betaFeedback

WorksPhysics and Astronomy

On the definition of the nucleon axial charge density

Abstract

We work out the spatial density distributions corresponding to the axial-vector charge density operator for spin-1/2 systems using states described by sharply localized wave packets in arbitrary Lorentz-frames. The static approximation, leading to the frequently assumed Breit-frame distributions, is also considered. We discuss the interpretation of the resulting spatial densities in terms of the axial charge density.

Introduction and summary

Following the seminal papers on electron-proton scattering in the 60s of the last century [1, 2, 3] the three-dimensional Fourier transform of the charge form factor in the Breit frame is traditionally interpreted as the electric charge density of the considered hadron. Analogous interpretation has also been suggested for the Fourier transforms of the gravitational form factors and for various local distributions [4, 5, 6]. On the other hand, in last two decades it has been repeatedly pointed out that the identification of spatial density distributions with the Fourier transform of the corresponding form factors in the Breit frame is problematic [7, 8, 9, 10, 11, 12, 13].

The issues of a proper definition and interpretation of the spatial density distributions of matrix elements of local operators have attracted much attention in last years, see, e.g., Refs. [7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]. A novel definition of the three-dimensional charge density using sharply localized wave packet states has been suggested in Ref. [20]. In fact, similar results have been published long ago in largely unknown work of Ref. [21]. In Refs. [20, 22, 23, 24, 25] the three-dimensional spatial densities, possessing the usual probabilistic interpretation, have been defined in the zero average momentum frame (ZAMF) for systems with various spins by using spherically symmetric sharply localized wave packets (here we deleted: ”without using any approximations”).1 New definition has advantage compared to the Breit frame approach that it is also applicable to systems whose characteristic radii are of the order of (or less than) their Compton wavelengths. By generalizing the definition to moving frames it was also shown that in the infinite-momentum frame (IMF), the new spatial densities turn into the well-known two-dimensional distributions in the transverse plane, multiplied by the delta-function in the longitudinal direction.

Recently, the spatial densities of spin-1/21/2 systems corresponding to the axial-vector current have been studied in the phase-space Wigner distributions approach [26]. The authors of this paper argued that the axial charge density vanishes identically in the Breit frame and thus does not yield a nontrivial axial charge radius. For a review on the determinations of the axial radius of the nucleon, defined in terms of the slope of the axial-vector form factor, see Ref. [27].

In this paper we work out the details of the novel definition of the spatial distribution corresponding to the zeroth component of the axial-vector current for spin-1/21/2 systems. We build upon our earlier work [20, 22, 23, 24, 25] and employ sharply localized wave packet states to obtain the corresponding spatial densities in the ZAMF as well as in IMF.2 We also consider an alternative definition of the densities by taking the static limit, which leads to the frequently used Breit-frame distributions, and compare our results to the findings of Ref. [26]. While the resulting spatial distributions cannot be directly interpreted as axial charge densities, see also Ref. [26], we argue that a proper definition of the axial charge densities follows by dropping in the obtained expressions the overall spin-dependent factor that does not encode information about the internal structure of the system. Similarly to the electromagnetic case, the axial charge density defined via the Breit-frame distribution is appropriate and, in fact, quite natural for heavy systems, however it is inappropriate for systems with radii of the order of, or smaller, than their Compton wave lengths. This motivates us to consider the ZAMF definition without relying on the static approximation that corresponds to considering the system in states characterized by wave packets whose spatial extent is much larger than the Compton wavelength, yet much smaller than the corresponding radius [11]. Spatial densities defined this way probe not only the internal structure of a system under consideration but also the effects of delocalization. In contrast, using sharply localized wave packet states allows one to reduce delocalization artifacts in the definition of spatial densities to, in principle, arbitrarily small amount.

Our paper is organized as follows. In sec. II, we apply the framework developed in Refs. [20, 22, 23, 24, 25] to work out the spatial distributions corresponding to the zeroth component of the axial-vector current in sharply-localized wave-packet states. The considerations in this section are limited to the ZAMF. We also consider here an alternative and more traditional approach that uses wave packets localized at distances much larger than the Compton wave length of the system which does, however, only provide wave-packet-independent distributions in the static limit. In line with the observations made in Ref. [26], we find vanishing results for the ZAMF densities in both approaches. To trace the origin of these findings, the considerations of sec. II are generalized to the spatial distributions in arbitrary moving frames in sec. III. The interpretation of the vanishing distributions in the ZAMF and a physically meaningful definition of the intrinsic axial charge densities of a spin-1/2 system are provided in sec. IV.

Spatial distributions in the ZAMF

The normalization of the four-momentum eigenstates ∣p,s⟩|p,s\rangle which we use to characterize the considered spin-1/2 system is specified by the equation

⟨p′,s′∣p,s⟩=2E(2π)3δs′sδ(3)(p′−p),(1)\langle p',s'|p,s\rangle= 2E(2\pi)^3\delta_{s's}\delta^{(3)}(\mathbf{p}'-\mathbf{p}), \tag*{(1)}

where p=(E,p)\mathbf{p}=(E,\mathbf{p}), with E=m2+p2E=\sqrt{m^2+\mathbf{p}^2}, and mm is the particle’s mass.

The matrix elements of the axial-vector current operator between momentum eigenstates of a spin-1/2 system is parameterized in terms of two form factors (utilizing Lorentz invariance, isospin conservation and the discrete symmetries C, P and T, and assuming the absence of second class currents):

⟨p′,s′∣jAμ(r,0)∣p,s⟩=e−i(p′−p)⋅ruˉ(p′,s′)[γμγ5GA(q2)+qμγ52MGP(q2)]u(p,s).(2)\langle p',s'|j_A^\mu(\mathbf{r},0)|p,s\rangle=e^{-i(p'-p)\cdot r}\bar{u}(p',s')\left[\gamma^\mu\gamma_5G_A(q^2)+\frac{q^\mu\gamma_5}{2M}G_P(q^2)\right]u(p,s). \tag*{(2)}

Here, pp and ss (p′p' and s′s') denote the momentum and the polarization of the initial (final) states, respectively, and the Dirac spinors are normalized as uˉ(p,s′)u(p,s)=2mδs′s\bar{u}(p,s')u(p,s)=2m\delta_{s's}. The momentum transfer is given as q=p′−pq=p'-p, while GA(q2)G_A(q^2) and GP(q2)G_P(q^2) are the axial and induced pseudoscalar form factors, respectively. The factor MM introduced in front of the form factor GPG_P is commonly set equal to the particle’s mass. However, as it mixes the orders of the 1/m1/m expansion used below, we distinguish between these two quantities. In the final results we set M=mM=m, meaning that these mass scales are numerically equal.

To obtain local density distributions we calculate the matrix element of the axial charge density operator in a state localized in coordinate space. For that purpose we consider a normalizable Heisenberg-picture state specified by a wave packet:

∣Φ,X,s⟩=∫d3p2E(2π)3ϕ(s,p)e−ip⋅X∣p,s⟩.(3)|\Phi,\mathbf{X},s\rangle=\int\frac{d^3p}{\sqrt{2E(2\pi)^3}}\phi(s,\mathbf{p})e^{-i\mathbf{p}\cdot\mathbf{X}}|p,s\rangle. \tag*{(3)}

From the normalization condition it follows that

∫d3p ∣ϕ(s,p)∣2=1.(4)\int d^3p\,|\phi(s,\mathbf{p})|^2=1. \tag*{(4)}

The state ∣Φ,X,s⟩|\Phi,\mathbf{X},s\rangle above depends on the vector X\mathbf{X}, whose interpretation will be given later. To define spatial density distributions we employ spherically symmetric wave packets with spin-independent profile functions ϕ(s,p)=ϕ(p)=ϕ(∣p∣)\phi(s,\mathbf{p})=\phi(\mathbf{p})=\phi(|\mathbf{p}|), which corresponds to ZAMF. This choice appears legitimate due to the absence of a preferred direction in space. For convenience, we also introduce a dimensionless profile function ϕ~\tilde{\phi} via

ϕ(p)=R3/2ϕ~(Rp),(5)\phi(\mathbf{p})=R^{3/2}\tilde{\phi}(R\mathbf{p}), \tag*{(5)}

where RR characterizes the size of the wave packet with small values of RR corresponding to a sharp localization.

The matrix element of the axial charge density operator jA0j_A^0 for the state defined in Eq. (3) is given by

jAϕ0(r)=⟨Φ,X,s∣jA0(x,0)∣Φ,X,s⟩=∫d3P d3q(2π)34EE′uˉ(p′,s′)[γ0γ5GA((E−E′)2−q2)+q0γ52MGP((E−E′)2−q2)]u(p,s)×ϕ(P−q2)ϕ∗(P+q2)e−iq⋅r,(6)\begin{aligned} j_{A\phi}^0(\mathbf{r})=\langle\Phi,\mathbf{X},s|j_A^0(\mathbf{x},0)|\Phi,\mathbf{X},s\rangle\\ &=\int\frac{d^3P\,d^3q}{(2\pi)^3\sqrt{4EE'}}\bar{u}(p',s')\left[\gamma^0\gamma_5G_A((E-E')^2-\mathbf{q}^2)+\frac{q^0\gamma_5}{2M}G_P((E-E')^2-\mathbf{q}^2)\right]u(p,s) \\ &\times\phi\left(\mathbf{P}-\frac{\mathbf{q}}{2}\right)\phi^*\left(\mathbf{P}+\frac{\mathbf{q}}{2}\right)e^{-i\mathbf{q}\cdot\mathbf{r}}, \tag*{(6)} \end{aligned}

where P=(p′+p)/2\mathbf{P}=(\mathbf{p}^{\prime}+\mathbf{p})/2, E=m2+P2−P⋅q+q2/4E=\sqrt{m^2+\mathbf{P}^2-\mathbf{P}\cdot\mathbf{q}+\mathbf{q}^2/4}, E′=m2+P2+P⋅q+q2/4E^{\prime}=\sqrt{m^2+\mathbf{P}^2+\mathbf{P}\cdot\mathbf{q}+\mathbf{q}^2/4}, and we have introduced r=x−X\mathbf{r}=\mathbf{x}-\mathbf{X}. The spatial vector coordinates x\mathbf{x}, X\mathbf{X} and r\mathbf{r} are defined hier in ZAMF, while in the next section the same variables refer to moving frame.

Using the methods of Refs. [28, 29], the R→0R\to0 limit of the expression in Eq. (6) can be calculated without specifying the form factors and the profile function of the packet. For GA(q2)G_A(q^2) and GP(q2)G_P(q^2) decreasing at large q2q^2 faster than 1/q21/q^2 and 1/(q2)21/(q^2)^2, respectively, the result of jAϕ0(r)j^0_{A\phi}(\mathbf{r}) in the R→0R\to0 limit is obtained using the method of dimensional counting [28] by substituting P=P~/R\mathbf{P}=\widetilde{\mathbf{P}}/R, expanding the integrand around R=0R=0 and keeping terms up to the zeroth order. Doing so we obtain

JAϕ0(r)=∫d3P~d3q(2π)3P~^⋅σ GA[(P~^⋅q)2−q2]∣ϕ~(P~)∣2e−iq⋅r,(7)J^0_{A\phi}(\mathbf{r})=\int\frac{d^3\widetilde{P}d^3q}{(2\pi)^3}\widehat{\widetilde{\mathbf{P}}}\cdot\boldsymbol{\sigma}\,G_A\left[\left(\widehat{\widetilde{\mathbf{P}}}\cdot\mathbf{q}\right)^2-\mathbf{q}^2\right]\left|\widetilde{\phi}(\widetilde{\mathbf{P}})\right|^2e^{-i\mathbf{q}\cdot\mathbf{r}}, \tag*{(7)}

where σ\boldsymbol{\sigma} is a three-vector of Pauli sigma matrices, P~^=P~/P~\widehat{\widetilde{\mathbf{P}}}=\widetilde{\mathbf{P}}/\widetilde{P} with P~≡∣P~∣\widetilde{P}\equiv|\widetilde{\mathbf{P}}|, and we use JA0J^0_A instead of jA0j^0_A for the density written as an operator in spin space. Using Eq. (4), the integration over P~\widetilde{\mathbf{P}} for spherically symmetric wave packets with ϕ~(P~)=ϕ~(∣P~∣)\widetilde{\phi}(\widetilde{\mathbf{P}})=\widetilde{\phi}(|\widetilde{\mathbf{P}}|) leads to

JA,ZAMF0(r)=0.(8)J^0_{A,\mathrm{ZAMF}}(\mathbf{r})=0. \tag*{(8)}

Next, in order to obtain the axial charge distribution in the static limit, which we refer to as “naive” following the terminology of Ref. [11], we expand the integrand in Eq. (6) in powers of 1/m1/m and subsequently localize the wave packet by taking the limit R→0R\to0 [11, 12]. The two leading-order contributions in the 1/m1/m-expansion of the integrand lead to the following expression:

JAϕ,naive0(r)=∫d3P d3q(2π)3[P⋅σmGA(−q2)−q⋅σ P⋅q4Mm2GP(−q2)]ϕ(P−q2)ϕ∗(P+q2)e−iq⋅r.(9)J^0_{A\phi,\mathrm{naive}}(\mathbf{r})=\int\frac{d^3P\,d^3q}{(2\pi)^3}\left[\frac{\mathbf{P}\cdot\boldsymbol{\sigma}}{m}G_A(-\mathbf{q}^2)-\frac{\mathbf{q}\cdot\boldsymbol{\sigma}\,\mathbf{P}\cdot\mathbf{q}}{4Mm^2}G_P(-\mathbf{q}^2)\right]\phi\left(\mathbf{P}-\frac{\mathbf{q}}{2}\right)\phi^*\left(\mathbf{P}+\frac{\mathbf{q}}{2}\right)e^{-i\mathbf{q}\cdot\mathbf{r}}. \tag*{(9)}

Also for the naive definition of the axial charge density in Eq. (9), the R→0R\to0 limit of the expression in Eq. (9) can be calculated by applying the method of dimensional counting without specifying the form factors and the profile function of the packet. Again, for GA(q2)G_A(q^2) and GP(q2)G_P(q^2) decreasing at large q2q^2-values faster than 1/q21/q^2 and 1/(q2)21/(q^2)^2, respectively, we obtain using the method of dimensional counting [28]

JA,naive0(r)≡0,(10)J^0_{A,\mathrm{naive}}(\mathbf{r})\equiv0, \tag*{(10)}

where we have used the spherical symmetry of the wave packet.3

At first sight, the vanishing results of Eqs. (10) and (8) seem to suggest that the Fourier transform of GA(q2)G_A(q^2) does not correspond to the axial charge density of the considered hadron, in agreement with the statements of Ref. [26]. To have a more detailed look at the problem we consider in the next section our system from the point of view of a moving observer.

Spatial distributions in moving frames

To obtain the expressions of spatial density distributions in moving frames, we follow the procedure of Ref. [20] and replace the packet in Eq. (6) with its boosted expression. For that we express a normalizable Heisenberg-picture state in a moving frame in terms of the spherically symmetric profile function ϕ(p)\phi(\mathbf{p}) (as (see also Ref. [30]):

∣Φ,X,s⟩v=∫d3p2E(2π)3γ(1−v⋅pE)ϕ(Λv−1p)∑s1Dss1[W(Λv,Λv−1pm)]e−ip⋅X∣p,s1⟩,(11)|\Phi,\mathbf{X},s\rangle_{\mathbf{v}}=\int\frac{d^3p}{\sqrt{2E(2\pi)^3}}\sqrt{\gamma\left(1-\frac{\mathbf{v}\cdot\mathbf{p}}{E}\right)}\phi(\Lambda_{\mathbf{v}}^{-1}\mathbf{p})\sum_{s_1}D_{s s_1}\left[W\left(\Lambda_{\mathbf{v}},\frac{\Lambda_{\mathbf{v}}^{-1}\mathbf{p}}{m}\right)\right]e^{-i\mathbf{p}\cdot\mathbf{X}}|\mathbf{p},s_1\rangle, \tag*{(11)}

where γ=(1−v2)−1/2\gamma=(1-v^2)^{-1/2}, E=m2+p2E=\sqrt{m^2+\mathbf{p}^2}, Λv\Lambda_{\mathbf{v}} is the Lorentz boost from the ZAMF to the moving frame, Λv−1p=v^(p⋅v^)+γ(p⋅v^−vE)v^\Lambda_{\mathbf{v}}^{-1}\mathbf{p}=\widehat{\mathbf{v}}(\mathbf{p}\cdot\widehat{\mathbf{v}})+\gamma(\mathbf{p}\cdot\widehat{\mathbf{v}}-vE)\widehat{\mathbf{v}}, and Ds1s[W]D_{s_1s}[W] is a spin-1/2 representation of the corresponding Wigner rotation WW [31]. In exact analogy to the case of the ZAMF, we find using the method of dimensional counting of Ref. [28] (or the ”strategy of regions” of Ref. [29]) that the only non-vanishing contribution to spatial density distributions is obtained by substituting P=P~/R\mathbf{P}=\widetilde{\mathbf{P}}/R, and expanding around R=0R=0. Doing so for the moving frame we obtain

JAϕ,v0(r)=∫d3P~ d3q(2π)3 γ(1−v⋅P~)P~⋅σ GA[(P~⋅q)2−q2]∣ϕ~(P~′)∣2e−iq⋅r,(12)J^{0}_{A\phi,v}(\mathbf{r})=\int\frac{d^{3}\widetilde{P}\,d^{3}q}{(2\pi)^{3}}\,\gamma\left(1-\mathbf{v}\cdot\widetilde{\mathbf{P}}\right)\widetilde{\mathbf{P}}\cdot\boldsymbol{\sigma}\,G_{A}\left[\left(\widetilde{\mathbf{P}}\cdot\mathbf{q}\right)^{2}-q^{2}\right]\left|\widetilde{\phi}(\widetilde{\mathbf{P}}^{\prime})\right|^{2}e^{-i\mathbf{q}\cdot\mathbf{r}}, \tag*{(12)}

where we have introduced P~′=v^×(P~×v^)+γ(P~⋅v^−vP~)v^\widetilde{\mathbf{P}}^{\prime}=\widehat{\mathbf{v}}\times\left(\widetilde{\mathbf{P}}\times\widehat{\mathbf{v}}\right)+\gamma\left(\widetilde{\mathbf{P}}\cdot\widehat{\mathbf{v}}-v\widetilde{P}\right)\widehat{\mathbf{v}} and a unit vector P~′≡m^\widetilde{\mathbf{P}}^{\prime}\equiv\widehat{\mathbf{m}}. Next, we change the integration variable P~→P′\widetilde{\mathbf{P}}\to\mathbf{P}^{\prime} and define a vector-valued function

n(v,m^)=v^×(m^×v^)+γ(m^⋅v^+v)v^.(13)\mathbf{n}(\mathbf{v},\widehat{\mathbf{m}})=\widehat{\mathbf{v}}\times\left(\widehat{\mathbf{m}}\times\widehat{\mathbf{v}}\right)+\gamma\left(\widehat{\mathbf{m}}\cdot\widehat{\mathbf{v}}+v\right)\widehat{\mathbf{v}}. \tag*{(13)}

Given that P~=v^×(P~′×v^)+γ(P~′⋅v^+vP~′)v^\widetilde{\mathbf{P}}=\widehat{\mathbf{v}}\times\left(\widetilde{\mathbf{P}}^{\prime}\times\widehat{\mathbf{v}}\right)+\gamma\left(\widetilde{\mathbf{P}}^{\prime}\cdot\widehat{\mathbf{v}}+v\widetilde{P}^{\prime}\right)\widehat{\mathbf{v}}, it follows that n^=P^\widehat{\mathbf{n}}=\widehat{\mathbf{P}}. Taking into account the Jacobian of the change of variables P~→P~′\widetilde{\mathbf{P}}\to\widetilde{\mathbf{P}}^{\prime} we obtain from Eq. (12):

JAϕ,v0(r)=∫dm^ P~′2dP~′ d3q(2π)3∣ϕ~(P~′)∣2e−iq⋅r n^⋅σ GA[(n^⋅q)2−q2].(14)J^{0}_{A\phi,v}(\mathbf{r})=\int\frac{d\widehat{\mathbf{m}}\,\widetilde{P}^{\prime2}d\widetilde{P}^{\prime}\,d^{3}q}{(2\pi)^{3}}\left|\widetilde{\phi}(\widetilde{\mathbf{P}}^{\prime})\right|^{2}e^{-i\mathbf{q}\cdot\mathbf{r}}\,\widehat{\mathbf{n}}\cdot\boldsymbol{\sigma}\,G_{A}\left[\left(\widehat{\mathbf{n}}\cdot\mathbf{q}\right)^{2}-q^{2}\right]. \tag*{(14)}

Using the spherical symmetry of ϕ~(P~′)=ϕ~(∣P~′∣)\widetilde{\phi}(\widetilde{\mathbf{P}}^{\prime})=\widetilde{\phi}(\left|\widetilde{\mathbf{P}}^{\prime}\right|) and Eq. (4), the integration over P~′\widetilde{\mathbf{P}}^{\prime} can be easily carried out, and we obtain the final expressions for the matrix element of the axial charge operator in moving frames:

JA,v0(r)=14π∫dm^d3q(2π)3e−iq⋅r n^⋅σ GA[(n^⋅q)2−q2].(15)J^{0}_{A,v}(\mathbf{r})=\frac{1}{4\pi}\int d\widehat{\mathbf{m}}\frac{d^{3}q}{(2\pi)^{3}}e^{-i\mathbf{q}\cdot\mathbf{r}}\,\widehat{\mathbf{n}}\cdot\boldsymbol{\sigma}\,G_{A}\left[\left(\widehat{\mathbf{n}}\cdot\mathbf{q}\right)^{2}-q^{2}\right]. \tag*{(15)}

The above result simplifies in two limiting cases. First, for v→0v\to0 we have n(v,m^)=m^+v+O(v2)\mathbf{n}(\mathbf{v},\widehat{\mathbf{m}})=\widehat{\mathbf{m}}+\mathbf{v}+O(v^{2}) and

JA,v0(r)=14π∫dm^ d3q(2π)3e−iq⋅r(m^⋅σ+v⋅σ)GA[(m^⋅q)2−q2]+12π∫dm^ d3q(2π)3e−iq⋅r m^⋅σ v⋅q m^⋅q GA′[(m^⋅q)2−q2]+O(v2).(16)\begin{aligned} J^{0}_{A,v}(\mathbf{r})={}&\frac{1}{4\pi}\int\frac{d\widehat{\mathbf{m}}\,d^{3}q}{(2\pi)^{3}}e^{-i\mathbf{q}\cdot\mathbf{r}}\left(\widehat{\mathbf{m}}\cdot\boldsymbol{\sigma}+\mathbf{v}\cdot\boldsymbol{\sigma}\right)G_{A}\left[\left(\widehat{\mathbf{m}}\cdot\mathbf{q}\right)^{2}-q^{2}\right]\\ &+\frac{1}{2\pi}\int\frac{d\widehat{\mathbf{m}}\,d^{3}q}{(2\pi)^{3}}e^{-i\mathbf{q}\cdot\mathbf{r}}\,\widehat{\mathbf{m}}\cdot\boldsymbol{\sigma}\,\mathbf{v}\cdot\mathbf{q}\,\widehat{\mathbf{m}}\cdot\mathbf{q}\,G_{A}^{\prime}\left[\left(\widehat{\mathbf{m}}\cdot\mathbf{q}\right)^{2}-q^{2}\right]+O(v^{2}). \tag*{(16)} \end{aligned}

For the integral over m^\widehat{\mathbf{m}} in the second line of Eq. (16) we choose the zz axis in the direction of q\mathbf{q}, and xx and yy axis such that σ\boldsymbol{\sigma} lies in xzxz plane. Introducing spherical coordinates m^=(sin⁡θcos⁡ϕ,sin⁡θsin⁡ϕ,cos⁡θ)\widehat{\mathbf{m}}=(\sin\theta\cos\phi,\sin\theta\sin\phi,\cos\theta) it is easily seen that the integral over ϕ\phi leads to vanishing result. Next, by taking into account in the first line that the integration over an odd function of m^\widehat{\mathbf{m}} vanishes, we are left with

JA,v0(r)=14πv⋅σ∫dm^ d3q(2π)3e−iq⋅rGA[(m^⋅q)2−q2]+O(v2).(17)J^{0}_{A,v}(\mathbf{r})=\frac{1}{4\pi}\mathbf{v}\cdot\boldsymbol{\sigma}\int\frac{d\widehat{\mathbf{m}}\,d^{3}q}{(2\pi)^{3}}e^{-i\mathbf{q}\cdot\mathbf{r}}G_{A}\left[\left(\widehat{\mathbf{m}}\cdot\mathbf{q}\right)^{2}-q^{2}\right]+O(v^{2}). \tag*{(17)}

The second limiting case corresponds to the IMF with v→1v\to1 and γ→∞\gamma\to\infty, for which the vector-valued function n^\widehat{\mathbf{n}} turns to v^\widehat{\mathbf{v}}. In this case the integration over m^\widehat{\mathbf{m}} becomes trivial. Thus, for the matrix element of the axial charge density operator in the IMF we obtain

JA,IMF0(r)=v^⋅σ∫d3q(2π)3e−iq⋅rGA[(v^⋅q)2−q2]=δ(r∥)v^⋅σ∫d2q⊥(2π)2e−iq⊥⋅r⊥GA[−q⊥2],(18)J^{0}_{A,\mathrm{IMF}}(\mathbf{r})=\widehat{\mathbf{v}}\cdot\boldsymbol{\sigma}\int\frac{d^{3}q}{(2\pi)^{3}}e^{-i\mathbf{q}\cdot\mathbf{r}}G_{A}\left[\left(\widehat{\mathbf{v}}\cdot\mathbf{q}\right)^{2}-q^{2}\right]=\delta(r_{\parallel})\widehat{\mathbf{v}}\cdot\boldsymbol{\sigma}\int\frac{d^{2}q_{\perp}}{(2\pi)^{2}}e^{-i\mathbf{q}_{\perp}\cdot\mathbf{r}_{\perp}}G_{A}\left[-q_{\perp}^{2}\right], \tag*{(18)}

where r∥≡∣r∥∣r_{\parallel}\equiv\left|\mathbf{r}_{\parallel}\right| with r∥≡r⋅v^v^\mathbf{r}_{\parallel}\equiv\mathbf{r}\cdot\widehat{\mathbf{v}}\widehat{\mathbf{v}} and r⊥≡∣r⊥∣r_{\perp}\equiv\left|\mathbf{r}_{\perp}\right| with r⊥≡r−r∥\mathbf{r}_{\perp}\equiv\mathbf{r}-\mathbf{r}_{\parallel}. The appearance of δ(r∥)\delta(r_{\parallel}) in these expressions reflects the fact that in the IMF, the system is Lorentz-contracted to a two-dimensional object perpendicular to the velocity v^\widehat{\mathbf{v}} of the moving frame.

To obtain the expressions of spatial density distributions for the static approximation in moving frames, we apply the 1/m1/m expansion to the integrand in Eq. (12) and obtain at leading order:

JAϕ,naive,v0(r)=∫d3P d3q(2π)3γP⋅σmGA[−q2]ϕ∗[Λv−1(P−q/2)]ϕ[Λv−1(P+q/2)]e−iq⋅r,(19)J^{0}_{A\phi,\mathrm{naive},v}(\mathbf{r})=\int\frac{d^{3}P\,d^{3}q}{(2\pi)^{3}}\gamma\frac{\mathbf{P}\cdot\boldsymbol{\sigma}}{m}G_{A}\left[-q^{2}\right]\phi^{*}\left[\Lambda_{v}^{-1}\left(\mathbf{P}-\mathbf{q}/2\right)\right]\phi\left[\Lambda_{v}^{-1}\left(\mathbf{P}+\mathbf{q}/2\right)\right]e^{-i\mathbf{q}\cdot\mathbf{r}}, \tag*{(19)}

where Λv−1p=v^×(p×v^)+γ(p⋅v^−mv)v^+O(1/m)\Lambda_{v}^{-1}\mathbf{p}=\widehat{\mathbf{v}}\times(\mathbf{p}\times\widehat{\mathbf{v}})+\gamma(\mathbf{p}\cdot\widehat{\mathbf{v}}-mv)\widehat{\mathbf{v}}+O(1/m). Next, we change the integration variable P→P′\mathbf{P}\to\mathbf{P}^{\prime} by substituting P′=v^×(P×v^)+γ(P⋅v^−mv)v^\mathbf{P}^{\prime}=\widehat{\mathbf{v}}\times(\mathbf{P}\times\widehat{\mathbf{v}})+\gamma(\mathbf{P}\cdot\widehat{\mathbf{v}}-mv)\widehat{\mathbf{v}} and find that the Jacobian of the change of variables cancels the factor γ\gamma in the integrand of Eq. (19). For sharply localized packets, the integration over P′\mathbf{P}^{\prime} can be carried out easily by substituting P′=P~′/R\mathbf{P}^{\prime}=\widetilde{\mathbf{P}}^{\prime}/R and expanding at R=0R=0. Doing so in the leading order of the 1/m1/m expansion we obtain:

JA,naive,v0(r)=v⋅σ∫d3q(2π)3GA[−q2]e−iq⋅r.(20)J^{0}_{A,\mathrm{naive},v}(\mathbf{r})=\mathbf{v}\cdot\boldsymbol{\sigma}\int\frac{d^{3}q}{(2\pi)^{3}}G_{A}\left[-q^{2}\right]e^{-i\mathbf{q}\cdot\mathbf{r}}. \tag*{(20)}

Interpretation in terms of the axial charge density and discussion

To shed light on the interpretation of the obtained spatial distributions it is instructive to look at the corresponding radial moments in moving frames. In three spatial dimensions, we find

⟨r2k⟩v=∫d3r r2kJA,v0(r)=v⋅σ[2kk!GA(k)(0)]+O(v2),(21)\langle r^{2k}\rangle_v = \int d^3r\,r^{2k}J^0_{A,v}(\mathbf{r}) = \mathbf{v}\cdot\boldsymbol{\sigma}\left[2^k k!G_A^{(k)}(0)\right] + O(v^2), \tag*{(21)}

where GA(k)(0)G_A^{(k)}(0) is the kkth derivative of GA(q2)G_A(q^2) with respect to q2q^2 taken at q2=0q^2=0. Similarly, for the naive distribution JA,naive,v0(r)J^0_{A,\mathrm{naive},v}(\mathbf{r}), one has

⟨r2k⟩v,naive=∫d3r r2kJA,naive,v0(r)=v⋅σ[(2k+1)!k!GA(k)(0)],(22)\langle r^{2k}\rangle_{v,\mathrm{naive}} = \int d^3r\,r^{2k}J^0_{A,\mathrm{naive},v}(\mathbf{r}) = \mathbf{v}\cdot\boldsymbol{\sigma}\left[\frac{(2k+1)!}{k!}G_A^{(k)}(0)\right], \tag*{(22)}

Notice that the expressions in the square brackets of Eqs. (21) and (22) coincide with the corresponding expressions for the moments of the electric charge distribution for a scalar particle of Ref. [20] upon replacing the form factor GA(q2)G_A(q^2) with F(q2)F(q^2).

We are now in the position to define the axial charge density. Clearly and as already mentioned above, the distributions JA,v0(r)J^0_{A,v}(\mathbf{r}) and JA,naive,v0(r)J^0_{A,\mathrm{naive},v}(\mathbf{r}) cannot be interpreted as densities of the axial charge. In particular, they change sign upon spatial reflections as follows from their very definition in Eq. (6), i.e., they behave as pseudoscalar densities.4 This also implies that the moments of these distributions are pseudoscalar quantities, which is certainly not what one expects from the density distribution of any quantity including the axial charge. In particular, we expect the properly defined axial charge density to be normalized to the axial charge gA=GA(0)g_A=G_A(0) in any frame. The expressions for the radial moments of the distributions JA,v0(r)J^0_{A,v}(\mathbf{r}) and JA,naive,v0(r)J^0_{A,\mathrm{naive},v}(\mathbf{r}) in Eqs. (21) and (22) have the form of the spin- and velocity-dependent prefactors, that do not depend on the internal structure of the system, times the frame-independent factors which incorporate the whole information about the axial charge distribution encoded in the form factor GAG_A. This suggests the definition of the properly normalized axial charge densities ρA,v(r)\rho_{A,v}(\mathbf{r}) and ρA,naive,v(r)\rho_{A,\mathrm{naive},v}(\mathbf{r}) by dropping the above-mentioned structure-independent prefactors via

JA,v0(r)=:v⋅σ ρA,v(r),JA,naive,v0(r)=:v⋅σ ρA,naive,v(r).(23)J^0_{A,v}(\mathbf{r}) =: \mathbf{v}\cdot\boldsymbol{\sigma}\,\rho_{A,v}(\mathbf{r}), \qquad J^0_{A,\mathrm{naive},v}(\mathbf{r}) =: \mathbf{v}\cdot\boldsymbol{\sigma}\,\rho_{A,\mathrm{naive},v}(\mathbf{r}). \tag*{(23)}

Notice that the normalization of densities defined via Eq. (23) is uniquely fixed by the condition that the axial charge of the considered system is given by GA(0)G_A(0). This way, the naive definition of the axial charge density becomes independent of the frame velocity and reduces to the traditional Breit-frame expression,

ρA,naive(r)=∫d3q(2π)3GA[−q2]e−iq⋅r.(24)\rho_{A,\mathrm{naive}}(\mathbf{r}) = \int\frac{d^3q}{(2\pi)^3}G_A\left[-\mathbf{q}^2\right]e^{-i\mathbf{q}\cdot\mathbf{r}}. \tag*{(24)}

On the other hand the density in the ZAMF is obtained from Eq. (17) by dropping the factor v⋅σ\mathbf{v}\cdot\boldsymbol{\sigma} and taking the limit v→0v\to0. The resulting expression reads:

ρA,ZAMF(r)=lim⁡v→0ρA,v(r)=14π∫d2m^ d3q(2π)3GA[(m^⋅q)2−q2]e−iq⋅r.(25)\rho_{A,\mathrm{ZAMF}}(\mathbf{r}) = \lim_{v\to0}\rho_{A,v}(\mathbf{r}) = \frac{1}{4\pi}\int\frac{d^2\hat{\mathbf{m}}\,d^3q}{(2\pi)^3}G_A\left[(\hat{\mathbf{m}}\cdot\mathbf{q})^2-\mathbf{q}^2\right]e^{-i\mathbf{q}\cdot\mathbf{r}}. \tag*{(25)}

Analogously in the infinite-momentum frame from Eq. (18) we obtain

ρA,IMF(r)=∫d3q(2π)3GA[(v^⋅q)2−q2]e−iq⋅r=δ(r∥)∫d2q⊥(2π)2GA[−q⊥2]e−iq⊥⋅r⊥.(26)\rho_{A,\mathrm{IMF}}(\mathbf{r}) = \int\frac{d^3q}{(2\pi)^3}G_A\left[(\hat{\mathbf{v}}\cdot\mathbf{q})^2-\mathbf{q}^2\right]e^{-i\mathbf{q}\cdot\mathbf{r}} = \delta(r_\parallel)\int\frac{d^2q_\perp}{(2\pi)^2}G_A\left[-\mathbf{q}_\perp^2\right]e^{-i\mathbf{q}_\perp\cdot\mathbf{r}_\perp}. \tag*{(26)}

Equations (26) and (25) represent our final result for the axial charge density in the IMF and ZAMF, respectively, defined by sharply localized spherically symmetric wave packets.5 Interestingly, these densities coincide with the corresponding expressions for the electric charge density of Ref. [20] upon replacing the electric form factor F(q2)F(q^2) by the axial one GA(q2)G_A(q^2). The obtained expressions for the axial charge density are applicable to systems regardless of their mass and, in particular, remain valid for light systems, whose Compton wave length is comparable to or larger than the characteristic scale(s) of the axial charge distribution. Last but not least, we emphasize that since the axial charge density depends on the relative distance r=x−Xr = x - X, XX should be interpreted as the position of the center of the axial charge distribution of the considered system.

We have also considered a naive definition of the axial charge density obtained by taking the static limit of the integrand in Eq. (12), which corresponds to localizing the system at distances much larger than the Compton wavelength, while much smaller than other characteristic scale(s) of the system. The resulting naive axial density coincides with the three-dimensional Breit-frame distribution and is frame-independent. Notice, however, that such a definition depends on the form of the wave packet the system is prepared in if one goes beyond the static approximation, see Ref. [20] for more details. Still, this way of defining the axial charge density is appropriate and, in fact, quite natural for heavy systems. Notice, however, that the sharp localization limit of R→0R \to0 does not commute with the static limit, which is why the resulting (mass-independent) expressions for the densities and their spherical moments in Eqs. (21) and (22) differ from each other, see also Ref. [20]. In particular, one has

⟨r2⟩A=∫d3r r2ρA,V(r)=4GA′(0),(27)\langle r^2 \rangle_A = \int d^3r\,r^2\rho_{A,V}(\mathbf{r}) = 4G_A'(0), \tag*{(27)}

while the conventional definition of the axial charge radius is given by

⟨r2⟩A,naive=∫d3r r2ρA,naive(r)=6GA′(0).(28)\langle r^2 \rangle_{A,\mathrm{naive}} = \int d^3r\,r^2\rho_{A,\mathrm{naive}}(\mathbf{r}) = 6G_A'(0). \tag*{(28)}

Acknowledgments

We thank Cédric Lorcé for discussions and critical comments. This work was supported in part by the MKW NRW under the funding code NW21-024-A, by the Georgian Shota Rustaveli National Science Foundation (Grant No. FR-23-856), by ERC NuclearTheory (grant No. 885150) and ERC EXOTIC (grant No. 101018170), by CAS through a President’s International Fellowship Initiative (PIFI) (Grant No. 2025PD0022), and by the EU Horizon 2020 research and innovation programme (STRONG-2020, grant agreement No. 824093).

References

  1. [1]R. Hofstadter, F. Bumiller, and M. R. Yearian, Rev. Mod. Phys. 30, 482 (1958).
  2. [2]F. J. Ernst, R. G. Sachs and K. C. Wali, Phys. Rev. 119, 1105-1114 (1960).
  3. [3]R. G. Sachs, Phys. Rev. 126, 2526-2260 (1962).
  4. [4]M. V. Polyakov and A. G. Shuvaev, [arXiv:hep-ph/0207153 [hep-ph]].
  5. [5]M. V. Polyakov, Phys. Lett. B 555, 57 (2003), [hep-ph/0210165].
  6. [6]M. V. Polyakov and P. Schweitzer, Int. J. Mod. Phys. A 33 (2018) no.26, 1830025.
  7. [7]M. Burkardt, Phys. Rev. D 62 (2000), 071503(R), [erratum: Phys. Rev. D 66 (2002), 119903(E)], [arXiv:hep-ph/0005108 [hep-ph]].arxiv.org/abs/hep-ph/0005108
  8. [8]G. A. Miller, Phys. Rev. Lett. 99, 112001 (2007), [arXiv:0705.2409 [nucl-th]].DOI
  9. [9]G. A. Miller, Phys. Rev. C 79, 055204 (2009), [arXiv:0901.1117 [nucl-th]].DOI
  10. [10]G. A. Miller, Ann. Rev. Nucl. Part. Sci. 60 (2010), 1-25, [arXiv:1002.0355 [nucl-th]].DOI
  11. [11]R. L. Jaffe, Phys. Rev. D 103 (2021) no.1, 016017, [arXiv:2010.15887 [hep-ph]].DOI
  12. [12]G. A. Miller, Phys. Rev. C 99, no.3, 035202 (2019), [arXiv:1812.02714 [nucl-th]].DOI
  13. [13]A. Freese and G. A. Miller, Phys. Rev. D 103, 094023 (2021), [arXiv:2102.01683 [hep-ph]].DOI
  14. [14]Y. Guo, X. Ji and K. Shiells, Nucl. Phys. B 969, 115440 (2021), [arXiv:2101.05243 [hep-ph]].DOI
  15. [15]J. Y. Panteleeva and M. V. Polyakov, Phys. Rev. D 104 (2021) no.1, 014008, [arXiv:2102.10902 [hep-ph]].DOI
  16. [16]A. Freese and G. A. Miller, Phys. Rev. D 105, no.1, 014003 (2022), [arXiv:2108.03301 [hep-ph]].DOI
  17. [17]C. Lorcé, Phys. Rev. Lett. 125, no.23, 232002 (2020), [arXiv:2007.05318 [hep-ph]].
  18. [18]C. Lorcé, P. Schweitzer and K. Tezgin, Phys. Rev. D 106, no.1, 014012 (2022), [arXiv:2202.01192 [hep-ph]].DOI
  19. [19]C. Lorcé, H. Moutarde and A. P. Trawiński, Eur. Phys. J. C 79, no.1, 89 (2019), [arXiv:1810.09837 [hep-ph]].DOI
  20. [20]E. Epelbaum, J. Gegelia, N. Lange, U.-G. Meißner and M. V. Polyakov, Phys. Rev. Lett. 129, no.1, 012001 (2022), [arXiv:2201.02565 [hep-ph]].DOI
  21. [21]N. G. Fleming, Physical Reality & Math. Descrip., 357 (1974).
  22. [22]J. Y. Panteleeva, E. Epelbaum, J. Gegelia and U.-G. Meißner, Phys. Rev. D 106, no.5, 056019 (2022), [arXiv:2205.15061 [hep-ph]].DOI
  23. [23]J. Y. Panteleeva, E. Epelbaum, J. Gegelia and U.-G. Meißner, Eur. Phys. J. C 83, no.7, 617 (2023), [arXiv:2211.09596 [hep-ph]].arxiv.org/abs/2211.09596
  24. [24]H. Alharazin, B. D. Sun, E. Epelbaum, J. Gegelia and U.-G. Meißner, JHEP 02, 163 (2023), [arXiv:2212.11505 [hep-ph]].DOI
  25. [25]J. Y. Panteleeva, E. Epelbaum, J. Gegelia and U.-G. Meißner, JHEP 07, 237 (2023). [arXiv:2305.01491 [hep-ph]].DOI
  26. [26]Y. Chen, Y. Li, C. Lorcé and Q. Wang, Phys. Rev. D 110, no.9, L091503 (2024), [arXiv:2405.12943 [hep-ph]].arxiv.org/abs/2405.12943
  27. [27]V. Bernard, L. Elouadhriri and U.-G. Meißner, J. Phys. G 28 (2002), R1-R35, [arXiv:hep-ph/0107088 [hep-ph]].DOI
  28. [28]J. Gegelia, G. S. Japaridze and K. S. Turashvili, Theor. Math. Phys. 101, 1313-1319 (1994).
  29. [29]M. Beneke and V. A. Smirnov, Nucl. Phys. B 522, 321-344 (1998).
  30. [30]S. E. Hoffmann, [arXiv:1804.00548 [quant-ph]].arxiv.org/abs/1804.00548
  31. [31]S. Weinberg, “The Quantum theory of fields. Vol. 1, : Foundations,” Cambridge University Press (2005-06-02).

Paper details

vlnc:W_4gkf3tt9
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