ISO 4967:2026

Steel — Determination of the non-metallic inclusion content — Micrographic method

This document specifies a micrographic method of determining the non-metallic inclusions in rolled or forged steel products having a reduction ratio of at least 3 using the images of a standard reference chart or direct measurement by image analysis technologies. The standard reference chart described in this document is not entirely applicable for certain types of steel (e.g. free cutting steels).

Acier — Détermination de la teneur en inclusions non métalliques — Méthode micrographique

L'ISO 4967:2013 spécifie une méthode micrographique à l'aide d'images types, permettant de déterminer les inclusions non métalliques dans les aciers laminés ou forgés ayant un taux de corroyage d'au moins 3. La présente méthode est largement utilisée pour évaluer l'aptitude de l'acier à un emploi donné. Cependant, comme il est difficile d'obtenir des résultats reproductibles en raison de l'influence de l'opérateur, même avec un grand nombre d'échantillons, il convient de prendre des précautions lors de l'utilisation de la méthode. L'ISO 4967:2013 s'applique également à la détermination des inclusions non métalliques à l'aide de techniques d'analyse d'images.

General Information

• Status: Published
• Publication Date: 28-May-2026
• ICS: 77.040.99 - Other methods of testing of metals
• Technical Committee: ISO/TC 17/SC 7 - Methods of testing (other than mechanical tests and chemical analysis)
• Drafting Committee: ISO/TC 17/SC 7 - Methods of testing (other than mechanical tests and chemical analysis)
• Current Stage: 6060 - International Standard published
• Start Date: 29-May-2026
• Due Date: 16-Mar-2026
• Completion Date: 29-May-2026

Relations

• Revises: ISO 4967:2013 - Steel — Determination of content of non-metallic inclusions — Micrographic method using standard diagrams
• Effective Date: 18-Mar-2023

Overview

ISO 4967: Steel - Determination of the Non-Metallic Inclusion Content - Micrographic Method is an internationally recognized standard developed by ISO. This standard outlines a micrographic method for evaluating non-metallic inclusion content in rolled or forged steel products with a reduction ratio of at least 3. By utilizing standard reference diagrams and image analysis technologies, ISO 4967 provides steel producers, quality assurance professionals, and end users with a consistent approach to assess steel cleanliness and suitability for specific applications. Accurate determination of inclusion content is vital, as non-metallic inclusions can significantly influence mechanical properties, reliability, and processability of steel products.

Key Topics

• Non-Metallic Inclusions in Steel: Includes types such as sulfides, aluminates, silicates, and globular oxides.
• Micrographic Examination: Assessment using optical microscopy or digital imaging at a magnification of 100×, comparing observed features with standard charts.
• Inclusion Classification: Inclusions are categorized into Types A (sulfide), B (aluminate), C (silicate), and D (globular oxide), further subdivided by morphology and size (fine/thick classes).
• Specimen Preparation and Sampling: Guidance on specimen cutting, polishing, and appropriate sampling locations to ensure representative results and consistency.
• Inclusion Rating Methods:
  • Method A: Worst-field rating-records the highest observed severity for each inclusion type across all examined areas.
  • Method B: Field frequency method-counts or measures inclusion features across multiple fields for statistical representation.
• Image Analysis Options: Permits digital image analysis, improving reproducibility and reducing operator bias, with recommendations for system resolution and calibration.

Applications

ISO 4967 is widely applied in the steel industry for:

• Product Quality Control: Ensuring steel meets cleanliness requirements for critical applications such as automotive, aerospace, and structural components.
• Supplier Approval and Auditing: Supporting procurement decisions and supplier evaluation by providing objective data on steel inclusions.
• Research & Development: Comparing inclusion content after new processing methods or alloying strategies.
• Failure Analysis: Investigating component failures linked to non-metallic inclusions by quantifying their content and morphology.
• Customer-Supplier Agreements: Serving as a reference in contractual agreements for acceptance or rejection based on inclusion ratings.

Adherence to ISO 4967 enhances process control, minimizes variability, and increases confidence in end-product reliability, especially where steel performance is critical.

Related Standards

• ASTM E45: Standard Test Methods for Determining the Inclusion Content of Steel-a widely used method with comparable scopes, though terminology and details differ.
• ISO 4968: Micrographic determination of the ferrite or austenite grain size in steels.
• ISO 14284: Steel and iron - Sampling and preparation of samples for the determination of chemical composition.
• ISO 10118: Steel - Determination of inclusion content by image analysis.
• ISO 643: Steels - Micrographic determination of the apparent grain size.

Using ISO 4967 in combination with related steel analysis and quality standards ensures robust, comprehensive evaluation of steel cleanliness and material suitability.

---

Keywords: steel inclusions, steel analysis, micrographic method, ISO 4967, non-metallic inclusions, steel cleanliness, quality control, inclusion rating, image analysis, international steel standards

Standards Content (Sample)

International
Standard
ISO 4967
Fourth edition
Steel — Determination of the
2026-05
non-metallic inclusion content —
Micrographic method
Acier — Détermination de la teneur en inclusions non métalliques
— Méthode micrographique
Reference number
© ISO 2026
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: [phone]
Email: [e-mail]
Website: www.iso.org
Published in Switzerland
ii
Contents Page
Foreword .iv
Introduction .vi
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Symbols . 3
5 Principle . 4
6 Sampling . 6
7 Preparation of specimens .11
8 Determination of the content of inclusions .11
8.1 Method of observation .11
8.2 Actual examination . 12
8.2.1 General . 12
8.2.2 Method A . . . 12
8.2.3 Method B . . 12
8.2.4 General rules for methods A and B . 13
9 Expression of results .15
9.1 General . 15
9.2 Case of method A . 15
9.3 Case of method B . 15
10 Test report .15
Annex A (normative) ISO chart images for inclusion types A, B, C, D and subgroup DS . 17
Annex B (normative) Examples of assessment of inclusions .31
Annex C (informative) Examples of assessment of complex inclusions .35
Annex D (informative) Illustration of the width definitions .37
Annex E (informative) Assessment result examples and global inclusion content metrics .39
Annex F (informative) Relationship between indices and inclusion measurements .52
Bibliography .56

iii
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).
ISO draws attention to the possibility that the implementation of this document may involve the use of (a)
patent(s). ISO takes no position concerning the evidence, validity or applicability of any claimed patent
rights in respect thereof. As of the date of publication of this document, ISO had not received notice of (a)
patent(s) which may be required to implement this document. However, implementers are cautioned that
this may not represent the latest information, which may be obtained from the patent database available at
www.iso.org/patents. ISO shall not be held responsible for identifying any or all such patent rights.
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 17, Steel, Subcommittee SC 7, Methods of testing
(other than mechanical tests and chemical analysis).
This fourth edition cancels and replaces the third edition (ISO 4967:2013), which has been technically
revised.
The main changes are as follows:
— added the mandatory clauses normative references (see Clause 2) and terms and definitions (see
Clause 3), and renumbered the subsequent clauses;
— modified the proximity conditions for stringers (allowing for legacy conditions): the new transverse
conditions mirror the conditions used longitudinally and remove ambiguity;
— changed the width definition (allowing for legacy/alternative definitions): the new definition avoids
the sensitivity to misalignment of the bounding box and the underestimation of the “largest particle”
approach;
— added further illustrations of width definitions, including the largest particle approach to inclusions
with overlapping particles;
— clarified the treatment of inclusions intersecting the field of view, particularly for long inclusions
(allowing for legacy treatment);
— clarified the treatment of B/C hybrid stringers;
— modified Tables 2 and 3;
— added sampling specifications and the possibility to use stacking and/or rectangular fields for cross-
section thicknesses under 0,71 mm;
— included the DS subgroup into D thick rating;
— clarified the treatment of DS inclusions in Method B;

iv
— clarified the averaging of cross sections in Method A;
— modified most of the chart diagrams;
— replaced the global metrics for Method B (allowing for legacy metrics);
— added more analysis examples.
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.

v
Introduction
Every routine inclusion rating by necessity applies the analysis of an incomplete sample to an entire heat.
On the one hand, only a very small fraction of the total material volume is analysed, and on the other hand
the analysis is performed on a two-dimensional section of three-dimensional inclusions. Standards like this
document cannot eliminate the associated statistical uncertainties but can strive to add as little uncertainty
as possible by defining the process as clearly as possible.
Despite the statistical shortcomings, methods like those described in this document are widely used to
assess the suitability of a steel product for a given use. However, since it is difficult to achieve reproducible
results owing to the distributional randomness of non-metallic inclusions, even with a large number of
specimens, precautions should be taken when using the method.
One way to reduce the scatter inherent to the method is to avoid relying on subjective human judgment. Image
analysis has shown itself to be a useful tool to improve reproducibility — if the hardware is appropriately
configured and if the rules in the standard are indisputably clear for the software developer. This document
addresses the minimal system requirements for resolution and reduces ambiguity in its rules compared to
the previous revision.
However, it is acknowledged that neither steel producers and customers can instantly change specifications,
nor can software developers immediately change the rules for evaluation. To allow for an adaptation period,
where methods and definitions have changed, it is permitted to continue to use the methods and definitions
that have been established in the past. This document refers to such methods and definitions as well as
derived interpretations as “legacy.” Because of the ambiguities of previous editions, there is no one legacy
approach, but instead a variety of legacy approaches.
Another clarification relative to the 2013 revision concerns the DS inclusions. There was much ambiguity
surrounding them because they were presented as another type of inclusions. This made it unclear whether
large globular particles were part of the D rating as well, since one important rule of inclusion rating is to
rate every inclusion once and only once. With the redefinition of DS as a subgroup of type D designed for
easy rating and reporting of oversized type D inclusions it is clear that every DS particle is rated in the D
thick
rating, just as every oversized sulfide is rated in the A rating.
thick
Historically, ISO 4967 has always shown a significant similarity to the ASTM E 45 standard. With the
revised definitions, particularly those defining proximity limits, there is a greater separation between the
standards, though due to the inherent statistical uncertainties the ratings will correlate in most instances.
However, these increased differences convinced the ISO/TC 17/SC 7 to continue using the terms “fine” and
“thick” in order to more clearly distinguish ISO 4967 results from ASTM E 45 results.
Revisions always take place on a strict timeline and often the deadline forces the publication of a standard
that is good enough, but not yet perfect. Topics that further revisions can address include the treatment
of particle clusters, easier oversized reporting for Types A to C, and more guidelines on computer-assisted
rating.
It is worth remembering that the changes in this inclusion rating method do not change a good steel into a
bad steel, but serve the goal of a clearer, more differentiated description of the steel.

vi
International Standard ISO 4967:2026(en)
Steel — Determination of the non-metallic inclusion content
— Micrographic method
1 Scope
This document specifies a micrographic method of determining the non-metallic inclusions in rolled or
forged steel products having a reduction ratio of at least 3 using the images of a standard reference chart or
direct measurement by image analysis technologies.
The standard reference chart described in this document is not entirely applicable for certain types of steel
(e.g. free cutting steels).
2 Normative references
There are no normative references in this document.
3 Terms and definitions
For the purposes of this document, the terms and definitions given in the following apply.
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at www .iso .org/ obp
— IEC Electropedia: available at www .electropedia .org
3.1
particle
single precipitate, in general non-metallic
3.2
stringer
arrangement of at least 2 particles for type A and C inclusions and 3 particles for type B, aligned in a plane
parallel to the hot working axis and offset by no more than 10 µm, with a separation of no more than 40 µm
between any two nearest neighbour particles
3.3
inclusion
general designation of a rateable feature composed of at least one particle, defined by the size and proximity
of its constituents
Note 1 to entry: The inclusion can describe a single particle or a single stringer.
3.4
length
l
dimension of a particle or an inclusion in the main deformation direction, usually larger than the width

3.5
width
w
largest local dimension of a particle or an inclusion measured perpendicular to the main deformation
direction (calliper width), as shown in Figure 1
Note 1 to entry: The calliper width clarifies an ambiguity in previous editions. A “largest particle” and a “bounding
box” approach were among the most frequent alternatives. Further illustrations are shown in Annex D.
w
Figure 1 — Schematic of particle and inclusion width
3.6
aspect ratio
ratio of length to width
3.7
diameter
d
dimension of a globular particle or inclusion in the main deformation direction
3.8
worst-field rating
rating in which the specimen is rated for each group and subgroup of inclusion by assigning the value for the
highest severity rating observed of that inclusion group and subgroup anywhere on the evaluated area of
the specimen
3.9
type
categorization of inclusions according to morphology, colour and proximity
3.10
class
categorization by width or diameter within types (fine and thick)
3.11
group
categorization by type and class
3.12
globular particle
particle with an aspect ratio less than 3
3.13
elongated particle
particle with an aspect ratio more than or equal to 3
3.14
hybrid stringer
stringer consisting of both globular and elongated particles

3.15
reduction ratio
ratio between original and final cross-sectional area after rolling or forging
3.16
calibration factor
parameter indicating the actual size on the specimen surface corresponding to the pixel length
4 Symbols
The symbols used are given in Table 1.
Table 1 — Symbols
Symbols Definitions Values
A Area of an individual inclusion

Alw

A bar above the symbol indicates an average.
Superscripts used: variable index i, “excess”, vari-
able class
Subscripts used: selected type/class/group, “tot”
(total), S (specimen), “ox” (“oxidic”), “glob” (globu-
lar), fixed index (e.g. “i = 0,5”)
t Wall thickness, as shown in Figure 9
C Weighted global severity indices
i
Superscripts used: selected type/class/group,
“tot” (total)
The subscript “i” stands for “inclusion” and is not
the index number.
d Dimension of a globular particle or inclusion in
the main deformation direction
A bar above the symbol indicates an average.
Superscripts used: variable class c, variable
index i
Subscripts used: “min”, “max”, and selected type/
class/group
e Longitudinal distance between particles as shown
in Figures 11 to 12
f Inclusion area fractions
Subscripts used: “tot” (total), “ox” (“oxidic”),
“glob” (globular)
f Inclusion number frequency
number
Superscripts used: selected type/class/group
i Index
l Dimension of a particle or an inclusion in the main
deformation direction, usually larger than the
width
A bar above the symbol indicates an average.
Superscripts used: index i
Subscripts used: “min”, “max”
n Number of type D globular oxides per field
A bar above the symbol indicates an average.
Superscripts used: index i
Subscripts used: “min”
N Number of rated fields
r Width of plate as shown in Figures 4 to 6

TTabablele 1 1 ((ccoonnttiinnueuedd))
Symbols Definitions Values
R Relative area fractions
Superscripts used: selected type/class/group or
combination thereof such as “ox” (“oxidic”)
Subscripts used: Ratio expressed, e. g. “B:C”, “f:t”
(fine to thick), “G:nG” (globular to elongated)
s Transverse distance between particles as shown
in Figures 11 to 12
S Specimen area
w Dimension of a particle or an inclusion perpendic-
ular to the main deformation direction
A bar above the symbol indicates an average.
Superscripts used: class c, index i
Subscripts used: “min”, “max”, selected type/
class/group
x Inclusion interdistance
Subscripts used: selected type/class/group
Legacy global metrics
C Cleanness index (“t” stands for “total”)
t
 
 
CnF
t ii

 
 
i0

S
F Weighting factor of index i, always rounded to one
i25,
i
significant digit
15,
F 10
i
i Mean index for the entire assessed surface
moy
i
tot
i =
moy
N
i Total index for the entire assessed surface
tot
iin

tot i
i0
n Total number of fields (A, B, C, and D) and inclu-
i
sions (DS) rated as index i
5 Principle
5.1 The method consists of comparing the observed field to the chart images defined in this document and
taking in consideration separately each group and subgroup of inclusions. In the case of image analysis, fields
are rated according to Table 2, Table 3, and the relationships given in Annex F. Automatic image analysis can
be used provided that the accuracy of the method has previously been validated. Digitized images should
have a calibration factor of 1 µm/pixel or preferably finer.
5.2 The chart images correspond to square fields of view, each with an area of 0,50 mm , as obtained with
a longitudinal plane-of-polish and as observed with bright field illumination at 100×.
5.3 According to the grey level, shape, and distribution of the inclusions, the chart images are divided
into four main types, bearing the reference A, B, C, and D. These four types represent the most commonly
observed inclusion types and morphologies:
— Type A (“sulfide” type): highly malleable, individual light grey elongated particles with a wide range of
aspect ratios and generally rounded ends;

— Type B (“aluminate” type): numerous non deformable, angular, low aspect ratio (< 3), black or bluish
particles (at least three) aligned in the deformation direction;
— Type C (“silicate” type): highly malleable, individual black or dark grey elongated particles with a wide
range of aspect ratios (≥ 3) and generally smooth outlines, often with sharp ends;
— Type D (globular “oxide” type): non deformable, angular or circular, low aspect ratio (< 3), black or
bluish, randomly distributed particles.
NOTE The chemical composition of the inclusions present in a steel sample cannot be determined with the
methods described in this document. The apparent chemical names attributed to the types A, B, C, and D derive from
the typical composition historically found when analysing inclusions of such morphology and colour. The same is true
for the non-traditional inclusion types in the paragraphs that follow.
5.4 Non-traditional inclusion types may also be rated based on their morphology compared to the above
four types and a statement about their apparent chemical nature. As an example, globular sulfides would be
rated as a D type and denoted with a descriptive subscript (e.g. D ) defined in the test report. Examples
sulf
of subscripts include D (globular calcium sulfides); D (globular rare earth sulfides); and D (globular
cas RES Dup
duplex inclusions, such as calcium sulfide surrounding an aluminate). The treatment of complex inclusions
should be separately agreed between the parties. Examples are given in Annex C.
5.5 Types of precipitate, such as borides, carbides, carbonitrides or nitrides may also be rated based on
their morphology compared to the above four types and a statement about their apparent chemical nature
as described in 5.4. Examination at a magnification greater than 100 × may be used to identify the nature of
the non-traditional inclusions before performing the test.
5.6 The categorization of inclusions into types A, B, C and D is based on their grey level and then, after
forming stringers, on their morphologies. Each of the four types shall be further categorized into two classes
based on their width or diameter, as specified in Annex A, with six images representing increasing inclusion
content. The images for indices 3,5 and higher and the limit values for indices 5,5 and higher are not given
in this document. For simplifying the frequent oversized reporting for Type D inclusions with a diameter >
13 μm an additional subgroup DS is defined. Oversized Type D inclusions are rated as D and also as DS.
thick
5.7 The chart images in Annex A carry an index number, i, from 0,5 to 3, the numbers increasing with the
inclusion lengths (Types A, B, C) or by the number (Type D) or by the diameter (Subgroup DS), as defined
in Table 2, and categorized by thickness, as defined in Table 3. The total length of inclusions, or number
of inclusions, or diameter of the inclusion in each chart image is the lower boundary value of Table 2. For
example, the images for A 2 depict inclusions with a morphology in accordance with type A and with the-
total length corresponding to the lower boundary value of i = 2. For indices larger than 3,0, the rating is
performed by comparing measured values with Table 2.

Table 2 — Inclusion rating limits
Type Subgroup
Index
A B C DS
D
i
total length total length total length diameter
count number
µm µm µm µm
0,5 ≥ 37 ≥ 17 ≥ 18 ≥ 1 > 13
1,0 ≥ 127 ≥ 77 ≥ 76 ≥ 4 ≥ 19
1,5 ≥ 261 ≥ 184 ≥ 176 ≥ 9 ≥ 27
2,0 ≥ 436 ≥ 343 ≥ 320 ≥ 16 ≥ 38
2,5 ≥ 649 ≥ 555 ≥ 510 ≥ 25 ≥ 53
3,0 ≥ 898 ≥ 822 ≥ 746 ≥ 36 ≥ 76
3,5 ≥ 1 181 ≥ 1 147 ≥ 1 029 ≥ 49 ≥ 107

TTabablele 2 2 ((ccoonnttiinnueuedd))
Type Subgroup
Index
A B C DS
D
i
total length total length total length diameter
count number
µm µm µm µm
4,0 ≥ 1 498 ≥ 1 530 ≥ 1 359 ≥ 64 ≥ 151
4,5 ≥ 1 848 ≥ 1 973 ≥ 1 737 ≥ 81 ≥ 214
5,0 ≥ 2 230 ≥ 2 476 ≥ 2 163 ≥ 100 ≥ 303
(< 2 641) (< 3 042) (< 2 639) (< 121) (< 429)
⁞ ⁞ ⁞ ⁞ ⁞ ⁞
Table 3 — Inclusion width and diameter parameters
Type Fine class Thick class
µm µm µm µm
A (width) ≥ 2 ≤ 4 > 4 ≤ 12
B (width) ≥ 2 ≤ 9 > 9 ≤ 15
C (width) ≥ 2 ≤ 5 > 5 ≤ 12
D (diameter) ≥ 2 ≤ 8 > 8 ≤ 13
NOTE Type D inclusions with a width less than 2 µm are also not included in the inclusion rating.
6 Sampling
6.1 The shape of the inclusion depends to a large extent on the reduction ratio of the steel; therefore,
comparative measurements should only be carried out on prepared specimens taken from samples with a
similar reduction ratio.
6.2 Unless something else is defined in the product standard or agreed by the parties involved, the
polished surface of the specimen used to determine the content of inclusions should be about 200 mm .
6.3 When the cross-section thickness is insufficient to prepare a single specimen of 200 mm , more than
one specimen shall be taken from the same sampling location to conform to 6.2. Where reaching 200 mm
or more is onerous, the total length of the longitudinal pieces taken from each sampling location shall not be
less than 100 mm. These specimens shall be analysed as one whole specimen.
6.4 In general, single sections less than 0,71 mm in thickness are not analysed using this document, since
this is restricted by the field side length. However, as part of a mutual agreement, narrower sections may be
assessed by
a) stacking two or more sections together and thereby creating a thicker aggregate section, making sure
that no small gaps or edge artefacts between the stacked sections are categorised as inclusions
b) using a rectangular field as long as the field area is 0,5 mm , e.g. 1,0 mm×0,5 mm, provided that the
rectangular field fits within the visual field of the microscope at the applied magnification.
6.5 The method of sampling, including the sampling location and the number of sampling locations, shall
be defined in the product standard or subject to agreement between the parties.
6.6 In the absence of such specifications, the sampling procedure should be as follows:
— bar, wire rod, or wire with a diameter less than or equal to 25 mm: the surface to be examined consists
of the full diametral section of length sufficient to obtain a surface conform to 6.2 (see Figure 2);

— bar, wire rod, wire, or billet with a diameter greater than 25 mm and less than or equal to 40 mm: the
surface to be examined consists of at least half the diametral section (from the centre to the edge of the
sample) (see Figure 3);
— bar, wire rod, wire, or billet with diameters greater than 40 mm: the surface to be examined consists of
a part of diametral section located halfway between the outer surface and the centre (see Figure 4);
— plate with a thickness less than or equal to 25 mm: the surface to be examined consists of the whole
thickness, and located at the quarter of the width (see Figure 5);
— plate with a thickness greater than 25 mm and less than or equal to 40 mm: the surface to be examined
consists of at least half the thickness from the surface to the centre and is located at the quarter of the
width (see Figure 6);
— plate with a thickness greater than 40 mm: the surface to be examined consists of quarter the thickness
and is located halfway between the outer surface and the middle of the thickness and at the quarter of
the width (see Figure 7);
— tube or pipe with a wall thickness less than or equal to 25 mm: the surface to be examined consists of
the full diametral section of a length sufficient to obtain a sufficient surface, and, for welded products,
located far away from the welding bead (see Figure 8);
— tube or pipe with a wall thickness greater than 25 mm: the surface to be examined consists of a part
of the diametral section located halfway between the outer diameter and the inner diameter, and, for
welded products, far away from the welding bead (see Figure 9).
6.7 The number of samples to be taken is defined in the product standard or by special agreement. For any
other product, the sampling procedures shall be subject to agreement between the parties.

Figure 2 — Sample from bar with a diameter ≤ 25 mm

Figure 3 — Sample from bar or billet with a diameter or length of side > 25 mm and ≤ 40 mm
Figure 4 — Sample from bar or billet with a diameter or length of side > 40 mm

Key
r width
a rolling direction
Figure 5 — Sample from plate with thickness ≤ 25 mm
Key
r width
a rolling direction
Figure 6 — Sample from plate with thickness > 25 mm and ≤ 40 mm

Key
r width
a rolling direction
Figure 7 — Sample from plate with thickness > 40 mm
Figure 8 — Sample from tube or pipe with a wall thickness ≤ 25 mm

Key
t wall thickness
Figure 9 — Sample from tube or pipe with a wall thickness > 25 mm
7 Preparation of specimens
7.1 The specimen shall be cut so as to obtain a surface for examination. In order to achieve a flat surface
and to avoid rounding the edges of the specimen when polishing, the specimen may be held mechanically or
may be mounted.
7.2 When polishing specimens, it is important to avoid any tearing out or deformation of the inclusions,
or contamination of the polished surface, so that the surface is as clean as possible, and the shape of the
inclusions is not affected. These precautions are of particular importance when the inclusions are small. It
is advisable to use diamond paste for polishing. In certain cases, it can be necessary for the specimen to be
heat treated before polishing in order to give it the maximum possible hardness.
8 Determination of the content of inclusions
8.1 Method of observation
8.1.1 Examination with the microscope may be by one of two methods:
— by images on the computer screen, ground glass, or other similar device;
— by observation by means of an eyepiece.
8.1.2 The method of observation chosen shall be maintained throughout the test.
8.1.3 If the image is displayed on a computer screen, ground glass, or other similar device, the magnification
shall be 100 × ± 2 × on the screen. It is recommended to place an overlay (see Figure 10) of a 71 mm ± 1 mm
square (0,50 mm true area) over or behind the computer screen, ground glass, or other projection screen.
The image within the 71 mm square is compared to the chart images of the standard chart as specified in
Annex A.
8.1.4 If the inclusions are examined through the microscope eyepieces, insert a reticle with the test
pattern equivalent to the one shown in Figure 10 in the microscope at the appropriate location so that the
test grid area is 0,50 mm at the image plane. In some cases, a magnification greater than 100 × may be used,

provided that the same reference field size (0,5 mm ) is applied for the standard images. This approach shall
be recorded in the test report.
NOTE Separating the scale (one eyepiece) from the square and the reference shapes (the other eyepiece) allows
for measuring the length and width of an inclusion by rotating the scale independently and leaving the field reference
fixed.
Figure 10 — Suggested test pattern for grid overlays or reticles
8.2 Actual examination
8.2.1 General
Two methods are defined as described in 8.2.2 and 8.2.3. For both methods, if different types of inclusions
or inclusions of the same type but of a different class are observed in the same field, they shall be assessed
separately.
8.2.2 Method A
Method A is a worst-field rating method. The entire polished surface is examined and, for each type and
class of inclusions, a note is made of the index number i of the chart image which corresponds to the worst
field examined.
8.2.3 Method B
8.2.3.1 Method B is a field frequency method. The entire polished surface is examined, and each field of
the specimen is compared with the chart images. The index number i of the field which best corresponds
to the field examined for each type and class of inclusions is noted. Each inclusion group shall be assessed
separately, as specified in the assessment example in Annex B, B.1.

8.2.3.2 In order to minimize the cost of examination, it may be agreed upon to make a partial examination
of the specimen by studying a reduced number of fields, distributed in accordance with a fixed scheme. Both
the number of fields examined, and their distribution shall be arranged by prior agreement. In general, at
least 100 fields are examined.
8.2.4 General rules for methods A and B
8.2.4.1 Each field observed is compared with the chart images. If a field of inclusions falls between two
chart images, it is rated following the image with the lower index i.
NOTE The presence of numerous elongated inclusions oriented in a consistently different direction from the
presumed main deformation direction indicates a need to rotate the sample. The presence of numerous elongated
inclusions oriented in various different directions indicates that the sample is not suitable for the standard.
8.2.4.2 Individual inclusions that have a length greater than the field side length (0,710 mm) or a width or
diameter greater than the thick class maximum (see Table 3) will be rated as oversized by length, width, or
diameter. The oversized dimensions of the inclusion shall be noted separately. However, the total length of
these inclusions shall be included in the overall rating of that field as specified in the assessment example of
oversized inclusions in Annex B, B.2.
Legacy processes of dealing with inclusions longer than the field side length are permitted but shall be
declared in the report.
NOTE Custom methods of rating the severity of the oversized inclusions A, B, and C are a matter of agreement
between the parties.
8.2.4.3 Any inclusion intersecting the edge of the field of view shall be assigned to a field of view if the
centre of gravity of the inclusion lies within the field for Method A. For Method B, it shall be assigned to a
single field using a process that assigns all inclusions to exactly one field (e.g. centre of gravity, first found).
Legacy processes of assigning inclusions to fields are permitted but shall be declared in the report.
8.2.4.4 All oversized type D inclusions shall be rated as D thick and shall also be rated as subgroup
DS inclusions. For Method A, the DS rating of a specimen is the DS rating of the largest oversized type D
inclusion. For Method B, all the oversized type D inclusions shall be individually tallied and rated, regardless
of the proximity of other DS inclusions. An assessment example of oversized inclusions can be seen in B.3 of
Annex B.
8.2.4.5 The reproducibility of measurements is improved if actual measurements (inclusion lengths of A,
B or C types, diameter of DS subgroup) and counts (D types) are made. Use a grid overlay or reticle, as shown
in Figure 10, the measurement limits in Tables 2 and 3, and the morphological descriptions in Clause 5, as
illustrated in the chart.
8.2.4.6 Non-traditional inclusion types are rated according to the chart type and subgroup (A, B, C, D,
DS) that best corresponds to their morphology. Compare the length, number, thickness, or diameter of the
inclusions to each type shown in Annex A or determine their total length, number, thickness, or diameter,
and use Tables 2 and 3 to assign the appropriate index and class. Where necessary, a higher magnification is
used to better discern the individual constituents of complex inclusions. An assessment example of complex
inclusions can be seen in Annex C.
8.2.4.7 Stringers are formed from single particles of the same grey level category that meet the following
conditions. For type A and C inclusions, two individual particles of lengths l and l are considered as one
1 2
inclusion or stringer if the longitudinal distance e is lower than or equal to 40 μm and if the distance s (the
distance between facing tangent lines of particles) is lower than or equal to 10 μm (see Figure 11 and D.3).
For type B inclusions, three individual particles of lengths l , l and l are considered as one inclusion or
1 2 3
stringer if the longitudinal distance e is lower than or equal to 40 μm and if the distance s, (the distance
between facing tangent lines of particles) is lower than or equal to 10 μm (see Figure 12 and D.3). The

distance between centreline and centreline of particles is allowed to be used as a legacy definition. It will
gradually be phased out in future revisions. The use of legacy definitions shall be noted in the report.
s ≤ 10 µm, 0 ≤ e ≤ 40 µm
Figure 11 — Stringer for type A and C inclusions
s ≤ 10 µm, 0 ≤ e ≤ 40 µm
Figure 12 — Stringer for type B inclusions
8.2.4.8 In the case of an inclusion with particles of different width, the width to be considered is the
calliper width of the inclusion (see Figures 11 and 12) unless the largest particle width or another legacy
definition is agreed on between the parties.
8.2.4.9 Hybrid stringers composed of both globular and elongated black “oxide” particles are classified
as type B or type C. If the number of particles with l/w < 3 is less than 3, the inclusion is categorized as
type C. If the number of particles with l/w < 3 is three or larger, the inclusion is categorized as type B if the

total length of all individual particles with l/w ≥ 3 constitutes less than 50% of the total inclusion length;
otherwise, the inclusion is categorized as type C.
NOTE Figure C.2 of Annex C shows examples of how to categorize hybrid stringers when specific information on
the chemistry is available.
9 Expression of results
9.1 General
Unless otherwise stated in the product standard or agreement in parties, the results may be expressed as
described in 9.2 and 9.3 depending on the method used (method A or method B). Any subscripts used to
identify non-traditional inclusion types shall be defined.
9.2 Case of method A
9.2.1 For each specimen and each type and class of inclusions (see Annex B), the index i of the worst field
shall be reported, with the presence of an oversized inclusion being indicated by the letter s. The dimension
of the largest oversized inclusion shall be noted as well. For a single specimen, the notation may be made
using an “e” to indicate the rating for the “thick” class. The specimen 1 in Table E.1 would be noted A 2,0 / Ae
1,0 / B 2,0 / D 1,5 / Des 0,5 / DS 1,0.
9.2.2 On the basis of the index numbers relating to each specimen, an arithmetic mean may be assessed
per cast for each type and class. A typical result is given in E.1 of Annex E.
9.2.3 If fewer than three specimens per cast are rated, or if the surveyed area differs significantly between
specimens, the individual specimen ratings should be reported.
9.3 Case of method B
9.3.1 The number of fields corresponding to each index i, per specimen and per group and subgroup, shall
be reported along with the total number of fields observed (N). It is recommended to formulate acceptance
criteria for Method B as the maximum number of fields of a certain group per inspected area.
9.3.2 The set of results from 9.3.1 may be used in special treatments for expressing results as global
metrics, subject to agreement between the parties. Calculations and typical results are given in Annex E.
10 Test report
The test report shall contain the following information, if available and applicable:
a) a dated reference to this document and the method used, e.g. ISO 4967:2026 Method A;
b) the use of legacy approaches or alternative definitions for
1) width,
2) proximity conditions,
3) oversized inclusion length,
4) assignation to field of view, and
5) global metrics;
c) the steel grade and the cast number;

d) the nature (sheet, wire, rod, etc.) and dimensions of the product;
e) the relevant sampling information;
f) the magnification if greater than 100 ×, or the calibration factor for digital assessments;
g) the number of observed fields or the total area examined;
h) the results of the examination (including the number, size and type of oversized inclusions or stringers);
i) statement of subscripts used to define any non-traditional inclusion type;
j) unique report identification;
k) name of operator or unique operator identification.

Annex A
(normative)
ISO chart images for inclusion types A, B, C, D and subgroup DS
The standard reference chart images from index 0,5 to index 3,0 are given in Figures A.1, A.2, A.3, A.4 and
A.5, for types A, B, C, D and subgroup DS, respectively. All images represent a 710 µm × 710 µm square field
of view.The minimum total length of each index is given in Table 2.
a) A 0,5 (2 μm≤w≤4 μm, minimum total b) Ae 0,5 (4 μm length≥37 μm) total length≥37 μm)
c) A 1,0 (2 μm≤width≤4 μm, minimum d) Ae 1,0 (4 μm total length 127 μm) total length 127 μm)
Figure A.1 — Standard reference chart images of A type (1 of 3)

e) A 1,5 (2 μm≤width≤4 μm, minimum f) Ae 1,5 (4 μm total length 261 μm) total length 261 μm)
g) A 2,0 (2 μm≤width≤4 μm, minimum h) Ae 2,0 (4 μm total length 436 μm) total length 436 μm)
Figure A.1 — Standard reference chart images of A type (2 of 3)

i) A 2,5 (2 μm≤width≤4 μm, minimum j) Ae 2,5 (4 μm total length 649 μm) total length 649 μm)
k) A 3,0 (2 μm≤width≤4 μm, minimum l) Ae 3,0 (4 μm total length 8
...