Chip最新文献

On-chip topological nanophotonic devices, which take photons as information carriers with topological protection during light propagation, have great application potential in the next generation photonic chips. The topological photonic states enable the nanophotonic devices to be robust and stable, immune to scattering even with imperfect structures. The development, opportunities and challenges of the on-chip topological nanophotonic devices have attracted great attention of scholars, and desired to be known. In this review, topological devices were introduced in the order of functionalities on an integrated photonic chip, i.e. topological light source, topological light waveguiding, topological light division and selection, topological light manipulation and topological light detecting. Finally, we gave outlooks for predicting and promoting the performances of on-chip topological nanophotonic devices from the angles of non-Hermitian systems, non-Abelian topology, metasurfaces, intelligent algorithms and multiple functional topological nanophotonic integration. This review provides rich knowledge about on-chip topological nanophotonic devices. The insights in this paper will spark inspiration and inspire new thinking for the realization of topological photonic chips.

Thanks to the increasingly high standard of electronics, the semiconductor material science and semiconductor manufacturing have been booming in the last few decades, with massive data accumulated in both fields. If analyzed effectively, the data will be conducive to the discovery of new semiconductor materials and the development of semicondulctor manufacturing. Fortunately, machine learning, as a fast-growing tool from computer science, is expected to significantly speed up the data analysis. In recently years, many researches on machine learning study of semiconductor materials and semiconductor manufacturing have been reported. This article is aimed to introduce these progress and present some prospects in this field.

A lithium-niobate-on-insulator (LNOI) electro-optic (EO) modulator based on a 2 × 2 FP-cavity was designed and realized with an ultra-compact footprint and an ultra-high bandwidth. A comprehensive analysis on the present LNOI FP-cavity modulator was conducted to reveal the dependence of modulation bandwidth and modulation efficiency on the cavity Q-factor and the operation wavelength detuning to the resonance. In particular, the 2 × 2 FP cavity was designed to achieve an optimal Q factor by reducing the reflectivity of reflectors and the cavity length, thus reducing the photon lifetime in the cavity . An ultra-short effective cavity length of only∼ 50 µm was achieved for the designed LNOI FP-cavity modulator, with itsfootprint being as compact as ∼ 4 × 500 µm². It was demonstrated theoretically that the modulation bandwidth could be improved significantly to be over 200 GHz by utilizing the peaking enhancement effect. The fabricated device exhibited an excess loss of ∼ 1 dB and an extinction ratio of ∼ 20 dB in experiments, while the measured 3-dB bandwidth was higher than 110 GHz (beyond the maximal range of the facilities in experiments). Up till now, to our best knowledge, this has been the first LNOI microcavity modulator with a bandwidth higher than 110 GHz. Finally, high-quality eye-diagrams of 100 Gbps on-off keying (OOK) and 140 Gbps 4-pulse amplitude modulation (PAM4) signals were demonstrated experimentally, and the energy consumption for the OOK signals was as low as 4.5 fJ/bit.

Smart dust, which refers to miniaturized, multifunctional sensor motes, would open up data acquisition opportunities for Internet of Things (IoT) and Environmental protection applications. However, critical obstacles remain challenging in the integration of high-density sensors, further miniaturization of device platforms, and reduction of cost. Here, we demonstrate the concept of smart digital dust to address these problems, the results of which combine the benefit of (i) maturity of complementary metal-oxide semiconductor (CMOS) processing approaches and (ii) unique form factors of emerging flexible electronics. As a prototype for smart digital dust, we present a millimeter-scale multifunctional optoelectronic sensor platform consisting of high-performance optoelectronic sensor cores and commercially available integrated-circuit components. The smart material-assisted optoelectronic sensing mechanism enables real-time, high-sensitivity hydrogen, temperature, and relative humidity (RH) sensing based on a single chip with ultralow power consumption. Such a microsystem presented here introduces a viable solution to the multifunctional sensing need of IoT and could serve as a building block for the rapidly evolving future framework of smart dust.

Metamaterial devices (metadevices) have been developed in progress aiming to generate extraordinary performance over traditional devices in the (sub-)terahertz (THz) domain, and their planar integration with complementary-metal-oxide-semiconductor (CMOS) circuits pave a new way to build miniature silicon plasmonics that overcomes existing challenges in chip-to-chip communication. In an effort towards low-power, crosstalk-tolerance, and high-speed data link for future exascale data centers, this article reviews the recent progress on two metamaterials, namely, the spoof surface plasmon polaritons (SPPs), and the split-ring resonator (SRR), as well as their implementations in silicon, focusing primarily on their fundamental theories, design methods, and implementations for future THz communications. Owing to their respective dispersion characteristic at THz, these two metadevices are highly expected to play an important role in miniature integrated circuits and systems toward compact size, dense integration, and outstanding performance. A design example of a fully integrated sub-THz CMOS silicon plasmonic system integrating these two metadevices is provided to demonstrate a dual-channel crosstalk-tolerance and energy-efficient on-off keying (OOK) communication system. Future directions and potential applications for THz metadevices are discussed.

A strategy for realizing a microchip-scale single-photon digital loop buffer controlled by low-voltage electronic signals was studied in the context of integrated photonics. A potential implementation for bridging a gap between other technologies used a recirculating loop architecture based on advances in low-loss passive waveguides and a fast electro-optic add-drop switch. Although the requirements of single-photon buffers are demanding, our analysis suggested that a voltage-controlled, room-temperature catch-and-store short-term quantum memory for light on a chip may be feasible in certain regimes.

Photodetectors are finding various potential applications in sensing and detection, information communication, light-emitting diode, optical modulators, ultrafast laser, etc. Molybdenum disulfide (MoS₂) has sparked great interest given its unique crystal phase, flexible preparation, structural stability, and regulable photoelectronic features. Therefore, the MoS₂-based photodetector is demonstrated to be an excellent device fabrication platform to explore underlying sensitive detection, broadband optical detection, high-speed response, low-power consumption, two-dimensional integrated circuit, and its synergetic mechanism, which is also proved to be an excellent candidate for next-generation optoelectronics. This review summarizes the structural, optical, and transport features of MoS₂. Then the working mechanisms and figures of merit are explored for the MoS₂ detector. Further, the detector modulation strategies are introduced in detail about layer-number engineering and chemical doping engineering. Afterward, the recent heterostructure assembling strategies (MoS₂/nD, n=0,1,2,3) of detector architectures are classified based on flexible van der Waals assembling. Finally, the future direction of MoS₂ photodetectors is discussed, which can be delivered as a feasible guideline in two-dimensional photodetector and integrated circuit fields.

The integration of qubits with long coherence times and functional quantum devices on a single chip, and thus the realization of an all-solid-state quantum computing chip, is an important goal in current experimental research on quantum information processing. Among various quantum platforms, a series of significant progresses have been made in photonic quantum chips and superconducting quantum chips, while both the number of qubits and the complexity of quantum circuits have been increasing. Although these two chip platforms have respective unique advantages and potentials, their shortcomings have been gradually revealed and need to be solved. By introducing phonon-integrated devices, it is possible to combine all unsuspended phononic, photonic, and superconducting quantum devices organically on the same chip to achieve coherent coupling among them. Here, we provide a prospect and a short review on the integrated photonic, superconducting, and hybrid quantum chips for quantum information processing.

With the development of artificial intelligence and the Internet of Things, the number of sensory nodes is growing rapidly, leading to the exchange of large quantities of redundant data between sensors and computing units. In-sensor computing schemes, which integrate sensing and processing, have provided a promising route to addressing the sensing/processing bottleneck by reducing power consumption, time delay and hardware redundancy. In this study, an in-sensor computing architecture involving a photoelectronic cell based on a wafer-scale two-dimensional MoS₂ thin film was demonstrated. The MoS₂ photodetector cell used a top-gate device structure with indium tin oxide (ITO) as the transparent gate electrode, which exhibited high air-stability and a high photoresponsivity (R) up to 555.8 A W^–1 at an illumination power density (P_in) of 16.0 µW cm^–2 (λ = 500 nm). Additionally, a MoS₂ photodetector array with uniform photoresponsive characteristics was achieved. Furthermore, logic gates, including inverter, NAND, and NOR, were achieved based on MoS₂ photodetector cells. Such multifunctional and robust in-sensor computing was ascribed to the uniform wafer-scale MoS₂ film grown by atomic layer deposition (ALD) and the unique device structure. Because the detection of optical signals and logic operations were achieved through MoS₂ photodetector cells with area efficiency, the proposed in-sensor computing device paves the way for potential applications in high-performance, integrated sensing and processing systems.

Invasive neural probes are one of the most critical components in the intracortical neural signal recording system. However, they can cause brain damage and tissue response during and after implantation. Thus, neural probes with high flexibility, biocompatibility, and simple implantation methods are required in brain research. Here we present a novel approach to fabricating a multilayer flexible tassel-type neural probe using low-cost maskless laser direct-write lithography, combined with straightforward release and assembly methods to prepare a whole implantation system. The probe has 32 recording electrodes with an area of 8 × 8 µm², arranged into two rows of different depths and 16 separated shanks, aiming at the neural signal recording in an extensive range. Polyimide and gold are used as the insulating and conductive layers, respectively. With the help of a polyethylene glycol (PEG) coating, the tassel structure was mechanically enhanced for successful implantation, and our morphology characterization showed that the diameter of the coated probe was less than 50 µm. Mechanical property tests also proved that it had the necessary stiffness for brain implantation. Afterwards, electrochemical tests were carried out, which showed that the probe had a rather low impedance after a simple gold electroplating. Finally, in vivo experiments demonstrated our probe can be successfully used in neural recording.