Evidence of $ZZγ$ production with the ATLAS detector

Using the full Run-2 dataset of proton-proton collisions at 13 TeV collected by the ATLAS detector, this paper presents the first evidence of simultaneous $ZZ\gamma$ production with a significance of 4.4$\sigma$, observing eight events against a background of 0.92 and measuring a fiducial cross-section consistent with Standard Model predictions.

The Gist

Imagine the Large Hadron Collider (LHC) as the world's most powerful, high-speed particle racetrack. Inside this 27-kilometer ring, scientists smash protons (tiny subatomic particles) together at nearly the speed of light. Usually, these crashes are messy, creating a chaotic cloud of debris. But sometimes, if the conditions are just right, the crash produces something rare and beautiful: a specific trio of particles appearing out of nowhere.

This paper is the announcement that the ATLAS experiment (a giant, cylindrical detector surrounding the racetrack) has finally caught a glimpse of a very rare event: the simultaneous production of two Z bosons and one photon.

Here is the story of how they found it, explained simply.

1. The "Holy Grail" Trio

In the Standard Model (the rulebook of physics), particles usually come in pairs or single units. Getting three specific heavy particles to show up at the exact same time is like winning the lottery three times in a row.

• The Z Boson: Think of these as the "heavyweight champions" of the particle world. They are massive and unstable, meaning they fall apart almost instantly into other particles (specifically, electrons or muons, which are like heavy cousins of electrons).
• The Photon: This is a particle of light. In this experiment, it's a high-energy "flash" of light.

The team was looking for a specific signature: Two Z bosons + One Photon.
Since the Z bosons fall apart immediately, the scientists couldn't see the Zs themselves. Instead, they looked for the "footprints" they left behind: four charged particles (leptons) and one photon.

2. The Search for a Needle in a Haystack

The LHC runs for years, creating billions of collisions. But the specific event the scientists wanted (ZZγ) is incredibly rare.

• The Haystack: The "background noise." Most collisions produce common particles, fake signals, or random glitches that look like the rare event but aren't. It's like trying to hear a specific whisper in a stadium full of cheering fans.
• The Needle: The actual signal.

The team analyzed data from 2015 to 2018 (a massive amount of data, equivalent to 140 "inverse femtobarns"—a unit of collision volume). They built a sophisticated filter to ignore the noise. They looked for events where:

1. There were exactly four leptons (electrons or muons).
2. There was exactly one high-energy photon.
3. The leptons came in two matching pairs (like two couples dancing).

3. The Result: Finding 8 Needles

After sifting through the massive haystack, the scientists found 8 events that matched their criteria perfectly.

• The Expectation: Based on their calculations of "noise" (background), they expected to find less than 1 event (about 0.92) just by chance.
• The Reality: They found 8.

This is a huge difference. To prove this wasn't just a lucky fluke or a statistical error, they calculated the "significance."

• The Analogy: Imagine flipping a coin. If you get heads 10 times in a row, it's suspicious. If you get heads 100 times, it's impossible by chance.
• The Score: This result had a significance of 4.4 sigma. In the world of particle physics, 3 sigma is "evidence" (we're pretty sure), and 5 sigma is "discovery" (we are absolutely certain). They are very close to the "discovery" threshold, officially calling this "Evidence."

4. Why Does This Matter?

You might ask, "So what? We found 8 rare events."
This is crucial for three reasons:

1. Testing the Rulebook: The Standard Model predicts exactly how often this should happen. The scientists measured the rate and found it matched the prediction almost perfectly. This confirms our current understanding of how the universe works at the most fundamental level.
2. The "Gauge Structure": The Z bosons and photons interact via forces called "gauge couplings." Seeing them together helps us understand the "glue" that holds the universe together.
3. Looking for New Physics: If the rate had been higher or lower than predicted, it would have been a smoking gun for "New Physics"—perhaps a hidden particle or a force we don't know about yet. Since it matched, it tells us the Standard Model is still holding strong, but it also sets the stage for future experiments (like the upcoming Run 3 of the LHC) to look for even rarer deviations.

5. The "Fake" Problem

One of the hardest parts of this job was dealing with "imposters."

• Fake Photons: Sometimes a jet of particles (like a spray of debris) looks like a single photon to the detector. The team used clever math to estimate how many of these fakes were in their data and subtracted them out.
• Fake Leptons: Sometimes a particle gets misidentified as an electron or muon. They used a "matrix method" (a statistical tool) to estimate and remove these errors.

The Bottom Line

The ATLAS collaboration has successfully spotted a very rare cosmic event: two heavy Z bosons and a photon appearing together. They found 8 real examples where they expected less than 1. This confirms the Standard Model's predictions and proves that our detectors and theories are working correctly. It's a victory for precision physics, showing that even in the chaotic storm of particle collisions, we can find the specific, rare patterns that tell us how the universe is built.

Technical Summary

1. Problem and Motivation

The Standard Model (SM) of particle physics describes fundamental interactions through gauge symmetries. While the production of single and double electroweak gauge bosons ($W$, $Z$, $\gamma$) has been well-studied, the simultaneous production of three gauge bosons (triboson production) is a rare process with a very low cross-section.

• Scientific Goal: The primary objective is to observe the $ZZ\gamma$ final state ($pp \to ZZ\gamma$) to test the gauge structure of the SM.
• Physics Implications: Measuring this process allows for:
  • Validation of SM predictions at the next-to-leading order (NLO) in perturbative QCD.
  • Probing of quartic gauge couplings (interactions between four gauge bosons), which are sensitive to potential new physics contributions beyond the SM.
  • Investigation of the Higgs boson's coupling to $Z$ bosons and photons via the $ZH \to ZZ\gamma$ channel (though the Higgs contribution is subdominant).

2. Methodology

Dataset and Detector

• Experiment: ATLAS detector at the Large Hadron Collider (LHC).
• Dataset: Full Run-2 dataset collected from 2015 to 2018.
• Conditions: Proton-proton ($pp$) collisions at a center-of-mass energy of $\sqrt{s} = 13$ TeV.
• Integrated Luminosity: $140 \text{ fb}^{-1}$.

Signal Definition and Final State

The analysis focuses on the fully leptonic decay channel:
$$pp \to ZZ\gamma \to \ell^+\ell^-\ell'^+\ell'^-\gamma$$
where $\ell, \ell' \in \{e, \mu\}$. This results in a final state of four leptons and one photon.

• Signal Components:
  1. Triboson Production: Dominant contribution via Feynman diagrams involving triple and quartic gauge couplings (Figure 1a).
  2. Higgs Production: Subdominant contribution from $ZH \to ZZ\gamma$ (Figure 1b), included as part of the signal.
  3. Excluded: Final-state radiation (FSR) from $ZZ$ production (Figure 1c) is explicitly rejected to isolate the genuine $ZZ\gamma$ process.

Event Selection (Signal Region)

• Leptons: Two opposite-sign, same-flavor (OSSF) lepton pairs.
  • Transverse momentum ($p_T$): Leading $> 30$ GeV, sub-leading $> 25$ GeV, others $> 10$ GeV.
  • Pseudorapidity ($|\eta|$): Electrons $< 2.47$ (excluding transition region $1.37 < |\eta| < 1.52$), Muons $< 2.5$.
  • Identification: Medium ID for electrons/muons; isolation criteria applied.
• Photon:
  • $p_T > 20$ GeV.
  • $|\eta| < 2.37$ (excluding transition region).
  • Tight identification and loose isolation.
  • Separation from leptons: $\Delta R(\ell, \gamma) > 0.4$.
• Kinematic Cuts:
  • Invariant mass of lepton pairs ($m_{\ell\ell}$) must be close to the $Z$ mass ($m_Z$), with a minimum of 40 GeV.
  • FSR Rejection: To suppress $ZZ$ events with radiated photons, the sum of the three-body invariant mass ($m_{\ell\ell\gamma}$) and the two-body invariant mass ($m_{\ell\ell}$) for each pair must exceed $2m_Z$.

Background Estimation

Due to the low expected signal yield, data-driven methods were crucial:

1. Fake Photons: Estimated using a fake photon ratio method. The ratio of fake photons in $Z\gamma$ events (Control Region) is extrapolated to $ZZ\gamma$ events, assuming independence from the number of $Z$ bosons.
2. Fake Leptons: Estimated using the matrix method, utilizing a $Z+1\ell$ control region to derive efficiencies for prompt vs. fake leptons.
3. Pileup: Estimated by overlaying inclusive $ZZ \to 4\ell$ events with photons from $\gamma + \text{jet}$ events at the particle level, corrected for detector resolution.
4. Irreducible Backgrounds: $t\bar{t}\gamma$, $WZ$, and $ZZ$ processes were estimated using Monte Carlo (MC) simulations.

Simulation

• Signal: Generated with Sherpa 2.2.11 (NLO QCD, LO EW) for $ZZ\gamma$ and Powheg-Box + Pythia 8 for $ZH \to ZZ\gamma$.
• Detector Response: Modeled with Geant4.

3. Key Contributions

• First Evidence: This paper reports the first evidence of $ZZ\gamma$ production in $pp$ collisions.
• Fiducial Cross-Section Measurement: Provides a precise measurement of the fiducial cross-section in a region minimizing photon final-state radiation.
• Background Modeling: Demonstrates a robust data-driven approach for estimating fake photon backgrounds in a low-statistics, multi-object final state.
• FSR Suppression: Implements a specific kinematic criterion to distinguish genuine $ZZ\gamma$ production from $ZZ$ events with final-state radiation.

4. Results

• Event Yield:

  • Observed Events: 8 events were selected in the signal region.
  • Expected Background: $0.92 \pm 0.15$ events.
  • Expected Signal: $7.04 \pm 0.18$ events (combining $ZZ\gamma$ and $ZH \to ZZ\gamma$).

• Significance:

  • The background-only hypothesis is rejected with an observed significance of 4.4$\sigma$ (expected: 4.4$\sigma$). This meets the threshold for "evidence" (typically $>3\sigma$) but falls just short of "observation" ($>5\sigma$).

• Cross-Section Measurement:
  The measured fiducial cross-section is:
  $$\sigma_{ZZ\gamma} = 0.144^{+0.064}_{-0.051} (\text{stat.}) ^{+0.007}_{-0.005} (\text{syst.}) \text{ fb}$$

  • Total Uncertainty: Dominated by statistical uncertainty (~40%). Total systematic uncertainty is 4.1%.
  • Comparison with SM: The result is in excellent agreement with the Standard Model prediction:
    $$\sigma_{SM}^{\text{fid.}} = 0.143^{+0.007}_{-0.004} \text{ fb}$$
  • Signal Strength: The measured signal strength relative to the SM is $\mu_{ZZ\gamma} = 1.01^{+0.45}_{-0.36} (\text{stat.}) \pm 0.05 (\text{syst.}) \pm 0.06 (\text{theory})$.

• Distributions: The kinematic distributions (photon $p_T$, $m_{4\ell\gamma}$, and dilepton masses) show good agreement between data and the combined signal-plus-background prediction.

5. Significance

• Validation of SM: The measurement confirms the SM prediction for the rare $ZZ\gamma$ production process, validating the electroweak sector's gauge structure at the 13 TeV scale.
• Foundation for Future Physics: This result establishes the methodology and sensitivity required to probe quartic gauge couplings (QGCs). With the larger datasets expected in LHC Run-3 and Run-4, this channel will become a powerful tool for searching for anomalous interactions and new physics phenomena that modify the $ZZ\gamma$ vertex.
• Technical Milestone: Successfully isolating a 4-lepton + 1-photon signal with such low background demonstrates the high performance of the ATLAS detector and the effectiveness of advanced data-driven background estimation techniques.