ISO/TS 21383:2021

TECHNICAL ISO/TS
SPECIFICATION 21383
First edition
2021-03
Microbeam analysis — Scanning
electron microscopy — Qualification
of the scanning electron microscope
for quantitative measurements
Reference number
©
ISO 2021
© ISO 2021
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting
on the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address
below or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii © ISO 2021 – All rights reserved

Contents Page
Foreword .v
Introduction .vi
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Symbols and abbreviated terms . 3
5 General principles . 5
5.1 Condition setting . 5
5.2 Contrast/brightness setting . 5
5.3 Sample preparation . 6
6 Measurement of image sharpness . 7
7 Measurement of drift and drift-related distortions (imaging repeatability) .8
7.1 Measurement of image drifts within specified time intervals. . 9
7.1.1 One-minute drift measurement .10
7.1.2 Ten-minute drift measurement .10
7.1.3 One-hour drift measurement .10
7.1.4 Long-term larger than one-hour drift measurement .10
7.2 Evaluation of the drift and the drift-related distortions by using image overlay .11
7.3 Evaluation of the drift and the drift-related distortions by using cross-correlation
function (CCF) .13
7.3.1 Measurement of the drifts by using the CCF .13
7.3.2 Measurement of the distortions by using the CCF .15
8 Measurement of electron-beam-induced contamination .15
8.1 Cleaning of the sample surface .16
8.2 Cleaning of the inner surfaces of the sample chamber .16
8.3 Measurement method of the contamination.17
8.3.1 Measurement of the height of the contamination growth .17
8.3.2 Measurement of relative carbon concentration of the contamination by
the X-ray analysis .18
8.3.3 Measurement of the surface contamination by the change of SEM signal
intensities .18
9 Measurement of the image magnification and linearity .19
9.1 Measurement of the image magnification .20
9.2 Measurement of the image linearity .21
10 Measurement of background noise .22
10.1 Evaluation methods by using noise profiles and processed images .22
10.2 Evaluation methods by calculating numerical image properties .28
11 Measurement of the primary electron beam current .30
11.1 Ten-minute primary electron beam current measurement .30
11.2 Long-term primary electron beam current measurement .30
12 Reporting Form .32
Annex A (informative) Measurement of image sharpness .34
Annex B (informative) Measurement of image drift and distortions caused by unintended
motions .36
Annex C (informative) Measurement of electron beam-induced contamination .47
Annex D (informative) Measurement of the image magnification and linearity .53
Annex E (informative) Measurement of the primary electron beam current .56
Bibliography .58
iv © ISO 2021 – All rights reserved

Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards
bodies (ISO member bodies). The work of preparing International Standards is normally carried out
through ISO technical committees. Each member body interested in a subject for which a technical
committee has been established has the right to be represented on that committee. International
organizations, governmental and non-governmental, in liaison with ISO, also take part in the work.
ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of
electrotechnical standardization.
The procedures used to develop this document and those intended for its further maintenance are
described in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the
different types of ISO documents should be noted. This document was drafted in accordance with the
editorial rules of the ISO/IEC Directives, Part 2 (see www .iso .org/ directives).
Attention is drawn to the possibility that some of the elements of this document may be the subject of
patent rights. ISO shall not be held responsible for identifying any or all such patent rights. Details of
any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received (see www .iso .org/ patents).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and
expressions related to conformity assessment, as well as information about ISO's adherence to the
World Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT), see www .iso .org/
iso/ foreword .html.
This document was prepared by Technical Committee ISO/TC 202, Microbeam analysis, Subcommittee
SC 4, Scanning electron microscopy.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www .iso .org/ members .html.
Introduction
The scanning electron microscope (SEM) is a very versatile instrument, which is widely used in
production, development and scientific research across the world. While they are easy to operate and
provide results quickly, there are a number of notorious problems, which hinder operating them at
their best performance. These are the reasons for lack of excellent repeatability in SEM imaging and
measurements. The most bothersome ones among these are unintended motions of the sample stage
and the primary electron beam, geometry distortions, wrong image magnification, image blur (lack
of sharp focus), noise and electron beam-induced contamination. Quantification of these essential
performance parameters is very useful to ensure that all SEMs perform at manufacturers specifications
and at users’ own purpose. Quantified knowledge helps in the evaluation of measurement uncertainties,
and necessary repairs.
This document pertains to measurement methods for the following SEM performance parameters:
— Image sharpness (spatial resolution, primary electron beam focusing ability).
— Drifts (the sample stage, the electron beam and the electron-optical column).
— Cleanliness (lack of beam-induced contamination).
— Image magnification and linearity (both in X and Y directions).
— Background noise.
— Primary electron beam current.
These parameters will also be influenced by the SEM conditions such as the lifetime of source (emitter
conditions), lifetime of liner tube and apertures (contamination of the electron optical parts), time and
intensity of last cleaning of vacuum chamber by the plasma cleaning or Ultra Violet irradiation, the
sample preparation and final surface cleaning.
vi © ISO 2021 – All rights reserved

TECHNICAL SPECIFICATION ISO/TS 21383:2021(E)
Microbeam analysis — Scanning electron microscopy
— Qualification of the scanning electron microscope for
quantitative measurements
1 Scope
This document describes methods to qualify the scanning electron microscope with the digital imaging
system for quantitative and qualitative SEM measurements by evaluating essential scanning electron
microscope performance parameters to maintain the performance after installation of the instruments.
The items and evaluating methods of the performance parameters are selected by users for their own
purposes.
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content
constitutes requirements of this document. For dated references, only the edition cited applies. For
undated references, the latest edition of the referenced document (including any amendments) applies.
ISO 16700:2016, Microbeam analysis — Scanning electron microscopy — Guidelines for calibrating image
magnification
ISO/IEC 17025:2017, General requirements for the competence of testing and calibration laboratories
ISO 22493, Microbeam analysis — Scanning electron microscopy — Vocabulary
ISO/TS 24597:2011, Microbeam analysis — Scanning electron microscopy — Methods of evaluating image
sharpness
3 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO 22493 and the following apply.
ISO and IEC maintain terminological databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at http:// www .electropedia .org/
3.1
scanning electron microscope
SEM
instrument that produces magnified images of a specimen by scanning its surface with an electron beam
[SOURCE: ISO 16700, 3.1]
3.2
image
two-dimensional representation of the specimen surface generated by SEM
[SOURCE: ISO 16700, 3.2]
3.3
image magnification
ratio of the linear dimension of the scan display to the corresponding linear dimension of the specimen
scan field
[SOURCE: ISO 16700, 3.3]
3.4
scale marker
line/generated line (intervals) on the image (3.2) representing a designated actual length in the
specimen
[SOURCE: ISO 16700, 3.4]
3.5
reference material
RM
material, sufficiently homogeneous and stable with respect to one or more specified properties, which
has been established to be fit for its intended use in a measurement process
[SOURCE: ISO Guide 30:2015, 2.1.1, modified — Note 1 to entry to Note 4 to entry are omitted.]
3.6
certified reference material
CRM
reference material (RM) (3.5) characterized by a metrologically valid procedure for one or more
specified properties, accompanied by an RM certificate that provides the value of the specified property,
its associated uncertainty, and a statement of metrological traceability
[SOURCE: ISO Guide 30:2015, 2.1.2, modified — Note 1 to entry to Note 4 to entry are omitted.]
3.7
calibration
set of operations which establish, under specified conditions, the relationship between the magnification
indicated by the SEM and the corresponding magnification determined by examination of an RM (3.5)
or a CRM
[SOURCE: ISO 16700, 3.7]
3.8
accelerating voltage
absolute acceleration potential V [V] of the final anode to the electron emitter
a
Note 1 to entry: For the electron charge q [C], the accelerated electron will obtain the energy qV [J] =
e ea
V [eV] ≡ E and enter the sample with this energy provided that the initial energy E from the emitter
a L I
is negligible. The “landing energy” to the specimen means this energy E , typically expressed in the unit
L
eV or keV.
Refer to Clause 4 concerning eV.
2 © ISO 2021 – All rights reserved

4 Symbols and abbreviated terms
A , A
areas of the binarized pictures FI and FI , I respectively, typically ex-
() ()
ACF,1 CCF,n
BAC1 BCC 1 n
pressed in pixel
ACF auto-correlation function
CCF cross-correlation function
D , D drift quantities of the n-th image I n=12,,  for the horizontal (H) and vertical
()
Hn Vn n
(V) directions respectively, typically expressed in pixel or in nm
D , D
displacements for X and Y directions respectively from the origin, typically ex-
X Y
pressed in nm
max ||D , max ||D mean the largest absolute values of displacements from the origin
X Y
in X and Y directions respectively
12/
D
O
distance DD=+D from the initial position XY, which is regarded as
()
()
OX X OO
the origin, typically expressed in nm
max D - the largest value of the distance D
O O
d
pitch length of the RM or the CRM, typically expressed in nm
d measured mean pitch length, typically expressed in nm
S
d averaged value of measured d , typically expressed in nm
SA S
CG contrast to gradient method for evaluating image sharpness
DR derivative method for evaluating image sharpness
FT Fourier transform method for evaluating image sharpness
−19
eV electronvolt, a unit of energy equal to approximately 1.6×10 joules (J). By definition,
it is the amount of energy gained (or lost) by the charge of a single electron moving
across an electric potential difference of 1 volt
auto-correlation function of the image I
FI
()
AC 1
cross-correlation function for the initial image I and n-th image I n=23,,  in
FI , I ()
()
1 n
CC 1 n
the measurements
FI binarized picture of the auto-correlation function FI of the initial image I by
() ()
BAC1 AC 1 1
using a thresholding level T
B
FI , I binarized picture of the cross-correlation function FI , I for the initial image I
() ()
BCC 1 n CC 1 n 1
and the n-th image I n=23,,  by using a thresholding level T
()
n B
HFW horizontal field width
HV,
horizontal (H) and vertical (V) peak positions of the auto-correlation function FI
()
PP11
AC 1
respectively
HV, horizontal (H) and vertical (V) peak positions of the cross-correlation function
PPnn
FI , I respectively
()
CC 1 n
I background noise image
BG
I flat image whose signal intensity of the element ij, is the mean value S of the
()
FB MEAN
background noise image I
BG
I processed images obtained from the background noise images I , for example, by a
PB BG
method of contrast enhancement
I original secondary electron (SE) or backscattered electron (BE) scanning image
OS
Ii , j signal intensity of the element ij, of the image I , where i and j mean horizontal and
() ()
vertical numbers of the element respectively measured from the initial element ()11,
Ii(), j signal intensity of the element ()ij, of the reference image I which is set to the flat
ref ref
image I when calculating the peak signal-to-noise ratio S
FB PSNR
max Ii, j means the maximum value of the given n -bit imaging mode. If
 ()
 
ref IM
n = 8-bit imaging mode, then max Ii, j =255
 ()
IM  ref 
Ii(), j signal intensity of the element ()ij, of the test image I which is set to the back-
test test
ground noise image I when calculating the peak signal-to-noise ratio S
BG PSNR
Ii(), j signal intensity of the element ()ij, of the reference image I which is be obtained
refS refS
from the test image Ii(), j by applying a suitable image filter to reduce the
testS
image noise
Ii, j signal intensity of the element ij, of the test image I which is usually not the
() ()
testS testS
background noise image I but actual SE or BE scanning image
BG
I primary electron beam current (probe current)
p
kV kilovolt
k ratio of the area A to the area A (/kA= A )
A,n CCF,n ACF,1 A,,nnCCF ACF,1
L
image size (total pixels of the image area), typically expressed in pixel such as LL×
HV
L , L horizontal (H) and vertical (V) image sizes (lengths) respectively, typically expressed
H V
in pixel
L pixel size, typically expressed in nm
p
L horizontal (H) line profile which is obtained vertically averaged for specified band
pHA
areas from the background noise image
L vertical (V) line profile which is obtained horizontally averaged for specified band
pVA
areas from the background noise image
l the total measured length by a scaler on the screen or the photograph ( ln= ∙d )
S SP S
M
image magnification for setting
N number of measurements for beam current
M
N number of measurements for image drift
MD
N
number of measurements for image magnification
MM
4 © ISO 2021 – All rights reserved

N number of measurements for image sharpness
MS
n
total number of pitches for measurement
P
R image sharpness, typically expressed in nm
L
R image sharpness, typically expressed in pixel
PX
S maximum value of intensities of pixels in an image
MAX
S mean value of intensities of pixels in an image
MEAN
S minimum value of intensities of pixels in an image
MIN
S peak signal-to-noise ratio
PSNR
S standard deviation of intensities of pixels in an image
STD
V accelerating voltage
a
WD working distance
5 General principles
The best performance of any SEM is at some optimized set of instrument settings; therefore, throughout
this document for the various assessments those imaging parameters and instrument settings should
be used if those are specified by the SEM’s manufacturer for achieving the best performance. These
basic principles are useful for users' own purpose in many cases.
5.1 Condition setting
Some SEMs have only one accelerating voltage in these specifications; others may have more (e.g.
15 kV and 1 kV). All the assessments should be performed at all specified accelerating voltages and
magnifications if those parameters and settings are applicable to user’s purposes and evaluations. If
optimization of SEM-based measurements requires different parameters (accelerating voltage, beam
current, etc.) for users' own purpose, then for these all the assessments should be performed and the
results recorded in the report.
Beyond setting the magnification, accelerating voltage, beam current, all pertinent parameters to the
values specified by the instrument manufacturer for proving the best resolution performance, it is
important to set the focus (astigmatism), contrast, and brightness to their SEM-specific, best settings
for taking images.
5.2 Contrast/brightness setting
Only images with properly set contrast and brightness should be used for the various measurements
and quantitative SEM assessments. The contrast and the brightness must be set so that all pixels have
grey-scale levels that never reach the lowest (dark, under-saturation) or the highest (bright, over-
saturation) level. This is important to make sure that no information is lost by setting contrast and
brightness values incorrectly.
If the system has the signal monitor or can generate the histogram for the relative signal range [0,1] in
the acquisition of SEM image, verify that the signals are within the range [0.2, 0.8] approximately.
In 8-bit imaging mode, in properly set images the intensities of the image pixels vary in between 0 and
255 grey levels. In 16-bit imaging mode the intensities of the image pixels vary in between 0 and 65535
grey levels.
Figure 1 shows examples of secondary electron (SE) images taken at accelerating voltage 5 keV, 19,5 μm
HFW. Contrast and brightness are properly set a) and wrong b), c), d) and e). Figure 2 shows their
corresponding histograms.
5.3 Sample preparation
Concerning the samples, there will not be the perfect or the almighty sample which are applicable to many
evaluation items. In general, the best samples are different for the measurements of image sharpness,
drift and drift-related distortions, electron beam induced contamination, image magnification and
linearity. Select the ideal sample which is suitable for the required evaluation accordingly.
a) Proper contrast and bright- b) Too little contrast c) Too much contrast
ness
d) Too much brightness e) Too little brightness
Figure 1 — Examples of SE images for various contrast and brightness setting
6 © ISO 2021 – All rights reserved

a) Proper contrast and bright- b) Too little contrast c) Too much contrast
ness
d) Too much brightness e) Too little brightness
Key
I signal intensity for 8-bit imaging mode
S
N number of counts
C
Figure 2 — Examples of histograms corresponding to the images in Figure 1
6 Measurement of image sharpness
The definition and the explanation of the term “image sharpness” are described in ISO TS 24597 for
the SEM. Image sharpness is strongly related to the focusing ability of the SEM in forming the primary
electron beam. It is one of most widely used, but certainly not sufficient performance parameter.
However, it is very useful to know the image sharpness because the algorithms do not depend on the
human sense and give quantitative results by using the procedures described below.
On the other hand, the term “Lateral resolution” or “Spatial resolution” are not defined strictly in SEM
even though these terms are popular in the surface chemical analysis.
[1]
NOTE Refer to ISO 18115-1:2013 for terms used in spectroscopy .
Even if various SEM manufacturers use these terms, the notion of resolution is not established
scientifically, it is sample-and method-dependent, and there is no accurate way of measuring it today.
The focusing ability is related to the size and shape of the primary electron beam at the surface of the
sample. Its measurement is also very difficult, especially for sub-nanometre focuses, i.e., beam sizes.
Furthermore, these values are not related to beam focusing only, but to interaction volume, stability of
probe scanning and external disturbances as well.
To acquire the images for the evaluation of the image sharpness, refer to the Clause 4 “Steps for
acquisition of an SEM image” of ISO TS 24597. “4.1 General”, “4.2 Specimen”, “4.4 Selection of the field
of view” and “4.7 Contrast-to-noise ratio of the image” will be useful information. The structure of the
sample should not be “periodic mesh” or “line and space” because some specific features or frequencies
are emphasized in the signal analysis. The samples as shown in Figure 3 a) is typical and appropriate
because the particles with different diameters are randomly distributed, and their surface are flat and
edges are sharp.
Set the focus and astigmatism to their best, the magnification, accelerating voltage, beam current,
working distance to the values specified by the instrument manufacturer for proving the best image
sharpness performance, and take several images.
To evaluate the SEM image sharpness performance, follow the procedure in ISO/TS 24597:2011. Select
the evaluation method from the 3-methods DR (Derivative) method, FT (Fourier transform) method
and CG (Contrast-to-gradient) method. For one image, plural methods can be used if necessary. Valuate
that the obtained results are allowable or not for the quantitative measurements. Report evaluated
results with the evaluation methods and the valuation.
Figure 3 shows the examples of the selected SEM images with the image size L=×512 512 for the
evaluation of the image sharpness R [pixel] or R [nm]. The obtained values of image sharpness R
PX L L
are 1,9 nm for Figure 3 a) and 3,3 – 4,2 nm for Figure 3 b). The example of the evaluation process for
these images is shown in Table A.1.
a) Accelerating Voltage V = 15 kV, beam b) Accelerating Voltage V = 1 kV, beam
a a
current I = 43 pA, 265 nm HFW. current I = 43 pA, 657 nm HFW.
P P
Figure 3 — Selected SEM images for the evaluation of image sharpness.
Sample: Evaporated gold on carbon.
See Clause 12 and Annex A for further pertinent information.
7 Measurement of drift and drift-related distortions (imaging repeatability)
Unintended motions make the primary electron beam land on wrong, unintended locations on
the sample, which results in poor repeatability and/or distorted, blurry images, especially at high
magnifications. These typically arise from mechanical and acoustical impacts, temperature variation
of the room and the cooling water, hysteresis, adverse external electromagnetic fields, and from the
noise in various circuits of the SEM. In the SEM with high working rate for many years, charging of
contaminated apertures and inner walls of the liner tube for electron beam could be possible reason for
unintended disturbances.
So, it is very useful to know the drifts and the drift related distortions quantitively and qualitatively
by applying the procedures described below because these properties tell us the upper limit of the
significant measurements.
8 © ISO 2021 – All rights reserved

Modern SEMs can have close to 1 nanometre image sharpness, so even very small unintended motions
can interfere or easily ruin the imaging and measurement performance of the SEM. Both the sample
stage with the sample and the primary electron beam make unintended motions and they result in a
combined error in the actual location where the electron beam-sample interaction takes place.
It is important to make sure that the geometry distortions of the SEM are sufficiently small for acquiring
images and for carrying out measurements within requirements. These, depending on the task at hand,
may be diverse, e.g. for high-quality, fine image sharpness imaging are a lot more stringent than for
simple microanalysis. Depending on the intended use, the drift-related performance should be valuated
for one-minute (typical secondary electron image acquisition time), ten-minute (analytical acquisition),
and for one and many hours-long time periods.
7.1 Measurement of image drifts within specified time intervals.
To determine the drift performance, set the focus to its best, magnification, accelerating voltage, beam
current, to the values specified by the instrument manufacturer for proving the best image sharpness
performance, and without intentionally changing the sample location, perform at least the 1- and 10-
minute assessments, and if it is relevant for the task at hand, add further, longer ones. For longer time
drift measurements, the magnification shall be lowered to keep the reference pattern always in the
central 80 % of the image area (refer to ISO 16700). The guiding principle is to measure the drift for as
long periods of time as the time of the measurement or procedure carried out by the SEM. Valuate that
the measured drifts are allowable or not in the quantitative SEM measurements. Report all results as
image drift performance values and the valuation.
When drift performance is measured by watching the displacements (movements) D and D of a
X Y
marked object (or a particle) in the recorded image, use the determined left end or the right end of the
object for the x – direction, and use the determined bottom end or the top end of the object for the
y-direction as shown in Figure 4. This is the reason why the definition of the displacement become
unclear under the drift-related distortions. For the drift measurement, the initial position XY, of
()
OO
the object should be near the centre of the field (the display). The initial position XY, is regarded
()
OO
12/
as the origin, and the distance D from the origin is defined as DD=+D .
()
O OX X
Key
displacement for X direction, expressed in nm
D
X
displacement for Y direction, expressed in nm
D
Y
Figure 4 — Measurement method of the displacement
After the installation of the SEM, the drifts may be enlarged owing to the increased movements of the
floor or the increased fluctuation of the external electromagnetic fields. Then, the obtained results
do not reflect original SEM performances but the condition of the environments, or the ability of the
isolation and the shielding. Measure the movements of the floor or the fluctuation of the external field
by the methods SEM manufacturer have performed if possible. Consider the source of the change,
and report the estimated results. Improper sample preparation, such as the sample surface with non-
conductor and the insufficiently grounded samples, will also cause the severe drifts.
The acquired images in the drift measurement are also utilized for measurement of drift-related
distortions (imaging repeatability). Refer to 7.2 and 7.3.
7.1.1 One-minute drift measurement
Take 5 consecutive images at every 15 second or so, play them in sequence to visualize the nature and
the extent of the drift, find the largest displacement max ||D and max ||D in X and Y directions
Y
X
respectively among them. Report them as 1-minute drift performance value with the largest X and Y
direction motion, both the largest values max ||D and max ||D and the largest value max D of the
X Y O
distance D from the origin.
O
A graphical example of the measurement is shown in Figure 5 a), and the numerical example for the
report is shown in Table B.1.
Image acquisition time t and image storing time t will be selected as users like within the total
a s
15 seconds. The point is to accomplish the image sequence acquisition within one minute.
If these fast images prove to be too noisy for reliable measurements, and sufficiently higher beam
currents are used, then this condition should be added to the report because the drift performance may
depend on the beam current.
7.1.2 Ten-minute drift measurement
Take 11 images, one at start and then at every minute or so, play them in sequence to visualize the
nature and the extent of the drift, find the largest displacement max ||D and max ||D in X and Y
X Y
directions respectively among them. Report them as 10-minute drift performance value, both the
largest values max ||D and max ||D and the largest value max D of the distance D from the origin.
Y
X O O
A graphical example of the measurement is shown in Figure 5 b), and the numerical example for the
report is shown in Table B.2.
The time t and the time t will be chosen as users like within the total one minute.
a s
7.1.3 One-hour drift measurement
Take 21 images, one at start and then every 3 min or so, play them in sequence to visualize the nature
and the extent of the drift, find the largest displacement max ||D and max ||D in X and Y directions
Y
X
respectively among them. Report them as 1-hour drift performance value, both the largest values max
||D and max ||D and the largest value max D of the distance D from the origin.
X Y O O
A graphical example of the measurement is shown in Figure 5 c), and the numerical example for the
report is shown in Table B.3.
7.1.4 Long-term larger than one-hour drift measurement
Obtain the standard interval t [min] of the recording (measurement) using Formula 1:
ID
x
tt=02. 5∙ (1)
ID DM
where
t [min] is the total time of drift measurements;
DM
x=0.605 is the approximate constant.

Take images at every t [min] or so, play them in sequence to visualize the nature and the extent of the
ID
drift, find the largest displacement max ||D and max ||D in X and Y directions respectively among
X Y
them. Report them as long-term drift performance value, both the largest values max ||D and max
X
||D and the largest value max D of the distance D from the origin.
Y O O
10 © ISO 2021 – All rights reserved

A graphical example of the long term (ten-hour) measurement is shown in Figure 5 d), and the numerical
example for the report is shown in Table B.4.
a) One-minute drift measurement b) Ten-minute drift measurement
c) One-hour drift measurement d) Long term (ten-hour) drift measurement
Key
displacement for X direction, expressed in nm
D
X
displacement for Y direction, expressed in nm
D
Y
Figure 5 — Examples of the measurement of the drift
[6],[7]
Some automatic software-based measurements are used for a drift-compensated imaging if the
SEM user requests these functions to decrease the effects of the drifts.
See Clause 12, Annex B and “B.2 Measurement of image drifts within specified time intervals” and “B.5
Measurement of the drifts and drift-compensated imaging”.
7.2 Evaluation of the drift and the drift-related distortions by using image overlay
To assess the slow scan imaging repeatability performance, set the focus and astigmatism to their best
to prove the best image sharpness for the accelerating voltage, electron beam current and working
distance which are selected by the users for their own purposes. Then take 4 consecutive images as
[7]
shown in Figure 6 at suitable location over the resolution reference sample without changing
anything or moving the sample. Select the areas from the original images so that their largest common
[4]
sections are nearly centred (drift adjusted) by using suitable image processing software . Add the 4
original images I ()n=12,, 34, to the report, and add their selected (drift adjusted) images for the
OSn
overlay, and the overlay image I that is generated by overlaying them so that their largest common
DAO
sections are nearly centred as shown in Figure 7 b). Valuate that the overlay image I is allowable or
DAO
not for the quantitative measurements and add the valuations to the report.
If the simple overlay image I is required to observe the magnitude of the drift, select the centre areas
SO
from the original images. Add the 4 original images I to the report, and add their selected images in
OSn
which drift are not adjusted for the simple overlay, and the overlay image I that is generated by
SO
overlaying them as shown in Figure 7 a).
NOTE If the size of an original image is 1280x960 pixels, then the size of the selected image is typically
512x512 pixels.
a) The first image I b) The 2nd image I
OS1 OS2
c) The 3rd image I d) The 4th image I
OS3 OS4
Figure 6 — Four consecutive, slow-scan images for qualitative assessment of drift-related
distortions. 256 nm HFW. Refer to the bibliography [7]
12 © ISO 2021 – All rights reserved

a) Simple overlay image I b) Drift adjusted overlay image I
SO DAO
Figure 7 — Overlaid images by using 4 simple consecutive images and 4 drift adjusted images.
Refer to the bibliography [7]
To assess the fast scan imaging repeatability performance, set the focus and astigmatism to their best
to prove the best image sharpness for the accelerating voltage, electron beam current and working
distance which are selected by the users for their own purposes. Then take many consecutive images at
a suitable location over the resolution reference sample without changing anything or moving the
sample. Select the areas from the 4 typical images (original images) I ()n=12,,  so that their
OSn
largest common sections are nearly centred (drift adjusted) as the assessment of the slow scan imaging
repeatability. Add the 4 original images I to the report, and add their selected (drift adjusted)
OSn
images for the overlay, and the overlay image I Valuate that the overlay image I is allowable or
DAO. DAO
not for the quantitative measurements, and add the valuations to the report.
Concerning the software-based technics, refer to the annexes “B.4 Measurement of the distortions
caused by high-frequency motions or stage vibration” and “B.5 Measurement of the drifts and drift-
compensated imaging”.
See Clause 12, Annex B and “B.3 Measurement of image drift and the drift-related distortions by using
image overlay” and “B.5 Measurement of the drifts and drift-compensated imaging”.
7.3 Evaluation of the drift and the drift-related distortions by using cross-correlation
function (CCF)
This method can be used to evaluate the quantities of the drifts and distortions with some numerical
indexes at the same time.
7.3.1 Measurement of the drifts by using the CCF
For the acquired images I , I ,  , I ,  , as shown in Figure 8, calculate the cross-correlation function
1 2 n
FI , I , where I is the image at the start of the measurement, and I ()n=12,, is the n-th image
()
CC 1 n 1 n
in the measurements. The function FI , I is same as the auto-correlation function FI of the
() ()
CC 11 AC 1
image I . Let HV, and HV, be the peak positions of the two-dimensional distribution of
() ()
1 PP11 Pn Pn
the FI and the FI , I respectively, and let 2 ≤ n , then Formula 2 and Formula 3
() ()
AC 1 CC 1 n
DH=−H (2)
HPnn P1
DV=−V (3)
VPnn P1
mean the drift quantities for horizontal (H) and vertical (V) directions. If D < 0, then the area of the
Hn
scan is shifted to the left and the obtained scanning image is shifted to the right. In a similar manner, if
D < 0, then obtained scanning image is shifted to the bottom. Calculate FI , I and obtain the
()
Vn CC 1 n
HV, for I n=23,,  then the drift DD, for each image I can be obtained.
() () ()
Pn Pn n HVnn n
In the case of SEM imaging, the electron beam is scanned from left to right in the horizontal (H) coordinate,
and scanned from top to bottom in the vertical (V) coordinate. Therefore, the larger value of the vertical
coordinate means lower position in the image. This coordinate system (H, V) is different from the usual
right-handed system (X, Y). Apart from the vertical direction, similar drifts data can be obtained as
shown in the tables from Table B.1 to Table B.4, and Figure 5 from subfigure a) to d). Valuate that the
drifts are allowable or not in the quantitative measurements for the specified (selected) images I .
n
Report the original specified images I ()n=12,, as shown in Figure 8, auto-correlation function
n
FI() and cross-correlation functions FI(), I as shown in Figure 9, the peak positions
AC 1 CC 1 n
HV, and HV, and the drifts DD, as listed in Table B.5 and the valuation.
() () ()
PP11 Pn Pn HVnn
See Clause 12, Annex B.6 and “B.6.1 Measurement of the drifts by using the CCF”.
a) The first image I b) The 2nd image I c) The 3rd image I d) The 4th image I
1 2 3 4
Figure 8 — Four consecutive, slow-scan images I n=12,,34, selected from Figure 6,
()
n
180 nm HFW
a) FI b) FI , I c) FI , I d) FI , I
() () () ()
AC 1 CC 12 CC 13 CC 14
Figure 9 — Examples for the auto-correlation function FI and the cross-correlation
()
AC 1
functions FI , I for n=23,,4
()
CC 1 n
14 © ISO 2021 – All rights reserved

7.3.2 Measurement of the distortions by using the CCF
Let FI and FI , I be the binarized pictures of FI and FI , I respectively by
() () () ()
BAC1 BCC 1 n AC 1 CC 1 n
using the thresholding level T which is the half of the maximum signal. If the signal range of the
B
function FI and FI , I is normalized to [0, 255], thenT = 255/2. Each binarized picture
() ()
CC 1 CC 1 n B
shows the cross section of the peak distribution at the signal intensity T = 255/2 as shown in Figure 10.
B
a) FI b) FI , I c) FI , I d) FI , I
() () () ()
BAC1 BCC 12 BCC 13 BCC 14
Figure 10 — E
...