Explore the vast range of titles available.

HighlightsA hybridized mechanical and solar energy-driven hydrogen production system was developed.A rotatory disc-shaped triboelectric nanogenerator (RD-TENG) enables to harvest mechanical energy from water flow and functions as a sufficient external power source.WO3/BiVO4 heterojunction is fabricated as photoanodes in the self-powered photoelectrochemical (PEC) cell, and the hydrogen production rate reaches to 7.27 μL min−1 under sunlight illumination with the energy conversion efficiency of 2.59%.Photoelectrochemical hydrogen generation is a promising approach to address the environmental pollution and energy crisis. In this work, we present a hybridized mechanical and solar energy-driven self-powered hydrogen production system. A rotatory disc-shaped triboelectric nanogenerator was employed to harvest mechanical energy from water and functions as a sufficient external power source. WO3/BiVO4 heterojunction photoanode was synthesized in a PEC water-splitting cell to produce H2. After transformation and rectification, the peak current reaches 0.1 mA at the rotation speed of 60 rpm. In this case, the H2 evolution process only occurs with sunlight irradiation. When the rotation speed is over 130 rpm, the peak photocurrent and peak dark current have nearly equal value. Direct electrolysis of water is almost simultaneous with photoelectrocatalysis of water. It is worth noting that the hydrogen production rate increases to 5.45 and 7.27 μL min−1 without or with light illumination at 160 rpm. The corresponding energy conversion efficiency is calculated to be 2.43% and 2.59%, respectively. All the results demonstrate such a self-powered system can successfully achieve the PEC hydrogen generation, exhibiting promising possibility of energy conversion.

Setting the number (or density) of pre-cracks as a variable, the effects of randomly-distributed pre-cracks on the mechanical properties and the corresponding failure process of sandstone are studied using Particle Flow Code software (PFC). The result shows that, as the number and total length of the pre-cracks increase, Unconfined Compressive Strength (UCS) of the sandstone specimen approximately shows a negative exponential variation, while the elastic modulus decreased near-linearly. With the increase of pre-crack number, both the UCS and the new crack number corresponding to the failure of the sandstone are gradually reduced; in addition, the angle of new cracks, as the data approximately obeyed the normal distribution, are mainly concentrated at about 90°. Partition damage phenomenon is more obvious for the specimen with less pre-cracks. And there are stress shielding region distributed above and below the near-horizontal pre-cracks, and the concentrated stress is more likely to exist at whose tips. The strength theory of rock damage is established based on the Mori–Tanaka method in this paper, by introducing a density coefficient η for the non-uniform arrangement of pre-cracks, the evolution strength of cracked sandstone can be well-described.

To address the challenges of weak model generalization and limited model capacity adaptation in traditional malware detection methods, this article presents a novel malware detection approach based on stacked depthwise separable convolutions and self-attention, termed CoAtNet. This method combines the strengths of the self-attention module’s robust model adaptation and the convolutional networks’ powerful generalization abilities. The initial step involves transforming the malicious code into grayscale images. These images are subsequently processed using a detection model that employs stacked depthwise separable convolutions and an attention mechanism. This model effectively recognizes and classifies the images, automatically extracting essential features from malicious software images. The effectiveness of the method was validated through comparative experiments using both the Malimg dataset and the augmented Blended+ dataset. The approach’s performance was evaluated against popular models, including XceptionNet, EfficientNetB0, ResNet50, VGG16, DenseNet169, and InceptionResNetV2. The experimental results highlight that the model surpasses other malware detection models in terms of accuracy and generalization ability. In conclusion, the proposed method addresses the limitations of traditional malware detection approaches by leveraging stacked depthwise separable convolutions and self-attention. Comprehensive experiments demonstrate its superior performance compared to existing models. This research contributes to advancing the field of malware detection and provides a promising solution for enhanced accuracy and robustness.

Highlights The heteroatom atomically doping strategy was reported to construct highly efficient sites on metal oxides for the detection of low-concentration gas. The atomically dispersed Sn atoms were intentionally incorporated into the Fe 2 O 3 lattice during the oxidative annealing of Fe-based metal organic framework, leading to specific Sn–O–Fe sites, porous structures, and abundant oxygen vacancies. The optimized Sn-Fe 2 O 3 exhibited exceptional sensing performance for NO 2 detection: ultra-high sensitivity ( Rg / Ra =2646.6 to 1 ppm NO 2 ), ultra-low limit of detection (10 ppb), and high selectivity. Conventional gas sensing materials (e.g., metal oxides) suffer from deficient sensitivity and serve cross-sensitivity issues due to the lack of efficient adsorption sites. Herein, the heteroatom atomically doping strategy is demonstrated to significantly enhance the sensing performance of metal oxides-based gas sensing materials. Specifically, the Sn atoms were incorporated into porous Fe 2 O 3 in the form of atomically dispersed sites. As revealed by X-ray absorption spectroscopy and atomic-resolution scanning transmission electron microscopy, these Sn atoms successfully occupy the Fe sites in the Fe 2 O 3 lattice, forming the unique Sn–O–Fe sites. Compared to Fe–O–Fe sites (from bare Fe 2 O 3 ) and Sn–O–Sn sites (from SnO 2 /Fe 2 O 3 with high Sn loading), the Sn–O–Fe sites on porous Fe 2 O 3 exhibit a superior sensitivity ( R g / R a  = 2646.6) to 1 ppm NO 2 , along with dramatically increased selectivity and ultra-low limits of detection (10 ppb). Further theoretical calculations suggest that the strong adsorption of NO 2 on Sn–O–Fe sites (N atom on Sn site, O atom on Fe site) contributes a more efficient gas response, compared to NO 2 on Fe–O–Fe sites and other gases on Sn–O–Fe sites. Moreover, the incorporated Sn atoms reduce the bandgap of Fe 2 O 3 , not only facilitating the electron release but also increasing the NO 2 adsorption at a low working temperature (150 °C). This work introduces an effective strategy to construct effective adsorption sites that show a unique response to specific gas molecules, potentially promoting the rational design of atomically modified gas sensing materials with high sensitivity and high selectivity.

Indium‐based materials can selectively reduce CO2 to formate, but their activities still fall short of expectations to be considered for practical applications. Structural engineering at the nanoscale offers a promising solution. However, it is challenging to directly prepare nanostructures of metallic indium because of its low melting point and high oxophilicity. Herein, a strategy to prepare highly dispersed indium oxide nanoparticles as the precatalyst supported on conductive carbon nanorods from annealing the MIL‐68 (In) precursor is proposed. When assessed in an H‐cell, the product enables CO2 reduction to formate with great faradaic efficiency of around 90% over a wide potential window in 0.5 m KHCO3. When applied in a gas‐diffusion‐electrode‐based flow cell, the catalyst delivers large current density of up to 300 mA cm−2 in 1 m KOH, great formate faradaic efficiency and decent stability. These results indicate the commercial viability of the catalyst even though the carbonate buildup at the gas diffusion electrode remains an issue of future research. Highly dispersed In2O3 nanoparticles supported on conductive carbon nanorods are prepared from annealing the MIL‐68 (In) precursor, and investigated as the precatalyst for electrochemical CO2 reduction to formate. The product exhibits great formate faradaic efficiency of around 90% over a wide potential window, large current density of up to 300 mA cm−2 and decent stability.

Potassium ion hybrid capacitors (PIHCs) have attracted considerable interest due to their low cost, competitive power/energy densities, and ultra‐long lifespan. However, the more sluggish insertion kinetics of battery‐type anodes than capacitor‐type cathodes in PIHCs seriously limits their practical application. Therefore, developing advanced anodes with high capacitor and suitable K+ intercalation is imperative and significant. A novel core–shell structure of NiCo oxide/NiCo oxyphosphide (NCOP) nanowires are designed and constructed in this study via efficient and facile strategy. Combining the merits of the core–shell structure and the massive active sites in the oxyphosphide layer, the as‐prepared NCOP composites manifest highly reversible capacitors and outstanding rate capability. Meanwhile, the insertion and conversion potassium storage mechanisms of the NCOP are successfully revealed through in situ X‐ray diffraction and density functional theory calculations, respectively. Furthermore, the PIHC was assembled with NCOP anode and borocarbonitride cathode, which displays a large energy density and high‐power density, along with an exceptional capacity retention of ≈90% over 10 000 cycles at 1.0 A g−1. This work provides the anion regulation strategy for modifying the transition metal oxide and constructing the advancing electrode materials for next‐generation energy storage and beyond. Novel bimetallic oxyphosphide nanowires (NCOP) as advanced anode for potassium ion hybrid capacitor are synthesized via a universal anion‐exchange strategy. Benefiting from the massive defects and multiple cation valence of the oxyphosphide and the stable core‐shell structure of the NCOP, the anode exhibits superior potassium storage capability. DFT calculations and In‐situ XRD reveal the potassium storage mechanism of the NCOP.

The accuracy of cosmic ray observations by the Large High Altitude Air Shower Observatory Wide Field-of-View Cherenkov/Fluorescence Telescope Array (LHAASO-WFCTA) is influenced by variations in aerosols in the atmosphere. The solar photometer (CE318-T) is extensively utilized within the Aerosol Robotic Network as a highly precise and reliable instrument for aerosol measurements. With this CE318-T 23, 254 sets of valid data samples over 394 days from October 2020 to October 2022 at the LHAASO site were obtained. Data analysis revealed that the baseline Aerosol Optical Depth (AOD) and Ångström Exponent (AE) at 440–870 nm (AE440–870nm) of the aerosols were calculated to be 0.03 and 1.07, respectively, suggesting that the LHAASO site is among the most pristine regions on Earth. The seasonality of the mean AOD is in the order of spring > summer > autumn = winter. The monthly average maximum of AOD440nm occurred in April (0.11 ± 0.05) and the minimum was in December (0.03 ± 0.01). The monthly average of AE440–870nm exhibited slight variations. The seasonal characterization of aerosol types indicated that background aerosol predominated in autumn and winter, which is the optimal period for the absolute calibration of the WFCTA. Additionally, the diurnal daytime variations of AOD and AE across the four seasons are presented. Our analysis also indicates that the potential origins of aerosol over the LHAASO in four seasons were different and the atmospheric aerosols with higher AOD probably originate mainly from Northern Myanmar and Northeast India regions. These results are presented for the first time, providing a detailed analysis of aerosol seasonality and origins, which have not been thoroughly documented before in this region, also enriching the valuable materials on aerosol observation in the Hengduan Mountains and Tibetan Plateau.

Sonocatalytic therapy (SCT) offers a non‐invasive and deep tissue‐penetrating approach to addressing the pathological challenges of atherosclerosis. However, its therapeutic efficacy remains limited by the lack of efficient sonosensitizers. A critical challenge in SCT is simultaneously leveraging beneficial plaque microenvironment factors, such as elevated H2O2 levels, while mitigating adverse conditions, including hypoxia. Herein, a microenvironment‐regulatable single‐atom sonozyme system is presented to enable effective SCT while simultaneously refining the lesion microenvironment. The single‐atom manganese catalyst (SMC) is synthesized via MOF‐derived precursor pyrolysis followed by ion implantation, yielding atomically precise four‐coordinated active sites. Functionalized with hyaluronic acid (HA) facilitates targeted delivery of SMC‐HA to M1 macrophages. Under ultrasound (US), SMC‐HA effectively eliminates M1 macrophages, thereby reducing plaque burden and promoting lesion regression in two ApoE−/− mice models. Overall, SMC‐HA reinforces its role as an advanced sonosensitizer for SCT. This study establishes SMC‐HA‐mediated SCT as a promising therapeutic strategy for atherosclerotic plaque treatment. A tetranitrogen‐coordinated single‐atom manganese catalyst (SMC) is fabricated with outstanding sonosensitization efficiency and multi‐enzyme‐like catalytic properties. To enable better targeting of M1 macrophages, hyaluronic acid (HA) is applied to modify the surface of SMC, yielding SMC‐HA nanozymes. The SMC‐HA exhibits superior sonocatalytic activity. Finally, the as‐synthesized SMC‐HA demonstrates significantly enhanced anti‐atherosclerotic effect in ApoE–/– mouse model.

Deep Neural Networks (DNNs) are integral to various aspects of modern life, enhancing work efficiency. Nonetheless, their susceptibility to diverse attack methods, including backdoor attacks, raises security concerns. We aim to investigate backdoor attack methods for image categorization tasks, to promote the development of DNN towards higher security. Research on backdoor attacks currently faces significant challenges due to the distinct and abnormal data patterns of malicious samples, and the meticulous data screening by developers, hindering practical attack implementation. To overcome these challenges, this study proposes a Gaussian Noise-Targeted Universal Adversarial Perturbation (GN-TUAP) algorithm. This approach restricts the direction of perturbations and normalizes abnormal pixel values, ensuring that perturbations progress as much as possible in a direction perpendicular to the decision hyperplane in linear problems. This limits anomalies within the perturbations improves their visual stealthiness, and makes them more challenging for defense methods to detect. To verify the effectiveness, stealthiness, and robustness of GN-TUAP, we proposed a comprehensive threat model. Based on this model, extensive experiments were conducted using the CIFAR-10, CIFAR-100, GTSRB, and MNIST datasets, comparing our method with existing state-of-the-art attack methods. We also tested our perturbation triggers using various defense methods and further experimented on the robustness of the triggers against noise filtering techniques. The experimental outcomes demonstrate that backdoor attacks leveraging perturbations generated via our algorithm exhibit cross-model attack effectiveness and superior stealthiness. Furthermore, they possess robust anti-detection capabilities and maintain commendable performance when subjected to noise-filtering methods.

BiPO4/Bi2S3 photocatalysts were successfully synthesized by a simple two-step hydrothermal process, which involved the initial formation of BiPO4 rod and then the attachment of Bi2S3 through ion exchange. The as-synthesized products were characterized by X-ray diffraction (XRD), scanning electron microscope (SEM), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), and UV-vis diffuse reflectance spectra (UV-vis DRS). It was found that BiPO4 was regular rods with smooth surfaces. However, BiPO4/Bi2S3 heterojunction had a rough surface, which could be attributed to the attachment of Bi2S3 on the surface of BiPO4 rods. The BiPO4/Bi2S3 composite exhibited better photocatalytic performance than that of pure BiPO4 and Bi2S3 for the degradation of methylene blue (MB) and Rhodamine B (RhB) under visible light. The enhanced photocatalytic performance could be ascribed to synergistic effect of BiPO4/Bi2S3 heterojunction, in which the attached Bi2S3 nanoparticles could improve visible-light absorption and the BiPO4/Bi2S3 heterojunction suppressed the recombination of photogenerated electron-hole pairs. Our work suggested that BiPO4/Bi2S3 heterojunction could be a potential photocatalyst under visible light.