Functorial properties of Schwinger–DeWitt expansion and Mellin–Barnes representation
A. O. Barvinsky, A. E. Kalugin, W. Wachowski · 2025
Read Functorial properties of Schwinger–DeWitt expansion and Mellin–Barnes representation by A. O. Barvinsky, A. E. Kalugin, W. Wachowski (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-03944.txt.
Introduction
Quantum field theory on curved spacetime needs the response of operators built from differential equations on that background. Andrei O. Barvinsky, Alexey E. Kalugin, and Władysław Wachowski study integral kernels for such operator functions in a 2025 preprint. Their starting point is the heat kernel and its classic DeWitt expansion.
The authors show the kernel for a general operator function can be expanded by integrating the heat kernel series term by term. The expansion neatly separates geometry, carried by the standard coefficients, from the choice of function, carried by new scalar kernels they define. For important function shapes, those kernels are computed as multiple Mellin-Barnes integrals. The paper also handles delicate points: when term-by-term integration is legitimate, how to regularise divergent integrals, and what the expansion means physically.
Catalog record
| CREATOR | A. O. Barvinsky, A. E. Kalugin, W. Wachowski |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 118 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-03944
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Functorial properties of Schwinger–DeWitt expansion and Mellin–Barnes representation A. O. Barvinsky, A. E. Kalugin, W. Wachowski Abstract We consider integral kernels for functions f ( F ^ ) f(\hat{F}) of a minimal second-order differential operator F ^ ( ∇ ) \hat{F}(\nabla) on a curved spacetime. We show that they can be expanded in a functional series, analogous to the DeWitt expansion for the heat kernel, by integrating the latter term-by-term. This procedure leads to a separation of two types of data: all information about the bundle geometry and the operator F ^ ( ∇ ) \hat{F}(\nabla) is still contained in the standard HaMiDeW coefficients a ^ k [ F | x , x ′ ] \hat{a}_{k}[F|x,x^{\prime}] (we call this property “off-diagonal functoriality”), while information about the function f f is encoded in some new scalar functions 𝔹 α [ f | σ ] \mathbb{B}_{\alpha}[f|\sigma] and 𝕎 α [ f | σ , m 2 ] \mathbb{W}_{\alpha}[f|\sigma,m^{2}] , which we call basis and complete massive kernels, respectively. These objects are calculated for operator functions of the form exp ( − τ F ^ ν ) / ( F ^ μ + λ ) \exp(-\tau\hat{F}^{\nu})/(\hat{F}^{\mu}+\lambda) as multiple Mellin–Barnes integrals. The article also discusses subtle issues such as the validity of the term-by-term integration, the regularization of IR divergent integrals, and the physical interpretation of the resulting expansions. 1 Introduction The heat kernel method is an important tool in the arsenal of modern mathematical physics DeWitt 1965 ; Barvinsky and Vilkovisky 1985 ; Gilkey 1975 ; Gilkey 1995 . Let us briefly remind the reader the basics of it while simultaneously fixing our conventions. Let \hat{F}(\nabla) be a non-negatively definite (pseudo)differential operator acting on sections of a vector bundle 1 1 1 Hats everywhere indicate the matrix nature of the objects under consideration, for example, \hat{1}=\delta^{A}_{B} , where A,B are indices in fibers (they run over the set of all fields of the theory and can have both internal, spacetime, and spinor components). Unless otherwise stated, we will assume that when an operator acts on a two-point function, it always acts with respect to its first argument. over a Riemannian manifold \mathcal{M} with metric g_{ab} . Then the integral kernel of the function of \hat{F}(\nabla) is \hat{K}[f(F)|x,x^{\prime}]=f(\hat{F})\,\delta(x,x^{\prime}), (1.1) where \delta(x,x^{\prime}) is the delta function. It is a two-point (i.e., depending on two points x,x^{\prime}\in\mathcal{M} ) matrix-valued function. Then the action of an operator f(\hat{F}) on the section of the bundle \bm{\varphi}=\varphi^{A} is represented by an integral convolution with the corresponding kernel: f(\hat{F})\,\bm{\varphi}(x)=\int\limits_{\mathcal{M}}d^{d}x^{\prime}\,\hat{K}[\,f(F)|x,x^{\prime}]\,\bm{\varphi}(x^{\prime}). (1.2) In particular, the heat kernel is the integral kernel of the operator exponential \hat{K}_{F}(\tau|x,x^{\prime})=\hat{K}[e^{-\tau F}|x,x^{\prime}], (1.3) where the additional parameter \tau is referred to as proper time. The heat kernel finds application in many areas of both pure mathematics and theoretical physics. In particular, its applications in quantum field theory (QFT) on a curved background are based on the fact that if \mathcal{M} is a physical spacetime, and \hat{F}[\bm{\Phi}|\nabla] is the wave operator for some field theory (depending on nontrivial background fields \bm{\Phi}=\Phi^{A} ), then the one-loop quantum effective action of this theory is given by \varGamma_{\text{1-loop}}[\bm{\Phi}]=\frac{1}{2}\Tr\ln\hat{F}=-\frac{1}{2}\int\limits_{0}^{\infty}\frac{d\tau}{\tau}\,\Tr e^{-\tau\hat{F}}, (1.4) where the functional trace \Tr is defined as \Tr f(\hat{F})=\int\limits_{\mathcal{M}}d^{d}x\;\tr\hat{K}[f(F)|x,x],\quad\tr\hat{K}=K^{A}_{A}. (1.5) The efficiency of the heat kernel method is based on the fact that for a second order minimal operator \hat{F}(\nabla)=-\hat{1}\Box+\hat{P}, (1.6) where \Box=-g^{ab}\nabla_{a}\nabla_{b} is the covariant Laplacian 2 2 2 We work in the Euclidean version of the theory related to the physical Lorentzian signature spacetime by Wick rotation. and \hat{P}(x) is a matrix depending on the point x (a potential term), there is a remarkable DeWitt’s expansion for the heat kernel \hat{K}_{F}(\tau|x,x^{\prime}) in the form of a functional series DeWitt 1965 3 3 3 A generalization of this expansion to the case of higher-order minimal operators \hat{F}(\nabla)=\hat{1}(-\Box)^{N}+\hat{P}(\nabla) was constructed in Barvinsky et al. 2019 ; Barvinsky and Wachowski 2022 ; Barvinsky et al. 2023a , and in the paper Barvinsky et al. 2025 it was shown how to obtain the corresponding expansion for causal (non-minimal) operators. , \hat{K}_{F}(\tau|x,x^{\prime})=\sum\limits_{k=0}^{\infty}B_{\tfrac{d}{2}-k}(\tau,\sigma)\cdot\hat{a}_{k}[F|x,x^{\prime}]. (1.7) Here the scalar function B_{\alpha}(\tau,\sigma)=\frac{\tau^{-\alpha}}{(4\pi)^{d/2}}\exp\left(-\frac{\sigma}{2\tau}\right) (1.8) depends neither on the bundle geometry nor on the potential term \hat{P} , and its dependence on the points x and x^{\prime} is only implicit, through its dependence on the Synge world function \sigma(x,x^{\prime}) (i.e., one half of the square of the geodesic distance between x and x^{\prime} ). Conversely, the off-diagonal heat kernel coefficients \hat{a}_{k}[F|x,x^{\prime}] (which we will also call HaMiDeW in honor of Hadamard, Minakshisundaram, and DeWitt Gibbons 1979 ) are matrix-valued two-point functions which do not depend on the proper time \tau , but contain information about the bundle geometry and the potential term \hat{P} . 4 4 4 It is always assumed that x and x^{\prime} are sufficiently close to each other so that all quantities under consideration are well-defined. In our notations the Pauli–van Vleck–Morette determinant \Delta^{1/2}(x,x^{\prime}) is absorbed into coefficients \hat{a}_{k}[F|x,x^{\prime}] of the expansion ( 1.7 ). The coincidence limits of the HaMiDeW coefficients and their covariant derivatives can be found from the chain of recurrent equations for \hat{a}_{k}[F|x,x^{\prime}] as some combinations of the background field curvatures \mathfrak{R}\in\{R_{a}{}_{bcd},\hat{\mathcal{R}}_{ab},\hat{P}\} and their derivatives DeWitt 1965 ; Barvinsky and Vilkovisky 1985 , which can be symbolically represented as \nabla^{n}\hat{a}_{k}[F|x,x^{\prime}]\,\Big|_{\,x=x^{\prime}}=\sum\limits_{2m+l=2k+n}\nabla^{l}\mathfrak{R}^{m}. (1.9) Then substituting ( 1.7 ) into the formula ( 1.4 ) yields a local expansion of the one-loop effective action \varGamma_{\text{1-loop}}[\bm{\Phi}] in a series in increasing powers of the background dimension, which underlies effective field theory. In many applications there is a need to obtain an off-diagonal (i.e., when x\neq x^{\prime} ) expansion similar to the DeWitt one ( 1.7 ) not only for the operator exponential \exp(-\tau\hat{F}) , but also of more complicated operator functions f(\hat{F}) . The physical application that motivates us most is the need to study models with a non-minimal wave operator \hat{H}(\nabla) . Non-minimality of the operator means that its principal term can no longer be represented as a power of the d’Alembertian \Box . Such models, in particular the Hořava–Lifshitz type, are getting more and more frequently encountered in high energy physics and quantum gravity Barvinsky et al. 2017a ; Barvinsky et al. 2017b ; Barvinsky et al. 2022 ; Barvinsky et al. 2023b . They are either too complicated to be handled by the usual Feynman diagrammatic technique or impossible to treat by the original Schwinger–DeWitt method. An important special case are causal models, i.e., those in which the characteristic surfaces of physical modes (in the Lorentzian version of the theory) are null surfaces in the metric g_{ab} . As shown in Barvinsky et al. 2025 ; Barvinsky and Kalugin 2024 , calculating the heat kernel \exp(-\tau\hat{H}) and the effective action ( 1.4 ) for such models can be reduced to finding expansions for operator functions of a more complex form f(\hat{F})=\frac{e^{\textstyle-\tau\hat{F}^{\nu}}}{\vphantom{L^{L}}\hat{F}^{\mu}}, (1.10) where the operator \hat{F}(\nabla) ( 1.6 ) is the second-order minimal, and the powers \mu and \nu are determined by the order of the operator \hat{H}(\nabla) and the spin-tensor content of the model. Therefore, the key problem that we solve further in the proposed paper is the construction of expansions similar to DeWitt’s ( 1.7 ) for kernels of operator functions of the form (