ISO 13067:2020
(Main)Microbeam analysis — Electron backscatter diffraction — Measurement of average grain size
Microbeam analysis — Electron backscatter diffraction — Measurement of average grain size
This document describes procedures for measuring average grain size derived from a two-dimensional polished cross-section using electron backscatter diffraction (EBSD). This requires the measurement of orientation, misorientation and pattern quality factor as a function of position in the crystalline specimen[1]. The measurements in this document are made on two dimensional sections. The reader should note carefully the definitions used (3.3) which draw a distinction between the measured sectional grain sizes, and the mean grain size which can be derived from them that relates to the three dimensional grain size. NOTE 1 While conventional methods for grain size determination using optical microscopy are well-established, EBSD methods offer a number of advantages over these techniques, including increased spatial resolution and quantitative description of the orientation of the grains. NOTE 2 The method also lends itself to the measurement of the grain size of complex materials, for example those with a significant duplex content. NOTE 3 The reader is warned to interpret the results with care when attempting to investigate specimens with high levels of deformation.
Analyse par microfaisceaux — Diffraction d'électrons rétrodiffusés — Mesurage de la taille moyenne des grains
General Information
Relations
Standards Content (Sample)
INTERNATIONAL ISO
STANDARD 13067
Second edition
2020-07
Microbeam analysis — Electron
backscatter diffraction —
Measurement of average grain size
Analyse par microfaisceaux — Diffraction d'électrons rétrodiffusés —
Mesurage de la taille moyenne des grains
Reference number
©
ISO 2020
© ISO 2020
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 2020 – All rights reserved
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
3.1 Terminology associated with EBSD measurement of grain size . 2
3.2 Terminology associated with grains and grain boundaries determined via EBSD . 4
3.3 Terminology associated within grain size measurement . 5
3.4 Terminology associated with data correction and uncertainty of EBSD maps . 6
4 Equipment for grain sizing by EBSD . 7
4.1 Hardware requirements . . 7
4.2 Software requirements . 7
5 Acquiring the map for grain sizing by EBSD . 7
5.1 Specimen preparation . 7
5.2 Defining specimen axes . 7
5.3 Stage positioning and calibration . 8
5.4 Linear calibration . 8
5.5 Preliminary examination . . 8
5.6 Choice of step size . 8
[7][8]
5.7 Determination of the level of angular accuracy needed .10
5.8 Choice of areas to be mapped and map size .10
5.9 Considerations when examining plastically deformed materials .11
6 Analytical procedure .11
6.1 Definition of boundaries .11
6.1.1 Grain boundary angles .11
6.1.2 Handling incomplete boundaries .12
6.1.3 Dealing with special boundaries .12
6.2 Post-acquisition treatment of raw data .12
6.3 Data-cleaning steps .12
6.4 Measurement of sectional grain size .16
6.5 Calculation of average grain size .16
6.6 Representation of data .17
7 Measurement uncertainty .17
8 Reporting of analysis results .18
Annex A (informative) Grain size measurement .20
Annex B (informative) Reproducibility.22
Bibliography .25
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.
This second edition cancels and replaces the first edition (ISO 13067:2011), which has been technically
revised. The main changes compared to the previous edition are as follows:
— Data from a round robin (Annex B) have been used to:
— Include information on expected precision (Clause 7 and Annex B);
— Include more detail on sources of errors (Clause 7);
— Clarify statements on minimum numbers of grains measured (5.8) and acceptable clean up
procedures (6.3–6.3);
— Clarify the distinction between sectional grain size measured on a 2D section and average
grain size determined from some 2D measurements of grain sections which can be related by
stereology to the 3D grain size;
— Additionally, improvements have been made to the description of calculation of average values
(6.5) and representation of the data (6.6).
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.
iv © ISO 2020 – All rights reserved
Introduction
The mechanical and electromagnetic properties of engineering materials are strongly influenced
by their crystal grain size and distribution. For example, strength, toughness and hardness are
all important engineering properties that are strongly influenced by these parameters. Both bulk
materials and thin films, even as narrow two-dimensional structures, are influenced by grain size. For
this reason, it is important to have standard methods for its measurement with commonly used and
agreed terminology. This document describes procedures for measuring average grain size from maps
of local orientation measurements using electron backscatter diffraction.
INTERNATIONAL STANDARD ISO 13067:2020(E)
Microbeam analysis — Electron backscatter diffraction —
Measurement of average grain size
1 Scope
This document describes procedures for measuring average grain size derived from a two-dimensional
polished cross-section using electron backscatter diffraction (EBSD). This requires the measurement
of orientation, misorientation and pattern quality factor as a function of position in the crystalline
[1]
specimen . The measurements in this document are made on two dimensional sections. The reader
should note carefully the definitions used (3.3) which draw a distinction between the measured
sectional grain sizes, and the mean grain size which can be derived from them that relates to the three
dimensional grain size.
NOTE 1 While conventional methods for grain size determination using optical microscopy are well-
established, EBSD methods offer a number of advantages over these techniques, including increased spatial
resolution and quantitative description of the orientation of the grains.
NOTE 2 The method also lends itself to the measurement of the grain size of complex materials, for example
those with a significant duplex content.
NOTE 3 The reader is warned to interpret the results with care when attempting to investigate specimens
with high levels of deformation.
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, Microbeam analysis — Scanning electron microscopy — Guidelines for calibrating image
magnification
ISO/IEC 17025, General requirements for the competence of testing and calibration laboratories
ISO 23833, Microbeam analysis — Electron probe microanalysis (EPMA) — Vocabulary
ISO 24173, Microbeam analysis — Guidelines for orientation measurement using electron backscatter
diffraction
3 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO 24173 and ISO 23833 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 Terminology associated with EBSD measurement of grain size
3.1.1
step size
distance between adjacent points from which individual EBSD patterns are acquired during collection
of data for an EBSD map
3.1.2
pixel
picture element
smallest area of an EBSD map, with the dimensions of the step size (3.1.1), to which is assigned the
result of a single orientation (3.1.3) measurement made by stopping the beam at a point at the centre of
that area
3.1.3
orientation
mathematical description of the angular relationship between the crystal axes of the analysis point and
a reference frame, usually the specimen axes
[SOURCE: ISO 24173:2009, 3.16, modified to include different reference frames.]
3.1.4
indexed
meets the predetermined threshold for reliability for the orientation (3.1.3) of a pixel (3.1.2) calculated
from the EBSD pattern acquired for that pixel
3.1.5
indexing reliability
numerical value that indicates the confidence/reliability that the indexing software places in an
automatic analysis
Note 1 to entry: This parameter varies between EBSD manufacturers, but can include:
a) the average difference between the experimentally determined angles between diffracting planes and those
angles calculated for the orientation determined by EBSD software;
b) the difference between the number of triplets (intersections of three Kikuchi bands) in the EBSD pattern
matched by the chosen orientation and the next best possible solution, divided by the total number of
triplets.
3.1.6
orientation map
crystal orientation map
map-like display of pixels (3.1.2) derived from the sequential measurement of crystal orientation (3.1.3)
at each point in a grid [see Figures 1 b) to 1 f)] showing the crystallographic relationship between the
pixels and the reference frame
[SOURCE: ISO 24173:2009, 3.17, modified to include reference to examples.]
3.1.7
pattern quality
measure of the sharpness of the diffraction bands or the range of contrast within a diffraction pattern
Note 1 to entry: Different terms are used in different commercial software packages, including, for example,
band contrast, band slope and image quality.
2 © ISO 2020 – All rights reserved
3.1.8
pattern quality map
map-like display of pixels (3.1.2) derived from the sequential collection of EBSD patterns at each point
in a grid [see Figure 1 a)] showing the pattern quality (3.1.7) of the individual pixels
Note 1 to entry: Since measures of pattern quality can change at features such as grain boundaries and with
orientation, the pattern quality map can give an indication of grain shape and size.
Note 2 to entry: Pattern quality maps can also indicate areas of heavy deformation and inadequate preparation,
such as residual scratches.
Note 3 to entry: Small particles and features also contribute to the pattern quality map.
3.1.9
pseudosymmetry
potential for an EBSD pattern to be indexed in several different ways due to internal similarities within
the EBSD pattern
Note 1 to entry: Pseudosymmetry is a problem with some crystal orientations, usually when a main zone axis is
in the centre of the pattern. Typical cases are a {0001} pole for a hexagonal structure and a <111> pole for a cubic
structure.
Note 2 to entry: Structures such as high-symmetry tetragonal crystals with an axial ratio, c/a, αpproximately
equal to 1 are also likely to exhibit pseudosymmetry in EBSD patterns.
[SOURCE: ISO 24173:2009, 3.22]
3.1.10
misorientation
rotation, often defined by an angle/axis pair, required to rotate one set of crystal axes into coincidence
with the other set of crystal axes, given two crystal orientations (3.1.3)
3.1.11
disorientation
due to crystal symmetry, there can be several axis/angle pairs which represent the same misorientation,
in which case the one having the smallest angle is called the disorientation
Note 1 to entry: For most crystal symmetries, there are multiple symmetrically equivalent axes for the
disorientation with the smallest misorientation angle.
Note 2 to entry: Misorientation and disorientation are terms which are often used interchangeably. Disorientation
is the more rigorous term here, but misorientation is the more frequently used.
3.1.12
forescatter imaging
orientation contrast produced from electrons which channel out of the specimen
Note 1 to entry: Other contrast mechanisms such as composition can also affect the contrast obtained.
3.1.13
electron-channelling contrast imaging
ECCI
orientation contrast produced from electrons which channel into the specimen
3.1.14
barrel distortion
difference in lateral magnification between the central and peripheral areas of an image such that the
lateral magnification is less at the periphery
Note 1 to entry: A square object in the centre of the field appears barrel-shaped (i.e. with convex sides).
[SOURCE: ISO 10934-1:2002, 2.4.5.1]
3.1.15
pincushion distortion
difference in lateral magnification between the central and peripheral areas of an image such that the
lateral magnification is greater at the periphery
Note 1 to entry: A square object in the centre of the field appears cushion-shaped (i.e. with concave edges).
[SOURCE: ISO 10934-1:2002, 2.4.5.2]
3.2 Terminology associated with grains and grain boundaries determined via EBSD
3.2.1
grain boundary
line separating adjacent regions of points in an EBSD orientation map with disorientation (3.1.11) across
the line greater than a minimum angle chosen to define the grain boundaries
3.2.2
grain
region of points with similar orientation (3.1.3) (within a tolerance), completely enclosed by grain
boundaries (3.2.1) and greater than the minimum size defined to exclude isolated (often badly indexed
(3.1.4)) points as small grains
3.2.3
sub-grain boundary
line separating adjacent regions of points in a grain (3.2.2) with a difference in orientation (3.1.2) across
the line smaller than that defining a grain (3.2.2) but greater than that defining a sub-grain (3.2.4)
Note 1 to entry: Effectively, sub-grain boundaries are grain boundaries with a smaller misorientation limit than
that defining a grain boundary. These boundaries can have a characteristic linear appearance and exhibit a
characteristic misorientation.
3.2.4
sub-grain
region of points with similar orientation completely enclosed by boundaries greater than the minimum
sub-grain boundary (3.2.3) angle
3.2.5
special boundary
boundary between two grains (3.2.2) having a special orientation (3.1.3) relationship within a tolerance
associated with identifying them in orientation maps (3.1.6)
3.2.6
twin boundary
particular case of a special boundary (3.2.5) between crystals oriented with respect to one another
according to some symmetry rule, in which the boundary itself is planar and is a characteristic
crystallographic plane (for both crystals) and, frequently, one crystal is the mirror image of the other
Note 1 to entry: For example, in face-centred-cubic structures, the characteristic misorientation defining a
common twin can be described as a 60° rotation about the <111> axis with the boundary plane normal to the
rotation axis.
3.2.7
recrystallized grains
new set of undeformed grains (3.2.2) formed by consuming deformed grains through nucleation and
growth processes
Note 1 to entry: Measurements of misorientation within grains by EBSD can be used to distinguish between
deformed and undeformed grains.
4 © ISO 2020 – All rights reserved
3.2.8
phase
physically homogeneous volume in a material having the same crystal structure and chemical
composition
3.3 Terminology associated within grain size measurement
There are a variety of ways of representing average grain size. This subclause outlines some of the
more common terms used, and the reader is referred to Annex A for more details about other terms,
about the standards available and about the applicability of methods for particular grain shapes and
distributions.
3.3.1
3D grain size
three-dimensional size of a grain (3.2.2) or crystal within a polycrystalline material, measured as
a volume
Note 1 to entry: In a strict stereological definition, just the term grain size is sufficient to denote this value, but
it is recommended to use the full description 3D grain size to avoid confusion with the sectional grain size (3.3.8)
which is often shortened to grain size as well.
3.3.2
average grain size
value determined from a two dimensional measurement which is related to the average three
dimensional size of a collection of grains or crystals forming a polycrystalline material by stereological
[2]
relationships . It can be reported as one or more of the following measurements:
a) average area
b) average diameter determined from average area
c) average linear intercept length
3.3.3
line intercept
distance between the points at which a straight line crossing a grain intersects the grain boundary
(3.2.1) on each side
[15]
Note 1 to entry: See ASTM E112 for more details.
3.3.4
equivalent circle diameter
D
circle
diameter of the circle with an area equivalent to the grain section area, given by:
1/2
D = (4A/π)
circle
where A is the area of the grain (3.2.2)
[15]
Note 1 to entry: The ASTM grain size number, G, is given by :
G = −6,64log D − 2,95
10 circle
where D is measured in mm.
circle
3.3.5
Feret diameter
perpendicular distance between two parallel lines drawn in a given direction tangential to the
perimeter of an object on opposite sides of the object
Note 1 to entry: It is also known as the calliper diameter.
Note 2 to entry: Different variants of the Feret diameter are used. For example, the Feret diameter can be
measured in the vertical and horizontal directions or in any two directions at right angles to each other.
3.3.6
grain shape
property whose value is determined by fitting an ellipse round the grain (3.2.2) and measuring the
aspect ratio, i.e. the ratio of the length of minor axis to the length of the major axis
Note 1 to entry: It is sometimes referred to as grain elongation.
Note 2 to entry: The value lies in the range 0 to 1.
Note 3 to entry: There are several ways of fitting the ellipse round the grain, and different methods can result in
small differences in the measured aspect ratio.
3.3.7
grain shape orientation
angle between the major axis of an ellipse fitted round the grain (3.2.2) and the horizontal direction,
usually measured counter clockwise
3.3.8
sectional grain size
two-dimensional size of a planar cross section through a grain (3.2.2), reported as
a) the area of the cross section
b) a diameter (see circle equivalent diameter (3.3.4) or Feret diameter (3.3.5))
Note 1 to entry: This is often shortened to grain size in common parlance, but can lead to confusion with the 3D
grain size (3.3.1).
Note 2 to entry: This is equivalent to the term projected grain size.
3.4 Terminology associated with data correction and uncertainty of EBSD maps
3.4.1
misindexing
assigning an incorrect orientation (3.1.3) or phase (3.2.8) to the measured EBSD pattern
Note 1 to entry: This can occur for a number of reasons, e.g. pseudosymmetry effects, attempting to index a poor
pattern or attempting to index a pattern from an unanticipated phase for which the indexing software is not
configured.
3.4.2
non-indexing
non-assignment of an orientation (3.1.3) due to insufficient quality of the EBSD pattern
Note 1 to entry: This can occur for a variety of reasons, such as roughness of the specimen, dust on the specimen,
overlapping patterns at the grain boundary, a poor-quality pattern due to the effects of strain, or the pattern is
from an unanticipated phase.
6 © ISO 2020 – All rights reserved
3.4.3
data cleaning
process chosen to accommodate non-indexed (3.4.2) and misindexed (3.4.1) data within the map, using
a given set of parameters, typically based on the characteristics (orientation (3.1.3), phase (3.2.8)) of a
certain number of nearest neighbours
Note 1 to entry: A wide range of terms (not necessarily mathematically precise) is used in the various
commercially available software packages for different data-cleaning operations, including noise reduction,
extrapolation, dilation and erosion.
Note 2 to entry: See Figures 1 b) to 1 f).
4 Equipment for grain sizing by EBSD
4.1 Hardware requirements
The reader is referred to ISO 24173 for equipment needed to acquire electron backscatter patterns,
index the patterns (determine the orientation) and either step the beam across the specimen surface or,
less commonly, step the stage, keeping the beam stationary to acquire a map.
4.2 Software requirements
4.2.1 The software shall allow the orientation data (or other parameters, such as pattern quality
derived from each diffraction pattern) to be displayed as a map.
4.2.2 The software shall correct misindexed pixels or fill in non-indexed pixels (see 6.2 and 6.3).
4.2.3 The software shall use orientation data to define the positions of boundaries in accordance with
the criteria selected.
4.2.4 The software shall identify grains as regions of connected pixels from the set of boundary points
and measure grain size parameters. Special treatment may be applied to grains that intercept the map
edges, e.g. removal or weighting.
5 Acquiring the map for grain sizing by EBSD
5.1 Specimen preparation
In order to achieve a high degree of indexing of individual pixels (a high indexing hit rate), it is necessary
to produce a surface finish which produces EBSD patterns of sufficient quality to be indexed reliably.
The criteria used for indexing reliability shall be defined and reported by the user.
The surface preparation method adopted will be dependent on the material and also on its condition
e.g. metallurgical heat treatment. The reader should refer to standard texts on polishing and etching
and ISO 24173:2009, Annex B. Over-etching of grain boundaries should be avoided since it leads to
increased numbers of non- and mis-indexed points and to low index reliability at the grain boundaries.
If necessary, the specimen may be coated with a thin conductive coating (such as carbon) to prevent
charging and electron beam drift and thus avoid distortion of the image.
5.2 Defining specimen axes
If the specimen is known to be strongly textured, e.g. from thermomechanical processing, the axes of
the specimen shall be identified prior to preparation for EBSD such that EBSD measurements can be
related to these axes. These axes are usually related to the rolling direction (for metals), to a growth
direction (e.g., in thin films) or to a principal applied stress.
5.3 Stage positioning and calibration
The procedures set out in ISO 24173 shall be followed. The specimen shall be fixed to the scanning
electron microscope (SEM) stage in the desired orientation with the specimen axes relative to the stage
axes and imaged at a working distance at which the SEM and EBSD image magnifications have been
calibrated and at which the EBSD system itself has been calibrated to index diffraction patterns.
The purpose of this calibration is to check that there is no influence of distortion on the recorded
patterns and to ensure that the tilt angle relative to the specimen is correct. Reference [4] discusses
distortion round the edges.
The specimen tilt has a significant effect on the image magnification in the direction on the specimen
surface normal to the tilt axis. Great care shall be taken to measure the tilt angle of the specimen
surface accurately.
NOTE A 1° change in tilt angle at a tilt angle at 70° will cause a change of ∼5 % in the size of the step used in
the direction on the specimen surface normal to the tilt axis when collecting data for the map.
5.4 Linear calibration
Follow the recommendations of ISO 16700.
5.5 Preliminary examination
An initial examination of the specimen shall be made to identify an initial set of operating parameters
needed to map the orientation of the specimen with an acceptable level of accuracy and within an
acceptable period of time over an area sufficient to give data on a statistically significant number of grains.
The reader is referred to ISO 24173 for information needed to measure the orientation.
5.6 Choice of step size
5.6.1 If the grain size and shape are not known already, an approximate grain size and shape estimation
shall be performed by a quick imaging technique. An optical microscope might work on a region with
[5]
only slight polishing relief or on an etched region adjacent to that to be examined by EBSD. Forescatter
or electron-channelling contrast imaging using diodes mounted on the EBSD detector, or imaging with
the specimen current, can also produce images relatively quickly.
As an alternative to mapping, some EBSD software offers a line intercept method as a mapping mode.
This can be used to quickly give an approximate grain size measurement.
5.6.2 The step size should be chosen in relation to the average grain size, unless information on a
particular minimum size is required. In either case, it shall be recognized that a judgement is being made
on the minimum number of pixels that are used to define a grain either by a lineal or areal method. See
also 6.3 and Figures 1 d), e) and f) for the effects of step size choice.
8 © ISO 2020 – All rights reserved
Figure 1 — An area of a Ni superalloy mapped by EBSD under different conditions
Figure 1 shows, using Ni as an example material to illustrate general principles:
a) a pattern quality map (grey-scale range covering 20 to 160 of 256 grey levels), generated using a
0,5 µm step size;
b) from the same data set, the raw orientation map (96,7 % indexed) with non-indexed points in white
and inverse pole figure colouring of orientations (specimen normal direction, with key bottom right);
c) Figure 1 b) after removing clusters of 3 pixels or less and replacing the unindexed pixels by
orientations based on their six nearest neighbours (99,3 % indexed);
d) similar to Figure 1 c), but based on two instead of six nearest neighbours (99,8 % indexed);
e) the same area mapped with a 1 μm step size;
f) the same area mapped with a 2 μm step size.
Figures 1 c) to f) all use the orientation key shown in Figure 1 b) and show grain boundaries (>10°) in
black and twin boundaries (60° ± 1°, [111] ± 1°) in grey.
A simple rule that can be applied to a preliminary scan is that the step size should be less than 10 % of
[3]
the approximate mean grain size . Annex B summarises some effects of choice of step size determined
[6]
by an interlaboratory comparison . To confirm the validity of the chosen step size, repeat the mapping
of a single area at several step sizes and determine the maximum size below which no significant
difference in average grain size is determined. This choice has a direct influence on the accuracy of the
grain size measurement.
5.6.3 In choosing the step size, the spatial resolution of the system needs to be considered. The step
size is preferably larger than the interaction volume, which will be determined both by the material
examined and the operating parameters of the SEM, such as the filament type, accelerating voltage and
aperture size.
[7][8]
5.7 Determination of the level of angular accuracy needed
The speed with which EBSD patterns are acquired (including any averaging of patterns) may affect
the precision with which band edges can be detected and thus the angular accuracy of the calculated
orientation. Other factors, such as the Hough resolution and the number of bands chosen to match the
calculated orientation, also affect the calculation time as well as the angular accuracy.
If too long a time is taken for acquisition and calculation, problems of specimen drift can be increased
significantly, and fewer points will be acquired in a given time, reducing the statistical significance of
the data acquired. To minimize drift, it is recommended that the specimen have a good earth (ground)
path and is securely fastened to the stage. Avoid carbon tabs. A thin carbon coating might also be
necessary for insulating specimens.
If the time taken is too short, then levels of indexing reliability will be reduced. The settings chosen as a
compromise between the two opposing factors above shall be recorded.
In some cases, saving of EBSD patterns without indexing during mapping may speed up map acquisition.
Subsequent indexing off-line enables investigation of the effect of some of the above parameters on
indexing accuracy. The approach of saving patterns for subsequent analysis is strongly recommended,
not just as a way of speeding up data acquisition but also to allow flexibility in data interpretation.
5.8 Choice of areas to be mapped and map size
The areas chosen for examination shall be representative of the microstructure as a whole, and, if there
is variation with position in the specimen, the positions examined shall be recorded in relation to the
specimen geometry.
For conventional linear-intercept measurements, standards such as ASTM E112 recommend
[6]
measurement of a minimum of 50 grains from a minimum of 3 fields. An EBSD interlaboratory study
(Annex B) demonstrated that measurement of 300 grains generally produces a stable, repeatable value
on a running-average plot and that measurement of too few grains with a small step size can lead to
significant errors by missing large grains. It is therefore recommended that a minimum of 300 grains
per field is measured for a minimum of 3 fields, and that grains at the edge of each field shall be excluded
from calculations.
10 © ISO 2020 – All rights reserved
If this minimum proves impossible (because for example of a very large grain size) then grains
cutting the edge of a field may be included if the “bias” incurred by including partial grain areas can
[9]
be compensated for by, for example, the Miles-Lantuéjoul correction which assigns each particle a
weight that is proportional to the chance it has of being contained within the measurement field.
With some equipment and software, it is possible to join together EBSD maps of adjacent areas. This
should be avoided since joining maps in this way can lead to errors of alignment and the creation of
false boundaries. Since grain size is a statistical quantity, it is better practice to take measurements on
several separate areas.
NOTE 1 If statistical tools can be used to reduce errors in joining maps together, this process of grouping maps
could be of interest.
NOTE 2 Difficulties in aligning images might be caused by using too low a magnification, giving rise to
aberrations in the images, such as radial distortions (e.g. pincushion and barrel distortion) and scan rotation or
an incorrectly set up SEM that shows poor orthogonality in the scan. Orthogonality errors can be observed and
corrected for with the aid of a rectangular grid.
5.9 Considerations when examining plastically deformed materials
Where there is a high degree of damage, e.g. from plastic deformation, it might be impossible to obtain
good diffraction patterns. This makes indexing impossible or leads to inaccurate measurement of
orientation or phase. 6.2 and 6.3 consider the treatment of maps where this occurs, but it should be
noted that, in cases where a substantial number (>10 %) of pixels are not indexed reliably, this treatment
can distort the results and introduce significant inaccuracies.
Furthermore, deformation often leads to the formation of new grains and sub-grain boundaries.
However, there is no universally agreed definition of the misorientation angles that define these
boundaries since the significance of the boundary angle will vary depending to material type and
the property under consideration. Thus, even if good indexing is achieved, it is essential that any
measurement of size in a deformed material specify the misorientation angle used to define a grain
boundary.
Heavily deformed microstructures can also show significant anisotropy, and several definitions might
be needed to give representative descriptions of the grain size.
A further possible consequence of deformation, particularly at elevated temperatures, is the formation
of strain-free recrystallized grains. In such cases, these grains might have a significantly larger grain
size than the initial grains, resulting in a bimodal grain size distribution and the need to map at different
step sizes to resolve the distribution.
6 Analytical procedure
6.1 Definition of boundaries
6.1.1 Grain boundary angles
After following the steps above and acquiring data to plot, for example, maps of orientation, grain
boundaries can be drawn on the maps. This requires the angles defining the various possible boundaries
to be chosen. Guidelines for this are given below but, whether these or other methods are used, the
definitions and procedures used for grain size values shall be stated with all results.
For relatively simple equiaxed grain structures, such as fully recrystallized metals or undeformed cast
metals, the misorientation that is used to define the grain boundary may be taken to be as little as 5°.
For these types of material, there is evidence that misorientation angles between 5° and 15° make little
[10]
difference to the average grain size .
For other materials, with more complicated grain structures, larger angles, typically 10° or 15°,
depending on material, are used. Measurement of grain size as a function of misorientation angle might
be useful in gathering information on the structure. Care shall be taken to ensure that the prescribed
angle is not so large that it results in two or more distinct, preferred orientations being encompassed
[11][12]
within the angular range .
6.1.2 Handling incomplete boundaries
In some materials, particularly after deformation, selected boundaries might not extend completely
between two regions to terminate at a triple point with another boundary because the measured
misorientation changes along the length of the boundary fall below the defined grain boundary
angle. In such cases, it might be possible to extrapolate the boundary by reducing the minimum angle
generally used elsewhere in the map (see 5.6) to a new, lower, value. If this is done, it shall be recorded
with the final result (preferably reporting the effect on mean size with and without extrapolation). It is,
however, preferable to measure size with reduced misorientation angles defining grain boundaries and
to note the effect of this reduction on the measured grain or sub-grain size.
6.1.3 Dealing with special boundaries
With conventional techniques, special boundaries such as twins in cubic materials, which can be
identified by their morphology, are frequently ignored for the purposes of grain size measurement.
Since EBSD measures angle/axis quantitatively, these boundaries can be easily determined by software
and excluded from EBSD measurements of grain size. However, since there will be some variation in
measured angle and axis about the idealized value, the tolerances used to define the boundaries shall
be recorded (e.g. ±2° from 60° about <111>). The following should also be noted:
a) EBSD will define some boundaries as twins because they meet the defined misorientation
tolerances, whereas conventional optical microscopy would not identify them because the typical
morphology of twinning is not obvious. This effect can be reduced by only including those boundary
[13]
segments with a trace, which also satisfies the requirements for a twin plane .
b) Removal of grain boundaries in a) will lead to larger grain sizes than if twins are included.
6.2 Post-acquisition treatment of raw data
Rarely will every pixel in an EBSD orientation map be correctly indexed. In addition to errors in
orientation measurement for each pixel indexed, some pixels will not be indexed. The relative
proportions of these pixels will depend on specimen preparation, the nature of the specimen, the SEM
operating conditions and the EBSD indexing parameters.
In simple recrystallized specimens, it is normally possible to achieve a level of 95 % of pixels with
acceptably high index reliability. This level of 95 % should be the target for all maps, but it was shown
[6]
in an Interlaboratory study , (Annex B) that if misindexing is random then a minimum of 80 % with
acceptably high index reliability (and following data-cleaning steps described in 6.3) will still produce
acceptable results for mean grain size. However, incorrect or excessive manipulation of the raw data
can alter the final measured sizes significantly.
NOTE In multiphase materials, it can be necessary to treat each phase differently, using a separate dataset
for each phase.
6.3 Data-cleaning steps
6.3.1 Remove data for all grain sections with a number of pixels lower than a user-defined value
[6]
(typically 3 to 5) . See also 6.3.4 for removal of the smallest grain sections after image processing. The
threshold value and the number removed shall be recorded.
6.3.2 Index any singly unindexed pixel where it is surrounded by x or more pixels of the same
orientation, where the value of x is dependent on the grid used for mapping (square or hexagonal). Care
needs to be taken if this process is repeated several times (a single pass is generally sufficient for non-
indexed points at grain boundaries). The percentage indexed should not be increased by more than,
12 © ISO 2020 – All rights reserved
[6]
typically, 5 % but it has been shown (Annex B) that, in the case of random misindexing, up to 10 % of
pi
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.