Controlled filamentation instability as a scalable fabrication approach to flexible metamaterials.

Long and flexible arrays of nanowires find impactful applications in sensing, photonics, and energy harvesting. Conventional manufacturing relies largely on lithographic methods limited in wafer size, rigidity, and machine write time. Here, we report a scalable process to generate encapsulated flexible nanowire arrays with high aspect ratios and excellent tunable size and periodicity. Our strategy is to control nanowire self-assembly into 2D and 3D architectures via the filamentation of a textured thin film under anisotropic stretching. This is achieved by coupling soft lithography, glancing angle deposition, and thermal drawing to obtain well-ordered meters-long nanowires with diameters down to 50 nanometers. We demonstrate that the nanowire diameter and period of the array can be decoupled and manipulated independently. We propose a filamentation criterion and perform numerical simulations implementing destabilizing long-range Van der Waals interactions. Applied to high-index chalcogenide glasses, we show that this decoupling allows for tuning diffraction. Finally, harnessing Mie resonance, we demonstrate the possibility of manufacturing macroscopic meta-grating superstructures for nanophotonic applications.

Novel insights into the design of stretchable electrical systems.

Soft electronics have recently gathered considerable interest because of their biomechanical compatibility. An important feature of deformable conductors is their electrical response to strain. While development of stretchable materials with high gauge factors has attracted considerable attention, there is a growing need for stretchable conductors whose response to deformation can be accurately engineered to provide arbitrary resistance-strain relationships. Rare studies addressing this issue have focused on deterministic geometries of single rigid materials, limiting the scope of these strategies. We introduce the novel concept of periodic stretchable patterns combining multiple conductive materials to produce tailored responses. Using shortest path algorithms, we establish a computationally efficient selection method to obtain the required resistance-strain relationship. Using this algorithm, we identify and experimentally demonstrate constant resistance-strain responses up to 50% elongation using a single microtextured material. Last, we demonstrate counterintuitive sinusoidal responses by integrating three materials, with interesting applications in sensing and soft robotics.

Publisher Correction: Self-assembly of nanostructured glass metasurfaces via templated fluid instabilities.

In the version of this Article originally published, the volume, article number and year of ref. 32 were incorrect; they should have read 31, 1802348 (2019). This has now been corrected.

Self-assembly of nanostructured glass metasurfaces via templated fluid instabilities.

Modern devices require the tuning of the size, shape and spatial arrangement of nano-objects and their assemblies with nanometre-scale precision, over large-area and sometimes soft substrates. Such stringent requirements are beyond the reach of conventional lithographic techniques or self-assembly approaches. Here, we show nanoscale control over the fluid instabilities of optical thin glass films for the fabrication of self-assembled all-dielectric optical metasurfaces. We show and model the tailoring of the position, shape and size of nano-objects with feature sizes below 100 nm and with interparticle distances down to 10 nm. This approach can generate optical nanostructures over rigid and soft substrates that are more than tens of centimetres in size, with optical performance and resolution on a par with advanced traditional lithography-based processes. To underline the potential of our approach, which reconciles high-performance optical metasurfaces and simple self-assembly fabrication approaches, we demonstrate experimentally and via numerical simulation sharp Fano resonances with a quality factor, Q, as high as ∼300 in the visible for all-dielectric nanostructures, to realize protein monolayer detection.