Notes on color reductions and γ \gamma traces
Oliver Schnetz · 2025
Read Notes on color reductions and γ \gamma traces by Oliver Schnetz (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-2504-05853.txt.
Introduction
Particle physics calculations drown in bookkeeping: color charges must be summed and gamma matrices traced, loop after loop. Oliver Schnetz writes the field a manual in 2025 notes, presenting tuned algorithms for both chores in the standard gauge theory of the strong force. The emphasis falls on color reductions, the simplifications that collapse huge sums into compact factors.
The account is self-contained and proved, aimed at readers who need the machinery rather than folklore. Fine-tuning of known methods pushes the reachable loop order higher, which matters because precision predictions for collider experiments live at those orders. The notes collect scattered tricks into one verified toolkit.
Catalog record
| CREATOR | Oliver Schnetz |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 34 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-05853
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Notes on color reductions and γ \gamma traces Oliver Schnetz Abstract . We present efficient algorithms to calculate the color factors for the S U ( N ) SU(N) gauge group and to evaluate γ \gamma traces. The aim of these notes is to give a self-contained, proved account of the basic results with particular emphasis on color reductions. We fine tune existing algorithms to make calculations at high loop orders possible. 1. Introduction Yang-Mills quantum field theories (QFTs) feature combinatorial factors in the Feynman integrals which come from the SU(N) color gauge group. The calculation of these color factors is explained in [ 4 ] (see also [ 1 , 5 ] ). In QFTs with fermions, a first step in the calculation of a Feynman integral often is the evaluation of the traces over the \gamma matrices which originate from fermion edges and from vertices. The evaluation of these \gamma traces is explained in [ 6 , 14 ] (and in any textbook on QFT, see e.g. [ 8 ] ). Today, one typically calculates to loop orders \leq 5 , where any implementation of these reductions is sufficient, see e.g. [ 9 , 7 ] . With the method of graphical functions [ 10 , 3 , 11 , 12 , 2 ] it may become possible to tackle higher loop orders in certain setups. At loop orders \geq 6 it becomes increasingly desirable to fine tune the algorithms for color and \gamma reductions. In these notes, we collect identities that are necessary to perform the reductions to high loop orders. We include the proofs of all relevant results. Particular emphasis is on color reductions, where the proof of the essential identity (Proposition 3 ) is not included in [ 4 ] (see [ 5 ] for the proof). We also prove some additional results about color reductions. The suggested algorithms are implemented in the Maple package HyperlogProcedures . Typical color reductions are more or less instant at relevant loop orders. The reduction of \gamma traces is slightly more time-consuming, mostly because it can produce lengthy results at high loop orders. For example, using HyperlogProcedures on a single core of an office PC, the average time for a \gamma reduction of a Feynman graph that contributes to the photon propagator is approximately 2 minutes at six loops and 30 minutes at seven loops. Acknowlegements The author is supported by the DFG-grant SCHN 1240/3-1. He thanks Sven-Olaf Moch for discussions and encouragement. The author also thanks Simon Theil for generating the figures in this work. 2. Color reduction 2.1. The color graph We follow the algorithm presented in [ 4 ] , which is particularly simple in the case of the group SU( N ). The complex Lie-algebra of SU( N ) is \mathrm{sl}(N) , the Lie-algebra of traceless matrices. For any representation of \mathrm{sl}(N) , the commutator of the basis T^{i}=(T^{i})_{ab} is 1 1 1 In [ 4 ] f_{ijk}=\mathrm{i}C_{ijk} . We do not see the necessity to pass to complex numbers. Euclidean QED and Yang-Mills theory can be formulated in a real setup. (1) [T^{i},T^{j}]=f_{ijk}T^{k}=\sum_{k=1}^{N^{2}-1}f_{ijk}T^{k}, where we sum over repeated indices (Einstein’s sum convention). The structure constants f_{ijk}=(f_{i})_{kj} define an adjoint representation of \mathrm{sl}(N) . By definition, the trace of T^{i} vanishes, (2) T^{i}_{aa}\equiv\sum_{a=1}^{N}T^{i}_{aa}=0. In some sense, we consider color factors as objects in a combinatorial differential geometry with Euclidean metric. The matrices T^{i} are chosen orthogonal and normalized, (3) \mathrm{Tr}\,T^{i}T^{j}=\delta_{i,j}. We use orthogonality to express the structure constants f_{ijk} in terms of the matrices T^{i} , (4) f_{ijk}=\mathrm{Tr}\,[T^{i},T^{j}]T^{k}. The cyclicity of the trace implies that the f_{ijk} are fully anti-symmetric in their indices, (5) f_{ijk}=f_{jki}=f_{kij}=-f_{ikj}=-f_{kji}=-f_{jik}. The first two identities of ( 5 ) allow us to use a graphical representation. a b i j k i i j a b T^{i}_{ab} f_{ijk} \delta_{i,j} \delta_{a,b} Figure 1. Feynman rules for color graphs To T^{i}_{ab} we associate a corolla (a vertex with three half-edges). The half-edge i has no orientation, while the half-edges a and b are ingoing and outgoing, respectively (see Figure 1 ). To f_{ijk} we associate a corolla of three non-oriented half-edges, where we fix a planar representation with counter-clockwise i,j,k . The sum over double indices glues the corresponding half-edges. Note that matrix products in the T^{i} preserve the orientation. We obtain a graph with fixed planar embedding: a ribbon graph. By anti-symmetry, flipping two edges in the vertex f_{ijk} gives a minus sign, see Figure 2 . So, the planar embedding determines the sign of the color factor. j k i j k i =\;- Figure 2. Flipping two edges at an adjoint vertex gives a minus sign. The graphical representations of ( 1 ) to ( 5 ) are in Figure 3 , where we swapped the sides of ( 1 ). To simplify the graphical notation it is customary to drop the labels of external half-edges using the convention that half-edges located at the same position have equal labels. a b k j i = a b j i - a b j i i =\;0 i =\;0 i j i j = j k i i j k i j k = - Figure 3. Basic identities for color graphs. Lemma 1 . An orthonormal basis of the fundamental (defining) representation of sl(N) are the N\times N matrices \displaystyle T^{\alpha\beta} \displaystyle=\frac{1}{\sqrt{2}}(E_{\alpha\beta}+E_{\beta\alpha}),\qquad\tilde{T}^{\alpha\beta}=\frac{\mathrm{i}}{\sqrt{2}}(E_{\alpha\beta}-E_{\beta\alpha})\quad\text{for }1\leq\alpha<\beta\leq N, (6) \displaystyle T^{k} \displaystyle=\frac{1}{\sqrt{k(k+1)}}\bigg(\Big(\sum_{r=1}^{k}E_{rr}\Big)-kE_{(k+1)(k+1)}\bigg)\quad\text{for }k=1,\ldots,N-1, where (E_{ab})_{cd}=\delta_{a,c}\delta_{b,d} are the elementary matrices. We fix any sequence of \alpha\beta in T^{\alpha\beta} and \tilde{T}^{\alpha\beta} to continue the labels 1,\ldots,N-1 of T^{k} to N-1+2N(N-1)/2=N^{2}-1 . Proof. The orhonormality of T^{\alpha\beta} and \tilde{T}^{\alpha\beta} is clear. Because for \alpha\neq\beta and any \gamma we have (no sum) \mathrm{Tr}\,E_{\gamma\gamma}E_{\alpha\beta}=\delta_{\alpha,\gamma}\mathrm{Tr}\,E_{\alpha\beta}=0 , the T^{\alpha\beta} and \tilde{T}^{\alpha\beta} are orthogonal to the T^{k} . From (T^{k})^{2}=\frac{1}{k(k+1)}\bigg(\Big(\sum_{r=1}^{k}E_{rr}\Big)+k^{2}E_{(k+1)(k+1)}\bigg) we obtain \mathrm{Tr}\,(T^{k})^{2}=1 . ∎ Definition 2 . Let G be a ribbon graph with oriented and non-oriented edges. We assume that G has two types of vertices: a (fundamental) vertex with two oriented edges and one non-oriented edge and an (adjoint) vertex with three non-oriented edges. We write e_{G} , v_{G} , h_{G} , c_{G} for the number of edges, the number of vertices, the number of independent cycles (loops), and the number of components of G , respectively. The empty (self-)loop \circ (top graphs in Figure 5 ) has e_{\circ}=v_{\circ}=0 and h_{\circ}=c_{\circ}=1 . The graph G can have external (non-paired) half-edges (hairs). Equivalently, we connect external vertices to the external half-edges and obtain a decomposition of the vertices into external (one-valent) and internal (three-valent) vertices, v_{G}=v_{G}^{\mathrm{ext}}+v_{G}^{\mathrm{int}} . After the removal of the non-oriented edges, G decomposes into a collection of f_{G} oriented cycles. The reduction of the color graph G is R_{G}(N) . From graph homology we get (7) h_{G}-e_{G}+v_{G}-c_{G}=0 and from counting half-edges we obtain (8) v_{G}^{\mathrm{ext}}=2e_{G}-3v_{G}^{\mathrm{int}}. 2.2. Fundamental identities \frac{1}{N} = - Figure 4. The two-term relation of SU(N) color reductions, Equation ( 9 ). The core identity for color reductions is depicted in Figure 4 . This identity is specific to the Lie algebras \mathrm{sl}(N) . Similar reduction formulae for other Lie algebras are in [ 4 ] . Proposition 3 . In any orthonormal basis T^{i} , we obtain for the sum \sum_{i}T^{i}\otimes T^{i}