A Feynman integral via higher normal functions

From MaRDI portal



Abstract: We study the Feynman integral for the three-banana graph defined as the scalar two-point self-energy at three-loop order. The Feynman integral is evaluated for all identical internal masses in two space-time dimensions. Two calculations are given for the Feynman integral; one based on an interpretation of the integral as an inhomogeneous solution of a classical Picard-Fuchs differential equation, and the other using arithmetic algebraic geometry, motivic cohomology, and Eisenstein series. Both methods use the rather special fact that the Feynman integral is a family of regulator periods associated to a family of K3 surfaces. We show that the integral is given by a sum of elliptic trilogarithms evaluated at sixth roots of unity. This elliptic trilogarithm value is related to the regulator of a class in the motivic cohomology of the K3 family. We prove a conjecture by David Broadhurst that at a special kinematical point the Feynman integral is given by a critical value of the Hasse-Weil L-function of the K3 surface. This result is shown to be a particular case of Deligne's conjectures relating values of L-functions inside the critical strip to periods.


The authors picked up the so-called three banana scalar Feynman integral in two space-time dimensions to convert it into a subject of purely mathematical investigation. I mention for the uninitiated that Feynman integrals appear in the perturbative expansion of scattering amplitudes in quantum filed theories. If we go beyond the zeroth order approximation in this expansion, we will encounter the famous (and often divergent) Feynman integrals. The paper under consideration discusses the mathematical aspects of one such integral. In the first section the authors relate the integral to the third order Picard-Fuchs differential equation (with the variable being the external four momentum squared \(t\) ) and discuss in detail the inhomogeneous solution of this equation. The latter is given in terms of Bellinson-Levin elliptic trilogarthmic functions. The section finishes with the evaluation of the Feynman integral at special arguments, \(t=0\) and \(t=1\). The denominator of the integrand is associated wit \(K3\) surfaces with a specific Picard number. After a brief introduction to the subject in section three, the authors use the surface theory to relate the integral to higher normal functions which gives an alternative proof of the results obtained via the Picard-Fuchs differential equation. Yet another method to express the integral by elliptic trilogarithms is presented in section 5. The subsequent section treats the Feynman integral via the Hodge theory. This section ends with a discussion of the integral values at \(t=0\) and \(t=1\) which is not a repetition, but offers a different point of view on this matter. The cycle closes here. This brief description here of the paper shows what the authors intend to do: they try and (succeed) to make many inter-connections between different mathematical theories with the banana Feynman integral as the binding element.NEWLINENEWLINEThe terminology of the paper is dense owing to the many inter-connections. One could also say that the pre-requisites are manifold. The reader should have heard about motivic cohomology, Dedekind eta function, Eisenstein series and symbols, L-Functions, the Milnor K-group, Higher Chow groups, Zariski closure, Deligne complex, Bellinson cycles, Gauss-Manin connection, Kuga variety and many other mathematical constructs and definitions. A physicist even acquainted with higher mathematics will wonder what has happened to the old Feynman integral, a mathematician might be delighted to see so many branches of mathematics united together around a physical object of interest.NEWLINENEWLINEThe rather big paper is self contained (provided the reader's diversity in mathematical education is vast enough) with proofs of the theorems and several discussions around them. The bibliography alone tells the unexpected: the banana Feynman integral has more ramifications than Feynman could imagine.



Cites work


Cited in
(only showing first 100 items - show all)








This page was built for publication: A Feynman integral via higher normal functions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2786464)