Search

In this work, flexible Cu2−xS films on nylon membranes are prepared by combining a simple hydrothermal synthesis and vacuum filtration followed by hot pressing. The films consist of Cu2S and Cu1.96S two phases with grain sizes from nano to submicron. Doping Se on the S site not only increases the Cu1.96S content in the Cu2−xS to increase carrier concentration but also modifies electronic structure, thereby greatly improves the electrical properties of the Cu2−xS. Specifically, an optimal composite film with a nominal composition of Cu2−xS0.98Se0.02 exhibits a high power factor of ~150.1 μW m−1 K−2 at 300 K, which increases by ~138% compared to that of the pristine Cu2−xS film. Meanwhile, the composite film shows outstanding flexibility (~97.2% of the original electrical conductivity is maintained after 1500 bending cycles with a bending radius of 4 mm). A four-leg flexible thermoelectric (TE) generator assembled with the optimal film generates a maximum power of 329.6 nW (corresponding power density of 1.70 W m−2) at a temperature difference of 31.1 K. This work provides a simple route to the preparation of high TE performance Cu2−xS-based films.

Cumin essential oil chitosan nanocapsules (CENPs) were prepared through the ionic gelation method by blending chitosan (CS) with cumin essential oil (CEO) in different proportions (1:0.8, 1:1, 1:2, 1:3, 1:4). Subsequently, these nanocapsules were characterized and evaluated for their antibacterial properties to determine the optimal cumin essential oil encapsulation and antibacterial efficacy. The outcomes demonstrated that the highest encapsulation efficiency of CENPs was 52%, achieved with a 1:3 CS/CEO ratio. At this point, the nanoparticles had the smallest particle size (584.67 nm) and a regular spherical distribution in the emulsion. Moreover, the CENPs could release the encapsulated CEOs slowly, leading to efficient inhibition of E. coli and L. monocytogenes over a relatively extended period (24–36 h) compared to the CS and CEO. This research offers a promising approach for the use of nanocapsules in food preservation.

This study aimed to investigate the release of volatile compounds in mutton shashliks (named as FxLy, x-fat cubes: 0-4; y-lean cubes: 4-0) with different fat–lean ratios before and during consumption, respectively. In total, 67 volatile compounds were identified in shashliks using gas chromatography/mass spectrometry. Aldehyde, alcohol, and ketone were the major volatile substances, accounting for more than 75% of the total volatile compounds. There were significant differences in the volatile compounds of mutton shashliks with different fat–lean ratios. With the increase of the fat content, the types and content of volatile substances released also increase. However, when the percentage of fat exceeded 50%, the number of furans and pyrazine, which were characteristic of the volatile compounds of roasted meat, was decreased. The release of volatiles during the consumption of mutton shashliks was measured using the exhaled breath test and the results showed that adding an appropriate amount of fat (2:2) shorten the mastication duration and weaken the breakdown of bolus particles in the consumption process, which is not conducive to the release potential of volatile substances. Therefore, setting the fat to lean ratio to 2:2 is the best choice for making mutton shashliks, as it (F2L2) can provide rich flavor substances for mutton shashliks before and during consumption.

Significant progress has been made in generating counterfeit images and videos. Forged videos generated by deepfaking have been widely spread and have caused severe societal impacts, which stir up public concern about automatic deepfake detection technology. Recently, many deepfake detection methods based on forged features have been proposed. Among the popular forged features, textural features are widely used. However, most of the current texture-based detection methods extract textures directly from RGB images, ignoring the mature spectral analysis methods. Therefore, this research proposes a deepfake detection network fusing RGB features and textural information extracted by neural networks and signal processing methods, namely, MFF-Net. Specifically, it consists of four key components: (1) a feature extraction module to further extract textural and frequency information using the Gabor convolution and residual attention blocks; (2) a texture enhancement module to zoom into the subtle textural features in shallow layers; (3) an attention module to force the classifier to focus on the forged part; (4) two instances of feature fusion to firstly fuse textural features from the shallow RGB branch and feature extraction module and then to fuse the textural features and semantic information. Moreover, we further introduce a new diversity loss to force the feature extraction module to learn features of different scales and directions. The experimental results show that MFF-Net has excellent generalization and has achieved state-of-the-art performance on various deepfake datasets.