Search

Aspect-based sentiment classification, a fine-grained sentiment analysis task, aims to predict the sentiment polarity for a specified aspect. However, the existing aspect-based sentiment classification approaches cannot fully model the dependency-relationship between words and are easily disturbed by irrelevant aspects. To address this problem, we propose a novel approach named Dependency-Relationship Embedding and Attention Mechanism-based LSTM. DA-LSTM first merges the word hidden vector output by LSTM with the dependency-relationship embedding to form a combined vector. This vector is then fed into the attention mechanism together with the aspect information which can avoid interference to calculate the final word representation for sentiment classification. Our extensive experiments on benchmark data sets clearly show the effectiveness of DA-LSTM.

With the rapid development of nanotechnology, the size of a device reaches sub-nanometer scale. The larger resistivity of interconnect leads to serious overheating of integrated circuits. Silicon-based electronic devices have also reached the physical limits of their development. The use of carbon nanotubes instead of traditional wires has become a new solution for connecting nano-structures. Nanocluster particles serving as brazing material play an important role in stabilizing the connection of carbon nanotubes, which places higher demands for nanoscale manipulation techniques. In this paper, the dynamic processes under different operating scenarios were simulated and analyzed, including probe propulsion nanoparticle operation, probe pickup nanoparticle operation and probe pickup nanocluster particle operation. Then, the SEM (Scanning Electron Microscope) was used for nanoparticle manipulation experiments. The smallest unit of carbon nanotube wire was obtained by three-dimensional (3D) construction of a carbon nanotube–silver nanocluster particle (CN-AgNP), which verified the feasibility of 3D manipulation of carbon nanotube wire construction. The experiments on the construction of carbon nanotube–nanocluster particle structures in three-dimensional operation were completed, and the smallest unit of carbon nanotube wire was constructed. This nano-fabrication technology will provide an efficient and mature technical means in the field of nano-interconnection.

Privacy-preserving cross-modal retrieval is a significant problem in the area of multimedia analysis. As the amount of data is exploding, cross-modal data analysis and retrieval is often realized on cloud computing environment. Therefore, the privacy protection of large-scale cross-modal data has become a problem that can not be ignored. To further improve the accuracy and efficiency of privacy-preserving search, this paper proposes a novel cross-modal hashing scheme, named deep adversarial privacy-preserving cross-modal hashing (DAP $$^2$$ CMH). This method consists of a deep cross-modal hashing model termed DACMH, and a secure index structure called CMH $$^2$$ -Tree. The former is a combination of deep hashing and adversarial learning to capture intra-modal and inter-modal correlation. The latter is a hierarchical hashing index structure that can provide efficient data organization based on cross-modal hash codes. We conduct comprehensive experiments on three common used benchmarks. The results show that the proposed approach DAP $$^2$$ CMH outperforms the state-of-the-arts.

White-light emitting (WLE) polymers have attracted continuous attention for their promising applications in solid-state lighting, flexible display and related fields. However, achieving dual-emission and pure white-light emission in a single-component polymer is still challenging. In this study, a brominated single-component polymer BrOD-TFB was designed and synthesized, which shows dual-emission and white light emission properties in solution and room-temperature phosphorescence (RTP) in thin films. The dual-emission properties can be tuned by concentration, solvent polarity, and excitation energy. Spectral analyses and theoretical calculations reveal that the origin of the high-energy emission band (HEB) is intramolecular charge transfer (ICT) along the polymer chain, whilst the low-energy emission band (LEB) originates from the excited-state related to the intra-chain and inter-chain C-Br⋯π interactions as demonstrated by the single-crystal structure of the model compound. Appropriate control of the formation and the destruction of the halogen-assisted interactions can initiate white-light emission in the single-component polymer. More interestingly, by dispersing BrOD-TFB (0.1 wt%) in a non-emissive, colorless and transparent polymer, the characteristics of this white-light emission can be fully demonstrated while exhibiting unexpected RTP properties, with photoluminescence quantum efficiency (ΦPL) of up to 23% and CIE coordinates of (0.32, 0.32).

In recent years, Hashing has become a popular technique used to support large-scale image retrieval, due to its significantly reduced storage, high search speed and capability of mapping high dimensional original features into compact similarity-preserving binary codes. Although effectiveness achieved, most existing hashing methods are still some limitations, including: (1) Many supervised hashing methods only transform the label information into pairwise similarities to guide the hash code learning process, which will lead to the loss of rich semantic information between image pairs. (2) Some pioneer hashing methods use a relaxation-based strategy to solve discrete problems, resulting in a large quantization error. (3) Some supervised hashing methods handle the hashing learning procedure based on an asymmetric learning manner, and although this partially solves the problems of low efficiency and accuracy of symmetric learning strategy, they are all based on the embeddings of equal dimension, which leads to the reduction in the models representation ability and an increase in potential noise. To overcome the above limitations, in this paper, we propose a novel yet simple but effective hashing method, named Scalable Deep Asymmetric Hashing (SDAH). Specifically, SDAH is an end-to-end deep hashing method based on a fast iterative optimization strategy, which utilizes two real-valued embeddings of unequal dimensions, i.e., real-valued embeddings of images and labels, to flexibly perform asymmetric similarity computation. It can circumvent the use of the large pairwise similarity matrix by introducing an intermediate label matrix term which results in a remarkable reduction in the memory space cost. By doing this, the learned hash codes are more semantically informative for image retrieval tasks. Experimental results on several benchmark datasets highlight the superiority of SDAH in comparison with many state-of-the-art hashing methods.

In previous research, interrelationships between green innovation and supply chain collaboration (SCC) are under-explored. This study explores the effects of SCC on green innovation performance and considers supply chain collaborative drivers as a crucial part of green innovation strategy. We construct a conceptual model containing a set of 17 drivers on the implementation of green innovation and analyze the contextual relationships between these drivers using an interpretive structural modeling (ISM) method to identify the driving and dependence powers in green innovation within the context of SCC. The model is validated on a case study involving an automobile firm in southwestern China. The study reveals that the relationships between supply chain partners have a very high influence on green innovation performance. SCC should be deemed as a core component in green innovation strategies. In addition, environmental regulations, top management commitment and social recognition are among high driving and dependence powers. We have confirmed the significance of SCC enablers for green innovation and identified several specific attributes through research. To further improve the value of this study, theoretical and managerial implications are discussed.

Persistent luminescence from metal-free organic materials is attractive for their ultralong exciton lifetimes. Color-tunable persistent luminescence from single-component organic materials is fascinating but still challenging. By utilizing an efficient approach of "self"-interface energy transfer (IET), the persistent luminescence color of an organic phosphor (CTXO) can be reversibly and continuously tuned by external physical stimuli. Its color circularly changes between green (lifetime = 0.24 s) and deep-yellow (lifetime = 0.10 s) when CTXO is repeatedly triggered with thermal annealing and mechanical grinding. Self-IET from the crystalline part (donor), which exhibits persistent room-temperature phosphorescence, to the amorphous part (acceptor) inside its semicrystal during these treatments is found to be the key exciton process for such novel color modulation. This also provides opportunity for designing stimuli-responsive smart materials with controlled persistent luminescence.

Red thermally activated delayed fluorescence (TADF) emitters with aggregation-induced emission (AIE) properties have been developed, based on strong electron-with-drawing 1,8-naphthalimide (NI) acceptor (A). With strong electron-rich 10-H-phenothiazine (Pz) or 9,9-dimethyl-9,10-dihydroacridine (DMAC) groups as electron donors (D), 2,6-dimethylphenyl (DM) as a π-bridge, these highly twist D-π-A architecture NI derivatives showed typical AIE feature and excellent solid-state TADF emission with high photoluminescence quantum yield (PLQY) up to 55%. Moreover, the NI derivatives exhibited excellent thermal morphological stability in their solid amorphous states, providing a stable EML for practical applications. Finally, a high external quantum efficiency (EQE) of 7.13% from Pz-DMNI, was achieved in the red OLEDs device with electroluminescence (EL) peak of 635 nm.

Summary This article proposed a high-performance triboelectric-electromagnetic hybrid wind energy harvester (WEH). By adopting the revolution and rotation movements of tapered rollers, which serve as both the rotor of the electromagnetic generator (EMG) part and freestanding layers of the triboelectric nanogenerator (TENG) part, the WEH can work as a sustainable power source and a self-powered wind speed sensor. When the wind speed is 12 m/s, super-high open-circuit voltage peaks of 47.4 and 683 V can be achieved by the EMG and TENG, respectively, corresponding to the high-power outputs of 62 and 1.8 mW. It was demonstrated that the WEH can easily light up over 600 red light-emitting diodes and even a 5-W globe light. A self-powered wireless temperature and humidity sensing network was also systematically demonstrated. In summary, the proposed WEH exhibits bright future toward IoT applications, such as in border detection, smart buildings, and so on., Graphical abstract, Highlights • A triboelectric-electromagnetic hybrid generator based on tapered rollers is proposed • The high-performance WEH has been systematically optimized and tested • A self-powered anemometer system has been successfully established • The WEH shows bright prospect in future IoT applications, Electromagnetics ; Energy Resources ; Energy Systems ; Device

Sensors based on triboelectric nanogenerators have been widely used for energy harvesting and sensing applications, however, the applications of multidimensional information perception and interactive control are insufficient. In this paper, we present an ultra-stretchable triboelectric strip sensor (TSS) using triboelectric mechanism for controlling the objects in three-dimensional space. This facile and low-cost TSS is mainly composed of a parallel structure including three symmetric sensor strips fixed on the base and a mobile stage connected with them. Based on the coupling effect of triboelectrification and electrostatic induction, the length changes of each strip with the same finger contacting point generate different signal output ratios from two terminal electrodes, functioning as the interactive interface for multi-dimensional sensing and controlling. Hence, the parallel sensor strips structure can realize the sensing and controlling in three degrees of freedom of linear motion and two degrees of freedom of rotational motion, and resultant sensing ranges are X, Y, Z, α and β (20 mm, 20 mm, 30 mm, 36°, and 36°). In terms of manipulation, this is simpler than the conventional controller with rigid structure and includes additional space dimensions. Furthermore, demonstration of the TSS as human-nanomachine terminal to control the nanomanipulator in scanning electron microscope (SEM) is successfully realized with the accuracy of 10 nm. The proposed TSS shows great potential for the applications in automated control, robotics control, and Internet of Things (IoT).

Escherichia coli O157: H7 (E. coli O157: H7) is a foodborne pathogenic bacterium which can cause fever, diarrhoea and vomiting in humans. Thus, a rapid, simple, and specific bioprobe for pathogen detection in contaminated foods has been attracted more and more attention. In this work, the strong fluorescent amino-functionalized graphene quantum dots (af-GQDs) were prepared by hydrothermal method. The microtopographic height, surface morphology and spectroscopic properties of af-GQDs are characterized by the high resolution transmission electron microscope (HRTEM), atomic force microscope (AFM), UV-vis, fluorescence, Raman spectroscopic techniques. All the results showed that the af-GQDs can be effectively applied in the preparation of biocompatible immunofluorescence probe and in the detection of E. coli O157: H7. The minimum detection limit is 100 cfu/mL. It is a simple, rapid, sensitive, low-cost and easy to be popularized method, which provides a feasible way to monitor E. coli O157: H7 in food safety.

The near-field enhancement underneath the AFM tip irradiated by a laser beam is widely utilized in various nanofabrication techniques, such as nano-welding, nano-manipulation, nano-etching and so on. In this paper, a numerical model of near-field enhancement on AFM tip irradiated by annular laser beam is established using COMSOL Multiphysics. The results show that when the AFM tip is irradiated by annular laser beam, strong near-field enhancement is induced at the apex of the tip. The field distribution curve with the substrate presents a saddle shape, which demonstrates that the annular beam irradiated the AFM tip produce an extremely high near-field enhancement between the particle and the tip. In addition, the results indicate that when the AFM tip is irradiated by the annular laser with higher frequency, the near-field enhancement underneath the apex of the tip increases correspondingly. Moreover, when the irradiation angle is 60 Deg and the irradiation distance is 65 nm, the near-field enhancement reaches the peak. Based on the numerical simulation of the near-field enhancement, the system of the AFM tip irradiated by annular laser beam can be anticipated for further application in the field of nanofabrication.

Due to its fast retrieval and storage efficiency capabilities, hashing has been widely used in nearest neighbor retrieval tasks. By using deep learning based techniques, hashing can outperform non-learning based hashing technique in many applications. However, we argue that the current deep learning based hashing methods ignore some critical problems (e.g., the learned hash codes are not discriminative due to the hashing methods being unable to discover rich semantic information and the training strategy having difficulty optimizing the discrete binary codes). In this paper, we propose a novel image hashing method, termed as \textbf{\underline{A}}symmetric \textbf{\underline{D}}eep \textbf{\underline{S}}emantic \textbf{\underline{Q}}uantization (\textbf{ADSQ}). \textbf{ADSQ} is implemented using three stream frameworks, which consist of one \emph{LabelNet} and two \emph{ImgNets}. The \emph{LabelNet} leverages the power of three fully-connected layers, which are used to capture rich semantic information between image pairs. For the two \emph{ImgNets}, they each adopt the same convolutional neural network structure, but with different weights (i.e., asymmetric convolutional neural networks). The two \emph{ImgNets} are used to generate discriminative compact hash codes. Specifically, the function of the \emph{LabelNet} is to capture rich semantic information that is used to guide the two \emph{ImgNets} in minimizing the gap between the real-continuous features and the discrete binary codes. Furthermore, \textbf{ADSQ} can utilize the most critical semantic information to guide the feature learning process and consider the consistency of the common semantic space and Hamming space. Experimental results on three benchmarks (i.e., CIFAR-10, NUS-WIDE, and ImageNet) demonstrate that the proposed \textbf{ADSQ} can outperforms current state-of-the-art methods., Accepted to IEEE ACCESS. arXiv admin note: text overlap with arXiv:1812.01404

A pair of chiral aggregation-induced emission (AIE) enantiomers and their racemate exhibit not only distinctively different circularly polarized luminescence behaviors, but also completely opposite mechanoluminescence (ML) activities. The data from single crystal analysis identifies that the bright ML of the AIE enantiomers is probably activated by their molecular chirality, which enables unique chiral crystal structures with a polar space group and compact molecular packing. These experimental findings provide a direction for harvesting prominent ML from AIE luminogens. Concurrently, high contrast mechanochromism with larger ΔλPL,max and ΔΦF,s values by up to 140 nm and 0.61 respectively have been achieved from this AIE system, which represents the state-of-the-art mechanochromic performance for ML emitters. Moreover, force-induced white-light emission and thermally activated delayed fluorescence have also been observed. The results of theoretical calculations suggest that the destruction of intermolecular interactions and conversion of molecular conformation account for the significant multiple-variations in emission properties triggered by a single mechanical stimulus.

With the rapid growth of multimedia data (e.g., image, audio and video etc.) on the web, learning-based hashing techniques such as Deep Supervised Hashing (DSH) have proven to be very efficient for large-scale multimedia search. The recent successes seen in Learning-based hashing methods are largely due to the success of deep learning-based hashing methods. However, there are some limitations to previous learning-based hashing methods (e.g., the learned hash codes containing repetitive and highly correlated information). In this paper, we propose a novel learning-based hashing method, named Deep Attention-guided Hashing (DAgH). DAgH is implemented using two stream frameworks. The core idea is to use guided hash codes which are generated by the hashing network of the first stream framework (called first hashing network) to guide the training of the hashing network of the second stream framework (called second hashing network). Specifically, in the first network, it leverages an attention network and hashing network to generate the attention-guided hash codes from the original images. The loss function we propose contains two components: the semantic loss and the attention loss. The attention loss is used to punish the attention network to obtain the salient region from pairs of images; in the second network, these attention-guided hash codes are used to guide the training of the second hashing network (i.e., these codes are treated as supervised labels to train the second network). By doing this, DAgH can make full use of the most critical information contained in images to guide the second hashing network in order to learn efficient hash codes in a true end-to-end fashion. Results from our experiments demonstrate that DAgH can generate high quality hash codes and it outperforms current state-of-the-art methods on three benchmark datasets, CIFAR-10, NUS-WIDE, and ImageNet., Comment: Accepted to IEEE ACCESS

Thermally activated delayed fluorescence (TADF) emitters integrated with aggregation-induced emission (AIE) property are expected to enable nondoped organic light-emitting diodes (OLEDs) with merits of high efficiency and simple fabrication process. We previously proposed a novel design strategy of combining AIE and TADF features to develop AIE–TADF emitters with asymmetric structure, leading to an AIE–TADF emitter 4,4-CzSPz exhibiting mechanoluminescence. Herein, photophysical, thermal, and electroluminescence properties of neat 4,4-CzSPz films have been systematically investigated. It was found that 4,4-CzSPz film exhibits a high photoluminescence quantum efficiency of 97.3% and excellent thermal properties. Significantly, 4,4-CzSPz afforded its doped and nondoped OLEDs with very high external quantum efficiencies of 26.2% and 20.7%, respectively, which are rarely reported among TADF-based OLEDs. The results demonstrate universal applications of 4,4-CzSPz in OLEDs with different device configurations fo...

Metal matrix syntactic foams are consisting of metal matrix and hollow spheres in closely or randomly packed, which own multifunctional properties with lightweight, damping, heat insulation, energy absorption and have a vast application prospect. Steel matrix can extend the potential of syntactic foams as a materials class to several new fields of application. In this paper, the hollow alumina spheres were introduced into the steel matrix by infiltration casting, the minimum diameter of hollow spheres for infiltration is analyzed in theory, the steel matrix syntactic foams were successfully prepared, which contain two different sphere types with average diameter sizes 3.97mm and 4.72mm, and the average densities of syntactic foams were calculated to be 4.39 (spheres occupy 43.7% of the volume) and 3.74 g/cm3 (spheres occupy 52.1% of the volume), respectively. The microstructure characteristics of the steel matrix syntactic foam were analyzed by means of scanning electron microscopy and energy spectrum.

As a new type composite material, steel-based porous materials presented the higher stiffness, strength and higher temperature resistance in comparison with the nonferrous metals. In order to realize the free-design of pore structure and free-control of porosity of steel-based porous material, indirect 3D printing technology combined with the investment casting process has been used to fabricate the steel-based porous materials. The finite element method has been carried out to study the relationship between the porosity and the mechanical properties by quasi-static compression simulation, which is consistent with compression test. The modified Gibson-Ashby formula is used to establish the quantitative relationship between porosity and elastic modulus and yield strength of steel-based porous materials based on Kelvin structure.

In mesoscopic scale, concrete is regarded as a heterogeneous three-phase material composed of mortar, aggregate and interfacial transition zone (ITZ). The effect of mesoscopic structure on the mechanical behaviors of concrete should be paid more attention. The fractal characteristics of aggregate were calculated, then the geometric models of aggregate were reconstructed by using fractal Brownian motion. Based on the random distribution of aggregates, the concrete mesoscopic structure model was established. And the numerical model was generated by using grid mapping technology. The dynamic compression experiments of concrete under Split Hopkinson Pressure Bar (SHPB) loading verify the reliability and validity of the mesoscopic structural model and the parameters of the constitutive model. Based on these, a numerical study of concrete under dynamic splitting is carried out. By changing the parameters of the constitutive model, the effects of tensile strengths of aggregate, mortar and ITZ on the dynamic tensile strength of concrete are discussed. The results show that the dynamic failure of specimen usually occurs at the interfacial transition zone, then extends to the mortar, and the aggregates rarely fail. However, the increase of strain rate intensifies this process. When the strain rate increases from 72.93 s−1 to 186.51 s−1, a large number of aggregate elements are deleted due to reaching the failure threshold. The variation of tensile strengths of each phase component have the same effect on the dynamic tensile strength and energy of concrete. The dynamic tensile strength and energy of concrete are most affected by the tensile strength of mortar, following by the ITZ, but the tensile strength of aggregate has almost no effect.

In situ scanning electron microscope (SEM) characterization have enabled the stretching, compression, and bending of micro/nanomaterials and have greatly expanded our understanding of small-scale phenomena. However, as one of the fundamental approaches for material analytics, torsion tests at a small scale remain a major challenge due to the lack of an ultrahigh precise torque sensor and the delicate sample assembly strategy. Herein, we present a microelectromechanical resonant torque sensor with an ultrahigh resolution of up to 4.78 fN∙m within an ultrawide dynamic range of 123 dB. Moreover, we propose a nanorobotic system to realize the precise assembly of microscale specimens with nanoscale positioning accuracy and to conduct repeatable in situ pure torsion tests for the first time. As a demonstration, we characterized the mechanical properties of Si microbeams through torsion tests and found that these microbeams were five-fold stronger than their bulk counterparts. The proposed torsion characterization system pushes the limit of mechanical torsion tests, overcomes the deficiencies in current in situ characterization techniques, and expands our knowledge regarding the behavior of micro/nanomaterials at various loads, which is expected to have significant implications for the eventual development and implementation of materials science. A microelectromechanical resonant sensor is proposed for small-scale torsion tests and demonstrated for the testing of silicon microbeams. It is important to understand the mechanical properties of small-scale materials used in MEMS and NEMS. Various systems have been reported for this under a range of loading conditions, but the application of torque to small-scale samples is particularly hard to measure. Here, a joint team led by Xueyong Wei from Xi’an Jiaotong University and Yajing Shen from City University of Hong Kong reports a MEMS-based resonator sensor that can measure bending and torsion over a wide dynamic range with a high force resolution. They demonstrate their system for the testing of silicon microbeams, which are found to be five times stronger than bulk silicon under pure torsion.

In the fast tool servo (FTS) system for microstructure surface cutting, the dynamic voltage hysteresis of piezoelectric actuators (PEAs) and the cutting force produced in the manufacturing affect the driving accuracy and the cutting performance. For a multi-input-single-output (MISO) cutting system, in this paper, a dynamic hysteresis model based on a rate-dependent Prandtl&ndash, Ishlinskii model is proposed. A backpropagation neural network (BPNN) is established to describe the cross-coupling effect between the applied voltage and external load. An inverse dynamic model is developed to compensate the nonlinearity of PEAs. The accuracy of the model and its inverse is discussed and the performance of the inverse feedforward compensator is validated through experiments.

Hashing techniques have recently been successfully applied to solve similarity search problems in the information retrieval field because of their significantly reduced storage and high-speed search capabilities. However, the hash codes learned from most recent cross-modal hashing methods lack the ability to comprehensively preserve adequate information, resulting in a less than desirable performance. To solve this limitation, we propose a novel method termed Nonlinear Robust Discrete Hashing (NRDH), for cross-modal retrieval. The main idea behind NRDH is motivated by the success of neural networks, i.e., nonlinear descriptors, in the field of representation learning, and the use of nonlinear descriptors instead of simple linear transformations is more in line with the complex relationships that exist between common latent representation and heterogeneous multimedia data in the real world. In NRDH, we first learn a common latent representation through nonlinear descriptors to encode complementary and consistent information from the features of the heterogeneous multimedia data. Moreover, an asymmetric learning scheme is proposed to correlate the learned hash codes with the common latent representation. Empirically, we demonstrate that NRDH is able to successfully generate a comprehensive common latent representation that significantly improves the quality of the learned hash codes. Then, NRDH adopts a linear learning strategy to fast learn the hash function with the learned hash codes. Extensive experiments performed on two benchmark datasets highlight the superiority of NRDH over several state-of-the-art methods.

The micro-nano robot can achieve precise targeting in the aspect of tumor drug delivery, so that the drug can enrich the focus area to the maximum extent and clear the focus, and the driving methods are varied, including chemical drive, physical drive and biological flagellum drive, etc. However, the development of micro-nano robot is mostly restricted by factors, and how to achieve accurate targeting is the premise of the application. Among many driving modes, magnetic drive has made a great breakthrough in precise control due to the existence of Helmholtz coil, and whether optical drive, which is also a physical driving method, can also realize three-dimensional trajectory motion control of micro-nano robot has become a key issue in current research. In order to solve this problem, spherical TiO 2 of different sizes were prepared by two different technological methods: hydrothermal method and stirring method, and then modeled by COMSOL simulation software, the distribution of the electric field intensity on the surface of TiO 2 microsphere under plane light irradiation was obtained by giving different incident light angles, wavelengths, microsphere sizes, etc. The simulation results were analyzed to obtain the correct control scheme.

The denoising of SEM images and extracting of image information have attracted attention in the micro-nano research fields. In this paper, we analyze the noise types and characteristics of SEM images and propose the method of deep convolutional neural network to denoise SEM images. On the basis of denoising, suitable methods are used to identify the position of single, multiple nanowires and positional relationship of multiple nanowires in the depth direction. Because of special imaging principles, there are many differences between SEM images and ordinary gray images. To the SEM image, noise has more significant influences on extracting information from SEM image. Samples which have various types and morphologies were observed under different types of electron microscopy and observation conditions to produce dataset. Then the supervised deep convolutional neural network was trained by this dataset, which has a good adaptive on denoising of SEM images. After denoising, the identification method of nanowires was further proposed which could accurately recognize single, multiple nanowires by canny edge detection and Hough transform, and quickly identify their positional relationship.

In this paper, we report a novel design of a microspherical lens assembly that can realize super-resolution optical imaging in air with super-wide field of view (FOV). The microspherical lens assembly is made from a barium titanate glass (BTG) microsphere that is glued to a microspherical crown lens through optical adhesive. Experimental results show that super-resolution optical imaging in air can be achieved under a regular optical microscope when the microspherical lens assembly is brought sufficiently close to the sample surface. Super-wide FOV is also achieved when a larger BTG microsphere is used and there seems no obvious limitation on the size of the BTG microsphere as long as the curvature of the microspherical crown lens is properly tuned. Simulation study suggests that the focus point of the microlens needs to close to the bottom surface of the BTG microsphere in order to achieve the super-resolution optical imaging. The novel microspherical lens assembly will make the microsphere based super-resolution imaging a step further close to practical applications.

Graphene is widely used as the basic materials of nano optical devices and sensors on account of its special structures and excellent photoelectric properties. Graphene is considered as an ideal material for photodetectors because of its ultra-wide absorption spectrum from the ultraviolet to the terahertz band, ultrahigh carrier mobility and ultrafast photoreaction speed. In this study, a photothermal nano-device was made using graphene that was transferred to an electrode using an all-dry viscoelastic stamping method. The nano-device has the advantages of simplicity, high efficiency and instant measurement. This nano-device was used to measure the light absorption of graphene, and the calculated light absorption rate of graphene is basically consistent with previous research results. Experiments on irradiation at different wavelengths and thermal heating at different temperatures show that the nano-device has an excellent response to near-infrared and mid-infrared light. The conclusions provide an experimental basis for the research, design and fabrication of nano-devices, and this device can provide an effective method for detecting light and temperature in areas such as electronic components and solar cells.

It is widely accepted that incineration can permanently eliminate plastic waste. However, unburned material still exists in the bottom ash that is a solid residue from incinerators. In this study, microplastics exacted from bottom ash in 12 mass burn incinerators, one bottom ash disposal center and four fluidized bed incinerators were identified by micro-Fourier transform infrared spectroscopy. The results showed that bottom ash was a neglected microplastics source with an abundance of 1.9-565 n/kg, which indicated that per metric ton waste produce 360 to 102,000 microplastic particles after incineration. Nine types of plastics were identified, of which polypropylene and polystyrene were the predominant types. Microplastics sized between 50 μm and 1 mm accounted for 74 %. Granules, fragments, film, and fibers accounted for 43 %, 34 %, 18 %, and 5 % of the microplastics, respectively. The abundance of microplastics differed significantly with whether the local waste was source-separated, the local gross domestic product per capita, and the types of furnace. The global microplastics emission from incineration bottom ash was then estimated. Our observations provide empirical evidence proving that incineration is not the terminator of plastic waste, and bottom ash is a potential source of microplastics released into the environment.

Triboelectric nanogenerators and sensors have been widely adopted for diversified energy harvesting and sensing applications, but the demonstrations of 3D information sensing and controlling are very limited. In this paper, we present a novel self-powered virtual reality 3D-control sensor (VR-3D-CS) based on triboelectric mechanism for controlling the attitude (both the position and rotation) of object in 3D virtual space. This innovative, cost-effective, simple-designed sensor has a symmetric 3D structure with eight separated sensing electrodes and two touching spheres as the interactive interface with human fingers for 3D force information sensing and VR controlling. Based on the coupling effect of triboelectrification and electrostatic induction, the VR-3D-CS generates different electric output signals in response to different operation manner that can be used to control the attitude of objects in 3D virtual space. The symmetrical 3D configuration design of the sensor enables the detection of 3D force from both the normal direction and shear direction. By employing vector properties of force and signal analysis from the eight sensing electrodes, detection of six-axis directions in 3D space is achieved by triboelectric mechanism for the first time. The VR-3D-CS has been demonstrated to be able to detect normal force in the range of 0–18 N. It can resolve the shear force direction with step resolution of at least 15°. Besides, due to the positive output voltage and low internal impedance, the VR-3D-CS is readily compatible with commercial portable signal processing system for signal analysis and controlling. Demonstration of the VR-3D-CS as interactive interface for Augmented Reality (AR) control is successfully realized. The robust structure, stable output performance and self-powered sensing property enable the device as an ideal human machine interface towards AR interface, batteryless and energy saving applications.

Commercial carbazole has been widely used to synthesize organic functional materials that have led to recent breakthroughs in ultralong organic phosphorescence1, thermally activated delayed fluorescence2,3, organic luminescent radicals4 and organic semiconductor lasers5. However, the impact of low-concentration isomeric impurities present within commercial batches on the properties of the synthesized molecules requires further analysis. Here, we have synthesized highly pure carbazole and observed that its fluorescence is blueshifted by 54 nm with respect to commercial samples and its room-temperature ultralong phosphorescence almost disappears6. We discover that such differences are due to the presence of a carbazole isomeric impurity in commercial carbazole sources, with concentrations

Due to the advances in mobile computing and multimedia techniques, there are vast amount of multimedia data with geographical information collected in multifarious applications. In this paper, we propose a novel type of image search named interactive geo-tagged image search which aims to find out a set of images based on geographical proximity and similarity of visual content, as well as the preference of users. Existing approaches for spatial keyword query and geo-image query cannot address this problem effectively since they do not consider these three type of information together for query. In order to solve this challenge efficiently, we propose the definition of interactive top-$k$ geo-tagged image query and then present a framework including candidate search stage , interaction stage and termination stage. To enhance the searching efficiency in a large-scale database, we propose the candidate search algorithm named GI-SUPER Search based on a new notion called superior relationship and GIR-Tree, a novel index structure. Furthermore, two candidate selection methods are proposed for learning the preferences of the user during the interaction. At last, the termination procedure and estimation procedure are introduced in brief. Experimental evaluation on real multimedia dataset demonstrates that our solution has a really high performance., 24 pages

Among numerous noble-metal-free electrocatalysts, molybdenum sulfides are recognized as promising candidates for hydrogen evolution reaction (HER). Owing to abundant under-coordinated sulfur atoms serving as catalytically active centers, intensive spectra studies have revealed that the HER property of amorphous molybdenum sulfide (a-MoSx) is superior to crystal molybdenum sulfide (c-MoS2). In this work, nitrogen-doped active carbon (NAC) is obtained through plasma treatment and amorphous MoSx/NAC hybrid catalyst films are electrodeposited on carbon fiber papers (CFP) which are employed as porous three-dimensional electrodes with low resistivity. The incorporation of NAC increases the electrochemically active surface area, enhances the electron transport, and facilitates the reaction kinetics. Moreover, the functionality durability benefits from synergistic effect between a-MoSx and NAC. Thus, the obtained hybrid catalyst delivers the excellent HER activity, requiring only 203 mV overpotential to achieve a geometrical current density of 100 mA cm−2 with a Tafel slope of ∼43.9 mV per decade.

Catalytic micro- and nanorobots exhibit self-propulsion in the presence of chemical fuels such as glucose, hydrogen peroxide. Generally, swimmers incorporate a catalyst as a part of their structure for the decomposition of a chemical fuel. Several applications such as active drug delivery, minimally invasive surgery, and water remediation, have been envisaged. In this letter, we investigated the influence of geometric variable of catalytic microrobot to their swimming behaviors in hydrogen peroxide (H2O2) solutions. V-shaped microrobots with various opening angles (45°, 90°, and 135°), and triangle-shaped microrobots that have similar dimensions with V-shaped microrobots, have been fabricated by focused-ion-beam etching on commercial platinum microwires. Swimming tests were performed in 15% and 30% H2O2 solutions. We have found that V-shaped microrobot with 45° opening angle showed a superior swimming speed of 4.3 μm/s in 15% H2O 2 solution, while 90° microrobot exhibits a speed of 1.52 μm/s. The 135° V-shaped microrobot and all triangle-shaped microrobots have not shown any obvious motions in 30% H2O2 solution. Our observation that the swimming performances of various shaped microrobots change in tandem proves the necessity of structural asymmetry in designing catalytic micro- and nanorobots.

Persistent luminescence from purely organic materials is basically triggered by light and electricity, which largely confines its practical applications. A purely organic AIEgen exhibits not only persistent photoluminescence, but also transient and persistent room-temperature mechanoluminescence. By simply turning on and off a UV lamp, tricolor emission switching between blue, white, and yellow was achieved. The data from single-crystal structure analysis and theoretical calculation suggest that mechanism of the observed persistent mechanoluminescence (pML) is correlated with the strong spin-orbit coupling of the bromine atom, as well as the formation of H-aggregates and restriction of intramolecular motions in noncentrosymmetric crystal structure. These results outline a fundamental principle for the development of new pML materials, providing an important step forward in expanding the application scope of persistent luminescence.

Thienyl containing triarylethylene derivatives, namely 2ThDpF and 3ThDpF, with photochromic properties have been designed and synthesized. These new photochromic molecules with simple chemical structures show fast-response and striking photochromic behaviors in the solution-state, in the solid-state and in polymer films. Based on 2ThDpF, a real-time and repeatable photoresponsive surface was fabricated. Upon UV-light irradiation for 1 minute, most of the 2ThDpF nano-aggregates on the SiO2 substrate changed from cone-shaped to hump-shaped and the contact angles of a water droplet drastically increased from 43° to 95° simultaneously. The surface morphology and wettability can easily be reverted by white-light irradiation for 5 minutes. The key issues to affect the real-time morphology and wettability changes are also discussed in-depth based on single-crystal analyses and TDDFT calculations. Therefore, these triarylethylene derivatives with simple molecular structures are attractive in the areas of photochromism and photoresponsive surfaces.

Thermally activated delayed fluorescence (TADF) emitters can harvest singlet and triplet excitons for light emission to afford highly efficient organic light-emitting diode (OLED) devices, especially for all-TADF white OLEDs (WOLEDs). However, the majority of TADF emitters suffer from concentration quenching effects and require complicated doping techniques for device fabrication. Herein, we demonstrate a yellow TADF emitter that enables non-doped OLED systems, which is beneficial for the simplification of the OLED device structure and fabrication. The non-doped yellow OLED achieves a maximum external quantum efficiency (EQE) of 16.7% at a high luminance of 3600 cd m−2, in which such a high magnitude is rarely reported. More significantly, by the combination of the yellow TADF emitter with a blue TADF emitter, a fully non-doped emissive layer (EML) strategy has been demonstrated for the fabrication of an all-TADF WOLED, in which the non-doped yellow and blue TADF emissive layers are closely stacked together without any interlayers. The fully non-doped WOLEDs show a maximum EQE of up to 19.8%, along with high quality white-light emission. This is the first report regarding the fully non-doped all-TADF WOLEDs, and the presented design strategy provides a universal route towards the fabrication of simple and efficient WOLEDs.

Vast amount of multimedia data contains massive and multifarious social information which is used to construct large-scale social networks. In a complex social network, a character should be ideally denoted by one and only one vertex. However, it is pervasive that a character is denoted by two or more vertices with different names; thus it is usually considered as multiple, different characters. This problem causes incorrectness of results in network analysis and mining. The factual challenge is that character uniqueness is hard to correctly confirm due to lots of complicated factors, for example, name changing and anonymization, leading to character duplication. Early, limited research has shown that previous methods depended overly upon supplementary attribute information from databases. In this paper, we propose a novel method to merge the character vertices which refer to the same entity but are denoted with different names. With this method, we firstly build the relationship network among characters based on records of social activities participating, which are extracted from multimedia sources. Then we define temporal activity paths (TAPs) for each character over time. After that, we measure similarity of the TAPs for any two characters. If the similarity is high enough, the two vertices should be considered as the same character. Based on TAPs, we can determine whether to merge the two character vertices. Our experiments showed that this solution can accurately confirm character uniqueness in large-scale social network.

The persistent room-temperature phosphorescence (RTP) of purely organic materials in the solid state has recently attracted much research interest and found promising, rapid and visual applications by the naked eye, after the removal of the excitation source. However, almost all reported organic persistent RTP processes are induced by using a UV-light irradiation source. In this report, persistent RTP with an emission lifetime as long as 0.15 s which can be induced not only by photoirradiation but also by mechanical action is presented, which merges mechano-induced luminescence and persistent RTP together. It is found that such persistent RTP is produced through intermolecular electronic coupling (IEC) of units with different excited state configurations. Interestingly, there are two different crystals with and without mechano-induced persistent RTP which can be grown from this organic material, as such a new type of mechano-luminescence (ML) is strongly dependent on their intermolecular interactions. Furthermore, the intensity of such mechano-induced persistent RTP can be increased by lowing the temperature as well, similar to that of photo-induced persistent RTP. Notably, these two crystals exhibit a ML enhancement and ML “turn on”, respectively, with decreasing temperature. Therefore, such mechano-induced persistent RTP provides an example of new types of organic luminescent materials, which is a missing jigsaw piece of organic luminescence and important for both fundamental research and practical applications of both persistent RTP and ML of organic materials.

Due to the coaction of supply harmonics and the harmonics resulting from the motor structures, the mechanism and distribution characteristics of iron losses become more complicated in inverter-fed induction motors. Therefore, accurate prediction and fine analysis of iron loss are very important at the design stage of high-efficiency inverter-fed induction motors. In order to predict the iron losses accurately, this paper proposes a piecewise iron loss model whose parameters vary with the magnitude and frequency of flux density, and two additional flux density terms are introduced to the classical iron loss model considering the nonlinearity of magnetic material and harmonic fields. With this model, the iron losses are calculated for an inverter-fed 5.5 kW induction motor. The results reveal the distribution characteristics of hysteresis and eddy current losses in stator and rotor cores, and the characteristics of additional iron loss caused by harmonic fields. By the comparison of predicted and measured no-load iron loss under different supply voltages and switching frequencies, the proposed model and the analysis results are validated.

The walking state monitoring is indispensable during the robot-aided walking of people with lower limb dysfunctions. In this paper, the existence of human–robot coordination state is first statistically verified in the process of using a walking-aid cane-type robot during walking. Based on this coordination, a new walking state monitoring method is proposed by using the principal component analysis (PCA). The abnormal or emergency walking state is promptly detected if the new sample data are found to deviate from an off-line PCA model, which is generated from plentiful normal walking data of different subjects. Furthermore, a state diagnosis algorithm based on the contribution plot is also developed for the walking state recognition and diagnosis. In this way, typical abnormal states like the leg restrictions can be distinguished from the emergency states including falls and the stumbling. Moreover, the human–robot coordination analysis can be performed using less sensors built-in the robot without needing the posture information of full human body. The effectiveness of the proposed method is proven by experiments. Better recognition rate and real-time performance of the method are also verified by comparing with conventional center of pressure based monitoring method.

Thermally activated delayed fluorescence (TADF) emitters have aroused serious attention for their use in organic light-emitting diodes (OLEDs). However, TADF emitters providing both doped and non-doped OLEDs with high device performance are rare. In this work, we present a yellow TADF emitter (OPDPO) meeting the above requirements. The OPDPO-based doped OLEDs show a maximum external quantum efficiency (EQE) of 26.7%, which is amongst the highest ever reported for doped yellow TADF-based OLEDs. The EQE remains as high as 22.1% at a luminance of 1000 cd m−2. Meanwhile, the OPDPO-based non-doped OLEDs afford a maximum EQE of 16.6% at a high luminance of 3600 cd m−2, which is amongst the best performances of non-doped yellow TADF-based OLEDs. Moreover, simple and efficient white OLEDs are implemented by combining a blue conventional fluorescent emitter with either a doped or non-doped OPDPO emissive layer, confirming the universal applications of OPDPO in high-performance OLEDs.

Warm-white organic light-emitting diodes (WOLEDs) fabricated with hybrid fluorescence-based emitters, consisting of traditional fluorescence and thermally activated delayed fluorescence (TADF) emitters can achieve high efficiency without the use of expensive organometallic phosphors. However, most of these types of OLEDs require complicated fabrication techniques, especially host–guest doping systems and multiple interlayers. Herein, an efficient design strategy for hybrid fluorescence WOLEDs is presented, in which a traditional blue fluorescence emitter is simply stacked together with a yellow TADF emitter without any interlayers. Due to the high triplet excited state of the chosen blue emitter, its intrinsically non-radiative triplet excitons can be efficiently utilized for light emission, by exoergic transfer to the adjacent yellow TADF emitter. Following this idea, a maximum external quantum efficiency of 20.8% at a high luminance of 400 cd m−2 is achieved for a fully non-doped WOLED when using a blue fluorescent emitter with the aggregation-induced emission property. This strategy provides a universal approach for the fabrication of highly-efficient fluorescence–TADF hybrid WOLEDs based on the fully non-doped device configuration.

Stretchable superhydrophobic materials are of potential interest for next-generation multi-functional haptic technologies particularly suited for wearable device and artificial skin applications. However, the practical applications of stretchable superhydrophobic materials are hindered by some issues, such as low mechanical robustness, harmful chemicals, etc. Here, we partially embedded perfluorosilane-coated graphene into thermoplastic polyurethane (TPU) by a dissolution and resolidification method. Due to the exceptional physical properties of graphene, the resulting nanocomposites could maintain their superhydrophobicity after toleration of strain up to 400%, man-made destruction by hands or sandpapers, pollution by oil, immersion in various corrosive liquids, and heat treatment at 150 °C for 24 h. It should be noted that even under an ultra-large load of 2 kg (32.5 kPa pressure), this graphene composite could withstand the abrasion by sandpaper for 20.00 m without losing superhydrophobicity. More remarkably, the electrical resistance of this graphene superhydrophobic composite is sensitive to material deformation, and can be directly applied to gloves for real-time detection of human motions.

Hashing has been shown to be successful in a number of Approximate Nearest Neighbor (ANN) domains, ranging from medicine, computer vision to information retrieval. However, current deep hashing methods either ignore both rich information of labels and visual linkages of image pairs, or leverage relaxation-based algorithms to address discrete problems, resulting in a large information loss. To address the aforementioned problems, in this paper, we propose an E nhanced D eep D iscrete H ashing (EDDH) method to leverage both label embedding and semantic-visual similarity to learn the compact hash codes. In EDDH, the discriminative capability of hash codes is enhanced by a distribution-based continuous semantic-visual similarity matrix, where not only the margin between the positive pairs and negative pairs is expanded, but also the visual linkages between image pairs is considered. Specifically, the semantic-visual continuous similarity matrix is constructed by analyzing the asymmetric generalized Gaussian distribution of the visual linkages between pairs with label consideration. Besides, in order to achieve an efficient hash learning framework, EDDH employs an asymmetric real-valued learning structure to learn the compact hash codes. In addition, we develop a fast discrete optimization algorithm, which can directly generate discrete binary codes in single step, and introduce an intermediate term before iterations to avoid the problems caused by directly the use of large semantic-visual similarity matrix, which results in a significant reduction in the computational overhead. Finally, we conducted extensive experiments on three datasets to show that EDDH has a significantly enhanced performance compared to the compared state-of-the-art baselines.

Interfacial engineering methods have been developed to solve defect issues of perovskite solar cells (PSCs). However, traditional surface passivation has limited effects on eliminating defect-forming residuals, while secondary grain growth (SGG) is restricted by limited choices of additives and intrinsic properties of perovskites. Here, a pincer strategy of taking advantages of surface passivation and SGG is proposed to modify both exterior and interior of CH3 NH3 PbI3 (MAPbI3 ) perovskite, by employing cyanoacetate-containing donor-acceptor compounds (CA-D-A) including 2-cyano-3-(3,4,5-trimethoxyphenyl)acrylic acid (CA), methanaminium 2-cyano-3-(3,4,5-trimethoxyphenyl)acrylate (CAMA), and aminomethaniminium (Z)-2-cyano-3-(3,4,5-trimethoxyphenyl)acrylate (CAFA). In comparison to untreated perovskite, CA-D-A treated perovskites present better crystallinity because of SGG, lower trap densities due to the synergistic effect of surface passivation and SGG, and tuned energy levels induced by CA-D-A. Accordingly, the CA-D-A treated MAPbI3 -based PSCs exhibit higher open-circuit voltage and fill factor than the control PSC without any treatment, leading to improved power conversion efficiency (PCE) and enhanced device stability, especially the CAMA treated PSCs with an average PCE promoted from 17.77 (control PSCs) to 18.71%, and importantly an excellent PCE of 19.71% through further optimization. This work provides an effective strategy for developing highly efficient and stable PSCs with the assistance of both surface passivation and SGG.

Online healthcare communities (OHCs) have become producers of medical information. Solving the issue of how to effectively reuse such a large amount of medical data and discover its potential value is of the utmost importance for alleviating the shortage of medical resources. Online consultation has received widespread attention and population since its first appearance in 1999, and as a result, many diagnostic multi-turn questions and answers (Q&A) documents have become available. This type of document is formed by multiple rounds of patient questions and doctors’ diagnostic answers and contains massive medical knowledge and doctors’ diagnostic experience. Few studies concentrate on the modeling and recommendation of this type of document, yet making these documents convenient for reuse reduces the cost of medical consultation for patients and saves time addressing common diseases for doctors. In this paper, we focus on the modeling and understanding of diagnostic multi-turn Q&A records and propose a deep-learning recommendation framework based on patient medical information needs, the contents of Q&A records and doctor background information. With the evaluation based on a real dataset that contains pediatric consultation dialogues fetched from DingXiangYuan, a famous online consultation application in China, we found that the proposed model achieved a good performance on the recommendation of diagnostic multi-turn Q&A records and outperformed baseline models. In addition, we discussed a potential application scenario of the recommendation model, suggesting that the proposed model can promote the reduction of patient costs and doctors’ work pressure in countries or regions with insufficient medical resources.

Hashing technology has been widely used in approximate nearest neighbor search algorithms for large-scale cross-modal retrieval due to its significantly reduced storage and high-speed search capabilities. However, most existing supervised cross-modal hashing methods either mainly rely on binary pairwise similarity and fail to exploit the rich semantic information contained in the label matrix, or the use of single linear projections which suffer from limited information completeness. In this paper, we propose a novel method, named Nonlinear Supervised Discrete Hashing (NSDH). Specifically, NSDH consists of two components, (1) a semantic enhancement descriptor consisting of multiple linear projections that is used to extract comprehensive latent representations of heterogeneous multimedia data, which aligns the original heterogeneous features and integrates the rich semantic label matrix; (2) a fast discrete optimization module used to learn discriminative compact hash codes, which preserves the similarity information using an inner product between the real-valued embeddings of the output of the semantic enhancement descriptors. Therefore, NSDH leverages both the label matrix and similarity information in order to enhance the semantic information of the learned hash codes. In this way, the representation learning capability of the output layer of semantic enhancement descriptors can be greatly enhanced and as a result the learned hash codes are more discriminative. In addition, we present a fast discrete optimization algorithm to efficiently learn the binary hash codes. Results from our experiments on two benchmark datasets highlight the superiority of NSDH in comparison to many state-of-the-art cross-modal hashing methods.

Since the amplitude and frequency of irregular waves change with time, great difficulties are brought for solving ship load responses in random waves. To take the effect of various frequencies of irregular waves into consideration in load responses of hull, the wave memory effect is necessary. A semi-analytical method is introduced for the time-domain retardation functions, and then a nonlinear hydroelastic method considering memory effect for ships in irregular waves is proposed. Segmented self-propelling model experiments of a container ship were carried out in a towing tank, a ship motion measuring device for self-propelling model test was designed. Whipping responses of the ship in regular and irregular waves are analyzed. Finally, the calculation results are compared with those measured by segmented model experiments, and the result indicates that the memory effect has little effect on load responses of ship in regular waves, but pronounced effect on results in irregular waves. Moreover, the presented method is reasonable for the prediction of ship load responses in irregular waves.

The widespread presence of humic acids (HAs) and chromium (Cr) in aquatic systems and drinking water sources is a serious threat to the environment and human being. However, no studies on the potential interaction and simultaneous removal of these two types of contaminants have been conducted. Here, a novel electro-Fenton process was reported for the effective and simultaneous removal of HAs and Cr(VI), where H2O2 was in situ produced from the electro-generated H2 and O2 on the Pd/Fe3O4 nanocatalyst surface and Fe(II) was provided from the support. We show that HAs exhibit as reductants and sorbents as well as electron shuttles for the removal of Cr(VI); HAs were efficiently mineralized, as revealed by a total organic carbon removal efficiency of beyond 90%, and Cr(VI) was completely reduced with 90% of total Cr removal. The Pd/Fe3O4 nanocatalysts, as electron transfer and hydrogen atom transfer carriers, were found to play fundamental roles for the removal of HAs and Cr(VI). The HAs degradation was driven by the sequentially involved O 2 - and OH radicals during electro-Fenton process. The reduction of Cr(VI) was primarily attributed to the contributions from Pd/Fe3O4 nanoparticles and atomic hydrogen on the catalyst surface. In a similar manner, a good recycle of Fe3+ to Fe2+ was accelerated, which favored the Fenton reaction to generate reactive oxidizing species. Importantly, the energy consumption of this electro-Fenton process for HAs degradation was almost 2 orders of magnitude lower than those of reported electrochemical oxidation. The Pd/Fe3O4 nanocatalysts showed excellent removal performances for HAs and Cr(VI) after eight times repeated treatment. This work reports a cost-effective methodology for simultaneous removal of HAs and Cr(VI) and provides a new insight into efficient elimination of complex contaminants by heterogeneous catalysis.

Energy harvesting technique has been identified as an innovative and environment-friendly way for powering micro sensors and electronic devices. In this paper, a hybrid generator based on triboelectric and electromagnetic mechanisms (TMHG), which provides two electrical power outputs at the same time, was proposed. The output performance in time and frequency domains are investigated at different initial gap distances and excitation accelerations. At an acceleration of 1 g and an initial gap distance of 5 mm, the maximum power outputs of electromagnetic and triboelectric parts can be obtained as 0.28 mW and 1.09 mW, respectively. The resonant frequency of the TMHG is as low as 23.5 Hz, making it more suitable to harvest energy from low frequency vibrations of less than 30 Hz. Besides, the operating bandwidth of TMHG can be adjusted by changing the initial gap distance according to the application requirements.

Non-skewed rotors can be used to eliminate the slot harmonic field, as well as, the noise and vibration can also be weakened effectively. However, it is difficult to determine the spatial distribution of the asymmetrical rotor bars and slot combinations, which can affect the loss and air-gap force significantly. Taking a 5.5kW, 4-pole induction motor as an example, this paper studies the influence of the above factors on the loss and air-gap force systematically, by using Finite Element Method (FEM), and the experiment validation on efficiency and no-load noise are also performed on a 5.5 kW prototype.