Institute of Intelligent Machines

Metal oxide gas sensors are predominant solid-state gas detecting devices for domestic, commercial and industrial applications, which have many advantages such as low cost, easy production, and compact size. However, the performance of such sensors is significantly influenced by the morphology and structure of sensing materials, resulting in a great obstacle for gas sensors based on bulk materials or dense films to achieve highly-sensitive properties. Lots of metal oxide nanostructures have been developed to improve the gas sensing properties such as sensitivity, selectivity, response speed, and so on. Here, we provide a brief overview of metal oxide nanostructures and their gas sensing properties from the aspects of particle size, morphology and doping. When the particle size of metal oxide is close to or less than double thickness of the space-charge layer, the sensitivity of the sensor will increase remarkably, which would be called "small size effect", yet small size of metal oxide nanoparticles will be compactly sintered together during the film coating process which is disadvantage for gas diffusion in them. In view of those reasons, nanostructures with many kinds of shapes such as porous nanotubes, porous nanospheres and so on have been investigated, that not only possessed large surface area and relatively mass reactive sites, but also formed relatively loose film structures which is an advantage for gas diffusion. Besides, doping is also an effective method to decrease particle size and improve gas sensing properties. Therefore, the gas sensing properties of metal oxide nanostructures assembled by nanoparticles are reviewed in this article. The effect of doping is also summarized and finally the perspectives of metal oxide gas sensor are given.

To detect trace trinitrotoluene (TNT) explosives deposited on various surfaces instantly and on-site still remains a challenge for homeland security needs against terrorism. This work demonstrates a new concept and its utility for visual detection of TNT particulates on various package materials. The concept takes advantages of the superior fluorescent properties of quantum dots (QDs) for visual signal output via ratiometric fluorescence, the feasibility of surface grafting of QDs for chemical recognition of TNT, and the ease of operation of the fingerprint lifting technique. Two differently sized CdTe QDs emitting red and green fluorescences, respectively, have been hybridized by embedding the red-emitting one in silica nanoparticles and covalently linking the green-emitting one to the silica surface, respectively, to form a dual-emissive fluorescent hybrid nanoparticle. The fluorescence of red QDs in the silica nanoparticles stays constant, whereas the green QDs functionalized with polyamine can selectively bind TNT by the formation of Meisenheimer complex, leading to the green fluorescence quenching due to resonance energy transfer. The variations of the two fluorescence intensity ratios display continuous color changes from yellow-green to red upon exposure to different amounts of TNT. By immobilization of the probes on a piece of filter paper, a fingerprint lifting technique has been innovated to visualize trace TNT particulates on various surfaces by the appearance of a different color against a yellow-green background under a UV lamp. This method shows high selectivity and sensitivity with a detection limit as low as 5 ng/mm(2) on a manila envelope and the attribute of being seen with the naked eye.

This paper reports a surface functional monomer-directing strategy for the highly dense imprinting of 2,4,6-trinitrotoluene (TNT) molecules at the surface of silica nanoparticles. It has been demonstrated that the vinyl functional monomer layer of the silica surface can not only direct the selective occurrence of imprinting polymerization at the surface of silica through the copolymerization of vinyl end groups with functional monomers, but also drive TNT templates into the formed polymer shells through the charge-transfer complexing interactions between TNT and the functional monomer layer. The two basic processes lead to the formation of uniform core-shell TNT-imprinted nanoparticles with a controllable shell thickness and a high density of effective recognition sites. The high capacity and fast kinetics to uptake TNT molecules show that the density of effective imprinted sites in the nanoshells is nearly 5 times that of traditional imprinted particles. A critical value of shell thickness for the maximum rebinding capacity was determined by testing the evolution of rebinding capacity with shell thickness, which provides new insights into the effectiveness of molecular imprinting and the form of imprinted materials. These results reported here not only can find many applications in molecularly imprinting techniques but also can form the basis of a new strategy for preparing various polymer-coating layers on silica support.

Abstract A multifunctional Au‐coated TiO 2 nanotube array is made via synthesis of a TiO 2 nanotube array through a ZnO template, followed by deposition of Au particles onto the TiO 2 surface using photocatalytic deposition and a hydrothermal method, respectively. Such arrays exhibit superior detection sensitivity with high reproducibility and stability. In addition, due to possessing stable catalytic properties, the arrays can clean themselves by photocatalytic degradation of target molecules adsorbed to the substrate under irradiation with UV light into inorganic small molecules using surface‐enhanced Raman spectroscopy (SERS) detection, so that recycling can be achieved. Finally, by detection of Rhodamine 6G (R6G) dye, herbicide 4‐chlorophenol (4‐CP), persistent organic pollutant (POP) dichlorophenoxyacetic acid (2,4‐D), and organophosphate pesticide methyl‐parathion (MP), the unique recyclable properties indicate a new route in eliminating the single‐use problem of traditional SERS substrates and show promising applications for detecting other organic pollutants.

A well-known gas sensing material SnO2 in combination with reduced graphene oxide was used in heavy metal ions detection for the first time. This work reports the detailed study on the SnO2/reduced graphene oxide nanocomposite modified glass carbon electrode, which could be used for the simultaneous and selective electrochemical detection of ultratrace Cd(II), Pb(II), Cu(II), and Hg(II) in drinking water. The SnO2/reduced graphene oxide nanocomposite electrode was characterized voltammetrically using redox couples (Fe(CN)63–/4–), complemented with electrochemical impedance spectroscopy (EIS). Square wave anodic stripping voltammetry (SWASV) has been used for the detection of Cd(II), Pb(II), Cu(II), and Hg(II). The detection limit (3σ method) of the SnO2/reduced graphene oxide nanocomposite modified GCE toward Cd(II), Pb(II), Cu(II) and Hg(II) is 1.015 × 10–10 M, 1.839 × 10–10 M, 2.269 × 10–10 M, and 2.789 × 10–10 M, respectively, which is very well below the guideline value given by the World Health Organization. The chemical and electrochemical parameters that exert influence on deposition and stripping of metal ions, such as supporting electrolytes, pH value, deposition potential, and deposition time, were carefully studied. An interesting phenomenon of mutual interference was observed. Most importantly, we pose a potential for the use of gas sensing material in heavy metal ions detection.

Abstract Low-density compressible materials enable various applications but are often hindered by structure-derived fatigue failure, weak elasticity with slow recovery speed and large energy dissipation. Here we demonstrate a carbon material with microstructure-derived super-elasticity and high fatigue resistance achieved by designing a hierarchical lamellar architecture composed of thousands of microscale arches that serve as elastic units. The obtained monolithic carbon material can rebound a steel ball in spring-like fashion with fast recovery speed (∼580 mm s −1 ), and demonstrates complete recovery and small energy dissipation (∼0.2) in each compress-release cycle, even under 90% strain. Particularly, the material can maintain structural integrity after more than 10 6 cycles at 20% strain and 2.5 × 10 5 cycles at 50% strain. This structural material, although constructed using an intrinsically brittle carbon constituent, is simultaneously super-elastic, highly compressible and fatigue resistant to a degree even greater than that of previously reported compressible foams mainly made from more robust constituents.

Here, we report the shell thickness-dependent Raman enhancement of silver-coated gold nanoparticles (Au@Ag NPs) for the identification and detection of pesticide residues at various fruit peels. The Raman enhancement of Au@Ag NPs to a large family of sulfur-containing pesticides is ~2 orders of magnitude stronger than those of bare Au and Ag NPs, and there is a strong dependence of the Raman enhancement on the Ag shell thickness. It has been shown for the first time that the huge Raman enhancement is contributed by individual Au@Ag NPs rather than aggregated Au@Ag NPs with "hot spots" among the neighboring NPs. Therefore, the Au@Ag NPs with excellent individual-particle enhancement can be exploited as stand-alone-particle Raman amplifiers for the surface identification and detection of pesticide residues at various peels of fruits, such as apple, grape, mango, pear, and peach. By casting the particle sensors onto fruit peels, several types of pesticide residues (e.g., thiocarbamate and organophosphorous compounds) have been reliably/rapidly detected, for example, 1.5 nanograms of thiram per square centimeter at apple peel under the current unoptimized condition. The surface-lifting spectroscopic technique offers great practical potentials for the on-site assessment and identification of pesticide residues in agricultural products.

Since the discovery of carbon nanotubes (CNTs), they have drawn considerable research attention and have shown great potential application in many fields due to their unique structural, mechanical, and electronic properties. However, their native insolubility severely holds back the process of application. In order to overcome this disadvantage and broaden the scope of their application, chemical functionalization of CNTs has attracted great interest over the past several decades and produced various novel hybrid materials with specific applications. Notably, the rapid development of functionalized CNTs used as electrochemical sensors has been successfully witnessed. In this featured article, the recent progress of electrochemical sensors based on functionalized CNTs is discussed and classified according to modifiers covering organic (oxygen functional groups, small organic molecules, polymers, DNA, protein, etc.), inorganic (metal nanoparticles, metal oxide, etc.) and organic-inorganic hybrids. By employing some representative examples, it will be demonstrated that functionalized CNTs as templates, carriers, immobilizers and transducers are promising for the construction of electrochemical sensors.

Remote sensing has more advantages than the traditional methods of land surface water (LSW) mapping because it is a low-cost, reliable information source that is capable of making high-frequency and repeatable observations. The normalized difference water indexes (NDWIs), calculated from various band combinations (green, near-infrared (NIR), or shortwave-infrared (SWIR)), have been successfully applied to LSW mapping. In fact, new NDWIs will become available when Advanced Land Imager (ALI) data are used as the ALI sensor provides one green band (Band 4), two NIR bands (Bands 6 and 7), and three SWIR bands (Bands 8, 9, and 10). Thus, selecting the optimal band or combination of bands is critical when ALI data are employed to map LSW using NDWI. The purpose of this paper is to find the best performing NDWI model of the ALI data in LSW map. In this study, eleven NDWI models based on ALI, Thematic Mapper (TM), and Enhanced Thematic Mapper Plus (ETM+) data were compared to assess the performance of ALI data in LSW mapping, at three different study sites in the Yangtze River Basin, China. The contrast method, Otsu method, and confusion matrix were calculated to evaluate the accuracies of the LSW maps. The accuracies of LSW maps derived from eleven NDWI models showed that five NDWI models of the ALI sensor have more than an overall accuracy of 91% with a Kappa coefficient of 0.78 of LSW maps at three test sites. In addition, the NDWI model, calculated from the green (Band 4: 0.525–0.605 μm) and SWIR (Band 9: 1.550–1.750 μm) bands of the ALI sensor, namely NDWIA4,9, was shown to have the highest LSW mapping accuracy, more than the other NDWI models. Therefore, the NDWIA4,9 is the best indicator for LSW mapping of the ALI sensor. It can be used for mapping LSW with high accuracy.

A simple, instrument-free, paper-based analytical device with dual-emission carbon dots (CDs) (blue CDs and red CDs) was developed for the semiquantitative, visual, and sensitive speciation analysis of lead ions in a real sample with a sensitive detection limit of 2.89 nM. When a paper strip was immersed into the sample solution, the blue fluorescence was quenched by Pb2+ in solution, while the red fluorescence served as a background reference without color change, and significant color evolutions from blue to red were observed under the ultraviolet lamp, resulting in a semiquantitative visual detection. Furthermore, a smartphone was used in the visual detection of lead ions by identifying the RGB value of the fluorescent probe solution and corresponding paper strip. The application of smartphones and fluorescent paper strips has greatly shortened the detection time and reduced the cost of detection, providing a new strategy for the on-site and semiquantitative detection of heavy-metal ions in water samples.

In this paper, a novel heuristic structure optimization methodology for radial basis probabilistic neural networks (RBPNNs) is proposed. First, a minimum volume covering hyperspheres (MVCH) algorithm is proposed to select the initial hidden-layer centers of the RBPNN, and then the recursive orthogonal least square algorithm (ROLSA) combined with the particle swarm optimization (PSO) algorithm is adopted to further optimize the initial structure of the RBPNN. The proposed algorithms are evaluated through eight benchmark classification problems and two real-world application problems, a plant species identification task involving 50 plant species and a palmprint recognition task. Experimental results show that our proposed algorithm is feasible and efficient for the structure optimization of the RBPNN. The RBPNN achieves higher recognition rates and better classification efficiency than multilayer perceptron networks (MLPNs) and radial basis function neural networks (RBFNNs) in both tasks. Moreover, the experimental results illustrated that the generalization performance of the optimized RBPNN in the plant species identification task was markedly better than that of the optimized RBFNN.

Mn2+-doped ZnS nanocrystals with an amine-capping layer have been synthesized and used for the fluorescence detection of ultratrace 2,4,6-trinitrotoluene (TNT) by quenching the strong orange Mn2+ photoluminescence. The organic amine-capped nanocrystals can bind TNT species from solution and atmosphere by the acid-base pairing interaction between electron-rich amino ligands and electron-deficient aromatic rings. The resultant TNT anions bound onto the amino monolayer can efficiently quench the Mn2+ photoluminescence through the electron transfer from the conductive band of ZnS to the lowest unoccupied molecular orbital (LUMO) of TNT anions. The amino ligands provide an amplified response to the binding events of nitroaromatic compounds by the 2- to approximately 5-fold increase in quenching constants. Moreover, a large difference in quenching efficiency was observed for different types of nitroaromatic analytes, dependent on the affinity of nitro analytes to the amino monolayer and their electron-accepting abilities. The amine-capped nanocrystals can sensitively detect down to 1 nM TNT in solution or several parts-per-billion of TNT vapor in atmosphere. The ion-doped nanocrystal sensors reported here show a remarkable air/solution stability, high quantum yield, and strong analyte affinity and, therefore, are well-suited for detecting the ultratrace TNT and distinguishing different nitro compounds.

A bright blue fluorescent graphene oxide that originates from passivation of surface reactive sites by amide formation and ring-opening amination of epoxide has been prepared. The surface polarity and charges of the fluorescent graphene oxide can synchronously be tuned by varying the used alkylamines.

Iron and 1,3,5-benzenetricarboxylic (Fe–BTC) metal–organic coordination polymers are synthesized via a simple solvothermal method. The as-synthesized Fe–BTC polymers exhibit gel behavior, which is stable in common organic solvents or in water. The Fe–BTC polymer as an adsorbent for arsenic removal from water is tested. The kinetics and thermodynamics of arsenic adsorption by the Fe–BTC polymer in aqueous solution are investigated comprehensively. The effect of pH on the adsorption is also investigated. Kinetic studies show that the kinetic data are well described by the pseudo-second-order kinetic model. The thermodynamic analysis indicates that the adsorption is spontaneous. The adsorption isotherms can be well described with the Langmuir equation. The Fe–BTC polymers show relatively high arsenic adsorption capacity, more than 6 times that of iron oxide nanoparticles with a size of 50 nm and 36 times that of commercial iron oxide powders. Hence, the as-synthesized Fe–BTC polymers possess relatively high stability and better adsorption characteristic than nanomaterials simultaneously. It also can be considered as a new method to conquer the dilemma between the excellent properties from nanoscale effect and the aggregation of small size particles in the adsorption application of nanoparticle materials.

The "fixed" or "flexible" design of plasmonic hotspots is a frontier area of research in the field of surface-enhanced Raman scattering (SERS). Most reported SERS hotspots have been shown to exist in zero-dimensional point-like, one-dimensional linear, or two-dimensional planar geometries. Here, we demonstrate a novel three-dimensional (3D) hotspot matrix that can hold hotspots between every two adjacent particles in 3D space, simply achieved by evaporating a droplet of citrate-Ag sols on a fluorosilylated silicon wafer. In situ synchrotron-radiation small-angle X-ray scattering (SR-SAXS), combined with dark-field microscopy and in situ micro-UV, was employed to explore the evolution of the 3D geometry and plasmonic properties of Ag nanoparticles in a single droplet. In such a droplet, there is a distinct 3D geometry with minimal polydispersity of particle size and maximal uniformity of interparticle distance, significantly different from the dry state. According to theoretical simulations, the liquid adhesive force promotes a closely packed assembly of particles, and the interparticle distance is not fixed but can be balanced in a small range by the interplay of the van der Waals attraction and electrostatic repulsion experienced by a particle. The "trapping well" for immobilizing particles in 3D space can result in a large number of hotspots in a 3D geometry. Both theoretical and experimental results demonstrate that the 3D hotspots are predictable and time-ordered in the absence of any sample manipulation. Use of the matrix not only produces giant Raman enhancement at least 2 orders of magnitude larger than that of dried substrates, but also provides the structural basis for trapping molecules. Even a single molecule of resonant dye can generate a large SERS signal. With a portable Raman spectrometer, the detection capability is also greatly improved for various analytes with different natures, including pesticides and drugs. This 3D hotspot matrix overcomes the long-standing limitations of SERS for the ultrasensitive characterization of various substrates and analytes and promises to transform SERS into a practical analytical technique.

In this study, a novel adaptive sparse representation (ASR) model is presented for simultaneous image fusion and denoising. As a powerful signal modelling technique, sparse representation (SR) has been successfully employed in many image processing applications such as denoising and fusion. In traditional SR‐based applications, a highly redundant dictionary is always needed to satisfy signal reconstruction requirement since the structures vary significantly across different image patches. However, it may result in potential visual artefacts as well as high computational cost. In the proposed ASR model, instead of learning a single redundant dictionary, a set of more compact sub‐dictionaries are learned from numerous high‐quality image patches which have been pre‐classified into several corresponding categories based on their gradient information. At the fusion and denoising processes, one of the sub‐dictionaries is adaptively selected for a given set of source image patches. Experimental results on multi‐focus and multi‐modal image sets demonstrate that the ASR‐based fusion method can outperform the conventional SR‐based method in terms of both visual quality and objective assessment.

Feature selection (FS) is an important component of many pattern recognition tasks. In these tasks, one is often confronted with very high-dimensional data. FS algorithms are designed to identify the relevant feature subset from the original features, which can facilitate subsequent analysis, such as clustering and classification. Structured sparsity-inducing feature selection (SSFS) methods have been widely studied in the last few years, and a number of algorithms have been proposed. However, there is no comprehensive study concerning the connections between different SSFS methods, and how they have evolved. In this paper, we attempt to provide a survey on various SSFS methods, including their motivations and mathematical representations. We then explore the relationship among different formulations and propose a taxonomy to elucidate their evolution. We group the existing SSFS methods into two categories, i.e., vector-based feature selection (feature selection based on lasso) and matrix-based feature selection (feature selection based on l <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">r,p</sub> -norm). Furthermore, FS has been combined with other machine learning algorithms for specific applications, such as multitask learning, multilabel learning, multiview learning, classification, and clustering. This paper not only compares the differences and commonalities of these methods based on regression and regularization strategies, but also provides useful guidelines to practitioners working in related fields to guide them how to do feature selection.

For detecting substances that are invisible to the human eye or nose, and particularly those biomolecules, the devices must have very small feature sizes, be compact and provide a sufficient level of sensitivity, often to a small number of biomolecules that are just a few nanometres in size. Electrical nanogap devices for biosensing have emerged as a powerful technique for detecting very small quantities of biomolecules. The most charming feature of the devices is to directly transduce events of biomolecules specific binding into useful electrical signals such as resistance/impedance, capacitance/dielectric, or field-effect. Nanogap devices in electrical biosensing have become a busy area of research which is continually expanding. A wealth of research is available discussing planar and vertical nanogap devices for biosensing. Planar nanogap devices including label-free, gold nanoparticle-labeled, nanoparticles-enhanced, nanogapped gold particle film, and carbon nanotube nanogap devices as well as vertical nanogap devices with two and three terminals for biosensing are carefully reviewed. The aim of this paper is to provide an updated overview of the work in this field. In each part, we discuss the principles of operation of electrical biosensing and consider major strategies for enhancing their performance and/or key challenges and opportunities in current stages, and in their further development.

In this brief, a novel local descriptor, named local binary count (LBC), is proposed for rotation invariant texture classification. The proposed LBC can extract the local binary grayscale difference information, and totally abandon the local binary structural information. Although the LBC codes do not represent visual microstructure, the statistics of LBC features can represent the local texture effectively. In addition, a completed LBC (CLBC) is also proposed to enhance the performance of texture classification. Experimental results obtained from three databases demonstrate that the proposed CLBC can achieve comparable accurate classification rates with completed local binary pattern.

Early detection of arrhythmia and effective treatment can prevent deaths caused by cardiovascular disease (CVD). In clinical practice, the diagnosis is made by checking the electrocardiogram (ECG) beat-by-beat, but this is usually time-consuming and laborious. In the paper, we propose an automatic ECG classification method based on Continuous Wavelet Transform (CWT) and Convolutional Neural Network (CNN). CWT is used to decompose ECG signals to obtain different time-frequency components, and CNN is used to extract features from the 2D-scalogram composed of the above time-frequency components. Considering the surrounding R peak interval (also called RR interval) is also useful for the diagnosis of arrhythmia, four RR interval features are extracted and combined with the CNN features to input into a fully connected layer for ECG classification. By testing in the MIT-BIH arrhythmia database, our method achieves an overall performance of 70.75%, 67.47%, 68.76%, and 98.74% for positive predictive value, sensitivity, F1-score, and accuracy, respectively. Compared with existing methods, the overall F1-score of our method is increased by 4.75~16.85%. Because our method is simple and highly accurate, it can potentially be used as a clinical auxiliary diagnostic tool.