Speaker
Description
Highly granular silicon-tungsten calorimeters, originally developed for precision measurements at future Higgs factories, provide excellent spatial sampling of electromagnetic showers together with compact transverse shower development. These features also make them attractive for experiments searching for feebly interacting particles, where the reconstruction and separation of nearby photons can be essential.
In this contribution, we present recent studies of a highly granular SiW calorimeter concept for the LUXE New Physics with Optical Dump (NPOD) proposal. Building on previous proof-of-concept studies, we investigate the interplay between detector geometry and the reconstruction of photon energy, impact position and incident direction. We also explore machine-learning-based reconstruction methods designed to exploit the full three-dimensional information of the calorimeter showers and compare their performance with conventional reconstruction techniques. These studies provide a framework for the joint optimization of detector geometry and reconstruction algorithms for compact calorimeters in dark-sector experiments.