New Approaches to Image Processing Based Failure Analysis of Nano-Scale ULSI Devices introduces the reader to transmission and scanning microscope image processing for metal and non-metallic microstructures.
Engineers and scientists face the pressing problem in ULSI development and quality assurance: microscopy methods can't keep pace with the continuous shrinking of feature size in microelectronics. Nanometer scale sizes are below the resolution of light, and imaging these features is nearly impossible even with electron microscopes, due to image noise.
This book presents novel "smart" image processing methods, applications, and case studies concerning quality improvement of microscope images of microelectronic chips and process optimization. It explains an approach for high-resolution imaging of advanced metallization for micro- and nanoelectronics. This approach obviates the time-consuming preparation and selection of microscope measurement and sample conditions, enabling not only better electron-microscopic resolution, but also more efficient testing and quality control. This in turn leads to productivity gains in design and development of nano-scale ULSI chips.
The authors also present several approaches for super-resolving low-resolution images to improve failure analysis of microelectronic chips.
1. Preface2. Introduction2.1 Basics of Image Processing2.2 The Problems of Shrinking Feature Size in ULSI Development and Failure Analysis2.3 High Resolution Imaging of Metallic Structures2.4 High Resolution Imaging of Non-Metallic Structures2.5 Fabrication techniques in ULSI industry3. New Image Processing Methods for Advanced Metallization in Micro- and Nano-Electronics3.1 Characteristics of Metal Ultra-Thin Films Microstructures3.2 Methods Based on Sample and Imaging System Knowledge3.3 Methods Based on Microstructure Grain Size and Shape Range Knowledge3.4 Increased Productivity by Obviating Steps of Selection of Measurement Conditions3.5 Demonstration of Method Capabilities3.5.1 Demonstration on Blurred HRSEM Images of Copper and Silver Films Microstructures3.5.2 Demonstration on Indistinct Images of Filled Trenches and Vias4. New Super Resolving Techniques and Methods for Micro-Electronics4.1 The basics of super resolution4.1.1 Introduction4.1.2 Super resolving hardware4.1.3 Super resolving numerics4.2 Numerical Approaches for super resolved imaging4.2.1 High-Resolution Layout Image Transform4.2.2 Low Resolution Image Transform (Experiment-based)4.2.3 Results of the Comparison4.3 Radon based super resolved imaging4.4 Numerical approaches for characterization of ULSI circuits4.5 Applications in Failure Analysis4.6 Applications in Manufacturing and Testing