Monolayer V_{2}MX_{4}: A New Family of Quantum Anomalous Hall Insulators.

We theoretically propose that the van der Waals layered ternary transition metal chalcogenide V_{2}MX_{4} (M=W, Mo; X=S, Se) is a new family of quantum anomalous Hall insulators with sizable bulk gap and Chern number C=-1. The large topological gap originates from the deep band inversion between spin-up bands contributed by d_{xz}, d_{yz} orbitals of V and spin-down band from d_{z^{2}} orbital of M at the Fermi level. Remarkably, the Curie temperature of monolayer V_{2}MX_{4} is predicted to be much higher than that of monolayer MnBi_{2}Te_{4}. Furthermore, the thickness dependence of the Chern number for few multilayers shows interesting oscillating behavior. The general physics from the d orbitals here applies to a large class of ternary transition metal chalcogenide such as Ti_{2}WX_{4} with the space group P-42m. These interesting predictions, if realized experimentally, could greatly promote the research and application of topological quantum physics.

Realization of quasi-1D topological magnetism at the V-alloyed MoS2 zigzag edge.

Topological magnetism in quasi-1D systems can be interesting because of the significant quantum confinement. However, the realization is missing. In this letter, we propose the use of 3× periodicities related edge reconstructions of MoS2 zigzag edges to construct a topological quasi-1D spin chain. Specifically, a trimer Su-Schrieffer-Heeger model can be applied to illustrate the topological and spin order when the inter-cell hopping integral is larger than the intra-cell ones. As a result, topological ferromagnetic order is achieved for S-oriented edge states magnetized by V atoms and confirmed by first-principles calculations and Wannier functions analysis. Finally, gap opening and spin-polarized end states are realized.

Quasicrystalline 30° twisted bilayer graphene as an incommensurate superlattice with strong interlayer coupling.

The interlayer coupling can be used to engineer the electronic structure of van der Waals heterostructures (superlattices) to obtain properties that are not possible in a single material. So far research in heterostructures has been focused on commensurate superlattices with a long-ranged Moiré period. Incommensurate heterostructures with rotational symmetry but not translational symmetry (in analogy to quasicrystals) are not only rare in nature, but also the interlayer interaction has often been assumed to be negligible due to the lack of phase coherence. Here we report the successful growth of quasicrystalline 30° twisted bilayer graphene (30°-tBLG), which is stabilized by the Pt(111) substrate, and reveal its electronic structure. The 30°-tBLG is confirmed by low energy electron diffraction and the intervalley double-resonance Raman mode at 1383 cm-1 Moreover, the emergence of mirrored Dirac cones inside the Brillouin zone of each graphene layer and a gap opening at the zone boundary suggest that these two graphene layers are coupled via a generalized Umklapp scattering mechanism-that is, scattering of a Dirac cone in one graphene layer by the reciprocal lattice vector of the other graphene layer. Our work highlights the important role of interlayer coupling in incommensurate quasicrystalline superlattices, thereby extending band structure engineering to incommensurate superstructures.

A two-dimensional ErCu2 intermetallic compound on Cu(111) with moiré-pattern-modulated electronic structures.

A rare-earth compound on a metal may form a two-dimensional (2D) intermetallic compound whose properties can be further modulated by the underlying substrate periodicity and coupling. Here, we present a combinational and systematic investigation using scanning tunneling microscopy/spectroscopy (STM/STS) and density functional theory (DFT) calculations on erbium (Er) on Cu(111). Experimentally, an intriguing growth mode transition from a branched island to a fractal-like island has been observed depending on whether the deposition process of Er is interrupted for a certain duration: post-deposition effects, such as nucleation and island growth controlled by diffusion, play an essential role in altering the Er island edge and its activity. Upon annealing, the branched Er islands become strands of amorphous surface alloy; in contrast, the fractal-like islands (with additional Er atoms on top) give rise to a monolayer thick 2D ErCu2 intermetallic compound and display a moiré pattern. Theoretically, using DFT calculations, we found that the characteristic energy states, particularly the state in the unoccupied region around 582-663 meV, of the 2D ErCu2 intermetallic compound are position-dependent, consistent with STS measurements. The moiré pattern originating from the mismatch of the periodicities of the ErCu2 layer and the Cu(111) surface was identified to be responsible for the observed periodic modulation on the coupling interaction that affects the electronic structures. Our further DFT calculations on a free-standing ErCu2 monolayer found it to be a 2D ferromagnet with topological band structures. Our work should stimulate further studies on such 2D rare-earth-based nanostructures and exploration of the use of the tunable electronic structures in such atomically-thin layers.

Parameter-Free and Electron Counting Satisfied Material Representation for Machine Learning Potential Energy and Force Fields.

We proposed a parameter-free volume element representation that satisfies the electron counting model and obtains accurate machine learning potential energy and direct force fitting of randomly perturbed hexagonal BN. Our method preserves permutational, translational, and rotational invariance and can be extended to three-dimensional systems, verified by a system of bulk Si. As a result, we obtained 0.57 meV/atom potential energy root mean squared error (RMSE) and 59 meV/Å force RMSE for perturbed bulk BN systems and 0.43 meV/atom potential energy RMSE and 36 meV/Å force RMSE for perturbed Si systems. In addition, an unbiased perturbation-based data set construction scheme is introduced and a continuous population distribution is obtained with a training data set of 4500, which is about 1 order of magnitude smaller than standard methods based on first-principles molecular dynamics simulations and saves a large amount of computing resources. General validity of our model is verified by structure optimization, molecular dynamics simulations, and extrapolations.