4D Characterization of Damage in Interconnects: Experiment and Simulation

随着超大规模集成电路的不断缩小，金属导体或互连受到越来越高的电流密度的影响。然而，足够大的电流密度可以触发原子扩散，称为电迁移（EM），导致小缺陷的产生，最终导致电路故障。目前，电子行业为减轻电磁损伤而采取的安全措施并不可靠，因为这些措施通常是基于微观结构降解过程的二维成像所提供的有限见解。为了完全理解这一过程，需要一个完整的三维互连图像，并结合时间的第四个维度，以了解降解过程的进行。为了实现这种“四维”（4D）成像，先进的x射线断层扫描与降解过程的计算模拟相结合。该奖项支持研究应用这种独特的4D成像，以确定集成电路中使用的无铅互连和焊料材料中em介导的损伤机制的基本机制，其中高电流密度普遍存在。从这项研究中获得的新知识将指导未来的策略，旨在减轻先进微处理器中使用的小型化电路的电磁诱发故障。新知识在微电子工业中有广泛的应用，具有重大的经济意义。该项目的研究将通过创建材料微观结构模拟和可视化软件来加强材料教育，并通过推出一个新的网站来揭示3D微观结构的复杂性，从而提高公众的意识。本研究项目的目的是设计一种集成的实验和计算方法，以实现对锡基和铟基，无铅互连和焊料中伴随微结构降解的潜在机制的基本理解。研究人员将利用实时4D x射线成像、衍射对比断层扫描和相场建模来预测多相、多组分电子材料中em介导的缺陷模式的发生和生长。应对以下挑战：(1) EM条件下的加速失效测试和微缺陷演化的4D（3个空间维度和时间）原位表征(2)建立具有各向异性扩散的BCT sn晶体中缺陷模式形成的大规模三维模拟相场模型，并结合焦耳加热、背应力和热迁移等多物理场；(3)从相同初始状态出发，利用加速电磁破坏试验现场4D数据集验证相场模型。从这项工作中获得的新知识将增强当前预测电磁损伤开始的能力，同时为有效控制电磁介导的焊料和互连失效提供新策略。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

With a continuous downscaling of very large-scale integrated circuits, metallic conductors or interconnects are subject to increasingly high current densities. However, a sufficiently large electric current density can trigger atomic diffusion, known as electromigration (EM), leading to the generation of small defects, ultimately causing circuit failure. Current safety measures adopted in the electronics industry to mitigate EM-induced damage are unreliable, as they are typically based on limited insights provided by two-dimensional imaging of the microstructural degradation process. For complete understanding of the process, a full three-dimensional image of the interconnects is needed, combined with the fourth dimension of time, to understand the degradation process as it proceeds. To achieve this "four-dimensional" (4D) imaging, advanced x-ray tomography is combined with computational simulations of the degradation process. This award supports research to apply this unique 4D imaging to determine the fundamental mechanisms of EM-mediated damage mechanisms in Pb-free interconnects and solder materials used in integrated circuits where high current densities are prevalent. New knowledge gained from this study will guide future strategies aimed at mitigating the EM-induced failure of miniaturized circuits used in advanced microprocessors. The new knowledge has broad application in the microelectronics industry, with substantial economic implications. Research in this project will enhance materials education via the creation of material-microstructure simulation and visualization software, and create public awareness through the launch of a new website to expose the complexity of 3D microstructures.The aim of this research project is to devise an integrated experimental and computational approach to achieve a fundamental understanding of the underlying mechanisms that accompany microstructural degradation in Sn- and In-based, Pb-free interconnects and solders. The investigators will utilize real-time 4D X-ray imaging, diffraction contrast tomography, and phase-field modeling to predict the onset and growth of EM-mediated defect patterns in multiphase, multicomponent electronic materials. The following challenges will be addressed: (1) Accelerated failure testing and 4D (3 spatial dimensions and time) in-situ characterization of microdefect evolution under EM conditions (2) Formulating phase-field models for large-scale 3D simulations of defect pattern formation in BCT Sn-crystals that exhibit anisotropic diffusion and incorporating the multiphysics of Joule heating, back-stress, and thermomigration, and (3) Validating the phase-field model using the in-situ 4D datasets obtained from accelerated EM-failure tests by starting from the same initial state. The new knowledge to be gained from this work will enhance current capabilities to predict the onset of EM damage and at the same time enable new strategies for controlling EM-mediated failure of solders and interconnects, efficiently.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.