Speaker
Description
COSINUS (Cryogenic Observatory for Signals seen in Next‑generation Underground Searches) develops cryogenic scintillating calorimeters with NaI target crystals to provide a model‑independent cross‑check of the DAMA/LIBRA result. Because hygroscopic NaI cannot withstand direct deposition of a tungsten transition‑edge sensor (W‑TES), COSINUS employs the remoTES design, in which the TES is thermally coupled to the crystal via a metallic phonon collector and a gold wire. The resulting phonon transport, and hence the pulse shape, is well described by a three‑node thermal model. In this poster we present a Bayesian pulse-shape inference framework using BAT.jl that we validated on several COSINUS remoTES prototypes. By fitting detector pulse shapes, the framework yields posteriors for physically meaningful parameters, such as thermal responses and collection efficiencies, that quantify phonon propagation and hence detector response. We report pulse‑shape study results for these prototypes and show how the extracted parameters provide guidance for detector R&D and deepen our understanding of the detector physics. The methodology is broadly applicable to other cryogenic rare‑event searches. In particular, we outline plans to adapt the framework to CRESST (Cryogenic Rare Event Search with Superconducting Thermometers) detectors; this application is prospective and will be pursued in forthcoming work. We also discuss a path to scaling the inference with machine‑learning surrogates, enabling fast, event‑by‑event parameter estimation on full detector datasets.
| Main Contribution topic | Direct detection |
|---|---|
| Secondary contribution topic | Theory / Phenomenology |