In Situ Artificial Hybrid SEI Layer Enabled High-Performance Prelithiated SiOx Anode for Lithium-Ion Batteries

Considered the promising anode material for next-generation high-energy lithium-ion batteries, SiO_x has been slow to commercialize due to its low initial Coulombic efficiency (ICE) and unstable solid electrolyte interface (SEI) layer, which leads to reduced full-cell energy density, short cycling lives, and poor rate performance. Herein, a novel strategy is proposed to in situ construct an artificial hybrid SEI layer consisting of LiF and Li₃Sb on a prelithiated SiO_x anode via spontaneous chemical reaction with SbF₃. In addition to the increasing ICE (94.5%), the preformed artificial SEI layer with long-term cycle stability and enhanced Li⁺ transport capability enables a remarkable improvement in capacity retention and rate capability for modified SiO_x. Furthermore, the full cell using Li(Ni_0.8Co_0.1Mn_0.1)O₂ and a pre-treated anode exhibits high ICE (86.0%) and capacity retention (86.6%) after 100 cycles at 0.5 C. This study provides a fresh insight into how to obtain stable interface on a prelithiated SiO_x anode for high energy and long lifespan lithium-ion batteries.     

抑制图像非语义信息的通用后门防御策略

目的 后门攻击已成为目前卷积神经网络所面临的重要威胁。然而,当下的后门防御方法往往需要后门攻击和神经网络模型的一些先验知识,这限制了这些防御方法的应用场景。本文依托图像分类任务提出一种基于非语义信息抑制的后门防御方法,该方法不再需要相关的先验知识,只需要对网络的输入进行编解码处理就可以达到后门防御的目的。方法 核心思想是在保持图像语义不改变的同时,尽量削弱原始样本中与图像语义不相关的信息,以此抑制触发器。通过在待保护模型前添加一个即插即用的U型网络(即信息提纯网络)来实现对图像非语义信息的抑制。其输入是干净的初始样本,输出命名为强化样本。具体的训练过程中,首先用不同的训练超参数训练多个结构不一的干净分类器,然后在保持强化样本被上述分类器正确分类的前提下,优化信息提纯网络使强化样本和原始样本之间的差异尽可能地大。结果 实验在MNIST、CIFAR10和Image Net10数据集上进行。实验结果显示,经过信息提纯网络编解码后,干净样本的分类准确率略有下降,后门攻击成功率大幅降低,带有触发器的样本以接近干净样本的准确率被正确预测。结论 提出的非语义信息抑制防御方法能够在不需要相关先验知识的...     

Development,integration, and field evaluation of an autonomous citrus-harvesting robot

Citrus harvesting is a labor-intensive and time-intensive task. As the global population continues to age, labor costs are increasing dramatically. Therefore, the citrus-harvesting robot has attracted considerable attention from the business and academic communities. However, robotic harvesting in unstructured and natural citrus orchards remains a challenge. This study aims to address some challenges faced in commercializing citrus-harvesting robots. We present a fully integrated, autonomous, and innovative solution for citrus-harvesting robots to overcome the harvesting difficulties derived from the natural growth characteristics of citrus. This solution uses a fused simultaneous localization and mapping algorithm based on multiple sensors to perform high-precision localization and navigation for the robot in the field orchard. Besides, a novel visual method for estimating fruit poses is proposed to cope with the randomization of citrus growth orientations. Further, a new end-effector is designed to improve the success and conformity rate of citrus stem cutting. Finally, a fully autonomous harvesting robot system has been developed and integrated. Field evaluations showed that the robot could harvest citrus continuously with an overall success rate of 87.2% and an average picking time of 10.9 s/fruit. These efforts provide a solid foundation for the future commercialization of citrus-harvesting robots.     

Effect of Nano Potassium Fertilizer on Cucumber Amino Acid Component,Volatile Metabolite Components and GLN Family Genes Expression

Wireless Personal Communications - To investigate the effect of nano-potassium silicate on the expression of glutamine (GLN) family gene, amino acid (AAs) componentand volatile metabolites (VMs)...     

Global contextual guided residual attention network for salient object detection

High-level semantic features and low-level detail features matter for salient object detection in fully convolutional neural networks (FCNs). Further integration of low-level and high-level features increases the ability to map salient object features. In addition, different channels in the same feature are not of equal importance to saliency detection. In this paper, we propose a residual attention learning strategy and a multistage refinement mechanism to gradually refine the coarse prediction in a scale-by-scale manner. First, a global information complementary (GIC) module is designed by integrating low-level detailed features and high-level semantic features. Second, to extract multiscale features of the same layer, a multiscale parallel convolutional (MPC) module is employed. Afterwards, we present a residual attention mechanism module (RAM) to receive the feature maps of adjacent stages, which are from the hybrid feature cascaded aggregation (HFCA) module. The HFCA aims to enhance feature maps, which reduce the loss of spatial details and the impact of varying the shape, scale and position of the object. Finally, we adopt multiscale cross-entropy loss to guide network learning salient features. Experimental results on six benchmark datasets demonstrate that the proposed method significantly outperforms 15 state-of-the-art methods under various evaluation metrics.

     

Characterisation of sol–gel method synthesised MgZnFe2 O4 nanoparticles and its cytotoxic effects on breast cancer cell line,MDA MB‐231 in vitro

In this study, nanocrystalline magnesium zinc ferrite nanoparticles were successfully prepared by a simple sol–gel method using copper nitrate and ferric nitrate as raw materials. The calcined samples were characterised by differential thermal analysis/thermogravimetric analysis, Fourier transform infrared spectroscopy and X‐ray diffraction. Transmission electron microscopy revealed that the average particle size of the calcined sample was in a range of 17–41 nm with an average of 29 nm and has spherical size. A cytotoxicity test was performed on human breast cancer cells (MDA MB‐231) and (MCF‐7) at various concentrations starting from (0 µg/ml) to (800 µg/ml). The sample possessed a mild toxic effect toward MDA MB‐231 and MCF‐7 after being examined with MTT (3‐[4, 5‐dimethylthiazol‐2‐yl]‐2, 5 diphenyltetrazolium bromide) assay for up to 72 h of incubation. Higher reduction of cells viability was observed as the concentration of sample was increased in MDA MB‐231 cell line than in MCF‐7. Therefore, further cytotoxicity tests were performed on MDA MB‐231 cell line.Inspec keywords: sol‐gel processing, nanoparticles, nanofabrication, magnesium compounds, zinc compounds, toxicology, biological organs, cancer, cellular biophysics, nanomedicine, calcination, differential thermal analysis, Fourier transform infrared spectra, X‐ray diffraction, transmission electron microscopy, particle size, organic compoundsOther keywords: sol‐gel method, cytotoxic effects, breast cancer cell line, MDA MB‐231 in vitro, nanocrystalline magnesium zinc ferrite nanoparticles, copper nitrate, ferric nitrate, raw materials, calcined samples, differential thermal analysis, thermogravimetric analysis, Fourier transform infrared spectroscopy, X‐ray diffraction, transmission electron microscopy, average particle size, cytotoxicity testing, human breast cancer cells, mild toxic effect, 3‐[4,5‐dimethylthiazol‐2‐yl]‐2,5 diphenyltetrazolium bromide) assay, cell viability, MCF‐7, MDA MB‐231 cell line, size 17 nm to 41 nm     

Sub‐Micron Anisotropic InP‐based III–V Semiconductor Material Deep Etching for On‐Chip Laser Photonics Devices

Two InP‐based III–V semiconductor etching recipes are presented for fabrication of on‐chip laser photonic devices. Using inductively coupled plasma system with a methane free gas chemistry of chlorine and nitrogen at a high substrate temperature of 250 °C, high aspect ratio, anisotropic InP‐based nano‐structures are etched. Scanning electron microscopy images show vertical sidewall profile of 90° ± 3°, with aspect ratio as high as 10. Atomic Force microscopy measures a smooth sidewall roughness root‐mean‐square of 2.60 nm over a 3 × 3 μm scan area. The smallest feature size etched in this work is a nano‐ring with inner diameter of 240 nm. The etching recipe and critical factors such as chamber pressure and the carrier plate effect are discussed. The second recipe is of low temperature (?10 °C) using Cl₂ and BCl₃ chemistry. This recipe is useful for etching large areas of III–V to reveal the underlying substrate. The availability of these two recipes has created a flexible III–V etching platform for fabrication of on‐chip laser photonic devices. As an application example, anisotropic InP‐based waveguides of 3 μm width are fabricated using the Cl₂ and N₂ etch recipe and waveguide loss of 4.5 dB mm^?1 is obtained.

     

Dendrite‐Free Lithium Deposition via Flexible‐Rigid Coupling Composite Network for LiNi0.5Mn1.5O4/Li Metal Batteries

Notorious lithium dendrite causes severe capacity fade and harsh safety issues of lithium metal batteries, which hinder the practical applications of lithium metal electrodes in higher energy rechargeable batteries. Here, a kind of 3D‐cross‐linked composite network is successfully employed as a flexible‐rigid coupling protective layer on a lithium metal electrode. During the plating/stripping process, the composite protective layer would enable uniform distribution of lithium ions in the adjacent regions of the lithium electrode, resulting in a dendrite‐free deposition at a current density of 2 mA cm^?2. The LiNi_0.5Mn_1.5O₄‐based lithium metal battery presents an excellent cycling stability at a voltage range of 3.5–5.0 V with the induction of 3D‐cross‐linked composite protective layer. From an industrial field application of view, thin lithium metal electrodes (40 µm, with 4 times excess lithium) can be used in LiNi_0.5Mn_1.5O₄ (with industrially significant loading of 18 mg cm^?2 and 2.6 mAh cm^?2)‐based lithium metal batteries, which reveals a promising opportunity for practical applicability in high energy lithium metal batteries.     

Erratum to: Charge transfer in graphene/polymer interfaces for CO2 detection

Nano Research - The order of the authors in the original version of this article was unfortunately incorrect on the first page and the first page of the ESM. Instead of Myungwoo Son1, Yusin Pak1,...