Spontaneous parametric down-conversion (SPDC) is the most widely used source of high-dimensional entangled two-photon states, and the entanglement in the spatial degree of freedom is considered best suited for harnessing high-dimensional advantages. Although the Schmidt basis provides a natural choice for state characterization of entangled two-photon states in any degree of freedom, the existing techniques can only reconstruct the Schmidt spectrum when the Schmidt basis is known a priori. There is currently no technique that can measure the unknown Schmidt basis of an entangled two-photon state. In this paper, we present a technique that measures the complete set of spatial Schmidt modes without any prior knowledge of the Schmidt basis, as long as the cross-spectral density functions corresponding to individual photons are of Schell-model type. Using this technique, we report measurement of states with over 3000 Schmidt modes—highest reported yet—with up to 98% fidelity. Our technique works even when the entangled field is very low-dimensional or entirely separable, as well as when the field pumping the SPDC is structured. We expect our work to significantly advance harnessing of spatial entanglement.

In optics, Brewster’s effect is a significant phenomenon widely used in photonics and laser physics to control radiation polarization at the interface between two dielectric media. In this work, we show that the Brewster effect is clearly evident when light passes through a small nanohole made in a thick metal film (zero-mode waveguide, ZMW). We develop a model to describe this phenomenon, involving increased excitation efficiency of the dipole moment associated with the ZMW at a specific angle—the Brewster angle—at the metal and dielectric interface. At this angle, the transmittance for $p$-polarized light through the ZMW is $20\pm 4$ times higher than for $s$-polarized light, opening new possibilities for applications of ZMW in single-molecule microscopy methods.

Microwave kinetic inductance detectors (MKIDs) are superconducting photon counting detectors that simultaneously measure the energy and arrival time of individual photons and are suitable for large-format arrays. The energy resolving power $R=E/\mathrm{\Delta }E$ is the primary figure of merit for spectrophotometric applications. It is limited in practice by the assumptions underlying the Wiener optimal filter used to estimate photon energy: stationary noise, linear response, and energy independent pulse shapes, all of which MKIDs structurally violate. We present Very Efficient Neural Optimal-filter for MKIDs (Venom), a selective state-space model based on the Mamba architecture that replaces the coordinate transform and optimal filter, operating directly on raw in-phase and quadrature timestream data, carrying 3314 trainable parameters, and targeting deployment on the MKIDGen3 and future system-on-chip readout platforms. To deal with the small calibration datasets typical of MKID experiments, we train on synthetic pulses drawn from a streaming principal component analysis that samples photon energy continuously between calibration wavelengths. All numbers are reported on a stratified 15% held-out validation set, with $R$ measured from a kernel density estimate. On an InHf bilayer resonator (ten wavelengths, 254–1310 nm), Venom reaches a mean $R$ of 26.4 vs 24.2 for the published per-wavelength optimal filter baseline, a 9% improvement from one model in place of ten per-laser filter templates. The $R$ of 38.6 at 254 nm is the highest ever recorded for an ultraviolet-to-near-infrared MKID suitable for use in a dense array. On a PtSi resonator (five wavelengths, 808–1310 nm), where the published analysis used one shared 920 nm template, Venom matches the optimal filter, with a mean $R$ of 9.1 vs 8.8 at the three interior wavelengths.

Efficient fluorescence collection is essential for precise quantum measurements based on nitrogen-vacancy (NV) centers in diamond. However, photon extraction is limited by total internal reflection at the diamond-air interface. To address this challenge, we propose a short-focal-length ${\text{Si}}_3{\mathrm{N}}_4$ metalens on a ${\text{SiO}}_2$ support for broadband fluorescence collection over 600–800 nm in an ensemble-NV ODMR system. The metalens is designed as a compact replacement for the conventional lens collection path. FDTD simulations show that the metalens improves the forward photon collection efficiency by a factor of 3.2 compared with bare diamond. Experimentally, ODMR measurements show fluorescence-related signal enhancements of 33% and 45% for microring and double-split-ring microwave antennas, respectively. These results confirm that the proposed design provides an effective and compact solution for enhancing fluorescence collection in diamond-based quantum sensing systems.

Erbium ions (${\mathrm{Er}}^{3+}$) provide a telecom-band optical transition with strong magnetic-dipole character, making them attractive for quantum communication and spin-photon interfaces. Identifying host environments that combine low decoherence with photonic compatibility, however, remains a central challenge. Here we investigate ${\mathrm{Er}}^{3+}$ emission in tungsten disulfide (${\mathrm{WS}}_2$) flakes, a layered host offering low nuclear-spin density and narrow telecom emission. Using time- and polarization-resolved photoluminescence under modest magnetic fields ($<0.2\mathrm{T}$), we observe pronounced dimming, lifetime extension, and rotation of the emission dipole when the field has an out-of-plane component, whereas in-plane fields produce little change. Effective model calculations of ${\mathrm{Er}}^{3+}$ in monolayer ${\mathrm{WS}}_2$ parametrized from density functional theory suggest that these effects arise primarily from Zeeman-induced mixing of near-degenerate crystal-field (CF) sublevels, which modifies the magnitude and orientation of the optical transition dipole moments. Comparative measurements in flakes of different thickness and numerical estimates of the local density of optical states further point to a secondary contribution from dipole coupling to the anisotropic photonic environment of thin ${\mathrm{WS}}_2$ layers. These findings identify layered ${\mathrm{WS}}_2$ as a platform where magnetic fields can tune telecom emission through an interplay of crystal-field physics and anisotropic photonic coupling.

This work investigates the noise characteristics of filamentary- and interfacial-type resistive random-access memory devices in a single material system based on ${\mathrm{La}}_2{\mathrm{NiO}}_{4+\delta }$, using low-frequency noise spectroscopy across different resistance states, read voltages, and device sizes. We observe a clear difference in the noise properties of the two device types: (i) filamentary devices exhibit overall elevated noise levels compared with the interfacial ones, (ii) the relative noise magnitude between high (HRS) and low resistance states (LRS) in the two device types is inverted: elevated noise is present in the HRS of filamentary devices and in the LRS of interfacial ones and vice versa, (iii) no area dependence of noise properties is observed for the filamentary devices, whereas interfacial devices show a size-dependent frequency exponent and noise magnitude scaling. For both switching mechanisms, a systematic two-stage noise suppression with increasing read voltages is observed, which is found to be correlated with the transition from ohmic to space-charge-limited conduction. Our study links noise characteristics to the underlying switching and conduction mechanisms of oxygen vacancy migration-based memories.

Direct laser acceleration leveraging transverse fields promises unparalleled gradients exceeding $1\mathrm{TeV}/\mathrm{m}$ yet is fundamentally constrained by phase slippage, which limits interaction length and final energy gain. Here, we overcome this dephasing limitation by employing a pulse-front-tilted laser interacting with a plasma grating. This configuration ensures sustained phase matching, facilitating extended acceleration and stable near-field confinement. Uniquely, this structure functions as a phase-selective sieve that generates dense electron bunches with durations of hundreds of attoseconds. Particle-in-cell simulations demonstrate that a 61 J laser pulse yields an 11 pC beam with energies surpassing 10 GeV at gradients of approximately $2\mathrm{TeV}/\mathrm{m}$. Derived for an idealized dephasingless regime, linear scaling laws predict that 1 kJ pulses can drive beams to 200 GeV. These findings present a promising pathway toward on-chip TeV-scale accelerators and attosecond sources, with potential applications for high-energy colliders, ultrafast electron imaging, and advanced light sources.

In the Taiji project—a Chinese space-borne gravitational wave detector—the drag-free and attitude control system (DFACS) is the core technology responsible for maintaining the inertial motion of the test mass (TM). A thorough understanding of the spacecraft (SC) dynamic behavior under complex gravitational perturbations and multisource nongravitational disturbances is crucial for optimizing control system design, studying data processing pipelines, and verifying the detector performance. In this paper, we propose a unified simulation framework integrating the constellation orbital motion with a full-state closed-loop DFACS system and develop a 60-degrees-of-freedom (DoF) DFACS simulator. Furthermore, benefiting from the introduction of high-precision numerical methods, the simulator is extended to 31 significant digits, which supports high-fidelity numerical simulations of the intersatellite laser interferometry measurements from picometers to million kilometers. Three different experiments are implemented to assess the performance of DFACS by analyzing the single TM’s residual force noise, moving optical subassembly’s jitter, and the arm-length variation of TM-TM and SC-SC. Moreover, we quantitatively demonstrate the feasibility of decoupling DFACS design and orbit optimization. For the Taiji project, this study enables a more realistic on-orbit simulation of the detector and provides a representative system response under closed-loop control, supporting more accurate detector sensitivity analysis and facilitating refined optimization for scientific objectives.

We investigate the Gilbert damping constant $\alpha$ and the inhomogeneity of the effective magnetic anisotropy field $H_K^{\mathrm{eff}}$ in nanoscale magnetic tunnel junctions (MTJs) with various free-layer diameters $D$, which have two different structures possessing different magnetic properties around the pattern edge. Homodyne-detected ferromagnetic resonance (FMR) is used, allowing to disentangle the inhomogeneous broadening $\mathrm{\Delta }{H}_0$ owing to the spatial variation of $H_K^{\mathrm{eff}}$ from the effects of damping in the linewidth of spectra. We find that $\mathrm{\Delta }{H}_0$ increases with decrease in $D$, while $\alpha$ is almost constant. The degree of the increase in $\mathrm{\Delta }{H}_0$ is found to change with the MTJ structure. We show that these results are explained by a model considering the distribution of a demagnetizing field that increases with the decrease in $D$. The obtained findings provide important insights for the development of nanoscale MTJs with high reliability and high efficiency.

Dimensionality reduction provides a powerful route to tune the electronic and magnetic properties of van der Waals materials, yet its influence on electronic transport remains complex due to competing effects from quantum confinement and modified scattering mechanisms. Here, we investigate this interplay in an antiferromagnetic semiconductor ${\text{NaCrTe}}_2$ using first-principles calculations combined with the Boltzmann transport equation beyond the constant relaxation time approximation. Our results show that the monolayer limit induces a coupled magnetostructural reconstruction, reducing the band gap from 0.44 eV (bulk) to 0.15 eV (monolayer) and significantly enhancing the static dielectric constant. This evolution triggers a fundamental scattering inversion: whereas bulk transport is limited by polar optical phonon scattering, the monolayer becomes dominated by acoustic deformation potential scattering. We show that this crossover originates from the simultaneous suppression of the Fröhlich interaction through enhanced dielectric screening and the amplification of acoustic scattering due to pronounced lattice softening. These results clarify how the interplay between dielectric screening, lattice stiffness, and band topology governs transport in low-dimensional magnetic semiconductors, providing a framework to optimize their electronic performance.

Iron oxide-based magnetic nanoparticles (MNPs) play a major role as contrast agents for magnetic resonance imaging (MRI). The introduction of ${\text{Zn}}^{2+}$ and ${\text{Mn}}^{2+}$ ions within their crystal lattice alters the magnetic anisotropy and the spin dynamical properties of the systems, which in turn influence the nuclear relaxation mechanisms that drive contrast enhancement in MRI. This work explored two sets of doped-magnetite MNPs (chemical formula ${\text{Zn}}_x{\text{Mn}}_y{\text{Fe}}_{3-x-y}{\mathrm{O}}_4$) with core diameters of $d\sim 4$ or 15 nm produced by the fast and reproducible microwave-assisted synthesis method, and using different solvents like water or polyols. The $d\sim 4\mathrm{nm}$ MNPs were doped with ${\text{Zn}}^{2+}$, from $x=0$ to 0.9; for ${\text{Zn}}^{2+}=0.5$, ${\text{Mn}}^{2+}$ was introduced at $y=0.2,0.4$. The $d\sim 15\mathrm{nm}$ MNPs were doped with ${\text{Zn}}^{2+}$ from $x=0$ to 0.4, with ${\text{Mn}}^{2+}$ added at $y=0.1$ for ${\text{Zn}}^{2+}=0.3$ and at 0.3 for ${\text{Zn}}^{2+}=0.4$. Both sets were dispersed in water with carboxymethyl-dextran or polyacrylic acid coatings to ensure their colloidal stability, even under high magnetic fields. We highlighted the influence of the size and of the ${\text{Zn}}^{2+}/{\text{Mn}}^{2+}$ doping on the static and dynamical magnetic properties, also through the ${}^1\mathrm{H}$-NMR-dispersion profile analysis over a wide frequency range and the application of the superparamagnetic relaxation heuristic model. Larger particles exhibited higher nuclear relaxometric efficiency and anisotropy compared to the smaller ones, consistently outperforming the commercial contrast agents. On the other hand, the magnetic response and the nuclear relaxivities are enhanced by a ${\text{Zn}}^{2+}$ content of $x=0.1$ ions for approximately $4\mathrm{nm}$ MNPs and of $x=0.4$ for approximately $15\mathrm{nm}$ MNPs, while a slight enhancing effect was obtained by ${\text{Mn}}^{2+}$ introduction. It is thus shown how a proper size and doping combination provide an attractive way to modify the static and dynamical magnetic properties of MNPs by acting on different spin-spin correlation times dominating in specific frequency intervals, thus enabling the fine-tuning of image contrast properties suitable for magnetic resonance imaging.

This study establishes a baseline for muonium (Mu) charge-exchange dynamics in n-type $4H\text{−}\mathrm{SiC}$ through a detailed low-energy muon spin rotation ($\mathrm{LE}\text{−}\mu \mathrm{SR}$) investigation. Epitaxially grown and ion-implanted samples with nitrogen and phosphorus donors were characterized to assess the effect of carrier concentration and doping method on defect formation. $\mathrm{LE}\text{−}\mu \mathrm{SR}$ enabled nanometer-scale depth profiling of near-surface and implanted regions, revealing variations in charge carrier concentration due to fixed surface charges. The temperature dependence of the diamagnetic fraction and phase provided direct evidence of the ${\mathrm{Mu}}^0$ to ${\mathrm{Mu}}^{-}$ transition, with extracted activation energies consistent with known donor ionization energies. Additionally, high-field $\mu \mathrm{SR}$ was used to analyze the Mu dynamics, and Monte Carlo simulations were used to model the ${\mathrm{Mu}}^0$ electron capture process. The simulation results offer a quantitative method to extract free electron concentrations from $\mathrm{LE}\text{−}\mu \mathrm{SR}$ data, enhancing its capability to characterize the activation of dopants and carrier depth profiles. We demonstrate that $\mathrm{LE}\text{−}\mu \mathrm{SR}$ is a powerful depth-resolved tool that can provide insights for optimizing the fabrication of reliable SiC devices for power electronics.

Self-sustained oscillators produce stable periodic motion robust to dissipation. Such motion is usually achieved by work fed back into the oscillator, but its performance is often limited by frequency-dependent operation. Here, we propose a scheme for self-sustained vibrations without external feedback. We consider a movable Cooper-pair box attached to the free end of a voltage-biased normal-metal pillar. The Cooper-pair box carries an Andreev current subject to an electric field applied perpendicular to the current. In the adiabatic limit, where the Cooper-pair box state follows its motion, vibrational instability occurs, pumped by inelastic Andreev tunneling. Nonlinearity of the Josephson coupling saturates the vibrational amplitude, resulting in two-dimensional self-vibrations. We discuss the advantage of this adiabatic scheme in comparison with feedback-induced self-oscillation.

Dispersive coupling between vibrational modes with different frequencies is a major nonlinear dynamical effect. We show, analytically and numerically, that in $p$-doped semiconductors such coupling can be strongly enhanced. Moreover, the relative coupling strength increases with the order of the nonlinearity. The doping-induced dispersive coupling becomes much stronger than the intrinsic one already for moderately strong doping. Its dependence on the hole density is nonmonotonic, and the temperature dependence becomes nonmonotonic for higher densities. Relevant mesoscopic frequency fluctuations are briefly discussed. The theory is applied to Si resonators, where doping is often used to compensate the temperature dependence of a clock mode, whereas another low-frequency mode is used as a thermometer to enable temperature stabilization.

Bound states in the continuum (BICs) enable elastic-wave resonances with ultralow radiation loss, but in multimode plates, localized modes readily couple to multiple radiation channels. Here we extend a symmetry-mismatch design recently implemented in one-dimensional Lamb-wave structures to a two-dimensional phononic crystal plate and realize elastic BICs and quasi-BICs by exploiting a mismatch between symmetric ($S$) and antisymmetric ($A$) Lamb modes. The plate is engineered such that the $A$ mode forms a passband while the $S$ mode exhibits a band gap; a central defect then hosts localized $S$ modes whose eigenfrequencies lie in this band gap yet overlap the $A$-mode continuum, forming $S$-mode BICs. Finite-element simulations and laser Doppler vibrometry show that the number of BIC modes can be controlled by varying the number of defect unit cells. By weakly breaking the midplane mirror symmetry through tilted hole sidewalls, we induce controlled $S\text{−}A$ hybridization and convert these strictly bound states into quasi-BICs that are efficiently excited by plane waves. Using this mechanism, we implement a planar multicavity structure where spatially separated quasi-BIC resonators are addressed at distinct frequencies, thereby realizing planar space-frequency multiplexing and providing a simple route to multifrequency elastic-wave signal processing and integrated phononic circuits.

Frequency-multiplexed entangled photon pair sources with narrow bandwidths and high pair generation efficiency are a key enabling technology for quantum networking. We present a simulation-based design study of an integrated photonic racetrack resonator source for spontaneous parametric down-conversion (SPDC) that simultaneously achieves all three properties. The central result is a simulated set of 90 doubly resonant signal and idler frequency-mode pairs with an effective frequency-space Schmidt number of 89.62, an average single-island purity of 0.960, an average bandwidth of 1.08 GHz, a mean free spectral range of 51.9 GHz, and a total internal pair-generation-rate efficiency of $1.16\mathrm{GHz}{\mathrm{mW}}^{-1}$. Under deterministic wavelength-based splitting, the folded 45-pair state has an effective frequency-space Schmidt number of 44.93. To support these predictions, we derive a closed-form analytical connection between classical cavity parameters (resonant frequencies, decay rates, and coupling coefficients) and the quantum joint spectral amplitude and pair generation rate, extending the dispersive-medium quantization formalism of Raymer to the nonlinear optical cavity case. We demonstrate how classical electromagnetic field simulations can be combined with this analytical framework to predict quantum figures of merit for an integrated photonic source prior to fabrication. Fabrication and experimental validation are left for future work.