Single-Neutron Photo-Nuclear Events for LDMX HCal Calibration

Abstract

The upcoming Light Dark Matter eXperiment (LDMX) aims to push new frontiers in the search for the nature of dark matter. Its electromagnetic calorimeter (ECal) can be calibrated directly from the electron beam in the experiment, however the hadronic calorimeter (HCal) will be installed behind and to the sides of the ECal. Thus, it cannot be calibrated directly and has to be done so through other means. The HCal acts as a veto device and calibration is important for ruling out uninteresting events. Additionally, if dark matter decays into luminous matter within the detectors, accurate energy measurements of their decay products are of great interest to determining the properties of dark matter. In this thesis, we investigate the feasibility of one alternative HCal calibration technique: namely the use of single-neutron photo-nuclear events. The idea is that the incident electrons radiate a single bremsstrahlung photon which in turn kicks out a single neutron in a photo-nuclear interaction in the ECal. Naïvely, one could assume that the missing energy of the electron, the energy of the photon and the kinetic energy of the neutron correlate linearly. We simulate 10^6 such events with GEANT4 and investigate the relations between photon energy, neutron energy, detected ECal energy and detected HCal energy. We find that a fraction of these events can be selected to provide well-defined linear relations between all these quantities while maintaining sufficient statistics for aiding the calibration of the HCal.

Popular Abstract

Astronomers discovered in the 1970's that there was not enough matter in the universe to explain certain cosmological phenomena. Modern estimates suggest there would need to be about 6 times as much matter as we could see. As this mysterious extra matter was invisible to us, they named it "dark matter" and the nature of what it is, how it works and where it came from is one of the most pressing questions in modern astrophysics and particle physics. Many experiments have since been designed to search for new particles that could account for dark matter, but to this day none have managed to pin it down. The latest in line for this undertaking will be the Light Dark Matter eXperiment (LDMX) at SLAC near Stanford, USA, which is scheduled for operation in a few years. The premise of LDMX is to send electrons with a known energy towards a target where they interact to hopefully produce dark matter. The leftover energy of the electron and other by-products is measured behind the target by several particle detectors. Since the dark matter particle cannot be detected, any energy is missing from the measurements could indicate production of dark matter.

Like with most measurement devices, the particle detectors must be calibrated to translate their electrical signals to particle energies. This is usually done by sending particles of a known energy into the detectors and characterize their response. For the detectors that are not directly in line-of-sight with the particle beam, this is pretty much impossible. In this thesis, we investigate one possible way to circumvent this issue. There is a series of particle interactions in which the incident electron produces a photon, and this photon kicks out a neutron from a nucleus in the experiment, that occurs relatively frequently. One could hope that the missing energy from the electron corresponds to the energy of the photon, which in turn corresponds to the energy of the neutron. Thus, by measuring the electron energy in the first detector (which is calibrated), one can infer what energy should be measured from the neutron in the second detector. In this thesis, we simulate one million such events to see if these relations hold and how consistently the detectors measure the energies of the particles.

Our results show that there is indeed a good correspondence between missing electron energy, photon energy, neutron energy and detector responses. Additionally, we find that there may be sufficient data over a wide enough energy range to calibrate the HCal well and achieve good energy resolution. However, previous work suggest that this sentiment may not hold between different simulation models, so more work is needed to assess the accuracy of these results.

• author: Sahlberg, Daniel
• supervisor: Ruth Pöttgen; Luis Sarmiento; Lene Bryngemark
• organization: Particle and nuclear physics; Department of Physics
• course: PHYM01 20251
• year: 2025
• type: H2 - Master's Degree (Two Years)
• subject: Physics and Astronomy
• language: English
• additional info: Code used for this thesis is available at https://github.com/bryngemark/LU-student-projects.; LDMX software library is available at https://github.com/LDMX-Software/ldmx-sw/tree/trunk.
• id: 9195513
• date added to LUP: 2025-06-10 08:49:21
• date last changed: 2025-06-10 08:49:21