The exfoliation and identification of the two-dimensional (2D) single atomic layer of carbon have opened the opportunity to explore graphene and related 2D materials due to their unique properties. 2D materials are regarded as one of the most exciting solutions for next generation electronics and optoelectronics in the technological evolution of semiconductor technology. In this review, we focus on the core concept of "structure-property relationships" to explain the state-of-the-art of 2D materials and summarize the unique electrical and light-matter interaction properties in 2D materials. Based on this, we discuss and analyze the structural properties of 2D materials, such as defects and dopants, the number of layers, composition, phase, strain, and other structural characteristics, which could significantly alter the properties of 2D materials and hence affect the performance of semiconductor devices. In particular, the building blocks principles and potential electronic and optoelectronic applications based on 2D materials are explained and illustrated. Indeed, 2D materials and related heterostructures offer the promise for challenging the existing technologies and providing the chance to have social impact. More efforts are expected to propel this exciting field forward.

No abstract

Metamaterial assisted terahertz (THz) label-free bio-sensing has promising applications. However, the sensitive THz detection of highly absorptive liquid samples remains challenging. Here, we present a novel multi-microfluidic-channel metamaterial biosensor (MMCMMB) for highly sensitive THz sensing of small volume liquid samples. The multi-channels are set mostly in the strong electric field enhancement area of the metamaterial, which significantly decreases the liquid amount and enhances interaction between the sensing targets and the THz wave (thus increasing the sensitivity). The sensing results of isopropyl alcohol (IPA)-water mixtures and bovine serum albumin (BSA) solutions based on the bow-tie array metamaterial with multi-channels demonstrate the effectiveness of this proposed design and the great potential in THz bio-sensing. This design has the advantages of being highly sensitive, label-free, cost-effective, easy to operate and only needing a tiny liquid volume. Thus our device provides a robust route for metamaterial assisted THz label-free bio-sensing of liquid-based substances. IndexTerms-Terahertz spectroscopy, biomedical spectroscopy, metamaterial, microfluidic.

Transition metal dichalcogenides (TMDCs) have recently attracted growing attention in the fields of dielectric nanophotonics because of their high refractive index and excitonic resonances. Despite the recent realizations of Mie resonances by patterning exfoliated TMDC flakes, it is still challenging to achieve large-scale TMDC-based photonic structures with a controllable thickness. Here, we report a bulk MoS2 metaphotonic platform realized by a chemical vapor deposition (CVD) bottom-up method, supporting both pronounced dielectric optical modes and self-coupled polaritons. Magnetic surface lattice resonances (M-SLRs) and their energy-momentum dispersions are demonstrated in 1D MoS2 gratings. Anticrossing behaviors with Rabi splitting up to 170 meV are observed when the M-SLRs are hybridized with the excitons in multilayer MoS2. In addition, distinct Mie modes and anapole-exciton polaritons are also experimentally demonstrated in 2D MoS2 disk arrays. We believe that the CVD bottom-up method would open up many possibilities to achieve large-scale TMDC-based photonic devices and enrich the toolbox of engineering exciton-photon interactions in TMDCs.

No abstract

Evolution strategies have been demonstrated to have the strong ability to roughly train deep neural networks and well accomplish reinforcement learning tasks. However, existing evolution strategies designed specially for deep reinforcement learning only involve the plain variants which can not realize the adaptation of mutation strength or other advanced techniques. The research of applying advanced and effective evolution strategies to reinforcement learning in an efficient way is still a gap. To this end, this paper proposes a restart-based rank-1 evolution strategy for reinforcement learning. When training the neural network, it adapts the mutation strength and updates the principal search direction in a way similar to the momentum method, which is an ameliorated version of stochastic gradient ascent. Besides, two mechanisms, i.e., the adaptation of the number of elitists and the restart procedure, are integrated to deal with the issue of local optima. Experimental results on classic control problems and Atari games show that the proposed algorithm is superior to or competitive with state-of-the-art algorithms for reinforcement learning, demonstrating the effectiveness of the proposed algorithm.

In this paper we propose a time-variant photonic crystal, which can be formed by a stream of wave-length-scale microdroplets flowing through a microfluidic channel. The functionality stems from the photonic bandgap generated from the 1D periodic perturbation of refractive index. The periodicity, volume fraction and composition of both the dispersed and the continuous phases can be conveniently tuned in real time by hydrodynamic or pneumatic methods. By simulation, it is found that the time-variant nature of the proposed structure can induce an abnormal energy evolution, which is distinct from any existing photonic crystal structures. As a basic component for optofluidic systems, the droplet-based photonic crystal may find potential applications in light modulation and light confinement, and could be an ideal model for pursuing physical insights into time-variant optofluidic systems.

No abstract