Traditional strengthening ways, such as strain, precipitation, and solid-solution, come into effect by pinning the motion of dislocation. Here, through first-principles calculations we report on an extra-electron induced covalent strengthening mechanism, which alters chemical bonding upon the introduction of extra-valence electrons in the matrix of parent materials. It is responsible for the brittle and high-strength properties of Al12W-type compounds featured by the typical fivefold icosahedral cages, which are common for quasicrystals and bulk metallic glasses (BMGs). In combination with this mechanism, we generalize ductile-to-brittle criterion in a universal hyperbolic form by integrating the classical Pettifor's Cauchy pressure with Pugh's modulus ratio for a wide variety of materials with cubic lattices. This study provides compelling evidence to correlate Pugh's modulus ratio with hardness of materials and may have implication for understanding the intrinsic brittleness of quasicrystals and BMGs.
We propose a doping method by using [6,6]-phenyl-C-butyric acid methyl ester (PCBM) to fill the grain boundary interstices of the methylammonium lead iodide (CHNHPbI) perovskite for the elimination of pinholes. A sandwiched PCBM layer is also used between the perovskite and TiO layers to improve the interfacial contact. By using these two methods, the fabricated perovskite solar cells show a low hysteresis effect and high current density, which result from the improved compactness at the grain boundaries of the perovskite surface and the interface between the TiO/perovskite layers. The theoretical and experimental results indicate that PCBM can effectively suppress carrier recombination, regardless of the interfacial layer or dopant. We also found that the dark current reduced during the analysis of dark state current-voltage ( I- V) characteristics. The slopes of the I- V curves for the fluorine-doped tin oxide/PCBM-doped perovskite/Au device reduce monotonically with the increase in the PCBM concentration from 0.01 to 0.1 wt %, which suggest the decreasing defects in the perovskite layer. By tuning the PCBM doping and controlling the preparation process, we have successfully fabricated a planar TiO/PCBM-based PCBM-doped perovskite photovoltaic device that reaches a high current density of 22.6 mA/cm and an outstanding photoelectric conversion efficiency up to 18.3%. The controllability of the PCBM doping concentration and interfacial preparation shed light on further optimization of the photoelectric conversion efficiency of perovskite solar cells.
Traffic flow forecasting is a critical task for urban traffic control and dispatch in the field of transportation, which is characterized by the high nonlinearity and complexity. In this paper, we propose an end-to-end deep learning based dual path framework, i.e., Spatial-Temporal Graph Attention Network (STGAT), for traffic flow forecasting. Specifically, different from previous structure-based approaches, STGAT can be directly generalized to the graph with arbitrary structure. Furthermore, STGAT is capable of handling long temporal sequence by stacking gated temporal convolution layer. The dual path architectures is proposed for taking both potential and existing spatial dependencies into account. By capturing potential spatial dependencies will naturally catch more useful information for forecasting. We design a gated fusion mechanism to combine the outputs from each path. The proposed model can be directly applicable to inductive learning tasks by introducing a graph attention mechanism into spatial-temporal framework, which means our model can be generalized to completely unseen graphs. Moreover, experimental results on two public real-world traffic network datasets, METR-LA and PEMS-BAY, show that our STGAT outperforms the state-of-the-art baselines. Additionally, we demonstrate the proposed model is competent for efficient migration between graphs with different structures.INDEX TERMS Traffic flow forecasting, spatial-temporal graph neural networks, intelligent transportation systems.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.