No abstract

Hole transport layers (HTLs) play a crucial role in the efficiency and stability of perovskite solar cells (PSCs). The most efficient PSCs based on spiro-OMeTAD (Spiro) generally have stability problems. Here, NiO x /Spiro HTL has been designed and implemented by combining the advantages of these two films. The results indicated that a device based on a NiO x /Spiro HTL has faster hole extraction ability and better energy alignment than that of a pure Spiro device, thus improving the PCE from 19.8 to 21.66%. Compared with the 60% initial efficiency of Spiro-based devices, the NiO x /Spiro bilayer devices have higher stability and maintain 90% initial efficiency over 1200 h. In this work, NiO x is applied to perovskite devices with N−I−P configuration, which provides a possible mitigation strategy to reduce the V OC deficit for efficient and stable devices.

Pattern recognition receptors (PRRs) and their signaling pathways have essential roles in recognizing various components of pathogens as well as damaged cells and triggering inflammatory responses that eliminate invading microorganisms and damaged cells. The zebrafish relies heavily on these primary defense mechanisms against pathogens. Here, we review the major PRR signaling pathways in the zebrafish innate immune system and compare these signaling pathways in zebrafish and humans to reveal their evolutionary relationship and better understand their innate immune defense mechanisms.

There appears to be a controversy on whether remnant PbI 2 is beneficial to the performance of perovskite solar cells (PSCs). We have shown that PSCs with residual PbI 2 deposited by one-step antisolvent solution and two-step evaporationsolution method both have shown better performance than those without excess PbI 2 . X-ray diffraction with diverse X-ray incident angles combined with scanning electron microscopy and secondary-ion mass spectrometry is employed to identify the position of remnant PbI 2 . It reveals that residual PbI 2 is located at grain boundaries near the perovskite/hole-transporting layer interface area for the one-step antisolvent solution method, and the two-step evaporation-solution method situates the excess PbI 2 at grain boundaries and the electron transport layer/perovskite interface. The cell performance implies that grain boundary passivation is beneficial for promoting short-circuit current density, while interface passivation is more favorable to enhance open-circuit voltage and fill factor. The revealed passivation process indicates a deep understanding of remnant PbI 2 and contributes to the development of PSCs.

In this work, we use periodic multilayered structures as scaffolds in order to magnify the effect of both the scaffold and the electron selective layer in perovskite solar cells, and understand their influence on cell performance.

No abstract

Deep neural networks provide unprecedented performance gains in many real world problems in signal and image processing. Despite these gains, future development and practical deployment of deep networks is hindered by their blackbox nature, i.e., lack of interpretability, and by the need for very large training sets. An emerging technique called algorithm unrolling or unfolding offers promise in eliminating these issues by providing a concrete and systematic connection between iterative algorithms that are used widely in signal processing and deep neural networks. Unrolling methods were first proposed to develop fast neural network approximations for sparse coding. More recently, this direction has attracted enormous attention and is rapidly growing both in theoretic investigations and practical applications. The growing popularity of unrolled deep networks is due in part to their potential in developing efficient, high-performance and yet interpretable network architectures from reasonable size training sets. In this article, we review algorithm unrolling for signal and image processing. We extensively cover popular techniques for algorithm unrolling in various domains of signal and image processing including imaging, vision and recognition, and speech processing. By reviewing previous works, we reveal the connections between iterative algorithms and neural networks and present recent theoretical results. Finally, we provide a discussion on current limitations of unrolling and suggest possible future research directions.

No abstract