No Arabic abstract
Borexino, a liquid scintillator detector at LNGS, is designed for the detection of neutrinos and antineutrinos from the Sun, supernovae, nuclear reactors, and the Earth. The feeble nature of these signals requires a strong suppression of backgrounds below a few MeV. Very low intrinsic radiogenic contamination of all detector components needs to be accompanied by the efficient identification of muons and of muon-induced backgrounds. Muons produce unstable nuclei by spallation processes along their trajectory through the detector whose decays can mimic the expected signals; for isotopes with half-lives longer than a few seconds, the dead time induced by a muon-related veto becomes unacceptably long, unless its application can be restricted to a sub-volume along the muon track. Consequently, not only the identification of muons with very high efficiency but also a precise reconstruction of their tracks is of primary importance for the physics program of the experiment. The Borexino inner detector is surrounded by an outer water-Cherenkov detector that plays a fundamental role in accomplishing this task. The detector design principles and their implementation are described. The strategies adopted to identify muons are reviewed and their efficiency is evaluated. The overall muon veto efficiency is found to be 99.992% or better. Ad-hoc track reconstruction algorithms developed are presented. Their performance is tested against muon events of known direction such as those from the CNGS neutrino beam, test tracks available from a dedicated External Muon Tracker and cosmic muons whose angular distribution reflects the local overburden profile. The achieved angular resolution is 3-5 deg and the lateral resolution is 35-50 cm, depending on the impact parameter of the crossing muon. The methods implemented to efficiently tag cosmogenic neutrons are also presented.
The solar neutrino experiment Borexino, which is located in the Gran Sasso underground laboratories, is in a unique position to study muon-induced backgrounds in an organic liquid scintillator. In this study, a large sample of cosmic muons is identified and tracked by a muon veto detector external to the liquid scintillator, and by the specific light patterns observed when muons cross the scintillator volume. The yield of muon-induced neutrons is found to be Yn =(3.10+-0.11)10-4 n/({mu} (g/cm2)). The distance profile between the parent muon track and the neutron capture point has the average value {lambda} = (81.5 +- 2.7)cm. Additionally the yields of a number of cosmogenic radioisotopes are measured for 12N, 12B, 8He, 9C, 9Li, 8B, 6He, 8Li, 11Be, 10C and 11C. All results are compared with Monte Carlo simulation predictions using the Fluka and Geant4 packages. General agreement between data and simulation is observed for the cosmogenic production yields with a few exceptions, the most prominent case being 11C yield for which both codes return about 50% lower values. The predicted {mu}-n distance profile and the neutron multiplicity distribution are found to be overall consistent with data.
The direct detection of dark matter particles requires ultra-low background conditions at energies below a few tens of keV. Radioactive isotopes are produced via cosmogenic activation in detectors and other materials and those isotopes constitute a background source which has to be under control. In particular, tritium is specially relevant due to its decay properties (very low endpoint energy and long half-life) when induced in the detector medium, and because it can be generated in any material as a spallation product. Quantification of cosmogenic production of tritium is not straightforward, neither experimentally nor by calculations. In this work, a method for the calculation of production rates at sea level has been developed and applied to some of the materials typically used as targets in dark matter detectors (germanium, sodium iodide, argon and neon); it is based on a selected description of tritium production cross sections over the entire energy range of cosmic nucleons. Results have been compared to available data in the literature, either based on other calculations or from measurements. The obtained tritium production rates, ranging from a few tens to a few hundreds of nuclei per kg and per day at sea level, point to a significant contribution to the background in dark matter experiments, requiring the application of specific protocols for target material purification, material storing underground and limiting the time the detector is on surface during the building process in order to minimize the exposure to the most dangerous cosmic ray components.
Cosmogenic radio-nuclei are an important source of background for low-energy neutrino experiments. In Borexino, cosmogenic $^{11}$C decays outnumber solar $pep$ and CNO neutrino events by about ten to one. Highly efficient identification of this background is mandatory for these neutrino analyses. We present here the details of the most consolidated strategy, used throughout Borexino solar neutrino measurements. It hinges upon finding the space-time correlations between $^{11}$C decays, the preceding parent muons and the accompanying neutrons. This article describes the working principles and evaluates the performance of this Three-Fold Coincidence (TFC) technique in its two current implementations: a hard-cut and a likelihood-based approach. Both show stable performances throughout Borexino Phases II (2012-2016) and III (2016-2020) data sets, with a $^{11}$C tagging efficiency of $sim$90 % and $sim$63-66 % of the exposure surviving the tagging. We present also a novel technique that targets specifically $^{11}$C produced in high-multiplicity during major spallation events. Such $^{11}$C appear as a burst of events, whose space-time correlation can be exploited. Burst identification can be combined with the TFC to obtain about the same tagging efficiency of $sim$90 % but with a higher fraction of the exposure surviving, in the range of $sim$66-68 %.
A new experiment to quantitatively measure neutrons induced by cosmic-ray muons in selected high-Z materials is introduced. The design of the Muon-Induced Neutron Indirect Detection EXperiment, MINIDEX, and the results from its first data taking period are presented as well as future plans. Neutron production in high-Z materials is of particular interest as such materials are used for shielding in low-background experiments. The design of next-generation large-scale experiments searching for neutrinoless double beta decay or direct interactions of dark matter requires reliable Monte Carlo simulations of background induced by muon interactions. The first five months of operation already provided a valuable data set on neutron production and neutron transport in lead. A first round of comparisons between MINIDEX data and Monte Carlo predictions obtained with two GEANT4- based packages is presented. The rate of muon-induced events is overall a factor three to four higher in data than predicted by the Monte Carlo packages. In addition, the time evolution of the muon-induced signal is not well described by the simulations.
A study on cosmogenic activation in germanium was carried out to evaluate the cosmogenic background level of natural and $^{70}$Ge depleted germanium detectors. The production rates of long-lived radionuclides were calculated with Geant4 and CRY. Results were validated by comparing the simulated and experimental spectra of CDEX-1B detector. Based on the validated codes, the cosmogenic background level was predicted for further tonne-scale CDEX experiment. The suppression of cosmogenic background level could be achieved by underground germanium crystal growth and high-purity germanium detector fabrication to reach the sensitivity requirement for direct detection of dark matter. With the low cosmogenic background, new physics channels, such as solar neutrino research and neutrinoless double-beta decay experiments, were opened and the corresponding simulations and evaluations were carried out.