Data processing: generic control systems or specific application – Specific application – apparatus or process – Product assembly or manufacturing
Reexamination Certificate
2001-09-17
2004-04-06
Picard, Leo (Department: 2125)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Product assembly or manufacturing
C700S110000, C438S005000
Reexamination Certificate
active
06718224
ABSTRACT:
FIELD OF THE INVENTION
The present invention generally relates to a system and method for estimating error in a manufacturing process and more particularly, to a system and method which estimates the error between a control model and a semiconductor manufacturing process and which attributes each portion of the error to a specific contributing context item.
BACKGROUND OF THE INVENTION
Semiconductors are made in production facilities, commonly referred to as “fabs.” A large fab may contain hundreds of automated tools that cooperatively work to convert circular silicon “wafers” (each consisting of dozens, hundreds, and even potentially thousands of chips) into functioning products. One of the challenges in these fabs is to control the manufacturing tools and equipment in a manner which minimizes variation and defects in the products being produced. The manufacturing equipment and tools used within these fabs, such as steppers and scanners, have a multitude of parameters which must be examined and controlled in order to minimize these variations and errors.
In addition to the errors caused by manufacturing equipment and tools, variations in product quality are often functions of the particular lot of products or wafers being processed. As a result, prior control systems for operating the manufacturing tools and processes have often been designed to use feedback data from lots that include the same type of product and that are in the same “streamline” (i.e., products that have experienced the same upstream process flow as the lot currently being processed). However, in fabs producing a relatively diverse variety of products or “high-mix” fabs, certain feedback data loops may operate with hours or even days between data points in the feedback loop. Process tool disturbances that occur during this “downtime” are not captured in the feedback data. As a result, these delays may cause a loss of information regarding the manufacturing tools' respective contributions to the variance in a specific product.
Moreover, prior fabrication control systems do not partition the error into its contributing context items or component parts, such as the specific error components related to the current and previous manufacturing tools and reticles that were used on a product. That is, these conventional systems only generally identify an overall or cumulative error, and do not provide information as to the various elements or items which cooperate to produce the error. Hence, the ability to optimize the manufacturing process and to provide corrective measures to prevent errors in future fabrications or processes is substantially limited.
Some efforts have been made to partition manufacturing errors into their respective components through the use of a conventional sliding mode control law (e.g., a control law implementing a sliding gain). These efforts have had limited success. Particularly, these prior methods have only been capable of identifying two context items. Furthermore, the sliding gains utilized within these prior methods result in a loss of information in one or more individual components of the error estimate, which increases as the sliding gain approaches zero (i.e., as more and more lots are processed through the system).
It is therefore desirable to provide a system and method for estimating error within a semiconductor fabrication process, which overcomes the drawbacks and limitations of prior systems and methods, which estimates the error between a control model and a semiconductor manufacturing process, and which attributes each portion of the error to a specific contributing context item.
SUMMARY OF THE INVENTION
The present invention provides many advantages over conventional error estimation methods and systems. By way of example and without limitation, the present invention accurately estimates the error within a semiconductor manufacturing process and attributes each portion of the error to a specific contributing context item (i.e., a specific component that contributes to the overall error). Furthermore, in one non-limiting embodiment, the present invention is effective to partition the overall error into four separate contributing context items. Moreover, the present invention does not require the use of a sliding gain and continues to re-estimate all components of variation throughout the application of the method.
According to one aspect of the present invention, a control system is provided for a manufacturing process. The control system includes a first portion which identifies a plurality of error components based upon data describing a lot of products being processed, and which estimates values for each of the error components based upon manufacturing feedback data; and a second portion which communicates with the first portion and which receives the estimated values from the first portion and utilizes the estimated values to control the manufacturing process.
According to a second aspect of the present invention, a system is provided for estimating errors within a semiconductor fabrication process. The system includes a controller which is adapted to receive context data, process data and numeric data regarding the semiconductor fabrication process, and which is further adapted to identify an optimal number of error components based upon the context data, and to estimate a value for each of the error components based upon the process data and numeric data.
According to a third aspect of the present invention, a method is provided for estimating errors in a semiconductor manufacturing process. The method includes the steps of: obtaining context items data describing a lot of wafers being processed; identifying a plurality of error components based upon the context items data; obtaining process data describing the manufacturing process; obtaining numeric data describing post-processing measurements taken on the lot of wafers; and estimating values for each of the plurality of error components by use of the process data and numeric data.
These and other features and advantages of the invention will become apparent by reference to the following specification and by reference to the following drawings.
REFERENCES:
patent: 5761093 (1998-06-01), Urbish et al.
patent: 5973766 (1999-10-01), Matsuura et al.
patent: 6197604 (2001-03-01), Miller et al.
patent: 6230069 (2001-05-01), Campbell et al.
patent: 6459941 (2002-10-01), Lobb et al.
patent: 6460002 (2002-10-01), Bone et al.
patent: 6535774 (2003-03-01), Bode et al.
patent: 2002/0177917 (2002-11-01), Polla et al.
patent: 2003/0014145 (2003-01-01), Reiss et al.
patent: 2003/0040830 (2003-02-01), Parikh et al.
patent: 2003/0199108 (2003-10-01), Tanaka et al.
Stuber, J., et al.; Device Dependant Run-to-Run Control of Transistor Critical Dimension by Manipulating Photolithography Exposure Settings; pp. 1-21; Proceedings of the AEC/APC Symposium XII (Sep. 2000).
Campbell W. Jarrett
Firth Stacy
Frank Elliot
Gray Cary Ware & Freidenrich LLP
Picard Leo
Yield Dynamics, Inc.
LandOfFree
System and method for estimating error in a manufacturing... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for estimating error in a manufacturing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for estimating error in a manufacturing... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3275045