Unbiased Estimation of Linewidth Roughness

Published

May 1, 2005

Author(s)

John S. Villarrubia, B Bunday

Abstract

Line width roughness (LWR) is usually estimated simply as three standard deviations of the line width. The effect of image noise upon this metric includes a positive nonrandom component. The metric is therefore subject to a bias or ?systematic error? that we have estimated can be comparable in size to the roughness itself for samples as smooth as required by the industry roadmap. We illustrate the problem using scanning electron microscope images of rough lines. We propose simple changes to the measurement algorithm that, if adopted by metrology instrument suppliers, would permit estimation of LWR without bias caused by image noise. (11 references)

Proceedings Title

Proceedings of SPIE

Volume

5752

Conference Dates

February 27-March 4, 2005

Conference Location

San Jose, CA, USA

Conference Title

Metrology, Inspection, and Process Control for Microlithography XIX, Richard M. Silver, Editor, May 2005

Pub Type

Conferences

Download Paper

Local Download

Keywords

line edge roughness (LER), line width roughness (LWR), measurement algorithms, measurement bias, Scanning Electron Microscopy (SEM)

Citation

Villarrubia, J. and Bunday, B. (2005), Unbiased Estimation of Linewidth Roughness, Proceedings of SPIE, San Jose, CA, USA, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=822373 (Accessed April 26, 2024)