ED Sciences Physiques et de l'Ingénieur
Simulation tool optimisation and analysis of early calibration and reactor neutrino data for the JUNO Experiment
by Matthieu LECOCQ (Laboratoire de Physique des 2 Infinis de Bordeaux)
The defense will take place at 14h00 - Marie Curie 19 Chemin du Solarium, Batiment CREATIF, 33170 Gradignan
in front of the jury composed of
- Cécile JOLLET - Maîtresse de conférences - Université de Bordeaux - Directeur de these
- David LHUILLIER - Directeur de recherche - IRFU CEA Paris-Saclay - Rapporteur
- Davide FRANCO - Directeur de recherche - APC - Rapporteur
- Margherita BUIZZA AVANZINI - Chargée de recherche - Laboratoire Leprince Ringuet - Examinateur
- Claudia NONES - Directrice de recherche - IRFU-CEA Paris-Saclay - Examinateur
- Nadezda SMIRNOVA - Professeure des universités - LP2I Bordeaux / Université de Bordeaux - Examinateur
Nearly a century after it was first theorised, the neutrino is at the forefront of research in particle physics. The discovery of neutrino oscillations has been an important milestone in the field, as it not only demonstrated that a neutrino produced in a given flavour could be detected in a different one due to mixing, but, most importantly, established that neutrinos are massive, in direct contradiction with the Standard Model of Particle Physics. This discovery opened a rich experimental program aimed at precisely measuring the oscillation parameters and answering the remaining open questions about neutrinos. Among them, the neutrino mass ordering (NMO) — knowing which neutrino is the heaviest (or lightest) — constitutes one of the parameters needed to unlock the full oscillation picture. While recent results seem to be leaning towards a normal ordering (neutrinos ordered in mass similarly to their corresponding leptons), the latest joint effort from the NOνA and T2K experiments was unable to provide a decisive conclusion on the ordering problem. The Jiangmen Underground Neutrino Observatory (JUNO) is a next-generation, 20-kton liquid scintillator detector with an unprecedented 3% energy resolution, designed to probe the neutrino mass ordering using reactor antineutrinos, while reaching sub-percent precision on oscillation parameters, and also providing a broad multi-purpose physics program including solar, atmospheric, and supernova neutrino studies. This thesis comes at a crucial period of the experiment, as the detector was nearing completion, and the start of data taking was drawing near. This work presents several detector commissioning and analysis studies conducted using the early data collected by JUNO, with a focus on the Inverse Beta Decay (IBD) detection channel central to the experiment's primary goal. Prior to the commissioning phase, the thesis focused on optimising JUNO's Monte Carlo simulation tools, followed by their validation against calibration source data collected throughout the detector commissioning, from the dual-phase filling period to the filled detector. Using radioactive and neutron sources deployed at various positions, the light yield, spatial uniformity, and energy non-linearity of the detector are characterized and compared between data and simulation, for both the large and small photomultiplier systems. These studies identify and help correct several discrepancies between data and simulations, improving the fidelity of the detector response model that underlies all subsequent physics analyses. Building on this validated detector understanding, the central analysis of this thesis presents the IBD event selection applied to the first 60 days of JUNO data and the characterisation of the 9Li/8He cosmogenic background, produced by muon spallation in the liquid scintillator and constituting one of the most significant correlated backgrounds to the IBD signal. Two complementary background suppression strategies are presented: a spherical veto around the accompanying neutron position and a cylindrical veto along the reconstructed muon track, exploiting the spatial and temporal correlation between the parent muon and the spallation products. Finally, this thesis explores ortho-positronium (o-Ps) formation as an alternative pulse-shape discrimination technique to distinguish positron from electron events, with potential applications for cosmogenic background rejection complementary to the veto-based methods. The formation fraction and lifetime of o-Ps in the JUNO liquid scintillator are estimated using both simulation and early data, and the prospects and limitations of this technique are discussed, including the role of photomultiplier timing resolution.
Single-Particle Orientation Tracking (SPoT) of Near-Infrared-Emitting Individual Carbon Nanotubes to Probe the Brain Extracellular Space
by LiMeng RUAN (Laboratoire Photonique, Numérique & Nanosciences)
The defense will take place at 14h00 - Amphitheater Institut d'Optique d'Aquitaine Rue François Mitterrand CS30006
in front of the jury composed of
- Laurent COGNET - Directeur de recherche - CNRS - Directeur de these
- Charles KERVRANN - Directeur de recherche - INRIA - Rapporteur
- François MARQUIER - Professeur des universités - Université ENS Paris-Saclay - Rapporteur
- Sophie BRASSELET - Directrice de recherche - CNRS - Examinateur
- Stéphane GRAUBY - Professeur des universités - Université de Bordeaux - Examinateur
The brain extracellular space (ECS) is a system of interconnected regions bounded by neuronal membranes and containing the interstitial fluid together with the extracellular matrix (ECM). The latter plays a fundamental role in the molecular transport that underlies the diffusion of biological and chemical signals and intercellular communication. Understanding its characteristics is essential to advance our knowledge of brain physiology and pathology. Single-particle tracking provides access to the local diffusive behavior of molecules in living brain tissue. In particular, single-walled carbon nanotubes are highly efficient emitters in the near-infrared and serve as probes for imaging and characterizing the ECS. While their translational diffusion in the ECS is well established, their rotational diffusion remains little explored, although it can report on the local viscosity of the ECS and the organization of the ECM. In this thesis, we focused on the rotational and translational diffusion of single color-center carbon nanotubes (CCNTs), whose length can be tuned to the specific environment, in order to better understand the brain ECS. We developed a radially and azimuthally polarized (raPol) microscope operating in the short-wave infrared, inspired by single-molecule orientation localization microscopy (SMOLM). When imaging CCNTs, the dipole spread functions proved highly sensitive to optical aberrations; we estimated them using a vectorial phase-retrieval algorithm. By incorporating a Deep-SMOLM algorithm, we achieved high-precision estimation of the three-dimensional orientation and two-dimensional localization of single molecules. On this basis, we established a Single Particle Orientation Tracking (SPoT) technique, enabling the simultaneous reconstruction of the rotational and translational trajectories of individual molecules. We applied a Brownian diffusion model describing the rotational and translational dynamics of anisotropic probes. To enable a quantitative analysis of the trajectories, we introduced the mean squared displacement, the mean squared angular displacement, and the reorientation time-correlation function as key observables at the single-particle scale. We first studied the diffusion of CCNTs in water–glycerol mixtures: the analysis of viscosity-dependent Brownian motion showed excellent agreement with theory. Experiments in 2% agarose gels and in dense emulsions then revealed how the microstructure of the networks and geometric confinement induce systematic deviations from ideal Brownian diffusion. Finally, the approach was applied to mouse brain slices, allowing a quantitative characterization of CCNT diffusion in the ECS. The comparative analysis of rotational and translational diffusion revealed a marked heterogeneity in the diffusion properties of the ECS. In addition, the ECM was enzymatically degraded using hyaluronidase: this degradation significantly altered the diffusive behavior, confirming the crucial role of the matrix as a physical barrier regulating molecular diffusion within the ECS. Overall, this thesis proposes an integrated experimental and analytical framework for characterizing complex biological microstructures, using the rotational and translational diffusion of single molecules as quantitative physical observables.