Electroweak Currents in Few-Body Systems

Few-body electroweak observables are matrix elements of a matched current between consistently computed nuclear states, not properties of a potential or a current operator in isolation. The leading one-nucleon charge and spin–isospin operators are supplemented by pion-range and contact many-body currents according to the chosen EFT. Regulator independence then requires the Hamiltonian and current to share their cutoff, symmetries, low-energy constants, and unitary convention.

Required background. Nuclear Forces in the Chiral Expansion supplies the force hierarchy, reducible iteration, regulator choices, and distinction between a potential and an observable.

Helpful background. The Fermi Limit of Weak Interactions supplies the normalization and Lorentz structure of low-energy semileptonic interactions.

From external sources to nuclear operators

At momentum transfer $q$ below the EFT breakdown scale, couple the low-energy theory to external vector, axial, scalar, and pseudoscalar sources. Functional differentiation defines the current, while matching fixes its coefficients to the Standard Model, single-nucleon form factors, or specified few-body data. For a weak process, the nuclear part of the amplitude has the form

$$\mathcal M_{fi}=\mathcal N_{\rm EW}\, L_\mu\langle\Psi_f^{(-)}|J^\mu(q)|\Psi_i^{(+)}\rangle,$$

where $\mathcal N_{\rm EW}$ contains the declared factors such as $G_F$, CKM elements, and radiative matching. The incoming and outgoing states solve the same strong Hamiltonian used to derive the current. Their scattering boundary conditions matter whenever a continuum channel is open.

The operator decomposes by particle rank,

$$J^\mu=\sum_i J_{i,\mathrm{1b}}^\mu +\sum_{i<j}J_{ij,\mathrm{2b}}^\mu +\sum_{i<j<k}J_{ijk,\mathrm{3b}}^\mu+\cdots .$$

With isospin generators $\tau^a/2$, representative zero-momentum one-body limits are

$$V^{0,a}_{\mathrm{1b}}=N^\dagger\frac{\tau^a}{2}N, \qquad \boldsymbol A^a_{\mathrm{1b}} =g_A N^\dagger\boldsymbol\sigma\frac{\tau^a}{2}N.$$

Finite-$q$ recoil, weak-magnetism, induced-pseudoscalar, and nucleon-size terms enter at their assigned orders. These are already matched one-body structures; adding a nuclear contact current must not refit or duplicate them.

For angular-momentum eigenstates it is useful to project $J^\mu$ onto Coulomb, longitudinal, transverse electric, and transverse magnetic multipoles $C_J$, $L_J$, $E_J$, and $M_J$. This exposes parity and angular-momentum selection rules and isolates the long-wavelength limit. Multipole conventions include phases and factors of $q$; a calculation must state them before comparing reduced matrix elements from different sources.

Counting one- and many-body currents

The relevant degrees of freedom and counting depend on resolution:

Regime	Strong states	Current hierarchy and matching
Chiral EFT, $Q\lesssim m_\pi$	Nucleons with explicit pion exchange; selected potentials iterated	One-body currents plus pion-range exchange and short-range contacts ordered in $Q/\Lambda_b$; pion–nucleon LECs can appear in both forces and currents
Pionless EFT, $Q\ll m_\pi$	Contact interactions with large S-wave scattering lengths resummed	One-body currents plus derivative and contact two-body currents; coefficients such as $L_{1,A}$ are scheme-dependent and require matching beyond elastic NN scattering

“Higher body” does not automatically mean negligible. A nominal suppression can be offset by a forbidden or accidentally small one-body matrix element, large scattering lengths, or coherent sums. Conversely, iterating the strong potential does not imply iterating a current insertion: an ordinary transition amplitude contains one insertion of the matched current between nonperturbative states.

In chiral EFT, long-range two-body axial currents contain pion exchange and pion–nucleon coefficients conventionally denoted $c_3$ and $c_4$. A short-range axial coupling can be related to the three-nucleon-force coefficient $c_D$ only after fixing the regulator, normalization, and unitary convention. In pionless EFT the analogous short-distance response is encoded in a different coupling, often $L_{1,A}$; numerical values cannot be transferred between these theories. The construction and renormalization of chiral nuclear currents are reviewed in Krebs 2020, §§ 2–5.

Current and potential are representation dependent. If a unitary transformation changes $H$ to $H'=UHU^\dagger$, consistency requires

$$J^{\mu\prime}=U J^\mu U^\dagger, \qquad |\Psi'\rangle=U|\Psi\rangle.$$

Only the complete matrix element is invariant. Combining wave functions from one regulator or unitary convention with currents from another leaves uncanceled short-distance dependence, even when both ingredients are individually called the same chiral order.

Symmetry identities and regulator consistency

For a conserved vector current, on-shell matrix elements satisfy

$$q_\mu\langle J_V^\mu\rangle=0.$$

In Hamiltonian language and a consistent Fourier convention this becomes

$$\boldsymbol q\cdot\boldsymbol J_V(\boldsymbol q) =[H,\rho_V(\boldsymbol q)],$$

whose matrix element gives $(E_f-E_i)\langle\rho_V\rangle$. A regulator applied to the potential but not to the exchange current generally spoils this identity; deriving or regulating both together makes the missing term visible. The axial current is not conserved: its divergence obeys the appropriate partially conserved axial-current relation, including the pion-pole and explicit chiral-symmetry-breaking terms. Imposing vector transversality on it would be an error.

The same consistency requirement connects long-wavelength electric multipoles to the charge density through the continuity equation. This “Siegert” organization can incorporate pieces of exchange currents fixed by charge conservation, but it does not eliminate independent transverse or axial contact operators. Krebs emphasizes that inconsistent regularization of forces and currents leaves symmetry-violating terms at the working order Krebs 2020, §§ 4.4–4.5.

Representative Gamow–Teller workflow

For an allowed low-momentum charged-current transition, strip off the leptonic normalization and form the reduced nuclear matrix element

$$M_{\rm GT} =\left\langle\Psi_f\left\|\, g_A\sum_i\boldsymbol\sigma_i\tau_i^- +\boldsymbol J_{A,\mathrm{2b}}^- +\cdots\right\|\Psi_i\right\rangle.$$

A controlled calculation proceeds as follows:

1. Declare kinematics and normalization. State $q^0$, $|\boldsymbol q|$, state normalization, reduced-matrix-element convention, and which electroweak and radiative factors are outside $M_{\rm GT}$.
2. Choose the common EFT setup. Fix chiral or pionless degrees of freedom, order, cutoff, Hamiltonian, and the current derived in the same scheme.
3. Match shared and current-only coefficients. Propagate the joint covariance of strong LECs, single-nucleon inputs, and current contacts. If the transition itself calibrates a contact, it is not a held-out validation observable. A concrete chiral calculation linking an axial contact to the three-nucleon interaction is Gazit, Quaglioni, and Navrátil 2009, pp. 1–4.
4. Solve the states and insert the current. Converge bound or scattering states numerically, then evaluate one- and many-body pieces and their interference—not just their separate squared magnitudes.
5. Attach corrections once. Combine recoil, radiative, isospin-breaking, Coulomb, and finite-size effects with explicit scheme labels so that terms already contained in form factors or matching coefficients are not counted again.

For outputs $y_i$ depending on shared parameters $\theta_a$, the linearized parameter covariance is

$$(C_y^{\rm param})_{ij} =\sum_{a,b} \frac{\partial y_i}{\partial\theta_a} (C_\theta)_{ab} \frac{\partial y_j}{\partial\theta_b}.$$

Add truncation, numerical, radiative, and experimental components only after identifying their cross-covariances. Treating a shared current LEC as independent in every process destroys precisely the correlation that makes a joint electroweak analysis informative. The full method is developed on Nuclear Predictions, Uncertainties, and Evidence Across Methods.

Independent checks and handoffs

• Charge and symmetry. Recover the correctly normalized conserved charge as $\boldsymbol q\to0$ and test the vector continuity equation at the operator or matrix-element level.
• Cutoff and order. Refit only the coefficients assigned as calibration inputs when varying the common cutoff. The residual variation and successive-order increments should decrease according to the declared counting; regulator spread alone is not an uncertainty distribution.
• Dimensions and units. Track powers of momentum in multipoles, phase space, and $\mathcal N_{\rm EW}$. Quote whether a reported number is a dimensionless reduced matrix element, a rate, or a cross section.
• Unitarity and final-state interactions. Below the first inelastic threshold, a transition into a two-body scattering channel must carry the strong final-state phase required by unitarity. Using a plane wave in one contribution and a distorted wave in another fails this check.
• Impulse limit. Turning off many-body currents should reproduce the declared one-body approximation, not a differently normalized observable.

When a current matrix element is extracted from a Euclidean finite-volume calculation, the conversion to an infinite-volume transition amplitude requires the finite-volume normalization and channel-mixing factors owned by Finite-Volume Matrix Elements and One-to-Two Transitions. A finite-volume matrix element should not be inserted directly into the continuum formula above.

Common pitfalls

Calling a current coefficient a force prediction. Shared symmetry and matching can relate coefficients, but the relation is convention dependent and can contain additional contact terms. State the regulator and operator basis before using a force fit in a current.

Mixing orders across ingredients. A high-order wave function does not upgrade a lower-order current. Quote the order of the Hamiltonian, current, single-nucleon input, and radiative correction separately, then use the lowest controlled combination.

Validating on calibration data. If a beta-decay rate fixes the axial contact, agreement with that same rate checks implementation, not prediction. Reserve an independent process or kinematic region for validation.

References

• Gazit, Doron, Sofia Quaglioni, and Petr Navrátil. “Three-Nucleon Low-Energy Constants from the Consistency of Interactions and Currents in Chiral Effective Field Theory.” Physical Review Letters 103 (2009): 102502. DOI.
• Krebs, Hermann. “Nuclear Currents in Chiral Effective Field Theory.” European Physical Journal A 56 (2020): 234. DOI.