PhD Day 2026
Amphi Dirac
IP2I Lyon
-
-
09:00
→
09:25
Coffee 🥐☕️ 25m
Coffee & croissants at the entrance of amphi Dirac
-
09:30
→
10:00
Detector R&D for Future High-Energy Colliders: Instrumentation and Deep Learning Developments for the Timing-SDHCAL 30m
As a brief introduction, the Semi Digital Hadronic Calorimeter (SDHCAL) is a concept of a gaseous hadronic calorimeter based on Resistive Plate Chamber (RPC) technology, originally designed for the future linear collider ILC. In the context of future circular colliders such as FCC and CEPC, an upgraded version - the Timing-SDHCAL (T-SDHCAL) - is currently under development. It incorporates precise timing capabilities provided by Multi-Gap RPC (MRPC) technology.
During this talk, I will give an overview of my PhD research. The first part focuses on the design, construction, and characterization of several 50cm × 30cm MRPC prototypes. The second part explores the use of deep learning techniques to improve the reconstruction of the energy of incoming particles.Orateur: William VAGINAY (CNRS - IP2I) -
10:00
→
10:30
Non-perturbative aspects of supersymmetric gauge theories 30m
Quantum field theories are currently our best framework to modelize fundamental interactions. While perturbative approaches in most theories are successful, non-perturbative effects need to be included to understand these theories at all scales, however this is technically very difficult. Gauge theories engineered from string theory are a rich class of examples in which powerful methods are available to compute such exact results.
Concretely, I will present the full partition function of 6-dimensional supersymmetric gauge theories on a 4-sphere, focusing on non-perturbative contributions. The latter are expected to capture non-trivial effects in lower dimensional dual theories which are very hard to compute with other means. An extension on how different boundary conditions for the gauge theory degrees of freedom are realized on the dual side is also explored.Orateur: Timothé ALEZRAA -
10:30
→
11:00
Preliminary measurement of the germanium ionisation yield and first studies of the new silicon crystal detectors of the Ricochet experiment 30m
Coherent elastic neutrino nucleus scattering (CEvNS) was first measured experimentally by the COHERENT experiment in 2017 and is currently being studied by many experiments all around the world. In this context, the Ricochet international collaboration aims to detect the CEvNS process in order to search for new physics. The detectors used are germanium crystals operated at cryogenic temperatures, which can simultaneously readout the ionization and heat energies resulting from particle interactions occurring in the crystal.This dual measurement allows both particle identification, by discriminating between electronic and nuclear recoils, and determination of the recoil energy of the interactions. This presentation focuses on a preliminary study dedicated to measure
the ionization yield of nuclear recoils in germanium detectors at the keV energy scale, based on the first commissioning data from the Ricochet experiment. Furthermore, as the Ricochet experiment is planning to use silicon as a new crystal material, this presentation is also covering the first hardware developments and preliminary performance analysis of the silicon detectors.Orateur: Tatiana LE BELLEC (IP2I) -
11:15
→
11:45
Study of the influence of radiolysis of an aqueous solution at the interface with 316L stainless-steel 30m
The objective of my thesis is to investigate the effect of charged particle induced-radiolysis on the evolution of the passive film formed at the 316L stainless steel/electrolyte interface. This thin passive film (few nanometers thick) exhibits a duplex structure: an inner dense layer rich in chromium oxide, which ensures corrosion protection, and an outer amorphous layer of iron hydroxide containing bound water.
For this purpose, several experimental cells were developed and installed on multiple accelerator beamlines delivering 3 to 30 MeV protons and 10 MeV pulsed electrons. In situ electrochemical measurements are used to monitor the evolution of the redox potential and the properties of the passive film. The elemental analysis of this film, including H, Fe, Cr, and Ni, is complex due to its very small thickness and the fast kinetics of water radiolysis at the solid–liquid interface. The solid interface is characterized using XPS and IBA techniques.Orateur: brice lathuiliere -
11:45
→
12:15
Searches for CP violation in the top-quark sector with CMS at the LHC, CPT violation phenomenology, and Tracker Endcap metrology for the HL-LHC upgrade. 30m
My thesis investigates new sources of CP violation in the top-quark sector within the CMS collaboration at the LHC, addressing the matter–antimatter asymmetry. Exploiting the top quark’s large mass as a probe of beyond-the-Standard-Model physics, the analysis uses Run 2 data at 13 TeV. The single top t-channel process is used to probe CPV within an
Effective Field Theory framework. A complementary phenomenology study explores CPT violation through reinterpretation of LHC data.
In parallel, I contribute to the CMS Phase-2 Tracker upgrade for the High-Luminosity LHC.
My work involves metrological studies of the endcap “Dees,” mechanical structures with two-phase CO₂ cooling, ensuring they meet the required precision for qualification.Orateur: Enzo FILLAUDEAU (IP2I, Lyon) -
14:15
→
14:45
Search for CP-violation in ttH multilepton process 30m
The top quark (t) is unique among the other quarks for a variety of reasons, among which its high coupling with the Higgs boson (H). This high coupling constant is suspected to be possibly tied to Beyond Standard Model physics, such as the possibilty of a break in CP-simmetry. My thesis has the objective of investigating this hypothesis by studying the ttH process within the CMS collaboration at the LHC. The analysis utilizes ttH processes with multileptonic decay, up to Run 3 at 13.6 TeV. For this purpose, multiple Boosted Decision Trees were tested, utilizing various combinations of CP-sensitive variables. In parallel, I collaborate with the CNRS for the update of CMSSW in preparation for the CMS upgrade for the High-Luminosity LHC.
Orateur: Giorgio Mauceri -
14:45
→
15:15
Cosmological implication of the joint observation of Gravitational Waves and Type Ia Supernovae 30m
The expansion of the Universe is well established, yet the current value of the Hubble constant, H0,
remains a subject of ongoing debate. Direct measurements based on Type Ia supernovae and Cepheid variables yield values significantly higher than those inferred from early-Universe observations of the cosmic microwave background, dating back approximately 13 billion years.An independent approach to measuring H0 relies on gravitational-wave (GW) signals from kilonova events, which can simultaneously provide both distance and redshift—two key observables for determining the Hubble constant. However, only a single such event has been detected so far, resulting in a measurement with large uncertainties.
In this thesis, we propose a novel method that can be applied to any gravitational-wave event. It is based on a statistical matching between the matter density distribution inferred from gravitational-wave observations and that obtained from supernovae and galaxy surveys.
In this presentation, I will outline the conceptual framework of this method and present the preliminary work required for its implementation on both the gravitational-wave and Type Ia supernova data.
Orateur: Léo Zimmermann (Onde Gravitationelle, Cosmologie)According to the standard model of cosmology, the Universe is expanding. First observed through galaxy redshifts, this expansion was later confirmed by Type Ia supernovae (SNIa) and the Cosmic Microwave Background (CMB), leading to the so-called Hubble tension.
This thesis aims to provide an alternative perspective on the inference of cosmic expansion. To this end, we use the density fields traced by SNIa and gravitational waves (GW) from Black Holes mergers, under the assumption that both probes are correlated to the same underlying matter distribution.
The main limitations of this approach lie, on the one hand, in the limited number and precision of GW detections, and on the other hand, in the intrinsic scatter of SNIa. To address these issues, I first investigated the impact of noise on GW observations and contributed to the detection of more than 100 new GW signals in LIGO-Virgo-KAGRA data from 2024-2025. In parallel, to reduce the intrinsic scatter of SNIa, I implemented an autoencoder to capture the subtle diversity in their spectra. -
15:15
→
15:45
Enhancing MBTA Efficiency and Template Banks Generation for Sub-Solar-Mass Black Hole Searches 30mOrateur: Gaspard Joubert
The first direct observation of gravitational waves (GW) in 2015 by the LIGO/Virgo/Kagra (LVK) collaboration opened a new era of astronomy. The GW group at IP2I in Lyon contributes to both the analysis of interferometers data and the characterization of the VIRGO detector. Most gravitational wave online search pipelines rely on matched filtering (MF), a method which compares the data to a large set of theoretical waveforms, called templates. The Multi-Band Template Analysis (MBTA) is one of those MF-based pipelines. It filters data across several frequency bands in parallel, thus cutting down the computational overhead induced by MF.This algorithm, on the other hand, relies on more complex sets of template banks, covering the phase space over different frequency bands. My work focuses on simplifying and accelerating template bank production for MBTA, and on exploring strategies to improve MBTA’s efficiency in detecting sub-solar-mass black hole binaries.
-
16:00
→
16:30
H → γγ search at low mass with CMS & Electroweak precision test of BSM theories for future colliders 30m
The first part of my thesis is dedicated to the search for a low-mass resonance around 95 GeV in the di-photon channel with the CMS experiment. In addition to the search, I will also present an update of the S-method for the extraction of the photon energy scale calibration with Z → µ+µ−γ events.
The second part is focused on the electroweak (EW) precision test of specific
BSM theories for future colliders. To begin, I will introduce the concept of electroweak radiative corrections and present the results I obtained in the case of an SU(5) asymptotic Grand Unified Theory. By this means, I will able to introduce a Python package that I created to handle specific integrals that emerge from loop calculations. I will then switch to two other BSM models in which I started to perform EW precision tests, one with Vector Like Quarks and one with a composite Higgs.Orateur: Christian Verollet (IP2I Lyon France) -
16:30
→
17:00
Nuclear structure: from mean-field to beyond mean-field calculations 30m
Nuclear structure calculations aim to describe elementary properties of atomic nuclei. Solving the Schrödinger equation for such a system is, however, computationally expensive, which prevents us from performing exact calculations. Several assumptions can be made in order to make numerical computations tractable. I will discuss the self-consistent mean-field model that has become a cornerstone in the description of nuclei. Within this approach, one is able to compute various observables with a reasonable precision. Nonetheless, it is known that this precision could be further improved and that the mean-field approximation lacks the ability to compute certain observables. Going beyond the mean-field is not straightforward since the currently existing effective interactions can not be used without ambiguity in such models. In the remainder of the talk I will present our attempt to build a new interaction suited for both mean-field and beyond mean-field calculations.
Orateur: Valentin Guillon (IP2I, Université Claude Bernard Lyon 1) -
17:00
→
17:30
CANCELLED : Linking ion interaction processes to biological effects in hadrontherapy with the NanOx biophysical model 30m
Accurate biological dose calculation in hadrontherapy remains a major challenge due to the complexity of inelastic processes governing energy deposition at the microscopic scale. My thesis project aims to improve biological dose modeling by establishing a direct link between ion-matter interactions and the predictions of the NanOx biophysical model. The work involves calculating ionization, excitation, and charge transfer cross sections of ions for biological targets, and assessing their impact on chemical yields and cell survival fractions. Correlations between these cross sections and nanodosimetric quantities are being investigated to identify the dominant mechanisms driving biological effects. These results are expected to provide insight into the role of fundamental interaction processes in biological damage. Based on this, a simplified version of NanOx will be developed, aiming to reduce computational cost and facilitate its integration into clinical treatment planning systems.
Orateur: Camila STRUBBIA MANGIARELLI (IP2I)
-
09:00
→
09:25