Phosphorene, the single layer counterpart of black phosphorus, is a novel two-dimensional semiconductor with high carrier mobility and a large fundamental direct band gap, which has attracted tremendous interest recently. Its potential applications in nano-electronics and thermoelectrics call for fundamental study of the phonon transport. Here, we calculate the intrinsic lattice thermal conductivity of phosphorene by solving the phonon Boltzmann transport equation (BTE) based on first-principles calculations. The thermal conductivity of phosphorene at 300 K is 30.15 W m⁻¹ K⁻¹ (zigzag) and 13.65 W m⁻¹ K⁻¹ (armchair), showing an obvious anisotropy along different directions. The calculated thermal conductivity fits perfectly to the inverse relationship with temperature when the temperature is higher than Debye temperature (Θ_D = 278.66 K). In comparison to graphene, the minor contribution around 5% of the ZA mode is responsible for the low thermal conductivity of phosphorene. In addition, the representative mean free path (MFP), a critical size for phonon transport, is also obtained.

A major challenge in materials design is how to efficiently search the vast chemical design space to find the materials with desired properties. One effective strategy is to develop sampling algorithms that can exploit both explicit chemical knowledge and implicit composition rules embodied in the large materials database. Here, we propose a generative machine learning model (MatGAN) based on a generative adversarial network (GAN) for efficient generation of new hypothetical inorganic materials. Trained with materials from the ICSD database, our GAN model can generate hypothetical materials not existing in the training dataset, reaching a novelty of 92.53% when generating 2 million samples. The percentage of chemically valid (charge-neutral and electronegativity-balanced) samples out of all generated ones reaches 84.5% when generated by our GAN trained with such samples screened from ICSD, even though no such chemical rules are explicitly enforced in our GAN model, indicating its capability to learn implicit chemical composition rules to form compounds. Our algorithm is expected to be used to greatly expand the range of the design space for inverse design and large-scale computational screening of inorganic materials.

New classes of two-dimensional (2D) materials beyond graphene, including layered and non-layered, and their heterostructures, are currently attracting increasing interest due to their promising applications in nanoelectronics, optoelectronics and clean energy, where thermal transport is a fundamental physical parameter. In this paper, we systematically investigated the phonon transport properties of the 2D orthorhombic group IV–VI compounds of GeS, GeSe, SnS and SnSe by solving the Boltzmann transport equation (BTE) based on first-principles calculations. Despite their similar puckered (hinge-like) structure along the armchair direction as phosphorene, the four monolayer compounds possess diverse anisotropic properties in many aspects, such as phonon group velocity, Young's modulus and lattice thermal conductivity (κ), etc. Especially, the κ along the zigzag and armchair directions of monolayer GeS shows the strongest anisotropy while monolayer SnS and SnSe show almost isotropy in phonon transport. The origin of the diverse anisotropy is fully studied and the underlying mechanism is discussed in details. With limited size, the κ could be effectively lowered, and the anisotropy could be effectively modulated by nanostructuring, which would extend the applications to nanoscale thermoelectrics and thermal management. Our study offers fundamental understanding of the anisotropic phonon transport properties of 2D materials, and would be of significance for further study, modulation and applications in emerging technologies.

Silicene, the silicon-based counterpart of graphene with a two-dimensional honeycomb lattice, has attracted tremendous interest both theoretically and experimentally due to its significant potential industrial applications. From the aspect of theoretical study, the widely used classical molecular dynamics simulation is an appropriate way to investigate the transport phenomena and mechanisms in nanostructures such as silicene. Unfortunately, no available interatomic potential can precisely characterize the unique features of silicene. Here, we optimized the Stillinger-Weber potential parameters specifically for a single-layer Si sheet, which can accurately reproduce the low buckling structure of silicene and the full phonon dispersion curves obtained from ab initio calculations. By performing equilibrium and nonequilibrium molecular dynamics simulations and anharmonic lattice dynamics calculations with the new potential, we reveal that the three methods consistently yield an extremely low thermal conductivity of silicene and a short phonon mean-free path, suggesting silicene as a potential candidate for high-efficiency thermoelectric materials. Moreover, by qualifying the relative contributions of lattice vibrations in different directions, we found that the longitudinal phonon modes dominate the thermal transport in silicene, which is fundamentally different from graphene, despite the similarity of their two-dimensional honeycomb lattices.