Heat kernel approach to the one-loop effective action for nonlinear electrodynamics

We develop a heat kernel method to compute the one-loop effective action for a general class of nonlinear electrodynamic (NLED) theories in four dimensional Minkowski spacetime. Working in the background field formalism, we extract the logarithmically divergent part of the effective action, the so-called induced action, corresponding to the DeWitt $a_2$ coefficient of the heat kernel. In NLED, quantisation yields non-minimal differential operators, for which standard heat kernel techniques are not immediately applicable. Considering the weak-field regime, we calculate the $a_0$, $a_1$ and $a_2$ contributions to leading order in the background electromagnetic field strength. Finally, we consider conformal NLED theories and compute the $a_0$ contribution to all orders. For this class, we comment on the role of causality being necessary and sufficient for the convergence of the exact $a_1$ and $a_2$ contributions.

💡 Research Summary

The paper presents a systematic heat‑kernel method for computing the one‑loop effective action of a broad class of nonlinear electrodynamics (NLED) theories in four‑dimensional Minkowski space. After a brief historical motivation—starting from the Born‑Infeld model and its resurgence in string theory—the authors introduce the general NLED Lagrangian L(α,β), where α = ¼ F_{ab}F^{ab} and β = ¼ \tilde F_{ab}F^{ab}. They discuss duality‑invariant (self‑dual) and conformal NLED models, and emphasize the role of causality conditions (both weak‑field and strong‑field) for physical consistency.

In Section 2 the authors perform the background‑field split A_a = A_{B,a}+A_{Q,a} and adopt the Lorenz gauge χ(A_Q)=∂^aA_{Q,a}. Integrating out the quantum vector field yields a functional determinant of a second‑order differential operator Δ_{ab}. Unlike the minimal operator of the Schwinger‑DeWitt formalism, Δ_{ab} contains a field‑dependent principal matrix H_{ab}=−L_α η_{ab} 𝟙₄+G_{ab} and a first‑derivative term V_a, making it genuinely non‑minimal. The tensor G_{acdb} involves second derivatives of the Lagrangian (L_{αα}, L_{αβ}, L_{ββ}) contracted with the background field strength F_{ab} and its dual.

Section 3 introduces a Volterra‑series expansion to treat the exponential of the non‑minimal operator. By Fourier transforming the delta function, the authors replace ordinary derivatives with X_a=∂a+i k_a and write
e^{isΔ}=e^{is H∂∂} ∑{n=0}^∞ ∫0^1 dy_1 … ∫0^{y{n-1}} dy_n Π{i=1}^n e^{-y_i H∂∂} V·∂ e^{y_i H∂∂}.
This representation allows a systematic perturbative expansion in the background field strength. Working in the weak‑field regime (α,β≪1) they compute the DeWitt coefficients a_0, a_1 and a_2 up to quartic order in F. The a_0 term is essentially the trace of the identity matrix, modified by the factor L_α^{-2}. The a_1 coefficient receives contributions from V_a and the gradients of L_α and L_β, producing terms of order F^2. The a_2 coefficient is the most intricate; it contains products of G_{acdb} and V_a, leading to F^4 structures that reproduce the known induced action of Born‑Infeld theory. The logarithmically divergent part of the effective action is identified as Γ_div = -(ln Λ)/(4π)^2 ∫ d^4x a_2(x).

Section 4 focuses on conformal NLED theories, defined by the condition α L_α+β L_β = L and L_{αα}L_{ββ}−L_{αβ}^2 = 0. In this case the principal matrix H_{ab} reduces to a scalar function times the metric, and the Volterra series converges exactly. The authors compute the a_0 coefficient to all orders in the background field, showing that it resums to a closed expression depending only on the invariant combination α^2+β^2. They then analyze the convergence of the a_1 and a_2 series and prove that the weak‑field causality constraints (L_{αα}≥0, L_{ββ}≥0, L_{αα}L_{ββ}−L_{αβ}^2≥0) are necessary for convergence, while the strong‑field condition (−L_α + α(L_{ββ}−L_{αα})−2βL_{αβ}−(L_{αα}+L_{ββ})(α^2+β^2) > 0) is sufficient to guarantee absolute convergence. Thus, causality, originally a classical requirement for subluminal wave propagation, emerges as a quantum consistency condition for the heat‑kernel expansion.

The conclusion (Section 5) emphasizes that the presented heat‑kernel framework provides a general tool for handling non‑minimal operators in NLED, enabling the calculation of one‑loop quantum corrections for a wide variety of models, including Born‑Infeld, ModMax, and higher‑derivative extensions. The link between causality and the convergence of the quantum effective action is highlighted as a novel insight. Appendices supply technical details: matrix identities specific to four dimensions, a basis for four‑derivative quartic‑field structures, and algebraic properties of the tensor G_{abcd}. Overall, the paper bridges a gap between classical nonlinear electrodynamics and its quantum treatment, offering both methodological advances and physical interpretations.