⚛️ phenomenology

Reinterpretation of ATLAS and CMS searches in monojet and mono- $V$ final states: prospects of limits on excited neutrinos

This paper reinterprets ATLAS and CMS monojet and mono- $V$ search results at 13 TeV to set upper limits on excited neutrino production, excluding masses up to approximately 4 TeV in benchmark scenarios and providing complementary constraints on low-mass, high-coupling regions.

Original authors: Gabriela Karkosova Martinovicova, Vojtech Pleskot

Published 2026-06-24

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Gabriela Karkosova Martinovicova, Vojtech Pleskot

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

Imagine the Large Hadron Collider (LHC) as the world's most powerful particle-smashing machine. Scientists use it to look for "new physics"—particles or forces that don't fit into our current rulebook, known as the Standard Model. One specific thing they are hunting for is the excited neutrino ( $\nu^*$ ).

Think of a regular neutrino as a calm, quiet ghost that rarely interacts with anything. An excited neutrino is like that same ghost, but it's been "supercharged" or "excited" to a higher energy state. It's heavier, more energetic, and might reveal that our fundamental particles aren't actually solid, indivisible dots, but rather complex structures made of even smaller, hidden building blocks (like a Lego brick made of smaller Lego pieces).

Here is what this paper does, broken down into simple concepts:

1. The Detective Work: "Mono" Searches

The scientists didn't build a new machine; they acted like detectives re-examining old crime scenes. They looked at data from two major detectors, ATLAS and CMS, which have been smashing protons together at record speeds.

They focused on a specific type of "crime scene" called a monojet or mono-V event.

The Analogy: Imagine a billiard table where you hit a cue ball (a proton) and it smashes into other balls. If a ball suddenly flies off the table at high speed, but you can't see what hit it or where it went, you know something invisible happened.
The Reality: In these collisions, a single, high-energy jet of particles (or a heavy particle called a "Vector Boson," like a W or Z) flies out one way, while a huge amount of "missing energy" flies the other way. This missing energy is the signature of a neutrino (or an excited neutrino) escaping undetected.

2. The Theory: The "BSZ" Model

To understand what they are looking for, the authors used a specific mathematical recipe called the BSZ model.

Think of this model as a "choose your own adventure" book for how these excited neutrinos are made and how they decay (break apart).
The model has two main ways these particles interact:
1. Contact Interactions (CI): Like two people bumping into each other in a crowded room without touching.
2. Gauge Interactions (GI): Like two people shaking hands or exchanging a gift.
The paper tests different "flavors" of this model by changing the strength of these interactions (represented by numbers $f$ and $f'$ ) and the size of the "composite" structure (represented by $\Lambda$ ).

3. The Simulation: A Digital Twin

Since they couldn't run new experiments, they built a digital twin of the LHC on their computers.

They simulated millions of collisions where excited neutrinos might have been created.
They programmed their computer to act exactly like the ATLAS and CMS detectors, seeing what these invisible particles would look like if they were there.
They then compared their "what-if" simulations against the actual data the real detectors collected.

4. The Results: How Heavy Can They Be?

The goal was to see if the real data contained any signs of these excited neutrinos. Since they found none, they could set a "limit" on how heavy these particles could possibly be before they would have been seen.

The "Monojet" Search (The Heavy Hitter):
- This search looked for a single jet of particles recoiling against missing energy.
- The Result: They found that if excited neutrinos exist, they must be extremely heavy—heavier than 4,000 GeV (about 4 TeV).
- The Metaphor: It's like saying, "If there is a giant monster hiding in the forest, it must weigh more than 4 tons, because if it weighed less, we would have heard it step on a twig." This pushes the limit of what we knew before (which was only up to 1.6 TeV) by a huge margin.
The "Mono-V" Search (The Specialized Hunter):
- This search looked for a specific type of heavy particle (W or Z boson) recoiling against missing energy.
- The Result: This method is better at finding lighter excited neutrinos, but only if they interact very strongly with the W and Z bosons. They found that for these specific conditions, they could rule out masses up to about 1.5 TeV.
- The Metaphor: This is like using a metal detector. It's not as good at finding everything, but if the object is made of a specific metal (strong interaction), it will find it even if it's smaller.

5. The "Safety Check"

A crucial part of the paper was making sure their detective work was logical.

The Problem: If the excited neutrinos were so common that they showed up in the "control rooms" (areas used to calibrate the background noise), it would mess up their calculations.
The Fix: They checked their simulations to ensure that for the masses they were testing, the "ghosts" wouldn't accidentally appear in the calibration zones. If the ghosts were too common in those zones, they had to ignore those specific mass ranges to keep their math honest.

Summary

In short, this paper is a re-analysis of existing data. By using new computer simulations based on the theory of "excited neutrinos," the authors showed that:

If these particles exist, they are likely much heavier than we previously thought (up to 4 TeV).
The ATLAS and CMS experiments are powerful enough to rule out a massive chunk of the "where could they be hiding" map.
Different search methods (Monojet vs. Mono-V) cover different parts of the map, working together to ensure no stone is left unturned.

They didn't find the particles, but they successfully narrowed down the search area, telling future physicists: "Don't look for light excited neutrinos here; if they exist, they are hiding in the heavy, high-energy zone."

Technical Summary: Reinterpretation of ATLAS and CMS Searches for Excited Neutrinos

Problem Statement
The search for excited neutrinos ( $\nu^*$ ), predicted by fermion compositeness models where quarks and leptons are composite objects, has been a long-standing objective in Beyond the Standard Model (BSM) physics. While limits on excited quarks ( $q^*$ ) and charged leptons ( $\ell^*$ ) have been extensively explored at the Large Hadron Collider (LHC), dedicated searches for $\nu^*$ have remained scarce. The most recent constraints were established by the ATLAS Collaboration in 2012 using 8 TeV data, excluding masses below 1.6 TeV. No comparable limits have been reported using the significantly larger Run 2 datasets collected at $\sqrt{s} = 13$ TeV, leaving the high-mass regime of the $\nu^*$ parameter space largely unexplored. This work addresses this gap by reinterpreting existing ATLAS and CMS monojet and mono-V searches to assess their sensitivity to $\nu^*$ production.

Methodology
The study employs a reinterpretation framework using the Baur-Spira-Zerwas (BSZ) effective model to describe $\nu^*$ phenomenology. In this framework, $\nu^*$ production occurs via contact interactions (CI) between quarks, while decays proceed through both CI and gauge-mediated interactions (GI).

Signal Generation: Signal events for $pp$ collisions at $\sqrt{s} = 13$ TeV were simulated using Pythia 8.310. The simulation covered a discrete mass range of $m_{\nu^*} = 600$ GeV to 5000 GeV. Two benchmark scenarios for the compositeness scale ( $\Lambda$ ) were considered: $\Lambda = m_{\nu^*}$ and $\Lambda = 10$ TeV. Various coupling parameters ( $f, f'$ ) were tested, with values ranging from 0.0 to 1.0.
Detector Simulation: Detector effects were modeled using Delphes 3.5.0 with specific configurations approximating the ATLAS and CMS detectors. Jet reconstruction utilized the anti- $k_t$ algorithm (AK4 with $R=0.4$ for both; AK8 with $R=0.8$ was added for CMS to emulate boosted object reconstruction).
Reinterpretation Strategy: The authors utilized published signal region (SR) selections, post-fit background estimates, and observed event yields from:
- ATLAS: A monojet search using 139 fb $^{-1}$ of Run 2 data.
- CMS: A combined monojet and mono-V search using 101 fb $^{-1}$ of 2017–2018 data.
  Statistical limits were derived using the profile likelihood ratio hypothesis test and the $CL_s$ method. The analysis included a rigorous consistency check to ensure that potential $\nu^*$ signals did not significantly contaminate the control regions (CRs) used to normalize backgrounds. Points where the signal-to-background ratio in high- $p_T$ CR subregions exceeded 10% were excluded from the statistical test to avoid biasing the background estimation.

Key Contributions and Results
The paper derives 95% confidence-level (CL) upper limits on the $\nu^*$ production cross-section as a function of mass and coupling parameters.

ATLAS Monojet Search:
- For the benchmark scenario $\Lambda = m_{\nu^*}$ , the monojet searches allow the exclusion of excited neutrino masses up to approximately 3.9 TeV (observed limit) for all three flavors ( $\nu^*_e, \nu^*_\mu, \nu^*_\tau$ ) and representative coupling values ( $f=f'$ ).
- The exclusion limits show negligible dependence on the coupling parameter $f$ in this regime, as the monojet signal regions are dominated by CI-mediated decays ( $\nu^* \to \nu q\bar{q}$ ).
- For the $\Lambda = 10$ TeV scenario, the monojet channel showed insufficient sensitivity to set competitive limits in the considered mass range.
CMS Monojet and Mono-V Searches:
- Monojet: Similar to the ATLAS results, the CMS monojet channel provided competitive limits for $\Lambda = m_{\nu^*}$ , excluding masses up to ~3.8 TeV.
- Mono-V: The mono-V channel (targeting events with a large-radius jet consistent with hadronic $W/Z$ decays) proved sensitive to a complementary region of the parameter space characterized by large couplings ( $f, f'$ ) and high $\Lambda$ (specifically $\Lambda = 10$ TeV).
- For $\Lambda = 10$ TeV and $f=f'=1$ , the CMS mono-V search excludes masses up to ~1.5 TeV. This channel is particularly sensitive to GI-mediated decays ( $\nu^* \to \ell W$ ), where the $W$ boson decays hadronically. The limits for $\nu^*_\tau$ are slightly stronger than for $\nu^*_e$ and $\nu^*_\mu$ due to the higher probability of misreconstructing hadronically decaying $\tau$ leptons as jets.

Significance and Claims
The authors claim that this work demonstrates the potential of existing LHC Run 2 mono-object searches to significantly extend the excluded parameter space for excited neutrinos. Specifically:

The study updates the 2012 limit of 1.6 TeV to nearly 4 TeV for the $\Lambda = m_{\nu^*}$ scenario using Run 2 data.
It highlights the complementarity between monojet and mono-V searches: monojet searches constrain the high-mass regime dominated by contact interactions, while mono-V searches constrain the low-mass, high-coupling regime dominated by gauge-mediated interactions.
The paper explicitly notes that while the systematic uncertainty treatment and signal simulation are less rigorous than those performed by the experimental collaborations themselves, the results provide a robust assessment of the sensitivity of current datasets to $\nu^*$ production.
The authors provide a GitLab repository containing the derived cross-section limits and expected exclusion contours for all tested scenarios to facilitate further phenomenological studies.