Search for the inert doublet model in a final state with two leptons at the CMS experiment
File(s)
Author(s)
Curtis, Edward
Type
Thesis
Abstract
This thesis describes two searches for neutral scalars of the Inert Doublet Model (IDM)
in the dilepton final state. The first search is performed using proton-proton
collision data collected from 2016-2022 with the CMS detector at centre of mass
energies of $\sqrt{s} = 13.0$ TeV and $\sqrt{s} = 13.6$ TeV, corresponding
to a total integrated luminosity of 172.2 fb$^{-1}$. No significant excess above the SM
prediction is found. 95\% CL upper limits are
placed on the masses of the two neutral IDM scalars, $H$ and $A$, resulting in
a maximum exclusion up to $M_H=$ 108 GeV at a mass splitting of approximately $M_A - M_H = 80$ GeV.
The second search is a projection study looking at the
sensitivity to the IDM at the Future Circular Collider with electron-positron
collisions at $\sqrt{s} = 240$ GeV and $\sqrt{s} = 365$ GeV, corresponding to 10.8
ab$^{-1}$ and 2.7 ab$^{-1}$ of integrated luminosity, respectively. Using
solely simulated events and a simplified detector model, expected limits are
derived for the two neutral IDM scalars, resulting in an expected exclusion of
nearly the entire kinematically-allowed parameter space.
In addition, a study is presented on the performance of
graph neural networks (GNNs) for electron and photon identification versus
pileup, carried out in the context of the High-Luminosity LHC CMS endcap
upgrade. Utilising only the raw energy deposits in the detector, the GNN is
capable of automatically extracting meaningful features from the data, and is
shown to outperform current methods based on boosted decision trees.
in the dilepton final state. The first search is performed using proton-proton
collision data collected from 2016-2022 with the CMS detector at centre of mass
energies of $\sqrt{s} = 13.0$ TeV and $\sqrt{s} = 13.6$ TeV, corresponding
to a total integrated luminosity of 172.2 fb$^{-1}$. No significant excess above the SM
prediction is found. 95\% CL upper limits are
placed on the masses of the two neutral IDM scalars, $H$ and $A$, resulting in
a maximum exclusion up to $M_H=$ 108 GeV at a mass splitting of approximately $M_A - M_H = 80$ GeV.
The second search is a projection study looking at the
sensitivity to the IDM at the Future Circular Collider with electron-positron
collisions at $\sqrt{s} = 240$ GeV and $\sqrt{s} = 365$ GeV, corresponding to 10.8
ab$^{-1}$ and 2.7 ab$^{-1}$ of integrated luminosity, respectively. Using
solely simulated events and a simplified detector model, expected limits are
derived for the two neutral IDM scalars, resulting in an expected exclusion of
nearly the entire kinematically-allowed parameter space.
In addition, a study is presented on the performance of
graph neural networks (GNNs) for electron and photon identification versus
pileup, carried out in the context of the High-Luminosity LHC CMS endcap
upgrade. Utilising only the raw energy deposits in the detector, the GNN is
capable of automatically extracting meaningful features from the data, and is
shown to outperform current methods based on boosted decision trees.
Version
Open Access
Date Issued
2025-09-30
Date Awarded
2026-02-01
Copyright Statement
Attribution-NonCommercial-ShareAlike 4.0 International Licence (CC BY NC-SA)
Advisor
Wardle, Nicholas
Magnan, Anne-Marie
Sponsor
Science and Technology Facilities Council (Great Britain)
Publisher Department
Department of Physics
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)