集成电路多层结构中的化学机械抛光技术

化学机械抛光技术（ＣＭＰ）已成功地应用于集成电路多层结构中的介质层和金属层的全局平面化。这是唯一能对亚微米经器件提供全局平面化的技术，对０．３５μｍ及以下的器件工艺是绝对必须的。ＣＭＰ面临的问题主要是难以维持高效的、稳定的、一次通过性的生产运转。     

新型前栅极CNT-FED的研究

传统的前栅极场致发射显示板由于介质层和栅极的制作在生成或制作阴极场致发射源之后，在制造过程中容易破坏场致发射源；另外阴极发射对介质层厚度、阳极电压和调制极开口等参数非常敏感，所以器件的发射均匀性难以保证。为了解决这些问题，引入了类似HOP玻璃的结构。阴极上只需要丝印碳纳米管，无需制作介质层和栅极，解决了场致发射源被破坏的问题。阴极与栅极之间真空取代了介质层，由于真空的绝缘性能高于介质层，所以阴极与栅极之间的距离可以减小，栅极的调制效果更显著。由于玻璃的平整度高于一般方法制作的介质层的平整度，所以器件的发射均匀性比较好。     

P—InPMIS结构界面陷阱的深能级研究

对经ＰＥＣＶＤ生长的Ｐ－ＩｎＰＭＩＳ结构的界面陷阱进行了研究。样品介质模生长是在特定条件下进行的，分别利用Ｃ－Ｖ和ＤＬＴＳ技术进行研究。结果表明，结果表明，在介质膜和ＩｎＰ之间的ＩｎＰ之间的ＩｎＰ－侧有界面陷阱存在，并获得了与之有关的深能级参数。这些陷阱可能是不同生长条件的介质膜淀积过程中等离体引进的有关辐照损伤。     

光纤接入网组网及保护方式探讨

将接入网分为主干层，分配层和引入层三个层面，并采用灵活点ＦＰ和分配点ＤＰ的概念，对接入网的组网及其业务保护方式有一些深入的探讨。提出了“次环形”和“五环形”拓扑结构的概念，对主干层和分配层光缆的组网方式，带状光缆选用，敷设芯数，业务保护方式等提出了建议。     

光子扫描隧道显微镜的四层介质理论

光子扫描隧道显微镜的四层介质理论夏德宽（大连理工大学物理系大连１１６０２３）詹卫伸（大连轻工业学院基础部大连１１６０３４）光子扫描隧道显微镜（ＰＳＴＭ）是９０年代初发展起来的新型表面探测仪器。但目前的探测理论是三层介质理论，不能讨论与被测样品的厚度和...     

彩色交流等离子体显示板辉光扩展的研究

交流等离子体显示板（简称ＡＣＰＤＰ）的显示像元的放电是通过气体作为耦合介质，电极间是通过电介质作为耦合介质的，因而属于容性结构，放电较为复杂。理论上描述电流和器件特性较为困难，国外曾经从实验中作过某些定性的讨论。本文从理论上分析了ＡＣＰＤＰ放电气体与介质、介质与发光像元间的相互作用及发光容量不同时发光像元之间的影响以及所产生的辉光扩展，成功地解释了实验现象。     

多频段天线罩功率传输系数的优化设计

按射线跟踪法分析天线罩的理论,把罩壁近似为均匀多层介质平板,对其传输特性进行了分析;根据天线罩须在多个频率段上有较高传输系统的特殊要求,按最大最小原则建立了二重优化模型,采用ＳＵＭＴ外点法和ＰＯＷＥＬＬ法相结合对天线罩的各层介质壁厚进行优化设计;对某型２０／３０ＧＨｚ双频段卫星通信天线罩的优化达到了预期效果。     

人体浑浊软介质和生物介质对激光和其它辐射透过率的提高

讨论并用实验研究了人体浑浊介质及生物介质一厚度高于特征衰减深度若千倍的人体组织和散射介质层——对激光和其它辐射的透过率。在挤压和刺穿软的散射介质时，观察到透过率急剧提高。发现压迫人体浑浊介质层时，透明度远高于对模拟生物组织局部加压的情况,这与血液消失以及把肌肉从加压处挤开有关。对无线电辐射——带电粒子、X射线及伽马射线等，也同样观察到散射和吸收的降低。指出可能应用本研究结果作深度辐照，以抑制内部病变（如脊髓)，治疗神经性传染病(侧面肌肉萎缩硬化、播散性硬化、脊髓灰白质炎等)，也可用于治疗肿瘤、溃疡等。     

HfO2／Si（001）界面层的TEM研究

随着微电子技术的迅猛发展，集成电路的集成度不断增大，器件的尺寸不断缩小。当MOSFET尺寸缩小到0．1nm的尺度以下时，栅氧化层的等效厚度（在保持栅电容值不变的条件下，以相对介电常数为3．9的SiO2作为标准得到的栅介质层厚度）需要小于3nm。如仍采用传统的氧化硅作为栅氧化层介质，电子的直接隧穿效应和栅介质层所承受的电场将变得很大，由此引起栅介质的漏电流增大和可靠性下降等严重问题，严重阻碍了MOS器件的进一步发展。因此，人们提出了采用高介电常数的栅介质（通常称为高K栅介质）替代传统氧化硅的解决方法。利用高K介质材料替代传统氧化硅作为栅介质，可以在保持等效厚度不变的条件下，增加介质层的物理厚度，因而可大大减小直接隧穿效应和栅介质层承受的电场强度。新型高K栅介质研究已成为国际微电子领域的热门研究课题之一。     

体积全息图的偏振特性

本文用多层介质膜系的特征 阵,讨论了体积全息的偏特性。用计算机模拟方法分析了体积全息图对Ｐ偏振和Ｓ偏振再现的光谱响应。并把理论分析与实验结果进行了比较。     

The effect of device geometry on lateral mode content of stripe geometry lasers

The dielectric profile of stripe geometry injection lasers is modeled with an objective of structure design requirements for fundamental lateral mode operation. Heterostructure lasers are modeled with a dielectric step profile using an effective dielectric discontinuity based on the gain/loss profile of the active layer as well as the overall geometrical structure. The analysis provides a quantitative comparison of the performance of two important double-heterostructure lasers: 1) the oxide-stripe geometry laser and 2) the channeled-substrate planar (CSP) laser. Modes of oxide-stripe lasers have lateral gain confinement, whereas, modes of CSP devices have strong lateral index confinement. To isolate the influence of geometry on the effective dielectric profile we assume that the real refractive index of the active layer is position independent. Resulting calculations show that a stripe geometry laser inherently has a depressed effective index in the active region below the metallic contact. This phenomenon alone produces index anti-guiding. In actual devices, both geometry and free carrier injection into the active region produce lateral index antiguiding. Lateral mode cut-off conditions are calculated as functions of the effective complex dielectric step and the stripe width. The results show that cutoff is related in a unique fashion to the ratio of the real and imaginary parts of the complex dielectric step; the ratio is positive for index guided modes and negative for gain guided ones.     

Reliability of InGaAsP light emitting diodes at high current density

InGaAsP LED's emitting at 1.3 µm are attractive sources for long distance transmission systems. These devices have been shown to have excellent reliability at 8 kA/cm². In order to launch more power into the optical fiber it is desirable to operate at the highest possible current densities consistent with system reliability requirements. In this work, the high temperature aging behavior of these LED's has been studied at current densities from 20 to 40 kA/². Both the standard LED structure, where the small diameter p-contact is isolated with a dielectric layer, and a structure in which a Schottky barrier is used for isolation are examined. Dark spot defect (DSD) formation is greatly enhanced at these high current densities in devices with dielectric isolation, limiting MTTF atT_{J} = 70degC to 2 times 10^{5}h. In contrast, devices with Schottky-barrier isolation remain essentially free of DSD's and haveMTTF > 10^{7}h at 70°C. These results suggest that stress from the dielectric layer promotes the growth of DSD's. Schottky-barrier devices in which the dielectric layer is eliminated are, thus, better suited for high current-density operation.     

Simulation of a periodic dielectric corrugation with an equivalent anisotropic layer

A periodic, corrugated, dielectric layer is simulated by an anisotropic dielectric layer of equal thickness. The tensor elements of the equivalent dielectric layer are given in terms of the permittivity of the dielectric material, the period of the surface corrugation, and the width of the corrugations. The validity of this technique is verified by comparing the reflection coefficient of the equivalent layer with that of the corresponding corrugated layer using the moment method. Employing a multiple layer approach, the technique is extended to handle periodic surfaces with arbitrary cross sections which can be used to design millimeter wave dielectric plate polarizers and absorbers.     

High-Performance MIM Capacitors Using HfLaO-Based Dielectrics

Metal-insulator-metal (MIM) capacitors fabricated with (8%) La-doped HfO₂ single layer as well as HfLaO/ LaAlO₃/HfLaO multilayer dielectric stack are demonstrated. While the La-doped HfO₂ single layer is crystallized at 420^°C annealing, HfLaO/LaAlO₃/HfLaO multilayer dielectric stack remains amorphous. A high dielectric-constant value of 38 can be obtained when 8% La-doped HfO₂ is crystallized into cubiclike structure. However, it is observed that the linearity of MIM capacitor is degraded upon crystallization. The multilayer film has lower average dielectric constant but shows low quadratic voltage linearity of less than 1000 ppm/V² up to a capacitance density of 9 fF/?m² . It is observed that the HfLaO single-layer MIM is suitable for the applications with requirements of high capacitance density and robust reliability, while the multilayer MIM is suitable for a precision circuit.     

厚介质防护层下天线方向图恶化机理分析及解决办法

定性的讨论了厚介质防护层下天线方向图的恶化机理,并结合电磁软表面提出了改善天线辐射性能的方法。将厚介质防护层等效为多层不同数值的阻抗层。通过分析电磁波和表面波的传播路径,解释了厚介质防护层下天线方向图恶化的原因。通过增加电磁软表面,扼制了表面波的传播及其引起的辐射,降低了厚介质防护层对天线方向图的影响。方向图测试结果显示,利用电磁软表面抑制表面波的传播,可以有效改善厚介质防护层下天线的方向图,在高超音速飞行器通信领域具有一定的应用前景。     

导体表面有耗介质层中垂直电偶极子激励的场

本文导了有耗介质层内垂直电偶极子激励场的表达式，通过积分路径变换的方法，对表达式中的索档菲积分进行数值。计算了激励场在介质层内外的分布曲线，并分析了介质损耗及厚度对场分布的影响。     

Coupling and radiation through “V-shaped” narrow slotsusing the FDTD method

A “V-shaped” slot in the narrow wall of a rectangular waveguide is presented as a coupling element. Three- and four-port coupler parameters and the radiation characteristics of the slot were investigated using the finite-difference time-domain (FDTD) method. Two approximate numerical-analytical procedures are proposed to reduce the computer requirements and predict multislot device characteristics. A clear improvement in four-port coupler parameters was observed following the incorporation of a thin dielectric layer     

电荷俘获存储器阻挡层研究进展

随着非挥发性存储器(NVM)存储单元的特征尺寸进入20 nm节点,使用单层SiO2作为阻挡层的传统电荷俘获存储器结构性能上逐渐受到限制。基于阻挡层在存储器栅堆栈中的作用与基本要求,首先,指出单层SiO2作为阻挡层存在的主要问题,然后对高介电常数材料作为阻挡层时,其禁带宽度、介电常数、内部的缺陷密度以及退火工艺等方面对存储特性的影响进行了分析,同时对近年来研究较多的阻挡层能带工程进行了详细介绍,如SiO2和Al2O3的复合阻挡层结构、多层高介电常数材料的阻挡层结构等。最后,对目前研究进展中存在的问题以及未来的研究方向和趋势进行了总结和展望。     

Employing vertical dielectric layers to improve the operation performance of flash memory devices

A number of new device structures have been reported recently to improve the operation performance of flash memory. In this work, a novel flash device with a vertical dielectric layer in the depletion region is proposed through simulation approach. The simulation results show that the employment of a vertical dielectric layer in the depletion region can improve the operation performance of flash memory. The improvement can be attributed to a lower potential in the central region of device channel and the increase of the potential drop in the channel direction near drain junction. Thus, this proposed vertical dielectric layer increases the electrical field of the channel and thus the probability and the momentum of electron injection. The operation characteristics of the flash device with a vertical dielectric layer in the depletion region of source and drain are superior to those without. In addition, it is found that a vertical dielectric layer with lower dielectric constant can enhance the operation performance of flash device even more.     

The importance of insulator properties in a thin-film electroluminescent device

AC-coupled thin-film electroluminescent devices employ insulating layers to limit the current density through the active layers. The requirements and limitations in driving such devices are then dependent upon the properties of the dielectric layers. We define an external efficiency for an ac thin-film electroluminescent device and consider two limiting cases, current response 1) fast, and 2) slow, compared to an applied voltage change. Relationships are then derived among external efficiency, relative luminance, and the two principal characteristics of the insulating layer, specifically charge density at breakdown and thickness. It is shown, for example, that charge density at breakdown for the insulators should be at least three times the corresponding quantity for the active layer.