Search

Seismic analysis of large tunnel systems using the continuum (Finite Element; FE) approach can be complex and computationally expensive. The inefficiency stems from the extended length of tunnels, compared to the tunnel diameter that controls the local ground-structure interaction, passing through different terrain and lithological profiles, with complex fixity conditions provided by the intermediate station boxes, and the increased significance of asynchronous ground motion effects given the length scale. This study proposes an uncoupled numerical methodology to model and analyse the seismic response of large tunnel systems that is able to consider various tunnel alignments. The method is capable of simplifying the complete tunnel system at global scale into a lower-order, practically affordable numerical approach while still retaining the ability to account for the aforementioned key features. This was achieved using a Beam-on-Nonlinear-Winkler Foundation (BNWF) approach. The (dynamic) ground-structure interaction was considered using springs and dashpots calibrated against 2D nonlinear plane strain FE analyses under quasi-static and dynamic conditions, respectively. The springs were subjected to a free-field displacement time history obtained from 1-D nonlinear wave propagation analyses. The proposed method was implemented for the case study of the circular Large Hadron Collider (LHC) tunnel network at CERN (in French: Conseil Européen pour la Recherche Nucléaire) in Geneva, Switzerland, which is 27km in circumference and runs 100m below the ground surface, where the tunnel is completely embedded within a competent layered rock. The tunnel complex contains six large underground cavern structures housing the particle detectors (similar to 'station-boxes') along its alignment. The study investigated the seismic actions developed through the circular tunnel alignment considering various ground motion schemes and examined the effect of alignment geometry on the magnitude of these forces by a comparison with a straight alignment possessing the same tunnel-ground interaction properties. The actions generated from the low-order BNWF model of the tunnel alignment were then utilised as time-varying boundary conditions in full 3D models of critical structural locations at local scale, specifically the tunnel-cavern connection where there is a change of fixity and stiffness and the largest induced seismic actions. The work was extended by using the methodology developed to comparatively study the co-seismic behaviour of common tunnel alignments that may be utilised in underground mass-transit systems, which are of a similar size to the LHC system. The effects of various degrees of curvature in plan and change in elevation and running depth were considered, assuming a ground profile, structure and interaction characteristics based on the LHC tunnel.

This thesis presents the observation of photon-induced W⁺W⁻ production using proton-proton collision data at a centre-of-mass energy of 13 TeV and with an integrated luminosity of 139 fb⁻¹, recorded by the ATLAS detector. Particular emphasis is given on data-driven corrections to pileup distributions in simulation and calculation of the signal efficiency to pass the exclusivity selection, which requires that no tracks be reconstructed within a ±1 mm region of the signal vertex. Studies into track multiplicity distributions due to multi-parton interactions are also presented, detailing corrections to leading-track transverse momentum. The γγ → WW process is then measured differentially in various kinematic distributions and an unfolding procedure is presented to remove detector effects from the data. Finally, prospects for future γγ → WW analyses at the High-Luminosity LHC are presented. The analyses are split into different categories, depending on whether selections are made on forward protons measured with the ATLAS Forward Proton detector. It is concluded that reconstructing tracks below nominal kinematic thresholds is an important area of research and development for such analyses to be fruitful in the future of LHC data-taking. A novel analysis method of matching reconstructed forward protons to vertex timing information is presented and shown to possibly improve signal to background ratios relative to an analysis in which no requirement on forward proton tags is made. Detector performance work on the inter-plane alignment of the ATLAS Forward Proton detector and forward proton reconstruction is also presented.

Results of a search for new phenomena in final states with b-jets and missing transverse momentum is presented using 139 fb-1 of proton- proton data collected at a centre-of-mass energy of 13 TeV by the ATLAS detector at the Large Hadron Collider. The search targets three models: pair-produced supersymmetric bottom squark decays into bottom quarks and a stable neutralino, dark matter particles produced through the decay of a scalar or pseudo scalar mediator in association with bottom quarks, and pair production of scalar third- generation down-type leptoquarks. Although no significant excess above standard model predictions are observed, limits set by previous searches are greatly improved upon. Bottom squark masses below 1270 GeV are excluded at 95% CL if the neutralino is assumed massless and excluded up to 660 GeV in the compressed region.

The decays B⁰ → K⁰_Sℓ⁺ℓ⁻ and B⁺ → K*⁺ℓ⁺ℓ⁻ - where ℓ⁺ℓ⁻ represents an electron-positron or muon-antimuon pair - are examples of b → sℓ⁺ℓ⁻ processes. Numerous analyses of such decays have found tensions with predictions under the Standard Model (SM) of particle physics. These tensions form a coherent pattern, which could be caused by a novel physical effect. Ratios of branching fractions for muon-mode and electron-mode b → sℓ⁺ℓ⁻ processes can be predicted with O(1%) precision under the SM, and can provide stringent experimental tests for novel physical processes which violate lepton universality. This dissertation presents new measurements of such ratios using B⁰ → K⁰_Sℓ⁺ℓ⁻ and B⁺ → K*⁺ℓ⁺ℓ⁻ decays. These measurements used data from pp collisions collected using the LHCb detector between 2011 and 2018, with a total sample size of 9 fb⁻¹.

The nuclear structure of ²⁹Mg was probed using the d(²⁸Mg,p)²⁹Mg transfer reaction to populate its single-particle states. The ISOLDE facility at CERN provided a 9.473·A beam of ²⁸Mg which was directed at a deuterated target within the ISOLDE solenoidal spectrometer. Exploiting the kinematic advantages of this technique allowed most states up to 5 MeV to be resolved, and angular distributions of the reaction cross section to be obtained. The DWBA code DWUCK5 was used to obtain spectroscopic factors for these states. Additionally, some higher-lying excited states were identified, and their possible properties were proposed. Theoretical calculations in this region broadly reproduced the observed behaviour in ²⁹Mg, as well as matching trends from other nuclides in the N=17 isotones. These calculations indicate that the nucleon-nucleon interaction between protons and neutrons, with the tensor interaction as a key component, is driving the evolution of shell structure, with the filling of the πd5/2 orbital of particular importance in this region. Finite geometry effects also play an important role in this evolution as the p orbitals approach the neutron-separation threshold, and the energy spacing between them reduces. The information gained from the nuclear structure will help to continue refining these interactions, and provide a valuable benchmark for nuclear-structure studies around the border of the island of inversion. Similar experiments to study ³⁰Al and ³¹Mg carried out recently will help to further the understanding of nuclear structure in this exotic region of the nuclear chart.

The formulation of the Standard Model of particle physics (SM) is one of the greatest scientific achievements of the 20th century. It is, however, incomplete (for example, it lacks a dark matter candidate) as well as the fact that the hierarchy problem violates naturalness arguments. This has motivated the construction of particle accelerators to probe fundamental particles at increasingly high center-of-mass energies and luminosities, the LHC at CERN being the latest to continue this legacy. This thesis covers both the enhancement of luminosity measurements of pp collisions at ATLAS, underpinning the accuracy of all measurements made by the detector, and a search for one of the most theoretically viable extensions to the SM: supersymmetry. ATLAS uses mainly event-counting algorithms to measure luminosity, which break down at higher luminosities. If the ATLAS SemiConductor Tracker (SCT) can be employed as a luminometer using hit-counting algorithms, this issue may be mitigated. It is established here that the SCT can feasibly operate as a luminometer when recording two-or-more strip clusters with the standard binary readout mode (01X). Thus, the SCT can measure the luminosity with an accuracy within 10% of two of ATLAS's existing luminometers: LUCID and TileCal. The discovery of the supersymmetric top (stop) would be fundamental for solving the hierarchy problem. An analysis of an experimentally challenging region of phase space, where stop decays have a compressed mass spectrum, complements the ATLAS one-lepton stop search using 13 TeV pp collisions at 139 fb−1 is presented. The aMT2 kinematic variable, designed to give a lower limit on pair-produced particle masses, is found to be effective at differentiating SUSY decays from the SM background, when used as an upper bound. No significant excess was observed above the Standard Model background and limits at 95% confidence level are set. Stop quarks are found to be excluded up to 500 GeV and mass splittings between the stop and the neutralino are found to be excluded up to 130 GeV, complimenting the exclusion limits found by other ATLAS searches for stop decays with one-lepton final states.

Dark matter is a poorly-understood phenomenon in nature. Though substantial evidence corroborates its existence, only few characteristics have been determined. One goal of the Large Hadron Collider (LHC) at CERN is to produce dark matter in high energy proton-proton collisions, potentially allowing insight into its currently-mysterious origins. Many models have been postulated regarding its nature, two of which are investigated in this thesis: invisible decays of the Higgs boson, and the production of semi-visible jets. The data used is from the LHC Run-2 era and recorded by the CMS experiment, corresponding to an integrated luminosity of 137 fb⁻¹ at a centre of mass energy of 13 TeV. The branching ratio of the Higgs boson to invisible states is predicted to be 0.1% in the standard model of particle physics. Enhancements from a coupling to dark matter may be observable at the LHC. A search is performed in final states comprising jets and missing transverse momentum targeting the ttH and VH Higgs boson production modes. With the full Run-2 dataset from CMS, no significant deviation from the standard model is observed. Results are presented as an upper limit on the measured cross section times branching ratio over the standard model Higgs boson cross section at the 95% confidence level. For the ttH-tagged events, observed and expected limits of 0.56 and 0.50, respectively, are achieved. For the VH-tagged events, observed and expected limits of 0.32 and 0.22, respectively, are found. A combined Run-2 limit of 0.28 is observed compared to 0.20 expected. These results are interpreted in simplified dark matter scenarios. Dark matter may exist in a Hidden Valley dark sector connected to the visible universe via a leptophobic mediator. Analogous to QCD, dark quarks may be produced at the LHC, hadronising and decaying into a mixture of visible and invisible particles: a semi-visible jet. The behaviour of simulated signal from s- and t-channel production modes in the CMS detector-and variables discriminating it from background-is explored. The transverse mass of the dijet system was found to be the most effective.

The LHCb RICH detection system at CERN has been established as an effective way to identify particles in high energy physics experiments. Future increases in collision rate and luminosity make limitations of both the particle identification (PID) algorithm and the Cherenkov radiators apparent. These two challenges are addressed in this thesis. In the first part, a convolutional neural network (CNN) approach to particle identification in LHCb RICH is introduced. Changes in the LHCb computing infrastructure towards multi-core architectures provide the opportunity to benefit from the quickly evolving field of machine learning. The calculations of the presented neural network are well suited for quasi- real-time reconstruction through parallel processing. It takes binary input images from the two RICH detectors to classify particle species. The model is hyperparameter-optimised and trained on classification accuracy with simulated collision data for the upcoming LHC opera- tion starting in 2022, reaching a PID performance comparable to the conventional algorithm. Under conditions of reduced combinatorial background, as expected from the introduction of timing resolution to the RICH detectors in future upgrades, the network achieves a particle identification performance close to 100 %, with simultaneous misclassification of the most prevalent particle species approaching 0 %. The second part of the thesis addresses the limitations of conventional Cherenkov detectors, which require large gas chambers to distinguish between particles at high energies and to reach sufficient photon yield. Photonic crystals have been shown to emit photons under specific angles similar to the conventional Cherenkov effect. The properties of this novel Cherenkov radiator offer the possibility to distinguish particles in a much wider range of energies while requiring significantly less space as a Cherenkov radiator. The work presented in this chapter introduces a framework to manufacture, simulate, and study photonic crystals under high-energy beam conditions at CERN. In particular, the manufacturing process of a one-dimensional photonic crystal, suitable for high-energy applications, is described, as well as its tests in a 200 MeV electron beam at the CERN CLEAR facility.

High energy particle colliders provide unique facilities to investigate the physics that take place at the smallest scales. The thesis present works for both future and upgraded colliders. CLIC - the Compact Linear Collider - is a proposed high energy electron-positron collider with collision energies between 350 and 3000 GeV. The LHC is a high energy proton-proton collider at CERN operating at 13 TeV. Over the next few years, it will be upgraded to the High-Luminosity LHC to operate at higher luminosities. The thesis presents an analysis of the ttH¯ production at CLIC leading to predicted precisions on top-Yukawa coupling and constraints on the CP properties of the Higgs boson. The thesis also presents work for the ATLAS experiment investigating the suitability of new sensors for use in the HL-LHC and studies of physics processes. The predicted precision on the top -Yukawa coupling using the e+e− → ttH¯ process is measured to be 2.7% at √s =1.5 TeV at CLIC with polarised beams and an integrated luminosity of 2.5 ab−1 using the CLIC SiD detector model. The H → b¯b final state of the Higgs boson and the 6-jet and 8-jet final states are used for the analysis. A multivariate selection is used to separate the ttH¯ signal from the backgrounds. b-tagging information is observed to be the strongest variable in the multivariate selection. The CP-proprties of the top -Higgs coupling are also investigated at √s =1.5 TeV using CLIC machine. A Higgs boson containing a mixture of scalar and pseudoscalar components is characterised by a mixing angle, φ. The precision of the mixing angle φ is determined to be ∆ sin2 φ ≃ 0.07 with cross-section measurements from both 6-jet and 8-jet final states. The results can be further improved up to ∆ sin2 ≃ 0.03 with an additional angular distribution, up-down asymmetry, in the semi-leptonic channel. The ATLAS experiment is upgrading its Inner Detector (ID) to an all-silicon system, the Inner Tracker (ITk). A new Front End chip, RD53A, has been developed using 65 nm CMOS technology. The thesis presents the work on the procedure of setting up the test for the RD53A chip by using testbeam at DESY with EUDET telescope. The EUTelescope software is used to reconstruct the tracks of particles and the TBMon2 framework is used to do the final analysis for the testing module. With data collected in December 2018 at DESY testbeam facility, the residual of the UK RD53A module with cell size of 50×50 µm2 is found to be symmetrical with 50.7 µm in x-axis and 51.3 µm in y-axis. The hit efficiency is determined to be 97.651%±0.035% which passes the requirement for the HL-LHC.

Beam profile and emittance monitoring is essential to understand the dynamic behaviour of the ensemble of particles in an accelerator beam. A non-invasive measurement of the beam profile ensures reliable measurements. One such device is the beam gas Ionisation Profile Monitor (IPM), which relies on detecting the ionisation products from the interaction between residual gas molecules in the vacuum of the accelerator and the beam particles. Traditional detectors in IPMs have limited the instruments reliability and performance. A novel approach using Timepix3 pixel detectors is explored in this thesis project which enables direct detection of ionisation electrons with a precise time resolution. A radiation hard readout system was developed to allow the maximum detection performance of the pixel detectors and beam measurements were recorded during 2018. A beam profile reconstruction method was developed that takes advantage of the information recorded by the Timepix3 detectors to identify and only select events that are associated with ionisation electrons. From these events, a beam profile was reconstructed and the beam size calculated by fitting a Gaussian model to the beam profile data or by calculating the RMS beam size directly. During 2018 a prototype IPM with pixel detectors was installed in the Proton Synchrotron at CERN. Beam profile measurements recorded with this instrument demonstrated the ability to measure the beam profile continuously throughout the beam cycle in the accelerator. Expected beam dynamics effects such as adiabatic damping and oscillations during transition crossing were observed with the instrument. The time resolution of the pixel detectors enabled bunch-by-bunch measurements of the beam profile by integrating the recorded events separately for each bunch over multiple turns.

A search for CP violation in baryonic four-body Λ⁰b → pπ⁻π⁺π⁻ decays is performed using the data collected by the LHCb detector during Run 1 (2011-2012) and Run 2 (2016-2017) of the LHC. The collected data corresponds to 7 fb⁻¹ of integrated luminosity and contains approximately 36 thousand signal candidates. The measurement is done using two independent methods - Triple Product Asymmetries and the energy test. The Triple Product Asymmetries method is sensitive to the P-odd type of CP violation and is based on the binned and integrated measurements of P-odd asymmetry variables. The energy test is a model independent technique, that can search for P-even and P-odd CP violation. The technique is based on a comparison of the distribution of the events in the decay phase space. Neither of the methods find a deviation from CP conservation hypothesis with a significance higher than 3 sigma. Parity violation is measured with both methods with significance higher than the 5σ observation threshold. Studies of the radiation damage of the LHCb VELO silicon-strip detector are presented. The data taken during special regular VELO performance tests, called Charge Collection Efficiency scans, is studied. Multiple VELO performance aspects are investigated, such as the bad strip detection and exclusion algorithms, Charge Collection Efficiency and Cluster Finding Efficiency. An investigation of unforeseen tracking efficiency reduction, caused by the Second Metal Layer effect, is carried out. The time dependence of the radiation damage effects of the VELO subdetector are presented. A novel procedure of beneficially annealing the VELO is performed and presented. The high voltage tests of the VELO are performed during the last runs of the LHC, testing the detector beyond its design limitations. The listed studies allowed the evolution of the radiation damage effects of the VELO sensors to be seen. In particular, approximately 2% of VELO strips are found to be bad, but the number is stable throughout the years. The required bias voltage of the sensors is seen to rise with increasing irradiation, as predicted by the Hamburg model. The Second Metal Layer effect evolution in time is quantified. The effect starts early in Run 1 with initial irradiation, but stabilizes for the rest of the VELO operation until the end of Run 2.

The STEFF spectrometer was used at the Neutron Time-of-Flight facility (n ToF) at CERN in 2016 to perform a 30-day long experimental campaign of measurements of fission fragments and gamma rays produced in 235U fission for a wide range of incident neutron energies. A pipeline for reading, correlating and database deposition of the experimental data from this experimental campaign as well as for future STEFF campaigns at n ToF has been constructed. The pipeline resulted in 70-fold data size reduction to an experimental database that can be fully processed in ≈7 hours. The collected gamma-ray data acquired using NaI and LaBr3 detectors have been analyzed in the < 1 eV neutron energy range and compared to prior STEFF 235U fission gamma-ray measurements. A method for correcting NaI signal amplitudes for n ToFspecific effects, such as rates and pulse types, based on fission gamma-ray spectrum shape has been developed. The correcting factors were the greatest for the dedicated proton pulses at neutron energies of ≈0.06 eV, increasing signal amplitude by approximately a factor of 2. Corrections to LaBr3 signals have also been considered and performed based on count rates, with the the largest correcting factors reducing signal amplitude by ≈15%. The corrected and calibrated energy spectra and calculated fold distributions have been prepared for extraction of gamma-ray multiplicity, average energy and total energy in thermal and epithermal fission of 235U. An experiment was conducted at the Lohengrin mass spectrometer at Institut LaueLangevin, France, using a FiFI spectrometer for measurement of masses and atomic numbers of selected 235U fission fragments. The details of the experiment and the data analysis are presented, and a method for calibrating Bragg detectors for atomic number extraction is proposed. The method is based on amplitudes, derivatives and risetimes of signals produced by fission fragments in isobutane fill gas. The extracted signal properties were used in conjunction with known fragment masses and energies to produce functional forms based on powers of fragment velocities and average atomic numbers. Furthermore, a comparison with simulations produced in SRIM-2013 was performed, assessing the accuracy of the simulations.

The increase in luminosity during the LHC upgrade programme causes a challenging rise in track multiplicity and hit occupancy in the LHCb detector. In order to mitigate this effect, the use of photon detector hit time information is presented in the context of the Ring-Imaging Cherenkov (RICH) detectors. The application of a time gate in the FPGA of the digital readout board for the Upgrade Ia photon detector, which is being installed for LHC Run 3, is described. Data recorded during SPS charged particle beam tests using a 6.25 ns time gate show a reduction of up to a factor of four in asynchronous detector noise compared to the original 25ns readout. A time-walk correction based on the time-over- threshold is proposed. Using the LHCb simulation framework, the intrinsic time resolution of the RICH detectors is demonstrated to be less than 10 ps. This is particularly relevant for the LHCb Upgrade II, which is scheduled for the year 2030 in preparation for a further order- of-magnitude rise in luminosity. Methods of time gating and scaling of the signal amplitude in the RICH reconstruction likelihood maximisation algorithm are presented. The results show that, considering improvements in the time-resolution only, a photon detector with an approximately 50 ps resolution can achieve today's particle ID performance in the high- luminosity LHC environment. In the second part of this thesis, the first published semiconductor tracker for cosmic-ray muon scattering tomography is presented. The tracker uses silicon strip sensors from the ATLAS Semiconductor Tracker (SCT) with an 80μm pitch. A novel electronic readout system for the sensors is designed, based on a scalable, inexpensive, flexible, FPGA-based solution. A high-precision mechanical structure with integrated cooling is built to align the SCT modules. This alignment is fine-tuned in software, and the tracker performance is compared with a Geant4 simulation. A scattering angle resolution compatible with 1.5 mrad at the 4 GeV average cosmic-ray muon energy is obtained. Data are recorded for plastic, iron and lead samples using 45000 muons. Images are reconstructed using the Angle Statistics Reconstruction algorithm, and demonstrate good contrast between low and high atomic number materials.