石英晶片外观缺陷对频率的影响

研究了石英晶片外观缺陷对频率的影响。石英晶片其自身存在的外观缺陷包括:崩边、玷污、缺角等,这些缺陷可能会影响到成品石英晶体电参数的性能,例如影响石英晶体的频率特性、电阻特性、DLD特性等。目前石英晶片在镀电极前均需要进行外观缺陷检测,国内大多采用人工目测检测的方式,光学方法的石英晶片的缺陷自动检测技术还不成熟。无论哪种检测方法,都具有比较大的主观性,目前还没有建立有关石英晶片对石英晶体电参数的影响关系的研究报道。因此需要研究石英晶片外观缺陷对石英晶体电参数的影响的对应关系,以便能更准确地分选出对石英晶片电参数有影响的存在外观缺陷的石英晶片。     

The nature of defect patterns on integrated-circuit wafer maps

The history of IC-yield models dates from those based on the simple Poisson distribution to current models based on the families of compound and generalized Poisson distributions. The latter are more complex because the IC chips have grown larger in area and circuit density, thereby unveiling the clustering (aggregation) properties of defects on wafers. These clustering properties are reflected by the parameters of the distributions on which the existing yield models are based. The possibility of any statistical distribution providing an exact representation of some actual population is very small. It is important, though, that the parameters of alternative statistical distributions describing some actual population should have some physical meaning. This paper considers 3 distributions used to model empirical defect distributions in IC manufacturing. The applicability of these distributions depends strongly on the area of the chips fabricated on a wafer. The authors discuss the cases where a study of the parameters alone might not provide conclusive evidence about the spatial properties of defect patterns on IC-wafer maps. They propose some measures to explain defect clustering: variance/mean ratio, turning point, mean crowding, and patchiness. Due to the complementary nature of these measures, no measure, by itself, can provide sufficient information about defect patterns or complete details of the differences between several patterns. It is therefore instructive to analyze defect patterns by as many measures as feasible     

Implications of a localised defect model for wafer level reliability measurements of thin dielectrics

In this paper a simple model of an oxide defect as a region of localised oxide thinning is used to explore the relationship between the most commonly used measurements of dielectric reliability. For each measurement it shows how the measured parameters depend on the area and effective thickness of the defect. The work shows that in constant voltage and ramped voltage stress the area and thickness of the defect may be easily separated in the measured parameters. However, in constant current and ramped current measurements all measured parameters are dependent on both area and thickness which makes the extraction of area and thickness more difficult. It is shown that, in order to be able to project from one measurement to any other, the defect area and thickness must be determined. In particular, if projections of charge to breakdown are required then the use of a model which only includes defect thinning as proposed by Lee et al, [1], is not sufficient.     

A reliability model for SAC solder covering isothermal mechanical cycling and thermal cycling conditions

This paper presents an alternative to the use of energy-based methodologies for life cycle predictions of solder interconnects. Isothermal mechanical cycling testing has been conducted using joint-scale solder samples on a novel testing apparatus. The test data shows that work as a single parameter is insufficient in predicting failure; nor does the inclusion of cyclic frequency and mean temperature improve work-based methodologies. Here, a novel semi-empirical approach is presented in which stress, strain, strain rate and temperature are individually treated to create a model capable of predicting material behaviour under arbitrary cyclic loading conditions. The model constants are fitted to the results of the isothermal mechanical cycling tests, using load drop as a measure of damage. The calibrated model is then employed to predict the failure of a BGA device under thermal cycling. The modelling results show state-of-the-art agreement with the test data and superiority over Morrow model constants from literature that have been applied to this data set.     

Prediction of thermo-mechanical integrity of wafer backend processes

More than 65% of IC failures are related to thermal and mechanical problems. For wafer backend processes, thermo-mechanical failure is one of the major bottlenecks. The ongoing technological trends like miniaturization, introduction of new materials, and function/product integration will increase the importance of thermomechanical reliability, as confirmed by the ITRS (International technology roadmap for semiconductors; [1]). Since most of the thermomechanical problems initiate in the design phase, failure prevention-designing for reliability, is strongly desired. To support wafer backend process development, it is necessary to develop reliable and efficient methodologies (both testing and modeling) to predict the thermal and mechanical behavior of backend processes.This paper presents our research results covering the backend process reliability modeling considering both thermal and mechanical (CMP) loading. The emphasis is particularly on the effect of using Cu/SiLK low-dielectric-constant (low-k) structure instead of the traditional Al/SiO₂. SiLK is a particular polymeric low-k material developed by the Dow Chemical Company [2] [Adv. Mater. 12 (2002) 1767].Our results shows that Cu/SiLK structures exhibit significantly different reliability characteristics than their aluminum predecessors, and that they are more critical from several design aspects. This not only makes the stress management in the stacks more difficult, but also strongly impacts packaging.     

基于GBDT的船舶油耗预测模型设计

建立准确的船舶油耗预测模型是实现船舶节能减排的基础.文中以某一远洋船舶作为研究对象,对船舶的实际运营数据进行分析和预处理,结合GBDT算法构建船舶油耗的预测模型.将模型在测试数据集上的预测值与实测数据进行对比,结果表明预测误差低于6.98％,优于随机森林和决策树的预测结果.该文验证了基于GBDT所建立的油耗预测模型的准...     

基于改进BP算法的软件缺陷预测模型研究

针对传统软件测试成本高及测试过程依赖于软件用例的设置等问题,设计了基于BP,JCUDA_BP和JCUDASA_BP的软件缺陷预测模型,并通过调研、实验的方式对基于改进BP算法的软件缺陷预测算法进行了相关的研究分析,探讨了JCUDA技术对于BP算法的影响,证明了模拟退火算法与JCUDA技术相结合的方式具有改进软件缺陷预测模型的可能性。     

Modeling of rarefied gas heat conduction between wafer andsusceptor

Gas-assisted wafer cooling/heating system is one of the key technologies for manufacturing microelectronic devices. The heat conduction in the gap between wafer and susceptor is modeled here as a one- or two dimensional (1-D or 2-D) rarefied gas problem. The simulation is performed for monatomic and diatomic gases by means of the Direct Simulation Monte Carlo (DSMC) method. In the 1-D case the gas heat conductivity is obtained for various factors (gas species, surface temperature, energy accommodation coefficient) as a function of the Knudsen number. All numerical data obtained can be summarized by a simple equation: Smoluchowski's equation extended to the free molecular regime. In the 2-D case, the DSMC method is applied to the heat transfer between rough and smooth surfaces. The effect of surface roughness on heat conduction is clarified     

基于纹理特征的太阳能晶片计数研究

针对工业生产中太阳能晶片人工计数的难以实施,本文提出了运用数字图像处理技术的自动计数方法。从具体应用角度出发,通过对晶片叠层图像的纹理分析,先用一阶统计量法和边缘检测投影法对晶片区域进行定位,然后通过极值点分析法进行计数。实验结果表明该算法的定位准确度可达到94%以上,计数正确率可以达到95%,而且算法简单可行,具有实用价值。     

A model for wafer scale variation of removal rate in chemical mechanical polishing based on elastic pad deformation

It is well known that within wafer non-uniformity (WIWNU), due to the variation in material, removal rate (MRR) in the whole wafer plays an important role in determining the quality of a wafer planarized by CMP. Various material removal models also suggest that the MRR is strongly influenced by the interface pressure. In the present work, an analytical expression for pressure distribution at the wafer and pad interface is developed. It is observed that depending on the wafer curvature and polishing conditions, the interface pressure may exhibit significant variation. The analytical model predictions are first verified against finite element method (FEM) simulations. The predicted analytical pressure profiles are then utilized in Preston's equation to estimate the MRR, and these MRR predictions are also compared to experimental observations. The analytical results suggest, that for a specified wafer curvature there exists a certain polishing condition (and vice versa) that will enable holding the WIWNU within a specified tolerance band. The proposed model facilitates the design space exploration for such optimal polishing conditions.     

基于灰色预测模型的电视频道热度预测

针对电视频道热度的预测可以为业务决策提供重要依据,电视频道热度受到多种复杂因素的影响,包括电视频道的播放内容及收视群体等因素.基于某地卫视频道一周的频道热度数据,利用灰色预测模型进行频道热度的短期预测,对预测值与实际采集值进行对比,结果证明在原始数据量少及样本种类单一的情况下,灰度预测模型能较好的对电视频道热度进行中短期预测.     

基于深度学习的太阳能电池板缺陷检测模型设计

太阳能电池板是光伏发电的核心部件,表面质量关系着其使用寿命和发电效率。针对目前太阳能电池板缺陷检测方法存在着检测效率低、周期长、检测缺陷单一等问题,文章设计了一种基于深度学习的太阳能电池板缺陷检测模型SSNBDL,其基本思想是,在Segnet网络框架基础上,使用空洞卷积替代其中的池化层,使得在增大感受野的同时还保留了图像的边缘信息,基于该模型对太阳能电池板缺陷进行检测,明显提高了检测的准确率。     

Prediction of network attack profit path based on NAPG model

The network attack profit graph (NAPG) model and the attack profit path predication algorithm are presentedherein to cover the shortage of considerations in attacker-s subjective factors based on existing network attack pathprediction methods. Firstly, the attack profit is introduced, with the attack profit matrix designed and the attackprofit matrix generation algorithm given accordingly. Secondly, a path profit feasibility analysis algorithm isproposed to analyze the network feasibility of realizing profit of attack path. Finally, an opportunity profit path andan optimal profit path are introduced with the selection algorithm and the prediction algorithm designed for accurateprediction of the path. According to the experimental test, the network attack profit path predication algorithm isapplicable for accurate prediction of the opportunity profit path and the optimal profit path.     

Prediction of wafer state after plasma processing using real-timetool data

Empirical models based on real-time equipment signals are used to predict the outcome (e.g., etch rates and uniformity) of each wafer during and after plasma processing. Three regression and one neural network modeling methods were investigated. The models are verified on data collected several weeks after the initial experiment, demonstrating that the models built with real-time data survive small changes in the machine due to normal operation and maintenance. The predictive capability can be used to assess the quality of the wafers after processing, thereby ensuring that only wafers worth processing continue down the fabrication line. Future applications include real-time evaluation of wafer features and economical run-to-run control     

基于涡球模型的湍流气动光学效应预测方法

摘 要: 气动光学效应工程预测中通常采用的RANS方法无法提供流场涡结构的细节,因而难以模拟脉动结构对光学传输的影响。改进了一种基于涡球模型的随机相位屏方法,根据流场的统计平均特性在其中布置涡球,以模拟瞬态涡结构分布对气动光学的影响。改进的主要思路是建立涡球参数与流动统计特性之间的关联,应用涡球所在位置的密度脉动均方根值和脉动尺度确定涡球与周围流体的密度差和涡球的直径,避免了原有方法在脉动较弱的区域布置大尺度涡球的可能,更接近于物理实际。对可压缩剪切层流场光学畸变模拟的结果表明:改进后的随机相位屏方法得到的光学畸变程度与光强直接积分结果更为一致,相对原始方法,计算精度有明显的提高。     

A model for wafer scale variation of material removal rate in chemical mechanical polishing based on viscoelastic pad deformation

It is well known that within-wafer nonuniformity (WIWNU) due to the variation in material removal rate (MRR) in chemical mechanical polishing (CMP) significantly affects the yield of good dies. The process control for a batch CMP operation is further complicated by wafer-to-wafer nonuniformity (WTWNU) caused by MRR decay when a number of wafers are polished with the same unconditioned pad. Accordingly, the present work focuses on modeling the WIWNU and WTWNU in CMP processes. Various material removal models suggest that the MRR is strongly influenced by the interface pressure. It is also well known that the viscoelastic properties of the pad play an important role in CMP. In the present work, an analytical expression for pressure distribution (and its associated MRR) at the wafer-pad interface for a viscoelastic pad is developed. It is observed that under constant load, which is typical during main polishing in CMP, the spatial distribution of the interface pressure profile may change with time from edge-slow to edge-fast, depending on the combination of wafer curvature, down pressure, and pad properties. For constant displacement operations, the pressure profile retains its edge-slow or edge-fast characteristics over time. The analytical model predictions of MRR based on viscoelastic pad properties also correlate very well to existing experimental observations of MRR decay when an unconditioned pad is used to polish a number of wafers. Based on these observations, it may be conjectured that the viscoelastic material properties of the pad play a primary role in causing the observed MRR decay. The analytical results obtained in the present work can also provide an estimation of evolution of thickness removal distribution over the entire wafer. This may be used for determining the optimum thickness of the overburden material and its polishing time, and for effective control of CMP processes.     

Fabrication of low defect density 3C-SiC on SiO2 structures using wafer bonding techniques

This paper reports on a process to fabricate single-crystal 3C-SiC on SiO₂ structures using a wafer bonding technique. The process uses the bonding of two polished polysilicon surfaces as a means to transfer a heteroepitaxial 3C-SiC film grown on a Si wafer to a thermally oxidized Si wafer. Transfer yields of up to 80% for 4 inch diameter 3C-SiC films have been achieved. Homoepitaxial 3C-SiC films grown on the 3C-SiC on SiO₂ structures have a much lower defect density than conventional 3C-SiC on Si films.     

基于图像处理的石英晶体片分选系统

介绍了石英晶体片分选系统的组成和工作原理,对石英晶体片缺陷进行了分类和特征分析,阐述了控制器的部分软件流程.最后,给出了目前石英晶体片分选系统达到的整体性能.     

Defect detection algorithm for wafer inspection based on laserscanning

A defect detection algorithm for wafer inspection based on laser scanning is presented. Microscopic anomalies, contaminants, and process induced pattern defects result in a two-dimensional (2-D) laser scattering signature, which closely resembles the coherent point-spread-function of the scanning laser beam. This point-spread-function is a 2-D Gaussian in the majority of cases and can be characterized by four parameters. The algorithm fits Gaussian surfaces to sampled data points. Events are accepted or rejected on the basis of how similar the Gaussian parameters are to that of the point-spread-function, known a priori. It is shown that the algorithm achieves a 95% capture for submicron particles and pattern defects on typical logic and array wafer regions. Results demonstrating the algorithm's performance relative to mechanical and electronic noise and to signal resolution are presented     

Prediction of product yield distributions from wafer parametricmeasurements of CMOS circuits

A technique for predicting the yield distribution of CMOS circuits based on electrical parameter distributions is presented. This technique uses the mean and standard deviation of the measured threshold voltage and mobility of NMOS and PMOS transistors to project the yield of the circuit in a specified design window. The method thus provides a quantitative means of carrying out tradeoffs between design windows and final product yield