Particle identification method by analyzing pulse shape with neural network

 Recent cluster-model calculations predict that α condensed states emerge in self-conjugate N = 4n nuclei. In the α condensed states, all of the α clusters are condensed in the lowest energy orbit, and their matter density is as low as 1/4 to 1/5 of normal nuclear states. Thus, observation of the α condensed states is important for clarifying physical properties of low-density nuclear matter.
 The α condensed states are expected to decay by emitting multiple α clusters. However, it is predicted that the emitted α particles have low energies about 1—3 MeV. It is difficult to identify such low-energy particles by conventional E - ΔE telescopes.because these particles cannot penetrate the ΔE detector.
 In the present study, we attempted to identify low-energy charged particles by pulse shape analysis with a machine  learning technique. We acquired pulse shapes for known particles and used them to train an AI. We will report details of our study and performance of the particle identification method with the AI.