The widespread pollution of water environments by microplastics (MPs) is a critical global issue. The identification of MPs remains challenging, primarily due to their limited texture and color information. To address this issue, this Letter proposes a learning-based approach that incorporates what we believe to be a novel geometric polarization characterization of MPs with quantum neural networks (QNNs). Our method comprises three stages. First, we establish a novel Polarization Angle-Weighted Geometric (PAWG) feature derived from the geometric relationship of polarization information. Second, a novel polarization-driven quantum convolutional network (PQCNet) is designed based on pure variational quantum circuits. Finally, the PAWG features are fed into the PQCNet to train an effective variational quantum classifier for intelligent identification of MPs. Experimental results show that the proposed method yields an average accuracy of approximately 95% across three MP types (i.e., PP, LDPE, and HDPE) in water. Further, the model with only 0.0023 M parameters enables rapid inference with an average prediction time of 11.3 ms per sample. Such a success rate and computational efficiency indicate that our method can be used to rapidly identify MPs.

In this study, we propose a novel, to the best of our knowledge, terahertz parametric generator capable of generating a widely tunable THz-wave using a single fixed-wavelength pump source. The pump beam is split into two beams, which are made to intersect inside a nonlinear optical crystal. The two resulting broadband parametric fluorescence components spatially overlap and mutually enhance each other at the intermediate angle between the two pump beams, and the corresponding wavelength component serves as a seed for Stokes beam generation. By varying the angle between the two pump beams, the wavelengths of the Stokes beam and the THz-wave can be controlled, and frequency tunability over the range of 1.0-2.2 THz was demonstrated.

A novel, to the best of our knowledge, millijoule-class femtosecond optical parametric amplification-based source is implemented as a seeder in the high-repetition-rate HF-PW laser of ELI ALPS. High energy seeding results in a bandwidth corresponding to a sub-19 fs transform limit, significant gain reduction in the Ti:Sa amplifier chain, and an extremely low amplified spontaneous emission relative intensity of 2 × 10[-14] with 10 J compressed energy. The combination of the achieved parameters with the routinely available 2.5 or 5 Hz repetition rate offers unprecedented opportunities for the generation of particles and high harmonics.

Modeling wavelength-dependent aberrations in broadband imaging systems requires high-dimensional parameter spaces that render Monte Carlo (MC) tolerance analysis intractable. We propose a low-dimensional spectral parameterization of Zernike coefficients that compresses broadband aberration modeling from 288 to 36 parameters and enables tractable MC tolerance analysis for manufactured systems. For a dispersion-dominated DD-CASSI system, we further introduce a Cauchy-derived Hybrid basis whose 1/λ[2] dependence captures defocus-driven dispersion. Compared with nominal-PSF training, MC-augmented training with the proposed parameterization improves reconstruction PSNR by 1.06 dB, of which 0.48 dB is attributable to the Hybrid basis over a polynomial baseline.

This Letter addresses Rayleigh-Brillouin scattering (RBS) modeling methodology for binary mixtures of nitrogen and gaseous hydrocarbon fuels in the kinetic scattering regime, with its extension to filtered Rayleigh scattering (FRS). Modeled mixture FRS signals were generated by summing mole-weighted FRS signals of mixture constituents, with the required RBS lineshapes for the pure species modeled using the widely accepted Tenti S6 gas kinetics model. The modeled FRS mixture signals were then compared with measurements, and noticeable discrepancies up to 21% were observed. Such deviations demonstrate that the conventional FRS modeling framework for the mixtures of complex hydrocarbon fuels in the kinetic scattering regime requires refinement, with a specific treatment of diffusive effects.

Lensless imaging is attractive for its compact and low-cost design. However, high-accuracy and robust phase retrieval from a single pattern remains challenging and critically relies on accurate priors. Here, we propose a parallel consensus optimization termed PaCO to achieve robust and high-accuracy lensless imaging. PaCO decomposes the ill-posed problem into parallel subproblems with a distributed computational architecture. Each subproblem is handled by dedicated solvers that respectively enforce data fidelity, Fourier support, and spatial smoothness. The local solutions are aggregated into a consensus estimate. Experiments on a mask-based system demonstrate that PaCO can reconstruct objects without precise support or distance priors and accurately retrieve vortex and binary-phase-plate wavefronts, while conventional methods suffer from reconstruction instability. Statistical tests confirm that PaCO improves the success rate from 74% to 96.8% while halving the processing time. The parallel architecture of PaCO is inherently CUDA-friendly, holding strong potential for real-time processing in extreme-scale imaging workflows.

We propose an environment-coupled dynamic model to analyze the mode splitting of orbital angular momentum (OAM) beams propagating in complex environments and suppress the mode splitting through pre-compensated OAM. The feasibility of the model in analyzing mode splitting and splitting suppression was demonstrated through spatial light modulators, generated turbulence, and water flow experiments. Furthermore, we designed the mode content and relative argument for multiple-emitted-beam multiplexing and realized binary encoding based on mode content control. This study provides a physically analyzable approach for the stable propagation of OAM beams passing through an inhomogeneous medium, and also provides a research paradigm for the propagation analysis of other structured beams in complex environments.

By taking advantage of the unique structure of Ba3W2O9, a positioning restriction is applied to precisely control doped Mn to form isolated Mn[4+] ions or Mn[4+]-Mn[4+] ion pairs, thereby achieving the dual emission centers. The optimized phosphor (Ba3W1.97O9:0.03Mn[4+]) boasts excellent performances of a broadband near-infrared emission with the full width at half maximum of 170 nm, internal quantum efficiency of 74%, and external quantum efficiency of 43%. The emission peaks with wavelengths greater than 800 nm were verified to originate from the luminescence of Mn[4+]-Mn[4+] ion pairs through electron paramagnetic resonance, concentration gradient, and lifetime decay.

Spectral mismatch between on-chip emitters and photodetectors remains a critical bottleneck for AlGaN-based deep-ultraviolet (DUV) monolithic photonic integration. Here, a 250 nm AlGaN DUV monolithic integrated photonic chip with improved emission-detection spectral matching is demonstrated. By tailoring the aluminum content in the quantum barriers, the detector absorption edge is redshifted due to the enhanced quantum-confined Stark effect (QCSE), while the emission peak is blueshifted owing to the carrier screening effect (CSE) and strengthened quantum confinement effect (QCE). This spectral engineering broadens the emission-detection overlap region from 19 to 35 nm and enables complete coverage of the defined electroluminescence (EL) spectral width by the detector responsivity window. In addition, a photocurrent model based on the interplay between photovoltaic current and external-field-induced current is established to describe the bias-dependent on-chip signal conversion. On-chip DUV optical communication is further verified by 50 Mbps pseudorandom binary sequence transmission. These results provide a practical route for overcoming spectral mismatch in AlGaN DUV monolithic photonics and open a route toward high-performance integrated systems for optical communication and integrated photonic sensing.

We demonstrate an LSTM-assisted optical frequency domain reflectometry (OFDR) method for differential modal group delay (DMGD) characterization of short few-mode fibers. Sparse phase-domain resampling is first applied to compensate for laser sweep nonlinearity with a low sampling rate. Then, the resampled beat signal is extrapolated by an LSTM network to improve the resolution. In a 5-m single-mode fiber, the distance resolution is improved from 4.95 mm to 0.44 mm with a 20.74-GHz sweep bandwidth. The method resolves the overlapped LP01 and LP11 modal peaks in 1-m and 6-m two-mode fibers, yielding a differential modal group delay of 1.911 ps/m for the 1-m fiber, in good agreement with the reference values. A 2.2-m four-mode fiber is further tested to demonstrate the applicability of the method. It provides a route for modal characterization of short fibers under a limited sweep bandwidth.

We report a high-power, high-repetition-rate picosecond optical parametric oscillator (OPO) based on cadmium silicon phosphide CdSiP2 (CSP) nonlinear crystal, synchronously pumped by a Yb-fiber laser at ~78 MHz, delivering record idler average power in the deep-infrared (deep-IR). The OPO generates 243 mW at 6249 nm, maintaining >150 mW across nearly 75% of the idler tuning range from 6199 to 6760 nm, along with simultaneous signal tuning from 1262 to 1284 nm. The generated idler exhibits an excellent passive power stability better than 1.3% rms over 6 h, while the near-IR signal shows <0.5% rms stability with 6.2 ps pulses centered at 1281 nm. The unprecedented output power in deep-IR is achieved by the systematic investigation and careful optimization to minimize thermal effects through focusing relaxation and efficient cavity design. This is the highest, to the best of our knowledge, deep-IR average power ever achieved in the 6-7 µm wavelength range.

To meet the demand for stable trapping of multiple particles in complex environments and efficient design of multiple optical bottles, Airyprime chain optical bottle (ACOB) beams are proposed, and their structural features are efficiently designed using a neural network. The evolution properties of ACOB beams and the formation mechanism of chain optical bottles are analyzed, revealing the relationships between the number, maximum size, and formation position of the optical bottles and the optical field parameters. The neural network is trained to learn the nonlinear mapping from the optical field parameters to the structural features of the chain optical bottles. Experimental results further verify the ability of the chain optical bottles to stably trap multiple particles. This study provides new insights into the efficient design and intelligent control of complex optical bottle beams.

The RGB (de)multiplexer is a key component in laser display systems and demands advanced design solutions that achieve high efficiency within a small form factor. Here, a CMOS-compatible silicon nitride (SiN) visible light photonic integrated circuits (PICs) platform is employed to successfully demonstrate an ultracompact RGB (de)multiplexer, featuring a footprint as small as 86 × 2.3 µm[2]. Leveraging a novel, to the best of our knowledge, dual-segmented multimode waveguide design, the device achieves a low insertion loss of ~1-2 dB at target wavelengths and sets the stage for monolithic integration of laser sources on the same PIC chip. This breakthrough further paves the way for the development of integrated, miniaturized CMOS-compatible RGB light engines offering superior brightness, high resolution, and a wide field-of-view. We anticipate that our design can contribute significantly to the development of immersive AR/VR/MR display technology, while simultaneously extending its impact into sensing and visible light communication applications.

Breast tumor margin assessment requires rapid identification of regions of interest (ROIs) and detailed visualization of their microstructural alterations. Here, we report position-tracked handheld optical coherence tomography (PH-OCT) guided by projected wide-field near-infrared autofluorescence (PW-NIRAF) imaging to meet these requirements. The PW-NIRAF system comprises a near-infrared (NIR) imaging camera and a projector with calibrated fields of view, enabling ROI identification and projection of the corresponding ROI map over a 10 × 10 cm[2] field of view (FOV). The PH-OCT system consists of a handheld gradient-index-rod-lens-based OCT probe for volumetric visualization of ROI microstructure, with each OCT volume covering 4 × 4 × 1 mm[3] at 9.66 μm axial and 17.54 μm lateral resolution. Blue and green tapes used as color markers on the OCT-probe housing, together with a color camera and a deep-learning-based YOLOv11-pose model, enable real-time probe-tip localization for OCT volume acquisition across PW-NIRAF-projected ROIs, providing depth-resolved microstructural information beyond wide-field NIRAF contrast. Color-camera/projector calibration further enables back-projection of OCT-derived tumor-region maps onto the tissue surface. Validation on ex vivo breast tissue lays the foundation for future intraoperative breast tumor margin assessment using PW-NIRAF-guided PH-OCT.

A quasi-bound state in the continuum (BIC) near-perfect absorption meta-surface for the deep ultraviolet (DUV) band, which is composed of a solid substrate fused silica/Al/SiO2/HfO2 multilayer structure, is proposed and numerically validated. The device exhibits an iso-spectral dual-Q response where the resonant wavelength remains nearly constant across all polarization angles, while the Q-factor gradually decreases with increasing polarization angle, ranging from ~140 to ~410 in the DUV band. It enables near-perfect absorption with tunable resonant peaks in the 240-280 nm solar-blind region. The underlying leakage channels and quasi-BIC characteristics are revealed by fitting with the temporal coupled-mode theory, combined with multipole decomposition analysis and eigenmode/band structure calculations. By using the iso-spectral dual-Q feature, we further propose a ratio-based health factor that can be used for spectrometer-free spectral purity detection. Fabrication tolerance analysis and empirical scaling formulas for rough design in the solar-blind band are also provided.