The JUNO experiment was proposed with the main aim of solving the problem related to the neutrino mass ordering through accurate measurements of the antineutrinos flow produced by nuclear reactors. Due to the extremely small cross-section of neutrinos, the number of expected signal events is very small, about 60 IBD events per day, and it is therefore essential to keep under control the rate of background events. This can be achieved by minimizing all the sources that contribute to the generation of spurious events and in the first place those generated by the radioactive background. For each progenitor of the natural chains (U238 and Th232), for the 40K, and for some key nuclides, such as 60Co and 210Pb, it is necessary to impose strong limits on the concentration that may be present within the materials of the detector. Given the structure of the JUNO detector, the most critical material is the liquid scintillator for which uranium and thorium concentrations below 1E-15 g/g and potassium below 1E-16 g/g are required. In this thesis, I present the work I did in this context with two main purposes. The first one is the validation of the Monte Carlo software of the JUNO experiment applied to the background simulations with the aim of verifying the radiopurity limits imposed for the materials and determining the background budget of the experiment. The second one is the implementation of a measurement technique that allows reaching the sensitivities required for the measurement of the content of uranium, thorium, and potassium in the liquid scintillator. The validation of the Monte Carlo software of the JUNO experiment (SNiPER) was performed by comparing its results with those of two other simulation codes, in particular with the software Arby, developed at the University of Milano-Bicocca. I was able to study different aspects and many critical issues of the simulation of the background and the results reported by the official tool, such as the application of the quenching factor and the shape of the radioactive β-decay spectra. The spectra of the deposited energy produced by the contaminations in the main components of the JUNO detector were computed with the Monte Carlo codes. The rate of events induced in the detector was assessed based on the imposed radiopurity limits, obtaining the expected total background event rate. The value obtained is lower than the limit set to ensure the final sensitivity of the experiment. This allowed correcting and validating the answer of the official software of the JUNO experiment and verifying the actuality of the radiopurity limits initially defined for the components of the detector. During my Ph.D. I completed the development of the new measurement system, called GeSparK that exploits the coincidence between a liquid scintillator and an HPGe detector to reduce the background of the single HPGe detector. I also worked on the development of a new delayed coincidence technique that exploits the nuclear structure of 239Np, the activation product of 238U, in order to obtain an extremely strong marker of this particular decay and significantly increase the measurement sensitivity compared to the traditional approach. The sensitivity obtained was still insufficient compared to the requests of JUNO and for this reason, it was decided to implement a series of radiochemical treatments. Different treatments have been proposed, tested, and implemented with the two aims of increasing the mass of the measurable sample and reducing the concentration of interfering nuclides. The technique developed for uranium and thorium involves a liquid-liquid extraction phase and the extraction chromatography with UTEVA and TEVA resins respectively before and after irradiation. Two measurements conducted on "blank samples” with the final procedure allowed us to achieve a sensitivity that is compatible with the limits imposed by JUNO for the liquid scintillator at the ppq level.
L'esperimento JUNO è stato proposto con l'obiettivo principale di risolvere il problema relativo all'ordinamento di massa dei neutrini attraverso misure accurate del flusso di antineutrini prodotto da reattori nucleari. A causa della sezione estremamente ridotta dei neutrini, il numero di eventi segnale previsti è molto piccolo, circa 60 eventi IBD al giorno, ed è quindi essenziale tenere sotto controllo il tasso di eventi di fondo. Ciò può essere ottenuto minimizzando tutte le sorgenti che contribuiscono alla produzione di eventi spuri e in primo luogo quelle generate dal fondo radioattivo. Per ogni capostipite delle catene naturali (U238 e Th232), per il 40K, e per alcuni nuclidi chiave, come 60Co e 210Pb, è necessario imporre forti limiti alla concentrazione che può essere presente all'interno dei materiali del rivelatore. Data la struttura del rivelatore JUNO, il materiale più critico è il liquido scintillante per il quale sono richieste concentrazioni di uranio e torio inferiori a 1E-15 g/g e potassio inferiori a 1E-16 g/g. In questa tesi, presento il lavoro che ho svolto in questo contesto con due scopi principali. Il primo è la validazione del software Monte Carlo dell'esperimento JUNO applicato alle simulazioni di fondo con l'obiettivo di verificare i limiti di radiopurezza imposti per i materiali e determinare il background-budget dell'esperimento. La seconda è l'implementazione di una tecnica di misura che permette di raggiungere le sensibilità richieste per la misura del contenuto di uranio, torio e potassio nel liquido scintillante. La validazione del software Monte Carlo dell'esperimento JUNO (SNiPER) è stata effettuata confrontandone i risultati con quelli di altri due codici di simulazione, in particolare con il software Arby, sviluppato presso l'Università degli studi di Milano-Bicocca. Ho potuto studiare diversi aspetti e molte criticità della simulazione del fondo e dei risultati riportati dallo strumento ufficiale, come l'applicazione del quenching factor e la forma degli spettri beta. Gli spettri dell'energia depositata prodotta dalle contaminazioni nei componenti principali del rivelatore JUNO sono stati calcolati con i codici Monte Carlo. Il tasso di eventi indotti nel rivelatore è stato valutato in base ai limiti di radiopurezza imposti, ottenendo il rate totale di eventi di fondo previsto. Il valore ottenuto è inferiore al limite imposto per garantire la sensibilità finale dell'esperimento. Ciò ha consentito di correggere e validare la risposta del software ufficiale dell'esperimento JUNO e di verificare l'attualità dei limiti di radiopurezza inizialmente definiti per i componenti del rivelatore. Durante il mio dottorato di ricerca ho completato lo sviluppo del nuovo sistema di misura, chiamato GeSparK che sfrutta la coincidenza tra uno scintillatore liquido e un rivelatore HPGe per ridurre il fondo del rivelatore HPGe singolo. È stata anche sviluppata una nuova tecnica di coincidenza ritardata che sfrutta la struttura nucleare del 239Np, il prodotto di attivazione del 238U, al fine di ottenere un marcatore estremamente forte di questo particolare decadimento e aumentare significativamente la sensibilità di misura rispetto all'approccio tradizionale. La sensibilità ottenuta era ancora insufficiente rispetto alle richieste di JUNO e per questo si decise di attuare una serie di trattamenti radiochimici. Diversi trattamenti sono stati proposti, testati e implementati con i due obiettivi di aumentare la massa del campione misurabile e ridurre la concentrazione di nuclidi interferenti. La tecnica sviluppata per uranio e torio prevede una fase di estrazione liquido-liquido e di estrazione cromatografia con resine UTEVA e TEVA rispettivamente prima e dopo l’irraggiamento. Due misurazioni condotte su campioni "bianchi" con la procedura finale hanno permesso di ottenere una sensibilità compatibile con i limiti imposti da JUNO per lo scintillatore liquido a livello del ppq.
(2023). Development of innovative techniques for ultra-trace elements analysis. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2023).
Development of innovative techniques for ultra-trace elements analysis
BARRESI, ANDREA
2023
Abstract
The JUNO experiment was proposed with the main aim of solving the problem related to the neutrino mass ordering through accurate measurements of the antineutrinos flow produced by nuclear reactors. Due to the extremely small cross-section of neutrinos, the number of expected signal events is very small, about 60 IBD events per day, and it is therefore essential to keep under control the rate of background events. This can be achieved by minimizing all the sources that contribute to the generation of spurious events and in the first place those generated by the radioactive background. For each progenitor of the natural chains (U238 and Th232), for the 40K, and for some key nuclides, such as 60Co and 210Pb, it is necessary to impose strong limits on the concentration that may be present within the materials of the detector. Given the structure of the JUNO detector, the most critical material is the liquid scintillator for which uranium and thorium concentrations below 1E-15 g/g and potassium below 1E-16 g/g are required. In this thesis, I present the work I did in this context with two main purposes. The first one is the validation of the Monte Carlo software of the JUNO experiment applied to the background simulations with the aim of verifying the radiopurity limits imposed for the materials and determining the background budget of the experiment. The second one is the implementation of a measurement technique that allows reaching the sensitivities required for the measurement of the content of uranium, thorium, and potassium in the liquid scintillator. The validation of the Monte Carlo software of the JUNO experiment (SNiPER) was performed by comparing its results with those of two other simulation codes, in particular with the software Arby, developed at the University of Milano-Bicocca. I was able to study different aspects and many critical issues of the simulation of the background and the results reported by the official tool, such as the application of the quenching factor and the shape of the radioactive β-decay spectra. The spectra of the deposited energy produced by the contaminations in the main components of the JUNO detector were computed with the Monte Carlo codes. The rate of events induced in the detector was assessed based on the imposed radiopurity limits, obtaining the expected total background event rate. The value obtained is lower than the limit set to ensure the final sensitivity of the experiment. This allowed correcting and validating the answer of the official software of the JUNO experiment and verifying the actuality of the radiopurity limits initially defined for the components of the detector. During my Ph.D. I completed the development of the new measurement system, called GeSparK that exploits the coincidence between a liquid scintillator and an HPGe detector to reduce the background of the single HPGe detector. I also worked on the development of a new delayed coincidence technique that exploits the nuclear structure of 239Np, the activation product of 238U, in order to obtain an extremely strong marker of this particular decay and significantly increase the measurement sensitivity compared to the traditional approach. The sensitivity obtained was still insufficient compared to the requests of JUNO and for this reason, it was decided to implement a series of radiochemical treatments. Different treatments have been proposed, tested, and implemented with the two aims of increasing the mass of the measurable sample and reducing the concentration of interfering nuclides. The technique developed for uranium and thorium involves a liquid-liquid extraction phase and the extraction chromatography with UTEVA and TEVA resins respectively before and after irradiation. Two measurements conducted on "blank samples” with the final procedure allowed us to achieve a sensitivity that is compatible with the limits imposed by JUNO for the liquid scintillator at the ppq level.File | Dimensione | Formato | |
---|---|---|---|
phd_unimib_775677.pdf
accesso aperto
Descrizione: Andrea Barresi - PhD Thesis
Tipologia di allegato:
Doctoral thesis
Dimensione
19.93 MB
Formato
Adobe PDF
|
19.93 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.