Tesis por compendio / [ES] La Tomografía por Emisión de Positrones (PET) es una de las técnicas más importantes en la medicina de diagnóstico actual y la más representativa en el campo de la Imagen Molecular. Esta modalidad de imagen es capaz de producir información funcional única, que permite la visualización en detalle, cuantificación y conocimiento de una variedad de enfermedades y patologías. Áreas como la oncología, neurología o la cardiología, entre otras, se han beneficiado en gran medida de esta técnica. A pesar de que un elevado número de avances han ocurrido durante el desarrollo del PET, existen otros que son de gran interés para futuras investigaciones. Uno de los principales pilares actualmente en PET, tanto en investigación como en desarrollo, es la obtención de la información del tiempo de vuelo (TOF) de los rayos gamma detectados. Cuando esto ocurre, aumenta la sensibilidad efectiva del PET, mejorando la calidad señal-ruido de las imágenes. Sin embargo, la obtención precisa de la marca temporal de los rayos gamma es un reto que requiere, además de técnicas y métodos específicos, compromisos entre coste y rendimiento. Una de las características que siempre se ve afectada es la resolución espacial. Como discutiremos, la resolución espacial está directamente relacionada con el tipo de centellador y, por lo tanto, con el coste del sistema y su complejidad. En esta tesis, motivada por los conocidos beneficios en imagen clínica de una medida precisa del tiempo y de la posición de los rayos gamma, proponemos configuraciones de detectores TOF- PET novedosos capaces de proveer de ambas características. Sugerimos el uso de lo que se conoce como métodos de "light-sharing", tanto basado en cristales monolíticos como pixelados de tamaño diferente al del fotosensor. Estas propuestas hacen que la resolución espacial sea muy alta. Sin embargo, sus capacidades temporales han sido muy poco abordadas hasta ahora. En esta tesis, a través de varios artículos revisados, pretendemos mostrar los retos encontrados en esta dirección, proponer determinadas configuraciones y, además, indagar en los límites temporales de éstas. Hemos puesto un gran énfasis en estudiar y analizar las distribuciones de la luz centellante, así como su impacto en la determinación temporal. Hasta nuestro conocimiento, este es el primer trabajo en el que se estudia la relación de la determinación temporal y la distribución de luz de centelleo, en particular usando SiPM analógicos y ASICs. Esperamos que esta tesis motive y permita otros muchos trabajos orientados en nuevos diseños, útiles para instrumentación PET, así como referencia para otros trabajos. Esta tesis esta organizada como se describe a continuación. Hay una introducción compuesta por tres capítulos donde se resumen los conocimientos sobre imagen PET, y especialmente aquellos relacionados con la técnica TOF-PET. Algunos trabajos recientes, pero aún no publicados se muestran también, con el objetivo de corroborar ciertas ideas. En la segunda parte se incluyen las cuatro contribuciones que el candidato sugiere para el compendio de artículos. / [CA] La Tomografia per Emissió de Positrons (PET) és una de les tècniques més importants en la medicina de diagnòstic actual i la més representativa en el camp de la Imatge Molecular. Esta modalitat d'imatge és capaç de produir informació funcional única, que permet la visualització en detall, quantificació i coneixement d'una varietat de malalties i patologies. Àrees com l'oncologia, neurologia o la cardiologia, entre altres, s'han beneficiat en gran manera d'aquesta tècnica. Tot i que un elevat nombre d'avanços han ocorregut durant el desenvolupament del PET, hi ha altres que són de gran interés per a futures investigacions. Un dels principals pilars actuals en PET, tant en investigació com en desenvolupament, és l'obtenció de la informació del temps de vol (TOF en anglès) dels raigs gamma detectats. Quan açò ocorre, augmenta la sensibilitat efectiva del PET, millorant la qualitat senyal-soroll de les imatges. No obstant això, l'obtenció precisa de la marca temporal dels raigs gamma és un repte que requerix, a més de tècniques i mètodes específics, compromisos entre cost i rendiment. Una de les característiques que sempre es veu afectada és la resolució espacial. Com discutirem, la resolució espacial està directament relacionada amb el tipus de centellador, i per tant, amb el cost del sistema i la seua complexitat. En aquesta tesi, motivada pels coneguts beneficis en imatge clínica d'una mesura precisa del temps i de la posició dels raigs gamma, proposem nouves configuracions de detectors TOF-PET capaços de proveir d'ambduess característiques. Suggerim l'ús del que es coneix com a mètodes de "light-sharing", tant basat en cristalls monolítics com pixelats de diferent tamany del fotosensor. Aquestes propostes fan que la resolució espacial siga molt alta. No obstant això, les seues capacitats temporals han sigut molt poc abordades fins ara. En aquesta tesi, a través de diversos articles revisats, pretenem mostrar els reptes trobats en aquesta direcció, proposar determinades configuracions i, a més, indagar en els límits temporals d'aquestes. Hem posat un gran èmfasi a estudiar i analitzar les distribucions de la llum centellejant, així com el seu impacte en la determinació temporal. Fins al nostre coneixement, aquest és el primer treball en què s'estudia la relació de la determinació temporal i la distribució de llum de centelleig, en particular utilitzant SiPM analògics i ASICs. Esperem que aquesta tesi motive i permeta molts altres treballs orientats en nous dissenys, útils per a instrumentació PET, així com referència per a altres treballs. Aquesta tesi esta organitzada com es descriu a continuació. Hi ha una introducció composta per tres capítols on es resumeixen els coneixements sobre imatge PET i, especialmente, aquells relacionats amb la tècnica TOF-PET. Alguns treballs recents, però encara no publicats es mostren també, amb l'objectiu de corroborar certes idees. La segona part de la tesi conté els quatre articles revisats que el candidat suggereix. / [EN] Positron Emission Tomography (PET) is one of the greatest tools of modern diagnostic medicine and the most representative in the field of molecular imaging. This imaging modality, is capable of providing a unique type of functional information which permits a deep visualization, quantification and understanding of a variety of diseases and pathologies. Areas like oncology, neurology, or cardiology, among others, have been well benefited by this technique. Although numerous important advances have already been achieved in PET, some other individual aspects still seem to have a great potential for further investigation. One of the main trends in modern PET research and development, is based in the extrapolation of the Time- Of-Flight (TOF) information from the gamma-ray detectors. In such case, an increase in the effective sensitivity of PET is accomplished, resulting in an improved image signal-to-noise ratio. However, the direction towards a precise decoding of the photons time arrival is a challenging task that requires, besides specific approaches and techniques, tradeoffs between cost and performance. A performance characteristic very habitually compromised in TOF-PET detector configurations is the spatial resolution. As it will be discussed, this feature is directly related to the scintillation materials and types, and consequently, with system cost and complexity. In this thesis, motivated by the well-known benefits in clinical imaging of a precise time and spatial resolution, we propose novel TOF-PET detector configurations capable of inferring both characteristics. Our suggestions are based in light sharing approaches, either using monolithic detectors or crystal arrays with different pixel-to-photosensor sizes. These approaches, make it possible to reach a precise impact position determination. However, their TOF capabilities have not yet been explored in depth. In the present thesis, through a series of peer-reviewed publications we attempt to demonstrate the challenges encountered in these kinds of configurations, propose specific approaches improving their performance and eventually reveal their limits in terms of timing. High emphasis is given in analyzing and studying the scintillation light distributions and their impact to the timing determination. To the best of our knowledge, this is one of the first works in which such detailed study of the relation between light distribution and timing capabilities is carried out, especially when using analog SiPMs and ASICs. Hopefully, this thesis will motivate and enable many other novel design concepts, useful in PET instrumentation as well as it will serve as a helpful reference for similar attempts. The present PhD thesis is organized as follows. There is an introduction part composed by three detailed sections. We attempt to summarize here some of the knowledge related to PET imaging and especially with the technique of TOF-PET. Some very recent but still unpublished results are also presented and included in this part, aiming to support statements and theories. The second part of this thesis lists the four peer-reviewed papers that the candidate is including. / This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement No 695536). It has also been supported by the Spanish Ministerio de Economía, Industria y Competitividad under Grants No. FIS2014-62341-EXP and TEC2016-79884-C2-1-R. Efthymios Lamprou has also been supported by Generalitat Valenciana under grant agreement GRISOLIAP-2018-026. / Lamprou, E. (2021). Development and Performance Evaluation of High Resolution TOF-PET Detectors Suitable for Novel PET Scanners [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/162991 / Compendio

233 U est le noyau fissile produit dans le cycle du combustible 232 T h/233 U qui a été proposé comme une alternative plus sûre et plus propre du cycle 238 U/239 P u. La connaissance précise de la section efficace de capture de neutrons de cet isotope est requise avec une haute précision pour la conception et le développement de réacteurs utilisant ce cycle du combustible. Les deux seuls jeux de données expérimentales fiables pour la section efficace de capture de l’233 U montrent des écarts important allant jusqu’à 20%. Ces différences peuvent être dues à desincertitudes systématiques associées à l'efficacité du détecteur, la correction du temps mort, la soustraction du bruit de fond et le phénomène d’empilement de signaux causé par la forteactivité α de l’échantillon. Un dispositif expérimental dédié a la mesure simultanée des sections efficaces de fission et de capture des noyaux fissiles radioactifs a été conçu, assemblé et optimiséau CENBG dans le cadre de ce travail. La mesure sera effectuée à l’installation de temps de vol de neutrons Gelina de l’IRMM, où les sections efficaces neutroniques peuvent être mesurées sur une large gamme d’énergie avec une haute résolution énergétique. Le détecteur de fission se compose d’une chambre à ionisation (CI) multi-plaque de haute efficacité. Les rayons γ produits dans les réactions de capture sont détectés par un ensemble de six scintillateurs C6 D6entourant la CI. Dans ces mesures, les rayons γ de la capture radiative sont masqués parle grand nombre de rayons γ de fission, ce qui représente le problème le plus délicat. Ces γ parasites doivent être soustraits par la détection des événements de fission avec une efficacité très bien connue (méthode de VETO). Une détermination précise de cette efficacité est assezdifficile. Dans ce travail, nous avons soigneusement étudié la méthode des neutrons prompts de fission pour la mesure de l'efficacité de la CI, apportant un éclairage nouveau sur la méthode, ce qui a permi d’obtenir une excellente précision sur l'efficacité de détection des fission d’une sourcede 252 Cf. Avec cette même source, plusieurs paramètres (pression du gaz, haute tension et la distance entre les électrodes) ont été étudiés afin de déterminer le comportement de la CI et detrouver le point de fonctionnement idéal : une bonne séparation énergétique entre les particulesα et les fragments de fission (FF) et une bonne résolution temporelle. Une bonne séparationα-FF a également été obtenue avec une cible d’233 U très radioactive. De plus, l’analyse deforme de signaux entre les rayons γ et les neutrons dans les détecteurs C6 D6 a été observée àGelina dans des conditions expérimentales réalistes. Pour conclure, le dispositif expérimentalet la méthode de VETO ont été soigneusement vérifiés et validés, ouvrant la voie à la mesure future des sections efficaces de capture et fission d’233 U . / 233U is the fissile nucleus produced in 232T h/233U fuel cycle which has been proposed as asafer and cleaner alternative to the 238U/239P u cycle. The accurate knowledge of the neutroncapture cross-section of this isotope is needed with high-precision for design and developmentof this fuel cycle. The only two reliable experimental data for the capture cross-section of233U show discrepancies up to 10%. These differences may be due to systematic uncertaintiesassociated with the detector efficiency, dead-time effects, background subtraction and signalpile-up caused by the α-activity of the sample. A special experimental set-up for simultaneousmeasurement of fission and capture cross sections of radioactive fissile nuclei was designed,assembled and optimized at CENBG in the frame of this work. The measurement will be per-formed at the Gelina neutron time-of-flight facility at IRMM, where neutron cross sectionscan be measured over a wide energy range with high energy resolution. The fission detectorconsists of a multi-plate high-efficiency ionization chamber (IC). The γ-rays produced in cap-ture reactions are detected by an array of six C6 D6 scintillators surrounding the IC. In thesemeasurements the radiative capture γ-rays are hidden in large background of fission γ-rays thatrepresents a challenging issue. The latter has then to be subtracted by detecting fission eventswith a very well known efficiency (VETO method). An accurate determination of this efficiencyis rather difficult. In this work we have thoroughly investigated the prompt-fission-neutronsmethod for the IC efficiency measurement, providing new insights on this method. Thanks tothis study the IC efficiency was determined with a very low uncertainty. Using a 252Cf source,several parameters (gas pressure, high voltage and the distance between the electrodes) havebeen studied to determine the behaviour of the IC in order to find the ideal operation point:a good energy separation between α-particles and fission fragments (FF) and a good timingresolution. A good α-FF separation has been obtained with a highly radioactive 233U target.Also, the pulse-shape discrimination between γ-rays and neutrons in the C6D6 detectors wasobserved at Gelina under realistic experimental conditions. To conclude, the experimentalset-up and the VETO method have been carefully checked and validated, opening the way tofuture measurements of the capture and fission cross sections of 233U.

A new measurement of the neutron electric dipole moment (nEDM) is being developed at TRIUMF, where a high density source of ultra cold neutrons (UCN) is currently under construction. A fast, high-efficiency UCN detector is needed for the experiment, and a 6-Li doped glass scintillation detector is being explored for this purpose. In this work, simulations and test measurements were carried out to optimize the light guide design for the new UCN detector. Acrylic and air-core light guides, the latter with two different reflecting surfaces, were considered. Three prototype light guides were constructed and tested, and results were compared with simulations. The best solution was found to be an acrylic guide, wrapped with mylar foil. For a guide 12 cm in length as required by the experimental layout, a lower limit of approximately 25 photoelectrons per neutron capture was established for the proposed geometry and photomultiplier configuration. / May 2016

First principle electronic structure theory is used to describe the effect of crystal binding on radiation detectors, electron transport properties, and structural damage induced by intense irradiation. A large database containing general electronic structure results to which data mining algorithms can be applied in the search for new functional materials, a case study is presented for scintillator detector materials. Inelastic cross sections for the generation of secondary electron cascades through impact ionization are derived from the dielectric response of an electron gas and evolved in time with Molecular Dynamics (MD). Qualitative and quantitive estimates are presented for the excitation and relaxation of a sample irradiated with Free Electron Laser pulses. A study is presented in where the structural damage on covalent bonded crystals following intense irradiation is derived from a Tight Binding approach and evolved in time with MD in where the evolution of the sample is derived from GW theory for the quasiparticle spectra and a dedicated Boltzmann transport equation for the impact ionization.

233U is the fissile nucleus produced in 232T h/233U fuel cycle which has been proposed as asafer and cleaner alternative to the 238U/239P u cycle. The accurate knowledge of the neutroncapture cross-section of this isotope is needed with high-precision for design and developmentof this fuel cycle. The only two reliable experimental data for the capture cross-section of233U show discrepancies up to 10%. These differences may be due to systematic uncertaintiesassociated with the detector efficiency, dead-time effects, background subtraction and signalpile-up caused by the α-activity of the sample. A special experimental set-up for simultaneousmeasurement of fission and capture cross sections of radioactive fissile nuclei was designed,assembled and optimized at CENBG in the frame of this work. The measurement will be per-formed at the Gelina neutron time-of-flight facility at IRMM, where neutron cross sectionscan be measured over a wide energy range with high energy resolution. The fission detectorconsists of a multi-plate high-efficiency ionization chamber (IC). The γ-rays produced in cap-ture reactions are detected by an array of six C6 D6 scintillators surrounding the IC. In thesemeasurements the radiative capture γ-rays are hidden in large background of fission γ-rays thatrepresents a challenging issue. The latter has then to be subtracted by detecting fission eventswith a very well known efficiency (VETO method). An accurate determination of this efficiencyis rather difficult. In this work we have thoroughly investigated the prompt-fission-neutronsmethod for the IC efficiency measurement, providing new insights on this method. Thanks tothis study the IC efficiency was determined with a very low uncertainty. Using a 252Cf source,several parameters (gas pressure, high voltage and the distance between the electrodes) havebeen studied to determine the behaviour of the IC in order to find the ideal operation point:a good energy separation between α-particles and fission fragments (FF) and a good timingresolution. A good α-FF separation has been obtained with a highly radioactive 233U target.Also, the pulse-shape discrimination between γ-rays and neutrons in the C6D6 detectors wasobserved at Gelina under realistic experimental conditions. To conclude, the experimentalset-up and the VETO method have been carefully checked and validated, opening the way tofuture measurements of the capture and fission cross sections of 233U.

SuperCDMS (Super Cryogenic Dark Matter Search) est une expérience de détection directe de la matière sombre qui utilise des détecteurs cryogéniques au silicium et au germanium capables de mesurer des dépôts d'énergie aussi petits que quelques eV. L'un des défis les plus importants de la recherche de la matière sombre est de réussir à bloquer les nombreux bruits de fond. Une de ces sources de bruit de fond est la production de plasmons par la diffusion Compton, un phénomène ne déposant généralement qu'une dizaine d’eV et affectant surtout les détecteurs à semi-conducteurs. Dans cet ouvrage, nous allons estimer le taux de production de ces plasmons et démontrer qu'il est primordial de tenir compte de ce bruit de fond avant de pouvoir affirmer avoir trouvé la matière sombre. Un autre enjeu majeur de SuperCDMS est l'étalonnage de ses détecteurs à très faible énergie. En effet, la matière sombre est supposée interagir avec le noyau des atomes. Toutefois, il n'existe présentement aucune donnée sur le degré d'ionisation pour des reculs nucléaires de moins de 0,7 keVnr dans le silicium, où keVnr représente l'énergie déposée en keV lors d'une collision nucléaire (nr). Dans la deuxième partie de cet ouvrage, nous allons montrer comment l'expérience IMPACT (Ionization Yield Measurement with Phonons At Cryogenic Temperature), une sous-expérience de SuperCDMS composée de membres de SuperCDMS, compte mesurer ce degré d'ionisation pour des reculs nucléaires entre 0,1 et 4 keVnr dans le silicium. Connaître ce degré d'ionisation est d'une grande importance pour SuperCDMS car il permet de reconstruire l'énergie déposée par une particule de matière sombre lorsqu'elle effectue une collision nucléaire dans ses détecteurs de matière sombre. / SuperCDMS (Super Cryogenic Dark Matter Search) is an experiment for the direct detection of dark matter that uses cryogenic silicon and germanium detectors which can measure energy deposition as low as a few eV. One of the biggest challenges in the search for dark matter is to block the various background events. One of those background source is the excitation of plasmons by Compton scattering, a phenomenon that generally deposits about ten eV and which affects mainly semiconductor detectors. In this work, we will estimate the production rate of plasmons and argue that we must understand this background in order to claim evidence of dark matter. One of SuperCDMS's other challenges is the calibration of its detectors at very low energy. Indeed, dark matter is known to interact with the atom's nucleus, but unfortunately, there is no data at the moment about the ionization yield of nuclear recoils below 0,7 keVnr in silicon, where keVnr is the energy deposited in keV in a nuclear recoil (nr). In the second part of this work, we will show how the IMPACT experiment (Ionization yield Measurement with Phonons At Cryogenic Temperature), a sub-experiment of SuperCDMS composed of members of SuperCDMS, aims to measure the ionization yield for nuclear recoils between 0,1 and 4 keVnr in silicon. It is important for SuperCDMS to know this ionization yield to reconstruct the energy deposited in their detectors by a nuclear recoil from a dark matter particle.