Search

Herein, we report surface coarsened titanium dioxide (TiO2) nanobelts with anatase/rutile heterophase junctions via a facile hydrothermal/calcination method for simultaneous hydrogen (H2) and oxygen (O2) productions from pure water, with excellent production rates of 0.614 and 0.297 mmol h−1 g−1 with platinum (Pt)/cobalt phosphide (CoP) as cocatalysts, respectively. Besides, the TiO2 nanobelts-900 °C with anatase/rutile heterophase junctions show a notable improvement in photocatalytic H2 and O2 production than pure anatase TiO2 nanobelts (TiO2 nanobelts-600 °C, 700 °C and 800 °C) and pure rutile TiO2 nanobelts (TiO2 nanobelts-1000 °C). The anatase/rutile heterophase junctions could effectively stimulate the transfer of electrons from rutile to anatase and then to Pt, and H2 generation on the surface of Pt. In the meantime, the holes can be transferred from anatase to rutile and then to CoP, and water oxidation on CoP’s surface.

The graphitic carbon nitride (g-C3N4) hollow nanotubes synthesized via a simple freeze-drying method are used for constructing Two-dimensional (2D)-one-dimensional (1D) molybdenum sulfide (MoS2) nanoflake/g-C3N4 hollow nanotube (MoS2/g-C3N4 nanotube) photocatalysts. The MoS2/g-C3N4 nanotube composite with 15 wt% MoS2 shows the highest hydrogen (H2) production rate (1124 μmol·h−1·g−1), much higher than bulk g-C3N4 (64 μmol·h−1·g−1) and g-C3N4 nanotubes (189 μmol·h−1·g−1). The excellent photocatalytic activity of MoS2/g-C3N4 nanotube composites can be ascribed to more exposed active edges of 2D-1D structure, multiple light reflection/scattering channels of 2D nanoflake/1D hollow nanotube composite structure and better carrier transfer and separation by heterojunction interface.

Metasurfaces have been verified as an ideal way to control electromagnetic waves within an optically thin interface. In this paper, a design method of a tunable metasurface integrated with vanadium dioxide (VO2) is proposed to realize independent control of geometric and propagation phase modulation. The reversible conversion of VO2 between insulator phase and metal phase can be realized by controlling the ambient temperature, which enables the metasurface to be switched quickly between split-ring and double-ring structures. The phase characteristics of 2-bit coding units and the electromagnetic scattering characteristics of arrays composed of different arrangements are analyzed in detail, which confirms the independence of geometric and propagation phase modulation in the tunable metasurface. The experimental results demonstrate that the fabricated regular array and random array samples have different broadband low reflection frequency bands before and after the phase transition of VO2, and the 10 dB reflectivity reduction bands can be switched quickly between C/X and Ku bands, which are in good agreement with the numerical simulation. This method realizes the switching function of metasurface modulation mode by controlling the ambient temperature, which provides a flexible and feasible idea for the design and fabrication of stealth metasurfaces.

Light harvesting and charge separation are both significant in the photocatalysis, but it is challenging to synchronously realize both in a single-component material. The novel porous TiO2 nanoflowers (NFs) photocatalysts with stable bronze (TiO2(B))/anatase heterophase junctions and large pore sizes are prepared via a hydrothermal/annealing method. The presence of porous nanoflower structure enhances the light absorption through reflection/refraction of light. The stable TiO2(B)/anatase heterophase junctions can efficiently promote the separation of photoinduced electrons and holes pairs and therefore suppress the charge recombination. The large pore sizes provide multi-level channels for the absorption and diffusion of reactants. With the increase of annealing temperatures from 350 to 550 °C, the H2 evolution activity is promoted. However, overhigh annealing temperature (650 °C) cause the broken of nanoflower structure and TiO2(B)/anatase heterophase junctions, thus inducing even decrease of H2 evolution activity. As a consequence, the obtained TiO2 NFs exhibit an enhanced photocatalytic H2 evolution activity at the optimal annealing temperature (550 °C) with Pt as co-catalysts (5.013 mmol h−1g−1), exceeding that of TiO2 NFs without annealing (0 mmol h−1g−1) and pure anatase TiO2 NFs (TiO2 NFs-650 °C, 4.722 mmol h−1g−1), respectively. Interestingly, TiO2 NFs-550 °C still show a high hydrogen evolution rate of 4.317 mmol h−1g−1 in the absence of co-catalysts.

A metal-organic frameworks-derived sulfuration and phosphorization method is developed for constructing a hydrogen evolution reaction (HER) electrocatalyst (MoWOSP@C), which is composed of MoxW1-xO2 nanoparticles, MoxW1-xS2 nanosheets, and MoxW1-xP nanoparticles coated with a porous hollow carbon matrix. The three-dimensional MoWOSP@C nano-octahedron shows excellent catalytic activity with a low overpotential of −118 mV at a current density of −10 mA cm−2 and a small Tafel slope of 74.1 mV dec−1, which are all higher than those of the single phase components. The outstanding performance of the MoWOSP@C is due to the highly exposed active sites and synergistic effects of the MoxW1-xS2 and MoxW1-xP, carbon conductive matrix and mesoporous hollow structures.

Day-ahead load forecasting is the key part in day-ahead scheduling of power system. Considering the uncertainty of load forecasting can improve the robustness of the system and reduce the risk cost. This paper proposes a distributed robust optimization (DRO) method for units dispatch with Monte Carlo (MC) dropout and Wasserstein ambiguity set. The model uncertainty is obtained by introduce MC dropout into deep learning model. The ambiguity set is obtained by taking the prediction results as the prior data of Wasserstein ambiguity set method. Finally, the optimal dispatch problem is solved by the Robust Stochastic Optimization Made Easy (RSOME) toolbox in MATLAB.

Deep learning plays an important role in short-term load forecasting. By combining deep learning model with expert knowledge, the accuracy and interpretation of load forecasting model can be further improved. Under the guidance of load forecasting theory, the effect of holidays on daily load was taken into account. This paper proposes a new convolutional LSTM (ConvLSTM) deep learning model with taking advantage of expert knowledge. Simulation results show that the proposed ConvLSTM network captures spatiotemporal correlations better and consistently outperforms LSTM for load forecasting, especially in holidays load forecasting.

An efficient and flexible method using femtosecond laser bursts assisted by wet etching is presented to fabricate large-area high-quality microlens arrays (MLAs) on a silica glass surface. In this method, femtosecond laser bursts can ablate micro craters on silica glass in a fast, single-step process by controlling the electron density and a high-speed scanning galvanometer, and the influence mechanism of the number of pulses within a burst on the accuracy and quality of micro craters is analyzed in detail. The experimental results show that the preparation efficiency of micro craters is significantly improved to approximately 32,700 per second. By subsequent acid etching, concave microlenses with controllable dimensions, shapes, and alignments are easily obtained. A large area close-packed hexagonal concave MLA is successfully fabricated by using this method and shows high surface quality and uniformity, which excellently demonstrates the feasibility and flexibility of rapidly fabricating MLAs in the burst regime.

The biggest challenging issue in photocatalytic hydrogen production is to efficiently separate the photoinduced electron–hole pairs, which requires the enrichment of photoinduced electrons on the photocatalyst's surface. Herein, TiO2 nanosheets (NSs) are in situ grown on highly conductive Ti3C2 MXene and then 1T-MoS2 nanopatches are uniformly distributed on the Ti3C2/TiO2 composite through a two-step hydrothermal method. Thus, a unique 2D structure of the Ti3C2/TiO2/1T-MoS2 composite with double metallic co-catalysts Ti3C2 and 1T phase MoS2 nanopatches is achieved, and the content of 1T phase reaches 84%. The photocatalytic H2 production rate of the Ti3C2/TiO2/1T-MoS2 composite with an optimized MoS2 loading amount (15 wt%) is nearly 132, 11, and 1.5 times higher than that of pure TiO2 NSs, Ti3C2/TiO2 composites and Ti3C2/TiO2/2H-MoS2 composites (15 wt%). The 1T-MoS2 nanopatches on the surface of the Ti3C2/TiO2 composites increase the specific surface area and boost the density of active sites. Besides, the presence of Ti3C2 MXene and 1T phase MoS2 is found to enhance the electronic conductivity, resulting in an increase in efficiency for electron transfer.

Herein, we report surface coarsened titanium dioxide (TiO

The graphitic carbon nitride (g-C

Herein, we report that 2D metallic 1T-WS2 acts as co-catalyst in assisting g-C3N4 nanotubes (NTs) obtained via a simple grinding method for promoting photocatalytic H2 production. The 1T-WS2/g-C3N4 composite with optimum 27% 1T-WS2 displays a significantly improved photocatalytic H2 production rate (1021 μmol h−1 g−1), 17.6 times higher than g-C3N4 NTs. Besides, the apparent quantum efficiency (AQE) of 1T-WS2/g-C3N4 composite (27%) achieves 11.23% under light at λ = 370 nm. Meanwhile, the 1T-WS2/g-C3N4 composites present an excellent photocatalytic H2 production stability. The possible photocatalytic mechanism over 1T-WS2/g-C3N4 composites is proposed. Specifically, the photogenerated electrons from 1D g-C3N4 tubular nanostructure with open mesoporous morphology are migrated to conductive 1T-WS2 NSs quickly as electron acceptors along the 1D path. 1T-WS2 NSs can also offer abundant active sites on the edge and basal planes.

A metal-organic frameworks-derived sulfuration and phosphorization method is developed for constructing a hydrogen evolution reaction (HER) electrocatalyst (MoWOSP@C), which is composed of Mo

Adhesion and delamination have been pervasive problems hampering the performance and reliability of micro-and nano-electronic devices. In order to understand, predict, and ultimately prevent interface failure in electronic devices, development of accurate, robust, and efficient delamination testing and prediction methods is crucial. Adhesion is essentially a multi-scale phenomenon: at the smallest scale possible, it is defined by the thermodynamic work of adhesion. At larger scales, additional dissipative mechanisms may be active which results in enhanced adhesion at the macroscopic scale and are the main cause for the mode angle dependency of the interface toughness. Undoubtedly, the macroscopic adhesion properties are a complex function of all dissipation mechanisms across the scales. Thorough understanding of the significance of each of these dissipative mechanisms is of utmost importance in order to establish physically correct, unambiguous, values of the adhesion properties, which can only be achieved by proper multi-scale techniques. The topic “Advances in Delamination Modeling” has been split into two separate chapters: this chapter discusses the atomistic aspects of delamination, while the preceding chapter deals with the atomistic aspects of interface separation. The chapter starts with a concise overview of molecular simulation strategies. Next, examples are provided which represent actual materials being developed for electronic packaging: (1) the prediction of thermomechanical properties of an epoxy molding compound (EMC) and the adhesion properties of an EMC/copper interface by means of MD and CG MD approaches; (2) the modeling of wetting, adhesion, and reliability cycling of die attach and via fills; (3) model scaling to discrete element modeling (DEM) for understanding underfill flow; (4) CG modeling of an epoxy molding compound which relates to the first example; (5) molecular modeling of silicate layers used in planarization and encapsulant layers for flat panel displays; (6) mesoscale modeling of diffusion of organic bases which is of concern to photoresist poisoning; and (7) the prediction of thermomechanical properties of a low-k dielectric material, SiOC:H.

We present a coarse-grained model for molecular dynamics simulations of an epoxy system composed of epoxy phenol novolac as epoxy monomer and bisphenol-A as the cross-linking agent. The epoxy and hardener molecules are represented as short chains of connected beads, and cross-linking is accomplished by introducing bonds between reactive beads. The interbead potential, composed of Lennard-Jones, bond stretching, and angle bending terms, is parametrized through an optimization process based on a particle swarm optimization method to fit certain key thermomechanical properties of the material obtained from experiments and previous full atomistic simulations. The newly developed coarse-grained model is capable of predicting a number of thermomechanical properties of the epoxy system. The predictions are in very good agreement with available data in the literature. More importantly, our coarse-grained model is capable of predicting tensile failure of the epoxy system, a capability that no other conventional molecular dynamic simulation model has. Finally, our coarse-grained model can speed up the simulations by more than an order of magnitude when compared with traditional molecular dynamic simulations.

Presented in this paper is a numerical study based on classical molecular dynamics simulation to understand the deformation and failure behavior of an epoxy/copper bimaterial under pure tension normal to the interface. The epoxy considered is a highly cross-linked epoxy phenol novolac, and the copper substrate is a standard face-center-cubic single crystal with its (1,1,1) surface as the epoxy/copper interface. Stress versus displacement/strain curves are obtained to understand the bimaterial behavior and to predict the epoxy/copper interfacial tensile strength. It is found that the interfacial failure is brittle caused by simultaneous detachment of epoxy atoms from the copper substrate, and the interfacial tensile strength is almost unaffected by the unloading and reloading before the failure strength is reached. Effects of temperature, epoxy cross-link density, and epoxy functionality are also investigated. Findings of this study provide significant insights into the deformation and failure behavior mechanisms of the epoxy/copper bimaterial interface.

This paper reports the use of molecular dynamics simulations to study the thermomechanical properties of an epoxy molding compound formed by curing tri/tetra-functionalized EPN1180 with Bisphenol-A. An interactive crosslinking-relaxation methodology is developed to construct the simulation cell. This crosslinking-relaxation methodology allows the construction of highly crosslinked polymer network from a given set of monomers. Based on this computational algorithm, three-dimensional simulation cells can be constructed. By using an existing polymer consistent force-field, several thermomechanical properties of the model epoxy are computed such as the curing induced shrinkage, gelation point, coefficient of thermal expansion, glass transition temperature, Young's modulus and Poisson's ratio. The dependence of these properties on crosslink density and temperature is also investigated. Simulated results are compared with existing theoretical or experimentally measured values when available. Good agreements are observed.

We derive a method to determine the effective local elasticity tensor by combining the local stress, local strain, and global strain in conjunction with a linear least-squares solver. This method reduces to the standard stress-strain fluctuation method for the estimation of global moduli if the local stress and the local strain are substituted by global counterparts. The quality of the developed method is verified by the analysis of an fcc Lennard-Jones single crystal. By using this method, surface elastic behaviors of three nanoplates are investigated. Simulation results prove that both softening and stiffening effects could be detected, depending on the surface orientations and loading directions. This approach could be especially valuable for complicated morphologies where the physical properties of the local material are challenging or impractical to obtain.

Using a previously developed coarse-grained model, we conducted large-scale (∼ 85 × 85 × 85 nm(3)) molecular dynamics simulations of uniaxial-strain deformation to study the tensile behavior of an epoxy molding compound, epoxy phenol novolacs (EPN) bisphenol A (BPA). Under the uniaxial-strain deformation, the material is found to exhibit cavity nucleation and growth, followed by stretching of the ligaments separated by the cavities, until the ultimate failure through ligament scissions. The nucleation sites of cavities are rather random and the subsequent cavity growth accounts for much (87%) of the volumetric change during the uniaxial-strain deformation. Ultimate failure of the materials occurs when the cavity volume fraction reaches ∼ 60%. During the entire deformation process, polymer strands in the network are continuously extended to their linear states and broken in the postyielding strain hardening stage. When most of the strands are stretched to their taut configurations, rapid scission of a large number of strands occurs within a small strain increment, which eventually leads to fracture. Finally, through extensive numerical simulations of various loading conditions in addition to uniaxial strain, we find that yielding of the EPN-BPA can be described by the pressure-modified von Mises yield criterion.

In this study, a coarse-grained model is developed to describe the interatomic interactions between a cross-linked epoxy and a copper substrate. Based on this model, the tensile deformation and failure of an epoxy/Cu bimaterial is studied. Attention is given to the microstructural evolution near the epoxy/Cu interface, and its effects on the overall stress–strain behavior of the bimaterial. It is found that, under uniaxial strain, plastic deformation in the epoxy/Cu bimaterial is localized to a thin layer of epoxy next to the epoxy/Cu interface. This discovery enables the definition of an interfacial zone. Consequently, a cohesive zone model at the continuum level could potentially be constructed where the separation of the interface is given by the stretching of the interfacial zone computed from the coarse grain simulations.

Epoxy molding compounds (EMC) have been widely used in electronic packaging industry as adhesives, for example, die-attach and underfill. However, interfaces between EMC and other components such as copper substrates are weak and prone to delamination, due to large interfacial stresses developed during thermal cycling processes. Cohesive zone finite element method is able to simulate fracture initiation and propagation along a prescribed path, thus is an ideal choice of interface performance evaluation. However, this continuum mechanics based method relies on the determination of a traction-separation law to define cohesive elements. To this end, molecular dynamics (MD) simulation is conducted in this work to extract traction-separation law from a fully-atomistic model for epoxy/Cu interface at finite temperature. The epoxy studied in this work is a novel epoxy molding compound synthesized by curing tri/tetra-functionalized EPN1180 with Bisphenol-A. A fully atomistic model for epoxy reflecting its network nature was created by applying a cross-linking algorithm to a confined layer with 2D periodic boundary condition assigned, containing a physical mixture of monomers. The interface model was built by laying epoxy slab with 2D crosslinked network structure on Cu substrate, and further optimized by energy minimization and MD simulations. MD simulation of tensile deformation was then conducted to extract traction-separation law for epoxy/Cu interfaces. Failure of the interface model was found to initiate within epoxy, localized within an interfacial zone, but transit to complete interface separation in the end.

A novel Epoxy Molding Compound (EMC) with a crosslinked network structure was formed by curing tri-/tetra-functionalized EPN1180 with Bisphenol-A. A full atomistic model reflecting the network nature of the material was constructed by applying an iterative crosslinking algorithm to an amorphous cell with 3D periodic boundary condition containing the stoichiometric mixture of constitutive monomers. The geometry of the model was then optimized using the COMPASS force-field in Materials Studio [1]. The variation of system density and volume against temperature was simulated using a cooling down profile, which was employed to derive the glass transition temperature and coefficient of thermal expansion of the system. Furthermore, the Young's modulus and Poisson's ratio were calculated by uni-axial tensile molecular statics simulations. The material properties computed by molecular dynamics/mechanics simulations were in good agreement with experiment measurements. An epoxy resin/copper interface model was constructed and the interfacial adhesion energy was calculated as the energy difference between the total energy of the entire system and the sum of the energies of individual materials. The traction-displacement law of the interface was derived when the system was subjected to a molecular statics uniaxial tension. The work of separation and the peak traction, considered as the two key parameters required by cohesive zone finite element simulation, were extracted from the traction-displacement law.

The strength of the stress singularities at the apex of a sharp V-notch for plane problems are investigated analytically in this paper. Three sets of boundary conditions, namely, free-free, fixed-fixed, and fixed-free, along the two faces of the notch are considered. The material is assumed to be governed by the deformation theory of plasticity with linear hardening characteristic. Results for both plane stress and plane strain conditions are presented.

The singular elastic-plastic stress, strain and the displacement fields at the tip of a sharp notch for both plane stress and plane strain conditions are investigated analytically. The material is assumed to be governed by the deformation theory of plasticity with linear strain hardening characteristic. Since the elastic strain is retained in the analysis, the final strain and displacement fields can be separated into the elastic and the plastic parts. In the case with zero notch angle, the results reduce to the classical crack problem. The relationship of the amplitude of the near crack tip elastic-plastic field to the elastic far field is obtained. Both mode I and mode II cases are investigated. The mixed mode case is also discussed.