Double virtual QCD corrections to $t\bar{t}+$jet production at the LHC

This paper presents the first leading-colour computation of the double virtual contributions to top-quark pair production with a jet at next-to-next-to-leading order in QCD, providing analytically extracted finite remainders via a differential equation-based special function basis and a publicly available C++ library for phenomenological applications.

The Gist

Imagine you are trying to predict exactly how a billiard table will look after a very complex shot. In the world of particle physics, the "billiard table" is the Large Hadron Collider (LHC), and the "balls" are top quarks (the heaviest known elementary particles) and jets of other particles.

Scientists want to know the precise probability of a top quark and an anti-top quark being created alongside a jet of particles. To do this, they need to perform incredibly complex mathematical calculations. This paper is about finishing a specific, very difficult layer of those calculations.

Here is a breakdown of what the authors did, using simple analogies:

1. The Goal: Predicting the "Double Virtual" Effect

Think of the collision between particles as a dance.

• The Basic Dance (Tree Level): This is the simplest version of the dance, where particles just bump into each other and bounce off.
• The One-Step Dance (One Loop): Sometimes, during the dance, a particle briefly turns into a different particle and then back again before finishing the move. This is a "loop."
• The Two-Step Dance (Two Loops): This paper focuses on the most complex version: the "double virtual" contribution. Imagine the particles briefly turning into a cloud of other particles, which then turn into another cloud, and then resolve back into the original particles.

Calculating this "double loop" is like trying to predict the outcome of a dance where the dancers are constantly splitting into ghosts and recombining twice before the music stops. It is mathematically messy and prone to errors. The authors have successfully calculated this specific "double virtual" layer for top quarks and jets.

2. The Problem: The "Elliptic" Monster

In previous years, scientists developed a toolkit to solve these dance calculations for simpler particles (like massless gluons). They used a set of standard "special functions" (like a universal dictionary of math words) to describe the results.

However, top quarks are heavy. When you introduce this heavy mass into the "double loop" calculation, the standard dictionary breaks down. The math starts involving "elliptic curves"—a type of geometry that is much more complex than the simple shapes the old toolkit could handle. It's like trying to use a map of a flat city to navigate a mountain range; the old tools just don't fit.

3. The Solution: Building a New Toolkit

The authors couldn't use the old "canonical" method because of these elliptic curves. So, they built a new, custom toolkit:

• The "Overcomplete" Dictionary: Instead of trying to force the math into a perfect, minimal dictionary, they created a slightly larger, "overcomplete" set of special functions. Think of it as having a few extra synonyms in your dictionary. It's not the most efficient way to write a sentence, but it allows you to describe the complex elliptic shapes without getting stuck.
• The Differential Equation Engine: These special functions are defined by "differential equations" (rules that describe how the function changes as the energy of the collision changes). The authors built a system to solve these rules.
• The "Square Root" Trap: A major hurdle was that these equations contained "square roots" that could flip signs or jump between different values (like a switch that flips on and off unpredictably). The authors wrote a new computer algorithm that acts like a careful guide, ensuring the math stays on the correct path and doesn't get lost in the "branches" of the square roots.

4. The Result: A Ready-to-Use Library

Once they solved the math, they didn't just leave it on a piece of paper. They turned their results into a C++ software library.

• Imagine they built a high-precision calculator that anyone at a university or research lab can plug into their own simulations.
• This library allows scientists to instantly calculate the "hard function" (the core probability) for top quark production with jets, including all the complex "double virtual" effects they just solved.

5. Why It Matters (According to the Paper)

The paper states that experimental data from the LHC is becoming incredibly precise. To match this precision, theoretical predictions must also be extremely accurate (specifically at "Next-to-Next-to-Leading Order").

• Without this calculation, our theoretical predictions would be like a blurry photo.
• With this calculation, the photo becomes sharp.
• This allows scientists to compare the theory directly with the real data to test the Standard Model of physics and potentially measure the mass of the top quark more accurately.

Summary

The authors successfully solved a notoriously difficult math problem involving heavy particles and complex geometry. They created a new method to handle the "elliptic" shapes that usually break these calculations, built a robust computer program to solve the equations, and released a free tool so other scientists can use these results to make sharper, more accurate predictions about how the universe works at the smallest scales.

Technical Summary

Technical Summary: Double Virtual QCD Corrections to $t\bar{t}+\text{jet}$ Production at the LHC

Problem Statement
This work addresses the computation of the double virtual QCD corrections (two-loop amplitudes) for top-quark pair production in association with a jet ($t\bar{t}+\text{jet}$) at hadron colliders. While $t\bar{t}$ production is well-studied at Next-to-Next-to-Leading Order (NNLO), the inclusion of an associated jet is phenomenologically critical, as approximately 50% of $t\bar{t}$ events contain a jet. Previous calculations for this process were limited to Next-to-Leading Order (NLO). Achieving NNLO precision requires the evaluation of two-loop amplitudes, which presents significant challenges due to the presence of massive internal propagators (top quarks) and the resulting emergence of elliptic structures in the Feynman integrals. These structures prevent the use of standard canonical differential equation (DE) systems based on polylogarithms, creating a bottleneck for the analytic and numerical evaluation of the finite remainders.

Methodology
The authors employ a hybrid approach combining modern algebraic and numerical techniques to overcome the complexity of massive two-loop five-point amplitudes:

1. Partial Amplitude Decomposition: The calculation is performed in the leading-colour approximation. The authors decompose the scattering amplitudes for the partonic processes $0 \to \bar{t}tggg$ and $0 \to \bar{t}t\bar{q}qg$ into partial amplitudes. They define finite remainders by subtracting universal ultraviolet (UV) and infrared (IR) singularities from the renormalised amplitudes.
2. Helicity Amplitude Construction: Using the four-dimensional projector method, the amplitudes are decomposed into massive spinor form factors. The contracted helicity amplitudes are computed by fixing the helicities of massless partons and treating the massive spinor dependence via form factors.
3. Finite-Field Reconstruction: The rational coefficients of the amplitudes are reconstructed analytically from numerical evaluations over finite fields. The authors utilize momentum-twistor parametrisation to express these coefficients as rational functions, avoiding large intermediate expressions typical of traditional symbolic approaches.
4. Special Function Basis: To handle the elliptic and nested square-root structures arising from the massive top quark, the authors construct a (potentially) overcomplete basis of special functions. Unlike canonical DE approaches, this basis is defined by a system of differential equations containing only algebraic functions. This allows the isolation of elliptic contributions to the finite remainder while enabling analytic subtraction of poles.
5. Numerical Evaluation Strategy: A new strategy for the numerical integration of the differential equations for the special functions is presented. This involves:
   • Solving a linear system of first-order DEs for a "DE basis" of 1282 functions (including polynomials of the generating set).
   • Using the Bulirsch-Stoer algorithm with complex deformation of the integration path to avoid singularities and ensure numerical stability.
   • Implementing a specific algorithm to track branch cuts of square roots along the integration curve.
   • Optimizing the evaluation of the connection matrix through variable permutations and partial fraction decomposition to reduce floating-point operations.

Key Contributions

• Analytic Reconstruction: The paper provides the first analytic reconstruction of the leading-colour two-loop helicity amplitudes for $t\bar{t}+\text{jet}$ production in all relevant channels.
• Special Function Basis: The authors extend the special function basis introduced in previous work (ref. [85]) to cover all 12 permutations of external momenta required for the cross-section calculation, resolving redundancies and ensuring the basis is closed under permutations.
• Numerical Implementation: A C++ library is developed and released, capable of evaluating the squared matrix elements (hard functions) for phenomenological studies. This library includes the numerical solver for the special functions and the assembly of the rational coefficients.
• Benchmark Results: The authors provide benchmark numerical values for the hard functions at a specific physical phase-space point for all relevant scattering channels ($gg \to t\bar{t}g$, $q\bar{q} \to t\bar{t}g$, etc.).

Results

• The finite remainders are expressed in terms of monomials of special functions and transcendental constants, with rational coefficients reconstructed from finite-field data.
• The numerical evaluation strategy demonstrates high stability and efficiency. Benchmarks indicate that evaluating the hard function for a single phase-space point takes approximately 10–35 seconds depending on the channel, with the majority of time spent evaluating rational coefficients.
• The implementation successfully reproduces previous benchmarks for the gluon channel and passes cross-checks against known one-loop results and dimensional analysis requirements.
• The code requires high-precision arithmetic (quadruple precision via dd_real or qd_real) for the rational coefficients to ensure numerical stability, particularly in regions where large cancellations occur.

Significance
The paper claims to provide the final ingredient necessary for producing differential cross-section predictions for $t\bar{t}+\text{jet}$ production at NNLO in QCD within the leading-colour approximation. This level of precision is required to match the decreasing experimental uncertainties at the LHC. The work demonstrates that processes involving massive fermions and elliptic structures can be computed phenomenologically by constructing a non-canonical but numerically tractable basis of special functions. The provided C++ library enables immediate application in Monte Carlo simulations for precision top-quark physics, including the extraction of the top-quark mass and tests of the Standard Model. The authors note that while the leading-colour approximation is a limitation compared to full-colour results available for other NNLO processes, the techniques developed here pave the way for future extensions to full colour and higher multiplicities.