parts of the action in the soft wall model. The Dirac part of the action in (A.21) yields

SDiracEM[j,f]=12g52∫dzd4y ge−ϕzR(Ψˉ1fγN(13VN0T0+VN3T3)Ψ1i+Ψˉ2fγN(13VN0T0+VN3T3)Ψ2i),=(2π)4δ4(p′−p−q)×FN(p′)×FN(p)×12g52×2g52×eN×uˉsf(p′)ε(q)usi(p)×12∫dzz2Me−ϕV(Q,z)(ψ~L2(z)+ψ~R2(z)).(48)\begin{aligned} S_{\mathrm{Dirac}}^{\mathrm{EM}}[j,f]=\frac{1}{2g_{5}^{2}}\int dzd^{4}y\,\sqrt{g}e^{-\phi}\frac{z}{R}\left(\bar{\Psi}_{1f}\gamma^{N}\left(\frac{1}{3}V_{N}^{0}T^{0}+V_{N}^{3}T^{3}\right)\Psi_{1i}+\bar{\Psi}_{2f}\gamma^{N}\left(\frac{1}{3}V_{N}^{0}T^{0}+V_{N}^{3}T^{3}\right)\Psi_{2i}\right), \\ &=(2\pi)^{4}\delta^{4}(p'-p-q)\times F_{N}(p')\times F_{N}(p)\times\frac{1}{2g_{5}^{2}}\times2g_{5}^{2}\times e_{N}\times\bar{u}_{s_{f}}(p')\varepsilon(q)u_{s_{i}}(p)\times\frac{1}{2}\int\frac{dz}{z^{2M}}e^{-\phi}V(Q,z)\left(\widetilde{\psi}_{L}^{2}(z)+\widetilde{\psi}_{R}^{2}(z)\right). \tag*{(48)} \end{aligned}

Here M=τ−32M=\tau-\frac{3}{2}, ϕ=κ2z2\phi=\kappa^{2}z^{2}, and ϵμ(q)\epsilon^{\mu}(q) is the polarization of the electromagnetic (EM) probe. In the last equality in (B.3), we substituted the bulk gauge fields by (A.5), and the bulk fermionic currents in terms of the fermionic fields (A.16). The charge assignments are eN=1e_{N}=1 for the proton, and eN=0e_{N}=0 for the neutron. We also use the nucleon (ground state) bulk wave functions given by (A.16)

ψ~R(z)=nRξτ−32L0(τ−2)(ξ)≡nRξτ−32,\widetilde{\psi}_{R}(z)=n_{R}\xi^{\tau-\frac{3}{2}}L_{0}^{(\tau-2)}(\xi)\equiv n_{R}\xi^{\tau-\frac{3}{2}},
ψ~L(z)=nLξτ−1L0(τ−1)(ξ)≡nLξτ−1,(49)\widetilde{\psi}_{L}(z)=n_{L}\xi^{\tau-1}L_{0}^{(\tau-1)}(\xi)\equiv n_{L}\xi^{\tau-1}, \tag*{(49)}

with the generalized Laguerre polynomials Ln(α)(ξ)L_{n}^{(\alpha)}(\xi), ξ=κ2z2\xi=\kappa^{2}z^{2}, nR=nLτ−1n_{R}=n_{L}\sqrt{\tau-1} and nL=κ−(τ−2)2/Γ(τ)n_{L}=\kappa^{-(\tau-2)}\sqrt{2/\Gamma(\tau)}. We also use the non-normalizable solution to the bulk U(1) gauge field (the bulk-to-boundary vector propagator) (A.5)

V(Q,z)=g52×∑n−fn2Q2+mn2×2ξL1n(ξ)=ξΓ(1+a)U(1+a;2;ξ)=ξ∫01dx(1−x)2xaexp⁡(−xξ1−x)(50)\mathcal{V}(Q,z)=g_{5}^{2}\times\sum_{n}\frac{-f_{n}^{2}}{Q^{2}+m_{n}^{2}}\times2\xi L_{1}^{n}(\xi)=\xi\Gamma(1+a)U(1+a;2;\xi)=\xi\int_{0}^{1}\frac{dx}{(1-x)^{2}}x^{a}\exp\left(-\frac{x\xi}{1-x}\right) \tag*{(50)}

with the decay constant of the vector meson resonances fn=2κg5f_{n}=\frac{\sqrt{2}\kappa}{g_{5}}, linear Regge trajectory for mass of vector meson resonances mn2=4κ2(n+1)m_{n}^{2}=4\kappa^{2}(n+1), a=Q2/(4κ2)a=Q^{2}/(4\kappa^{2}), and ξ=κ2z2\xi=\kappa^{2}z^{2}. The bulk-to-boundary propagator satisfies V(0,z)=V(Q,0)=1\mathcal{V}(0,z)=\mathcal{V}(Q,0)=1.

The Dirac part of the electromagnetic current (B.2) follows from (B.3) by variation

WEM(Dirac)μ(Q2)=uˉs′(p′)γμus(p)×eN×C1(Q)≡1FN(p′)FN(p)δSDiracEMδμ(q)(51)W_{EM(\mathrm{Dirac})}^{\mu}(Q^{2})=\bar{u}_{s'}(p')\gamma^{\mu}u_{s}(p)\times e_{N}\times C_{1}(Q)\equiv\frac{1}{F_{N}(p')F_{N}(p)}\frac{\delta S_{\mathrm{Dirac}}^{EM}}{\delta\mu(q)} \tag*{(51)}

or more explicitly

C1(Q)=12∫dzz2Me−ϕV(Q,z)(ψ~L2(z)+ψ~R2(z))=g52×∑n−1Q2+mn2×[fn2∫dzz2Me−ξξn1(ξ)(ψ~L2(z)+ψ~R2(z))]=g52×∑n−1Q2+mn2×GVnBˉB=∫01dx(τ(1−x2)−12)xa(1−x)τ−2=(a+2τ)Γ(a+1)Γ(τ)2Γ(a+τ+1)=(a2+τ)B(τ,a+1),(52)\begin{aligned} C_{1}(Q)&=\frac{1}{2}\int\frac{dz}{z^{2M}}e^{-\phi}\mathcal{V}(Q,z)\left(\widetilde{\psi}_{L}^{2}(z)+\widetilde{\psi}_{R}^{2}(z)\right)\\ &=g_{5}^{2}\times\sum_{n}\frac{-1}{Q^{2}+m_{n}^{2}}\times\left[f_{n}^{2}\int\frac{dz}{z^{2M}}e^{-\xi\xi_{n}^{1}(\xi)}\left(\widetilde{\psi}_{L}^{2}(z)+\widetilde{\psi}_{R}^{2}(z)\right)\right]\\ &=g_{5}^{2}\times\sum_{n}\frac{-1}{Q^{2}+m_{n}^{2}}\times\mathcal{G}_{V_{n}\bar{B}B}=\int_{0}^{1}dx\left(\tau\left(1-\frac{x}{2}\right)-\frac{1}{2}\right)x^{a}(1-x)^{\tau-2}=\frac{(a+2\tau)\Gamma(a+1)\Gamma(\tau)}{2\Gamma(a+\tau+1)}\\ &=\left(\frac{a}{2}+\tau\right)B(\tau,a+1), \tag*{(52)} \end{aligned}

where, in the last line, we have defined the Euler beta function B(x,y)=Γ(x)Γ(y)Γ(x+y)B(x,y)=\frac{\Gamma(x)\Gamma(y)}{\Gamma(x+y)}. In the second line, we have also defined the vector meson resonances VnV_{n} coupling to the proton (baryon) BB as

GVnBˉB≡fn2∫dzz2Me−ξξn1(ξ)(ψ~L2(z)+ψ~R2(z))(53)\mathcal{G}_{V_{n}\bar{B}B}\equiv f_{n}^{2}\int\frac{dz}{z^{2M}}e^{-\xi\xi_{n}^{1}(\xi)}\left(\widetilde{\psi}_{L}^{2}(z)+\widetilde{\psi}_{R}^{2}(z)\right) \tag*{(53)}

from which we can conclude that GVnBˉB∼Nc\mathcal{G}_{V_{n}\bar{B}B}\sim N_{c}, as expected in the large-NcN_{c} limit. Note that normalizing C1(0)=1C_{1}(0)=1 for the proton, fixes 1+O(Nc−2)=11+O(N_{c}^{-2})=1. The form factor readily crosses to time-like (modulo 1/Nc1/N_{c} widths), with all meson-baryon-antibaryon couplings (B.8) fixed, in the double limit of large NcN_{c} and strong ’t Hooft coupling! This clearly shows the relationship with constructions based on dispersive analyses [22] (and references therein).

The nucleon electromagnetic form factors including the Pauli contribution follows a similar reasoning, with the result for the proton

F1P(Q)=C1(Q)+ηPC2(Q),F2P(Q)=ηPC3(Q),(54)\begin{aligned} F_{1}^{P}(Q)&=C_{1}(Q)+\eta_{P}C_{2}(Q),\\ F_{2}^{P}(Q)&=\eta_{P}C_{3}(Q), \tag*{(54)} \end{aligned}

and the neutron

F1N(Q)=ηNC2(Q),F2N(Q)=ηNC3(Q).(55)\begin{aligned} F_{1}^{N}(Q)&=\eta_{N}C_{2}(Q),\\ F_{2}^{N}(Q)&=\eta_{N}C_{3}(Q). \tag*{(55)} \end{aligned}

The additional invariant form factors stem from the Pauli contribution. They are structurally similar to the Dirac one in (B.7), with

C2(Q)=12∫dzz2Me−ϕz∂zV(Q,z)(ψ~L2(z)−ψ~R2(z))=∫01dx(τx(τx+x−3)+1)xa(1−x)τ−2=a(a(τ−1)−1)Γ(a+1)Γ(τ)Γ(a+τ+2)=(τ−1)a(a(τ−1)−1)(a+τ)(a+τ+1)B(τ−1,a+1),(56)\begin{aligned} C_{2}(Q)&=\frac{1}{2}\int\frac{dz}{z^{2M}}e^{-\phi}z\partial_{z}\mathcal{V}(Q,z)\left(\widetilde{\psi}_{L}^{2}(z)-\widetilde{\psi}_{R}^{2}(z)\right)\\ &=\int_{0}^{1}dx\left(\tau x(\tau x+x-3)+1\right)x^{a}(1-x)^{\tau-2}\\ &=\frac{a(a(\tau-1)-1)\Gamma(a+1)\Gamma(\tau)}{\Gamma(a+\tau+2)}\\ &=(\tau-1)\frac{a(a(\tau-1)-1)}{(a+\tau)(a+\tau+1)}B(\tau-1,a+1), \tag*{(56)} \end{aligned}

and

C3(Q)=2∫dzz2Me−ϕmNzV(Q,z)ψ~L(z)ψ~R(z),=∫01dx  4(τ−1)τxa(1−x)τ−1,=4(τ−1)τΓ(a+1)Γ(τ)Γ(a+τ+1),=4(τ−1)τB(τ,a+1),(57)\begin{aligned} C_3(Q)=2\int\frac{dz}{z^{2M}}e^{-\phi}m_N zV(Q,z)\widetilde{\psi}_L(z)\widetilde{\psi}_R(z), \\ &=\int_0^1dx\;4(\tau-1)\tau x^a(1-x)^{\tau-1}, \\ &=\frac{4(\tau-1)\tau\Gamma(a+1)\Gamma(\tau)}{\Gamma(a+\tau+1)}, \\ &=4(\tau-1)\tau B(\tau,a+1), \tag*{(57)} \end{aligned}

with mN2=4κ2(τ−1)m_N^2=4\kappa^2(\tau-1). Note that the contributions C1,2(Q)C_{1,2}(Q) to the Dirac form factors are chirality-spin preserving with LLLL and RRRR following the γμ\gamma^\mu assignment, while the contribution C3(Q)C_3(Q) to the Pauli form factor is chirality-spin flipping with LRLR following the σμν\sigma^{\mu\nu} assignment.

Hard scattering rule and the dual resonance model

The hard scaling behavior for the electric form factor [39] is recovered at high energy by fixing τ→3\tau\to3. Indeed, the contribution C1(Q)C_1(Q) to the Dirac form factor F1(Q)F_1(Q) in (B.7) with a≫1a\gg1 and fixed τ\tau, asymptotes

C1(Q)=(a2+τ)×B(τ,a+1)≈12Γ(τ)aτ−1,(58)C_1(Q)=\left(\frac{a}{2}+\tau\right)\times B(\tau,a+1)\approx\frac{1}{2}\frac{\Gamma(\tau)}{a^{\tau-1}}, \tag*{(58)}

in agreement with the hard scattering rule for τ=3\tau=3. Similarly, the Pauli contribution C2(Q)C_2(Q) to the Dirac form factor F1(Q)F_1(Q) in (B.11), at high momentum with a≫1a\gg1 and fixed τ\tau, asymptotes

C2(Q)=(τ−1)a(a(τ−1)−1)(a+τ)(a+τ+1)B(τ−1,a+1),≈(τ−1)Γ(τ)aτ−1,(59)\begin{aligned} C_2(Q)=(\tau-1)\frac{a(a(\tau-1)-1)}{(a+\tau)(a+\tau+1)}B(\tau-1,a+1), \\ &\approx(\tau-1)\frac{\Gamma(\tau)}{a^{\tau-1}}, \tag*{(59)} \end{aligned}

also in agreement with the hard scattering rule for τ=3\tau=3. Note that The Pauli term contribution C3(Q)C_3(Q) to the Pauli form factor F2(Q)F_2(Q) in (B.12), at high momentum with a≫1a\gg1 and fixed τ\tau, asymptotes to

C3(Q)=4τ(τ−1)B(τ,a+1)≈4τ(τ−1)Γ(τ)aτ,(60)C_3(Q)=4\tau(\tau-1)B(\tau,a+1)\approx4\tau(\tau-1)\frac{\Gamma(\tau)}{a^\tau}, \tag*{(60)}

which is subleading in 1/Q21/Q^2.

This scaling law originates from the contribution of the R-spinor (B.4) in the form factor near the UV boundary, as it dwarfs that of the L-spinor (B.4) when inserted in the integral (B.7), with the bulk-to-boundary propagator V(Q,z∼1/Q)≈1V(Q,z\sim1/Q)\approx1. More specifically, the integrand in (B.7) gives

(zz2M)ξ2τ−3(1+ξ)≈4τ−6z2τ−4≈(1Q2)τ−1(61)\left(\frac{z}{z^{2M}}\right)\xi^{2\tau-3}(1+\xi)\approx\frac{4^{\tau-6}}{z^{2\tau-4}}\approx\left(\frac{1}{Q^2}\right)^{\tau-1} \tag*{(61)}

Note that the Pauli contributions in (B.11) carry integrands with extra z∼1/Qz\sim1/Q suppression near the UV boundary, and are sub leading in 1/Q21/Q^2. This is expected since the Pauli term is helicity-flipping.

Finally, we note the close relationship between the nucleon form factor obtained in the old resonance dual model [35]

B(rV,12+a)B(rV,12)(62)\frac{B(r_V,\frac{1}{2}+a)}{B(r_V,\frac{1}{2})} \tag*{(62)}

and the holographic form factors (C.1–C.3). In (C.5), the spin rho Regge trajectory is

12+a=12+α′t,\frac{1}{2}+a=\frac{1}{2}+\alpha't,

with rV=τ−1r_V=\tau-1 and α′=1/4κ2\alpha'=1/4\kappa^2 the open string or rho meson Regge slope. Since the dual resonance description is rooted in the Veneziano amplitude for open string scattering, it is not surprising that the holographic description which captures the dominant stringy excitations in bulk, yields similar results.

References

References

  1. [1]C. E. Carlson, Prog. Part. Nucl. Phys. 82, 59 (2015), 1502.05314.
  2. [2]J. C. Bernauer, EPJ Web Conf. 234, 01001 (2020).
  3. [3]H. Gao and M. Vanderhaeghen, Rev. Mod. Phys. 94, 015002 (2022), 2105.00571.
  4. [4]J. Arrington et al. (2021), 2112.00060.
  5. [5]S. Durr et al., Science 322, 1224 (2008), 0906.3599.
  6. [6]J. C. Bernauer et al. (A1), Phys. Rev. Lett. 105, 242001 (2010), 1007.5076.
  7. [7]H. Fleurbaey, S. Galtier, S. Thomas, M. Bonnaud, L. Julien, F. Biraben, F. Nez, M. Abgrall, and J. Guéna, Phys. Rev. Lett. 120, 183001 (2018), 1801.08816.
  8. [8]A. Antognini et al., Science 339, 417 (2013).
  9. [9]N. Bezginov, T. Valdez, M. Horbatsch, A. Marsman, A. C. Vutha, and E. A. Hessels, Science 365, 1007 (2019).
  10. [10]A. Grinin et al., Science 370, 1061 (2020).
  11. [11]W. Xiong et al., Nature 575, 147 (2019).
  12. [12]T. Mart and A. Salakmonu, Phys. Rev. C 87, 025807 (2013), 1302.6012.
  13. [13]I. T. Lorenz and U.-G. Meißner, Phys. Lett. B 737, 57 (2014), 1406.2962.
  14. [14]K. Griffioen, C. Carlson, and S. Maddox, Phys. Rev. C 93, 065207 (2016), 1509.06676.
  15. [15]D. W. Higinbotham, A. A. Kabir, V. Lin, D. Meekins, B. Norum, and B. Sawatzky, Phys. Rev. C 93, 055207 (1980). (2016), 1510.01293.
  16. [16]M. Horbatsch, E. A. Hessels, and A. Pineda, Phys. Rev. C 95, 035203 (2017), 1610.09760.
  17. [17]S. Zhou, P. Giuliani, J. Piekarewicz, A. Bhattacharya, and D. Pati, Phys. Rev. C 99, 055202 (2019), 1808.05977.
  18. [18]G. Lee, J. R. Arrington, and R. J. Hill, Phys. Rev. D 92, 013013 (2015), 1505.01489.
  19. [19]A. V. Gramolin and R. L. Russell, Phys. Rev. D 105, 054004 (2022), 2102.13022.
  20. [20]J. M. Alarcón, D. W. Higinbotham, and C. Weiss, Phys. Rev. C 102, 035203 (2020), 2002.05167.
  21. [21]H. Atac, M. Constantinou, Z. E. Meziani, M. Paolone, and N. Sparveris, Eur. Phys. J. A 57, 65 (2021), 2009.04357.
  22. [22]Y.-H. Lin, H.-W. Hammer, and U.-G. Meißner, Phys. Lett. B 816, 136254 (2021), 2102.11642.
  23. [23]Z.-F. Cui, D. Binosi, C. D. Roberts, and S. M. Schmidt (2021), 2102.01180.
  24. [24]J. Zhou, V. Khachatryan, H. Gao, S. Gorbaty, and D. W. Higinbotham (2021), 2110.02557.
  25. [25]A. Karch, E. Katz, D. T. Son, and M. A. Stephanov, Phys. Rev. D 74, 015005 (2006), hep-ph/0602229.
  26. [26]Z. Abidin and C. E. Carlson, Phys. Rev. D 79, 115003 (2009), 0903.4818.
  27. [27]R. Pohl et al., Nature 466, 213 (2010).
  28. [28]A. Beyer et al., Science 358, 79 (2017).
  29. [29]P. J. Mohr, D. B. Newell, and B. N. Taylor, CODATA recommended values of the fundamental physical constants: 2018. (2019), URL http://physics.nist.gov/constants.
  30. [30]X. Zhan et al., Phys. Lett. B 705, 59 (2011), 1102.0318.
  31. [31]P. J. Mohr, D. B. Newell, and B. N. Taylor, Journal of Physical and Chemical Reference Data 45, 043102 (2016), URL https://doi.org/10.1063/1.4954402.
  32. [32]A. J. Krasznahorkay et al., Phys. Rev. Lett. 116, 042501 (2016), 1504.01527.
  33. [33]C. E. Carlson and B. C. Rislow, Phys. Rev. D 86, 035013 (2012), 1206.3587.
  34. [34]I. Zahed (2021), 2102.08191.
  35. [35]P. H. Frampton, Phys. Rev. D 1, 3141 (1970).
  36. [36]J. M. Maldacena, Adv. Theor. Math. Phys. 2, 231 (1998), hep-th/9711200.
  37. [37]O. Aharony, S. S. Gubser, J. M. Maldacena, H. Ooguri, and Y. Oz, Phys. Rept. 323, 183 (2000), hep-th/9905111.
  38. [38]E. Witten, Adv. Theor. Math. Phys. 2, 505 (1998), hep-th/9803131.
  39. [39]G. P. Lepage and S. J. Brodsky, Phys. Rev. D 22, 2157 (1980).
  40. [40]A. Vega, I. Schmidt, T. Gutsche, and V. E. Lyubovitskij, Phys. Rev. D 83, 036001 (2011), 1010.2815.
  41. [41]J. Arrington, W. Melnitchouk, and J. A. Tjon, Phys. Rev. C 76, 035202 (2007), 0707.1861.
  42. [42]J. C. Bernauer et al. (A1), Phys. Rev. C 90, 015206 (2014), 1307.6227.
  43. [43]M. Paolone et al., Phys. Rev. Lett. 105, 072001 (2010), 1002.2188.
  44. [44]J. Arrington and I. Sick, J. Phys. Chem. Ref. Data 44, 031204 (2015), 1505.02680.
  45. [45]P. A. Zyla et al. (Particle Data Group), PTEP 2020, 083C01 (2020).
  46. [46]H. Atac, M. Constantinou, Z. E. Meziani, M. Paolone, and N. Sparveris, Nature Commun. 12, 1759 (2021), 2103.10840.
  47. [47]J. J. Kelly, Phys. Rev. C 70, 068202 (2004).
  48. [48]H. R. Grigoryan and A. V. Radyushkin, Phys. Rev. D 76, 095007 (2007), 0706.1543.
  49. [49]P. Colangelo, F. De Fazio, F. Giannuzzi, F. Jugeau, and S. Nicotri, Phys. Rev. D 78, 055009 (2008), 0807.1054.
  50. [50]H. Forkel, Phys. Rev. D 78, 025001 (2008), 0711.1179.
  51. [51]N. R. F. Braga and A. Vega, Eur. Phys. J. C 72, 2236 (2012), 1110.2548.
  52. [52]K. A. Mamo and I. Zahed, Phys. Rev. D 104, 066010 (2021), 2102.00608.
  53. [53]K. A. Mamo and I. Zahed, Phys. Rev. D 101, 086003 (2020), 1910.04707.
  54. [54]K. A. Mamo and I. Zahed, Phys. Rev. D 103, 094010 (2021), 2103.03186.
  55. [55]K. A. Mamo and I. Zahed, Phys. Rev. D 104, 066023 (2021), 2106.00722.
  56. [56]K. A. Mamo and I. Zahed, Phys. Rev. D 101, 066013 (2020), 1807.07969.
  57. [57]K. A. Mamo and I. Zahed, Phys. Rev. D 101, 066014 (2020), 1905.07864.
  58. [58]J. Hirn and V. Sanz, JHEP 12, 030 (2005), hep-ph/0507049.
  59. [59]S. K. Domokos, H. R. Grigoryan, and J. A. Harvey, Phys. Rev. D 80, 115018 (2009), 0905.1949.
  60. [60]A. Cherman, T. D. Cohen, and E. S. Werbos, Phys. Rev. C 79, 045203 (2009), 0804.1096.
  61. [61]D. K. Hong, T. Inami, and H.-U. Yee, Phys. Lett. B 646, 165 (2007), hep-ph/0609270.
  62. [62]H. Nastase, Introduction to the ADS/CFT Correspondence (Cambridge University Press, 2015), ISBN 978-1-107-08585-5, 978-1-316-35530-5.

Paper details

vlnc:W_bsxwe38b
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