$e^+e^- \to ZH$ at NLO EW matched to a QED parton shower

This paper presents an automated, process-independent NLO-matched QED parton shower method that resolves challenges posed by lepton structure function singularities and provides the first EW NLO+PS predictions for $e^+e^- \to ZH$ production at proposed FCC-ee energies.

The Gist

Imagine you are trying to predict the outcome of a high-stakes billiard game, but instead of billiard balls, you are smashing electrons and positrons (particles of light and matter) together at nearly the speed of light. The goal is to create a Higgs boson and a Z boson (heavy particles that are the "stars" of this show).

For decades, physicists have been getting better and better at predicting these collisions. However, as our detectors become more precise, the old math isn't good enough anymore. It's like trying to measure the thickness of a human hair with a ruler meant for measuring football fields; you need a micrometer.

This paper introduces a new, ultra-precise "micrometer" for particle physics, specifically designed for the next generation of particle colliders (like the proposed FCC-ee). Here is the breakdown of what they did, using simple analogies.

1. The Problem: The "Sneeze" Effect

When you smash two electrons together, they don't just hit and stop. They are like two people wearing static-charged sweaters who, just before they touch, start screaming and flinging sparks (photons) everywhere.

In physics, this is called Initial-State Radiation. The electrons "sneeze" out energy before the collision even happens. This sneeze changes the energy of the collision. If you don't account for it perfectly, your prediction of what happens next (like creating a Higgs boson) will be wrong.

For a long time, physicists had two ways to handle this:

• The "Fixed" Method: A very precise calculation for the main event, but it ignores the sneezing.
• The "Shower" Method: A simulation that handles the sneezing well, but is a bit rough on the main event details.

The challenge was to combine them: get the precision of the fixed method and the realism of the shower method.

2. The Solution: A New "Traffic Cop"

The authors created a new system called MC@NLO matched to a QED Parton Shower.

Think of the collision as a busy highway intersection.

• The Fixed Calculation (NLO) is the traffic engineer who knows exactly how many cars should pass through based on the rules.
• The Parton Shower is the traffic cop on the ground, directing individual cars, handling sudden stops, and managing the flow of traffic (the sneezing photons).

Usually, these two don't get along. The engineer says "100 cars," and the cop says "Actually, 105 because of that red light." If you just add them together, you double-count the cars.

The authors built a new traffic cop (a QED dipole shower) that knows exactly how to talk to the engineer. They developed a special "handshake" protocol (the MC@NLO matching method) that ensures:

1. The engineer's precise count is respected.
2. The cop's realistic flow of traffic is added on top without double-counting.
3. The "sneezes" (photons) are handled correctly, even when they are very weak or very strong.

3. The Big Hurdle: The "Infinity" Singularity

Here is where it gets tricky. In the world of electrons, there is a mathematical "trap."

Imagine a speed limit sign that says "Speed Limit: 100 mph." But right at 100 mph, the sign says "Speed Limit: INFINITY." In math, this is called a singularity. When electrons get very close to their maximum energy, the math blows up.

In the world of protons (used in the Large Hadron Collider), this isn't a huge problem. But for electrons, this "infinity" is right in the middle of the action. The authors had to invent a new way to "smooth out" this infinity so the computer could calculate it without crashing. They used a technique called linear rescaling, which is like putting a speed bump right before the infinity sign so the cars slow down smoothly instead of flying off the road.

4. The Test Drive: The "Neutrino" and the "Higgs"

To prove their new system works, they did two things:

1. The Test Drive (Neutrinos): They simulated a simpler collision (making neutrinos) at different energies. They tweaked the "speed bumps" and "traffic lights" (mathematical parameters) to make sure the results were stable and didn't change wildly just because they turned a dial slightly. The system passed with flying colors.
2. The Real Race (Higgs Production): They applied their system to the main event: creating a Higgs boson at a future collider energy (240 GeV and 365 GeV).

The Results:

• At 240 GeV (just enough energy to make the Higgs), the "sneezing" isn't too wild. The new system confirmed that the standard calculations were mostly right, but added a tiny bit of extra precision.
• At 365 GeV (more energy), the electrons "sneeze" much harder. Here, the difference was huge. The old methods (just the shower or just the fixed math) would have been off by a lot. The new MC@NLO method showed exactly how the extra energy creates a cascade of photons that changes the shape of the final result.

5. Why Does This Matter?

We are building the next generation of particle colliders (like the FCC-ee) to find new physics. But to find something new, we must be absolutely sure we understand the old stuff (the Standard Model).

If we don't understand the "sneezing" of the electrons perfectly, we might mistake a math error for a new particle. This paper provides the ultimate calculator to ensure that when we see something strange at the next collider, it's actually a discovery, not just a glitch in our math.

In short: The authors built a super-precise simulation tool that perfectly combines the "big picture" math with the "nitty-gritty" details of how electrons behave, solving a tricky mathematical infinity problem along the way. This tool is ready for the next big leap in particle physics.

Technical Summary

Here is a detailed technical summary of the paper "e+e−→ZH at NLO EW matched to a QED parton shower" by Lois Flower and Marek Schönherr.

1. Problem Statement

As the particle physics community prepares for next-generation high-energy precision frontier colliders (specifically the proposed FCC-ee), theoretical predictions must match the anticipated experimental precision. Current Standard Model predictions often have theoretical uncertainties larger than experimental errors.

• The Specific Challenge: For electron-positron ($e^+e^-$) colliders, Initial State Radiation (ISR) from leptons is critical. Unlike hadron colliders where QCD radiation dominates, $e^+e^-$ collisions are heavily affected by QED radiation, which alters the effective center-of-mass energy of the collision (the "radiative return" effect).
• Limitations of Existing Methods:
  • Structure Function (SF) Method: Solves DGLAP equations analytically but struggles with the integrable singularity at $x=1$ (where $x$ is the momentum fraction) when combined with numerical event generation.
  • YFS Method: Exponentiates soft divergences but is distinct from the standard parton shower paradigm used in modern Monte Carlo generators.
  • Gap: There was a lack of an automated, process-independent framework that combines Next-to-Leading Order (NLO) Electroweak (EW) calculations with a QED parton shower (PS) for initial-state leptons. Existing tools like BABAYAGA use forward evolution and are limited to specific processes, while general-purpose generators lacked this specific matching capability.

2. Methodology

The authors developed a novel framework within the SHERPA event generator to match NLO EW calculations with a QED dipole parton shower using the MC@NLO method.

A. Construction of the QED Dipole Shower

• Backward Evolution: The shower utilizes a backward-evolution paradigm (standard for initial-state radiation in hadron colliders) adapted for leptons.
• Splitting Functions: They employ Catani-Seymour dipole splitting functions translated to QED.
• Handling Negative Weights: A major technical hurdle is that QED splitting functions can be negative when multiple charged particles are present (due to charge correlations). The standard veto algorithm fails here. The authors implemented a weighted veto algorithm using two overestimates ($g(t)$ and $h(t)$) to handle negative splitting probabilities while maintaining unitarity.
• Evolution Variable: They use modified transverse momentum ($\bar{k}_T^2$) for soft-photon singularities and reduced virtuality ($\bar{q}^2$) otherwise to ensure correct coherence and angular ordering.

B. The Electron Structure Function and Singularity Handling

• The Singularity: The electron structure function $f_e(x, Q^2)$ behaves as $(1-x)^{\beta-1}$ near $x=1$, creating an integrable singularity that causes numerical instability in Monte Carlo integration.
• Linear Rescaling Solution: Instead of the standard constant reweighting or $\delta$-function condensation, the authors introduced a linear rescaling ansatz ($ax+b$) in the region $[1-\delta, 1-\epsilon]$.
  • This ensures the structure function remains continuous and differentiable (up to the cutoff), allowing the parton shower to evolve particles into this region without generating infinite weights.
  • The shower is then forbidden from evolving further once a particle crosses the threshold $x_{max} = 1-\delta$.

C. MC@NLO Matching at NLO EW

• Formalism: They extended the MC@NLO formalism to Electroweak corrections.
  • S-events: Born-level configurations modified by the shower to subtract infrared divergences.
  • H-events: Real-emission configurations ($n+1$ particles) that provide the hard corrections.
• Scale Choices: Careful selection of factorization and shower starting scales ($\mu_F, \mu_Q$) is crucial to avoid double counting and ensure logarithmic accuracy. They use the invariant mass of the uncharged final state as the factorization scale.
• Input Schemes: They utilize a mixed scheme: the $G_\mu$ scheme for the Born process and $\alpha(0)$ for photon emissions in the shower, consistent with physical expectations for long-distance photons.

3. Key Contributions

1. First Automated NLO EW + QED PS: This is the first implementation of an automated, process-independent MC@NLO matching of NLO EW corrections with a QED parton shower for initial-state leptons.
2. Weighted Veto Algorithm: Development of a robust algorithm to handle negative splitting functions inherent in QED dipole showers with multiple charged particles.
3. Linear Rescaling of Structure Functions: A new method to regularize the $x \to 1$ singularity in electron structure functions that preserves continuity and allows for stable parton shower evolution.
4. Validation Framework: A comprehensive validation suite testing infrared parameters ($\epsilon, \delta, t_c$) and verifying the cancellation of divergences between fixed-order NLO and the matched prediction.

4. Results

The method was validated using $e^+e^- \to \nu_\mu \bar{\nu}_\mu$ and applied to Higgs production in association with a Z boson ($e^+e^- \to ZH$) at two FCC-ee energy points: 240 GeV (threshold) and 365 GeV (radiative production).

• Validation ($e^+e^- \to \nu_\mu \bar{\nu}_\mu$):

  • The MC@NLO prediction reproduces the fixed-order NLO total cross-section and differential distributions (e.g., neutrino transverse momentum) to within per-mille precision.
  • The method is stable under variations of the infrared cutoff ($t_c$) and structure function parameters ($\epsilon, \delta$), provided $\epsilon \le 10^{-8}$ and $\delta \le 10^{-4}$.
  • The "S-event" and "H-event" decomposition correctly reproduces the NLO behavior, with H-events dominating hard radiation tails.

• Phenomenology ($e^+e^- \to ZH$):

  • 240 GeV (Threshold): The phase space for hard radiation is restricted. The NLO EW correction is significant (-24% to the cross-section). The MC@NLO prediction matches the fixed-order NLO results for the $ZH$ invariant mass and $Z$ transverse momentum ($p_T$) with sub-percent precision.
  • 365 GeV (Radiative): Hard quasi-collinear radiation is abundant. The parton shower significantly impacts observables like the $ZH$ transverse momentum and photon multiplicity.
  • Comparison: The standard LO+PS (Leading Order + Shower) significantly overestimates hard radiation (by ~40-80%) compared to NLO. The MC@NLO method corrects this, providing a smooth transition between the soft/collinear region (resummed by the shower) and the hard region (described by the NLO matrix element).
  • Photon Multiplicity: The method correctly predicts the suppression of events with multiple hard photons compared to LO+PS, aligning with fixed-order NLO expectations.

5. Significance

• Precision Physics: This work provides the first tool capable of generating unweighted events for $e^+e^-$ collisions with full NLO EW accuracy and QED resummation. This is essential for the FCC-ee program, where luminosity measurements and Higgs coupling extractions require theoretical uncertainties at the $10^{-4}$ level.
• General Applicability: While demonstrated on $ZH$ production, the method is general and applicable to any colorless process at lepton colliders (e.g., Bhabha scattering, $e^+e^- \to f\bar{f}$).
• Future Outlook: The framework paves the way for combining QCD and QED showers for processes with colored final states and supports the development of Lepton PDFs for future high-energy lepton colliders. It bridges the gap between fixed-order precision and the phenomenological necessity of parton showers for detector-level simulations.