Density Functional Theory Study of the Point Defects on KDP (100) and (101) Surfaces

Surface defects are usually associated with the formation of other forms of expansion defects in crystals, which have an impact on the crystals’ growth quality and optical properties. Thereby, the structure, stability, and electronic structure of the hydrogen and oxygen vacancy defects (V_H and V_O) on the (100) and (101) growth surfaces of KDP crystals were studied by using density functional theory. The effects of acidic and alkaline environments on the structure and properties of surface defects were also discussed. It has been found that the considered vacancy defects have different properties on the (100) and (101) surfaces, especially those that have been reported in the bulk KDP crystals. The (100) surface has a strong tolerance for surface V_H and V_O defects, while the V_O defect causes a large lattice relaxation on the (101) surface and introduces a deep defect level in the band gap, which damages the optical properties of KDP crystals. In addition, the results show that the acidic environment is conducive to the repair of the V_H defects on the surface and can eliminate the defect states introduced by the surface V_O defects, which is conducive to improving the quality of the crystal surface and reducing the defect density. Our study opens up a new way to understand the structure and properties of surface defects in KDP crystals, which are different from the bulk phase, and also provides a theoretical basis for experimentally regulating the surface defects in KDP crystals through an acidic environment.     

HeH+ under Spatial Confinement

In the present study, the influence of spatial confinement on the bond length as well as dipole moment, polarizability and (hyper)polarizabilities of HeH+ ion was analyzed. The effect of spatial confinement was modelled by cylindrically symmetric harmonic oscillator potential, that can be used to mimic high pressure conditions. Based on the conducted research it was found that the spatial confinement significantly affects the investigated properties. Increasing the confinement strength leads to a substantial decrease of their values. This work may be of particular interest for astrochemistry as HeH+ is believed to be the first compound to form in the Universe.     

基于改进VGG网络的单体热电池X光图像无损检测方法研究

为解决单体热电池生产中出现的安装错误、人工检测耗时耗力的问题,提出一个结合迁移学习和卷积 神 经网络(convolutional neural network,CNN) 的单体热电池缺陷检测模型。首先,对数据集图像进行裁剪、加噪等预处理,以VGG16(visual geometry group 16) 网络作为 模型的骨干架构,在瓶颈层后增添选择性核(selective kernel,SK) 卷积;然后,增添全局平均池化(global average pooling,GAP) 层, 增加Dropout层及添加 L2 正则化等微调操作,得到单体热电池缺陷检测模型Q-VGGNet;最后,在大型公开数据集ImageNet上进 行预训练学习,将获得的权重参数迁移到单体热电池图像识别模型Q-VGGNet上。测试实验表明:6种 网络模型对数据集缺陷图像的总体识别准确率分别达到了98.39%、94.44%、97.27%、96.34%、93.71%、 95.61%,Q-VGGNet网 络模型 对合格图像和 漏装负极、极耳断裂、漏装集流片3种缺陷图像 识别准确率 分别达到了99.6%,95.9%,99.6%和98.4%。检测结果表明:该方法能够更准确、快速地检测热电池缺陷, 拥有良好的缺陷诊断能力,较传统方法提高近3%,为人工检测单体热电池缺陷提供了良好的解决途径。     

基于双通道注意力的轻量化PCB缺陷检测研究

印刷电路板 (printed circuit board,PCB)在实际生产过程中存在缺陷样式多种多样、缺陷小、缺陷位置难以定位的问题,而一个巨大的模型难以实现实时检测的要求,且大量的深度可分离卷积层建立的轻量级模型也不能达到足够的精度,为此提出一种基于YOLOv5s的PCB缺陷检测算法。 将原始Backbone的Conv模块跟C3模块用GhostConv替换,在Neck部分则引入了一种新的轻量级卷积技术GSConv,减轻模型大小的同时保持精度,GSConv在模型的准确性和速度之间完成了一个极好的权衡,针对许多注意力模块无法关注全局信息同时模型大的问题,提出了多尺度的轻量化双通道注意力模块(double channel depthwise attention module,DWAM),进一步提高模型精度。通过多组实验, 结果表明,改进算法所有类别的平均mAP为99.14%,且模型的GFLOPs为7.194 G,Params为7.175,原始的YOLOv5s平均mAP为96.86%,GFLOPs为6.89 G,Params为6.596,虽然Params以及GFLOPs有所增大,但是还是满足轻量网络的要求,并且精度相对于YOLOv5s提高了2.25%,且对于每个类别的缺陷识别准确率都有改善,大幅减少计算量和模型参数的同时保证了准确率,满足工业检测生产需求的同时便于移动端部署。     

基于LSDANet的手机芯片屏蔽壳表面缺陷检测方法

为了解决手机芯片屏蔽壳表面白印缺陷微小、尺度各异等因素影响检测快速性和准确性的问题,本文提出一种基于长短连接通路和双注意力网络(long short link and double attention network, LSDANet)的手机芯片屏蔽壳表面缺陷检测方法。首先,通过构建基于编码和解码的语义分割模型和利用长短距离连接通路,提高网络模型对尺度各异缺陷的特征提取能力。其次,分别设计基于通道和空间的注意力机制,增大5—10 pixel尺寸的白印缺陷在空间和通道上的特征权重。最后,融合双注意力机制和长短距离连接通路分割模型,构建LSDANet缺陷检测网络,应用于手机芯片屏蔽壳表面缺陷检测。实验数据表明,LSDANet网络能够达到96.21%的平均像素精度、66.13%的平均交并比和39.03的每秒检测帧数,相比多种语义分割算法均具有更高的检测精度和速度。     

硼磷硅玻璃缺陷分析与研究

在晶圆制造中广泛运用掺杂有硼和磷的二氧化硅即硼磷硅玻璃作为绝缘介质,一般用于金属布线前的绝缘层。掺杂硼磷的作用是降低回流温度,减少热预算。但在实际运用中,由于硼磷硅玻璃性质不够稳定,常常受到环境的影响,其中的三氧化二硼和五氧化二磷容易和空气中的水汽反应生成硼酸和磷酸继而形成缺陷,造成电路短路,最终导致产品低良率,给公司造成损失。文章就硼磷硅玻璃受环境影响的异常现象进行分析研究,找到缺陷形成的根本原因及预防方法。     

光梯度力与激光捕获

从电偶极子在非均匀电磁场中所受梯度力出发,用光的电磁理论和电解质小球的电偶极子模型,论述了介质小球在光场中所受光梯度力与光强梯度成正比,并以此来定量说明激光对介质小球的捕获原理.     

一种小型化宽频带卫星导航终端天线的设计

研究了一种小型化宽频带交叉偶极子结构卫星导航终端天线。天线结构由交叉偶极子臂、耦合贴片和侧面开斜槽的反射腔组成,反射腔的侧面斜槽可调节电流分布,实现了天线宽阻抗带宽、宽轴比带宽和尺寸的小型化。测量结果表明,■dB阻抗带宽实测为0.96～2.12 GHz,轴比≤3 dB的圆极化带宽为1.06～1.84 GHz,在北斗卫星导航系统B1、B2、B3频段中心频率点的实测增益分别为6.07 dBi、5.25 dBi、6.2 dBi。该天线频带宽、结构简单,不需要复杂的圆极化馈电网络,适合批量生产。     

一维光子晶体微腔在硅基材料发光中的应用研究

硅基材料的高效发光对未来硅基光电子集成的发展极其关键,含微腔的一维光子晶体可以显著提高其发光强度、窄化其发光峰.介绍了几种硅基材料发光的一维光子晶体微腔结构,包括单缺陷模式的对称与非对称结构、多缺陷模结构及电注入结构.利用传输矩阵法计算其缺陷模透射谱,以间接分析其发光谱.     

Defect Engineering on Boron Nitride for O2 Activation and Subsequent Oxidative Desulfurization

The rational design of highly active hexagonal boron nitride (h-BN) catalysts at the atomic level is urgent for aerobic reactions. Herein, a doping impurity atom strategy is adopted to increase its catalytic activities. A series of doping systems involving O, C impurities and B, N antisites are constructed and their catalytic activities for molecular O₂ have been studied by density functional theory (DFT) calculations. It is demonstrated that O₂ is highly activated on O_N and B_N defects, and moderately activated on C_B and C_N defects, however, it is not stable on N_B and O_B defects. The subsequent application in oxidative desulfurization (ODS) reactions proves the O_N and C-doped (C_B, C_N) systems to be good choice for sulfocompounds oxidization, especially for dibenzothiophene (DBT). While the B_N antisite is not suitable for such aerobic reaction due to the extremely stable B−O^*−B species formed during the oxidation process.