A tunable metasurface broadband absorber based on single-layer graphene is proposed for the terahertz (THz) frequency band, which consists of a gold bottom layer (0.5 mu m), a silica dielectric middle layer (12 mu m) and a graphene top layer with symmetric rectangular patches. The graphene patches are designed with proportional and symmetrical spacing from periodic edges to realize efficient tunable broadband absorption in the THz band. By virtue of the surface plasmon resonance effect of graphene and the synergy of electromagnetic coupling and precise equivalent impedance matching of structural units, the absorber achieves an absorption rate of over 90% in the 2.48-5.92 THz frequency range with an effective bandwidth of 3.44 THz and an average absorption rate of 94.2%, and perfect absorption (99.99% and 99.44%) at 2.88 THz and 5.04 THz respectively. The intrinsic physical mechanism of broadband absorption is systematically revealed by quantitative analysis of electric field distribution: the superposition of plasmon resonance in different structural regions and the strong electric field coupling effect between regions lead to the overlap of absorption peaks, thus expanding the effective absorption bandwidth. Dynamic tuning of the absorption frequency band is realized by regulating the graphene Fermi level via gate voltage, which changes the surface plasmon resonance frequency and structural equivalent impedance. This absorber exhibits excellent process fault tolerance and low angle tolerance, meaning it maintains stable absorption even at an incident angle ranging from 0 degrees to 30 degrees. These superior performances endow the proposed absorber with important practical application potential in THz optoelectronic devices, high-speed wireless communication, substance detection and identification, and electromagnetic stealth.
In order to explore the influence of different strains on the flavor characteristics of fermented sufu, this study focuses on Mucor racemosus (MR), Rhizopus oryzae (RO), and Aspergillus candidus (AC). The metabolic products and flavor profile differences were systematically analyzed using gas chromatography-mass spectrometry (GC-MS), electronic nose detection, total acidity measurement, and sensory evaluation. The results indicated that the MR group achieved the highest sensory score owing to its elevated ester and substantial alcohol contents, which imparted a distinctive floral-fruity aroma with balanced flavor. In contrast, the RO group exhibited dominant acids, resulting in a strong sour taste, weaker esterification, and a loose texture. The AC group showed high aldehyde/ketone proportions along with ammonia-related metabolites and the highest total acidity, collectively generating pronounced off-flavors and the lowest sensory score. Principal component analysis of electronic nose data confirmed that divergent metabolic pathways among the strains constituted the core factor for flavor differentiation. Variable correlation analysis further quantified the regulatory mechanisms: esters and alcohols exhibited significant positive correlations with sensory scores, whereas acids and aldehydes/ketones showed strong negative correlations with off-flavor intensity. In conclusion, MR, due to its high production of esters and alcohols, is more suitable for traditional sufu flavor; RO and AC require improvement through strain combinations or process optimization to address their sourness and off-flavor defects. This study provides theoretical guidance for precise strain selection and flavor-directed regulation in industrial sufu production.
With the increasing demands for high-temperature resistance and superior electromagnetic wave absorption in complex environments, developing effective low-frequency (2-8 GHz) absorbers remains challenging, especially for magnetic composites with high magnetic content and dispersibility. This study proposes a hummingbird acoustic resonance mixing (HAM) strategy to fabricate FeSi@SiC composites. Compositing soft magnetic FeSi with high-surface-area nano-SiC enhances electromagnetic loss and improves the high-temperature stability of the composite material due to the partial protection of FeSi by SiC. The optimized composite achieves a minimum reflection loss of-35.9 dB and a qualified absorption efficiency of 3.9 GHz at only 3 mm thickness, along with maintained structural integrity at 800 degrees C. Enhanced low-frequency absorption is revealed to stem from a large-small size magnetic-dielectric structure, a high density of confined magnetic units and strong magnetic coupling, which collectively promote magnetic loss. This work introduces a new approach to designing low-frequency electromagnetic wave absorbers with high-temperature oxidation resistance by precisely tuning their dielectric and magnetic constituents, and elucidates the underlying low-frequency magnetic attenuation mechanism along with its potential applications.
As a cornerstone of national digital transformation strategies, digital infrastructure offers a promising pathway to address urban–rural dualism, narrow income disparities, and advance common prosperity. Using provincial panel data from 30 Chinese regions over 2011–2022, this study empirically examines how digital infrastructure affects the urban–rural income gap. Our results show that digital infrastructure significantly reduces this gap. Mediation analysis identifies three key transmission channels: expansion of non-agricultural employment, enhancement of technological innovation, and upgrading of industrial structure. The gap-reducing effect is notably stronger in central China and in provinces with higher levels of economic development. Importantly, the application layer of digital infrastructure demonstrates substantially greater effectiveness than the physical construction layer. These findings provide actionable insights for policymakers seeking to optimize digital infrastructure investments to promote inclusive growth and reduce spatial gaps.
This paper investigates sampled-data synchronization control of chaotic neural networks subject to actuator saturation. System states are sampled based on a monotonically-increasing-function-based event-triggered sampling scheme. The key feature of this scheme is that an event is triggered only when a prescribed threshold is exceeded by the accumulated square of errors rather than the instantaneous error. With this feature, Zeno phenomenon does not occur, which is proven theoretically. By constructing an event-triggering-function-dependent Lyapunov functional, a sufficient criterion is formulated to ensure asymptotic synchronization between the master and slave systems. Building upon this criterion, an LMI-based algorithm is presented to jointly design both control gains and event-triggered parameters. Finally, a well-studied numerical example is taken to further demonstrated the effectiveness of the proposed results.
