No abstract

No abstract

No abstract

Employing molecular dynamics, we have simulated collisions of various sizes of 5.7-keV silicon clusters with a silicon surface. At this energy, the simulation provides an atomistic description of the evolution of the multiple scattering process of atoms in the cluster as well as the substrate where atoms interact with each other by covalent forces. The change from the collision cascade that is characteristic in the single-ion case to the simultaneous collision process that leads to the collective motion of particles, such as macroscopic vibration of the surface and shock waves, is found by increasing the number of the atoms in the cluster. With an increase in cluster size, the impurity profile and the damaged region for the cluster ion implantation is also found to be shallower. In contrast to single-ion implantation, cluster implantation may not induce defects in deep regions, making it suitable for device design. The channeling effect is found to be mainly suppressed by the lowfrequency macroscopic vibrations of the surface rather than by the amorphization of the surface caused by the impact.

Accurate assessment of social interactions in mammalian models are necessary to elucidate the pathogenesis of psychiatric and neurodegenerative diseases. The common marmoset (Callithrix jacchus) is a useful model for understanding the mechanisms of these disorders. However, behavioral measurements of free-moving group animals, including assessment of sociality have not been conducted. Here, we developed a new animal behavioral analysis system for three-dimensional (3D) trajectory and behavior of free-moving individuals. Each marmoset was identified using deep learning facial recognition (accuracy ≥ 97%). Their trajectories were tracked independently by combining video, 3D coordinates detected by light detection and ranging (lidar) systems, and facial identification. Location preferences and distances between individuals were calculated using the 3D trajectories and grooming behavior was detected by deep learning. This novel system will allow the quantification of individual captive animals within groups, facilitating the quick and automatic measurement of social behavior, and enhancing development of objective behavioral indices.