betaFeedback

WorksPhysics and Astronomy

Light-quark mass dependence of the nucleon axial charge and pion-nucleon scattering phenomenology

Abstract

The light-quark mass dependence of the nucleon axial isovector charge (gAg_A) has been studied up to next-to-next-to-leading order, O(p4)O(p^4), in relativistic chiral perturbation theory using extended-on-mass-shell renormalization, without and with explicit Δ(1232)Δ(1232) degrees of freedom. We show that in the ΔΔ-less case, at this order, the flat trend of gA(Mπ)g_A (M_π) exhibited by state-of-the-art lattice QCD (LQCD) results cannot be reproduced using low energy constants (LECs) extracted from pion-nucleon elastic and inelastic scattering. A satisfactory description of these LQCD data is only achieved in the theory with ΔΔ. From this fit we report gA(Mπ(phys))=1.260±0.012g_A (M_{π\rm{(phys)}}) = 1.260 \pm 0.012, close to the experimental result, and d16=−0.88±0.88d_{16}= -0.88\pm 0.88 GeV−2^{-2}, in agreement with its empirical value. The large uncertainties are of theoretical origin, reflecting the difference between O(p3)O(p^3) and O(p4)O(p^4) that still persists at large MπM_π in presence of the ΔΔ.

Introduction

The axial isovector charge gAu−dg_A^{u-d} (gAg_A from now on) is a fundamental property of the nucleon related to the difference in the spin fractions carried by uu and dd quarks. With a magnitude precisely determined from neutron β\beta decay [1], the nucleon axial charge represents a benchmark for non-perturbative studies of Quantum Chromodynamics in the lattice (LQCD) alongside with other nucleon properties, such as scalar and tensor charges, electromagnetic form factors and parton distribution functions (see Ref. [3] and Sec. 10 of Ref. [4] for recent reviews). Over the last decade, progress in this direction has been significant [4], owing to better computational resources, improved algorithms and techniques to reduce systematic errors, in particular those induced by excited-state contamination, which can be considerable in the baryon sector for currently available lattice ensembles [5, 6].

As the effective theory of QCD in the non-perturbative regime, chiral perturbation theory (ChPT) plays an important role. It allows to determine the light (uu, dd) quark mass dependence of low-energy hadronic observables in terms of low-energy constants (LECs) and perform model-independent extrapolations of LQCD results to the physical point. ChPT also allows to account for finite lattice-volume and lattice spacing corrections in a systematic way [7, 8]. It has also proved helpful to deal with the contamination from excited states [5, 6, 9]. The interplay between ChPT and LQCD can also be used to determine poorly known LECs, which are difficult to access with experimental data. This is the case of d16d_{16} present in the O(p3)\mathcal{O}(p^3) part of the πN\pi N Lagrangian. Via a 4d16Mπ24d_{16}M_\pi^2 term, this LEC controls the slope of the light-quark mass dependence of gAg_A. Therefore, its extraction from LQCD results at low pion masses is only natural. LEC d16d_{16} has been identified as one of the most significant sources of uncertainty in quark mass dependence of nuclear properties such as ground-state and binding energies through long-range nuclear forces [10, 11, 12].

Alternatively, LECs have been extracted from experimental information [3] on various processes such as pion photo- and electroproduction [13, 14, 15, 16] but, predominantly, from pion-nucleon (πN\pi N) scattering [17, 18, 19, 20, 21, 22, 23, 24, 25]. Although d16d_{16} cannot be extracted from πN\pi N elastic scattering, it contributes significantly to inelastic, πN→ππN\pi N \to\pi\pi N, scattering. Indeed, in a combined fit to both elastic and inelastic πN\pi N scattering data d16d_{16} has been recently extracted, albeit with a large uncertainty [26]. Baryon ChPT (BChPT) has been used to calculate the axial charge within the heavy-baryon [27, 28, 29, 30, 31] and relativistic [22, 23, 30, 32, 33, 34, 35] approaches. ChPT approaches both without and with the Δ(1232)\Delta(1232) as an explicit degree of freedom. In the relativistic framework, both infrared and extended-on-mass-shell (EOMS) regularization methods have been applied. Some of these studies also address the nucleon axial form factor at low four-momentum transfers squared [32, 33, 34, 35]. The MπM_\pi dependence of gAg_A has been specifically investigated in Refs. [28, 29, 30, 31, 34, 35].

Armed with relativistic baryon ChPT, we revisit this problem up to next-to-next-to-leading order (NNLO) in the perturbative expansion. The EOMS scheme ensures that not only power counting but also analytic properties of loop functions are properly preserved. After describing in § 1.1 the details of the ChPT calculation and introducing the relevant terms of the effective Lagrangian we show in Sec. III the MπM_\pi dependence of gAg_A predicted with the phenomenological LECs obtained in Ref. [26] from πN\pi N elastic and inelastic scattering. In order to obtain a better description of recent LQCD data we have performed fits from which some of the LECs and, in particular, d16d_{16}, are determined. These are presented in § 1.3. A meaningful representation of the flat trend exhibited by the LQCD results is only achieved with explicit Δ(1232)\Delta(1232). Differences between orders, which are considerable between NLO and NNLO, are used to provide a measure of the systematic error arising from the truncation of the perturbative series. These differences are the main source of uncertainty in the extracted d16d_{16} value.

The nucleon axial charge in relativistic BChPT

In the isospin limit, the matrix element of the axial isovector current

Aμa(x)=qˉ(x)γμγ5τa2q(x),(1)A_\mu^a(x)=\bar{q}(x)\gamma_\mu\gamma_5\frac{\tau^a}{2}q(x), \tag*{(1)}

with q=(u,d)Tq=(u,d)^T the quark-field doublet, taken between on-shell nucleon states of equal four-momenta pp (p2=mN2p^2=m_N^2), can be written as

⟨N(p)∣Aμa(0)∣N(p)⟩=uˉ(p)gAγμγ5τa2u(p).(2)\langle N(p)|A_\mu^a(0)|N(p)\rangle=\bar{u}(p)g_A\gamma_\mu\gamma_5\frac{\tau^a}{2}u(p). \tag*{(2)}

The isovector character of the current is manifest given the presence of the Pauli isospin matrices τa\tau^a. We calculate the axial charge gAg_A employing relativistic BChPT with pions, nucleons and Δ\Delta as degrees of freedom. We use the standard power counting [36], extended to diagrams with Δ\Delta baryons following small scale expansion (SSE) of Refs. [37, 38]. In this power counting, the mass difference δ=mN−mΔ\delta=m_N-m_\Delta is O(p)\mathcal{O}(p).

Relevant terms of the effective Lagrangian

In this section the terms in the effective Lagrangian, Leff\mathcal{L}_{\mathrm{eff}}, required for our calculation are presented. We need

Leff⊃LπN(1)+LπΔ(1)+LπNΔ(1)+LπN(2)+LπNΔ(2)+LπΔ(2)+LπN(3),(3)\mathcal{L}_{\mathrm{eff}}\supset\mathcal{L}_{\pi N}^{(1)}+\mathcal{L}_{\pi\Delta}^{(1)}+\mathcal{L}_{\pi N\Delta}^{(1)}+\mathcal{L}_{\pi N}^{(2)}+\mathcal{L}_{\pi N\Delta}^{(2)}+\mathcal{L}_{\pi\Delta}^{(2)}+\mathcal{L}_{\pi N}^{(3)}, \tag*{(3)}

where superscripts indicate the chiral order and subscripts show the present degrees of freedom. The terms in LπN(1,3)\mathcal{L}_{\pi N}^{(1,3)}, LπΔ(1)\mathcal{L}_{\pi\Delta}^{(1)} and LπNΔ(1)\mathcal{L}_{\pi N\Delta}^{(1)} that are relevant for our study have been displayed in Ref. [34] using the same notation adopted here. In addition, O(p2)\mathcal{O}(p^2) contributions are entailed to give gAg_A at NNLO. Following the notation of Ref. [39], the required terms of the πN\pi N Lagrangian are

LπN(2)⊃Ψˉ(c1⟨χ+⟩−c28m˚2(⟨uμuν⟩{Dμ,Dν}+h.c.)+c32⟨uμuμ⟩+ic44[uμ,uν]σμν)Ψ,(4)\mathcal{L}_{\pi N}^{(2)}\supset\bar{\Psi}\left(c_1\langle\chi_+\rangle-\frac{c_2}{8\mathring{m}^2}\left(\langle u_\mu u_\nu\rangle\{D^\mu,D^\nu\}+\mathrm{h.c.}\right)+\frac{c_3}{2}\langle u_\mu u^\mu\rangle+\frac{\mathrm{i}c_4}{4}[u_\mu,u_\nu]\sigma^{\mu\nu}\right)\Psi, \tag*{(4)}

where Ψ\Psi is the isospin doublet of nucleon fields. Including only isovector axial external fields aμaτa/2≡aμa_\mu^a\tau^a/2\equiv a_\mu, we end up with

Dμ=∂μ+Γμ,Γμ=12[u†(∂μ−iaμ)u+u(∂μ+iaμ)u†],uμ=i[u†(∂μ−iaμ)u−u(∂μ+iaμ)u†],(5)\begin{aligned} D_\mu&=\partial_\mu+\Gamma_\mu,\\ \Gamma_\mu&=\frac{1}{2}\left[u^\dagger(\partial_\mu-\mathrm{i}a_\mu)u+u(\partial_\mu+\mathrm{i}a_\mu)u^\dagger\right],\\ u_\mu&=\mathrm{i}\left[u^\dagger(\partial_\mu-\mathrm{i}a_\mu)u-u(\partial_\mu+\mathrm{i}a_\mu)u^\dagger\right], \tag*{(5)} \end{aligned}

χ+=u†χu†+uχ†u\chi_+=u^\dagger\chi u^\dagger+u\chi^\dagger u, with χ=diag⁡(Mπ2,Mπ2)\chi=\operatorname{diag}(M_\pi^2,M_\pi^2), and ⟨…⟩\langle\ldots\rangle denotes a trace over isospin. In LπNΔ(2)\mathcal{L}_{\pi N\Delta}^{(2)}, after redundant terms are eliminated from Eq. (67) of Ref. [40] (see also Sec. 3.1 of Ref. [41] and the Appendix of Ref. [42]) only the following monomials

LπNΔ(2)⊃Ψˉkξki3/2{b42ωαiωβjγβγ5τj+b52ωαiωβjγβγ5τj}Ψ+h.c.(6)\mathcal{L}_{\pi N\Delta}^{(2)}\supset\bar{\Psi}_k\xi_{ki}^{3/2}\left\{\frac{b_4}{2}\omega_\alpha^i\omega_\beta^j\gamma^\beta\gamma_5\tau^j+\frac{b_5}{2}\omega_\alpha^i\omega_\beta^j\gamma^\beta\gamma_5\tau^j\right\}\Psi+\mathrm{h.c.} \tag*{(6)}

contribute to the nucleon axial charge. Here Ψμ\Psi_\mu denotes the Rarita-Schwinger field of the Δ\Delta resonance. States {ξij3/2Ψμj}i=1−3\{\xi_{ij}^{3/2}\Psi_\mu^j\}_{i=1-3}, where ξij3/2=δij−τiτj/3\xi_{ij}^{3/2}=\delta_{ij}-\tau_i\tau_j/3 are isospin-3/2 projectors, are also isospin doublets, whose explicit expressions in terms of the physical Δ\Delta states are derived, for example, in Appendix A of Ref. [43]. The isovector projection of uμu^\mu is represented by ωμ,i=12⟨τiuμ⟩\omega^{\mu,i} = \frac{1}{2}\langle\tau^i u^\mu\rangle (see for instance Ref. [40]). Finally, also at O(p2)\mathcal{O}(p^2) [40, 44],

LπΔ(2)⊃Ψˉμiξij3/2{a1⟨χ+⟩δjkgμν}ξkl3/2Ψνl.(7)\mathcal{L}_{\pi\Delta}^{(2)} \supset\bar{\Psi}_{\mu}^{i}\xi_{ij}^{3/2}\left\{a_1\langle\chi_+\rangle\delta^{jk}g^{\mu\nu}\right\}\xi_{kl}^{3/2}\Psi_{\nu}^{l}. \tag*{(7)}

introduces an MπM_\pi dependent correction to the Δ\Delta mass, in the same way as the term proportional to c1c_1 does for the nucleon mass.

Perturbative calculation

The set of Feynman diagrams that contribute to the axial form factor and, in particular, to gAg_A are displayed in Figure 1 and Figure 2. The LECs that each of these diagrams brings about are listed in Table I. At O(p4)\mathcal{O}(p^4), there are contributions from O(p2)\mathcal{O}(p^2) vertices but also baryon (NN, Δ\Delta) mass insertions. The later are calculated perturbatively, i.e. we evaluate directly diagrams (i), (j), (l)-(n) of Figure 2, avoiding Dyson re-summations to all orders in the propagators. Alternative choices have been considered in Refs. [33, 35].

Figure 1. Diagrams at orders O(p)\mathcal{O}(p), (a), and O(p3)\mathcal{O}(p^3), (b)-(g), contributing to the nucleon axial charge. Dashed, solid single and double lines denote pions, nucleons and Δ\Delta resonances in that order; wiggly lines stand for external axial fields. The open circle represents an O(p3)\mathcal{O}(p^3) vertex, while the rest of the vertices are O(p)\mathcal{O}(p). Permutations of diagrams (d) and (f) have been omitted in the figure.

Figure 2. O(p4)\mathcal{O}(p^4) diagrams contributing to the nucleon axial charge. Line styles are the same as in Figure 1. Filled circles denote O(p2)\mathcal{O}(p^2) vertices. Permutations of all these diagrams are taken into account but not explicitly represented.

Although not diagrammatically represented, nucleon wave-function renormalization is taken into account in the standard way. To the order of our calculation, only the O(p)\mathcal{O}(p) diagram (a), giving just the axial charge in the chiral limit g˚A\mathring{g}_A, should be multiplied by the wave-function renormalization constant ZNZ_N, calculated from the O(p4)\mathcal{O}(p^4) nucleon self-energy.

ZN≈1+∂ΣN(4)∂p̸∣p̸=mN.(8)Z_N \approx1+\left.\frac{\partial\Sigma_N^{(4)}}{\partial\not{p}}\right|_{\not{p}=m_N}. \tag*{(8)}

Additional O(p3)\mathcal{O}(p^3) and O(p4)\mathcal{O}(p^4) contributions to gAg_A are generated in this way. The LECs that enter in them are also listed in Table I.

In order to absorb the ultraviolet divergencies generated by loops, we perform dimensional renormalization in the MS‾\overline{\mathrm{MS}} scheme [45], at the scale μ=m˚\mu=\mathring{m}, where m˚\mathring{m} denotes the nucleon mass in the chiral limit. An additional renormalization is then performed to account for the power-counting breaking caused by the presence of baryon masses which do not vanish in the chiral limit. Among the available schemes we employ EOMS renormalization [45] which consists in the absorption of the power counting breaking (PCB) terms in a redefinition of the LECs. In this way, power counting is restored without altering the analytic properties of the loops and preserving covariance. Note that, as in earlier studies [24, 34], PCB terms are identified and subtracted in an expansion in powers or MπM_\pi but not in δ\delta. The EOMS renormalization shifts for the LECs are lengthy, so we have included them in the supplementary material.

DiagramsO(p)\mathcal{O}(p)O(p2)\mathcal{O}(p^2)O(p3)\mathcal{O}(p^3)
(a), (c), (d)g˚A\mathring{g}_A--
(b)--d16d_{16}
(e)g˚A\mathring{g}_A--
(f)g˚A\mathring{g}_A, hAh_A--
(g)g1g_1, hAh_A--
(h)g˚A\mathring{g}_Ac2−4c_{2-4}-
(i)g˚A\mathring{g}_Ac1c_1-
(j)g˚A\mathring{g}_Ac1c_1-
(k)hAh_Ab4,5b_{4,5}-
(l)g˚A\mathring{g}_A, hAh_Aa1a_1-
(m)g1g_1, hAh_Aa1a_1-
(n)g˚A\mathring{g}_A, hAh_Ac1c_1-
wfrg˚A\mathring{g}_A, hAh_Ac1c_1, a1a_1-

Table 1. LECs introduced by Feynman diagrams in Figure 1, Figure 2 and by wave-function renormalization (wfr).

In the way outlined above we obtain the axial charge within the EOMS renormalization scheme up to O(p4)\mathcal{O}(p^4) with explicit Δ\Delta. Our result has the following structure, with superindices indicating the chiral order:

gA=g˚A+4d16Mπ2+gA(loop)(3)Δ(g˚A;Mπ)+gA(loop)(3)Δ(g˚A,hA,g1;Mπ)+gA(loop)(4)Δ(g˚A,c1,c2,c3,c4;Mπ)+gA(loop)(4)Δ(g˚A,hA,g1,c1,a1,b4,b5;Mπ)+O(p5).(9)g_A = \mathring{g}_A + 4d_{16}M_\pi^2 + g_{A(\mathrm{loop})}^{(3)\Delta}(\mathring{g}_A;M_\pi) + g_{A(\mathrm{loop})}^{(3)\Delta}(\mathring{g}_A,h_A,g_1;M_\pi) + g_{A(\mathrm{loop})}^{(4)\Delta}(\mathring{g}_A,c_1,c_2,c_3,c_4;M_\pi) + g_{A(\mathrm{loop})}^{(4)\Delta}(\mathring{g}_A,h_A,g_1,c_1,a_1,b_4,b_5;M_\pi) + \mathcal{O}(p^5). \tag*{(9)}

The structures arising from the loops are preserved at the price of keeping higher order terms. The O(p3)\mathcal{O}(p^3) part of gAg_A is given in Eq. (21) of Ref. [34]. In Appendix § 1.1.2 we provide the O(p4)\mathcal{O}(p^4) contribution from wave-function renormalization. The length of the rest of the O(p4)\mathcal{O}(p^4) expression, which depends on several LECs, discourages us from giving it explicitly but is available in a Mathematica notebook as supplementary material. We nonetheless identify that at O(p4)\mathcal{O}(p^4), i. e. at O(Mπ3)\mathcal{O}(M_\pi^3) for gAg_A, c3c_3 and c4c_4 enter in the following combination c~4=c4−c3/2\tilde{c}_4 = c_4 - c_3/2, while c2c_2 enters only at O(p5)\mathcal{O}(p^5) [O(Mπ4)\mathcal{O}(M_\pi^4) in gAg_A]. Although the rather involved contribution of diagram (k) does not, a priori, show the factorization of any mixture of b4b_4 and b5b_5 LECs, after expanding in MπM_\pi, one finds that the combination which actually enters at O(p4)\mathcal{O}(p^4), is b~4=b4+(12/13)b5\tilde{b}_4 = b_4 + (12/13)b_5. At O(p3)\mathcal{O}(p^3) we reproduce the results of Eq. (21) of Ref. [34]. At O(p4)\mathcal{O}(p^4) we agree with the Δ\Delta EOMS expressions given in Refs. [22, 33] except for the different treatment of nucleon mass insertions, which in our case, do not include resummations. As a result, c1Mπ2c_1M_\pi^2 factors appear only linearly in our calculation but not at all orders. Furthermore, a PCB term proportional to c4c_4 present in Ref. [22] after the EOMS renormalization is absent in our final result.

Light-quark mass dependence of the axial coupling from phenomenological input

Once the framework and the chiral order are established, the MπM_\pi dependence of gAg_A is ultimately determined by the LEC values. In Ref. [26], elastic πN\pi N and inelastic πN→ππN\pi N \to\pi\pi N scattering has been studied up to O(p4)\mathcal{O}(p^4) in covariant ChPT using a modified version of the EOMS approach [25]. The LECs that enter gA(Mπ)g_A(M_\pi) at O(p4)\mathcal{O}(p^4) in the Δ\Delta-less model were extracted, including d16d_{16}, owing to the inclusion of low energy total and double differential πN→ππN\pi N \to\pi\pi N cross section data in the combined analysis. To make a prediction of gA(Mπ)g_A(M_\pi) based on this phenomenological input, we should transform the LECs from the modified EOMS of Refs. [25, 26] to the conventional one adopted here. To the order we are working at, this transformation, whose details are disclosed in Appendix A, affects the numerical value of d16d_{16} but not of c1−4c_{1-4}. The axial coupling in the chiral limit, g˚A\mathring{g}_A, which is not extracted in Refs. [25, 26], is determined as

gA(Mπ=Mπ(phys),g˚A,d16,ci)=gA(phys)=1.2754(13)exp⁡(2)RC(10)g_A(M_\pi= M_\pi(\mathrm{phys}),\mathring{g}_A,d_{16},c_i) = g_A(\mathrm{phys}) = 1.2754(13)_{\exp}(2)_{\mathrm{RC}} \tag*{(10)}

from the experimental value, precisely known from β\beta decay [1]. Up to higher orders for gAg_A, the remaining parameters, i.e. the pion decay constant and the nucleon mass in the chiral limit, have been fixed to $F_\pi \simeq F_\pi(\mathrm{phys}) = 92.2\ \mathrm{MeV} and m˚∼mN(phys)+4c1Mπ(phys)2\mathring{m} \sim m_N^{(\mathrm{phys})} + 4c_1M_{\pi(\mathrm{phys})}^2, with Mπ(phys)=135M_{\pi(\mathrm{phys})} = 135 MeV and mN(phys)=939m_N^{(\mathrm{phys})} = 939 MeV, respectively. The numerical values for d16d_{16}, c1−4c_{1-4} as well as g˚A\mathring{g}_A and their uncertainties are summarized in Table II.

O(p3)\mathcal{O}(p^3)O(p4)\mathcal{O}(p^4)
g˚A\mathring{g}_A1.251±0.0511.251 \pm0.0511.089±0.0301.089 \pm0.030
d16d_{16} (GeV−2^{-2})−2.2±1.1-2.2 \pm1.1−1.86±0.80-1.86 \pm0.80
c1c_1 (GeV−1^{-1})-−0.89±0.06-0.89 \pm0.06
c2c_2 (GeV−1^{-1})-3.38±0.153.38 \pm0.15
c3c_3 (GeV−1^{-1})-−4.59±0.09-4.59 \pm0.09
c4c_4 (GeV−1^{-1})-3.31±0.133.31 \pm0.13

Table II. LECs used to obtain the gA(Mπ)g_A(M_\pi) dependence in the Δ\Delta case.

The resulting gA(Mπ)g_A(M_\pi) at O(p3)\mathcal{O}(p^3) and O(p4)\mathcal{O}(p^4) are displayed in Figure 3 together with a subset of recent LQCD determinations 4. The curves are accompanied by 1σ1\sigma statistical error bands arising from the uncertainties in the LECs of Table II assuming they are Gaussian-distributed. Unaccounted correlations are negligible because the uncertainty is strongly dominated by the d16d_{16} error. The O(p3)\mathcal{O}(p^3) result shows agreement with the lattice determinations, albeit with

Figure 3. Pion-mass dependence of gAg_A at O(p3)\mathcal{O}(p^3) (red) and O(p4)\mathcal{O}(p^4) (blue) using phenomenological input from Ref. [26] and 1σ1\sigma error bands. The LQCD data points are from CalLat 18 [46] (black circles), Mainz 19 [47] (red crosses), RQCD 19 [9] (green triangles) and NME 21 [48] (blue squares).

increasing tension as MπM_\pi grows and a curvature in the central value that is absent in the data. On the other hand, it is immediately apparent that the O(p4)\mathcal{O}(p^4) prediction not only does not improve the O(p3)\mathcal{O}(p^3) but plainly fails to describe the MπM_\pi dependence of LQCD data. We have checked that alternative c1−4c_{1-4} determinations from earlier πN\pi N analyses using EOMS [22, 24] do not mitigate the steep rise of gA(Mπ)g_A(M_\pi) at O(p4)\mathcal{O}(p^4). Therefore, the inability to reconcile the light-quark mass dependence of g˚A\mathring{g}_A at O(p4)\mathcal{O}(p^4) with phenomenology, earlier observed in non-relativistic heavy-baryon ChPT [29, 30], is also a feature of the relativistic version of the theory.

The problem with the O(p4)\mathcal{O}(p^4) result might be caused by an accidental slow convergence of gAg_A in BChPT. Instead, one could argue that additional degrees of freedom such as the Δ\Delta [30] or even the Roper N(1440) resonance might solve the issue. In Ref. [26], a version of the theory incorporating the Δ\Delta pole is also considered. However, in order to predict gA(Mπ)g_A(M_\pi) up to O(p4)\mathcal{O}(p^4) with explicit Δ\Delta, additional LECs absent in that study are needed. For this reason we do not tackle the role of the Δ\Delta in the gA(Mπ)g_A(M_\pi) dependence from purely empirical input but postpone it to the next § 1.3 where our fits to recent LQCD data are reported.

Analysis of LQCD results and LEC determination

Data set and fit strategy

The axial charge has been historically underpredicted in the lattice (see for instance Figure 2 of Ref. [49]), but, as a result of conceptual and technical improvements, the majority of LQCD results nowadays agree with the experimental value at the level of a few percent [4]. Most significantly, the analysis of excited-state contamination has notably evolved in last years. Therefore, we only include in our data set results from studies with an improved treatment of these effects. We take into account renormalized {gAi}\{g_A^i\} data at different {Mπi,ai}\{M_\pi^i,a_i\} values, where aa stands for the lattice spacing, from 5 CalLat 18 [46], Mainz 19 [47], RQCD 19 [9] and NME 21 [48].567 Our analysis treats 2+1+1 (CalLat 18) and 2+1 (Mainz 19, RQCD 19, NME 21) ensembles on the same footing, assuming that the cc-quark sea content plays a negligible role. We disregard LQCD determinations of gAg_A coming from q2q^2, finite volume, aa or MπM_\pi extrapolations. Since we do not correct the finite volume effects, we consider only large volumes, satisfying MπL≥3.5M_\pi L \ge3.5, so that the neglected extrapolation is small and can be absorbed in the errors.

In order to gauge the ability of BChPT at O(p3)\mathcal{O}(p^3) and O(p4)\mathcal{O}(p^4) to describe the MπM_\pi dependence of gAg_A, and to extract g˚A\mathring{g}_A and d16d_{16} LECs, we perform fits to the LQCD data set defined above. With this aim we define the following χ2\chi^2:

χ2=∑i(gA(Mπi,ai)−gAi)2(ΔgAi)2+χprior2.(11)\chi^2 = \sum_i \frac{\left(g_A(M_\pi^i,a_i)-g_A^i\right)^2}{\left(\Delta g_A^i\right)^2}+\chi^2_{\mathrm{prior}}. \tag*{(11)}

In addition to gA(Mπ)g_A(M_\pi) from BChPT, our theoretical parametrization incorporates lattice discretization corrections as

gA(Mπi,ai)=gA(Mπi)+xjainj.(12)g_A(M_\pi^i,a_i)=g_A(M_\pi^i)+x_j a_i^{n_j}. \tag*{(12)}

Free parameters xjx_j, with j=1,2,3,4j=1,2,3,4 when lattice data point i∈{CalLat 18,Mainz 19,RQCD 19,NME 21}i\in\{\mathrm{CalLat\ 18},\mathrm{Mainz\ 19},\mathrm{RQCD\ 19},\mathrm{NME\ 21}\} respectively, control the leading aa-dependence of the LQCD data, which is action specific: n1,4=1n_{1,4}=1, while n2,3=2n_{2,3}=2. These discretization corrections are small and do not substantially change the extracted LECs, but result in an appreciable reduction of the fits’ χ2/dof\chi^2/\mathrm{dof}.

Some of the LECs upon which gA(Mπ)g_A(M_\pi) depends are well determined in other studies and are kept fixed, while others are treated as free parameters together with g˚A\mathring{g}_A and d16d_{16}. To improve our description of the LQCD data and reduce correlations [50], for free LECs we assume naturalness Λn−1cn∼1\Lambda^{n-1}c_n\sim1, encoded in Gaussian priors

χprior2=∑free LECs(Λn−1cn5)2.(13)\chi^2_{\mathrm{prior}}=\sum_{\mathrm{free\ LECs}}\left(\frac{\Lambda^{n-1}c_n}{5}\right)^2. \tag*{(13)}

Here, cnc_n generally denotes a LEC of chiral order O(pn)\mathcal{O}(p^n); the breakdown scale is set to Λ=1 GeV∼4πFπ\Lambda=1\ \mathrm{GeV}\sim4\pi F_\pi [51, 52]. We anticipate that a prior for g˚A\mathring{g}_A is superfluous, since its value is always driven to a natural one by low MπM_\pi gALQCDg_A^{\mathrm{LQCD}} data.

The large relative contribution of O(p4)\mathcal{O}(p^4) terms discussed in the previous section and illustrated by Figure 3 is a clear indication that the theoretical error associated with truncation of the perturbative expansion should be taken into account. We follow the method proposed in Ref. [53]. Let XX be an observable with a chiral perturbative expansion

X=X(0)+∑m=1∞ΔX(m);X=X^{(0)}+\sum_{m=1}^{\infty}\Delta X^{(m)};
ΔX(m)=X(m)−X(m−1)\Delta X^{(m)}=X^{(m)}-X^{(m-1)}

encompasses all the monomials that start at order mm. If XX is calculated up to order nn, X≈X(n)X\approx X^{(n)}, assuming that the truncation error is dominated by order n+1n+1, its contribution ΔX(n+1)\Delta X^{(n+1)} can be estimated in a conservative way as [25, 53]

∣ΔX(n+1)∣=max⁡{Qn+1∣X(0)∣,Qn∣ΔX(1)∣,…,Q∣ΔX(n)∣}.\left|\Delta X^{(n+1)}\right|=\max\left\{Q^{n+1}\left|X^{(0)}\right|,Q^n\left|\Delta X^{(1)}\right|,\ldots,Q\left|\Delta X^{(n)}\right|\right\}.

where QQ denotes the expansion variable. For gAg_A, Q=Mπ/ΛQ=M_\pi/\Lambda. Recalling that in this case O(p5)=O(Mπ4)\mathcal{O}(p^5)=\mathcal{O}(M_\pi^4), for our O(p4)\mathcal{O}(p^4) calculation, Eq. (15) gives

ΔgAχ(5)=max⁡{(MπΛ)4∣g˚A∣,(MπΛ)2∣ΔgA(3)∣,MπΛ∣ΔgA(4)∣}∼Q4=O(p5).(14)\Delta g_{A\chi}^{(5)}=\max\left\{\left(\frac{M_\pi}{\Lambda}\right)^4|\mathring{g}_A|,\left(\frac{M_\pi}{\Lambda}\right)^2|\Delta g_A^{(3)}|,\frac{M_\pi}{\Lambda}|\Delta g_A^{(4)}|\right\}\sim Q^4=\mathcal{O}(p^5). \tag*{(14)}

Connecting to Eq. (9), ΔgA(3)=4d16Mπ2+gA(3)(loop)\Delta g_A^{(3)}=4d_{16}M_\pi^2+g_A^{(3)}{}_{\mathrm{(loop)}} and ΔgA(4)=gA(4)(loop)\Delta g_A^{(4)}=g_A^{(4)}{}_{\mathrm{(loop)}}. Based on the results in Sec. III it is easy to foresee that at larger MπM_\pi, ΔgAχ(5)\Delta g_{A\chi}^{(5)} will be determined by the last term in Eq. (14). In our O(p3)\mathcal{O}(p^3) fits we do not assume any prior knowledge about O(p4)\mathcal{O}(p^4) and, therefore, the truncation error is given by

ΔgAχ(4)=max⁡{(MπΛ)3∣g˚A∣,MπΛ∣ΔgA(3)∣}∼Q3=O(p4).(15)\Delta g_{A\chi}^{(4)}=\max\left\{\left(\frac{M_\pi}{\Lambda}\right)^3|\mathring{g}_A|,\frac{M_\pi}{\Lambda}|\Delta g_A^{(3)}|\right\}\sim Q^3=\mathcal{O}(p^4). \tag*{(15)}

In our χ2\chi^2, Eq. (11), this theoretical error is added in quadrature to the one of LQCD points:

(ΔgAi)2=(ΔgALQCDi)2+(ΔgAχ(Mπi))2.(16)(\Delta g_A^i)^2=(\Delta g_{A\mathrm{LQCD}}^i)^2+(\Delta g_{A\chi}(M_\pi^i))^2. \tag*{(16)}

This recipe assigns larger uncertainties to points at high MπM_\pi, where the convergence of the chiral expansion is poorer, therefore reducing their impact on the fit. We should also mention that, as discussed in Refs. [25, 53, 54], although this theoretical error estimate does not have a clear statistical interpretation, results are similar to those obtained in a preferable but beyond the scope of the present study Bayesian approach [50].

Fits are carried out iteratively until convergence is achieved. The first minimization is performed without theoretical errors, which are subsequently evaluated using the LECs determined in the previous step. The lattice discretization parameters xjx_j are fixed in the first iteration because evolving them results in overfitting. With the described procedure, LQCD and truncation errors are not independent and it is not obvious how to combine them in the error for a given observable. Consequently, following Ref. [25], we plot them separately in the error bands for gA(Mπ)g_A(M_\pi). Furthermore, as a quantitative measure of the agreement of our best-fit curve with the LQCD data, we also provide the χ02\chi_0^2 value, defined as

χ02=∑i(gA(Mπi,ai)−gAi)2(ΔgALQCDi)2.(17)\chi_0^2=\sum_i\frac{\left(g_A(M_\pi^i,a_i)-g_{A}^{i}\right)^2}{(\Delta g_{A\mathrm{LQCD}}^i)^2}. \tag*{(17)}

Finally, we have tested the convergence range of our calculation by varying the maximum MπM_\pi of the lattice data included in the fits in the range of Mπcut∈[200,402]M_{\pi\mathrm{cut}}\in[200,402] MeV. A plateau in the χ2\chi^2 and the fitted LECs is found towards the end of the interval. In consequence we report results including all points in the chosen data sets up to Mπ=402M_\pi=402 MeV8. As discussed in Sec. IV C 2, the theoretical error becomes larger at higher MπM_\pi where the convergence is poorer and, therefore, the corresponding LQCD points weight less in the fits.

B. Δ\Delta case

O(p3)\mathcal{O}(p^3)

At O(p3)\mathcal{O}(p^3) the only free parameters (besides the xjx_j governing the aa-dependence) are g˚A\mathring{g}_A and d16d_{16} LECs. The results of the fit are shown in the upper left panel of Figure 4 and the first column of Table III. At the first sight, this model describes LQCD data well, with a good χ2/dof\chi^2/\mathrm{dof}, relatively small errors and natural g˚A\mathring{g}_A and d16d_{16}. As apparent from the comparison of the first two columns of Table II and Table III, the d16d_{16} value is slightly above the phenomenological one extracted in Ref. [26] and with much smaller error. Instead, g˚A\mathring{g}_A is in tension with the value obtained from experimental input, Eq. (10). As a consequence gA(Mπ(phys))=1.205±0.010g_A(M_{\pi(\mathrm{phys})})=1.205\pm0.010, well below the experimental result. However, an inspection of the results for the O(p4)\mathcal{O}(p^4) model described below reveals a large contribution from the O(p4)\mathcal{O}(p^4) terms at Mπ≳200M_\pi\gtrsim200 MeV, in line with the findings of Sec. III. The fact that these terms are considerably larger than the error band of the O(p3)\mathcal{O}(p^3) result implies that, in this case, the theoretical error estimated from O(p1)\mathcal{O}(p^1) and O(p3)\mathcal{O}(p^3) terms falls short in accounting for the O(p4)\mathcal{O}(p^4) contribution. The agreement of gA(Mπ)g_A(M_\pi) at O(p3)\mathcal{O}(p^3) with LQCD points appears then as misleadingly good, while the uncertainties in the LEC values can be regarded as underestimated. In other words, to be realistic, an O(p3)\mathcal{O}(p^3) analysis should be limited to small Mπ<200M_\pi<200 MeV, which would be unfeasible with the small amount of presently available LQCD data in this region.8

Figure 4. Pion-mass dependence of gAg_A obtained by fitting LQCD results with O(p3)O(p^3) and O(p4)O(p^4) relativistic BChPT without and with Δ(1232)\Delta(1232) as explicit degree of freedom. Gray (dark) bands correspond to errors determined by propagating LEC uncertainties. Blue bands represent the estimated theoretical uncertainties ΔgAχ(4,5)\Delta g_{A\chi}^{(4,5)}. The LQCD points from CalLat 18 [46] (black circles), Mainz 19 [47] (red crosses), RQCD 19 [9] (green triangles) and NME 21 [48] (blue squares) are shown at their finite aa values i.e. without (small) discretization corrections.

O(p4)O(p^4)

In the O(p4)O(p^4) Δ\Delta model one has the additional contribution of NLO LECs c1−4c_{1-4}. They were initially allowed to evolve in the fits under the constrains set by their empirical determination [26] (second column of Table II) implemented as Gaussian priors. However, as these LECs are quite well determined, this procedure yields substantially the figures favored by the priors. Therefore, we report here the results of a simpler fit with c1−4c_{1-4} held fixed to their central phenomenological values (second column of Table III).

In any case, as apparent from the lower left panel of Figure 4 the O(p4)O(p^4) Δ\Delta model fails to describe the light-quark mass dependence of gAg_A.9 The small χ2\chi^2 is merely a consequence of the large theoretical error, which reduces the impact of high MπM_\pi points on the fit. Nevertheless, the poor agreement is reflected in the magnitude of χ02\chi_0^2 and also in the quite unnatural d16d_{16} in spite of its prior. This large (in absolute value) d16d_{16} is inconsistent with its phenomenological value and nonetheless unable to correct the MπM_\pi dependence at Mπ≳300 MeVM_\pi\gtrsim300\ \mathrm{MeV}, which is largely dominated by the O(p4)O(p^4) terms and, therefore, very similar to the one displayed in Figure 3.

The fact that the very wide theoretical error band encompasses the lattice points reflects that the disagreement would be removed by a (large) contribution of O(p5)O(p^5) counterbalancing the O(p4)O(p^4) ones. Actually, within heavy-baryon ChPT it has been shown [29] that the curve can be bent down by additional contributions of orders O(p5,6)O(p^{5,6}) with LECs of natural size (see also Fig. 1 of Ref. [11]). Here we take a different avenue and introduce the Δ(1232)\Delta(1232) explicitly as advocated in Ref. [30] based on the Adler-Weisberger sum rule [55, 56] and a heavy-baryon ChPT calculation for gAg_A.

TABLE III: LEC values, both fixed and fitted to LQCD data, in the four different models under study. The χ2/dof\chi^2/\mathrm{dof} for each fit is given at the bottom; χ02\chi^2_0, defined in Eq. (17), does not include theoretical errors or naturalness priors.

O(p3) Δ\mathcal{O}(p^3)\ \DeltaO(p4) Δ\mathcal{O}(p^4)\ \DeltaO(p3) Δ\mathcal{O}(p^3)\ \DeltaO(p4) Δ\mathcal{O}(p^4)\ \Delta
g˚A\mathring{g}_A (free)1.1302±0.00981.1302 \pm0.00981.453±0.0481.453 \pm0.0481.1383±0.00991.1383 \pm0.00991.240±0.0461.240 \pm0.046
d16d_{16} (GeV−2\mathrm{GeV}^{-2}) (free)−0.925±0.055-0.925 \pm0.055−9.77±0.87-9.77 \pm0.871.224±0.0401.224 \pm0.040−0.88±0.88-0.88 \pm0.88
hAh_A--1.351.351.351.35
g1g_1--−2.29-2.29−2.29-2.29
c1c_1 (GeV−1\mathrm{GeV}^{-1})-−0.89±0.06-0.89 \pm0.06-−1.15±0.05-1.15 \pm0.05
c2c_2 (GeV−1\mathrm{GeV}^{-1})-3.38±0.153.38 \pm0.15-1.57±0.101.57 \pm0.10
c3c_3 (GeV−1\mathrm{GeV}^{-1})-−4.59±0.09-4.59 \pm0.09-−2.54±0.05-2.54 \pm0.05
c4c_4 (GeV−1\mathrm{GeV}^{-1})-3.31±0.133.31 \pm0.13-2.61±0.102.61 \pm0.10
a1a_1 (GeV−1\mathrm{GeV}^{-1})---0.900.90
b~4\widetilde{b}_4 (GeV−2\mathrm{GeV}^{-2}) (free)---−12.3±1.0-12.3 \pm1.0
x1x_1 (fm−1\mathrm{fm}^{-1})0.39±0.680.39 \pm0.68-0.38±0.070.38 \pm0.070.21±0.070.21 \pm0.07
x2x_2 (fm−2\mathrm{fm}^{-2})−8.10±1.80-8.10 \pm1.80-−8.17±1.80-8.17 \pm1.80−7.27±1.80-7.27 \pm1.80
x3x_3 (fm−2\mathrm{fm}^{-2})2.25±1.832.25 \pm1.83-2.17±1.832.17 \pm1.833.28±1.843.28 \pm1.84
x4x_4 (fm−1\mathrm{fm}^{-1})0.61±0.110.61 \pm0.11-0.60±0.110.60 \pm0.110.51±0.110.51 \pm0.11
m˚\mathring{m} (GeV\mathrm{GeV})0.8740.8740.8740.8740.8550.8550.8550.855
m˚Δ\mathring{m}_{\Delta} (GeV\mathrm{GeV})--1.1661.1661.1661.166
χ2/dof\chi^2/\mathrm{dof}36.06/(43−6)=0.9836.06/(43-6)=0.9813.31/(43−2)=0.3313.31/(43-2)=0.3337.60/(43−6)=1.0237.60/(43-6)=1.0211.14/(43−7)=0.3111.14/(43-7)=0.31
χ02/dof\chi^2_0/\mathrm{dof}424.87/(43−6)=11.48424.87/(43-6)=11.48122820.67/(43−2)=2995.63122820.67/(43-2)=2995.63439.19/(43−6)=11.87439.19/(43-6)=11.87501.62/(43−7)=13.93501.62/(43-7)=13.93

Table 3. LEC values, both fixed and fitted to LQCD data, in the four different models under study. The χ2/dof\chi^2/\mathrm{dof} for each fit is given at the bottom; χ02\chi^2_0, defined in Eq. (17), does not include theoretical errors or naturalness priors.

C case

With the Δ(1232)\Delta(1232) as explicit degree of freedom, new contributions with additional LECs arise. We fix the LπNΔ(1)\mathcal{L}^{(1)}_{\pi N\Delta} coupling hAh_A to its large-NcN_c value hA=1.35h_A=1.35 [26], which is close to its empirical value [24]. For the LπΔ(1)\mathcal{L}^{(1)}_{\pi\Delta} coupling g1g_1, whose impact on gAg_A is small, the large-NcN_c limit yields ∣g1∣=2.29\lvert g_1\rvert=2.29 [24, 26]. We are unable to discern its sign in our fits to gAg_A LQCD data and opt for g1=−2.29g_1=-2.29, preferred both by πN\pi N elastic scattering [24] and our own studies of the nucleon axial form factor at low q2q^2 to be reported elsewhere. Finally, in analogy to the nucleon case, m˚Δ≃mΔ(phys)−4a1Mπ(phys)2\mathring{m}_{\Delta}\simeq m_{\Delta(\mathrm{phys})}-4a_1M_{\pi(\mathrm{phys})}^2 with mΔ(phys)=1232 MeVm_{\Delta(\mathrm{phys})}=1232\ \mathrm{MeV} and a1=0.90 GeV−1a_1=0.90\ \mathrm{GeV}^{-1} from the LQCD MπM_{\pi} dependence of the Δ\Delta mass [57].

O(p3)

With this model, the result of the fit for gA(Mπ)g_A(M_{\pi}) closely resembles the O(p3) Δ\mathcal{O}(p^3)\ \Delta one as can be seen in the comparison of the upper panels of Figure 3. However, the value of d16d_{16} changes considerably, including its sign, with respect to the O(p3) Δ\mathcal{O}(p^3)\ \Delta one. The same feature was obtained when this LEC was extracted from LQCD results for the axial form factor at low q2q^2 with O(p3)\mathcal{O}(p^3) relativistic BChPT [34]. The O(p4)\mathcal{O}(p^4) fit described in the next section produces O(p4)\mathcal{O}(p^4) terms larger than the theoretical error estimated from O(p1)\mathcal{O}(p^1) and O(p3)\mathcal{O}(p^3) ones (although to a lesser extent than in the Δ\Delta case) making the extension to O(p4)\mathcal{O}(p^4) desirable for a more realistic determination of LECs and their uncertainties.

O(p4)

In the fit of the O(p4)\mathcal{O}(p^4) model with explicit Δ\Delta we fix the c1−4c_{1-4} LECs to the values extracted from the πN\pi N scattering [25] (see the last column of Table III). They take the Δ\Delta pole into account and are in good agreement with the joint πN+πN\pi N+\pi N fit from Ref. [26]. In addition to g˚A\mathring{g}_A and d16d_{16} we now have O(p2)\mathcal{O}(p^2) b4b_4 and b5b_5 LECs as free parameters. As mentioned in § 1.1 B, these LECs appear at O(p4)\mathcal{O}(p^4) [more precisely, at O(Mπ4/δ)\mathcal{O}(M_{\pi}^4/\delta)] in the combination b~4=b4+(12/13)b5\widetilde{b}_4=b_4+(12/13)b_5. Therefore we keep only b~4\widetilde{b}_4 as a free parameter of the fit and neglect remaining higher order monomials proportional to b5b_5.

The result of the fit, depicted in the lower right panel of Figure 4, satisfactorily describes the trend of gA(Mπ)g_A(M_{\pi}) as predicted by LQCD up to relatively large MπM_{\pi}. The theoretical error is large and rapidly increasing with MπM_{\pi} due to the large contribution of O(p4)O(p^4) terms. Nevertheless, the description of LQCD data and convergence are notably improved with respect to the O(p4)O(p^4) Δ\Delta model.

The extracted LECs are given in Table III. The b~4\tilde b_4 value might seem unnatural but one should keep in mind that it is a combination of LECs. As shown in Table IV the correlations among LECs are sizable; they are an indication of degeneracy among the parameters that could be partially lifted by adding a new dimension to the fit (i.e. studying the axial form factor at finite q2q^2). The d16d_{16} value obtained from this model, d16=−0.88±0.88 GeV−2d_{16}=-0.88\pm0.88\ \mathrm{GeV}^{-2}, is in good

g~A\tilde g_Ad16d_{16}b~4\tilde b_4
g~A\tilde g_A1-0.970.79
d16d_{16}1-0.92
b~4\tilde b_41

Table 4. Correlation matrix for the fit with the O(p4)O(p^4) Δ\Delta model.

agreement with the determination from inelastic πN→ππN\pi N\to\pi\pi N with explicit Δ\Delta pole [26] which, translated to standard EOMS, is d16(pheno)=−1.0±1.0 GeV−2d_{16}(\mathrm{pheno})=-1.0\pm1.0\ \mathrm{GeV}^{-2}. Although, as argued in the Introduction, the MπM_{\pi} dependence of gAg_A is in principle better suited to extract d16d_{16} than the available experimental πN→ππN\pi N\to\pi\pi N data, the convergence issues of the former lead to large errors, comparable to those of the phenomenological result. The g~A\tilde g_A value is higher than in the O(p3)O(p^3) fits and with a larger error. Furthermore, at the physical point gA(Mπ(phys))=1.260±0.012g_A(M_{\pi(\mathrm{phys})})=1.260\pm0.012 is now close to the experimental value although with a much larger uncertainty.

The stability of the results for g~A\tilde g_A and d16d_{16} as a function of the maximum MπM_{\pi} considered, MπcutM_{\pi\mathrm{cut}}, is shown in Figure 5. One can see that for both quantities the numerical values and their errors stabilize for Mπ≳300 MeVM_{\pi}\gtrsim300\ \mathrm{MeV}.

Results of the fits for the $O(p^4)$ model with explicit $\Delta$ as a function of the maximum $M_{\pi\mathrm{cut}}$

Figure 5. Results of the fits for the O(p4)O(p^4) model with explicit Δ\Delta as a function of the maximum MπcutM_{\pi\mathrm{cut}}.

Owing to the consideration of theoretical errors, the analysis could be extended to a broader range of MπM_{\pi} because points at higher MπM_{\pi}, where the convergence is worse, have a larger combined uncertainty, Eq. (16), and contribute less to the fit.

Conclusions and outlook

We have investigated the pion mass dependence of the nucleon axial coupling up to O(p4)≡O(Mπ3)O(p^4)\equiv O(M_{\pi}^3) (NNLO) in relativistic BChPT with EOMS renormalization. We have shown that, at this order and without including Δ(1232)\Delta(1232) explicitly, the pion mass dependence of gAg_A obtained using LECs extracted from phenomenological analyses of pion-nucleon elastic and inelastic (πN→ππN\pi N\to\pi\pi N) scattering cannot describe the rather flat behavior predicted by state-of-the-art LQCD simulations. The disagreement is manifest from pion masses right above the physical point. This feature, earlier observed in non-relativistic heavy-baryon ChPT [29, 30], is therefore also present in the relativistic theory. The fact that O(p4)O(p^4) terms become large from relatively low Mπ≳200 MeVM_{\pi}\gtrsim200\ \mathrm{MeV} implies that O(p3)O(p^3) analyses of gA(Mπ)g_A(M_{\pi}) likely underestimate theoretical uncertainties.

In line with the findings of [30] within heavy-baryon ChPT, we can satisfactorily describe the LQCD data for gA(Mπ)g_A(M_{\pi}) at O(p4)O(p^4) only after the Δ\Delta is incorporated as an explicit degree of freedom. However, although in a much smaller degree than in the Δ\Delta case, a fast increase in the relative size of O(p4)O(p^4) terms with MπM_{\pi} is observed and reflected by the estimate of theoretical uncertainties. This fact jeopardizes the precision of the ChPT description of gA(Mπ)g_A(M_{\pi}) at Mπ≳300M_\pi\gtrsim300 MeV and negatively influences the extraction of LECs based on LQCD data at such MπM_\pi. Together with the sizable correlations, which can be reduced by considering both the MπM_\pi and q2q^2 dependence of the axial form factor in the fits, this feature implies that heavier resonances and/or O(p5)\mathcal{O}(p^5) terms are still required to reach a good convergence and reduce theoretical errors. The impact of setting the baryon masses in the loops to the values obtained by the LQCD simulations may be also worth exploring in view of the results of [35] although this would correspond to the resummation of baryon selfenergy insertions of higher order. For this purpose it would be relevant to have more information about the Δ\Delta pole position for the different lattice ensembles.

From our O(p4)\mathcal{O}(p^4) analysis of recent LQCD data we have obtained gA(Mπ(phys))=1.260±0.012g_A(M_{\pi(\mathrm{phys})}) = 1.260 \pm0.012, close to the experimental value, and d16=−0.88±0.88 GeV−2d_{16} = -0.88 \pm0.88\ \mathrm{GeV}^{-2}, in agreement with πN\pi N phenomenology. As a consequence of the previously discussed issues, errors are still large, particularly for d16d_{16} that is naturally extracted from the light-quark mass dependence of the nucleon axial coupling. New LQCD results at Mπ≲250M_\pi\lesssim250 MeV will improve the precision of the extracted LECs. Besides, our ongoing effort to extend the analysis to the whole axial form factor (at low q2q^2) may shed more light on d16d_{16}, as well as other LECs such as d22d_{22}.

Acknowledgments

We thank G. Guerrero, S. Leupold and M. J. Vicente Vacas for useful discussions, and D. Yao for sharing details of his related calculation with us. E. Epelbaum, A. Gasparyan and H. Krebs kindly answered our questions about their renormalization prescription. We are also indebted to the RQCD team for making their axial form-factor results available to us and to R. Gupta for providing relevant information about the NME ones. This research has been partially supported by the Spanish Ministerio de Ciencia e Innovación under contracts FIS2017-84038-C2-1-P and PID2020-112777GB-I00, the EU STRONG-2020 project under the program H2020-INFRAIA-2018-1, grant agreement no. 824093 and by Generalitat Valenciana under contract PROMETEO/2020/023.

A LEC conversion

The covariant renormalization prescription adopted in Refs. [25, 26] differs from EOMS. In EOMS only power-counting breaking terms are absorbed in the LECs. To obtain an equivalence between covariant and heavy-baryon results, the prescription of Ref. [25] also subtracts infrared-regular terms at the order of the calculation, as well as pieces proportional to log⁡(Mπ2/mN2)\log(M_\pi^2/m_N^2). After setting λ=0\lambda= 0 and μ=mN\mu= m_N in Eq. (21) of Ref. [25] one is left with a dimensionally regularised LEC, xrx_r:

xr=xˉcov+1Fπ2(δxˉf(3)+δxˉf(4))+βx32π2Fπ2log⁡(Mπ(phys)2mN(phys)2)=xEOMS+δxfEOMS(18)x_r = \bar{x}^{\mathrm{cov}} + \frac{1}{F_\pi^2}\left(\delta\bar{x}_f^{(3)} + \delta\bar{x}_f^{(4)}\right) + \frac{\beta_x}{32\pi^2 F_\pi^2}\log\left(\frac{M_{\pi(\mathrm{phys})}^2}{m_{N(\mathrm{phys})}^2}\right) = x^{\mathrm{EOMS}} + \delta x_f^{\mathrm{EOMS}} \tag*{(18)}

which allows to express the renormalized EOMS LECs xEOMSx^{\mathrm{EOMS}} in terms of the corresponding ones, xˉcov\bar{x}^{\mathrm{cov}}, from Refs. [25, 26]. The β\beta-functions βx\beta_x required to cancel the mesonic tadpole terms are listed in Appendix B of Ref. [25] as well as finite shifts δxˉf(3,4)\delta\bar{x}_f^{(3,4)}. Our EOMS finite shifts δxfEOMS\delta x_f^{\mathrm{EOMS}} are given in the supplementary material. In particular, for the relevant LECs in this study we obtain that

d16EOMS=d16cov+1Fπ2δdˉ16f(3)+βd1632π2Fπ2log⁡(Mπ(phys)2mN(phys)2),(19)d_{16}^{\mathrm{EOMS}} = d_{16}^{\mathrm{cov}} + \frac{1}{F_\pi^2}\delta\bar{d}_{16f}^{(3)} + \frac{\beta_{d_{16}}}{32\pi^2 F_\pi^2}\log\left(\frac{M_{\pi(\mathrm{phys})}^2}{m_{N(\mathrm{phys})}^2}\right), \tag*{(19)}

with

δdˉ16f(3)=132π2(g˙A+gA3)andβd16=g˙A2+gA3(20)\delta\bar{d}_{16f}^{(3)} = \frac{1}{32\pi^2}(\dot{g}_A + g_A^3)\quad\text{and}\quad\beta_{d_{16}} = \frac{\dot{g}_A}{2} + g_A^3 \tag*{(20)}

because, in this case, δdˉ16f(4)=δd16fEOMS\delta\bar{d}_{16f}^{(4)} = \delta d_{16f}^{\mathrm{EOMS}}. Furthermore,

c1−4EOMS=c1−4cov(21)c_{1-4}^{\mathrm{EOMS}} = c_{1-4}^{\mathrm{cov}} \tag*{(21)}

once βc1−4=0\beta_{c_{1-4}} = 0 [25] and the finite shifts coincide with those in EOMS. Equations (19)-(21) are derived for the Δ\Delta case but hold also for the model with the Δ\Delta pole of Ref. [25] as it does not involve additional renormalizations.

Contribution to gAg_A from wave function renormalization

To the order of the present study ZNZ_N, defined in Eq. (8), can be written as ZN=1+116π2Fπ2(δZN(3)loop+δZN(4)loop)Z_N=1+\frac{1}{16\pi^2F_\pi^2}\left(\delta Z_N^{(3)\mathrm{loop}}+\delta Z_N^{(4)\mathrm{loop}}\right). The contribution to gAg_A that starts at O(p4)\mathcal{O}(p^4) is then gˉAδZN(4)loop\bar{g}_A\delta Z_N^{(4)\mathrm{loop}}, with δZN(4)loop\delta Z_N^{(4)\mathrm{loop}}, including Δ\Delta, given by

δZN(4)loop=−24c1gA2Mπ5f(Mπ)m3a3/2{6m˙4−6Mπ2m˙2+Mπ4}+2Mπ427m˙3{a1hA2m˙m˙Δ4(18m˙4m˙Δ−4m˙3(Mπ2−9m˙Δ2)+m˙2(36m˙Δ3−103Mπ2m˙Δ)−36m˙Δ2(Mπ2+4m˙Δ2))+54m˙Δ(Mπ4+Mπ2m˙Δ2−5m˙Δ4)+18c1m˙Δ2(18gA2m˙2m˙Δ2(Mπ2−2m˙2)a+hA2(8m˙4+6m˙3m˙Δ+m˙2(Mπ2−2m˙Δ2)6m˙m˙Δ(Mπ2−2m˙Δ2)−6(Mπ4−3Mπ2m˙Δ2+3m˙Δ4)))}+4Mπ49m˙3m˙Δ4log⁡(m˙2Mπ2){a1hA2m˙(−9m˙4m˙Δ+m˙3(4Mπ2−6m˙Δ2)+13m˙2Mπ2m˙Δ−3Mπ4m˙Δ+3m˙5)−27c1gA2m˙Δ4Mπ2}+83hA2Mπ2log⁡(m˙){a1(2m˙Δ3m˙Δ2−2m˙+m˙Δ)−6c1Mπ2(2m˙+m˙Δ)m˙Δ2}−8c1hA2Mπ2log⁡(Mπ)3m˙5m˙Δ2{5m˙8+6m˙7m˙Δ−3m˙6(2Mπ2+m˙Δ2)−3m˙5m˙Δ(Mπ2+m˙Δ2)+m˙2(2Mπ6−3Mπ4m˙Δ2+m˙Δ6)+3m˙m˙Δ(Mπ2−m˙Δ2)3−3(Mπ2−m˙Δ2)4}−4a1hA2Mπ23m˙2m˙Δ3log⁡(m˙2m˙Δ2){9m˙4Mπ2+m˙3(4Mπ2m˙Δ−2m˙Δ3)+m˙2(m˙Δ4−3Mπ4)−Mπ6+Mπ2m˙Δ4+2m˙Δ6}+2a1hA2Mπ23m˙4m˙Δ3log⁡(Mπ2m˙Δ2){−5m˙8−4m˙7m˙Δ+6m˙6Mπ2+2m˙5m˙Δ(Mπ2−3m˙Δ2)+2m˙2(Mπ6+Mπ4m˙Δ2−2Mπ2m˙Δ4)+2m˙m˙Δ(Mπ2−m˙Δ2)2(Mπ2+5m˙Δ2)−3(Mπ2−m˙Δ2)3(Mπ2+3m˙Δ2)}−8hA2Mπ23m˙5m˙Δ3log⁡(m˙Δ){a1m˙(5m˙8+4m˙7m˙Δ+4m˙5m˙Δ3+6m˙2m˙Δ6−10m˙m˙Δ7−9m˙Δ8)−c1m˙Δ(−5m˙8−6m˙7m˙Δ+3m˙6(2Mπ2+m˙Δ2)+3m˙5m˙Δ(Mπ2+m˙Δ2)+m˙2(2Mπ6−3Mπ4m˙Δ2−m˙Δ6)+3m˙m˙Δ(Mπ2−m˙Δ2)3−3(Mπ8−4Mπ6m˙Δ2+6Mπ4m˙Δ4−4Mπ2m˙Δ6−m˙Δ8))}+4hA2Mπ2(m˙−m˙Δ)(m˙+m˙Δ)33m˙5m˙Δ3log⁡(m˙Δ2−m˙2){a1m˙(5m˙8−6m˙7m˙Δ+12m˙6m˙Δ2−8m˙5m˙Δ3+9m˙4m˙Δ4)+2c1m˙Δ(5m˙4−4m˙3m˙Δ+5m˙2m˙Δ2−3m˙m˙Δ3+3m˙Δ4)}+F(Mπ)3m˙5m˙Δ3{4hA2(m˙2+2m˙m˙Δ−Mπ2+m˙Δ2)(a1m˙(5m˙8−6m˙7m˙Δ+2m˙6(m˙Δ2−6Mπ2)+4m˙5(2Mπ2m˙Δ+m˙Δ3)+2m˙4(3Mπ4−5m˙Δ4)+2m˙3m˙Δ(Mπ4−6Mπ2m˙Δ2+5m˙Δ4)+m˙2(4Mπ6−2Mπ4m˙Δ2+4Mπ2m˙Δ4−6m˙Δ6)−4m˙(Mπ6m˙Δ−3Mπ2m˙Δ5+2m˙Δ7)−3(Mπ2−m˙Δ2)3(Mπ2+3m˙Δ2))+2c1m˙Δ(5m˙8−4m˙7m˙Δ−m˙6(6Mπ2+5m˙Δ2)+m˙5(5m˙Δ3−Mπ2m˙Δ)+2m˙4m˙Δ2(Mπ2−m˙Δ2)+2m˙3m˙Δ(Mπ2−m˙Δ2)2−m˙2(2Mπ6−3Mπ4m˙Δ2+m˙Δ6)+3m˙m˙Δ(Mπ2−m˙Δ2)3+3(Mπ2−m˙Δ2)4))}.(22)\begin{aligned} \delta Z_N^{(4)\mathrm{loop}}={}&-\frac{24c_1g_A^2M_\pi^5f(M_\pi)}{m^3a^{3/2}}\{6\dot{m}^4-6M_\pi^2\dot{m}^2+M_\pi^4\}\\ &+\frac{2M_\pi^4}{27\dot{m}^3}\{\frac{a_1h_A^2\dot{m}}{\dot{m}_\Delta^4}(18\dot{m}^4\dot{m}_\Delta-4\dot{m}^3(M_\pi^2-9\dot{m}_\Delta^2)+\dot{m}^2(36\dot{m}_\Delta^3-103M_\pi^2\dot{m}_\Delta)-36\dot{m}_\Delta^2(M_\pi^2+4\dot{m}_\Delta^2))\\ &\qquad+54\dot{m}_\Delta(M_\pi^4+M_\pi^2\dot{m}_\Delta^2-5\dot{m}_\Delta^4)+\frac{18c_1}{\dot{m}_\Delta^2}(\frac{18g_A^2\dot{m}^2\dot{m}_\Delta^2(M_\pi^2-2\dot{m}^2)}{a}+h_A^2(8\dot{m}^4+6\dot{m}^3\dot{m}_\Delta\\ &\qquad\qquad+\dot{m}^2(M_\pi^2-2\dot{m}_\Delta^2)6\dot{m}\dot{m}_\Delta(M_\pi^2-2\dot{m}_\Delta^2)-6(M_\pi^4-3M_\pi^2\dot{m}_\Delta^2+3\dot{m}_\Delta^4)))\}\\ &+\frac{4M_\pi^4}{9\dot{m}^3\dot{m}_\Delta^4}\log(\frac{\dot{m}^2}{M_\pi^2})\{a_1h_A^2\dot{m}(-9\dot{m}^4\dot{m}_\Delta+\dot{m}^3(4M_\pi^2-6\dot{m}_\Delta^2)+13\dot{m}^2M_\pi^2\dot{m}_\Delta-3M_\pi^4\dot{m}_\Delta+3\dot{m}^5)\\ &\qquad-27c_1g_A^2\dot{m}_\Delta^4M_\pi^2\}\\ &+\frac{8}{3}h_A^2M_\pi^2\log(\dot{m})\{a_1(\frac{2\dot{m}_\Delta^3}{\dot{m}_\Delta^2}-2\dot{m}+\dot{m}_\Delta)-\frac{6c_1M_\pi^2(2\dot{m}+\dot{m}_\Delta)}{\dot{m}_\Delta^2}\}\\ &-\frac{8c_1h_A^2M_\pi^2\log(M_\pi)}{3\dot{m}^5\dot{m}_\Delta^2}\{5\dot{m}^8+6\dot{m}^7\dot{m}_\Delta-3\dot{m}^6(2M_\pi^2+\dot{m}_\Delta^2)-3\dot{m}^5\dot{m}_\Delta(M_\pi^2+\dot{m}_\Delta^2)\\ &\qquad+\dot{m}^2(2M_\pi^6-3M_\pi^4\dot{m}_\Delta^2+\dot{m}_\Delta^6)+3\dot{m}\dot{m}_\Delta(M_\pi^2-\dot{m}_\Delta^2)^3-3(M_\pi^2-\dot{m}_\Delta^2)^4\}\\ &-\frac{4a_1h_A^2M_\pi^2}{3\dot{m}^2\dot{m}_\Delta^3}\log(\frac{\dot{m}^2}{\dot{m}_\Delta^2})\{9\dot{m}^4M_\pi^2+\dot{m}^3(4M_\pi^2\dot{m}_\Delta-2\dot{m}_\Delta^3)+\dot{m}^2(\dot{m}_\Delta^4-3M_\pi^4)-M_\pi^6+M_\pi^2\dot{m}_\Delta^4+2\dot{m}_\Delta^6\}\\ &+\frac{2a_1h_A^2M_\pi^2}{3\dot{m}^4\dot{m}_\Delta^3}\log(\frac{M_\pi^2}{\dot{m}_\Delta^2})\{-5\dot{m}^8-4\dot{m}^7\dot{m}_\Delta+6\dot{m}^6M_\pi^2+2\dot{m}^5\dot{m}_\Delta(M_\pi^2-3\dot{m}_\Delta^2)\\ &\qquad+2\dot{m}^2(M_\pi^6+M_\pi^4\dot{m}_\Delta^2-2M_\pi^2\dot{m}_\Delta^4)+2\dot{m}\dot{m}_\Delta(M_\pi^2-\dot{m}_\Delta^2)^2(M_\pi^2+5\dot{m}_\Delta^2)-3(M_\pi^2-\dot{m}_\Delta^2)^3(M_\pi^2+3\dot{m}_\Delta^2)\}\\ &-\frac{8h_A^2M_\pi^2}{3\dot{m}^5\dot{m}_\Delta^3}\log(\dot{m}_\Delta)\{a_1\dot{m}(5\dot{m}^8+4\dot{m}^7\dot{m}_\Delta+4\dot{m}^5\dot{m}_\Delta^3+6\dot{m}^2\dot{m}_\Delta^6-10\dot{m}\dot{m}_\Delta^7-9\dot{m}_\Delta^8)\\ &\qquad-c_1\dot{m}_\Delta(-5\dot{m}^8-6\dot{m}^7\dot{m}_\Delta+3\dot{m}^6(2M_\pi^2+\dot{m}_\Delta^2)+3\dot{m}^5\dot{m}_\Delta(M_\pi^2+\dot{m}_\Delta^2)+\dot{m}^2(2M_\pi^6-3M_\pi^4\dot{m}_\Delta^2-\dot{m}_\Delta^6)\\ &\qquad+3\dot{m}\dot{m}_\Delta(M_\pi^2-\dot{m}_\Delta^2)^3-3(M_\pi^8-4M_\pi^6\dot{m}_\Delta^2+6M_\pi^4\dot{m}_\Delta^4-4M_\pi^2\dot{m}_\Delta^6-\dot{m}_\Delta^8))\}\\ &+\frac{4h_A^2M_\pi^2(\dot{m}-\dot{m}_\Delta)(\dot{m}+\dot{m}_\Delta)^3}{3\dot{m}^5\dot{m}_\Delta^3}\log(\dot{m}_\Delta^2-\dot{m}^2)\{a_1\dot{m}(5\dot{m}^8-6\dot{m}^7\dot{m}_\Delta+12\dot{m}^6\dot{m}_\Delta^2-8\dot{m}^5\dot{m}_\Delta^3+9\dot{m}^4\dot{m}_\Delta^4)\\ &\qquad+2c_1\dot{m}_\Delta(5\dot{m}^4-4\dot{m}^3\dot{m}_\Delta+5\dot{m}^2\dot{m}_\Delta^2-3\dot{m}\dot{m}_\Delta^3+3\dot{m}_\Delta^4)\}\\ &+\frac{F(M_\pi)}{3\dot{m}^5\dot{m}_\Delta^3}\{4h_A^2(\dot{m}^2+2\dot{m}\dot{m}_\Delta-M_\pi^2+\dot{m}_\Delta^2)(a_1\dot{m}(5\dot{m}^8-6\dot{m}^7\dot{m}_\Delta+2\dot{m}^6(\dot{m}_\Delta^2-6M_\pi^2)\\ &\qquad+4\dot{m}^5(2M_\pi^2\dot{m}_\Delta+\dot{m}_\Delta^3)+2\dot{m}^4(3M_\pi^4-5\dot{m}_\Delta^4)+2\dot{m}^3\dot{m}_\Delta(M_\pi^4-6M_\pi^2\dot{m}_\Delta^2+5\dot{m}_\Delta^4)\\ &\qquad+\dot{m}^2(4M_\pi^6-2M_\pi^4\dot{m}_\Delta^2+4M_\pi^2\dot{m}_\Delta^4-6\dot{m}_\Delta^6)-4\dot{m}(M_\pi^6\dot{m}_\Delta-3M_\pi^2\dot{m}_\Delta^5+2\dot{m}_\Delta^7)\\ &\qquad-3(M_\pi^2-\dot{m}_\Delta^2)^3(M_\pi^2+3\dot{m}_\Delta^2))\\ &\qquad+2c_1\dot{m}_\Delta(5\dot{m}^8-4\dot{m}^7\dot{m}_\Delta-\dot{m}^6(6M_\pi^2+5\dot{m}_\Delta^2)+\dot{m}^5(5\dot{m}_\Delta^3-M_\pi^2\dot{m}_\Delta)+2\dot{m}^4\dot{m}_\Delta^2(M_\pi^2-\dot{m}_\Delta^2)\\ &\qquad+2\dot{m}^3\dot{m}_\Delta(M_\pi^2-\dot{m}_\Delta^2)^2-\dot{m}^2(2M_\pi^6-3M_\pi^4\dot{m}_\Delta^2+\dot{m}_\Delta^6)+3\dot{m}\dot{m}_\Delta(M_\pi^2-\dot{m}_\Delta^2)^3+3(M_\pi^2-\dot{m}_\Delta^2)^4))\}. \tag*{(22)} \end{aligned}

with the following definitions:

F(Mπ)≡Mπ2blog⁡(−m˚2+b+Mπ2+m˚Δ22Mπm˚Δ),f(Mπ)≡ilog⁡(Mπ+ia2m˚)(23)F(M_{\pi}) \equiv\frac{M_{\pi}^{2}}{b}\log\left(\frac{-\mathring{m}^{2}+b+M_{\pi}^{2}+\mathring{m}_{\Delta}^{2}}{2M_{\pi}\mathring{m}_{\Delta}}\right), \qquad f(M_{\pi}) \equiv i\log\left(\frac{M_{\pi}+i\sqrt{a}}{2\mathring{m}}\right) \tag*{(23)}

where b≡m˚4−2m˚2(Mπ2+m˚Δ2)+(Mπ2−m˚Δ2)2b \equiv\sqrt{\mathring{m}^{4}-2\mathring{m}^{2}(M_{\pi}^{2}+\mathring{m}_{\Delta}^{2})+(M_{\pi}^{2}-\mathring{m}_{\Delta}^{2})^{2}} and a≡4m˚2−Mπ2a \equiv4\mathring{m}^{2}-M_{\pi}^{2}. This result, but also the full expression for the O(p4)\mathcal{O}(p^{4}) nucleon selfenergy ΣN\Sigma_{N} with explicit Δ\Delta, from which ZNZ_{N} is derived using Eq. (8), can be found in the Mathematica notebook provided as supplementary material.

References

References

  1. [1]M. Gorchtein and C.-Y. Seng, Dispersion relation analysis of the radiative corrections to g_A in the neutron β-decay, JHEP 10, 053, arXiv:2106.09185 [hep-ph].DOI
  2. [2]P. Zyla et al. (Particle Data Group), Review of Particle Physics, PTEP 2020, 083C01 (2020).
  3. [3]H.-W. Lin, Frontiers in lattice nucleon structure, Int. J. Mod. Phys. A 35, 2030006 (2020).DOI
  4. [4]Y. Aoki et al., FLAG Review 2021, (2021), arXiv:2111.09849 [hep-lat].
  5. [5]B. C. Tiburzi, Time Dependence of Nucleon Correlation Functions in Chiral Perturbation Theory, Phys. Rev. D 80, 014002 (2009), arXiv:0901.0657 [hep-lat].DOI
  6. [6]O. Bar, Chiral perturbation theory and nucleon–pion–state contaminations in lattice QCD, Int. J. Mod. Phys. A 32, 1730011 (2017), arXiv:1705.02806 [hep-lat].DOI
  7. [7]S. R. Beane and M. J. Savage, Nucleon properties at finite lattice spacing in chiral perturbation theory, Phys. Rev. D 68, 114502 (2003), arXiv:hep-lat/0306036.
  8. [8]S. R. Beane and M. J. Savage, Baryon axial charge in a finite volume, Phys. Rev. D 70, 074029 (2004), arXiv:hep-ph/0404131.DOI
  9. [9]G. S. Bali, L. Barç̧a, S. Collins, M. Gruber, M. Löffler, A. Schäfer, W. Söldner, P. Wein, S. Weishäupl, and T. Wurm (RQCD), Nucleon axial structure from lattice QCD, JHEP 05, 126, arXiv:1911.13150 [hep-lat].DOI
  10. [10]S. R. Beane and M. J. Savage, The Quark mass dependence of two nucleon systems, Nucl. Phys. A 717, 91 (2003), arXiv:nucl-th/0208021.DOI
  11. [11]J. C. Berengut, E. Epelbaum, V. V. Flambaum, C. Hanhart, U. G. Meissner, J. Nebreda, and J. R. Pelaez, Varying the light quark mass: impact on the nuclear force and Big Bang nucleosynthesis, Phys. Rev. D 87, 085018 (2013), arXiv:1301.1738 [nucl-th].DOI
  12. [12]E. Epelbaum, H. Krebs, T. A. Lähde, D. Lee, and U.-G. Meißner, Dependence of the triple-alpha process on the fundamental constants of nature, Eur. Phys. J. A 49, 82 (2013), arXiv:1303.4856 [nucl-th].DOI
  13. [13]V. Bernard, N. Kaiser, T. S. H. Lee, and U.-G. Meissner, Threshold pion electroproduction in chiral perturbation theory, Phys. Rept. 246, 313 (1994), arXiv:hep-ph/9310329.
  14. [14]M. Hilt, B. C. Lehnhart, S. Scherer, and L. Tiator, Pion photo- and electroproduction in relativistic baryon chiral perturbation theory and the chiral MAID interface, Phys. Rev. C 88, 055207 (2013), arXiv:1309.3385 [nucl-th].
  15. [15]G. H. Guerrero Navarro and M. J. Vicente Vacas, Threshold pion electro- and photoproduction off nucleons in covariant chiral perturbation theory, Phys. Rev. D 102, 113016 (2020), arXiv:2008.04244 [hep-ph].DOI
  16. [16]N. Rijneveen, A. M. Gasparyan, H. Krebs, and E. Epelbaum, Pion photoproduction in chiral perturbation theory with explicit treatment of the Δ(1232) resonance, (2021), arXiv:2108.01619 [nucl-th].arxiv.org/abs/2108.01619
  17. [17]M. Mojzis, Elastic pi N scattering to O(p**3) in heavy baryon chiral perturbation theory, Eur. Phys. J. C 2, 181 (1998), arXiv:hep-ph/9704415.DOI
  18. [18]N. Fettes and U. G. Meissner, Pion nucleon scattering in chiral perturbation theory. 2: Fourth order calculation, Nucl. Phys. A 676, 311 (2000), arXiv:hep-ph/0002162.DOI
  19. [19]N. Fettes and U. G. Meissner, Pion - nucleon scattering in an effective chiral field theory with explicit spin 3/2 fields, Nucl. Phys. A 679, 629 (2001), arXiv:hep-ph/0006299.DOI
  20. [20]K. Tarkoshing and P. J. Ellis, Low-energy pion nucleon scattering in the heavy baryon and infrared schemes, Phys. Rev. C 67, 015208 (2003), arXiv:nucl-th/0208049.
  21. [21]M. Hoferichter, J. Ruiz de Elvira, B. Kubis, and U.-G. Meißner, Roy–Steiner-equation analysis of pion–nucleon scattering, Phys. Rept. 625, 1 (2016), arXiv:1510.06039 [hep-ph].
  22. [22]Y.-H. Chen, D.-L. Yao, and H. Q. Zheng, Analyses of pion-nucleon elastic scattering amplitudes up to O(p^4) in extended-on-mass-shell subtraction scheme, Phys. Rev. D 87, 054019 (2013), arXiv:1212.1893 [hep-ph].DOI
  23. [23]J. M. Alarcon, J. Martin Camalich, and J. A. Oller, Improved description of the πN-scattering phenomenology in covariant baryon chiral perturbation theory, Annals Phys. 336, 413 (2013), arXiv:1210.4450 [hep-ph].DOI
  24. [24]D.-L. Yao, D. Siemens, V. Bernard, E. Epelbaum, A. M. Gasparyan, J. Gegelia, H. Krebs, and U.-G. Meißner, Pion-nucleon scattering in covariant baryon chiral perturbation theory with explicit Delta resonances, JHEP 05, 038, arXiv:1603.03638 [hep-ph].
  25. [25]D. Siemens, V. Bernard, E. Epelbaum, A. M. Gasparyan, H. Krebs, and U.-G. Meißner, Elastic pion-nucleon scattering in chiral perturbation theory: A fresh look, Phys. Rev. C 94, 014620 (2016), arXiv:1602.02640 [nucl-th].DOI
  26. [26]D. Siemens, V. Bernard, E. Epelbaum, A. M. Gasparyan, H. Krebs, and U.-G. Meißner, Elastic and inelastic pion-nucleon scattering to fourth order in chiral perturbation theory, Phys. Rev. C 96, 055205 (2017), arXiv:1704.08988 [nucl-th].DOI
  27. [27]J. Kambor and M. Mojzis, Field redefinitions and wave function renormalization to O(p**4) in heavy baryon chiral perturbation theory, JHEP 04, 031 (2003), arXiv:hep-ph/9901235.
  28. [28]T. R. Hemmert, M. Procura, and W. Weise, Quark mass dependence of the nucleon axial vector coupling constant, Phys. Rev. D 68, 075009 (2003), arXiv:hep-lat/0303002.DOI
  29. [29]V. Bernard and U.-G. Meißner, The Nucleon axial-vector coupling beyond one loop, Phys. Lett. B 639, 278 (2006), arXiv:hep-lat/0605010.DOI
  30. [30]M. Procura, B. U. Musch, T. R. Hemmert, and W. Weise, Chiral extrapolation of g(A) with explicit Delta(1232) degrees of freedom, Phys. Rev. D 75, 014503 (2007), arXiv:hep-lat/0610105.DOI
  31. [31]A. Calle Cordon and J. L. Goity, Baryon Masses and Axial Couplings in the Combined 1/N_c and Chiral Expansions, Phys. Rev. D 87, 016019 (2013), arXiv:1210.2364 [nucl-th].DOI
  32. [32]M. R. Schindler, T. Fuchs, J. Gegelia, and S. Scherer, Axial, induced pseudoscalar, and pion-nucleon form-factors in manifestly Lorentz-invariant chiral perturbation theory, Phys. Rev. C 75, 025202 (2007), arXiv:nucl-th/0611083.DOI
  33. [33]S.-i. Ando and H. W. Fearing, Ordinary muon capture on a proton in manifestly Lorentz invariant baryon chiral perturbation theory, Phys. Rev. D 75, 014025 (2007), arXiv:hep-ph/0608195.DOI
  34. [34]D.-L. Yao, L. Alvarez-Ruso, and M. J. Vicente-Vacas, Extraction of nucleon axial charge and radius from lattice QCD results using baryon chiral perturbation theory, Phys. Rev. D 96, 116022 (2017), arXiv:1708.08776 [hep-ph].DOI
  35. [35]M. F. M. Lutz, U. Sauerwein, and R. G. E. Timmermans, On the axial-vector form factor of the nucleon and chiral symmetry, Eur. Phys. J. C 80, 844 (2020), arXiv:2003.10158 [hep-lat].DOI
  36. [36]S. Weinberg, Effective chiral Lagrangians for nucleon - pion interactions and nuclear forces, Nucl. Phys. B 363, 3 (1991).DOI
  37. [37]T. R. Hemmert, B. R. Holstein, and J. Kambor, Systematic 1/M expansion for spin 3/2 particles in baryon chiral perturbation theory, Phys. Lett. B 395, 89 (1997), arXiv:hep-ph/9606456.DOI
  38. [38]T. R. Hemmert, B. R. Holstein, and J. Kambor, Chiral Lagrangians and delta(1232) interactions: Formalism, J. Phys. G 24, 1831 (1998), arXiv:hep-ph/9712496.
  39. [39]N. Fettes, U.-G. Meißner, M. Mojzis, and S. Steininger, The Chiral effective pion nucleon Lagrangian of order p**4, Annals Phys. 283, 273 (2000), [Erratum: Annals Phys. 288, 249–250 (2001)], arXiv:hep-ph/0001308.DOI
  40. [40]S.-Z. Jiang, Y.-R. Liu, and H.-Q. Wang, Chiral Lagrangians with Δ(1232) to one loop, Phys. Rev. D 97, 014002 (2018), arXiv:1707.09690 [hep-ph].DOI
  41. [41]M. Holmberg and S. Leupold, The relativistic chiral Lagrangian for decuplet and octet baryons at next-to-leading order, Eur. Phys. J. A 54, 103 (2018), arXiv:1802.05168 [hep-ph].DOI
  42. [42]Y. Ün al, A. Küçükarslan, and S. Scherer, Axial-vector nucleon-to-delta transition form factors using the complex-mass renormalization scheme, Phys. Rev. D 104, 094014 (2021), arXiv:2104.03125 [hep-ph].DOI
  43. [43]C. Hacker, N. Wies, J. Gegelia, and S. Scherer, Including the Delta(1232) resonance in baryon chiral perturbation theory, Phys. Rev. C 72, 055203 (2005), arXiv:hep-ph/0505043.DOI
  44. [44]D. Siemens, Elastic pion-nucleon scattering in chiral perturbation theory: Explicit Δ(1232) degrees of freedom, (2020), arXiv:2001.03906 [nucl-th].arxiv.org/abs/2001.03906
  45. [45]T. Fuchs, J. Gegelia, J. Gaparidze, and S. Scherer, Renormalization of relativistic baryon chiral perturbation theory, Phys. Rev. D 68, 056005 (2003), arXiv:hep-ph/0302117.DOI
  46. [46]C. C. Chang et al., A per-cent-level determination of the nucleon axial coupling from quantum chromodynamics, Nature 558, 91 (2018), arXiv:1805.12130 [hep-lat].DOI
  47. [47]T. Harris, G. von Hippel, P. J. Bunce, H. B. Meyer, K. Ottnad, J. Wilhelm, H. Wittig, and L. Wrang, Nucleon isovector charges and twist-2 matrix elements with N_f = 2 + 1 dynamical Wilson quarks, Phys. Rev. D 100, 034513 (2019), arXiv:1905.01291 [hep-lat].DOI
  48. [48]S. Park, R. Gupta, B. Yoon, S. Mondal, T. Bhattacharya, Y.-C. Jang, B. Joó, and F. Winter (Nucleon Matrix Elements (NME)), Precision Nucleon Charges and Form Factors Using 2+1-flavor Lattice QCD, (2021), arXiv:2103.05599 [hep-lat].DOI
  49. [49]M. Constantinou, Lattice Generalized Parton Distributions and Form Factors of the Nucleon, in 22nd International Symposium on Spin Physics (2017) arXiv:1701.02855 [hep-lat].arxiv.org/abs/1701.02855
  50. [50]S. Wesolowski, N. Klco, R. J. Furnstahl, D. R. Phillips, and A. Thapaliya, Bayesian parameter estimation for effective field theories, J. Phys. G 43, 074001 (2016), arXiv:1511.03618 [nucl-th].DOI
  51. [51]A. Manohar and H. Georgi, Chiral Quarks and the Nonrelativistic Quark Model, Nucl. Phys. B 234, 189 (1984).
  52. [52]S. Scherer and M. R. Schindler, A Primer for Chiral Perturbation Theory, Vol. 830 (2012).DOI
  53. [53]E. Epelbaum, H. Krebs, and U. G. Meißner, Improved chiral nucleon-nucleon potential up to next-to-next-to-next-to-leading order, Eur. Phys. J. A 51, 53 (2015), arXiv:1412.0142 [nucl-th].DOI
  54. [54]R. J. Furnstahl, N. Klco, D. R. Phillips, and S. Wesolowski, Quantifying truncation errors in effective field theory, Phys. Rev. C 92, 024005 (2015), arXiv:1506.01343 [nucl-th].DOI
  55. [55]S. L. Adler, Calculation of the axial vector coupling constant renormalization in beta decay, Phys. Rev. Lett. 14, 1051 (1965).
  56. [56]W. I. Weisberger, Renormalization of the Weak Axial Vector Coupling Constant, Phys. Rev. Lett. 14, 1047 (1965).DOI
  57. [57]L. Alvarez-Ruso, T. Ledwig, J. Martin Camalich, and M. J. Vicente-Vacas, Nucleon mass and pion-nucleon sigma term from a chiral analysis of lattice QCD data, Phys. Rev. D 88, 054507 (2013), arXiv:1304.0483 [hep-ph].

Paper details

vlnc:W_63dnuuqv
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