Electronic structure simulations enable the calculation of a wide variety of fundamental materials properties. However, they consume a significant portion of scientific HPC resources worldwide. Artificial intelligence and machine learning, which have emerged as a powerful tool for analyzing complex datasets, have the potential to accelerate electronic structure calculations such as density functional theory. The combination of these two fields enables highly efficient simulations at unprecedented scales. In this review, the authors present a comprehensive analysis of research articles in chemistry and materials science that employ machine-learning techniques and outline the current trends at the intersection of these fields.

The recently discovered kagome metal CsV${}_3$Sb${}_5$ displays a superconducting transition at low temperature accompanied by a charge density wave ordering at higher temperature, among many other interesting features that arise from nested saddle points near the Fermi energy. Through careful hole doping via partial substitution of Sn in the in-plane kagome Sb site, double-dome superconductivity and suppressed charge density wave order were observed. These phenomena can be partially explained by modeling the evolution of electronic band structure and changes in Fermi surface.

A newly discovered material offers a platform to study exotic spin structures and transport mechanisms that could be relevant to future spin-based electronic devices.

Synopsis on:
Hongrui Zhang et al.
Phys. Rev. Materials 6, 044403 (2022)

The spatial complexity of cross-scale atomistic simulations renders them unsuitable for simple human visual inspection. Instead, specialized structure characterization techniques are required to aid interpretation. These have historically been challenging to construct, requiring significant intuition and effort. In this article the authors introduce a data-centric framework that favors the employment of machine learning over heuristic rules of classification. It is demonstrated that the data-centric framework outperforms all of the most popular heuristic methods while introducing a systematic route for generalization to new crystal structures.

The authors study the divacancy in 3C-SiC, a promising system for quantum information or sensing applications, using large-scale GW plus Bethe-Salpeter equation simulations of nearly 1000 atoms. Notably, in contrast to the widely studied diamond NV center, low-energy excitonic states of 3C-SiC divacancy show substantial characters of transitions from localized defect states to continuum states. Some defect states that contribute to the low-energy excitations significantly hybridize with conduction bands. This work quantitatively determines the quasiparticle energies of defect states and zero-phonon line energy, emphasizing the importance of frontier conduction bands on the low-energy excitons of 3C-SiC divacancy.

While the materials family of electron-doped Bi${}_2$Se${}_3$ has been established as nematic topological superconductors, the observability of superconductivity on their surface has been controversial for over 10 years. Here, the authors try to resolve this longstanding issue with extensive STM study of high-quality Sr${}_x$Bi${}_2$Se${}_3$ crystals. Based on their results they propose that superconductivity cannot reach the surface when the topological surface states are intact, but it becomes observable when the topological surface states are destroyed due to strain. In particular, contamination of the STM tip with micrometer-sized flakes of strained Sr${}_x$Bi${}_2$Se${}_3$ can cause spurious observation of superconductivity.

Inversion-asymmetric stacks of metallic magnetic layers have often been exploited to control the chiral noncollinear ordering of their magnetic moments. Here, the authors investigate the interfacial aspects of the Dzyaloshinskii-Moriya interaction, giving rise to this chiral magnetic ordering, and quantify its contributions to within a couple atomic layers. This observation is further supported by first-principles calculations. The confirmation of the short spatial extent of the interfacial DMI is expected to enable the synthesis of dense magnetic multilayers and to offer further possibilities for engineering their spintronic properties.

The authors present a promising machine learning model, which focuses on site-magnetic-properties for rapid screening in materials design and accelerates computational screening of candidate materials that possess high magnetizations and large magnetic anisotropy energies.

Understanding and controlling interlayer hybridization in layered van der Waals materials is an important prerequisite for developing efficient and highly tunable spin- and valleytronic devices.Here, the authors spectroscopically investigate the vibrational and orbital coupling between layers for the intricate case of bilayer and trilayer MoS${}_2$. The application of an external electric field manifests itself in field-activated phonon modes along with strongly tunable circular dichroism in both bilayer and trilayer MoS${}_2$. First-principles calculations in combination with rate equation modeling suggest that interlayer charge transfer via the Q point dominates the electron population reflected in the tunable circular dichroism. This work contributes to the understanding of the complex interplay between crystal symmetry and interlayer charge transfer in van der Waals materials.

Guest Editor David Tománek reflects on the joint Collection at its closing.

At the beginning of 2021, eight Physical Review journals began publishing Letters which are intended for the accelerated publication of important new results targeted to the specific readership of each journal.

APS Editor in Chief, Michael Thoennessen, discusses a new opportunity for communicating authors to include their pronouns together with their contact email in order to promote a more respectful, inclusive, and equitable environment.

Physical Review Applied and Physical Review Materials are pleased to present the Collection on Two-Dimensional Materials and Devices, highlighting one of the most interesting fields in Applied Physics and Materials Research. Papers belonging to this collection will be published throughout 2020. The invited articles, and an editorial by the Guest Editor, David Tománek, are linked in the Collection.

Guest Editor David Tománek introduces a collection of papers in Physical Review Applied and Physical Review Materials on two-dimensional materials and devices, in a snapshot of the leading edge of this hot field.

A discussion of the focus on materials related research in the Physical Review journals.