We approach the inverse problem of imaging as a task of automated feature learning of the underlying wave equation. While feature extraction with unsupervised machine learning is widely used in analyzing complex data on clustering, classification, and visualization, we show its direct usage on discovering interpretable physical concepts and then obtaining images from wave-propagation data. By using a spring-mass lattice as a simplified model in acoustic imaging, a variational autoencoder is trained to extract features that govern the dynamics of wave propagation from configurations with uncorrelated random material parameters. The extracted features are transformed to images of spring constants and masses by an additional linear regression. The current scheme of extraction-based inverse imaging of features is robust against noise in wave-propagation data and incomplete accessibility in the case of inverse scattering in practice. Our approach requires minimal prior knowledge of the wave-scattering mechanism with applications from the extraction of physical constants and defect detection to discovering physical phenomena.

We study the impact of experimental imperfections in integrated optical circuits. We discuss the emergence of a moderate biased error in path encoding, and investigate its correlation with properties of the optical paths. Our analysis connects and deepens previous studies in this direction, revealing potential issues in high-precision tests and optical implementations of machine learning.

Nonlocal point-to-point correlations between two photons have been used to produce “ghost” images without placing the camera towards the object. Here we theoretically demonstrate and analyze the advantage of non-Gaussian quantum light in improving the image quality of a ghost imaging system over a traditional Gaussian light source. For any squeezing degree, the signal-to-noise ratio (SNR) of the ghost image can be enhanced by the non-Gaussian operations of photon addition and subtraction on a two-mode squeezed light source. We find striking evidence that using non-Gaussian coherent operations, the SNR can be promoted to a high level even within the extremely weak squeezing regime. The resulting insight provides experimental recipes of quantum imaging using non-Gaussian light for illumination.

The canonical way to sustain photonic topological edge states relies on the interface formed by topologically trivial and nontrivial photonic crystals that share a common band gap to prevent the coupling of edge states from the radiation continuum of free space. Herein, we report a scheme to achieve bound topological edge states in the continuum at the air edge of a single topologically nontrivial photonic crystal utilizing the physics of bound states in the continuum. These localized states, though existing in the continuum, exhibit topological features inherited from the nontrivial topology of bulk photonic bands, such as pseudospin-momentum locking unidirectional propagation. Microwave experiments are provided to further validate the theoretical predictions. Our work not only has practical importance but also opens an avenue where the nontrivial topology of bulk photonic bands could be exploited to enrich the physics of bound states in the continuum.

Conventional acoustic metasurfaces are constructed with “locally” gradient phase-shift profiles provided by subunits. The local strategy implies the ignorance of the mutual coupling between subunits, which limits the efficiency of targeted sound manipulation, especially in complex environments. By taking into account the “nonlocal” interaction among subunits, nonlocal metasurface offers an opportunity for accurate control of sound propagation, but the requirement of the consideration of gathering coupling among all subunits, not just the nearest-neighbor coupling, greatly increases the complexity of the system and therefore hinders the explorations of functionalities of nonlocal metasurfaces. In this work, empowered by deep-learning algorithms, the complex inverse gathering coupling can be learned efficiently from the preset dataset so that the inverse mechanism of nonlocal metasurfaces can be described effectively. As an example, we demonstrate that nonlocal metasurfaces, which can redirect an incident wave into multichannel reflections with arbitrary energy ratios, can be accurately predicted by deep-learning algorithms. Compared to the theory, the relative error of the energy ratios is less than 1%. Furthermore, experiments witness three-channel reflection with three types of energy ratios of (1, 0, 0), (1/2, 0, 1/2), and (1/3, 1/3, 1/3), proving the validity of the deep-learning-enabled nonlocal metasurfaces. Our work might blaze an alternative trail in the design of acoustic functional devices, especially for the cases containing complex wave-matter interactions.

Planar structure and small atomic mass usually induce high lattice thermal conductivity in a crystalline solid. Here, we present a first-principles study on the lattice dynamics and phonon-transport properties of a series of ABX honeycomb compounds and show that replacing cations with lighter atoms may yield lower lattice thermal conductivity $({\kappa }_l)$. The anomalous mass-trend ${\kappa }_l$, which is experimentally observed, is found to be inherent to the lattice vibration of the specific layered honeycomb structure, using A$\mathrm{Cu}\mathrm{Sb}$ (A = $\mathrm{Ca}$, $\mathrm{Sr}$, and $\mathrm{Ba}$) as prototypical compounds. We reveal that a lighter A atom bonds more weakly with the $\mathrm{Cu}\text{−}\mathrm{Sb}$ honeycomb ring, giving rise to stronger vibrational anharmonicity, which induces ultralow ${\kappa }_l$ in $\mathrm{Ca}\mathrm{Cu}\mathrm{Sb}$ that is unexpected of light elements and a high-symmetry planar structure. Combined with the high degeneracy and strong anisotropy of the bottom conduction band, $\mathrm{Ca}\mathrm{Cu}\mathrm{Sb}$ exposes an ultrahigh n-type zT of about 2.2 at 700 K. These findings elucidate the mechanism governing anomalous phonon-transport behavior, which also offers guidance in discovering low-cost and low-mass-density materials for advanced thermoelectric applications.

We demonstrate a method to reduce number fluctuations in an ultracold atomic sample using real-time feedback. By measuring the Faraday rotation of an off-resonant probe laser beam with a pair of avalanche photodetectors in a polarimetric setup, we produce a proxy for the number of atoms in the sample. We iteratively remove a fraction of the excess atoms from the sample to converge on a target proxy value in a way that is insensitive to environmental perturbations and robust to errors in light polarization. Using absorption imaging for out-of-loop verification, we demonstrate a reduction in the number fluctuations from $3\mathrm{\%}$ to $0.45\mathrm{\%}$ for samples at a temperature of $16.4\mu \mathrm{K}$ over the time scale of several hours, which is limited by temperature fluctuations, beam-pointing noise, and photon shot noise.

We develop a concept of metasurface-assisted ghost imaging for nonlocal discrimination between a set of polarization objects. The specially designed metasurfaces are incorporated in the imaging system to perform parallel state transformations in general elliptical bases of quantum-entangled or classically correlated photons. Then, only four or fewer correlation measurements between multiple metasurface outputs and a simple polarization-insensitive bucket detector after the object can allow for the identification of fully or partially transparent polarization elements and their arbitrary orientation angles. The approach can find applications for real-time and low-light imaging across diverse spectral regions in dynamic environments.

Quantum holonomic gates hold built-in resilience to local noises and provide a promising approach for implementing fault-tolerant quantum computation. We propose to realize high-fidelity holonomic $(N+1)$-qubit controlled gates using Rydberg atoms confined in optical arrays or superconducting circuits. We identify the scheme, deduce the effective multibody Hamiltonian, and determine the working condition of the multiqubit gate. Uniquely, the multiqubit gate is immune to systematic errors, i.e., laser parameter fluctuations and motional dephasing, as the $N$ control atoms largely remain in the very stable qubit space during the operation. We show that $C_N$-not gates can reach the same level of fidelity at a given gate time for $N\le 5$ under a suitable choice of parameters, and the gate tolerance against errors in systematic parameters can be further enhanced through optimal pulse engineering. In the case of Rydberg atoms, the proposed protocol is intrinsically different from typical schemes based on Rydberg blockade or antiblockade. Our study paves an alternative way to build robust multiqubit gates with Rydberg atoms trapped in optical arrays or with superconducting circuits. It contributes to current efforts to develop scalable quantum computation with trapped atoms and fabricable superconducting devices.

We report on the observation of a complex nonlinear intensity dependence of the circular and linear photogalvanic currents induced by infrared radiation in compensated (${\mathrm{Bi}}_{0.3}{\mathrm{Sb}}_{0.7}{)}_2$(${\mathrm{Te}}_{0.1}{\mathrm{Se}}_{0.9}{)}_3$ three-dimensional topological insulators. The photocurrents are induced by direct optical transitions between topological surface and bulk states. We show that an increase in the radiation intensity results first in a highly superlinear rise in the amplitude of both types of photocurrents, whereas at higher intensities the photocurrent saturates. Our analysis of the observed nonlinearities shows that the superlinear behavior of the photocurrents is caused by heating of the electron gas, while the saturation is induced by slow relaxation of the photoexcited carriers, resulting in absorbance bleaching. The observed nonlinearities give access to the Fermi-level position with respect to the Dirac point and the energy relaxation times of Dirac fermions, providing an experimental room-temperature probe of topological surface states.

Scholte modes that are localized between a submerged axisymmetric structured elastic plate and surrounding fluid can undergo mode conversion via Umklapp diffraction into radiative modes; this radiative response is verified by experiments that show focusing of underwater sound across a broad range of frequencies. The diffracted beams, that form a cone, are engineered to exist at a desired spatial position, associated with an abrupt change in the patterning of the plate. These structures take the form of grooves present only on one side of the plate, yet the focusing phenomena is achieved on both sides, even as viewed from the flat surface.

We present an analysis of gated $(\mathrm{In},\mathrm{Ga})\mathrm{As}$/$(\mathrm{In},\mathrm{Al})\mathrm{As}$ heterostructures, a device platform to realize spin-orbitronic functionalities in semiconductors. The phenomenological model deduced from our magnetotransport experiments allows us to correlate the gate response of the two-dimensional electron systems with the design parameters of the heterostructure, in particular the indium concentration. We explain the occurrence of metastable electrostatic configurations showing reduced capacitive coupling and provide gate-operation strategies to reach classical field-effect control in such heterostructures. Our study highlights the role of the intrinsic $(\mathrm{In},\mathrm{Al})\mathrm{As}$ deep donor defects, as they govern the dynamics of the electrostatic response to gate-voltage variations through charge trapping and unintentional tunneling.

A coupled magnetostructural transition (MST) from a high-temperature low-magnetic ${\mathrm{Ni}}_2\mathrm{In}$-type hexagonal phase to a low-temperature high-magnetic $\mathrm{Ti}\mathrm{Ni}\mathrm{Si}$-type orthorhombic phase, and the associated large magnetocaloric effect, are explored for the $({\mathrm{Mn}}_{0.6}{\mathrm{Fe}}_{0.4})({\mathrm{Ni}}_{1-x}{\mathrm{Co}}_x)\mathrm{Si}$ system. Room-temperature structural analysis is carried out to investigate the orthorhombic distortion of the hexagonal phase initiated by introducing $\mathrm{Co}$ at the $\mathrm{Ni}$ site. The indirect isothermal magnetic measurement technique is employed to estimate the magnetocaloric effect across the coupled phase transition. Fascinatingly, a relatively large value of isothermal magnetic entropy change (Δ$S_{M}$) of (−12.02 ± 0.34) J ${\rm{kg}}^{-1}\;{\rm{K}}^{-1}$ at (332.5 ± 0.2) K and relative cooling power (RCP) of (119.6 ± 3.5) J ${\rm{kg}}^{-1}$ is achieved for the composition $({\mathrm{Mn}}_{0.6}{\mathrm{Fe}}_{0.4})({\mathrm{Ni}}_{0.6}{\mathrm{Co}}_{0.4})\mathrm{Si}$, for a field change of 30 kOe. The complex hysteresis behavior around the phase transition determines the efficiency of the material, which is revealed by both temperature- and field-sweep-rate dependence magnetization studies. Superheating (or supercooling) of the parent phase observed at higher temperature and field-sweep rates leads to broad thermal- and field-induced hysteresis. Lower sweep rates of temperature and magnetic field remove the quenching phenomenon across the MST and enhance the working efficiency. A large magnetocaloric effect and negligible variation of hysteresis with the temperature- and magnetic-field-sweep rates compared with other alloys makes this nominal low-cost material a promising candidate for practical solid-state cooling applications.

The goal of quantum metrology is the precise estimation of parameters using quantum properties such as entanglement. This estimation usually consists of three steps: state preparation, time evolution during which information of the parameters is encoded in the state, and readout of the state. Decoherence during the time evolution typically degrades the performance of quantum metrology and is considered to be one of the major obstacles to realizing entanglement-enhanced sensing. We show, however, that under suitable conditions, this decoherence can be exploited to improve the sensitivity. Assume that we have two axes, and our aim is to estimate the relative angle between them. Our results reveal that the use of Markovian collective dephasing to estimate the relative angle between the two directions affords Heisenberg-limited sensitivity. Moreover, our protocol based on Markovian collective dephasing is robust against environmental noise: it is possible to achieve the Heisenberg limit by applying the collective dephasing even under the effect of independent dephasing. Our counterintuitive proposal with the decoherence leads to alternative applications in quantum metrology.

For field emission (FE), it is widely expected that its emitting current density $J$ will become space-charge-limited current (SCLC) within a gap spacing $D$ biased at sufficiently large voltage $V$. In this paper, we reveal a peculiar finding in which this expected two-stage transition (from FE to SCLC) is no longer valid for FE not obeying the traditional Fowler-Nordheim (FN) law. By employing a generalized FN scaling of $\ln (J/E_s^k)\propto -1/E_s$, where $E_s$ is the surface electric field at the cathode, we show the existence of a critical exponent $k_c\equiv 3/2$ where unusual behaviors occur for $k<k_c$: (a) only FE at small $D$ (no transition to SCLC possible) and (b) three-stage transition from FE first to SCLC then back to FE at large $D$. For $k>k_c$, the conventional two-stage transition from FE to SCLC will always occur for all $D$, including the classical case of $k$ = 2. Using various unconventional FE models with $k\ne 2$, we specifically demonstrate these peculiar transitions. Under a normalized model, our findings uncover the rich interplay between the source-limited FE and bulk-limited SCLC over a wide range of operating conditions.

The emergence of a photonic band gap in $\mathrm{Ge}$-on-$\mathrm{Si}$ micropillars ordered in a two-dimensional square lattice is demonstrated by the finite-element method. Candidate architectures are fabricated through epitaxy and the opening of the photonic band gap experimentally proved by photoluminescence spectroscopy. When the direct-gap emission of $\mathrm{Ge}$ is resonantly driven into the photonic gap, light propagation in the lattice plane is inhibited. Emission is eventually funneled out of plane, yielding a giant increase, i.e., about one order of magnitude, in the observed intensity. The demonstration of light routing in microcrystals’ lattices opens interesting possibilities for $\mathrm{Si}$ photonics. The epitaxial self-assembled microstructures introduced here can be monotonically integrated on $\mathrm{Si}$ to improve the performances of group-IV lasers or engineered to optimize the working wavelength of future quantum photonic circuits.

We analyze the complex impact of the local magnetic spin texture on the transverse Hall-type voltage in device structures utilized to measure magnetoresistance effects. We find a highly localized and asymmetric magnetic sensitivity in the eight-terminal geometries that are frequently used in current-induced switching experiments, for instance, to probe antiferromagnetic materials. Using current-induced switching of antiferromagnetic $\mathrm{Ni}\mathrm{O}/\mathrm{Pt}$ as an example, we estimate the change in the spin Hall magnetoresistance signal associated with switching events based on the domain-switching patterns observed via direct imaging. This estimate correlates with the actual electrical data after subtraction of a nonmagnetic contribution. Here, the consistency of the correlation across three measurement geometries with fundamentally different switching patterns strongly indicates a magnetic origin of the measured and analyzed electrical signals.

Hyperbolic metamaterials (HMMs) support propagating waves with arbitrarily large wavevectors over broad spectral ranges, and are uniquely valuable for engineering radiative thermal transport in the near field. Here, by employing a rational-design approach based on the electromagnetic local density of states, we demonstrate the ability of HMMs to substantially rectify radiative heat flow. Our idea is to establish a forward-biased scenario where the two HMM-based terminals of a thermal diode feature overlapped hyperbolic bands which result in a large heat current, and suppress the reverse heat flow by creating spectrally mismatched density of states as the temperature bias is flipped. As an example, we present a few high-performance thermal diodes by pairing HMMs made of polar dielectrics and metal-to-insulator transition materials in the form of periodic nanowire arrays, and considering three representative kinds of substrates. Upon optimization, we theoretically achieve a rectification ratio of 324 at a 100-nm gap, which remains greater than 148 for larger gap sizes up to 1 µm over a wide temperature range. The maximum rectification represents an almost 1000-fold increase compared with a bulk diode using the same materials, and is twice that of state-of-the-art designs. Our work highlights the potential of HMMs for rectifying radiative heat flow, and may find applications in advanced thermal-management and energy-conversion systems.

Optical high-finesse cavities are a well-known means to enhance light-matter interactions. Despite much progress in the realization of strongly coupled light-matter systems, the controlled positioning of single solid emitters in cavity modes remains a challenge. We pursue the idea of using nanofibers with subwavelength diameter as a substrate for such emitters. This paper addresses the question of how strongly optical nanofibers influence the cavity modes. We analyze the influence of the fiber position for various fiber diameters on the finesse of the cavity and on the shape of the modes.