1063-7397/02/3104- $27.00 © 2002 MAIK “Nauka /Interperiodica”
Russian Microelectronics, Vol. 31, No. 4, 2002, pp. 207–223. Translated from Mikroelektronika, Vol. 31, No. 4, 2002, pp. 243–262.
Original Russian Text Copyright © 2002 by Volk, Gornev, Novikov, Ozerin, Plotnikov, Prokhorov, Rakov.
According to the International Technology Road-
map for Semiconductors, critical dimensions and other
feature sizes should be measured to an accuracy of 1–
2% of the minimum feature size, which is expected to
decrease from 115 to 22 nm during the period 2002–
2016 . Although the SEM [2–7] remains an instru-
ment of choice for on-line dimensional inspection, the
AFM [8–13] has recently proven a strong competitor.
The above considerations suggest designing a universal
standard in the sense that it would be suitable for both
the SEM and the AFM and would be useful over the
range from tens of micrometers down to tens of nanom-
This paper reviews existing designs of linear stan-
dards, proposes a standard that conforms to the Road-
map, and reports some experimental results in support
of the proposal.
2. REVIEW OF LINEAR STANDARDS
Modern SEM dimensional metrology uses three
types of surface pattern: period structures [14, 15],
pitch structures [3, 16], and linewidth structures [5, 17].
The name of a structure type indicates the certiﬁed
dimension: period, pitch, or linewidth. Certiﬁcation is
performed by diffractometry [14, 18], interferometry
[19, 20], or ellipsometry [17, 21], respectively. How-
ever, the process technology employed may introduce
serious errors in the certiﬁed dimension. Some exam-
ples of linear standards are given in Table 1.
AFM linear standards fall into two general classes.
Those for the atomic scale are based on crystal lattices
certiﬁed by x-ray diffraction. Standards for the
micrometer and submicrometer regions are similar to
SEM regular structures.
2.1. Period Structures
The period structure has so many basic units that it
is impossible to pick the same unit twice. The period is
quantiﬁed by averaging over the basic units. Novikov
 conducted a Monte Carlo study of the effect
of the spread in interedge distance (the distance
between similar edges of two adjacent basic units) on
the accuracy to which the period
is measured by dif-
fraction. It was shown that the measurement error
relates to the standard deviation
is the total number of basic units contributing
to the diffraction pattern. It has been reported that
as large as 10 to 30 nm [23, 24]. On the other hand, the
practical value of
increases from ~10
passes from the micrometer to the nanometer region, so
is below 1 nm in most cases .
We see that the calibration of SEMs and AFMs must
involve many measurements of interedge distance
[according to Eq. (1)] performed at different locations
on the structure . However, industrial SEMs and
AFMs have to be calibrated on the basis of few (three
at most) values of interedge distance, so that
Also note that the magniﬁcation and beam diameter
of an SEM change in passing from one surface feature
to another [25, 26], which prevents us from improving
Linear Standard for SEM–AFM Microelectronics Dimensional
Metrology in the Range 0.01–100
Ch. P. Volk*, E. S. Gornev*, Yu. A. Novikov**, Yu. V. Ozerin*, Yu. I. Plotnikov*,
A. M. Prokhorov
**, and A. V. Rakov**
* AOOT NIIME and Mikron Plant, Zelenograd, Moscow, Russia
** Natural Sciences Research Center, Institute of General Physics, Russian Academy of Sciences, Moscow, Russia
Received February 19, 2002
—Linear standards for the calibration of SEMs and AFMs are reviewed. Requirements to a surface
pattern designed to serve as a universal standard for the above purpose are deﬁned. A trapezoidal pitch structure
is proposed, in which the sidewalls of basic units essentially make a large angle with the normal to the surface.
Its uses in SEM–AFM dimensional metrology are considered. A new, universal standard implementing the
structure is described. Its certiﬁcation is brieﬂy reported. SEM and AFM experiments with the standard are pre-