Zhejiang University

About: Zhejiang University is a education organization based out in Hangzhou, Zhejiang, China. It is known for research contribution in the topics: Catalysis & Population. The organization has 161257 authors who have published 183264 publications receiving 3417592 citations. The organization is also known as: Chekiang University & Zheda.

Efficient alkaline hydrogen evolution on atomically dispersed Ni–Nx Species anchored porous carbon with embedded Ni nanoparticles by accelerating water dissociation kinetics

Abstract: Developing inexpensive and efficient electrocatalysts for hydrogen evolution reaction (HER) during alkaline water electrolysis is crucial for renewable and sustainable energy harvesting. Herein, we report a novel hybrid electrocatalyst comprising atomically dispersed Ni–Nx species anchored porous carbon (Ni–N–C) matrix with embedded Ni nanoparticles for HER. This new catalyst is synthesized via pyrolysis of hydrothermally prepared supermolecular composite of dicyandiamide and Ni ions followed by an acid etching treatment. The achieved hybrid exhibits superior catalytic performance toward HER with a small overpotential of 147 mV at 10 mA cm−2 and a low Tafel slope of 114 mV dec−1, comparable to those of state-of-the-art heteroatom-doped nanocarbon catalysts and even outperforming other reported transition-metal-based compounds in basic media. Experimental observations and theoretical calculations reveal that the presence of Ni nanoparticles can optimize surface states of Ni−Nx active centers and reduce energy barriers of dissociated water molecules, which synergistically improve OH− adsorption and promote HER kinetics. When served as electrodes for both cathode and anode, an alkaline water electrolyzer could afford a current density of 10 mA cm−2 at a low cell voltage of 1.58 V, rivalling the sufficiently high overpotentials of integrated Pt/C–Ir/C benchmark electrodes.

Snubberless Bidirectional DC–DC Converter With New CLLC Resonant Tank Featuring Minimized Switching Loss

Abstract: A bidirectional DC-DC converter (BDC) with a new CLLC-type resonant tank, which features zero-voltage switching (ZVS) for the input inverting choppers and zero-current switching (ZCS) for the output rectifier switches, regardless of the direction of the power flow, is proposed in this paper. Possessing the very optimal ZVS +ZCS soft-switching feature, this proposed converter will have a minimized switching loss if all of the main switches are implemented with metal-oxide-semiconductor field-effect transistors, and thereby, the proposed converter is fully soft switched and totally snubberless. The detail operation principles, as well as the design considerations, are presented. The methodologies to develop a unidirectional ZVS+ZCS dc-dc converter for the corresponding pulsewidth modulation and frequency modulation converters are proposed. The approach on how to construct a fully soft-switched BDC has also been proposed and analyzed. Finally, a topology extension is made, and another fully soft-switched BDC is derived. A prototype, which interfaces the 400-48-V dc buses for the uninterrupted power supply system with a power rating of 500 VA, was developed to verify the validity and applicability of this proposed converter. The highest applicable conversion efficiencies for the bidirectional operational modes are exceeding 96%.

Hierarchical Recurrent Neural Encoder for Video Representation with Application to Captioning

Abstract: Recently, deep learning approach, especially deep Convolutional Neural Networks (ConvNets), have achieved overwhelming accuracy with fast processing speed for image classification. Incorporating temporal structure with deep ConvNets for video representation becomes a fundamental problem for video content analysis. In this paper, we propose a new approach, namely Hierarchical Recurrent Neural Encoder (HRNE), to exploit temporal information of videos. Compared to recent video representation inference approaches, this paper makes the following three contributions. First, our HRNE is able to efficiently exploit video temporal structure in a longer range by reducing the length of input information flow, and compositing multiple consecutive inputs at a higher level. Second, computation operations are significantly lessened while attaining more non-linearity. Third, HRNE is able to uncover temporal tran-sitions between frame chunks with different granularities, i.e. it can model the temporal transitions between frames as well as the transitions between segments. We apply the new method to video captioning where temporal information plays a crucial role. Experiments demonstrate that our method outperforms the state-of-the-art on video captioning benchmarks.

Ultrahigh Thermal Conductive yet Superflexible Graphene Films

Abstract: Electrical devices generate heat at work. The heat should be transferred away immediately by a thermal manager to keep proper functions, especially for high-frequency apparatuses. Besides high thermal conductivity (K), the thermal manager material requires good foldability for the next generation flexible electronics. Unfortunately, metals have satisfactory ductility but inferior K (≤429 W m−1 K−1), and highly thermal-conductive nonmetallic materials are generally brittle. Therefore, fabricating a foldable macroscopic material with a prominent K is still under challenge. This study solves the problem by folding atomic thin graphene into microfolds. The debris-free giant graphene sheets endow graphene film (GF) with a high K of 1940 ± 113 W m−1 K−1. Simultaneously, the microfolds render GF superflexible with a high fracture elongation up to 16%, enabling it more than 6000 cycles of ultimate folding. The large-area multifunctional GFs can be easily integrated into high-power flexible devices for highly efficient thermal management.