Using nonequilibrium molecular dynamics and lattice dynamics, we investigate phonon conduction on twodimensional graphene/boron nitride superlattices with varying periods and interface structures. As the period of superlattice increases to a critical value near 5 nm the lattice thermal conductivity drops sharply to a minimum, and beyond that it smoothly increases with the period. We show that the minimum in the thermal conductivity arises from a competition between lattice dispersion and anharmonic effects such as interface scattering. The initial reduction of thermal conductivity can partially be accounted for by harmonic wave effects induced by interfacial modulation, such as the opening of phononic band gaps and reduction of group velocity. Beyond the minimum, reduced inelastic interface scattering is responsible for the recovery. The overall range of thermal conductivity exhibited by the superlattices is substantially reduced with respect to the parent materials. A universal scaling of the thermal conductivity with total superlattice length is found, suggesting that the critical period is independent of total length and that long-wavelength phonons are dominant carriers. Furthermore, we demonstrate the ultrasensitivity of thermal conductivity to interfacial defects and superlattice periodicity disorder.

First-principles and classical modeling reveal a glass-crystal duality in the nature of vibrational transport in lead halide hybrid perovskites.

Vacancy-ordered lead-free perovskites with more-stable crystalline structures have been intensively explored as the alternatives for resolving the toxic and long-term stability issues of lead halide perovskites (LHPs). The dispersive energy bands produced by the closely packed halide octahedral sublattice in these perovskites are meanwhile anticipated to facility the mobility of charge carriers. However, these perovskites suffer from unexpectedly poor charge carrier transport. To tackle this issue, we have employed the ultrafast, elementalspecific X-ray transient absorption (XTA) spectroscopy to directly probe the photoexcited electronic and structural dynamics of a prototypical vacancy-ordered lead-free perovskite (Cs3Bi2Br9). We have discovered that the photogenerated holes quickly self-trapped at Br centers, simultaneously distorting the local lattice structure, likely forming small polarons in the configuration of Vk center (Br2dimer). More significantly, we have found a surprisingly long-lived, structural distorted state with a lifetime of ∼59 μs, which is ∼3 orders of magnitude slower than that of the charge carrier recombination. Such long-lived structural distortion may produce a transient "background" under continuous light illumination, influencing the charge carrier transport along the lattice framework.

Thermoelectricity produced from usually negative-valued heat is a green and promising candidate on the future energy landscape. The most effective thermoelectric materials exhibit low thermal conductivity κ. However, less than...

Charge-compensation-induced Frenkel defects were proven as the major reason for phonon scattering of p-type half-Heusler ZrCoSb.

Recently, the domains of low-dimensional (low-D) materials and disordered materials have been brought together by the demonstration of several new low-D, disordered systems. The thermal transport properties of these systems are not well-understood. Using amorphous graphene and glassy diamond nanothreads as prototype systems, we establish how structural disorder affects vibrational energy transport in low-dimensional, but disordered, materials. Modal localization analysis, molecular dynamics simulations, and a generalized model together demonstrate that the thermal transport properties of these materials exhibit both similarities and differences from disordered 3D materials. In analogy with 3D, the low-D disordered systems exhibit both propagating and diffusive vibrational modes. In contrast to 3D, however, the diffuson contribution to thermal transport in these low-D systems is shown to be negligible, which may be a result of inherent differences in the nature of random walks in lower dimensions. Despite the lack of diffusons, the suppression of thermal conductivity due to disorder in low-D systems is shown to be mild or comparable to 3D. The mild suppression originates from the presence of low-frequency vibrational modes that maintain a well-defined polarization and help preserve the thermal conductivity in the presence of disorder.

Ultrathin transition metal carbides with high capacity, high surface area, and high conductivity are a promising family of materials for applications from energy storage to catalysis. However, large-scale, cost-effective, and precursor-free methods to prepare ultrathin carbides are lacking. Here, we demonstrate a direct pattern method to manufacture ultrathin carbides (MoC x , WC x , and CoC x ) on versatile substrates using a CO 2 laser. The laser-sculptured polycrystalline carbides (macroporous, ~10–20 nm wall thickness, ~10 nm crystallinity) show high energy storage capability, hierarchical porous structure, and higher thermal resilience than MXenes and other laser-ablated carbon materials. A flexible supercapacitor made of MoC x demonstrates a wide temperature range (−50 to 300 °C). Furthermore, the sculptured microstructures endow the carbide network with enhanced visible light absorption, providing high solar energy harvesting efficiency (~72 %) for steam generation. The laser-based, scalable, resilient, and low-cost manufacturing process presents an approach for construction of carbides and their subsequent applications.

Electron charge density distribution of materials is one of the key quantities in computational materials science as theoretically it determines the ground state energy and practically it is used in many materials analyses. However, the scaling of density functional theory calculations with number of atoms limits the usage of charge-density-based calculations and analyses. Here we introduce a machine learning scheme with local-environment-based graphs and graph convolutional neural networks to predict charge density on grid-points from crystal structure. We show the accuracy of this scheme through a comparison of predicted charge densities as well as properties derived from the charge density, and the scaling is O(N). More importantly, the transferability is shown to be high with respect to different compositions and structures, which results from the explicit encoding of geometry.