ACS Photonics最新文献

Silicon photonic integrated circuits have been extremely well developed and have gradually moved toward large-scale production. However, the limitations of current scaling have forced researchers to explore new avenues to achieve more compact integration and to develop more cost-effective silicon photonics components. Silicon photonic FPGAs are more area-efficient and flexible compared to traditional on-chip optical circuits due to their reconfigurable nature, which allows for the optimization of silicon photonic devices after fabrication. This feature enables a wide range of applications and performance requirements to be met with a single chip design, thereby reducing costs and enabling the rapid prototyping of new photonic circuits. Here, leveraging ferroelectric-doped graphene into a silicon field programmable gate array, we propose a compact reconfigurable electro-optical device with superior nonvolatility and reconfigurability, broadening the range of applications for programmable silicon photonics. Nonvolatile multilevel memory with electrical write and optical readout is implemented. This innovative memory system supports 10 distinct levels of electro-optical storage, providing enhanced capacity and flexibility. Carrier-enhanced and -depleted modes can be reconfigured by electrical programming on the same optical logic gate. Reconfigurable logic computing in the electronic and optical domain that takes advantage of this feature is demonstrated. Our work provides a compact new approach for programmable electro-optic field programmable gate arrays with low power consumption.

An all-optical diffractive deep neural network (D²NN) consists of deep-learning-based design of passive diffractive layers and uses light to perform massive computations at the speed of light with zero extra power consumption, exhibiting advantages of large bandwidth, high interconnection, and parallel processing capability. In this paper, we introduce a novel approach utilizing a 5-layer all-optical D²NN constructed with photoinduced liquid crystal (LC) alignment technology to create LC-based tunable phase retarders as artificial neural layers. The D²NN architecture leverages microscale multidomain LC retarders as optical neurons to manipulate the geometric phase of incident light. We systematically simulate pixel-level displacements to enhance alignment tolerance during experiments, achieving robust resilience against misalignment interference with a 2-pixel tolerance in the x and y directions. By actively tuning the LC retarders with external voltage, we optimize the alignment strategy for all network layers, incorporating specially designed concave or convex lenses at each LC retarder for precise alignment in the x, y, and z directions. Through training with a handwritten dataset from MNIST, the D²NN demonstrates a simulated accuracy of 94.17% with a 2 pixel misalignment tolerance. Experimental validation achieves a classification accuracy of 89% with 500 random digits from the test dataset. This research showcases the potential for network miniaturization, integration, and compatibility with visible light, underscoring the practical applicability of an all-optical D²NN for diverse real-world applications.

The relaxation of organic polaritons is a key aspect for understanding nonequilibrium bosonic condensation in organic microcavities. In this work, dual-branch vibrational quanta-assisted polariton condensation is experimentally observed in organic single-crystal-filled microcavities. By precisely modulating the thickness of the planar optical resonator, we can tune the ground states of two lower polariton branches to perfectly match the energies of two vibrational modes and consequently trigger polariton condensation in both branches. These condensates have nearly identical thresholds. Dynamical analysis indicates that efficient energy relaxation of the photogenerated excitons to the two vibrational modes through the nonradiation of two separately vibrational quanta enables polaritons to populate the ground states of these two lower polariton branches. Our work is evidence of the importance of the vibrational quanta relaxation mechanism for polariton condensation and provides a pathway for multicolor polariton condensation and future laser displays.

Scattering spectra from radiative non-Hermitian systems often exhibit intricate line shapes, where peaks typically garner the most attention for mode identification. However, in multimode systems, the valleys between these peaks may contain valuable information. This “coupling” feature arises from the nonorthogonality of modes in both far and near fields, giving rise to diverse and complex spectra-hole-burning (SHB) patterns. Traditionally, the interpretation of these SHBs has focused on Rabi splitting or Fano resonances, often concentrating solely on either far-field interference or near-field coupling. However, it is essential to recognize that both phenomena coexist in non-Hermitian scatterings. In this study, we develop a quantitative quantum model to probe scattering SHB by simultaneously extracting near-field coupling rates between system quasinormal modes, nonradiative decay rates into a heat reservoir, and radiative decay rates into a vacuum reservoir for far-field interference. We apply our model to illustrate the concept of geometric engineering in tuning the ratio of far-field interference and near-field coupling, exemplified by a silver dimer transitioning from cube-dimer to sphere-dimer or cube-dimer to nanocube-on-mirror configurations. Through this, we establish a universal design guideline for non-Hermitian scattering by creating a two-mode SHB library based on arbitrarily tunable far-field interference and near-field coupling. The developed model serves as a generalized diagnostic tool for probing the SHB mechanisms in all types of non-Hermitian scattering problems, promising to advance our understanding of intricate phenomena and facilitate the design of tailored optical devices with enhanced performance and functionality.