广东以色列理工学院

This study proposes an AI-assisted wave-based inverse reconstruction framework for characterizing spatially varying fiber orientation fields in thin-walled short-fiber-reinforced composite plates. The problem is formulated as an inverse mapping from time-domain signals collected by multiple sensors to a discretized fiber orientation field, enabling direct prediction without iterative inversion. To generate training data efficiently, several homogenized finite element models are developed and compared with a detailed finite element model that explicitly resolves the microstructure. The results show that all homogenized models provide substantial reductions in computational cost, while their accuracy differs. Among them, the unit-cell-based reinforcement-part averaging model achieves the most favorable balance between accuracy, efficiency, and robustness, and is therefore selected for large-scale forward simulations. Based on the generated dataset, a multilayer perceptron is trained to predict the fiber orientation field from boundary wave signals. A double-angle representation is adopted to account for the periodic nature of orientation. The results demonstrate that the model can accurately reconstruct both random and structured orientation distributions, with stable training behavior and good generalization performance.

Reversible photoswitches that operate directly inside the DNA π-stack itself have remained elusive. Here, we introduce one or more azobenzene moieties into a DNA duplex while leaving the opposite position baseless. Reversible trans-cis photoisomerization strongly modulates longitudinal π-conjugation: the planar trans isomer partially preserves π-overlap across the defect, whereas the nonplanar cis isomer disrupts it dramatically. Both isomers substantially lower junction conductance compared to pristine DNA, but the cis-rich state induces a far more severe reduction. Conductive-probe atomic force microscopy reveals robustly reversible photoswitching: the cis-rich state lowers current by up to 2 orders of magnitude relative to the trans-rich state. On/off ratios increase superlinearly with the number of azobenzene units (up to ∼ 150 for three units) and remain >50 even at bias voltages of 1.5 Vexceptionally stable for a molecular photoswitch. Ultraviolet photoelectron spectroscopy confirms that the HOMO level of the DNA π-system is virtually unaffected both by azobenzene incorporation and by trans-cis isomerization. The observed switching therefore originates from a stronger suppressed rate of charge transfer across the cis defect, fully consistent with enhanced backscattering and possible destructive quantum interference due to its higher asymmetry location.

In this paper, we study the characterization of inner uniformity of bounded domains G in ℝn, and prove that the following three conditions are equivalent: (1) G is inner uniform; (2) G is Gromov hyperbolic and its inner metric boundary is naturally quasisymmetrically equivalent to the Gromov boundary; (3) G is Gromov hyperbolic and linearly locally connected with respect to the inner metric. The equivalence between the conditions (1) and (2), and the implication from (2) to (3) affirmatively answer three questions raised by Bonk, Heinonen, and Koskela in 2001.

Carbon molecular sieve (CMS) membranes have attracted significant research attention owing to their exceptional gas transport properties, attributed to their tailorable pore structures that facilitate precise molecular discrimination based on subtle differences in molecular size and shape. Thereby, a novel hybrid CMS (HCMS) membrane was prepared via incorporating ionic liquid (IL) 1-butyl-3-methylimidazolium trifluoromethanesulfonate ([Bmim][OTF]) within a polyimide (PI) matrix, followed by low-temperature pyrolysis. The addition of ILs enabled precise modulation of the pore structures of CMS membrane, resulting in concurrent enhancement of both CO2 permeability and CO2/CH4 selectivity. Specifically, compared to the pristine PI-550 CMS membrane, the optimized PI-3%[Bmim][OTF]-550 HCMS membrane demonstrated a 2.8-fold improvement in CO2/CH4 selectivity (from 20.5 to 57.6) alongside a remarkable increase in CO2 permeability from 2871.9 to 4329.8 Barrer, surpassed the 2019 Robeson upper bound for gas separation membranes.

The gelatinization behavior of starch is critically influenced by its amylose content (AC), yet the underlying mechanisms at the lamellar level remains incompletely understood, especially for starches with high AC (>50 %). In this study, we investigated the real-time structural evolution of maize starches with a wide range of AC (3.2 %-79.6 %) during gelatinization using in situ small-angle X-ray scattering (SAXS). Upon heating from 40 degrees C to 90 degrees C: (1) the lamellar thickening occurred initially but was followed by rapid disruption at low AC (3.2 % and 26.3 %); (2) the lamellar structure remained stable with minimal changes in lamellar peak intensity and lamellar thickness at high AC (58.6 %), (3) the electron density contrast between crystalline and amorphous phases increased, and both lamellar types exhibited thickening at very high AC (71.2 % and 79.6 %). Furthermore, the correlation length (xi) decreased with increasing AC up to 60 %, suggesting a more compact viscoelastic structure due to greater chain release and entanglement. Our data provide evidence that maize starch with about 60 % amylose has especially stable crystalline lamellae. This stability likely explains why starches with moderately high amylose content tend to gelatinize at higher temperatures.