Abstract: The DarkSide collaboration aims for a dark matter direct detection using Argon-based dual-phase Time
Projection Chambers (TPCs), capable of observing both the scintillation and the ionization signals of
recoiling nuclei. The DarkSide “zero-background” strategy exploits the pulse shape discrimination
intrinsic of liquid argon, which permits to separate unambiguously nuclear recoils, the class of events
possibly induced by dark matter interactions, from electron recoil events induced by natural radioactivity.
Building on the compelling science results of the DarkSide-50 prototype detector, in operation at LNGS
since 2013, the 20 tonnes of fiducial mass experiment DarkSide-20k will push the sensitivity of dark
matter searches close to the ultimate experimental limit, the neutrino floor.
Two key technological breakthroughs achieved in these years by the DarkSide Collaboration were crucial
to convince researchers from all other liquid argon dark matter programs (DEAP-3600 and MiniCLEAN
at SNOLab, ArDM) to join the DarkSide-20k experiment and to form the “Global Argon Dark Matter”
Collaboration. These breakthroughs are the extraction of low-radioactivity argon from underground
sources and the development of Silicon Photomultiplier (SiPM) based replacements for cryogenic PMTs,
which is the central matter of my research.
I will present DarkSide-50 latest physics results. I will then detail the R&D activity that allowed to
assemble the first Photo-Detector Module (PDM) prototype for DarkSide-20k. I will provide a detailed
description of SiPMs’ performances at cryogenic temperatures, outcome of several characterization
campaigns. In particular I will show pulse features, as well as dark count rate and correlated noises and
compare them with the set of tight requirements compiled for the detector. I will describe the design
process of a ultra low-noise, high bandwidth pre-amplifier capable of operation in liquid argon. The
resulting amplifier allows the readout of a 10 × 10 mm2 SiPM with a signal to noise ratio (SNR) in
excess of 30 and a timing resolution better than 1 nanosecond at liquid argon temperature. Starting from
these results, I will show the development of a SiPM connection scheme to read an active surface of 25
cm2 as a single channel, while fulfilling the challenging SNR and timing resolution limits. I will finally
discuss the PDM performance requirements set for DarkSide-20k. In particular I will explain the SNR
constraints, the dark count rate and correlated noise impact on the physics reach of the detector and the
importance of the timing resolution for the pulse shape discrimination.