Re-Evaluation of the Accuracy of NIST Photomask Linewidth Standards
James E. Potzick
Every artifact measurement standard has some uncertainty associated with its calibration, and the NIST Photomask Linewidth Standards are no exception. This uncertainty is caused by a combination of those factors which influence the calibration measurement. The methods of estimating measurement uncertainty, however, and the interpretation of its meaning, have been subjects of discussion for many years. The International Organization for Standardization (ISO) has recently published a Guide to the Expression of Uncertainty in Measurement, which outlines an operational procedure for combining the effects of the various contributing factors. The value of a measurement standard lies in its accuracy--the estimated calibration uncertainty claimed by its purveyor, and its authority--the ultimate user's confidence in the purveyor's accuracy claim. These qualities are then transferred to the user's subsequent in-house measurements and add to their value, thus justifying the cost of the standard. A new NIST policy on measurement uncertainty has led to a change in the method of determining the uncertainty of its standards to comply with ISO recommendations. This has changed the value of the NIST Photomask Linewidth Standards in the field. This paper outlines the ISO method of determining measurement uncertainty, compares the old and the new calibration uncertainties for these standards and the methods for their determination, and outlines the method for determining the uncertainty of subsequent measurements.
Proceedings of SPIE
Santa Clara, CA
Integrated Circuit Metrology, Inspection, and Process Control IX, Marylyn H. Bennett, Editor May 1995, Mask Metrology