Collaborative Research: Nonparametric Bayesian Modeling of Reliability of Nonelectronics

这项拨款为奈米电子学的可靠性研究提供经费。纳米电子学是美国强劲经济增长的推动力。可靠性被认为是决定纳米制造成功与否的最关键因素之一。该合作项目将由一名纳米器件工程师和一名可靠性专家领导。本跨学科研究的目的是研究有效和实用的非参数贝叶斯方法来评估纳米电子产品在早期设计和开发阶段的可靠性。特别地，提出的研究任务包括两个密切相关的组成部分：(i)分析研究，开发实用的非参数贝叶斯方法，用于估计纳米电子器件的重要可靠性指标；（ii）实验研究，通过制作和表征新开发的基于纳米晶体嵌入高k薄膜的非易失性存储器来验证模型。在这个项目的成功完成后，设计和制造可靠的纳米电子产品的新方法和工具将被开发出来。这将是基于实验数据和非参数贝叶斯方法的纳米产品可靠性建模和预测的第一个系统研究，该方法提供了极大的灵活性和能力，以解决影响产量和成本的实际产品中的挑战。该结果将使制造商能够在早期设计和开发阶段评估新纳米产品的可靠性和涉及的成本，这将加快产品进入市场。此外，新型非易失性存储器件在下一代集成电路（ic）和数据存储中具有实际应用价值。这个跨学科的研究项目将为学生提供可靠性、半导体器件、薄膜材料和高科技制造工艺的理论和实验教育。

This grant provides funding for the investigation of reliability in nanoelectronics. Nanoelectronics is a driving force for strong economic growth in the United States. Reliability is recognized as one of the most critical factors in determining the success of nano fabrication and manufacturing. This collaborative project will be led by a nano device engineer and a reliability specialist. The objective of this interdisciplinary research is to investigate effective and practical nonparametric Bayesian methods to assess the reliability of nanoelectronic products during the early design and development stages. Specially, the proposed research tasks consist of two closely-related components: (i) an analytical study to develop practical nonparametric Bayesian methods for estimating important reliability measures of nanoelectronic devices, and (ii) an experimental study to validate the models by fabricating and characterizing newly developed nonvolatile memories based on nanocrystals embedded high-k thin films. Upon the successful completion of this project, new methods and tools that are critical to design and manufacture reliable nanoelectronic products will be developed. This will be the first systematic study in modeling and predicting reliability of nano products based on experimental data and nonparametric Bayesian methods which offer great flexibility and capability to address challenges in real products influencing yield and cost. The result will enable manufacturers to assess reliability of new nano products and involved costs in the early design and development stages, which would expedite products to the market. Furthermore, the new nonvolatile memory devices have practical applications in future generation integrated circuits (ICs) and data storage. This multidisciplinary research project will provides students with both theoretical and experimental education in reliability, semiconductor devices, thin film materials, and high-tech fabrication processes.