«Early development of a test-bed to measure fractoluminescence in scintillators & simulation of a 24Na source for the SNO+ experiment by Emilie Mony A ...»
Early development of a test-bed to measure
fractoluminescence in scintillators
simulation of a 24Na source for the SNO+
A thesis submitted to the
Department of Physics, Engineering Physics & Astronomy
in conformity with the requirements for
the degree of Master of Science
Kingston, Ontario, Canada
Copyright c Emilie Mony, 2014
This thesis consists of two parts; the ﬁrst part pertains to fractoluminescence as a potential background in crystal scintillator detectors, and the second part bears on the simulation of a Na source to be used during the liquid scintillator phase of the SNO+ experiment.
I participated in early work to develop a test-bed to study fractoluminescence in scintillators, and report here on preliminary results I obtained before I shifted my focus to SNO+. Full results obtained by the group have since been reported in PRL 111 154301 . This project follows the discovery that mechanical stress on a dark matter detector’s crystals was causing a background signal. The response of inorganic crystal scintillators (Bi4 Ge3 O12, ZnWO4, CdWO4 ) compressed to the point of rupture was studied. The double cleavage drilled compression geometry was used to create controlled cracks in 20×5×3 mm3 samples. A correlation between a sudden drop of the force, a burst of photonic and of acoustic emissions was discovered and a lower bound was set on the conversion eﬃciency from strain energy to light energy.
SNO+ is a large underground experiment that aims primarily to search for neutrinoless double beta decay. The SNO+ detector consists of an acrylic vessel of liquid scintillator surrounded by light detectors. A tagged Na source was proposed as one of several radioactive sources to be deployed within the vessel to calibrate the i detector. To achieve this an activated NaI(Tl) crystal would be coupled to a photomultiplier tube and lowered into the center of the vessel. The second half of this thesis explores options for implementing this plan and presents the detector response to a Na source as simulated by the Monte Carlo software developed by SNO+. The size of the crystal inﬂuences the type of information that can be gleaned from using this source so four diﬀerent crystal sizes are presented for comparison. The simulations show that the source can be used to test the linearity of the energy scale and the simulation’s quenching model.
ii Acknowledgments Project 1: Fractoluminescence I would like to express my gratitude to my supervisor, Philippe Di Stefano, for guiding me through this project and thesis carefully and patiently. I’m also grateful to Wolfgang Rau for his guiding questions and suggestions. Marc-Antoine Verdier, Carlos Martinez, and Oleg Kamaev were very supportive and generous with their time and I thank them sincerely for that.
I want to thank the other students who worked on this project; Yuan Wei who designed several improvements to the hardware setup and Alexis Tantot with whom I had the pleasure of working during the last stages of my data collection and analysis and who is now developing this project in interesting new directions.
I owe thanks to the scientists from LPMCN and ENS Lyon who helped collect my ﬁrst data set: Lo¨ Vanel, St´phane Santucci, Osvanny Ramos and Sergio Ciliberto.
ıc e Thank you to Gary Contant for fabricating various iterations of the sample holder and to Steve Gillen for making the integrating ampliﬁer.
I also want to thank the CDMS graduate students who warmly welcomed me to Queen’s and who were all very fond of seedy hummus.
iii Project 2: Calibration Source for SNO+
I would like to express my gratitude to my supervisor, Mark Chen, for encouraging me and for helping me make sense of my results by explaining physics processes with clarity and joy. I’m also grateful to Alex Wright for his excellent recommendations;
this thesis wouldn’t be half as good if I hadn’t followed up on his suggestions.
I’d like to extend my gratitude to Nasim Fatemighomi who was extremely helpful, especially learning how to use RAT. Thank you also to Szymon Manecki, who was instrumental in moving the plan for the source forward.
It was a privilege to be part of the SNO+ collaboration and I’d like to thank Logan Sibley and Phil Jones in particular for kindly and promptly answering the many questions I emailed them. It was also an honour to receive guidance from Art McDonald before his retirement and it was a pleasure to work alongside the other SNO+ graduate students at Queen’s; Matt, Satoko, Maryam, and Erin.
This project built on the work started by Hugh Evans for the SNO experiment, so I am very much obliged to him for not having to start from scratch.
Lastly, I want to thank my friends and family for their invaluable love and support.
Project 1: Fractoluminescence The author’s supervisor, Philippe Di Stefano, along with collaborators from LPMCN and ENS Lyon, conceived of the project, developed an early version of the setup and performed a fracture test before the author was involved in the project. The geometry of the crystals (DCDC) was also decided on before the author joined the project.
The author was involved in the discussions pertaining to the dimension of the crystals and to improvements to the setup. Drawings of parts to be fabricated for the setup were not the work of the author, but rather of Yuan Wei, mainly, who worked on this project as an undergraduate student for his ﬁnal project. All the data presented in this thesis were collected by the author with Philippe Di Stefano and either with the aforementioned French collaborators or with Alexis Tantot, who took on this project for his PhD dissertation. Analysis of the data presented here is the sole work of the author.
v Project 2: Calibration Source for SNO+
The geometry of the sources were modelled by the author, based on an existing NaI source designed by Prof. Hugh Evans and on Ortec assemblies. The activation of a NaI source at Queen’s was undertaken by the author and the subsequent calculations of the activity of Na for diﬀerent crystal sizes and of other activated nuclei within the source was also the work of the author. The simulations of the decays within the sources inside the SNO+ detector were done by the author using the software developed by the SNO+ collaboration (RAT). Analysis of the simulated data is the sole work of the author.
16.1 Annotated Plot of the Energy Distribution in LAB.......... 109
16.2 Distributions of the Energy Deposited in the LAB........... 111
16.3 Comparison of the Four Sources from their Energy Distributions... 112
16.4 Distribution of the number of PMTs hit................ 113
16.5 2D Histograms of Energy Deposited in LAB and Energy Deposited in NAI.................................... 115
16.6 Diagram of the 2D Histograms used to Determine Cuts........ 116
16.7 Cut Nhit distributions.......................... 118
16.8 Energy Resolution Before and After Cuts................ 120
16.9 Centroids Before and After Cuts..................... 121
16.10Nhits/MeV Without the Source..................... 123
16.11Resolution With and Without the Source................ 125
16.12Nhits/MeV With and Without the Source............... 126
Early development of a test-bed to measure fractoluminescence in scintillators Chapter 1 Introduction Background discrimination is a primary challenge for rare event searches like the direct detection of dark matter. It is therefore important to understand all possible sources of background signals. The ﬁrst experiment described in this thesis was envisioned following the discovery that fractures in dark matter detectors could produce unwanted signals . To produce similar results in a controlled manner we compressed inorganic crystal scintillators in the double cleavage drilled compression (DCDC) geometry by means of a linear actuator so as to produce cracks in the samples which propagated in a well deﬁned manner. The samples were made of bismuth germanate (Bi4 Ge3 O12 ), commonly abbreviated to BGO, cadmium tungstate (CdWO4 ), and zinc tungstate (ZnWO4 ). Each sample was 20 mm long, 5 mm wide and 3 mm thick and each had a 1.5 mm diameter hole drilled through its center. The result of these brittle fractures is the emission of light and sound. A fraction of the light emitted is captured by a photomultiplier tube (PMT) and the energy released in the form of light is estimated by calibrations performed using radioactive sources. Two piezo electric sensors are placed on either side of the sample and measure acoustic phonons
CHAPTER 1. INTRODUCTION 3and a force gauge set between the linear actuator and the sample allows the force applied to the sample to be known. A correlation in time and in amplitude was observed between a drop in the force, bursts of light and of acoustic emissions.
The main objective of this experiment is to help identify a fracture signal in scintillator detectors used for rare event searches, therefore I will give a brief overview of dark matter focussing on dark matter detectors and their results. I will then present certain elements of fracture mechanics related to deformation, cracking, and the DCDC geometry. We calibrated the photon signal by making an energy spectrum of the scintillation light produced by four diﬀerent radioactive gamma sources (137 Cs, Na, 57 Co, 241 Am), so I will summarize the scintillation process in crystal scintillators and will discuss some aspects of gamma spectroscopy. I will then describe the setup and how it has evolved. I will subsequently describe the experimental procedure involved in the fracture of each crystal. Finally, I will discuss the results of this study.
Chapter 2 Dark Matter
2.1 Dark Matter Detection Evidence for dark matter was ﬁrst discovered in the 1930s when Zwicky measured the velocity dispersion of galaxies within the Coma galaxy cluster and found that the galaxies were moving too rapidly for the cluster to remain bound, considering the amount of mass within the cluster deduced from the luminous matter. The simplest explanation is that there is matter that is invisible, i.e. that doesn’t interact with the electromagnetic force, but that does interact gravitationally. Evidence for dark matter has continued to come to light: a similar phenomenon as the one Zwicky discovered in the Coma cluster is also observed in rotation curves of spiral galaxies which show that stars at suﬃciently large distances from their galactic centre are moving too quickly given the amount of visible mass in the galaxy interior to them.
Gravitational lensing, i.e. the bending of light due to a massive object gravitationally bending the path light travels, has been observed where the mass necessary to create the lensing eﬀect isn’t visible. Fluctuations (or anisotropies) in the cosmic microwave CHAPTER 2. DARK MATTER 5 background are also nicely explained when dark matter is included in models of the early Universe.
Recent measurements by the Planck collaboration indicate that only 4.9% of the mass-energy content of the Universe, as it is now, is made of baryonic matter, while dark matter accounts for 26.8% and dark energy constitutes the remaining 68.3% .
Very little is known about the nature of dark matter, but currently the most popular candidate is the weakly interacting massive particle (WIMP) which stems from an extension of the standard model. WIMPs would interact only via the gravitational force, the weak force, and possibly new interactions and would be relatively massive particles (m 1 GeV/c2 ).
There are several types of experiments which are attempting to detect WIMPS either by producing them in a particle collider (e.g. Large Hadron Collider), by observing their annihilation or decay products (e.g. Alpha Magnetic Spectrometer on the International Space Station), or by detecting their weak interaction with matter (direct detection experiments). A WIMP would interact with matter by scattering oﬀ a nucleus, causing that nucleus to recoil. Experiments that aim to detect the small amount of energy imparted to the target nuclei do so by measuring the heat signal (phonons), the scintillation light signal (photons), the ionization signal (current), or a combination of these signals coming from their detectors.
Direct detection experiments are often located underground to be shielded from cosmic rays. The underground laboratories that house them include SNOLAB in Sudbury Ontario, the Grand Sasso National Laboratory in Italy and the Soudan Underground Laboratory in Minnesota. These experiments can be grouped into those that use crystals as their target material, such as CDMS (Ge, and previously Si), CHAPTER 2. DARK MATTER 6 CRESST (CaWO4 ), CoGent (Ge), DAMA (NaI(Tl)) and EDELWEISS (Ge), those that use noble liquid and/or gases as their target such as XENON and ZEPLIN, which both use Xenon, and those that use bubble chamber technology such as COUPP and SIMPLE which are superheated droplet detectors. Results from all of these direct detection experiments are shown in Fig. 2.1.
Figure 2.1: Results from various dark matter experiments as of 2012 .
Exclusion limits (90% conﬁdence level) on spin-independent WIMP-nucleon cross sections and detection claims (2σ) are shown.
This plot shows the spin-independent WIMP-nucleon cross section as a function of WIMP mass. The closed curves indicate the regions where experiments (DAMA, CoGent, and CRESST) claim to have seen a signal or an excess of events that could be interpreted as a signal. The open curves show the upper limit on the WIMP-nucleon cross section. This means the experimenters have ruled out cross sections by failing to CHAPTER 2. DARK MATTER 7 conclusively see a WIMP-nucleon interaction in the region above the curve, and this also means the detector of the given experiment is insensitive to regions below the curve. There is therefore a discrepancy between many of the limits and the claimed signal regions. As a result dark matter has yet to be conclusively detected.