Timing characterization for T2K ND280Upgrade SuperFGD detector XXI International Workshop on Neutrino Telescopes Gioele Reina, for the T2K Collaboration. JGU Mainz, contact:
[email protected] References: [1] K. Abe et al., The T2K experiment, arXiv:1106.1238; [2] K. Abe et al., T2K ND280Upgrade: technical design report, arXiv:1901.03750; [3] S. Abe et al., Introducing a Markov Chain-Based Time Calibration Procedure for Multi-Channel Particle Detectors: Application to the SuperFGD and ToF Detectors of the T2K Experiment, arXiv:2508.07846; [4] K.L. Chung, Markov Chains, doi:10.1007/978-3-642-62015-7; [5] C.M. Sha and N.V. Dokholyan, Simple exponential acceleration of the power iteration algorithm, arXiv:2109.10884 The T2K experiment and its Near Detector Upgrade The Super Fine Grained Detector Time Calibration Algorithm Application to the Super Fine Grained Detector Results and Performances Tokai 2 Kamioka T2K is a long-baseline neutrino oscillation experiment located in Japan[1]. Accelerator νμ and anti-νμ are produced at the J-PARC proton accelerator facility, and are then detected by a set of near detectors and the Super-Kamiokande far detector. It measures νμ disappearance and νe appearance to estimate neutrino oscillation parameters. ND280 is one of T2K’s near detector. It is located off-axis and is needed to constrain flux and cross section. For the upgrade[2], the upstream section is replaced by three new sub-detectors: ●Super Fine Grained Detector (SuperFGD) ●2 High Angle Time Projection Chambers (HA-TPCs) ●6 Time Of Flight planes (ToF) Upgrade improvements ●Lower proton detection threshold ●Higher detection efficiencies ●Higher angular acceptance ●Neutron kinematics via time of flight The SuperFGD is the new active target: ●~2 million 1 cm3 scintillator cubes ●3 WLS fibres crossing each cube ●SiPMs readout at one end of each fibre ●~58’000 readout channels SuperFGD purposes: ●PID improvement ●Low energy threshold ●Neutron reconstruction The SuperFGD is the first scintillator detector that allows neutron kinetic energy reconstruction, by measuring the time of flight of a neutron produced in a neutrino interaction inside the detector. The time characterization is crucial for this purpose. The time measurement is affected mostly by two contributions: time offsets and time walk. Times ti of the signals produced in the same cube are highly correlated - Matching hit pairs[3]. With a granular detector it is possible to know the distances di between the channels. Each matching hit pair will have two channels (α, β), two times (t1, t2) and two distances (d1, d2). Measured difference: Δt = t1-t2 Expected difference: Δd = (d1-d2)/v, with v speed of light in the fibre Goal: minimize Δ = Δt-Δd with an iterative procedure Time offsets Mis-synchronization and cable length Time walk Smaller amplitudes appear to be later 1. For each channel α evaluate average Δα = ∑βΔαβ/N 2. Δα is the offset correction for channel α 3. Apply correction at the next iteration tk+1 = tk+Δα/2 4. Repeat until convergence Mathematically equivalent to the convergence of a row-stochastic matrix (Markov matrix)[4]. Mathematical convergence dealt with power method[5]. Monte-Carlo simulation of random deposits of energy shows the efficiency in extrapolating the correct offsets T0 previously applied. Generated offsets with two gaussian distribution centered in -2 and 2 ns (left) and with a uniform distribution between -1 and 1 ns (right). Time offsets ●SuperFGD time offsets map obtained with sample of cosmics and neutrino-beam data ●~200 matched hits per channel ●Time offsets firstly obtained using matching hit pairs having comparable charges ●Evident dependence on Front-End Board (FEB) electronic units ●FEB clock phase up to 2.5 ns Time walk ●Time walk overall contribution ●After time offset extrapolation ●Obtained using matching hit pairs with charges with large difference ●High charge time walk negligible wrt low charge time walk ●Improvement possible by obtaining time walk for each channel separately Fibre time resolution ●Single fibre time resolution after applying calibration ●Similar sample to validate calibration ●σt=σ(Δt-Δd)/√2 for matching hits with similar charges ●Considerable improvement after calibration result ●Time resolution expected behavior as charge increases Cube time resolution ●Sample of cosmic muon data ●Each cube raw time reconstructed with all the three fibres crossing it ●Similar behavior as a function of cube charge ●Cube time resolution below 1 ns at 120 p.e., typical cube charge from a MIP Neutron candidate event from beam data ●Sub-ns time resolution needed to unlock detection of neutron produced in the detector ●Capability of tagging and reconstructing neutrons Many thanks to Lorenzo Giannessi and Haowei Zheng for working with me in this project! T2K work in progress T2K work in progress T2K work in progress T2K work in progress T2K work in progress T2K work in progress Neutron candidate
