Lightweight, nanoporous aerogel fibers are crucial for personal thermal management and specialized heat protection. However, wet-spinning methods, exemplified by aramid aerogels, inevitably form a dense outer layer, significantly reducing the volume fraction of efficient thermal barrier nanovoids and limiting the development of ultimate thermal resistance in fibers. Herein, we develop a microfluidic spinning method to prepare gradient all-nanostructure aramid aerogel fibers (GAFs). Benefiting from the simultaneous shear alignment and diffusion dilution of a good solvent within the channels, the precursor gel fibers assemble into a structure with a sparse exterior and dense interior, which reverses during supercritical drying to form sheath and core layers with average pore diameters of 150 nm and 600 nm, respectively. Experiments and simulations reveal that the gradient nanostructure creates high interfacial thermal resistance at heat transfer interfaces, resulting in a radial thermal conductivity as low as 0.0228 W m–1 K–1, far below that of air and wet-spun aerogel fibers. Moreover, GAF’s unique nano-entangled network efficiently dissipates stress, achieving exceptionally high tensile strength (29.5 MPa) and fracture strain (39.2%). This work establishes a correlation between multiscale nanostructures and superlative performance, thereby expanding the scope of aerogel applications in intricate environments.
To take full advantage of the electronic properties of transition-metal dichalcogenides and their vdW layered structures, it will be necessary to control the local electronic structure, on which the effect of lattice deformation is significant. Nevertheless, a general approach to programming nanoscale morphology in TMD materials, which would permit local strain engineering, has proven elusive. In this work, we propose a general moiré-templated nanoscale morphology engineering method based on bilayer TMDs. The moiré superlattice plays the key role in enforcing in-plane periodical variations in local interlayer spacing and potential energy. Upon global in-plane compression, the high-energy, large-interlayer-separation stacking domains serve as periodic buckling initiation sites. The buckled features can be thus precisely correlated to the moiré periodicity. The spatial profile of the buckled morphology and strain field are possible to be pre-determined, providing a bridge to the electronics and optoelectronics design. We take twisted bilayer MoS2 to demonstrate our approach. We further demonstrate how the morphology can modulate band gap and optical absorption of a MoS2 monolayer, envisioned as a potential constituent layer in a Moiré-templated, strain-engineered vdW heterostructure of TMDs. The robustness and programmable nature of our approach arise from superlattice symmetry, energetics and mechanics. Our approach provides a new strategy for on-demand design of morphology and local strain in TMDs under mechanical deformation.
The quest for efficient and robust deep learning models for molecular systems representation is increasingly critical in scientific exploration. The advent of message passing neural networks has marked a transformative era in graph-based learning, particularly in the realm of predicting chemical properties and expediting molecular dynamics studies. We present the Moment Graph Neural Network (MGNN), a rotation-invariant message passing neural network architecture that capitalizes on the moment representation learning of 3D molecular graphs, is adept at capturing the nuanced spatial relationships inherent in three-dimensional molecular structures. From benchmark tests on public datasets, MGNN delivers multiple state-of-the-art results on QM9, revised MD17 and MD17-ethanol. Its generalizability and efficiency are also tested in additional systems including 3BPA and 25-element high-entropy alloys. The prowess of MGNN also extends to dynamic simulations, accurately predicting the structural and kinetic properties of complex systems such as amorphous electrolytes, with results that closely align with those from ab-initio simulations. The application of MGNN to the simulation of molecular spectra exemplifies its potential to offer a promising alternative to traditional electronic structure methods.