The 50 most-cited articles in Optics Letters, 2020

An ultrafast laser delivering 10.4 kW average output power based on a coherent combination of 12 step-index fiber amplifiers is presented. The system emits close-to-transform-limited 254 fs pulses at an 80 MHz repetition rate, and has a high beam quality ( M 2 ≤ 1.2 ) and a low relative intensity noise of 0.56% in the frequency range of 1 Hz to 1 MHz. Automated spatiotemporal alignment allows for hands-off operation.

Recovering the high-resolution three-dimensional (3D) surface of an object from a single frame image has been the ultimate goal long pursued in fringe projection profilometry (FPP). The color fringe projection method is one of the technologies with the most potential towards such a goal due to its three-channel multiplexing properties. However, the associated color imbalance, crosstalk problems, and compromised coding strategy remain major obstacles to overcome. Inspired by recent successes of deep learning for FPP, we propose a single-shot absolute 3D shape measurement with deep-learning-based color FPP. Through "learning" on extensive data sets, the properly trained neural network can "predict" the high-resolution, motion-artifact-free, crosstalk-free absolute phase directly from one single color fringe image. Compared with the traditional approach, our method allows for more accurate phase retrieval and more robust phase unwrapping. Experimental results demonstrate that the proposed approach can provide high-accuracy single-frame absolute 3D shape measurement for complicated…

We developed a type of toroidal multi-pass cell with multi-layer patterns based on the off-axis model. The effective path length of the original toroidal multi-pass cell is extended several roundtrips in comparison with the single-layer pattern, since the inner surface of the toroidal multi-pass cell is more efficiently utilized. The light pattern has been achieved by using the simple ring surface, which is easy to fabricate. The exact analytical equations for the design of the toroidal multi-pass cell were derived based on analytical vector calculations. A series of numerical ray tracing simulations is presented, and the maximum theoretical optical path length that can be reached is 30 m with a setup of 5 cm column radius. Furthermore, two practical spot patterns are demonstrated with a path length of 8.3 m for a two-layer pattern and 10 m for a three-layer pattern, with respective effective volumes of 63 mL and 94 mL. Furthermore, the fringe effect is substantially reduced to less than 0.5% by the usage of our designed mask.

We experimentally demonstrate all-optical reconfigurable nonlinear activation functions in a cavity-loaded Mach-Zehnder interferometer device on a silicon photonics platform, via the free-carrier dispersion effect. Our device is programmable to generate various nonlinear activation functions, including sigmoid, radial-basis, clamped rectified linear unit, and softplus, with tunable thresholds. We simulate benchmark tasks such as XOR and MNIST handwritten digit classifications with experimentally measured activation functions and obtain accuracies of 100% and 94%, respectively. Our device can serve as nonlinear units in photonic neural networks, while its nonlinear transfer function can be flexibly programmed to optimize the performance of different neuromorphic tasks.

The transport-of-intensity equation (TIE) is one of the most well-known approaches for phase retrieval and quantitative phase imaging. It directly recovers the quantitative phase distribution of an optical field by through-focus intensity measurements in a non-interferometric, deterministic manner. Nevertheless, the accuracy and validity of state-of-the-art TIE solvers depend on restrictive pre-knowledge or assumptions, including appropriate boundary conditions, a well-defined closed region, and quasi-uniform in-focus intensity distribution, which, however, cannot be strictly satisfied simultaneously under practical experimental conditions. In this Letter, we propose a universal solution to TIE with the advantages of high accuracy, convergence guarantee, applicability to arbitrarily shaped regions, and simplified implementation and computation. With the "maximum intensity assumption," we first simplify TIE as a standard Poisson equation to get an initial guess of the solution. Then the initial solution is further refined iteratively by solving the same Poisson equation, and thus the…

Compact beam steering in the visible spectral range is required for a wide range of emerging applications, such as augmented and virtual reality displays, optical traps for quantum information processing, biological sensing, and stimulation. Optical phased arrays (OPAs) can shape and steer light to enable these applications with no moving parts on a compact chip. However, OPA demonstrations have been mainly limited to the near-infrared spectral range due to the fabrication and material challenges imposed by the shorter wavelengths. Here, we demonstrate the first chip-scale phased array operating at blue wavelengths (488 nm) using a high-confinement silicon nitride platform. We use a sparse aperiodic emitter layout to mitigate fabrication constraints at this short wavelength and achieve wide-angle beam steering over a 50° field of view with a full width at half-maximum beam size of 0.17°. Large-scale integration of this platform paves the way for fully reconfigurable chip-scale three-dimensional volumetric light projection across the entire visible range.

Valley photonic crystals (VPhCs) are an attractive platform for the implementation of topologically protected optical waveguides in photonic integrated circuits (PICs). The realization of slow light modes in the topological waveguides may lead to further miniaturization and functionalization of the PICs. In this Letter, we report an approach to realize topological slow light waveguides in semiconductor-slab-based VPhCs. We show that a bearded interface of two topologically distinct VPhCs can support topological kink modes with large group indices over 100 within the topological bandgap. We numerically demonstrate robust light propagation in the topological slow light waveguide with large group indices of ∼60, even under the presence of sharp bends. Our work opens a novel route to implement topological slow light waveguides in a way compatible with current PIC technology.

In this work, we demonstrate postcompression of 1.2 ps laser pulses to 13 fs via gas-based multipass spectral broadening. Our results yield a single-stage compression factor of about 40 at 200 W in-burst average power and a total compression factor >90 at reduced power. The employed scheme represents a route toward compact few-cycle sources driven by industrial-grade Yb:YAG lasers at high average power.

A low voltage operation electro-optic modulator is critical for applications ranging from optical communications to an analog photonic link. This paper reports a hybrid silicon nitride and lithium niobate electro-optic Mach-Zehnder modulator that employs 3 dB multimode interference couplers for splitting and combining light. The presented amplitude modulator with an interaction region length of 2.4 cm demonstrates a DC half-wave voltage of only 0.875 V, which corresponds to a modulation efficiency per unit length of 2.11 V cm. The power extinction ratio of the fabricated device is approximately 30 dB, and the on-chip optical loss is about 5.4 dB.

Unlike the conventional spin Hall effect of light (SHEL) originating from the light-matter interaction, the spin-dependent splitting in the geometric SHEL is purely a geometric effect and independent from the properties of matter. Here it is shown that the geometric SHEL is not only of fundamental theoretical interest in understanding the spin-orbit interaction of light, but also sheds light on important technological applications. This Letter describes the theoretical foundation and experimental realization of optical differential operation and one-dimensional edge detection based on the geometric SHEL.

We demonstrate a C-band wavelength-tunable microlaser with an E r 3 + -doped high quality ( ∼ 1.8 × 10 6 ) lithium niobate microdisk resonator. With a 976 nm continuous-wave pump laser, lasing action can be observed at a pump power threshold lower than 400 µW at room temperature. Furthermore, the microdisk laser wavelength can be tuned by varying the pump laser power, showing a tuning efficiency of ∼ − 17.03 p m / m W at low pump power below 13 mW, and 10.58 pm/mW at high pump power above 13 mW.

We demonstrate the generation of 1.1 J pulses of picosecond duration at 1 kHz repetition rate (1.1 kW average power) from a diode-pumped chirped pulse amplification Yb:YAG laser. The laser employs cryogenically cooled amplifiers to generate λ = 1030 n m pulses with average power of up to 1.26 kW prior to compression with excellent beam quality. Pulses are compressed to 4.5 ps duration with 90% efficiency. This compact picosecond laser will enable a variety of applications that require high energy ultrashort pulses at kilohertz repetition rates.

We demonstrate the reliable generation of 1-mJ, 31-fs pulses with an average power of 1 kW by post-compression of 200-fs pulses from a coherently combined Yb:fiber laser system in an argon-filled Herriott-type multi-pass cell with an overall compression efficiency of 96%. We also analyze the output beam, revealing essentially no spatiospectral couplings or beam quality loss.

Noninvasive, objective measurement of rod function is as significant as that of cone function, and for retinal diseases such as retinitis pigmentosa and age-related macular degeneration, rod function may be a more sensitive biomarker of disease progression and efficacy of treatment than cone function. Functional imaging of single human rod photoreceptors, however, has proven difficult because their small size and rapid functional response pose challenges for the resolution and speed of the imaging system. Here, we describe light-evoked, functional responses of human rods and cones, measured noninvasively using a synchronized adaptive optics optical coherence tomography (OCT) and scanning light ophthalmoscopy (SLO) system. The higher lateral resolution of the SLO images made it possible to confirm the identity of rods in the corresponding OCT volumes.

We propose and numerically demonstrate a scheme of coherent optical chaos communication using semiconductor lasers for secure transmission of optical quadrature amplitude modulation (QAM) signals. In this scheme, a laser intensity chaos and its delayed duplicate are used to amplitude-quadrature modulate a continuous-wave light to generate a chaotic carrier. High-quality chaotic carrier synchronization between the transmitter and receiver is guaranteed by laser intensity chaos synchronization, avoiding laser phase fluctuation. Decryption is implemented by a 90 deg optical hybrid using the synchronous chaotic carrier as local light. Secure transmission of an optical 40 Gb/s 16QAM signal is demonstrated by using a laser intensity chaos with a bandwidth of 11.7 GHz. The system performances are evaluated by analyzing a bit error ratio with different masking coefficients, signal rates, synchronization coefficients, parameter mismatches, and dispersion compensation. It is believed that this scheme can pave a way for high-speed optical chaos communication.

Artificial neural networks have shown effectiveness in the inverse design of nanophotonic structures; however, the numerical accuracy and algorithm efficiency are not analyzed adequately in previous reports. In this Letter, we demonstrate the convolutional neural network as an inverse design tool to achieve high numerical accuracy in plasmonic metasurfaces. A comparison of the convolutional neural networks and the fully connected neural networks show that convolutional neural networks have higher generalization capabilities. We share practical guidelines for optimizing the neural network and analyzed the hierarchy of accuracy in the multi-parameter inverse design of plasmonic metasurfaces. A high inverse design accuracy of $\pm 8\;{\rm nm}$±8nm for the critical geometrical parameters is demonstrated.

Time-varying metamaterials are artificial materials whose electromagnetic properties change over time. Similar to a spatial medium discontinuity, a sudden change in time of the metamaterial refractive index induces the generation of reflected and refracted light waves. The relationship between the incident and emerging fields at one temporal interface has been subject of investigation in earlier studies. Here, we extend the study to a temporal slab, i.e., a uniform homogeneous medium that is present in the whole space for a limited time. The scattering coefficients have been derived as a function of the refractive indices and application time, demonstrating that the response of the temporal slab can be controlled through the application time, which acts similarly to the electrical thickness of conventional spatial slabs. The results reported in this Letter pave the way to creating novel devices based on temporal discontinuities, such as temporal matching networks, Bragg grating, and dielectric mirrors, which exhibit zero space occupancy by exploiting the time dimension, instead of…

Image fusion is the key step to improve the performance of object detection in polarization images. We propose an unsupervised deep network to address the polarization image fusion issue. The network learns end-to-end mapping for fused images from intensity and degree of linear polarization images, without the ground truth of fused images. Customized architecture and loss function are designed to boost performance. Experimental results show that our proposed network outperforms other state-of-the-art methods in terms of visual quality and quantitative measurement.

Global acquisition of atmospheric wind profiles using a spaceborne direct-detection Doppler wind lidar is being accomplished following the launch of European Space Agency's Aeolus mission. One key part of the instrument is a single-frequency, ultraviolet laser that emits nanosecond pulses into the atmosphere. High output energy and frequency stability ensure a sufficient signal-to-noise ratio of the backscatter return and an accurate determination of the Doppler frequency shift induced by the wind. This Letter discusses the design of the laser transmitter for the first Doppler wind lidar in space and its performance during the first year of the Aeolus mission, providing valuable insights for upcoming space lidar missions.

High optical quality (Q) factors are critically important in optical microcavities, where performance in applications spanning nonlinear optics to cavity quantum electrodynamics is determined. Here, a record Q factor of over 1.1 billion is demonstrated for on-chip optical resonators. Using silica whispering-gallery resonators on silicon, Q-factor data is measured over wavelengths spanning the C/L bands (100 nm) and for a range of resonator sizes and mode families. A record low sub-milliwatt parametric oscillation threshold is also measured in 9 GHz free-spectral-range devices. The results show the potential for thermal silica on silicon as a resonator material.

Multispectral endomicroscopy provides tissue functional information in addition to structural information for accurate disease diagnosis. In this Letter, we propose a snapshot multispectral endomicroscope that employs a fiber bundle to deliver an in-body tissue spatial-spectral datastream to an external compressive spectral imager. Equipped with an end-to-end deep-learning-based reconstruction algorithm, we are able to capture tissue multispectral data in video rates and reconstruct high-resolution multispectral images with up to 24 spectral channels in near-real time.

Resonator fiber optic gyroscope (RFOG) performance has hitherto been limited by nonlinearity, modal impurity, and backscattering in the sensing fibers. The use of hollow-core fiber (HCF) effectively reduces nonlinearity, but the complex interplay among glass and air-guided modes in conventional HCF technologies can severely exacerbate RFOG instability. By employing high-performance nested anti-resonant nodeless fiber, we demonstrate long-term stability in a hollow-fiber RFOG of 0.05 deg/h, nearing the levels required for civil aircraft navigation. This represents a ${{3}} \times$ improvement over any prior hollow-core RFOG and a factor of ${{500}} \times$ over any prior result at integration times longer than 1 h.

We demonstrated a high-power Q -switched two-stage Ho:YAG master-oscillator power-amplifier (MOPA) system dual-end pumped by Tm:YLF lasers. A new method was introduced by rotating and swapping spatial axial directions of pump beams to improve the beam quality of the Ho:YAG oscillator and first-stage amplifier. Two parallel second-stage Ho:YAG amplifiers were employed to output high power. A total maximum average output power of 332 W at 2091 nm with pulse repetition frequency of 20 kHz was achieved. Then a Z n G e P 2 MOPA system was demonstrated using the Ho:YAG MOPA as the pump source. A maximum average output power of 161 W at 3–5 µm was obtained with 290 W incident Ho pump power, corresponding to beam quality factors M 2 of 3.42 and 3.83 for horizontal and vertical directions, respectively.

We demonstrate numerically and experimentally a gain-managed nonlinear amplifier with a large mode area fiber. The amplifier delivers 1.2 µJ and sub-40 fs pulses with the spectrum spanning from ∼ 1000 to ∼ 1180 n m . We show that longitudinal gain-loss evolution plays an essential role in pulse formation by comparing simulations with different gain models to the experimental results.

The diffractive deep neural network ( D 2 N N ) has demonstrated its importance in performing various all-optical machine learning tasks, e.g., classification, segmentation, etc. However, deeper D 2 N N s that provide higher inference complexity are more difficult to train due to the problem of gradient vanishing. We introduce the residual D 2 N N s (Res- D 2 N N ), which enables us to train substantially deeper diffractive networks by constructing diffractive residual learning blocks to learn the residual mapping functions. Unlike the existing plain D 2 N N s , Res- D 2 N N s contribute to the design of a learnable light shortcut to directly connect the input and output between optical layers. Such a shortcut offers a direct path for gradient backpropagation in training, which is an effective way to alleviate the gradient vanishing issue on very deep diffractive neural networks. Experimental results on image classification and pixel super-resolution demonstrate the superiority of Res- D 2 N N s over the existing plain D 2 N N architectures.

Spontaneous emission of an excited atom in a featureless continuum of electromagnetic modes is a fundamental process in quantum electrodynamics associated with an exponential decay of the quantum emitter to its ground state accompanied by an irreversible emission of a photon. However, such a simple scenario is deeply modified when considering a "giant" atom, i.e., an atom whose dimension is larger than the wavelength of the emitted photon. In such an unconventional regime, non-Markovian effects and strong deviations from an exponential decay are observed owing to interference effects arising from nonlocal light-atom coupling. Here we suggest a photonic simulation of non-Markovian giant atom decay, based on light escape dynamics in an optical waveguide nonlocally coupled to a waveguide lattice. Major effects, such as nonexponential decay, enhancement, or slowing down of the decay, and formation of atom-field dark states can be emulated in this system.

We propose a digital micromirror device (DMD) holographic display, where speckleless holograms can be observed in the expanded viewing zone. Structured illumination (SI) is applied to expand the small diffraction angle of the DMD using a laser diode (LD) array. To eliminate diffraction noise from SI, we utilize an active filter array for the Fourier filter and synchronize it with the LD array. The speckle noise is reduced via temporal multiplexing, where the proposed system supports a dynamic video of 60 Hz using the DMD's fast operation property. The proposed system is verified and evaluated with experimental results.

We report, for the first time, to the best of our knowledge, a femtosecond mode-locked Fe:ZnSe laser. Passive mode locking is implemented using graphene as a saturable absorber. The laser operates at 4.4 µm with a repetition frequency of 100 MHz and 415 mW output power pumped by a fiber 7 W Er:ZBLAN laser. The pulse duration of about 732 fs is retrieved from the first-order autocorrelation function. Additionally, we observe pulsed nanosecond oscillation under continuous-wave pumping and strong amplitude modulation caused by Kerr self-focusing. This Letter fills the gap in operating regimes of Fe:ZnSe lasers and paves the way for the development of powerful ultrafast high-repetition-rate mid-IR sources for the most advanced fields of science.

Optics is a promising platform in which to help realize the next generation of fast, parallel, and energy-efficient computation. We demonstrate a reconfigurable free-space optical multiplier that is capable of over 3000 computations in parallel, using spatial light modulators with a pixel resolution of only 340×340. This enables vector-matrix multiplication and parallel vector-vector multiplication with vector size of up to 56. Our design is, to the best of our knowledge, the first to simultaneously support optical implementation of reconfigurable, large-sized, and real-valued linear algebraic operations. Such an optical multiplier can serve as a building block of special-purpose optical processors such as optical neural networks and optical Ising machines.

The conventional Shack-Hartmann wavefront sensor (SHWS) requires wavefront slope measurements of every micro-lens for wavefront reconstruction. In this Letter, we applied deep learning on the SHWS to directly predict the wavefront distributions without wavefront slope measurements. The results show that our method could provide a lower root mean square wavefront error in high detection speed. The performance of the proposed method is also evaluated on challenging wavefronts, while the conventional approaches perform insufficiently. This Letter provides a new approach, to the best of our knowledge, to perform direct wavefront detection in SHWS-based applications.

A Mach–Zehnder silicon photonic switch with low random phase errors is proposed and demonstrated for the first time, to the best of our knowledge, by incorporating judiciously widened and shortened phase shifter waveguides. With a 180 nm complementary metal–oxide–semiconductor (CMOS) foundry process, more than one hundred 2 × 2 thermo-optic Mach–Zehnder switches (MZSs) with varied phase shifter widths have been designed, fabricated, and characterized on 14 silicon chips. The mean and standard deviation of the random phase errors of the MZSs with phase shifters widened to 2 µm are less than a third of those of the conventional design with 0.45-µm-wide single-mode phase shifters. This validates the improved fabrication tolerance and results in considerable reduction of the power consumption for the phase error compensation. Such elegant methodology paves the way to further scaling up N × N silicon thermo-optic switches and can be generalized for other phase-sensitive integrated photonic devices as well.

We show the tight-focusing properties of a linearly polarized circular Airy Gaussian vortex beam (CAiGVB) with a high-numerical-aperture objective lens; the light intensity distributions exhibit diversity with different positions of the vortex pairs (on-axis or off-axis vortex pairs). By choosing different optical distribution factors, the CAiGVB turns into a circular Airy vortex beam or Gaussian vortex beam, and the depth of focus can also be controlled. It is known that the vortex beam possesses both orbital and spin angular momentum. The spin density vector changes its direction in three-dimensional space during beam propagation, as long as it is not purely transverse or longitudinal, which would cause 3D polarization ellipse topologies. In contrast, the polarization topologies degenerate into 2D when the spin density vector is purely transverse or longitudinal. Furthermore, the direction of the spin density vector is closely related to the Gouy phase difference between longitudinal and transverse electric field components of the vectorial beam.

In mask-based lensless imaging, iterative reconstruction methods based on the geometric optics model produce artifacts and are computationally expensive. We present a prototype of a lensless camera that uses a deep neural network (DNN) to realize rapid reconstruction for Fresnel zone aperture (FZA) imaging. A deep back-projection network (DBPN) is connected behind a U-Net providing an error feedback mechanism, which realizes the self-correction of features to recover the image detail. A diffraction model generates the training data under conditions of broadband incoherent imaging. In the reconstructed results, blur caused by diffraction is shown to have been ameliorated, while the computing time is 2 orders of magnitude faster than the traditional iterative image reconstruction algorithms. This strategy could drastically reduce the design and assembly costs of cameras, paving the way for integration of portable sensors and systems.

In this Letter, the broadband operation in wavelengths from 520 nm to 980 nm is demonstrated on silicon nitride nanophotonic phased arrays. The widest beam steering angle of 65° on a silicon nitride phased array is achieved. The optical radiation efficiency of the main grating lobe in a broad wavelength range is measured and analyzed theoretically. The optical spots radiated from the phased array chip are studied at different wavelengths of lasers. The nanophotonic phased array is excited by a supercontinuum laser source for a wide range of beam steering for the first time to the best of our knowledge. It paves the way to tune the wavelength from visible to near infrared range for silicon nitride nanophotonic phased arrays.

We propose and demonstrate that optical analog computing of spatial differentiation and edge detection can be realized with a single layer of dielectric metasurface. The optical transfer function for second-order derivation is obtained by engineering the spatial dispersion of electric dipole resonance supported by the silicon nanodisks in the metasurface. Benefiting from this unique mechanism of electric dipole resonance, spatial differentiation can be performed for two dimensions and arbitrary polarization with a large spatial bandwidth and high efficiency at the visible wavelength. Explicitly, we have numerically validated the application with one-dimensional spatial functions as well as an image, and the results show excellent performance. Our study can facilitate the research of optical computing with artificial nanostructures.

We demonstrate supercontinuum generation over an octave spaning from 1055 to 2155 nm on the highly nonlinear aluminum gallium arsenide (AlGaAs)-on-insulator platform. This is enabled by the generation of two dispersive waves in a 3-mm-long dispersion-engineered nano-waveguide. The waveguide is pumped at telecom wavelengths (1555 nm) with 3.6 pJ femtosecond pulses. We experimentally validate the coherence of the generated supercontinuum around the pump wavelength (1450-1750 nm), and our numerical simulation shows a high degree of coherence over the full spectrum.

Optical analog computing has attracted widespread attention in recent decades due to its advantages of lower consumption, higher efficiency, and real-time imaging in image processing. Here, we propose a two-dimensional optical analog computing scheme based on the Brewster effect. We experimentally demonstrate two-dimensional edge detection with high efficiency. By combining microscopy, our approach may develop some significant applications in cellular and molecular imaging.

We demonstrate a high-efficiency thermo-optic (TO) tunable micro-ring resonator in thin-film lithium niobate. Thermal insulation trenches around the heated micro-ring resonator and the underlying silicon substrate significantly reduce the heating power consumption and improve the tuning efficiency. Compared to conventional TO devices without thermal insulation trenches, the proposed device achieves a full free spectral range wavelength shift with a 14.9 mW heating power, corresponding to a thermal tuning efficiency of 53.7 pm/mW, a more than 20-fold improvement of tuning efficiency. The approach enables energy-efficient high-performance TO devices such as optical switches, wavelength routers, and other reconfigurable photonic devices.

We demonstrate coherent supercontinuum generation spanning over an octave from a silicon germanium-on-silicon waveguide using ∼ 200 f s pulses at a wavelength of 4 µm. The waveguide is engineered to provide low all-normal dispersion in the TM polarization. We validate the coherence of the generated supercontinuum via simulations, with a high degree of coherence across the entire spectrum. Such a generated supercontinuum could lend itself to pulse compression down to 22 fs.

We demonstrate terahertz (THz) generation using the tilted pulse front method in lithium niobate, driven at an unprecedented high average power of more than 100 W and at a 13.3 MHz repetition rate, provided by a compact amplifier-free mode-locked thin-disk oscillator. The conversion efficiency was optimized with respect to the pump spot size and pump pulse duration, enabling us to generate a maximum THz average power of 66 mW, which is, to the best of our knowledge, the highest reported to date from a laser-driven, few-cycle THz source. Furthermore, we identify beam walk-off as the main obstacle that currently limits the conversion efficiency in this excitation regime (with moderate pulse energies and small spot sizes). Further upscaling to the watt level and beyond is within reach, paving the way for linear and nonlinear high average power THz spectroscopy experiments with an exceptional signal-to-noise ratio at megahertz repetition rates.

We propose a reflective terahertz (THz) metalens with four focal points for polarization detection of THz beams. The metalens is composed of Z -shaped resonators with spatially variant orientations, a reflective gold layer, and a dielectric spacer between them. The polarization states of the focal points include left circular polarization, right circular polarization, an incident polarization state, and a polarization state whose major axis is rotated π / 4 in comparison with that of the incident polarization. The handedness, ellipticity, and major axis of the polarization state can be determined based on the light intensities of the focal points. The uniqueness of the designed device renders this technique very attractive for applications in compact THz polarization detection and information processing.

Terahertz (THz) sensing of molecular fingerprint enables wide applications in biomedicine and security detection. Conventional detection approaches face big barriers in trace analysis of analyte due to the difficulties of enhancing the broadband molecular absorption. In order to achieve strong broadband wave-matter interaction for the analyte, we propose a method based on THz wave angular scanning on a dielectric metagrating. In virtue of the guided-mode resonance, one can strengthen the local electric field in various trace-amount analytes by tuning the polarization and incident angle, which leads to significant enhancement on the broadband signal of molecular fingerprint. The study paves the way for more applications of THz trace-amount detection.

In this Letter, we report high-speed integrated 14 µm in diameter micro-light-emitting diode (μLED) arrays with the parallel configuration, including 2 × 2 , 2 × 3 , 2 × 4 , and 2 × 5 arrays. The small junction area of μLED ( ∼ 191 µ m 2 ) in each element facilitates the operation of higher injection current density up to 13 k A / c m 2 , leading to the highest modulation bandwidth of 615 MHz. The optical power of 2 × 5 array monotonically increases ( ∼ 10 times higher) as the number of arrays increases (1 to 10), while retaining the fast modulation bandwidth. A clear eye diagram up to 1 Gbps without any equalizer further shows the capability of this high-speed transmitter for VLC. These results mean that tailoring the optical power of μLEDs in a parallel-biased integrated array can further enhance the data transmission rate without degradation of the modulation bandwidth.

We propose a simple hardware architecture for solving exclusive OR (XOR) tasks in a single step by using a single photonic spiking neuron based on vertical-cavity surface-emitting lasers with an embedded saturable absorber (VCSEL-SA) subject to dual-polarized pulsed optical injection. We model the inhibitory photonic spiking neuron by extending the Yamada model and spin-flip model to incorporate the two polarization-resolved modes and the saturable absorber. It is shown that, by carefully adjusting the temporal difference according to the inhibitory window, the XOR operation can be realized in a single photonic spiking neuron, which is interesting and valuable for the photonic neuromorphic computing and information processing.

We report, to the best of our knowledge, the first optical trapping experimental demonstration of microparticles with frozen waves. Frozen waves are an efficient method to model longitudinally the intensity of nondiffracting beams obtained by superposing copropagating Bessel beams with the same frequency and order. Based on this, we investigate the optical force distribution acting on microparticles of two types of frozen waves. The experimental setup of holographic optical tweezers using a spatial light modulator has been assembled and optimized. The results show that it is possible to obtain greater stability for optical trapping using frozen waves. The significant enhancement in trapping geometry from this approach shows promising applications for optical tweezers micromanipulations over a broad range.

High-efficiency chirped grating couplers (GCs) with coupling efficiencies (CE) approaching 90%/coupler were designed by using a particle swarm optimization algorithm. These GCs were fabricated on $z$-cut lithium niobate on insulator (LNOI) with an Au layer on the lithium niobate substrate. The maximum CEs for transverse electric and transverse magnetic polarization input were measured up to ${\sim}{72.0}\%$/coupler and ${\sim}{61.6}\%$/coupler, respectively, which are the state-of-the-art values for LNOI GCs as far as we know. These GCs contribute to the realization of high-efficiency LNOI on-chip integrated optics.

Imaging in low light is significant but challenging in many applications. Adding the polarization information into the imaging system compromises the drawbacks of the conventional intensity imaging to some extent. However, generally speaking, the qualities of intensity images and polarization images cannot be compatible due to the characteristic differences in polarimetric operators. In this Letter, we collected, to the best of our knowledge, the first polarimetric imaging dataset in low light and present a specially designed neural network to enhance the image qualities of intensity and polarization simultaneously. Both indoor and outdoor experiments demonstrate the effectiveness and superiority of this neural network-based solution, which may find important applications for object detection and vision in photon-starved environments.

An ultra-broadband polarization beam splitter (PBS) with low excess loss (EL) and a high extinction ratio (ER) is proposed and demonstrated for the case with 340 nm thick silicon-on-insulator waveguides. Here the PBS is realized by using cascaded adiabatic dual-core tapers, which consist of a strip core and a subwavelength-grating core. For the designed PBS, which has a 33.6 µm long mode-evolution region, the ELs are ${ \lt }{0.3}\;{\rm dB}$<0.3dB, and the ERs are ${ \gt }{20}\;{\rm dB}$>20dB for both TE and TM polarizations in an ultra-broad bandwidth of ${ \gt }{270}\;{\rm nm}$>270nm (1400-1670 nm) in theory. For the fabricated PBS, the measured bandwidths for achieving ERs of ${\sim}{20}$∼20 and ${\sim}{25}\;{\rm dB}$∼25dB are 240 and 220 nm, while the 1 dB bandwidth is as large as 230 nm, which are the largest bandwidths reported to date, to the best of our knowledge.

In this study, we aimed to measure material dispersion in fused silica using a low coherence interferometric method. The measurement was carried out quickly and efficiently in a wide spectral range using this method. The refractive index and group index of fused silica were determined by capturing a few interferograms. The material dispersion was modeled using a Sellmeier equation with three resonances. Three different fits were investigated; the most appropriate fit was the one that used both the measured refractive and group indexes to model the dispersion. Second-order dispersion was also quantified, and zero-dispersion wavelength was determined.

Acoustic wave sensors with a high sensitivity and small size are highly desired for a wide variety of important and emerging applications such as photoacoustic gas sensing and bio-imaging. Here we present an ultracompact optical fiber acoustic sensor based on an optomechanical resonator that is directly in situ printed on the end face of a standard single-mode optical fiber by using an optical 3D μ-printing technology. The fiber-top optomechanical microresonator is composed of a microscale suspended polymer micro-disk that forms a Fabry–Perot interferometric cavity, together with the optical fiber end face, and acts as the acoustic wave-sensitive micromechanical resonator simultaneously. The microbeams for suspending the micro-disk are devised with a spiral structure to overcome the small-size imposed low deflection amplitude so as to improve its sensitivity to acoustic waves. The sensor with a high sensitivity of 118.3 mV/Pa and low noise equivalent acoustic signal level of 0.328 µ P a / H z 1 / 2 at audio frequency is experimentally demonstrated. Moreover, with a resonance…