Phaset 1

Phase Transformations and
Heat Treatments of Steels

Taylor & Francis
Taylor & Francis Group
http://taylorandfrancis.com
Phase Transformations and
Heat Treatments of Steels
Bankim Chandra Ray, Rajesh Kumar Prusty, and

Deepak Nayak
First edition published 2020
by CRC Press
6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742
and by CRC Press

2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
© 2020 Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, LLC
Reasonable efforts have been made to publish reliable data and information, but the author and
publisher cannot assume responsibility for the validity of all materials or the consequences of
their use. The authors and publishers have attempted to trace the copyright holders of all material
reproduced in this publication and apologize to copyright holders if permission to publish in this
form has not been obtained. If any copyright material has not been acknowledged, please write and
let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying, microfilming, and recording, or in any information
storage or retrieval system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, access www.copyright.
com or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923,
978-750-8400. For works that are not available on CCC, please contact mpkbookspermissions@
tandf.co.uk
Trademark notice: Product or corporate names may be trademarks or registered trademarks and are
used only for identification and explanation without intent to infringe.
ISBN: 978-0-367-02868-8 (hbk)

ISBN: 978-0-429-01921-0 (ebk)
Typeset in Times
by codeMantra
Contents
Preface....................................................................................................................... ix
Authors....................................................................................................................... xi
Chapter 1 An Introduction to Metals .................................................................... 1

1.1 Elements, Atoms, and Isotopes .................................................1
1.2 Types of Bonding between Atoms............................................. 2
1.3 Crystal Structures ......................................................................5
Further Reading .................................................................................. 19
Chapter 2 Diffusion............................................................................................. 21
2.1 Atomic Diffusion Mechanisms ............................................... 21
2.2 Types of Diffusion ................................................................... 22
Further Reading .................................................................................. 36
Chapter 3 Defects in Crystalline Solids .............................................................. 37

3.1 Introduction
............................................................................. 37
3.2 Classification ........................................................................... 37
Reference ............................................................................................ 50
Further Reading .................................................................................. 50
Chapter 4 Solid Solutions.................................................................................... 51

4.1 Introduction ............................................................................. 51
4.2 Types of Solid Solutions .......................................................... 51
4.3 Electron-to-Atom
Ratio ........................................................... 53
4.4 Enthalpy of Formation of a Solid Solution .............................. 54
4.5 Entropy of Formation of a Solid Solution................................ 56
4.6 Free Energy Change upon Formation of a Solid Solution....... 58
4.7 Ordered and Random Solid Solutions .....................................60
4.8 Intermediate Phases ................................................................. 61
Further Reading .................................................................................. 62
Chapter 5 Phase Diagrams and Phase Transformations ..................................... 63

5.1 Thermodynamic Considerations of Phase Diagrams .............. 63
5.2 Gibb’s Phase Rule ....................................................................64
5.3 Lever
Rule ............................................................................... 65
5.4 Types of Phase Diagrams and Phase Transformations............66
5.5 Some Other Solid-Phase Transformations in
Metals and Alloys .................................................................... 76
v
vi Contents
5.6 Roles of Defects and Diffusion ............................................... 77

Further Reading .................................................................................. 78
Chapter 6 Iron–Carbon Phase Diagram ............................................................. 79

6.1 Introduction
............................................................................. 79
6.2 Allotropic Transformations in Iron .........................................80
6.3 Solubility of Carbon in Iron ....................................................80
6.4 Iron–Iron
Carbide Phase Diagram .......................................... 82
6.5 Effect of Alloying Elements on the
Iron–Carbon Equilibrium Diagram ........................................ 86
Further Reading ..................................................................................90
Chapter 7 Thermodynamics and Kinetics of Solid-State Phase Transformation .....91

7.1
Introduction ............................................................................. 91
7.2 Nucleation ................................................................................ 91
7.3 Growth Kinetics ...................................................................... 98
7.4 Time–Temperature Transformation and Continuous
Cooling Transformation Diagrams........................................ 102
References ........................................................................................ 106
Further Reading ................................................................................ 106
Chapter 8 Phase Transformation in Steels ........................................................ 107

8.1 Introduction ........................................................................... 107
8.2 Formation of Austenite .......................................................... 107
8.3 Pearlitic Transformation ........................................................ 108
8.4 Bainitic Transformation......................................................... 111
8.5 Martensitic Transformation ................................................... 113
Further Reading ................................................................................ 116
Chapter 9 Heat Treatment Furnaces ................................................................. 117

9.1 Introduction
........................................................................... 117
9.2 Classification of Furnaces ..................................................... 117
9.3 Batch
Furnace ........................................................................ 118
9.4 Continuous Furnace............................................................... 123
9.5 Salt Bath Furnace .................................................................. 125
Further Readings .............................................................................. 125
Chapter 10 Heat Treatment Atmosphere............................................................. 127

10.1 Introduction
........................................................................... 127
10.2 Reactions between Atmosphere and Material ....................... 127
10.3 Types of Furnace Atmospheres ............................................. 129
Further Reading ................................................................................ 131
Contents vii
Chapter 11 Common Heat Treatment Practices.................................................. 133

11.1 Introduction ........................................................................... 133
11.2 Typical Heat Treatment Processes......................................... 133
11.3 Hardenability ......................................................................... 153
11.4 Case Hardening and Surface Hardening ............................... 158
11.5 Thermomechanical
Treatment............................................... 164
11.6 Heat Treatment of Carbon and Alloy Steels.......................... 166
Further Reading ................................................................................ 168
Chapter 12 Special Steels.................................................................................... 169

12.1 Stainless
Steels ...................................................................... 169
12.2 Hadfield Manganese Steels.................................................... 172
12.3 High-Strength Low-Alloy or Microalloyed Steels ................ 172
12.4 Transformation-Induced Plasticity Steels .............................. 172
12.5 Maraging Steels ..................................................................... 173
12.6 Dual-Phase
Steels .................................................................. 173
12.7 Tool
Steels ............................................................................. 173
12.8 Electric Grade Steels ............................................................. 177
Further Reading ................................................................................ 177
Chapter 13 Some In Situ Postweld Heat Treatment Practices ............................ 179

13.1 Necessity................................................................................ 179
13.2 Conventional Postweld Heat Treatment Process ................... 179
13.3 In Situ Postweld Heat Treatment of Transformation-
Induced Plasticity Steel ......................................................... 180
13.4 Postweld Heat Treatment of Duplex Stainless Steel.............. 182
References ........................................................................................ 184
Chapter 14 Heat Treatment of Cast Iron ............................................................. 185

14.1 Introduction
........................................................................... 185
14.2 Types of Cast Iron ................................................................. 185
14.3 Heat Treatment of Gray Cast Iron ......................................... 192
14.4 Heat Treatment of Malleable Cast Iron ................................. 193
14.5 Heat Treatment of Spheroidal Graphite Irons ....................... 194
Reference .......................................................................................... 196
Further Reading ................................................................................ 196
Chapter 15 Heat Treatment Defects and Their Determination ........................... 197

15.1 Distortion
............................................................................... 198
15.2 Warping ................................................................................. 198
15.3 Residual Stresses ................................................................... 199
15.4 Quench Cracking ................................................................... 199
viii Contents
15.5 Soft Spots...............................................................................200

15.6 Oxidation and Decarburization ............................................. 201
15.7 Low Hardness and Strength after Hardening ........................202
15.8 Overheating of Steel .............................................................. 203
15.9 Burning of Steel.....................................................................204
15.10 Black
Fracture .......................................................................205
15.11 Deformation and Volume Changes after Hardening ............. 205
15.12 Excessive Hardness after Tempering ....................................205
15.13 Corrosion and Erosion ...........................................................206
References ........................................................................................206
Further Reading ................................................................................206
Chapter 16 Some Special Heat Treatment Practices...........................................207

16.1 Automobile
Industries ...........................................................207
16.2 Aerospace Industries ............................................................. 210
16.3 Medical Equipment ............................................................... 214
16.4 Defense Industries ................................................................. 216
References ........................................................................................ 217
Index ...................................................................................................................... 219

Preface
The study of phase transformations, in terms of subtly underlying thermodynamics
and kinetics with the probability of attachment of atoms to critical nuclei, forms
the scientific aspect of formation of microstructures, and subsequently, the practical
practices of phase transformations are commercially achieved through heat treat-
ments. As a matter of fact, heat treatment solely involves managing the kinetics
of phase transformation to achieve desirable properties required in technological
applications. This book explains mechanisms, thermodynamics, and kinetics of
phase transitions in materials engineering field of steels and provides concepts to
commercialization-oriented understanding of heat treatment practices and principles
on the knowledge of phase transformations. Phase transformation and heat treatment
are indeed connected concurrently and, in the authors’ opinion, should be treated
as complementary subjects as opposed to distinct subjects. Phase Transformations
and Heat Treatments of Steels attempts to fill this gap, prevalent in the available
literature. The book’s unique presentation links basic understanding of theory with
application in a perpetually progressive yet exciting manner.
The book focuses on the processing–structure–properties triangle/structure–
property correlation, as it introduces the fundamental principles of physical met-
allurgy, phase transformations, and heat treatments of steel legacy as a structural
material. It supplies a broad overview of specific types of phase transformations,
supplemented by practical case studies of engineering alloys. The lucid, well-orga-
nized, and eminently readable text not only provides a proficient analysis of all the
relevant topics but also makes them comprehensible to the readers through the adept
use of numerous diagrams and illustrations. The book takes a pedagogical approach
and analyzes all concepts systematically and logically, making it ideal for those new
to the field.
While attempting to provide a comprehensive understanding of phase transfor-
mations with the practice of heat treatment and their interrelationships, the authors
have tried to provide insights into specialized processes and practices, and to convey
the excitement of the atmosphere in which new and different properties are intro-
duced and tailored.
Divided into 16 chapters, the material is organized in a logical progression,
beginning with fundamental principles and then building to more complex concepts
involved in phase transformation and heat treatment of almost all types of steels as
well as cast irons.
The book begins with a clear exposition of the basic concepts and definitions
related to metals, chemical bonding, and the structure of solids in Chapter 1.
Chapters 2 and 3 provide a theoretical background to solid-state diffusion and crys-
tal imperfections. These chapters also provide a deep insight into the structural con-
trol necessary for optimizing the various properties of materials. Then, a detailed
discussion on solid solutions with their types and the associated thermodynamics
follows in Chapter 4. The first part of Chapter 5 offers an in-depth analysis of phase
diagrams, while the latter part covers the structure and change of structure through
ix
x Preface
phase transformations. Chapter 6 discusses, in rich detail, the iron–carbon phase

diagram, which is indispensable in the field of metallurgy and materials engineering.
Chapter 7 emphasizes the thermodynamic and kinetic aspects of solid-state phase
transformation. This includes a comprehensive discussion on nucleation and growth
in the context of phase transformations in general. Chapter 8 extends the theory of
phase transformations to steels. Chapter 9 provides succinct summaries of common
furnaces used for heat treatment. This is followed by a thorough discussion on heat
treatment atmosphere in Chapter 10. Chapter 11 is devoted to a detailed descrip-
tion of the prevalent heat treatment practices and their purposes. It also highlights
the topics of hardenability, hardening treatment, and thermomechanical treatment.
Chapter 12 introduces certain special steels in order to pique readers’ interest for
understanding the current and futuristic aspects of steels application. Chapters 13
and 14 describe in situ postweld heat treatment practices and heat treatments of
cast irons, respectively. Chapter 15 is dedicated to discussing defects produced due
to heat treatment and their characterization. Chapter 16 contributes toward round-
ing out readers’ knowledge regarding certain practical and industrial heat treatment
techniques. The 16 chapters of the book are organized for almost linear purview in
a graduate-level course, in order to provide concrete understanding through steps.
Accordingly, each chapter serves a different purpose, but all chapters are connected
to provide an assimilated knowledge of phase transformation through proper cor-
relation between structural evolution and corresponding thermodynamic kinetics.
Designed primarily as an introductory text for undergraduate and postgraduate
students of metallurgy, the book also serves the needs of allied scientific disciplines
at the undergraduate and graduate levels. With its excellent balance between the
fundamentals and advanced information, it can also serve as an invaluable guide for
practicing professional engineers and scientists.
The authors are indeed privileged to extend their sincere appreciation and
acknowledgment to students, staff, scientists, and faculties of National Institute of
Technology, Rourkela, and CSIR-Institute of Minerals and Materials Technology,
Bhubaneswar, for their insightful suggestions incorporated into this scholarly manu-
script. We shall remain grateful to all around us here for their contributions to meet
the need of present students and scholars.
All family members of all authors are truly indebted in our hearts for their uncon-
ditional soulful support.
Authors
Bankim Chandra Ray, Phd, has worked at the National
Institute of Technology (NIT), Rourkela, India, since
1989. A dedicated academician with more than three
decades of experience, Bankim Chandra Ray is a full
professor since 2006 at the NIT, Rourkela. He earned
his PhD from the Indian Institute of Technology,
Kharagpur, India, in 1993. Apart from instructing
students in the field of Phase Transformation and Heat
Treatment, he has also guided many master’s degree and
PhD scholars. He has made seminal contributions in the
field of Phase Transformation and Heat Treatment and
Composite Materials.
An adept administrator, he has also served as the Dean of Faculty, Head of the
Department of the Metallurgical and Materials Engineering, and also an incumbent
Coordinator of Steel Research Center at the NIT, Rourkela.
His research interests are mainly focused on the mechanical behavior of FRP
composites. He is leading the Composite Materials Group at the NIT, Rourkela,
a group dedicated to realizing the technical tangibility of FRP composites
(https://www.frpclabnitrkl.com). With numerous highly cited publications in promi-
nent international journals, he has contributed extensively to the world literature in
the field of material science and engineering. He also holds a patent deriving from
his research. With nearly 155 publications in international journals, Prof. Ray also
authored many books/book chapters from leading publishers. He has been associated
with several prestigious societies such as Indian Institute of Metals, Indian National
Academy of Engineering, and several government and private institutes and organiza-
tions. As an advisor to New Materials Business, Tata Steel Ltd., he has been instru-
mental in facilitating the steel magnate’s foray into the FRP composites business.
His constant endeavors toward academics and his field of specialization have been
unparalleled, and yet thoroughly inspiring for many of the young engineering minds.
Rajesh Kumar Prusty, PhD, is an enthusiastic and

focused teacher who is committed to safeguarding and
promoting the education and well-being of students
and young people at all times in a friendly and con-
ducive atmosphere. Presently he is an assistant profes-
sor at the Department of Metallurgical and Materials
Engineering, NIT, Rourkela, India, from 2014. He has
graduated with a master’s degree (ME) in Materials
Engineering from Indian Institute of Science,
Bangalore, and earned a PhD from the NIT, Rourkela.
He is fascinated in opening up new dimensions of
xi
xii Authors
research particularly in the domain of structural materials with the ambition of mak-
ing these materials very reliable in the Indian market. He has authored more than 25
SCI/Scopus journal articles and one book related to FRP composite materials, and
at present, he is a principal investigator of three ongoing projects sponsored by the
DRDO, CSIR, and SERB. He actively contributes to the FRP composite lab at the
NIT, Rourkela (https://www.frpclabnitrkl.com/) in terms of visualizing and identi-
fying real-time problems associated with FRP composites followed by conceptual-
izing and designing experiments and work plan accordingly to address those issues
through constructive and logical techniques. He has supervised 16 master’s and 16
bachelor’s theses up to the present. Currently, he has five doctoral and five master’s
scholars working on FRP composites.
Deepak Nayak is a scientist at the CSIR-Institute of

Minerals and Materials Technology, Bhubaneswar,
since 2015. Born in 1989 in Bhubaneswar, he graduated
from the NIT, Rourkela, with a BTech in Metallurgical
and Materials Engineering before working as a gradu-
ate engineer trainee at JSL Stainless Ltd., New Delhi,
in 2010. Afterward, he returned to Bhubaneswar,
where he earned his MTech from Academy of
Scientific & Innovative Research (CSIR-IMMT) in
2013. He manages projects from concept to commer-
cialization in the areas of ferrous metallurgy, including
mineral beneficiation of low-grade iron ores, micro-
wave processing, and extraction of valuable metals.
1 An Introduction
to Metals
1.1 ELEMENTS, ATOMS, AND ISOTOPES

Chemical elements are the fundamental matters of all the materials. These elements
are chemically distinct and exhibit unique physical and mechanical properties. The
basic representative block of any element is again the atoms, which is comprised of
electrons (negatively charged), protons (positively charged), and neutrons (neutral).
Both proton and neutron are of almost similar weight, which is around 1.67 × 10 −27 kg,
whereas electron is much lighter in weight around 9.11 × 10 −31 kg. The weight of an
atom is almost the same as that of the nucleus, which contains the neutrons and
protons. However, the diameter of an atom (~10 −10 m or 1 Å) is quite larger than
that of the nucleus (~10 −14 m). The magnitude of charge of an electron and a proton
being equal, the atom contains exactly the same number of both the entities in order
to maintain electrical neutrality. The atomic number of an element indicates the
number of protons it possesses in a single atom which is also the same as the number
of electrons. However, the difference between atomic weight and atomic number
usually indicates the average number of neutrons in the atom. The periodic table
designed by the Russian scientist Mendeleev is the ideal tool to find out these num-
bers for any specific element. Almost all the empty cells in the original Mendeleev’s
periodic table have been filled due to discovery of new elements in subsequent time.
Some elements may have a higher or lower number of neutrons than that of the
electrons/protons. As these elements have the same number of electrons/protons,
the atomic number is not changed. However, due to different neutrons, the atomic
weight becomes different. These are called isotopes. The most common example of
isotopes is hydrogen. A high fraction of hydrogen atoms comprises only one proton
without any neutron, thus having an atomic weight of 1. A small fraction of hydro-
gen atoms contains one proton and one neutron giving rise to an atomic weight of 2,
which are most commonly known as deuterium. Another small fraction of hydrogen
atoms known as tritium (atomic weight 3) possess one proton and two neutrons. In all
these cases, the number of protons is 1, and all these elements have the same atomic
number, but with different atomic weight. Most of the elements in nature are mixture
of such multiple isotopes, and thus, the atomic mass is not always a whole number
and is the weighted average of the atomic weights of these isotopes. Continuing with
the example of hydrogen, in commonly available hydrogen, the isotopes are mixed
in such a proportion that the average atomic weight is 1.008. Taking the examples
of iron (Fe), it is naturally available in the form of four stable isotopes. The most
abundantly available form of iron is 56Fe (~91.754%) followed by 54Fe (~5.845%), 57Fe
(2.119%), and 58Fe (0.282%), giving rise to an average atomic mass of 55.85.
1
2 Phase Transformations and Heat Treatments of Steels
1.1.1 Types of elemenTs

Based on the broad physical and mechanical properties of elements, they are catego-
rized into three groups, i.e., (i) metals, (ii) metalloids, and (iii) nonmetals. Typically,
metals are of shiny lustrous appearance, when prepared fresh or fractured. They
are good conductors of heat and electricity. Metals can be plastically deformed and
are normally malleable (can be made to thin sheets) and ductile (can be drawn into
wires). Except mercury, all other metals remain in their solid state at normal room
temperature and exhibit crystalline arrangement of atoms. Around 91 out of 118 ele-
ments in the periodic table are metal. However, the exact number is not available, as
the boundaries between metals, nonmetals, and metalloids fluctuate due to lack of
globally accepted basis of categorization. Metals constitute around 25% of the earth’s
crust and are inseparable from the present era of civilization. To a large extent, the
development of civilization is driven by development in the field of metals and asso-
ciated products. In the same line, a nonmetal is defined as an element that lacks in
the metallic properties. Low density, boiling temperature, and melting temperature
are some key physical properties of nonmetals. Most of the nonmetals are gases at
room temperature and usually poor conductors of heat and electricity. Some nonmet-
als are brittle solids at room temperature but good conductors of electricity and heat,
e.g., carbon. Metalloids exhibit properties that are in between metals and nonmetals
or a mixture of metals and nonmetals. Boron, silicon, germanium, arsenic, antimony,
and tellurium are the well-accepted metalloids. They usually have metal-like lustrous
appearance but are only moderate conductors of heat and electricity.
1.2 TYPES OF BONDING BETWEEN ATOMS

In general, materials in their solid state exhibit a well-arranged array of atoms form-
ing a regular geometric pattern to reduce their free energy. However, solids such
as glass, wax, and paraffin do not follow this trend, and the arrangement of atoms
is similar to that of the liquid state and thus termed as amorphous solids. The next
important point now is to consider the types of forces responsible to have bonds
between adjacent atoms in order to complete the structure of the solid. Generally,
there exist four types of interatomic bonding.
1.2.1 IonIc Bond

Ionic bonding is preferable for elements having large difference in electron negativ-
ity. All the elements tend to become stable upon achieving octet configuration, i.e.,
exactly eight electrons on the outermost shell. However, many of the pure elements
do not satisfy this criterion. For example, sodium (atomic number 11) is having only
one electron in its outermost shell. Hence, the easiest way for sodium to achieve
octet configuration is to donate this single electron. However, as the process does not
involve transfer of protons, upon donating the single electron, it becomes positively
charged and is termed as a positive ion or cation. For sodium, it is denoted as Na+. All
the elements in group IA of the periodic table in this way tend to be single positive
cations due to their single valency (e.g., Na+, K+, Cs+).
An Introduction to Metals 3
In a similar way, group IIA elements become double positively charged upon
donating two electrons to acquire their respective octet configuration (e.g., Mg2+,
Ca2+, Ba2+). On the contrary, the number of electrons in nonmetals generally is close
to eight. Hence, the easiest way for them to achieve the stable octet configuration is
to accept the required number of electrons (eight – number of electrons in outermost
shell) from somewhere else to have exactly eight electrons at the outermost shell.
Considering the case of chlorine, which is having seven electrons in its outermost
shell, a single electron is required to be stable. If it gets that electron from some-
where, it becomes single negatively charged (as there is one more electron than the
number of protons). This negatively charged ion is termed as anion and denoted as
Cl−. Elements corresponding to group VIIA thus tend to be single negatively charged
(e.g., F−, Cl−, Br−). Similarly, elements in group VIA require two electrons to achieve
octet configuration and thus become doubly charged anions (e.g., O2−, S2−). Now,
this process of obtaining octet configuration is possible only when there is mutual
exchange of electrons, such that electrical neutrality is maintained. For example,
the electron donated by Na is accepted by the nearby Cl atom in order to neutralize
the charge and form a compound named sodium chloride (NaCl) or normal table
salt (Figure 1.1). One point here is to note that the properties of the ionic compound
formed have nothing to do with the properties of the individual elements, i.e., sodium
and chloride. Looking at NaCl, sodium is known as a very reactive element and
chlorine is a poisonous gas, but the table salt has neither of these properties, and
almost everyone uses this every day without any bad effect. This indicates that ionic
bond is a very strong bond. In case of oppositely charged ions with different charge
magnitude, a stoichiometry is maintained so as to establish electrical neutrality. For
example, one Mg2+ ion is bonded to either one O2− ion or two Cl− ions forming MgO
and MgCl2, respectively.
FIGURE 1.1 Formation of NaCl by ionic bonding between Na+ and Cl−.
1.2.2 covalenT Bond

In contrast to ionic bonding, in a covalent bonding, some common electrons are
basically shared among two atoms. This bond is mostly formed between atoms of
the same element or atoms of the elements that are closely placed in the periodic
table. To be more specific, this type of bonding is usually feasible for nonmetals and
between metals and nonmetals. Electronegativity of both the atoms being almost the
same, neither of them are expected to donate electron(s), and at the same time, these
elements normally exhibit high ionization energy. Hence, the best possible way for
them is to share some of their outermost shell electrons with each other in order to
obtain the stable octet configuration. Carbon is one of the most suited elements for
covalent bond, as it is having four electrons in its outer shell. Donating or accepting
four electrons to achieve the octet configuration is highly energetically unfavorable.
Hence, the carbon atom shares its four electrons with four other electrons by making
covalent bonds to achieve stability. For example, hydrogen is having only a single
electron in its outer shell and thus requires another electron to achieve stable gas
configuration. In order to satisfy this, the four outermost electrons of a single carbon
atom form covalent bonds with four such hydrogen atoms as shown in Figure 1.2. By
doing so, both carbon and hydrogen obtain the stable gas configuration.
Similarly, another common example of covalent bond is formation of hydrogen
gas (H2). Each hydrogen atom shares its electron with another atom to achieve stable
gas configuration. The strong attraction force or bonding in case of a covalent bond is
due to the electrostatic attraction of the shared pair of electrons to the nuclei of both
the atoms. For most of the gases, covalent bond is quite preferable.
1.2.3 meTallIc Bond

This type of bond is typically prominent in the case of metals in which there is lack
both in the availability of oppositely charged ions and in the number of electrons
required for covalent bond. Usually, metals are having a tendency to throw their
electrons from valence band to the conduction band owing to their high electrical
conductivity. This leads to formation of a negatively charged electron cloud around
FIGURE 1.2 Formation of methane by covalent bonds between carbon and hydrogen atoms.
FIGURE 1.3 Illustration of metallic bonding.
the atom. Metallic bond in this case is thus defined as the electrostatic attraction
between the positively charged nucleus and the negatively charged electron cloud.
In other words, it can also be stated as the sharing of the conduction electrons in a
positively charged cation’s structure. The metallic atoms thus are bonded together
due to their attraction toward the shared sea/cloud/swarm of free electrons around
them as shown in Figure 1.3. As this is a kind of sharing of free electrons among the
atoms as explained in covalent bonding, metallic bond very often can be considered
as extended covalent bond.
Many of the physical and mechanical properties of metals, such as density,
strength, ductility, and electrical and thermal conductivity, are due to this metallic
bonding among the atoms.
1.2.4 van der Waals forces

van der Waals forces are not of the nature of chemical bonds and usually distance
dependent. Unlike other atomic bonds explained earlier, these bonds are relatively
weak forces and gradually diminished with increasing distance between participat-
ing atoms/molecules. If no other force is present among neighboring atoms, then
there exists a critical distance between them after which their nuclei start repelling
each other due to similar nature of charge. Very often, van der Waals forces are
also termed as intermolecular forces. Dipole–dipole forces, dispersion forces, and
hydrogen bonding are different types of van der Waals forces. van der Waals forces
are operative at lower temperatures where thermal agitation of atoms is minimum.
1.3 CRYSTAL STRUCTURES
In a crystalline material, atoms occupy some specific points in the space. An exam-
ple of atomic arrangement is shown in Figure 1.4a. Atomic positions (center of the
atom) can be thought of the points of intersection of an array of infinite length lines.
FIGURE 1.4 Crystalline arrangement: (a) an array of atoms in space, (b) construction
of space lattice (a network of lines), (c) a representative volume element (unit cell), and (d)
parameters of a unit cell.
This array of straight lines is termed as space lattice (Figure 1.4b). These atomic
positions are identical to each other, or in other words, all the atoms are having the
same surroundings. A basis or motif can be defined as a group of atoms that are
located at a particular order with respect to each other. Combination of both the lat-
tice and motif gives rise to the complete crystal structure of the atoms. A unit cell
in a crystal is further defined as the smallest representative volume element (Figure
1.4c), which upon repetition in all three dimensions yields the complete crystal struc-
ture. A unit cell is a fundamental entity of all the crystalline materials. A unit cell
is constituted by its six parameters, three intersecting axial lengths (sides), and cor-
respondingly three angles between each pair of axes. The unit cell dimensions in
the X, Y, and Z (not necessarily perpendicular to each other) directions are normally
represented by a, b, and c, respectively. α, β, and γ represent the angles between Y and
Z, Z and X, and X and Y, respectively, as shown in Figure 1.4d.
All the possible space lattices can be formed from seven different crystal
systems or unit cells. These seven crystal systems are different from each other
based on the values of a, b, c, α, β, and γ. These possible crystal systems are
reported in Table 1.1.
Interestingly, most of the important metals crystallize in two particular crys-
tal systems, i.e., cubic and hexagonal, and only three Bravais lattices are common
such as face-centered cubic (FCC), body-centered cubic (BCC), and close-packed
hexagonal (CPH).
TABLE 1.1
Various Possible Crystal Systems and Corresponding Bravais Lattices
Crystal
System Sides Angles Possible Bravais Lattices
Triclinic a≠b≠c α ≠ β ≠ γ ≠ 90°
Simple triclinic
Monoclinic a≠b≠c α = γ = 90° ≠ β
Simple monoclinic
Base-centered monoclinic
Hexagonal a=b≠c α = β = 90°, γ = 120°
Simple hexagonal
(Continued )
TABLE 1.1 (Continued )

Crystal
Rhombohedral/ a=b=c α = β = γ ≠ 90°
trigonal
Simple trigonal
Orthorhombic a≠b≠c α = β = γ = 90°
Simple orthorhombic
Body-centered orthorhombic
Face-centered orthorhombic
(Continued )

Crystal
Base-centered orthorhombic
Tetragonal a=b≠c α = β = γ = 90°
Simple tetragonal
Body-centered tetragonal
Cubic a=b=c α = β = γ = 90°
Simple cubic
(Continued )

Crystal
Body-centered cubic
Face-centered cubic
1.3.1 face-cenTered cuBIc sTrucTure

In this structure, the unit cell is cubic in nature (a = b = c and α = β = γ = 90°), and
the atoms occupy the cube corners and face centers as shown in Table 1.1. An infinite
array of atoms with FCC arrangement is shown in Figure 1.5a. The smallest repeat-
ing unit of this can thus be extracted as shown in Figure 1.5b. Consider the right-
hand-side face of the cube in Figure 1.5b. This face is common to two unit cells, one
shown in the figure and the other just immediate next right to it.
As can be seen, the corner atoms are not touching each other. The face cen-
ter atom of this right-side face (shown as atom “X”) touches the four corner atoms
on that particular face, four face center atoms (top, bottom, front, and back) of the
shown unit cell, and similarly four face center atoms of the adjacent right unit cell.
Thus, an atom is having 12 nearest neighboring atoms, and therefore, the coordina-
tion number of an FCC atom is 12. A representative of the atomic location in the unit
cell is shown in Figure 1.5c. However, note that in this figure the spheres (corner and
face center) represent the center of the atom (readers should not assume the size of
this sphere to be the same as that of the corresponding atom; it is just for the sake of
simple representation).
Many of the ductile metals such as copper, aluminum, nickel, silver, and γ-iron
crystallize in the FCC form.
FIGURE 1.5 Face-centered cubic structure: (a) arrangement of atoms, (b) atomic position in
a single unit cell, and (c) representative unit cell with atomic locations.
FIGURE 1.6 Body-centered cubic structure: (a) arrangement of atoms, (b) atomic position
in a single unit cell, and (c) representative unit cell with atomic locations.
1.3.2 Body-cenTered cuBIc sTrucTure

In a similar way, a BCC crystal exhibits a cubic unit cell with atoms located at all
the cube corners and at the body center. The corner atoms do not touch each other.
However, the body-centered atom (shown as atom “Y”) touches all the eight corner
atoms of the unit cell. Thus, the number of nearest neighbors or coordination number
of a BCC atom is eight. The array of atoms, atomic arrangement in a unit cell, and a
representative unit cell have been represented in Figure 1.6a, b, and c, respectively.
Metals such as vanadium, sodium, tungsten, barium, and α-iron crystallize in the
BCC form.
1.3.3 close-packed Hexagonal

A CPH or hexagonal close-packed (HCP) structure usually comprises two parallel
planes containing 2-D hexagonal lattices with atoms placed at the corners and cen-
ter of the hexagons (Figure 1.7a). These closely packed planes are known as basal
planes (or A plane). In addition, on the horizontal plane (or B plane), which is located
FIGURE 1.7 Close-packed hexagonal structure: (a) arrangement of atoms, (b) atomic
position in a single unit cell, and (c) representative unit cell with atomic locations.
exactly at the vertical center of these two basal planes, three more atoms are pres-
ent in the form of a triangle as shown in Figure 1.7c. The basal plane can be divided
into six identical equilateral triangles, and the projections of the centers of the three
atoms of the middle plane coincide with the centroid of the alternate equilateral
triangles on the basal plane. As one can visualize, the central atom of a hexagonal
lattice on the basal plane (top face of the hexagonal lattice shown in Figure 1.7b,
shown as atom “Z”) makes bond with each of the six hexagonal corner atoms. In
addition, it also forms bonds with each three atoms of the B plane enclosed in the
unit cell. In a similar way, there also exists another B plane above the top face of the
hexagon corresponding to the next unit cell. Three atoms of this B plane form bonds
with atom “Z.” Thus, the coordination number of a CPH atom is 12. The vertical line
joining the face-centered atoms of the top and bottom surface passes through the
centroid of the equilateral triangle formed by the three atoms in the middle plane.
Metals such as cadmium, cobalt, zinc, magnesium, and α-titanium crystallize in the
CPH structure.
If one closely observes the atomic arrangements of FCC and CPH, there is a
very little difference. Figure 1.8 is provided to figure out the structural difference
in arrangement of atoms in FCC and CPH. Let us first consider a layer of closely
packed atoms as shown in Figure 1.8a. This is the most closely packed configura-
tion of atoms, or atomic plane with least possible free space. In this arrangement,
each atom is touching and surrounded by six atoms. Let this position be called as
A plane, i.e., the centers of all the atoms are placed in “A” position as shown in
Figure 1.8b. Now, there are two types of void location in this plane, i.e., “B” and
“C.” That means for close packing in the z-direction, the centers of the atoms of the
next layer must be placed either vertically above “B” or above “C.” Let us assume
the next layer to be of “B” type as shown in Figure 1.8c. After this, the next or third
layer will decide whether it will be of FCC or CPH structure. After placement of the
second atomic layer at “B” position, the centers of the atoms of the third layer may
be placed either vertically above “A” or above “C.” If it is placed at “A” position, it
is termed as “ABCABC…” stacking or FCC configuration as shown in Figure 1.8d.
On the contrary, if the third layer of atoms are placed at “A” position, it is termed as
“ABAB…” stacking, and consequently, the arrangement is known as CPH packing
as shown in Figure 1.8e.
FIGURE 1.8 Layer-by-layer arrangement of atomic planes in the case of FCC and CPH.
FCC, face-centered cubic; CPH, close-packed hexagonal.
Hence, the stacking sequence of atomic planes or the position of the atomic
centers of the atomic planes makes FCC different from CPH. However, it is very
evident from the previous explanation that the atomic packing is the same for both
the structures.
1.3.4 alloTropy
The term “Allotropes” comes from the Greek, where “allos” means another and
“tropes” means manner or style. Hence, allotropes means that an element may crys-
tallize in different crystal structure in its solid form. The best example of allotropes is
iron. When a liquid iron is cooled down at normal atmospheric pressure, it solidifies
at 1539°C, has BCC crystal structure, and is known as δ-iron. Upon further cooling,
this δ-iron converts to γ-iron at 1394°C. γ-Iron has FCC crystal structure and is stable
up to 912°C after which it again converts to a BCC structure known as α-iron. β-Iron
also exists but mostly becomes obsolete these days having the same crystal structure
as α-iron, but having paramagnetic behavior. Upon heating the ferromagnetic α-iron,
it loses its magnetism above the Curie temperature, which is 770°C, and becomes
paramagnetic β-iron. However, the crystal structures of both α-iron and β-iron are
identical. Another well-known element for allotropes is carbon. Carbon crystallizes
with several allotropic forms, i.e., diamond having tetrahedral structure, graphite
having hexagonal structure, and fullerenes including Bucky balls (C60). Other metals
such as tin, cobalt, and polonium also exhibit allotropes.
1.3.5 packIng facTor/effIcIency

Packing factor or packing fraction (PF) or packing efficiency of a crystal structure
is defined as the ratio of the volume of atoms enclosed in a unit cell to the total vol-
ume of the unit cell. This is basically an entity to estimate the extent of packing or,
in other words, the extent of available free space in the crystal structure. Let us thus
find out the PF of the aforementioned three most common crystal structures.
As per the definition of PF, it can be mathematically represented as
Total volume of the atoms enclosed in an unit cell

PF =
Total volume of the unit cell
The next term now comes is the effective number of atoms. Very often, an atom in
a unit cell does not completely belong to that particular atom; rather, it is shared
among some adjoining unit cells. For example, a face-centered atom is shared among
two unit cells. From this, the PF can be further simplified as
No. of effective atoms × volume of a single atom
PF =
Total volume of the unit cell
The number of effective atoms is basically the total number of atoms that solely
correspond to only a single unit cell. This can be further mathematically defined as
No.of effective atoms = ∑C n

i
i i
Ci = contribution of the ith type of atoms to the unit cell, and

ni = the total number of ith type of atoms in the unit cell.
In this way, let us find out the PF of the crystal structures.
1.3.5.1 Packing Fraction of a Body-Centered Cubic Structure

Let us consider a BCC single unit cell ABCDEFGH as discussed earlier and shown
in Figure 1.9a. Here, there are two types of atoms: (i) cube corner atoms and (ii) body
center atoms.
A cube corner is basically common to eight adjacent cubes. Hence, the atom at
the cube corner is shared by the same eight unit cells. As a consequence, 81 th volume
1
of the atom is enclosed by a single unit cell. Thus, CC = for the cube corner atom
8
FIGURE 1.9 Atomic arrangement of BCC crystal structure. BCC, body-centered cubic.
in this BCC structure, and there are eight such cube corners making nC = 8. Then,
consider the body center atom. The entire body center atom is enclosed in a single
unit cell. Thus, C BC = 1, and there is only one such body center atom in a unit cell
(nBC = 1).
No. of effective atoms = ∑C n = C

i
i i C × nC + C BC × nBC =
1
8
× 8 +1×1= 2
Assuming atoms to be of spherical shape and taking the atomic radius to be r and
lattice parameter to be a,
4 3
No. of effective atoms × volume of a single atom 2 × 3 πr
PF = =
Total volume of the unit cell a3
Now, we require a relation between r and a in order to find out the exact PF for the
structure. For this, consider the right angle triangle AEG as shown in Figure 1.9b.
EG corresponds to the face diagonal of the side EFGH having square shape. Thus,
the length of EG becomes “√2a” (Pythagorean theorem) as shown in the figure. The
size and location of the atoms are also clearly visible in the figure (i.e., the body cen-
ter atom touches all other atoms). Thus, the diagonal AG of the right angle triangle
AEG is “√3a.” From the figure, it can also be seen that AG is four times the radius
of the atom, i.e., AG = 4r. Hence,
3 a = 4r
4 3 8 3 8
2× πr πr π
Hence, PF = 3 = 3 = 3 = 0.68
a3  4 
3 64
 r 3 3
3 
Hence, for a BCC structure, the packing efficiency is 0.68 or 68%.

1.3.5.2 Packing Fraction of a Face-Centered Cubic Structure

In a similar way, the PF of an FCC crystal structure can also be determined. Consider
the unit cell of FCC crystal structure ABCDEFGH as shown in Figure 1.10a. The
atoms are located at the cube corners and at the face centers. Each face center atom is
1
shared by two adjacent unit cells having a common face. Hence, C BC = , and there
2
are total such face center atoms in each unit cell, i.e., nBC = 6. As explained in the
1
case of BCC, in the FCC structure also, CC = and nC = 8. The number of effective
atoms is thus 8
∑C n = C
1 1
No. of effective atoms = i i C × nC + CFC × nFC = ×8+ ×6 = 4
i
8 2
Assuming atoms to be hard spheres, the packing factor now is

4 3
No.of effective atoms × volume of a single atom 4× πr
PF = = 3
Total volume of the unit cell a3
Now, to establish a relation between a and r, consider the right angle triangle AEF as
shown in Figure 1.10b, where it can be clearly noticed that the face-centered atom is
touching all the corner atoms of the face. Hence, the diagonal AF of the right angle
triangle AEF is
AF = 2 a = 4r
Hence,
4 3 4
4× πr 4 × πr 3
PF = 3 = 3
3 = 0.74
a3 (
2 √ 2r )
Therefore, FCC is a denser structure than BCC having a PF of 0.74 or 74%.
FIGURE 1.10 Atomic arrangement of FCC crystal structure. FCC, face-centered cubic
1.3.5.3 Packing Fraction of a Close-Packed Hexagonal Structure

For obtaining the PF of a CPH structure, one first needs to determine the volume
of the unit cell. Unlike a cubic structure, the sides of a hexagonal structure are not
equal, i.e., a = b ≠ c.
If one clearly observes the arrangement of atoms in the CPH structure, it can be
noticed that in a unit cell, there exists a middle horizontal plane containing three
atoms, which is exactly midway between the two parallel hexagons (basal planes) as
shown in Figure 1.11a. The length of each side of the hexagon and the vertical dis-
tance between the hexagons are represented by “a” and “c,” respectively. As the basal
plane is the most closely packed plane, each atom touches six neighboring atoms.
Thus, the side of the hexagon is equal to the diameter of the atoms (a = 2r , r is the
radius of the atom). Furthermore, each of the hexagons of the unit cell can be divided
into six equilateral triangle as shown in Figure 1.11b. And the centers of the atoms
in the middle plane exist exactly at the middle points of the vertical lines joining the
centroids of each pair of the alternate equilateral triangles as shown in Figure 1.11b.
A simplified version of this arrangement has been presented in Figure 1.11c. The
atom “D” (an atom of the middle plane) sits just vertically above the centroid of the
equilateral triangle formed by A, B, and C atoms. However, all these four atoms
FIGURE 1.11 Arrangement of atoms in CPH structure and determination of its PF. CPH,
close-packed hexagonal; PF, packing fraction.
(A, B, C, and D) touch each other (discussed earlier). The volume of the hexagonal
unit cell can be obtained by multiplying the area of the hexagon with the height.
Therefore, a relation must be established between the lattice parameters “c” and “a”
in order to represent the volume only in terms of “a.” For that, consider the equilat-
eral triangle obtained by the centers of atoms A, B, and C as shown in Figure 1.11d.
E is the centroid of this triangle, and hence, the center of atom D is exactly vertically
above the point E. As the distance between the vertex and centroid of the triangle is
two-thirds of the length of the median,
2 a2 2 3 a
AE = × a2 − = × a=
3 4 3 2 3
Now, again a right angle triangle can be visualized, which is obtained by the points
A, E, and D as shown in Figure 1.11e. As the atom D touches all the atoms A, B,
and C, the distance between the centers of atoms D and A is equal to the diameter
of the atoms, which is again equal to the lattice parameter “a.” The vertical distance
between the centers of the atoms A and D is half of the height of the hexagon, i.e.,
c
DE =
2
As, DE = AD 2 − AE 2
c a2  a 
= a2 − As AD = a and AE =
2 3  3 
c 2
Or, = a = 0.816a
2 3
Or, c = 1.632a
Each atom at the corner of the hexagon is shared by six such unit cells, and thus, the
contribution of a single corner atom to a single unit cell is 16 , and there are six such
corner atoms in each of the hexagon making a total of 12 corner atoms. The face
center atom at the center of each basal hexagonal plane is shared among two such
unit cells; thus, its contribution toward the unit cell is 12 , and there are two such face
center atoms in a single unit cell. All the three atoms in the middle plane belong to
only a single unit cell. Thus,
1 1
No.of effective atoms = × 12 + × 2 + 1 × 3 = 6
6 2
Therefore,
4 3
No.of effective atoms × volume of a single atom 6× πr
PF = = 3
Total volume of the unit cell 33.92 r 3
Thus, PF = 0.74 or 74%
From the previous discussion, it is very clear that both FCC and CPH have the same
PF, which is also quite expected due to the layer-by-layer stacking of closely packed
atomic planes as discussed earlier. However, the BCC structure is less densely packed
in comparison to FCC and CPH.
FURTHER READING
Avner, S. H. Introduction to Physical Metallurgy. (Tata McGraw-Hill Education, New York,
1997).
Azaroff, L. V. Introduction to Solids. (Tata McGraw-Hill Education, New York, 1960).
Taylor & Francis
2 Diffusion
2.1 ATOMIC DIFFUSION MECHANISMS

Atomic diffusion in solids only occurs when atoms jump from one site to another.
Normally, atoms in a crystal oscillate around their equilibrium positions, and when
these oscillations become large enough, atoms change sites. The nature of atomic
movement in the solid decides the diffusion mechanism, i.e., interstitial and vacancy
mechanism.
Interstitial mechanism – an atom is said to diffuse by an interstitial mechanism
when it passes from one interstitial site to one of its nearest-neighbor interstitial
sites without permanently displacing any of the matrix atoms. For example, let us
consider the interstitial diffusion mechanism in the (100) plane of an FCC lattice,
as shown in Figure 2.1. The atom labeled 3 represents the position of the interstitial
atom before diffusion whereas the atom labeled 4 represents the position after dif-
fusion. However, the jump is only possible if the matrix atoms labeled 1 and 2 move
apart to allow the atom labeled 3 to go through them and occupy the position of the
atom labeled 4. The movement of matrix atoms will cause distortion, and it is this
distortion that constitutes the barrier to an interstitial atom changing sites. Generally,
the interstitial diffusion mechanism occurs in alloys, e.g., C in α- and γ-irons. It also
occurs in substitutional alloys like in radiation damage studies, energetic particles,
e.g., neutrons, can knock atoms off normal lattice sites into interstitial positions to
form “self-interstitials”.
Vacancy mechanism – the atom is said to have diffused by a vacancy mechanism
if one of the atoms on an adjacent site jumps into the vacancy. Let us consider a
close-packed plane of spheres in an FCC lattice, as shown in Figure 2.2. The atom
FIGURE 2.1 A schematic representation of interstitial diffusion mechanism on a FCC (100)

plane. FCC, face-centered cubic.
21
FIGURE 2.2 Movement of an atom by vacancy diffusion in an FCC lattice. FCC,

face-centered cubic.
diffuses to the vacant site, creating a vacancy on the site that the atom has just left.
Hence, the diffusion will depend upon the presence of a vacancy on an adjacent site,
and the rate of diffusion is dependent upon how easily vacancies can form in the
lattice and how easy it is for an atom to move into a vacancy.
2.2 TYPES OF DIFFUSION

2.2.1 InTersTITIal dIffusIon
There are two common types of diffusion process. Substitutional diffusion occurs
when substitutional atoms diffuse by the vacancy mechanism. On the other hand,
in the case of interstitial atoms, diffusion takes place by forcible movement of the
interstitial atom through the parent larger atoms, known as interstitial diffusion.
The interstitial diffusion process is discussed in detail in the following.
Consider a simple cubic arrangement of atoms and the solute atoms occupying the
interstitial positions. The atomic planes 1 and 2 contain n1 and n2 atoms per square
meter, respectively. Consider a very dilute solution such that each solute atom is
having six neighboring interstitial sites, which are unoccupied. In such case, if each
interstitial atom jumps γ times per second, then the number of atoms that will jump
from plane 1 to 2 in 1 s is given by
1
J1− 2 = γ n1 atoms m 2 s (2.1)
6
In the same way, the number of atoms that will jump from plane 2 to 1 is given by
1
J2−1 = γ n2 atoms m 2 s (2.2)
6
As n1 > n2, a net atom flux will be evolved from plane 1 to 2, which is represented by
1
Jnet = J2−1 − J1− 2 = γ (n2 − n1 ) atoms m 2 s (2.3)
6
Diffusion 23
If the jump distance or the separation between planes 1 and 2 is α, then the relation
between n1 and the concentration of solute atom at the position of plane 1, C1 will be
n1
C1 = (2.4)
α
Similarly,
n2
C2 = (2.5)
α
Hence, from equations 2.4 and 2.5, we have
n1 − n2 = α (C1 − C2 ) (2.6)
But, from Figure 2.3, we have
∂C
C1 − C2 = −α (2.7)
∂x
Therefore, substituting equations 2.6 and 2.7 in equation 2.3 will give
1 ∂C
Jnet = −  γα 2  atoms m 2 s (2.8)
6  ∂x
FIGURE 2.3 Interstitial diffusion process in a concentration gradient (the derivation of

Fick’s first law).
The concentration gradient in the above case ∂C / ∂ x is a partial derivative and can
be a function of time. Hence, it is obvious that there will be a flow of atoms in the
downward direction of the concentration gradient by a random jumping way.
1
Replacing D = γα 2 in equation 2.8 gives
6
∂C
Jnet = − D (2.9)
∂x
This equation is popularly known as Fick’s first law of diffusion, and D represents
the diffusion coefficient or intrinsic diffusivity. D has the unit of m2/s. Steady-state
diffusion is described by Fick’s first law which states that the concentration does
not change with time. Flux, which is the net number of atoms crossing a unit area
perpendicular to a given direction per unit time, is constant with time. However,
for most practical situations, concentration varies with distance and time. Such a
diffusion is called non–steady-state diffusion. In such cases, Fick’s first law can no
longer be valid, and therefore, we need to find out the variation of concentration with
distance and time. Let us consider a simple case, where there exists a concentration
profile in a single direction ( x ) only (Figure 2.4).
The number of interstitial atoms that diffuse into the thin slice of area A and
thickness δ x along the plane 1 in a very small time interval δ t may be represented by
J1 Aδ t. Similarly, the number of atoms that leave the slice across the plane 2 during this
time δ t is J2 Aδ t. Since some amounts of interstitial atoms get accumulated in the thin
slice, J2 < J1. So, the concentration of interstitial atoms within the thin slice is given by
( J1 − J2 ) Aδ t
δC = (2.10)
Aδ x
FIGURE 2.4 The derivation of Fick’s second law.

Diffusion 25
But δ x is small. So, we have
∂J
J2 = J1 + δx (2.11)
∂x
∂J
⇒ J1 − J2 = − δx (2.12)
∂x
Substituting the value of J1 − J2 in equation 2.10, we get
 ∂J 
δC =  −  δt (2.13)
 ∂x 
As δ t → 0, we have
∂C ∂J
=− (2.14)
∂t ∂x
From Fick’s first law, we know that
∂C
J = −D (2.15)
∂x
Combining equations 2.14 and 2.15, we get
∂C ∂  ∂C 
= D  (2.16)
∂t ∂ x  ∂ x 
This expression is commonly called as Fick’s second law of diffusion. In some cases,
D can be assumed to be independent of concentration, and under such condition,
equation (2.16) can further be simplified as
∂C ∂2 C
=D 2 (2.17)
∂t ∂x
A graphical representation of equation is presented in Figure 2.5. Figure 2.5a shows

∂2 C
that the concentration at all points increases with time, i.e., > 0, whereas
∂x 2
Figure 2.5b shows that the concentration at all points decreases with time, i.e.,
∂2 C
< 0.
∂x 2
2.2.1.1 Solution to the Diffusion Equation

The solution will be discussed in case of carburization of steel, which is commonly
encountered and has significant practical importance. During carburization of steel,
the carbon content at the surface is maintained at some constant value by controlling
the amounts of gas mixtures in which the steel is held for a certain period of time.
FIGURE 2.5 Concentration profile with time: (a) positive curvature and (b) negative
curvature.
Simultaneously, diffusion of carbon also takes place from the surface to the core of
the steel. The concentration profiles along distance were then plotted at different
time periods as shown in Figure 2.6. A more generalized mathematical formulation
then can be developed by taking into account the solved Fick’s second diffusion law
with some boundary conditions as stated in the following:
C (at x = 0) = CS (2.18)
C ( ∞ ) = C0 (2.19)
where C0 is the original carbon concentration of the steel.

Consider the case where the diffusivity of carbon in austenite has a direct rela-
tion with the concentration, and taking an infinite long specimen, an average value
may be helpful in order to attain the solution. In this case, the solution can be
written as
 x 
C = CS − (CS − C0 ) erf  (2.20)
 2 √ Dt 
FIGURE 2.6 Concentration profiles at different time intervals. C0 is the initial concentra-
tion of the steel, and CS is the surface concentration, which is maintained constant.
Diffusion 27
where “erf” is the usual notation for error function. The mathematical expression for
“erf” can be defined as
z
2
erf ( z ) =
√π ∫ exp ( − y ) dy
0
2
(2.21)
The error function values can be found from books of standard mathematical
functions. Since erf (0.5)  0.5, the depth of the carbon concentration is midway
between CS and C0 , which is given by
x
 0.5 (2.22)
2 √ Dt
⇒ x  √ Dt (2.23)
From this expression, it can clearly be visualized that there exists a direct relation-
ship between the depth of isoconcentration and √ Dt. For example, for a diffusivity
of carbon in austenite at 1000°C, D  4 × 10 −11 m 2 s, a carburized layer of 0.2 mm
( )
2
depth will take a time duration of 1000 s, i.e., 0.2 ×10 −3 4 × 10 −11. Similarly, there
exist many other examples apart from carburization of steel where the solution of the
diffusion equation can be applied. A summary of the solutions to different practical
situations is described in Table 2.1.
2.2.1.2 Effect of Temperature

Temperature has a vital influence on the diffusivity and diffusion rates. To find out
the effect of temperature on the interstitial diffusion in solids, let us consider the
jump technique of an interstitial atom, as shown in Figure 2.7.
TABLE 2.1
Solutions of the Diffusion Equations Encountered in
Different Practical Situations
Carburization  x 
C = CS − (CS − C0 ) erf 
 2 √ Dt 
Decarburization  x 
C = C0 erf 
 2 √ Dt 
Diffusion couple  C + C2   C1 − C2   x 
C= 1  −  erf  
 2   2  2 √ Dt 
C1 = concentration of steel 1
C2 = concentration of steel 2
Homogenization  πx   t
C = Cmean + β 0 sin   erf  − 
 λ   τ
β 0 = initial concentration amplitude
λ = half wavelength
τ = relaxation time
FIGURE 2.7 Effect of position of interstitial atom on the free energy curve.
The atom from the ground state jumps to another ground state through an
activated state, where it has to move its neighboring atoms elastically. When the
interstitial atom moves through the activated state, the parent atoms are forced apart
into higher energy positions. As a result, the net free energy increment for the system
is given by ∆Gm (m stands for migration). If the interstitial atom at the initial ground
state has a mean frequency of vibration ν along the x direction, then it attempts ν
∆Gm 
jumps per second. Out of this, exp  − attempts become eventually successful.
 RT 
If we consider a 3-D geometry of vibration, each atom, let us say, is surrounded by
z identical sites, where accommodation of this atom is possible, and the frequency of
this atomic jumping may be mathematically determined by the following expression.
∆Gm 
γ = zν exp  − (2.24)
 RT 
1
But we know that ∆Gm = ∆H m − T∆Sm and D = γ d 2 . So,
6
1 ∆Gm  2
D= zν exp  − d (2.25)
6  RT 
1 2 ∆S ∆H m 
⇒D= d zν exp  m  exp  −  (2.26)
6  R   RT 
∆Sm is due to thermal entropy contribution and independent of temperature. So,
D = D0 exp  ID 
Q
(2.27)
 RT 
Diffusion 29
FIGURE 2.8 Plot of plot log D vs. 1/T .
1 2 ∆S
where D0 = d zν exp  m  , and QID = ∆H m is called the activation enthalpy or
6  R 
interstitial diffusion. This activation energy can be determined from the diffusion
coefficients calculated at different temperatures.
D = D0 exp  ID 
Q
(2.28)
 RT 
QID
⇒ log D = log D0 − (2.29)
2.3 RT
So if we plot log D vs. 1/T, we can determine the activation energy for diffusion QID
Q
from the slope which is equal to – ID (Figure 2.8).
R
2.2.2 suBsTITuTIonal dIffusIon

In the case of substitutional diffusion, atom jumps from its position to a nearby
vacant lattice position. Hence, a free lattice site in the neighboring sites is the essen-
tial criterion for such diffusion to take place. The simplest form of substitutional
diffusion is the self-diffusion of atoms in a pure metal.
2.2.2.1 Self-Diffusion
The rate at which self-diffusion in metals takes place may be experimentally deter-
mined by incorporating some radioactive atoms (A*) to a pure substance (A). And
then the penetration rate may be elucidated at different temperatures. In terms of
chemical nature, both A* and A can be treated as identical. Hence, it is fair to assume
their jump frequency to be more or less same. Therefore,
1
DA* = DA = γ d 2 (2.30)
6
In substitutional diffusion, once the atom jumps into the vacancy, the atoms are also
having tendency to return back to the earlier location. Under such conditions, these
jumps do not contribute to the diffusive flux. So,
1
DA* = fDA = γ d 2 (2.31)
6
where f is the correlation factor. However, the effect of this parameter is negligible,
as f is having a value close to unity.
In the case of interstitial diffusion, the probability of successful jumps was given by
 −∆Gm 
exp 
 RT 
But the adjacent site may not be vacant for substitutional diffusion most of the time.
Therefore, the probability for having a vacant neighboring site is zX v , where z is the
number of nearest neighbors. The probability of having any vacant site is referred as
X v , i.e., vacancy mole fraction. Hence, the probability of a successful jump in case of
substitutional diffusion is given by
 −∆Gm 
zX v exp 
 RT 
Now, the number of such successful jump attempts per unit time performed by an
atom, which is vibrating with frequency ν, is given by
−∆Gm 
γ = ν zX v exp  (2.32)
 RT 
As discussed in earlier chapter, vacancies are thermodynamically stable imperfec-
tions, so
−∆Gv 
X v = X veq = exp  (2.33)
 RT 
Substituting equations 2.32 and 2.33 in equation 2.30, we have
1 2 − ( ∆Gm + ∆Gv )
DA = d z exp (2.34)
6 RT
Equation 2.34 is identical to that of interstitial diffusion with an additional term
representing the activation energy required for self-diffusion, i.e., ∆Gv . With this, let
us now find out the diffusion coefficient in a body-centered cubic lattice (assuming
(110) plane), as shown in Figure 2.9. It is considered that atom 9 lies on the plane of
the BCC lattice. The atom can jump to either 5 or 6, which are present on adjacent
plane. So, the probability of any atom on (110) plane of a BCC lattice to jump to its
adjacent plane is 2/8 = 1/4, where 8 is the coordination number of the atom. The flux
of atoms on plane is, therefore,
1
J1 = γ n1 (2.35)
4
Diffusion 31
FIGURE 2.9 Self-diffusion process in a (110) plane of a BCC lattice. BCC, body-centered
cubic.
where γ is the jump frequency and n1 is the number of atoms on plane 1. Similarly,
the flux of atoms on plane 2 is
1
J2 = γ n2 (2.36)
4
where n2 is the number of atoms on plane 2. Since, n1 > n2,
1
Jnet = J = J1 − J2 = γ ( n1 − n2 ) (2.37)
4
We know that if C1 and C2 are the concentrations of atoms on planes 1 and 2, respec-
tively, then
n1 n
C1 = ; C2 = 2 (2.38)
d d
From Figure 2.9, we get
∂C
C1 − C2 = − d (2.39)
∂x
Therefore, we have
∂C
n1 − n2 = − d 2 (2.40)
∂x
But, d is given by
a
d= (2.41)
h + k2 + l2
2
where a is the lattice parameter. So,
a2
d2 = (2.42)
2
Now, a is related to the jump distance α by
3a
α= (2.43)
2
Putting the value of a in terms of α in equation 2.42, we have
2α 2
d2 = (2.44)
3
Hence, equation 2.40 now becomes
∂C 2 ∂C
n1 − n2 = − d 2 = − α2 (2.45)
∂x 3 ∂x
So,
1  2 2  ∂C   1 2  ∂C  ∂C
J= γ  α −   = γα  −  = −D (2.46)
4  3  ∂x   6  ∂x  ∂x
1
where D = γα 2 is the diffusion coefficient.
6
2.2.2.2 Diffusion in Substitutional Alloys
In substitutional alloys, there is a difference in the rate at which solute atoms and
solvent atoms jump into the nearby available vacant sites. So, each of these atoms
must be having its own distinct intrinsic coefficient of diffusion DA or DB . According
to Fick’s first law, we have
∂CA
J A = − DA (2.47)
∂x
∂CB
J B = − DB (2.48)
∂x
where J A and J B represent the fluxes corresponding to the atoms of A and B across
a defined crystallographic plane.
Diffusion 33
FIGURE 2.10 Diffusion in substitutional alloys: (a) concentration profile after interdiffu-
sion of A and B, (b) variation of J A , J B, and JV with respect to position x, and (c) variation
in the formation and destruction rates of vacancies or destroyed across the diffusion couple.
Consider a welded joint made up of pure metals A and B. In such case, interdif-
fusion takes place among A and B across the welded zone. By allowing diffusion to
take place at relatively elevated temperature, a concentration gradient is expected to
develop, as represented in Figure 2.10a.
Assumption 1: There is no change in the net number of atoms existing per unit
volume of the metal. So,
CO = CA + CB (2.49)
∂CA ∂C
=− B (2.50)
∂x ∂x
This indicates that for both A and B, the driving force for diffusion, i.e., the
existing concentration gradient, remains the same in magnitude. However,
their gradients are directed in opposite directions. Now,
∂CA
J A = − DA (2.51)
∂x
∂CA
J B = DB (2.52)
∂x
Assumption 2: DA > DB . Hence, we have
JA > JB
Assumption 3: When an atom jumps to a nearby vacancy, it can also be

assumed that the earlier vacancy has moved to the earlier location of atom
as shown in Figure 2.11.
So, if there exists a net flux of atoms in a particular direction, then it is evident that
there must be a net flux of vacancies in the opposite direction. Now, the total flux of
vacancies is −J A − J B , because of the atomic migration of A ( −J A ) in addition to the
vacancy flux arising due to the diffusion B atoms (−J B). Since J A > J B , net flux of
vacancies is
JV = − J A − J B (2.53)
∂CA
JV = ( DA − DB ) (2.54)
∂x
FIGURE 2.11 Schematic representation of mutual exchange of locations among an atom

and a vacancy.
Diffusion 35
Now, there will be variation in JV across the diffusion couple, as shown in Figure 2.10b.
In order to maintain the vacancy concentration near equilibrium, vacancies must be
created on the B-rich side and destroyed on the A-rich side. The rate at which the
vacancies are created or destroyed is given by
∂CV ∂J
=− V (2.55)
∂t ∂t
This rate of vacancy creation and destruction varies as a function of distance across
the couple as represented in Figure 2.10c. Hence, there exists a net flux of vacan-
cies across the central plane of the diffusion couple. It results in the movement of
the lattice planes present in the nearby vicinity of the central plane of the couple to
the left. The flux of vacancies crossing a lattice plane hence affects the velocity of
movement of this lattice plane. Let us suppose that the lattice plane has an area S .
So, it will sweep out a volume of Sνδ t during a small time interval δ t. This volume
will contain Sνδ tCO atoms. Exactly this amount of atoms are displaced by an equal
number of vacancies passing through the plane at that time duration, i.e., JV Sδ t.
So, we have
JV Sδ t = Sνδ tCO (2.56)
JV = ν CO (2.57)
Combining equations 2.54 and 2.57, we have
1 ∂C
ν= ( DA − DB ) A (2.58)
CO ∂x
CA
But, we know that X A = . Hence, substituting the value of CA in terms of X A gives
CO
∂XA
ν = ( DA − DB ) (2.59)
∂x
However, in reality, compositional change may take place at a defined location
relative to the specimen ends.
Therefore, there is a need of deriving Fick’s second law of diffusion for substitu-
tional alloys. Let us consider a thin slice of material having thickness δ x present at
a fixed distance x from any of the ends of the diffusion couple. Let J A′ be the total
atomic flux through the constant plane having two contributions: one is diffusive
∂CA
flux, JV = − DA and the other is the flux due to the velocity of the lattice in
∂x
which diffusion is occurring, ν CA. So, we have
∂CA
J A′ = − DA + ν CA (2.60)
∂x
Substituting the value of ν in equation (2.60), we have
∂CA ∂X
J A′ = − DA + CA ( DA − DB ) A (2.61)
∂x ∂x
CA C
Since, X A = and X B = B , we get
CO CO
∂CA
J A′ = ( − DA + X A DA − X A DB ) (2.62)
∂x
Simplifying equation (2.62), we have
∂CA
J A′ = − D (2.63)
∂x
where D is the interdiffusion coefficient, which is defined as follows:
D = X A DB + X B DA (2.64)
Earlier, we had derived the following relation (see equation 2.14):
∂CA ∂J ′
=− A (2.65)
∂t ∂x
Hence, substituting equation 2.65 into equation 2.63, we have
∂CA ∂  ∂CA 
= D  (2.66)
∂t ∂x  ∂x 
For substitutional alloys, equation (2.66) is popularly known as Fick’s second law
for diffusion.
FURTHER READING
Porter, D. A. & Easterling, K. E. Phase Transformations in Metals and Alloys, Third Edition
(Revised Reprint). (CRC Press, Boca Raton, FL, 1992).
3 Defects in
Crystalline Solids
3.1 INTRODUCTION
Any discrepancy from ideal crystalline arrangement in a crystalline solid is known
as crystalline defect/imperfection. Imperfections are invariably associated with
crystalline solids. The dimension of this defect might be microscopic or macro-
scopic. Defect might influence most of the properties of the solid, e.g., the type and
density of the defects affects the strength, ductility, electrical conductivity, corrosion
resistance, etc. Not only this, defect also influences the thermodynamics and kinetics
of phase transformations in solids, Hence, it is very important to identify the defect
and the possible ways to control them to have a control over the overall properties
and performance of the material.
3.2 CLASSIFICATION
Classification of the defects has been made based upon the dimensionality of the
defect. The various classes of defects are as follows:
• Zero dimensional (or point defect)

• One dimensional (or line defect)
• Two dimensional (or interfacial defect)
• Three dimensional (or pores, voids, cavities, etc.)
A schematic representation to distinguish these defects from each other is given in

Figure 3.1. We shall discuss these defects in more detail in this chapter.
3.2.1 Zero-dImensIonal or poInT defecT

This defect has a size range of one or two atomic sizes of the same crystal.
3.2.1.1 Vacancy
Vacancy is the simplest point defect in case of metal. When any of the lattice point
of a crystalline solid is vacant, a vacancy is said to be created. A pictorial representa-
tion of vacancy is shown in Figure 3.1. Among all of the existing crystalline defects,
vacancy is the only equilibrium or thermodynamic defect (i.e., creation of vacancy
leads to decrease in free energy of the material). The decrease in free energy is
associated with the increase in conformational entropy.
To create a vacancy, one of the atoms needs to be removed from its original lattice
position. Hence, certain amount of energy is required for this. Let “Ev” be the energy
37
FIGURE 3.1 Schematic representation of crystalline defects: (a) 0-D defect (vacancy),
(b) 1-D defect (dislocation), (c) 2-D defect (grain boundary), and (d) 3-D defect (void).
required for creating a single vacancy. Hence, the total increase in energy of the
system for n vacancies corresponds to “nEv” (enthalpic contribution toward free
energy).
Next, consider after creation of “n” vacancies, there are “N − n” atoms occupy-
ing their original lattice positions. Hence, “n” vacancies and “N − n” atoms can be
N!
arranged over “N” lattice positions by “Ω” ways, where Ω = .
n!( N − n )!
Hence, the increase in conformational entropy (ΔS) can be calculated as
∆S = k ln Ω (3.1)
Now, the net change in free energy can be determined as
N!
∆G = nEv − T∆S = nEv − kT ln (3.2)
n!( N − n )!
Defects in Crystalline Solids 39
Incorporating Stirling’s theorem (ln a! = a ln a),
∆G = nEv − kT [ N ln N − n ln n − ( N − n) ln( N − n) ] (3.3)
To find out the equilibrium vacancy amount (ne), the differentiation of “ΔG” with
respect to “ne” should be zero.
∂( ∆G )
= 0 = Ev − kT  ln ( N − ne ) − ln ne 
∂ne
N − ne
Ev = kT ln
ne
N − ne Ev
ln =
ne kT
ne E
ln =− v
N − ne kT
ne E
= exp  − v  (3.4)
N − ne  kT 
Usually, the relative number of vacancy (ne) is very less in comparison to the number
of lattice positions (N) in any crystal, so “N − ne” can be considered as “N”. Hence,
ne E
= exp  − v  (3.5)
N  kT 
This is called the equilibrium vacancy concentration.

A schematic of the variation of free energy with respect to vacancy concentration
is shown in Figure 3.2. Vacancy may be created during solidification of the crystal
due to hindrance to crystal growth from neighboring crystal nuclei. Atomic diffusion
FIGURE 3.2 Variation of free energy with vacancy concentration.

may also be responsible for creation of vacancy in solid state itself. In solid state also,
heavy plastic deformation, rapid quenching, high energy particle bombardment, and
so forth are some ways of incorporating more vacancies to the system. However,
these nonequilibrium vacancies have a tendency to cluster, resulting formation of
divacancy or trivacancy and so on. Position exchange of a vacancy with its neighbor-
ing atom is an important phenomenon, which is known as diffusion in solids, and is
accelerated upon excursion to elevated temperature.
3.2.1.2 Self-Interstitial
Another point defect in case of pure metal is known as self-interstitial defect. When
an extra atom of the same species occupies any of the interstitial voids in the crystal,
it is known as interstitial defect. This results in a high amount of distortion, as the
size of the interstitial free space is significantly smaller than the size of the atom.
Hence, these interstitials are generally not produced naturally, rather due to external
events like irradiation. Figure 3.3 depicts the difference between vacancy and inter-
stitial defect in crystals.
FIGURE 3.3 Point defects in pure metal: (a) perfect crystal, (b) vacancies, (c) interstitials,
and (d) equal number of vacancies and interstitials.
3.2.1.3 Point Defect in Ionic Compound

Analysis of point defect in the case of ionic compound is complex, as charge
neutrality is an essential criterion in this regard. The ratio of the number of missing
cations and anions should be such that the overall charge is balanced. For simplic-
ity, let us assume an ionic compound where both cation and anion are having same
valency such that the formula for the compound is MX (Mn+ is the cation and Xn− is
the anion).
When an equal number of vacancies are created on M (or X) sublattice and the
same number of interstitial M (or X) are formed without interruption of X (or M)
sublattice, the defect is known as Frankel defect. When an equal number of vacan-
cies are present both in M and X sublattices, the defect is termed as Schottky defect.
This is shown in Figure 3.4.
3.2.1.4 Extrinsic Point Defect

Practically, it is impossible to obtain a perfect pure metal (100% purity) without any
impurity or foreign atoms. These foreign atoms constitute extrinsic defect in the
host metal, and if the foreign atom gets dissolved in the parent metal, it is known as
solid solution. Depending on the position of the foreign atom in the parent metal, it
is further classified into two classes. If the foreign atom occupies the interstitial void
FIGURE 3.4 Point defects in ionic compound structure: (a) perfect crystal structure, (b and c)
Frankel defect, and (d) Schottky defect.
FIGURE 3.5 Solid solution: (a) interstitial and (b) substitutional.
position in the metal, then it is referred to interstitial solid solution, and if the foreign
atom occupies the lattice position of the parent metal (in other words, replaces one
of the parent atoms from its position), it is known as substitutional solid solution. An
excellent example of interstitial solid solution is steel where the solute foreign atom
(carbon in this case) occupies the interstitial position between the parent iron atoms.
Typically, when the foreign atom is of very small size in comparison to the parent
metal atom, interstitial solid solution is formed. In the case of substitutional solid
solution, both the parent and the foreign atoms should have almost the same atomic
size. For example, Cu atom can substitute Ni atom without disturbing the crystal
arrangement of Ni lattice forming a substitutional solid solution. A schematic of the
interstitial and substitutional solid solution is shown in Figure 3.5.
3.2.2 one-dImensIonal or lIne defecT

In crystalline solids, line defect refers to those defects producing lattice distortion
around a line. This class of defect is very important as far as mechanical proper-
ties of the material are concerned. The one-dimensional or line defect in case of
crystalline solids is known as dislocation. Dislocations may be produced during
solidification or during plastic deformation or by vacancy condensation. Presence
of dislocations allows plastic deformation of the material far below the fracture
stress of the material without disturbing the crystal structure. Dislocation line
is defined as the boundary between the disturbed and undisturbed portion of the
crystal. Another important characteristic of dislocation is the Burgers vector. The
Burgers vector is the vector having slip direction, and magnitude is the interatomic
distance between two consecutive atoms along the slip direction. Based on the
geometry of dislocation, it is divided into two groups: (i) edge dislocation and (ii)
screw dislocation.
3.2.2.1 Edge Dislocation
When an extra portion of an atomic plane is present (or in other words, a portion
of an atomic plane is missing) in a crystal, the defect formed is known as edge
FIGURE 3.6 Schematic of edge dislocation.1
dislocation. This extra plane is known as extra half plane. A schematic diagram of
edge dislocation is shown in Figure 3.6.
To get an idea about the formation of edge dislocation in crystal, consider
Figure 3.7. Application of small shear force to a perfect crystal causes localized
plastic deformation on the slip plane (Figure 3.7a).
Because of the local deformation, the crystallographic arrangement is disturbed
in that vicinity, and formation of an extra half plane takes place (Figure 3.7b).
Dislocation line in case of edge dislocation is the edge of the extra half plane shown
as line AB. With increase in the magnitude of the applied shear force, the dislocation
line moves toward left, i.e., perpendicular to itself. The Burgers vector in this case is
shown in Figure 3.7b, which has existence on the slip plane along the slip direction.
Hence, in case of edge dislocation, the Burgers vector is perpendicular to the disloca-
tion line (from Figure 3.7).
FIGURE 3.7 Formation of edge dislocation: (a) perfect crystal and (b) crystal with an edge
dislocation.
A positive edge dislocation is represented by an inverted “T” symbol ( ) where

the extra half plane exists at the top of the crystal. When the extra half plane exists
at the bottom part of the crystal, it is referred to negative edge dislocation and
represented as ( ). In case of an edge dislocation, there exists a compressive stress,
where the extra half plane is present, and a tensile stress in the other part of the
dislocation line.
3.2.2.2 Screw Dislocation
When a part of the crystal has a relative angular orientation with respect to the other
part, the defect is referred as screw dislocation. The word screw is derived from the
spiral nature of the atomic arrangement around the dislocation line.
Because of the applied shear stress, a part of the crystal undergoes deforma-
tion as shown in Figure 3.8. A portion of the upper part of the crystal has been
moved partially with respect to the rest part of the crystal. The line XY represents
the boundary between deformed and undeformed portions of the crystal; hence,
it is the dislocation line. With a higher applied shear stress, the boundary moves
toward left. The Burgers vector is shown in Figure 3.8b, which represents the
direction of slip. From the figure, it is very clear that in case of a screw disloca-
tion, the Burgers vector is parallel to the dislocation line unlike in case of edge
dislocation. Similar to that of edge dislocations, two types of screw dislocations do
exist, namely positive or clockwise ( ) and negative or anticlockwise ( ) screw
dislocation.
3.2.2.3 Dislocation Movement
A common mechanism by which dislocation (dislocation line) moves is known as
dislocation glide. An essential criterion for dislocation glide is that, along with the
movement of dislocation line, the dislocation line and the Burgers vector should
always be present on the same slip plane. To understand this, again consider
Figures 3.7 and 3.8. With increase in the magnitude of applied stress, the dislocation
line moves in the direction of applied stress, but it always remains on the slip plane.
FIGURE 3.8 Formation of screw dislocation: (a) perfect crystal and (b) crystal with a screw
dislocation.
Similarly, although the magnitude of the Burgers vector increases, still it exists on
the same slip plane where the dislocation line is present.
Cross-slip is a mechanism through which a screw dislocation changes its slip
plane. The schematic of cross-slip is shown in Figure 3.9.
Consider a pair of slip planes ABCD and CDEF having line of intersection CD.
Let us assume that there is a screw dislocation with dislocation line XY (parallel
to CD) and the Burgers vector as shown in Figure 3.9 (parallel to XY) on the slip
plane ABCD. With the application of force, the dislocation line moves toward CD.
When the dislocation line coincides with CD, the dislocation line has now a provi-
sion to further move on the plane CDEF, as the Burgers vector at this stage will also
be present on the plane CDEF fulfilling the criterion of dislocation glide on CDEF
plane. This phenomenon of changing the glide plane of the dislocation is known as
cross-slip. But cross-slip is only possible for screw dislocation. In case of an edge
dislocation, cross-slip is not possible, which can be analyzed from Figure 3.10.
When the dislocation line will coincide with CD, the Burgers vector (perpendicu-
lar to XY and along BC) will not reside on the plane CDEF; rather, it would be at
FIGURE 3.9 Schematic of cross-slip of screw dislocation.
FIGURE 3.10 Schematic showing cross-slip is impossible in case of edge dislocation.

some angle to the slip plane CDEF. Hence, movement of dislocation is not possible on
the slip plane CDEF. Hence, in case of an edge dislocation, cross-slip is not possible.
Another mechanism through which an edge dislocation leaves its slip plane is known
as climb. This is related to vacancy or atomic diffusion. When a series of vacancies
replace the line of atoms at the edge of the extra half plane or when a series of atomic
line is added below the extra half plane (above the extra half plane in case of negative
edge dislocation), the dislocation line leaves the earlier slip plane and moves to another
slip plane parallel to earlier one. This phenomenon is known as dislocation climb. As
climb involves diffusion, the process becomes feasible only at higher temperature.
3.2.2.4 Existence of Dislocations in Crystals

In reality, pure edge or pure screw dislocation is rare to find in crystals. Rather,
most of the dislocations in crystalline materials are combination of both edge and
screw dislocations. A schematic representation of a dislocation in a crystal is shown
in Figure 3.11. It can be observed that the dislocation has different characters at
different locations. However, the Burgers vector is invariant and hence has same
magnitude and direction irrespective of the location. A dislocation also cannot be
terminated inside a crystal suddenly; it can only be terminated either in itself (loop)
or at a node or at the surface.
3.2.2.5 Dislocation Energy
Any dislocation has inherently a stress field associated with it. This stress is normally
within the elastic limit, and thus, an elastic strain energy is developed at the disloca-
tion. The elastic strain energy per unit length of dislocation can be determined from
the following expression:
1 2
E= Gb (3.6)
2
where G represents the shear modulus and b is the Burgers vector.
FIGURE 3.11 A schematic of existence of dislocation in a crystal.

3.2.3 TWo-dImensIonal or surface defecTs

Surface defects refer to regions of crystals, where distortions lie about a surface.
The thickness of the region might have dimensions around few atomic diameters.
The atoms at this surface do not satisfy the minimum energy criterion, and hence, the
energy of this surface is higher than that of the perfect crystal. Surface defect is
classified into two classes.
3.2.3.1 External Surface Defect

The free surface of any crystalline solid represents a higher energy state. This free
external surface represents the termination of the crystal. The coordination number
of the atoms on this surface is less than the ideal coordination number of the crystal
structure.
Consider Figure 3.12a, which represents a part of a perfect crystal, and assume
that it has infinite size. Hence, every atom is surrounded by four nearest neighbor
atoms in 2-D. To make a free surface, one has to cut into two parts along the line AB,
as shown in Figure 3.12b. To cut, all the bonds along that line are to be broken which
needs some energy, and that energy is now stored on both the surfaces, known as sur-
face energy. Thus, the coordination numbers of corresponding atoms whose bonds
are broken possess a lower coordination number less than the ideal one. Hence, the
free surface is a defect.
The free surface being of higher energy acts as catalyst for many reactions. Thus,
the process of corrosion starts from the free surface. Powder samples (higher free
surface area) thus are more prone to chemical reactions. During sintering of green
pellets, this high surface energy is accounted for the agglomeration of the fine pow-
der particles to reduce the free surface area and thus net reduction in surface energy.
3.2.3.2 Internal Surface Defect

This refers to surface defect that exists inside the solid. The most frequently
encountered internal surface defect is grain boundary. The interface between neigh-
boring grains is termed as grain boundary. During solidification of a polycrystal-
line material from its molten liquid bath, several nuclei are formed and they grow
simultaneously in their own crystallographic directions, and later they start touching
each other toward the end of solidification. Now, the atoms exactly at this interface
possess high energy, as they deviate from the ideal crystallographic arrangement.
FIGURE 3.12 Free surface as a defect: (a) perfect crystal and (b) cutting along AB to make
two free surfaces.
FIGURE 3.13 Grain boundaries in polycrystalline metal.
This orientation difference between two consecutive grains gives rise to grain
boundary defect. Although the structure of each gain is same (the values of a, b, c
and α, β, γ for each grain are the same), but the crystallographic axes of each grain
make some angle with the other, e.g., the “a” axes of grain 1 and grain 2 are not
unidirectional as shown in Figure 3.13. The angle between these two axes (“a” axis
of grain 1 and “a” axis of grain 2), known as orientation difference represents the
angle of the grain boundary.
The boundary between two adjacent grains is generally distorted and less packed
as that of bulk grains. The width of this zone is quite narrow and ranges up to five
atomic diameters. Due to less packing of the grain boundary, it exhibits a higher
energy than the grain. The extent of this energy further increases with more distor-
tion, i.e., more free space.
The orientation mismatch between two adjacent grains is defined as the misalign-
ment angle. Grain boundaries are further divided into two groups depending on the
misalignment angle. A low-angle grain boundary (LAGB) is defined as the grain
boundary with a misalignment angle less than 15°. These grains are also known
as subgrains. The LAGBs are generally formed by an array of dislocations. If the
LAGB is formed by edge dislocations, it is known as tilt boundary, and if it is
formed by screw dislocations, it is known as twist boundary. In contrast, in case of
a high-angle grain boundary, the grains are misoriented by an angle more than 15°.
3.2.3.3 Twin Boundary
Another example of two-dimensional defect is twins or twin boundary. A twin
boundary is defined as the boundary between a part of the grain and its mirror
image as shown in Figure 3.14. Generally, a twin boundary exhibits much lower
energy than that of grain boundary. Twin is again divided into two categories:
(i) mechanical/deformation twin and (ii) annealing twin. Deformation twins are
FIGURE 3.14 A schematic representation of twin boundary.
generally evolved during high strain rate cold working of the metals having few slip
systems. It happens when a part of the crystal is sheared by a specific magnitude
such that it becomes the mirror image of the remaining part of the crystal. Out of
the common crystal structures, deformation twins are more prominent in hexagonal
close-packed (HCP) (zinc, magnesium, etc.) and tetragonal (tin) structures due to
availability of less slip systems.
Annealing twins are mostly formed in face-centered cubic (FCC) metals during
recrystallization of deformed metals due to growth accidents. It is quite prominent
in copper, nickel, α-brass, and austenitic iron.
3.2.3.4 Stacking Fault
As the name suggests, it is the fault in the stacking sequence of a crystalline solid. For
example, the stacking sequence of an FCC crystal is …ABCABCABC…. However,
at instances, the stacking sequence of a part of the crystal may be in the form of …
ABCACABC…, due to some missing plane. Now, in this part of the crystal, the
stacking CACA represents an HCP system, but in reality, it is the stacking fault in the
FCC crystal. In another example, the stacking sequence of the same FCC crystal may
change to ABCACBCABC, due to existence of an additional plane. In this case, ACB
forms a twin. Therefore, this kind of stacking fault is also referred as microscopic
twin in FCC metals. The stacking fault energies (SFEs) of metals normally lie in the
range of 0.01–0.1 J/m2. A lower SFE metal has wider stacking fault, and thus, strain
hardens rapidly. SFEs of some common metals and alloys are provided in Table 3.1.
TABLE 3.1
SFEs of Some Common Metals and Alloys
Metal/Alloy Stacking Fault Energy (mJ/m2)
Brass <10
303 stainless steel 8
Silver ~25
Gold ~50
Copper ~80
Magnesium 125
Nickel ~150
Aluminum ~200
SFEs, stacking fault energies
3.2.4 THree-dImensIonal or volume defecT

Any defect having a three-dimensional structure is a volume defect. In metals,
various commonly observed volume defects are pores/voids, crack, impurities, and
so on. Pores/voids are normally introduced into the material during the fabrication
stage. Cracks are formed due to localized stress, which may be formed during han-
dling or during service condition. Both pores and cracks act as stress concentrators
and thus decrease the strength of the materials. Impurities also may be entrapped
in the metal during its processing, but in some cases, impurities are intentionally
introduced into the metals. These impurities sometimes obstruct the motion of the
dislocation and thus increase the strength of the metal. Such types of impurities are
termed as dispersoids. Sometimes, a new phase like precipitate may also be formed
in the material as a result of phase transformation, which also improves the strength
of the material by the precipitation hardening mechanism.
REFERENCE

FURTHER READING
Dieter, G. E. Mechanical Metallurgy. (McGraw-Hill, New York, 1988).
Abbaschian, R. & Reed-Hill, R. E. Physical Metallurgy Principles. (Cengage Learning,
Massachusetts, 2008).
4 Solid Solutions
4.1 INTRODUCTION
A solution is basically a homogeneous mixture of one or more types of solutes in a
solvent. A solvent is further defined as the component where a solute is dissolved,
and a solute is the entity that is dissolved in the solvent to obtain the homogeneous
mixture. Solute and solvent may be either solid or liquid or gas. Homogeneous in
this context is generally treated as a single-phase substance having uniform physi-
cal, chemical, and mechanical properties. As per the nature of the solvent, a solution
is classified as either solid or liquid or gaseous solution. The maximum limit of the
weight of the solute that can be dissolved in a solvent is termed as solubility and is
usually a function of temperature and pressure. Depending on this, a solution may
be termed as unsaturated, saturated, or supersaturated. For most of the solutions, the
solubility limit increases with increase in temperature. For example, the solubility
of sugar increases in water upon heating. Furthermore, the solubility limit generally
is higher in the liquid state than that of the solid state. In metallurgy, the focus is
usually on solid solutions/alloys, where the solvent is primarily a solid-state material
at ambient environment, e.g., steel is a solid solution of carbon in iron. Here, carbon
acts as the solute, whereas iron behaves like the solvent. In case of a solid solution,
the crystallographic arrangement of atoms of the solvent phase remains unaltered by
the process of solution formation. In context to solutions, we usually encounter the
term solubility. Solubility is defined as the maximum amount of the solute that can
be dissolved in a solvent at a given temperature and pressure combination. It comes
from the concept of free energy change, i.e., a solution is evolved only when there is
a decrement in the net free energy of the system. The properties of solid solutions to
a large extent are dependent on their composition. In addition, the type of phases and
their morphologies may also influence the bulk performance of the solid solution to
a significant degree.
4.2 TYPES OF SOLID SOLUTIONS

Solid solutions are classified into two groups based on the atomic arrangement of the
solute atoms in the solvent lattice.
4.2.1 InTersTITIal solId soluTIon

An interstitial solid solution is said to be formed only when the solute atoms occupy
the interstitial free space among the atoms of the solvent crystal lattice. Usually, ele-
ments with very small atomic size (less than 1 Å) are suitable for having an interstitial
solid solution, which can fit into the available interstitial sites in the solvent phase.
Hydrogen, carbon, boron, and nitrogen are among the solutes that can be accommo-
dated at such interstitial sites. However, hydrogen (atomic size ~0.46 Å) is the only
51
element that has a size usually lower than that of the interstitial site. Other elements,
being relatively larger than the interstitial sites, cause a volume expansion of the lat-
tice. Therefore, only a limited number of metals (such as Fe, Ni, and Ti) do have
significant solubility for these elements. Steel is a classic example of interstitial solid
solution/alloy, where the carbon atoms do occupy the interstitial sites of iron lattice.
The simplest form of difference between an interstitial alloy and interstitial
compound is the concentration of solute. The usual concentration of solute in an
interstitial alloy is much lesser than that of the interstitial compound. With a large
difference in the atomic radii of solute and solvent, the mobility of the solute atoms
is significant at lower solute concentration. With increase in the solute concentration,
the solution approaches toward saturation, and eventually, the mobility of the solute
atoms is lowered. Beyond saturation, the mobility of the solute atoms is drastically
reduced, and formation of interstitial compound having a composition stoichiometry
starts forming. The concentration of both the solute and the solvent is fixed in case of
an interstitial compound, and thus, it has a fixed chemical formula. On the contrary,
an interstitial solid solution has a range of composition.
Coming to the solubility of interstitial alloy, as the solute atoms occupy only the
free space available in the solvent lattice, the solubility of interstitial solute in such
alloy is limited. With phase transformation and temperature, this solubility limit
may further be altered depending on the volume of free space and size of the inter-
stitial voids. For example, the maximum solubility of carbon in α-iron is 0.025% (at
723°C), whereas in the case of γ-iron, the solubility limit is 2.06% (at 1147°C).
4.2.2 suBsTITuTIonal solId soluTIon

If the solute atoms substitute/replace the atoms of the solvent phase without destroy-
ing the crystal structure of the solvent in a solution, it is termed as substitutional solid
solution. Daily life examples of substitutional alloys include brass which is com-
prised of Cu and Zn, and gold–silver alloys used for making jewelry. Au (or Ag) may
replace FCC Ag (or Au), resulting a FCC alloy. A schematic representation of pure
metal, interstitial alloy and substitutional alloy is shown in Figure 4.1. In contrast to
the interstitial solid solutions, the solubility of substitutional solid solutions is mostly
higher. However, there are several factors that combinedly govern the solubility of an
FIGURE 4.1 Arrangement of atoms in a (a) pure metal, (b) interstitial alloy, and (c) substi-
tutional alloy.
Solid Solutions 53
alloy system. For complete solubility, several conditions must be met as formulated
by English metallurgist and materials scientist William Hume-Rothery and known
as the Hume-Rothery rules.
• Crystal structure: If both the constituents exhibit the same crystal structure,
then the chance of having complete solubility is higher.
• Relative size difference: For achieving a substantial solubility, the relative
atomic size difference between the solvent and solute must not exceed 15%.
rsolute − rsolvent
Mismatch = × 100% < 15%
rsolvent
• Chemical affinity: There should not exist a strong chemical affinity between
the solute and the solvent in order to have a higher solubility. It can also be
expressed in terms of electron negativity. There is a higher chance of com-
pound formation with solute–solvent combination having a higher differ-
ence in electron negativity. For maximum solubility, ideally the difference
between the electron negativity of solute and solvent should be close to 0.
• Valency: A metal can be dissolved in a metal of lower valence to a larger
extent than that of higher valence. For example, Al can be dissolved in Ni
to a higher extent than Ni in Al. For having a high solubility, typically both
the solute and the solvent should have similar valency.
For complete solubility, all these aforementioned criteria must be met. An excellent
example of complete solubility is copper–nickel alloy system. Both Cu and Ni crys-
tallize in FCC structure, and the difference in their atomic radii is less than even 1%.
Both Cu and Ni have very close electron negativity and valency. All these factors
together make it possible to have a Cu–Ni alloy system having complete solubility
in each other. However, it is fairly easy to check whether a particular combination of
metals exhibits favorable size difference. This only factor is sufficient to restrict the
solubility to a very low limit. Upon fulfillment of the favorable size conditions, other
criteria must also be considered to predict the solubility of the alloy system.
4.3 ELECTRON-TO-ATOM RATIO
This is somehow related to the valency factor as discussed in the Hume-Rothery
rules of solubility. The electron-to-atom ratio (e/a) is an important parameter which
influences the solubility of an alloy system. The solute that is having a tendency to
increase the overall e/a of the alloy is expected to have a higher solubility. As per
the earlier example, Cu (higher valency) has a higher solubility in Ni (lower valency)
than the reverse. Similarly, Mg (+2) has 70 atom% solubility in Li(+1), whereas Li
has only 24.5 atom% solubility in Mg. Interestingly, in the case of copper- and silver-
based alloys, the solubility limit of many elements typically corresponds to a value
where the e/a approaches 1.35–1.4, which corresponds to the critical concentration of
1.36 at which the Fermi sphere tends to reach the first Brillouin zone. For example,
the maximum solubility of Zn in Cu is 42 atom%, i.e., in a representative alloy of
100 atoms, there will be 42 atoms of Zn and thus 58 atoms of Cu. Corresponding
to the valency of Zn (+2) and Cu (+1), the total number of valency electron thus is
142 (= 42 × 2 + 58 × 1). With this, the e/a ratio becomes ~1.4.
4.4 ENTHALPY OF FORMATION OF A SOLID SOLUTION

Enthalpy is the measure of the heat/internal energy change during any reaction/
solution/transformation. In case of a solid solution, there may be release of heat or
absorption of heat or no change in heat due to the mixing process. For an ideal solu-
tion, the net change in enthalpy should be 0, i.e., ∆H mix = 0. However for nonideal
solutions, (i) ∆H mix < 0, which indicates an exothermic reaction or heat is evolved
during the reaction, or (ii) ∆H mix > 0, which indicates an endothermic reaction or
heat is absorbed during the reaction. Enthalpy of mixing depends on the likeliness
of an atom toward similar kind of atoms or dissimilar atoms. In an alloy system of A
and B, there may exist three different kinds of bonds, A–A, B–B, and A–B as shown
in Figure 4.2.
The number of A–A, B–B, and A–B bonds depends on the concentration and
distribution of the solute atoms in the solution. If there are MAA number of A–A
bonds, MBB number of B–B bonds, and MAB number of A–B bonds, then the internal
energy of the A–B solid solution (Ess) may be represented as,
Ess = M AA EAA + M BB EBB + M AB EAB (4.1)
where E represents the bond energy, and the subscripts AA, BB, and AB denote the
respective bonds. In the A–B alloy, let us now consider that there are NA number of
A atoms and NB number of B atoms, and each atom is having coordination number
of z. The total number of bonds involving A atoms is
N A z = 2 M AA + M AB (4.2)
N A z − M AB
M AA = (4.3)
2
FIGURE 4.2 Atomic bonding in (a) A lattice, (b) B lattice, and (c) A–B lattice (solid
solution).
Solid Solutions 55
Similarly,
N B z = 2 M BB + M AB (4.4)
And,
N A z − M AB
M BB = (4.5)
2
Now the expression for internal energy becomes
N A z − M AB N z − M AB
Ess = EAA + A EBB + M AB EAB
2 2
(4.6)
N z N z E + EBB 
Ess = A EAA + B EBB +  EAB − AA M
2 2  2  AB
The net energy of pure components A and B before formation of solid solution can
be presented as
NA z N z
E1 = EAA + B EBB (4.7)
2 2
Hence, the enthalpy change due to formation of solid solution is
∆H mix = Ess − E1 (4.8)
EAA + EBB 
∆H mix =  EAB − M (4.9)
 2  AB
The previous expression indicates that
EAA + EBB
a. For an ideal solution, ∆H mix = 0, or EAB = . The bond energy
2
A–B is the average of the bond energies A–A and B–B.
E + EBB
b. For an exothermic reaction, ∆H mix < 0, or EAB < AA . A atoms want
2
to be surrounded by B atoms and vice versa.
E + EBB
c. For an endothermic reaction, ∆H mix > 0, or EAB > AA . Both A and
2
B atoms want to be surrounded by similar atoms.
EAA + EBB
However, when EAB and values are close to each other, a random
2
arrangement of atoms may be expected in the A–B alloy leading to a regular solution.
In such cases, a good approximation of enthalpy of mixing is
∆H mix = X A X BΩ (4.10)
FIGURE 4.3 Enthalpy of mixing vs. mole fraction of B in an A–B solid solution.
NA
XA = , or mole fraction of A.
NT
N
X B = B , or mole fraction of B.
NT
N T is the total number of atoms.
EAA + EBB  zN T
Ω =  EAB − (4.11)
 2  2
The expression for ∆H mix may further be rewritten as
(
∆H mix = (1 − X B ) X BΩ = X B − X B2 Ω) (4.12)
For a positive Ω, the variation in the enthalpy of mixing is plotted in Figure 4.3 as a
function of mole fraction of B.
4.5 ENTROPY OF FORMATION OF A SOLID SOLUTION

Entropy is conventionally defined as the degree of randomness of a system. A math-
ematical expression for the change in entropy (S) can be presented by the Boltzmann
equation mentioned in the following:
∆S = k B ln ω (4.13)
Where k B is Boltzmann’s constant and ω is typically known as number of possible

distinguishable microstates or arrangements.
Generally, entropy has two contributions for a solid solution. The first one is
the thermal entropy that comes from the thermal energy of the system. Secondly,
the different ways that the atoms of solvent and solute arrange themselves to form the
solution are known as configurational entropy. Consider one mole of a solid solution,
i.e., N0 having NA number of “A” atoms and NB number of “B” atoms.
N0 = NA + NB (4.14)
Solid Solutions 57
In the crystal structure of the solid solution, there are now total N 0 lattice sites, which
are to be occupied by N A number of A atoms and N B number of B atoms. Let us now
fill the lattice by A and B atoms. The first A atom now can occupy any of the N 0
sites, the second can occupy the rest N 0 − 1 sites, and the third then can occupy the
remaining N 0 − 2 sites. In this way, the last A atom can occupy any of the available
N 0 − N A + 1 sites. As all the A atoms are similar, the total number of ways that the
N A number of A atoms can fill the lattice can be represented as
N 0 × ( N 0 − 1) × ( N 0 − 1) × ( N 0 − 1) ×  × ( N 0 − N A + 1)
ωA = (4.15)
NA !
In a similar way, the first atom of B can occupy any of the N 0 − N A available free
sites, the second atom of B can occupy any of the N 0 − N A − 1 available free site and
so on, and the last B atom will occupy the last available free lattice site. In this way,
the total number of possible ways that the N B number of B atoms can fill the lattice
can be represented as
ωB =
( N 0 − N A ) × ( N 0 − N A − 1) × ( N 0 − N A − 2 ) ×  × 3 × 2 × 1 (4.16)
NB !
Now combining both A and B atoms, the total number of ways that these N A number
of A atoms and N B number of B atoms can fill the crystal with N 0 lattice points is
ω = ωA × ωB
N 0 × ( N 0 − 1) × ( N 0 − 1) × ( N 0 − 1) ×  × ( N 0 − N A + 1)
ω=
NA !
( N 0 − N A ) × ( N 0 − N A − 1) × ( N 0 − N A − 2 ) ×  × 3 × 2 × 1
×
NB !
N0 ! (4.17)
ω=
NA ! × NB !
Following equations 4.13 and 4.17,
 N0 ! 
∆Smix = k B ln  
 NA ! × NB !
∆Smix = k B ( ln N 0 !− ln N A !− ln N B !)
Using Sterling’s approximation, i.e., ln ( x !) ≈ x ln x − x
∆Smix = k B ( N 0 ln N 0 − N 0 − N A ln N A + N A − N B ln N B + N B )
As N 0 = N A + N B
FIGURE 4.4 Variation in entropy of mixing as a function of composition of the solid

solution.
∆Smix = k B ( N 0 ln N 0 − N A ln N A − N B ln N B )
Again as N A = N 0 X A and N B = N 0 X B,
∆Smix = k B ( N 0 ln N 0 − N 0 X A ln ( N 0 X A ) − N 0 X B ln ( N 0 X B ))
∆Smix = k B N 0 ( ln N 0 − X A ln ( N 0 X A ) − X B ln ( N 0 X B ))
∆Smix = k B N 0 ( ln N 0 − X A ln N 0 − X A ln X A − X B ln N 0 − X B ln X B )
∆Smix = k B N 0 ( ln N 0 − X A ln N 0 − X B ln N 0 − X A ln X A − X B ln X B )
∆Smix = k B N 0 ( ln N 0 (1 − X A − X B ) − X A ln X A − X B ln X B )
As k B N 0 = R (universal gas constant) and X A + X B = 1,
∆Smix = − R ( X A ln X A + X B ln X B ) (4.18)
Equation 4.18 indicates that ∆Smix is positive, as the mole fractions X A and X B are
less than unity, and logarithm of these values must be negative. A schematic repre-
sentation of the entropy of mixing of a solid solution as a function of composition is
shown in Figure 4.4.
4.6 FREE ENERGY CHANGE UPON FORMATION

OF A SOLID SOLUTION
As per Gibb’s free energy, the free energy of formation of a solution can be repre-
sented by the following expression:
∆Gmix = ∆H mix − T∆Smix (4.19)
In case of an ideal solution (i.e., ∆H mix = 0), the free energy of formation of a solid
solution is solely governed by the entropic contribution, i.e.,
Solid Solutions 59
FIGURE 4.5 Effects of composition of the solution and temperature on the free energy
change for an ideal solution.
∆Gmix
ideal
= −T∆Smix (4.20)
∆Gmix
ideal
= RT ( X A ln X A + X B ln X B ) (4.21)
Hence, the free energy of an ideal solution seems to be more negative, or the process
of solution formation is more spontaneous when the temperature increases, which
can also be seen from Figure 4.5.
If we consider the actual free energy (not free energy change) of the solution
before (G1) and after (G 2) mixing (Figure 4.6),
G1 = GA X A + GB X B (4.22)
G2 = G1 + ∆Gmix
ideal
= GA X A + GB X B + RT ( X A ln X A + X B ln X B ) (4.23)
For real solutions, although the second component (i.e., T∆Smix or entropic contribu-
tion) of equation 4.19 is negative, ∆H mix may be either positive or negative. Thus,
∆Gmix
real
may be positive or negative. The expression for ∆Gmix
real
may be represented as
∆Gmix
real
= X A X BΩ + RT ( X A ln X A + X B ln X B ) (4.24)
FIGURE 4.6 Free energy of the solution before and after mixing, and free energy of mixing
as functions of temperature.
FIGURE 4.7 Enthalpy, entropy, and free energy change upon solid solution formation at
various Ω and T combinations.
If formation of solid solution is as exothermic process, i.e., ∆H mix < 0, ∆Gmix real
will be
always negative irrespective of temperature and composition; however, the magnitude
of the free energy change will depend on both temperature and composition of the
solid solution as presented in Figure 4.7a and b. In case it is an endothermic reaction,
it will be difficult to confirm whether it will yield a positive or negative free energy
change. For a fixed composition, ∆Gmix real
will be negative at relatively higher tempera-
ture, as the magnitude of the entropic contribution will be higher than that of enthal-
pic contribution (Figure 4.7d). At relatively lower temperature, the variation of ∆Gmix real
is further complex as shown in Figure 4.7c. From all these cases, it indicates that
∆Gmix will become negative upon addition of small amount of solute to the solvent.
4.7 ORDERED AND RANDOM SOLID SOLUTIONS

Atoms of solute and solvent in a substitutional solid solution may have random or
preferred locations. In a random solid solution, all the lattice sites are identical to
each other. Thus, the probability of any of the lattice site being occupied by A atom
is equal to the mole fraction of A, i.e., XA. The number and bond energies of A–A,
B–B, and A–B bonds are useful in order to find out whether there is any sort of
ordering in the solution. If the mean of the bond energies of A–A and B–B bonds
 i.e., EAA + EBB 
  is less than EAB, or in other words, if Ω > 0, it indicates that A
2
likes A more than B. Hence, A atom will be having a more number of A atoms as the
Solid Solutions 61
nearest neighbors than that of a random solid solution. This is known as clustering.
E + EBB
However, when AA is greater than EAB (i.e.,Ω < 0), A atoms have a prefer-
2
ence toward B atoms, and thus, there may be a short-range ordering or A atoms get
surrounded by more B atoms than that of the random solid solution. To know the
extent of ordering in a solid solution, an ordering parameter(s) is used, which is
evaluated from the following expression:
M AB − M AB
ran
s= (4.25)
max
M AB − M AB
ran
M AB represents the number of A–B bonds in the solid solution, and the superscript
ran represents the number of A–B bonds in a random solution and max represents
the highest possible number of A–B bonds. A perfect random solution exhibits s = 0,
while negative and positive values of s indicate clustering and short-range ordering
in the solid solution.
4.8 INTERMEDIATE PHASES
Sometimes, at certain composition of an alloy, the crystal structure of the alloy is
different from both the crystal structures of the parent elements due to the least free
energy. In such cases, the evolved phase is known as an intermediate phase. The con-
cept of lowest free energy arises from the composition of the alloy, which tends to
exhibit a lower free energy than the terminal solid solutions. Various possible bonds
and packing arrangement of atoms in the solid solution are very often responsible for
formation of these intermediate phases.
In general, for an alloy system having a possibility of intermediate phase,
generation forms a typical “U”-shaped free energy vs. composition curve near the
composition of the intermediate phase. The depth and/or width of this “U”-shaped
curve give information related to the composition range and stability of the interme-
diate phase.
In case of an intermediate phase, having a very narrow composition range with
a sharp reduction in the free energy is termed as intermediate compound having
almost fixed composition, i.e., A xBy with a stoichiometric combination of A and B as
shown in Figure 4.8a. Figure 4.8b represents the existence of an intermediate phase
with a broad composition variation (i.e., no stoichiometry).
Based on the type of interaction between A and B, the intermediate phase may
be covalent or ionic as well. In both these cases, the intermediate phase is having a
fixed composition (intermediate compound). AlSb and ZnSb are among the covalent
compounds having zinc-blend structure, whereas Mn2Sn, CoSb, and Fe3Sn2 are few
among the covalent compounds having nickel arsenide crystal structure. Sometimes,
the intermediate compound looks like a vertical straight line due to its extreme zero
tolerance for additional solute and solvent concentration. There are also various ionic
intermediate phases available such as MgSe, Mg2Sn, and so on. There also exists an
interstitial compound, where small atoms like hydrogen, carbon, or nitrogen occupy
the interstitial voids, and formation of covalent bonds takes place. The Laves phases
are also another type of intermediate phase having composition AB2. Generally, the
FIGURE 4.8 Free energy vs. composition diagram for (a) intermediate compound and (b)
intermediate phase.
relative size difference between the atoms of solute and solvent in the case of the
Laves phase is about 1.1–1.6. Atomic packing becomes most efficient if the atomic
arrangement follows any of the Laves phase structure of MgCu2, MgNi2, or MgZn2.
FURTHER READING

5 Phase Diagrams and
Phase Transformations
As discussed earlier, the existence of a particular phase of an alloy depends on the
thermodynamics/free energy of the mixing process, which is altered by the tempera-
ture of the surrounding, atmospheric pressure, and composition of the alloy. To pre-
dict the existence of a possible phase or mixture of phases at certain combination of
aforementioned factors, phase diagrams are useful. Phase diagrams are also referred
as constitutional diagrams or equilibrium diagrams.
5.1 THERMODYNAMIC CONSIDERATIONS OF PHASE DIAGRAMS

As stated earlier, a phase diagram is basically a graph that indicates the type of phase
or mixture of phases under a given circumstance. Let us consider some basic features
of a phase diagram.
5.1.1 coordInaTes of a BInary pHase dIagram

The simplest phase diagram is a binary phase diagram having only two constituents.
Hence, one possible variable here is the composition. As the metallurgical phase
diagrams are constructed at normal atmospheric pressure, the environmental vari-
able is only temperature. Hence, a binary phase diagram is constructed taking the
composition of the alloy and its temperature into consideration. Conventionally, the
composition (normally mentioned in terms of atom fraction or weight fraction) and
temperature are presented in the abscissa (x coordinate) and ordinate (y coordinate),
respectively. The weight fraction (or wt%) or atom fraction (or at%) are interchange-
able based on the following expressions for an alloy of A and B.
wt.% of A
at.%of A = × 100
a 
wt.%of A + wt.%of B ×  A  (5.1)
 aB 
at.% of B = 100 − at.% of A
where aA and aB are the atomic weights of A and B, respectively.
5.1.2 consTrucTIon of pHase dIagram

A phase diagram is usually constructed concentrating on the solid state and early liq-
uid state (just after melting). The traditional method of constructing a phase diagram
is to cool a given composition of an alloy very slowly (to approach the equilibrium)
63
from its melt state to the desired range of temperature, which is usually the room
temperature. During this whole range of cooling, the microstructure (e.g., liquid and
solid with phase α and β, mixture of α and β, and so on) of the alloy is continuously
monitored as a function of temperature. This process is repeated for various alloy
compositions, after which the complete phase diagram is developed. In subsequent
section, a detailed discussion on construction of some simple phase diagrams will
be done.
5.2 GIBB’S PHASE RULE

Gibb’s phase rule gives the number of possible ways to vary the controllable param-
eters independently in order to obtain a particular phase or phase mixture at equi-
librium. Such number of ways is termed as degree of freedom (F). For being in
equilibrium, the free energy of the system is minimum, and thus, there is no driving
force for any transformation. In order to calculate the degrees of freedom, we first
should consider the number of possible combinations among the controllable vari-
ables (compositional and environmental variables) and equalities. Consider a system
containing “m” number of phases formed by “n” number of components. Let the
phases be P1, P2, P3, …, Pm, and similarly the components are C1, C2, C3, …, Cn. Now
let us first determine the number of compositional variables. Any phase, e.g., P1, is
comprised of C1, C2, C3, …, Cn. The possible ways of varying the composition of P1
is by varying the respective percentage of C1, C2, C3, …, Cn. This can be done by
(n − 1) ways, as the percentage of Cn is automatically fixed (i.e., 100 − (%C1 + %C2 +
%C3 + ⋯ + %Cn−1)) after fixing the percentages of C1, C2, C3, …, Cn−1. For each of
the phases P1, P2, P3, …, Pm, thus the number of compositional variables is (n−1). So,
the total number of compositional variable for all phases is m(n−1).
No.of compostional variables = m(n − 1) (5.2)
Next coming to the environmental variables, the effects of all possible environmen-
tal variables on the stability of the phase or phase mixture should be considered.
Very often, pressure and temperature are considered as the environmental variables.
But it is not strictly restricted only for these two variables, and hence, other environ-
mental factors may also be considered having significant influence on the phases.
No.of environmental variables = E (5.3)
Putting all the variables together,
Total no.of variables = No.of compostional variables

+ No.of environmental variables
Hence,
Total no.of variables = m(n − 1) + E (5.4)

Phase Diagrams and Phase Transformations 65
The next thing to consider here is the number of equalities. For the phases P1, P2,
P3, …, Pm to be in equilibrium, the chemical potential or vapor pressure of each of
the elements/components should be the same in each of the phases. Hence,
µCP11 = µCP12 = µCP31 = … = µCP1m
µCP12 = µCP22 = µCP32 = … = µCPm2

(5.5)
   
µCP1n = µCP2n = µCP3n = … = µCPmn
where µCP11 is the chemical potential of the component C1 in the phase P1. It can be
observed from the previous expressions that, for each row of expression (i.e., involv-
ing a particular component), there are (m − 1) equal signs, and in expression 5.5,
Total no.of equalities = n(m − 1) (5.6)
The degree of freedom (DOF or F) now can be determined by subtracting the total
number of equalities from the total number of variables.
F = m(n − 1) + E − n(m − 1)
Simplifying,
F = n−m+E (5.7)
Conventionally, the number of phases, i.e., m is designated as P, and similarly the

number of components, i.e., n is designated as C. If pressure and temperature are the
only environmental variables, then the number of environmental variables, i.e., E
becomes 2. Hence, expression 5.7 may be rewritten as
F =C−P+2 (5.8)
This expression is known as Gibb’s phase rule. For most of the metallurgical prac-
tices, the number of environmental variables is 1, which corresponds to tempera-
ture, as processes are carried out at normal atmospheric pressure (which is constant).
Thus, for metallurgical purpose, equation 5.8 is modified as
F = C − P +1 (5.9)
5.3 LEVER RULE

For a system comprising two phases, the Lever rule is very often used to determine
the weight fraction of each phase at a given combination of temperature and alloy
composition. Let us take 1 g of an alloy with composition wB, which consists of two
phases α and β at a given temperature T. Thus,
Total weight of B in the alloy = wB (5.10)

The weight fractions of α and β are wα and wβ, respectively. Hence,
wα + wβ = 1 (5.11)
α
Let the composition of α be wB. Thus,
Weight of B in α phase = wα × wαB (5.12)
Similarly, let the composition of β be wBβ . And,
Weight of B in β phase = wβ × wBβ = (1 − wα ) × wBβ (5.13)
The total amount of B in the alloy must be equal to the sum of the B in α and β
phases. Accordingly,
wB = wα × wαB + (1 − wα ) × wBβ
Simplifying and rearranging,
wBβ − wβ
wα =
wBβ − wαB
and (5.14)
wβ − wαB
wβ = 1 − wα =
wBβ − wαB
Applications of the Lever rule in various phase diagrams are discussed later.
5.4 TYPES OF PHASE DIAGRAMS AND

PHASE TRANSFORMATIONS
There exist several types of phase diagrams depending on the solubility of the con-
stituents of the alloy system in liquid and solid states as mentioned in Table 5.1.
5.4.1 IsomorpHous pHase dIagram: BoTH componenTs are

compleTely soluBle In BoTH lIquId and solId sTaTes
For satisfying such criteria, the Hume-Rothery rules must be obeyed. Both the met-
als should have the same crystal structure and have less than 8% difference in their
atomic radius. Understanding the construction of the phase diagram for such system
is simplest.
5.4.1.1 Construction of Type I Phase Diagram

Cooling curves are the origin of obtaining such phase diagram. It is well known that
a pure metal has sharp melting/freezing point, i.e., the entire pure metal specimen
TABLE 5.1
Characteristics of Different Types of Phase Diagrams
Type Characteristics
Type I (isomorphous) Both the constituents are completely soluble in each other in liquid
as well as solid states.
Type II (eutectic) Both metals are completely soluble in liquid state and having zero
solubility in each other in solid state.
Type III (eutectic) Both metals are completely soluble in liquid state and having
limited solubility in each other in solid state.
Type IV (congruent melting There exists an intermediate compound of fixed chemical (or very
intermediate compound) narrow range) composition that virtually divides the phase
diagram into two parts. This intermediate compound is a
congruent melting one.
Type V (peritectic) In this type of phase diagram, there exists a transformation where
a mixture of solid and liquid upon cooling give rise to formation
of a new solid.
Type VI (monotectic) Two liquids are partially soluble in each other.
Type VII Two metals are insoluble in both liquid and solid states.
freezes/solidifies or melts at a particular temperature. Hence upon slow cooling, the

cooling curve (temperature vs. time) of such liquid metal specimen yields a horizon-
tal step exactly at the freezing temperature (Tm) of the metal as shown in Figure 5.1a.
On the contrary, an alloy has a range of melting temperature. As shown in Figure
5.1b, the cooling curve of an alloy of given composition does not yield a horizontal
step; rather, freezing starts at Tms temperature and is finished at Tmf. These Tms and Tmf
are composition dependent, and by capturing these Tms and Tmf for different composi-
tions of the alloy, the entire phase diagram can be prepared.
Let us discuss this process of construction in a more detailed way. Consider an
alloy system made up of A and B metals having melting points TmA and TmB, respec-
tively, and TmA < TmB (Figure 5.2).
For pure metals A and B, as there is a sharp melting point, transition from liquid
to solid (or vice versa) phase takes place at that particular temperature. The degree
FIGURE 5.1 Cooling curves of (a) pure metal and (b) an alloy.
FIGURE 5.2 Schematic representation of cooling curves of (a) pure A, and A + B alloys
with (b) 20% B, (c) 40% B, (d) 60% B, (e) 80% B, and (f) pure B.
of freedom at melting (where both liquid and solid phases coexist) of a pure metal
is thus 0 (one component and two phases). Hence, melting of a pure metal is an
invariant transformation. However, for any alloy, the DOF for melting is 1, sug-
gesting a little change in the melting point of the metal by a small change in its
composition.
5.4.1.2 Compositions of the Phases in a Phase Mixture

A type I phase diagram is generally having three distinct zones, i.e., (i) solid (single
phase), (ii) liquid (single phase), and (iii) solid and liquid together (two-phase mix-
ture). Unlike the single-phase alloy, a two-phase mixture is composed of two differ-
ent phases (one solid and one liquid) of different compositions (Figure 5.3).
Assume an alloy of composition CA is solidified from its melt (Figure 5.4).
Upon cooling in the liquid state, it remains a single-phase system with the same
composition. However, cooling below its liquidus temperature makes the system
comprising of a mixture of both solid and liquid phases. However, the composi-
tions of this solid and liquid in the phase mixture are different and change with
temperature. The composition of these two phases at any temperature can be
determined from the phase diagram by constructing a tie line (horizontal line) at
that particular temperature. The intersection of the tie line with the liquidus and
solidus lines gives the composition of the liquid and solid phases, respectively. In
this example, when the temperature of the alloy is T1, the solid and liquid phases
are of composition CS1 and C 1L, respectively (refer Figure 5.4). Further lowering in
temperature to T2 results in having a solid of CS2 and liquid of C L2 compositions as
FIGURE 5.3 Construction of type I phase diagram from cooling curves.
FIGURE 5.4 Concept of tie line for determination of the composition of phases.
5.4.1.3 Percentage of the Phases in a Phase Mixture

The next important thing in a phase mixture of given alloy composition is to deter-
mine the percentage of different phases that coexist with each other (Figure 5.5).
Reconsider the same alloy system, and take 100 gm of an alloy of composition CA.
At a temperature T (which corresponds to a two-phase region), let us try to determine
the percentage of solid and liquid in the phase mixture. As per the concept of tie line,
the compositions of the solid and liquid phases in the mixture are CS and C L, respec-
tively. If we assume there is x gm of liquid, then the amount of solid will be (100 − x)
gm. As the quantity of B (composition) of the alloy remains unchanged during the
entire cooling process (mass conservation), the amount of B in the liquid state (single
phase) must be equal to the sum of the amounts of B in the solid and liquid phases
present in the phase mixture. Mathematically it may be represented as
100 × CA = x × C L + (100 − x ) × CS
FIGURE 5.5 Determination of percentage of phases in a phase mixture by tie line.
This gives
CA − CS C − CA
x= or S
C L − CS CS − C L
If we look at the phase diagram, CS − CA essentially represents the length of the line
BC, and CS − C L is the length of the line AC. Hence,
Length of BC
The amount of liquid (wt.%) =
Length of AC
Similarly, it can also be obtained that
Length of AB
The amount of solid (wt.%) =
Length of AC
This is known as the Lever rule.
5.4.1.4 Properties of Type I Alloys

Very often, properties of a metal change upon alloying, and they are mostly sensi-
tive to the composition of the alloy. The changes in these properties are mostly
associated with the distortion caused to the parent crystal structure of the metal
(Figure 5.6).
It can be observed from the figure that there exists a maximum in the strength vs.
composition diagram, indicating the alloy to be stronger than both the pure metals.
Apparently, it turns out that the alloy having 60%–70% Ni exhibits the maximum
strength with an appreciable ductility.
FIGURE 5.6 Changes in (a) lattice parameter and (b) strength and % elongation due to
alloying.1
5.4.2 euTecTIc pHase dIagram: BoTH THe consTITuTIve meTals

are 100% soluBle In eacH oTHer In THe lIquId sTaTe BuT
compleTely InsoluBle or lImITed soluBIlITy In THe solId sTaTe
This is the situation when the solubility of the solute metal is negligible or limited
in the solvent metal. This arises when one or more of the Hume-Rothery rules are
unsatisfied. In most of the cases, the solubility is limited, and only in some rare cases,
the solubility is absolute zero. In this case also, the phase diagram can be constructed
by a series of cooling curves of pure terminal metals and alloys of varying composi-
tions. As per Raoult’s law, if the solute is soluble in the solvent in liquid state, but
insoluble in solid state, then addition of the solute decreases the freezing point of
the solvent. Hence, considering an A–B alloy system, addition of B decreases the
freezing point of A, and similarly, addition of A decreases the freezing point of B.
Now, it is very obvious from this statement that there exists an alloy composition
corresponding to a minimum freezing point in the entire range of composition, as
illustrated in Figure 5.7.
It can be noted from Figure 5.7 that upon addition of B to A, solidification starts at a
lower temperature than that of pure A. However, for a broad range of the composition
of the alloy, completion of the solidification is achieved at a constant temperature cor-
responding to TE, known as the eutectic temperature of the alloy. Interestingly, only
for a single composition (i.e., CE) in the entire alloy system, solidification is completed
only at a single temperature instead of a range. This is known as eutectic composition,
where melting/solidification takes place as that of a pure metal. However, this melt-
ing cannot be termed as congruent melting as the compositions of the parent liquid
and the solids formed are different. Now the phase diagram consists of four distinct
zones: (i) a liquid phase (denoted by L) where both metals are completely dissolvable
in each other; two phase mixtures of (ii) liquid and solid A (denoted by L + A); (iii)
liquid and solid B (denoted by L + B); and (iv) solid A and solid B (denoted by A + B).
Conventionally, the alloy having composition less than CE is termed as hypoeutectic
and greater than CE is termed as hypereutectic. Let us now consider a liquid solution
of eutectic composition is being solidified. The phase diagram indicates complete
FIGURE 5.7 An eutectic phase diagram showing no solubility of the constituents in each
other in the solid state.
solidification of the entire alloy at TE temperature, which essentially means that dur-
ing the entire process of solidification, the temperature of the alloy system does not
change. When the temperature drops just below TE, the liquid phase will transform
into two solids represented by the end points of the horizontal strain line passing
through TE, i.e., solid A and solid B. Initially, when a tiny crystal of solid A is formed,
the adjacent liquid bath becomes rich in B. To bring down the bath concentration
again to CE, a layer of solid B is formed leaving the adjacent bath with local concen-
tration less than CE, making conditions favorable for solidification of another layer
of A. This process continues till completion of the solidification process, making the
solid comprising alternate layers of pure A and pure B. This transformation is known
as eutectic transformation as stated in the following:
Cooling
−−−−−
Liquid L ↽−−−− ⇀
− Solid A + Solid B
Heating
5.4.2.1 Microstructural Evolution During Cooling of a Eutectic Alloy

Let us consider a hypoeutectic alloy of composition C0 being solidified from the liq-
uid melt as shown in Figure 5.8. In the liquid stage, e.g., at temperature T1, the alloy
comprises only liquid phase having composition C0. Upon cooling, no microstruc-
tural change takes place till the liquidus temperature. However, upon crossing the
liquidus temperature, i.e., in the two-phase region, the alloy comprises both liquid (L)
and solid A. In this two-phase zone, as temperature is lowered, the weight fraction
of A continues to increase, and consequently, the liquid weight fraction continues
to drop as per the Lever rule. As liquid amount decreases, the concentration of B in
the liquid increases. For example, at temperature T2, the system contains A in solid
form and liquid L of composition C1 (refer Figure 5.8). Just before the eutectic tem-
perature (i.e. T3 which is just above TE), the system contains again pure A in solid
form and liquid of eutectic composition (CE), which can also be visualized from the
phase diagram. At TE, eutectic transformation takes place, and the liquid transforms
to the eutectic mixture of solid A and solid B. Hence, at a temperature just below
FIGURE 5.8 Microstructural evolution during solidification of a hypoeutectic alloy.
TE, (T4 as shown in Figure 5.8), the microstructure of the alloy contains a phase mix-
ture of A (known as proeutectic phase) and eutectic mixture of A and B as shown in
Figure 5.8. Solidification of a hypereutectic alloy can also be understood from the
same principle. In such case, cooling below the liquidus temperature, crystals of solid
B are formed in the liquid bath, and after the eutectic transformation, the microstruc-
ture of the alloy contains a phase mixture of proeutectic B and eutectic mixture.
There exist several eutectic alloy systems having almost negligible solubility of
the constituents in each other in their solid state, such as Bi-Cd, Al-Si, and so on.
Phase diagram of Bi–Cd alloy system has been presented in Figure 5.9. When the
alloy of eutectic composition (i.e., 40% Cd) was cooled from liquid state, the micro-
structure contains 100% eutectic mixture at solid state (i.e., below 144°C) as shown
in Figure 5.9 (point C). It is very obvious that this eutectic microstructure has 40%
Cd and 60% Bi in it. In case of a hypoeutectic alloy, upon cooling below the liquidus
temperature, proeutectic Bi is started solidifying in a liquid pool. The microstructure
now consists of liquid and solid Bi (point A in Figure 5.9). Upon further cooling in
this two-phase region, the amount of solid Bi increases, and gradually, the amount of
liquid decreases (as from the Lever rule) and the liquid is enriched in Cd. Eventually,
just above the eutectic temperature (i.e., 144°C), the liquid achieves the eutectic com-
position. The liquid upon crossing 144°C transforms to the solid eutectic mixture of
Bi and Cd without causing any change to the proeutectic Bi. Hence, below 144°C
temperature, the microstructure comprises proeutectic Bi and eutectic mixture of
Bi and Cd as shown in Figure 5.9 (point B). Let us consider an alloy with 10% Cd.
The Lever rule can be used to determine the weight fraction of proeutectic Bi in the
solid state.
FIGURE 5.9 Phase diagram of Bi–Cd alloy system with microstructures of hypoeutectic,
eutectic, and hypereutectic compositions during solidification.
40 − 10
Weight fraction of proeutectic Bi = = 0.75
40
Weight fraction of eutectic mixture = 1 − 0.75 = 0.25
Hence, the composition of the alloy is 90%Bi–10%Cd, and the microstructure con-
tains 75% proeutectic Bi and 25% eutectic mixture.
Similarly, in case of a hypereutectic alloy, proeutectic Cd and liquid are the micro-
structural constituents in the two-phase region as shown in Figure 5.9 (point D).
Upon cooling below the eutectic temperature, now the microstructure consists of
proeutectic Cd and eutectic mixture of Bi and Cd (point F in Figure 5.9). Consider a
hypereutectic alloy containing 70% Cd. From the phase diagram, it can be observed
that at solid state, the microstructure contains proeutectic Cd and the eutectic mix-
ture. In the microstructure of this alloy in solid state,
70 − 40
Weight fraction of pro-eutectic Cd = = 0.50
100 − 40
Weight fraction of eutectic mixture = 1 − 0.50 = 0.50
Another variation of eutectic alloy exists, where the constituents are having limited
solubility in each other, and in reality, this is very common. A typical phase diagram
of such type is shown in Figure 5.10.
As pointed out, B is having limited solubility in A. However, a solid solution is
formed when the amount of B in the system remains below a critical value, and this
is shown in Figure 5.10 where α is the resulted solid solution also known as terminal
solid solution. Similarly, β is also another terminal solid solution of A in B. An inter-
esting point here to note is that at eutectic temperature, B has maximum solubility
FIGURE 5.10 Eutectic phase diagram where constituents are partially soluble in each other.
in A and vice versa. Solubility here, as can be seen, is a function of temperature. As

the temperature decreases below TE , the solubility limit decreases, and the boundary
between the single-phase and two-phase region below the TE temperature is known
as solvus line. In this phase diagram, also there exists a eutectic transformation. But
the basic difference from the earlier one is that upon cooling the liquid of eutectic
composition, it yields a eutectic mixture comprising both the terminal solid solutions
(earlier one contains both pure metals).
Cooling
_____
Liquid L ↽____ ⇀
_ Solid α + Solid β
Heating
It should also be noticed from the phase diagram that the composition of the eutectic
mixture changes slightly, as there is a change in solubility, when cooled from TE to
room temperature.
5.4.2.2 Properties of Eutectic Alloys

It is a well-known fact that the properties of an alloy are controlled by the composi-
tion to a large extent. It certainly holds true in the case of eutectic alloys as well.
Perhaps in all the eutectic alloy system, the alloy having eutectic composition exhib-
its the highest strength among all others. The parent metals are usually soft and plas-
tic, whereas the eutectic phase is hard and brittle. Furthermore, as the concentration
of the solute increased from pure solvent to eutectic composition, there is a linear
increment in the amount of eutectic phase in the alloy. This indicates increment in
strength of the alloy, as the eutectic composition is approached.
5.4.3 perITecTIc pHase dIagram

If the combination of a solid and a liquid upon cooling yields a new solid, the trans-
formation is known as peritectic reaction. Usually, the new solid formed upon this
transformation is either a terminal solid solution or an intermediate phase.
Cooling
−−−−−
Liquid L + Solid α ↽−−−− ⇀
− Solid β
Heating
5.5 SOME OTHER SOLID-PHASE TRANSFORMATIONS

IN METALS AND ALLOYS
Apart from the transformations discussed earlier, there also exists several other
phase transformations that take place in solid state only. Some of these are dis-
cussed here.
5.5.1 alloTropIc TransformaTIon

The basis of allotropic transformation in context to crystalline solids has been
explained in Chapter 1, while discussing the crystal structure of metals. Some of the
metals such as iron, cobalt, and manganese undergo this allotropic transformation.
Allotropic transformation is usually represented in a binary phase diagram in form
of points on the vertical line indicating pure metal. The respective solid solution of
these allotropes further may be represented in phase diagram in form of closed loop.
Iron–carbon phase diagram is a nice example of such transformation, and the details
of this will be explained in the next chapter.
5.5.2 order–dIsorder TransformaTIon

The concept of ordered and random solid solutions has been explained in Chapter 4.
In general, the solute atoms in a substitutional solid solution occupy random loca-
tions yielding disordered solid solutions. However, in certain cases, precise control
over the cooling rate may result a particular arrangement of solute atoms in the alloy,
known as ordered solid solution. In most of the cases, ordering takes place with a
particular composition corresponding to certain atomic ratio. As a convention, the
ordered phases in a phase diagram are usually denoted by α′, β′, γ′ or α′, α″, and so
on. And the phase boundary of these order phases is drawn by dot–dash lines.
5.5.3 euTecToId reacTIon

This is a most commonly occurring phase transformation in solid state, where a solid
upon cooling produces two different solids as represented in the following. A very
familiarized eutectoid reaction exists in iron–carbon diagram, where austenite con-
verts to a mixture of ferrite and cementite (known as pearlite) and will be discussed
in detail in the next chapter.
Cooling
−−−−−
Solid α ↽−−−− ⇀
− Solid β + Solid γ
Heating
5.5.4 perITecToId reacTIon

When a mixture of two different solids upon cooling yields a new solid, the transfor-
mation is known as peritectoid transformation as presented in the following:
Cooling
−−−−−
Solid α + Solid β ↽−−−− ⇀
− Solid γ
Heating
Usually, the new solid formed is a terminal solid solution or an intermediate phase.
Ni–Zn and Cu–Si alloys are some of examples showing peritectoid reaction.
5.6 ROLES OF DEFECTS AND DIFFUSION

Diffusion has a very important role to play in the case of solid-state phase trans-
formation. The movement of atoms in a solid body is usually random in nature.
However, in case where a large number of such diffusing atoms are involved, the
flow may trace a systematic path. As discussed in eutectic reaction, alternate layers
of solids form the eutectic mixture. And this requires diffusion of the solute atoms
to ensure appropriate concentration in each layer. The case is also similar in the case
of eutectoid reaction.
Diffusion in solid state usually takes place by three mechanisms: (i) vacancy
diffusion, (ii) interstitial diffusion, and (iii) atom exchange mechanism. As stated
earlier, vacancy is a thermodynamic feature in a crystalline solid, and the presence
of vacancy sites is usually preferred for ease diffusion. As shown in Figure 5.11,
an atom sitting next to the vacancy site may jump the vacancy site, thus creating
vacancy in its earlier location.
In a similar way, an interstitial atom may occupy a normal lattice site repelling
an atom from its lattice site to a nearby interstitial location as shown in Figure 5.12.
FIGURE 5.11 Schematic representation of vacancy diffusion in crystalline solid.
FIGURE 5.12 Schematic representation of interstitial diffusion in crystalline solids.

The other, i.e., atom mutual interchange making a close loop is also a possible
diffusion mode. But various experimental evidences have shown vacancy diffusion
to be the primary mode of diffusion in solids. The rate at which diffusion takes place
is affected by several parameters, such as temperature, concentration, surface area,
and so on.
FURTHER READING
1997).
Clark, D. S. & Varney, W. R. Physical Metallurgy for Engineers. (Litton Educational
Publishing Inc., New York, 1962).
6 Iron–Carbon
Phase Diagram
6.1 INTRODUCTION
Iron and carbon alloys, particularly steels, are widely used engineering alloys
because these are relatively cheaper with an extensive range of properties. A study of
the iron–carbon system is therefore valuable, as it helps us to understand the phase
transformations that occur in steels. Before we consider steels, it is equally impor-
tant to know the behavior of pure iron. The purest form of iron, which is commonly
used in industries, is called as ingot iron. It has a typical tensile strength of about
300 N/mm2 and is extensively used for drainage culverts, roofing, ducts, washing
machines, and stoves. Later, wrought iron, which is a mixture of high-purity iron and
slag, was manufactured by the Byers process. In this process, the iron metal is melted
and purified to a highly refined state; an iron silicate slag is prepared; finally, the iron
metal is disintegrated and mechanically incorporated in the slag to form sponge-like
balls of iron globules coated with the slag. These balls are hammered to squeeze out
the excess slag and then rolled to bars, plates, rods, and tubes. A comparison of the
composition and typical properties of ingot and wrought iron is given in Table 6.1.
The characteristic properties of wrought iron are because of the nature of the slag
distribution, i.e., the presence of threadlike slag fibers in the soft ferrite matrix, such
as good machinability, excellent shock and corrosion resistance, and good weld-
ability. Wrought iron has a wide range of industrial applications, including railroads,
ship buildings, oil industries, crane hooks, anchors, and architectural purposes.
TABLE 6.1
Composition and Mechanical Properties of Different Forms of Iron
% Composition Ingot Iron Wrought Iron
Carbon 0.012 0.06
Manganese 0.017 0.045
Silicon Trace 0.101
Phosphorous 0.005 0.068
Sulfur 0.025 0.009
Slag 1.97
Mechanical Properties Longitudinal
Tensile strength (N/mm2) 300 350
Yield strength (N/mm2) 200 215
% elongation 75 18–25
% reduction in area 30–40 35–45
79
6.2 ALLOTROPIC TRANSFORMATIONS IN IRON

Iron exists in more than one type of lattice structure, depending upon the tempera-
ture. To understand the same, let us look at the typical cooling curve for pure iron,
which is shown in Figure 6.1. Iron first solidifies at 1539°C. The iron at this stage
is in the body-centered cubic (BCC) structure and is called delta iron (δ-Fe). Upon
more cooling, δ-Fe changes to gamma iron (γ-Fe) at 1394°C. γ-Fe is face-centered
cubic (FCC) and nonmagnetic. Furthermore, at 910°C, γ-Fe transforms to alpha iron
(α-Fe), which is nonmagnetic, but the structure is changed to BCC. Finally, the α-Fe
becomes magnetic without a change in crystal structure at 768°C. This temperature
is commonly known as the Curie temperature (named after Madame Curie).
6.3 SOLUBILITY OF CARBON IN IRON

Carbon, when dissolved in iron, forms an interstitial solid solution. The solubility
of carbon depends upon the type and nature of the interstitial voids and the crystal
structure of iron (BCC or FCC). In the case of FCC γ-Fe, carbon enters the octahe-
dral voids because the radius of the largest interstitial sphere that would just fit in
octahedral voids is 0.052 nm as against 0.028 nm in tetrahedral voids. Figure 6.2a
illustrates the octahedral holes in FCC lattice, which are also symmetrical. However,
the radius of the carbon atom being 0.077 nm, some distortion of the lattice occurs
FIGURE 6.1 A typical cooling curve of pure iron.

Iron–Carbon Phase Diagram 81
FIGURE 6.2 (a) Symmetrical octahedral void in FCC structure and (b) unsymmetrical
octahedral void in BCC structure. The black spheres indicate the octahedral voids. It can
also be seen that the size of the octahedral voids is larger in FCC than in BCC. BCC, body-
centered cubic; FCC, face-centered cubic.
when the carbon atom enters the iron lattice. This lattice distortion limits the solubil-
ity of carbon in γ-Fe up to 2% at 1147°C and 0.8% at 727°C.
In the case of BCC α-Fe, the radius of the largest interstitial sphere that would
just fit in tetrahedral voids (0.036 nm) is more than the octahedral voids (0.019 nm).
Nevertheless, it is essential to note that the carbon atoms prefer to enter the smaller
octahedral voids. In octahedral voids of BCC lattice, the carbon atom has only two
nearest iron atoms, while four other iron atoms are at a larger distance, as shown in
Figure 6.2b. Hence, the carbon atom displaces these two nearest atoms by 0.053 nm
in one of the <100> directions to maintain symmetry (Figure 6.3a). It results in
enlargement of the c-axis as compared with the a-axis, thereby leading to tetragonal
distortion of the lattice. In contrast, when the carbon atom tends to occupy the larger
tetrahedral voids, it has all the four atoms as its nearest neighbors, and displacing
these four atoms causes more strain energy, as illustrated in Figure 6.3b. Therefore,
the carbon atoms prefer to enter the smaller octahedral voids, as they cause less dis-
tortion than bigger tetrahedral voids.
It can also be seen in Figure 6.4 that the solubility of carbon in austenite is more
than in ferrite, although the vacant space in the BCC lattice is greater than that in the
FCC lattice. As discussed earlier, the presence of carbon in BCC α-Fe lattice causes
tetragonal distortion, which restricts the iron to accommodate only a limited amount
of carbon atoms. Hence, the maximum solubility of carbon in α-Fe is only 0.02% at
727°C. On comparison, the size of the octahedral voids in FCC γ-Fe lattice where
carbon sits is larger, resulting in fewer distortions.
FIGURE 6.3 (a) Tetragonal distortion caused by the carbon atom (black sphere) in octa-
hedral voids in BCC α-Fe and (b) carbon atom placed in the tetrahedral void in BCC α-Fe
displaces more iron atoms resulting in more strains. BCC, body-centered cubic
6.4 IRON–IRON CARBIDE PHASE DIAGRAM

The iron–iron carbide phase diagram is only a part of the iron–carbon phase dia-
gram, where the influence of carbon up to 6.67% by weight on the allotropic trans-
formation of iron is studied. Iron forms an interstitial compound, iron carbide or
cementite, Fe3C, which contains 6.67% carbon. Therefore, this diagram is called
the iron–iron carbide phase diagram. Many books prefer to call this diagram as an
equilibrium diagram, but it is not an exact equilibrium diagram since equilibrium
means no change of phase with time. Moreover, in this case, cementite is a meta-
stable phase, and it decomposes to the stable form of carbon, i.e., graphite. This
decomposition takes a very long time, and therefore, the metastable conditions are
considered to represent the equilibrium changes.
6.4.1 InvarIanT poInTs In fe–fe3c pHase dIagram

The iron–iron carbide phase diagram schematically illustrated in Figure 6.4 shows
three horizontal lines where three important invariant reactions are described. The
first invariant reaction, on cooling, occurs at 1495°C, which is represented as
L +δ  γ (6.1)
While gamma, γ, is an interstitial solid solution of carbon in γ-Fe, δ is an interstitial

solid solution of carbon in δ-Fe. The maximum solubility of carbon in δ-Fe is only
0.10% (point A in Figure 6.5), whereas it is much more in γ-Fe. On increasing the
FIGURE 6.4 The iron–iron carbide phase diagram labeled in general terms.
amount of carbon up to 0.10%, the temperature of the allotropic change increases

from 1400°C to 1495°C. The line AB represents the start of the δ to γ allotropic
transformation for alloys containing less than 0.10% carbon. On the other hand, the
horizontal line AC marks the beginning of the above transformation by the peritectic
reaction (equation 6.1) for alloys containing carbon between 0.10% and 0.18%. The
line BC represents the end of this δ to γ transformation for alloys containing less
than 0.18% carbon. Similarly, line CD marks the start and end of the transformation
through the peritectic reaction for alloys between 0.18% and 0.50% carbon.
The second horizontal line EFG in Figure 6.6 represents a eutectic reaction at
1147°C given by
L  γ + Fe3 C (6.2)
Whenever the liquid crosses this line EFG at the eutectic point F, the eutectic reac-
tion takes place, and the liquid solidifies into a mixture of two phases, namely,
austenite (γ) and cementite (Fe3C). This eutectic mixture is commonly known as
ledeburite. It is usually not seen in microstructures since austenite undergoes another
transformation during further cooling. In general, the alloys that are having more
than 2.11% carbon are categorized under cast irons. Furthermore, the cast irons are
subdivided into two classes. The alloys having carbon between 2.11% and 4.3% are
FIGURE 6.5 The peritectic region of the iron–iron carbide phase diagram.
FIGURE 6.6 The eutectic region of the iron–iron carbide phase diagram.
called as hypoeutectic cast irons, whereas those having carbon between 4.3% and
6.67% are called as hypereutectic cast irons. Depending upon this classification,
which is based on the amount of carbon, the composition of the alloy at the eutectic
temperature varies. For example, the hypoeutectic cast irons (say 3% C) consist of
proeutectic austenite along with liquid before the eutectic reaction takes place. In
contrast, hypereutectic cast irons (say 5% C) comprise proeutectic cementite. The
exact amount of these proeutectic phases is determined by using the Lever rule. The
remaining amount of liquid transforms into the mixture of austenite and cementite,
i.e., ledeburite upon cooling at the eutectic temperature.
The third horizontal line represents the eutectoid invariant reaction, and it is
given by
γ  α + Fe3 C (6.3)
The eutectoid point is at 0.8% carbon, and the eutectoid temperature is 727°C
(Figure 6.7). As described in equation 6.3, austenite transforms into a mixture of
ferrite and cementite, commonly known as pearlite, because of its pearly appearance
under the optical microscope. The detailed microstructure evolution of pearlite is
discussed in Chapter 8. Like cast irons, steels that contain less than 2.11% carbon are
FIGURE 6.7 The eutectoid region of the iron–iron carbide phase diagram.
also subdivided into two different categories based on the carbon content. Steels that
contain carbon between 0.02% and 0.8% are called hypoeutectoid steels, whereas
hypereutectoid steels contain carbon between 0.8% and 2.11%. This steel portion
of the iron–iron carbide phase diagram is of great interest because of the ample
application areas of various types of steels. The properties of steels depend on their
microstructures, and therefore, the microstructural changes that occur during cool-
ing from the austenite range will be discussed in Chapter 11.
6.4.2 crITIcal TemperaTures In fe–fe3c pHase dIagram

In Figure 6.7, we have discussed the phase transformation upon slow cooling from
austenite. The allotropic change from FCC γ-Fe to BCC α-Fe occurs when the hypo-
eutectoid steel crosses the IJ line. This line is known as the upper critical tempera-
ture and is often labeled in many books as A3. Similarly, there are other critical
temperature lines in the iron–iron carbide phase diagram, where phase transforma-
tions happen. For example, the eutectoid temperature line HI is called as the lower
critical temperature line or the A1 line, where austenite transforms to pearlite upon
cooling. The line IK, which represents the maximum amount of carbon that can be
dissolved in austenite of hypereutectoid steels as a function of temperature, is known
as the upper critical temperature or Acm line. When the hypereutectoid steels cross
this Acm line, the excess carbon above the amount required to saturate austenite pre-
cipitates as cementite along the grain boundaries of austenite. The Curie temperature
line, i.e., the line at 768°C in Figure 6.8, where the α-Fe becomes magnetic on cool-
ing, is called as the A2 line.
However, the critical lines on cooling are not the same as for heating. The phase
transformations always occur below the equilibrium temperature during cooling and
above the equilibrium temperature during heating for which the critical line during
cooling is always lower than the critical line during heating (Figure 6.8). The critical
temperature line during heating is designated as Ac (c is taken from the French word
chauffage, which means heating), whereas during cooling, it is designated as Ar (r is
taken from the French word refroidissement, which means cooling). The difference
between these two lines decreases with the decrease in heating rate. In other words,
the two lines approach each other as the rate of heating and cooling becomes slower.
Hence, at infinitely slow heating and cooling, the two lines appear precisely at the
same temperature.
6.5 EFFECT OF ALLOYING ELEMENTS ON THE

IRON–CARBON EQUILIBRIUM DIAGRAM
The presence of alloying elements affects the iron–carbon equilibrium diagram in
such a way that the diagram no longer represents equilibrium conditions. The alloy-
ing elements change the eutectoid temperature, the position of the eutectoid point,
and the location of the α- and γ-fields. The influence of different alloying elements
on eutectoid temperature is shown in Figure 6.9. As evident from the figure, Mn and
Ni tend to lower the eutectoid temperature. Other alloying elements such as Cr, V, W,
and Si raise the eutectoid temperature, as their concentration increases. Ti and Mo
FIGURE 6.8 Hypoeutectoid region of the iron–iron carbide phase diagram showing various
critical temperature lines.
FIGURE 6.9 Influence of several alloying elements on the eutectoid temperature.
are most effective in increasing the eutectoid temperature. This shift in the eutectoid
temperature is essential to note since it decides whether it will raise or lower the
proper hardening temperature as compared to plain carbon steels.
In contrast, all the alloying elements tend to lower the carbon content of the eutec-
toid composition (Figure 6.10). Ti and Mo are the most effective in lowering the
eutectoid composition. It may also be noted that increasing amounts of Mn and Ni
FIGURE 6.10 Influence of several alloying elements on the eutectoid composition.
lower the critical temperature sufficiently so that austenite does not transform even
on slow cooling, and hence, it is retained at room temperature. These elements, along
with Co, Cu, and Zn, are, therefore, categorized as austenite-stabilizing elements or
austenite stabilizers. On the other hand, alloying elements such as Cr, W, Mo, V, and
Si reduce the austenitic region and enlarge the field in which α- or δ-iron is found.
Since they tend to stabilize ferrite, they are called ferrite stabilizers. These elements
are more soluble in α-iron than in γ-iron and thus favor the formation of a larger
quantity of carbide in the steel for given carbon content.
6.5.1 open γ-fIeld

Mn and Ni enlarge the austenite region, make it a stable phase even at room tem-
perature, and shift the critical points. While the A4 point shifts upward, the A3 point
moves downward, thus narrowing the range of α-phase and enlarging that of γ-phase
(Figure 6.11a). Alloys with more than X% of solute have a solid solution of alloying
elements in γ-iron at all temperatures and, hence, are called austenitic alloys. These
two elements, Mn and Ni, help obtain austenitic steels. For example, austenitic stain-
less steel with 8% Ni and Hadfield manganese steel with about 12% Mn have a wide
range of industrial applications. Some other elements such as Rh, Pd, Cr, Ru, Co, Pt,
Os, and Ir have similar effects.
6.5.2 expanded γ-fIeld

C and N expand γ -region, but due to their limited solubility in iron and by the for-
mation of a compound, the range of γ -field reduces, although the two-phase region
containing γ-phase remains (Figure 6.11b).
FIGURE 6.11 Different categories of the iron–iron carbide phase diagram: (a) open γ-field,
(b) expanded γ-field, (c) closed γ-field, and (d) contracted γ-field.
6.5.3 closed γ-fIeld

Elements such as Si, Al, Be, and P lower A4 and raise A3 to contract the γ-region to
a small area called γ-loop, as shown in Figure 6.11c. Alloys with more than X% of
solute in Figure 6.11c have a solid solution of alloying elements in γ-iron at all tem-
peratures and, therefore, are called ferritic alloys. Influential carbide formers such as
Cr, W, Mo, V, and Ti have similar effects.
The austenite phase may not appear in steel at any temperature when sufficiently
large quantities of such elements are added. For example, by adding 12.8% Cr or
more, the austenitic phase disappears, and the δ-ferrite and α-ferrite merge to give a
continuous ferrite from room temperature up to the melting point. Such steels cannot
be heat-treated conventionally because the austenite phase is not available for phase
transformation.
6.5.4 conTracTed γ-fIeld

B, Ta, Nb, and Zr strongly contract the γ-region, but due to their low solubility in iron
and by the formation of a compound, two-phase alloys are formed before the γ-phase
region is wholly enclosed (Figure 6.11d).
Zener and Andrews have developed a thermodynamic relationship to explain the
formation of open γ-field and closed γ-field in phase diagrams. The relationship is
as follows:
Cα
= β e ∆H / RT (6.4)
Cγ
or,
Cα ∆H
ln = (6.5)
Cγ RT
where Cα and Cγ are the fractional concentrations of an alloying element in α-Fe and
γ-Fe, respectively; β is a constant; ∆ H is the enthalpy change, i.e., Hγ − Hα ; Hγ is
the heat absorbed per unit of solute dissolving in γ-Fe; and Hα is the heat absorbed
per unit of solute dissolving in α-Fe. The solute forms a closed γ-loop if the enthalpy
change, ∆ H is positive, and an open γ-loop if it is negative.
FURTHER READING
Abbaschian, R. & Reed-Hill, R. E. Physical Metallurgy Principles. (Cengage Learning,
Massachusetts, 2008).
Raghavan, V. Physical Metallurgy: Principles and Practice. (PHI Learning Pvt. Ltd., New
Delhi, 2006).
1997).
7 Thermodynamics and
Kinetics of Solid-State
Phase Transformation
7.1 INTRODUCTION
A phase transformation always needs a driving force and happens only when there is
a net decrement in the free energy of the system. In this regard, the thermodynamics
of the feasibility of a phase transformation is essential to understand to comprehend
the presence of possible phases under certain combinations of temperature, pres-
sure, and so on. Furthermore, the conversion is not abrupt. It takes place at a fixed
or dynamic rate, which is again governed by a set of parameters. Hence, a phase
transformation is divided into two parts, namely, nucleation and growth.
7.2 NUCLEATION
Most phase transformations start by the formation of a large number of small par-
ticles of the new phase(s), and these particles subsequently grow in size until the
transformation is complete. The appearance of the particles can be termed as nucle-
ation, while growth involves the increase in the size of these nuclei. There are two
types of nucleation: homogenous and heterogeneous. In homogenous nucleation, the
probability of nucleation at any given site is almost similar to that of any other place
within the assembly, whereas, in heterogeneous nucleation, the likelihood of nucle-
ation at specific preferred locations in the assembly is much higher than that at other
locations. For example, in gases, heterogeneous nucleation can occur at container
walls or at any other impurity particles (Figure 7.1).
7.2.1 Homogenous nucleaTIon

In the case of the solidification of a liquid phase, which is undercooled below the
melting temperature, a solid nucleus forms leading to a Gibbs free energy change of
ΔG, which can be written as
∆G = −Vs ∆Gv + Asl γ (7.1)
where Vs is the volume of the solid sphere of radius r, ∆Gv is the free energy
difference per unit volume between the solid and liquid phases, or the volume free
energy, Asl is the solid/liquid interfacial area, and γ is the solid/liquid interfacial free
energy.
91
FIGURE 7.1 Two types of nucleation.
We can write ∆Gv as
∆Gv = Gvl − Gvs (7.2)
where Gvl and Gvs are the free energies per unit volume of liquid and solid, respec-
tively. It is apparent from Figure 7.2 that at T < Tm, Gvs < Gvl. Therefore, a driving
force for solidification, ∆Gv , always exists whenever a liquid is cooled below the
melting temperature.
At temperature T:
Gvl = H vl − TSvl (7.3)
Gvs = H vs − TSvs (7.4)
∆Gv = ∆H v − T∆Sv (7.5)
At temperature Tm:
∆Gv = ∆H mv − Tm ∆Smv = 0 (7.6)
FIGURE 7.2 Variation of free energy of liquid and solid as a function of temperature.
Solid-State Phase Transformation 93
∆H mv
∆Smv = (7.7)
Tm
∆H v and ∆Sv can be considered as independent of temperature for small undercool-

ing ΔT:
T∆H mv ∆H mv ∆T
∆Gv ≈ ∆H mv − = (7.8)
Tm Tm
where ∆H mv is the latent heat of fusion per unit volume.

Below Tm, ∆Gv is positive. Hence, the free energy change due to the formation of a
small volume of liquid has a negative input. However, γ is always positive due to the cre-
ation of a solid/liquid interface. Now, for the solid nucleus having radius r, we can have
4
∆G = − πr 3 ∆Gv + 4 πr 2γ (7.9)
3
This function first rises, passes through a maximum, and finally decreases, as shown
in Figure 7.3. In other words, the free energy increases when a solid particle begins
to form. When the solid cluster reaches a size equivalent to the critical radius r*, it
grows with a decrease in free energy. On the other hand, a solid cluster of radius r < r*
shrinks and redissolves. This particle is called an embryo, whereas the particle of
radius r > r* is termed a nucleus. Hence, the critical or the maximum free energy,
ΔG*, corresponds to an activation free energy, which is the free energy required for
the formation of a stable nucleus.
d∆G
Setting = 0, we have
dr
2γ
r* = − (7.10)
∆Gv
16πγ 3
∆G * = (7.11)
3 ( ∆Gv )
2
FIGURE 7.3 Free energy change of the nucleus as a function of its radius.
FIGURE 7.4 Critical radii and total free energy at different temperatures.
Substituting the value for ∆Gv , we have
 2γ Tm  1
r* =  (7.12)
 ∆H m  ∆T
 16πγ 3 ( Tm )2  1
∆G * =  2 
(7.13)
 3 ( ∆H m )  ( ∆T )
2
It can be noted that both r* and ΔG* decrease with increasing undercooling (ΔT),
7.2.1.1 Rate of Homogenous Nucleation

The nucleation process can be defined as the addition of one atom to a critical-
sized particle to make it just supercritical. Let us suppose that there are s* atoms in
the liquid phase facing the critical-sized particle across the interface. If anyone of
these atoms jumps from the liquid phase to the solid phase, the particle becomes just
supercritical and is said to be nucleated. Growth continues if additional atoms are
added to the supercritical particles.
The Maxwell–Boltzmann statistics can be used to know the number of critical-
sized particles in the liquid phase. Let us suppose that the total number of particles
per unit volume of the liquid phase be N T. The number of critical-sized particles N*
is approximated by
 ∆G * 
N * = N T exp  − (7.14)
 KT 
where G* is the nucleation barrier. The frequency with which the s* atoms
neighboring the critical-sized particle can cross the interface to join the particle is
given by
 ∆GD 
ν ′ = s*ν exp  − (7.15)
 KT 
FIGURE 7.5 The homogenous nucleation rate as a function of temperature.
where ν is the lattice vibration frequency (which is ~1013/s above the Debye tempera-
ture) and ∆GD is the activation barrier for diffusion across the interface. Now, the
 dN 
rate of nucleation I  = is given by
 dt 

(
 ∆G * + ∆GD
I = N × ν ′ = N T ν s* exp  −
)  (7.16)
 KT 
At the temperature of equilibrium between two phases, T0, when undercooling is zero
(T = 0), ∆G * is infinite. Therefore, the nucleation rate I, as given by equation 7.16, is
zero at the equilibrium temperature. As the temperature decreases, ΔT > 0 and ΔG
becomes finite. As ΔG decreases with decreasing temperature, I increases. But as the
temperature decreases, this at the same time diminishes the probability of attach-
 ∆GD 
ment of atoms to a critical nucleus. Therefore, as the term exp  − decreases
 KT 
rapidly, the nucleation rate decreases with decreasing temperature. It becomes zero
at 0 K. Because of these two opposing factors, I reaches a maximum value, as shown
in Figure 7.5.
The temperature of the maximum rate of nucleation TI max , can be found by setting
dI
= 0, from which the following expression for TI max can be obtained.
dt

d∆G *
=
(
∆G * + ∆GD ) (7.17)
dT TI max
7.2.2 HeTerogeneous nucleaTIon

Nucleation in the solid state, as in the liquid state, is often heterogeneous or nonran-
dom. Preferred nucleation sites include grain boundaries, grain corners, grain edges,
dislocations, and surfaces of inclusions or precipitates embedded in the matrix, as
In the case of heterogeneous nucleation of a spherical cap on a wall of a container
(Figure 7.7), three interfacial energies γ lc – liquid–container interface, γ ls – liquid–
solid interface, and γ sc – solid–container interface can be considered.
FIGURE 7.6 Schematic of heterogeneous nucleation.
FIGURE 7.7 Quasi-equilibrium between three interfacial energies associated with the solid
nucleus.
By balancing the interfacial forces in the plane of the container wall, we have
γ lc = γ sc + γ ls cos θ (7.18)
cos θ = (γ lc − γ sc ) γ ls (7.19)
where θ is the wetting angle.

The creation of the nucleus leads to a Gibbs free energy change ∆Ghet. Hence,
∆Ghet = −Vs ∆Gv + Asl γ sl + Ascγ sc − Ascγ lc (7.20)
Here,
Vs = πr 3 ( 2 + cos θ )(1 + cos θ ) 3

2
(7.21)
Asl = 2πr 2 (1 − cos θ ) (7.22)
Asc = πr 2 sin 2 θ (7.23)
Therefore,
∆Ghet = − { 4 3
3 }
πr ∆Gv + 4 πr 2γ sl S (θ ) (7.24)
∆Ghet = ∆Ghom S(θ ) (7.25)
where S(θ ) is a shape factor given by
S(θ ) = ( 2 + cos θ )(1 − cos θ ) 4

2
(7.26)
FIGURE 7.8 The excess free energy of solid for different types of nucleation. Note that r*
is independent of the nucleation site.
At r = r *,
d∆Ghet

dr
(
= −4 πr 2 ∆Gv + 8πrγ sl S (θ ) = 0 ) (7.27)
2γ sl
r* = (7.28)
∆Gv
16π (γ sl )
3
∆G *
het = S (θ ) (7.29)
3 ( ∆Gv )
2
∆Ghet
*
= ∆Ghom
*
S(θ ) (7.30)
S(θ) has a numerical value ≤1 dependent only on θ, i.e., the shape of the nucleus.
Hence, it is often referred to as a shape factor. Two specials cases need consideration:
1
Case I: θ → 90°, cos θ → 0, ∆Ghet *
→ ∆Ghom *
.
2
The hemispherical-shaped solid is still effective since the energy barrier is half of
that of the homogenous nucleation.
Case II: θ → 0°, cos θ → 1, ∆Ghet*
→ 0.
Since there is no energy barrier to the heterogeneous nucleation, it can start just
at the freezing temperature.
It can be concluded that when the contact angle is small, heterogeneous nucleation
becomes easier. The barrier energy required for heterogeneous nucleation decreases,
7.2.2.1 Rate of Heterogeneous Nucleation

The rate of heterogeneous nucleation can be expressed on analogous arguments for
homogenous nucleation. The number of nucleation sites can be given by
 ∆Ghet
*

*
N het = N T exp  − (7.31)
 KT 
FIGURE 7.9 Variation of homogenous and heterogeneous nucleation rates with undercool-
ing assuming the same critical value of ΔG*.
 ∆Ghom
*

*
N hom = N T exp  − (7.32)
 KT 
* *
N het  N hom (7.33)
It is clear from Figure 7.9 that the heterogeneous nucleation starts at a lower under-
cooling. Now, the rate of homogenous nucleation can be written as
I het 
= N T ν s exp −
* (
 ∆Ghet
*
+ ∆GD )  (7.34)
 KT 
Thus, I het > I hom , because ∆Ghet

*
< ∆Ghom
*
.
7.3 GROWTH KINETICS
Once the solid nucleus has exceeded the critical size and becomes a stable nucleus,
growth occurs. In line with nucleation kinetics, there are also many facets to
study the growth kinetics. In this context, we shall discuss two types of growth:
interface-controlled growth and diffusion-controlled growth. In the case of
interface-controlled growth, a nucleated particle can grow spontaneously by adding
atoms or molecules at its surface or interface. One such example is the freezing of
water that involves the movement of water molecules across the liquid/solid inter-
face. The movement needs thermal activation and is short range.
In contrast, diffusion-controlled growth requires long-range diffusion. For exam-
ple, during precipitation from a supersaturated solution, the critical nucleus gets
enriched with one of the components in relation to the average composition of the
neighboring matrix.
7.3.1 InTerface-conTrolled groWTH

We shall assume here that the interface is nonsingular and incoherent. The precipitate
differs in composition from the matrix. However, the rate of growth is controlled
by the mechanism that allows the solute atoms to cross over from the matrix to
the precipitate. Let ∆GD be the activation energy for an atomic jump across the inter-
face. The change in the chemical free energy per atom in the α → β transformation
is v∆g , where v is the volume per atom. The free energy barrier for an atomic jump
from α to β is ∆GD, while the jump from β back to α is ∆GD − v∆g, as shown in
Figure 7.10.
The net rate of atom transfer from α to β is equal to the difference in the forward
and backward rate. Therefore, the net rate of atomic jumping from α to β per unit
area of the interface is
 ∆GD    v∆g  
I = sν exp  − 1 − exp  (7.35)
 kT    kT  
where s is the number of interfacial atoms in each phase per unit area of interface
and ν is the atomic vibration frequency.
Let us suppose that λ is the jump distance across the interface. The velocity of the
boundary will, therefore, be given by
λI
u= (7.36)
s
 ∆GD    v∆g  
u = λν exp  −  1 − exp  (7.37)
 kT    kT  
For a sufficiently small undercooling, we may assume v∆g  kT. Hence, the growth
velocity becomes
 ∆GD 
u = λν exp  − (7.38)
 kT 
FIGURE 7.10 The free energy barrier for atom jump across the interface during growth.1
The diffusion coefficient Db for the diffusional jump across the interface can be
defined as
 ∆GD 
Db = λ 2ν exp  − (7.39)
 kT 
Expression (7.39) for the growth velocity shows that as T becomes very small, it
approaches zero. Again, it is zero at the transformation temperature, T0. It means
that the growth velocity reaches a maximum value at some intermediate temper-
ature. This is verified experimentally for the transformation of white to gray tin
(T0 = +13°C). The maximum in the growth rate (~1 mm/h) occurs at −32°C, where
ΔT = 45°C. After that, the growth rate decreases exponentially with temperature, as
the boundary diffusion Db, falls, as given by equation 7.39.
7.3.2 dIffusIon-conTrolled groWTH

It refers to the formation of a new phase with a composition different from that of the
old phase, and the rate is controlled by the long-range diffusion of one of the com-
ponent atoms through the matrix phase. To understand this phenomenon, let us take
a simple example of precipitation of β-phase particles from a supersaturated solid
solution of α in a binary system of A and B. Zener proposed a simple theory for this
type of growth. Let us consider Zener’s theory for nonspherical precipitates, which
grow as a plate in the direction normal to its surface. Figure 7.11 shows the precipi-
tate particles of the β-phase formed in the matrix. The particles being abundant in B,
the composition of the matrix changes toward the equilibrium value cαβ (< c ).
FIGURE 7.11 A schematic binary phase diagram, showing the decreasing solid solubility of
B in A, as the temperature decreases.1
FIGURE 7.12 The concentration of B atoms as a function of distance during continuous

precipitation of β .1
The concentration–distance profile during the precipitation is shown in Figure 7.12

along the distance axis, at y = 0. The precipitation starts in the matrix of the initial
composition c. The precipitate has grown to a size r, where the interface separating
the precipitate from the matrix is located. The composition within the precipitate
particle from y = 0 to y = r is uniform and is equal to cβα . Equilibrium is assumed
to prevail at the interface so that the concentration of B in the α -phase can be shown
as cαβ . Due to the jumping of atoms from the matrix into the precipitate, there is
a drop in the concentration of B atoms in the matrix, as one moves away from the
interface and reaches c. As the precipitate particle grows to larger sizes, the region
of the surrounding matrix depleted of B atoms extends to larger distances, as shown
by the dotted line profile in Figure 7.12. cα denotes the concentration of B atoms in
the depleted region.
Let us suppose that in small time t, the boundary of the precipitate was moved
forward into the matrix through a distance dx. As the composition at the boundary or
interface remains independent of time, the number of solute atoms that arrive at the
interface from the matrix must be equal to the number of solute atoms that are added
to the growing particle. Hence, we have
(Cβα − Cαβ ) Adx = AD  dC α 

dx 
 dt (7.40)
where D is the diffusion coefficient which is assumed to be independent of concen-

dCα
tration and is the concentration gradient of B in the matrix phase at the inter-
dx
face. Now, the growth rate of the particle is given by
dx D dCα
u= = (7.41)
dt Cβα − Cαβ dx
7.4 TIME–TEMPERATURE TRANSFORMATION AND

CONTINUOUS COOLING TRANSFORMATION DIAGRAMS
The iron–iron carbide equilibrium phase diagram has very little significance
in practice as it does not include the phase transformation kinetics into account.
As most of the heat treatment practices are governed by nonequilibrium or fast cool-
ing, the finished or final product carries the effect of the cooling rate and hence
exhibits different properties. The equilibrium phase diagram also does not include
the metastable phases like bainite and martensite. Thus, the requirement of an under-
standing of the correlation between temperature and time on the phase transforma-
tion is realized.
The first idea regarding the phase transformation kinetics was established by
Davenport and Bain (Trans. AIME, vol. 90, p. 117, 1930). The diagram that reveals
the phase transformation of austenite into various phases at a certain combination
of temperature and time is known as a TTT curve. The TTT curve is based upon
isothermal cooling of the steel specimen. Each composition of steel has its character-
istic TTT curve. Although grain size and microstructural inclusions have some role
to play in this context, in general, their effects are neglected.
Steel with the eutectoid composition of about 0.8 wt% carbon is chosen for a
more straightforward interpretation of the TTT curve, as this does not contain any
proeutectoid phase in it. Several thin pieces (as small pieces will respond to the
temperature soon) of this composition are taken. All these specimens are heated
to a sufficiently high temperature (~775°C) to get a completely austenitic structure
in the furnace or molten bath. Another furnace or molten bath is kept ready at a
specific subcritical temperature, i.e., below the lower critical temperature (~675°C).
After completion of austenitization in the earlier furnace, all the specimens were
transferred immediately to the later furnace. In this furnace, each sample is kept for
different periods (e.g., 30 s, 5 h, 15 h, 25 h, 50 h, and 75 h) and then quenched in iced
brine. The microstructural characterization of all the samples is done to analyze the
overall pearlite fraction. A suitable way of representing both the time and tempera-
ture dependence of this transformation is shown in Figure 7.13.
For the sample, which was kept for very little time at the subcritical temperature,
there is little chance of formation of pearlite, as it involves diffusional transforma-
tion, which requires enough time. Hence, quenching at this stage results in a mar-
tensitic structure. Increasing the time of stay at the subcritical temperature yields a
higher pearlite content and lower martensite upon quenching; finally, beyond some
time, the structure becomes pearlite entirely. Figure 7.14 represents the variation of
pearlite percent in the final product with time at the subcritical temperature. The
same methodology is followed for the same composition specimen but at different
subcritical temperatures. It is exciting to know that at any particular temperature,
the transformation rate is inversely proportional to the time required to complete the
transformation to 100% pearlite. From Figure 7.14, it can be seen that at a tempera-
ture of 675°C, more time is required to complete the transformation. The transforma-
tion rate increases with decreasing temperature such that at 550°C, only about 10 s
is required for 100% pearlite transformation. Thus, the shorter the time, the higher
is the transformation rate.
FIGURE 7.13 A typical TTT diagram for eutectoid steel. TTT, time–temperature
transformation.
FIGURE 7.14 Variation of pearlite percent with the time stay at the subcritical temperature.
TTT diagrams are only strictly applicable to transformations carried out at con-
stant temperatures. However, in industrial practices, steel is rarely quenched to a
constant temperature. Instead, it is heated into the austenite range and continuously
cooled to room temperature at different cooling rates. The relationships between
transformation temperature and time can easily be depicted from continuous cool-
ing transformation (CCT) diagrams. For simplicity, let us choose the steel of eutec-
toid composition. Figure 7.15 shows two cooling curves at different cooling rates to
understand the CCT diagram. The specimens are cooled from above the eutectoid
temperature to room temperature. As seen from Figure 7.15, curve 1 crosses the
TTT diagram at point a. This point indicates the time required to nucleate pearlite
isothermally at the temperature of point a, i.e., around 650°C.
At temperatures above 650°C, more time is required to start the pearlite transfor-
mation. Therefore, the point in continuous cooling where actually conversion starts
lies to the right and below point a (designated as b).
Similarly, the finish of the pearlitic transformation can be shown as point d,
whereas point c indicates the finish of the isothermal transformation. So, it can be
concluded that the CCT lines are shifted to right with respect to the corresponding
isothermal transformation lines. This also explains why there is no bainitic transfor-
mation in the metal in case of continuous cooling. The fact is that the pearlite range
overshadows the bainitic range. In other words, austenite is converted to pearlite
before the cooling curve reaches the bainitic transformation range. This happens
in case of slow or moderate cooling rates (curve 1). Alternatively, in case of high
heating rates (curve 2), the specimen stays in the bainitic transformation range for a
shorter time to allow any significant bainite to form. Hence, in eutectoid steel, con-
tinuous cooling does not yield bainite.
It is clear that pearlite has its transformation start temperature (Ts) and transfor-
mation finish temperature (Tf) and, for bainite and martensite, the Ts (i.e., Bs and
Ms) is independent of the cooling rate. For simplicity, we have marked Ts as heavy
lines in Figure 7.16.
With decreasing cooling time or increasing cooling rate, Ts gradually decreases
from the A3 temperature to the Bs temperature. When a higher cooling rate is
reached, the Ts jumps to the martensitic plateau. These plateaux signify the forma-
tion of metastable products, and they indicate that each metastable product possesses
its Ts (Ms, Bs, and Ma), and it can only be formed within a certain cooling rate range.
FIGURE 7.15 A typical continuous cooling transformation diagram.

FIGURE 7.16 CCT diagram of 0.24C-1.67Mn-0.39Si-0.14Ni-0.17Cr-0.22Mo-0.11V.2 CCT,

continuous cooling transformation.
There also exists a minimum rate of quenching (critical cooling rate, CCR) that will
produce a martensitic structure. This CCR misses the nose at which pearlitic trans-
formation begins (Figure 7.17).
Recent advancements have proved that this is a misleading diagram as it shows
that continuous cooling has to do something with the S-curve in TTT diagrams,
FIGURE 7.17 Effect of cooling rates on transformation.

FIGURE 7.18 TTT diagram superimposed on a CCT diagram (dashed lines) to emphasize
the difference in kinetics between the cooling transformations and isothermal transforma-
tions.2 TTT, time–temperature transformation; CCT, continuous cooling transformation.
which is erroneous. The TTT curve is superimposed over the CCT diagram to under-
stand the real mechanism, as shown in Figure 7.18.
( )
It is not the CCR but the lower CCR limit of the martensite TLM that decides the
formation of martensite during cooling. Depending upon the relative position and
shape of the TTT and CCT diagrams, the CCR can be above or below the TLM . In
this case, CCR is lower than TLM . Therefore, to analyze a complicated heat treatment
procedure involving both cooling and isothermal transformations, CCT and TTT
diagrams should be used together.2
REFERENCES

FURTHER READING
Porter, D. A. & Easterling, K. E. Phase Transformations in Metals and Alloys, Third Edition
(Revised Reprint). (CRC Press, Boca Raton, FL, 1992).
8 Phase Transformation
in Steels
8.1 INTRODUCTION
The properties of steel are mostly dependent on the transformed product(s) from aus-
tenite, which are obtained during different heat treatment practices, and therefore, it
is essential to understand the phase transformation in steels. Not only the presence of
phases like pearlite, bainite, and martensite affect the properties, but also their mor-
phology is equally significant in deciding the resultant properties. Hence, this chap-
ter deals with the formation of austenite and its subsequent transformation behavior.
Indeed, this study is essential in order to study and understand the theoretical aspects
of available standard and modified heat treatment methods, which is described in
detail in the subsequent chapter.
8.2 FORMATION OF AUSTENITE

The carbon steel contains a mixture of ferrite and pearlite, only pearlite, or pearlite
and cementite at the room temperature depending on the carbon content of the steel.
When the steel is heated into the austenite range, the formation of austenite occurs.
For example, for hypoeutectoid steels, the formation of austenite starts when the
temperature of the steel crosses the lower critical temperature, AC1. Similarly, aus-
tenite starts to form just at the eutectoid temperature, i.e., 723°C for eutectoid steels.
The transformation begins with the nucleation of austenite at the interfaces between
the ferrite and cementite of pearlite, as shown in site 1 of Figure 8.1.
However, the more preferred site for nucleation is the boundary of the pearlite col-
ony (site 2). This transformation process continues until all the ferrite and cementite
dissolve to form austenite. While ferrite transforms to austenite by changing the
crystal structure, cementite transforms to austenite by diffusion of carbon atoms.
However, the process of dissolution of ferrite is faster and, therefore, gets completed
before that of cementite. This results in the formation of inhomogeneous austenite.
FIGURE 8.1 Different stages of austenite formation from lamellar pearlite.
107
FIGURE 8.2 Effect of temperature and time on the austenite formation from pearlite in
eutectoid steel.
Homogenization of austenite happens either by increasing the temperature or by

holding the steel for a longer time so that all the carbon atoms diffuse. The complete
transformation process is illustrated in Figure 8.2.
The grain size of austenite decides the properties of steel. So, different prop-
erties such as tensile strength, toughness, machinability, and hardenability can be
changed by varying the austenite grain size. Based on the tendency of austenite
grain to grain growth, there are two types of steels, namely, inherently fine-grained
steels and inherently coarse-grained steels. Inherently fine-grained steels resist the
grain growth of austenite with increasing temperature, whereas inherently coarse-
grained steels grow abruptly with rising temperature (1000°C–1050°C), and hence,
steels with coarse grains are obtained. However, once the temperature increases
beyond 1050°C, the grain coarsening is quite rapid in steels with finer grain size,
and at around 1150°C temperature, both the steel types have the same grain size
(Figure 8.3). Since the inherently fine-grained steels are able to retain a fine austenite
grain size for more extended periods, these are often used for the carburizing process
at around 950°C with long carburizing cycles.
8.3 PEARLITIC TRANSFORMATION
Pearlite can be defined as the mixture of two distinct phases, i.e., ferrite and cement-
ite in a well-defined pattern. When austenite having composition same as eutectoid
one is cooled below the lower critical temperature, pearlite is usually formed. In
case of a eutectoid steel, the active nuclei may be any of the two phases-ferrite or
cementite, but usually it is the cementite. The rate at which pearlitic transformation
takes place is governed by both (i) nucleation rate (N) and (ii) growth rate (G). Both
these parameters are further controlled by the temperature at which transformation
takes place. Nucleation rate is usually defined by the number of fresh nuclei that are
formed per unit volume in a unit time in the untransformed austenite. It becomes
Phase Transformation in Steels 109
FIGURE 8.3 Effect of temperature on the austenite grain size in inherently fine-grained
and coarse-grained steels.
higher as the temperature of transformation is lowered and at around the tempera-

ture close to 555°C, it apparently becomes maximum. At a constant temperature, the
pearlitic growth is not a function of time. As temperature decreases below the eutec-
toid temperature, it increases and eventually reaches a maximum value close to the
nose of the “S” curve. With further decrement in temperature, the rate again drops
down. The morphology of pearlite is dependent on the relative rate of nucleation and
growth, i.e. N/G. If this ratio is high, large numbers of nodules form but grow only
a small distance as impingement occurs. With the fall of temperature, N increases
at a faster rate than G, resulting in several pearlite nodules in one grain of austenite.
These pearlitic colonies tend to grow almost equally in both longitudinal and
transverse directions of the lamella, resulting in a near spherical nodule. The most
famous and widely accepted physical model for the formation of pearlite from aus-
tenite was proposed by F.C. Hull and R.F. Mehl and hence popularized as the Hull–
Mehl model. Figure 8.4 shows different stages involved in the formation of pearlite
from austenite. At first, the cementite nucleates at the austenite grain boundary and
then grows. This growth occurs only by the removal of carbon from austenite, and
consequently, austenite transforms to ferrite. Ferrite thus recently evolved starts
growing along the cementite plate surface and thereby enriches the adjacent austen-
ite with carbon. This process of nearby austenite enrichment by carbon continues till
the concentration next to it approaches as that of cementite, and thus, a new cement-
ite is generated next to the ferrite plate. In this process, other plates of ferrite and
cementite are formed, and they form a colony.
8.3.1 InTerlamellar spacIng

The interlamellar spacing of pearlite can be measured as the distance from the
center of one ferrite (or may be taken as cementite) plate to immediate next fer-
rite (or cementite) plate in one pearlite colony. It is measured when the accounted
FIGURE 8.4 Different stages involved in the nucleation and growth of pearlite colony from
austenite.
plates are perpendicular to the plane of the sample on which imaging is carried
out. Assuming all other parameters to be the same, the interlamellar spacing is a
function of the transformation temperature. At relatively higher transformation
temperature, the interlamellar spacing is higher giving rise to formation of coarse
pearlite. However, a finer pearlitic structure (i.e., smaller interlamellar spacing) usu-
ally possesses a higher strength. Alloying elements, except cobalt, increase the inter-
lamellar spacing because they transform at higher temperatures. This process is not
dependent on the prior grain size of austenite and the extent of concentration and
structural homogeneity of austenite and hence can be termed as structure intensive.
It decreases with decreasing transformation temperature; thus, pearlite becomes fine
and finer.
8.3.2 kIneTIcs of pearlITe TransformaTIon

Johnson and Mehl related the fraction of austenite transformed to pearlite as a
function of time by the following equation:
f (t ) = 1 − e( −π 3) NG 3 + t 4 (8.1)
Here, f (t ) is the fraction of austenite converted to pearlite. This equation has been
developed based upon the following assumptions:
• N and G are constant with time.

• Nucleation is considered as a random process.
• The shape of the nodules remains spherical as they grow bigger (until they
impinge on each other).
When the fraction transformed f (t ) is plotted against 4 NG 3 t, a sigmoidal curve

appears (Figure 8.5), which indicates that the kinetic behavior of pearlite formation
is a nucleation and growth process.
FIGURE 8.5 Reaction curve for nucleation.
8.4 BAINITIC TRANSFORMATION
Bainite is composed of ferrite and carbide with ferrite as the nucleus. However,
unlike pearlite, bainite is not having a lamellar morphology, and cementite is dis-
tributed over the ferrite in a much finer scale. In case of a eutectoid steel, formation
of bainite takes place at a temperature range of 200°C–500°C from undercooled
austenite. In steels having disconnected pearlite and bainitic zones (in case of some
alloy steels), bainitic transformation is possible by both isothermal and continuous
cooling from austenite, whereas in the case of plain carbon steels and some other
alloyed steels, overlapping of these pearlitic zones with the bainitic zone makes
bainitic transformation not so easy. Depending on the cooling rate, either pearlite or
bainite or a mixture may be formed. In case of very high rate of cooling, martensitic
transformation may also be possible.
In such kind of steels, 100% bainitic steel may not be produced by continuous
cooling. Rather, a faster cooling rate should be adopted in order to suppress pearlitic
transformation and to make the cooling curve intersect the continuous cooling trans-
formation curve in the bainitic region.
8.4.1 upper BaInITe

When the transformation temperature remains in the upper region of the bainitic
zone, i.e., usually 550°C–400°C, upper bainite is formed. The morphology of upper
bainite comprises lath-shaped ferrite, and the carbides that are precipitated out from
the austenite matrix are nucleated along the needle axis of the bainite as shown in
Figure 8.6. Upper bainite is also called feathery bainite due to its resemblance with
the feather of a bird.
8.4.1.1 Formation of Upper Bainite

For formation of the upper bainite, ferrite is the active nucleus, and thus, there is a
coherency between ferrite and the prior austenite. After nucleation of this ferrite,
further growth takes place when the transformation is allowed to take place in
FIGURE 8.6 Different stages involved in the formation of upper bainite.
the upper bainite zone of the bainitic region. Due to rejection of carbon from this
structure, the nearby austenite becomes eventually carbon rich and makes condi-
tions favorable for cementite formation at the interface between the bainite and
austenite. Growth of this generated carbide results in a carbon-depleted austenite
next to it, again making conditions favorable for ferrite by the sheer mechanism.
This process continues, and the growth of already-formed bainite and nucleation
of fresh bainite take place simultaneously. Hereby, it can be inferred that diffusion
and distribution of carbon in the austenitic region next to the bainitic structure is
the dominant rate-controlling step in this case. It has also been postulated that
various mechanisms of surface relief are also associated with the formation of
upper bainite.
8.4.2 loWer BaInITe

From the nomenclature, it is obvious that formation of the lower bainite is facilitated
when the transformation is allowed to take place in the lower zone of the bainitic
region, i.e., 250°C–400°C. Interestingly, in case of lower bainite, the carbides are
generally precipitated at 55°–60° angle to the major bainitic axis as presented in
Figure 8.7. It is also termed as plate bainite. It has a martensite-like acicular appear-
ance as it forms as individual plates adopting a lenticular shape. Carbides are on finer
scale and have shapes of rods or blades.
8.4.2.1 Formation of Lower Bainite

The rate at which the carbon is diffused from the austenite having low carbon con-
tent is the rate-controlling step in case of formation of lower bainite. The nature of
carbide formed is a function of the chemical composition of the steel and the tem-
perature at which transformation takes place.
FIGURE 8.7 Different stages involved in the formation of lower bainite.
8.5 MARTENSITIC TRANSFORMATION
Martensite is formed by quenching austenite. Martensitic transformation is a diffu-
sionless transformation. Since there is no diffusion of carbon in this transformation,
there is no change in the chemical composition, i.e., both alpha and gamma irons are
having the same composition. Hence, α-iron becomes saturated with carbon as it is
well known that the solubility limit of carbon is considerably more in γ-iron than that
of α-iron. Accordingly, martensite is usually defined as a supersaturated solid solu-
tion of carbon in α-iron. Martensite has a body-centered tetragonal structure. The
tetragonality is due to the trapping of carbon in the solution. While two dimensions
of the unit cell are same, the third one is slightly expanded because of the trapped
carbon.
There are several other important characteristics of the martensitic transforma-
tion, which are described in the following.
8.5.1 ImporTanT cHaracTerIsTIcs

8.5.1.1 Diffusionless Transformation
Martensitic transformation is a diffusionless transformation, as martensite has the
same composition as the parent austenite. Carbon remains in dissolved solid solution
state in martensite, even if the other alloying elements are present.
8.5.1.2 Surface Relief
Martensite crystal is displaced partly above and partly below the surface of the par-
ent austenite. It indicates that the shear (or displacive) mechanism is responsible
for this transformation. Surface relief is an essential characteristic of martensitic
transformation.
FIGURE 8.8 Effect of carbon in the martensite range.
8.5.1.3 Ms –Mf Temperatures

Martensitic transformation starts at Ms temperature and ends at Mf temperature. The
Ms temperature is a specific temperature for a given steel. This temperature depends
on the chemical composition of the steel. The relationship between Ms temperature
and the chemical composition is as shown in the Andrews equation:
M s (°C) = 539 − 423(%C) − 30.4 (%M n ) − 12.1(%Cr) − 17.7(%Ni) − 7.5 (% M 0 ) (8.2)
Almost all alloying elements, except cobalt and aluminum, lower the Ms tempera-
ture. Carbon has the most intense effect on Ms temperature. The higher the carbon
content of steel, the lower the Ms temperature (Figure 8.8). The temperature at which
martensitic transformation ends, i.e., Mf, is a function of the cooling rate and, hence,
can be changed by controlling the cooling rate. The Mf temperature is usually low-
ered by reducing the rate of cooling. Intuitively, this can be attributed to the fact that
Mf is mostly not mentioned in the TTT diagrams.
8.5.1.4 Athermal Transformation
The transformation of austenite to martensite proceeds on continuous cooling below
the Ms temperature and stops as soon as cooling is interrupted, i.e., the steel is
allowed to stay at a temperature between the start and finish temperatures of mar-
tensitic transformation.
Further transformation is possible only upon further cooling. Hence, martensitic
transformation is usually ascribed as athermal transformation. If the room tempera-
ture lies between Ms and Mf temperatures, quenching the steel to room tempera-
ture leads to a large amount of untransformed austenite along with martensite. This
untransformed austenite is called retained austenite. If the cooling of steel is arrested
between Ms and Mf temperatures and again the cooling is resumed by lowering the
temperature, the transformation does not start immediately but starts after austen-
ite is undercooled to a much lower temperature, and in the end, a large amount of
austenite remains untransformed even after crossing Ms temperature. This extra

untransformed austenite because of the arrest is called stabilized austenite.
8.5.1.5 Effect of Applied Stress on Transformation

If stress from outside is applied on austenite above the Ms temperature, the Ms tem-
perature gets raised, i.e., martensite forms above its Ms temperature. The plastically
deformed steel may undergo martensitic transformation by cooling below a critical
temperature, known as Md.
For a given plastic deformation, the percentage of martensite formed becomes
higher as the transformation temperature is brought down from Md to Ms.
8.5.2 Hardness of marTensITe

The presence of carbon is the main reason behind the high hardness of martensite in
the case of plain carbon steels. It increases as carbon content in the steel increases.
Both Ms and Mf temperatures are controlled by the steel composition, which even-
tually decide the extent of hardening of the martensite. With lowering in carbon
content, the Ms and Mf temperatures are increased. Hence, in high-carbon steel, the
retained austenite content usually remains at a higher side.
8.5.3 morpHology of marTensITe

8.5.3.1 Lath Martensite
It has the shape of the strip, the length of which has the largest dimension and is
limited by the grain boundary of austenite. Laths are grouped in packets in parallel
fashion laths and have a high density of dislocations in the range of 1015–1016 m−2 as
in a heavily cold worked metal and are created by the process of slip (Figure 8.9a).
Lath martensite is formed when Ms temperature is high, such as in low and medium
carbon steels.
8.5.3.2 Plate Martensite
Plate martensite is also called acicular or lenticular martensite (lens-shaped).
It resembles the shape of mechanical twins, as shown in Figure 8.9b. It is formed
in steels having low Ms temperatures, i.e., steels having higher carbon as well as
alloying elements.
FIGURE 8.9 Schematic micrograph of (a) lath martensite and (b) plate martensite.
FURTHER READING
Singh, V. Physical Metallurgy. (Standard Publishers Distributors, New Delhi, 2005).
1997).
Sharma, R. C. Principles of Heat Treatment of Steels. (New Age International, New Delhi,
2003).
9 Heat Treatment Furnaces
9.1 INTRODUCTION
The achievement of excellent heat treatment properties relies upon the appropriate
choice of heat treatment furnaces and the atmosphere maintained inside the furnace.
Heat treatment furnaces mostly contain heating chambers in the form of refractory
vessels that hold the charge and contain the heat. There are different sources of heat
through which the temperature inside the furnace is controlled. Depending upon the
size of the charge, heating mechanism, charge handling system, and several other
parameters, heat treatment furnaces are designed.
9.2 CLASSIFICATION OF FURNACES

The heat treatment furnaces are generally classified according to the source of heat,
use, type of operation, and working environment.
Based on the sources of heat, the furnaces can be classified as fuel-fired fur-
naces and electrically heated furnaces. The choice of the right fuel depends on the
availability and cost of fuel. Depending on the type of fuel, the fuel-fired furnaces
can be further classified as solid fuels, liquid fuels, and gaseous fuels. Commonly
used solid fuels are coal, pulverized coal, and coke. The advantage of using coal
as a solid fuel is its cheap availability, although smoke and lack of temperature
control are associated with it as limitations. The problem of smoke can be avoided
by using coke since it is less volatile and has a calorific value less than that of
coal. Among liquid fuels, fuel oil, gasoline, and kerosene are typically used. These
liquid fuels can be easily stored and fired at any time, but they are not being used
nowadays due to their high cost. Nevertheless, oil-fired furnaces are used where a
high temperature (>1000°C) is required. A heat circulation arrangement is neces-
sary for such types of furnaces, as the temperature inside the furnace is not uni-
form, and hence, these are not economical. On the other hand, gas-fired furnaces
are economical and have specific advantages over other furnaces. They possess
better control of temperature, have a simpler design, and can be used up to 1500°C.
Some examples of gaseous fuels are natural gas, coal gas, producer gas, cracked
oil gas, and refinery gas.
Electrically heated furnaces offer certain advantages over other furnaces such as
more straightforward design, uniformity of temperature inside the furnace cham-
ber, highest efficiency in heat utilization, and clean working conditions. These are
the furnaces that can attain very high temperatures and are convenient to start
and switch off at any moment. Based on the method of heating, electric furnaces
are again classified as resistance furnace, induction furnace, plasma arc furnace, arc
furnace, and electron beam furnace. Resistance furnaces are typically used for the
heat treatment of metals and alloys. The temperature in such types of furnaces can
117
TABLE 9.1
Common Resistors Used in Electrical Resistance Furnaces
Maximum Working
Name Composition Temperature (°C)
Constantan Cu, 40% Ni 900
Nichrome I Ni, 20% Cr 1100
Nichrome II Ni, 24% Fe; 16% Cr 950
Alumel Ni, 3% Mn; 2% Al; 1% Si 1200
Chromel Ni, 10% Cr 1200
Chromel C Ni, 23% Fe; 15% Cr; 2% Mn 900
Kanthal Fe, 25% Cr; 5% Al; 3% Co 1400
Tungsten 2400
Molybdenum 1800
Tantalum 2200
Platinum 1500
Pt–Rh alloy, Pt, 10% Rh 1700
Thoria 2400
Graphite 2000
be controlled easily to a high degree of accuracy. Various resistors are used in these
furnaces depending upon the desired temperature. A list of resistors used in electri-
cal resistance furnaces is given in Table 9.1. Induction furnaces are generally used
for surface hardening purposes. Other furnaces such as plasma arc furnace, arc fur-
nace, and electron beam furnace are mainly used for melting metals and alloys.
Depending on the use or heat treatment types, the furnaces can also be classified.
For example, salt bath furnaces are used for treatments such as annealing, normal-
izing, and hardening. Sealed quench furnaces are used for carburizing or carboni-
triding purposes. According to the type of operation, the furnaces are categorized
into two groups, namely, batch furnace and continuous furnace. Both these types of
furnaces are described in detail in the next two sections.
9.3 BATCH FURNACE
A batch furnace consists of a refractory-lined insulated chamber enclosed in a steel
shell with one or more access doors. These furnaces are widely accepted because
of their flexibility to varying sizes of the workpiece to be heat-treated and different
heat treatments to be performed. The term “batch” signifies that the heat treatment
of the workpiece is carried out and completed in different batches. A batch furnace
can be horizontal or vertical type depending upon the size of the workpiece and type
of heat treatment adopted. These types of furnaces are beneficial for carrying out
laboratory-scale experiments. Some very commonly used batch furnaces are box-
type furnace, muffle furnace, bell furnace, vacuum furnace, fluidized bed furnace,
pit furnace, and bogie hearth furnace.
Heat Treatment Furnaces 119
FIGURE 9.1 Schematic diagram of a box-type furnace.
9.3.1 Box-Type BaTcH furnace

One of the simplest types of batch furnaces is a box-type batch furnace. The chamber
is rectangular in section, and it has an opening just as in a box, as schematically illus-
trated in Figure 9.1. These furnaces can be used for annealing, pack carburizing, and
hardening of low-alloy steels. Generally, the loading and unloading process of the
samples is done manually from the front end of the furnace. For heavy components,
the bottom loading furnaces are used, in which the bottom cover and the transfer cab
are lowered, and the charge is placed on it. The bottom cover is closed after loading
of the charge. These box-type furnaces are heated through electricity or fuel.
9.3.2 muffle furnace

A muffle furnace is an empty cuboid or tube-shaped furnace made of unique refrac-
tory material or nonscaling steel. Any fuel or electrical energy can be used to heat
the muffle furnace. In this case, the heat source does not legitimately reach the mate-
rial being heat-treated. The samples are charged, and electrical energy or gas firing
can be utilized to heat the furnace externally. Electrically insulated furnaces are
used extensively for the heat treatment of small parts. In electric muffle furnaces, as
shown in Figure 9.2, the muffle is surrounded by different heating elements such as
SiC heating rods, MoSi2 heating rods, nichrome, or kanthal wires, depending upon
the maximum temperature desired.
In contrast, in the case of gas-fired muffle furnaces, the muffle gets heated by
the hot gases generated outside the muffle (Figure 9.3). The hot gases are made to
circulate through the space between the interior wall and the exterior muffle wall.
Since the combustion products of the gas do not enter the heating chamber or muffle,
scaling of the components is prevented. Muffle furnaces are used for annealing,
nitriding, and hardening purposes.
FIGURE 9.2 Schematic diagram of an electric muffle furnace.
FIGURE 9.3 Schematic diagram of a gas-fired muffle furnace.
9.3.3 pIT furnace

Pit furnace is mainly used for heavy and large parts such as tubes, spindles, shafts,
and rods. It comprises the furnace put in a pit and a cover or lid put over the furnace
(Figure 9.4). Most of the portion of the pit furnace lies below the ground level. Fans
placed inside the pit furnace promote uniformity of temperature and gas composi-
tion, thereby reducing distortion and warpage. Pit furnaces are especially appropriate
FIGURE 9.4 Schematic diagram of a pit furnace.
for parts to be cooled in the furnaces as direct quenching with such huge charge, and
the enormous furnace is not possible.
9.3.4 BogIe HearTH furnace

It is a modified variant of the box-type batch furnace. The furnace is especially
suitable for the heat treatment of large and heavy components. The bogie acting as
a hearth with the full charge is put inside the furnace, which is refractory topped.
A schematic illustration of the bogie hearth furnace is shown in Figure 9.5. The heat-
ing is done by a fuel or by some electric resistance heating elements. Proper sealing
controls the atmosphere inside the furnace. These types of furnaces are typically
used for stress-relieving, annealing, and hardening purposes.
9.3.5 Bell furnace

Bell furnaces, also referred to as liftoff cover furnaces, consist of two contain-
ers (Figure 9.6). The inner container comprises the charge, which is placed on the
hearth, and a retort on the top of the charge. The retort is sealed at the base with
sand. The outer bell-shaped container, which carries the heating elements, covers
the inner container. Protective gas is supplied to the sealed retort for constant pro-
tection. After heating of the first charge, the bell-shaped container is lifted off and
placed on the second assembly to carry out the heating process. In the meantime,
FIGURE 9.5 Schematic diagram of a bogie hearth furnace.
FIGURE 9.6 Schematic diagram of a bell furnace.
the first assembly and its charge cool down under the protective atmosphere, as the
gas is continuously being fed all through the sealed retort. Such types of furnaces
are widely used in wire industries for annealing coils and strips under a controlled
atmosphere to avoid decarburization or oxidation.
9.4 CONTINUOUS FURNACE
In a continuous furnace, the parts to be heat-treated are continuously introduced from
one end of the furnace. After the heat treatment of the parts is over, the heat-treated
parts are forced to move to the other end of the furnace where they are discharged.
These furnaces are meant for typically mass production of components and are,
therefore, mostly found in industries or pilot plants of national laboratories. A con-
tinuous furnace can have different zones depending upon the type of heat treatment
adopted. For example, a continuous carburizing furnace has separate chambers for
heating, carburizing, and the diffusion process. A rotary kiln has different zones
for preheating, reaction, and cooling purposes. Modern continuous furnaces include
rotary hearth furnace, conveyor furnace, tunnel furnace, shaker hearth furnace,
rotary retort furnace, and rotary kiln.
9.4.1 roTary HearTH furnace

The rotating hearth furnace consists of a rotating hearth that rotates along its verti-
cal axis inside a stationary roof, as shown in Figure 9.7. Components with different
compositions can be heat-treated in such types of furnaces. The components to be
heat-treated are charged through an opening and discharged from the same opening
or adjacent opening after the completion of the heat treatment cycle. The speed of
rotation is varied to such an extent that the heat treatment cycle is finished by the
time the hearth finishes one complete revolution.
9.4.2 conveyor furnace

In conveyor furnaces, endless belts are utilized to move the segments through the
heater. Such furnaces are suitable for smaller parts. The parts are charged continuously
into the belt at one end of the furnace and discharged from the other end. Inside the
furnace, the movement of the belts carrying the parts is regulated in such a way
FIGURE 9.7 Schematic diagram of a rotary hearth furnace.

FIGURE 9.8 Schematic diagram of a conveyor furnace.
that the heat treatment is completed by the time the parts are discharged. Figure 9.8
shows the schematic illustration of the conveyor furnace. These furnaces are appli-
cable for solidifying, hardening, tempering, and normalizing purposes.
9.4.3 Tunnel furnace

In a tunnel furnace, as shown in Figure 9.9, several cars carrying the charge are
moved into the furnace at one end. The cars are pushed through the furnace with
the help of some mechanical means. The movement of cars is so arranged that when
a car is removed at the discharged end, another car is loaded simultaneously at the
charging end. These types of furnaces are typically used for annealing purposes.
FIGURE 9.9 Schematic diagram of a tunnel furnace.

FIGURE 9.10 Schematic diagram of a salt bath furnace.
9.5 SALT BATH FURNACE

A salt bath furnace (Figure 9.10) consists of an oval or rectangular container made
of ceramic or metal. This container holds the molten salt in which the samples to
be heat-treated are immersed. The container is heated and maintained at a required
temperature either by the combustion of fuel or by electric resistors. Nitrates, chlo-
rides, carbonates, cyanides, and caustic soda are the most commonly used salts in
this furnace. The heat transfer takes place very quickly, as salt bath comes in the best
intimate contact with the charge since the mode of heat transfer is mainly by convec-
tion. Also, the molten salts possess high heat capacity, resulting further in rapid heat-
ing of the charge as compared to conventional furnaces. Thus, heat treatment time
is reduced, resulting in a good economy. Salt bath furnaces can be used for various
heat treatment operations like austempering, martempering, hardening, cyaniding,
carburizing, and tempering. Salt bath furnaces hold certain advantages over other
furnaces. Oxidation and decarburization can be avoided easily, as the sample is in
direct contact with the molten bath. Worn out electrode can be easily replaced while
the furnace is in operation. Complicated shapes with variable section thickness can
also be heated along with the simply shaped object. In the same way, light and heavy
objects can be handled in the same bath.
FURTHER READINGS
Singh, V. Heat Treatment of Metals. (Standard Publishers Distributors, New Delhi, 2006).
Rajan, T. V., Sharma, C. P., & Sharma, A. Heat Treatment: Principles and Techniques.
(PHI Learning Pvt. Ltd., New Delhi, 2011).
Taylor & Francis
10 Heat Treatment
Atmosphere
10.1 INTRODUCTION
Heat treatment furnace atmospheres have played an essential part in successful
heat treatment practices. The furnace atmosphere that protects the components
from oxidation or decarburization during the process of heat treatment and also
maintains the required surface properties of the components is called a controlled
atmosphere. The composition and distribution of the furnace atmosphere are
adjusted or controlled to achieve specific surface properties. Depending on the
type of heat treatment adopted, controlled atmospheres can be either protective or
chemically active.
A protective atmosphere protects the material from chemically reacting with
harmful elements, which may lead to oxidation or decarburization during heat
treatment. Examples of heat treatment practices where protective atmospheres are
used include bright annealing, normalizing of ferrous and nonferrous metals, and
annealing of malleable cast iron. In contrast, in the case of chemically active atmo-
spheres, the surface chemically reacts with the atmosphere to achieve improved
surface characteristics. Examples include carburizing and carbonitriding of steel
components, decarburization, nitriding, and hardening processes.
Several gases are used in heat treatment furnaces as controlled atmospheres,
including hydrogen, nitrogen, helium, argon, ammonia, propane, carbon monoxide,
carbon dioxide, methane, and butane. In some cases, a mixture of gases is used,
which decides the final properties of the controlled atmosphere. The inert gases,
particularly argon and nitrogen, are used as protective gases that prevent unwanted
chemical reactions like oxidation or decarburization from occurring on the metal
surface.
10.2 REACTIONS BETWEEN ATMOSPHERE AND MATERIAL

The reactions that occur between the controlled atmosphere containing gases and
the material are classified into three categories: the reaction between oxygen and the
material, the reaction between carbon and the material, and the reaction between
gases.
127
10.2.1 reacTIons BeTWeen oxygen and maTerIal

The reactions between oxygen and the material to be heat-treated form the basis for
protecting the materials from oxidation. Hence, understanding of these reactions is
very much essential. The reactions that control oxidation in furnaces, like in the case
of bright annealing of low-carbon steel, are as follows:
Material + H 2 O  Oxide + H 2 (10.1)
Material + CO 2  Oxide + CO (10.2)
The equilibrium constants for the above reactions can now be written as
pH2
K1 = (10.3)
pH2O
pCO
K2 = (10.4)
pCO2
Hence, the equilibrium constant at temperature T, K T , is given as
K1 pCO × pH2O
KT = = (10.5)
K2 pCO2 × pH2
The equilibrium constant K T is also related to temperature as follows:
∆G 0
log10 K T = − (10.6)
4.576T
where ∆G 0 is the free energy of the reaction. Therefore, it can be seen that the
oxidation/reduction potential depends not only on K1 and K 2 but also on the oxide
dissociation pressure of the particular material at the given temperature.
10.2.2 reacTIons BeTWeen carBon and maTerIal

Carburization and decarburization are the two basic heat treatment methods where
reactions between the carbon and the material are involved. While carbon monoxide
and methane are the gases used for carburizing, the gases responsible for decarbu-
rization are carbon dioxide, water vapor, and hydrogen. The reactions involved are
as follows:
2CO + 3Fe  CO 2 + Fe3 C (10.7)
CH 4 + 3Fe  2H 2 + Fe3 C (10.8)

Heat Treatment Atmosphere 129
The equilibrium constants for the above reactions can now be written as
pCO2
K3 = (10.9)
p2 CO
p2 H2
K4 = (10.10)
pCH4
p2 CO pCH
Therefore, at any given temperature, the or 2 4 ratios can be used to assess
pCO2 p H2
the carbon potential of the atmosphere. In the case of decarburization, water vapor
plays a vital role, which is evident from the following reaction:
H 2 O + Fe3 C  H 2 + CO + 3Fe (10.11)
Another reaction (equation 10.12) is very significant for nonferrous metals where the
atmosphere is contaminated with sulfur.
H 2 S + Metal → H 2 + Metal Sulphide (10.12)
10.2.3 reacTIons BeTWeen gases

Gases like oxygen, carbon, and hydrogen are often used to obtain equilibrium com-
positions at various temperatures. For example, the equilibrium is established by the
following reaction at temperatures above 800°C:
CO + H 2 O  CO 2 + H 2 (10.13)
It may be noted that carbon monoxide has a higher affinity for oxygen at lower tem-
peratures, whereas hydrogen has a higher affinity for oxygen at higher temperatures.
The affinities of both these gases toward oxygen are equal at about 850°C. Because
of the water–gas reaction (equation 10.13), the concentration of carbon monoxide is
reduced at low temperature wherever the atmosphere contains water vapor.
10.3 TYPES OF FURNACE ATMOSPHERES

10.3.1 exoTHermIc aTmospHere
Exothermic atmospheres are produced by the exothermic combustion of gases and
air and are low-cost prepared furnace atmosphere. Depending upon the air-to-gas
ratio, exothermic atmospheres are again subgrouped to lean (totally burnt) and
rich (flammable) types. Nominal composition in volume percentage for the rich
atmosphere is
N 2 = 71.5%; CO = 10.5%; H 2 = 12.5%; CH 4 = 0.5%

The lean atmosphere generally contains 0%–3% CO, 0%–4% H2, and rest N2. The
nominal composition in volume percentage is
N 2 = 86.8%; CO = 1.5%; H2 = 1.2%; CO 2 = 10.5%
It principally constitutes nitrogen and, therefore, is used as an inert atmosphere in

most of the heat treatment processes.
Rich atmospheres are commonly used for bright annealing, normalizing, and
tempering of steel. On the other hand, lean atmospheres are employed for the
heat treatment of nonferrous metals and alloys like bright annealing of copper.
Exothermic atmospheres cannot be used where decarburization is involved. It is
because these atmospheres have low carbon potential due to the presence of carbon
dioxide. Therefore, gases like water vapor and carbon dioxide are removed to pro-
duce exothermic atmospheres. Examples include pure nitrogen and nitrogen-based
atmospheres containing carbon monoxide and hydrogen up to 25% of the total.
10.3.2 endoTHermIc aTmospHere

Endothermic atmospheres are produced when a hydrocarbon-containing fuel reacts
with oxygen to oxidize the hydrocarbon to carbon monoxide and hydrogen. The reac-
tions involved are as follows:
2CH 4 + O 2 + 4N 2 = 2CO + 4H 2 + 4N 2 (10.14)
2C3 H8 + 3O 2 + 12N 2 = 6CO + 8H 2 + 12N 2 (10.15)
The endothermic gas atmosphere can be used in any furnace where reducing con-
ditions are required, but it is typically used as a carrier gas in gas carburizing and
carbonitriding applications. Endothermic generators find applications in heat treat-
ment and brazing of carbon steels, sintering, and as a base gas for gas carburizing
and carbonitriding of steels.
10.3.3 ammonIa-Based aTmospHere

The anhydrous liquid ammonia is dissociated into its constituents by heating in an
electrically heated or gas-fired chamber having one or more catalyst (iron at 560°C
or nickel at 900°C). The following reaction takes place during the dissociation:
2NH 3 → N 2 + 3H 2 (10.16)
The product of the above reaction is called cracked ammonia that comprises 75%
hydrogen and 25% nitrogen. Based on the grade of ammonia used, the dew point of
the cracked ammonia gas varies from −15°C to −40°C.
Burnt ammonia is produced by passing ammonia or air mixture over a catalyst at
850°C. The reaction that occurs during the burning is as follows:
4NH 3 + 3O 2 + 12N 2 → 14N 2 + 6H 2 O (10.17)

Heat Treatment Atmosphere 131
Burnt ammonia is generally used for bright annealing of strip and wire, whereas
cracked ammonia is used for bright annealing of stainless steels, sintering, and other
applications.
10.3.4 cHarcoal-Based aTmospHere

Such type of atmosphere is prepared by burning charcoal with air or flue gas. Air is
passed through a bed of hot charcoal, which burns the charcoal to N2, CO2, and H2O
at the bottom-most part of the bed. The highly heated charcoal then converts CO2 to
CO and H2O to H2 gas as the gas mixture moves up the bed. These atmospheres find
applications in annealing, normalizing, and hardening high-carbon steels without
scale or decarburization. Because of high operating costs and the inability to make it
automatic, these atmospheres are less used.
10.3.5 Hydrogen aTmospHere

Hydrogen is commercially available at the purity of 98%–99.9% with traces of water
vapor and oxygen. It is obtained by the electrolysis of water, or by the decomposition
of ammonia, or by the catalytic conversion of hydrocarbons, or by the water–gas
reaction. Hydrogen produced by the electrolysis of distilled water is best suited for
metallurgical purposes. Hydrogen is a powerful deoxidizer, but in the dry state, it
decarburizes high-carbon steels to form methane at high temperatures. Hydrogen,
when adsorbed, causes hydrogen embrittlement, especially in high-carbon steel. Dry
hydrogen is often used as an atmosphere in annealing of low-carbon steels, stainless
steels, direct reduction of metal ores, and sintering of metal powder components.
FURTHER READING
Taylor & Francis
11 Common Heat
Treatment Practices
11.1 INTRODUCTION
A metal or alloy must be tuned to the desired properties before use in several
engineering applications. Most of these properties depend upon the structure, and
suitable heat treatment processes can produce the preferred phases. According to
Metals Handbook, heat treatment may be defined as “a judicious combination of
both heating and cooling operations, including heating/cooling rate, heat treatment
temperature performed on metals or alloys in their solid-states in a synchronized way
that will yield desired properties.” The initial step is heating the material to some
temperature to form austenitic structure and then holding at that temperature for the
required time, followed by cooling down to room temperature. These transformed
products define the physical and mechanical properties of the material. Therefore,
looking at the importance of a detailed understanding of the heat treatment process,
this chapter mainly focuses on different heat treatment practices that are commonly
used for steels since steel is one of the most essential and versatile materials for
numerous engineering applications. Subsequent sections describe various heat treat-
ment methods for other metals and alloys.
11.2 TYPICAL HEAT TREATMENT PROCESSES

11.2.1 annealIng
In simple words, annealing treatment comprises subjecting the steel to a predefined
austenitic temperature and holding for some time depending on its thickness fol-
lowed by slow cooling, preferably furnace cooling. The treatment method depends
not only on the steel composition but also on other dimensional parameters such
as size, shape, and ultimately the final properties desired. The annealing treatment
serves numerous purposes; some of them are as follows:
• Relieve internal stresses generated due to casting, welding, or other mechan-

ical working operations
• Better machinability
• Achieve chemical uniformity
• Restore toughness of the metallic component
• Refinement of grain size
• Decrease the gaseous contents of steel
133
11.2.1.1 Full annealing
The term annealing, without any prefix or suffix, usually means full annealing. It
comprises heating the specimen to a high temperature, normally above its upper crit-
ical temperature, holding for a sufficient time so as to ensure homogenous austenite,
and finally, slow cooling that is usually performed in the furnace. Steel specimen for
this purpose is heated to a temperature, which is usually around 20°C–40°C higher
than the upper critical temperature (A3) in case of hypoeutectoid steels. Heating
at this temperature results in austenite with finer grains, which upon slow cooling
below A1 yields ferrite and pearlite of fine grains structure. On the other hand, if the
austenitizing temperature is about 100°C above A3, grain growth of austenite occurs,
and the coarse grains of austenite are not refined upon cooling, which in turn impairs
the properties of the steel. The previous discussions are illustrated in Figure 11.1.
Steel is heated to around 20°C–40°C above the lower critical temperature (A1) in
case of a hypereutectoid steels. Heating at this temperature results in fine grains of
austenite with partly or entirely spheroidized Fe3C, which on slow cooling through A1
produces fine grains of pearlite and spheroidized cementite. The reason behind such
spheroidization of cementite may be ascribed to the reduction in the interfacial area
between austenite and cementite, giving rise to a lower interfacial energy. Heating
slightly above Acm temperature results in a single-phase homogenous austenite.
FIGURE 11.1 Schematic illustration of microstructural changes on annealing of a typical

hypoeutectoid steel (0.2% carbon steel).
Common Heat Treatment Practices 135
FIGURE 11.2 Schematic illustration of microstructural changes on annealing of hypereu-

tectoid steel (1% carbon steel).
Such steel on slow cooling gets proeutectoid cementite formed at the grain bound-
aries of austenite as a thicker network, and pearlite has coarse grains, as shown in
Figure 11.2. Generally, heating to such temperature ranges for annealing is avoided
because of the following reasons:
• The cementite (which is very brittle) network facilitates brittle fracture by

providing path for crack propagation
• Grain coarsening of austenite occurs
• High energy consumption
• More time and less productivity
• More scaling and decarburization with no advantage
11.2.1.1.1 Purpose of Full Annealing

• Steel castings, in particular, have invariably coarse austenite grains,
which result in coarse ferrite or pearlite grains, commonly called as
“Widmanstatten” structures (Figure 11.3). These structures have feeble
impact strength and low toughness. Hence, annealing is performed in such
steel castings or hot worked steels to improve the mechanical properties by
refining the grain size.
FIGURE 11.3 Formation of Widmanstatten structure in steel castings.
• Annealing is also done to soften the steel. The slow cooling causes the
transformation of austenite at A1 or close to it to coarse lamellar pearlite,
which is very soft and, therefore, results in softening of steel with a reduc-
tion in hardness.
• Internal stresses of different types, also known as residual stresses, are
usually developed because of high strains associated with welding, cold
working, grinding, machining, case hardening, phase transformation, and
so on. Annealing is done to relieve these internal stresses. Slow cooling
during and after phase transformation induces no internal stresses.
• Steels with 0.3%–0.4% carbon have better machinability in the annealed
state. As annealing results in coarse pearlite and ferrite, which increase the
softness and ductility, it results in improved machinability.
• Annealing reduces some defects such as aligned sulfide inclusions or bands
in steels.
11.2.1.2 Homogenizing Annealing/Diffusion Annealing

Typically, heavy plain carbon steel castings, steel ingots, and high-alloy steel cast-
ings tend to have columnar grains, dendrites, and nonhomogeneous structures,
which render brittleness or, in other words, decrease the ductility and toughness of
steel. Homogenizing annealing is used to get rid of such nonuniform structures in
steels. In this process, the steel component is heated much above the upper critical
temperature and is soaked for prolonged periods at that temperature, followed by
slow cooling. As the diffusion of most of the alloying elements in steels is much
lower than carbon, the alloy steel ingots are usually homogenized at higher tempera-
tures of 1150°C–1200°C for 10–20 h, followed by slow cooling. The hypoeutectoid
inhomogeneous alloy steels are held at 1000°C for 1–2 h, whereas hypereutectoid
alloy steels are held for 5–6 h. Homogenization impairs the properties because it
coarsens the austenite grains. Thus, steels after such heat treatment undergo either
normalizing or full annealing.
Homogenization annealing also produces thick scales on the steel surfaces.
Scaling results in loss of metal. A controlled heating thus is a recommended method
where a precise temperature is fixed and the required soaking period is considerably
low. Higher temperatures, prolonged holding time, slow cooling rates in addition to
higher amount of scaling, and further heat treatment make this process highly eco-
nomically expensive.
11.2.1.3 Partial Annealing
It is also termed as intercritical annealing or incomplete annealing. It is comprised
of subjecting the steel to an intermediate temperature between the lower critical
temperature (A1) and upper critical temperature (A3 or Acm). Subsequent cooling has
to be carried out slowly till room temperature or may be in air from an intermediate
temperature such that all the austenite gets converted to pearlite.
11.2.1.4 Subcritical Annealing
It is a heat treatment operation in which the maximum heating temperature is always
below the lower critical temperature (A1). Here, no phase transformation is expected
to take place. This is only a thermally activated phenomenon, which facilitates pro-
cesses like recovery, recrystallization, grain growth, and agglomeration of carbides.
11.2.1.5 Recrystallization Annealing
Practically, all steels that are heavily cold-worked are processed by this annealing.
In this process, the steel component is heated just above its recrystallization tem-
perature followed by holding for homogenization and subsequent cooling. It results
in improved ductility at the expense of hardness or strength. The final microstruc-
ture now consists of fine equiaxed ferritic grains that are almost strain-free. The
primary target for performing a recrystallization annealing is to restore ductility
and toughness and refinement of coarse grains and to improve the electric and mag-
netic properties in case of grain-oriented steels. To initiate recrystallization, a crit-
ical amount of deformation causing formation of crystalline imperfections is the
driving force. These imperfections facilitate initiation of nucleation for formation of
new grains. As the amount of imperfections increase, the temperature required for
recrystallization tends to decrease. Recrystallization temperature is the temperature
at which a cold-worked metal gets 50% recrystallized in 1 h. On average, it is given
by the following equation:
tr = (0.3 − 0.5)Tmp (11.1)
where tr is the recrystallization temperature and Tmp is the melting temperature, both
in the Kelvin scale. There are some essential points to remember related to recrystal-
lization temperature, which is summarized in the following:
• Recrystallization temperature increases with the purity of the metal. Pure

metals always have low recrystallization temperature than impure metals.
• Zinc, tin, and lead possess recrystallization temperatures below the room
temperature. Hence, these metals cannot be cold-worked at room tempera-
ture since they recrystallize spontaneously, reforming a strain-free lattice
structure.
• Recrystallization temperature decreases with increasing annealing time.
• The finer the initial grain size, the lower the recrystallization temperature.
11.2.1.5.1 Hot and Cold Working of Metals

When a material is worked or deformed plastically above its recrystallization tem-
perature, the process is called as hot working. On the other hand, cold working
is the process of plastic deformation of material below the recrystallization. Let
us understand these two processes in line with the process of plastic deformation.
Two opposing phenomena cooccur when a material is plastically deformed at a high
temperature. One is the hardening phenomenon, which happens due to the plastic
deformation, and the other is the softening phenomenon, which happens due to the
recrystallization. At a specific temperature, these two cooccurring phenomena bal-
ance themselves, and the process of working the material above this specific tem-
perature is known as hot working, whereas cold working is nothing but working
below this temperature. Examples of hot working include working of lead and tin at
room temperature since their recrystallization temperature is below the room tem-
perature. However, working a metal like tungsten at 1200°C is still considered as
cold working, as its recrystallization temperature is higher than 1200°C.
11.2.1.6 Stress-Relieving Annealing
Internal stresses (residual stresses or locked-in stresses) are generated in the course
of different operations like welding, solidification, grinding, machining, shot peer-
ing, case hardening, and precipitation. Tensile stresses, particularly in surface layers,
are most dangerous to cause warpage or even cracks at low or without any external
stresses. When steel is heated to below A1 temperature to eliminate such type of
residual stresses, it is then called stress-relieving annealing. The primary aim is to
remove the harmful tensile stresses, allow higher external loads, increase fatigue life
and prevent intercrystalline corrosion, increase the impact resistance and lower the
susceptibility to brittle fracture, reduce the chance of cracking, prevent stress corro-
sion, and achieve dimensional stability.
11.2.1.7 Isothermal Annealing
Steel austenitized at temperatures of 20°C–40°C above A3 is cooled quickly to the
temperature of isothermal holding, which is below A1 temperature in the pearlitic
range. As a result, all the austenite is converted into pearlite. Although the target
of this isothermal annealing is more or less similar to that of full annealing, it pos-
sesses certain advantages over full annealing. The process of isothermal annealing
is cheaper than full annealing because of a shorter heat treatment cycle. As at a con-
stant temperature the transformation takes place, the microstructure obtained in case
of isothermal annealing is more uniform as compared to full annealing where trans-
formation occurs over a range of temperatures. The hardness of steel can be thus
controlled better in isothermal annealing. Isothermal annealing improves machin-
ability with an excellent surface finish.
11.2.1.8 Patenting
Patenting is an application of isothermal transformation, which consists of austen-
itizing steel in a continuous furnace to temperatures of 150°C–200°C above AC3
(usually 870°C–930°C) to get completely homogenous austenite. The steel is then
soaked for a fairly long period of time and then quenched in a bath (lead or salt
bath) maintained at a constant temperature. The typical temperature of the bath var-
ies from about 450°C to 550°C. For steels, the temperature of quenching medium
is chosen close to the nose of the time–temperature transformation (TTT) curve,
which results in the transformation of austenite to pearlite. Once the transformation
is complete, the steel is cooled down to room temperature by either keeping in air or
water spraying.
In some cases, a minor amount of upper bainite can be observed. Eutectoid
steel after patenting may have an interlamellar spacing of 40 nm of strength
1240–1450 MPa, which is cold drawn to the interlamellar spacing of 10 nm. Fine
interlamellar spacing easily blocks the motion of dislocations, increasing the strength
drastically with high toughness in twisting and bending. In most practical situations,
this patenting process is used for producing springs, high strength ropes, and piano
wires of usually 0.45%–1.0% carbon steel.
11.2.1.9 Spheroidization Annealing
It is the method of heating to a specific temperature, typically the austenitizing tem-
perature in most cases, and holding at that temperature followed by prolonged cool-
ing to produce spheroidal pearlite or carbides in globular forms in steels. Usually, the
austenitizing temperatures used for different types of steels are as follows:
Eutectoid steels – 750°C–760°C

Hypoeutectoid steels – 770°C–790°C
Hypereutectoid steels – 770°C–820°C
High-speed steels – 875°C
The main aim of obtaining spheroids structure is to obtain maximum softness, duc-
tility, and machinability with minimum hardness in the material. It is applied to
high-alloy tool steels and high-carbon steels to improve machinability and ductility.
Low-carbon steels may be spheroidized for cold forming, such as tubing. A micro-
structure of coarse spheroidized cementite (or alloy carbides) particles embedded in
a ferrite matrix is generally seen in this case.
Various types of annealing treatments are discussed, and each treatment has a
different heating profile depending upon the application areas. The temperature
ranges for these different annealing processes are illustrated in Figure 11.4 for better
understanding.
FIGURE 11.4 Different types of annealing treatments and their respective temperature
ranges.
11.2.2 normalIZIng
It is another heat treatment process, where the steel is taken to a temperature above
the upper critical temperature (A3 or Acm), soaked there for a proper time followed by
air cooling. The austenitizing temperature ranges for various steels are
It can be seen that the temperature for the normalizing treatment is higher than that
for annealing. It increases the homogeneity of austenite, thereby leading to a bet-
ter dispersion of ferrite and cementite resulting in enhanced mechanical properties.
Practically, this treatment is used to refine the coarse grains of steel castings and
forgings, which have been not worked at high temperatures. In particular, it elimi-
nates dendritic and nonuniform structures. Normalizing is also used to eliminate
or reduce microstructural irregularities. As normalizing reduces the grain size of
phases, the resultant microstructure has greater uniformity with excellent mechani-
cal properties. It reduces even bonding in carbon steels. In the case of hypereutectoid
steels, it helps break the coarse cementite network. The cementite network does not
appear in the normalized structure because of the less availability of time during
cooling. A schematic illustration of changes in the microstructures on normalizing
at different austenitizing temperatures is shown in Figure 11.5. There are specific
other necessary points to remember while conducting the normalizing heat treat-
ment process, which is summarized in the following:
• Austenite formed during normalizing treatment should be homogenous

to have excellent stability so that it could be supercooled to temperatures
much below A1 by air cooling before it transforms into fine pearlite.
• Lower is the temperature of transformation; finer is the grain size of ferrite
and pearlite with smaller interlamellar spacing.
11.2.3 HardenIng
As the title suggests, this treatment is used often to harden the material. In this treat-
ment, the steel component is subjected to an appropriate austenitizing temperature,
soaked there to develop homogenous and fine-grained austenite, and then followed
by cooling at a faster rate than its critical cooling rates to suppress pearlite and
bainitic transformations. Hence, austenite transforms into martensite, and the steel
becomes hard; such cooling is called quenching. Usually, carbon steels are quenched
in water, and alloy steels in oil. The hardness of the martensite depends mainly on
the carbon content of the steel. The critical cooling rate is defined as the slowest
cooling rate, which produces a full martensite structure, i.e., the cooling rate touch-
ing the nose of the “S” curve. In real cases, this treatment is often used to develop
high hardness in tool steels so that it can cut other metals. The cutting ability of tool
steel is proportional to its hardness. Besides, many machine parts such as gears,
FIGURE 11.5 Schematic illustration of changes in the microstructure on normalizing.

(a) Higher austenitizing temperature results in coarse austenite grains (more homogeneous).
(b) Austenitizing in normalizing range results in slight coarser grains than that obtained in
annealing range. (c) Normalizing results in Widmanstatten structure. (d) Finer ferrite and
pearlite is obtained by normalizing at a lower temperature.
shafts, cams, and bearings are hardened to induce high wear resistance. Another
essential objective of hardening is to develop high yield strength in machine compo-
nents made of structural steels with excellent toughness and ductility to bear higher
working stresses. The more the yield strength, the higher the magnitude of stress,
which a part can carry during service.
11.2.3.1 Hardening Temperature for Different Types of Steels

The hardening temperature depends on the chemical constituents of different kinds
of steel and is depicted in Figure 11.6. For example, hypoeutectoid steels are heated
to about 20°C–40°C more than the critical temperature (A3), whereas eutectoid and
hypereutectoid steels are subjected to about 20°C–40°C more than the lower critical
temperature (A1). Hypoeutectoid and eutectoid steels, when subjected to the tem-
perature range mentioned earlier, convert to homogenous and fine-grained austenite,
which transforms into martensite on rapid quenching from the hardening temperature
(Figure 11.7). The presence of martensite increases the hardness of the steel.
For hypoeutectoid steel if the hardening temperature is chosen in the same way
as that of hypereutectoid steel (i.e., in between A1 and A3), a microstructure com-
prised of ferrite and austenite is formed. It transforms to ferrite and martensite on
FIGURE 11.6 Hardening temperature ranges for different types of carbon steels.
A, austenite; F, ferrite; P, pearlite.
FIGURE 11.7 Critical cooling rate for martensite formation in eutectoid steel. A, austenite;
M; martensite.
quenching. Ferrite, a very soft phase, lowers the hardness of hardened steel consider-
ably. This phenomenon is known as incomplete hardening and is not usually used
in practice.
Hypereutectoid steels, when heated at a temperature slightly above A1, consist
of fine grains of proeutectoid cementite and austenite. Upon quenching, this aus-
tenite changes to martensite, whereas the cementite remains unchanged. Hence,
cementite being harder than martensite, its presence improves the hardness, abra-
sion resistance, and wear resistance as compared to the only martensite structure.
Now, if the hardening or austenitizing temperature is increased and is just below
the Acm temperature, a portion of the proeutectoid cementite in the austenite gets
dissolved to increase the carbon content of austenite. This excessively high tem-
perature also tends to form undesirable coarse austenitic grain size. On quench-
ing, the hardness of the as-quenched material is less because of the following
reasons:
• A lesser amount of proeutectoid cementite is present.

• A more considerable amount of soft retained austenite is produced
since the dissolved carbon of cementite has lowered both the Ms and Mf
temperatures.
• The coarser martensite has a lesser hardness.
Furthermore, heating hypereutectoid steels above Acm results in the austenite of

coarser grains and surface decarburization. The as-quenched hardness is again low
in this case because of the following reasons:
• Lack of harder cementite.

• A higher amount of dissolved carbon in the austenite results in a higher
retained austenite content.
• Coarse acicular martensite has poor mechanical properties.
• Decarburized surface responds poorly to hardening treatment.
• Very high magnitude of internal stresses is introduced into the hardened
structure as a result of quenching from a high temperature.
In the austenitic type of alloy steels, the room temperature structure may consist
of austenite and some precipitate alloy carbides. These steels have Ms temperatures
below the room temperature, and thus, heating these steels to temperatures higher
than Acm helps to dissolve the alloy carbides completely to obtain homogenous
austenite.
In carbide type of alloy steels, heating is done at higher temperatures, i.e., between
A1 and Acm to dissolve as much of alloy carbides as possible, leaving some carbides
(such as vanadium carbide) in the undissolved state to inhibit the grain growth. The
dissolved carbides are made to precipitate as fine and uniformly dispersed carbides
during tempering, which induces in such steels the important characteristic called
red hardness or secondary hardening. More is the volume of fine carbide precipi-
tates; better is the red hardness. It makes these tool steels to cut or machine materials
at high speeds.
11.2.4 quencHIng
Quenching is the process that determines the cooling rate to decide the properties
of the hardened material like structure, hardness, and strength. Quenching may be
done by air, water, brine, oils, polymer quenchant, and salts or with the help of gases,
liquids, and solids. Usually, a quenching medium is selected that provides a faster
cooling rate than the critical cooling rate. Quenching medium characteristics are as
follows:
• Quenchant temperature
• Specific heat capacity of the quenchant
• The viscosity of the quenchant
• Thermal conductivity
• Latent heat of vaporization
• Extent of agitation of the quenching medium
When a heated (say at 840°C) steel specimen is quenched in a stationary bath of cold
water, a typical cooling curve forms with three stages (Figure 11.8).
• Vapor blanket stage (stage A)

The surface of the hot metal gets vaporized as soon as the metal comes
in contact with the quenching medium because of the high temperature of
the metal. Hence, a thin stable film of vapor forms around the hot metal
(Figure 11.9a). Since vapor films are poor conductors of heats, the cool-
ing rate is relatively slow. At this stage, cooling is done by conduction and
radiation through the vapor film.
• Liquid boiling stage (stage B)
In this stage, heat is removed very rapidly in the form of heat of vapor-
ization, and so the vapor blanket is broken, as shown in Figure 11.9b. After
that, the quenchant comes directly in contact with hot metal surface, and
violent boiling occurs. Very rapid cooling, mostly by convection, occurs in
this stage as the quenching medium always remains in contact with the steel
surface. This stage of quenching corresponds to the nose of the continuous
cooling temperature (CCT) curve of the steel, i.e., 500°C–100°C.
FIGURE 11.8 Cooling curve of a heated steel object when quenched in cold water.
FIGURE 11.9 Schematic illustration of different stages of cooling: (a) vapor blanket stage,
(b) liquid boiling stage, and (c) direct contact stage.
• Direct contact stage (stage C)

It is also known as the liquid cooling stage or convection stage. It begins
when the temperature of the metal surface is below the boiling point of
the quenching liquid (Figure 11.9c). Cooling in this stage occurs by both
conduction and convection processes through the liquid. The cooling rate is
slowest in this stage as vapors do not form.
11.2.4.1 Quenching Media
Given the discussions about heat removal, which includes three stages, there are
practically no such quenching media that can follow these three stages ideally. Let
us discuss some of the industrially accepted quenching media in detail. Before we
begin, the industrial quenching media in order of decreasing quenching severity are
listed in the following:
• Brine
• Water
• Fused or liquid salts
• Soluble oil and water solutions
• Oil
• Air
11.2.4.1.1 Brine
Aqueous solutions of salts like 10% sodium chloride or calcium chloride are denoted
as brine solutions. In brine, due to the heating of the quenching medium in contact
with the metal surface, crystals of the salt tend to get deposited on the metal surface.
The layer of solid crystals so formed disrupts with minor explosive violence and
chucks off a cloud of crystals. The vapor film gets destroyed, which allows direct
contact of the brine with the metal surface. Brines are used where faster cooling
rates are required.
11.2.4.1.2 Water
It is the most popular quenching medium because of its low cost, availability in
abundance, secure handling, no pollution problems, and maximum cooling rate.
Water quenching is only applicable for plain carbon steels and some low-alloy steels.
Water quenching often results in a cooling rate, which is greater than the critical
cooling rate. Such a high cooling rate usually leads to the creation of large amount of
internal stresses in the quenched product, leading to distortion or even to the advance
of cracks. Long-term stability of the vapor blanket stage is another noteworthy draw-
back of water quenching. Agitation of the water bath during quenching may help in
reduction or elimination of this stage.
11.2.4.1.3 Salt Bath
Salt bath is especially used as a quenchant for tool steels. It is the best-suited quench-
ing medium for steel with good hardenability and thin sections. Its advantages are a
uniform temperature throughout the bath, selective hardening, no danger of oxida-
tion, carburization, or decarburization during cooling. Some common salt baths are
NaNO3, 5% NaNO3 + 50% KNO3, and 50% NaNO3 + 50% KNO2.
11.2.4.1.4 Oil
Oils have high boiling points, lower quenching power than water or brine, lower sta-
bility, and high cost. Mineral oil is the most common oil used as quenchant.
11.2.4.2 As-Quenched Structures
At first, martensite forms as soon as the quenching starts. The depth of martensite
formation depends on the time the part remains in the quenching bath. In quenched
steel, the quantity of martensite formed depends on the location of Ms, Mf, and
the temperature of the quenching medium, which usually is the room temperature
(Figure 11.10). As long as the room temperature lies between Ms and Mf tempera-
tures, austenite does not change completely to martensite since it has not been cooled
to below the Mf temperature. The untransformed austenite is called retained austen-
ite. The amount of retained austenite rises with the increase of carbon content of
steel, as shown in Figure 11.11.
The substructure of the retained austenite consists of a higher density of disloca-
tions and stacking faults.
FIGURE 11.10 Influence of carbon on the Ms and Mf temperatures.
FIGURE 11.11 Influence of carbon on the amount of retained austenite.
11.2.5 TemperIng
Tempering consists of heating the hardened steel to a temperature below the A1 tem-
perature, holding at that temperature, and then followed by prolonged cooling. The
aim of tempering is to relieve quenching stresses developed during hardening and
increase ductility and toughness of the steel by sacrificing the hardness or strength.
11.2.5.1 Stages of Tempering

Based on the tempering temperature, the treatment proceeds through four distinct
but then again overlapping stages.
11.2.5.1.1 First Stage of Tampering (up to 200°C)

During the early stage of tempering, precipitation of a hexagonal close-packed tran-
sition carbide (epsilon carbide, Fe2.4C) occurs due to a decrease in the tetragonality
of martensite. The formation of epsilon carbide decreases the c/a ratio as carbon
precipitates from martensite.
Steels having carbon up to 0.2%, as a rule, are not hardened and tempered.
However, martensite in such steels has a BCC structure, which does not show
any change in the first stage of tempering except carbon atoms that segregate
to dislocations. Steels having more than 0.2% carbon are highly unstable due to
supersaturation. So, in this case, tempering results in the formation of low-carbon
martensite and epsilon carbide by the transformation of high-carbon martensite. The
contraction in volume occurs due to the rejection of carbon. Decrease of tetragonal-
ity decreases the hardness, but precipitation of epsilon carbide increases the hard-
ness of the steel proportional to its amount formed. Thus, the net effect is that the
hardness of steel usually decreases continuously but only slightly. In high-carbon
steels (1.2% C steel), a slight increase in hardness is observed due to a relatively large
volume of epsilon carbide.
11.2.5.1.2 Second Stage of Tempering (200°C–300°C)

During this stage, retained austenite transforms to lower bainite. This lower
bainite consists of ferrite and epsilon carbide and, therefore, is different from
the conventional bainite. A slight increase in the volume of steel occurs. When
the carbon content of the steel is high, the amount of retained austenite being
large transforms to more hard lower bainite in vast proportions. Ductility and
toughness increase slightly by this treatment, with a corresponding decrease in
hardness and strength.
11.2.5.1.3 Third Stage of Tempering (200°C–350°C)

In the third stage, the epsilon carbides dissolve, and the low-carbon martensite
loses both its carbon and tetragonality and becomes ferrite. Carbon thus released
combines with epsilon carbide, which in turn transforms to rod-shaped cementite.
All these changes occur with the help of diffusion and nucleation. The contraction of
volume occurs at this stage. The hardness decreases continuously and sharply with
a rapid increase in toughness.
11.2.5.1.4 Fourth Stage of Tempering (350°C–700°C)

Heating the hardened steel in this range of temperature results in coarsening or
spheroidization of cementite along with recovery and recrystallization of ferrite.
The growth of cementite starts at around 300°C, but spheroidization occurs above
400°C. Reduction in surface energy provides the necessary condition for spheroidi-
zation of cementite. Above 600°C, equiaxed grains of ferrite (by recrystallization)
form have coarse globules of cementite. It is the spheroidized or globular pearlite,
which is softest with the highest ductility and best machinability.
11.2.5.1.5 Secondary Hardening and the Fifth Stage of Tempering

When alloying elements are added to steels, they may enter the ferrite or the carbides in
varying amounts depending on the alloying element concerned. Non–carbide-forming
elements, such as Al, Cu, Si, P, Ni, and Zr, enter into ferrite. They have minimal impact
on the tempered hardness of steel. Other elements are found in both the ferrite and the
carbides. A number of these elements in the order of their tendency to form carbides
(Mn having the least and Ti the greatest) are Mn, Cr, W, Mo, V, and Ti. Most alloy-
ing elements in steels tend to increase the resistance of the steel to softening when it
is heated, which means that for a given time and temperature of tempering, alloy steel
will possess a higher hardness after tempering than plain carbon steel of the same car-
bon content. For example, there is an increase in hardness with the addition of 5% Mo
in a 0.35% carbon steel (Figure 11.12). This effect is especially significant in steels that
contain considerable amounts of carbide-forming elements. When the carbide-forming
elements are tempered below 540°C, the tempering reactions tend to form cementite
particles based on Fe3C, or more accurately (Fe, M)3C where M represents any of the
substitutional atoms in the steel. In general, the alloying elements are present in the
cementite particles only in about the same ratio as they are present in the steel as a
whole. When the tempering temperatures exceed 540°C, appreciable amounts of alloy
carbides are precipitated, which replace the coarse cementite particles. The precipita-
tion of alloy carbides is identified as the fifth stage of tempering. The precipitation of
these new carbides, in general, does not conform to the formula (Fe, M)3C.
11.2.5.1.6 Secondary Hardening
When the amount of strong carbide-forming elements is large, and the highly alloyed
steels are tempered at 500°C–600°C, the general decrease of hardness with temper-
ing temperature is not only arrested, but an increase in hardness with improved
toughness occurs. This increase in hardness is called secondary hardness or red
hardness. Above 500°C, these elements have high diffusivity to nucleate and grow
FIGURE 11.12 Secondary hardening caused by the presence of a 5% molybdenum in plain

carbon steel.
to form fine dispersion of alloy carbides to cause secondary hardening. Because of

the lower diffusion rates of alloying elements, these alloy carbides are formed only
at the high temperatures. Besides, at low temperatures, diffusion of solute atoms is
sluggish to result in formation of any carbides.
Secondary hardening is a process in which coarse cementite particles are replaced
by new and much finer alloy carbide dispersion of V4C3, Mo2C, and W2C. The critical
dispersion causes a peak in the hardness, but as carbide dispersion slowly coarsens,
the hardness decreases. The peak is realized at 500°C in chromium steels, 550°C in
molybdenum steels, 550°C–600°C in vanadium steels, and 600°C in titanium steels.
The amount of secondary hardening is directly proportional to the volume fraction
of alloy carbides and, hence, is related to the concentration of the strong carbide-
forming elements present in the steel.
11.2.6 ausTemperIng and marTemperIng

Austempering is a hardening treatment in which austenite transforms isothermally
to lower bainite and is used to reduce distortion and cracks in high-carbon steels. It
involves heating the steel to above the austenitizing temperature, quenching in mol-
ten salt bath held at a particular temperature above Ms point and within the bainitic
range (300°C–400°C), and then keeping the steel at this temperature to let austenite
transform completely to lower bainite. After full transformation, steel is removed
out of the bath and cooled in the air up to the room temperature. A schematic illus-
tration of the austempering process superimposed on the TTT diagram is shown in
Figure 11.13. The equalization of temperature throughout the cross section of the
part before bainite formation minimizes the stresses developed during austempering,
FIGURE 11.13 Schematic illustration of the austempering process.

which are negligible compared to stresses developed during hardening. Steel to be

austempered should have adequate hardenability to avoid pearlite formation. TTT
diagram is also useful for determining the suitability of given steel for austemper-
ing. The rate of cooling should be higher than the upper critical cooling rate, i.e.,
care must be taken such that there is no austenite to pearlite transformation. Steels
having considerable soaking time for the bainite transformation are not suitable for
this treatment.
Martempering is also a hardening treatment used to minimize distortion and
cracking. In regular hardening operation, when the heated steel is plunged in cool-
ant, a vast difference of temperature develops between the surface and the center of
the part, which introduces two types of stresses:
11.2.6.1 Thermal Stresses
A significant difference in temperature results in differential volume contraction,
which produces volumetric internal stresses called thermal stresses.
11.2.6.2 Structural Stresses
Due to a significant difference in temperature between the surface and the center, the
martensite transformation range is attained at a different time from the surface to the
center. This differential expansion causes internal stresses called structural stresses.
When both these types of stresses develop simultaneously, the nature of stress
becomes quite complicated at places, and therefore, the magnitude of stresses may
become large enough to cause distortion and even cracking. If the magnitude of
the tensile stress exceeds the yield point of the material, the steel distorts, but if it
exceeds the tensile strength, the steel develops cracks.
Martempering treatment involves heating the steel to the austenitizing tempera-
ture followed by quenching in a constant temperature bath maintained above Ms
point, typically between 180°C and 250°C. Steel is then held in the bath until the
temperature throughout the cross section (from the surface to the center) becomes
uniform and identical to the bath temperature. After that, steel is withdrawn and
cooled in air through Ms to room temperature to obtain martensite simultane-
ously across the whole section. Finally, tempering is done as required for the part.
Figure 11.14 explains the martempering process superimposed on the TTT diagram.
The cooling rate should be substantially high, and the soaking period quite short
to restrict the conversion of austenite to pearlite or even bainite. In martempering,
as the bath temperature is not the room temperature but above Ms temperature, the
steel temperature gradient is reduced, and so, the magnitude of the thermal stress is
reduced. TTT is useful to fix the correct bath temperature, to fix the isothermal hold-
ing time without bainite formation, and to find out the maximum size of the part of
that steel, which can be martempered.
11.2.7 suBZero TreaTmenT

Subzero treatment is mainly done for hardened steels that contain retained austen-
ite. As discussed in the quenching section, retained austenite drastically reduces
the mechanical properties and leads to unpredictable dimensional changes, and
FIGURE 11.14 Schematic illustration of the martempering process.
therefore, treatment is required to convert them to martensite, which is done by the

subzero treatment. The process involves cooling the steel to the subzero temperature
that is lower than the Mf temperature of the steel. Usually, Mf temperatures for steels
lie between −30°C and −70°C. This treatment is followed by tempering to remove
the internal stresses developed in the steel. The subzero treatment is not useful for
such steels that are slightly heated to 100°C–200°C or kept at room temperature for
several hours after hardening because austenite becomes stable, and then it does
not transform to martensite completely. Hence, subzero treatment should always
be executed immediately after completion of hardening treatment. Subzero treat-
ment is mostly preferred in high-carbon and high-alloy steels, making them suitable
for bearings, gauges, tools, and components that require high impact and fatigue
strength in addition to dimensional stability.
11.3 HARDENABILITY
11.3.1 sIgnIfIcance
Steels are typically graded based on chemical composition. However, the compo-
sition is not a fixed number; instead, it is a range. For example, AISI 4340 steel
has 0.38%–0.43% carbon, 0.60%–0.80% manganese, 0.20%–0.35% silicon,
1.65%–2.00% nickel, 0.70%–0.90% chromium, and 0.20%–0.30% molybdenum.
This variation in the chemical composition causes a difference in the critical cooling
rate during quenching. Hence, the rates of cooling at the surface and center are not
the same, which in turn results in a gradient of hardness across the cross section.
Purchasing steel based on chemical composition will not guarantee the full hard-
ness, and therefore, it is necessary to know the hardenability of the steel.
Hardenability may be defined as the susceptibility of the steel to hardening

when quenched and is related to the depth and distribution of hardness across a
cross section. It depends on the addition of alloying elements and the grain size
of austenite. If rapid cooling is performed, martensitic transformation may even
be feasible at the center of the steel if the corresponding cooling rate exceeds
the critical one. In this way, a completely hardened steel would be obtained
with martensite throughout the cross section. However, such rapid quenching
may further be associated with unwanted effects like warping and ultimately
cracking of the specimen. Therefore, the hardenability of steel is revealed by
its ability to harden throughout its cross section while avoiding radical quench-
ing. The regions of martensite and pearlite can be easily differentiated in the
steel section in many ways. The presence of pearlite results in a sudden fall
of hardness. The fracture surface of the sample, as schematically illustrated in
Figure 11.15, is another way of identifying the martensite or pearlite content.
While martensite breaks with brittle fracture, pearlite breaks with a ductile frac-
ture. Microstructurally, steel contains 50% pearlite and 50% martensite at the
brittle-to-ductile transition region.
FIGURE 11.15 Variation of hardness across the cross section of a quenched cylinder.
Microstructural examination reveals that the fracture surface has two distinct appearances,
i.e., ductile and brittle fracture.
11.3.2 deTermInaTIon of HardenaBIlITy

The following methods determine the hardenability of steel:
• Grossman’s critical diameter

• Jominy end quench test
11.3.2.1 Grossman’s Critical Diameter

The test is named after M.A. Grossman, who measured hardenability in terms of
critical diameter. In this method, many steel rods of different diameters are quenched
under the same conditions (Figure 11.16). While the rod with the smallest diameter
gets hardened throughout the cross section, the hardness decreases in the center
with an increase in the diameter of the rod. As the diameter of the rod increases, the
cooling rate at the center of the steel rod decreases, thereby giving a softcore con-
taining pearlite. The position of the steepest drop in hardness, or when the rod has a
50% martensitic structure at the center, is called the critical diameter (Dc). Its value
depends on the steel composition and the quenching method. For example, the value
of critical diameter is 1.83″ for water and 1.25″ for oil. It is always essential to spec-
ify the severity of the quench when hardenability is to be expressed in terms of Dc.
11.3.2.2 Severity of Quench

Grossman defined an ideal quenching medium as a medium that removes the heat
from the steel surface as soon as the heat flows from the center of the rod to the
surface. The critical diameter of the specimen corresponding to this ideal quenchant
is termed as the ideal critical diameter (DI). On the other hand, the severity of the
quench is designated by the heat transfer equivalent H and is given by
S
H= (11.2)
K
FIGURE 11.16 Effect of the diameter of the specimen on the depth or hardness while
quenching in the same medium.
FIGURE 11.17 Grossman’s master graph, which shows the relationship between the critical
diameter (Dc), ideal critical diameter (DI), and the severity of quench (H).
where S is the heat transfer coefficient between the steel and the quenching medium,
and K is the thermal conductivity of the steel. The relationship between the critical
diameter (Dc), ideal critical diameter (DI), and the severity of quench (H) can be
obtained from thermodynamic considerations, as shown in Grossman’s master graph
(Figure 11.17). From the figure, for Dc = 1″ and H = 5 (agitated brine quench), DI or
hardenability comes out to be 1.2″.
11.3.2.3 Jominy End Quench Test

The Grossman method of determining the hardenability of steel is not so popular as
it takes a long time to quench and test a large number of steel rods. A comparatively
faster and popular method of hardenability determination of steel is the Jominy end
quench test. In this test, a round specimen having a diameter of 1″ and a length of
4″ is heated uniformly to the proper austenitizing temperature (Figure 11.18a). After
that, the sample is quickly placed in a fixture where a jet of water impinges the bot-
tom part of the specimen. Specific parameters such as the size of the orifice, the
distance from the orifice to the bottom of the specimen, and the temperature and
circulation of the water are standardized so that every specimen in this fixture can
receive the same rate of cooling. After cooling, the specimen is removed from the
fixture, and two parallel flat surfaces are ground longitudinally to a depth of 0.015″.
The hardness is obtained at intervals of 1/16″ from the quenched end. A curve is
then drawn between the hardness values and the distance from the quenched end
FIGURE 11.18 Schematic illustration of the Jominy end quench test: (a) experimental setup
and (b) a typical Jominy curve showing the hardness variations in the quenched bar.
(Figure 11.18b). In general, ASTM Designation A225-48T End Quench Test is

followed for determining the hardenability of steel.
11.3.3 facTors affecTIng HardenaBIlITy

Hardenability is a measure of the amount of martensite present in the structure.
Hence, steel is said to have high hardenability if it is completely transformed to mar-
tensite. The factors that promote the formation of martensite are those who shift the
nose of the TTT curve to the right. Alloying elements, carbon content, and austenitic
grain size affect hardenability. All alloying elements, except cobalt, have a tendency
to shift the nose of the TTT curve to the right. This ensures a more hardenable
steel as the probability that the cooling curve will not touch the nose of CCT curve
is higher. However, undissolved inclusions, such as carbides or nitrides, nonmetal-
lic inclusions, and inhomogeneity of austenite, which may happen due to the pres-
ence of alloying elements, decrease the hardenability of the steel. Carbon content
dramatically influences the hardenability. The size of austenite also plays a vital role
in determining the hardenability. The finer the austenite grain size, the lower the
hardenability. It may be attributed to the fact that with a decrease in grain size, more
sites become available for pearlite nucleation. Once pearlite formation occurs, mar-
tensite transformation is suppressed, and hence, hardenability decreases. Although
coarse-grained steel increases the hardenability, it is not always recommended to use
the same to increase hardenability because it may lead to specific other undesirable
properties such as poor impact properties, decrease in ductility, and quench crack
susceptibility.
11.4 CASE HARDENING AND SURFACE HARDENING

A large number of machine components like gears require a combination of various
properties such as high surface hardness and wear resistance along with excellent
toughness and impact resistance. High-carbon steels can provide high surface hard-
ness and wear resistance after suitable heat treatment, but impact strength is poor.
On the other hand, low-carbon steels provide high impact strengths but very low sur-
face hardness and wear resistance. Surface hardening and case hardening methods
are commonly used to impart this combination of properties. While case hardening
changes the chemical composition of the surface layers, surface hardening does not
alter the chemical composition of the steel but includes phase transformation by fast
heating and cooling of the outer surface. Conventional case hardening methods are
carburizing, nitriding, and cyaniding or carbonitriding. Surface hardening processes
are flame hardening, induction hardening, laser hardening, and electron beam hard-
ening. Let us discuss some of these methods in detail.
11.4.1 carBurIZIng
It is a heat treatment operation in which the carbon content of the surface of low-
carbon steel, usually about 0.20% carbon or lower, or low-carbon-containing alloy
steel is enriched by exposing the steel to a carburizing atmosphere in the austenitic
region followed by quenching and subsequent tempering. Generally, carburizing is
carried out by keeping the low-carbon steel in contact with solid, liquid, or gaseous
atmosphere of high-carbon activity at high temperatures (850°C–950°C) to obtain
surface carbon content of 0.6%–1.10%. Even with higher surface carbon content, the
case becomes very brittle due to the coarse cementite network formed in it. So the
steel is subjected to quenching. As-quenched structures formed are martensite and
retained austenite. However, the surface is still brittle. Therefore, the steel is subse-
quently subjected to tempering.
11.4.1.1 Mechanism of Carburizing

Carburizing occurs in two steps:
• Absorption of free carbon on the steel surface occurs due to a large differ-
ence between the carbon potential of the atmosphere and the carbon content
of the steel surface. The absorption rate of the additional carbon content on
FIGURE 11.19 Variation of carbon concentration with distance.
the surface increases with an increase in the surface carbon content until it
becomes equal to that of the atmosphere.
CO 2 (g) + C(s)  2CO(g) (11.3)
• The diffusion of carbon from the surface to the interior occurs in the second
step. As soon as the steel comes in contact with the carburizing atmosphere,
the carbon on the surface attains its maximum value (Cs). It remains con-
stant throughout the process. The carbon content of the interior is the same
as the original carbon content in the steel, Co. At an instant t > 0, the carbon
concentration varies linearly with distance (Figure 11.19). Fick’s first law of
diffusion then governs the diffusion of carbon into the steel.
11.4.1.2 Fick’s First Law of Diffusion

Fick’s first law of diffusion is given by
dc C − Co
J = − Dγc = Dγc s (11.4)
dx dx
where J is the flux of diffusing species. Here, J is the net carbon flux per unit area
dc
per second. Dγc is the diffusivity of carbon in austenite, and is the concentration
dx
gradient. The negative sign indicates that the concentration of carbon decreases with
distance. Let us assume that the case depth increases from x to x + dx on increasing
the duration from t to t + dt. In the time interval dt, the carbon supplied by diffusion
is Jdt, which is shown by the shaded area in Figure 11.19. So,
(Cs − Co ) ( x + dx ) (Cs − Co ) ( x )
Jdt = − (11.5)
2 2
On combining equations 11.4 and 11.5, we have
( Cs − Co ) (Cs − Co ) ( x + dx ) − (Cs − Co ) ( x )
Dγc dt = (11.6)
x 2
Dγc
(Cs − Co ) (C − Co ) (dx )
dt = s (11.7)
x 2
2 Dγc (Cs − Co ) dt = (Cs − Co ) x (dx ) (11.8)
xdx = 2 Dγc dt (11.9)
On integrating equation 11.9, we have
x2
= 2 Dγc t (11.10)
2
x = ∅√t (11.11)
where x is total case depth in milimeters, t is the time in hours, and ∅ parameter is
a function of temperature, as illustrated in Figure 11.20.
11.4.1.3 Heat Treatment after Carburizing (Postcarburizing Treatment)

Carburized components are given heat treatment to develop hard and wear-resistant
surfaces, to refine the grain size of the case as well as core, and to break the cementite
network to reduce brittleness. Typically, a double heat treatment cycle is performed
in case the steel is having a coarse-grained structure throughout its cross section.
The steel is first subjected to above A3 temperature to refine the grain size of the
core and dissolve the proeutectoid cementite of the case. Normalizing is then done,
which does not allow the network of cementite to form. It is then reheated to above
AC1 to refine the grain size of the case, and quenching produces fine martensite in
FIGURE 11.20 Variation of ∅ parameter with temperature.

FIGURE 11.21 Schematic representation of postcarburizing heat treatment cycles.
the case. Tempering is done at 180°C–200°C to relieve the stresses and induce some
toughness. A schematic representation of this treatment is provided in Figure 11.21.
11.4.2 nITrIdIng
Nitriding is another case hardening technique where nitrogen enrichment of the
surface of the steel specimen is done in the temperature range of 500°C–575°C
(when the steel is ferritic) in an atmosphere of 15%–30% dissociated ammonia for
an extended time. Nitriding does not require heating of the steel into austenitic field
region and subsequent quenching to form martensite. So, it involves minimum dis-
tortion and provides excellent dimensional control. Steels subjected to nitriding are
medium carbon steels. Main reasons for the high surface hardness of nitrided steels
are as follows:
• Fine (size 5–15 nm) and uniformly dispersed particles act as strong barriers
to block the motions of dislocations.
• There is a high density of dislocations of 1010 cm−2 as in a heavily cold-
worked metal, which increases the hardness.
• Nitrogen atoms form Cottrell-type atmospheres to increase the hardness.
11.4.2.1 Gas Nitriding
Gas nitriding is the most common method of nitriding, where the components to
be nitrided are heated in an atmosphere of dissociated ammonia gas. At nitriding
temperatures, 500°C–575°C, ammonia reacts at the surface of the steel to introduce
nitrogen in iron, i.e.;
3
NH 3 = N(Fe) + H2 (11.12)
2
with a constant equilibrium K given by

3
a N pH22
K1 = (11.13)
pNH3
K1 pNH3
%N = 3
(11.14)
pH22
%N refers to nitrogen dissolved in iron at a given temperature. During nitriding, %N

at the surface in solution is given by equation 11.14, only when it does not exceed
the solubility limit of nitrogen in steel at the nitriding temperature. Ammonia gas is
itself unstable at nitriding temperatures and dissociates according to the reaction, as
given in the following:
1 3
NH 3 = N2 + H2 (11.15)
2 2
This reaction is relatively slow. Molecular nitrogen present in the gas atmosphere
may also decompose at the surface of the steel to introduce nitrogen in steel, i.e.,
1
N 2 = N(Fe) (11.16)
2
1
%N = K 2 pN22 (11.17)
However, at nitriding temperatures, the constant K 2 is orders of magnitude smaller

than K1 in equation 11.13. NH3–H2 mixture containing 18% NH3 and 82% H2 (total
pressure of 1 atm) gives the same percentage nitrogen dissolved in iron as 5 × 103 atm
of N2 at 500°C. Hence, it is not practical to use nitrogen gas for nitriding; instead,
we go for NH3 + N2 + H2 mixture. The atomic nitrogen in equation 11.12, which
is absorbed by the steel surface, can form nitrides following the phase diagram of
the iron–nitrogen system. The first phase is α ss − N, which is called nitrous ferrite.
Another vital phase that forms is α ′′ (low concentration of N). It is a fine and coher-
ent precipitate (Fe16N2). In case the level of nitrogen exceeds 0.1%, γ ′-nitride forms,
which is Fe4N. This γ ′-nitride contributes to the brittle white layer of nitride on steel
surface. If nitrogen >6%, then ∈-nitride forms, which is a solid solution, and when it
is combined with carbon, it produces a tribologically desirable phase. The duration of
nitriding usually varies from 48 to 96 h to achieve 1-mm-deep layer. It also depends
upon the temperature (Figure 11.22), alloying elements, and the depth required.
11.4.3 cyanIdIng and carBonITrIdIng

The cyaniding and carbonitriding processes involve the hardening of the surface
layer of the steel by the addition of both carbon and nitrogen. If the hardening is car-
ried out in liquid salt baths, it is known as cyaniding, and if the process is carried out
in a gaseous atmosphere, it is called as carbonitriding.
FIGURE 11.22 Effect of nitriding time and temperature on the case depth.
In the cyaniding process, the parts to be hardened are immersed in a liquid bath
of NaCN with the concentration changing between 25% and 90% at 800°C–960°C.
A controlled flow of air is passed through the molten bath. Molten NaCN decom-
poses in the presence of air to produce sodium cyanate (NaNCO), which in turn
decomposes to produce carbon and nitrogen. The reactions involved are as follows:
2NaCN + O 2 → 2NaNCO (11.18)
2NaNCO + O 2 → Na 2 CO3 + CO + 2N (11.19)
2CO → CO 2 + C (11.20)
The carbon content of the surface increases with an increase in the cyanide concen-
tration of the bath. Generally, the hardened parts are quenched in water or oil after
cyaniding, followed by a low-temperature tempering.
The carbonitriding process involves the hardening of the steel in a gas mixture
of carrier gas, enriching gas, and ammonia so that carbon and nitrogen are diffused
simultaneously into the surface of the steel. While the carrier gas is typically a mix-
ture of nitrogen, hydrogen, and carbon monoxide produced in an endothermic gen-
erator, as in gas carburizing, the enriching gas is usually propane or natural gas and
is the primary source for the carbon added to the surface. Nitrogen content diffused
into the steel surface depends on the ammonia content and temperature. This process
can be carried out at a lower temperature since carbon–nitrogen austenite is stable at
lower temperatures than plain carbon austenite. Here, quenching is performed by oil
rather than water to avoid distortion and cracking.
11.4.4 flame HardenIng

In flame hardening, heat is applied by an oxy-acetylene or oxy-fuel blowpipe, fol-
lowed by spraying of a jet of water as the coolant. Skill is required to avoid over-
heating; otherwise, it can result in cracking after quenching and grain growth in
the areas just below the hardened surface. Since selected areas are hardened in this
treatment and then quenched to form martensite, it is essential to begin with a steel
that is capable of being hardened. Typically, the carbon content of steel used for
flame hardening varies from 0.3% to 0.6%. In general, four methods are used for
flame hardening:
• Stationary
• Progressive
• Spinning
• Progressive–spinning
In the stationary method, both torch and workpiece are stationary, and therefore, this
method is used for the spot hardening of small parts such as valve stems and open-
end wrenches. In the progressive method, the torch moves over the large stationary
workpiece. This method is utilized for the hardening of large parts like teeth of giant
gears. In the spinning method, the torch is fixed while the workpiece rotates. This
method is used for circular cross section, such as precision gears and pulleys. In the
progressive spinning method, the torch moves over the rotating workpiece, and the
method is used to surface-harden large circular parts such as rolls and shafts.
11.4.5 InducTIon HardenIng

As the name suggests, this treatment involves heating of the components by electro-
magnetic induction. A coil or conductor carries a high-frequency alternating current,
which is then induced into the steel component enclosed within the magnetic field of
the coil. As a result, induction heating takes place, and only the outer surface of the
steel component is hardened due to the skin effect. However, the heat applied to the
surface tends to flow toward the center of the component by conduction, and there-
fore, the time of hardening is an essential parameter in controlling the depth of the
hardened zone. Typically, the parts are heated for a few seconds only to a depth that
is inversely proportional to the square root of the frequency followed by quenching
by a jet of water. Frequencies in the range of 10,000–500,000 Hz are commonly used
in this hardening method.
11.5 THERMOMECHANICAL TREATMENT
Thermomechanical treatment is a combination of heat treatment and a deforma-
tion process to change the shape and refine the microstructure of the material by
affecting the phase transformation. The mechanism behind this treatment is that
the plastic deformation process induces various crystal defects such as dislocations,
vacancies, stacking faults, and subgrain boundaries in the material. These defect
sites act as nucleation sites and thereby alter the kinetics of phase transformation and
morphology of the phases developed in the specimen. Hot rolling of metals, a well-
established industrial process, is a thermomechanical treatment that plays an integral
part in the processing of many sheets of steel from low-carbon mild steels to highly
alloyed stainless steels.
11.5.1 classIfIcaTIon Based on crITIcal TemperaTures

Table 11.1 gives the critical temperatures–based classification of thermomechanical
treatments of steel. Three classes of thermomechanical treatments have been identi-
fied, namely, supercritical, intercritical, and subcritical treatment. While supercriti-
cal treatment involves deforming the steel in the austenitic condition, the intercritical
treatment comprises deforming the steel in a phase mixture of ferrite and austenite.
On the other hand, subcritical treatment comprises deforming the steel below the
lower critical temperature (A1). This treatment also comes under the low-temperature
thermomechanical treatment (LTMT) since the austenite is supercooled to a tem-
perature below the recrystallization temperature. Accordingly, the supercritical
treatment that includes deformation of austenite at a temperature above the recrystal-
lization temperature falls under the high-temperature thermomechanical treatment
category. This section mostly focuses on the subcritical treatment or LTMT such as
ausforming and isoforming processes.
11.5.1.1 Ausforming
The process involves the deformation of metastable austenite at a constant tem-
perature between the ferrite and bainite curves of the TTT diagram, as shown in
Figure 11.23. No transformation occurs during mechanical treatment. The deformed
steel is then quenched to obtain martensite, and subsequent tempering is done to
achieve appropriate mechanical properties. It provides high strength without affect-
ing ductility and toughness. Usually, a 4.7% Cr, 1.5% Mo, 0.4% V, and 0.34% C steel
has a tensile strength of about 2000 MPa after conventional quenching and temper-
ing treatment, whereas after ausforming, the strength increases to over 3000 MPa.
Steels that possess a sufficient gap between the pearlite and bainitic curves are
subjected to this treatment. Alloying elements such as Cr, Mo, Ni, and Mn develop a
deep metastable austenite bay by displacing the TTT curve to longer transformation
times (Figure 11.23). In this case, a high austenitization temperature is necessary
for the dissolution of alloying elements. Nevertheless, it should not be very high;
otherwise, the coarsening of austenite will take place, which ultimately results in
coarse martensite (low toughness) on quenching. Cooling from this austenitizing
temperature to the metastable austenite bay must be rapid to avoid the occurrence of
ferrite, and after deformation, the cooling must be fast enough to prevent the forma-
tion of bainite. The amount of deformation is the most critical variable. The higher
TABLE 11.1
Critical Temperature–Based Classification of Thermomechanical Treatments
Class Temperature Range Examples
Supercritical TMT or HTMT Above A3 Hot–cold working, controlled rolling
Intercritical TMT Between A3 and A1
Subcritical TMT or LTMT Below A1 Isoforming, ausforming
TMT, thermomechanical treatment; LTMT, low-temperature thermomechanical treatment;

HTMT, high-temperature thermomechanical treatment.
FIGURE 11.23 Schematic illustration of the ausforming process.
the degree of working, the greater the ultimate strength achieved. Ductility increases
with increasing deformation, although it only becomes significant at deformations
above 30% reduction in thickness. Steels subjected to substantial deformation dur-
ing ausforming process exhibit very high dislocation densities (up to 1013 cm−2). This
high dislocation density along with the fine dispersion of alloy carbides is respon-
sible for the strengthening effect.
11.5.1.2 Isoforming
Unlike ausforming, isoforming involves deformation of metastable austenite until
the transformation of austenite is complete at the deformation temperature. The steel
is first heated to above A3 followed by immediate quenching to a temperature of
about 650°C, as shown in Figure 11.24. Mechanical deformation is applied at this
temperature during the phase transformation so that the structure produced com-
prises fine ferrite subgrains with spheroidized cementite particles instead of a ferrite/
pearlite aggregate. Due to the refinement of the ferrite grain size and replacement of
lamellar cementite by spheroidized particles, substantial improvement in toughness
and strength occurs.
11.6 HEAT TREATMENT OF CARBON AND ALLOY STEELS

In general, carbon steels are used in applications where moderate strength is required.
They have the inherent properties derived from carbon only. Heat treatment of car-
bon steels is only possible in thinner sections, as it is challenging to harden thicker
sections (thicker than 1.5 cm) because of their low hardenability. Moreover, these car-
bon steels cannot be used in corrosive environments. These limitations put engineers
FIGURE 11.24 Schematic illustration of the isoforming process.
to think about alloy steels, which not only enhances the properties of carbon steels
but also induces other specific properties in the steels. Heat treatment of carbon
steels has been mostly discussed while describing standard heat treatment practices.
Hence, this section focuses on various types of alloy steels and their heat treatment
methods. However, emphasis on special steels and their heat treatment practices is
deliberated in the subsequent chapter.
Alloy steels are categorized based on the alloying element present in the steel.
The most common are manganese steels, silicon steels, chromium steels, nickel
steels, and molybdenum steels. Manganese is added to carbon steels to improve the
tensile strength, hardenability, and hot workability of steel. Heat treatment meth-
ods like hardening and tempering result in the best possible mechanical proper-
ties, while normalizing treatment improves the impact strength, after which they
are used for large forgings and castings. Silicon, like manganese, is present as a
cheap deoxidizer in all steels. When the steels contain more than 0.6% silicon, they
are grouped as silicon steels. They possess improved elastic properties, excellent
electrical and magnetic properties, and enhanced resistance to scaling at high tem-
peratures. Mainly, steels with 3%–4% silicon and less than 0.5% carbon, popularly
known as electrical steel, is used in the cores and poles of electrical machinery.
The desired properties are derived from a coarse-grained and textured structure,
which is obtained by repeated cold rolling and annealing at 1100°C–1200°C under
hydrogen atmosphere.
On the other hand, chromium steels are well known for high hardness and wear
resistance. The low-carbon (<0.2% carbon), low-chromium (<1% chromium) steels
are usually case carburized. The presence of chromium increases the wear resistance
of the case, but the toughness of the core is reduced. Hence, these low-carbon alloy
steels are often subjected to postcarburizing heat treatment processes to refine both
the case and the core. With an increase in carbon content up to 0.4%, these medium-
carbon alloy steels are exposed to oil or water quenching, followed by tempering.
These types of steels are mostly used in springs, engine bolts, axles, and studs. High-
carbon (~1%) and high-chromium (~1.5%) alloy steels find application extensively in
the ball and roller bearings and for crushing machinery. Nickel steels are best suited
where high strength and toughness, improved fatigue strength, impact resistance,
and shear strength are required, like for high-strength structural steels. These alloy
steels, unlike carbon steels, do not need rapid quenching for obtaining high hardness
since they are heat-treated at much lower temperatures. Low-carbon (~0.2%) and
low-nickel (~3.5%) steels, typically from 23xx series, are widely used for carburiz-
ing of drive gears, studs, and kingpins, whereas the 5% nickel steels (25xx series)
have improved toughness and are, therefore, used for heavy-duty applications such
as bus and truck gears, cams, and crankshafts. Molybdenum steels require high heat
treatment temperatures, typically 20°C higher than carbon steels of same carbon
content, as they form complex carbides that are more stable than cementite. The
alloy steels with low-carbon content (40xx and 44xx series) are generally used for
case carburizing. With higher carbon content, they find applications in automotive
coil and leaf springs.
Nevertheless, alloy steels with more than one alloying element, such as
chromium–molybdenum steels, nickel–molybdenum steels, and nickel–chromium
steels, have the added advantage of improved properties and are widely used in
various sectors. Since a large number of such alloy steels are available, it is not pos-
sible to describe all the alloy steels types; however, a detailed description of special
steels, which have unique applications, is given in the next chapter.
FURTHER READING
Avala, L. Concepts in Physical Metallurgy. (Morgan & Claypool Publishers, San Rafael, CA,
2017).
12 Special Steels
12.1 STAINLESS STEELS
Stainless steels have a minimum of 11.5% chromium and some amounts of nickel,
molybdenum, and manganese to enhance corrosion resistance. The corrosion-
resisting property is due to the thin, protective, and stable chromium or nickel oxide
film on the surface. A three-digit numbering system is used to identify stainless
steel, as given in Table 12.1. While the first digit designates the group, the last two
digits have no specific importance.
We have already discussed the fact that chromium is a suitable ferrite stabilizer,
and hence, in the Fe–Cr phase diagram (Figure 12.1), it can be seen that chromium
stabilizes ferrite, leading to the formation of a closed γ-loop with a maximum of
12.7% chromium at around 1000°C. The ferrite phase is stable above this chromium
content of 12.7% over the entire temperature range. However, certain additions of
austenite stabilizers such as carbon or nitrogen increase the solubility of chromium
in austenite by expanding the γ-loop. Similarly, nickel further modifies the phase
diagram, as it strongly stabilizes austenite. Figure 12.2 explains the effect of carbon
on the phase diagram of 18/8 steel (18% chromium and 8% nickel). For a low-carbon
18/8 steel, carbides like Cr23C6 can be obtained at room temperatures. This carbide
dissolves in austenite if the steel is heated to above 1050°C, and quenching from
this temperature will avoid the formation of carbide and result in only austenite at
room temperature. However, if the steel is heated in the range of 500°C–750°C, the
carbides precipitate at the grain boundaries, thereby making the steel susceptible
to intergranular corrosion as it depletes chromium from the adjacent grains. Thus,
depending upon the composition, the stainless steels behave differently to heat treat-
ment and accordingly are classified into three general groups.
12.1.1 ferrITIc sTaInless sTeels

These are iron–chromium alloys having chromium approximately from 12% to
27%. These may contain small addition of manganese, silicon, nickel, molybdenum,
TABLE 12.1
Various Series of Stainless Steels
Series Designation Groups
2xx Chromium–nickel–manganese, nonhardenable, austenitic, nonmagnetic
3xx Chromium–nickel, nonhardenable, austenitic, nonmagnetic
4xx Chromium, hardenable, martensitic, magnetic
4xx Chromium, nonhardenable, ferritic, magnetic
5xx Chromium, low chromium, heat resisting
169
FIGURE 12.1 Iron–chromium phase diagram.
FIGURE 12.2 Influence of carbon on the phase diagram of 18/8 stainless steel.
Special Steels 171
and titanium. The carbon content is low (0.08%–0.2%), which helps to increase
toughness and reduce sensitization. Annealing is the only heat treatment practice
applied to such types of steels. In the annealed state, the strength of these ferritic
stainless steels increases by ~50% higher than that of the carbon steels. These
steels are extensively used for deep drawing applications such as vessels, kitchen
sinks, and utensils. However, this variety of stainless steel suffers from brittle-
ness upon slow cooling in the temperature range of 400°C–500°C. The reason
behind the decrease in ductility as well as impact properties is probably due to the
precipitation of very fine coherent chromium-rich, alpha prime particles arising
from the miscibility gap in the Fe–Cr system. The effect is significant at higher
chromium content.
12.1.2 marTensITIc sTaInless sTeels

These are heat-treatable stainless steels and typically contain 12%–18% chromium
and 0.10%–1.20% carbon. They have excellent corrosion resistance and toughness
and can be easily hot-worked. These steels are hardened and tempered to obtain the
best corrosion resistance but are not as good as the ferritic stainless steel. Like plain
carbon steels, the strength and hardness of these steels also depend on the carbon
content. The only difference is that the high alloy content of the stainless steels
causes the transformation to be slow, and hence, maximum hardness is produced
by air cooling. Based on the carbon content, these steels are further classified into
two categories: low-carbon martensitic stainless steels and high-carbon martensitic
stainless steels. Low-carbon martensitic steels are applicable where good weldabil-
ity, formability, and impact strength are required. These steels are heated to above
1050°C to obtain a fully austenitic structure, then quenched in oil or air, and finally
tempered. Tempering is done at a higher temperature (>600°C) to obtain higher
toughness, although there is a slight reduction in corrosion resistance due to the
precipitation of some carbides. In contrast, high-carbon martensitic stainless steels
possess high strength and hardness due to the presence of more carbon, but at the
expense of toughness and weldability. Higher austenitizing temperature is required
to dissolve the excess carbides formed, which ultimately results in poor impact
properties.
12.1.3 ausTenITIc sTaInless sTeels

Austenitic stainless steels have 16%–25% chromium and austenite-stabilizing
elements such as nickel, manganese, or nitrogen. Typical examples include
chromium–nickel (3xx type) and chromium–nickel–manganese stainless steels
(2xx type). These steels find a wide range of applications due to excellent corro-
sion resistance than the ferritic and martensitic stainless steels. Type 304 steels,
having a carbon content of 0.08% maximum, have good weldability and are used
for chemical and food processing equipment due to the restriction in carbide pre-
cipitation during welding. Later, type 304L stainless steel is developed to avoid
further carbide precipitation during welding, which contains only 0.03% carbon.
Type 316 stainless steels show better corrosion resistance property than 302 or
304 steels because they contain 2%–3% molybdenum, and hence, they are used for
pulp-handling purposes and photographic and food equipment. However, owing
to the presence of high nickel content, type 3xx stainless steels are expensive.
Alternatively, chromium–nickel–manganese stainless steels (type 201 and 202)
have been developed with the substitution of manganese for nickel. The presence
of manganese reduces the rate of work hardening.
12.2 HADFIELD MANGANESE STEELS

Hadfield manganese steel contains manganese above 10%. This steel is heated to
around 1100°C, and then it is water quenched to obtain high strength and ductil-
ity, and excellent resistance to wear. These steels find applications in power shovel
buckets and teeth, gears, spline shafts, axles, riffle barrels, grinding, and railway
track work. If this steel is slow-cooled from 950°C, the structure consists of large
brittle carbides neighboring the austenite grains, thereby decreasing the strength and
ductility. Therefore, such type of steel is always heated to above 1000°C and then
quenched to have much higher ductility and strength as compared to the annealed
condition. The steel is ordinarily reheated to below 260°C to reduce the quenching
stresses.
12.3 HIGH-STRENGTH LOW-ALLOY OR MICROALLOYED STEELS

High-strength low-alloy (HSLA) steels are prepared to have high fuel efficiency by
offering automobile components with high strength-to-weight ratios. These are mild
steels with carbon varying from 0.03% to 0.15%, manganese around 1.5%, and small
amounts of other alloying elements. Such types of steels are subjected to controlled
rolling to obtain ultrafine grains of size below 5 μm to attain a yield strength of about
550 MPa. The presence of low carbon with ultrafine ferrite grains is responsible for
high strength and good weldability.
12.4 TRANSFORMATION-INDUCED PLASTICITY STEELS

These steels are produced by strain-induced transformation of austenite to martens-
ite. Martensite forms in the metastable austenitic steels by the application of plastic
deformation at room temperature. The steel composition is so chosen that its Md
temperature is above room temperature, such as 0.3% C, 2% Mn, 2% Si, 9% Cr,
8.5% Ni, and 4% Mo. The steel is then heated to 1200°C and quenched. It is heavily
cold-worked above the Md temperature (450°C) to deform the austenite and allow the
precipitation of fine carbides to happen. This thermomechanical treatment with fine
precipitation increases the Md temperature. The steel is finally cold-rolled to below
Md temperature, i.e., at room temperature. It ultimately raises the Ms temperature of
the steel, and hence, some austenite transforms to martensite. Depending upon the
heat treatment steps, yield strengths from 700 to 2200 MPa with 90% to 20% ductil-
ity can be obtained. These steels are used where extremely high mechanical proper-
ties are desired, like in flat or wire forms.
Special Steels 173
12.5 MARAGING STEELS
Maraging steels are highly alloyed and low-carbon iron–nickel martensites that offer
an excellent combination of strength, ductility, and toughness. These are mostly used
for aerospace applications. Because of low carbon content, the martensite formed is
soft with about 30 HRc. Hence, age hardening is done to strengthen the steel further.
The composition of maraging steels is so chosen that lath martensite forms upon
cooling the steel down to the room temperature. It is because lath martensite has
several dislocations that are uniformly distributed and hence support age hardening
by providing several preferred nucleation sites for the intermetallic compounds to
precipitate. Besides, lath martensites provide favorite diffusion paths for substitu-
tional solute atoms during aging. The Ms temperature of such types of steels is kept
high around 200°C–300°C to avoid the formation of retained austenite. In some
cases, nitriding of maraging steels is done to induce hardness of about 60–70 HRc
up to a depth of 0.15 mm.
12.6 DUAL-PHASE STEELS
HSLA steels offer high strength-to-weight ratios, but their formability is poor.
Hence, they are not suitable for deep cold pressing. HSLA exhibits a sharp yield
point (stretcher strain), indicating nonuniform plastic deformation behavior. As a
result, in cold-worked steel, surface finish is poor, and hence, these are not suitable
for automobile applications.
Dual-phase steels overcome these problems. They have microstructures consisting
of islands of martensite (10%–20%) or distribution of martensite in a ferrite matrix.
It is a product of intercritical annealing, i.e., the steel is heated to a temperature of
about 790ºC in between A1 and A3 temperatures to obtain a microstructure consisting
of ferrite and austenite, as shown in Figure 12.3. On quenching, austenite transforms
into fine martensite. It is then tempered at low temperatures to improve the ductility.
The formation of martensite depends on the alloy and carbon content of the austenite
and the rate of cooling. Dual-phase steels offer several advantages such as low cost,
excellent formability, a high strength of about 500–700 M/Nm2, a low yield stress of
about 300–350 MPa, and no yield point elongation.
12.7 TOOL STEELS
Tool steels are high-quality special steels used for forming and cutting purposes.
They possess high hardness and wear resistance with economic tool life and dimen-
sional stability. There are a variety of tool steels available ranging from cheap high-
carbon steels to expensive high-speed steels, depending upon the applications. It is
the alloying elements that improve the properties and thereby increase the cost of the
tool steel. Various methods of classifying tool steels are summarized in Table 12.2.
The tool steels are rated according to different properties such as toughness, red
hardness, wear resistance, nondeforming properties, machinability, safety in hard-
ening, and resistance to decarburization. Based on the application areas, tool steels
with particular properties are selected.
FIGURE 12.3 Schematic illustration of obtaining a dual-phase steel.
TABLE 12.2
Classification of Tool Steels
Criteria Types
Quenching media Water hardening
Oil hardening
Air hardening
Alloy content Carbon tool
Low-alloy tool
Medium-alloy tool
Application Hot work
Shock resisting
High speed
Cold work
12.7.1 WaTer-HardenIng Tool sTeels

These are mostly plain carbon tool steels with carbon varying from 0.60% to
1.40%. They also contain some quantities of chromium and vanadium, which help
to enhance hardenability and wear resistance. Depending upon the carbon content,
these steels are further subdivided into three categories, as given in Table 12.3.
Special Steels 175
TABLE 12.3
Classification of Water-Hardening Tool Steels
Carbon Content Properties Applications
0.60%–0.75% High toughness and wear resistance Hammers, river sets, and concrete breakers
0.75%–0.95% Good toughness as well as good Chisels, punches, dies, and shear blades
wear resistance
0.95%–1.40% High wear resistance Woodworking tools, drills, taps, reamers,
and turning tools
These steels are generally water-quenched for obtaining high hardness and are
given spheroidization annealing treatment to enhance the machinability. They have
low red hardness and, hence, cannot be used as cutting conditions in cases where a
considerable amount of heat is generated at the edge.
12.7.2 sHock-resIsTIng Tool sTeels

Shock-resisting steels have high toughness and, as the name suggests, can withstand
shock. They contain carbon between 0.45% and 0.65%. The primary alloying ele-
ments in these steels are silicon, chromium, and tungsten. Silicon strengthens and
toughens the ferrite, while chromium rises the hardenability and wear resistance.
Tungsten improves the red hardness property of these steels. Sometimes, molybde-
num is added to increase the hardenability. The steels in this group are used in the
manufacturing of forming tools, punches, chisels, pneumatic tools, and shear blades.
12.7.3 cold-Work Tool sTeels

This group of steels is the most essential group of tool steels, as it offers a wide range
of applications. The key alloying elements added to this group of steels are chro-
mium, vanadium, and tungsten. High toughness, abrasion and wear resistance, and
impact resistance are some of the vital properties of cold-worked steels. These steels
are subdivided further into three classes – oil-hardening steels, air-hardening steels,
and high-carbon high-chromium steels.
Oil-hardening steels contain manganese, chromium, and tungsten. They enhance
hardness and wear resistance. As the name suggests, these are hardened by oil
quenching. They possess excellent nondeforming properties and are, therefore,
unlikely to bend, distort, twist, or crack during the heating process. Sometimes,
silicon is added to improve the machinability in the annealed state and increase the
resistance to decarburization of the steels. They are used for taps, form tools, and
expansion reamers.
Air-hardening steels are hardened by air cooling. They mostly contain about 1%
carbon, 3% manganese, 5% chromium, and 1% molybdenum. Among the alloying
elements, manganese and molybdenum impart air-hardening properties and induce
high hardenability. They have excellent nondeforming properties, wear resistance,
toughness, red hardness, and resistance to decarburization. These are mostly used
for blanking, forging, and trimming purposes.
High-carbon high-chromium steels contain carbon up to 2.25% and 12%–14%
chromium. Besides, they may contain other alloying elements such as molybdenum,
vanadium, and cobalt. They offer superb wear resistance, nondeforming properties,
excellent abrasion resistance, and minimum dimensional change in hardening. This
group of steels are used for blanking and piercing dies, drawing dies for wires, bars,
shear blades, and cold-forming rolls.
12.7.4 HoT-Work Tool sTeels

This group of steels is mostly applicable for high-temperature metal-forming pro-
cesses such as hot forging, hot stamping, hot extrusion, die casting, and plastic
molding. The operating temperature varies from 200°C to 800°C. They have good
red hardness due to the alloying elements such as chromium, molybdenum, and tung-
sten. The red hardness property becomes significant when the sum of these alloying
elements is at least 5%. Besides, they possess high wear resistance, toughness, ero-
sion resistance, and resistance to softening at high temperatures. Depending upon
the alloying elements, hot-worked tool steels are subdivided into three categories –
hot-work chromium-base, hot-work tungsten-base, and hot-work molybdenum-base.
Hot-work chromium-base steels usually contain a minimum of 3.25% chromium
and minor amounts of other alloying elements such as tungsten, vanadium, and
molybdenum. Because of these alloying elements, they have high red hardness
and high hardenability. These are mostly used for extrusion dies, die-casting dies,
and hot shears. A specific hot-work tool steel, H11 tool steel, finds applications in
highly stressed structural parts, predominantly for supersonic aircraft, as it can resist
softening upon continuous exposure to temperatures up to 500°C.
Hot-work tungsten-base steels have at least 9% tungsten and 2%–12% chromium.
With higher alloying elements content, these steels have high resistance to high-
temperature softening as compared to hot-work chromium-base steels. These are
used for mandrels, punches, and extrusion dies for brass, nickel alloys, and steels.
Hot-work molybdenum-base steels contain 8% molybdenum, 4% chromium
with small amounts of tungsten, and vanadium. These steels possess almost similar
characteristics as those of hot-work tungsten-base steels but have higher toughness
because of low carbon content. They are poor in resistance to decarburization, and
therefore, proper care needs to be taken during heat treatment.
12.7.5 HIgH-speed Tool sTeels

High-speed steels contain large amounts of tungsten or molybdenum along with
some chromium, vanadium, and cobalt. They consist of carbon varying from 0.70%
to 1.5%. They have good nondeforming properties, wear resistance, machinability,
and resistance to decarburization. These are mostly used for cutting tools, burnishing
tools, preparing extrusion dies, and blanking punches and dies. Based on the alloying
elements content, these steels are subdivided into two groups – tungsten-base high-
speed steels and molybdenum-base high-speed steels. Among the tungsten-base
Special Steels 177
high-speed steels, 18-4-1 (T1) is the most commonly used high-speed steel, which
contains 18% tungsten, 4% chromium, and 1% vanadium. Since tungsten is expen-
sive, it is replaced by molybdenum, which gives rise to molybdenum-base high-speed
steels. Sometimes, cobalt is added when high red hardness is required. For example,
T15 steel, which contains vanadium and cobalt, provides dominance in both red
hardness and abrasion resistance. However, the addition of cobalt makes the steel
brittle, and hence, the high-cobalt steels must be protected against excessive shock
or vibration during service.
12.8 ELECTRIC GRADE STEELS

Electric grade steels may contain up to 6.5% silicon, but commercial grades usually
have silicon content up to 3.2% because higher concentrations may lead to brittleness
during cold rolling. Also, other alloying elements such as manganese and aluminum
up to 0.5% are added. While silicon helps in lowering the core loss, carbon causes
magnetic aging by precipitating carbides. Thus, carbon content is kept very low to
around 0.005%. Electrical steels are generally manufactured in cold-rolled strips,
which are used as cores of transformers, stator, and rotor of electric motors. There
are two varieties of electrical steels – cold-rolled grain-oriented steel (CRGO) and
cold-rolled non–grain-oriented steel (CRNGO). While CRNGO steels have similar
magnetic properties in all directions, CRGO steels are treated in such a way that
the optimal features are developed only in the rolling direction. CRGO steels are
mostly used in the cores of power and distribution transformers because of their
ability to increase the magnetic flux density in the coil rolling direction. On the other
hand, CRNGO electrical steels find applications where magnetic flux density is not
constant, such as in electric motors and generators with moving parts.
FURTHER READING
1997).
Taylor & Francis
13 Some In Situ Postweld
Heat Treatment Practices
Very often, it is required to join steel components in order to prepare a structure.
Automobile industries are the major players in this regard. Welding of sheet metal is
one of the critical concerns for automobile industries, as it targets for improved crash-
worthiness and fuel economy. In fact, welding is unavoidable in almost every structural
application, may it be mobile structures such as automobile and railways or immobile
structures such as roofs, bridges, and so on. However, welding causes structural distor-
tions to the components and usually the weld zone, which is known as heat-affected
zone (HAZ), and more specifically, fusion zone (FZ) exhibits poor mechanical per-
formance than that of the unwelded structure. Usually, this FZ failure is a common
phenomenon during a collision, which reduces the crashworthiness of the automobile
and thus should be taken care of in order to have a reliable and durable long-term appli-
cation. Welding is not only a requirement while fabricating a new structure but also
essential during maintenance or replacement of a component in an assembly. In order
to restore the properties of the metal at FZ by modifying the microstructure and reliev-
ing the internal stresses, certain postweld heat treatment (PWHT) is essential. In case
of certain pressure vessels and pipes, PWHT is mandatory to ensure safe operation of
the components in both metallurgical and mechanical aspects.
13.1 NECESSITY
During a fusion welding operation, a part of the component is melted and resolidi-
fied. This process takes place only at a very small localized region of the component.
The high heating rate associated with the welding operation followed by high cool-
ing rate brings several changes in the microstructure in the HAZ and thus exhibits
different characteristics than that of the base metal. This evolved temperature gradi-
ent also leads to generation of residual stresses, which is also required to be released
in order to preserve the performance of the welded joints. These residual stresses
further become very critical in case of thick cross-sectioned components. In certain
cases, where the cooling rate surpasses a critical value, martensitic transformation
also may take place reducing the ductility of the joint drastically. In certain sensi-
tive microstructures, hydrogen-induced cracking (HIC) also may be resulted with
substantial amount of hydrogen and applied stress. Hence, it is a usual practice to
perform PWHT in critical applications, where unprecedented failure is not tolerable.
13.2 CONVENTIONAL POSTWELD HEAT TREATMENT PROCESS

Typically, after welding, the component is heated to an elevated temperature followed
by some period of holding and then controlled cooling. Two types of PWHT are
179
normally practiced: one in which there is a possibility of HIC, known as postheating,

and the other in which a high magnitude of residual stress exists, known as stress
relieving.
13.2.1 posTHeaTIng
Combination of a sensitive microstructure along with a substantial amount of hydro-
gen and structural load gives rise to a possibility of hydrogen-induced cracking.
In general, this hydrogen embrittlement usually happens close to the ambient tem-
perature. Hence, suitable heat treatment must be done immediately after welding
before the structure reaches the room temperature. From an intermediate tempera-
ture (just after welding), the steel should be heated to the postheating temperature
(~200°C–300°C) followed by a suitable residence time and then cooling.
13.2.2 sTress relIevIng

This heat treatment is adopted when the steel structure experiences a high magnitude
of residual stress as a consequence of fabrication process. In dome of the fabrication/
welding operations, the residual stresses may even go rise to the yield strength of the
material. Hence, for suitable in-service performance, these residual stresses must be
released. A slow heating is often required to a temperature below the lower critical
temperature of the steel followed by some holding time and then slow cooling. Most
of the time, the holding temperature is 600°C–700°C, and holding time is usually 1 h
for 15 mm thickness. Stress relieving benefits the structure by several possible ways,
i.e., by improving the dimensional stability and reducing the possibility of stress
corrosion cracking.
13.3 IN SITU POSTWELD HEAT TREATMENT OF

TRANSFORMATION-INDUCED PLASTICITY STEEL
As discussed in the earlier chapter, transformation-induced plasticity (TRIP) steel
is a special steel having a combination of ferrite, martensite, and retained austenite.
The TRIP effect is due to transformation of austenite to martensite upon loading.
Hence, under tensile kind of loading, the tendency of necking is minimized with
improved rate of strain hardening. This also enhances the energy absorption capac-
ity of the material. While all these aforementioned characteristics are desirable for
automobile applications, weldability of these steels overshadows its application in
such area. Interfacial failure through crack propagation across the FZ reduces its
load bearing energy absorption ability, while loaded in a cross tension mode. The
reduced weldability of such TRIP steel may be attributed to the brittle martensite
formed in the FZ. Very often, resistance spot welding is carried out for such steels
in various automobile applications. In situ PWHT thus would be an appropriate
method in order to restore the physical and mechanical properties of the weldment.
The martensite formed at the FZ and HAZ is required to be tempered in order to
reduce the brittleness. This is conventionally achieved by appropriate hardening
In Situ Postweld Heat Treatment Practices 181
and tempering operations. The weldment should come to a temperature below the
Mf temperature before start of the tempering process. In certain cases, reduction
of hardness of the HAZ is not the only requirement. In addition, if there is inco-
herency among the dendritic structure after solidification due to segregation of S
and P, then the edges of the weldment should be remelted to modify the unwanted
dendritic structure. Here, let us discuss a suitable technique for in situ PWHT of
TRIP steels after resistance spot welding.1
A welding operation was carried out using resistance spot welding using a PLC-
controlled machine. This process produced a weldment with a conventional recom-
mended FZ size of 4√t. This process was followed by application of a second pulse
current (3.5, 5.5, and 7.5 kA amplitude) for PWHT purpose.
Figure 13.1a displays a conventional welding microstructure comprising FZ,
HAZ, and base metal (BM). The microstructure of the metal in FZ was found to
be martensite as shown in Figure 13.1b. This martensite is quite hard with hardness
425 HV, much higher than that of the base metal. Mode I loading during CT testing
further indicates failure of the joint by propagation of the crack into the FZ, resulting
in interfacial failure as evident from Figure 13.1c. Semicleavage fracture mode can
be noticed from Figure 13.1d indicating the presence of hard brittle phase in the FZ.
Furthermore, the carbon equivalent of the steel plays an important role in driving
the crack to the FZ.
Effect of PWHT with different current is shown in Figure 13.2. All the PWHT
operations have increased both the peak load and failure energy substantially in
comparison to the as-welded structure with maximum of 60% and 50%, respectively.
FIGURE 13.1 (a) Weldment comprising FZ, HAZ, and BM, (b) martensite in the FZ,
(c) interfacial failure, and (d) fractography of the fractured surface.1 FZ, fusion zone; HAZ,
heat-affected zone.
FIGURE 13.2 Effect of PWHT on CT test, (a) load vs. displacement, (b) peak load, and
(c) failure energy.1 (A: as-welded, B: PWHT with 3.5 kA, C: PWHT with 5.5 kA; D: PWHT
with 7.5 kA). PWHT, postweld heat treatment.
FIGURE 13.3 Microstructure of FZ in its (a) as-welded and (b) tempered conditions.1 FZ,
fusion zone.
Visual inspection of the fractured specimens indicates pullout failure to be the domi-
nating failure mode rather than interfacial failure. This may be attributed to trans-
formation of the hard and brittle martensite at the HAZ to relatively soft and ductile
tempered martensite as shown in Figure 13.3. Lath martensite present in the FZ after
welding (Figure 13.3a) has been transformed to tempered martensite (Figure 13.3b)
after performing the PWHT process.
13.4 POSTWELD HEAT TREATMENT OF DUPLEX STAINLESS STEEL

Duplex stainless steels are optimal combination of mechanical properties and cor-
rosion resistance and very often the preferred structural material in various load
bearing applications serving at harsh and corrosive environments, such as oil and gas
industries, marine, chemical container, waste drainage pipes, and so on. The micro-
structural constituents, i.e., ferrite and austenite render this special combination of
properties in these duplex stainless steels. For achieving this optimal combination
of properties, ideally a 50% ferrite–50% austenite is required to be present in the
microstructure. While performing welding on such structures due to any in-service
or assembly or maintenance requirement, this ratio is altered due to associated phase
transformation. Due to presence of substantial high volume fraction of the alloying
elements, these steels tend to form several undesirable phases such as chi phase,
In Situ Postweld Heat Treatment Practices 183
carbide phase, sigma phase, secondary austenite, and so on in addition to ferritiza-

tion, which may negatively affect the properties and performance of the weldment.
Hence, suitable PWHT operation is essential for long-term reliable applications.
Here, the effects of short PWHT on the properties of a tungsten inert gas (TIG)–
welded duplex stainless steel are discussed.2 Cold-rolled duplex stainless steel sheets
of 1.5 mm thickness were taken and welded by single TIG welding method without
use of any filler material. This is followed by heat treatment of the joints in a muffle
furnace at 1000°C–1100°C (in the range of temperature where precipitation does not
take place) in normal atmosphere with a holding time of 1.5 min followed by water
quenching.
The base metal (Figure 13.4a) comprising ferrite (bright region) and austenite
(dark region) is a signature of duplex stainless steel having austenite embedded in
a continuous ferritic matrix. Columnar ferrite and equiaxed ferrite grain are visible
in the FZ at near the fusion line and center, respectively, as shown in Figure 13.4b
and c. Widmanstätten-type austenite (WA), intragranular austenite (IGA), and grain
boundary austenite (GBA) are embedded in the ferrite grains as shown in the figures.
The HAZ contains a gradient of microstructural changes due to temperature gradi-
ent in the region from melting point of the metal near the FZ to ambient temperature
at the base metal along with differential heating and cooling rates.
The microstructure of the FZ after the PWHT was also obtained in order to ver-
ify the possible changes in the microstructural features as shown in Figure 13.5.
Widmanstätten austenite and allotriomorphs austenite at the grain boundaries tend
FIGURE 13.4 Microstructure of duplex stainless steel, (a) base metal, (b) FZ near the fusion
line, (c) center of FZ, and (d) HAZ (in between two white vertical lines).2 FZ, fusion zone;
HAZ, heat-affected zone.
FIGURE 13.5 Microstructure of the FZ after PWHT at (a) 1020°C and (b) 1080°C for
15 min.2 FZ, fusion zone; PWHT, postweld heat treatment.
FIGURE 13.6 Ferrite content in base metal (BM), HAZ, and FZ in the as-welded (AW)
condition and after PWHT at different temperatures.2 FZ, fusion zone; HAZ, heat-affected
zone; PWHT, postweld heat treatment.
to increase as the PWHT is increased. This increment indeed is associated with

spheroidization of the intragranular austenite. Overall, all these facts point out to
have a higher austenite content on the FZ due to the PWHT process.
The ferrite content at different zones of the welded metal after PWHT is shown in
Figure 13.6. As can be seen, the base metal has almost 50% ferrite in the as-welded
condition and increases slowly as the PWHT temperature is increased. On the other
hand, the ferrite content is highest in the FZ followed by HAZ in the as-welded con-
dition. This gradually decreases as the temperature of PWHT is elevated till 1080°C,
and all the zones more or less contain equivalent ferrite content. However, beyond
that temperature, the ferrite content further tends to increase.
REFERENCES

14 Heat Treatment
of Cast Iron
14.1 INTRODUCTION
Cast irons are mostly alloys of iron and carbon that contain carbon between 2.1%
and 6.67%. At the eutectic temperature, such high levels of carbon are essential to
saturate austenite so that carbon precipitates in the form of carbides and graphite
in the structure of the alloy. Being brittle, they cannot be forged, rolled, drawn, or
worked at room temperature. However, they can be quickly melted and “cast” into
desired shapes, and hence, they are commonly known as cast irons. Melting units
for producing liquid melt for casting may be an electric arc, cupola, reverberatory,
or induction furnace.
Typical cast irons are cheap and, therefore, are of importance, as the properties of
cast iron can be varied over a broader range by proper alloying, reasonable foundry
control, and suitable heat treatment practice. Although they are brittle and have less
strength compared to steel, they have other useful features.
14.2 TYPES OF CAST IRON

The most appropriate method of classification of cast iron is dependent on the metal-
lographic structures. Carbon content, the alloy and impurity content, the rate of
cooling during and after solidification, and the heat treatment after casting are four
variables that need to be considered for differentiating the cast irons from each other.
These variables control the physical form and condition in which carbon is present
in the alloy. Carbon can present in a combined state as iron carbide or cementite or
occur as free carbon in graphite. Physical and mechanical properties of cast iron are
greatly influenced by the shape and distribution of the carbon in the structure of cast
irons. The classification of cast irons which is best done based on their microstruc-
ture is as follows:
• White cast iron: Carbon exists in the combined form as cementite.

• Gray cast iron: Carbon exists in the uncombined form, such as graphite
flakes.
• Malleable cast iron: It is a product of heat treatment of white cast iron, and
carbon is present in uncombined form as irregular round particles, also
called as temper carbon.
• Spheroidal graphite (S.G.) iron or nodular iron: Alloy addition leads to the
formation of spheroids of uncombined carbon during cooling. It is unlike
malleable cast iron, where temper carbon is formed not only during cooling
but also during heat treatment after casting.
185
14.2.1 WHITe casT Iron

According to the best method of classification of cast iron via metallographic struc-
ture, white cast iron is one in which all the carbon is present in a combined state, also
known as cementite. They have more than 2.11% carbon and consist of hard carbide
phases surrounded by a hardenable metal matrix. Both constituents grow from the
melt and differ in crystallographic structure, and therefore, the hardness depends on
alloying and heat treatment. White cast irons have been the doers of wear protection
in the mining and cement industry, as well as in road building where abrasion by
mineral grains prevails.
The changes occurring in the microstructure during solidification and subse-
quent cooling can be corroborated from the iron–iron carbide phase diagram, as
discussed in Chapter 6. A hypoeutectic white cast iron containing about 2.5% car-
bon when cooled from the molten state (point 1 in Figure 14.1) under equilibrium
conditions begins solidification at the liquidus line with the formation of dendrites of
austenite (point 2). Solid dendrites of proeutectic austenite continue to form by los-
ing carbon to the liquid until eutectic temperature (1147°C). At this point (point 3),
the alloy consists of proeutectic austenite having 2.11% carbon and liquid having
4.3% carbon. The liquid undergoes eutectic reaction forming ledeburite, which is
FIGURE 14.1 The iron–iron carbide phase diagram.

Heat Treatment of Cast Iron 187
a eutectic mixture of austenite and cementite. Usually, this newly formed eutec-
tic austenite gets added to the primary austenite leaving behind massive layers of
cementite. The Acm line of the iron–iron carbide phase diagram indicates that the
solid solubility of carbon in austenite decreases with temperature (1147°C–727°C),
and this results in the deposition of proeutectoid cementite (from both the primary
and eutectic austenite) onto the already-existing eutectic cementite. The dendrites of
primary austenite break due to the secondary (proeutectoid) cementite precipitating
out from them. At the eutectoid temperature (727°C), all the austenites having 0.77%
carbon undergo eutectoid reaction forming pearlite, i.e., ferrite and cementite. The
structure essentially remains unchanged after cooling from this temperature to room
temperature. The stages of formation of white cast iron are schematically illustrated
in Figure 14.2.
The characteristic microstructure of white cast iron consists of black dendrites
of transformed austenite (pearlite) in a white interdendritic network of cement-
ite. Cementite is brittle and has a high hardness (~ 1000 VPN). Large amounts of
cementite make the white cast irons hard and wear resistant, but because of extreme
FIGURE 14.2 Stages of formation of white cast iron.

brittleness, these are difficult to machine. Therefore, they are only used where resis-
tance to wear is mostly needed, and ductility is not a necessary material attribute,
like liners used in cement mixers, extrusion nozzles, ball mills, and specific types
of drawing dies.
14.2.2 gray casT Iron

Gray cast irons have flakes of graphite embedded in a matrix of steel. They show
a gray-blackish-colored fracture. The size and character of the steel matrix play
an essential role in the strength of gray cast iron. Gray cast irons are hypoeutectic
cast irons having a total carbon content between 2.4% and 3.8%. Silicon content
is between 1.2% and 3.5%. By adequately controlling the composition and cooling
rates, cementite tends to separate into graphite and austenite or ferrite.
Initially, these alloys solidify to form primary austenite. The first appearance
of combined carbon in the form of cementite occurs from the eutectic reaction at
1143°C. However, the graphitization process is aided by various factors like high car-
bon content, high temperature, and the proper amount of graphitizing elements, such
as silicon. It has been experimentally proved that with proper control of the above
factors, the alloy follows the stable iron–graphite equilibrium diagram, forming
austenite and graphite at the eutectic temperature of 1143°C. The graphite appears
as several irregulars, elongated and rounded plates that provide gray cast iron its
natural grayish or blackish fracture. It may be noted that the flakes are three-dimen-
sional particles. During the continued cooling process, additional carbon precipi-
tates because of the decline in carbon solubility in austenite. The carbon precipitates
as graphite or as proeutectoid cementite, which quickly graphitizes. The matrix
that embeds the graphite decides the strength of the gray cast. The condition of the
eutectoid cementite largely determines the nature of this matrix. For example, if the
composition and rate of cooling are so adjusted that the graphitization of eutectoid
cementite happens, then a fully ferritic matrix is produced.
In contrast, if the eutectoid cementite graphitization is avoided by controlling
the composition and cooling rate, the matrix is entirely pearlitic. The matrix com-
position varies from pearlite, through the mixtures of pearlite and ferrite in various
quantities, down to pure ferrite. The graphite–ferrite mixture is the weakest and soft-
est gray cast iron. Hence, the hardness and strength of this mixture can be increased
by increasing the combined carbon until it reaches a maximum with the pearlitic
gray cast iron.
14.2.3 malleaBle casT Iron

Malleable cast iron is a type of cast iron whose structure consists of a metastable
carbide phase in a pearlitic matrix phase. This metastable cementite phase tends to
undergo decomposition to stable iron and carbon phases. It does not occur under
normal conditions, and the metastable carbide phase persists indefinitely in its origi-
nal form. Cementite phase takes a long time to decompose into stable phases, and
the reaction Fe3 C  3Fe + C is favored by several factors, including high tempera-
tures, high carbon content, presence of impurities, and elements that support Fe3C
decomposition. The stable iron–carbide (graphite) system put on top of the meta-
stable systems can be seen in Figure 14.3. The ideal composition that is suitable for
the conversion of white iron to the malleable cast iron is given in Table 14.1.
Production of malleable cast iron requires the conversion of the carbon present in
the white cast iron into two separate phases of ferrite and irregular nodules of tem-
pered carbon (graphite). This conversion takes place in two stages of annealing. In
the first stage, annealing is done where the white cast iron is heated to a temperature
range of 900°C–950°C. During heating, the pearlite phase is converted to the aus-
tenite phase. The austenite dissolves an additional cementite phase as its temperature
reaches the austenitization temperature. It is shown in Figure 14.2 that the austenite
formed in the metastable iron–iron carbide system tends to dissolve more carbon than
the austenite of the stable iron–graphite system. As a result, a driving force arises,
which results in the precipitation of the carbon as a free elemental graphite. This
precipitation leads to the growth of the graphite nucleus, which grows at a uniform
rate in all directions. These graphite precipitates appear as irregular nodular shapes
or spheroids, commonly known as temper carbon. For the precipitation, graphite
particles need to nucleate and then further grow, which depends on many factors,
including the presence of any defects, foreign particles, and nucleating surfaces.
FIGURE 14.3 The iron–graphite system (dotted lines) placed over the iron–iron carbide
phase diagram.
TABLE 14.1
Typical Chemical Composition of Malleable
Cast Iron
Element Composition (%)
Carbon 2.16–2.90
Silicon 0.90–1.90
Manganese 0.15–1.25
Sulfur 0.02–0.20
Phosphorus 0.02–0.15
Some elements like silicon tend to promote this reaction, but it should not exceed
beyond a permissible limit; otherwise, the iron may not solidify as white cast iron,
which is undesirable. These nuclei are obtained through proper annealing cycles
and methods. Several parameters, such as temperature, composition, and nucleation
tendency determine the rate of annealing. Increasing temperature has both positive
and negative impacts. While increasing the temperature increases the decomposi-
tion tendency of the carbide phase leading to increased graphite precipitation, it has
specific limits, above which distortions occur in the annealed product. Hence, the
temperature needs to be chosen to account for these considerations and, therefore,
are controlled between 900°C and 950°C. In this temperature range, the cast iron is
maintained for an extended period until most of the primary carbide phase decom-
poses. The final microstructure at the end of the first stage consists of tempered
carbon nodules distributed throughout the matrix of saturated austenite. After the
first stage, the cast iron is cooled at a rather rapid rate.
During the second stage, the castings are cooled at a slow rate of 5–10°C/h
through the eutectoid region. The carbon dissolved in the austenite phase converts
to the graphite phase during the cooling process. These graphite phases get depos-
ited on the tempered carbon particles. The remaining austenite transforms to ferrite.
Once the graphitization is finished, no microstructural change takes place on further
cooling to room temperature. The final microstructure consists of tempered carbon
nodules in the ferritic matrix. This type of iron is known as standard or ferritic mal-
leable iron.
When the carbon nodules are compact, discontinuity in the ferritic matrix does
not take place, which helps in improving the strength and the ductility as compared
to gray cast iron. The graphite nodules serve as lubrication purposes for use in cutting
tools, resulting in good machinability of the tools. The ferritic malleable irons find
applications in automotive and agricultural equipment, railing casting on bridges,
industrial casters, and pipe fittings.
14.2.4 spHeroIdal grapHITe Iron

Spheroidal Graphite (S.G.) iron, also called nodular cast iron or ductile iron, contains
spheroids of the graphite phase embedded in the steel matrix. The matrix can be
ferritic or pearlitic, which governs the final properties of the casting. The general
composition of S.G. iron is given in Table 14.2. The microstructure of S.G. iron,
as shown in Figure 14.4, shows that the spheroids are more smooth-edged than the
irregular aggregates of temper carbon of the malleable iron.
S.G. iron is produced by controlled treatment of the cast iron using spheroidiz-
ing elements like magnesium or cerium, which helps in nodule formation of the
graphite phase. Based on the alloying proportion, there are different types of ductile
irons. Ferritic irons are those who have a matrix with a maximum of 10% pearlite.
This type of iron has maximum ductility, toughness, and machinability. The matrix
can be changed to martensite by quenching in oil or water from 850°C to 950°C.
The quenched structures are generally tempered to obtain the required strength and
hardness.
On the other hand, austenitic ductile irons contain alloying elements responsible
for maintaining the austenitic structure. These irons are useful where high corro-
sion resistance and excellent creep properties at high temperatures are required.
The properties of different types of S.G. iron are given in Table 14.3.
TABLE 14.2
Typical Composition of Elements in
Spheroidal Graphite Cast Iron
Element Composition (%)
Carbon 3.2–4.1
Silicon 1.0–2.8
Phosphorus <0.10
Sulfur <0.03
Manganese 0.3–0.8
Magnesium 0.04–0.06
FIGURE 14.4 Microstructure of S.G. cast iron.1 S.G., spheroidal graphite.

TABLE 14.3
Different Properties of Various Types of Spheroidal Graphite Iron
Tensile Hardness
Type Strength (MPa) (BHN) Elongation (%) Heat Treatment
Ferrite 350 130 22 Annealed
Ferrite 420 150 12 Annealed
Ferrite + pearlite 500 275 7 Annealed
Tempered Martensite 800 320 2 Quenched and tempered
Austenitic 400 200 40 Cast
14.3 HEAT TREATMENT OF GRAY CAST IRON

Among the heat treatment cycles, stress relieving is probably the most frequently
applied heat treatment to gray cast irons. As cooling proceeds at different rates
through various sections of the casting, it leads to the development of residual
stresses in the as-cast gray iron. These residual stresses reduce strength, cause dis-
tortions, and in extreme cases, may also lead to cracking. Hence, stress-relieving
treatment is done to remove unwanted stresses. Stress-relieving temperature is gen-
erally well below the pearlitic to austenitic transformation range. These stresses are
wholly removed by soaking the alloy in the range of 550°C–650°C.
Annealing treatment of gray cast iron is carried out to graphitize carbides and
to homogenize the castings. The alloy is softened, and its ductility and machin-
ability increase. At machine-building plants, gray iron castings are most frequently
annealed to improve their machinability. The annealing treatment involves soaking
the castings for up to 10 h at 850°C–950°C. Annealing of gray cast irons leads to
cementite decomposition. The intensity of graphitization increases with heating tem-
perature. Up to 600°C temperature, the treatment has an insignificant effect on the
structure of gray iron. Above 600°C, there is a marked increase in the rate at which
iron carbide decomposes to graphite and ferrite, which reaches a maximum at 760°C
for unalloyed and low-alloy gray cast irons. Small castings of simple shapes may be
heated to a temperature of 1050°C–1150°C in a salt bath for a minimum holding time
for the decomposition of cementite.
Normalizing treatment may be done by increasing the temperature to above the
transformation range (900°C–950°C) and soaking it for a particular time depend-
ing upon the thickness of casting followed by air cooling. Normalizing enhances
mechanical properties like hardness and tensile strength and restore as-cast proper-
ties that have been altered by other heat-treating processes, such as graphitizing or
preheating and postheating associated with repair welding.
Like steel, gray iron can be hardened by cooling rapidly or quenching in water,
oil, or salt bath from temperatures of about 800°C–850°C. Oil quenching is pre-
ferred over water quenching for through hardening as the latter may cause distortion
and cracking. For flame or induction hardening, water quenching is utilized, as only
the outer surface needs to be hardened. Quenched iron can be tempered in the range
of 150°C–650°C to relieve stresses and improve toughness. Although tempering

enhances strength and toughness, it decreases hardness. A minimum temperature
of about 370°C is required to restore the impact strength to the as-cast level. Heat
treatment is not customarily used commercially to increase the strength of gray iron
castings. Specific alloying elements can improve the strength of the as-cast metal.
Besides, a reduction in the amounts of silicon and total carbon increases the load-
bearing capacity of the castings. Gray iron is generally quenched and then tempered
to increase the wear and abrasion resistance by increasing the hardness. The treat-
ment produces a structure consisting of graphite embedded in a hard martensitic
matrix. The combination of hardness and excellent wear resistance makes the gray
cast irons suitable for some applications such as sprockets, diesel cylinder liners,
farm implement gears, and automotive camshafts.
14.4 HEAT TREATMENT OF MALLEABLE CAST IRON

We have established in the earlier section that the malleable irons of pearlite and ferrite
type are obtained after subsequent malleable iron formation and annealing treatment
of the white cast iron where the primary carbide phase undergoes decomposition reac-
tion. The procedure takes a considerable amount of time and involves a rapid rate of
cooling; hence, specimen with large dimensions and broad cross section may become
challenging to achieve the malleable cast iron morphology. Malleable cast irons are
used in applications that require good malleability of the tool along with appreciable
machinability. The following heat treatments can be given for malleable cast iron.
Increasing the austenitization time and temperature increases the dissolved car-
bon content in the austenite. It leads to the homogenization of the austenitic phase.
However, high temperatures also create distortion and stress generation, thereby
leading to crack formation. The suitable austenitization temperature for pearlitic
malleable iron is 885°C and that for ferritic malleable cast iron is 900°C. The hard-
ening steps for pearlitic malleable iron are as follows:
• The as-cast specimen are quenched in the air first after the first stage of
graphitization and carbide decomposition. Uneven distribution of carbon
may still be present in the matrix.
• The castings are again heated for 1 h at a suitable austenitization tempera-
ture, which leads to a homogenous structure.
• Oil quenching is done in a hot oil bath (50°C–55°C) with agitation, result-
ing in the formation of martensite of about 600 BHN hardness.
• Based on the hardness and ductility requirements, the castings are tem-
pered by heating to the temperature range of 600°C–700°C for 2 h. It leads
to the homogenization of properties. Heat treatment cycles are extremely
temperature dependent, and the higher the temperature, the higher the ten-
dency of carbon precipitation.
The ferritic malleable cast iron can also be given similar heat treatment cycles to
produce the same microstructures. The austenitization temperature is slightly
higher than that of pearlitic malleable cast iron, which is in the temperature range
of 900°C–930°C. The holding or soaking time is also higher. The resulting casting
has a lower carbon content than that of the pearlitic malleable iron. As a result, the
tempering temperature is also at a lower level.
Martempering is another heat treatment given to specimens that are susceptible
to crack formation due to stresses generated during quenching. The casting is aus-
tenitized at 885°C for 1 h, soaked in a salt bath at the temperature of 200°C followed
by air cooling. Tempering of the castings can be done to get the desired ductility and
strength. Bainite phase can be achieved after austempering pearlitic malleable cast
iron, which leads to increased tensile strength, yield strength, and ductility. Surface
hardening methods, including flame induction, can be used for increasing the hard-
ness of the specimen containing small dimensions and parts. The operation parame-
ters control the depth of hardening and composition. Other methods can also harden
the desired area of both ferritic and pearlitic malleable cast irons.
14.5 HEAT TREATMENT OF SPHEROIDAL GRAPHITE IRONS

S.G. cast iron consists of a graphite phase embedded in the ferrite or pearlite matrix,
which is dependent on the composition and other various factors. The nodular shape
of the graphite phase helps in avoiding any stress raising effect, and hence, this
proves beneficial in improving the mechanical properties like ductility and tensile
strength.
The as-cast iron develops internal stresses during the casting process, which can
be reduced by heating the casting and holding it at a suitable temperature. The time
of holding is dependent on the composition, complexity, final properties desired, and
dimensions of the cast. Stress-relieving temperatures for various irons are given in
Table 14.4.
Higher temperatures help in eliminating the stresses, but with higher tempera-
tures, hardness and tensile strength values decrease. After holding the castings at
the given temperature, they are furnace cooled and then cooled in the air to avoid
reintroduction of residual stresses.
Annealing treatment is required when good machinability and ductility is desir-
able without the need of high tensile strength – the annealing treatment results in
a microstructure with graphite nodules present in a fully ferritic matrix. For better
machinability, carbide-forming elements should be as low as possible. There are
three types of annealing treatments for different kinds of S.G. iron.
TABLE 14.4
Temperature Range of Stress Relieving for Various Types of S.G. Iron
Type of S.G. Iron Unalloyed Austenitic Low-Alloy High-Alloy
Temperature range (°C) 510–565 620–675 565–595 595–650
S.G., spheroidal graphite.

• Subcritical annealing – This treatment converts pearlite to the ferritic

phase. The presence of the carbide phase may result in a loss of ductility.
For this treatment, the temperature range is 705°C–720°C. After heating
and soaking, the castings are furnace cooled, followed by air cooling.
• Full annealing for unalloyed irons – This treatment is suitable for S.G iron
with no alloying elements and eutectic carbide present. The temperature
range is 870°C–900°C. After heating, the castings are furnace cooled till
350°C and then cooled in the air.
• Full annealing for S.G. iron–containing eutectic carbides – This treatment
is given for castings having eutectic carbides. For heavier sections, larger
soaking time is required, so that uniform temperature is achieved. After
soaking, the castings are furnace cooled to about 700°C and then held for
2 h, followed by furnace cooling to 350°C and finally air cooling to room
temperature.
Normalizing is another heat treatment process of heating S.G. iron castings to a tem-
perature range of 870°C–940°C, holding for 1 h followed by air cooling. If alloying
elements are present, then longer holding time is required. As a result of normal-
izing, tensile strength increases because of the formation of uniform and homog-
enous pearlite, and nodular graphite provided that the silicon content is low. The
cooling rate depends on the weight of the casting specimen. Generally, more heavy
castings shift the continuous cooling transformation curve toward the right so that
the formation of pearlite and nodular graphite takes place even at a slow cooling
rate. Tempering is done after normalizing in cases where martensite forms, which
ensures to relieve the stresses due to the formation of the martensitic phase and get
the desired hardness levels, tensile strength, and impact strength. Tempering is done
by heating the castings to a temperature of around 400°C–600°C, soaking for 1 h
followed by air cooling.
Hardening is the process of heating the cast to a temperature of austenitization
temperature of 845°C–925°C. As the austenitization temperature increases, the
hardness of the cast increases to the highest values up to 870°C, after which it starts
decreasing due to the presence of a large amount of retained austenite. Oil quench-
ing is generally preferred to avoid crack formation. Tempering is done immediately
after the hardening or quenching process to relieve the stresses formed during the
process at a temperature range of 420°C–600°C for at least a 1-h duration. The soak-
ing time depends on the cast dimensions. The hardness, strength, and ductility of the
tempered cast iron depend on the soaking time, composition, and temperature. The
tempering takes place in two stages. The first stage consists of the precipitation of
carbide phases followed by the second phase, which includes the nucleation followed
by growth of the graphite nodules. This leads to a decrease in hardness as well as
tensile strength.
The process of austempering for S.G. iron is similar to that of the austempering
process of steel. The method includes heating the ductile iron to an austenitization
temperature, quenching rapidly to temperature before the Ms temperature, holding it
for a while, and cooling to room temperature. Usually, after austempering in steel,
the bainite phase is formed, which is undesirable due to its low ductility and tough-
ness. However, in ductile iron, acicular ferrite is uniformly dispersed in a carbon-
rich metastable phase of austenite along with graphite nodules; this leads to good
ductility as compared to other heat treatment processes.
Surface hardening treatment is done to improve surface properties like hardness
and wear resistance. The heating time is short in these methods with little holding
times; hence, pearlitic matrix is preferred. Caution is required in surface hardening
of pearlitic irons to avoid crack formation. The cooling rate plays a vital role in these
treatments. Water quenching of pearlitic cast irons has higher hardness than cast iron
with the ferritic matrix. A fully ferritic matrix can form martensite only near the
graphitic nodules, resulting in lower hardness values. Various parameters control the
microstructure of the cast iron, such as temperature, time of holding, composition,
and rate of cooling. Polymer quenchants can be preferred instead of water quenching
to avoid the quench crack formation. The amount of alteration in the surface prop-
erties of the cast iron depends on the pearlite content. S.G. iron, which has already
been hardened and tempered, contains secondary graphite that can act as a source to
supply carbon during the surface hardening even for cast irons with ferritic matrix
up to 90%–95% extent.
As already discussed, nitriding is a type of case hardening process that involves
the diffusion process of nitrogen at a high temperature in the range of 550°C–650°C
in the presence of an ammonia atmosphere. The treatment, in this case, leads to the
formation of a thin layer with a high hardness value of 1100 VPN. The presence of
alloying elements can increase the extent of case hardening. The nitriding process is
done to improve the hardness, wear resistance, and corrosion resistance.
REFERENCE

FURTHER READING
Avner. Introduction to Physical Metallurgy. (Tata McGraw-Hill Education, New York, 1997).
Rajan, T. V., Sharma, C. P. & Sharma, A. Heat Treatment: Principles and Techniques. (PHI
Learning Pvt. Ltd., New Delhi, 2011).
15 Heat Treatment Defects
and Their Determination
Heat treatment plays an important role in improving the performance of steel by sig-
nificantly enhancing its mechanical properties, leading to a longer life span. It can be
described as the heating and cooling process that aims at improving the microstruc-
ture as well as the mechanical properties of metals and alloys. Various mechanical
components such as gears and crankshafts are subjected to continuous friction as
well as dynamic load. This requires the surface to possess high hardness and wear
resistance, while the core should have sufficient toughness. Despite the adoption
of requisite procedures of heat treatment, the heat-treated components may exhibit
undesired and detrimental properties that can seriously affect the quality of the com-
ponent in operation. If the latter occurs, then the material is said to be defective, and
such kind of defects are known as heat treatment defects.
Heat treatment defects may also arise due to unsuitable chemical composition of
the alloy and may pose a great threat during its operation, as the component may fail
anytime without any prior indication. Therefore, it is necessary to know the possible
causes of the heat treatment defects, so that it can be prevented on time. Inherent
defects in the material, inadequate design of the tool, lack of precise control over
furnace temperature and cooling/heating rates, and wrong selection of the material
may be some of the causes of heat treatment defects. Thus, preventive measures such
as proper selection of the defect free material, appropriate design of heat treatment
cycle, improved automation, and adequate tools will be beneficial to get the opti-
mized properties in the material after heat treatment.
The common types of defects that are usually observed in the heat treated com-
ponents are as follows:
• Distortion
• Warping
• Residual stress
• Quench cracking
• Soft spots
• Oxidation and decarburization
• Low hardness and strength after hardening
• Overheating
• Burning
• Black fracture
• Deformation and volume changes after hardening
• Excessive hardness after tempering
• Corrosion and erosion
197
15.1 DISTORTION
The change in the shape and size (or dimensional change) of the heat-treated com-
ponent due to thermal and structural stresses is known as distortion. Distortion can
be classified as follows:
• Size distortion: change in volume. Mainly takes place due to expansion or

contraction in the component.
• Shape distortion: change in geometrical form without any volumetric
change. This is manifested by change in curvature such as twisting or bend-
ing in the component.
Distortion mainly occurs due to the following:
• Distortion in steel mainly occurs during hardening and tempering.

• Combined effect of thermal and transformation stresses causes distortion.
• It is also influenced by the heating rate, size, shape, thickness of the wall,
and geometry of the component. Chemical and structural inhomogeneity,
cooling rate, and subzero treatment are also certain influencing parameters.
• The presence of residual stress in the component before heat treatment may
also cause distortion.
The risk or chances of distortion can be controlled by certain parameters that are as
follows:
• Design: Sharp corners or sharp projections and thin walls should be avoided
in the component.
• Composition: Proper selection of steel component in terms of its composi-
tion helps in the minimization of distortion.
• Uniform microstructure of the component and uniform temperature in the
furnace helps in the minimization of distortion.
• Machining helps in reducing the risk of distortion.
• Subcritical annealing or normalizing operation helps in relieving the inter-
nal stresses and, thus, reduces the occurrence of distortion.
• Heating rate has to be very fast.
• Preheating reduces shape distortion in steels by reducing the thermal
stresses produced due to the temperature gradient between the surface and
the core.
15.2 WARPING
Warping can be defined as the asymmetrical deformation of heat-treated component
during quenching. Nonuniform heating or cooling may lead to volumetric change,
which further causes warping. Inadequate support of the work piece at high tempera-
tures, presence of internal stresses, and inclination of the component in the quench
bath are some of the causes of warping. Figure 15.1 shows a typical warping defect
due to galvanization (submersed in a liquid zinc bath at around 450°C).
Heat Treatment Defects 199
FIGURE 15.1 Warping steel plates due to galvanization.1
The chances of warping can be minimized by the following:
• Addition of more alloying elements in the steel. This shifts the nose of
the continuous cooling temperature (CCT) curve toward the right side,
thus permitting a slow cooling rate. This reduces the generation of internal
stresses, and thus, chance of warping gets reduced.
• Slow cooling in the martensitic range.
• Appropriate surface hardening treatments also may reduce warping.
• Annealing, normalizing, and tempering at high temperature before harden-
ing also decrease the risk of warping.
• Uniform heating and quenching as well as proper alignment of the compo-
nent in quenching bath also acts as a corrective measure for warping.
15.3 RESIDUAL STRESSES
Stress may be produced during transformation of austenite to martensite. Mainly, resid-
ual stresses arise due to the temperature gradient and uneven transformation across the
cross section of a component. Residual stress can be minimized by the following:
• Slow down the cooling process so that temperature gradient can be low-
ered, and thus, transformation occurs at the same time throughout the cross
section. Thus, it reduces the residual stress generation.
• Addition of more alloying elements in the steel shifts the nose of the CCT
curve toward the right side, thus permitting a slow cooling rate.
• Residual stress can also be eliminated by stress-relieving process such as
reheating steel to a relatively low temperature, around 250°C with a suitable
holding time depending on the thickness followed by controlled cooling. At
this temperature, 80%–85% of residual stress can be eliminated.
• Tempering at high temperature can also relieve the residual stress.
15.4 QUENCH CRACKING
Quench cracks generally appear in zigzag form at the grain boundaries, and they
may be external or internal as well as small or large. They appear as straight lines
that run from the surface toward the center of the quenched specimen as shown in
FIGURE 15.2 A quench crack extending from the surface to the core.2
Figure 15.2. This may be formed due to the presence of stresses produced during
the transformation of austenite to martensite. They often appear after a steel sample
undergoes quenching. The differential cooling in the surface and core during mar-
tensitic transformation is accompanied by the volume mismatch, which results in the
generation of compressive stresses and thus, causes cracking. Generation of quench
crack is detrimental, as the steel with quench cracks cannot be used further and,
therefore, has only scrap value.
The preventive measures that can be followed include the following:
• Avoid manufacturing tools with sharp corners or sharp projections.

• There should be no residual stress, and if there is any, it should be relieved
by annealing.
• Cool the component slowly in the martensitic range.
• Can follow the route of interrupted quenching, oil quenching, austemper-
ing, and martempering.
• Immediately temper after the quenching process to avoid quench crack
formation.
15.5 SOFT SPOTS
When the hardness of the component surface is not uniform, even after the harden-
ing process, it may lead to varying hardness at different points on the surface of
hardened steels. Such kind of a defect is known as soft spots. The microstructure of
such soft spot seems as dark region in white martensitic matrix. The various reasons
pertaining to soft spots are as follows:
• Localized decarburization of steel

• Formation of vapor blanket between the quenchant and component during
quenching, which hinders rapid heat dissipation
• Inhomogeneity of microstructure
• Presence of foreign particles on the surface such as scales and dirt

• Heating of large components in furnaces, which may lead to nonuniformity
in heating of the component
• Improper handling of component during quenching
Soft spot formation can be prevented by adopting spray quenching over conventional
quenching.
15.6 OXIDATION AND DECARBURIZATION

Furnace atmospheric gases such as oxygen, carbon dioxide, and water vapor may
react with the steel surface, whereas the latter is heated at a high temperature in an
open furnace (open to the atmosphere). As a result, it gives rise to two surface phe-
nomena: oxidation and decarburization.
Oxidation of the steel occurs when it reacts with oxygen, carbon dioxide, or water
vapor. The possible series of chemical reaction are as follows:
2Fe + O 2 = 2FeO
4FeO + O 2 = 2Fe 2 O3
Fe + CO 2 = FeO + CO

3FeO + CO2 = Fe3 O4 + CO
Fe + H 2 O = FeO + H 2
3FeO + H 2 O = Fe3 O 4 + H 2
The gases also react with each other, and such kind of a reaction is known as water
gas reaction.
CO + H 2 O = CO 2 + H 2
The equilibrium relationship of iron and iron oxide with the gases at the operating
temperature determines the degree of oxidation. When there is a high content of CO
in a low content of water vapor in the furnace atmosphere, it leads to another reaction
known as producer gas reaction as mentioned in the following:
2CO = CO 2 + C
Steel behaves like a catalyst in the producer gas reaction, and thus, oxidation of
the steel surface occurs. At elevated temperatures (around 425°C), a porous and
oxide layer grows, which leads to continuous disintegration of the component. This
adversely affects the quality of the surface with consequential dimensional changes.
Decarburization involves the removal of carbon from the steel surface when
it is heated at elevated temperatures (around 650°C). Carbon reacts with oxygen
or hydrogen at this temperature, and this depth of carburization is a function of
time, temperature, and furnace atmosphere. As shown in Figure 15.3, there is a
FIGURE 15.3 Decarburized surface showing a variation in the amount and depth of ferrite
in as as-rolled eutectoid steel.3
variation in ferrite content on the surface of as-rolled eutectoid steel. This phenom-
enon gravely reduces the quenched hardness, wear resistance, and fatigue strength.
Decarburization involves the following reactions:
C + O 2 = CO 2
Fe3 C + O 2 = 3Fe + CO 2
C + CO 2 = 2CO
Fe3 C + CO 2 = 3Fe + 2CO
Fe3 C + H 2 O = 3Fe + H 2 + CO
The measures that can be followed to reduce decarburization and oxidation include
the following:
• Steel components can be heated in molten salts, vacuum, or controlled

atmosphere.
• Machining can be done to remove the decarburized surface layer.
• Steel components may be heated with carburizing agents with surface coat-
ings like borax.
• Components may be coated with ceramic material before the process of
heat treatment.
15.7 LOW HARDNESS AND STRENGTH AFTER HARDENING

Sometimes, when a material or component is heated till the austenitizing tempera-
ture and then quenched, it may not develop the desired hardness and strength. The
hardness is imparted by the martensite (hard phase) formation upon quenching.
But it requires a further heat treatment, i.e., tempering to reduce or eliminate the
brittleness of the as-quenched martensite. Tempering causes softening of the hard
phase to enhance the other mechanical properties such as toughness, formability,
and machinability. This greatly reduces the chance of crack nucleation. But there are
certain reasons that may lead to low hardness and strength after quenching, such as
the following:
• Lower hardening temperature and insufficient soaking time that may lead
to the formation of products like bainite and pearlite, which have lower
hardness value than that of martensite.
• Delayed quenching and slow cooling rate may also result the same as the
latter.
• Higher tempering temperature may induce unwanted excessive softness in
the component.
• Presence of large amount of retained (or untransformed) austenite, which is
much softer than the martensite.
In such kind of a situation, subzero treatment can be carried out at around −30°C to
−120°C, as the Mf transformation temperature may be below the room temperature.
Furthermore, improper surface hardening treatments like carburization, cyanid-
ing, and nitriding process may lead to reduced hardness. Thus, improper carburizing
atmosphere and temperature causes the aforementioned phenomena, and it can be
overcome by considering proper carburizing and postcarburizing heat treatment.
15.8 OVERHEATING OF STEEL

Overheating involves the heating of steel above the upper critical temperature for a
long period of time, which leads to coarsening of the grains. It adversely affects the
mechanical properties and may lead to premature failing of gears due to fatigue. The
optical and SEM images of a boiler steel tube are provided in Figures 15.4 and 15.5.
The cooler side, which experiences relatively lower temperature, shows conventional
lamellar pearlite structure. However, upon short-term excursion to high temperature
(~750°C), distortion of the pearlitic structure results, leading to its spherodization
resulting in poor strength.4
FIGURE 15.4 Optical microscope images of a boiler steel tube at (a) cooler side and (b)
hotter side.4
FIGURE 15.5 SEM images of a boiler steel tube at (a) cooler side and (b) hotter side.4
Effects of overheating are summarized as follows:
• Coarse-grained microstructure formation

• Widmanstätten structure formation in annealed steel
• Reduction in toughness and ductility
• Reduced hardness and coarse-grained martensite formation in the case of
hardened steels
• Scale formation and surface decarburization
• Segregation of impurities like sulfide inclusions along grain boundaries,
thus causing damage to the component
On the basis of composition, high-carbon and low-alloy steels, Ni–Cr–Mo, and high-
alloy steels are more prone to overheating as compared to mild steel. On the basis
of melting techniques, basic electric steels are more susceptible to overheating than
the steels manufactured by open hearth process. Steels with higher inclusion content
have higher overheating temperatures.
Overheated steels can be recovered by repeated normalizing process at tempera-
tures 60°C–100°C higher than the conventional temperature. Repeated hardening by
oil and tempering after prolonged soaking at 950°C–1120°C in carburizing atmo-
sphere may also help in recovering the steel from its overheated condition. But the
disadvantages of both these methods include dimension alteration and uneconomical
process.
15.9 BURNING OF STEEL

If the temperature exceeds the solidus temperature while heating the steel, liquation
may occur at the grain boundaries. In presence of an oxidizing condition, oxides may
form and penetrate the grain boundaries. This phenomenon is known as burning of
the steel, and its effect is permanent. Burnt steel becomes scrap. Effects of burning
are summarized as follows:
• Phosphorous gets segregated at the grain boundaries, and it subsequently

gets precipitated as iron phosphide. This causes embrittlement of steels.
The same kind of embrittlement may occur in the case of sulfur as well.
• High sulfur content has the ability to enhance both overheating and burning
of steels.
• The components that are burned are unserviceable and cannot be reclaimed.
It can hardly be used as a scrap.
On the basis of melting techniques, basic electric steels are more susceptible to burn-
ing than the steels manufactured by open hearth process.
15.10 BLACK FRACTURE
Black fracture can be characterized as carbon (specifically, graphite) inclusions in
the steel, which renders undesirable properties to the steel. This is mainly caused due
to excessive heating time and slow cooling after annealing. The corrective measure
of such kind of a defect includes heating of the steel to a high temperature and forg-
ing it thoroughly and uniformly.
15.11 DEFORMATION AND VOLUME

CHANGES AFTER HARDENING
Higher hardenability of the steels after the hardening process may result in severe
deformation as well as undesirable volumetric changes. Such kinds of defects are
detrimental to the mechanical properties of the component. Volumetric change, i.e.,
increase in volume, mainly occurs due to the martensitic transformation. The defor-
mation is more severe in case of higher hardenability of the steels. Uneven heating,
rapid cooling, improper support of the component in the furnace, improper dipping
in the quenching bath, and presence of residual stresses may also result in deforma-
tion and volumetric changes.
Certain measures that can be considered to prevent such kind of a defect include
the following:
• Use alloy steel that can undergo slight deformation by hardening.

• Cool the component slowly in the martensitic range.
• Apply surface-hardening treatments whenever possible.
• Perform stress-relieving treatment to reduce the residual stress.
15.12 EXCESSIVE HARDNESS AFTER TEMPERING

Insufficient or excessive holding time during tempering may lead to enhanced hard-
ness or insufficient hardness, respectively. This may occur due to low temperature
and/or insufficient soaking time while tempering. Maintenance of high temperature
during tempering may also cause such kind of a defect. This defect may also arise if
the hardening temperature is too low or if the cooling rate is too low.
Such kind of a defect can be prevented by the following:
• A second tempering treatment with proper temperature and soaking time

• Subsequent normalizing or annealing followed by hardening
15.13 CORROSION AND EROSION

Pitting-type corrosion of the component or localized corrosion may occur during
heat treatment due to the presence of high content of sulfuric salts (0.7%–0.8%) in
the molten salt bath. Additionally, when the bath is rich in oxygen or iron oxides, it
causes pitting corrosion of the component. The same conditions may also cause ero-
sion. Erosion can be defined as the reduction in size of the component due to the loss
of material from its surface.
Thus, its preventive measures are as follows:
• Controlling the salt composition in the bath

• Deoxidizing the bath
REFERENCES

FURTHER READING
Rajan, T. V., Sharma, C. P., & Sharma, A. Heat Treatment: Principles and Techniques. (PHI
Learning Pvt. Ltd., New Delhi, 2011).
Sinha, A. K. Defects and distortion in heat-treated parts. ASM Handb. 4, 601–619 (1991).
16 Some Special Heat
Treatment Practices
In present days, steel is an irreplaceable material being used in a wide spectrum
of applications. However, the requirements for various applications may differ in a
gross scale. Hence, appropriate selection of steel is the key to achieve the desirable
in-service performance. It is well established, at this point of time, that heat treat-
ment of steel plays a vital role in altering its properties. Hence, a well-planned heat
treatment cycle must be designed in order to comply with the stringent application
requirements. In this chapter, a brief knowledge has been shared about the type of
steels and respective heat treatment operations carried out by different industries.
16.1 AUTOMOBILE INDUSTRIES
Heat treatment is a key technology to enhance the operative use of materials and to
achieve the desirable combination of properties used in the automobile components.
The manufacture of automobiles involves the utilization of a variety of materials
and technologies. Vehicle weight reduction via the use of high-strength and light-
weight structural materials is a necessity in the present day as it aims towards energy
conservation. Though, aluminum and plastics have seen an increase in use, ferrous
alloys constitute about 70% of a current vehicle. Some of the steel components used
in automobiles are shown in Figure 16.1. Metallic materials are preferably matched
for specific applications where the material is expected to work in a very reliable
way even under heavy stresses. Effective heat treatment is necessary for optimizing
properties of nearly all types of metallic components with durability being a feature
of prominence in a great number of applications. Starting with raw metal products
to final component assembly, with various stages of processing, several types of
heat treatments are performed in the manufacturing of different types of automotive
components. Appropriate heat treatment processes render the required combination
strength and/or hardness with toughness, which is dictated by the respective compo-
nents’ applications.
Figure 16.1 shows the various automotive components utilizing iron and steel as
the building material. The 70% steel components used in automobiles include the
drive shafts, camshafts, crankshafts, cylinder heads, bearings, valves/outlets, rocker
arms, suspensions, gears, clutch parts and many more. Specific heat treatments for
parts are as follows:
16.1.1 drIvesHafTs/JoInTs
Driveshafts are the automotive components where safety has top priority as it func-
tions to transfer torque from the gearbox to the wheels (Figure 16.2). Heat treatment
207
FIGURE 16.1 Some components of a vehicle.1
FIGURE 16.2 Driveshafts as used in automobile applications.2
is necessary to improve service life of the component and to counteract wear. Case-
hardening treatments like vacuum hardening (vacuum carburizing) or gas nitriding
as well as plasma nitriding are suitable.
16.1.2 suspensIon
Suspensions connect each wheel to the body of the vehicle (Figure 16.3). Shock
absorbers, handlebars, wheel carriers, joints and springs are the various components
of the suspension system. To ensure longevity of these components case-hardening
treatments like inert gas hardening and vacuum hardening are the heat treatments
performed.
16.1.3 cluTcH
Clutch acts as the link between the power train and the drive train (Figure 16.4) and
thereby, plays a crucial role in the transfer of power and torque from the engine to the
wheels. Clutch plates are subjected to high wear stresses over frequent gear changes
Some Special Heat Treatment Practices 209
FIGURE 16.3 Suspension shock absorber used in automobile applications.3
FIGURE 16.4 Clutch as used in automobile applications.4
and are the most likely to be worn out the earliest. Good strength and toughness are
vital attributes given the stresses to which these components are subjected to. Plasma
nitriding is a suitable heat treatment process.
16.1.4 crank sHafTs

Crank shafts are responsible for the conversion of linear motion of pistons to rotary
motion (Figure 16.5). It ensures that this component is subject to tensile, compres-
sive, and radial forces. Sufficient bending and torsional strength of the shaft and
good loading capacity of the bearing points thus becomes a necessity. Gas nitriding,
nitrocarburizing, plasma nitriding, and case hardening are suitable thermochemi-
cal processes. Nitrogen addition leads to increase in strength and development of
compressive forces. This results in increased load-bearing capacity. Carburization of
surface layer is chiefly used in racing cars.
FIGURE 16.5 Crank shaft as used in automobile applications.5
16.1.5 BearIngs
Bearings transmit forces and movement between machine parts, as they slide or
roll over each other. These components are subjected to high loads. To limit resis-
tance generated during loading gas nitriding, plasma nitriding, case hardening, and
vacuum hardening are the treatment used to increase the strength.
16.1.6 gearWHeels/planeTary gears

Gearwheels (Figure 16.6) are used to transmit torque. Gears stand a chance to shat-
ter under extreme loads. In order to show elastic behavior throughout life cycle and
withstand load shocks surface strength and inner toughness of the gears are required.
Case hardening and nitriding treatments improve strength and wear behavior.
The various types of heat treatments used to treat automotive parts are given in
Table 16.1.
16.2 AEROSPACE INDUSTRIES
Heat treatment process plays an important and is frequently used to alter the prop-
erties of materials used for aerospace industries. Aerospace industry requires high
quality and precise dimension requirements for the components used. Materials used
in this sector are required to have high strength along with good fatigue life. Stress
reduction is required for achieving appreciable strength and fatigue life. This will
help in ensuring that the materials tolerate extreme temperature and pressure condi-
tions without failing, also precise and specific dimension requirements for the mate-
rials and components are there, which requires the materials to sustain these extreme
conditions without changes in their shape or size. Heat treatment proves to be an
essential step in this regard.
FIGURE 16.6 Schematic presentation of gears as used in automobile applications.6
TABLE 16.1
Various Automotive Components and Respective Heat Treatment Processes7
Types of Heat Treatment Purpose Typical Components
Annealing Softening and removing residual Forged blanks for gearing
stress and imparting substantial and miscellaneous parts
ductility for post processes
Quenching and tempering Optimized combination of Fasteners, rods, and arms
hardness for strength and
toughness
Austempering Optimize microstructure and Cast iron brackets,
hardness via isothermal high-carbon springs
transformation
Case hardening: For improving fatigue strength Gears and shafts, oil pump
Carburizing and wear resistance gears, brake pad liner
Nitriding plates, cam shafts
Carbonitriding
Nitro-carburizing
Oxy-nitro-carburizing
Commonly heat-treated components for aerospace industries include engines,

frame parts, cooling systems, and hydraulic fittings. Heat treatment processes and
techniques include a number of methods that improve the properties of the material
used in this industry namely, metals and alloys along with extending the life of these
components. These treatments along with the aerospace industry can be used also in
aviation, commercial, and business aircraft carriers and helicopters. Different heat
treatments can be given depending on the metals and alloys used and according
to the final property requirement of the end product. Aerospace industry is a vital
field of research where new engine designs, various R&D process, new materials
performance research along with testing of the joining systems, and novel materials
are highly dependent on the production technique and the post processing methods
given to it. Controlled atmosphere heat treatments like case hardening, tempering,
etc. are commonly done for components like flap tracks, landing gear parts, and air-
frame during the manufacturing process. The nitriding of engine components such
as gears leads to improved and desirable properties suiting to aerospace industry.
The key heat treatment processes are as follows:

a. Ageing

c. Homogenizing

e. Plasma nitriding

a. Carbonitriding

c. Case hardening

e. Hardening and tempering

In aerospace industry, safety of the passengers and the pilot is utmost important for
which the components should be reliable and as per the safety regulations. Hence,
reliable operations of the components are given to components to avoid fatal results.
Following are the components where heat treatment is commonly done.
16.2.1 seaT eJecTors

Ejector seats (Figure 16.7) for fighter aircrafts are manufactured to be as precise and
as close to the intended properties as possible. Heat treatment of these ensures that
this essential equipment operates properly without any malfunctioning. A countless
number of lives of pilots are saved and will continue to be saved by the efficient
deployment of these seats in case of any uncontrolled flights and accidents. Seat ejec-
tors are small yet an important part of these aircrafts.
16.2.2 TurBIne Blades

Aircraft blades (Figure 16.8) are subjected to extreme temperatures and conditions
while they are in operation. These components operate frequently at temperatures
tending to their melting point and hence can cause failure of these components while
FIGURE 16.7 Seat ejectors as used in aircrafts.8
FIGURE 16.8 Turbine blades as used in engines of aircrafts.9

in use. Hence, to avoid early failure, heat treatment operations along with other suit-
able methods and techniques are prominently used allowing these turbine blades to
withstand these high temperatures for prolonged period of time.
16.2.3 landIng gear

Safety components like aircraft landing gear (Figure 16.9) are expected to perform
well without any fault and should withstand environmental effects and avoid failure
mechanisms like fatigues every time the aircraft takes a flight. Several combinations
of heat treatment processing techniques are given to ensure the properties of the steel
are optimized which can help to protect it during the service life. Landing gears were
used to be surface-treated using hard chrome plates which are now replaced by other
environmentally friendly processes and thermal spraying techniques. These help the
landing gears to withstand wear and corrosion during their working life.
16.3 MEDICAL EQUIPMENT
Various medical equipment and supplies namely medical implants, dental tools, sur-
gical devices, and so on are frequently heat-treated and given thermal processing to
tailor the required specific properties of these equipment. Commonly used mate-
rial for these medical equipment includes stainless steel which can be given ideal
heat treatments from plenty of other available options. Medical industry demands in
FIGURE 16.9 Landing gear in aircrafts.10

choosing the right material with specific properties like good mechanical strength
and corrosion resistance along with biocompatibility. Biocompatibility is extremely
essential when medical industry is involved. Material lacking in this property can
lead to compromise of the life and health of people rendering these materials. Other
metals, namely, titanium are also used for hip implants that need to be given special
treatments to achieve stringent properties and proper finish.
16.3.1 surgIcal equIpmenT

Surgical equipment falls under the category of nonimplant medical devices. This
also includes dental instruments, surgical staples, guide pins, needs, trays, and stor-
age cabinets. The materials used in these applications should have good corrosion
resistance, good mechanical properties, formability, and biocompatibility. Along
with biocompatibility, surgical equipment also need good corrosion resistance to
withstand the attack of bodily fluids and inhibit reactions avoiding formation of
harmful product. Surgical equipment are typically made from 420 stainless steel
grade which is a high-carbon steel that can be heat-treated leading to hardening.
This material exhibits excellent corrosion resistance properties upon hardening and
offers appreciable ductility upon annealing condition. In hardened state, it can with-
stand water, alkali, and mild acid attacks; however, in annealed state, the corrosion
resistance tends to decrease.
16.3.2 medIcal devIce fasTeners

Fasteners are used in a number of applications in the medical industry namely in the
dental and orthopedic implants. They help to keep all the different components and
parts of the implant together in an assembly. These components and parts have dif-
ferent shapes and sizes. Usually, these components in medical industries are small
and tiny, but failure of a fastener will result in failure of the whole implant. This
needs the fastener to be suitable, precise, and compatible with other equipment and
parts along with good mechanical properties and biocompatibility. Only then, the
fastener can be expected to perform as desired. Choice of suitable fastener can cut
costs of the whole implant, improving the overall quality of the final product.
16.3.3 ImplanTs
Surgical implants are often made from specific grades of stainless steel, i.e. austen-
itic type. Examples are bone plates, screws, femoral fixation device, clips, nails, and
pins. Titanium is also gaining major interest of the researchers for use in implants
along with cobalt-based alloys. For dental and hip implants, chemical bonding
between the implant and the body/bone does not occur. This is because of the lack
of formation of chemical bond between them. Hence, the surface of the implant is
made with high porosity such that bone tissues can grow into the pores leading to a
stable fixture and integration. This is done by surface modification of the implant,
which creates a good connection between the implant and the body. Various metals
used for different implants are tabulated in Table 16.2.
TABLE 16.2
Various Metals Used in Different Implants11
Type of Implant Specific Use Metal Used
Cardiovascular Stents, artificial wires 316L SS, CoCr, Ti
Otorhinology Artificial eardrum 316L SS
Craniofacial Plate and screw 316L SS, CoCrMo, Ti
Orthopaedic Bone fixation, artificial joints 316L SS, Ti, CoCrMo
Dentistry Orthodontic wire, fillinf 316L SS, CoCrMo, TiMo
16.4 DEFENSE INDUSTRIES
The standards of materials used for mission-critical applications in the defense
industry are very high. When working with a metallic material, precision and qual-
ity are considered two of the most important factors for manufacturers in the defense
industries. Before adding to the defense equipment, it is essential that the metal parts
be treated correctly. Tanks, fighter jets, stealth bombers, and helicopters all need
precision-treated metallic components.
16.4.1 armor of comBaT veHIcles

The armor of combat vehicles needs to be light in weight and has great ballistic
protection. Ultra-high strength (UHS) steel is the most preferred metallic armor
used today due to the properties of good toughness with high strength and hardness
and ease of heat treatment that they possess. Quenching and tempering produce
strengthening and increase impact toughness of UHS steel. Martensitic structure
formed during quenching has the highest strength, but internal stresses developed
during transformation limit its use in untempered condition. To boost absorption
of impact energy, the ductility and toughness of material need to be increased.
Tempering serves this purpose.
Austenitization of UHS steel is carried out at 910°C for 1 h followed by oil or
water quenching. Tempering up to 600°C is done, with different tempering tempera-
tures producing varying properties. Table 16.3 gives the values of various mechani-
cal properties of UHS steel tempered at different temperatures after quenching.
From the varying mechanical properties, suitable tempering temperature for
specific applications may be used.
16.4.2 fIrearms
Different types of firearms are used for various purpose nowadays. Warping of
barrels is a problem in rifles, pistols, and shotguns. Warping occurs due to continu-
ous stress during constant firing that heats up the barrel. Heat treatment processes
such as case hardening, tempering, gas nitriding, and surface hardening increase
performance and accuracy of firearms. Maintenance and cleaning becomes
easier as well.
TABLE 16.3
Effects of Tempering Temperature on the Mechanical and Ballistic
Properties12
Tempering Yield Strength UTS Impact YS/UTS Hardness %Reduction
Temperature (°C) (MPa) (MPa) (VHN) (J) (mm) %Elongation
0 1367 1900 0.72 586 44 9
200 1417 1808 0.78 555 48 12
300 1463 1700 0.86 518 43 7.5
400 1433 1587 0.90 490 48 10
500 1286 1409 0.91 442 53 12
600 1146 1247 0.92 381 60 16
16.4.3 mIssIles
Missiles are normally made of materials having very high resistance to environ-
mental conditions. Thermal treatment is necessary in the manufacturing of missile
systems. All the components starting from the mounting hardware to the guidance
systems require treatment. Hardening, induction heat treatment and age hardening
are treatments that can be performed on missile motor cases, wing fins, mounting
hardware, and guidance components.
REFERENCES

Taylor & Francis
Index
A Close Packed Hexagonal (CPH), 6, 12–13
Clustering, 61
Age hardening, 173, 217 Clutch, 208–209
Ageing, 212 Coarsening, 108, 149
Allotropic transformation, 76, 80 Cold rolling, 1, 67, 177
Alloy carbides, 140, 144, 150–151, 166 Cold working, 138
Alloying elements, 86–89, 110, 113–115, 154, Component, 211
157, 162, 165, 172–173 Configuration, 2–4
Annealing, 133–142, 167 Configurational entropy, 56
Annealing twin, 48–49 Continuous cooling transformation (CCT), 102
Atomic bonding, 54 Continuous furnace, 123
Atomic diameter, 47–48 Contracted γ-field, 90
Atomic number, 1–2 Conveyor furnace, 123
Atomic weight, 1 Coordination number, 10–11, 30, 47
Ausforming, 165–166 Corrosion, 206
Austempering, 151, 195 Covalent bond, 4–5
Austenite, 26–27, 83 Crank shafts, 209
Austenite stabilizer, 88, 169 Creep, 191
Austenitic alloys, 88 Critical cooling rate, 105, 141
Austenitic stainless steel, 171 Crystallographic direction, 47
Austenitization, 102, 193 Crystallographic plane, 32
Automobile industries, 179, 207 Crystal structure, 5–6, 13–14
Crystal system, 6–10
B Cubic, 9
Bainite, 111–113 Curie temperature, 80, 86
Bainitic transformation, 111
Bainitic zone, 111 D
Batch furnace, 118
Bearing, 209–210 Degree of freedom, 65
Bell furnace, 121–122 Dendrite, 186–187
Black fracture, 205 Density, 5, 115
Body centred cubic (BCC), 6, 11 Density of dislocation, 115, 161
Bogie hearth furnace, 121 Diffusion, 21
Box-type batch furnace, 119 Diffusionless transformation, 113
Brass, 52 Dislocation, 42–46
Bravais lattice, 6–10 Dispersion, 141, 151
Brine, 146–147 Distortion, 198
Burgers vector, 42–46 Driveshaft, 207–208
Burning of steel, 204 Dual-phase steel, 173
Ductile fracture, 154
C Ductile iron, 190–191
Ductile tempered martensite, 182
Carbide, 82–89 Duplex stainless steel, 182–183
Carbonitriding, 162–163
Carburizing, 158–161
E
Case hardening, 158
Cast iron, 185 Edge dislocation, 42–43
Cementite, 82–83 Elastic strain energy, 46
Climb, 46 Electrical conductivity, 4, 37
Closed γ-field, 89 Electrical neutrality, 1, 3
Close packed, 6, 11–12 Electrical resistance furnace, 118
219
220 Index
Electric arc, 185 Homogenization, 27, 108

Electric grade steel, 177 Homogenous Nucleation, 91
Electron to atom ratio, 53 Hot worked steels, 135
Enthalpy of formation of a solid solution, 54 Hume-Rothery rules, 53, 66, 71
Entropy of formation of a solid solution, 56 Hypereutectic, 73–74, 85
Equilibrium, 63–65 Hypereutectoid, 86, 134–135
Erosion, 206 Hypoeutectic, 72–74, 85, 186
Eutectic phase, 72–73 Hypoeutectoid, 86–87, 134
Eutectic phase diagram, 75
Eutectic transformation, 72–73 I
Eutectoid composition, 87–88, 102
Eutectoid steel, 103–104, 107–108 Implants, 215
Eutectoid temperature, 85–87, 102 Inclusion, 157, 204
Expanded γ-field, 88–89 Induction furnace, 117–118
Induction hardening, 192
F Intergranular corrosion, 169
Interlamellar spacing, 109
Face centred cubic (FCC), 10 Intermediate phase, 61
Fatigue, 138, 153 Intermetallic, 173
Ferrite, 76, 85 Interstitial atom, 21–22
Ferrite stabilizer, 88 Interstitial diffusion, 21–22
Fick’s first law, 24 Interstitial site, 51–52
Fick’s second law, 25 Interstitial solid solution, 42, 51–52
Flame hardening, 163–164 Invariant, 68, 82
Fracture, 154 Isoforming, 165–167
Free energy, 28, 58 Isomorphous, 66–67
Full annealing, 134 Isothermal, 104, 106
Isotope, 1
G
J
Gamma iron, 80, 113
Gas carburizing, 130, 163 Jominy curve, 157
Gearwheels, 210 Jominy end quench test, 156
Gibb’s Phase rule, 64
Glide, 44–45 L
Grain, 47
Grain boundary, 47–48 Lamellar, 107, 111
Grain growth, 108, 134 Lath martensite, 115
Grain size, 108 Laves, 61–62
Graphite, 82 Ledeburite, 83, 186
Graphite flake, 185 Lever rule, 65
Graphitization, 188 Liquidus, 68, 72–73
Gray cast iron, 188 Lower bainite, 112
Grossman’s critical diameter, 155
Growth kinetics, 98 M
H Malleable cast iron, 188

Manganese Steel, 172
Hardenability, 153 Martensite, 113–115
Hardness, 115, 141 Martensitic transformation, 113
Heat affected zone (HAZ), 179 Medical equipment, 214
Heterogeneous Nucleation, 95 Metallic bond, 5
Hexagonal, 6–7 Metastable, 82, 105
Hexagonal Close Packed (HCP), 11, 49 Mf temperature, 114
High Strength Low Alloy (HSLA) steels, 172 Microstructure, 64, 73, 85, 140, 200
Index 221
Mild steel, 164, 172, 204 Recrystallization, 137, 138

Missiles, 217 Residual stresses, 199
Monoclinic, 7 Rotary hearth furnace, 123
Morphology, 109, 111, 164
Morphology of martensite, 115
Ms temperature, 114 S
Muffle furnace, 119 Salt bath furnace, 125
Screw dislocation, 44
N Seat ejector, 212
Nickel steels, 167–168 Secondary austenite, 183
Nodular cast iron, 190 Secondary graphite, 196
Nodular iron, 185 Secondary hardening, 150–151
Nodules, 109–110, 189–190 Sensitization, 171
Normalizing, 141 Sigmoidal curve, 110
Nucleation, 91 Slip, 42–46
Nucleation barrier, 94 Solid solution, 51
Solidus line, 68
O Spheroidal graphite (S.G.) iron, 190
Stabilized austenite, 115
Octahedral voids, 80–82 Stainless steel, 169
Open γ-field, 88 Strain energy, 81
Order - disorder transformation, 76 Strain hardening, 180
Orthorhombic, 8–9 Strain-induced transformation, 172
Overheating, 203 Strain rate, 49
Oxidation, 201 Stress- relieving annealing, 138
Stretcher-strain, 173
P Substitutional solid solution, 52
Surface hardening, 158, 196
Pack carburizing, 119
Surgical equipment, 215
Packing factor/efficiency, 14
Partial annealing, 137
Partially soluble, 67, 75 T
Patenting, 139
Pearlite, 102–104 Temperature-time-transformation (TTT)
Pearlitic transformation, 108 diagram, 102–106
Peritectic phase diagram, 75 Tempering, 148
Peritectic reaction, 83 Tetragonal, 9
Peritectoid reaction, 76 Tetragonality, 113, 149
Phase diagram, 63 Tetrahedral, 14, 80–82
Phase rule, 64–65 Thermomechanical treatment, 164
Plate martensite, 115 Tie line, 68–70
Polymer quenchant, 145, 196 Tilt boundary, 48
Precipitation, 100–101, 150, 172 Tool steels, 173
Proeutectic, 73–74, 85 Toughness, 137, 148–151
TRIP (Transformed Induced Plasticity) steels,
172
Q
Tunnel furnace, 124
Quenchant, 145, 147 Turbine blade, 213
Quench cracking, 199 Twin boundary, 48
Quenching, 145 Twist boundary, 48
R U
Raoult’s law, 71 Upper Bainite, 111
Recovery, 137, 149 Upper critical temperature, 86, 134, 137
222 Index
V Y
Vacancy, 21–22, 37 Yield point, 173
Van der Waals force, 5 Yield strength, 79, 142, 172
Void, 40–41
Z
W
Zener’s theory, 100
White cast iron, 186
Work hardening, 172

Phaset 1

Uploaded by

Copyright:

Available Formats

Phaset 1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Phaset 1

Uploaded by

Copyright:

Available Formats

Phase Transformations and

Heat Treatments of Steels

Bankim Chandra Ray, Rajesh Kumar Prusty, and

and by CRC Press

© 2020 Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, LLC

ISBN: 978-0-367-02868-8 (hbk)

Chapter 1 An Introduction to Metals .................................................................... 1

Chapter 3 Defects in Crystalline Solids .............................................................. 37

Chapter 4 Solid Solutions.................................................................................... 51

Chapter 5 Phase Diagrams and Phase Transformations ..................................... 63

5.6 Roles of Defects and Diffusion ............................................... 77

Chapter 6 Iron–Carbon Phase Diagram ............................................................. 79

Chapter 7 Thermodynamics and Kinetics of Solid-State Phase Transformation .....91

Chapter 8 Phase Transformation in Steels ........................................................ 107

Chapter 9 Heat Treatment Furnaces ................................................................. 117

Chapter 10 Heat Treatment Atmosphere............................................................. 127

Chapter 11 Common Heat Treatment Practices.................................................. 133

Chapter 12 Special Steels.................................................................................... 169

Chapter 13 Some In Situ Postweld Heat Treatment Practices ............................ 179

Chapter 14 Heat Treatment of Cast Iron ............................................................. 185

Chapter 15 Heat Treatment Defects and Their Determination ........................... 197

15.5 Soft Spots...............................................................................200

Chapter 16 Some Special Heat Treatment Practices...........................................207

Index ...................................................................................................................... 219

phase transformations. Chapter 6 discusses, in rich detail, the iron–carbon phase

Rajesh Kumar Prusty, PhD, is an enthusiastic and

Deepak Nayak is a scientist at the CSIR-Institute of

1.1 ELEMENTS, ATOMS, AND ISOTOPES

1.1.1 Types of elemenTs

1.2 TYPES OF BONDING BETWEEN ATOMS

1.2.1 IonIc Bond

1.2.2 covalenT Bond

1.2.3 meTallIc Bond

FIGURE 1.3 Illustration of metallic bonding.

1.2.4 van der Waals forces

TABLE 1.1 (Continued )

TABLE 1.1 (Continued )

TABLE 1.1 (Continued )

1.3.1 face-cenTered cuBIc sTrucTure

1.3.2 Body-cenTered cuBIc sTrucTure

1.3.3 close-packed Hexagonal

1.3.5 packIng facTor/effIcIency

Total volume of the atoms enclosed in an unit cell

No.of effective atoms = ∑C n

Ci = contribution of the ith type of atoms to the unit cell, and

In this way, let us find out the PF of the crystal structures.

1.3.5.1 Packing Fraction of a Body-Centered Cubic Structure

No. of effective atoms = ∑C n = C

Hence, for a BCC structure, the packing efficiency is 0.68 or 68%.

1.3.5.2 Packing Fraction of a Face-Centered Cubic Structure

Assuming atoms to be hard spheres, the packing factor now is

1.3.5.3 Packing Fraction of a Close-Packed Hexagonal Structure

Thus, PF = 0.74 or 74%

2.1 ATOMIC DIFFUSION MECHANISMS

FIGURE 2.1 A schematic representation of interstitial diffusion mechanism on a FCC (100)

FIGURE 2.2 Movement of an atom by vacancy diffusion in an FCC lattice. FCC,

2.2 TYPES OF DIFFUSION

Hence, from equations 2.4 and 2.5, we have

15.5 Soft Spots...............................................................................200