Molecular dynamics simulations play an important role in studying heat transport in complex materials. The lattice thermal conductivity can be computed either using the Green-Kubo formula in equilibrium MD (EMD) simulations or using Fourier's law in nonequilibrium MD (NEMD) simulations. These two methods have not been systematically compared for materials with different dimensions and inconsistencies between them have been occasionally reported in the literature. Here we give an in-depth comparison of them in terms of heat transport in three allotropes of Si: three dimensional bulk silicon, two-dimensional silicene, and quasi-one-dimensional silicon nanowire. By multiplying the correlation time in the Green-Kubo formula with an appropriate effective group velocity, we can express the running thermal conductivity in the EMD method as a function of an effective length and directly compare it with the length-dependent thermal conductivity in the NEMD method. We find that the two methods quantitatively agree with each other for all the systems studied, firmly establishing their equivalence in computing thermal conductivity. arXiv:1805.00269v1 [physics.comp-ph] 1 May 2018

No abstract

Traditional metal-oxide semiconductor devices are inadequate for use in artificial neural networks (ANNs) owing to their high power consumption, complex structures, and difficult fabrication techniques. Resistive random access memory (RRAM) is a promising candidate for ANNs owing to its simple structure, low power consumption, and excellent compatibility with CMOS. Moreover, it can mimic synaptic functions because of its multilevel resistive switching (RS) behavior. Herein, we demonstrate highly uniform RS and a high on/off ratio of RRAM based on graphene oxide by embedding gold nanoparticles into the device. This allowed reliable multilevel storage. Further, multilevel RRAM based on spike-timing-dependent-plasticity learning rules was used for image pattern recognition. These findings may offer a route to develop reliable digital memristors for ANNs.

Recently, a novel two-dimensional (2D) semiconductor, InSe, has attracted great attention due to its potential applications in optoelectronic devices and field effect transistors. In this study, phonon-limited mobility is investigated by the first-principles calculation. At 300 K, the intrinsic electron mobilities calculated from the electron-phonon coupling (EPC) matrix element are as high as µ x = 9.85 × 10 2 (zigzag direction) and µ y = 1.06 × 10 3 cm 2 V −1 s −1 (Armchair direction), respectively. The deformation potential theory (DPT) based on longitudinal acoustic and optical phonon scattering is also employed to investigate electron mobility. The mobility from optical phonon scattering is much higher than that from longitudinal acoustic phonon scattering. If the polarization characteristics of InSe are not considered, the electron mobility calculated from EPC matrix element is closed to that from the longitudinal acoustic phonon DPT. In this study, we have also investigated the effect of polarization properties in 2D InSe on electron mobility. At 300 K, the electron mobility for including Fröhlich interaction is reduced to µ x = 2.96 × 10 2 and µ y = 3.34 × 10 2 cm 2 V −1 s −1 . Therefore, the electron mobility for InSe is controlled by the scattering from polar phonons. The mobility can be increased to µ x = 3.46 × 10 2 and µ y = 3.78 × 10 2 cm 2 V −1 s −1 under 4% biaxial strain. This result is compared with the experiment, and some disagreements are explained.

No abstract