The edge states in a one-dimensional Su-Schrieffer-Heeger (SSH) chain have been extensively studied for their topologically protected properties when there are onsite dissipations induced by independent environments. We here show that common environments shared by nearest neighboring sites in the SSH chain can result in dissipative couplings between sites, and thus change the topologically trivial phase to a nontrivial one. In contrast to usual hybridization of two edge states in the SSH chain with finite size, we find that the dissipative couplings result in oscillations of these hybridized edge states, which are similar to those of Majorana modes in the semiconductor-superconductor nanowire. The parameter for controlling the topological phase of the SSH chain plays the role of the magnetic field in a nanowire for controlling Majorana oscillation. A measurement method for oscillations is proposed. Our study provides a useful way to manipulate and ascertain edge states, and is experimentally feasible within current technology of superconducting quantum circuits.

Fundamental aspects of spin-orbit interaction in commercially relevant $\mathrm{Ga}\mathrm{N}$/${\mathrm{Al}}_x{\mathrm{Ga}}_{1-x}\mathrm{N}$ heterostructures hosting two-dimensional electron systems are extensively studied by electron spin resonance (ESR). This unprecedentedly accurate experimental technique allows access to the fine details of coupling between the spin degree of freedom and the quantized orbital motion of an electron in the quantum Hall regime through the precise measurement of the electron $g$-factor. Filling-factor-dependent changes of the $g$-factor value in strong magnetic fields allow extraction of the Rashba spin-orbit interaction constant $\alpha$ in various $\mathrm{Ga}\mathrm{N}$/${\mathrm{Al}}_x{\mathrm{Ga}}_{1-x}\mathrm{N}$ heterojunctions with electron sheet densities in the range $0.8–5.2\times {10}^{12}{\mathrm{cm}}^{-2}$. Despite three significantly different approaches used to tune the electron density, the extracted value of $\alpha$ is on the order of $5.3\pm 0.4$ meV $\text{Å}$ for all experimental realizations. This striking finding can be explained by assuming that the spin-orbit interaction is of bulk origin. Theoretical calculations confirm this observation, as the bulk cubic in-plane wave-vector term of the spin-orbit interaction compensates the rising contribution due to the change in the quantum well shape. Finally, the value of $\alpha$ is cross-checked and confirmed by weak antilocalization measurements in the longitudinal magnetoresistance at substantially lower magnetic field values compared to the ESR approach. The presented experimental findings provide knowledge for estimating the true scales of spin-orbit interaction in GaN-based spintronic devices.

We demonstrate the storage and retrieval of an on-demand single photon generated by a collective Rydberg excitation in a low-noise Raman quantum memory located in a different cold atomic ensemble. We generate single photons on demand by exciting a cold cloud of rubidium atoms off resonantly to a Rydberg state, with a generation probability up to 15% per trial. We then show that the single photons can be stored and retrieved with a storage-and-retrieval efficiency of 21% and a low noise floor in the Raman quantum memory. This leads to a signal-to-noise ratio ranging from 11 to 26 for the retrieved single photon, depending on the input-photon-generation probability, which allows us to observe significant antibunching. We also evaluate the performance of the Raman memory as a built-in unbalanced temporal beam splitter, tunable by varying the write-in control-pulse intensity. In addition, we demonstrate that the Raman memory can be used to control the single-photon wave shape. These results are a step forward in the implementation of efficient quantum repeater links using single-photon sources.

We design a pair of resonators to realize multiple spoof plasmonically induced transparency (PIT) with high-Q group delays, which show potential in sensing applications. To reveal the mechanism of the multiple spoof PIT scheme, we deduce mathematical equations of a simplified model composed of multiple harmonic oscillators, and we further propose an equivalent-circuit model to analyze the metamaterials in an intuitive way. A prototype consisting of a broadband stub resonator and a spoof localized surface-plasmonic resonator is further fabricated to detect the complex permittivity of high-lossy liquids and solid samples in microwave bands. Experimental results show that the group-delay-based sensor can work well even when high-order modes are weakly excited, and the trade-off between sensitivity and properties of samples is discussed. The proposed double-resonator scheme with multiple spoof PIT is universal, and we design a metasurface working in terahertz bands to show that the concept can be widely used in various surface-sensing applications.

We achieve a compact optical frequency standard with an extended cavity diode laser locked to the 459-nm $6S_{1/2}$-$7P_{1/2}$ transition of thermal ${}^{133}\mathrm{Cs}$ atoms in a $\varphi 10\mathrm{mm}\times 50\mathrm{mm}$ glass cell, using modulation transfer spectroscopy (MTS). The self-estimated frequency stability of this laser is $1.4\times {10}^{-14}/\sqrt{\tau }$. With heterodyne measurement, we verify the linewidth-narrowing effect of MTS locking and measure the frequency stability of the locked laser. The linewidth of each laser is reduced from the free-running 69.6 to 10.3 kHz after MTS stabilization, by a factor of 6.75. The Allan deviation measured via beat detection is $2.1\times {10}^{-13}/\sqrt{\tau }$ for each MTS-stabilized laser. In addition, we measure the hyperfine structure of the $7P_{1/2}$ energy level based on the heterodyne measurements, and calculate the magnetic dipole constant A of the Cs $7P_{1/2}$ level to be 94.38(6) MHz, which agrees well with previous measurements. This compact optical frequency standard can also be used in other applications that require high-stability lasers, such as laser interferometry, laser cooling, geodesy, and so on.

Engineering desired Hamiltonians in quantum many-body systems is essential for applications such as quantum simulation, computation, and sensing. Many widely used quantum Hamiltonian engineering sequences are designed using human intuition based on perturbation theory, which may not describe the optimal solution and is unable to accommodate complex experimental imperfections. Here we numerically search for Hamiltonian engineering sequences using deep reinforcement learning (DRL) techniques and experimentally demonstrate that they outperform celebrated decoupling sequences on a solid-state nuclear magnetic resonance quantum simulator. As an example, we aim at decoupling strongly interacting spin-1/2 systems. We train DRL agents in the presence of different experimental imperfections and verify robustness of the output sequences both in simulations and experiments. Surprisingly, many of the learned sequences exhibit a common pattern that had not been discovered before, to our knowledge, but has a meaningful analytical description. We can thus restrict the searching space based on this control pattern, allowing us to search for longer sequences, ultimately leading to sequences that are robust against dominant imperfections in our experiments. Our results not only demonstrate a general method for quantum Hamiltonian engineering, but also highlight the importance of combining black-box artificial intelligence with an understanding of the physical system in order to realize experimentally feasible applications.

We demonstrate the use of radio-frequency (rf) resonators to measure the capacitance of nanoscale semiconducting devices in field-effect transistor configurations. The rf resonator is attached to the gate or the lead of the device. Consequently, tuning the carrier density in the conducting channel of the device affects the resonance frequency, quantitatively reflecting its capacitance. We test the measurement method on InSb and InAs nanowires at dilution-refrigerator temperatures. The measured capacitances are consistent with those inferred from the periodicity of the Coulomb blockade of quantum dots realized in the same devices. In an implementation of the resonator using an off-chip superconducting spiral inductor we find the measurement sensitivity values reaching down to $75\mathrm{zF}/\sqrt{\mathrm{Hz}}$ at 1 kHz measurement bandwidth, and noise down to 0.45 aF at 1 Hz bandwidth. We estimate the sensitivity of the method for a number of other implementations. In particular, we predict a typical sensitivity of about $40\mathrm{zF}/\sqrt{\mathrm{Hz}}$ at room temperature with a resonator composed of off-the-shelf components. Of several proposed applications, we demonstrate two: the capacitance measurement of several identical 80-nm-wide gates with a single resonator, and the field-effect mobility measurement of an individual nanowire with the gate capacitance measured in situ.

Speckle illumination is extensively applied in various imaging scenarios due to its advantages in enhancing the performance of imaging. Generally, the existing speckle-illumination methods are mainly based on external spatial modulation of light through a rotating diffuser or spatial light modulator. Both the lighting efficiency and the system’s compactness are limited for these conventional external modulation methods, due to their free-space light coupling and insertion or diffraction loss. Here, a multimode fiber laser with self-reconfigurable speckle illumination is proposed and realized, benefiting from the temporal-spectral randomness of a homemade coherent random-fiber laser and the spectral-sensitive output pattern of an out-coupling multimode fiber. Output characteristics in the spectral and spatial domain of the proposed laser are analyzed and show dynamically varying speckles within a moderate bandwidth. These features pave efficient ways for superresolution imaging through an opaque scattering medium as well as ghost imaging, which is realized and demonstrated based on the lighting of the proposed laser.

Magnetic skyrmions are topologically protected particlelike spin textures initially discovered in chiral magnetic materials. When driven by an electric current, magnetic skyrmions move at a certain angle (i.e., skyrmion Hall angle) with respect to the driving current due to the Magnus force, giving rise to the so-called skyrmion Hall effect (SHE). The SHE often leads to skyrmion annihilation at the sample edge, which is detrimental for practical applications. In this work, we study, experimentally and through micromagnetic simulations, the current-driven dynamics of skyrmion bubbles (SBs) in $\mathrm{Pt}/\mathrm{Co}/\mathrm{Ru}$ multilayer stacks. It is observed that, under certain circumstances, a long stripe domain will form and align along the sample edge, which separates the skyrmion bubbles from the edge and avoids their annihilation at the edge. Furthermore, the velocity of the skyrmion bubbles can be increased by the stripe domain wall (DW) through the DW-SB interaction, as predicted in our early theory. Our results deepen our understanding of skyrmion dynamics and pave the way to the realization of highly efficient information processing and computing devices based on ultrafast skyrmion motion.

Vortex-field acoustic levitation is an application of acoustic vortices that allows the manipulation of a wide range of materials in various media. However, to date, its levitation capability in air is limited to relatively low-density objects. Here, by optimizing an array-tube setup, we levitate iridium ($22.56\mathrm{g}/{\mathrm{cm}}^3$) in a first-order acoustic vortex in air. This technique makes it possible to observe the orbital angular momentum transfer carried by acoustic vortices to freely levitated high-density objects, of which the highest rotation speed reaches up to 6500 revolutions per second and increases with a decrease in the object size and an increase in the source amplitude. It is revealed that the thermal-viscous boundary layer dissipation is the dominant effect and leads directly to the rapid rotation of high-density objects. This work can expand the application scope of acoustic vortices, especially in the fields of containerless processing of various materials and rapid rotation of objects, and help to understand the mechanisms of orbital angular momentum transfer to matter.

In contrast to software simulations of neural networks, hardware implementations have limited or no tunability. While such networks promise great improvements in terms of speed and energy efficiency, their performance is limited by the difficulty of applying efficient training. We propose and realize experimentally an optical system where highly efficient backpropagation training can be applied through an array of highly nonlinear, nontunable nodes. The system includes exciton-polariton nodes realizing nonlinear activation functions. We demonstrate high classification accuracy in the MNIST handwritten digit benchmark in a single-hidden-layer system.

Nonreciprocal transmission of structured light has possible applications in high-capacity optical communication and high-dimensional signal processing. Here, we demonstrate the magnet-free and cavity-free optical nonreciprocity of an orbital-angular-momentum (OAM) beam using optical-pumping technology with the assistance of the Doppler effect in hot atoms. By using optical pumping to induce a direction-dependent population transfer of the ground-state atoms, we realize optical nonreciprocity for the signal laser that is carrying OAM. Interestingly, an isolation ratio close to 30 dB and maximum transmission over 0.86 can be achieved simultaneously without the use of an optical cavity. Due to the unique amplitude and helical phase profile of the structured light, our proposal may open an avenue for exploring the singularity characteristics of optical nonreciprocity using structured light.

Broadband precision spectroscopy is an invaluable tool for high-fidelity species diagnostics as well as multispecies sensing in complex real-world environments. However, concurrent demands for broad spectral coverage and high spectral resolution present great challenges to both optical measurement systems and data acquisition and processing. We report methodological and experimental demonstration of a midinfrared direct optical frequency comb (OFC) spectrometer, using Fourier-transform spectroscopy (FTS) with a compressed sensing (CS) algorithm. Efficient data sampling and accurate spectral recovery are demonstrated with comb-teeth-resolving resolution of $0.0083{\mathrm{cm}}^{-1}$ (250 MHz) over a spectral range of $300{\mathrm{cm}}^{-1}$. The $v_3$ band of ${\mathrm{CH}}_4$ is measured, and validated against conventional FTS results and theoretical calculations. Studies on the effect of compression ratio show that a compressed interferogram using as few as 2% of the regular sample points can effectively recover the high-resolution broadband spectrum with an accuracy of 1.9%. The developed diagnostic system bears promising potential for accurate, real-time, multispecies sensing that is pressingly needed for studying the dynamics of complex multicomponent systems.

A coherent-synchrotron-emission regime of forward high-order harmonics generation (HHG) is proposed for the emission of an isolated unipolar half-cycle attosecond pulse by a three-color laser pulse impinging on ultrathin foil. A theoretical model is proposed for the electron nanobunching mechanism and the forward radiation, which is consistent with the numerical-simulation results. As the forward HHG does not need to penetrate from the front to the rear side of the target, the spectrum of the forward HHG has no low-frequency cutoff. The robustness of this regime is verified with different laser and foil plasma conditions as well as the two-dimensional effects. The robustness is also checked with similarity theory, which confirms that isolated attosecond pulses can be efficiently generated when the plasma density and the laser amplitude change simultaneously such that their ratio remains unchanged.

The ion microwave frequency standard is a candidate for the next generation of microwave frequency standard with the potential for very wide applications. The Dick effect and second-order Doppler frequency shift (SODFS) limit the performance of ion microwave frequency standards. The introduction of sympathetic cooling technology can suppress the Dick effect and SODFS and improve the stability and accuracy of the frequency standard. However, the sympathetically cooled ion microwave frequency standard has seldom been studied before. This paper reports a high-performance sympathetically cooled ion microwave frequency standard in a Paul trap. Using laser-cooled ${}^{40}{\mathrm{Ca}}^{+}$ as coolant ions, ${}^{113}{\mathrm{Cd}}^{+}$ ion crystal is cooled to below 100 mK and has a coherence lifetime of over 40 s. The short-term frequency stability reached $3.48\times {10}^{-13}/{\tau }^{1/2}$, which is comparable to that of the mercury ion frequency standard. Its uncertainty is $1.5\times {10}^{-14}$, which is better than that of directly laser-cooled cadmium ion frequency standard.

Deep learning is found to be a powerful data-driven force to transform the way we discover, design, and utilize photonics and metasurfaces. More recently, there has been growing interest in deep-learning-enabled on-demand structural design, as it can ease the limitations of low efficiency, time-consuming, and experience navigation in conventional design. However, training data is a valuable source, especially for high-dimensional scatterers. It is extremely challenging and costly to keep the pace of data collection with the increasing degrees of freedom. Here, we propose a transfer-learning-assisted inverse-metasurface-design method to relieve the data dilemma. A flexible transferrable neural network composed of an encoder-decoder network and a physical assistance network is constructed, the latter of which is attached to solve the nonuniqueness problem. Starting from the 5 × 5 metasurface, we successfully migrate the inverse design to a 20 × 20 metasurface, with a Pearson correlation coefficient that reaches 97%. Compared with direct learning, the data requirement is reduced by over 30%. In the experiment, we validate the concept via wave-front customization. Our work constitutes a green and efficient inverse-design paradigm for fast far-field customization and provides a key advance for the next generation of large-scale intelligent metadevices.

Gradient-index (GRIN) lenses have paved the way for electromagnetic wave tailoring due to their spatially varying permittivities, which enable continuous manipulation of the wave as it passes through the lens volume. However, traditional GRIN design methodologies, such as transformation optics or geometrical optics, often generate lens profiles that are limited to single-band performance. Here, we propose a multiobjective optimization strategy customized for designing dual-band GRIN lenses. Utilizing this powerful inverse-design tool, in conjunction with an advanced manufacturing technique, a GRIN lens with compact size and nonintuitive permittivity distribution is designed, fabricated, and characterized. The lens is capable of realizing two unique gain enhancement objectives at the L and C bands simultaneously, where different feeding sources are employed at the two bands.

A microscopic understanding of the mixed-functionalized MXenes ($M_2{X-T}_x$; $T_x=\text{O, OH,}$ and $\mathrm{F}$) are extremely important to the design of efficient ${\mathrm{CO}}_2$ catalytic activity. Here, we report a first-principles study of the ${\mathrm{CO}}_2$ activation on pure $\mathrm{O}$, $\mathrm{OH}$, and $\mathrm{F}$ and mixed-functionalized ${\mathrm{Sc}}_2\mathrm{C}$ MXene surfaces. We find that ${\mathrm{CO}}_2$ adsorption energy can be tuned by changing the coverage of $\mathrm{O}$, $\mathrm{OH}$, and $\mathrm{F}$ functional groups on the surface. Fully terminated $\mathrm{O}$ and $\mathrm{F}$ ${\mathrm{Sc}}_2\mathrm{C}$ forms weak interactions with ${\mathrm{CO}}_2$ molecules (binding energy $-0.136$ and $-0.168\mathrm{eV}$) whereas mixed-functionalized ${\mathrm{Sc}}_2\mathrm{C}$ surface exhibits higher binding energy($-0.364\mathrm{eV}$). In the mixed-functionalized ${\mathrm{Sc}}_2\mathrm{C}$ case, only $\mathrm{O}$ sites allow ${\mathrm{CO}}_2$ reduction ($\mathrm{F}$ and $\mathrm{OH}$ are inactive) and finally converts into methane (${\mathrm{CH}}_4$). Ab-initio-based Bader charge analysis and projected density of state calculations reveal strong bonding between the C atom of ${\mathrm{CO}}_2$ and $\mathrm{O}$ functional group. The Gibbs free-energy calculation confirms the conversion of $\mathrm{HCO}$ into ${\mathrm{H}}_2\mathrm{CO}$ to be a rate-limiting step with the limiting potential 1.387 eV. In the mixed-functionalized surface, as we increase the number of $\mathrm{OH}$ groups in the vicinity of $\mathrm{O}$ sites, the binding energy increases (transiting from a physisorption to a chemisorption regime). However, increasing the amount of O coverage turns out to be detrimental to the catalytic activity. Our study highlights the role of different functional groups in achieving efficient ${\mathrm{CO}}_2$ catalytic activity on ${\mathrm{Sc}}_2\mathrm{C}$ MXene, which can further help us to design experiments accordingly.

Multiport interferometers based on integrated beamsplitter meshes are widely used in photonic technologies. While the rectangular mesh is favored for its compactness and uniformity, its geometry resists conventional self-configuration approaches, which are essential to programming large meshes in the presence of fabrication error. Here, we present a configuration algorithm, related to the $2\times 2$ block decomposition of a unitary matrix, that overcomes this limitation. Our proposed algorithm is robust to errors, requires no prior knowledge of the process variations, and relies only on external sources and detectors. We show that self-configuration using this technique reduces the effect of fabrication errors by the same quadratic factor observed in triangular meshes. This relaxes a significant limit to the size of multiport interferometers, removing a major roadblock to the scaling of optical quantum and machine-learning hardware.