Showing papers on "Cluster (physics) published in 2022"

Multi-view spectral clustering by simultaneous consensus graph learning and discretization

Abstract: Multi-view spectral clustering has drawn much attention due to its excellent performance in grouping arbitrarily shaped data. Most of the multi-view spectral clustering methods perform clustering, relying on multiple predefined similarity graphs. Unfortunately, they require three separate steps in sequence, i.e., similarity graph learning, cluster label relaxing, and discretization of continuous labels, resulting in a compromised clustering performance. The reason is that the predefined similarity graph may not be optimal for the subsequent clustering, and the relax-and-discretize strategy may cause significant information loss. To this end, in this work, we disentangle the above issue by simultaneous consensus graph learning and discretization, where the similarity graph and the discrete cluster label matrix are learned in a unified framework. Specifically, the consensus graph shared by all views is adaptively learned with the guidance of the discrete cluster label matrix. In contrast, the cluster information hidden in the discrete label matrix can effectively boost the quality of the consensus graph. As a result, information loss among independent steps is effectively obviated, and better performance can be achieved. Experimental results on several challenging data sets validated the effectiveness of the proposed method compared to the state-of-the-art approaches.

A systematic investigation of structural growth patterns and electronic properties of [LuGen]+/0 and [Gen+1]+/0 (n=1-17) nanoalloy clusters

Abstract: The geometries, stabilities, evolutions, and electronic properties of lutetium-doped germanium nanoalloy clusters LuGen (n = 1–17) in the neutral and cationic state are studied and compared with Gen+1 clusters by using unbiased global search method combined with a hybrid density functional approach. The results reveal that neutral and cationic structural growth patterns prefer Lu-substituted Gen+1 for n = 1–8, Lu-linked configuration for n = 9–14 (for cation n = 9–15) in which Lu atom links two Ge subclusters, and Lu-encapsulated Ge cage-like motif for n = 15–17 (for cation n = 16–17). A binding energy analysis reveals that Lu-encapsulated motifs enhance the nanocluster stability. Lu-doped neutral and cationic germanium clusters can decrease their ionization potential. Natural population analysis shows that Lu in Lu-substituted and Lu-linked configurations acts as electron donor, but acts as electron acceptor in Lu-encapsulated structures. HOMO-LUMO energy gap analysis discloses that Lu-doped Ge clusters can lessen their chemical stability excluding LuGe12 and LuGe17. Thorough analysis of stability, chemical bonding, density of states, iso-chemical shielding surface and UV-Vis spectra revealed that [LuGe17]+ is magic nanoalloy cluster, and has an ideal chemical and thermal stability, making it the most preferable building block for optoelectronic materials. Additionally, Infrared and Raman spectra of [LuGe17]+ are also reported. The accordance between theoretical and experimental values for binding energies of Ge2–8 and ionization potentials of Ge2–16,18 clusters suggest that the calculations and analysis of results in this study are successful.

Exploration of the icosahedral clusters in Ni–Nb binary metallic glasses via first-principles theory

Abstract: The binding energy, chemical ordering parameters, and electronic structure of Ni-centered NinNb13−n(n = 6, 7, 8, 9) icosahedral clusters were studied in accordance with the first-principles theory. These calculations demonstrated that the binding energy of the icosahedral clusters decreased when the homogeneous shell atoms preferred to bond together. The density of state at the EF of most icosahedral clusters was near a gap, indicating that all icosahedral clusters were in a relatively stable state. For the icosahedral cluster, the states near the Fermi level were dominated by Ni- and Nb-d electrons. The electron-localization function analysis of the most stable and unstable structures in icosahedral clusters showed that chemical bonding favored the formation of icosahedral clusters by introducing a string-like or net-like structure and further stabilized icosahedral clusters. The origin of glass-forming ability (GFA) was the synergistic effect of electronic stability and topological stability of icosahedral clusters in Ni–Nb metallic glasses.