Μελέτη ενός ανιχνευτή MICROMEGAS κατάλληλο για μετρήσεις σχάσης και νετρονικής ροής με χρήση του κώδικα FLUKA

Abstract

Οι ανιχνευτές αερίου, βασισμένοι στην τεχνολογία Micromegas[1], χρησιμοποιούνται ευρέως σε διάφορα πειράματα ατομικής, πυρηνικής και σωματιδιακής φυσικής. Επιπλέον, έχουν ιδιαίτερα χαμηλό κόστος κατασκευής, παρουσιάζουν ανθεκτικότητα σε περιβάλλον υψηλής ακτινοβολίας, ενώ συνδυάζουν ικανότητες σκανδαλισμού και προσδιορισμού τροχιάς. Με βάση αυτές τις ιδιότητες κατασκευάστηκε στο CERN μια συστοιχία ανιχνευτών Micromegas με την τεχνολογία Microbulk [2] κλεισμένων μέσα σε ένα κυλινδρικό αλουμινένιο θάλαμο με σκοπό την ταυτόχρονη έμμεση μέτρηση νετρονίων διαφόρων ενεργειών καθώς και θραυσμάτων σχάσης. Ο συγκεκριμένος ανιχνευτής χρησιμοποιείται σε δύο διαφορετικά πειράματα. Το πρώτο πείραμα είναι το n_TOF [3,4] που πραγματοποιείται στο CERN, με σκοπό την προσδιορισμό της ροής των νετρονίων της «λευκής» (όλων των ενεργειών) δέσμης του πειράματος που παράγεται από τη διαδικασία κατακερματισμού (spallation) στόχου μολύβδου που βομβαρδίζεται από δέσμη πρωτονίων ενέργειας 1 GeV. Το δεύτερο πείραμα πραγματοποιήθηκε με την συνεργασία της ομάδας πυρηνικής φυσικής του τομέα φυσικής του Εθνικού Μετσόβιου Πολυτεχνείου και του Ε.Κ.Ε.Φ.Ε. ΔΗΜΟΚΡΙΤΟΣ [5,6] και είχε ως σκοπό τον υπολογισμό της ενεργού διατομής της σχάσης του 237Np με νετρόνια 5-8 MeV. Βασικός σκοπός της παρούσας διπλωματικής ήταν η πραγματοποίηση προχωρημένων προσομοιώσεων με χρήση του κώδικα προσομοίωσης Monte Carlo FLUKA [7] με σκοπό την θεωρητική πρόβλεψη της απόκρισης του ανιχνευτή στην ανίχνευση των προϊόντων της αντίδρασης 10Β(n,α)7Li, καθώς και των θραυσμάτων σχάσης του 237Np. Τέλος για τις ανάγκες της προσομοίωσης αναπτύχτηκε και χρησιμοποιήθηκε μια συμπληρωματική ρουτίνα στην γλώσσα προγραμματισμού gFortran με σκοπό την σωστή παραγωγή των αρχικών σωματιδίων προς προσομοίωση.

Detectors based on the Micromegas principle have already been used in several atomic, nuclear and particle physics experiments. Moreover, they have low construction cost and are resistant to high levels of radiation. Consequently, very recently a new transparent neutron detector based on the Micromegas Micro-Bulk [2] technology has been developed. The challenge was to obtain a neutron transparent detector as a monitor of a neutron beam that was big enough to avoid significant neutron background generated by neutrons scattering from the chamber structure. Thus a large detector was developed at CERN within the context of the n_TOF collaboration [3,4], that was also capable of performing neutron induced fission and (n,α) cross section measurements. In this specific study, Monte Carlo simulations were carried out using the FLUKA [7] code, as part of computational support to two different experiments. The first part of the simulations was a study carried out for the n_TOF experiment which included the simulation of energy deposited by the products of the 10Β(n,α)7Li reaction inside the active gas area of the Micromegas detector. In order to better integrate the above reaction to the FLUKA code, an external routine was developed in the FORTRAN programming language in order to create a separate input file which contained information about the particles that had to be simulated. More to the point the routine was responsible for creating a file that included the energy and the position and direction coordinates of the α and 7Li particles. The routine took into consideration the kinematics of the reaction as well as the cross section which was taken from the ENDF/B-VII.1 [8] library. Furthermore, a modified version of the FLUKA source routine was used in order to feed the information of the particles to the FLUKA code and simultaneously to simulate the energy deposition of both products inside the active gas area. The final result was an event by event histogram of energy deposition which modelled the expected spectra of the Micromegas detector. After the study was accomplished, the final results were then compared to the actual experimental data and a fair agreement between the simulation and the experiment was found. The second part of the simulations was related to the experimental measurement of the 236Np(n,f) cross section that was conducted by the Nuclear Physics group of the Physics Department of the NTUA and took place at the Tandem accelerator facility of NCSR “Demokritos” in Athens [5,6]. The purpose of the study was to once again provide the experiment with information about the expected experimental spectra as well as to aid in the specification of the detector’s limits in the lower energy part of the spectra, where the noise from the electronics cannot be decoupled from the experimental data. Moreover, this was a two scale study, since the 236Np de-excites to 233Pa with the emittance of α particles of 4.8 MeV energy. Consequently there was a need of studying both the energy deposition, in the active gas volume of the detector, of α-particles as well as the fission fragments of the above mentioned reaction. The same strategy as for the previous simulations was adopted and an external routine was implemented as input to the FLUKA code. The α-particles were isotropically produced and simulated with precise energies. The results of the α-particles study were afterwards compared to the experimental spectra with a satisfactory agreement and the values of the simulated α efficiencies agreed with the experimental ones within 2-3%. For the second scale of this study the external routine integrated experimental data of the reaction based on the published studies [9] and [10] , in order to create fission fragments in the following way. A heavy fission fragment was selected from a Gaussian distribution with mean value μ=140 and standard deviation σ=6.5. Then 1-3 neutrons were assumed to be emitted in order to calculate the mass number of the light fragment and an average total kinetic energy of 174 MeV was distributed to their mass. The atomic number Z of the fission fragments was sampled from ±5 around the mean value given in [9]. The fission fragments produced were isotropically distributed in the volume of the target and their energy deposition was scored in the active gas area of the detector. The result was, once again, an event by event energy deposition histogram that was used as reference for the calibration of the detector. In a nutshell, the results of the Monte Carlo simulations using the FLUKA code that were conducted in the frame of this Master thesis, were more than satisfactory when compared to the experimental spectra. Moreover the study that was conducted clearly contributed in the success of the experiments and in the final publication of the experimental results. Last but not least, the outcome of this study showed that simulations seem to play a significant role in the experimental area and significantly aid in better understanding the physics that is involved behind any experiment.