Sub-bandgap photodetection on all-silicon platforms through harvesting hot electrons from the nonradiative decay of surface plasmons presents significant potential for CMOS-compatible optoelectronics, enabling self-powered room-temperature operation. However, the device performance is still limited by the high noise equivalent power (NEP), and conventional single narrowband photodetection responses cannot meet the requirements for encrypted wavelength-division demultiplexing in telecommunication. We address these challenges by developing a tunable dual-wavelength hot-electron photodetector with a record-low NEP performance across the telecommunication bands. Our design integrates plasmonic nanoholes with silicon nanowires, establishing a synergistic dual-resonance mechanism─enhancing plasmonic field intensity by 2 orders of magnitude while forming quasi-omnidirectional Schottky interfaces for efficient hot carrier confinement and collection. The device exhibits a dual narrowband photoresponse with spectral tunability across telecommunication wavelengths (O to L bands), achieving unprecedentedly low NEP values ranging from 8.8 × 10–13 W·Hz–0.5 to 2.3 × 10–12 W·Hz–0.5. As a proof of concept, we demonstrate encrypted demultiplexing in telecommunications using a dual-wavelength photodetector. This work paves the way for low-cost, all-silicon sub-bandgap photodetection and provides a scalable platform for integrated silicon photonics in secure communication systems and on-chip spectral sensing applications.

Visualizing the behavior of complex laser systems isan outstandingchallenge, especially in the presence of nonlinear mode interactions.Hidden features, such as the gain distributions and spatial localizationof lasing modes, often cannot be revealed experimentally, yet theyare crucial to determining the laser action. Here, we introduce anexperimental lasing spectroscopy method that visualizes the gain profilesof the modes in a complex, disorderly coupled microring array laserusing an artificial neural network. The spatial gain distributionsof the lasing modes are reconstructed without prior knowledge of thelaser device. We further extend the neural network to a tandem neuralnetwork that can control the laser emission by matching the modalgain/loss profile to selectively enhance the targeted modes. Thismode visualization method offers a new approach to extracting hiddenspatial mode features from photonic structures, which could improveour understanding and control of complex photonic systems.

The electron–hole exchange interaction and phonon-induced spin relaxation are critical factors for the depolarization of the valley information in two-dimensional systems. As for the transition metal dichalcogenide (TMDC) ternary alloys, prominent phonon-mediated intervalley scattering is expected due to the piezo-effect and lattice deformation. In our work, we demonstrate the dissimilar lifetime of the valley depolarization and intervalley scattering of the interlayer excitons in the TMDC ternary-based vdWHSs, where the intervalley scattering time of ∼20 ns in the 1L-WSSe/1L-WSe2 vdWHS is significantly greater than the valley depolarization lifetime of ∼2 ns. Phonon-mediated pathways are crucial for intervalley scattering in the presence of a magnetic field. The intervalley scattering can be enhanced by interlayer hopping and suppressed by encapsulation. The encapsulation-suppressed intervalley scattering enables the exciton valley Hall effect in the presence of a pseudomagnetic field in the 1L-WS2/1L-WSSe/1L-WSe2 vdWHS. Our results provide deeper insights into the valley depolarization dynamics in TMDC ternary-based vdWHSs for developing valleytronic devices.

Neuromorphic vision systems have the capacity to simulate the perception and processing of visual information by visual cells in the retina. However, when confronted with the challenges posed by a substantial volume of complex data and complex environments, traditional neuromorphic vision systems are unable to handle redundant signals of overenhancement and suppression. These systems are required to face the considerable challenges posed by complicated circuits and algorithms. In this paper, we present an adaptive synaptic transistor with a built-in heterojunction that can switch between three modes of synaptic excitation-inhibition effect, excitation-adaptive, and inhibition-adaptive photoconductivity effect by utilizing light switching and wavelength change. The device can complete the entire adaptation process from excitation sensitization to self-adaptation to the initial current in 1 s, and from excitation sensitization to adaptation in 3.2 s. The adaptation speed is superior to that of the human eye (5 min). The combination of convolutional neural networks (CNNs) with adaptive synaptic transistors has yielded the development of an advanced neuromorphic vision system. This system exhibits fast self-adaptation and static image recognition and classification capabilities, with a recognition rate that exceeds 90%, thereby facilitating the advancement of next-generation neuromorphic vision systems.

Self-trapped exciton (STE) emission stands out for its large Stokes shift, broad emission spectrum, and high photoluminescence quantum yield (PLQY). In this work, we developed an Sb3+-doped low-dimensional layered halide, Cs2CdBr2Cl2, as a system to achieve efficient broadband emission. By fine-tuning the Sb3+ doping concentration, we induced moderate lattice distortion, resulting in a suitable Huang–Rhys factor (S = 18.00) and a high exciton binding energy (Eb = 144.73 meV), which collectively enabled strong yellow STE emission with a PLQY of nearly 80%. Experimental and theoretical investigations reveal that the emission originates from STE states associated with distorted [SbCl4Br2]3– octahedra. This highly efficient luminescent material, when combined with commercial phosphors, achieves a high color rendering index (Ra) of 90.8 and exhibits excellent radiation resistance and great potential for X-ray imaging applications.

Achieving stray-light-free on-chip spectral detection with high resolution across a broad wavelength range is of paramount importance for practical applications, as stray light in spectral measurements often introduces spurious signals that obscure genuine spectral detection. To address this challenge, we propose a novel architecture that combines narrow-band filtering using a tunable high-Q microring resonator with wide-passband filtering through cascaded multimode-waveguide grating (MWG) filters. The high-Q microring, realized with broadened waveguides, generates multiple high-fineness transmission peaks spaced by the free spectral range (FSR) across the entire wavelength range. These peaks are selectively isolated one by one using corresponding wide-passband MWG filters with precisely tailored central wavelengths and bandwidths, effectively overcoming the FSR limitation of the microring. A key point lies in the design of the MWG filters, which exhibit a single passband over an exceptionally large wavelength range of 1200–1700 nm, ensuring the complete suppression of out-of-band stray light. Furthermore, the monolithic integration of on-chip Ge/Si photodetector arrays minimizes the disturbance of environmental stray light. Experimentally, we demonstrate a fully packaged silicon photonic spectrometer system featuring a high spectral resolution of 0.15 nm across the wavelength range of 1263–1355 nm, with a minimum detectable optical power of 1 nW. The system successfully identifies the absorption peaks of hydrogen fluoride gas, demonstrating its capability for practical applications such as smartphone-based sensing modules, gas monitoring, and optical coherence tomography (OCT). This work represents a significant advancement in on-chip spectrometers, offering a stray-light-free, high-performance, and compact solution for practical applications.

Terahertz (THz) spectroscopy can obtain characteristic fingerprint spectra by detecting the structural vibrations of molecules. Nanophotonic enhancement and subwavelength field confinement have expanded the application scope of this technology. However, current methods still require complex spectroscopic equipment and face obstacles in the trace analysis of analytes and broadband detection. Here, we introduce a novel metasurface design approach to detect the broadband absorption spectra of molecules, which combines the simplest detection method (transmission metasurface chipset) with the chemical specificity of THz spectroscopy. Specifically, we designed a metasurface chipset based on Fano resonance, capable of providing frequency-selective spectra and surface-sensitive resonances between 0.18 and 1.20 THz. We used this method to detect distinct absorption signatures of different interacting analytes, including l-tyrosine (5 and 10 mg/mL) and bovine hemoglobin. By combining the molecular absorption spectra with the transmission spectra of each metasurface chip, broadband detection was achieved through the design of the metasurface chipset, paving the way for the field-deployable applications of miniature THz spectroscopy devices.

Displays for future technologies such as augmented and virtual reality require ultrahigh resolution and efficient power consumption. While micro-LEDs can attain the required sizes for these displays, at present, their efficiencies are severely lacking, especially the red-emitting ones. Here, we show that by utilizing a combination of strategies for enhancing the carrier injection efficiencies, including nitrogen polarity, Mg-doped AlGaN electron-blocking layer, and a p-GaN layer with gradient doping, unprecedentedly high wall-plug efficiency for a submicron red LED has been achieved. Light-emitting diodes with submicrometer dimensions exhibited an emission wavelength of ∼650 nm, reaching a peak external quantum efficiency of ∼12.3%, and a wall-plug efficiency of ∼11.4%, corresponding to a peak electrical efficiency of ∼92%. Through this work, we demonstrate that the hitherto low electrical efficiency of nanowire-based devices can be overcome through careful design of the device heterostructure and that such devices can form the foundation of future micro-LED displays.

Spectral filters with wide bandwidth are required by many critical photonics applications such as communications, nonlinear optics, and spectroscopy. In this study, an adiabatic spectral filter with a flat-top spectral response and wide operational bandwidth is demonstrated on the lithium-niobate-on-insulator platform. By utilizing the spectrally selective waveguide cross-section, we achieve high extinction spectral filtering through a conventional adiabatic asymmetric directional coupler, which is connected to a mode demultiplexer for downstream processing. The fabricated devices exhibit an excess loss as low as 0.4 dB and a sidelobe suppression ratio of 20 dB. A remarkable 1-dB bandwidth of 75 nm is also achieved. By cascading two adiabatic spectral filters, we further demonstrate a wavelength triplexer for the 1577:1490:1342 nm downstream channel separation in next-generation passive optical networks, achieving excess losses of 0.4:0.7:0.4 dB with crosstalk levels of −25:–35:–16 dB. Our adiabatic spectral filter does not include any interference structures, making it inherently suitable for ultrabroadband filtering operations and facilitating future innovations in on-chip ultrabroadband signal processing.

In this work, we demonstrate a reconfigurable cavity-optomechanical array based on multiple mechanically coupled microbubble resonators (MBRs). By manipulating movable microdroplets in the microfluidic channels between adjacent MBRs, we dynamically tune both the mechanical coupling rates and the effective motional masses, thereby achieving parametric control over the optomechanical radial-breathing supermodes (RBSMs). This fluidic approach circumvents the fixed-coupling limitations of conventional solid-state resonator arrays, enabling phonon-mediated energy transfer and strong multimode cooperativity. Our platform thus represents a conceptual leap from single-resonator optomechanics to an optofluidically reconfigurable multiresonator architecture, opening new pathways for ultrasensitive detection. Experimentally, our tunable MBR array attains a record aerostatic pressure resolution of 1.6 μbar─an improvement of nearly 3 orders of magnitude─while suppressing temperature cross-sensitivity by a factor of 10. These advances not only demonstrate unprecedented control in cavity optomechanics but also lay the groundwork for applications such as topologically robust phononic sensing, distributed photonic sensor networks, and multiphase sensor schemes. Moreover, this platform offers a versatile testbed for exploring collective multimode optomechanical phenomena.