Detector response and analysis

• Applicant: Professorin Dr. Kathrin Valerius
• Subject Area: Nuclear and Elementary Particle Physics, Quantum Mechanics, Relativity, Fields
• Term: since 2026
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 561595909

DELight is a new particle physics experiment to search for light dark matter (LDM) particles with masses well below 1 GeV/c2 by detecting elastic LDM scattering off a cryogenic liquid helium (LHe) target. Its low mass number, the low threshold achievable with magnetic microcalorimeters (MMCs), its intrinsic radiopurity, and the availability of several signal channels render LHe an ideal target for LDM searches. DELight can only be realized through the confluence of its six projects. Each project plays a fundamental and unique role, where this project is responsible for the calibration scheme, high-level event building, and simulation and analysis pipelines. In addition, it will conduct sensitivity studies and contribute to the development of models targeting LDM interactions. Potential LDM signals in DELight are composed of multiple signatures present in superfluid helium: phonons and rotons (collectively referred to as quasiparticles), scintillation photons, and long-lived excimers. Based on these signal channels, it will develop a simulation framework to model detector response, signal transport, timing, and waveform formation. These simulations are essential for understanding performance, informing design, and supporting background rejection strategies. A key objective of the project is to conceive and implement a calibration scheme to validate the response model and to ensure energy scale accuracy. The calibration system will utilize both electronic and nuclear recoil sources (low-energy X-ray and gamma-ray sources, as well as suitable neutron sources) in addition to heat-pulsers integrated with the LAMCALs to ensure stability and uniformity across the detector modules. The analysis pipeline begins with optimal filtering (OF) of digitized traces to extract amplitudes and identify pulse components. Events are reconstructed with estimators for total energy and signal composition. Energy calibration uses signal yields derived from dedicated measurements, while machine learning is explored for position reconstruction as an additional background suppression measure. Noise characterization across all sensor channels supports both offline OF analysis and online triggering. Cross-correlated noise studies will optimize performance and enable multi-channel reconstruction techniques. Initially simulation-driven waveform templates will be gradually replaced with data-driven models derived from calibration data, ensuring a precise and validated description of detector response. Finally, it develops a statistical framework to evaluate DELight’s sensitivity across a wide range of LDM models, including contact, long-range, and exotic interactions. Multidimensional likelihood analysis and machine learning will guide optimal observable selection and background rejection. DELight’s liquid target will uniquely enable tests of low-energy excesses reported in other detectors, shedding light on their origin and advancing the field of rare-event searches.

• DFG Programme: Research Units
• Subproject of: FOR 6006: DELight: Direct Search Experiment for Light Dark Matter with Superfluid Helium
• Co-Investigators: Dr. Klaus Eitel; Professor Dr. Felix Kahlhoefer; Professor Dr. Markus Klute