Conductive electrodes and electric circuits that can remain active and electrically stable under large mechanical deformations are highly desirable for applications such as flexible displays, field-effect transistors, energy-related devices, smart clothing and actuators. However, high conductivity and stretchability seem to be mutually exclusive parameters. The most promising solution to this problem has been to use one-dimensional nanostructures such as carbon nanotubes and metal nanowires coated on a stretchable fabric, metal stripes with a wavy geometry, composite elastomers embedding conductive fillers and interpenetrating networks of a liquid metal and rubber. At present, the conductivity values at large strains remain too low to satisfy requirements for practical applications. Moreover, the ability to make arbitrary patterns over large areas is also desirable. Here, we introduce a conductive composite mat of silver nanoparticles and rubber fibres that allows the formation of highly stretchable circuits through a fabrication process that is compatible with any substrate and scalable for large-area applications. A silver nanoparticle precursor is absorbed in electrospun poly (styrene-block-butadiene-block-styrene) (SBS) rubber fibres and then converted into silver nanoparticles directly in the fibre mat. Percolation of the silver nanoparticles inside the fibres leads to a high bulk conductivity, which is preserved at large deformations (σ ≈ 2,200 S cm(-1) at 100% strain for a 150-µm-thick mat). We design electric circuits directly on the electrospun fibre mat by nozzle printing, inkjet printing and spray printing of the precursor solution and fabricate a highly stretchable antenna, a strain sensor and a highly stretchable light-emitting diode as examples of applications.

Conjugated polymers, such as polyfluorene and poly(phenylene vinylene), have been used to selectively disperse semiconducting single-walled carbon nanotubes (sc-sWnTs), but these polymers have limited applications in transistors and solar cells. Regioregular poly(3-alkylthiophene)s (rr-P3ATs) are the most widely used materials for organic electronics and have been observed to wrap around sWnTs. However, no sorting of sc-sWnTs has been achieved before. Here we report the application of rr-P3ATs to sort sc-sWnTs. Through rational selection of polymers, solvent and temperature, we achieved highly selective dispersion of sc-sWnTs. our approach enables direct film preparation after a simple centrifugation step. using the sorted sc-sWnTs, we fabricate high-performance sWnT network transistors with observed charge-carrier mobility as high as 12 cm 2 V − 1 s − 1 and on/off ratio of > 10 6 . our method offers a facile and a scalable route for separating sc-sWnTs and fabrication of electronic devices.

Piezomaterials are known to display enhanced energy conversion efficiency at nanoscale due to geometrical effect and improved mechanical properties. Although piezoelectric nanowires have been the most widely and dominantly researched structure for this application, there only exist a limited number of piezomaterials that can be easily manufactured into nanowires, thus, developing effective and reliable means of preparing nanostructures from a wide variety of piezomaterials is essential for the advancement of self-powered nanotechnology. In this study, we present nanoporous arrays of polyvinylidene fluoride (PVDF), fabricated by a lithography-free, template-assisted preparation method, as an effective alternative to nanowires for robust piezoelectric nanogenerators. We further demonstrate that our porous PVDF nanogenerators produce the rectified power density of 0.17 mW/cm3 with the piezoelectric potential and the piezoelectric current enhanced to be 5.2 times and 6 times those from bulk PVDF film nanogenerators under the same sonic-input.

A stretchable polymer channel layer for organic field-effect transistors is obtained by spin-coating a blend solution of polythiophene and rubber polymer. A network of the polythiophene nanofibril bundles surface-embedded in the rubber matrix allows large stretchability of the polythiophene film layer.

Deep neural networks with millions of parameters may suffer from poor generalization due to overfitting. To mitigate the issue, we propose a new regularization method that penalizes the predictive distribution between similar samples. In particular, we distill the predictive distribution between different samples of the same label during training. This results in regularizing the dark knowledge (i.e., the knowledge on wrong predictions) of a single network (i.e., a self-knowledge distillation) by forcing it to produce more meaningful and consistent predictions in a class-wise manner. Consequently, it mitigates overconfident predictions and reduces intra-class variations. Our experimental results on various image classification tasks demonstrate that the simple yet powerful method can significantly improve not only the generalization ability but also the calibration performance of modern convolutional neural networks.