Neutron source-based event reconstruction algorithm in large liquid scintillator detectors
Akira Takenaka e1,addr1, Zhangming Chen addr2, Arran Freegard addr2, Junting Huang addr2, Jiaqi Hui addr1, Haojing Lai addr2, Rui Li addr2, Yilin Liao addr2, Jianglai Liu addr1, addr2, Yue Meng addr2, Iwan Morton-Blake addr1, Ziqian Xiang addr2, Ping Zhang addr2 · 2025
Read Neutron source-based event reconstruction algorithm in large liquid scintillator detectors by Akira Takenaka e1,addr1, Zhangming Chen addr2, Arran Freegard addr2, Junting Huang addr2, Jiaqi Hui addr1, Haojing Lai addr2, Rui Li addr2, Yilin Liao addr2, Jianglai Liu addr1, addr2, Yue Meng addr2, Iwan Morton-Blake addr1, Ziqian Xiang addr2, Ping Zhang addr2 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-19236.txt.
Introduction
Akira Takenaka and co-authors present a 2025 paper on reconstructing particle events in giant liquid scintillator detectors. These detectors, like the JUNO observatory in China, catch faint light flashes from neutrinos. Turning flashes into an event's position, energy, and particle type is the core computing task.
Their algorithm leans on neutron calibration sources rather than generic models. A likelihood method combines the timing and charge seen by each light sensor, using response tables built from neutron captures, cosmic ray byproducts, and laser runs. In JUNO simulations it located events within about four centimeters and kept energy readings uniform to under half a percent. It also rejected eighty percent of alpha-particle background while keeping the signal.
Catalog record
| CREATOR | Akira Takenaka e1,addr1, Zhangming Chen addr2, Arran Freegard addr2, Junting Huang addr2, Jiaqi Hui addr1, Haojing Lai addr2, Rui Li addr2, Yilin Liao addr2, Jianglai Liu addr1, addr2, Yue Meng addr2, Iwan Morton-Blake addr1, Ziqian Xiang addr2, Ping Zhang addr2 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 80 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-19236
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Neutron source-based event reconstruction algorithm in large liquid scintillator detectors Akira Takenaka e1,addr1, Zhangming Chen addr2, Arran Freegard addr2, Junting Huang addr2, Jiaqi Hui addr1, Haojing Lai addr2, Rui Li addr2, Yilin Liao addr2, Jianglai Liu addr1, addr2, Yue Meng addr2, Iwan Morton-Blake addr1, Ziqian Xiang addr2, Ping Zhang addr2 Abstract We developed an event reconstruction algorithm, applicable to large liquid scintillator detectors, built primarily upon neutron calibration data. We employ a likelihood method using photon detection time and charge information from individual photomultiplier tubes. Detector response tables in the likelihood function were derived from americium-carbon neutron source events, 2.2 MeV γ \gamma -ray events from cosmic-ray muon spallation neutrons, and laser calibration events. This algorithm can reconstruct the event position, energy, and also has the capability to differentiate particle types for events within the energy range of reactor neutrinos. Using the detector simulation of the Jiangmen Underground Neutrino Observatory (JUNO) experiment as a large liquid scintillator detector example, we demonstrate that the presented reconstruction algorithm has a reconstructed position accuracy within ± \pm 4 cm, and a reconstructed energy non-uniformity under 0.5% throughout the central detector volume. The vertex resolution for positron events at 1 MeV is estimated to be around 9 cm, and the energy resolution is confirmed to be comparable to that in the JUNO official publication. Furthermore, the algorithm can eliminate 80% (45%) of α \alpha -particle (fast-neutron) events while maintaining a positron event selection efficiency of approximately 99%. ∎ 1 1 1 e-mail: [email protected] Journal: Eur. Phys. J. C 1 Introduction Large liquid scintillator detectors have played a crucial role in particle physics, particularly in neutrino physics. Recent examples include KamLAND [ 1 ] , Daya-Bay [ 2 ] , Double Chooz [ 3 ] , and RENO [ 4 ] , that made precise measurements of reactor neutrino oscillations, as well as Borexino [ 5 , 6 , 7 ] , which made comprehensive solar neutrino measurements. Currently, several new-generation liquid scintillator-based neutrino detectors have begun or will begin taking data soon, such as Kamland-Zen [ 8 ] , SNO+ [ 9 ] , Jinping Neutrino Experiment [ 10 ] , and JUNO [ 11 ] . These detectors use photomultiplier tubes (PMTs), mounted on the detector support structure, to detect the scintillation and Cherenkov photons produced in the liquid scintillator region due to the energetic, charged particles generated in neutrino interactions. In this paper, we discuss a data-driven algorithm capable of reconstructing the event position, energy, and particle type of charged particles based on the photon detection times and the amount of detected photons at each photomultiplier tube. As an application example of the algorithm, we developed and evaluated the performance of the algorithm using the detector simulation of the JUNO experiment [ 12 ] ; however, the method itself can be directly applied to events observed in the actual detector. JUNO is a 20-kiloton liquid scintillator detector, currently under construction in Jiangmen, China. Its primary physics goal is to determine the neutrino mass ordering by precisely measuring the reactor neutrino energy spectrum from the reactors located roughly 52.5 km (on average) from the experimental site. The detector will be equipped with 17,612 20-inch large PMTs (LPMT, 75% photo coverage) and 25,600 3-inch small PMTs (SPMT, 3% photo coverage) oriented towards the inner detector volume, which consists of a 17.7 m radius spherical acrylic vessel containing the 20 ktons of liquid scintillator, serving as the neutrino interaction target, as shown in Figure 1 . Electron antineutrinos from the nuclear reactors interact with protons in the liquid scintillator volume via the inverse beta decay process; \overline{\nu}_{\rm e}{\rm p}\rightarrow{\rm e}^{+}{\rm n} . The positron deposits its kinetic energy in the liquid scintillator and annihilates with an electron, producing two 511 keV \gamma -rays at the annihilation point. In addition, the emitted neutron thermalizes in the liquid scintillator and is eventually captured with a typical capture time of approximately 200 \mu sec. In the liquid scintillator, neutrons are mostly (about 99% of the time) captured by hydrogen nuclei, and a 2.2 MeV \gamma -ray is emitted from the de-excitation process of the resulting deuteron. In the event of a neutron being captured on carbon ( \sim 1% of the time), one or two \gamma -rays are emitted with a total energy of 4.9 MeV. To determine the neutrino mass ordering in JUNO, key points closely related to the event reconstruction algorithm are as follows: 1. An optimized energy resolution for these outgoing particles from the neutrino interaction, through correcting the event position-dependent non-uniform response of the detector. Detector calibration and simulation studies [ 13 ] indicated that the energy scale inside the detector could exhibit more than 10% non-uniformity. This non-uniformity arises from variations in effective PMT coverage as a function of event position, as well as complex light propagation processes, such as light absorption, scattering within the detector medium, and reflection/refraction at the detector boundaries. Such non-uniform energy responses introduce additional spread in energy estimation, requiring the event reconstruction algorithm to remove these effects to achieve optimal energy resolution. 2. Detector fiducialization. One of the primary background sources in the reactor neutrino analysis in JUNO is accidental background caused by radioactive impurities contaminating the detector components. Past radioactive assessments of each detector component [ 14 ] suggested that higher background event rates are expected near the boundary of the liquid scintillator, induced by detector components outside the liquid scintillator region. Additionally, particles produced near the boundary of the liquid scintillator are more likely to escape, leading to energy leakage. To eliminate backgrounds from external radioactive impurities and ensure accurate energy measurement, it is crucial to precisely reconstruct the event position and fiducialize the detector volume for physics analysis. 3. Background removal with pulse-shape discrimination. Besides backgrounds from the ambient detector materials, \alpha -particles produced by radioactive impurities in the liquid scintillator and fast-neutrons generated by cosmic-ray muons also pose challenges in the reactor neutrino analysis. Pulse-shape information can aid in distinguishing these backgrounds. Figure 1: A schematic view of the JUNO detector, a 20-kton liquid scintillator volume, contained in a spherical acrylic vessel and viewed by PMTs mounted on the stainless steel support structure. The acrylic vessel and stainless steel structure are connected by 590 acrylic nodes (connecting bars), and the volume outside of the acrylic vessel is filled with purified water. The detector X - and Z -axes are defined in the top right corner of this figure, and the origin of the coordinates is the center of the spherical liquid scintillator volume. The asterisk mark (*) in the figure represents the event vertex position, \bm{x}=(R,\Theta,\Phi) . The relative angle between the vertex position and PMT is expressed as \theta_{\rm PMT} . The radius of the spherical liquid scintillator volume is 17.7 m, and the distance between the detector center and PMT surface plane is approximately 19.4 m. PMTs mounted on the outer surface of the stainless steel support structure help to detect cosmic-ray muons from the outside of the detector. These PMTs are not involved in the reconstruction algorithm presented in this study. The algorithm presented in this paper employs a maximum likelihood method, in which the position and energy of the particle