Browse Articles

DescriptorMedSAM: language-image fusion with multi-aspect text guidance for medical image segmentation

Pressure-dependent photoluminescence study of band structure in germanium

Germanium (Ge) has long been regarded as a promising laser material for silicon photonics due to its quasi-direct bandgap to make up for the deficiency of indirect bandgap silicon, but its energy band structure under pressure remains puzzling. Here, we study the pressure-dependent photoluminescence (PL) of Ge compressed in a diamond anvil cell to reveal its energy band structure up to 3.89 GPa. Unlike the earlier reported results studied by absorption, the effect of high pressure on bandgaps can be studied for the same sample and the determinations of bandgap positions are not influenced by the sample thickness and the interference pattern. The PL peak related to X–Γ bandgap transition was observed with pressure coefficient of −12.4 ± 2.1 meV/GPa. We unambiguously show that the L- and Γ-valleys move upward while the X–valley moves downward in energy with increasing pressure, with a Γ–X crossover observed at an onset pressure of 0.74 GPa, and then a L–X crossover takes place near 2.85 GPa. These findings provide experimental evidence for identification of the band structure of Ge, deepening the understanding of pressure-induced bandgap modification and conduction band valley crossover.

Premature pollen development hinders autonomous self-pollination and promotes insect pollination in soybean (Glycine max L.)

Quantum structure for efficient p-doping of AlGaN with Al content over 70%

The low p-doping efficiency of AlGaN due to its high activation energy limits the advancement of deep ultraviolet optoelectronics. The quantum engineering method has provided a new insight to address the challenge of acceptor activation in Al-rich AlGaN. As its unique valence band offset, the distribution of acceptor and band edge states in the barrier and well determines the acceptor activation efficiency. Here, we proposed a model for reducing the acceptor activation energy in quantum-structured p-AlGaN and provided experimental validation. By tuning the reactor pressure, we prepared periodic quantum-structured p-AlGaN with different cycles successfully. Compared to long-period quantum structures, short-period quantum structures introduce more band offsets, enhancing the wavefunction overlap between the acceptor states and the band edge states, thereby significantly promoting acceptor activation. Ultimately, we achieved an ultrathin quantum-structure p-AlGaN with an average Al content of ∼77% and a periodic thickness of 2.7 nm, reaching a record-low resistivity value of about 5.4 Ω·cm. Additionally, it is demonstrated that short-period quantum structures can facilitate hole injection in deep ultraviolet light-emitting diodes, effectively improving their external quantum efficiency. This study offers an effective p-doping strategy for Al-rich AlGaN and paves the way for applications of deep ultraviolet optoelectronics.

A novel variant p.Y250C of FZD4 influences Norrine/β-catenin signaling pathway that associates with familial exudative vitreoretinopathy (FEVR)

Negative in-plane Poisson's ratio in [001]-textured PMN-PZT ceramics

[001]-textured 0.4P(Mg1/3Nb2/3)O3-0.25PbZrO3-0.35PbTiO3 (PMN-PZT) ceramics were fabricated by templated grain growth using 3 vol. % BaTiO3. Full matrices of dielectric (εij), elastic (sij, cij), and piezoelectric (dij) parameters were obtained by the resonance–antiresonance method. The ceramics exhibit a quasi-static piezoelectric coefficient d33 of 1490 pC/N and an electromechanical coupling factor k33 of 0.92, approaching the performance levels of piezoelectric single crystals. Notably, the textured PMN-PZT demonstrates an in-plane negative Poisson's ratio of ν12 = −0.13, representing a strong auxetic behavior that has been experimentally confirmed through direct strain–stress measurement. This distinctive characteristic is further corroborated by both laser scanning vibrometer and finite element analysis. Theoretical interpretation suggests that the negative Poisson's ratio stems primarily from stress-induced polarization rotation along the ⟨110⟩ direction. In [001]-textured ceramics, the transverse direction inherently contains this orientation, typically resulting in a low or negative Poisson's ratio. This abnormal Poisson's ratio property may affect device design approaches for transducers, sensors, and energy harvesting applications.

A validation of 3D imaging for non-invasive, tech-assisted diagnosis of caries and erosive tooth wear in primary teeth – an in vitro study

Red micro-LEDs on glass substrates for ultrahigh-luminance (>105 nits) transparent displays

AlGaInP-based red micro-light-emitting diodes (micro-LEDs) with lateral dimensions of 28 × 52 µm2 were fabricated on transparent glass substrates via solder bonding for high-performance transparent displays. The fabricated devices exhibited excellent spectral stability, featuring a record-low wavelength redshift coefficient of less than 0.126 nm/K. Notably, the glass-substrate-integrated micro-LED arrays achieved an optical transmittance of 52.6% alongside a record-high luminance of 1.5 × 105 nits, underscoring their great potential for practical transparent display applications.

DFT-based exploration of XMnCrZ (X = Ni, Ti; Z = Sn, Sb) quaternary heusler alloys for structural and multifunctional properties

Spin wave interference-based efficient neuromorphic computing

We demonstrate the design of a neuromorphic hardware, spin wave interference device (SWID), utilizing micromagnetic simulations, for performing feature extraction and classification of binary digit patterns. The SWID aims to reduce the weight computations in artificial neural network (ANN) implementations allowing for low power computing and faster inference. We achieve the direct classification of multibit binary input pulse schemes through synaptic behavior and interference of spin waves. We showcase the versatility of SWID's information processing capabilities across two-bit ranges, 4-bit and 6-bit binary digit data, by effectively controlling the nonlinearity and interference of spin waves with external input current pulses. The performance of the SWID with 4-bit and 6-bit digit pattern classification ability is tested for image recognition tasks with the Modified National Institute of Standards and Technology handwritten image database in a feed forward neural network. Though achieving 84.7% accuracy in image recognition, this SWID-based network reduces the weight computation by 99.4% as compared to the software-ANN, showcasing its capability for faster decision making. This huge reduction in computations offers great benefits to ANN applications in edge devices and memory constraint devices. These results underscore the potential of spin wave-based SWID in designing power efficient neuromorphic hardware.

Reliable and efficient solar radiation estimation with the insights of XAI

Coherent transport in strongly correlated perovskite-manganite quantum wells

Perovskite transition metal oxides (TMOs) are hallmark systems for studying electron correlations, with strong Coulomb interactions reaching the electron volt scale. Such interactions generally hinder coherent charge transport, limiting its observation to only moderately correlated TMOs. Among TMOs with strong electron correlations, the ferromagnetic perovskite manganite La1−xSrxMnO3 (LSMO) has attracted significant attention for spintronics applications due to its half-metallic nature and robust ferromagnetism, with a Curie temperature above room temperature. In this Letter, we report the emergence of oscillatory conduction in tunnel diodes incorporating an epitaxial thin LSMO layer—a phenomenon not previously observed in strongly correlated oxides. The observed oscillations originate from discrete quantum-well states formed via quantum confinement, indicating coherent transport across the LSMO layer. These quantum-well states are quantitatively explained using a tight-binding model tailored for the electronic structure of LSMO. Our findings demonstrate that high-quality epitaxial perovskite manganites can sustain coherent transport, even in the presence of strong electron correlations, offering avenues for oxide-based quantum and spintronics devices.

Improved cryopreservation of cardiomyocyte aggregates differentiated from GMP iPSC in a 3D culture format

Abstract Human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) are in the focus of clinical research for cell-based therapies for heart failure patients. However, challenges include long-term storage of cardiomyocyte aggregates (CMAs). Current freeze/thawing protocols have been established for single cells, indicating a substantial gap for process development and optimization for cryopreservation. Here CMA differentiation was achieved by the combination of WNT-pathway modulation and TGF-β/SMAD- and FGF-pathway targeting. An efficient CMA differentiation with cardiac marker expression of ≥ 95% ACTN2/TNNT2 was established. For process optimization, different freezing media were combined with a pre-treatment strategy and comprehensive assessment of cell recovery, viability, physiology, and purity was performed before and after cryopreservation. Development and optimization of cryopreservation strategies for CMAs led to a freezing medium termed “10% human serum albumin (HSA)” representing the best option tested. Analysis at 5-days post thawing revealed maintenance of high cardiac marker expression (> 90%), spontaneous contraction activity, and overall recovery of > 80% vital cell counts compared to sample analysis before the freezing procedure. Extensive modification and optimization of established single cell CM cryopreservation protocols was achieved with successful recovery of CMs within the complex structure of aggregates.

Twist angle control of charge density wave transitions in 1T–TaS2 homostructures

Twisted homostructures provide a versatile platform to engineer interlayer coupling and collective orders via moiré superlattices. While the effectiveness of the moiré potential in modulating charge density waves (CDWs) has been established through spectroscopic techniques, systematic investigations through electrical transport measurements remain scarce. Here, by leveraging interfacial moiré superlattices, we characterize the phase transition between the nearly commensurate CDW state and the incommensurate CDW state in twisted 1T-TaS2 homostructures. A two-step transition is observed in twisted 1T-TaS2 homostructures, which is in marked contrast to the single and smooth transition of pristine sample. The two–step transition behavior persists over twist angles from 0° to 58° and for excitation currents spanning two orders of magnitude. These features are consistent with a moiré potential induced periodic pinning landscape for nearly commensurate CDW domain walls at the twisted interface. Our results establish twist angle as an effective control knob for engineering CDWs in layered materials and open routes to moiré superlattice based device concepts.

Lithium-ion battery waste as a robust oxygen evolution reaction electrocatalyst for seawater splitting

Abstract Electrocatalytic seawater splitting seems to be the most promising and urgent demand strategy for clean hydrogen energy production. Utilizing low-cost electrocatalysts is pivotal in the hydrogen economy, as seawater splitting can be made highly efficient and more economical. To meet these expectations, we proposed a novel utilization for the black carbon mass left over from hydrometallurgical metal recovery as an efficient and stable electrocatalyst for oxygen evolution reaction (OER) performed in alkaline media. The SEM-EDS, XPS, XRD, XRF, and Raman analyses revealed that the composition and structure of the post-leached battery powders depend on the hydrometallurgical waste recycling conditions, which in turn affect their OER electrocatalytic activity. In particular, the material leached with sulfuric acid (BAT 1) retained a higher content of cobalt-based compounds (mainly LiCoO 2 and Co 3 O 4 ) embedded within a porous carbon matrix and, resulting in the best catalytic performance among all samples. The enhanced performance of BAT 1 is attributed to the synergy of its cobalt-based phases and the well-developed porous carbon structure, which collectively result in a high electrochemically active surface area. The electrochemical tests proved that Li-ion battery waste has remarkable OER catalytic performance with an overpotential of 226 mV and 225 mV, reaching 10 mA cm − 2 in water splitting and in seawater splitting, respectively, which is only 14 mV and 95 mV higher than for benchmark RuO 2 in water splitting and seawater splitting, respectively.

Temperature-dependent wake-up phenomena in AlScN ferrodiode memory devices

Ferroelectric aluminum scandium nitride (AlScN) is a promising material for use in nonvolatile digital memory operating at temperatures well above the limits of current commercial technology. Ferrodiodes consisting of thin films of AlScN sandwiched between metal contacts exhibit polarization-dependent electrical conduction that can distinguish the on/off memory state with bias voltages below 10 V. However, the reliability and repeatability of key device parameters such as switching voltage and on/off ratio can change significantly with temperature. Understanding the temperature dependence of the material parameters that govern these phenomena is critical to the development of practical memory devices operating reliably at high temperature. We have systematically studied the changes in wake-up-like behavior in 40 nm AlScN films from room temperature to 700 °C. Above 300 °C, an anomalous decrease in device current arises when the applied voltage exceeds the minimum switching voltage. The temperature and rate dependence of the anomalous current loss suggests that thermally activated changes to the interlayer near the top electrode alter the local charged defect compensation. This causes the leakage current to decrease even while the net remnant polarization in the films increases.

In silico study of a bilayer titanium dental implant with a porous titanium and hydroxyapatite composite outer layer for enhanced osseointegration

Abstract This study aims to develop an innovative bilayered dental implant design featuring a titanium alloy core with a porous composite titanium (Ti) and hydroxyapatite (HA) outer layer to enhance implant stability and patient outcomes. Using SolidWorks 2017, 3D models of the implants and a mandibular bone segment were created. A Finite Element (FE) analysis was then conducted with ANSYS Workbench to assess the mechanical behavior under a 250 N axial compressive load, comparing the bilayered implant to a conventional titanium implant. Variables like porosity (ranging from 10 to 90% in 10% increments), HA content (ranging from 10 to 50% in 5% intervals), and outer layer thickness (2 mm, 1.5 mm and 1 mm) were systematically analyzed. Each configuration was evaluated based on von Mises stress distribution and interfacial strain in peri-implant bone. Results indicated that all porous designs of bilayered implants had significantly lower von Mises stress than traditional implant, with reductions ranging from approximately 69 to 94%, depending on HA/Ti composition and shell thickness. The non-porous bilayer configurations also showed clear stress reductions, with decreases from approximately 72 to 90%, depending on the HA/Ti composition and shell thickness. However, these reductions were slightly lower than those observed in porous designs, with maximal reductions occurring in the porous core of some 2 mm bilayered implant configurations. The combined evaluation of strain and von Mises stress analyses identified the 2 mm core diameter with a 2 mm porous shell as the optimal design, providing favorable microstrain, improved load transfer, and reduced stress concentrations. This modification promotes a more favorable mechanical interaction between the implant and surrounding bone. These findings underscore the potential of bilayered porous implants to improve stability and bone integration, marking a significant step forward in dental implant technology. Further research, including experimental validation, is encouraged to verify these results and investigate other loading conditions, promoting the development of more effective and sustainable dental implant solutions.

Pressure-induced polar–nonpolar–polar phase conversion in Na0.5Bi0.5TiO3

Na0.5Bi0.5TiO3 (NBT) is a promising lead-free ferroelectric material and has a unique A-site substitution structure with rhombohedral (R3c) symmetry at ambient conditions. Combining in situ polarization–electric field (P–E) hysteresis loops, Raman spectroscopy, x-ray diffraction measurements, and ab initio calculations, here we report a pressure-induced polar–nonpolar–polar phase crossover in NBT with a two-stage structural transition from R3c to P21/m and then to Pmn21 symmetry. Electrical resistance and absorption spectroscopy measurements demonstrate the abnormal changes around 5 and 13 GPa, corresponding to the two-stage phase transitions. Detailed charge density difference analyses reveal that the structural transition causes a synergistic change of Ti cation off-center displacement and octahedron tilting angle, leading to the ferroelectric–paraelectric–ferroelectric transition under pressure. These results establish a phase transition sequence of NBT with intriguing ferroelectric and electronic evolution, which helps to resolve the controversial structural conversion process in NBT and expand our understanding of lead-free ferroelectric under extreme conditions.

Combustion performance prediction in oil and gas plants using integrated neural network models and SAP S4HANA sensor analytics

Atomistic nonequilibrium Green's function study of radiation-induced charge loss in floating gate flash memories

This study presents an atomistic nonequilibrium Green's function (NEGF) approach for modeling radiation-induced charge loss in floating-gate flash memories, which become increasingly vulnerable to radiation effects as device features shrink. Nonradiative charge-carrier recombination at localized deep-level defect centers is treated by coupling the defect Green's function to delocalized interface states derived from a two-probe tight-binding description, via multiphonon-scattering self-energies. Using oxygen vacancies as representative deep-level traps, the trap-assisted tunneling current under retention conditions is calculated by integrating the NEGF formalism within a drift-diffusion solver. This NEGF framework significantly improves agreement with experimental data from heavy-ion irradiation, and provides insights beyond semiclassical models. These findings underscore the necessity of atomistic, fully quantum treatments for reliable assessment and design of radiation-hardened nonvolatile memory technologies.