Noise filtering of scanning-electron-microscope images for accurate analysis of line-edge and line-width roughness

Atsushi Hiraiwa, Akio Nishida

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The control of line-edge or line-width roughness (LER/LWR) is a challenge, especially for future devices that are fabricated using extremeultraviolet (EUV) lithography. Accurate analysis of the LER/LWR plays an essential role in this challenge and requires the noise involved in scanning- electron-microscope (SEM) images to be reduced by appropriate noise filtering prior to analysis. To achieve this, we simulated the SEM images using a Monte Carlo method, and detected line edges in both experimental and theoretical images after noise filtering using new imageanalysis software. The validity of this software and these simulations was confirmed by a good agreement between the experimental and theoretical results. In the case when the image pixels aligned perpendicular (crosswise) to line edges were averaged, the variance varof that was additionally induced by the image noise decreased with a number NPIX;X of averaged pixels, with exceptions when NPIX;X was relatively large, whereupon the variance increased. The optimal NPIX;X to minimize var(f) was formulated based on a statistical mechanism of this change. LER/LWR statistics estimated using the crosswise filtering remained unaffected when NPIX;X was smaller than the aforementioned optimal value, but monotonically changed when NPIX;X was larger contrary to expectations. This change was possibly caused by an asymmetric scan-signal profile at edges. On the other hand, averaging image pixels aligned parallel (longitudinal) to line edges not only reduced var(f) but smoothed real LER/LWR. As a result, the nominal variance of real LWR, obtained using simple arithmetic, monotonically decreased with a number NPIX;L of averaged pixels. Artifactual oscillations were additionally observed in power spectral densities. Var(f) in this case decreased in inverse proportion to the square root of NPIX;L according to the statistical mechanism clarified here. In this way, the noise filtering has a marked effect on the LER/LWR analysis and needs to be appropriately and carefully applied. These results not only constitute a solid basis, but also considerably improve previous empirical instructions for accurate analyses. The most important lesson from this work is to crosswise average an optimized number of image pixels consulting the aforementioned equation.

Original languageEnglish
Article number043010
JournalJournal of Micro/Nanolithography, MEMS, and MOEMS
Volume11
Issue number4
DOIs
Publication statusPublished - 2012

Fingerprint

Linewidth
Electron microscopes
roughness
electron microscopes
Surface roughness
Pixels
Scanning
scanning
pixels
Power spectral density
Lithography
Monte Carlo methods
consulting
computer programs
Statistics
Monte Carlo method
proportion
education
lithography
statistics

Keywords

  • filtering
  • image
  • line edge roughness
  • line width roughness
  • noise
  • power spectral density
  • scanning electron microscope
  • standard deviation
  • variance

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Mechanical Engineering
  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Atomic and Molecular Physics, and Optics

Cite this

Noise filtering of scanning-electron-microscope images for accurate analysis of line-edge and line-width roughness. / Hiraiwa, Atsushi; Nishida, Akio.

In: Journal of Micro/Nanolithography, MEMS, and MOEMS, Vol. 11, No. 4, 043010, 2012.

Research output: Contribution to journalArticle

@article{6b81c8f09bb74e4da3e3a7729a40152d,
title = "Noise filtering of scanning-electron-microscope images for accurate analysis of line-edge and line-width roughness",
abstract = "The control of line-edge or line-width roughness (LER/LWR) is a challenge, especially for future devices that are fabricated using extremeultraviolet (EUV) lithography. Accurate analysis of the LER/LWR plays an essential role in this challenge and requires the noise involved in scanning- electron-microscope (SEM) images to be reduced by appropriate noise filtering prior to analysis. To achieve this, we simulated the SEM images using a Monte Carlo method, and detected line edges in both experimental and theoretical images after noise filtering using new imageanalysis software. The validity of this software and these simulations was confirmed by a good agreement between the experimental and theoretical results. In the case when the image pixels aligned perpendicular (crosswise) to line edges were averaged, the variance varof that was additionally induced by the image noise decreased with a number NPIX;X of averaged pixels, with exceptions when NPIX;X was relatively large, whereupon the variance increased. The optimal NPIX;X to minimize var(f) was formulated based on a statistical mechanism of this change. LER/LWR statistics estimated using the crosswise filtering remained unaffected when NPIX;X was smaller than the aforementioned optimal value, but monotonically changed when NPIX;X was larger contrary to expectations. This change was possibly caused by an asymmetric scan-signal profile at edges. On the other hand, averaging image pixels aligned parallel (longitudinal) to line edges not only reduced var(f) but smoothed real LER/LWR. As a result, the nominal variance of real LWR, obtained using simple arithmetic, monotonically decreased with a number NPIX;L of averaged pixels. Artifactual oscillations were additionally observed in power spectral densities. Var(f) in this case decreased in inverse proportion to the square root of NPIX;L according to the statistical mechanism clarified here. In this way, the noise filtering has a marked effect on the LER/LWR analysis and needs to be appropriately and carefully applied. These results not only constitute a solid basis, but also considerably improve previous empirical instructions for accurate analyses. The most important lesson from this work is to crosswise average an optimized number of image pixels consulting the aforementioned equation.",
keywords = "filtering, image, line edge roughness, line width roughness, noise, power spectral density, scanning electron microscope, standard deviation, variance",
author = "Atsushi Hiraiwa and Akio Nishida",
year = "2012",
doi = "10.1117/1.JMM.11.4.043010",
language = "English",
volume = "11",
journal = "Journal of Micro/ Nanolithography, MEMS, and MOEMS",
issn = "1932-5150",
publisher = "SPIE",
number = "4",

}

TY - JOUR

T1 - Noise filtering of scanning-electron-microscope images for accurate analysis of line-edge and line-width roughness

AU - Hiraiwa, Atsushi

AU - Nishida, Akio

PY - 2012

Y1 - 2012

N2 - The control of line-edge or line-width roughness (LER/LWR) is a challenge, especially for future devices that are fabricated using extremeultraviolet (EUV) lithography. Accurate analysis of the LER/LWR plays an essential role in this challenge and requires the noise involved in scanning- electron-microscope (SEM) images to be reduced by appropriate noise filtering prior to analysis. To achieve this, we simulated the SEM images using a Monte Carlo method, and detected line edges in both experimental and theoretical images after noise filtering using new imageanalysis software. The validity of this software and these simulations was confirmed by a good agreement between the experimental and theoretical results. In the case when the image pixels aligned perpendicular (crosswise) to line edges were averaged, the variance varof that was additionally induced by the image noise decreased with a number NPIX;X of averaged pixels, with exceptions when NPIX;X was relatively large, whereupon the variance increased. The optimal NPIX;X to minimize var(f) was formulated based on a statistical mechanism of this change. LER/LWR statistics estimated using the crosswise filtering remained unaffected when NPIX;X was smaller than the aforementioned optimal value, but monotonically changed when NPIX;X was larger contrary to expectations. This change was possibly caused by an asymmetric scan-signal profile at edges. On the other hand, averaging image pixels aligned parallel (longitudinal) to line edges not only reduced var(f) but smoothed real LER/LWR. As a result, the nominal variance of real LWR, obtained using simple arithmetic, monotonically decreased with a number NPIX;L of averaged pixels. Artifactual oscillations were additionally observed in power spectral densities. Var(f) in this case decreased in inverse proportion to the square root of NPIX;L according to the statistical mechanism clarified here. In this way, the noise filtering has a marked effect on the LER/LWR analysis and needs to be appropriately and carefully applied. These results not only constitute a solid basis, but also considerably improve previous empirical instructions for accurate analyses. The most important lesson from this work is to crosswise average an optimized number of image pixels consulting the aforementioned equation.

AB - The control of line-edge or line-width roughness (LER/LWR) is a challenge, especially for future devices that are fabricated using extremeultraviolet (EUV) lithography. Accurate analysis of the LER/LWR plays an essential role in this challenge and requires the noise involved in scanning- electron-microscope (SEM) images to be reduced by appropriate noise filtering prior to analysis. To achieve this, we simulated the SEM images using a Monte Carlo method, and detected line edges in both experimental and theoretical images after noise filtering using new imageanalysis software. The validity of this software and these simulations was confirmed by a good agreement between the experimental and theoretical results. In the case when the image pixels aligned perpendicular (crosswise) to line edges were averaged, the variance varof that was additionally induced by the image noise decreased with a number NPIX;X of averaged pixels, with exceptions when NPIX;X was relatively large, whereupon the variance increased. The optimal NPIX;X to minimize var(f) was formulated based on a statistical mechanism of this change. LER/LWR statistics estimated using the crosswise filtering remained unaffected when NPIX;X was smaller than the aforementioned optimal value, but monotonically changed when NPIX;X was larger contrary to expectations. This change was possibly caused by an asymmetric scan-signal profile at edges. On the other hand, averaging image pixels aligned parallel (longitudinal) to line edges not only reduced var(f) but smoothed real LER/LWR. As a result, the nominal variance of real LWR, obtained using simple arithmetic, monotonically decreased with a number NPIX;L of averaged pixels. Artifactual oscillations were additionally observed in power spectral densities. Var(f) in this case decreased in inverse proportion to the square root of NPIX;L according to the statistical mechanism clarified here. In this way, the noise filtering has a marked effect on the LER/LWR analysis and needs to be appropriately and carefully applied. These results not only constitute a solid basis, but also considerably improve previous empirical instructions for accurate analyses. The most important lesson from this work is to crosswise average an optimized number of image pixels consulting the aforementioned equation.

KW - filtering

KW - image

KW - line edge roughness

KW - line width roughness

KW - noise

KW - power spectral density

KW - scanning electron microscope

KW - standard deviation

KW - variance

UR - http://www.scopus.com/inward/record.url?scp=84879635390&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879635390&partnerID=8YFLogxK

U2 - 10.1117/1.JMM.11.4.043010

DO - 10.1117/1.JMM.11.4.043010

M3 - Article

AN - SCOPUS:84879635390

VL - 11

JO - Journal of Micro/ Nanolithography, MEMS, and MOEMS

JF - Journal of Micro/ Nanolithography, MEMS, and MOEMS

SN - 1932-5150

IS - 4

M1 - 043010

ER -