Dithizone-functionalized C18 online solid-phase extraction-HPLC-ICP-MS for speciation of ultra-trace organic and inorganic mercury in cereals and environmental samples.

A simple and efficient dithizone-functionalized solid-phase extraction (SPE) procedure, online coupled with high-performance liquid chromatography (HPLC)-inductively coupled plasma mass spectrometry, was developed for the first time for enrichment and determination of ultra-trace mercury (Hg) species (inorganic divalent Hg (Hg(II)), methylmercury (CH3Hg(II)) and ethylmercury (C2H5Hg(II)) in cereals and environmental samples. In the proposed method, functionalization of the commercial C18 column with dithizone, enrichment, and elution of the above Hg species can be completed online with the developed SPE device. A simple solution of 2-mercaptoethanol (1% (V/V)) could be used as an eluent for both the SPE and HPLC separation of Hg species, significantly simplifying the method and instrumentation. The online SPE method was optimized by varying dithizone dose, 2-mercaptoethanol concentration, and sample volume. In addition, the effect of pH, coexisting interfering ions, and salt effect on the enrichment was also discussed. Under the optimized conditions, the detection limits of Hg species for 5 mL water sample were 0.15 ng/L for Hg(II), 0.07 ng/L for CH3Hg(II), and 0.04 ng/L for C2H5Hg(II) with recoveries in the range of 85%-100%. The developed dithizone-functionalized C18 SPE column can be reused after a single functionalization, which significantly simplifies the enrichment step. Moreover, the stability of Hg species enriched on the SPE column demonstrated its suitability for field sampling of Hg species for later laboratory analysis. This environment-friendly method offers a robust tool to detect ultra-trace Hg species in cereals and environmental samples.

Enhanced Human Activity Recognition Using Wearable Sensors via a Hybrid Feature Selection Method.

The study of human activity recognition (HAR) plays an important role in many areas such as healthcare, entertainment, sports, and smart homes. With the development of wearable electronics and wireless communication technologies, activity recognition using inertial sensors from ubiquitous smart mobile devices has drawn wide attention and become a research hotspot. Before recognition, the sensor signals are typically preprocessed and segmented, and then representative features are extracted and selected based on them. Considering the issues of limited resources of wearable devices and the curse of dimensionality, it is vital to generate the best feature combination which maximizes the performance and efficiency of the following mapping from feature subsets to activities. In this paper, we propose to integrate bee swarm optimization (BSO) with a deep Q-network to perform feature selection and present a hybrid feature selection methodology, BAROQUE, on basis of these two schemes. Following the wrapper approach, BAROQUE leverages the appealing properties from BSO and the multi-agent deep Q-network (DQN) to determine feature subsets and adopts a classifier to evaluate these solutions. In BAROQUE, the BSO is employed to strike a balance between exploitation and exploration for the search of feature space, while the DQN takes advantage of the merits of reinforcement learning to make the local search process more adaptive and more efficient. Extensive experiments were conducted on some benchmark datasets collected by smartphones or smartwatches, and the metrics were compared with those of BSO, DQN, and some other previously published methods. The results show that BAROQUE achieves an accuracy of 98.41% for the UCI-HAR dataset and takes less time to converge to a good solution than other methods, such as CFS, SFFS, and Relief-F, yielding quite promising results in terms of accuracy and efficiency.

Arsenic Speciation on Silver Nanofilms by Surface-Enhanced Raman Spectroscopy.

Surface-enhanced Raman spectroscopy (SERS), as a nondestructive and fast detection technique, is a promising alternative approach for arsenic detection, particularly for in situ applications. SERS-based speciation analysis according to the fingerprint SERS signals of different arsenicals has the potential to provide a superior technique in species preservation over the conventional chromatographic separation methods, albeit with some difficulties due to the similarity in SERS patterns. In this study, we explored a novel SERS method for arsenic speciation by using the separation potential of the coffee ring effect on negatively charged silver nanofilms (AgNFs). Four arsenic species, including arsenite (AsIII), arsenate (AsV), monomethylarsonic acid (MMAV), and dimethylarsinic acid (DMAV), were measured for fingerprint SERS signals in solution and on the films. Significant enhancement of SERS signals on the dried coffee ring stains by the AgNFs were observed except for AsIII, and more importantly, arsenicals migrated varying distances during coffee ring development, promoting better speciation. Sodium dodecyl sulfate was then introduced into the droplet to reduce the droplet surface tension, facilitating the migration of solution into the peripheral region. Under the combined interactions of arsenicals with the AgNFs, solvent, and surfactant, enhanced separation between arsenicals was observed as a result of the formation of two concentric rings. Combining the SERS fingerprint signals and physical separation of arsenicals on the surface, arsenic speciation was achieved using the AgNFs substrate-based SERS technology, demonstrating the potential of the coffee ring effect for rapid separation and analysis of small molecules by SERS.

Study on Tandem Polymer Light Emitting Devices.

We report tandem polymer light emitting devices by using the PEDOT∶PSS/ZnO/PEIE charge generation layer (CGL) and investigate the influences of the conductance and thickness of PEDOT∶PSS layer on the properties of the devices. The results indicate that the conductance and thickness of PEDOT∶PSS layer have marginal impact on the J-V characteristics of the devices, while significant influences of device efficiency upon utilization of different PEDOT∶PSS specimens mainly come from their different strengths on exciton quenching. Luminance efficiency of TOLEDs with the PEDOT∶PSS thickness of 60 nm in CGL is better than TOLEDs with the PEDOT∶PSS thickness of 30 nm in CGL, the reason is that PEDOT∶PSS thickness of 60 nm the surface topography is more even . Luminance efficiency and driving voltage of the tandem devices match the sum of the luminance efficiency and driving voltage of the component light-emitting units, respectively, indicating that charges generated in the CGL can be injected efficiently into the adjacent light-emitting units. Incorporation of a V2O5 layer into the CGL structure only slightly affects the J-V and LE-I characteristics of the tandem devices, suggesting that the utilization of the PEDOT∶PSS/ZnO/PEIE CGL enables the simplification of the CGL structure without compromising device performance. The luminescence spectra of TOLEDs obviously involves two light emitting unit of spectrum, which shows that two light emitting unit in TOLEDs is normal work. Measurements on the capacitance-voltage characteristics of the CGL-based devices confirm that under negative bias (ITO anode) charges are accumulated and displaced in the CGL, which is totally in line with the full operation of light emitting units in the tandem devices. PEDOT∶PSS/ZnO/PEIE layer is evidenced the effective CGL. On this basis, for the first time we report tandem polymer light emitting devices containing three SY-PPV light-emitting units，which show the mixture of luminance efficiency and external quantum efficiency of 21.7 cd·A-1 and 6.95%, similar to the total luminance efficiency and external quantum efficiency of constituent LEUs. At 5 000 cd·m-2, the luminance efficiency and external quantum efficiency of the tandem devices are 20.5 cd·A-1 and 6.6%. Thus, the increase in the number of light emitting units leads to almost no performance losses, implying the robustness of the PEDOT∶PSS/ZnO/PEIE CGL. Tandem polymer light emitting devices containing three SY-PPV light-emitting units of the luminescent spectra is close to the light emitting unit. Further efforts on the optimization of hole injection layer in the CGL to minimize exciton quenching are underlying to promote the luminance efficiency of tandem polymer light emitting devices.

Photochromic copolymers containing 3-indolylfulgides/indolylfulgimides: synthesis and photochemical properties in toluene and as films.

Photochromic indolylfulgimides covalently attached to polymers have beneficial properties for optical switching. A 3-indolylfulgide and two 3-indolylfulgimides with one or two polymerizable styrene groups attached on the nitrogen atom(s) were synthesized. Copolymerization with methyl methacrylate (MMA) provided linear copolymers (one styrene group) or a cross-linked copolymer (two styrene groups). The properties of the monomers and copolymers in toluene or as thin films were characterized. The new copolymers were photochromic (reversible Z-to-C isomerization), absorbed visible light, and revealed good thermal and photochemical stability. At room temperature, all copolymer films showed no loss of absorbance after 5 weeks. At 80 °C in either toluene or as films, the Z-forms copolymers were less stable than the C-form copolymers, which showed little or no degradation after 400 h. The degradation rate due to repeated ring-closing - ring opening cycles was less than 3% per 100 cycles. The cross-linked copolymer showed photochemical stability comparable to monomeric fulgides in toluene, <1% per 100 cycles. In general, the properties of the linear and cross-linked copolymers were similar to the corresponding monomers in toluene. In films, the conformations of the Z-form were restricted due to the matrix indicating that the preparation of films from the C-form is advantageous.

Knowledge Graph Alignment Under Scarce Supervision: A General Framework With Active Cross-View Contrastive Learning.

Over recent years, a number of knowledge graphs (KGs) have emerged. Nevertheless, a KG can never reach full completeness. A viable approach to increase the coverage of a KG is KG alignment (KGA). The majority of previous efforts merely focus on the matching between entities, while largely neglect relations. Besides, they heavily rely on labeled data, which are difficult to obtain in practice. To address these issues, in this work, we put forward a general framework to simultaneously align entities and relations under scarce supervision. Our proposal consists of two main components, relation-enhanced active instance selection (RAS), and cross-view contrastive learning (CCL). RAS aims to select the most valuable instances to be labeled with the guidance of relations, while CCL contrasts cross-view representations to augment scarce supervision signals. Our proposal is agnostic to the underlying entity and relation alignment models, and can be used to improve their performance under limited supervision. We conduct experiments on a wide range of popular KG pairs, and the results demonstrate that our proposed model and its components can consistently boost the alignment performance under scarce supervision.

Searching for spin glass ground states through deep reinforcement learning.

Spin glasses are disordered magnets with random interactions that are, generally, in conflict with each other. Finding the ground states of spin glasses is not only essential for understanding the nature of disordered magnets and many other physical systems, but also useful to solve a broad array of hard combinatorial optimization problems across multiple disciplines. Despite decades-long efforts, an algorithm with both high accuracy and high efficiency is still lacking. Here we introduce DIRAC - a deep reinforcement learning framework, which can be trained purely on small-scale spin glass instances and then applied to arbitrarily large ones. DIRAC displays better scalability than other methods and can be leveraged to enhance any thermal annealing method. Extensive calculations on 2D, 3D and 4D Edwards-Anderson spin glass instances demonstrate the superior performance of DIRAC over existing methods. The presented framework will help us better understand the nature of the low-temperature spin-glass phase, which is a fundamental challenge in statistical physics. Moreover, the gauge transformation technique adopted in DIRAC builds a deep connection between physics and artificial intelligence. In particular, this opens up a promising avenue for reinforcement learning models to explore in the enormous configuration space, which would be extremely helpful to solve many other hard combinatorial optimization problems.

Strategy evaluation and optimization with an artificial society toward a Pareto optimum.

Strategy evaluation and optimization in response to troubling urban issues has become a challenging issue due to increasing social uncertainty, unreliable predictions, and poor decision-making. To address this problem, we propose a universal computational experiment framework with a fine-grained artificial society that is integrated with data-based models. The purpose of the framework is to evaluate the consequences of various combinations of strategies geared towards reaching a Pareto optimum with regards to efficacy versus costs. As an example, by modeling coronavirus 2019 mitigation, we show that Pareto frontier nations could achieve better economic growth and more effective epidemic control through the analysis of real-world data. Our work suggests that a nation's intervention strategy could be optimized based on the measures adopted by Pareto frontier nations through large-scale computational experiments. Our solution has been validated for epidemic control, and it can be generalized to other urban issues as well.

Finding key players in complex networks through deep reinforcement learning.

Finding an optimal set of nodes, called key players, whose activation (or removal) would maximally enhance (or degrade) certain network functionality, is a fundamental class of problems in network science1,2. Potential applications include network immunization3, epidemic control4, drug design5, and viral marketing6. Due to their general NP-hard nature, those problems typically cannot be solved by exact algorithms with polynomial time complexity7. Many approximate and heuristic strategies have been proposed to deal with specific application scenarios1,2,8-12. Yet, we still lack a unified framework to efficiently solve this class of problems. Here we introduce a deep reinforcement learning framework FINDER, which can be trained purely on small synthetic networks generated by toy models and then applied to a wide spectrum of influencer finding problems. Extensive experiments under various problem settings demonstrate that FINDER significantly outperforms existing methods in terms of solution quality. Moreover, it is several orders of magnitude faster than existing methods for large networks. The presented framework opens up a new direction of using deep learning techniques to understand the organizing principle of complex networks, which enables us to design more robust networks against both attacks and failures.

Speciation of thioarsenicals through application of coffee ring effect on gold nanofilm and surface-enhanced Raman spectroscopy.

Thioarsenicals, such as dimethylmonothioarsinic acid (DMMTAV) and dimethyldithioarsinic acid (DMDTAV), have been increasingly discovered as important arsenic metabolites, yet analysis of these unstable arsenic species remains a challenging task. A method based on surface-enhanced Raman spectroscopy (SERS) detection in combination with the coffee ringeffect for separation is expected to be particularly useful for analysis of thioarsenicals, thanks to minimal sample pretreatment and unique fingerprint Raman identification. Such a method would offer an alternative approach that overcomes limitations of conventional arsenic speciation techniques based on high performance liquid chromatography separation and mass spectrometry detection. A novel analytical method based on combination of the coffee ringeffect and SERS was developed for the speciation of thiolated arsenicals. A gold nanofilm (AuNF) was employed not only as a SERS substrate, but also as a platform for the separation of thioarsenicals. Once a drop of the thioarsenicals solution was placed onto the AuNF and evaporation of the solvent and the ring stamp formation onto AuNF began, the SERS signal intensity substantially increased from center to edge regions of the evaporated droplet due to the presence of the coffee ring effect. Through calculating the pKa's of DMMTAV and DMDTAV and accordingly manipulating the chemical environment, separation of these thioarsenicals was realized as they travelled different distances during the development of the coffee ring. The migration distances of individual species were influenced by a radial outward flow of a solute, the thioarsenicals-AuNF interactions and a thermally induced Marangoni flow. The separation of DMMTAV (center) and DMDTAV (edge) on the coffee ring, in combination with fingerprint SERS spectra, enables the identification of these thioarsenicals by this AuNF-based coffee ring effect-SERS method.

Geographical origin differentiation of Chinese Angelica by specific metal element fingerprinting and risk assessment.

Traceability offers significant information about the quality and safety of Chinese Angelica, a medicine and food homologous substance. In this study, a systematic four-step strategy, including sample collection, specific metal element fingerprinting, multivariate statistical analysis, and benefit-risk assessment, was developed for the first time to identify Chinese Angelica based on geographical origins. Fifteen metals in fifty-six Chinese Angelica samples originated from three provinces were analyzed. The multivariate statistical analysis model established, involving hierarchical cluster analysis (HCA), principal component analysis (PCA), and self-organizing map clustering analysis was able to identify the origins of samples. Furthermore, benefit-risk assessment models were created by combinational calculation of chemical daily intake (CDI), hazard index (HI), and cancer risk (CR) levels to evaluate the potential risks of Chinese Angelica using as traditional Chinese medicine (TCM) and food, respectively. Our systematic strategy was well convinced to accurately and effectively differentiate Chinese Angelica based on geographical origins.

Thiolation in arsenic metabolism: a chemical perspective.

In recent years, methylated thioarsenicals have been widely detected in various biological and environmental matrices, suggesting their broad involvement and biological importance in arsenic metabolism. However, very little is known about the formation mechanism of methylated thioarsenicals and the relation between arsenic methylation and thiolation processes. It is timely and necessary to summarize and synthesize the reported information on thiolated arsenicals for an improved understanding of arsenic thiolation. To this end, we examined the proposed formation pathways of methylated oxoarsenicals and thioarsenicals from a chemical perspective and proposed a novel arsenic metabolic scheme, in which arsenic thiolation is integrated with methylation (instead of being separated from methylation as currently reported). We suggest in the new scheme that protein-bound pentavalent arsenicals are critical intermediates that connect methylation and thiolation, with protein binding of pentavalent methylated thioarsenical being a key step for arsenic thiolation. This informative review on arsenic thiolation from the chemical perspective will be helpful to better understand the arsenic metabolism at the molecular level and the toxicological effects of arsenic species.

High-Efficiency Phosphorescent Hybrid Organic-Inorganic Light-Emitting Diodes Using a Solution-Processed Small-Molecule Emissive Layer.

The morphology and optical and electrical properties of solution-processed and vacuum-deposited 4,4',4″-tris(carbazol-9-yl)triphenylamine (TCTA):2,2'-(1,3-phenylene)bis[5-(4-tert-butylphenyl)-1,3,4-oxadiazole] (OXD-7) composite films are investigated. All of the films exhibit smooth and pinhole-free morphology, while the evaporated films possess enhanced carrier-transport properties compared to solution-processed ones. The close correlation between the carrier-transport feature and the packing density of the film is established. High-efficiency monochromatic and white phosphorescent hybrid organic-inorganic light-emitting diodes with solution-processed small-molecule emissive layers are reported: the maximum external quantum efficiencies of blue, yellow, and red devices are 18.9, 14.6, and 10.2%, respectively; white devices show a maximum luminance efficiency of 40 cd A(-1) and a power efficiency of 20.8 lm W(-1) at 1000 cd m(-2). The efficiencies of blue, red, and white devices represent significant improvement over previously reported values.