⚛️ high-energy experiments

Observation of electroweak production of pairs of Z bosons in proton-proton collisions at 13 TeV

The CMS experiment reports the first evidence of electroweak production of Z boson pairs in association with two jets using 138 fb $^{-1}$ of proton-proton collision data at 13 TeV, observing a significance of 3.1 standard deviations that rises to 5.0 standard deviations when combined with the four-lepton channel.

Original authors: CMS Collaboration

Published 2026-06-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: CMS Collaboration

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Catching a Ghostly Double-Decker

Imagine the Large Hadron Collider (LHC) at CERN as a massive, high-speed racetrack where tiny particles called protons are smashed together at nearly the speed of light. When they crash, they create a chaotic explosion of new particles, like shrapnel from a firework.

Physicists from the CMS experiment (a giant, high-tech camera surrounding the crash site) have been looking for a very specific, rare type of "shrapnel": two Z bosons (heavy particles that act like messengers of the weak nuclear force) appearing together, accompanied by two jets (sprays of particles).

The tricky part? One of these Z bosons is "invisible." It decays into neutrinos, which are ghost-like particles that pass right through the detector without leaving a trace. The other Z boson decays into two visible particles (electrons or muons). So, the scientists are looking for a scene with:

Two visible charged particles.
Two sprays of particles (jets).
A "missing" amount of energy (because the ghosts escaped).

The Main Discovery: "Evidence" of a Rare Dance

The paper reports the first evidence of these Z bosons being created through a process called Electroweak (EW) production.

The Analogy:
Think of the collision as a crowded dance floor.

The "QCD" background: Most of the time, Z bosons are created by the "strong force" (like a bouncer shoving people together). This is messy, common, and creates a lot of noise. It's like a mosh pit where everyone is bumping into each other randomly.
The "Electroweak" signal: The scientists are looking for a specific, elegant dance where two Z bosons are created by the "weak force" (like two dancers gently tossing a ball to each other from across the room). This is rare and happens when two quarks (the dancers) exchange a W or Z boson and scatter apart, leaving two jets behind them.

The team analyzed 138 "years" of data (a massive amount of collision records) and found that the number of these "elegant dances" they saw matches what the Standard Model (the rulebook of physics) predicts.

The Result: They found the signal with a statistical significance of 3.1 standard deviations. In the world of particle physics, this is like hearing a faint whisper in a noisy room and being 99.9% sure it's a voice and not just wind. It's strong "evidence," but not quite a "discovery" (which usually requires a 5.0 whisper).

The "Missing" Energy Puzzle

How do you find a particle that leaves no trace?
The scientists used a clever trick. They measured the total energy of everything visible in the crash. If the math doesn't add up—if there is a "hole" in the energy balance—they know something invisible (neutrinos) must have flown out. They required this "missing energy" to be very large to filter out the noise.

The Detective Work: Filtering the Noise

To find this rare signal, the team had to filter out millions of "fake" events. They used a Graph Neural Network (GNN).

The Analogy: Imagine trying to find a specific conversation in a stadium full of people shouting. A normal filter might just look at how loud a person is. The GNN is like a super-intelligent detective who looks at the relationships between people. It sees that the two jets are far apart and moving in a specific way, and the two electrons are paired up, and the missing energy is in a specific direction. It connects the dots to say, "This specific group of particles is likely the rare signal, not just random noise."

The "What If" Test: Looking for New Physics

The scientists also asked: "Is there anything weird here? Are there new forces we don't know about?"
They looked for Anomalous Quartic Gauge Couplings (aQGCs).

The Analogy: Imagine the Standard Model is a set of traffic laws. The scientists checked to see if the Z bosons were breaking the speed limit or driving on the wrong side of the road (interacting in ways the laws don't allow).
The Result: They found no evidence of rule-breaking. The Z bosons behaved exactly as the traffic laws predicted. They set new limits on how much "rule-breaking" could possibly be happening in the future.

The Grand Finale: Combining Forces

The paper mentions a previous study by the same team that looked at a different version of this event: where both Z bosons decay into visible particles (four charged leptons).

The Combination: When they combined the results from the "invisible neutrino" search (this paper) and the "all-visible" search (the previous paper), the evidence became much stronger.
The Final Score: The combined significance reached 5.0 standard deviations.
What this means: In the language of physics, this crosses the threshold from "evidence" to "Observation." The CMS experiment has now officially observed the electroweak production of Z boson pairs.

Summary

The CMS team has successfully spotted a rare, elegant interaction between two Z bosons created by the weak force, accompanied by two jets. They used advanced AI to filter out the noise, confirmed the invisible neutrinos were there by measuring missing energy, and found that nature is behaving exactly as the Standard Model predicts. By combining this with a previous result, they have officially observed this phenomenon for the first time, completing the picture of how all massive gauge boson pairs (WW, WZ, and ZZ) are produced via the electroweak force.

Technical Summary: Observation of Electroweak Production of Pairs of Z Bosons in Proton-Proton Collisions at 13 TeV

Problem and Motivation
In the Standard Model (SM) of particle physics, the Brout–Englert–Higgs mechanism is responsible for the masses of the W and Z bosons. A critical consequence of this mechanism is that Higgs boson exchange plays an essential role in vector boson scattering (VBS) processes, ensuring that scattering amplitudes remain within unitarity limits at high energies. While the electroweak (EW) production of same-sign WW, opposite-sign WW, and WZ pairs has been previously observed, the EW production of Z boson pairs ($ZZ$) accompanied by two jets ($jj$) represents a significant gap in the experimental verification of all massive gauge boson pair productions. This analysis targets the VBS process $pp \to ZZjj$ , specifically in the final state $ZZjj \to \ell\ell\nu\nu jj$ (where $\ell = e, \mu$ ), where one Z boson decays into charged leptons and the other into a neutrino pair. This channel is challenging due to the presence of large missing transverse momentum ( $p^{\text{miss}}_T$ ) and significant backgrounds from Quantum Chromodynamics (QCD) induced processes and other electroweak diboson productions.

Methodology
The analysis utilizes a dataset of proton-proton collisions collected by the CMS detector at the CERN LHC at a center-of-mass energy of $\sqrt{s} = 13$ TeV, corresponding to an integrated luminosity of 138 fb $^{-1}$ (2016–2018).

Event Selection: Candidate events are selected requiring exactly two prompt, same-flavor, opposite-sign (OSSF) leptons with transverse momentum ( $p_T$ ) thresholds of 25 GeV (leading) and 20 GeV (subleading), and an invariant mass consistent with a Z boson (76–106 GeV). Events must exhibit large missing transverse momentum ( $p^{\text{miss}}_T > 120$ GeV) and two jets with a large rapidity separation ( $|\Delta\eta_{jj}| > 2.5$ ) and a large dijet invariant mass ( $m_{jj} > 400$ GeV). Additional vetoes are applied to reject events with additional leptons, hadronically decaying taus, or b-tagged jets to suppress backgrounds from WZ, WW, top-quark, and Drell–Yan (DY) processes.
Background Estimation:
- Processes with neutrinos (WW, WZ, $t\bar{t}$ ) are modeled using Monte Carlo (MC) simulations (POWHEG v2, MCFM) with normalizations constrained by dedicated control regions (CRs): a 3-lepton CR (dominated by WZ) and an e $\mu$ CR (dominated by WW and $t\bar{t}$ ).
- The DY background, which lacks genuine $p^{\text{miss}}_T$ but can mimic the signal due to instrumental effects, is estimated using a data-driven method. A control sample of single-photon events ( $\gamma + \text{jets}$ ) is used as a proxy for Z + jets events, reweighted to match the kinematic distributions of the DY process in the signal region.
Signal Extraction: To enhance the separation between the rare EW signal and backgrounds, a Graph Neural Network (GNN) discriminator is employed. The GNN operates on a graph representation of physics objects (leptons and jets) and their pairwise correlations. The analysis utilizes a binned maximum likelihood fit to the flattened distribution of the GNN classifier score ( $D^{\text{flat}}_{\text{GNN}}$ ), combining all data-taking periods. Systematic uncertainties (theoretical modeling, jet energy scale/resolution, lepton efficiencies) are treated as nuisance parameters.
Effective Field Theory (EFT) Interpretation: The data is also used to probe anomalous quartic gauge couplings (aQGCs) within the Standard Model Effective Field Theory (SMEFT) framework, specifically using dimension-8 operators.

Key Contributions and Results

Cross Section Measurement: The analysis measures the EW production cross section in a fiducial volume defined by generator-level quantities. The result is $\sigma_{\text{EW}}(pp \to ZZjj \to \ell\ell\nu\nu jj) = 0.37^{+0.14}_{-0.12} (\text{stat}) ^{+0.06}_{-0.06} (\text{syst})$ fb. This is in agreement with the Standard Model prediction of $0.39 \pm 0.06$ fb.
Significance of Observation: The observed significance of the signal in the $\ell\ell\nu\nu$ channel is 3.1 standard deviations ( $\sigma$ ), with an expected significance of 2.8 $\sigma$ .
Combined Significance: By combining these results with a previously reported CMS measurement in the four-lepton ( $\ell\ell\ell'\ell'$ ) channel, the analysis achieves an observed (expected) significance of 5.0 (4.5) standard deviations for the EW production of Z boson pairs.
Anomalous Couplings: Limits on aQGCs are derived in terms of dimension-8 operators ( $S_0$ – $S_2$ , $M_0$ – $M_7$ , $T_1$ – $T_9$ ). The observed and expected 95% confidence level limits are consistent with previous measurements and show no evidence of beyond-the-Standard-Model effects.

Significance
This work reports the first evidence for the electroweak production of Z boson pairs in association with two jets in the $\ell\ell\nu\nu$ final state. When combined with the four-lepton channel, the result constitutes the first observation of EW ZZ production by the CMS experiment. This achievement completes the experimental observation of the EW production of all pairs of massive gauge bosons (WW, WZ, and ZZ) at the LHC, providing a crucial test of the SM mechanism for vector boson scattering and the unitarization of scattering amplitudes. The paper further establishes constraints on dimension-8 operators, contributing to the global search for new physics in quartic gauge couplings.