Efficient Machine Learning Model Focusing on Active Sites for the Discovery of Bifunctional Oxygen Electrocatalysts in Binary Alloys.

The distinctive characteristics of alloy catalysts, encompassing composition, structure, and modifiable adsorption sites, present significant potential for the development of highly efficient electrocatalysts for oxygen evolution/reduction reactions [oxygen evolution reactions (OERs)/oxygen reduction reactions (ORRs)]. Machine learning (ML) methods can quickly establish the relationship between material features and catalytic activity, thus accelerating the development of alloy electrocatalysts. However, the current abundance of features presents a crucial challenge in selecting the most pertinent ones. In this study, we explored seven intrinsic features directly derived from the material's structure, with a specific focus on the chemical environment of active sites and their nearest neighbors. An accurate and efficient ML model to predict potential bifunctional oxygen electrocatalysts based on the intrinsic features of AB-type alloy active sites and intermediate free energies in the OERs/ORRs was established. These features possess clear physical and chemical meanings, closely linked to the electronic and geometric structures of active sites and neighboring atoms, thereby providing indispensable insights for the discovery of high-performance electrocatalysts. The ML model achieved R2 scores of 0.827, 0.913, and 0.711 for the predicted values of the three intermediate (OH, O, OOH) free energies, with corresponding mean absolute errors of 0.175, 0.242, and 0.200 eV, respectively. These results indicate that the ML model exhibits high accuracy in predicting the intermediate free energies. Furthermore, the ML model exhibited a prediction efficiency 150,000 times faster than traditional density functional theory calculations. This work will offer valuable insights and a framework for facilitating the rapid design of potential catalysts by ML methods.

Electrical-gate-controlled giant tunneling magnetoresistance and its quasi-periodic oscillation in an interlaced magnetic-electric silicene superlattice.

In this work, we propose a silicene-based lateral resonant tunneling device by placing silicene under the superlattices interlaced, arranged by ferromagnetic gates and electric gates. Its ballistic transport properties are calculated by the transfer matrix method. Combined with the unique electrically tuned energy gap of silicene, its magnetoresistance (MR) can be exaggeratedly modulated over a wide range by applying electrostatic potential and the on-site potential difference. It is interestingly found that there is a quasi-periodic oscillation of the MR in silicene-based superlattice devices from the quantum resonant confinement of the band splitting by the electrostatic field. Moreover, the peak of the MR in a single-period structure can reach more than 104, while the peak of the MR in an interlaced alternating magnetic-electric silicene superlattice can reach more than 1017, which is one of the best-reported values. This may originate from the enhancement effect of the wave vector filtering by the controlled field. Our studies indicate that the silicene superlattices alternately arranged by the ferromagnetic gate and electric gate not only have giant MR (GMR) properties, but also exhibit the periodic oscillation characteristics of MR in which electric gates can be modulated. Therefore, this work provides a more flexible strategy for the construction of silicene-based nanoelectronic devices.

Stack-dependent ion diffusion behavior in two-dimensional bilayer C3B.

In recent years, two-dimensional (2D) C-based materials have been intensively studied due to their excellent physicochemical properties. Meanwhile, extensive research has revealed that the electrical properties of layered materials can be tuned by changing the stacking pattern. However, the tuning of ion diffusion properties through stacking remains to be explored. In this work, bilayer C3B with different stackings as a lithium-ion battery anode material is systematically investigated by first-principles calculations. The calculated results show that bilayer C3B has better electronic properties (with a band gap of 0.44 eV to 0.54 eV) and enhanced bonding strength of Li (-2.82 to -3.27 eV) compared to monolayer C3B. Moreover, the intralayer migration barrier of Li can be regulated by stacking. Interestingly, the AB stacked configuration has the lowest migration barrier of 0.100 eV, which is significantly lower than those of other stacking configurations and monolayer C3B. Further studies revealed that the formation of fast ion diffusion channels in the AB stacked configuration is due to the combined effect of layer distance and in-plane charge transfer. These results offer a new strategy for the regulation of ion diffusion properties in 2D van der Waals materials.

GaOx@GaN Nanowire Arrays on Flexible Graphite Paper with Tunable Persistent Photoconductivity.

Flexible optoelectronic synaptic devices that functionally imitate the neural behavior with tunable optoelectronic characteristics are crucial to the development of advanced bioinspired neural networks. In this work, amorphous oxide-decorated GaN nanowire arrays (GaOx@GaN NWAs) are prepared on flexible graphite paper. A GaOx@GaN NWA-based flexible device has tunable persistent photoconductivity (PPC) and shows a conversible fast/slow decay process (SDP). Photoconductivity can be modulated by single or double light pulses with different illumination powers and biases. PPC gives rise to the high-performance SDP such as a long decay time of 2.3 × 105 s. The modulation mechanism is proposed and discussed. Our results reveal an innovative and efficient strategy to produce decorated NWAs on a flexible substrate with tunable optoelectronic properties and exhibit potential for flexible neuromorphic system applications.

Fault Diagnosis of Motor Bearings Based on a One-Dimensional Fusion Neural Network.

Deep learning has been an important topic in fault diagnosis of motor bearings, which can avoid the need for extensive domain expertise and cumbersome artificial feature extraction. However, existing neural networks have low fault recognition rates and low adaptability under variable load conditions. In order to solve these problems, we propose a one-dimensional fusion neural network (OFNN), which combines Adaptive one-dimensional Convolution Neural Networks with Wide Kernel (ACNN-W) and Dempster-Shafer (D-S) evidence theory. Firstly, the original vibration time-domain signals of a motor bearing acquired by two sensors are resampled. Then, four frameworks of ACNN-W optimized by RMSprop are utilized to learn features adaptively and pre-classify them with Softmax classifiers. Finally, the D-S evidence theory is used to comprehensively determine the class vector output by the Softmax classifiers to achieve fault detection of the bearing. The proposed method adapts to different load conditions by incorporating complementary or conflicting evidences from different sensors through experiments on the Case Western Reserve University (CWRU) motor bearing database. Experimental results show that the proposed method can effectively enhance the cross-domain adaptive ability of the model and has a better diagnostic accuracy than other existing experimental methods.

Large-Area Synthesis of Layered HfS2(1- x )Se2 x Alloys with Fully Tunable Chemical Compositions and Bandgaps.

Alloying transition metal dichalcogenides (TMDs) with different compositions is demonstrated as an effective way to acquire 2D semiconductors with widely tunable bandgaps. Herein, for the first time, the large-area synthesis of layered HfS2(1- x )Se2 x alloys with fully tunable chemical compositions on sapphire by chemical vapor deposition is reported, greatly expanding and enriching the family of 2D TMDs semiconductors. The configuration and high quality of their crystal structure are confirmed by various characterization techniques, and the bandgap of these alloys can be continually modulated from 2.64 to 1.94 eV with composition variations. Furthermore, prototype HfS2(1- x )Se2 x photodetectors with different Se compositions are fabricated, and the HfSe2 photodetector manifests the best performance among all the tested HfS2(1- x )Se2 x devices. Remarkably, by introducing a hexagonal boron nitride layer, the performance of the HfSe2 photodetector is greatly improved, exhibiting a high on/off ratio exceeding 105, an ultrafast response time of about 190 µs, and a high detectivity of 1012 Jones. This simple and controllable approach opens up a new way to produce high-quality 2D HfS2(1- x )Se2 x layers, which are highly qualified candidates for the next-generation application in high-performance optoelectronics.

Temperature-Sensitive Structure Evolution of Lithium-Manganese-Rich Layered Oxides for Lithium-Ion Batteries.

Cathodes of lithium-rich layered oxides for high-energy Li-ion batteries in electrically powered vehicles are attracting considerable attention by the research community. However, current research is insufficient to account for their complex reaction mechanism and application. Here, the structural evolution of lithium-manganese-rich layered oxides at different temperatures during electrochemical cycling has been investigated thoroughly, and their structural stability has been designed. The results indicated structure conversion from the two structures into a core-shell structure with a single distorted-monoclinic LiTMO2 structure core and disordered-spinel/rock salt structure shell, along with lattice oxygen extraction and lattice densification, transition- metal migration, and aggregation on the crystal surface. The structural conversion behavior was found to be seriously temperature sensitive, accelerated with higher temperature, and can be effectively adjusted by structural design. This study clarifies the structural evolution mechanism of these lithium-rich layered oxides and opens the door to the design of similar high-energy materials with better cycle stability.

An atomistic mechanism study of GaN step-flow growth in vicinal m-plane orientations.

Elucidation of homoepitaxial growth mechanisms on vicinal non-polar surfaces of GaN is highly important for gaining an understanding of and control thin film surface morphology and properties. Using first-principles calculations, we study the step-flow growth in m-plane GaN based on atomic row nucleation and kink propagation kinetics. Ga-N dimer adsorption onto the m-plane is energetically more favorable than that of Ga and N isolated adatoms. Therefore, we have treated the dimers as the dominant growth species attached to the step edges. By calculating the free energies of sequentially attached Ga-N dimers, we have elucidated that the a-step edge kink growth proceeds by parallel attachment rather than by across the step edge approach. We found a series of favorable configurations of kink propagation and calculated the free energy and nucleation barriers for kink evolution on five types of step edges (a, +c, -c, +a + c, and -a - c). By changing the chemical potential µGa and the excess chemical potential Δµ, the growth velocities at the five types of edges are controlled by the corresponding kink pair nucleation barrier E* in their free energy profiles. To explore the kink-flow growth instability observed at different Ga/N flux ratios, calculations of kink pairs on the incompact -c and +c-step edges are further performed to study their formation energies. Variations of these step edge morphologies with a tuned chemical environment are consistent with previous experimental observations, including stable diagonal ±a ± c-direction steps. Our work provides a first-principles approach to explore step growth and surface morphology of the vicinal m-plane GaN, which is applicable to analyze and control the step-flow growth of other binary thin films.

Elastic Properties, Defect Thermodynamics, Electrochemical Window, Phase Stability, and Li(+) Mobility of Li3PS4: Insights from First-Principles Calculations.

The improved ionic conductivity (1.64 × 10(-4) S cm(-1) at room temperature) and excellent electrochemical stability of nanoporous ß-Li3PS4 make it one of the promising candidates for rechargeable all-solid-state lithium-ion battery electrolytes. Here, elastic properties, defect thermodynamics, phase diagram, and Li(+) migration mechanism of Li3PS4 (both γ and ß phases) are examined via the first-principles calculations. Results indicate that both γ- and ß-Li3PS4 phases are ductile while γ-Li3PS4 is harder under volume change and shear stress than ß-Li3PS4. The electrochemical window of Li3PS4 ranges from 0.6 to 3.7 V, and thus the experimentally excellent stability (>5 V) is proposed due to the passivation phenomenon. The dominant diffusion carrier type in Li3PS4 is identified over its electrochemical window. In γ-Li3PS4 the direct-hopping of Lii(+) along the [001] is energetically more favorable than other diffusion processes, whereas in ß-Li3PS4 the knock-off diffusion of Lii(+) along the [010] has the lowest migration barrier. The ionic conductivity is evaluated from the concentration and the mobility calculations using the Nernst-Einstein relationship and compared with the available experimental results. According to our calculated results, the Li(+) prefers to transport along the [010] direction. It is suggested that the enhanced ionic conductivity in nanostructured ß-Li3PS4 is due to the larger possibility of contiguous (010) planes provided by larger nanoporous ß-Li3PS4 particles. By a series of motivated and closely linked calculations, we try to provide a portable method, by which researchers could gain insights into the physicochemical properties of solid electrolyte.

Defects in AIN/GaN Superlattice: First Principle Calculations.

In this paper we investigate the atomic configurations, electronic structure and formation energies of native point defects, (such as vacancies and self-interstitials), in an AIN/GaN superlattice (SL) constructed on a wurtzite structure along a [0001] growth direction. Comprehensive first-principle calculations based on the density functional theory (DFT) are used. Cation and anion vacancies in the neutral charge state are calculated. For the native defects, the results showed that the most favorable configurations are the cation vacancies at the interface of the SL, or the anion vacancies in the GaN wells. Considering the formation energies of different vacancies, the results show that the nitrogen vacancy has the lowest formation energy, indicating that they are significantly the most stable configuration, and thus should be expected to be the major defect in a AIN/GaN superlattice.

Synthesis of Tetracyclic Quinazolinones Using a Visible-Light-Promoted Radical Cascade Approach.

A practical approach for the synthesis of tetracyclic pyrroloquinazolines using photoredox strategy has been developed. The visible-light-promoted intramolecular single-electron-transfer process between photocatalyst and N-(2-iodobenzyl)-N-acylcyanamides is considered to be involved in this transformation. Targeted pyrroloquinazoline derivatives (15 examples) are presented in good isolated yields (30%-88%).

Engineering of hydrogenated two-dimensional h-BN/C superlattices as electrostatic substrates.

Hybridized two-dimensional materials incorporating domains from the hexagonal boron nitride (h-BN) and graphene is an interesting branch of materials science due to their highly tunable electronic properties. In the present study, we investigate the hydrogenated two-dimensional (2D) h-BN/C superlattices (SLs) with zigzag edges using first-principles calculations. We found that the domain width, the phase ratio, and the vertical dipole orientation all have significant influence on the stability of SLs. The electronic reconstruction is associated with the lateral polar discontinuities at the zigzag edges and the vertically polarized (B2N2H4)(m) domains, which modifies the electronic structures and the spatial potential of the SLs significantly. Furthermore, we demonstrate that the hydrogenated 2D h-BN/C SLs can be applied in engineering the electronic structure of graphene: laterally-varying doping can be achieved by taking advantage of the spatial variation of the surface potential of the SLs. By applying an external vertical electric field on these novel bidirectional heterostructures, graphene doping levels and band offsets can be tuned to a wide range, such that the graphene doping profile can be switched from the bipolar (p-n junction) to unipolar (n(+)-n junction) mode. It is expected that such bidirectional heterostructures provide an effective approach for developing novel nanoscale electronic devices and improving our understanding of the fundamentals of low-dimensional materials.

A low cost, green method to synthesize GaN nanowires.

The synthesis of gallium nitride nanowires (GaN NWs) by plasma enhanced chemical vapor deposition (PECVD) are successfully demonstrated in this work. The simple and green synthesis route is to introduce gallium oxide (Ga2O3) and nitrogen (N2) for the growth of nanowires. The prepared GaN nanowires have a single crystalline wurtzite structure, which the length of some nanowires is up to 20 µm, with a maximum diameter about 140 nm. The morphology and quantity of the nanowires can be modulated by the growth substrate and process parameters. In addition, the photoluminescence and field emission properties of the prepared GaN nanowires have been investigated, which were found to be largely affected by their structures. This work renders an environmentally benign strategy and a facile approach for controllable structures on nanodevice.

Bipolar doping of double-layer graphene vertical heterostructures with hydrogenated boron nitride.

Using first-principles calculations, we examined the bipolar doping of double-layer graphene vertical heterostructures, which are constructed by hydrogenated boron nitride (BN) sheets sandwiched into two parallel graphene monolayers. The built-in potential difference in hydrogenated BN breaks the interlayer symmetry, resulting in the p- and n-type doping of two graphene layers at 0.83 and -0.8 eV, respectively. By tuning the interlayer spacing between the graphene and hydrogenated BN, the interfacial dipole and screening charge distribution can be significantly affected, which produces large modulations in band alignments, doping levels and tunnel barriers. Furthermore, we present an analytical model to predicate the doping level as a function of the average interlayer spacing. With large interlayer spacings, the "pillow effect" (Pauli repulsion at the highly charge overlapped interface) is diminished and the calculated Dirac point shifts are in good accordance with our prediction models. Our investigations suggest that this double-layer graphene heterostructures constructed using two-dimensional Janus anisotropic materials offer exciting opportunities for developing novel nanoscale optoelectronic and electronic devices.

Synthesis of isoquinolines via visible light-promoted insertion of vinyl isocyanides with diaryliodonium salts.

A synthetic strategy for multi-substituted isoquinoline derivatives has been developed using visible light-promoted vinyl isocyanide insertion with diaryliodonium salts at room temperature. The methodology presented here represents the first example of isoquinoline synthesis via somophilic isocyanide insertion.

Two dimensional Dirac carbon allotropes from graphene.

Using a structural search method in combination with first-principles calculations, we found lots of low energy 2D carbon allotropes and examined all possible Dirac points around their Fermi levels. Three amazing 2D Dirac carbon allotropes have been discovered, named as S-graphene, D-graphene and E-graphene. By analyzing the topology correlations among S-, T, net W graphene and graphene, we found that a general rule is valuable for constructing 2D carbon allotropes that are keen to possess Dirac cones in their electronic structures. Based on this rule, we have successfully designed many new 2D carbon allotropes possessing Dirac cones. Their energy order can be well described by an Ising-like model, and some allotropes are energetically more stable than those recently reported. The related electronic structures of these Dirac allotropes are anisotropy distinguished from those of graphene. Moreover, the fact that D- and E-graphene present Dirac cones suggests that sp hybridization or sp(3) hybridization could not suppress the emerging of Dirac features. Our results demonstrate that the Dirac cone and carrier linear dispersion is a very common feature in 2D carbon allotropes and can exist beyond the limitations of fundamental structure features of graphene.

Photodegradable supramolecular hydrogels with fluorescence turn-on reporter for photomodulation of cellular microenvironments.

Photodegradable hydrogels that allow 3D encapsulation of cells are important biomaterials to modulate cellular microenvironments with temporal and spatial resolution. Herein we report a photodegradable hydrogel formed by the self-assembly of short peptides modified with a novel phototrigger. The phototrigger is a biaryl-substituted tetrazole moiety that, upon mild light irradiation, undergoes rapid intramolecular photoclick ligation to form a highly fluorescent pyrazoline moiety. Short peptides linked with a tetrazole-containing moiety, Tet(I) or Tet(II), are able to self-assemble into hydrogels, among which the Tet(I)-GFF and Tet(II)-GFRGD gels show good mechanical strength and biocompatibility for 3D encapsulation and prolonged culture of live cells. The phototriggered tetrazole-to-pyrazoline transformation generates a highly fluorescent reporter and induces the disassembly of the hydrogel matrix by disturbing the balance between hydrophilic interaction and π-π stacking of the self-assembled system. Photomodulation of cellular microenvironments was demonstrated not only for the cells grown on top of the gel but also for stem cells encapsulated inside the hydrogels.

A photoelectrochemical investigation on the synergetic effect between CdS and reduced graphene oxide for solar-energy conversion.

CdS modified with reduced graphene oxide (RGO) has been widely demonstrated to be effective in the field of solar-energy conversion. However, the inherent mechanism of this superior property is still not thoroughly understood. Thus the photoelectrochemical method was employed to systemically investigate the synergetic effect between CdS and RGO. The result shows that the photoelectrochemical properties of RGO/CdS samples are sensitive to the relative ratio of RGO to CdS, and the photoelectrode with 1.0 wt% ratio of RGO possesses the best photoelectrochemical performance. Further investigation demonstrates that the synergetic effect between CdS and RGO directly influences the charge-transport property and band-structure of the composite, which is also supported by the X-ray photoelectron spectroscopy data and first-principle simulation, respectively.

Photocycloadditions of substituted oxazoles with isoquinoline-1,3,4-trione--chemo-, regio-, diastereoselectivities and transformation of the photocycloadducts.

Photoreactions of isoquinoline-1,3,4-triones and oxazoles with different substituents were found to give different chemo-, regio- and diastereoselectivities. The substituent at the C5 on the oxazole ring showed great influence on the chemoselectivity of the photoreaction as well as on the transformation of the photocycloadducts. The 2-methyl-5-methoxyoxazoles reacted with isoquinoline-1,3,4-triones rapidly and gave spirooxetanes with high regio- and diastereo-selectivity. Diastereoselectivity in the reaction of 2-phenyl-5-methoxyoxazoles with isoquinoline-1,3,4-triones was relevant to the substituent on the 4-position on the oxazole ring. Replacement of the 5-methoxy group with 5-methyl or 5-phenyl resulted in significant decrease on the reactivity of the oxazole as well as change on the diastereoselectivity in photocycloaddition with isoquinoline-1,3,4-triones. Acid-mediated transformations of the photocycloadduct spirooxetanes was found to give different type of products including ß-hydroxy-α-aminocarbonyl compounds and spiroisoquinolineoxazolines under different reaction conditions. Substituents on the spirooxetanes as well as the type and amount of acid used in the reaction played important roles in determining the type and diastereoselectivity of the products in the transformations.