Machine learning is one of the most popular fields in computer science and has a vast number of applications. In this work we will propose a method that will use a neural network to locally identify crystal structures in a mixed phase Yukawa system consisting of fcc, hcp, and bcc clusters and disordered particles similar to plasma crystals. We compare our approach to already used methods and show that the quality of identification increases significantly. The technique works very well for highly disturbed lattices and shows a flexible and robust way to classify crystalline structures that can be used by only providing particle positions. This leads to insights into highly disturbed crystalline structures.
Three-dimensional plasma crystals are often described as Yukawa systems for which a phase transition between the crystal structures fcc and bcc has been predicted. However, experimental investigations of this transition are missing. We use a fast scanning video camera to record the crystallization process of 70 000 microparticles and investigate the existence of the fcc-bcc phase transition at neutral gas pressures of 30, 40, and 50 Pa. To analyze the crystal, robust phase diagrams with the help of a machine learning algorithm are calculated. This work shows that the phase transition can be investigated experimentally and makes a comparison with numerical results of Yukawa systems. The phase transition is analyzed in dependence on the screening parameter and structural order. We suggest that the transition is an effect of gravitational compression of the plasma crystal. Experimental investigations of the fcc-bcc phase transition will provide an opportunity to estimate the coupling strength Γ by comparison with numerical results of Yukawa systems.
Applying an external electric AC field to a dusty plasma, the micro‐particles arrange in strings or chains. In analogy to electrorheological fluids, such a system is called electrorheological plasma. Turning gradually the AC field into a DC field, the string formation is diminished until it vanishes completely in the DC case. In this way, a cross‐over transition from a string‐like to an isotropic micro‐particle many‐body system can be investigated. Experimental investigations of electrorheological plasmas are performed under microgravity conditions in parabolic flights. For analysing the image data, a supervised machine learning code was developed and a continuous cross over was found. A molecular dynamics simulation showed qualitatively similar results but also some deviations from the experimental results.
The influence of neutral gas pressure for crystallization of cylindrical complex plasmas under laboratory conditions is investigated. For the analysis of the complex plasma structure, different methods are adopted: First, the pair correlation and a criterion based on the shape of the Voronoi cells are applied. Besides this, a new implementation, which connects the Minkowski structure metric with the benefits from the scalar product of the local bond order parameter, is presented. In addition, the bcc sensitive Minkowski structure metric is used to identify the crystalline structures. All criteria display the same behavior: Decreasing the neutral pressure leads to crystallization. This is the opposite behavior to that observed in former ground based experiments.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.