Probing new physics with four top quarks

Projektinformationen

4TopAndMore

ID Finanzhilfevereinbarung: 101107774

DOI

10.3030/101107774

Projekt abgeschlossen

EK-Unterschriftsdatum 3 April 2023

Startdatum 16 Oktober 2023

Enddatum 15 März 2026

Finanziert unter

• Marie Skłodowska-Curie Actions (MSCA)

Gesamtkosten

Keine Daten

EU-Beitrag

€ 181 152,96

Koordiniert durch

INSTITUTO DE FISICA DE ALTAS ENERGIAS
Spain

Periodic Reporting for period 1 - 4TopAndMore (Probing new physics with four top quarks)

Berichtszeitraum: 2023-10-16 bis 2026-03-15

Zusammenfassung vom Kontext und den Gesamtzielen des Projekts

The Standard Model of particle physics is an extremely successful description of known fundamental particles and interactions, but it remains incomplete. It does not explain, for example, the origin of the Higgs-boson mass scale, the nature of dark matter, or the observed matter–antimatter asymmetry of the universe. Many theories beyond the Standard Model predict new particles or interactions, referred to as new physics, that couple strongly to the top quark, the heaviest known elementary particle. Processes involving several top quarks are therefore especially powerful probes of new physics.

This project focused on the production of four top quarks, tttt, using data collected by the ATLAS experiment at the Large Hadron Collider (LHC). Four-top-quark production is one of the rarest and most energetic processes accessible at the LHC. It can reveal new physics in several ways: through a production rate larger than predicted, through altered event kinematics, through new heavy particles that decay into top quarks, or through indirect effects described by effective field theories.

The original objective of the project was to use tttt production to search for new physics through three connected steps: discovery and inclusive cross-section measurement, differential measurements of kinematic distributions, and interpretation of the results in terms of possible new physics. The project also planned to develop improved reconstruction tools for complex multi-top-quark events.

During the fellowship, the scientific context changed significantly. Four-top-quark production was discovered before the start of the project in a previous analysis led by the researcher using LHC Run 2 data. The project was therefore re-focused towards the next scientific step: using the established tttt process as a platform for dedicated new-physics searches and preparing the analysis tools needed for LHC Run 3 data at the new energy frontier of 13.6 TeV.

The project pathway to impact was therefore to improve the experimental and methodological foundations for future tttt measurements and searches. This included preparing Run 3 analysis infrastructure, improving object selection and electron calibration, implementing improved simulations for rare backgrounds, developing machine-learning tools, contributing to new-physics searches, and producing sensitivity projections for the High-Luminosity LHC (HL-LHC). These activities contribute to the long-term European strategy for particle physics by clarifying what future LHC and HL-LHC data can reveal about the fundamental structure of matter.

Arbeit, die ab Beginn des Projekts bis zum Ende des durch den Bericht erfassten Berichtszeitraums geleistet wurde, und die wichtigsten bis dahin erzielten Ergebnisse

The project carried out scientific and technical work in several connected areas. A major focus was the preparation of tttt analyses for LHC Run 3. This included software-framework preparation, processing of simulation samples, validation of analysis workflows and optimisation of object and event selections under the new run’s condition, in particular in events with two same-sign leptons or more than two leptons. These channels are critical because they provide strong sensitivity to rare multi-top-quark processes while suppressing many Standard Model backgrounds.

One of the major achievements was the implementation of state-of-the-art three-top-quark (ttt) simulation samples in the ATLAS framework. Three-top production was identified as a critical background for tttt measurements during the discovery of this process. The implementation of the first available next-to-leading-order prediction for this process improved the modelling foundation for future tttt and other multi-top analyses. Associated theoretical uncertainties were also studied in consultation with theory experts.

The project also developed and integrated a graph neural network (GNN) for multi-top-quark event classification. The network used information from reconstructed objects and their pairwise relations to improve the separation of tttt events from background processes. The GNN tool demonstrated superiority against traditional multivariate analysis method such as one based on boosted decision trees.

The project contributed directly to dedicated new-physics searches. A search for heavy scalar and pseudo-scalar particles in tttt events using Run 2 data was completed and published. This search provided new constraints on models predicting additional Higgs-like particles. Substantial progress was also made towards the first dedicated new physics search in tttt events including Run 3 data.

Although the full differential four-top-quark cross-section measurement was not completed during the fellowship, the project built foundations for future inclusive and differential measurements. These include Run 3-ready analysis infrastructure, improved object and event selections, rare-background modelling, and machine-learning classification.

The project also contributed to combined effective-field-theory studies by including a tttt-sensitive signal region as an orthogonal input to a broader multi-top-quark interpretation using Run 2 data. This tested how four-top information can be integrated into future global interpretations of top-quark-sector measurements.

In addition, the project produced HL-LHC sensitivity projections for tttt cross-section measurements, heavy-resonance searches and effective-field-theory sensitivity using differential observables. These studies provided input to the 2026 European Strategy Update for Particle Physics.

Further scientific outputs included a complementary search for heavy vector-like quark, a group of new particles predicted by theories addressing the question on the smallness of the Higgs boson mass, and an ATLAS detector-performance study of the Tile Calorimeter radiation hardness on a tile-by-tile basis. The latter contributed to understanding detector performance under future HL-LHC conditions.

Fortschritte, die über den aktuellen Stand der Technik hinausgehen und voraussichtliche potenzielle Auswirkungen (einschließlich der bis dato erzielten sozioökonomischen Auswirkungen und weiter gefassten gesellschaftlichen Auswirkungen des Projekts)

The project advanced the state of the art in several ways.

The project contributed to the transition of four-top-quark physics from discovery to exploitation. Once four-top production had been established, the key scientific question became how to use this rare process to search for new particles and interactions. The project addressed this by contributing to dedicated new physics searches in tttt events, including a published LHC Run 2 result and major progress towards a result including LHC Run 3 data.

The project improved the modelling of rare top-quark backgrounds. The implementation of the state-of-the-art next-to-leading-order ttt simulation in the ATLAS framework was an important advance because ttt production had been identified as a critical background in studies of tttt and other rare multi-top production. This improves the reliability of future analyses in rare multi-top final states.

The project developed a GNN-based approach for classifying complex multi-top events. This method uses relational information between reconstructed particles, making it well suited to final states with many jets, leptons and possible top-quark decay products. The approach improved signal-background separation and simplified the analysis workflow by combining classification and reconstruction-related information in a single model.

The project contributed to physics cases of future tttt studies at the HL-LHC. The sensitivity projections showed how future datasets can improve measurements of the tttt cross-section, extend searches for heavy resonances, and probe effective-field-theory operators through differential observables such as HT (the scalar sum of transverse momenta in the event). These results go beyond immediate Run 2 and Run 3 analyses by assessing the long-term discovery and interpretation potential of the HL-LHC.

The project also produced outputs that support further uptake by the scientific community. The Run 3 analysis infrastructure, improved object-selection studies, ttt simulation implementation, and tttt-sensitive regions for combined interpretations can be reused in future ATLAS measurements and searches. Further research will be needed to complete the Run 3 inclusive and differential tttt measurements, finalise the Run 2+Run 3 new-physics search, and develop full global EFT interpretations using multiple top-quark processes.