Aligned Network Research Articles

Centrifugal Inertia-Induced Directional Alignment of AgNW Network for Preparing Transparent Electromagnetic Interference Shielding Films with Joule Heating Ability.

Transparent electromagnetic interference (EMI) shielding is highly desired in specific visual scenes, but the challenge remains in balancing their EMI shielding effectiveness (SE) and optical transmittance. Herein, this study proposed a directionally aligned silver nanowire (AgNW) network construction strategy to address the requirement of high EMI SE and satisfactory light transmittance using a rotation spraying technique. The orientation distribution of AgNW is induced by centrifugal inertia force generated by a high-speed rotating roller, which overcomes the issue of high contact resistance in random networks and achieves high conductivity even at low AgNW network density. Thus, the obtained transparent conductive film achieved a high light transmittance of 72.9% combined with a low sheet resistance of 4.5 Ω sq-1 and a desirable EMI SE value of 35.2dB at X band, 38.9dB in the K-band, with the highest SE of 43.4dB at 20.4GHz. Simultaneously, the excellent conductivity endowed the film with outstanding Joule heating performance and defogging/deicing ability, ensuring the visual transparency of windows when shielding electromagnetic waves. Hence, this research presents a highly effective strategy for constructing an aligned AgNW network, offering a promising solution for enhancing the performance of optical-electronic devices.

3D Printing-Induced Hierarchically Aligned Nanocomposites With Exceptional Multidirectional Strain Sensing Performance.

High-performance sensors capable of detecting multidirectional strains are indispensable to understand the complex motions involved in flexible electronics. Conventional isotropic strain sensors can only measure uniaxial deformations or single stimuli, hindering their practical application fields. The answer to such challenge resides in the construction of engineered anisotropic sensing structures. Herein, a hierarchically aligned carbon nanofiber (CNF)/polydimethylsiloxane nanocomposite strain sensor is developed by one-step 3D printing. The precisely controlled printing path and shear flow bring about highly aligned nanocomposite filaments at macroscale and orientated CNF network within each filament at microscale. The periodically orientated nanocomposite filaments along with the inner aligned CNF network successfully control the strain distribution and the appearance of microcracks, giving rise to anisotropic structural response to external deformations. The synergetic effect of the multiscale structural design leads to distinguishable gauge factors of 164 and 0.5 for applied loadings along and transverse to the alignment direction, leading to an exceptional selectivity of 3.77. The real-world applications of the hierarchically aligned sensors in multiaxial movement detector and posture-correction device are further demonstrated. The above findings propose new ideas for manufacturing nanocomposites with engineered anisotropic structure and properties, verifying promising applications in emerging wearable electronics and soft robotics.

Development of biomass aerogels with aligned channels: For continuous treatment of oily wastewater and solar-powered oil evaporation

A biomass CS/CNTs@MTMS (MCCS) aerogel with both aligned channel network, superhydrophobicity, and photothermal conversion ability was prepared by a green and facile strategy of directed freeze-drying and chemical vapor deposition using chitosan (CS), carbon nanotubes (CNTs), and methyltrimethoxysilane (MTMS) as the building materials. Capacity to adsorb a large variety of oils and organic solvents, with an adsorption capacity of up to 34–83 g/g. After 10 cycles, the adsorption capacity of MCCS remained at 94 % of the initial capacity, providing excellent reusability. In addition, due to its unique network of aligned channels, the MCCS can continuously separate oil and water, making it a sustainable oil–water separator. More interestingly, the MCCS aerogel has excellent photothermal conversion capabilities, and it was utilized to evaporate oil collected during the oil–water separation process using solar energy. This work provides an opportunity to design novel self-cleaning photothermally driven oil–water separation biomass materials with superhydrophobicity-strong lipophilicity.

Highly conductive solid-solid phase change composites and devices enhanced by aligned graphite networks for solar/electro-thermal energy storage

Phase change materials (PCMs) are widely considered as promising energy storage materials for solar/electro-thermal energy storage. Nevertheless, the inherent low thermal/electrical conductivities of most PCMs limit their energy conversion efficiencies, hindering their practical applications. Herein, we fabricate a highly thermally/electrically conductive solid-solid phase change composite (PCC) enabled by forming aligned graphite networks through pressing the mixture of the trimethylolethane and porous expanded graphite (EG). Experiments indicate that both the thermal and electrical conductivities of the PCC increase with increasing mass proportion of the EG because the aligned graphite networks establish highly conductive pathways. Meanwhile, the PCC4 sample with the EG proportion of 20 ​wt% can achieve a high thermal conductivity of 12.82 ​± ​0.38 ​W·m−1·K−1 and a high electrical conductivity of 4.11 ​± ​0.02 ​S·cm−1 in the lengthwise direction. Furthermore, a solar-thermal energy storage device incorporating the PCC4, a solar selective absorber, and a highly transparent glass is developed, which reaches a high solar-thermal efficiency of 77.30 ​± ​2.71% under 3.0 suns. Moreover, the PCC4 can also reach a high electro-thermal efficiency of 91.62 ​± ​3.52% at a low voltage of 3.6 ​V, demonstrating its superior electro-thermal storage performance. Finally, stability experiments indicate that PCCs exhibit stabilized performance in prolonged TES operations. Overall, this work offers highly conductive and cost-effective PCCs, which are suitable for large-scale and efficient solar/electro-thermal energy storage.

Understanding how junction resistances impact the conduction mechanism in nano-networks

Networks of nanowires, nanotubes, and nanosheets are important for many applications in printed electronics. However, the network conductivity and mobility are usually limited by the resistance between the particles, often referred to as the junction resistance. Minimising the junction resistance has proven to be challenging, partly because it is difficult to measure. Here, we develop a simple model for electrical conduction in networks of 1D or 2D nanomaterials that allows us to extract junction and nanoparticle resistances from particle-size-dependent DC network resistivity data. We find junction resistances in porous networks to scale with nanoparticle resistivity and vary from 5 Ω for silver nanosheets to 24 GΩ for WS2 nanosheets. Moreover, our model allows junction and nanoparticle resistances to be obtained simultaneously from AC impedance spectra of semiconducting nanosheet networks. Through our model, we use the impedance data to directly link the high mobility of aligned networks of electrochemically exfoliated MoS2 nanosheets (≈ 7 cm2 V−1 s−1) to low junction resistances of ∼2.3 MΩ. Temperature-dependent impedance measurements also allow us to comprehensively investigate transport mechanisms within the network and quantitatively differentiate intra-nanosheet phonon-limited bandlike transport from inter-nanosheet hopping.

Splay aligned liquid crystal network photoactuators for integrated microfluidic pumps

As lab-on-a-chip devices aim to incorporate complex and multi-step microfluidic workflows, active fluid control within these platforms is critical. We use a splay aligned liquid crystal network (LCN) photoactuator, capable of deformation upon blue light illumination, as a stimuli responsive material to create integrated fluidic elements. After optimization of the manufacturing parameters, a well-defined actuator material is attained. Illumination with a 455 nm light emitting diode (LED) at 41 mW cm−2 achieves deflection of a 3 × 8 × 0.05 mm LCN film by 380 µm, generating forces of 1.1 mN. The actuator response is stable over time and scalable by light intensity, temperature, and size of the LCN film. By combining with a polydimethylsiloxane (PDMS) membrane, integration into a microfluidic chip is demonstrated and fluid movement of 77 nL in a 2 s stroke is attained. In a pumping setup, the LCN film functions as pumping membrane and two passive PDMS check valves complete the integrated micropump. Constant pumping rates of 0.1 µL min−1 are achieved. An advantage of this micropump setup lies in the non-contact actuation method, which allows for easy integration into microfluidic chips, without the need for chip-to-world connections. Furthermore, with the simple manufacturing procedure and the low operating power, important requirements for application in point-of-care settings are fulfilled.

Learning Aligned Vertex Convolutional Networks for Graph Classification.

Graph convolutional networks (GCNs) are powerful tools for graph structure data analysis. One main drawback arising in most existing GCN models is that of the oversmoothing problem, i.e., the vertex features abstracted from the existing graph convolution operation have previously tended to be indistinguishable if the GCN model has many convolutional layers (e.g., more than two layers). To address this problem, in this article, we propose a family of aligned vertex convolutional network (AVCN) models that focus on learning multiscale features from local-level vertices for graph classification. This is done by adopting a transitive vertex alignment algorithm to transform arbitrary-sized graphs into fixed-size grid structures. Furthermore, we define a new aligned vertex convolution operation that can effectively learn multiscale vertex characteristics by gradually aggregating local-level neighboring aligned vertices residing on the original grid structures into a new packed aligned vertex. With the new vertex convolution operation to hand, we propose two architectures for the AVCN models to extract different hierarchical multiscale vertex feature representations for graph classification. We show that the proposed models can avoid iteratively propagating redundant information between specific neighboring vertices, restricting the notorious oversmoothing problem arising in most spatial-based GCN models. Experimental evaluations on benchmark datasets demonstrate the effectiveness.

A gradient aligned domain adversarial network for unsupervised intelligent fault diagnosis of gearboxes

Unsupervised domain adaptation alleviates the dependencies of conventional fault diagnosis methods on sufficient labeled data and strict data distributions. Nonetheless, the current domain adaptation methods only concentrate on the data distributions and ignore the feature gradient distributions, leading to some samples being misclassified due to large gradient discrepancies, thus affecting diagnosis performance. In this paper, a gradient aligned domain adversarial network (GADAN) is proposed. First, the discrepancies of the marginal and conditional distribution between the source and target domain are reduced by minimizing the joint maximum mean discrepancy. Then, a pseudo-labeling approach based on a clustering self-supervised strategy is utilized to attain high-quality pseudo-labels of target domains, and most importantly in the dimension of the data gradient, the feature gradient distributions are aligned by adversarial learning to further reduce the domain shift, even if the distributions of the two domains are close enough. Finally, experiments and engineering applications demonstrate the effectiveness and superiority of GADAN for transfer diagnosis between various working conditions or different machines.

Enhancing the Thermal Conductivity of Epoxy Composites via Constructing Oriented ZnO Nanowire-Decorated Carbon Fibers Networks.

With the miniaturization and high integration of electronic devices, high-performance thermally conductive composites have received increasing attention. The construction of hierarchical structures is an effective strategy to reduce interfacial thermal resistance and enhance composite thermal conductivity. In this study, by decorating carbon fibers (CF) with needle-like ZnO nanowires, hierarchical hybrid fillers (CF@ZnO) were rationally designed and synthesized using the hydrothermal method, which was further used to construct oriented aligned filler networks via the simple freeze-casting process. Subsequently, epoxy (EP)-based composites were prepared using the vacuum impregnation method. Compared with the pure CF, the CF@ZnO hybrid fillers led to a significant increase in thermal conductivity, which was mainly due to the fact that the ZnO nanowires could act as bridging links between CF to increase more thermally conductive pathways, which in turn reduced interfacial thermal resistance. In addition, the introduction of CF@ZnO fillers was also beneficial in improving the thermal stability of the EP-based composites, which was favorable for practical thermal management applications.

Incorporating attribute-level aligned comparative network for generalized zero-shot learning

The key challenge of zero-shot learning (ZSL) is to sufficiently disentangle each latent attribute from the class-level semantic annotations of images, thereby achieving a desirable semantic transfer to unseen classes with the disentangled attributes. However, most existing studies tackle the ZSL task with a strict class-level alignment strategy that may yield insufficient disentanglement: (1) this strategy simply aligns holistic visual feature with its associated class-level semantic vector for each image; (2) the class-level semantic vectors have limited diversity and complex compositions of attributes. To address these issues, we propose an incorporating attribute-level aligned comparative network, i.e., IAAC-net, that develops the alignment strategy of ZSL to the attribute level. IAAC-net aims to establish diversified attribute-level and refined class-level alignments to facilitate attribute disentanglement and simultaneously improve zero-shot generalization. By further proposing a confusion-aware loss, the model is forced to rectify the disentanglement of indistinguishable attributes, which leads to a more accurate attribute disentanglement. The proposed IAAC-net yields significant improvements over the strong baselines, leading to new state-of-the-art performances on three popular challenging benchmarks, i.e., CUB, SUN, and AWA2.

Enhanced composite thermal conductivity by percolated networks of in-situ confined-grown carbon nanotubes

Despite the ever-increasing demand of nanofillers for thermal enhancement of polymer composites with higher thermal conductivity and irregular geometry, nanomaterials like carbon nanotubes (CNTs) have been constrained by the nonuniform dispersion and difficulty in constructing effective three-dimensional (3D) conduction network with low loading and desired isotropic or anisotropic (specific preferred heat conduction) performances. Herein, we illustrated the in-situ construction of CNT based 3D heat conduction networks with different directional performances. First, to in-situ construct an isotropic percolated conduction network, with spherical cores as support materials, we developed a confined-growth technique for CNT-core sea urchin (CNTSU) materials. With 21.0 wt.% CNTSU loading, the thermal conductivity of composites reached 1.43 ± 0.13 W/(m·K). Secondly, with aligned hexagonal boron nitride (hBN) as an anisotropic support, we constructed CNT-hBN aligned networks by in-situ CNT growth, which improved the utilization efficiency of high density hBN and reduced the thermal interface resistance between matrix and fillers. With ~ 8.5 wt.% loading, the composites possess thermal conductivity up to 0.86 ± 0.14 W/(m·K), 374% of that for neat matrix. Due to the uniformity of CNTs in hBN network, the synergistic thermal enhancement from one-dimensional (1D) + two-dimensional (2D) hybrid materials becomes more distinct. Based on the detailed experimental evidence, the importance of purposeful production of a uniformly interconnected heat conduction 3D network with desired directional performance can be observed, particularly compared with the traditional direct-mixing method. This study opens new possibilities for the preparation of high-power-density electronics packaging and interfacial materials when both directional thermal performance and complex composite geometry are simultaneously required.

Reactive Cell Electrospinning of Anisotropically Aligned and Bilayer Hydrogel Nanofiber Networks.

Structured hydrogels that incorporate aligned nanofibrous morphologies have been demonstrated to better replicate the structures of native extracellular matrices and thus their function in guiding cell responses. However, current techniques for nanofiber fabrication are limited in their ability to create hydrogel scaffolds with tunable directional alignments and cell types/densities, as required to reproduce more complex native tissue structures. Herein, we leverage a reactive cell electrospinning technique based on the dynamic covalent cross-linking of poly(ethylene glycol methacrylate (POEGMA) precursor polymers to fabricate aligned hydrogel nanofibers that can be directly loaded with cells during the electrospinning process. The scaffolds were found to support high C2C12 myoblast viabilities greater than 85% over 14 days, with changes in the electrospinning collector allowing for the single-step fabrication of nonaligned, aligned, or cross-aligned nanofibrous networks. Cell aspect ratios on aligned scaffolds were found on average to be 27% higher compared to those on nonaligned scaffolds; furthermore, cell-loaded bilayer scaffolds with perpendicular fiber alignments showed evidence of enabling localized directional cell responses to individual layer fiber directions while avoiding delamination between the layers. This fabrication approach thus offers potential for better mimicking the structure and thus function of aligned and multilayered tissues (e.g., smooth muscle, neural, or tendon tissues).

Integration of migration and attention flow data to reveal association of virtual–real dual intercity network structure

The study of intercity network structure from the perspective of flow space can effectively reveal the connection patterns, hierarchical laws, and spatial organization of cities. However, current studies mostly focus on a single real or virtual space, neglecting the comprehensive perspective. Furthermore, the research subject of network relationships is usually the complex original aligned network rather than the spatially cohesive network structure. In this study, spatial network structures are constructed and extracted based on the migration flow and attention flow, which represent people's objective behavior and subjective will, respectively. This study then reveals the association between virtual and real networks from three aspects: node location, node attributes, and local network structure characteristics. Results indicate that compared with the real network structure, the virtual network structure has a strong same-level node agglomeration and a significant node size difference. Spatial dislocations are observed in the high association of the various measures of China's real and virtual network structures. The degree of differences in network structures between virtual and real spaces varies spatially. This paper reveals the dynamics of urban development in different spaces, helping build a human-centered, high-quality space governance system for virtual and real spaces.

Multifunctional nanocomposites reinforced by aligned graphene network via a low-cost lyophilization-free method

Compared with a disordered network, the ordered framework is more beneficial to improving the multifunctional properties (including thermal, electrical, and mechanical properties) of epoxy-based composites. However, current methods for building aligned structures are both energy and time-consuming on account of the tedious and lengthy lyophilization process. Herein, a new strategy is proposed to obtain low-density graphene aerogels (GAs) with long-range oriented structures through a lyophilization-free processes. Due to the minimum thermal boundary resistance, the prepared aligned graphene/epoxy composite exhibits an ultrahigh thermal conductivity of ∼11.6 W/(m·K) at a graphene content of 1.84 vol%, i.e., an enhancement efficiency of 3640% per 1 vol%. In addition, the compressive strength of the composite is increased by 2.6 times compared to epoxy (from 0.29 MPa to 0.76 MPa). Because the composite shows a modulus as low as 0.5–1.0 MPa, it has excellent deformability and compression performance. Moreover, the obtained 3-mm-thick composite achieves an electromagnetic interference shielding effectiveness (EMI SE) of 40 dB due to the interconnected graphene network structure, which meets the requirements of commercial EMI shielding applications. Hence, this strategy can be used to synthesize aligned graphene/epoxy composites with excellent multifunctional performances, which can be simultaneously used as elastic thermal interface materials (TIMs) and EMI shielding materials in the fields of advanced electronic packaging.

Network alignment and link prediction using event-based embedding in aligned heterogeneous dynamic social networks

Network alignment and link prediction using event-based embedding in aligned heterogeneous dynamic social networks

Simplified geodata models for integrated urban and public transport planning

Abstract. The current division between urban and transport planning is a significant obstacle to achieving sustainable urban development. To transform cities towards sustainability, both fields must adopt shared or at least compatible models of the urban systems, namely transport, street and public space networks for all users and urban activities. Although several models and tools have emerged in recent years to facilitate this integration, there are still usability gaps that hinder their wider adoption. One of the gaps is a lack of flexibility to operate at different stages of integrated planning. To address this gap, the study aims to develop a set of aligned and flexible multimodal urban network models and tools to support different stages of planning. This paper focuses on the public transport geodata models, which were built by aggregating a General Transit Feed Specification (GTFS) dataset at various spatial and temporal levels. The aggregation levels range from a baseline data model that is useful for detailed planning stages, up to a topological data model that is suitable for macro scale and strategic planning. By using this unified set of models, the dialogue between the two fields at different integrated planning phases can be facilitated, and decision-making can be enhanced.

Transition from Diffusive to Superdiffusive Transport in Carbon Nanotube Networks via Nematic Order Control.

The one-dimensional confinement of quasiparticles in individual carbon nanotubes (CNTs) leads to extremely anisotropic electronic and optical properties. In a macroscopic ensemble of randomly oriented CNTs, this anisotropy disappears together with other properties that make them attractive for certain device applications. The question however remains if not only anisotropy but also other types of behaviors are suppressed by disorder. Here, we compare the dynamics of quasiparticles under strong electric fields in aligned and random CNT networks using a combination of terahertz emission and photocurrent experiments and out-of-equilibrium numerical simulations. We find that the degree of alignment strongly influences the excited quasiparticles' dynamics, rerouting the thermalization pathways. This is, in particular, evidenced in the high-energy, high-momentum electronic population (probed through the formation of low energy excitons via exciton impact ionization) and the transport regime evolving from diffusive to superdiffusive.

Transparent and Stretchable Electromagnetic Interference Shielding Film with Fence-like Aligned Silver Nanowire Conductive Network.

Flexible transparent conductive electrodes (TCEs) that can be used as electromagnetic interference (EMI) shielding materials have a great potential for use as electronic components in optical window and display applications. However, development of TCEs that display high shielding effectiveness (SE) and good stretchability for flexible electronic device applications has proven challenging. Herein, this study describes a stretchable polydimethylsiloxane (PDMS)/silver nanowire (AgNW) TCE with a fence-like aligned conductive network that is fabricated via pre-stretching method. The fence-like AgNW network endowed the PDMS/AgNW film with excellent optoelectronic properties, i.e., low sheet resistance of 7.68Ωsq-1 at 73.7% optical transmittance, thus causing an effective EMI SE of 32.2dB at X-band. More importantly, the fence-like aligned AgNW conductive network reveals a high stability toward tensile deformation, thus gives the PDMS/AgNW film stretch-stable conductivity and EMI shielding property in the strain range of 0-100%. Typically, the film can reserve ≈70% or 80% of its initial EMI SE when stretching at 100% strain or stretching/releasing (50% strain) for 128 cycles, respectively. Additionally, the film exhibits a low-voltage driven and stretchable Joule heating performance. With these overall performances, the PDMS/AgNW film should be well suited for use in flexible and stretchable optical electronic devices.

Integration of reconfigurable microchannels into aligned three-dimensional neural networks for spatially controllable neuromodulation.

Anisotropically organized neural networks are indispensable routes for functional connectivity in the brain, which remains largely unknown. While prevailing animal models require additional preparation and stimulation-applying devices and have exhibited limited capabilities regarding localized stimulation, no in vitro platform exists that permits spatiotemporal control of chemo-stimulation in anisotropic three-dimensional (3D) neural networks. We present the integration of microchannels seamlessly into a fibril-aligned 3D scaffold by adapting a single fabrication principle. We investigated the underlying physics of elastic microchannels' ridges and interfacial sol-gel transition of collagen under compression to determine a critical window of geometry and strain. We demonstrated the spatiotemporally resolved neuromodulation in an aligned 3D neural network by local deliveries of KCl and Ca2+ signal inhibitors, such as tetrodotoxin, nifedipine, and mibefradil, and also visualized Ca2+ signal propagation with a speed of ~3.7 μm/s. We anticipate that our technology will pave the way to elucidate functional connectivity and neurological diseases associated with transsynaptic propagation.

High-Mobility Flexible Transistors with Low-Temperature Solution-Processed Tungsten Dichalcogenides

The investigation of high-mobility two-dimensional (2D) flakes beyond molybdenum disulfide (MoS2) will be necessary to create a library of high-mobility solution-processed networks that conform to substrates and remain functional over thousands of bending cycles. Here we report electrochemical exfoliation of large-aspect-ratio (>100) semiconducting flakes of tungsten diselenide (WSe2) and tungsten disulfide (WS2) as well as MoS2 as a comparison. We use Langmuir-Schaefer coating to achieve highly aligned and conformal flake networks, with minimal mesoporosity (∼2-5%), at low processing temperatures (120 °C) and without acid treatments. This allows us to fabricate electrochemical transistors in ambient air, achieving average mobilities of μMoS2 ≈ 11 cm2 V-1 s-1, μWS2 ≈ 9 cm2 V-1 s-1, and μWSe2 ≈ 2 cm2 V-1 s-1 with a current on/off ratios of Ion/Ioff ≈ 2.6 × 103, 3.4 × 103, and 4.2 × 104 for MoS2, WS2, and WSe2, respectively. Moreover, our transistors display threshold voltages near ∼0.4 V with subthreshold slopes as low as 182 mV/dec, which are essential factors in maintaining power efficiency and represent a 1 order of magnitude improvement in the state of the art. Furthermore, the performance of our WSe2 transistors is maintained on polyethylene terephthalate (PET) even after 1000 bending cycles at 1% strain.