Electrochemical splitting of water is an attractive way to produce hydrogen fuel as a clean and renewable energy source. However, a major challenge is to accelerate the sluggish kinetics of the anodic half‐cell reaction where oxygen evolution reaction (OER) takes place. Here, a seamlessly conductive 3D architecture is reported with a carbon‐shelled Ni‐Co nanowire network as a highly efficient OER electrocatalyst. Highly porous and granular Ni‐Co nanowires are first grown on a carbon fiber woven fabric utilizing a cost‐effective hydrothermal method and then conductive carbon shell is coated on the Ni‐Co nanowires via glucose carbonization and annealing processes. The conductive carbon layer surrounding the nanowires is introduced to provide a continuous pathway for facile electron transport throughout the whole of the integrated 3D catalyst. This 3D hierarchical structure provides several synergistic effects and beneficial functions including a large number of active sites, easy accessibility of water, fast electron transport, rapid release of oxygen gas, enhanced electrochemical durability, and stronger structural integrity, resulting in a remarkable OER activity that delivers an overpotential of 302 mV with a Tafel slope of 43.6 mV dec−1 at a current density of 10 mA cm−2 in an alkaline medium electrolyte (1 m KOH).

No abstract

Hot electrons and surface-plasmon-driven chemistry are amongst the most actively studied research subjects because they are deeply associated with energy dissipation and the conversion processes at the surface and interfaces, which are still open questions and key issues in the surface science community. In this topical review, we give an overview of the concept of hot electrons or surface-plasmon-mediated hot electrons generated under various structural schemes (i.e. metals, metal-semiconductor, and metal-insulator-metal) and their role affecting catalytic activity in chemical reactions. We highlight recent studies on the relation between hot electrons and catalytic activity on metallic surfaces. We discuss possible mechanisms for how hot electrons participate in chemical reactions. We also introduce controlled chemistry to describe specific pathways for selectivity control in catalysis on metal nanoparticles.

No abstract

Enhancement of the photocatalytic activity of water splitting on 5–30 nm Au nanoparticles/TiO2 nanotube arrays was demonstrated.

This paper describes a new mechanism for boundary migration, namely the mixed control mechanism (either diffusion or interface reaction), and the principle of microstructural evolution in polycrystals. The basis of the mechanism and the microstructural prediction of the principle are explained, including key experimental results that support the mechanism and the principle. The solid state single crystal growth method, a new technique of single crystal fabrication, is described as an application example of the microstructural evolution principle. Future research directions in the subject areas are also given.

No abstract

Many real-world reinforcement learning tasks require multiple agents to make sequential decisions under the agents' interaction, where well-coordinated actions among the agents are crucial to achieve the target goal better at these tasks. One way to accelerate the coordination effect is to enable multiple agents to communicate with each other in a distributed manner and behave as a group. In this paper, we study a practical scenario when (i) the communication bandwidth is limited and (ii) the agents share the communication medium so that only a restricted number of agents are able to simultaneously use the medium, as in the state-of-the-art wireless networking standards. This calls for a certain form of communication scheduling. In that regard, we propose a multi-agent deep reinforcement learning framework, called SchedNet, in which agents learn how to schedule themselves, how to encode the messages, and how to select actions based on received messages. SchedNet is capable of deciding which agents should be entitled to broadcasting their (encoded) messages, by learning the importance of each agent's partially observed information. We evaluate SchedNet against multiple baselines under two different applications, namely, cooperative communication and navigation, and predator-prey. Our experiments show a non-negligible performance gap between SchedNet and other mechanisms such as the ones without communication and with vanilla scheduling methods, e.g., round robin, ranging from 32% to 43%.