Dark Matter Search with Cryogenic Phonon-Light Detectors

The CRESST (Cryogenic Rare Event Search with Superconducting
Thermometers) Dark Matter search is aiming to directly detect
Weakly Interacting Massive Particles (WIMPs) via their elastic
scattering off nuclei in a target. Due to the low event rate
expected for WIMP-nucleus scattering the suppression of background
which would hide or mimic the WIMP signal is of crucial importance.
Moreover, since the energy transferred to a nucleus in a
WIMP-nucleus elastic scattering is extremely low (a few tens of
keV), the energy threshold and the sensitivity of detectors are
additional fundamental issues in a Dark Matter search. CRESST
detectors consist of a 300g CaWO4 scintillating crystal operated as
a cryogenic calorimeter in close proximity to a second much smaller
cryogenic calorimeter used to detect the scintillation light
produced in the target crystal. The lower light yield of nuclear
recoils, caused by neutrons and WIMPs, with respect to electron
recoils resulting from alpha, beta and gamma interactions is used
to identify the event in the scintillating absorber. In the second
phase of the CRESST experiment, active background suppression is
achieved by the simultaneous measurement of a phonon and a light
signal from a scintillating cryogenic calorimeter. Passive
background suppression is achieved by operating CRESST detectors in
a low background facility located in a deep underground site. About
1% of the energy deposited in CaWO4 by beta or gamma interactions
can be detected as scintillation light; therefore the sensitivity
of light detectors is a fundamental issue for the discrimination of
electron recoils from nuclear recoils at energies relevant for WIMP
searches. This work reports on the development of extremely
sensitive cryogenic calorimeters for the detection of the
scintillation light. These detectors applied a new thermometer
geometry characterized by phonon collectors and a thin film thermal
coupling to the heat sink. This concept allows a high sensitivity
by decoupling the area required for the collection of non-thermal
phonons and the heat capacity of the sensor and permits to make the
thermal relaxation time of the thermometer long enough to allow for
the integration of the scintillation light, despite the slow light
emission of CaWO4 at low temperature. Results on new materials to
be used as absorbers for a new generation of light detectors are
presented. First competitive limits on WIMP Dark Matter established
by the CRESST experiment running scintillating CaWO4 cryogenic
detectors in association with the light detectors developed in the
course of this work are reported together with main results of the
detector prototyping phase.