Share Email Print

Spie Press Book

Lithography Process Control
Author(s): Harry J. Levinson
Format Member Price Non-Member Price

Book Description

This text covers lithography process control at several levels, from fundamental through advanced topics. The book is a self-contained tutorial that works both as an introduction to the technology and as a reference for the experienced lithographer. It reviews the foundations of statistical process control as background for advanced topics such as complex processes and feedback. In addition, it presents control methodologies that may be applied to process development pilot lines.

Book Details

Date Published: 12 March 1999
Pages: 204
ISBN: 9780819430526
Volume: TT28

Table of Contents
SHOW Table of Contents | HIDE Table of Contents
1. Introduction to the use of statistical process control in lithography
1.1. The assumptions underlying statistical process control
1.2. The properties of statistical process control
1.3. Situations in lithography where statistical process control cannot be applied naively
1.4. Non-normal distributions
1.5. Process capability
2. Sampling
2.1. Choosing the proper sample size
2.2. Measurement location considerations
2.3. Correlations
2.4. Measurement frequency
2.5. Systematic sources of variation
3. Simple and complex processes
3.1. Definitions
3.2. Why test wafers are useful
3.3. How to address complex processes in lithography
3.4. Distinguishing between layer-specific and equipment-specific effects
4. Linewidth control
4.1. Cause and effect
4.2. Independent variables
4.2.1. Exposure dose
4.2.2. Resist thickness
4.2.3. Focus
4.2.4. Bake temperatures
4.2.5. Resist development
4.2.6. Humidity
4.2.7. DUV resists special considerations
4.2.8. Contributions from reticles
4.3. Maximizing the process window
5. Overlay
5.1. Overlay models
5.2. Matching
5.3. Contributions from processing and alignment mark optimization
5.4. Addressing the problem of non-normal distributions
5.5. Outliers
6. Yield
6.1. Yield monitor strategy
6.2. Yield models
6.3. Parameters which affect yield
7. Process drift and automatic process control
7.1. Adjusting for process drift
7.2. The exponentially-weighted moving average
7.3. Automatic process control
8. Metrology
8.1. The need for understanding the measurement process: defect detection
8.2. Linewidth measurement using scanning electron microscopes
8.3. Electrical linewidth measurement
8.4. Measurement error budgets
8.5. Measurement of overlay
9. Control of operations
9.1. Self-control
9.2. Documentation
9.3. ISO 9000


In the mid-1980s the semiconductor industry underwent a fundamental change. With superior yields, reliability, and efficiency, Japanese manufacturers of integrated circuits surpassed their American competitors in terms of market share. The American semiconductor industry responded with significant quality improvement and regained the leadership market position by 1993. Today the integrated circuit business is highly competitive and global.

During the transition years I found that I was frequently consulted by lithography engineers who were attempting to apply quality improvement methods, such as statistical process control (SPC), to which they had been recently introduced. Through discussions with these engineers it became clear that there were subtle reasons why SPC could not be applied in a straightforward way to many situations which occurred in lithography. My explanations to the engineers evolved into a set of class notes, and now this Tutorial Text.

Many of the quality problems that we were trying to solve arose in conventional manufacturing situations, while others involved development pilot lines. Methods for addressing the problems that occur in the context of process development are rarely addressed in texts on process control. Another objective of this Tutorial Text is to present control methodologies applicable to development pilot lines.

To understand this text there are some prerequisites. A basic foundation in lithography science is assumed. The SPIE Handbook on Microlithography, Micromachining and Microfabrication. Volume 1: Microlithography provides a suitable introduction. It is also assumed that the reader has had some introduction to basic statistical concepts and statistical process control. It is my intention that this text be a self-contained tutorial on lithography process control for readers familiar with the prerequisite lithography science and basic statistical process control, although some subjects may involve a higher level of mathematical sophistication than others. The text covers the subject of lithography process control at several levels. Discussions of some very basic elements of statistical process control and lithography science are included, because, when trying to control a lithography process, a number of subtle problems arise that are related to fundamental issues. To most readers, the information presented on the foundations of statistical process control should be familiar. Nevertheless, it is useful to review the foundations of statistical process control, in order to clearly identify those circumstances in which the method may be applied, and where it needs to be applied with particular care. This inclusion of basic topics also allows the text to serve as an introduction to process control for the novice lithography engineer and as a reference for experienced engineers. More advanced topics are also included to varying levels of detail. Some of these topics, such as complex processes and feedback, are discussed in considerable detail, because there is no comparable presentation available. Other topics are only introduced briefly, and the reader is referred to other texts which cover the subject quite well.

The text also contains numerous references to the extensive literature on the subject of this book. These references are intended as a guide for further study by the interested reader and are also meant to serve as an acknowledgment to the many people who have contributed over the years to improving our understanding of the lithography process and how better to control it.

A few special acknowledgments are in order. First, I want to thank my wife, Laurie Lauchlan, who tolerated the many hours I spent in the study writing this book, and who shared many of her insights on metrology. Dr. David C. Joy helpfully provided information on recent developments in the understanding of charging in low voltage SEMs. I also want to express my gratitude to Chuck DeHont, who first allowed many of these ideas first to be implemented. Finally, I want to thank the many people with whom I have worked at Advanced Micro Devices, Sierra Semiconductor, and IBM, through whose efforts the world has been improved.

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research