Search

Polaritons are a hybrid class of quasiparticles originating from the strong and resonant coupling between light and matter excitations. Recent years have witnessed a surge of interest in novel polariton types, arising from directional, long-lived material resonances, and leading to extreme optical anisotropy that enables novel regimes of nanoscale, highly confined light propagation. While such exotic propagation features may also be in principle achieved using carefully designed metamaterials, it has been recently realized that they can naturally emerge when coupling infrared light to directional lattice vibrations, i.e., phonons, in polar crystals. Interestingly, a reduction in crystal symmetry increases the directionality of optical phonons and the resulting anisotropy of the response, which in turn enables new polaritonic phenomena, such as hyperbolic polaritons with highly directional propagation, ghost polaritons with complex-valued wave vectors, and shear polaritons with strongly asymmetric propagation features. In this Review, we develop a critical overview of recent advances in the discovery of phonon polaritons in low-symmetry crystals, highlighting the role of broken symmetries in dictating the polariton response and associated nanoscale-light propagation features. We also discuss emerging opportunities for polaritons in lower-symmetry materials and metamaterials, with connections to topological physics and the possibility of leveraging anisotropic nonlinearities and optical pumping to further control their nanoscale response.

Following the discovery of moir\'e-driven superconductivity in twisted graphene multilayers, twistronics has spurred a surge of interest in tailored broken symmetries through angular rotations, enabling new properties from electronics to photonics and phononics. Analogously, in monoclinic polar crystals a nontrivial angle between non-degenerate dipolar phonon resonances can naturally emerge due to asymmetries in their crystal lattice, and its variations are associated with intriguing polaritonic phenomena, including axial dispersion, i.e., a rotation of the optical axis with frequency, and microscopic shear effects that result in asymmetric loss distributions. So far these phenomena were restricted to specific mid-infrared frequencies, difficult to access with conventional lasers, and fundamentally limited by the degree of asymmetry and the strength of light-matter interactions available in natural crystals. Here, we leverage twistronics to demonstrate giant axial dispersion and loss asymmetry of hyperbolic waves in elastic metasurfaces, by tailoring the angle between coupled pairs of anisotropic metasurfaces. We show extreme control over elastic wave dispersion via the twist angle, and leverage the resulting phenomena to demonstrate reflection-free negative refraction, as well as the application of axial dispersion to achieve diffraction-free non-destructive testing, whereby the angular direction of a hyperbolic probe wave is encoded into its frequency. Our work welds the concepts of twistronics, non-Hermiticity and extreme anisotropy, demonstrating the powerful opportunities enabled by metasurfaces for tunable, highly directional surface acoustic wave propagation, of great interest for applications ranging from seismic mitigation to on-chip phononics and wireless communications, paving the way towards their translation into emerging photonic and polaritonic metasurface technologies.

Among different phases of the life cycle of a building or facility, design is of the utmost importance to ensure safety, efficiency and sustainability of the building or facility. How to control and improve design quality and efficiency has been explored for years, and more studies emerged with the popularization of Building Information Modelling (BIM). However, most of them focused on the extraction of design behaviors, while paying less attention to how a design is formed. Therefore, this study proposes an approach to tracking and replaying the BIM-based design process by integrating data logging and 4D visualization techniques. First of all, potential design behaviors and procedures are analyzed and extracted by observing how a designer designs a BIM model. Meanwhile, the required data for logging design process is defined and a relevant method to collect these data is developed based on the APIs of BIM software. Then, strategies on how to visualize different design procedures are designed and implemented via 4D visualization. Finally, a prototype system is developed based on Autodesk Revit and validated through a case study. Result shows that the proposed approach enables intuitively and interactively review of the design process, and makes it easier to understand design behaviors and even identify potential pitfalls, thus improving the design efficiency and quality.

Recently, higher-order topological phases have endowed materials many exotic topological phases. For three-dimensional (3D) higher-order topologies, it hosts topologically protected 1D hinge states or 0D corner states, which extend the bulk-boundary correspondence of 3D topological phases. Meanwhile, the enrichment of group symmetries with exploration of projective symmetry algebras redefined the fundamentals of nontrivial topological matter with artificial gauge fields, leading to the discovery of new topological phases in classical wave systems. In this Letter, we construct an acoustic topological semimetal characterized by both the first and the second Stiefel-Whitney (SW) topological charges by utilizing the projective symmetry. Different from conventional high-order topologies with multiple bulk-boundary correspondences protected by different class topological invariants, acoustic high-order Stiefel-Whitney semimetal (HOSWS) has two different bulk-edge correspondences protected by only one class (SW class) topological invariant. Two types of topological hinge and surface states are embedded in bulk bands at the same frequency, featuring similar characteristics of bound states in the continuum (BICs). In experiments, we demonstrate the existence of high-quality surface state and hinge state at the interested frequency window with polarized intensity field distributions.

Hydrogel adhesion that can be easily modulated in magnitude, space, and time is desirable in many emerging applications ranging from tissue engineering, and soft robotics, to wearable devices. In synthetic materials, these complex adhesion behaviors are often achieved individually with mechanisms and apparatus that are difficult to integrate. Here, we report a universal strategy to embody multifaceted adhesion programmability in synthetic hydrogels. By designing the surface network topology of a hydrogel, supramolecular linkages that result in contrasting adhesion behaviors are formed on the hydrogel interface. The incorporation of different topological linkages leads to dynamically tunable adhesion with high-resolution spatial programmability without alteration of bulk mechanics and chemistry. Further, the association of linkages enables stable and tunable adhesion kinetics that can be tailored to suit different applications. We rationalize the physics of chain slippage, rupture, and diffusion that underpins emergent programmable behaviors. We then incorporate the strategy into the designs of various devices such as smart wound patches, fluidic channels, drug-eluting devices, and reconfigurable soft robotics. Our study presents a simple and robust platform in which adhesion controllability in multiple aspects can be easily integrated into a single design of a hydrogel network.

One of the most significant breakthroughs in physics of the last decade has been the discovery that materials with non-trivial topological properties for electronic, electromagnetic, acoustic and mechanical responses can be designed and manufactured at our will through engineered metamaterials (MMs). Here, we review the foundation and the state-of-the-art advances of topological photonics, acoustics and mechanical MMs. We discuss how topological MMs enable nontrivial wave phenomena in physics, engineering, of great interest for a broad range of interdisciplinary science disciplines such as classical and quantum chemistry. We first introduce the foundations of topological materials and the main concepts behind their peculiar features, including the concepts of topological charge and geometric phase. We then discuss the topology of electronic band structures in natural topological materials, like topological insulators and gapless Dirac and Weyl semimetals. Based on these concepts, we review the concept, design and response of topologically nontrivial MMs in photonics and phononics, including topological phases in 2D MMs with and without time-reversal symmetry, Floquet TIs based on spatial and temporal modulation, topological phases in 3D MMs, higher-order topological phases in MMs, non-Hermitian and nonlinear topological MMs and the topological features of scattering anomalies. We also discuss the topological properties emerging in other related contexts, such as the topological aspects of chemical reactions and polaritons. This survey aims at connecting the recent advances in a broad range of scientific areas associated with topological concepts, and highlights opportunities offered by topological MMs for the chemistry community at large.

Recently, Graph Neural Network (GNN) has achieved remarkable success in various real-world problems on graph data. However in most industries, data exists in the form of isolated islands and the data privacy and security is also an important issue. In this paper, we propose FedVGCN, a federated GCN learning paradigm for privacy-preserving node classification task under data vertically partitioned setting, which can be generalized to existing GCN models. Specifically, we split the computation graph data into two parts. For each iteration of the training process, the two parties transfer intermediate results to each other under homomorphic encryption. We conduct experiments on benchmark data and the results demonstrate the effectiveness of FedVGCN in the case of GraphSage.

The rapid deployment of Internet of Things (IoT) applications leads to massive data that need to be processed. These IoT applications have specific communication requirements on latency and bandwidth, and present new features on their generated data such as time-dependency. Therefore, it is desirable to reshape the current IoT architectures by exploring their inherent nature of communication and computing to support smart IoT data process and analysis. We introduce in this paper features of IoT data, trends of IoT network architectures, some problems in IoT data analysis, and their solutions. Specifically, we view that software-defined edge computing is a promising architecture to support the unique needs of IoT data analysis. We further present an experiment on data anomaly detection in this architecture, and the comparison between two architectures for ECG diagnosis. Results show that our method is effective and feasible.

Tough adhesive hydrogels find broad applications in engineering and medicine. Such hydrogels feature high resistance against both cohesion and adhesion failure. The superior fracture properties may, however, deteriorate when the hydrogels swell upon exposure of water. The underlying correlation between the polymer fraction and fracture properties of tough adhesive hydrogels remains largely unexplored. Here we study how the cohesion and adhesion energies of a tough adhesive hydrogel evolve with the swelling process. The results reveal a similar scaling law of the two quantities on the polymer fraction. Our scaling analysis and computational study reveal that it stems from the scaling of shear modulus. The study will promote the investigation of scaling of hydrogel fracture and provide development guidelines for new tough adhesive hydrogels., Comment: submitted to ACS Macro Letters

The properties of topological systems are inherently tied to their dimensionality. Higher-dimensional physical systems exhibit topological properties not shared by their lower dimensional counterparts and, in general, offer richer physics. One example is a d-dimensional quantized multipole topological insulator, which supports multipoles of order up to 2^d and a hierarchy of gapped boundary modes with topological 0-D corner modes at the top. While multipole topological insulators have been successfully realized in electromagnetic and mechanical 2D systems with quadrupole polarization, and a 3D octupole topological insulator was recently demonstrated in acoustics, going beyond the three physical dimensions of space is an intriguing and challenging task. In this work, we apply dimensional reduction to map a 4D higher-order topological insulator (HOTI) onto an equivalent aperiodic 1D array sharing the same spectrum, and emulate in this system the properties of a hexadecapole topological insulator. We observe the 1D counterpart of zero-energy states localized at 4D HOTI corners - the hallmark of multipole topological phase. Interestingly, the dimensional reduction guarantees that one of the 4D corner states remains localized to the edge of the 1D array, while all other localize in the bulk and retain their zero-energy eigenvalues. This discovery opens new directions in multi-dimensional topological physics arising in lower-dimensional aperiodic systems, and it unveils highly unusual resonances protected by topological properties inherited from higher dimensions., Comment: 13 pages, 4 figures

Topological photonic systems represent a new class of optical materials supporting boundary modes with unique properties, not found in conventional photonics. While the early research on topological photonics has focused on edge and surface modes in 2D and 3D systems, respectively, recently higher-order topological insulators (HOTIs) supporting lower-dimensional boundary states have been introduced. In this work we design and experimentally realize a photonic kagome metasurface exhibiting a Wannier-type higher-order topological phase. We demonstrate and visualize the emergence of a topological transition and opening of a Dirac cone by directly exciting the bulk modes of the HOTI metasurface via solid-state immersion spectroscopy. The open nature of the metasurface is then utilized to directly image topological boundary states. We show that, while the domain walls host 1D edge states, their bending induces 0D higher-order topological modes confined to the corners. The demonstrated metasurface hosting topological boundary modes of different dimensionality paves the way to a new generation of universal and resilient optical devices which can controllably scatter, trap and guide optical fields in a robust way., Comment: 14 pages, 5 figures

This paper considers the problem of resource allocation in stream processing, where continuous data flows must be processed in real time in a large distributed system. To maximize system throughput, the resource allocation strategy that partitions the computation tasks of a stream processing graph onto computing devices must simultaneously balance workload distribution and minimize communication. Since this problem of graph partitioning is known to be NP-complete yet crucial to practical streaming systems, many heuristic-based algorithms have been developed to find reasonably good solutions. In this paper, we present a graph-aware encoder-decoder framework to learn a generalizable resource allocation strategy that can properly distribute computation tasks of stream processing graphs unobserved from training data. We, for the first time, propose to leverage graph embedding to learn the structural information of the stream processing graphs. Jointly trained with the graph-aware decoder using deep reinforcement learning, our approach can effectively find optimized solutions for unseen graphs. Our experiments show that the proposed model outperforms both METIS, a state-of-the-art graph partitioning algorithm, and an LSTM-based encoder-decoder model, in about 70% of the test cases., Comment: Accepted by AAAI 2020

Recently extended from the modern theory of electric polarization, quantized multipole topological insulators (QMTIs) describe higher-order multipole moments, lying in nested Wilson loops, which are inherently quantized by lattice symmetries. Overlooked in the past, QMTIs reveal new types of gapped boundaries, which themselves represent lower-dimensional topological phases and host topologically protected zero-dimensional (0D) corner states. Inspired by these pioneering theoretical predictions, tremendous efforts have been devoted to the experimental observation of topological quantized quadrupole phase in a variety of two dimensional (2D) metamaterials. However, due to stringent requirements of anti-commuting reflection symmetries in crystals, it has been challenging to achieve higher-order quantized multipole moments, such as octupole moments, in a realistic three-dimensional (3D) structure. Here, we overcome these challenges, and experimentally realize the acoustic analogue of a quantized octupole topological insulator (QOTIs) using negatively coupled resonators. We confirm by first-principle studies that our design possesses a quantized octupole topological phase, and experimentally demonstrate spectroscopic evidence of a topological hierarchy of states in our metamaterial, observing 3rd order corner states, 2nd order hinge states and 1st order surface states. Furthermore, we reveal topological phase transitions from higher- to lower-order multipole moments in altered designs of acoustic TIs. Our work offers a new pathway to explore higher-order topological states (HOTSs) in 3D classical platforms., Comment: 15 pages, 7 figures

The discovery of topological phases has recently led to a paradigm shift in condensed matter physics, and facilitated breakthroughs in engineered photonics and acoustic metamaterials. Topological insulators (TIs) enable the generation of electronic, photonic, and acoustic modes exhibiting wave propagation that is resilient to disorder, irrespective of manufacturing precision or unpredictable defects induced by the operational environment, known as topological protection. While originally limited to a dimensionality of the protected states that is one dimension lower than the host TI material, the recent discovery of higher-order topological insulators (HOTIs) provides the potential to overcome this dimensionality limitations by offering topological protection over an extended range of dimensionalities. Here we demonstrate 2D photonic HOTI (PHOTI) with topological states two dimensions lower than the one of the host system. We consider a photonic metacrystal of distorted Kagome lattice geometry that exhibits topological bulk polarization, leading to the emergence of 1D topological edge states and of higher order 0D states confined to the corners of the structure. Interestingly, in addition to corner states due to the nearest neighbour interactions and protected by generalized chiral symmetry 1, we discover and take advantage of a new class of topological corner states sustained by long-range interactions, available in wave-based systems, such as in photonics. Our findings demonstrate that photonic HOTIs possess richer physics compared to their condensed matter counterparts, offering opportunities for engineering novel designer electromagnetic states with unique topological robustness., Comment: 13 pages, 4 figures

In the past years classical wave-systems have constituted an excellent platform for emulating complex quantum phenomena. This approach has been especially fruitful in demonstrating topological phenomena in photonics and acoustics: from chiral edge states of Chern insulators and helical edge states of topological insulators to higher-dimensional topological states of quasiperiodic systems and systems with synthetic dimensions. Recently, a new class of topological states localized in more than one dimension of a D-dimensional system, referred to as higher-order topological (HOT) states, has been reported, offering an even more versatile platform to confine and control classical radiation and mechanical motion. However, because experimental research of HOT states has so far been limited to two-dimensional (2D) systems, third and higher-order states have evaded experimental observation. Studying higher-dimensional classical systems therefore opens an opportunity to emulate higher-order topological insulators and explore HOT states beyond second order. In this letter, we design and experimentally study a 3D acoustic metamaterial supporting third-order (0D) topological corner states, along with second-order (1D) edge states within the same topological bandgap, thus establishing a full hierarchy of HOT states in three dimensions, co-existing robustly within the same topological bandgap. The metamaterial is implemented over a versatile additive manufacturing platform, which enables rapid prototyping of metaatoms and metamolecules, which can be snapped together to form 3D metamaterials with complex geometries. The assembled 3D topological metamaterial represents the acoustic analogue of a pyrochlore lattice made of interconnected molecules, and is shown to exhibit topological bulk polarization, leading to the emergence of topological HOT states localized in all three or in two dimensions., Comment: 12 pages, 3 figures