NNLO QCD predictions for $t\bar t W$ production at hadron colliders

This paper presents the first NNLO QCD predictions for $t\bar t W$ production at hadron colliders based on a direct computation of the required two-loop amplitudes in the generalised leading-colour limit, addressing previous reliance on dynamical approximations for this complex process.

The Gist

Imagine the Large Hadron Collider (LHC) as the world's most powerful particle smasher. Inside, it smashes protons together to create a chaotic storm of new particles. Among the most interesting "debris" from these crashes is a specific trio: a top quark, an anti-top quark, and a W boson. This is a heavy, rare, and complicated event.

For a long time, scientists have measured how often this trio appears. The problem? The real-world measurements keep showing up more often than our best theoretical recipes predicted. It's like a chef following a recipe perfectly, but the cake keeps rising higher than the instructions say it should. To fix this, scientists need to upgrade their recipe from a "good guess" to a "perfectly precise calculation."

The Challenge: A Mathematical Mountain

Calculating how these particles interact is like trying to predict the exact path of every single raindrop in a hurricane. The math gets incredibly messy, especially when you try to account for the invisible "glue" (called QCD) that holds the particles together.

To get a truly accurate prediction, scientists need to calculate effects that happen at the "Next-to-Next-to-Leading Order" (NNLO). Think of this as calculating the recipe not just for the main ingredients, but also for the tiny, invisible interactions between them. The hardest part of this calculation involves a "two-loop" diagram. If a standard calculation is like drawing a simple line, a two-loop calculation is like trying to draw a knot that twists through itself in four dimensions.

For years, scientists had to use "shortcuts" (approximations) to solve this knot. They assumed the W boson was very light or the top quarks were very heavy to make the math manageable. While these shortcuts were good enough to get a rough idea, they left a tiny bit of uncertainty, like measuring a room with a tape measure that has a slightly stretched rubber band.

The Breakthrough: A New Way to Tie the Knot

This paper announces a major breakthrough. The team has finally solved the "knot" exactly, without relying on those heavy-handed shortcuts.

Instead of guessing the shape of the knot, they used a powerful new method called the "Generalised Leading-Colour Limit."

• The Analogy: Imagine the particles are wearing colored shirts (Red, Green, Blue). In the real world, they interact in all possible color combinations, which is a chaotic mess of math. The "Leading-Colour" limit is like saying, "Let's assume the Red shirts are the most popular and dominate the party, while the other colors are just background noise."
• Why it works: This isn't a wild guess; it's a controlled mathematical simplification. It strips away the most confusing parts of the math while keeping the most important physics intact. It's like listening to the lead singer of a band to understand the song, rather than trying to hear every single instrument perfectly at once.

The Result: A Clearer Picture

By using this new method, the team calculated the production rate of the top-anti-top-W trio with unprecedented precision.

1. The Numbers: Their new, more precise calculation predicts that this trio should appear slightly more often than the previous "shortcut" calculations suggested. Specifically, the new prediction is about 3% higher than the previous best estimate.
2. The Comparison: When they compared their new "exact" (within the color limit) result to the old "shortcut" results, they found they agreed very well. The old shortcuts were actually doing a decent job, but the new method confirms the numbers with much higher confidence.
3. The Uncertainty: The team estimates that their new method is accurate to within about 2.5%. This is a tiny margin of error, far better than the previous estimates.

Why This Matters

This isn't just about fixing a number on a chart.

• The Background: This specific particle trio is a "background noise" for many other experiments. If you are trying to find a new, rare particle (like a new type of Higgs boson), you have to know exactly how much "noise" the top-anti-top-W trio makes so you can subtract it. If your noise estimate is off, you might think you found a new particle when you didn't, or miss a real discovery.
• The Method: The biggest achievement here is the method. The team proved that they can solve these incredibly complex, multi-layered math problems using this new "color-focused" approach. It's like proving a new type of drill can bore through the hardest rock. This paves the way for solving other impossible-looking physics problems in the future.

In short, the scientists have taken a messy, complicated math problem, applied a clever new lens to simplify it, and produced a much sharper, more reliable prediction for how often nature creates these heavy particle trios. This helps ensure that when we look for new physics at the LHC, we aren't being fooled by a blurry picture.

Technical Summary

Technical Summary: NNLO QCD predictions for $t\bar{t}W$ production at hadron colliders

Problem Statement
The associated production of a top–antitop quark pair with a W boson ($t\bar{t}W$) is a critical process at the Large Hadron Collider (LHC), serving as a major background for searches involving same-sign lepton pairs (e.g., $t\bar{t}t\bar{t}$, $t\bar{t}H$) and a probe for physics beyond the Standard Model. Experimental measurements of the $t\bar{t}W$ production rate have consistently exceeded Standard Model predictions. While Next-to-Leading Order (NLO) QCD and Electroweak (EW) corrections are available, achieving the precision required to resolve these discrepancies necessitates Next-to-Next-to-Leading Order (NNLO) QCD calculations.

Previous NNLO predictions for the inclusive $t\bar{t}W$ cross section (Ref. [32]) relied on dynamical approximations for the two-loop virtual amplitudes: treating the W boson as soft (SA) or the top-quark mass as small compared to other scales (MA). While these approximations were deemed subleading compared to residual perturbative uncertainties, an exact computation of the two-loop contribution is ultimately required to eliminate systematic uncertainties and confirm previous results. The calculation of the two-loop amplitude for this $2 \to 3$ process is exceptionally demanding due to complex kinematics, multiple mass scales (top quark and W boson), and the resulting extreme algebraic complexity.

Methodology
This work presents the first NNLO QCD prediction for $t\bar{t}W$ production based on a direct computation of the required two-loop amplitudes, evaluated in the generalised leading-colour (LC) limit. The calculation utilizes the Matrix framework, extended for $t\bar{t}W$ production, employing the $q_T$-subtraction formalism for infrared singularities and dipole subtraction for real-emission contributions.

The core technical achievement lies in the computation of the two-loop virtual amplitude $M^{(2)}$:

1. Leading-Colour Approximation: The amplitude is computed keeping only terms proportional to $N_c^2$, $N_c n_l$, and $n_l^2$, where $N_c=3$ and $n_l=5$. This is a parametric approximation controlled by a fixed expansion parameter, distinct from the dynamical approximations used previously.
2. Amplitude Decomposition: The bare amplitude is decomposed into 24 independent tensor structures using physical projectors in the 't Hooft-Veltman scheme.
3. Integral Reduction and Evaluation: The projections are expressed as linear combinations of scalar Feynman integrals. To handle the complex analytic structures (including elliptic curves and nested square roots), the authors expand the integral bases around $\epsilon = 0$ and express coefficients as polynomials in special functions defined by systems of algebraic partial differential equations (DEs). These functions are evaluated numerically using AMFlow.
4. Numerical Reconstruction: To manage the algebraic complexity of the rational coefficients, the authors perform operations over finite fields using FiniteFlow. They reconstruct the rational numbers point-by-point using the Chinese remainder theorem, requiring up to 400 primes. This approach avoids numerical errors in coefficient evaluation while maintaining performance.
5. Validation: The results are validated against an independent numerical calculation using conventional dimensional regularisation and the Blade package for integration-by-parts (IBP) reduction. Agreement within five significant digits is confirmed at representative phase-space points.
6. Phenomenological Integration: The two-loop finite remainder is evaluated on a grid of ~224,000 phase-space points and interpolated using quadratic B-splines to compute the hard-virtual coefficient $H^{(2)}$.

Key Contributions

• First Direct Two-Loop Computation: This work provides the first NNLO QCD prediction for $t\bar{t}W$ based on a direct evaluation of the two-loop amplitude, rather than dynamical approximations.
• Methodological Milestone: It represents the first application of a numerical approach based on the evaluation of special functions and finite-field reconstruction for rational coefficients in a complete phenomenological calculation of a $2 \to 3$ process with multiple mass scales.
• Uncertainty Quantification: The paper provides a rigorous estimate of uncertainties associated with the leading-colour approximation, distinct from the dynamical approximations of previous work.

Results
The authors present NNLO QCD predictions for the inclusive $t\bar{t}W$ cross section at $\sqrt{s} = 13$ TeV.

• Comparison with Previous Work: The new results, based on the Leading-Colour Approximation (LCA), are consistent with the previous SA+MA predictions (Ref. [32]) within estimated uncertainties. The new LCA prediction for the inclusive cross section is approximately 3% higher than the SA+MA result but remains compatible.
• Impact of Two-Loop Contribution: The LC two-loop virtual contribution is found to be substantial, accounting for about 11% of the total NNLO cross section.
• Kinematic Dependence: The LCA consistently overestimates the exact one-loop result by about 22% at the inclusive level, decreasing to roughly 10% in high transverse momentum regions ($p_{T,t/\bar{t}} > 1$ TeV). The uncertainty assigned to the NNLO LCA result is estimated at the 2.5% level, derived from the relative difference between exact and LCA results at NLO.
• Cross Section Values: For $t\bar{t}W^-$, the NNLO cross section is predicted to be $241.9^{+6.4\%}_{-7.3\%} \pm 2.3\%$ fb (LCA), compared to $235.4^{+5.1\%}_{-6.6\%} \pm 1.9\%$ fb (SA+MA).

Significance
The paper claims that this work constitutes a crucial step toward eliminating systematic uncertainties in $t\bar{t}W$ predictions. By validating the previous dynamical approximations (Ref. [32]) with a direct, parametrically controlled calculation, the authors confirm the robustness of the existing theoretical benchmarks used for experimental comparisons.

Methodologically, the work pushes multi-loop amplitude techniques to their current limits, demonstrating the feasibility of handling extreme algebraic complexity in processes with multiple mass scales. The authors state that this achievement paves the way for NNLO QCD predictions for other complex processes. They note that while the current calculation is limited to the leading-colour limit, the basic ingredients are now available to combine this with approximate treatments of subleading-colour contributions in the short term, with exact full-colour contributions remaining a target for future methodological advances.