What is the $i\varepsilon$ for the S-matrix?

Holmfridur S. Hannesdottir, Sebastian MizeraarXiv 2022 · DOI · arXiv

TL;DR

This paper tackles the long-standing problem of how to mathematically extend the S-matrix (which describes particle scattering) into the complex plane for realistic theories like the Standard Model. The authors propose a new "branch cut deformation" prescription that ensures causality even when traditional methods fail due to massless or unstable particles, creating a "causal strip" around physical reality.

What Problem Does This Paper Address?

In particle physics, the S-matrix tells us the probability of various scattering outcomes. To unlock its full power—using tools like dispersion relations (relating the real and imaginary parts of an amplitude) or crossing symmetry (relating a process like $e^+e^- \to \gamma\gamma$ to its "crossed" version $\gamma e^- \to \gamma e^-$)—physicists must treat the S-matrix as an analytic function of complex variables, specifically the Mandelstam invariants ($s, t, u$).

Since the 1960s, we’ve known how to do this for simple cases: $2 \to 2$ scattering of stable, massive particles. In those cases, we use a "prescription" called $i\varepsilon$—adding a tiny imaginary "nudge" to the energy—to avoid mathematical singularities and ensure the physics remains causal (causes precede effects).

However, the Standard Model is messy. It contains massless particles (photons) and unstable ones (W and Z bosons). In these realistic scenarios, the traditional $i\varepsilon$ prescription can break down because different parts of a calculation (different Feynman diagrams) might require contradictory "nudges." This paper seeks a consistent, algebraic way to define the complex S-matrix for these difficult cases.

Methods

The authors move away from the traditional loop-momentum space and instead utilize the worldline formalism and Schwinger parameters.

  • Schwinger Parametrization: This technique represents Feynman integrals as integrals over "proper time" variables ($\alpha_e$) for each internal line in a diagram. This transforms the problem into an algebraic one governed by an "on-shell action" $\mathcal{V}$.
  • Branch Cut Deformations: Instead of just adding $i\varepsilon$ to external variables or masses, the authors propose deforming the integration contours of the Schwinger parameters themselves. This has the effect of "bending" branch cuts in the kinematic space away from the physical region.
  • Lefschetz Thimbles: They use this sophisticated mathematical framework to identify which classical "saddle points" (configurations where particles propagate classically) actually contribute to the final result on the "physical sheet."
  • Holomorphic Cutting Rules: They introduce a new version of the Cutkosky rules that doesn't rely on complex conjugation, making the analytic structure of the scattering amplitude much easier to track.

Datasets

As a theoretical physics paper, it does not use traditional training datasets. Instead, it evaluates its proposed prescription on several "case studies" of fundamental Feynman diagrams:

  • The Bubble Diagram: Used as a primer to illustrate how $i\varepsilon$ affects branch points.
  • The $su$-Box Diagram: A complex case with overlapping normal thresholds.
  • The $u$-Triangle Diagram: Used to show how unstable external particles can effectively "disconnect" the upper and lower halves of the complex plane.

Key Insights and Ideas

  • Singularities as Saddle Points: The authors show that all kinematic singularities (thresholds) can be understood as classical saddle points of the worldline action. If you can't "deform" your way around a point where the action vanishes, you've hit a branch point.
  • The Causal Strip: For realistic theories, we shouldn't expect the S-matrix to be analytic everywhere. Instead, the authors demonstrate that their prescription reveals a "complex strip" around the real axis where the S-matrix remains causal and well-defined.
  • Unstable Particles Disconnect the Plane: A major "aha" moment is the realization that when external particles are unstable, the upper-half and lower-half complex planes of the energy variable $s$ can become entirely disconnected by branch cuts, meaning they are represented by two different analytic functions.
  • Bound on Singularity Strength: Using analyticity arguments, they prove that in local quantum field theories, singularities can't be "worse" than simple poles, square roots, or logarithms.

Key Results

  • Validation of $s + i\varepsilon$: They provide a new proof that for $2 \to 2$ scattering of stable particles, the standard $s + i\varepsilon$ approach is always valid, regardless of mass or momentum transfer—closing a gap in 1960s-era proofs.
  • Generalized Dispersion Relations: They derive new dispersion relations that work even when the "Euclidean region" (a safe mathematical zone) does not exist, which is common in massless theories.
  • Numerical Convergence: They show that their contour deformation method is numerically much more stable than the traditional Feynman $i\varepsilon$ approach, which often suffers from precision errors near branch cuts.

Limitations and Future Work

  • Isolated Saddles Only: The current work focuses on "isolated and non-degenerate" saddle points. In theories with massless particles, one often encounters "continuous manifolds" of saddles, which require even more complex localization techniques.
  • Higher Multiplicity: While the paper touches on higher-point amplitudes (more than 4 particles), a full systematic treatment for $n$-point scattering in the Standard Model is still a work in progress.
  • Non-Perturbative Extension: The results are derived within perturbation theory (Feynman diagrams). Extending these insights to a fully non-perturbative S-matrix remains the "holy grail" of this sub-field.

Summary updated

Abstract

Can the S-matrix be complexified in a way consistent with causality? Since the 1960's, the affirmative answer to this question has been well-understood for $2 \to 2$ scattering of the lightest particle in theories with a mass gap at low momentum transfer, where the S-matrix is analytic everywhere except at normal-threshold branch cuts. We ask whether an analogous picture extends to realistic theories, such as the Standard Model, that include massless fields, UV/IR divergences, and unstable particles. Especially in the presence of light states running in the loops, the traditional $i\varepsilon$ prescription for approaching physical regions might break down, because causality requirements for the individual Feynman diagrams can be mutually incompatible. We demonstrate that such analyticity problems are not in contradiction with unitarity. Instead, they should be thought of as finite-width effects that disappear in the idealized $2\to 2$ scattering amplitudes with no unstable particles, but might persist at higher multiplicity. To fix these issues, we propose an $i\varepsilon$-like prescription for deforming branch cuts in the space of Mandelstam invariants without modifying the analytic properties. This procedure results in a complex strip around the real part of the kinematic space, where the S-matrix remains causal. In addition to giving a pedagogical introduction to the analytic properties of the perturbative S-matrix from a modern point of view, we illustrate all the points on explicit examples, both symbolically and numerically. To help with the investigation of related questions, we introduce a number of tools, including holomorphic cutting rules, new approaches to dispersion relations, as well as formulae for local behavior of Feynman integrals near branch points.
BibTeX
@misc{hannesdottir2022what,
  author = {Holmfridur S. Hannesdottir and Sebastian Mizera},
  title = {{What is the $i\textbackslash\{\}varepsilon$ for the S-matrix?}},
  year = {2022},
  eprint = {2204.02988},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2204.02988},
}

Loading related content...

Figure