We present a surface-passivation method that reduces surface-related losses by almost 2 orders of magnitude in a highly miniaturized $\mathrm{Ga}\mathrm{As}$ open microcavity. The microcavity consists of a curved dielectric distributed Bragg reflector with radius of approximately $10\mu \mathrm{m}$ paired with a $\mathrm{Ga}\mathrm{As}$-based heterostructure. The heterostructure consists of a semiconductor distributed Bragg reflector followed by an $n$-$i$-$p$ diode with a layer of quantum dots in the intrinsic region. Free-carrier absorption in the highly-$n$-doped and highly-$p$-doped layers is minimized by our positioning them close to a node of the vacuum electromagnetic field. The surface, however, resides at an antinode of the vacuum field and results in significant loss. These losses are much reduced by surface passivation. The strong dependence on wavelength implies that the main effect of the surface passivation is to eliminate the surface electric field, thereby quenching below-band-gap absorption via a Franz-Keldysh-like effect. An additional benefit is that the surface passivation reduces scattering at the $\mathrm{Ga}\mathrm{As}$ surface. These results are important in other nanophotonic devices that rely on a $\mathrm{Ga}\mathrm{As}$-vacuum interface to confine the electromagnetic field.

Calibration of sensors is a fundamental step in validating their operation. This can be a demanding task, as it relies on acquiring detailed modeling of the device, which can be aggravated by its possible dependence upon multiple parameters. Machine learning provides a handy solution to this issue, operating a mapping between the parameters and the device response, without needing additional specific information on its functioning. Here, we demonstrate the application of a neural-network-based algorithm for the calibration of integrated photonic devices depending on two parameters. We show that a reliable characterization is achievable by carefully selecting an appropriate network training strategy. These results show the viability of this approach as an effective tool for the multiparameter calibration of sensors characterized by complex transduction functions. Furthermore, the approach is proven to be versatile and promising for mass production, as the same neural network is able to calibrate different devices that have the same structure.

We have fabricated an array of subgap kinetic inductance detectors (SKIDs) made of granular aluminum ($T_c\sim 2$ K) sensitive in the 80–90 GHz frequency band and operating at 300 mK. We measure a noise equivalent power of $1.3\times {10}^{-16}{\mathrm{W}/\mathrm{Hz}}^{0.5}$ on average and $2.6\times {10}^{-17}{\mathrm{W}/\mathrm{Hz}}^{0.5}$ at best, for an illuminating power of 50 fW per pixel. Even though the circuit design of SKIDs is identical to that of the kinetic inductance detectors, the SKIDs operating principle is based on their sensitivity to subgap excitations. This detection scheme is advantageous because it avoids having to lower the operating temperature proportionally to the lowest detectable frequency. The SKIDs presented here are intrinsically selecting the 80–90 GHz frequency band, well below the superconducting spectral gap of the film, at approximately 180 GHz.

Deep learning is the backbone of artificial-intelligence technologies, and it can be regarded as a kind of multilayer feedforward neural network. An essence of deep learning is information propagation through layers. This suggests that there is a connection between deep neural networks and dynamical systems in the sense that information propagation is explicitly modeled by the time evolution of dynamical systems. In this study, we perform pattern recognition based on the optimal control of continuous-time dynamical systems, which is suitable for physical hardware implementation. The learning is based on the adjoint method to optimally control dynamical systems, and the deep (virtual) network structures based on the time evolution of the systems are used for processing input information. As a key example, we apply the dynamics-based recognition approach to an optoelectronic delay system and demonstrate that the use of the delay system allows for image recognition and nonlinear classifications using only a few control signals. This is in contrast to conventional multilayer neural networks, which require a large number of weight parameters to be trained. The proposed approach provides insight into the mechanisms of deep network processing in the framework of an optimal control problem and presents a pathway for realizing physical computing hardware.

Using a coupled parametric resonator array for generating and propagating a topological soliton in its rotating-frame phase space is theoretically and numerically investigated. In an analogy with the well-known ${\varphi }^4$ model, the existence of a soliton is topologically protected, as the boundary of two different phase domains of parametric oscillation. Numerical simulation indicates that propagation can be triggered by switching of the phase state of one specific resonator, and the effects of damping, collision, and symmetry lifting by the harmonic drive on the propagation dynamics are studied. The topological soliton can be implemented by using electromechanical resonators, which allow its propagation dynamics to be precisely electrically controlled and provide a fully controlled on-chip test bed for the study of a topological soliton.

In quantum optomechanics, finding materials and strategies to limit losses has been crucial to the progress of the field. Recently, superfluid ${}^4\mathrm{He}$ was proposed as a promising mechanical element for quantum optomechanics. This quantum fluid shows highly desirable properties (e.g., extremely low acoustic loss) for a quantum optomechanical system. In current implementations, superfluid optomechanical systems suffer from external sources of loss, which spoils the quality factor of resonators. In this work, we propose an alternate implementation, exploiting nanofluidic confinement. Our approach, based on acoustic resonators formed within phononic nanostructures, aims at limiting radiation losses to preserve the intrinsic properties of superfluid ${}^4\mathrm{He}$. In this work, we estimate the optomechanical system parameters. Using recent theory, we derive the expected quality factors for acoustic resonators in different thermodynamic conditions. We calculate the sources of loss induced by the phononic nanostructures with numerical simulations. Our results indicate the feasibility of the proposed approach in a broad range of parameters, which opens prospects for more complex geometries.

The recent experiment [Y. Wang et al., Science 366, 1125 (2019)] on magnon-mediated spin-transfer torque (MSTT) was interpreted in terms of a picture where magnons are excited within an antiferromagnetic insulator (AFI), by applying nonequilibrium electronic spin density at one of its surfaces, so that their propagation across AFI deprived of conduction electrons eventually leads to reversal of magnetization of a ferromagnetic metal (FM) attached to the opposite surface of an AFI. However, microscopic (i.e., Hamiltonian-based) understanding of how magnonic and electronic spin currents, both of which can exert torque on localized magnetic moments within FM, are generated and interconverted at multiple junction interfaces is lacking. We employ recently developed time-dependent nonequilibrium Green functions combined with the Landau-Lifshitz-Gilbert equation (TDNEGF+LLG) formalism to evolve conduction electrons quantum mechanically while they interact via self-consistent backaction with localized magnetic moments described classically by atomistic spin dynamics solving a system of LLG equations. Upon injection of square current pulse as the initial condition, TDNEGF+LLG simulations of FM-polarizer/AFI/FM-analyzer junctions show that reversal of localized magnetic moments within a FM analyzer is less efficient, in the sense of requiring larger pulse height and its longer duration, than conventional electron-mediated STT (ESTT) driving magnetization switching in a standard FM-polarizer/normal-metal/FM-analyzer spin valve. Since both electronic, generated by spin pumping from AFI, and magnonic, generated by direct transmission from AFI, spin currents are injected into the FM analyzer, its localized magnetic moments will experience combined MSTT and ESTT. Nevertheless, by artificially turning off ESTT we demonstrate that MSTT plays a dominant role whose understanding, therefore, paves the way for all-magnon-driven magnetization switching devices with no electronic parts.

Temperature plays an important role in spin-torque switching of magnetic tunnel junctions, causing magnetization fluctuations that decrease the switching voltage but also introduce switching errors. Here we present a systematic study of the temperature dependence of the spin-torque-switching probability of state-of-the-art perpendicular-magnetic-tunnel-junction nanopillars (40–60 nm in diameter) from room temperature down to 4 K, sampling up to a million switching events. The junction temperature at the switching voltage—obtained from the thermally assisted spin-torque-switching model—saturates at temperatures below about 75 K, showing that junction heating is significant below this temperature and that spin-torque switching remains highly stochastic down to 4 K. A model of heat flow in a nanopillar junction shows this effect is associated with the reduced thermal conductivity and heat capacity of the metals in the junction.

Many developments in science and engineering depend on tackling complex optimizations on large scales. The challenge motivates an intense search for specific computing hardware that takes advantage of quantum features, nonlinear dynamics, or photonics. A paradigmatic optimization problem is to find low-energy states in classical spin systems with fully random interactions. To date, no alternative computing platform can address such spin-glass problems on a large scale. Here, we propose and realize an optical scalable spin-glass simulator based on spatial light modulation and multiple light scattering. By tailoring optical transmission through a disordered medium, we optically accelerate the computation of the ground state of large spin networks with all-to-all random couplings. Scaling of the operation time with the problem size demonstrates an optical advantage over conventional computing. Our results highlight optical vector-matrix multiplication as a tool for spin-glass problems and provide a general route toward large-scale computing that exploits speed, parallelism, and coherence of light.

In this work, we fabricate a nitride-based photonic crystal (PhC) nanocavity with an embedded active layer for unexplored red regions using simplified selective wet etching of an ${\mathrm{Al}}_{0.82}{\mathrm{In}}_{0.18}\mathrm{N}$ sacrificial layer. Eu,O-codoped $\mathrm{Ga}\mathrm{N}$ ($\mathrm{Ga}\mathrm{N}$:$\mathrm{Eu}$,$\mathrm{O}$) is embedded in L7-type two-dimensional-PhC (2D-PhC) nanocavities with hexagonal holes. Room-temperature photoluminescence (PL) shows $\mathrm{Eu}$ emission coupled to the cavity modes at wavelengths around 620 nm. The quality factor of the fundamental mode in an L7 nanocavity is 5400, which is comparable to those of state-of-the-art 2D-PhC nanocavities with embedded nitride-based active layers for the blue and ultraviolet regions. We show that the Purcell effect due to the small mode volumes of the 2D-PhC nanocavities and enhanced light extraction lead to a 54-fold increase in the integrated PL intensity per unit area with respect to the typical $\mathrm{Ga}\mathrm{N}$:$\mathrm{Eu}$,$\mathrm{O}$ emission.

We propose an approach for wavelength-selective excitation of perpendicular standing spin waves by optical guided modes in a magnetic dielectric film. TM-polarized guided waves with elliptical polarization induce an effective magnetic field due to the inverse transverse magneto-optical Kerr effect (ITMOKE). The ITMOKE is similar to the inverse Faraday effect, except for the induced magnetic field direction that is oriented in the film plane perpendicular to guided-mode propagation. Switching between different guided modes by tuning the wavelength or angle of the laser pulse significantly modifies the spatial distribution of the induced ITMOKE field and, therefore, selectively launches the standing spin-wave modes of different orders. Consequently, one may perform a tunable optical excitation of a certain spin-wave mode via a femtosecond laser pulse of the corresponding wavelength. The proposed approach broadens the spin-wave manipulation capabilities via optical means.

Recently, nonreciprocal charge transport has emerged as an important experimental tool for detecting symmetry breaking in noncentrosymmetric crystals. Despite extensive studies on systems with intrinsic symmetry breaking from crystal structures or interfaces, nonreciprocal transport induced by extrinsic mechanisms, such as the asymmetric design of device geometry, remains largely unexplored. Here, we show that a giant nonreciprocal resistance arises extrinsically from the thermoelectric effect due to nonuniform current heating in mesoscopic devices. Taking $\mathrm{Nb}\mathrm{As}$ nanobelts as an example, we observe giant second-harmonic resistances induced by the Seebeck and Nernst effects, with a nonreciprocal coefficient value up to 5.4 × ${10}^{-6}{\mathrm{cm}}^2/\mathrm{A}$, which is higher than that of most noncentrosymmetric crystals. The symmetry of second-harmonic resistance in magnetic fields can be further controlled by designing the electrode geometry and is decoupled from that of the first-harmonic resistance or the crystal symmetry. We show that such thermoelectric related nonreciprocal resistances also exist in macroscopic samples up to hundreds of microns. Our results not only introduce a simple yet effective approach to probe the thermoelectric effect, but also reveal a previously underestimated mechanism that strongly modifies the second-order transport coefficients.

Slowing down, stopping, and reversing a signal is a core functionality for information processing. Here, we show that this functionality can be realized by tuning the dispersion of a periodic system through a dispersionless, or flat, band. Specifically, we propose a phononic metamaterial based on plate resonators, in which the phonon band dispersion can be modified from an acousticlike to an optical character by modulating a uniform prestress. The switch is enabled by the change in sign of an effective coupling between fundamental modes, which generically leads to a nearly dispersion-free band at the transition point. We demonstrate how adiabatic tuning of the band dispersion can immobilize and reverse the propagation of a sound pulse in simulations of a one-dimensional resonator chain. Our study relies on the basic principles of thin-plate elasticity independently of any specific material, making our results applicable across varied length scales and experimental platforms. More broadly, our approach could be replicated for signal manipulation in photonic metamaterials and electronic heterostructures.

Microwave resonators with a keyhole profile (KHRs) operating in the C band and the X band are designed and studied in numerical simulations and experiments. KHR structures concentrate a microwave magnetic field in a subwavelength volume, while suppressing microwave electric fields. This microwave magnetic field is focused at a finite working distance from KHR metal structures, allowing convenient optical excitation of the sample in both the Faraday geometry and the Voigt geometry. By means of room-temperature optically detected electron spin resonance on $\mathrm{Si}\mathrm{C}$ quantum defects, the conversion factor $B_1{P}_{\mathrm{MW}}^{-1/2}$ for conversion of microwave power into a microwave magnetic field is measured to be approximately $1.0\times {10}^{-3}\mathrm{T}{\mathrm{W}}^{-1/2}$ at a frequency of approximately 7 GHz and a working distance of approximately 0.5 mm from the KHR structure. Numerical simulations match the experimental observations, and an example model code for use with the finite-element-method program elmer is provided. The KHR structures are most promising for fast coherent electron-spin control in solid-state spin qubits, where a large microwave magnetic field needs to be achieved with simultaneous suppression of microwave heating and electric fields, while permitting efficient optical spin initialization and readout.

The fabrication of quantum key distribution (QKD) systems typically involves several parties, thus providing Eve with multiple opportunities to meddle with the devices. As a consequence, conventional hardware and/or software hacking attacks pose natural threats to the security of practical QKD. Fortunately, if the number of corrupted devices is limited, the security can be restored by using redundant apparatuses. Here, we report on the demonstration of a secure QKD setup with optical devices and classical postprocessing units possibly controlled by an eavesdropper. We implement a 1.25 GHz chip-based measurement-device-independent QKD system secure against malicious devices on both the measurement and the users’ sides. The secret key rate reaches 137 bps over a 24 dB channel loss. Our setup, benefiting from a high clock rate, miniaturized transmitters, and a cost-effective structure, provides a promising solution for widespread applications requiring uncompromising communication security.

Efforts to scale-up quantum computation have reached a point where the principal limiting factor is not the number of qubits, but the entangling gate infidelity. However, the highly detailed system characterization required to understand the underlying error sources is an arduous process and impractical with increasing chip size. Open-loop optimal control techniques allow for the improvement of gates but are limited by the models they are based on. To rectify the situation, we provide an integrated open-source tool set for control, calibration, and characterization (${\mathbb{C}}^3$), capable of open-loop pulse optimization, model-free calibration, model fitting, and refinement. We present a methodology to combine these tools to find a quantitatively accurate system model, high-fidelity gates, and an approximate error budget, all based on a high-performance, feature-rich simulator. We illustrate our methods using simulated fixed-frequency superconducting qubits for which we learn model parameters with less than $1\mathrm{\%}$ error and derive a coherence-limited cross-resonance gate that achieves $99.6\mathrm{\%}$ fidelity without the need for calibration.

Atomically sharp tips are a requirement for scanning-probe microscopy, such as scanning tunneling microscopy (STM) and atomic force microscopy (AFM). Compared with STM, AFM imaging is more sensitive to the sharpness of tip apices because long-range forces act as a background signal on the high-resolution AFM images originating from short-range forces. Here we report the investigation of in situ reproducible sharp tips for AFM. We make an ${\mathrm{Ag}}_2\mathrm{S}$ crystal, a mixed ionic and electronic conductor, on a conventional $\mathrm{Si}$ cantilever, and controllably grow and shrink the $\mathrm{Ag}$ nanoprotrusion by changing the polarity of the bias voltage between the tip and the sample. We are able to reduce the contribution of long-range forces by growing a $\mathrm{Ag}$ nanoprotrusion on the ${\mathrm{Ag}}_2\mathrm{S}$ tip, and obtain atomic-resolution AFM images. We also confirm that the ${\mathrm{Ag}}_2\mathrm{S}$ tip with a $\mathrm{Ag}$ nanoprotrusion, the end of which presumably terminates in $\mathrm{Si}$ atoms, is capable of simultaneous AFM and STM measurements.

We exploit the enhancement of liquid transport through a graphene-epoxy film actuated by surface acoustic waves to demonstrate effective and efficient nanoparticle filtration. Unlike previously reported findings using graphene films, the addition of silver epoxy to create a composite film allows higher excitation powers to be sustained over extended operation durations without disintegration of the film such that an increase in the liquid flow rate through the film by up to fourfold can be achieved. Counter to intuition that suggests that the epoxy would fill the interstitial voids between the layers that make up the graphene film so as to reduce the efficiency of liquid transport through the film, we show instead that the higher density and elastic stiffness of the composite film enhances the transmission of acoustic energy into the film that, in turn, leads to more efficient liquid transport through it. Filtration efficiencies above $99$% are obtained for 26-nm particles, decreasing to 99% for 10-nm and 5-nm particulates, which is a significant improvement compared to that obtained with a pure graphene film. While a decrease in filtration rate over time is observed (up to 50% over 3 h of continuous operation) typical to that of graphene and other filter media, the introduction of a simple backwash process allowed for up to 87% recovery of the initial filtration rate. The significant performance enhancement, namely the fourfold increase in throughput, filtration efficiencies of up to 99% for nanoparticles down to 5 nm, and the ability to sustain extended operation at higher powers without disintegration through such a facile and cost-effective modification to the film highlights its potential applicability in practical nanofiltration systems.