Abstract: In dc magnetrons the electrons emitted from the cathode may return there due to the applied magnetic field. When that happens, they can be recaptured or reflected back into the discharge, depending on the value of the reflection coefficient (RC). A 2d3v (two-dimensional in coordinate and three-dimensional in velocity space) particle-in-cellMonte Carlo model, including an external circuit, is developed to determine the role of the electron recapture in the discharge processes. The detailed discharge structure as a function of RC for two pressures (4 and 25mtorr) is studied. The importance of electron recapture is clearly manifested, especially at low pressures. The results indicate that the discharge characteristics are dramatically changed with varying RC between 0 and 1. Thus, the electron recapture at the cathode appears to be a significant mechanism in magnetron discharges and RC a very important parameter in their correct quantitative description that should be dealt with cautiously.

Abstract: Molecular dynamic simulations are performed to investigate the melting process of small three-dimensional clusters (i.e., systems with one and two shells) of classical charged particles trapped in an isotropic parabolic potential. The confined particles interact through a repulsive potential. We find that the ground-state configurations for systems with N=6, 12, 13, and 38 particles interacting through a Coulomb potential are magic clusters. Such magic clusters have an octahedral or icosahedral symmetry and are found to have a large stability against intrashell diffusion leading to an intershell melting transition prior to the intrashell and radial melting process. For systems with two shells a local radial melting of subshells is found at low temperatures resulting in a structural transition leading to an increased symmetry of the ordered system. Using Lindemanns criterion the different melting temperatures are determined and the influence of the screening of the interparticle interaction was investigated. A normal mode analysis is performed and some of the normal modes are found to be determinantal for the melting process.

Abstract: Similar to the spontaneous formation of colonies of bacteria, flocks of birds, or schools of fish, “living crystals” can be formed by artificial self-propelled particles such as Janus colloids. Unlike usual solids, these “crystals” are far from thermodynamic equilibrium. They fluctuate in time forming a crystalline structure, breaking apart and re-forming again. We propose a method to stabilize living crystals by applying a weak confinement potential that does not suppress the ability of the particles to perform self-propelled motion, but it stabilizes the structure and shape of the dynamical clusters. This gives rise to such configurations of living crystals as “living shells” formed by Janus colloids. Moreover, the shape of the stable living clusters can be controlled by tuning the potential strength. Our proposal can be verified experimentally with either artificial microswimmers such as Janus colloids, or with living active matter.

Abstract: The autoregressive neural networks are emerging as a powerful computational tool to solve relevant problems in classical and quantum mechanics. One of their appealing functionalities is that, after they have learned a probability distribution from a dataset, they allow exact and efficient sampling of typical system configurations. Here we employ a neural autoregressive distribution estimator (NADE) to boost Markov chain Monte Carlo (MCMC) simulations of a paradigmatic classical model of spin-glass theory, namely, the two-dimensional Edwards-Anderson Hamiltonian. We show that a NADE can be trained to accurately mimic the Boltzmann distribution using unsupervised learning from system configurations generated using standard MCMC algorithms. The trained NADE is then employed as smart proposal distribution for the Metropolis-Hastings algorithm. This allows us to perform efficient MCMC simulations, which provide unbiased results even if the expectation value corresponding to the probability distribution learned by the NADE is not exact. Notably, we implement a sequential tempering procedure, whereby a NADE trained at a higher temperature is iteratively employed as proposal distribution in a MCMC simulation run at a slightly lower temperature. This allows one to efficiently simulate the spin-glass model even in the low-temperature regime, avoiding the divergent correlation times that plague MCMC simulations driven by local-update algorithms. Furthermore, we show that the NADE-driven simulations quickly sample ground-state configurations, paving the way to their future utilization to tackle binary optimization problems.