1992
DOI: 10.1117/12.59785
|View full text |Cite
|
Sign up to set email alerts
|

<title>Numerical reference models for optical metrology simulation</title>

Abstract: ABSTRACFOptical modeling on the computer can aid R&D efforts to enhance metrology methods, and similarly for lithography, alignment, and particulate monitoring. However, full exploitation of optical modeling is hindered by the lack of appropriate benchmarks for verifying algorithms and evaluating approximations. To help remedy this situation we describe a preliminary set of scalar, 2D numerical reference models (NRMs). These include isolated thin and thick lines, periodic lines, and an isolated trench. Scatter… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
5
0

Year Published

1994
1994
2015
2015

Publication Types

Select...
3
1
1

Relationship

0
5

Authors

Journals

citations
Cited by 5 publications
(5 citation statements)
references
References 13 publications
0
5
0
Order By: Relevance
“…Unfortunately, many of the popular methods are only low order accurate and agree in test problems to only 1 digit of accuracy [45]. Finite difference approximation to the time-domain wave equation (FDTD) [16,42] is common; it has the advantage of being quite general, but has the disadvantage of requiring large regions of free space to be simulated, with significant dispersion error for electrically large domains.…”
Section: Introduction and Problem Formulationmentioning
confidence: 99%
See 1 more Smart Citation
“…Unfortunately, many of the popular methods are only low order accurate and agree in test problems to only 1 digit of accuracy [45]. Finite difference approximation to the time-domain wave equation (FDTD) [16,42] is common; it has the advantage of being quite general, but has the disadvantage of requiring large regions of free space to be simulated, with significant dispersion error for electrically large domains.…”
Section: Introduction and Problem Formulationmentioning
confidence: 99%
“…Moving to the frequency domain, finite element methods [5,30] require non-trivial grid generation, especially in three dimensions, and also demand the discretization of substantial regions of free space together with the imposition of non-local radiation conditions constructed from (1.6) and (1.7). The engineering community has developed many other methods-including coupled wave analysis, the waveguide method [45], the C method [24], and cylindrical (multipole) expansions [27,37]-which are often based on uncontrolled approximations.…”
Section: Introduction and Problem Formulationmentioning
confidence: 99%
“…Secondly, we exploit the fact that filling (rather than solving) the matrix blocks in (56) often accounts for the bulk of the solution time. Consider one of the integral operators used in the BIE matrix (defined in (19)),…”
Section: Accelerated Sweep Over Multiple Incident Angles At One Frequ...mentioning
confidence: 99%
“…There are many low-order numerical methods used to solve multilayer scattering problems, which in test problems may only agree to 1 digit of accuracy [56]. Finite difference time-domain (FDTD) [55,27] is easy to code but has dispersion errors, and requires artificial absorbing boundary conditions and arbitrarily long settling times near resonances.…”
Section: Introductionmentioning
confidence: 99%
“…[1][2][3][4][5] However, the mark selection procedure using real process wafers was time consuming because of numerous experimental evaluations of the mark signal and overlay accuracy. Though mark selection is done with great care, issues concerning the mark frequently arise in production.…”
Section: Introductionmentioning
confidence: 99%