You are on page 1of 154




Кафедра металловедения и физики прочности

В.Ю. Турилина

Механические свойства металлов.
Термическая обработка металлов.
Специальные стали и сплавы

Учебное пособие
Под редакцией профессора С.А. Никулина

Допущено учебно-методическим объединением по образованию

в области металлургии в качестве учебного пособия
для студентов высших учебных заведений,
обучающихся по направлению Металлургия

Москва 2013
УДК 669.017

д-р техн. наук, проф. С.В. Добаткин (ИМЕТ им. А.А. Байкова РАН)

Турилина, В.Ю.
Т86 Материаловедение : механические свойства металлов. Тер-
мическая обработка металлов. Специальные стали и сплавы :
учеб. пособие / В.Ю. Турилина ; под ред. С.А. Никулина. – М. :
Изд. Дом МИСиС, 2013. – 154 с.
ISBN 978-5-87623-680-7

Пособие содержит материал, необходимый для самостоятельной подго-

товки иностранных студентов к лекциям и практическим занятиям по дисци-
плинам «Теория и технология термической обработки металлов», «Специ-
альные стали и сплавы», «Механические свойства металлов», «Конструкци-
онные материалы». Рассмотрены следующие разделы: деформация, разруше-
ние и механические свойства; фазовые и структурные превращения при на-
греве и охлаждении; основные виды термической обработки; подробно рас-
смотрены основные виды специальных сталей и сплавов и области их приме-
нения в технике. В каждом разделе пособия приведены вопросы для само-
проверки освоения материала.
Этот материал даст студентам целостное представление о процессах, про-
исходящих в сталях при термическом и деформационном воздействии, о взаи-
мосвязи структуры и свойств, об основных принципах легирования сталей и
сплавов, о способах обеспечения требуемой структуры и комплекса свойств
методами термической обработки в сталях и сплавах различного назначения.
Пособие предназначено для бакалавров и магистров, обучающихся по
направлениям 150100 «Материаловедение и технологии материалов»,
150400 «Металлургия», 011200 «Физика».

УДК 669.017

ISBN 978-5-87623-680-7 © В.Ю. Турилина, 2013


Department of Physical Metallurgy and the Physics of Strength

V.Yu. Turilina

Material Science
Mechanical properties of metals.
Heat treatment of metals.
Special steels and alloys

Edited by professor S.A. Nikulin


Moscow 2013

Dr. Sc., Professor S.V. Dobatkin (IMET RAN)

Turilina, V.Yu.
Material science : mechanical properties of metals. Heat treatment
of metals. Special steels and alloys : textbook / V.Yu. Turilina ;
edited S.A. Nikulin. – М. : Publishing House "MISiS", 2013. – 154 p.

The textbook contains the material needed for self-training foreign students
to lectures and practical classes on academic disciplines "Theory and technology
of heat treatment of metals", Special Steels and Alloys", "Mechanical properties
of metals", "Engineering Materials". It includes the following sections:
deformation, fracture and mechanical properties, phase and structural
transformations during heat treatment, the main types of heat treatment, the main
types of special steels and alloys and their applications in engineering. Each
section of the book contains questions for self-learning.
This material will provide students with a complete picture of the processes
occurring in the steel under thermal effect and deformation, the relationship
of structure and properties, the basic principles of alloying steels and alloys, how
to provide the required structure and properties by heat treatment in steels and al-
loys for various purposes.
The textbook is intended for undergraduate and graduate students in areas of
150100 "Materials Science and Technology of Materials", 150400 "Metallurgy",
011200 "Physics".

Part 1 Mechanical properties of metals 6
1.1 Basic concepts and definitions 6
1.2 Deformation and fracture 9
1.2.1 Elastic deformation 9
1.2.2 Plastic deformation 14
1.2.3 Strengthening of metals 16
1.2.4 The fracture of materials 18
1.3 Mechanical testing of metals 28
1.3.1 Classification of mechanical tests 29
1.3.2 The main types of mechanical tests 29
Part 2 Management structure during heat treatment of steel 62
2.1 Basic concepts and definitions 62
2.2 Transformations in steel during heating and cooling 70
2.3 Heat treatment 87
2.3.1 Classification of heat treatment 87
2.3.2 First-order annealing 88
2.3.3 Second-order annealing 91
2.3.4 Hardening (quenching) 95
2.3.5 Tempering of steel 98
2.3.6 Precipitation hardening 101
2.3.7 Surface hardening 101
Part 3 Special steels and alloys 102
3.1 Ferrous alloys 105
3.1.1 General classification 106
3.1.2 Designation of steels 110
3.1.3 Constructional steels 110
3.1.4 Tool Steels 126
3.1.5 Steels with special physical properties 138
3.2 Non-ferrous alloys 147
References 153

1.1 Basic concepts and definitions
In the course of operation or use, all articles and structures are sub-
ject to the action of external forces which create internal stresses in the
metal. And that is inevitably cause deformation.
To keep these stresses, and, consequently, deformations within
permissible limits (to prevent structural failure), it is necessary to select
suitable materials for the components of various designs and to apply the
most effective heat treatment.
In the production of any product or design of metal is very impor-
tant to know the basic characteristics of workpieces and fabricated metal
product (strength, stiffness, hardness, toughness, and ductility).
Mechanical properties – these are the characteristics that define the
behavior of metal under the applied external forces. Mechanical properties
of metals used in the manufacture of various products and designs, are
determined by mechanical testing.
Mechanical tests are those in which specially prepared specimens
(test pieces) of standard form and size are tested on special machines.
A result of mechanical testing is numerical values of mechanical
properties, i.e. values of stress or strain, at which changes the physical and
mechanical condition of the material.
The mechanical properties are about the behavior of materials when
subject to forces. When a material is subject to external forces, then inter-
nal forces are set up in the material, which oppose the external forces.
The material can be considered to be rather like a spring. A spring,
when stretched by external forces, sets up internal opposing forces which
are readily apparent when the spring is released and they force it to con-
tract. A material subject to external forces which stretch it is said to be in
tension (Figure 1.l, a). A material subject to forces which squeeze it is
said to be in compression (Figure 1.1, b). If a material is subject to forces
which cause it to twist or one face slide relative to an opposite face then it
is said to be in shear (Figure 1.1, c).
An object, in some situations, can be subject to both tension and
compression, e.g. a beam (Figure 1.2) which is being bent, the bending
causing the upper surface to contract and so be in compression and the
lower surface to extend and be in tension.

Figure 1.1 – Scheme of load application:
a – tension; b – compression; c – shear

Figure 1.2 – Scheme of load application: bending

Stress and strain

In discussing the application of forces to materials an important as-
pect is often not so much the size of the force as the force applied per unit
area. Thus, for example, if we stretch a strip of material by a force P ap-
plied over its cross-sectional area F (Figure 1.3), then the force applied
per unit area is F / A.

Figure 1.3 – Scheme of the force applied per unit area

The term stress, symbol σ, is used for the force per unit area:
σ = P / F. (1.1)
Stress has the units of pascal (Pa), with 1 Pa being a force of 1 new-
ton per square metre, i.e. 1 Pa = 1 N/m2.
The stress is said to be direct stress when the area being stressed is
at right angles to the line of action of the external forces, as when the ma-
terial is in tension or compression. Shear stresses are not direct stresses
since the forces being applied are in the same plane as the area being
stressed. The area used in calculations of the stress is generally the origi-
nal area that existed before the application of the forces. The stress is thus
sometimes referred to as the engineering stress, the term true stress being
used for the force divided by the actual area existing in the stressed state.
When a material is subject to tensile or compressive forces, it
changes in length (Figure 1.4).

Figure 1.4 – Change in the length of the sample:

a – tensile strain; b – compressive strain
The term strain, symbol ε, is used for: Strain = change in length /
original length or
ε = Δl / l0, (1.2)
where Δl – change in length, mm;
l0 – original length, mm.
Since strain is a ratio of two lengths it has no units. Thus we might,
for example, have a strain of 0.01. This would indicate that the change in
length is 0.01 × the original length. However, strain is frequently ex-
pressed as a percentage:
Strain as ε % = (change in length / original length) · 100 %.
Thus the strain of 0.01 as a percentage is 1 %, i.e. this is when the
change in length is 1 % of the original length.

1.2 Deformation and fracture

1.2.1 Elastic deformation
Elasticity modulus
Deformation is the ability of a material to change its shape and size
under stress-loading. Elastic deformation is the material ability to return to
its shape and size after loading is removed. If the material changes its shape
after removal of loading then this kind of deformation is called plastic.
Material response at elastic deformation is well described by Hooke
law that determines direct proportion of stress and elastic deformation.
Figure 1.5 shows elastic range of dependence “strain – stress” for such
different types of deformation as tension, torsion (shear) and hydrostatic
compression. Slopes of these lines characterize the E, G, K, respectively.
So E is called Young’s module, G – shear module, and K – bulk (volume
elasticity) module. Elasticity modules define increasing stress intensity in
the process of elastic deformation.

Figure 1.5 – Elasticity modulus

Elasticity modules are related with stress and strain by following
E = S / e, (1.3)
G = t / g, (1.4)
K = P / χ, (1.5)
where S and t corresponds to normal and tangent stresses respectively;
P – hydrostatic pressure;
e and g corresponds to strains;
χ – relative reduction of volume.
If we consider atomic conception of elastic deformation then its
mechanism is implemented in atoms reversible movements from equilib-
rium position in crystal lattice. An increase in atomic displacement leads
to elastic deformation growth. Normally elastic deformation in metals can
not be large (for example less then 0.1 % of relative elongation) because
atoms are able to move reversibly just for a very small distance.
Physically, elasticity module characterizes the material resistance to
atoms displacement from equilibrium positions and as a result to elastic
deformation, elasticity modules define stiffness of the materials.
In no stress condition metal atoms oscillate near equilibrium posi-
tions in crystal lattice. Transaction forces between adjacent atoms on the
one hand are made up of attraction forces between positive ions and elec-
trons and on the other hand – repulsion forces between ions due to the
electron covers deformation. Figure 1.6 shows the curves which character-
ize transaction forces dependence on a distance between atoms. Curve 1
corresponds to repulsion forces, curve 2 – attraction forces and 3 – out-
come curve. This shows that the repulsion forces grow with the approach
of atoms to each other. Naturally attraction forces smoothly decrease with
the increase in the distance between atoms.

Figure 1.6 – Transaction forces dependence on a distance between atoms

The outcome force becomes equal to zero at some “A” distance,
which corresponds to atoms equilibrium position in crystal lattice. Slope
angle tangent of linear part on curve 3 characterizes intensity of stress
build-up that is necessary for elastic displacement from equilibrium posi-
tions or in other words the elasticity module.
The modulus of normal elasticity E depends only to a compara-
tively small extent on the structure, heat treatment, and composition of an
alloy. It is determined, mainly, by the type of crystal lattice. The modulus
of elasticity are as follows for certain important metals (Table 1.1).
Table 1.1– The modulus of elasticity of pure polycrystalline metals at room temperature
Metal Fe Ni Cu Al Ti Cr Mo Zn Co
E×10–5, MPa 2.17 2.05 1.25 0.72 1.08 2.40 8.47 0.94 2.04
G×10–5, MPa 0.89 0.78 0.46 0.27 0.41 0.90 1.22 0.37 0.76

Iron, which has a high modulus of elasticity, is the most important

of engineering materials.
There is one more parameter for elastic deformation and that is
called Poisson ratio (marked as ν). This important parameter is defined as
relationship of transversal elastic deformation to axial one. All four pa-
rameters are well related between each other by following equations:
E = 2G (1 + ν); (1.6)
E = 3K (1 – 2ν). (1.7)
Expressions (1.3) – (1.5) determine relation between stresses and
strains in the same direction. Although in some cases stresses can not fit in
with strains. For example a three axial deformation takes place at one ax-
ial tension. Then elementary Hooke law (exp (1.3) – (1.5)) must be
changed for the generalized one establishing linear relationship between
stresses and strains in any directions or in other words between all stress
tensor and strain tensor components:
Sx = K · χ + 2G · ex,
Sy = K · χ + 2G · ey, (1.8)
Sz = K · χ + 2G · ez,
K = E · ν / [(1 + ν)(1 – 2ν)].
This generalised version of Hooke law is appropriate for isotropic
material. However metals and alloys have a crystal structure and therefore
they are mainly anisotropic bodies. Their elastic properties are not the
same in different crystallographic directions. For anisotropic material
generalised Hooke law gets more complicated because it represents direct
proportion between each component of strain tensor and six independent
components of stress tensor. Proportion coefficients in these equations are
elasticity modules. Besides of those equations Hooke’s law contains rela-
tion equations of stress tensor and strain tensor. In these equations so
called elasticity coefficients are used. They relate with elasticity modules
by special relationship.
The influence of various factors on the elasticity modules. Consider
impacts of different factors on elasticity constants. For example, in pure
metals and in the majority of alloys an increase in temperature from 0 K to
melting point leads to a decrease in elasticity modules by 2–2.5 times. The
reason of such an effect has to do with thermal expansion of metals and
alloys. Poisson ratio depends very weakly on temperature and indicates a
slow growth with its increase. Elasticity modules depends weakly on ma-
terials microstructure parameters. For example grain size doesn’t effect
them, on the other hand severe cold plastic deformation leads to their de-
creasing. Modules are quite sensitive to materials texture because of ap-
peared anisotropy.
Alloying of metals is accompanied by changes in elasticity modules
in different ways. In the case of alloying by solid solutions the elasticity
modules are change according to the linear law. Solid solutions elasticity
modules increase if binding forces of alloying atoms and matrix ones are
higher than in pure metals, and visa versa. Lattice distortion near alloying
atoms leads to a decrease in modules. During alloying process in the case
of phase forming the elasticity modules increase if the phase module is
higher then the matrix one. However if the phase is softer than the matrix
then this leads to a decrease in alloy module.
Figure 1.7 shows the stress–strain curve of a material exhibiting
perfectly linear elastic behavior. This is the behavior characterized by
Hooke's law. All solids are linear elastic at small strains – by which we
usually mean less than 0.001, or 0.1 percent. The slope of the stress-strain
line, which is the same in compression as in tension, is of course Young's
Modulus, E. The area (shaded) is the elastic energy stored, per unit vol-
ume: since it is an elastic solid, we can get it all back if we unload the
solid, which behaves like a linear spring.
Figure 1.8 shows a nonlinear elastic solid. Rubbers have a stress –
strain curve like this, extending to very large strains (of order 5). The ma-
terial is still elastic: if unloaded, it follows the same path down as it did
up, and all the energy stored, per unit volume, during loading is recovered
on unloading– that is why catapults can be as lethal as they are.

Figure 1.7 – Stress–strain behavior for a linear elastic solid.
The axes are calibratedfor a material such as steel

Figure 1.8 – Stress – strain behavior for a nonlinear elastic solid.

The axes are calibratedfor a material such as rubber

Finally, Figure 1.9 shows a third form of elastic behavior found in

certain materials. This is called anelastic behavior. All solids are anelastic
to a small extent: even in the regime where they are nominally elastic, the
loading curve does not exactly follow the unloading curve, and energy is
dissipated (equal to the shaded area) when the solid is cycled. Sometimes
this is useful – if you wish to damp out vibrations or noise, for example;
you can do so with polymers or with soft metals (like lead) which have a
high damping capacity (high anelastic loss). But often such damping is
undesirable – springs and bells, for instance, are made of materials with
the lowest possible damping capacity (spring steel, bronze, glass).

Figure 1.9 – Stress – strain behavior for an anelastic solid.

The axes are calibratedfor fiberglass

1.2.2 Plastic deformation

Plastic deformation is the result of irreversible displacement of at-
oms. In crystals these displacement of atoms in most cases occur by
movement of dislocations that is the basic nuclear mechanism of plastic
deformation. Movement of dislocations can cause macroplastic deforma-
tion of the sample by sliding (slipping) or twinning. As a result of such
movement of dislocations there is a shift of one separate parts of a crystal
concerning others (sliding), or shift and turn of atomic lines in separate
sites of the sample under some corner to a shift direction (twinning).

Types of plastic deformation

Materials can be plastically deformed in the line of two main
modes: sliding and twinning.
The sliding deformation mode represents an atomic plain surface
shear similar to the movement of cards in a pack. Consider in detail the
sliding mode in the following scheme: there is a crystal where an edge
dislocation at its movement produces a partial shear of the crystal top part
against the lower one. Then combined deformation can be calculated as:
g = ρ · b · l, (1.9)
where ρ – dislocation density;
b – Burgers vector;
l – dislocation running length.
In real metals and alloys the dislocation sliding can occur in certain
crystallographic plain surfaces and directions which are typical for each
type of lattice. The sliding direction takes place always in its own sliding
plain surface. Such combination of plain surfaces and directions is named
the sliding system. One or several sliding systems can occur simultane-
ously in real metals.
Figure 1.10 shows an example of a typical sliding system for such
three widespread metals lattices as face centered cubic (f.c.c.), body cen-
tered cubic (b.c.c.) and hexagonal closest packed (h.c.p.). F.c.c. and h.p.c.
lattices are classified as closest packed ones. It is quite clear from this
schema that, directions and planes of each sliding system are the most
tightly packed. The most tightly packed plain surfaces have the biggest
intersurface distance that explains the easiest shear along them in com-
parison to other surfaces. Sliding directions correspond to Burgers unit
vector dislocation that is typical for each lattice.
For the most tightly packed lattices the sliding process goes mainly
on surfaces with the following indexes: (111) in f.c.c. and (0001) in h.c.p.
Packing in b.c.c. lattice is not tight. This type of lattice has several
sliding surfaces: (110), (211), (321).

a b c
Figure 1.10 – Slip systems in different lattices:
a – f.c.c.; b – h.c.p.; c – b.c.c.
Deformation by twinning occurs in the case when sliding is compli-
cated. Normally twinning is observed at low temperatures and high defor-
mation velocities. Under such conditions this deformation mode is typical
for metals possessing by and b.c.c. lattices. In pure f.c.c. metals twinning
occurs just at negative temperatures and high deformation velocities.
Deformation by twinning is a sort of shear of some material volume
that is called twin. The shear is caused by sliding of partial dislocations in
all parallel atomic surfaces of this material volume. It creates the same
lattice but with a mirrored symmetrical orientation to the twinning sur-
face. Figure 1.11 shows a scheme that illustrates the principal of twinning.

Figure 1.11– Deformation by twinning:

a – scheme of twinning; b – deformation twins in a zinc crystal

This scheme points out to the important particularity of twinning:

symmetrical surfaces of lattices can not be twinning surfaces because their
mirrored reflection doesn’t change orientation. Twinning goes with a spe-
cial combination of crystallographic directions and surfaces that are called
the twinning system. Normally twinning surfaces should be tight but pro-
viding for a possible dislocation splitting.

1.2.3 Strengthening of metals

The fundamental approach to strengthening metals is to devise
methods that increase the resistance to the motion of dislocations respon-
sible for the plastic deformation. Reducing the grain size by rapid cooling
of casting, or by adding inoculants or “seed” crystals to a solidifying alloy
to promote nucleation and grain refinement is one method to strengthen
metals. This is because fine-grained materials have a large-grain boundary
area, and the motion of dislocation is hindered by grain boundaries be-
cause of crystallographic misorientation between grains, and because of
the discontinuity of slip planes between neighboring grains. The strength-
ening by grain size reduction leads to an increase in the yield strength of
the material. The grain size dependence of the yield strength, σ0, of mono-
lithic crystalline solids is described by the Hall – Petch equation according
to which,
σ0 = σi + k ⋅ D −1/2 ,
where σi is a friction stress that opposes dislocation motion;
k is a material constant;
D is the average grain diameter.
Thus, a plot of the yield strength as a function of the inverse square
root of the average grain diameter will yield a straight line. The Hall –
Petch equation correctly describes the grain size dependence of yield
stress for micrometer size grains but fails at grain diameters smaller than
about 10 nm. This is because the equation was derived for relatively large
dislocation pile-ups at grain boundaries, with about 50 dislocations in the
pile-up. Clearly, at nanometric grain sizes, the pile-ups would contain
fewer dislocations, thus violating the key assumption of the Hall – Petch
The properties and behavior of materials change, often dramatically
and in surprising ways, when the grain size approaches nanoscale, with
the lower limit approaching the size of several crystal units. At nanometric
grain sizes the fraction of the disordered interfacial grain boundary area
becomes very large, and the grain diameter approaches some characteris-
tic physical length such as the size of the Frank-Reed loop for dislocation
slip. Furthermore, interfacial defects such as grain boundaries, triple
points, and solute atoms segregated at an interface begin to increasingly
influence the physical and mechanical behaviors. In addition, interfacial
defects from processing such as micropores, and minute quantities of ox-
ide contaminants and residual binders mask the grain size effects in
nanomaterials, making it very difficult to isolate the true effect of grain-
size reduction in such materials.
The other methods to strengthen metals include alloying, work
hardening, and precipitation hardening. Pure metals are almost always
soft, and alloying strengthens a metal. This is because the solute atoms
(interstitial or substitutional) locally distort the crystal lattice, and generate
stresses that interact with the stress field around dislocations, which hin-
der the dislocation motion. Impurity atoms segregate around the disloca-
tion line in configurations that lower the total energy. This causes the dis-
location to experience a drag due to the impurity “atmosphere” surround-
ing it, which must be carried along if the dislocation must move to cause
deformation. Besides alloying, plastic deformation can also strengthen
metals, a process called work-hardening or strain-hardening. Plastic de-
formation rapidly increases the density of dislocations, whose stress fields
inhibit the motion of other dislocations. Frequently, dislocation tangles
form provides considerable resistance to continued plastic deformation.
The strain-hardening exponent n in equation
σt = K ⋅ ε tn ,
introduced in the preceding section is a measure of the ability of a metal to
work harden (i.e. strengthen through plastic deformation); metals with a
large value of n strain harden more than metals with a low-value of и for a
given amount of plastic deformation or degree of cold work. Certain al-
loys can be strengthened via a special heat treatment that creates finely
dispersed hard second-phase particles, which resist the motion of disloca-
tions. This is discussed in the section on heat treatment. Finally, hard sec-
ond-phase particles and fibers may be added to metals from outside for
strengthening by creating a composite material.

1.2.4 The fracture of materials

Ability of materials to be deformed and maintain the big loadings
without destruction is the major quality defining their adaptability to
manufacture and operational properties. In most cases deformation at
achievement enough high pressure comes to an end with destruction. De-
struction process begins with formation of cracks of the submicroscopic
sizes and comes to an end with macroscopical division of the sample or a
design into separate parts. Many major mechanical properties of metals
and alloys characterized their resistance to destruction, size or work of
deformation before destruction.
The materials fracture process depends on different conditions such
as load type, working temperature, atmosphere and microstructure pa-
There are three kinds of stress: compressing (negative normal),
stretching (positive normal) and tangents. Destruction always occurs un-
der the influence of stretching or tangents of stresses. Compressing stress
in itself cannot cause destruction.

According to fundamental concepts, there are only two modes in
which materials can fracture under a single or monotonic load. These two
modes are called shear (under the influence of tangents stress) and cleav-
age (as a result of action of normal stretching stress) and differ primarily
in the way the basic material crystal structure behaves under load. On ap-
pearance of the destroyed samples it is possible to define a destruction
kind (a shear or a cleavage), that in some cases has practical value.
On the characteristic mechanism of development, speed and power
consumption of development of destruction distinguish brittle and ductile
destruction. It is considered, that the separation can occur without pre-
liminary macroplastic deformation while by a cut such deformation al-
ways precedes destruction. Therefore the cleavage often corresponds to
brittle fracture, and a shear – to ductile fracture. Consider the shear frac-
ture mode. If in the course of plastic deformation a fracture occurs, the
shear causes initialization of micropores in the most highly stressed re-
gion. Pores are formed and merged under load and as a result, the frac-
tured surface is formed and consists of microvoids. Such fracture surface
is called ductile.
Ductile fracture occurs usually after considerable plastic deforma-
tion (ten percent). Its main features are slow development of cracks and
high the power consumption, caused necessity of an expense of consider-
able work of plastic deformation at crack top. Therefore ductile fracture is
the least dangerous kind of destruction. Nevertheless the analysis of duc-
tile fracture is very important, in particular at the analysis of behavior of
metals in the conditions of processing by pressure where considerable
plastic deformations and where destruction, including ductile are created,
is inadmissible.
If a metal fractures in a cleavage mode its separation occurs very
suddenly and goes along two adjacent cells. There is no plastic deforma-
tion of metal at cleavage at least on macroscale. A cleavage fracture usu-
ally occurs in relatively hard, strong metals that have b.c.c. or h.c.p. lat-
tices. However under certain conditions (for example at a low tempera-
ture) metals that normally fracture in the shear mode may fracture in the
cleavage mode. On a micro scale, cleavage fractures occur along cells in-
terfaces, but can be seen as a grains splitting with no relation to the grain
boundaries. This looks like a brick wall fracture along the bricks in. In
many cases the crack direction can be determined from the study of the
cleavage fracture surfaces by scanning electronic microscope. Brittle frac-
ture is the most dangerous kind of destruction as it occurs catastrophically
quickly even in the conditions of rather low stresses. Therefore data on the
mechanism of brittle fracture and conditions which promote it or compli-
cate, are especially important.
The main distinctive features of both modes are shown in Table 1.2.
Table 1.2 – The main distinctive features of both modes – shear and cleavage
Distinctive features Shear Cleavage
Movement Sliding Snapping apart
Occurrence Gradual Sudden
Deformation Yes No
Behavior Ductile Brittle
Visual appearance of fracture Dull and fibrous Bright and sparkling
Microfractography Dimpled rupture Cleavage

To any, including brittle fracture of metals and alloys is preceded by

the plastic deformation necessary for origin of a crack. The deformation
previous a brittle fracture, usually is much less, than at ductile fracture.
Ductile and brittle fractures consist of two stages: origin of a crack
and its distribution. On the mechanism of origin of cracks ductile and brit-
tle fracture essentially do not differ.
Mechanisms of origin of cracks at brittle and ductile destruction
qualitatively do not differ. For crack origin the misfit dislocation at an
obstacle or crossing of the twin with border always is necessary. It is con-
sidered, that microcracks at the moment of origin have length ~10–4 mm.
Most often cracks arise at tops of misfit dislocation near to any barriers:
inclusions of superfluous phases, grain boundaries, twin etc. In immediate
proximity from a barrier (Figure 1.12) edge dislocation in a misfit disloca-
tion can appear under the influence of stress so closely pressed to each
other, that under them the germinal microcrack is formed.

Figure 1.12 – Scheme of nucleation of microcrack because

of accumulation of dislocations at barriers

In metals with the b.c.c. lattice the crack can be formed on the
model offered by Kotrell. Scheme of Kotrell does not demand presence of
barriers to dislocations in an initial condition. Barriers, and then mistif
dislocations and cracks are formed as a result of plastic deformation.
In the conditions of strongly developed intergranular deformations
the probability of origin of cracks on so-called triple junction grain's
boundaries increases.
The considered schemes of origin of cracks show, that destruction
of metals with a different lattice and a microstructure can begin differ-
ently. But as a result resistibility of metal or an alloy to destruction and
character of destruction are defined by conditions in which there is a mi-
crocrack. Therefore the second stage of destruction – a stage of distribu-
tion of a crack – is solving.
Qualitative distinction between ductile and brittle fracture is con-
nected with power consumption and speed of distribution of a crack. At
brittle fracture this speed is very great it reaches 0.4–0.5 speeds of distri-
bution of a sound in a material of the sample. In a case of ductile fracture
the crack extends basically with rather small speed, commensurable with
speed of deformation of the sample.
At distribution (growth) of a crack before its top in some zone there
is a plastic deformation. The size of this plastic zone is defined by yield
strength of a material. Usually in materials with high strength a zone of
plastic deformation in crack top small and destruction develops on the brit-
tle mechanism. In materials with low strength its size much more and de-
struction develops on the ductile mechanism. Therefore power consumption
of ductile fracture much more as at development of a ductile crack plastic
deformation goes not only near to its top, but also on considerable volume
of a detail or the sample. As a result the work necessary for advancement of
a crack, in this case much more, than at development of a brittle crack when
plastic deformation is localized in a narrow zone at its top.
Crack distribution is the most important stage of destruction. This
process, basically, defines resistance of a material to destruction. The
stage of distribution of a crack too consists of two stages – subcritical
crack growth, concerning slow development of a crack when destruction
process still it is possible to supervise, and overcritical crack distributions
when definitive destruction becomes very fast, difficultly operated and
often irreversible.
In polycrystals the crack at destruction can extend on a body of
grain or along borders. Accordingly distinguish transgranular fracture and
intergranular fracture. At low temperatures intergranular fracture is usu-
ally observed in brittle materials and caused by presence on a surface of
borders of grains of particles of fragile superfluous phases or impurity
segregation. Such destruction can occur also at the raised temperatures, in
the conditions of intensive development intergranular deformations. Ten-
dency to intergranular fracture increases with decreasing strain rate.
Brittle fracture for any metal material is observed only under cer-
tain conditions tests, processing or operation. Propensity to brittle struc-
ture especially strongly depends on: the temperature more low the usually
is more probability of brittle structure. Therefore mark the interval of
temperatures of transition ΔТd–b from ductile fracture (with enough high
plasticity) to brittle fracture (is allocated at low plasticity) on temperature
dependence of indicators of ductility and plasticity of technical alloys
(Figure 1.13).

Figure 1.13 – Dependence of toughness and percentage shear

on the temperature: ΔTbr – temperature range of transition from ductile
fracture to brittle; Tbr – ductile-brittle transition temperature corresponding
to the middle of the interval ΔTbr; T50 – ductile-brittle transition
temperature corresponding to the presence of 50 % of the percentage shear

Instead of an interval of temperatures often use any one tempera-

ture of ductile-brittle transition Тbr: the top or bottom borders of an inter-
val ΔТbr, or the temperature corresponding to the middle of this interval
(see fig. 1.13). Sometimes Тbr estimate as the temperature corresponding
to a certain share of brittle fracture of the sample. Size Тbr is widely used
as the characteristic of propensity of a material to fragile destruction: the
above Tbr, the more this propensity.
Brittle fracture is especially dangerous at temperatures near to room
and above. Metals and alloys at which Тbr lays at such temperatures, name
cold-short. On the contrary, alloys at which ductile-brittle transition oc-
curs in the field of negative temperatures, name cold-resistant. Cold-
shortness – a problem especially sharp for many metals and alloys with
the b.c.c. lattice.
For an explanation of possibility of transition from a ductile frac-
ture in brittle often use A.F. Ioffe's classical scheme (Figure 1.14).

Figure 1.14 – Transition from a ductile fracture in brittle fracture

The temperature of transition Тbr corresponds here to a point of

crossing of curves Scl (cleavage fracture strength) and Sy (yield strength).
The lower breaking point Тbr is reached earlier, than the yield strength Sy,
and destruction occurs is brittle, without preliminary plastic deformation.
Above Тbr at loading the sample in the course of test in the beginning it is
reached Sy, there is a plastic deformation, and then there is a destruction,
which in these conditions basically the ductile.
At materials with sharp temperature dependence of yield strength,
characteristic for alloys with the body-centred cubic (b.c.c.) lattice, strong
sensitivity Sy to speed of deformation is usually observed also: the in-
crease in speed causes growth of yield strength. It also promotes brittle
fracture (raises Тbr). Speed of plastic deformation near to top of an extend-
ing crack is close to speed of its development. At brittle fracture this speed
is great, that defines high yield strength at crack top. As a result before
quickly moving crack plastic deformation is complicated its work is small
also to a crack easier to extend as brittle.
For metals and alloys with face-centred cubic (f.c.c.) and close-
packed hexagonal (c.p.h.) lattices dependence of the yield strength on
temperature much less sharp and consequently the specified materials
usually more cold-resistant (Тbr lower).
Feature of destruction consists that it is much more local and struc-
turally-sensitive, than all kinds of deformation. Really, crack development
is defined by structure and properties of a material in immediate proximity
(on micron distances) from its top. Thus, characteristics of macrodestruc-
tion of the sample or a design are defined by local processes in micro-
There are other fracture modes different from shear and cleavage.
These include intergranular and a quasi-cleavage fracture mode under a sin-
gle stress load and a fatigue fracture mode under a multiple cycling load.
An intergranular fracture can occur if the grain boundaries are
weaker then the grains themselves. Fracture then goes along the grains
rather than through the grains. As a result in a fractured surface we can
see the grains sides in many areas. An intergranular fracture is usually
caused by such factors as hydrogen absorption, contact with liquid metals,
grains oxidation, grain boundary atomic segregations and so on.
A quasi-cleavage fracture is seen frequently in fractured surfaces of
quenched and tempered steels. This mode can be considered as combina-
tion of the shear and cleavage fracture modes because microdimples ap-
pear with cleavage fracture surfaces.
Fatigue fracture is caused by a multiple stress load. The repetitive
forces, which can reach millions of cycles lead to inevitable structural mi-
crodefects in crystals until a crack develops in certain crystals. The cy-
cling stressing automatically locates crystals with the weakest orientation
in the highest stress areas. A continued cycling causes cracks, which de-
velop and lead to fracture of a material or a detail of equipment.

Examination of failures
The following are illustrations of the types of failure that can be
found with metals, polymers, ceramics and composites:
1 Ductile failure with metals
Ductile fracture is due to tangential stresses which have reached a
definite fracture value called the shear strength (τk). This type of fracture
is preceded by considerable plastic deformation.
Proceeding from the fact that ST and τk are constant values for a
given material consider that brittle fracture, due to normal stresses, will
occur in cases where the cohesive strength is lower than the shear
strength. If the tangential stresses produce considerable plastic deforma-
tion before the normal stresses reach the cohesive strength, the fracture
will be of the ductile type. Thus, fracture due to normal stresses by separa-
tion or cleavage may be either brittle or ductile.
Fracture due to tangential stresses, resulting from plastic deforma-
tion, will, of course, always be ductile. Most frequently, metal fracture
does not occur as pure separation or shear but as a complex combination
of these two types of fracture (see table 1.2).
A ductile material is characterised by having a significant plastic
region to its stress-strain graph. When a ductile material has a gradually
increasing tensile stress applied then when yielding starts the cross-
sectional area of the material becomes reduced, necking being said to oc-
cur (Figure 1.15).

a b c d

Figure 1.15 – Stage in ductile fracture: a – necking occurring;

b – small cavities forming; c – cavities link up to form horizontal crack;
d – crack propagates to surface

Eventually after a considerable reduction in cross-sectional area the

material fails. The resulting fracture surfaces show a cone and cup forma-
tion (Figure 1.16).
This occurs because, under the action of the increasing stress, small
internal cracks form which gradually grow in size until there is an inter-
nal, almost horizontal, crack. The final fracture occurs when the material
shears at an angle of 45° to the axis of the applied stress. Such a type of
failure is referred to as a ductile fracture. Materials can also fail in a duc-
tile manner in compression, such fractures resulting in a characteristic
bulge and series of axial cracks around the edge of the material.
Figure 1.16 – Ductile failure

2 Brittle failure with metals

Brittle fracture consists in destroying the interatomic bonds by nor-
mal stresses. Brittle fracture is not accompanied by noticeable plastic de-
formation. The resistance to brittle fracture is called cohesive strength (ST).
Brittle fracture is comparatively rare. For example, this type of fracture
may be observed in zinc and its alloys or in iron and low-alloy steel at low
temperatures, as well as when brittle layers segregate on the grain bounda-
ries. It must be stressed, however, that pure brittle fracture is practically
never encountered. Fracture is always preceded by plastic deformation.
A brittle material has virtually no plastic region to its stress-strain
graph. Thus when a brittle material fractures there is virtually no plastic
deformation. Figure 1.17 shows possible forms of fracture in such a situa-
tion. The surfaces of the fractured material appear bright and granular due
to the reflection of light from individual crystals; the fracture has grains
within the material which have cleaved along planes of atoms.

Figure 1.17 – Brittle fracture

We can consider the sequence of events leading to brittle fracture to

be that when stress is applied the bonds between atoms and between
grains in the material are elastically strained, then at some critical stress
the bonds break, remember the material is brittle and there is no plastic
deformation and hence small-scale slip, and a crack propagates through
the material to give fracture. Such failure is known as brittle fracture.
3 Fatigue failure with metals
Fatigue failure often starts at some point of stress concentration.
This point of origin of the failure can be seen on the failed material as a
smooth, flat, semicircular or elliptical region, often referred to as the nu-
cleus. Surrounding the nucleus is a burnished zone with ribbed markings.
This smooth zone is produced by the crack propagating relatively slowly
through the material and the resulting fractured surfaces rubbing together
during the alternating stressing of the component. When the component
has become so weakened by the steadily spreading crack that it can no
longer carry the load, the final abrupt fracture occurs. This region of
abrupt failure has a crystalline appearance.
4 Failure with polymers
Brittle failure with polymeric materials is a common form of failure
with materials below their glass transition temperature, i.e. amorphous
polymers. The resulting fracture surfaces show a mirror-like region, where
the crack has grown slowly, surrounded by a region which is rough and
coarse where the crack has propagated at speed.
In an amorphous polymer the chains are arranged randomly with no
orientation. When stress is applied it can cause localized chain slippage
and an orientation of molecule chains with the result that the applied stress
causes small voids to form between the aligned molecules and fine cracks,
termed crazing, are formed. This is what constitutes the mirror-like region.
Because of the inherent weakness of the material in the crazed region it
serves as a place for cracks to propagate from and cause the material to
fracture. Initially the crack grows by the growth of the voids along the
midpoint of the craze. These then coalesce to produce a crack which then
travels through the material by the growth of voids ahead of the advancing
crack tip. This part of the fracture surface shows as the rougher region.
With crystalline polymers, the application of stress results in the
folded molecular chains becoming unfolded and aligned. The result is
then considerable, permanent deformation and necking. Prior to the mate-
rial yielding and necking starting, the material is quite likely to begin to
show a cloudy appearance. This is due to small voids being produced
within the material. Further stress causes these voids to coalesce to pro-
duce a crack which then travels through the material by the growth of
voids ahead of the advancing crack tip. Figure 7.8 shows typical forms of
stress-strain graphs for polymers showing this form of failure.
5 Failure with ceramics
Ceramics are brittle materials, whether glassy or crystalline. Typi-
cally fractured ceramic shows around the origin of the crack a mirror-like
region bordered by a misty region containing numerous micro cracks. In
some cases the mirror-like region may extend over the entire surface.
6 Failure with composites
The fracture surface appearances and mechanisms for composites
depend on the fracture characteristics of the matrix and reinforcement ma-
terials and on the effectiveness of the bonding between the two. Thus, for
example, for a glass-fibre reinforced polymer, depending on the strength
of the bonds between fibres and polymer, the fibres may break first and
then a crack propagate in shear along the fibre-matrix interface. Eventu-
ally the load which had been mainly carried by the fibres is transferred to
the matrix which then fails. Alternatively the matrix may fracture first and
the entire load is then transferred to the fibres which carry the increasing
load until they break; the result is a fracture surface with lengths of fibre
sticking out from it, rather like bristles of a brush.

1.3 Mechanical testing of metals

All solids have an elastic limit beyond which something happens. A
totally brittle solid will fracture, either suddenly (like glass) or progres-
sively (like cement or concrete). Most engineering materials do something
different; they deform plastically or change their shapes in a permanent
way. It is important to know when, and how, they do this – both so that
we can design structures which will withstand normal service loads with-
out any permanent deformation, and so that we can design rolling mills,
sheet presses, and forging machinery which will be strong enough to im-
pose the desired deformation onto materials we wish to form. To study
this, we pull carefully prepared samples in a tensile testing machine, or
compress them in a compression machine (which we will describe in a
moment), and record the stress required to produce a given strain.
When metals are deformed plastically at temperatures below those
at which recrystallization occurs, they are said to be cold worked. Besides
the shape change that results, strength and hardness usually increase and
the terms work hardening or strain hardening refer to such strengthening
by cold working. Another consequence is the lessening of the remaining
The most common method used to describe the work-hardening be-
havior of ductile metals in a quantitative way is by means of a tensile test.
Other tests that have been used are direct compression, torsion, balanced
biaxial or bulge testing, and plane-strain compression. Because it is the
simplest, the tensile test has found major use. For completeness, the vari-

ous mechanical properties determined from such a test will be defined, but
the major emphasis in this chapter will be placed upon the change in yield
strength that results from cold working. It is this aspect of work hardening
that is of greatest interest in plasticity studies.

1.3.1 Classification of mechanical tests

The first principle of the mechanical tests classification depends on
the scheme of the stress or strain state.
For example, the sample is stretched along one axis – the scheme of
uniaxial tension, stretch along two axes-biaxial tension (example – canvas
stretched on a frame by the artist), if something under pressure from three
sides (fish, floating deep in the ocean) – a scheme of the stress state under
triaxial compression, etc.
The next principle – classification by a method of the sample load-
ing: basically use two ways of loading: 1) strain at a given rate and meas-
uring the resistance of the sample of this strain, 2) applying a constant
load (stress) on the sample with measurement occurring at the same strain
Mechanical tests in which loading is changeable, it is possible to
classify on character of its change over time.
Tests may be:
1) static, when the load is increased slowly and gradually and the
metal is loaded on the scheme tension, compression, torsion, or bending;
2) dynamic, when the load increases rapidly as in an impact (for ex-
ample, the impact test);
3) repeated or fatigue, when the load repeatedly varies in the course
of the test either in value or both in value and in direction (tension-
compression, tension-compression, etc).
There are other principles of classification of mechanical tests.
They are less common and we do not consider them in this course.

1.3.2 The main types of mechanical tests

Uniaxial tensile test

Tension tests are of the static type, i.e., they are tests in which the
load is increased comparatively slowly, from zero to a certain final value.
For tension tests used the specimens of round or flat cross-section. The ends
of the specimen are secured in the grips of the testing machine. Tensile test-
ing equipment is standard in all engineering laboratories. The mechanical
properties in tension are determined on the gauge length of the specimen.

Such testing machines as Instron, MTS, ZWICK, and others are extensively
used for tension tests. All tension testing machines have two chief parts:
1 Unit for applying a load to the specimen (with a hydraulic or me-
chanical drive), and
2 Unit for measuring the applied load.
Testing machines are often equipped with recording devices which
automatically record changes in length of the specimen in accordance with
the applied load (stress–strain diagram).
The results determined by tensile testing are influenced by the con-
ditions that prevail during such a test; in general, the following are typical:
- Strain rate is of the order of 10–2 to 10–4 sec–1;
- Temperature is between 20 and 30 °C;
- Measurements are restricted to a gage section that experiences a
state of uniaxial tensile stress during uniform deformation.
Almost all materials, when strained by more than about 0.001
(0.1 percent), do something irreversible: and most engineering materials
deform plastically to change their shape permanently. If we load a piece
of ductile metal (like copper), for example in tension, we get the following
relationship between the load and the extension (Figure 1.18).
This can be demonstrated nicely by pulling a piece of plasticine (a
ductile nonmetallic material). Initially, the plasticine deforms elastically,
but at a small strain begins to deform plastically, so that if the load is re-
moved, the piece of plasticine is permanently longer than it was at the be-
ginning of the test: it has undergone plastic deformation. If you continue
to pull, it continues to get longer at the same time getting thinner because
in plastic deformation volume is conserved (matter is just flowing from
place to place). Eventually, the plasticine becomes unstable and begins to
neck at the maximum load point in the force-extension curve (Pu, see
fig. 1.18). Necking is instability. The neck then grows quite rapidly, and
the load that the specimen can bear through the neck decreases until
breakage takes place. The two pieces produced after breakage have a total
length that is slightly less than the length just before breakage by the
amount of the elastic extension produced by the terminal load.
Returning to fig. 1.18 it can be seen that after initial yielding has
taken place, further plastic deformation requires an increasing load but at
a decreasing rate. Work hardening strengthens the material but at the same
time the area is decreasing, and the combined effect of these two factors
produces the resulting shape of the load–extension curve. From initial
yielding up to ultimate load, the strengthening effect offsets the area re-
duction; at ultimate load a condition of tensile instability occurs. Note that
the deformation of the gage section is uniform up to this point.
Figure 1.18 – Typical load–extension curve
for a ductile metal during a tension test

At some location along the gage section a local constriction begins

and this is called necking. From that time on, the continual work harden-
ing in the necked region no longer compensates for the continual reduc-
tion of the smallest cross section in the neck. As a consequence, the load
required for further extension decreases. It is important to understand that
it is the load-carrying capacity of this now non-uniform specimen that de-
creases. Even though the material in the neck has been work hardened
more than any other section (and is therefore stronger), the smaller area of
the neck causes it to have the lowest load-carrying capacity. With continu-
ing extension, practically all further plastic deformation is restricted to the
region of the neck, so a highly non-uniform deformation continues along
the full gage section until fracture occurs. The fracture area is then based
upon the smallest diameter of the neck.
The basic information obtained is the load or force, P, N, required
to cause a given extension, Δl, mm. From these data, values of stress and
strain are computed since they provide more general information. Fig-
ure 1.18 is a schematic of a typical load–extension curve. Since the elastic
behavior occurs over a very small range of extension as compared with
the full test, it appears as a nearly vertical line if all data are plotted to a
common scale.
Before determining the traditional properties indicated by a tensile
test, the data from Fig. 1.18 are converted to a plot of nominal or engi-
neering stress and strain. These are defined as
σi = ; (1.10)
Δl l − l0
ε= = , (1.11)
l0 l0

where P and Δl are any pair of coordinate points on fig. 1.18 and F0 and l0
are the original area and gage length of the unloaded specimen. Since the
conversion from load – extension to nominal stress – strain utilizes the con-
stants F0 and l0, this amounts to nothing more than a scale factor adjustment.
Consequently, the shape of the σ–ε curve is identical to the P–Δl curve.
The stress, calculated by the formula
σp = , (1.12)
where F0 – initial cross-sectional area of the sample, is called
the limit of proportionality or proportional limit.
The proportional limit is the stress at which the deviation from a
linear relationship between the stress and deformation is almost com-
pletely absent. This stress is very low and causes elastic deformation only,
i.e. deformation, which disappears when the load is removed. For this rea-
son, the limit of proportionality is often identified with the elastic limit.
This is actually not so but accurate enough for practical purposes.
Usually the elastic limit σe is defined as the stress at which the per-
manent set is equal to 0.005 percent of the initial length of the specimen.
Thus, σp and σe characterize the resistance of the metal to small plastic
It’s important to remember, that the proportional and elastic limits
are essential characteristics of a metal. Any structure or machine compo-
nent must be designed so that its working stresses do not exceed σp and σe.
When a particular level of stress is reached, plastic flow begins; this
stress is called the yield strength and is defined as
σy = (1.13)
the subscript indicating a particular point on the curve.
Tensile strength, also called ultimate strength is defined as
σu = , (1.14)
where Pu is the maximum or ultimate load carried by the specimen
(onset of necking).
Ductility, which is a measure of the extent to which a metal can be
deformed plastically, is commonly defined by two parameters; both are
based upon measurements made after the specimen has fractured. They are

δu =
100 l f − l0 )
; (1.15)

100 F0 − F f ), (1.16)
where the subscript/relates to the values at fracture. Although standard
values of l0 and F0 are generally used, it should be realized that both per-
cent elongation (δ) and area reduction (ψ) will vary if non-standard values
of l0 and F0 are used. For this reason, these properties are not really fixed
for a given metal. ψ, the reduction of area at fracture in a tension test,
should not be confused with the area reduction in processing.
For a large number of metals and alloys, the region between the in-
ception of plastic deformation (yield point) and the onset of necking (ten-
sile strength) on a true stress-true strain diagram can be described by a
power law relationship of the form σt = K ⋅ εtn , where K and n are mate-
rial-specific constants, and n is called the strain-hardening exponent. The
strain-hardening exponent, n, is a measure of the ability of a metal to work
harden (i.e., strengthen through plastic deformation). A large value of
n indicates that strain hardening will be large for a given amount of plastic
deformation. Thus, annealed copper (n = 0.54) will strain-harden more
than annealed Al (n = 0.20) as would 70/30 brass (n = 0.49) compared to
4340 alloy steel (n = 0.15). Table 1.3 gives the values of n and K for
common metals and alloys.
Table 1.3 – Values of exponents n and K for different alloys
Alloy n K, MPa
Low-carbon steel (annealed) 0.26 530
4340 alloy steel (annealed) 0.15 640
304 stainless steel (annealed) 0.45 1275
2024 Al Alloy (annealed) 0.16 690
Brass (70 Co–30 Zn, annealed) 0.49 895
True stress-strain curves for plastic deformation. Naturally, be-
cause F0 and l0 are constant, the shape of the σ–ε (nominal stress–nominal
strain) curve is identical to that of the P–Δl (load–extension) curve. But
the σ–ε plotting method allows one to compare data for specimens having
different (though now standardized) F0 and l0, and thus to examine the
properties of material, unaffected by specimen size. The advantage of
keeping the stress in nominal units and not converting to true stress is that
the onset of necking can clearly be seen on the σ–ε curve (Figure 1.19).

Figure 1.19 – Comparison of nominal and true stress–strain curves

The apparent difference between the curves for tension and com-
pression is due solely to the geometry of testing. If, instead of plotting
load, we plot load divided by the actual area of the specimen, A, at any
particular elongation or compression, the two curves become much more
like one another. In other words, we simply plot true stress as our vertical
coordinate. This method of plotting allows for the thinning of the material
when pulled in tension, or the fattening of the material when compressed.
But the two curves still do not exactly match. The reason is a dis-
placement of (for example) Δl = l0 / 2 in tension and compression gives
different strains; it represents a drawing out of the tensile specimen from
l0 to 1.5l0, but a squashing down of the compressive specimen from l0 to
0.5l0. The material of the compressive specimen has thus undergone much
more plastic deformation than the material in the tensile specimen, and
can hardly be expected to be in the same state, or to show the same resis-
tance to plastic deformation. The two conditions can be compared prop-
erly by taking small strain increments
de = . (1.17)
about which the state of the material is the same for either tension or com-
pression. This is the same as saying that a decrease in length from
100 mm (l0) to 99 mm (l), or increase in length from 100 mm (l0) to
101 mm (l) both represent a 1 percent change in the state of the material.
Actually, they do not quite give exactly 1 percent in both cases, of course,
but they do in the limit
de = , (1.18)
Then, if the stresses in compression and tension are plotted against
dl ⎛l ⎞ F ⎛ 1 ⎞
e=∫ = ln ⎜ ⎟ = ln ( ε + 1) = ln 0 = ln ⎜ ⎟, (1.19)
l ⎝ l0 ⎠ Ff ⎝1− ψ ⎠

the two curves exactly mirror one another. The quantity e is called the true
strain and the matching curves are true stress/true strain (S / e) curves.
Further from the original load–extension or load–compression
curves easily calculate e, simply by knowing l0 and taking natural logs.
But how do we calculate S? Because volume is conserved during
plastic deformation we can write, at any strain,
F0 ⋅ l0 = F ⋅ l
provided the extent of plastic deformation is much greater than the extent
of elastic deformation (this is usually the case, but the qualification must
be mentioned because volume is only conserved during elastic deforma-
tion if Poisson's ratio v = 0.5, it is near 0.33 for most materials). Thus
F0 ⋅ l0
F= , (1.20)
P P⋅l
S= = = σ (1 + e ) , (1.21)
F F0 ⋅ l0
all of which we know or can measure easily.
Plastic work. When metals are rolled or forged, or drawn to wire, or
when polymers are injection molded or pressed or drawn, energy is ab-
sorbed. The work done on a material to change its shape permanently is
called the plastic work; its value, per unit volume, is the area under curves
1–2–3–4 (or 1–2′–3′–4′) shown in Figure 1.19; it may easily be found (if
the stress–strain curve is known) for any amount of permanent plastic de-
formation. Plastic work is important in metal- and polymer-forming op-
erations because it determines the forces that the rolls, or press, or mold-
ing machine must exert on the material.
There are different types of deformation curves.
The yielding behavior of an annealed brass and low-carbon steel is
illustrated in Figure 1.20.

Figure 1.20 – Offset method used to define yield strength

Brass is typical of most ductile metals in that yielding occurs so

gradually that it is difficult to define where the initial plastic deformation
takes place. For that reason, the yield stress is usually defined by the off-
set method. Using a 0.2 % offset, a line is constructed parallel to the linear
elastic portion of the stress-strain curve but displaced by a strain of 0.002
from the origin, as shown in Figure 1.20. The 0.2 % yield strength, shown
as Yb, is defined by the intersection of the offset line with the stress-strain
curve. With the low-carbon steel there is an upper yield point, A, and a
lower yield point, B. The latter is much less affected by specimen align-
ment and rate of loading than is A; for that reason, the stress level at B,
shown as Ys, is used to define the yield strength and there is no need to
resort to any offset.
If we load a material in compression, the force–displacement curve is
simply the reverse of that for tension at small strains, but it becomes differ-
ent at larger strains. As the specimen squashes down, becoming shorter and
fatter to conserve volume, the load needed to keep it flowing rises. No in-
stability such as necking appears, and the specimen can be squashed almost
indefinitely, this process only being limited eventually by severe cracking in
the specimen or the plastic flow of the compression plates.
Why this great difference in behavior? After all, we are dealing
with the same material in either case.
Validity of tensile test data. The purpose of taking tensile test
pieces and carrying out the tests is to obtain data which enables judge-
ments to be made about the material from which the test piece was cut.
The samples of a material have to be taken in such a way that the proper-
ties deduced from the tensile test are representative of the material as a
whole. There may, however, be problems in assuming this. The following
paragraphs outline some of these problems:
1 The properties of a product may not be the same in all parts of it.
With a casting there may be different cooling rates in different parts of a
casting, e.g. the surface compared with the core, or thin sections compared
with thick sections. As a result the internal structure of the material may
differ and as a consequence the tensile properties differ. A tensile test
piece cut from one part may not thus represent the properties of the entire
casting. For the same reason, the properties of a separately cast test piece
may not be the same as those of the cast product because the different
sizes of the two lead to different cooling rates.
2 The size of an item affects its properties after heat treatment. If
the mechanical properties of metals are looked up in tables you will often
find that different values of the properties are quoted for different limiting
ruling sections. The limiting ruling section is the maximum diameter of
round bar at the centre of which the specified properties may be obtained.
The reason for the difference of mechanical properties of the same mate-
rial for the different diameter bars is that during the heat treatment differ-
ent rates of cooling occur at the centre of such bars due to their differences
in sizes. Consequently there are differences in microstructure and hence
differences in mechanical properties. For example, the steel O7OM55
with a limiting ruling section of 19 mm may have tensile strengths of 850
to 1000 MPa, with a limiting ruling section of 63 mm the strengths are
777 to 930 MPa and for a limiting ruling section of 100 mm the strengths
are 700 to 830 MPa.
3 The properties of a product may not be the same in all directions.
For example, with rolled sheet there is a directionality of properties with
the tensile properties in the longitudinal, transverse and through the thick-
ness of the sheet differing. Thus, for example, with rolled brass strip we
might have tensile strengths of 740 MPa in the direction of the rolling and
850 MPa at right angles to it.
4 The temperature in service of the product may not be the same as
that of the test piece when the tensile test data was obtained. The tensile
properties of metals depend on temperature. In general, the tensile
modulus and tensile strength both decrease with an increase in tempera-
ture, the percentage elongation tends to increase.
5 The rate of loading of a product may differ from that used with
the test piece. The data obtained from a tensile test is affected by the rate
at which the test piece is stretched, so in order to give standardised result
the tests are carried out at a constant stress rate, between 2 and 20 MPa/s
if the tensile modulus is less than 150 GPa and between 6 and 30 MPa/s if
equal to or greater than 150 GPa.
Interpreting tensile test data. The results from tensile tests can be
used to determine the safe stresses to which a material can be subject.
Thus, the higher the yield stress of a metal the higher the stresses that it
can be exposed to in service without yielding. Another important deduc-
tion that can be made is whether the material is brittle or ductile. A brittle
material will show little plastic behavior and have a low percentage elon-
gation. A ductile material will shows considerable plastic behavior and
have a high percentage elongation.
A consequence of the heat treatment and working of a material that
occur during the fabrication of products is a change in mechanical proper-
ties. Thus tensile test data enable the effectiveness of heat treatments and
the effects of working to be monitored.

Hardness tests
Hardness is the resistance of a metal to the penetration of another
harder body which does not receive a permanent set. The hardness of a
material may be specified in terms of some standard test involving indent-
ing or scratching of the surface of the material, the harder a material the
more difficult it is to make an indentation or scratch. There is no absolute
scale for hardness, each hardness form of test having its own scale.

Though some relationships exist between results on one scale and those
on another, care has to be taken in making comparisons because the dif-
ferent types of test are measuring different things.
Hardness tests consist in measuring the resistance to plastic defor-
mation of layers of metal near the surface of the specimen. In the process
of hardness determination when the metal is indented by a special tip (steel
ball or diamond cone), the tip first overcomes the resistance of the metal to
elastic deformation and then, a small amount of plastic deformation. Upon
deeper indentation of the tip, it overcomes large plastic deformation.
Therefore, hardness is not an independent characteristic of the mechanical
properties. It determines the same properties as in other testing methods,
for example, tension tests, but under different loading conditions.
This fact enables a relation to be established between the hardness
and ultimate tensile strength of ductile metals. For metals of low ductility,
in which the tensile strength is characterised by the cohesive strength and
not by the resistance to considerable plastic deformation, the relation be-
tween hardness and ultimate tensile strength is not sufficiently reliable.
It is necessary to distinguish: 1) macrohardness, which is the hard-
ness of a material determined by its resistance to plastic deformation in a
volume so large that the differences in actual hardness of its component
microvolumes are of no influence, and 2) microhardness, which is the
hardness of materials in microscopically small volumes (hardness of the
separate structural components).
Hardness measurements for determining the properties of a ma-
chine component have found extensive application in the quality control
of metals and metal products in all branches of industry due to the rapidity
and simplicity of the tests and their nondestructive character.
So, the most common form of hardness tests for metals involves
standard indentors being pressed into the surface of the material con-
cerned. Measurements associated with the indentation are then taken as a
measure of the hardness of the surface. The Brinell test, the Vickers test
and the Rockwell test are the main forms of such tests.
The most widely employed hardness testing methods are: Ball-
indentation tests (Brinell principle). This method consists in pressing a
hardened steel ball for a time of 10 to 15 s under a constant standard force
P into a specially prepared flat surface on the test specimen (Figure 1.21).
After removing the load, an indentation remains on the surface of the test

Figure 1.21 – Brinell hardness test

At standard measurement of hardness on Brinell a steel ball in di-

ameter D under loading P is pressed in the examinee the sample during
certain time, τ. After the load and ball have been removed, the diameter d
of the indentation is measured.
The Brinell hardness number, signified by HB, is the ratio of the
load, applied to the ball during the test, to the area of the indentation pro-
duced by pressing the ball into the test specimen:
Applied force
Brinell hardness number = .
Surface area of indentation
The units used for the area are mm2 and for the force kg · f
(1 kg · f = 9.8 N and is the gravitational force exerted by 1 kg). The area
can be obtained, from the measured diameter of the indentation and ball
diameter, either by calculation or the use of tables:

Area =
πD ⎡ D −
⎢⎣ (D 2
− d2 ⎤ .
If the indented area F is expressed by the ball diameter D and the
diameter of the indentation d, then the Brinell hardness number may be
determined from the formula
HB = , [kg · f/mm2] or [MPa].
( )
πD D − D − d 2 2

The ball diameter and applied load are constant and are selected to
suit the composition of the metal, its hardness, and the thickness of the
test specimen. The diameter of the indentation is measured with a special
magnifying glass containing a scale graduated in tenths of a millimeter.
The diameter D (mm) of the ball used and the size of the applied
force P (kgf units) are chosen, for the British Standard, to give P / D2 val-
ues of 1, 5, 10 or 30 with the diameters of the balls being 1, 2, 5 or
10 mm. In principle, the same value of P / D2 should give the same hard-
ness value, regardless of the diameter of the ball used. It is necessary for
the impression to have a diameter of between 0.25 D and 0.50 D if accu-
rate values of the hardness are to be obtained. The P / D2 value is thus
chosen to fit the materials concerned, the harder the material the higher
the value used. For steels and cast iron the value used for P / D2 is 30, for
copper alloys and aluminium alloys 10, for pure copper and aluminium 5
and for lead, tin and tin alloys 1.
Diameter of a ball (indenter) and loading choose from several factors:
- The thickness of the sample should be not less tenfold depth of a
- The distance from edge of the sample to the indentation centre
should be more or equally 2,5d (9d – diameter of indentation), and centre
to centre – it is more than two next indentation or equally 4d (on the sam-
ple do not less than two indentation);
- The more the indentation, the more exact its measurement, there-
fore we choose the largest size of indenter, satisfies first two requirements.
The thickness of the material being tested should be at least ten
times the depth of the indentation if the results are not to be affected by
the thickness of the material. Also, because of the large depth of the in-
dentation, it cannot be used on plated or surface hardened materials since
the result will be affected by the underlying material.
The Brinell test cannot be used with very soft or very hard materi-
als. In the one case the indentation becomes equal to the diameter of the
ball and in the other there is either no or little indentation on which meas-
urements can be based. Also, with very hard materials the material de-
forms the indenter. The Brinell test is thus limited to materials with hard-
nesses up to about 450 HB with a steel ball and 600 HB with a tungsten
carbide ball. It is not advisable to apply Brinell tests to materials having a
hardness which exceeds 450 HB since the ball may be easily deformed
and this will introduce errors into the test results. The hardness of the ball
should be at least 1.7 times higher than the test specimen to prevent per-
manent set in the ball.
When during the Brinell test a hardened steel ball D = 10 mm is
pressed for a time 10 s into the surface of the material by a standard force
P = 3000 kg · f, hardness number write down like 400 HB, 250 HB etc. At
use of other conditions of test it is recommended to supplement designa-
tion HB with the figures specifying diameter of the used ball (mm), load-
ing (kg) and time (s). For example, 350 HB 5/750/30 is a number of hard-
ness on Brinell (350), received at indentation of ball D = 5 mm under
loading P = 750 kg · f for a time τ = 30 s.
In practice, the Brinell numbers corresponding to a given observed
diameter of indentation d are taken from tables to simplify calculations.
Determining hardness by the depth of indentation of a diamond
cone or small steel ball (Rockwell principle). This test, like the Brinell
test, is based on the indentation of a hard tip or indenter under a static load
into the test specimen. In the Rockwell test the hardness is determined not
by the diameter but by the depth of the indentation or impression. The
Rockwell hardness test differs from the Brinell hardness tests in not ob-
taining a value for the hardness in terms of the area of an indentation but
using the depth of indentation, this depth being directly indicated by a
pointer on a calibrated scale.
The indenter in a Rockwell test may be either a conical-shaped
diamond (called a “brale”) with a 120° apex angle or a hardened steel ball
1.5875 mm in diameter (Figure 1.22). The brale is used for testing materi-
als with a high hardness and the steel ball for softer materials. The brale
and the ball are indented by two consecutive loads: the preliminary load
P0 (equal to 10 kg · f) and the additional load P which equals 90 kg · f for
the ball (Scale B) and 140 kg · f for the brale (Scale C). A brale with an
applied load of 50 kg · f (Scale A) is employed for very hard materials and
thin test specimens. The Rockwell superficial-hardness tester is applied
for testing thin layers near the surface or thin specimens (nitrided steel,
safety-razor blades, etc.). In this machine, the preliminary load is 3 kg and
the total loads may be 15, 30, or 45 kg · f.

Figure 1.22 – Rockwell hardness test

There are a number of Rockwell scales (Table 1.4), the scale being
determined by the indenter and the additional force used. In any reference
to the results of a Rockwell test the scale letter must be quoted. At use
indenter a diamond, cone hardness on Rockwell define on three scales: A,
C, D (HRA, HRC, HRD). At work with steel balls hardness HR define on
six scales – B, E, F, G, H, K, L, M, R. More often than others scales A, C
and B are used.
After loading, the additional load is removed. The Rockwell hard-
ness number is the difference in depths of the indentations made by apply-
ing the total and preliminary loads, measured after removing the addi-
tional load.
This is the permanent increase in penetration e due to the additional
force. The Rockwell hardness number is then given by:
Rockwell hardness number (HR) = E – e,
where E is a constant determined by the form of the indenter. For
the diamond cone indenter E is 100, for the steel ball 130.
The Rockwell hardness number is read directly on the dial of the
instrument having a scale graduated in hardness units.
Table 1.4 – Rockwell hardness scales
Scale Indenter Typical applications
load, kg · f
A Diamond 60 Extremely hard materials, e.g. tool steels
B Ball 1.588 mm dia. 100 Softer materials, e.g. Cu alloys, Al alloys, mild steel
C Diamond 150 Hard materials, e.g. steels, hard cast irons, alloy steels
D Diamond 100 Medium case hardened materials
E Ball 3.175 mm dia. 100 Soft materials, e.g. Al alloys, Mg alloys, bearing metals
F Ball 1.588 mm dia. 60 As E, the smaller ball being more appropriate
where inhomogeneities exist
G Ball 1.588 mm dia. 150 Malleable irons, gun metals, bronzes
H Ball 3.175 mm dia. 60 Soft aluminium, lead, zinc, thermoplastics
K Ball 3.175 mm dia. 150 Aluminium and magnesium alloys
L Ball 6.350 mm dia. 60 Soft thermoplastics
M Ball 6.350 mm dia. 100 Thermoplastics
R Ball 12.70 mm dia. 60 Very soft thermoplastics

Rockwell tests are widely applied in industry due to the rapidity and
simplicity with which they may by performed, high accuracy achieved.
And for the most commonly used indenters with the Rockwell test the size
of the indentation is rather small. Localized variations of structure, com-
position and roughness can thus affect the results. The Rockwell test is,
however, more suitable for workshop or 'on-site' use as it less affected by
surface conditions than the Brinell or Vickers tests which require flat and
polished surfaces to permit accurate measurements. A variation of the
Rockwell test has to be used for thin sheet, this test being referred to as
the Rockwell superficial hardness test. Smaller forces are used and the
depth of indentation which is correspondingly smaller is measured with a
more sensitive device. The initial force used is 29.4 N. Table 1.5 indicates
the scales given by this test.

Table 1.5 – Rockwell superficial hardness scales
Scale Indenter Additional load, kg · f
15-N Diamond 15
30-N Diamond 30
45-N Diamond 45
15-T Ball 1.588 mm dia. 15
30-T Ball 1.588 mm dia. 30
45-T Ball 1.588 mm dia. 45
Note: the N scales are used for materials that if thick enough would have been tested on the
С scale, the T scales for those on the В scale.

The Rockwell hardness number may be converted into the Brinell

number using special tables or charts.
Vickers principle – is determining hardness by indentation of a
diamond pyramid. This testing method is used for determining hardness of
specimens of small cross-section or of their external layers on case-
hardened, nitrided, etc., specimens having a very high hardness. The
Vickers hardness test involves a diamond indenter, in the form of a
square-based pyramid with an angle of 136° between opposite faces, be-
ing pressed under load for 10 to 15 s into the surface of the material under
test (Figure 1.23). The result is a square-shaped impression. After the load
and indenter are removed the diagonals d of the indentation are measured.

Figure 1.23 – Vickers hardness test

The Vickers hardness number (HV) is obtained by dividing the size

of the force P, in units of kg · f, applied by the surface area, in mm2, of the
Applied force
Viskers hardness number = .
Surface area of indentation
The surface area can be calculated from the mean diagonal value,
the indentation being assumed to be a right pyramid with a square base
and an apex angle в of 136°, or obtained by using tables.

The hardness number is determined from the formula:
2 P sin
HV = 2 = 1.854 P , (1.23)
d d2
where P is the applied load (5, 10, 20, 30, 50, 100, or 120 kg);
α is the angle between opposite faces of the pyramid
(α = 136°);
d is the arithmetical mean, in mm, of the two diagonals
measured after removing the load.
The Vickers test has the advantage over the Brinell test of the in-
creased accuracy that is possible in determining the diagonals of a square
as opposed to the diameter of a pyramid. The square indentation produced
for a particular material depends on the force used, but because the inden-
tation is always the same square shape regardless of how big the force and
the surface area is proportional to the force, the hardness value obtained is
independent of the size of the force used. Typically a load of 30 kg is used
for steels and cast irons, 10 kg for copper alloys, 5 kg for pure copper and
aluminium alloys, 2.5 kg for pure aluminium and 1 kg for lead, tin and tin
alloys. The load is selected in accordance with the size and hardness of the
specimen. The thinner the specimen, the less load required. On the other
hand, the higher the load, the more accurate the results will be. As a rule,
the Vickers hardness number is determined from special tables in accor-
dance with the measured value of d (length of the diagonal).
Up to a hardness value of about 300, the hardness value number
given by the Vickers test is the same as that given by the Brinell test.
At higher hardness, the Vickers number is larger than the Brinell number.
Microhardness. At the present time microhardness tests are widely
used to determine the hardness of exceedingly thin layers, very small
specimens, and even separate structural components of alloys. This method
is based on combining a hardness tester, using a diamond pyramid indenter
at low loads, with a metallurgical microscope. Loads from 1 to 200 g are
applied. The microhardness number is determined from the formula:
H = 1.854 , (1.24)
where P is the load, in g;
d is the length of the diagonal, in microns.

Specimens for microhardness tests must be prepared in the same
manner as microsections. It is advisable to use electropolishing as this re-
moves the thin layer of cold-hardened metal, that is inevitable in grinding
the surface and which may affect the results of the test.
Comparison of the different hardness scales. The Brinell and Vick-
ers tests both involve measurements of the surface area of indentations,
the forms of the indenters used being different. The Rockwell test in-
volves measurements of the depth of penetration of indenters. Thus the
various tests are concerned with different measurements as an indication
of hardness. Consequently the values given by the different methods differ
for the same material. There are no simple theoretical relationships be-
tween the various hardness scales, though some simple approximate, ex-
perimentally derived, relationships have been obtained. Different relation-
ships, however, hold for different metals. The relationships are often pre-
sented in the form of tables.
There is an approximate relationship between hardness values and
tensile strengths:
tensile strength = k × hardness,
where k is a constant for a particular material. Thus for annealed
steels the tensile strength in MPa is about 3.54 times the Brinell hardness
value, and for quenched and tempered steels 3.24 times the Brinell hard-
ness value. For brass the factor is about 5.6 and for aluminium alloys
about 4.2.

Fracture toughness test

Crack development. The theoretical fracture strength of brittle ma-
terials, calculated on the basis of atomic bonding considerations, is very
high, typically on the order of E/10, where E is the modulus of elasticity.
For example, the theoretical strengths of aluminum oxide (Al2O3) and zir-
conia (ZrO2) are approximately 39.3 GPa and 20.5 GPa, respectively
( )
EAl2O3 = 393 GPa, and EZrO2 = 205 GPa . The actual strengths are in the
range 0.80–1.50 GPa for zirconia and 0.28–0.70 GPa for alumina. The
actual fracture strength of most brittle materials is 10 to 1000 times lower
than their theoretical strength. The discrepancy arises because of minute
flaws (cracks) that are universally present at the surface and in the interior
of all solids. These cracks locally amplify or concentrate the stress, espe-
cially at the tip of the crack. The actual stress experienced by the solid
near the crack tip could exceed the fracture strength of the material even
when the applied stress is only a fraction of the fracture strength. For a
long, penny-shaped crack of length 2a, with a tip of radius pt, the distribu-
tion of stress in the material is schematically shown in Figure 1.24.

Figure 1.24 – The geometry of surface and internal cracks (a)

and schematic stress profile along the line x–x′ in (a),
demonstrating stress amplification at the crack tip (b)

The maximum stress σm occurs at the crack tip, and is given from
the formula:
⎛ a ⎞
σm = 2σ0 ⎜ ⎟ , (1.25)
⎝ pt ⎠
where σ0 is the applied stress, and the stress ratio (σm/σ0) is called the
stress concentration factor K. Thus, long cracks with sharp tips will raise
the stress more than small cracks with large tip radii. For fracture to occur,
a crack must extend through the solid, a process that creates additional
free surface within the material. The elastic strain energy released on
crack propagation is consumed in creating the new surface that has an ex-
cess energy (surface energy) associated with it. The critical stress σc (i.e.,
the fracture strength), required to propagate a crack of length 2a through a
solid of elastic modulus E, and surface energy γsv is given from the Grif-
fith equation, applicable to brittle solids, which is
⎛ 2 E γ sν ⎞
σc = ⎜ ⎟ . (1.26)
⎝ πa ⎠
In the Griffith equation, the parameter a is the depth of a surface
crack, or half-length of an internal crack. For metals (and some polymers),
fracture is accompanied by plastic deformation, so that a portion of the
stored mechanical energy is used up in the plastic deformation accompa-
nying crack propagation. For such materials, the Griffith equation has
been modified to account for the material plasticity in terms of a plastic
energy γp, so that

( ) ⎤⎥
⎡ 2 E γ sν + γ p
σc = ⎢ . (1.27)
⎢⎣ πa ⎥⎦

Fracture toughness. A useful parameter to characterize a material's

resistance to fracture is the fracture toughness of the material. Fracture
toughness is most commonly specified for the tensile fracture mode, or
Mode I, in which a preexisting crack extends under a normal tensile load.
Two other modes of fracture are encountered, albeit to a lesser degree, in
solids. These are shear failure and failure by tearing (Figure 1.25).

Figure 1.25 – The three modes of crack displacement:

mode I (crack opening or tensile mode), Mode II (sliding mode),
and Mod III (tearing mode)

Mode I fracture toughness defined for the tensile fracture mode is

also called the plain-strain fracture toughness K1c, and is given from
K1c = Y σ πa , (1.28)
where Y is a geometric factor.
Fracture toughness can be defined as a measure of the ability of a
material to resist the propagation of a crack. The toughness is determined by
loading a sample of the material which contains a deliberately introduced
crack of length 2a and recording the tensile stress σ at which the crack
propagates. The smaller the value of the toughness the more readily a crack
propagates. The value of the toughness depend on the thickness of the mate-
rial, high values occurring for thin sheets and decreasing with increasing
thickness to become almost constant with thick sheets (Figure 1.26).

Figure 1.26 – Effect of sample thickness on fracture toughness Kс

Fracture toughness test are doing on the special samples, according

to the scheme of eccentric tension loading or three-point bending. The
samples have rectangular cross-section and one-sided sharp notch at the
apex of which the pulsator induced fatigue crack (Figure 1.27). For a cor-
rect determination K1c ratio of sample size, the notch and the crack should
ensure that conditions of plane deformation at the crack tip and elastic-
stress state away from her, the thickness of the sample section (B) should
be not less than 2.5 (K1c / σ0.2) 2, where σ0.2 – yield strength of the mate-
rial in the normal tension in the same conditions (see fig. 1.26).
For this reason a value called the plane strain fracture toughness K1c
is often quoted. This is the value of the toughness that would be obtained
with thick sheets. Typical values are of the order of 1 MPa⋅m1/2 for glass,
which readily fractures when there is a crack present, to values of the or-
der of 50 to 150 MPa·m1/2 for some steels and copper alloys. In such mate-
rials cracks do not readily propagate.
Table 1.6 gives the K1c values of several materials. Brittle materials
have a low K1c, whereas ductile materials have a high K1c. The mode I
toughness, K1c, depends also on the microstructure, temperature, and
strain rate. Generally, fine-grained materials exhibit a high K1c, and high
temperatures and low strain rates increase the K1c.
Figure 1.27 – Fracture toughness test pieces:
a and b – for three-point bending; c and d – for tensile loading

Table 1.6 – Mode I fracture toughness of selected metallic, polymeric and ceramic materials
Material K1c, MPa·m1/2
Cu alloys 30–120
Ni alloys 100–150
Ti alloys 50–100
Steels 80–170
Al alloys 5–70
Polyethylene 1–5
Polypropylene 3–4
Polycarbonate 1–2.5
Nylons 3–5
GFRP 20–60
Alumina 3–5
Si3N4 4–5
MgO 3
SiC 3
ZrO2 (PSZ) 8–13
Soda glass (Na2O–SiO2) 0.7–0.8
Concrete 0.2–1.4

The engineering value of K1c is in calculating the critical crack size

that would cause catastrophic failure in a component under a given ap-
plied stress σ. This critical size is given from the formula:
1⎛K ⎞
ac = ⎜ 1c ⎟ . (1.29)
π ⎝ Yσ ⎠
Alternatively, a maximum stress σ can be specified for a given ma-
terial containing cracks of known size such that the material will not fail.
This stress is
σ= . (1.30)
Y πα
Sophisticated tests and analytical methods have been developed to
measure the K1c of materials in various geometrical configurations. A
qualitative assessment of the relative fracture toughness of different mate-
rials, especially at high strain rates, is readily obtained from a simple test,
called the Charpy V-notch test. The test utilizes a standard specimen with
a V-notch machined on one of its faces. A pendulum weight is allowed to
fall from a fixed height and strike the sample securely positioned on a fix-
ture to expose its back-face (i.e., the face opposite the V-notch) to the
striking pendulum. The energy needed to fracture the specimen is ob-
tained from the difference in the height of the swinging pendulum before
and after the impact. A semiquantitative assessment of toughness can also
be obtained from the measurement of total crack length in a material when
a microscopic indentation is made on its surface under a fixed load. The
plastic yielding and stress accommodation in a ductile material will pre-
vent cracking, whereas fine, hairline cracks will emanate from the inden-
tation in a brittle material. Examination of the fracture surface of a mate-
rial can also reveal whether the failure is predominantly brittle or ductile.
The scanning electron micrographs in Figure 1.28, a shown that the in-
termetallic compound NiAl fractures in a predominantly intergranular
manner, whereas the same material with tungsten alloying exhibits trans-
granular fracture (Figure l.28, b), which is indicative of a predominantly
ductile mode of fracture in this material.

a b

Figure 1.28 – Scanning electron micrograph (SEM) view of the fracture

surface of an extruded Ni–46 Al alloy, compression tested at 300 K,
showing intergranular fracture (a); SEM view of the fracture surface
of a directionally solidified and compression deformed Ni–48.3 Al–1 W
alloy showing evidence of transgranular fracture (b)

Impact test
Impact tests are designed to simulate the response of a material to a
high rate of loading and involve a test piece being struck a sudden blow.
The main form of test is the Charpy test. This test involves the same type
of measurement but differ in the form of the test pieces. Involve a pendu-
lum (Figure 1.29) swinging down from a specified height h0 to hit the test
piece and fracture it.

Figure 1.29 – Impact testing

The height h to which the pendulum rises after striking and break-
ing the test piece is a measure of the energy used in the breaking. If no
energy were used the pendulum would swing up to the same height h0 it
started from, i.e., the potential energy mgh0 at the top of the pendulum
swing before and after the collision would be the same. The greater the
energy used in the breaking, the greater the “loss” of energy and so the
lower the height to which the pendulum rises. If the pendulum swings up
to a height h after breaking the test piece then the energy used to break it
is mgh0 – mgh.
Charpy test pieces. With the Charpy test, the energy absorbed in
breaking a test piece in the form of a beam is measured (Figure 1.30). The
standard machine has the pendulum hitting the test piece with energy of
300 ± 10 J. The test piece is supported at each end and notched at the
midpoint between the two supports. The notch is on the face directly op-
posite to where the pendulum strikes the test piece. The British and Euro-
pean Standard is BSEN 10045.
Figure 1.30 – Charpy test

For metals, the test piece generally has a square cross-section of

side 10 mm and length 55 mm with there being 40 mm between the sup-
ports. Figure 1.31 shows details of such a test piece and the forms of notch
commonly used.

Figure 1.31 – Charpy metal test pieces (a) and notch details (b)

With the V-notch, reduced width specimens of 7.5 and 5 mm can be

used. For plastics, the test pieces may be unnotched or notched. A standard
test piece is 120 mm long, 15 mm wide and 10 mm thick in the case of
moulded plastics. With sheet plastics the width can be the thickness of the
sheet. The notch is U-shaped with a width of 2 mm and a radius of 0.2 mm
at its base. For moulded plastics the depth below the notch is 6.7 mm, for
the sheet plastics either 10 mm or two-thirds of the sheet thickness.
Impact test results. In stating the results of impact tests it is vital
that the form of test is specified. There is no reliable relationship between
the values obtained by the two forms of test and so values from one test
cannot be compared with those from the other. In addition there is no reli-
able relationship between the impact energies given for breaking test
pieces of different sizes or different notches with the same test method.
The impact energy value obtained for a material is influenced by such fac-
tors as the temperature, the speed of impact, any degree of directionality
occurring in the properties of the material from which the test piece was
cut, and the thickness of the test piece.
For both the Izod and Charpy tests, the impact strengths for metals
are expressed in the form of the energy absorbed, i.e. as, for example 30 J.
For plastics, with the Izod test the results are expressed as the energy ab-
sorbed in breaking the test piece divided by the width of notch and with
the Charpy test as the energy absorbed divided by either the cross-
sectional area of the specimen for unnotched test pieces or by the cross-
sectional area behind the notch for notched test pieces, e.g. 2 kJ/m2.
Interpreting impact test results. When a material is stretched energy
is stored in the material. Think of stretching a spring or a rubber band.
When the stretching force is released the material springs back and the
energy is released. However, if the material suffers a permanent deforma-
tion then all the energy is not released. The greater the amount of such
plastic deformation the greater the amount of energy not released, Thus
when a ductile material is broken, more energy is 'lost' than with a brittle
material. Thus the impact test can be used to give information about the
type of fracture that occurs. For example, Figure 1.32 shows the effect of
temperature on the Charpy V-notch impact energies obtained for test
pieces of the 0.2 % carbon steel. Above about 0 °C the material gives duc-
tile failures, below that temperature – brittle failures. Such graphs have a
great bearing on the use that can be made of the material, since at low
temperatures the steel can be easily shattered by impact.
The appearance of the fractured surfaces after an impact test also
gives information about the type of fracture that has occurred. With a brittle
fracture of metals, the surfaces are crystalline in appearance. With a ductile
fracture, the surfaces are rough and fibrous in appearance. Also with ductile
failure there is a significant reduction in the cross-sectional area of the test
piece, but with brittle fracture there is virtually no such change.
The impact properties of plastics vary quite significantly with tem-
perature, changing in many cases from brittle to tough at some particular
transition temperature.
With plastics, a brittle failure gives fracture surfaces which are
smooth and glassy or somewhat splintered with a ductile failure the sur-
faces often have a whitened appearance. Also, the change in cross-
sectional area can be considerable with a ductile failure but negligible
with brittle failure.
Figure 1.32 – Effect of temperature for the 0.2 % carbon steel

One use of impact tests is to determine whether heat treatment has

been successfully carried out. A comparatively small change in heat
treatment can lead to quite noticeable changes in impact test results. The
changes can be more pronounced than changes in other mechanical prop-
erties such as percentage elongation or tensile strength.

Fatigue test
Failure caused by a fluctuating stress (periodic or random) acting
on a material that is smaller in magnitude than the material's tensile or
yield strength under static load, is called the fatigue failure. Parameters
used to characterize fatigue include: stress amplitude σa, mean stress σm,
and stress range σr. These parameters are defined as follows:
σ max − σ min σ + σ min
σa = , σ m = max , σ r = σ max − σ min .
2 2
Fatigue behavior is most conveniently characterized by subjecting a
specimen to cyclic tensile and compressive stresses until the specimen
fails (Figure 1.33).

Figure 1.33 – Principle of fatigue test

The results (Figure 1.34) of a series of such tests depict the number
of cycles to failure N, as a function of the applied stress S (usually ex-
pressed as stress amplitude σa).

Figure 1.34 – Fatigue response depicted as the stress amplitude (S) versus
number (N) of cycles to failure curves: a – the fatigue limit or endurance limit
is reached at a specific N value; b – material does not reach a fatigue limit

The general response of most materials is represented by one of the

two S – N curves shown in Figure 1.31. Some iron and titanium alloys
attain a limiting stress, called the fatigue limit or endurance limit (Fig-
ure l.34, a), below which they survive an infinite number of stress cycles.
In contrast, many nonferrous alloys do not exhibit a fatigue limit, and the
number of cycles to failure N, continuously decreases with increasing
stress S (Figure l.34, b). For these materials, fatigue strength is specified
as the value of stress at which the material will fail after a specified num-
ber of cycles N (usually, 107). Alternatively, fatigue life of a material can
be defined as the number of cycles to failure at a specified stress level.
Fatigue life is influenced by a large number of variables. For example,
minute surface or bulk imperfections (e.g., cracks, microporosity) in the
material, design features that could raise the stress locally, surface treat-
ments, and second-phase particles all influence the fatigue life.
In reality, the S – N curves of the type shown in Figure 1.34 are
characterized by a marked scatter in the data, usually caused due to the
sensitivity of fatigue to metallurgical structure, sample preparation, proc-
essing history, and the test conditions. As a result, for design purpose, fa-
tigue data are statistically analyzed and presented as probability of failure
curves. Fatigue failures occur in three distinct steps: crack initiation, crack
propagation, and a rather catastrophic final failure. Introducing compres-
sive stresses in the surface of the part (through surface hardening such as
carburizing or shot peening) improves the fatigue life because the com-
pressive stresses counter the tensile stresses during service.
A typical fracture, containing two distinct zones (Figure 1.35), is
obtained as a result of fatigue failure. The first zone, called the fatigue
zone, has a smooth surface and is gradually formed. First, a microcrack is
formed in the weakest region of the cross-section. This gradually develops
into a macrocrack. After the crack has propagated to occupy a consider-
able part of the cross-section, brittle fracture occurs in the remaining part.
The fracture in this second zone is granular or crystalline in appearance as
is typical for brittle fracture.

Figure 1.35 – Typical fatigue failure

Tests at elevated temperatures
Many components of modern machinery operate under high-tempe-
rature conditions. An increase in temperature has a large effect on all me-
chanical properties. Research has shown that elevated temperatures reduce
the modulus of elasticity (by decreasing interatomic binding), yield point,
ultimate strength, and especially the modulus of strain hardening in the
process of plastic deformation.
If a metal is heated above the recrystallization temperature, the
yield point and ultimate strength may be reduced several times while the
modulus of strain hardening may be reduced tens and hundreds of times.
At elevated temperatures, the mechanical properties of metals are a
result of the simultaneous processes of strain hardening, due to plastic de-
formation, and the softening effect of recovery and recrystallization.
Continuous plastic deformation takes place in metals and alloys, sub-
ject to constant loads at elevated temperatures, and it may lead to failure.
Creep is the time-dependent deformation at elevated temperatures
under constant load or constant stress. Creep deformation occurs in all
type of materials, including metals, alloys, ceramics, plastics, rubbers, and
Special tests are conducted to determine the strength of metals at
elevated temperatures. The following chief types of high temperature tests
are most frequently conducted:
Short-time tension tests are similar to tension tests at room tempera-
tures except that the specimen is placed in a furnace. The aim of these
tests is to determine the strength and ductility at a specified temperature.
Creep tests. The purpose of these tests is to determine the creep
limit, i.e., the maximum stress that may be applied for a long period of
time at a given temperature if the rate of creep (over this given period of
time) is not to exceed a specified value.
The specimen is placed in an electric furnace where it is heated to
the given temperature and is constantly subject to a load applied by a lever
and weights. The strain in the specimen is measured by an optical exten-
someter with an accuracy of 0.001 mm. Four or five specimens are tested
at each temperature under different loads and elongation vs time curves
(creep curves) are plotted for each specimen.
The section 0a on creep curves (Figure 1.36) corresponds to elastic
and plastic deformation due to the momentary application of the load. In
the following section ab, the metal is elongated at a nonuniform (decreas-
ing) rate and in section bc the rate of creep is constant.

The creep limit may be characterized, for example, by the stress
which produces a total elongation of 1 per cent in 300 hrs. On the creep
curves, this will correspond to the stress σ3. If the specifications require
that the creep limit be determined for an elongation not exceeding 1 per
cent in 100 hrs, the stress value will be σ4 (see fig. 1.36).
Only the creep rate for a steady process is taken into consideration
for machine components that are designed for continuous operation at ele-
vated temperatures. Limiting conditions are specified for such compo-
nents. In practice the stresses are most frequently determined for a total
elongation of 1 percent in 1000, 10000, and 100 000 hrs. These corre-
spond to average creep rates of 10–3, 10–4 and 10–5 percent per hour, re-

Figure 1.36 – Creep curves

Creep–rupture tests (Figure 1.37) are creep tests which are carried
on to the failure of the specimen. In these tests, the time to fracture the
specimen, at a given temperature and stress, is measured. Often permanent
elongation and reduction of cross-sectional area are also measured.
The deformation strain versus time behavior for high-temperature
creep under constant load is characterized by three distinct regions, fol-
lowing an initial instantaneous (elastic) deformation region. These three
creep regimes are (1) primary creep with a continuously decreasing creep
rate; (2) secondary or steady-state creep, during which the creep rate is
constant; and (3) tertiary creep, during which creep rate is accelerated,
leading to eventual material failure.
Figure 1.37 – Creep curve after creep–rupture tests

In primary creep, the decrease in the creep rate is caused by the ma-
terial's strain hardening. In the steady-state regime, usually the longest in
duration, equilibrium is attained between the competing processes of
strain hardening and recovery (softening). The creep failure toward the
end of the third stage is caused by the structural and chemical changes
such as de-cohesion at the grain boundaries, and nucleation and growth of
internal cracks and voids. From the material and component design per-
spectives, the steady-state creep rate (dεt/dt) is of considerable impor-
tance. The steady-state creep rate depends on the magnitude of applied
stress, σ, and temperature, T, according to
d εt ⎛ Q ⎞
= K ′σ n exp ⎜ ⎟, (1.31)
dt ⎝ RT ⎠
where Q is the activation energy for creep;
K' and n are empirical constants;
R is the gas constant.
The exponent n depends on the dominant mechanisms of creep un-
der a given set of experimental conditions, which include vacancy
diffusion in a stress field, grain boundary migration, dislocation motion,
grain boundary sliding, and others. The materials used in modern aircraft
engines provide a classic example of how high-temperature creep
resistance is enhanced through microstructure design in Ni-base alloys by
dispersing nanometer-size oxide particles.

Control questions
1 How many independent variables contain stress tensor?
2 Why yield strength in fine-grained polycrystals is above, than in
the coarse-grained?
3 Why destruction is structurally more-sensitive process, than plas-
tic deformation?
4 In what the basic difference of brittle fracture from the ductile
fracture consists?
5 What mechanism of nucleation of brittle and ductile cracks?
6 What are general features of a structure of ductile fracture?
7 What is definition of brittleness temperature of a material and
how it to define experimentally?
8 How to define critical brittleness temperature Т50?
9 Why the same alloy with small grain has brittleness temperature
the lower, than with large grain?
10 Which characteristics of strength and plasticity define at tension
tests are?
11 Why true stresses and deformations have physical sense only?
12 How to define effective elongation of the sample if its settlement
length has doubled?
13 Why unit elongation is always more than effective elongation?
14 Two identical samples are tested by the same machine on a ten-
sion and compression. Whether will their strength limit and limit of pro-
portionality be identical?
15 How ductility of a material is connected with plasticity? Which
characteristic is more structurally-sensitive? Why?
16 Why at measurement of hardness НВ different indenters on the
same sample identical results turn out only under condition of a constancy
of the relation of loading to diameter of a ball?
17 What is the difference of hardness numbers НВ 5/250/30-110,
and НВ 110?
18 On the same sample hardness by a Rockwell test HRA, HRC,
and HRCэ is measured. How their numerical values differ?
19 What does the impact strength size characterize? On what sam-
ples it define?
20 In what case temperature ductile-brittle transition more low – at
its definition on value KCU or KCV?
21 What aim of the tests at elevated temperatures of metals and al-

2.1 Basic concepts and definitions
The structure-sensitive properties like strength, ductility or tough-
ness depend critically on things like the composition of the metal and on
whether it has been heated, quenched or cold formed. Both alloying and
heat treatment are controlling the structure of the metal.

Crystal and glass structures

All bodies in nature, including, of course, metals, are composed of
During crystallization in a slow cooling, the disordered structure of a
liquid transforms into an ordered structure characteristic of crystalline solids.
Materials which have their particles arranged in a regular repetitive
arrangement which extends throughout the material are termed crystalline;
materials where there is no such orderly arrangement are said to be amor-
Metals usually are polycrystalline substances and they consist of
small crystals or grain in which atoms are packed in regular, repeating,
three-dimensional patterns.
Growing grain of one phase has the same crystal structure, but the
orientation of their crystal lattices differ. Where they meet, at the grain
boundaries, the regular pattern is disturbed.
Many attributes of metals are determined by the type of crystal lat-
tice, i.e. geometric arrangement of their constituent atoms.
A useful model to visualize the atomic packing in crystalline solids
is to first liken each atom as a “hard sphere”, and then identify the small-
est repeating cluster of atoms (unit cell) that could be stacked in three di-
mensions to generate the long-range atomic order.
For each metallic element, the hard spheres (with a characteristic
atomic radius) represent a positive ion in a cloud of electrons.
Looking at the smallest scale of controllable structural feature – the
way in which the atoms in the metals are packed together to give either a
crystalline or a glassy (amorphous) structure. Table 2.1 lists the crystal struc-
tures of the pure metals at room temperature. In nearly every case the metal
atoms pack into the simple crystal structures of face-centred cubic (f.c.c),
body-centred cubic (b.c.c.) or close-packed hexagonal (c.p.h.) (Figure 2.1).
Metal atoms tend to behave like miniature ball-bearings and tend to
pack together as tightly as possible. F.c.c. and c.p.h. give the highest pos-
sible packing density, with 74 % of the volume of the metal taken up by
the atomic spheres. However, in some metals, like iron or chromium, the
metallic bond has some directionality and this makes the atoms pack into
the more open b.c.c. structure with a packing density of 68 %. Some met-
als have more than one crystal structure.
Table 2.1 – Crystal structures of pure metals at room temperature
Unit cell dimensions (nm)
Pure metal Structure
a c
Aluminium f.c.c. 0.405
Beryllium c.p.h. 0.229 0.358
Cadmium c.p.h. 0.298 0.562
Chromium b.c.c. 0.289
Cobalt c.p.h. 0.251 0.409
Copper f.c.c. 0.362
Gold f.c.c. 0.408
Hafnium c.p.h. 0.320 0.506
Indium Face-centered tetragonal
Iridium f.c.c. 0.384
Iron b.c.c. 0.287
Lanthanum c.p.h. 0.376 0.606
Lead f.c.c. 0.495
Magnesium c.p.h. 0.321 0.521
Manganese Cubic 0.891
Molybdenum b.c.c. 0.315
Nickel f.c.c. 0.352
Niobium b.c.c. 0.330
Palladium f.c.c. 0.389
Platinum f.c.c. 0.392
Rhodium f.c.c. 0.380
Silver f.c.c. 0.409
Tantalum b.c.c. 0.331
Thallium c.p.h. 0.346 0.553
Tin Body-centered tetragonal
Titanium c.p.h 0.295 0.468
Tungsten b.c.c. 0.317
Vanadium b.c.c. 0303
Yttrium c.p.h. 0.365 0.573
Zinc c.p.h. 0.267 0.495
Zirconium c.p.h. 0.323 0.515

The most important examples are iron and titanium. Iron changes
from b.c.c. to f.c.c. at 914 °C but goes back to b.c.c. at 1391 °С; and tita-
nium changes from c.p.h. to b.c.c. at 882 °C. This multiplicity of crystal
structures is called polymorphism. But it is obviously out of the question to
try to control crystal structure simply by changing the temperature (iron is
useless as a structural material well below 914 °C). Polymorphism can,
however, be brought about at room temperature by alloying. Indeed, many
stainless steels are f.c.c. rather than b.c.c. and, especially at low tempera-
tures, have much better ductility and toughness than ordinary carbon steels.

Figure 2.1 – Main types of the simple crystal structures

This is why stainless steel is so good for cryogenic work: the fast
fracture of a steel vacuum flask containing liquid nitrogen would be em-
barrassing, to say the least, but stainless steel is essential for the vacuum
jackets needed to cool the latest superconducting magnets down to liquid
helium temperatures, or for storing liquid hydrogen or oxygen.
The main parameters for describing the crystal lattice are the crys-
tallographic planes and crystallographic directions.
Crystallographic direction it’s the line between two points in the
unit cell, and denoted as [uvw] where u, v, and w are the projections along
the x, y, and z axes. Crystal planes or atomic planes are specified relative
to the unit cell as (hkl) and for h.c.p. as (hklm).
If molten metals (or, more usually, alloys) are cooled very fast –
faster than about 106 K·s–1 – there is no time for the randomly arranged
atoms in the liquid to switch into the orderly arrangement of a solid crys-
tal. Instead, a glassy or amorphous solid is produced which has essentially
a “frozen-in” liquid structure. This structure – which is termed dense ran-
dom packing (drp) – can be modelled very well by pouring ball-bearings
into a glass jar and shaking them down to maximise the packing density. It
is interesting to see that, although this structure is disordered, it has well-
defined characteristics. For example, the packing density is always 64%,
which is why corn was always sold in bushels (1 bushel = 8 UK gallons):
provided the corn was always shaken down well in the sack a bushel al-
ways gave 0.64 · 8 = 5.12 gallons of corn material! It has only recently
become practicable to make glassy metals in quantity but, because their
structure is so different from that of “normal” metals, they have some very
unusual and exciting properties.

Structures of solutions and compounds

Few metals are used in their pure state – they nearly always have
other elements added to them which turn them into alloys and give them
better mechanical properties. The alloying elements will always dissolve
in the basic metal to form solid solution, although the solubility can vary
between < 0.01 % and 100 % depending on the combinations of elements
we choose. As examples, the iron in carbon steels can only dissolve
0.007 % carbon at room temperature; the copper in brass can dissolve
more than 30 % zinc; and the copper–nickel system – the basis of the mo-
nels and the cupronickels – has complete solid solubility.
There are two basic classes of solid solution. In the first, small at-
oms (like carbon, boron and most gases) fit between the larger metal at-
oms to give interstitial solid solutions (Figure 2.2, a). Although this inter-
stitial solubility is usually limited to a few per cent it can have a large ef-
fect on properties. Indeed, as we shall see later, interstitial solutions of
carbon in iron are largely responsible for the enormous range of strengths
that we can get from carbon steels. It is much more common, though, for

the dissolved atoms to have a similar size to those of the host metal. Then
the dissolved atoms simply replace some of the host atoms to give a sub-
stitutional solid solution (Figure 2.2, b). Brass and cupronickel are good
examples of the large solubilities that this atomic substitution can give.
Solutions normally tend to be random so that one cannot predict
which of the sites will be occupied by which atoms (Figure 2.2, c). But if
A atoms prefer to have A neighbors, or В atoms prefer В neighbors, the
solution can cluster (Figure 2.2, d); and when A atoms prefer В neighbors
the solution can order (Figure 2.2, e).

Figure 2.2 – Solid-solution structures: A – light atoms, B – dark atoms

Many alloys contain more of the alloying elements than the host
metal can dissolve. Then the surplus must separate out to give regions that
have a high concentration of the alloying element. In a few alloys these
regions consist of a solid solution based on the alloying element. (The
lead–tin alloy system, on which most soft solders are based, the lead can
only dissolve 2 % tin at room temperature and any surplus tin will sepa-
rate out as regions of tin containing 0.3 % dissolved lead.) In most alloy
systems, however, the surplus atoms of the alloying element separate out
as chemical compounds. An important example of this is in the alumin-
ium-copper system where surplus copper separates out as the compound
CuAl2. CuAl2 is hard and is not easily cut by dislocations. And when it is
finely dispersed throughout the alloy it can give very big increases in
strength. Other important compounds are Ni3Al, Ni3Ti, Mo2C and TaC (in
super-alloys) and Fe3C (in carbon steels).

Metal crystals, amorphous metals, solid solutions, and solid com-
pounds are all phases. A phase is a region of material that has uniform
physical and chemical properties. Water is a phase – any one drop of wa-
ter is the same as the next. Ice is another phase – one splinter of ice is the
same as any other. But the mixture of ice and water in your glass at dinner
is not a single phase because its properties vary as you move from water
to ice. Ice + water are two-phase mixture.
Grain and phase boundaries. A pure metal, or a solid solution, is
single-phase. It is certainly possible to make single crystals of metals or
alloys but it is difficult and the expense is only worth it for high-
technology applications such as single-crystal turbine blades or single-
crystal silicon for microchips. Normally, any single-phase metal is poly-
crystalline – it is made up of millions of small crystals, or grains, “stuck”
together by grain boundaries (Figure 2.3).

Figure 2.3 – The structure of a typical grain boundary

Because of their unusual structure, grain boundaries have special

properties of their own. First, the lower bond density in the boundary is
associated with a boundary surface-energy: typically 0.5 Joules per square
metre of boundary area (0.5J·m–2). Secondly, the more open structure of
the boundary can give much faster diffusion in the boundary plane than in
the crystal on either side. And finally, the extra space makes it easier for
outsized impurity atoms to dissolve in the boundary. These atoms tend to
segregate to the boundaries, sometimes very strongly. Then an average
impurity concentration of a few parts per million can give a local concen-
tration of 10 % in the boundary with very damaging effects on the fracture
In order tо “bridge the gap” between two crystals of different orien-
tation the atoms in the grain boundary have to be packed in a less ordered
way. The packing density in the boundary is then as low as 50 %.
When an alloy contains more of the alloying element than the host
metal can dissolve, it will split up into two phases. The two phases are
“stuck” together by interphase boundaries which, again, have special

properties of their own. Two phases which have different chemical compo-
sitions but the same crystal structure are illustrated in Figure 2.4, a. Pro-
vided they are oriented in the right way, the crystals can be made to match
up at the boundary. Then, although there is a sharp change in chemical
composition, there is no structural change, and the energy of this coherent
boundary is low (typically 0.05 J·m–2). If the two crystals have slightly dif-
ferent lattice spacings, the boundary is still coherent but has some strain
(and more energy) associated with it (Figure 2.4, b). The strain obviously
gets bigger as the boundary grows sideways: full coherency is usually pos-
sible only with small second-phase particles. As the particle grows, the
strain builds up until it is relieved by the injection of dislocations to give a
semi-coherent boundary (Figure 2.4, c). Often two phases which meet at
the boundary are large, and differ in both chemical composition and crystal
structure. Then the boundary between them is incoherent; it is like a grain
boundary across which there is also a change in chemical composition
(Figure 2.4, d). Such a phase boundary has a high energy – comparable
with that of a grain boundary – and around 0.5 J·m–2.

Figure 2.4 – Structures of interphase boundaries:

a – coherent; b – coherency strain; c – semi-coherent; d – incoherent

Shapes of grains and phases. Grains come in all shapes and sizes,
and both shape and size can have a big effect on the properties of the
polycrystalline metal (a good example is mild steel – its strength can be
doubled by a ten-times decrease in grain size). Grain shape is strongly
affected by the way in which the metal is processed. Rolling or forging,
for instance, can give stretched-out (or “textured”) grains; and in casting
the solidifying grains are often elongated in the direction of the easiest
heat loss. But if there are no external effects like these, then the energy of
the grain boundaries is the important thing. This can be illustrated very
nicely by looking at a “two-dimensional” array of soap bubbles in a thin
glass cell. The soap film minimises its overall energy by straightening out;
and at the corners of the bubbles the films meet at angles of 120° to bal-
ance the surface tensions. Of course a polycrystalline metal is three-
dimensional, but the same principles apply: the grain boundaries try to
form themselves into flat planes, and these planes always try to meet at
120°. A grain shape does indeed exist which not only satisfies these con-
ditions but also packs together to fill space. It has 14 faces, and is there-
fore called a tetrakaidecahedron. This shape is remarkable, not only for
the properties just given, but because it appears in almost every physical
science (the shape of cells in plants, of bubbles in foams, of grains in met-
als and of Dirichlet cells in solid-state physics).
If the metal consists of two phases then we can get more shapes.
The simplest is when a single-crystal particle of one phase forms inside a
grain of another phase.
Then, if the energy of the interphase boundary is isotropic (the
same for all orientations), the second-phase particle will try to be spheri-
cal in order to minimise the interphase boundary energy.
Naturally, if coherency is possible along some planes, but not along
others, the particle will tend to grow as a plate, extensive along the low-
energy planes but narrow along the high-energy ones. Phase shapes get
more complicated when interphase boundaries and grain boundaries meet.
The particle is shaped by two spherical caps which meet the grain
boundary at an angle θ. This angle is set by the balance of boundary tensions
2γαβ · cosθ = γgb, (2.1)
where γαβ is the tension (or energy) of the interphase boundary;
γgb is the grain boundary tension (or energy).
In some alloys γαβ can be ≤ γgb / 2 in which case θ = 0. The second
phase will then spread along the boundary as a thin layer of β. This “wet-
ting” of the grain boundary can be a great nuisance – if the phase is brittle
then cracks can spread along the grain boundaries until the metal falls
apart completely. A favorite scientific party trick is to put some alumin-
ium sheet in a dish of molten gallium and watch the individual grains of
aluminium come apart as the gallium whizzes down the boundary.
The second phase can, of course, form complete grains. But only if
γαβ and γgb are similar will the phases have tetrakaidecahedral shapes
where they come together. In general, γαβ and γgb may be quite different
and the grains then have more complicated shapes.

Summary: constitution and structure

The structure of a metal is defined by two things. The first is the
a) The overall composition – die elements (or components) that the
metal contains and the relative weights of each of them;
b) The number of phases, and their relative weights;
c) The composition of each phase.
The second is the geometric information about shape and size:
d) The shape of each phase;
e) The sizes and spacings of the phases.
Armed with this information, we are in a strong position to re-
examine the mechanical properties, and. explain the great differences in
strength, or toughness, or corrosion resistance between alloys. But where
does this information come from? The constitution of an alloy is summa-
rized by its phase diagram – the subject of the next chapter. The shape
and size are more difficult, since they depend on the details of how the
alloy was made. But, as we shall see from later chapters, a fascinating
range of microscopic processes operates when metals are cast, or worked
or heat-treated into finished products; and by understanding these, shape
and size can, to a large extent, be predicted.

2.2 Transformations in steel during heating and cooling

Heat treatment is done to transform the metallurgical structure in al-
loys to develop the desired properties. Iron-carbon alloys serve as a classic
example of how thermal treatments can be designed to transform the met-
allurgical structure for enhanced properties. Heat treatment practices make
use of the thermodynamic data provided by the Fe–C (more often the Fe–
Fe3C) phase diagram, and the experimental measurements of the trans-
formation kinetics.
Many (but not all) phase transformations involve atomic diffusion
and occur in a manner somewhat similar to the crystallization of a liquid
via nucleation and growth.
Table 2.2 lists the phases in the Fe–Fe3C system gives details of the
composite eutectoid and eutectic structures that occur during slow cooling.
Iron–carbon alloys are the most widely used structural materials,
and the phase transformations in these alloys are of considerable impor-
tance in heat treatment. The Fe–Fe3C diagram (Figure 2.5) exhibits three
important phase reactions: (1) a eutectoid reaction at 727 °C, consisting of
decomposition of austenite (γ-phase, f.c.c.) into ferrite (α-Fe, b.c.c.) and
cementite, i.e., γ ↔ α + Fe3C; (2) a peritectic reaction at 1493 °C, involv-
ing reaction of the b.c.c. δ-phase and the liquid to form austenite
(δ + L ↔ γ); and (3) a eutectic reaction at 1147 °C, resulting in the forma-
tion of austenite and cementite (L ↔ γ + Fe3C).
Table 2.2 – Phases in the Fe–Fe3C system and composite structure produced
during the slow cooling of Fe–C alloys
Phase Atomic
Description and comments
and structure packing
Liquid d.r.p. Liquid solution of С in Fe
δ Random interstitial solid solution of С in b.c.с. Fe. Maximum
b.c.c. solubility of 0.08 wt % С occurs at 1492 °С. Pure δ-Fe
is the stable polymorph between 1391 °С and 1536 °С
γ (also called Random interstitial solid solution of С in f.c.с. Fe. Maximum
“austenite”) f.c.c. solubility of 1.7 wt % С occurs at 1130 °С. Pure γ-Fe
is the stable polymorph between 914 °С and 1391 °С
α (also called Random interstitial solid solution of С in b.с.с. Fe Maximum
“ferrite”) b.c.c. solubility of 0.025 wt % С occurs at 723 °С. Pure α-Fe
is the stable polymorph below 914 °С
Fe3C (also called A hard and brittle chemical compound of Fe
“cementite”) and С containing 25 atomic % (6.7 wt %) С
Pearlite The composite eutectoid structure of alternating plates of α
and Fe3C produced when γ containing 0.8 wt % С is cooled
below 723 °C. Pearlite nucleates at γ grain boundaries.
It occurs, in low, medium and high carbon steels.
It is sometimes, quite wrongly, called a phase. It is not
a phase but is а mixture of the two separate phases α
and Fe3C in the proportions of 88.5 % by weight of α
to 11.5 % by weight of Fe3C. Because grains are single
crystals it is wrong to say that pearlite forms in grains:
we say instead that it forms in nodules
Ledeburite The composite eutectic structure of alternating plates of γ
and Fe3C produced when liquid containing 4.3 wt % С
is cooled below 1130 °C. Again, not a phase! Ledeburite
only occurs during the solidification of cast irons, and even
then the γ in ledeburite will transform to α + Fe3C at 723 °C

Figure 2.5 – The iron–iron carbide (Fe–Fe3C) diagram

Transformations in steel during heating
In practice for many types of heat treatment steel is heated to tem-
peratures corresponding to the existence of austenite.
Austenite grain size significantly affects the properties of steel.
It is known that the main transformation that occurs in heating steel
with structure of ferrite and pearlite or perlite and cementite is the trans-
formation of perlite (ferrite-cementite mixture) in austenite.
Formation of austenite during heating is a diffusion process. In its
initial state (before heating), the steel consists of cementite and ferrite.
The process of transforming of the ferrite-cementite mixture into austenite
occurs at the highest rate if the initial structure is finely lamellar pearlite,
at a lower rate if the pearlite is coarsely lamellar, and slowest of all for
granular pearlite.
This is explained by the fact that higher carbide dispersion in-
creases the interface with the ferrite. This, in turn, increases the rate of
austenite nucleation and crystal growth.
Austenite may be formed from perlite only after heating slightly
above the equilibrium temperature (Ac1), because at these temperatures the
free energy of the austenite is less than perlite and all bodies in nature tend
towards equilibrium, to a minimum internal energy.
Consider at the example of the transformation of eutectoid steel
(0.8 % C).
At temperatures slightly above 723 °C (Ac1), due to allotropic trans-
formation α-Fe → γ-Fe formed more energetically favorable than pearlite,
phase – austenite.
Extensive research has shown that the austenite nuclei always ap-
pear on the boundaries between crystals of ferrite and cementite.
Further growth of the austenite nuclei occurs by a diffusion mecha-
nism in both parties – by absorbing parts of ferrite and cementite in perlite.
The austenite nuclei develop due to the dissolution of cementite and
transformation of the ferrite. But after the ferrite has disappeared and even
after the cementite is completely dissolved, the austenite will not be ho-
mogeneous in carbon content.
Cementite contains 6.67 % C and ferrite – only 0.025 % C.
Therefore regions of austenite adjacent to the cementite particles
and the places the particles occupied before being dissolved are always
richer in carbon than regions occupied by the ferrite.
Additional time or higher temperatures, which increase the rate of
carbon diffusion is required to homogenize the carbon concentration
throughout the austenite crystals.
So, the pearlite-austenite transformation consists of three stages:
1) formation of the austenite nuclei and their growth due to the dis-
solution of cementite and ferrite transformation;
2) complete disappearance of ferrite and cementite;
3) equalizing the austenite composition throughout the crystals (ho-
mogenization of austenite).
All the carbon located in the perlite is dissolved in austenite.
In hipoeutektoid and hypereutectoid steels to these basic stages of
the transformation upon heating are added to the stage of dissolution of
free ferrite (in hipoeutektoid steel) and cementite (in hypereutectoid steel)
and the stage of homogenization of the austenite in these zones.
The point lying on the line SG indicate the Ac3, the same on the line
SE – Acm (see fig. 2.5).
With further heating of steels, since the temperature Ac3 (Acm) + 30–
50 °C, there is a growth of austenite grains.
Heating temperature during heat treatment is usually limited to these
values. Although in order to reduce the processing time is desirable to ac-
celerate the diffusion processes, heating the steel to higher temperatures.
The effect of austenite grain growth at an increase in temperature
must be taken into consideration in selecting the rate at which steel is to be
heated. The size of the austenite grains greatly affects the mechanical prop-
erties and, particularly, the impact strength of the steel after heat treatment.
A grain of austenite appears immediately upon the completion of
the pearlite-to-austenite transformation. It is usually called the original
austenite grain. Its size depends on the number of nuclei appearing in a
unit time in a given volume and their rate of growth. The more the carbide
phase is dispersed in the previous structure, the smaller the original grain
of austenite will be.
The austenite grain is ordinarily very small at the moment it appears
but it grows rapidly if held for appreciable time at the given temperature
or at increasing temperatures.
Austenite grain growth is spontaneous and is due to the tendency to
reduce the free energy of the system by decreasing the grain surface.
Grain growth develops by the enlargement of certain grains at the
expense of other smaller grains which are, consequently, less sta-
ble thermodynamically. Holding the austenite at a constant temperature
does not at first cause detectable grain growth. This is the so-called incu-
bation period. The higher the temperature, the less time required for the
incubation period after which intensive growth begins. In this second pe-
riod, considerable nonuniformity of grain size is observed. After this,
grain growth practically ceases and holding the austenite further at this
temperature will only equalize grain size.
Tendency to grain growth may differ even for steels of the same
grade but of different melts.
In this respect, two types of steel are distinguished: 1) inherently
fine-grained and 2) inherently coarse-grained.
Inherently fine-grained steel is characterized by its small tendency
toward austenite grain growth in heating; inherently coarse-grained steel,
on the opposite, is disposed to considerable grain growth under the same
conditions. The inclusions which are available in a steel, (nonmetallic in-
clusions, oxides, nitrides, carbides, etc.) block the growth of austenite
grains. The quantity of the inclusions, their character and distribution de-
pend on metallurgical features of melt of the steel.
At a sufficiently high temperature, an inherently fine-grained steel
may have austenite grains of even larger size than an inherently coarse-
grained steel. For this reason, the term actual grain is employed to indi-
cate the grain obtained in steel at a given temperature or as a result of a
certain type of heat treatment.
The grain size of the structural components of steel is closely re-
lated to that of the prior austenite grains. Consequently, actual grain size,
observed at normal temperatures, depends only on the temperature and
time in heating; upon subsequent cooling, the grain size does not change.
Therefore, actual grain size is determined exclusively by the temperature
to which steel is heated in heat treatment and the tendency toward grain
growth of the given steel in heating.
Actual grain size of steel determines its final properties. Coarsening
of the grain has a comparatively small effect on tensile strength, hardness,
or relative elongation, but sharply reduces impact strength.
It should be noted that when hypoeutectoid steels are heated to
temperatures considerably exceeding the point A3, in addition to austenite
grain growth, excess ferrite may precipitate, on subsequent cooling, as
long plates (or needles) cutting across the pearlite grains. This is called a
Widmanstatten structure. The formation of a Widmanstatten structure in
overheating steel further reduces impact strength. Overheated steel is
characterized by a coarse-grained fracture since the fracture passes along
the boundaries of the large grains. If the austenite grains are small, the
fracture will have a fine-grained structure.
The size of the inherent grain mainly affects the fabrication proper-
ties of steel. Thus, for example, inherently fine-grained steel may be
heated to higher temperatures without the fear of overheating, i.e., appre-
ciable coarsening of the grains.
There are an overheating and burning of steels. The overheating is
characterized by a coarse-grained structure of a steel and can be eliminated
by heat treatment – annealing, normalization or hardening plus tempering.
Burning takes place in case of heating of a steel to the temperature
close to melting point. It is characterized melting and in this connection
metal oxidation on boundaries of grains and cannot be corrected by heat
treatment. Burning is a waste.
The overheating is characterized by a coarse-crystalline brilliant
The overheating is accompanied by sharp reduction of impact
strength – it decreases to 0.1–0.2 МJ/m2.

Transformations in steel during cooling

Figure 2.6 depicts the development of microstructure in three repre-
sentative steels at different stages of cooling: a hypoeutectoid steel, a steel
of eutectoid composition (0.76 % C), and a hypereutectoid steel.
Consider phase transformations in the carbon steels with different
carbon content during slow cooling and how the room temperature micro-
structure of carbon steels depends on the carbon content.
The limiting case of pure iron is straight-forward: when γ-iron cools
below 9l4 °C α-grains nucleate at γ-grain boundaries and the microstruc-
ture transforms to α.
If we cool a steel of eutectoid composition (0.80 wt % C) below
723 °C pearlite nodules nucleate at grain boundaries and the microstruc-
ture transforms to pearlite. Initially cementite nucleates on the boundaries
of the austenite grains (see fig. 2.6, b). Because of the concentration fluc-
tuations, required for forming cementite, on the grains boundaries of aus-
tenite more easily occur and less energy is required for nucleation. The
cementite nucleuses grow due to diffusion of carbon from the adjacent
austenite. The austenite, surrounding the pearlite plate, loses its carbon
and is more easily transformed into ferrite. The ferrite formed does not
dissolve carbon to any appreciable extent, so the carbon is rejected to the
ferrite-austenite boundary where it facilitates the formation of the next
cementite plate. This forms what is called a pearlite colony. Plates of fer-
rite and cementite grow simultaneously. Repetition of this process forms
the pearlite nodule or grain. A cementite nuclei with a different orientation
appears on the boundary of the first pearlite grain and starts a new grain.
The process continues until the grains impinge upon each other and all of
the austenite is transformed into pearlite.

a b c

Figure 2.6 – Microstructure evolution in a Fe–C alloys of hypoeutectoid

composition (a), eutectoid composition (b), hypereutectoid composition (c)
during cooling from within the austenite region to below
the eutectoid temperature

When pearlite is cooling to room temperature, the concentration of

carbon in the α decreases slightly, following the α / α + Fe3C boundary.
The excess carbon reacts with iron at the α-Fe3C interfaces to form more
Fe3C. This is “plates out” on the surfaces of the existing Fe3C plates
which become very slightly thicker. The composition of Fe3C is depend-
ent of temperature.
If the steel contains less than 0.80 % С (a hypoeutectoid steel) then
the γ starts to transform as soon as the alloy enters the (α + γ)-field (see
fig. 2.6, a). Standard labeling for the “primary” α nucleates at γ grain
boundaries and grows as the steel is cooled from A3 to A1 (A3 is the stan-
dard labelling for the temperature at which α first appears, and A1 is stan-
dard for the eutectoid temperature). At A1 the remaining γ (which is now
of eutectoid composition) transforms to pearlite as usual. The room tem-
perature microstructure is then made up of primary α + pearlite.
If the steel contains more than 0.80 % С (a hypereutectoid steel)
then we get a room-temperature microstructure of primary Fe3C plus pear-
lite instead. Acm is the standard labelling for the temperature at which Fe3C
first appears (see fig. 2.6, c).
If you heat up the steel at a temperature slightly below the A1, the
cementite plates, trying to reduce its surface, to be transformed into grain. A
structure of granular pearlite corresponds to the maximum softness of steel.

It is possible sufficiently accurately determine the carbon content in
according to the structure of the steel. In hypoeutectoid steels such crite-
rion is the amount of perlite, in hypereutectoid steel – a thickness of the
cementite network. The cementite network allows us to estimate the car-
bon content of only approximately. A necessary condition for determining
the composition of the steel on the structure, it is a slow cooling from the
austenite region, as at high cooling rate the relation between structural
components may change significantly.
The structure of steel with granular pearlite (obtained in steel with a
carbon content above 0.6 %) the carbon content can be determined by
methods of quantitative metallography, but Uniform distribution of grains
of cementite in the structure of the steel annealed at granular perlite makes
it impossible to determine the carbon content in steel with the same cer-
tainty as can be done in the presence of the structure of lamellar pearlite.
Description of the structure of the annealed carbon steels with the
different carbon content. In alloys with the low carbon content ferrite and
pearlite are present in the form of grains. In alloys with the average carbon
content ferrite located around pearlite in the form of almost continuous
network along the boundaries of former austenite grains. At a further in-
crease of carbon content network is becoming more intermittent, until,
finally, it does not remain only a few small areas of ferrite. When the car-
bon content in the steel is 0.8 % it has purely pearlitic structure. Above
0.8 % C structure of the steel consists of pearlite and hypereutectoid (sec-
ondary) cementite, which appears as a fine network along the boundaries
of austenitic grains. The more the carbon content in steel the thicker a ce-
mentite network. When the carbon content is more than 1.7 % in the
structure appears ledeburite.
The mechanical properties are strongly dependent on both the car-
bon content and on the type of heat treatment. Both the yield strength and
tensile strength increase linearly with carbon content. This is what we
would expect: the Fe3C acts as a strengthening phase, and the proportion
of Fe3C in the steel is linear in carbon concentration. The ductility, on the
other hand, falls rapidly as the carbon content goes up because the α-Fe3C
interfaces in pearlite arc good at nucleating cracks.
Properties of the structural components of the annealed steel in a
relatively equilibrium state are listed in the Table 2.3.
Consider a eutectoid steel that is cooled from a high temperature
where it is fully austenitic to a temperature below 727 °C, and isother-
mally held for different times at the temperature. The isothermal hold will
transform the austenite into pearlite, which is a mixture of ferrite (α-Fe)
and cementite (iron carbide Fе3С). Experiments show that transformation
(conversion) kinetics at a fixed temperature exhibit an S-shaped curve.
Table 2.3 – Mechanical properties of of the structural components of the annealed steel
Mechanical properties
Hardness HB σu, МPа ψ, % δ, %
Ferrite 50–90 186–245 60–75 40–50
Cementite 750–820 30 – –
Lamellar pearlite 190–230 843–882 10–15 9–12
Granular pearlite 160–190 637–686 18–25 18–25

The Avrami equation applies to a variety of diffusion-driven phase

transformations, including recrystallization of cold-worked metals and to
many nonmetallurgical phenomena whose kinetics displays a sigmoid be-
havior. Usually the isothermal transformation data of S-shaped curve is
organized as temperature – versus – time map (with time on a logarithmic
scale) to delineate the fraction (0 to 100 %) transformed. The approach to
construct an isothermal transformation map is shown in Figure 2.7.

Figure 2.7 – Procedure to develop an isothermal transformation diagram

from an experimental Avrami curve

A complete isothermal transformation diagram for an eutectoid

steel is shown in Figure 2.8.
Figure 2.8 – The complete isothermal transformation diagram
for a Fe–C alloy of eutectoid composition

The different iron-carbon phases (e.g. coarse and fine pearlite, up-
per and lower bainite, etc.) that form under different isothermal conditions
are also displayed on this diagram. The physical appearance of some of
the phases as revealed under a microscope is shown in the photomicro-
graphs of Figure 2.9.

Figure 2.9 – Microstructures of α-Fe (ferrite) (a), γ-Fe (austenite) (b),

1.4 wt % C – steel having a microstructure of network cementite
surrounding the pearlite colonies (c), plate martensite (d)
In practice, steels are rapidly cooled in a continuous fashion from
the fully austenitic state to the ambient temperature, rather than held iso-
thermally to trigger the transformation. In continuous cooling of steel, the
time required for a transformation to start and end is delayed, and the iso-
thermal transformation curves all shift to longer times. Figure 2.10 shows
a continuous cooling transformation (CCT) diagram for a eutectoid steel
with lines representing different cooling rates superimposed on the dia-
gram (the constant cooling rate lines appear as curves rather than straight
lines because time is plotted on a logarithmic scale).

Figure 2.10 – The complete I–T diagram for a eutectoid steel:

A – austenite; B – bainite; M – martensite; P – pearlite

Heat treated steels harden because of the formation of a hard and

brittle phase, martensite, which forms only on rapid cooling below a cer-
tain temperature, called the martensite start temperature Ms, which is
shown in Figure 2.10. Martensite is a non-equilibrium phase and does not
appear on the equilibrium Fe–Fe3C phase diagram. Martensite has a body-
centered tetragonal (b.c.t.) structure and a density lower than that of the
f.c.c. austenite. Thus, there is volume expansion on martensite formation
that could cause thermal stresses and cracks in rapidly cooled parts. The
situation is exacerbated for relatively large parts in which the center cools
more slowly than the surface. Special surface-hardening treatments for
steels such as nitriding do not, however, require martensite formation; in-
stead, hard nitride compounds from the alloying elements in steel, and the
solid solution hardening due to nitrogen dissolution, result in a hard and
wear-resistant exterior in nitrided steels.
Steels hardened using martensitic transformation are frequently
given a secondary treatment called tempering to impart some toughness
and ductility. Hardened steels are tempered by heating and isothermal
holding at temperatures in the range 250 to 650 °C; this allows for diffu-
sional processes to form tempered martensite, which is composed of sta-
ble ferrite and cementite phases.

It is known that carbon steels could be strengthened by quenching
and tempering. To get the best properties we must quench the steel past
the nose of the C-curve. The cooling rate that just misses the nose is called
the critical cooling rate (CCR). If we cool at the critical rate, or faster, the
steel will transform to 100 % martensite. The CCR for a plain carbon steel
depends on two factors – carbon content and grain size. Adding carbon
decreases the rate of the diffusive transformation by orders of magnitude:
the CCR decreases from ≈105 °C·s–1 for pure iron to ≈200 °C·s–1 for 0.8 %
carbon steel (Figure 2.11).

Figure 2.11 – Effect of carbon content and grain size

on the critical cooling rate

Also it is known 8 that the rate of a diffusive transformation was

proportional to the number of nuclei forming per m3 per second. Since
grain boundaries are favorite nucleation sites, a fine-grained steel should
produce more nuclei than a coarse-grained one. The fine-grained steel will
therefore transform more rapidly than the coarse-grained steel, and will
have a higher CCR (see fig. 2.11).
Quenching and tempering is usually limited to steels containing
more than about 0.1 % carbon. Figure 2.11 shows that these must be
cooled at rates ranging from 100 to 2000 °C·s–1 if 100 % martensite is to
be produced. There is no difficulty in transforming the surface of a com-
ponent to martensite – we simply quench the red-hot steel into a bath of
cold water or oil. But if the component is at all large, the surface layers
will tend to insulate the bulk of the component from the quenching fluid.
The bulk will cool more slowly than the CCR and will not harden prop-
erly. Worse, a rapid quench can create shrinkage stresses which are quite
capable of cracking brittle, untempered martensite.
These problems are overcome by alloying. The entire TTT curve is
shifted to the right by adding a small percentage of the right alloying ele-
ment to the steel – usually molybdenum (Mo), manganese (Mn), chro-
mium (Cr) or nickel (Ni). Numerous low-alloy steels have been developed
with superior hardenability – the ability to form martensite in thick sec-
tions when quenched. This is one of the reasons for adding the 2–7 % of
alloying elements (together with 0.2–0.6 % C) to steels used for things
like crankshafts, high-tensile bolts, springs, connecting rods, and span-
ners. Alloys with lower alloy contents give martensite when quenched
into oil (a moderately rapid quench); the more heavily alloyed give mart-
ensite even when cooled in air. Having formed martensite, the component
is tempered to give the desired combination of strength and toughness.
Hardenability is so important that a simple test is essential to measure
it. The Jominy end-quench test, though inelegant from a scientific stand-
point, fills this need. A bar 100 mm long and 25.4 mm in diameter is heated
and held in the austenite field. When all the alloying elements have gone
into solution, a jet of water is sprayed onto one end of the bar (Figure 2.12).
The surface cools very rapidly, but sections of the bar behind the
quenched surface cool progressively more slowly (see fig. 2.12). When
the whole bar is cold, the hardness is measured along its length. A steel of
high hardenability will show a uniform, high hardness along the whole
length of the bar. This is because the cooling rate, even at the far end of
the bar, is greater than the CCR; and the whole bar transforms to marten-
site. A steel of medium hardenability gives quite different results. The
CCR is much higher, and is only exceeded in the first few centimetres of
the bar. Once the cooling rate falls below the CCR the steel starts to trans-
form to bainite rather than martensite, and the hardness drops off rapidly.

Figure 2.12 – The Jominy end-quench test for hardenability

After cooling, shallow flats are ground along the specimen length
along which the hardness is measured, to develop a hardness distribution
curve (hardenability curve). Because a distribution of cooling rates is
achieved along the length of the specimen (whose one end is cooled), the
hardness variation along the specimen length becomes a measure of cool-
ing rate. The maximum hardness occurs at the quenched end where nearly
100 % martensite forms. Away from the quenched end, cooling rate pro-
gressively decreases, thus allowing more time for diffusion of carbon to
form softer phases (e.g., pearlite). The steel with a high hardenability will
exhibit high hardness to relatively large distances from the quenched end.
The hardenability of plain carbon steels increases with increasing
carbon content. Likewise, alloying elements such as Ni, Cr, and Mo in
alloy steels improve the hardenability by delaying the diffusion-limited
formation of pearlite and bainite, thus aiding martensite formation.

Details of martensite formation

As Figure 2.13 shows, martensite the lenses are coherent with the
parent lattice. Martensites are always coherent with the parent lattice and
they grow as thin lenses on preferred planes and in preferred directions in
order to caus the least distortion of the lattice. Figure 2.14 shows how the
b.c.c. lattice is produced by atomic movements of the f.c.c. atoms in the
“switch zone”. Martensite lenses form and grow almost instantaneously
t ≈ 550 oC. Two adjacent f.c.c. cells make a distorted b.c.c. cell. If this is
subjected to the “Bain strain” it becomes an undistorted b.c.c. cell. This
atomic “switching” involves the least shuffling of atoms. As it stands the
new lattice is not coherent with the old one. But we can get coherency by
rotating the b.c.c. lattice planes as well. As the lenses grow the lattice
planes distort (see fig. 2.13) and some of the driving force for the
f.c.c. → b.c.c. transformation is removed as strain energy.

Figure 2.13 – The crystallographic relationships of martensite for pure iron

Figure 2.14 – The unit cells of f.c.c. and b.c.c. iron

(two adjacent f.c.c. cells make a distorted b.c.c. cell)

Fewer lenses nucleate and grow, and eventually the transformation
stops. In other words, provided we keep the temperature constant, the dis-
placive transformation is self-stabilizing (Figure 2.14, 2.15). The volume
of martensite produced is a function of temperature only, and does not
depend on time. To get more martensite we must cool the iron down to a
lower temperature (which gives more driving force). Even at this lower
temperature, the displacive transformation will stop when the extra driv-
ing force has been used up in straining the lattice. In fact, to get 100 %
martensite, we have to cool the iron down to ≈ 350 °C (see fig. 2.15).
Note that the temperature at which martensite starts to form is labelled Ms
(martensite start); the temperature at which the martensite transformation
finishes is labelled Mf (martensite finish).

Figure 2.15 – The displacive f.c.c. → b.c.c. transformation in iron

The martensite transformation in steels

As Table 2.4 shows, most characteristics of displacive transforma-
tions are quite different from those of diffusive transformations.
To make martensite in pure iron it has to be cooled very fast: at
about 105 °C·s–1. Metals can only be cooled at such large rates if they are
in the form of thin foils. How, then, can martensite be made in sizeable
pieces of 0.8 % carbon steel? A 0.8 % carbon steel is a “eutectoid” steel:
when it is cooled relatively slowly it transforms by diffusion into pearlite
(the eutectoid mixture of α + Fe3C).
The eutectoid reaction can only start when the steel has been cooled
below 723 °C. The nose of the C-curve occurs at ≈ 525 °C, about 175 °C
lower than the nose temperature of perhaps 700 °С for pure iron. Diffu-
sion is much slower at 525 °C than it is at 700 °C. As a result, a cooling
rate of ≈ 200 °C·s–1 misses the nose of the 1% curve and produces marten-
site. Pure iron martensite has a lattice which is identical to that of ordinary
b.c.c. iron. But the displacive and diffusive transformations produce dif-
ferent large-scale structures: myriad tiny lenses of martensite instead of
large equiaxed grains of b.c.c. iron. Now, fine-grained materials are
harder than coarse-grained ones because grain boundaries get in the way
of dislocations. For this reason pure iron martensite is about twice as hard
as ordinary b.c.c. iron. The grain size argument cannot, however, be ap-
plied to the 0.8 % carbon steel because pearlite not only has a very fine
grain size but also contains a large volume fraction of the hard, iron car-
bide phase. Yet 0.8 % carbon martensite is five times harder than pearlite.
The explanation lies with the 0.8 % carbon. Above 723 °C the carbon dis-
solves in the f.c.c. iron to form a random solid solution. The carbon atoms
are about 40 % smaller in diameter than the iron atoms, and they are able
to squeeze into the space between the iron atoms to form an interstitial
solution. When the steel is quenched, the iron atoms transform dis-
placively to martensite. It all happens so fast that the carbon atoms are
frozen in place and remain in their original positions. Under normal condi-
tions b.c.c. iron can only dissolve 0.035 % carbon.
Table 2.4 – Characteristics of transformations
Displacive (also called diffusionless,
shear, or martensitic)
Atoms move over distances ≤ interatomic Atoms move over distances
spacing. of 1 to 106 interatomic spacings.
Atoms move by making and breaking Atoms move by thermally activated
interatomic bonds and by minor “shuffling”. diffusion from site to site.
Atoms move one after another in precise Atoms hop randomly from site to site
sequence (“military” transformation). (although more hop “forwards” than
“backwards”) (“civilian transformation”).
Speed of transformation ≈ velocity of lattice Speed of transformation depends strongly
vibrations through crystal (essentially on temperature; transformation does not
independent of temperature); transformation occur below 0.3 Тm to 0.4 Tm.
can occur at temperatures as low as 4 K.
Extent of transformation (volume Extent of transformation depends
transformed) depends on temperature only. on time as well as temperature.
Composition cannot change (because atoms have Diffusion allows compositions
no time to diffuse, they stay where they are). of individual phases to change
in alloyed systems.
Always specific crystallographic relationship Sometimes have crystallographic
between martensite and parent lattice relationships between phases

This may seem a strange result – after all, only 68 % of the volume
of the b.c.c. unit cell is taken up by atoms, whereas the figure is 74 % for
f.c.c. Even so, the largest holes in b.c.c. (diameter 0.0722 nm) are smaller
than those in f.c.c. (diameter 0.104 nm).
The martensite is thus grossly oversaturated with carbon and some-
thing must give. The carbon atoms make room for themselves by stretch-
ing the lattice along one of the cube directions to make a body-centred
tetragonal unit cell. Dislocations find it very difficult to move through
such a highly strained structure, and the martensite is very hard as a result.

2.3 Heat treatment

2.3.1 Classification of heat treatment
One way in which the properties of a metal can be changed is by
the use of heat treatment; heat treatment can be defined as the controlled
heating and cooling of metals in the solid state for the purpose of altering
their properties. A heat-treatment cycle consists normally of three parts:
1 Heating the metal to the required temperature for the changes in
structure within the material to occur.
2 Holding at that temperature for a long enough time for the entire
material to reach the required temperature and the structural changes to
occur through the entire material.
3 Cooling, with the rate of cooling being controlled since it affects
the structure and hence properties of the material.
Heat treatment is divided into:
1 the proper heat treatment (is only the thermal effect on metal or
2 thermo mechanical processing (a combination of thermal effects
and plastic deformation) and
3 chemical heat treatment (a combination of thermal and chemical
In turn, proper heat treatment includes the following main types:
- first-order annealing;
- second-order annealing;
- hardening;
- tempering;
- aging.
Thermo-mechanical treatment includes, for example, low and high
temperature TMP.
The concept of chemical-heat treatment processes includes surface
diffusion saturation of steel:
1) by non-metals – nitrogen and carbon (carburizing, nitriding, ni-
triding) and
2) by metals – aluminum, chromium, etc.
2.3.2 First-order annealing
The purpose of the first-order annealing – partly or completely
eliminate the deviations from the equilibrium state, which arose under the
previous treatment.
It is important to point out that the heating temperature for first-
order annealing is not associated with phase transformation temperatures
and only depends upon the recovery or recristallization temperatures pe-
culiar to each metal.
Thus the influence of phase transformations during this annealing
does not take into account.
There are homogenizing annealing and recrystallization annealing.

Diffusion annealing (homogenizing)

Diffusion annealing (homogenizing) is applied to alloy steel ingots
and heavy complex castings for eliminating the chemical inhomogeneity
(the not uniform distribution of chemical elements) within the separate
crystals by diffusion.
Homogenizing is carried out at temperatures from 1100 to l200° C
(the optimum temperature is l150 °C) at which diffusion proceeds quite
easily and, to some extent, equalizes the composition of steels having de-
veloped dendritic segregations. The steel is heated to the maximum tem-
perature of l150 °C at the highest rate that the given furnace can provide
for. Scaling is very intensive at high temperatures and this leads to exces-
sive losses of metal. Holding time, therefore, should be at a minimum.
Holding is followed by cooling with the furnace for 6 to 8 hours to 800–
850 °C and then further cooling in air. After this annealing, the second-
order annealing is used to correct the structure of overheating.

Recrystallization annealing
Recrystallization annealing used for relieve internal stresses, reduce
the hardness, and increase the ductility of strain-hardened metal.
Recovery (when elastic distortions of the crystal lattice are elimi-
nated) and recrystallisation (when new grains are formed and begin to
grow) takes place with increasing annealing temperature during the sof-
tening. In result of this processes the metal regains its high ductility.
Heating is performed at temperature higher than the critical point of
recrystallization (often about 600–720 °С).

Plastic deformation, which distorts the crystal lattice and breaks up
the blocks of initial equiaxed grains to produce a fibrous structure or thin
plates, increases the free energy level of a metal (Figure 2.16).

a b

Figure 2.16 – Effect of recrystallization on the microstructure of a metal

(schematic), × 300 (after S.S. Steinberg): a – after deformation;
b – after recrystallization

Due to plastic deformation a texture is form in steel. Texture is a

regular orientation of the grains in relation to the external forces of defor-
mation. The formation of a texture is accompanied by anisotropy of me-
chanical and physical properties.
Before plastic deformation (in the pre-cold-worked state), the dislo-
cation density will be from 106 to 108 dislocations per cubic centimetre.
When metals are deformed plastically at room temperature the dislocation
density goes up enormously (t0 ≈ 1013 m–2). Each dislocation has a strain
energy of about Gb2/2 per unit length and the total dislocation strain en-
ergy in a cubic metre of deformed metal is about 2 MJ, equivalent to
15 J·mol–1.
An increase in the degree of deformation increases those properties
which characterise the resistance of the metal to plastic deformation (ulti-
mate strength σu, and hardness). The capability for plastic deformation
(relative elongation (δ), and relative reduction of area (ψ)), on the other
hand, is reduced. This phenomenon is called strain or work hardening.
Cold plastic deformation decreases the specific weight and the elec-
trochemical potential but increases the chemical reactivity of a metal. De-
formed metal possesses less corrosion resistance than un-deformed metal.
Cold deformation of iron substantially increases its coercive force and
decreases its magnetic permeability. The residual induction is decreased
with small degrees of deformation and increased with high degrees.
1 Recovery. If cold-worked metal is heated to comparatively low
temperatures (it is about 0.3Tm, where Tm is the melting point on the Kel-
vin temperature scale of the metal concerned), then the internal stresses
resulting from the working start to become relieved. This heating will
slightly lower the strength of the strain-hardened metal but the elastic
limit and ductility will increase, though they will not reach the values pos-
sessed by the initial material (before strain hardening).
During recovery, some of the physical properties (thermal and elec-
trical conductivities) are recovered, although mechanical properties do not
revert to the pre-cold-worked state.
There are no changes in grain structure during this but just some
slight rearrangement of atoms in order that the stresses become relieved.
This process is known as recovery. Copper has a melting point of
1083 °C, or 1356 K. Hence stress relief with copper requires heating to
about 407 K, i.e. 134 °C.
2 Recrystallization. A further rise in temperature increases the mo-
bility of the atoms. If the heating is continued to a temperature of about
0.3 to 0.5Tm there is a very large decrease in hardness, decrease in strength
and increase in elongation. The grain structure of the metal changes. With
recrystallization, crystals begin to grow from nuclei in the most heavily
deformed parts of the metal. Strain-free grains nucleate within the cold-
worked material and slowly consume the entire cold-worked structure.
Complete restoration of mechanical properties to their pre-cold-
worked state occurs during recrystallization. Figure 2.17 shows the effect
on the hardness.

Figure 2.17 – Effect of heat treatment on cold-worked metals (schematic)

Highly cold-worked metals recrystallize faster, and the recrystalli-
zation temperature decreases with increasing degree of cold-working.
Pure metals recrystallize faster than alloys, and alloying raises the recrys-
tallization temperature. The recrystallization temperature Tcr is defined as
the temperature at which the cold-worked structure fully recrystallizes in
1 h. For pure metals Tcr ~ 0.3Tm (Tm is the absolute melting temperature),
but for alloys Tcr ~ 0.7Tm. High-melting-point metals have a high Tcr.
Many deformation processes use hot-working to shape parts; hot-working
is done above Tcr.
The new stress-free recrystallized grains continue to grow if a high
temperature is maintained for a long period. This is because there is a dis-
tribution of grain sizes in the recrystallized material, and the need to de-
crease the grain boundary area provides the driving force for grain growth
(or grain coarsening). Grain coarsening occurs by competitive dissolution
of small grains and growth of larger grains in the distribution through
mass transport via atomic diffusion.
The onset of recrystallization is about 150 °C for aluminium,
200 °C for copper, 450 °C for iron and 620 °C for nickel.
As the temperature is further increased from the recrystallization
temperature so the crystals grow until they have completely replaced the
original distorted cold-worked structure. The hardness, tensile strength
and percentage elongation change little during this phase, the only change
being that the grains grow.

2.3.3 Second-order annealing

Annealing is one of the most important widely-used operations in
the heat treatment of steel.
The purpose of annealing is to obtain softness, improve machinabil-
ity, increase or restore ductility and toughness, relieve internal stresses,
reduce or eliminate structural inhomogeneity, refine grain size, and to
prepare steel for subsequent heat treatment.
There are several types of annealing.

Full annealing
Full annealing consists in heating a hypoeutectoid steel 30–50 °C
above the critical point Ac3, holding it at this temperature, and then slowly
cooling (at a rate of 30–200 °C/h, depending on the composition).
The rate of heating for annealing rolled stock or forgings in heavy
charges is as high as the given furnace can provide. The holding time at
the annealing temperature for these conditions is taken as one-half to one
hour per ton of the charge. Measures should be taken to prevent heavy
scaling and decarburization of the steel in annealing.
Slow cooling is required in annealing to enable the austenite to de-
compose at low degrees of supercooling so as to form a pearlite + ferrite
structure in hypoeutectoid steel, a pearlite structure, in eutectoid steel, and
a pearlite + cementite structure in hypereutectoid steel.
The higher the austenite stability, the slower the cooling must be to
ensure austenite decomposition in the high temperature range. Thus, alloy
steels, in which austenite is very stable, should be cooled much slower (30
to 100 °C/h) after annealing than carbon steel (150 to 200 °C/h).
The form in which the excess ferrite or cementite precipitates in
hypo- and hypereutectoid steels depends on the cooling rate. At high rates,
the excess components precipitate in the form of a network on the grain
boundaries of the prior austenite. Retarded cooling facilitates ferrite pre-
cipitation as separate clusters.
The formation of separate ferrite regions in the structure of a steel is
undesirable since, in subsequent heating for hardening, it is difficult to
equalize the carbon content throughout the austenite and soft spots may
remain after hardening.
Due to phase recrystallization, grains of a steel are also refined after
full annealing.
Complete phase recrystallization of hypereutectoid steel (full anneal-
ing) is required only in cases when hot-working (rolling or forging) was
finished at high temperatures and the steel is, consequently, coarse-grained.
If hot-working is finished at a normal temperature (slightly above Ac3)
incomplete annealing will be sufficient to ensure the required properties.
Hypoeutectoid hot-worked steel (rolled stock, sheet, forgings, etc.)
as well as castings of carbon and alloy steels may undergo full annealing.
Annealing reduces the tensile strength, yield point, and hardness of
rolled (or forged) steel while elongation and reduction of area are increased.
When cast steel is annealed, its strength is also increased because a
fine-grained structure is obtained and the Widmanstatten structure is
Full annealing considerably improves the machinability and form-
ability of steel.

Incomplete annealing
Incomplete annealing consists in heating steel to a temperature
somewhat above the critical point Ac1 holding it at this temperature, and

then slowly cooling it. Incomplete annealing is used to relieve internal
stresses and to improve the machinability of steel. It is associated with
only partial recrystallization; excess ferrite of hypoeutectoid steel or ex-
cess cementite of hypereutectoid steel does not pass over into the solid
solution and is not recrystallized. Incomplete annealing is applied chiefly
to eutectoid and hypereutectoid steels in which heating above point Ac1
causes practically complete recrystallization.
Sometimes under conditions of excessively high heating tempera-
tures and slow cooling, a cementite network appears in hypereutectoid
steels. In this case, incomplete annealing will not ensure a high toughness
and proper grain refinement. Prior to incomplete annealing, it will be nec-
essary to eliminate the cementite network by heating above point Acm fol-
lowed by rapid cooling (hardening or normalization).

Isothermal annealing
In this operation, steel is heated as for ordinary annealing and then
cooled comparatively rapidly (in air or by a blast in a furnace) to a tem-
perature below point Ar1 (usually 50 to 100 °C below in accordance with
the character of the TTT-curve for austenite decomposition). The steel is
held isothermally at this temperature during a certain period of time to
provide for complete austenite decomposition. This is followed by com-
paratively rapid cooling.
The main advantage of isothermal annealing is that it reduces the
time required for heat treatment of the steel. This is especially true for al-
loyed steels which must be cooled very slowly to obtain the required re-
duction in hardness. Isothermal annealing produces good results in treat-
ing relatively small charges of rolled stock or small forgings.
Large forgings or charges of rolled stock have different cooling
conditions on the surface and in the core so that the annealed structure
will not be uniform when passing over to isothermal holding. Therefore,
this type of annealing is not applied in such cases.

Annealing to obtain granular cementite (spheroidizing)

Annealing to obtain granular cementite (spheroidizing) is exten-
sively practized for high-carbon (tool) steels to transform lamellar pearlite
into the granular type. This process is performed by heating the steel
slightly above the critical point Ac1 (730–770 °C) with subsequent holding
at this temperature followed by slow cooling, at a rate of 25 to 30 per
hour, to 600 °C.
The annealing temperature should be only slightly above Ac1. Heating
to higher temperatures will make granular cementite difficult to obtain and
will facilitate the formation of lamellar pearlite. Slow cooling is required to
obtain austenite decomposition at low degrees of supercooling and so that
the carbide particles coagulate to a sufficient extent. This will ensure low
hardness after annealing the steel. Annealing procedures for steels contain
from 0.7 to 0.9 % С (grades У7 and У8) and from 1.0 to 1.3 % С (grades
У10, У12, and У13) following: the range of heating temperatures, at which
granular cementite is formed, is very narrow in steels with near-eutectoid
compositions. The optimum annealing temperature is 750 °C for eutectoid
steels and 770 °C for hypereutectoid steels. Frequently, to heat the charge
more uniformly throughout its height, the temperature is raised to 800 °C at
the beginning and end of the heating operation.
The heating time is determined from the weight (Q) of the charge.
Spheroidized steel has a lower hardness and tensile strength and a
correspondingly higher relative elongation and reduction of area than steel
subject to normal annealing.
Steel having a granular cementite structure is easily machined.

Normalizing of steel
This process consists in heating steel to a temperature from 40 to
50 °C above point Ac3 (Acm), holding at this temperature for a short time,
and subsequent cooling in air. Normalizing is used to eliminate coarse-
grained structures obtained in previous working (rolling, forging, or
stamping, to increase the strength of medium-carbon steels to a certain
extent (in comparison with annealed steel), to improve the machinability
of low-carbon steels, to improve the structure In welds, to reduce internal
stresses, to eliminate the cementite network in hypereutectoid steels, etc.
Normalized carbon steel consists of pearlite and ferrite in hypoeu-
tectoid steels and of pearlite and cementite in hypereutectoid steels.
More rapid cooling (in air), used in normalizing, causes the austen-
ite to decompose at lower temperatures. This increases the dispersity of
the ferrite-cementite mixture (pearlite) and increases the amount of eutec-
toid. Therefore, normalized steel has a higher strength and harder than
annealed steel.
Cooling in air employed in normalizing alloy steels, in which the
austenite is very stable, results in austenite decomposition at high degrees
of supercooling. It is even possible to obtain a hardened steel structure,
i.e., martensite, in this case. After being normalized, such steels will be

very hard and must undergo high-temperature tempering at 550–650 °C to
enable them to be machined.
It is essential to note that two heat-treating operations, normalizing
and high tempering, require less time than annealing. Therefore, these two
operations are often substituted for annealing in the treatment of alloy steels.
Normalizing is frequently applied as a final heat treatment for items
which are to operate at relatively high stresses.
Normalizing is extensively used for improving the properties of
steel castings. Normalized castings have a higher yield point, tensile
strength, and impact strength than annealed castings.

2.3.4 Hardening (quenching)

If after heating to a high enough temperature to cause recrystalliza-
tion, the heated carbon steel is cooled very rapidly, for example by being
dropped into cold water, there is not time for the austenite to loose its ex-
cess carbon atoms and they become trapped in the structure. The result is
a new structure called martensite. The carbon trapped in this structure
considerably distorts the structure. As a consequence, martensite is very
hard and brittle and the steel becomes harder and more brittle. The hard-
ness and strength increases quite significantly with an increase in carbon
content, this being because more carbon is trapped in the structure and so
there is more distortion. This form of heat treatment is called quenching.
A problem with the severe cooling that occurs in quenching is that
cracking can occur. The cracks are a consequence of the stresses occurring
as a result of the distortion produced by the structural changes and also
differential expansion as a result of different parts of a product cooling at
different rates.
Hardening is a heat treating process in which steel is heated to a
temperature above the critical point, held at this temperature, and then
quenched (rapidly cooled) in water, oil, or molten salt baths.
Hypoeutectoid steels are heated from 30 to 50 °C above point Ac3
while hypereutectoid steels are heated above point Ac1. In the first case, fer-
rite + pearlite, and in the second, pearlite + cementite, are transformed into
austenite upon heating. A considerable part of the cementite is retained.
Cooling at a rate higher than the critical value should enable the
austenite to be supercooled to the martensite point.
Hardened steel is in a stressed condition and is very brittle so that it
cannot be employed for practical purposes. After hardening steel must be
tempered to reduce the brittleness, relieve the internal stresses caused by
hardening, and to obtain predetermined mechanical properties.
Hardening followed by tempering is intended for improving the
mechanical properties of steel. The chief aim in hardening and tempering
tool steel is to increase the hardness and wear-resistance, retaining suffi-
cient toughness at the same time. For structural steels, the purpose is to
obtain a combination of high strength, ductility, and toughness.
Selecting the hardening temperature. The hardening temperature of
steel depends upon its chemical composition and, principally, upon its
carbon content.
Hypoeutectoid steels, containing pearlite and excess ferrite, are
hardened by heating to a temperature slightly (30–50 °C) above point Ac3.
This will enable homogeneous austenite to be obtained which will be
transformed into martensite. Upon heating within the interval between the
critical points (Ac1–Ac3), ferrite is partially retained with the austenite. The
austenite is transformed into martensite after subsequent quenching but
the ferrite remains unchanged. The result will be incomplete hardening
since the segregates of free ferrite in the structure of the hardened steel
reduce its hardness. The presence of ferrite, in addition to the brittle
“framework” of martensite crystals, does not allow noticeable toughness
to be obtained.
Ferrite zones lower the mechanical properties of steel not only after
hardening, but after high tempering as well.
Overheating steel above its critical point will cause considerable
austenite grain growth and coarse acicular martensite will be obtained af-
ter quenching. Such martensite has the low impact strength. Overheating
also increases the tendency of a steel to warp and crack during the
quenching operation.
Hypereutectoid steels are heated in hardening to a temperature of
Аc1 + (30 to 50 °C). A certain amount of cementite remains in the structure
of the steel heated to this temperature, in addition to the austenite. There-
fore, this cementite, which was not dissolved in heating, is retained in the
structure of the hardened steel in addition to the martensite. Such a struc-
ture has a higher hardness and wear-resistance than that obtained upon
quenching from a temperature above Acm, i.e., in the region of homogene-
ous austenite. This is explained by the reduction in the amount of retained
austenite and because cementite is harder than martensite. Besides, heat-
ing to temperatures above Acm will inevitably lead to coarsening of the
grain and warping of the part during quenching.
A necessary condition in hardening hypereutectoid steels is the
presence of excess cementite as separate small grains. Excess cementite,
having the form of a network, will increase the brittleness of hardened
steel and promote the formation of hardening cracks.
The heating rate and the heating time depend on the composition of
the steel, its structure, residual stresses, the form and size of the part to be
hardened, etc.
The more carbon and alloying admixtures in a steel, and the more
intricate and larger the part being hardened, the slower it should be heated
to avoid stresses due to temperature differences between the internal and
external layers of the metal, warping, and even cracking.
The practically attainable heating rate depends upon the thermal ca-
pacity of the furnace, the bulk of the charged parts, their arrangement in
the furnace, and other factors.
The first heating method provides a low rate of heating and may be
recommended only for high-alloy steels with the low heat conductivity or
for very large articles.
A high heating rate may be achieved if the articles are charged into
a furnace previously heated to the temperature specified by the heat treat-
ing procedure. This method is applied chiefly for heating small parts in
box furnaces or in continuous furnaces.
The third method is an example of forced heating. Here the articles
are charged into a furnace with a temperature, at the moment of charging,
even higher than the final heating temperature.
In all cases, whenever it is feasible, it is preferable to heat steel rap-
idly to the given temperature since this increases the output of the fur-
naces, reduces fuel consumption, and cuts the time required for heat
treatment. The heating rate for low-carbon steels may be very high with-
out regard for their initial condition. Unannealed high-carbon cast or
forged steels should be heated somewhat slower to prevent additional
stresses, due to rapid heating, which may cause excessive warping and
even cracking.
The heating time for carbon tool steels and medium-alloy structural
steels should be from 25 to 50 % more than for carbon structural steels.
The heating time for high-alloy structural and tool steels should be from
50 to 100 % higher.
The heating rate is usually reduced, not by reducing the furnace
temperature, but by preheating the articles.
The more uniform the heating is, the higher the permissible rate
may be. For this reason, heating in salt baths may be conducted more rap-
idly than heating in box furnaces.
Small items of any structural steel may be heated at the highest rate
the furnace will provide.
When the specified heating temperature is reached, the parts to be
hardened are held at this temperature until they are heated throughout,
until all phase transformations are completed, and until the austenite com-
position becomes equalized throughout the full volume. The higher the
heating temperature, the shorter the holding time may be.

2.3.5 Tempering of steel

Tempering, as mentioned above, leads to the decomposition of
martensite into a ferrite-cementite mixture and strongly affects all proper-
ties of steel.
At low tempering temperatures (up to 200 or 250 °C), the hardness
changes only to a small extent and the true tensile strength and bending
strength are increased. This may be explained by the separation of carbon from
the martensite lattice and the corresponding reduction in its stressed state.
A further increase in the tempering temperature reduces the hard-
ness, true tensile strength, proportional limit, and yield point, while the
relative elongation and reduction of area are increased. The properties of
steel after structural improvement, i.e., hardening followed by high tem-
pering, are always higher than those of annealed steel. This is due to the
difference in structure of the ferrite-cementite mixture.
Impact strength varies in a somewhat different manner. Tempering
at 250 to 400 °C reduces the impact strength of steel (temper brittleness).
All steels are subject to some extent to this tempering embrittlement
which is called temper brittleness of the first order. Some steels, for ex-
ample, carbon steels display only a slight loss in toughness. The impact
strength of alloy steels may be reduced by 50 or 60 %.
The reason for temper brittleness is not yet sufficiently clear. This
reduction in impact strength is most probably associated with the precipi-
tation of carbides from the martensite in tempering. It may also be due to
decomposition of the retained austenite.
It is quite clear from the aforegoing that the temperature range 250
to 400 °C should be avoided in assigning tempering temperatures.
Austempering gives good results for steels with a marked suscepti-
bility to nonrecurring brittleness. This will provide the considerably
higher impact strength than hardening followed by tempering in the range
from 250 to 400 °C.
In assigning tempering procedures it must be noted that longer tem-
pering times are required for heavy parts than for small ones to obtain the
same results.
Effect of tempering on the magnitude of residual (internal) stresses.
The volume increase and the fact that the martensitic transformation does not
take place at the same time, throughout the cross-section of the part being
hardened, inevitably lead to high internal stresses. Purely thermal stresses,
due to nonuniform cooling of surface and internal layers, also appear.
The simultaneous action of thermal compression and phase trans-
formations results in a very complex distribution of residual stresses in a
hardened article. Residual stresses may reach high values. They cause dis-
tortion (warping) and cracking in quenching.
In the most favorable cases, residual compressive stresses appear at
the surface of the article and tensile stresses of a somewhat less magni-
tude-in the core.
A less favorable case may occur, however, in which tensile stresses
appear at the surface. This may lead to quenching cracks.
The higher the cooling rate (especially in the martensitic range), the
heavier the cross-section of the article, and the coarser the actual austenite
grains, the larger the residual stresses will be after hardening.
Tempering is the principal method for relieving residual stresses in
hardened steel. The higher the tempering temperature, the more com-
pletely the internal stresses, caused by quenching, will be relieved.
After tempering a cylindrical specimen at 550 °C, the maximum ax-
ial stresses were reduced from 600 to 80 MPa.
Shear and radial stresses were reduced by approximately the same
amount. Residual stresses are most intensively relieved in the first 15 to
30 minutes of the tempering operation. After a holding time of 1.5 hours,
the stresses reach the minimum value for the given temperature.
The cooling rate after tempering has a great effect on the residual
stresses. The slower an article is cooled, the less the residual stresses will
be. Rapid cooling in water from 600 °С will produce stresses of a value
near to those produced by hardening. Cooling in air after tempering will
result in stresses 7 times less than those obtained by water cooling; cool-
ing in oil will give stresses 2.5 times less than in water.
Thus, articles of complex form, of carbon and certain alloy steels,
should be cooled slowly, after tempering at a high temperature, to prevent
the appearance of new stresses.
Tempering procedures are classified in accordance with the heating
Law-temperature tempering is performed in the range from 150 to
250 °C and its purpose is to reduce internal stresses and to increase the
toughness without any appreciable loss in hardness. This type of temper-
ing is usually used for measuring and cutting tools of carbon and low-
alloy steels, as well as for parts that have been surface hardened, case car-
burized, etc. The following holding times at the tempering temperature
may be recommended for carbon and alloy steel tools:
After low-temperature tempering, the martensite produced by
quenching is transformed into tempered martensite.
Temperature control in low tempering is achieved by watching the
so-called temper colours which appear on the ground surface of steel at
temperatures from 200 to 300 °C.
At temperatures from 220 to 240 °C, the oxide layer which forms
on carbon steel will have a straw-yellow colour (thickness of the oxide
layer 450 Å); in the range 240–260 °C, it will be of an orange colour (ox-
ide layer thickness 500 Å); in the range 260–280 °C, it will be violet in
colour (layer 650 Å), and from 280 to 300 °С the colour will be blue
(layer approx. 700 Å).
The temper colour is the colour of the thin layer of oxides that
forms on the surface of heated steel. The colour depends on the thickness
of the oxide layer.
Medium-temperature tempering at 350 to 450 °C is employed for
coil and laminated springs and provides the highest attainable elastic limit
in conjunction with ample toughness. Steel has a troostite structure after
this tempering procedure.
High-temperature tempering is performed in the range from 500 to
650 °C. It almost completely eliminates internal stresses and provides the
most favorable ratio of strength to toughness for structural steels. The
tempered steel has a sorbite structure after this treatment.
So, in the quenched state carbon steels have such a low ductility as
to be very difficult to use. The process known as tempering can, however,
be used to improve the ductility without loosing all the hardness gained by
the quenching.
Tempering involves heating the steel to a temperature at which
some of the carbon trapped in the martensite structure can diffuse out and
form cementite, so reducing the distortion of the structure. The amount of
carbon that diffuses out depends on the temperature used for the temper-
ing. Thus the mechanical properties depend on the tempering temperature.
Annealing the material results in recrystallization with the result
that large grains are produced with a drop in the number of dislocations.
As a result the material is weaker, softer and more ductile. In results of
quenching, martensite in which carbon atoms are trapped within a dis-
torted structure formation. As a consequence the material is much harder
and more brittle. Tempering allows some of the carbon atoms to diffuse
out and reduce the distortion, the higher the tempering temperature the
greater the reduction. Hence tempering restored some of the ductility but
reduces the strength.
2.3.6 Precipitation hardening
A wide range of alloys used in engineering depend on a treatment
called precipitation hardening for improvements in their hardness and
strength. This type of treatment is widely used with aluminium alloys and
nickel alloys. The process involves heating the alloy to above the recrys-
tallisation temperature, then quenching. The result is a distorted crystal
structure. However, with time, atoms diffuse out of the structure of this
type of alloy to give a fine precipitate. This precipitate lodges at grain
boundaries and in dislocations and as a consequence makes slip more dif-
ficult. The consequence is an increase in hardness and strength. For ex-
ample, the aluminium-copper alloy (2014) might have a tensile strength of
185 MPa and hardness 45 HB in the annealed state and after precipitation
hardening 425 MPa and 105 HB.

2.3.7 Surface hardening

There is often a need for the surface of a piece of steel to be hard, e.g.
to make it wear resistant, without the entire component being made hard
and often too brittle. Several methods are available for surface hardening.
For carbon steels surface hardening can be achieved by just heating
the surface layers to above the recrystallisation temperature and then
quenching to give a martensitic structure for these surface layers. This
selective heating of the surface layers can be heating them with an oxya-
cetylene flame, so called flame hardening, or by placing the steel compo-
nent in a coil carrying a high frequency current and allowing the induced
currents in the surface layers to do the heating, so called induction heat-
ing. Another method that can be used to produce martensite in the surface
layers is to increase the carbon content of the surface layers, this method
being known as case hardening. This can be done by heating the steel
while it is packed in charcoal and barium carbonate, pack carburizing, or
in a furnace in an atmosphere of a carbon-rich gas, gas carburizing, or
alternatively in a bath of liquid sodium cyanide, cyanide carburizing.
These methods might, for example, result in a steel having an inner core
containing 0.2 % carbon and surface layers with 0.9 % carbon.
Other processes that can be used involve changing the surface com-
position by diffusing nitrogen into it to produce hard compounds, nitrides.
This is done by heating the steel in an atmosphere of ammonia gas and
hydrogen. The process is known as nitriding. Carbonitriding involves
heating the steel in an atmosphere containing both carbon and ammonia
and allowing both carbon and nitrogen to diffuse into the surface layers.
The term engineering material is used for a material designed, made
and used for a practical purpose, e.g. carrying a mechanical load or per-
haps providing a path for an electric current. You are surrounded by prod-
ucts involving a wide range of such materials, e.g. the motorcar with its
shaped metal exterior surface, an electric cable with its metal core to carry
the current and plastic sheath for electric insulation, the mobile phone
with its plastic casing, a house with its brick structure and glass windows,
a desk made of wood.
There are many materials in use and it is convenient to classify
them into four main groups, these being determined by their internal struc-
ture and consequential properties. In everyday use, we identify materials
in these groups and distinguish between them by their characteristic prop-
erties rather than the internal structure responsible for the properties. Thus
we might look at a particular material and handle it and come to the con-
clusion that it is easily bent (the term used is ductile), cold to touch (this
means a high thermal conductivity since it readily conducts the heat away
from you hand), not easily broken, etc.

1 Metals
Essentially, metals are based on metallic chemical elements and, in
general, have high electrical conductivities, high thermal conductivities
can be ductile and thus permit products to be made by being bent into
shape, and have a relatively high stiffness and strength. Thus the body-
work of a car is generally made of metal which is bent into the required
shape and has reasonable stiffness and strength so that it retains that
shape. The core of an electric cable is made of a metal to give the high
electrical conductivity required.
Engineering metals are generally alloys, alloys being metallic mate-
rials formed by mixing two or more elements. For example, mild steel is
an alloy of iron and carbon, stainless steel an alloy of iron, chromium,
carbon, manganese and possibly other elements. The reason for adding
elements to the iron is to improve the properties since pure metals are very
weak materials. The carbon improves the strength of iron, chromium in
the stainless steel improves the corrosion resistance.
Metals are also classified as follows:

a) Ferrous alloys. These are iron-based alloys, e.g. steels and cast
b) Non-ferrous alloys. These are not iron-based alloys, e.g. alumin-
ium-based and copper-based alloys.

2 Polymers (plastics)
Essentially, polymers are based on large organic molecules, i.e.
carbon containing molecules. They can be classified as:
a) Thermoplastics soften when heated and become hard again when
the heat is removed. The term implies that the material becomes “plastic”
when heat is applied. Thus thermoplastic materials can be heated and bent
to form required shapes, thermosets cannot. Thermoplastic materials are
generally flexible and relatively soft. Polythene is an example of a ther-
moplastic, being widely used as films or sheets for such items as bags,
“squeezy” bottles, and wire and cable insulation.
b) Thermosets do not soften when heated, but char and decompose.
Thennosets are rigid and hard. Phenol formaldehyde, known as Bakelite,
is a thermoset. It is widely used for electrical plug casings, door knobs and
The term elastomer is used for polymers which by their structure al-
low considerable extensions that are reversible, e.g. rubber bands.
In general, polymers have low electrical conductivity and low
thermal conductivity, hence their use for electrical and thermal insulation.
Compared with metals, they have lower densities, expand more when
there is a change in temperature, are generally more corrosion resistant,
have a lower stiffness, stretch more and are not as hard. When loaded they
tend to creep, i.e. the extension gradually changes with time. Their proper-
ties depend very much on the temperature so that a polymer which may be
tough and flexible at room temperature may be brittle at 0 °C and show
considerable creep at 100 °C.

3 Ceramics
Ceramics are based on inorganic compounds and were originally
just clay-based materials. They tend to be brittle, relatively stiff, stronger
in compression than tension, hard, chemically inert and bad conductors of
electricity and heat. The non-glasses tend to have good heat and wear re-
sistance and high-temperature strength. Ceramics include:
a) Glasses: soda line glasses, borosilicate glasses, pyroceramics.

b) Domestic ceramics: porcelain, vitreous china, earthenware,
stoneware, cement.
Examples of domestic ceramics and glasses abound in the home in
the form of cups, plates and glasses.
c) Engineering ceramics: alumina, carbides, nitrides.
Because of their hardness and abrasion resistance, such ceramics
are widely used as the cutting edges of tools.
d) Natural ceramics: rocks.

4 Composites
Composites are materials composed of two different materials
bonded together. For example, there are composites involving glass fibres
or particles in polymers, ceramic particles in metals (cermets), and steel
rods in concrete (reinforced concrete). Wood is a natural composite con-
sisting of tubes of cellulose in a polymer called lignin. Composites made
with fibres embedded, all aligned in the same direction in some matrix,
will have properties in that direction markedly different from properties in
other directions. Composites can be designed to combine the good proper-
ties of different types of materials while avoiding some of their drawbacks.
If the above classification was solely in terms of properties it would
be deemed rather crude. Within each group there is a large variation in
properties and there are no clear property boundaries between them.
Consider steel and alloys a more detailed.
One way of modifying the properties of a metal is alloying. An al-
loy is a metallic material made by mixing two or more elements. The eve-
ryday metallic objects around you will be made, almost invariably, from
alloys rather then the pure metals themselves. Pure metals do not always
have the appropriate combinations of properties needed; alloys can, how-
ever, be designed to have them.
Making alloys is rather like baking a cake. The basic ingredients of
flour, sugar, fat, eggs and water are mixed together and then cooked. The
result is a cake which has a texture and properties quite different from those
of the individual ingredients. The type of cake produced depends on the rela-
tive amounts of the ingredients and the way it is cooked. In making alloys,
the ingredients are mixed and heated and the resulting alloy can have proper-
ties quite different from those of the ingredients. The properties will depend
on the relative amounts and nature of the ingredients as well as how they are
“baked”. An alloy is a particular mixture of components and so has a particu-
lar chemical composition, e.g. one carbon steel may be 99.0 % iron com-
bined with 1.0 % carbon while another is 99.5 % iron with 0.5 % carbon.
The coins in your pocket are made of alloys. Coins need to be made
of a relatively hard material which does not wear away rapidly, i.e. the
coins have to have a life of many years. Coins made of, say, pure copper
would be very soft; not only would they suffer considerable wear but they
would bend in your pocket. The copper-looking British coins are made of
an alloy of copper with 2.5 % by weight of zinc and 0.5 % of tin, the term
coinage bronze being used for the alloy. The silver-looking British coins
are made of an alloy of copper with 25 % by weight of nickel, the term
cupro-nickel being used for the alloy.
Pure metals tend to be soft with high ductility, low tensile strength
and low yield strength. Because of this they are rarely used in engineering.
Alloying can produce harder materials with higher tensile strength, higher
yield stress and a reduction in ductility. Such materials are more useful in
engineering. There are, however, some circumstances in which the proper-
ties of pure metals are useful. They are where high electrical conductivity
is required (alloying reduces conductivity); where good corrosion resis-
tance is required (alloying can result in less corrosion resistance); and
where very high ductility is required.
We can think of the structure of alloys in terms of the constituent
metals, say A and B, being mixed in the liquid state. Then when the mix-
ture solidifies, there is the possibility that solid alloy will have a crystal
structure in which some of the atoms in the crystal structure of A have
been replaced by atoms of В (substitutional solid solution). Alternatively,
because there are spaces between the atoms of A in its crystal structure,
some atoms of A, if small enough, might lodge in these spaces (interstitial
solid solution). Another possibility is that elements A and В combine to
form a chemical compound. With a compound there will be a particular
structure for that compound with atoms of A and В assuming specific po-
sitions, rather than just popping into any gap. Another possibility is that
when the liquid mixture cools A and В separate out, with В forming its
own crystal structure independent of A. The structure then becomes a mix-
ture of two types of crystals.

3.1 Ferrous alloys

Pure iron is a relatively soft material and is hardly ever used. Alloys
of iron with carbon are, however, very widely used, the term ferrous al-
loys being used for alloys with iron. Pure iron at room temperature exists
as a body-centred cubic structure, this being commonly referred to as fer-
rite. This form continues to exist up to 912 °C. At this temperature the

structure changes to a face-centred cubic one, known as austenite. Iron
atoms have a diameter of 0.256 nm (1 nanometre = 10–9 m); carbon atoms
are much smaller with a diameter of 0.154 nm. The face-centred cubic
structure is a more open structure than the body-centred structure; the
face-centred structure of austenite has voids which can accommodate
spheres up to 0.104 nm in diameter while the body-centred cubic structure
voids between the atoms which are 0.070 nm in diameter. Thus, carbon
atoms can be more easily accommodated within austenite, without severe
distortion of the lattice, than ferrite. Austenite can take up to 2.0 % of car-
bon while ferrite can only take 0.2 %. Thus when iron containing carbon
is cooled from the austenite state to the ferrite state, there is a reduction in
the amount of carbon that can be accommodated within the iron and so
some of the carbon atoms come out of the crystals and form a compound,
another crystal structure, between iron and carbon called cementite. Ce-
mentite is hard and brittle. The result can be a structure consisting of
purely ferrite grains mixed with grains which have a laminated structure
of ferrite and cementite. Such a laminated structure is termed pearlite.
Pure cementite is harder than pearlite, which in turns is harder than pure
ferrite. Thus the structure, and hence the properties, of the iron alloy is
determined by the amount of carbon present.
The percentage of carbon alloyed with iron has a profound effect on
the properties of the alloy. The terms used for the alloys produced with
different percentages of carbon are: wrought – iron 0 to 0.05 % carbon;
steel – 0.05 to 2 % carbon; cast iron – 2 to 4.5 % carbon.
The term carbon steel is used for those steels in which essentially
just iron and carbon are present. The term alloy steel is used where other
elements are included.

3.1.1 General classification

All steels which find application in the Russia at the present time
may be classified on the basis of the following principal characteristics
(Figure 3.1):
I As to their chemical composition, they are classified as:
a) plain-carbon steels;
b) alloy (or special) steels.
Besides carbon and the inevitable minor constituents (Mn, Si, S, P,
N, and others), alloy steels contain certain other specially-added elements,
such as Mn (> 1 %), Si (> 0.5 %), Ni, Cr, Mo, W, V, Ti, and others which
impart the required properties to the steel.

Figure 3.1 – Classification of steels
Carbon steels are grouped according to their carbon content with
the designations being roughly: mild steel – 0.10 to 0.25 % carbon; me-
dium-carbon steel – 0.20 to 0.50 % carbon; high-carbon steel – more than
0.50 % carbon.
Mild steel has a structure consisting predominantly of ferrite, me-
dium-carbon steels tend to have about equal amounts of ferrite and pear-
lite, while high-carbon steels have predominantly pearlite with some free
cementite occurring at high carbon contents.
Increasing the percentage of carbon, within the range considered,
increases the amount of pearlite at the expense of the softer ferrite, and
1 Increases the tensile strength;
2 Increases the hardness;
3 Decreases the percentage elongation;
4 Decreases the impact strength.
Mild steel is a general purpose steel and is used where hardness and
tensile strength are not the most important requirements but ductility is
often required. Typical applications are sections for use as joists in build-
ings, bodywork for cars and ships, screws, nails, wire. Medium-carbon
steel is used for agricultural tools, fasteners, dynamo and motor shafts,
crankshafts, connecting rods, gears. With such steels the lower ductility
puts a limit on the types of processes that can be used. Medium-carbon
steels are capable of being quenched and tempered to develop reasonable
toughness with strength. High-carbon steel is used for withstanding wear,
where hardness is a more necessary requirement than ductility. It is used
for machine tools, saws, hammers, cold chisels, punches, axes, dies, taps,
drills, razors. The main use of high-carbon steel is mainly as a tool steel.
High-carbon steels are usually quenched and tempered at about 250 °C to
develop their high strength with some slight ductility.
II As to the method of manufacture, they are classified as:
a) Ordinary and improved quality steels. These are smelted either
in converters (acid or basic) or in large open-hearth furnaces. In composi-
tion, these are carbon steels with the carbon content from 0.6 to 0.65 %.
The manufacturers of these steels guarantee either their specified me-
chanical properties or chemical composition. They contain slightly more
harmful impurities than the higher quality grades (sulphur up to 0.055 %
and phosphorus up to 0.05–0.085 %).
b) Quality steels, smelted in basic open-hearth furnaces. Here,
stricter conditions are observed in regard to the composition of the charge
and smelting and pouring procedures. Much higher requirements are
specified for these steels than the ordinary quality grades in respect to
their chemical composition, presence of nonmetallic inclusions and other
defects, as well as to their mechanical properties.
c) High-quality steels, smelted in acid or basic open-hearth fur-
naces or in electric furnaces. The composition of these steels is main-
tained within narrow limits; they contain only very small amounts of
harmful impurities (sulphur and phosphorus) and almost no nonmetallic
inclusions, their mechanical properties are strictly specified. These steels
are poured into ingot moulds of the smallest permissible size.
III As to their purpose, all steels may be classified into the follow-
ing groups:
a) Structural steels with the sub-groups:
- Engineering (or constructional) steels designed for structural mem-
bers and steel structures. This sub-group includes ordinary quality carbon
steels possessing good weldability and satisfactory mechanical properties
in the as-received condition, i.e., without additional heat treatment.
- Engineering steels designed for the manufacture of components of
industrial equipment. The main requirements made to these steels are high
strength under static and dynamic loads and good fabrication properties
(pronounced hardenability, minimum strain, satisfactory machinability,
etc.). It is also important that their mechanical and fabrication properties
vary within narrow limits in different melts or batches of the same steel.
This is necessary to provide machine parts of uniform quality in cases
when an established manufacturing process is strictly conformed to. This
is a vital factor for mass production plants.
Carbon, alloy, quality, and high-quality grades find applications as
engineering steels. Ordinary quality carbon steels are also used for low-
stress applications. Most engineering steels undergo heat treatment.
b) Tool steels intended for the manufacture of cutting and measur-
ing tools, as well as dies. Tool steels must have a high hardness and wear
resistance. They must have the high strength required by the cutting edge
of a tool to prevent its destruction by the high stresses due to metal cutting
and to prevent breakage in those parts of the tool that are subject to large
torque and bending stresses, as well as ample toughness to withstand im-
pacts and vibration. Steel designed for the manufacture of hot-forging dies
must have satisfactory mechanical properties at high temperatures.
This sub-group includes carbon steels (over 0.65–0.75 % C) and
quality and high-quality alloy steels with a high carbon content.
c) Steels with special physical properties which suit their applica-
tion. These include magnetic and nonmagnetic steels, steel with a definite
coefficient of expansion, stainless and heat-resistant steels, etc. In their
chemical composition, these special-purpose grades are quality and high-
quality alloy steels.

3.1.2 Designation of steels

The designation system for steels, standardized in Russia, utilizes
numbers and Russian letters.
Ordinary quality carbon steels are identified by the letters Ст. fol-
lowed by the numbers 1, 2, 3, etc. (such as Ст.1, Ст.2, Ст.3, Ст.5, etc.).
Increasing numbers indicate an increase in tensile strength and carbon
content and a decrease in ductility.
Quality structural steels are designated by the numbers 05, 08, 10,
20, 30, 45, etc., which indicate the average carbon content in hundredths
of one percent.
Carbon tool steel is designated by the Russian letter У and a num-
ber indicating the average carbon content in tenths of one per cent. Exam-
ples are У7 (containing 0.7 %С), У8, У9, У10, У11, У12, etc.
The various alloy steels are designated by a combination of Russian
letters and numbers. The letters indicate the presence of a certain alloying
element. Accordingly, А – nitrogen; Б – niobium; В – tungsten; Г – man-
ganese; Д – copper; Е – selenium; К – cobalt; М – molybdenum; Н –
nickel; П – phosphor; Р – boron; С – silicon; Т – titanium; Ф – vana-
dium; Х – chromium; Ц – zirconium; Ю – aluminium.
The number preceding the letter indicates the carbon content in
hundredths of one per cent while the number after the letter indicates the
approximate percentage of the alloying element, if its content exceeds
1.5 %. In all cases, the letter A at the end of the designation indicates that
the steel is of high quality (free of harmful impurities) and that high re-
quirements are specified in respect to metallurgical quality control.
For example, steel 15X is a chromium steel containing ~ 0.15 % С
and ~ 1 % Cr and steel 12X2H4A is a high-quality chromium-nickel steel
containing ~ 0.12 % С, ~ 2% Cr, and ~ 4 % Ni.

3.1.3 Constructional steels

Ordinary and improved carbon structural steels

Ordinary and improved hot rolled carbon steels are divided into the
following three groups:

1) Steels of Group I are supplied with their mechanical properties
guaranteed (БСт0, БСт3, БСт4, БСт5, БСт6, Ст0, Ст1, Ст2, Ст3, Ст4,
Ст5, Ст6 and Ст7).
Normally these steels do not undergo heat treatment and they are
utilised in the as-received condition. The mechanical properties of these
steels are determined by the treatment they are subject to by the user.
2) Steels of Group II are supplied with their chemical composition
guaranteed (Б09, Б16, Б23, БЗЗ, М12кп, М18, M21, M26, etc.).
Ordinary steels are made either by the open-hearth (M) or Bessemer
(Б) processes. In comparison with open-hearth steel, Bessemer steel has a
higher gas saturation, contains more sulphur and phosphorus, and, at the
same carbon content, has lower ductility and higher strength and hardness.
Bessemer steels have a greater susceptibility to ageing and cold-shortness.
They may be recommended only or statically loaded structures.
In respect to the degree of deoxidation, ordinary quality steels may
be classified as killed, semi-killed, and rimmed.
Killed steels are deoxidised in their manufacture and have a low gas
content. They are used in making machine parts subject to dynamic loads.
Rimmed steels are among the cheaper varieties since the losses in their
manufacture are insignificant (a rimmed steel ingot does not have a pipe). At
the same time, they possess essential disadvantages which include: 1) high
gas saturation, 2) lower impact strength, and 3) high tendency to ageing.
3) Steels of Group III are of improved quality. They are produced,
as a rule, in open-hearth furnaces and are degasified and deoxidised. They
are supplied with guaranteed composition and mechanical properties.
Ordinary and improved steels are extensively employed in con-
struction and general engineering.
For example, rolled bars, structural shapes, sheets, and plates of
steels Ст3, М18кп, and БСт4 are widely used for construction compo-
nents and trusses. These same steels are expediently used in engineering
for manufacturing various less important machine parts (bushings, liners,
levers, rods, yokes, etc.), not subject to heat treatment, as well as parts to
be carburized, such as wrist pins of low-power engines, gears, worms, etc.
Steels Ст5, M26, M31, and БСт5 are used for reinforcement and various
machine parts (shafts, pivots, nuts, washers, etc.), not subject to heavy
loads in operation.
Improved steels are widely employed for welded (M16C) and rivet-
ted (МСт3 мост) bridge structures, in ship-building (Ст1С, Ст2C, Ст3С,
Ст4С), for prestressed concrete constructions, etc.

Quality carbon structural steels
Table 3.1 lists the composition of quality carbon hypoeutectoid
steels and their mechanical properties. These steels are used as bar stock,
forgings, sheet, plate and other semifinished products.
Table 3.1 – Composition and mechanical properties of quality carbon steels
Carbon δ, %, ψ, %,
Steel grade σ , MPa σu, MPa a, kg⋅m/sm3
content, % y minimum minimum
08кп ≤0.06 200 330–410 35 60 –
10 0.05–0.10 220 360–450 32 55 –
15 0.07–0.13 240 400–490 29 55 –
20кп 0.17–0.24 250 430–530 27 55 –
20 0.17–0.24 260 440–540 26 55 –
25 0.22–0.29 280 480–580 24 50 10
30 0.27–0.34 300 520–620 22 50 10
35 0.32–0.39 320 560–660 21 45 9
40 0.37–0.44 340 600–720 19 45 9
45 0.42–0.49 360 640–760 17 40 8
50 0.47–0.55 380 680–800 15 40 7
55 0.52–0.60 400 710–830 13 35 –
60 0.57–0.65 420 730–850 12 35 –
65 0.62–0.70 430 760–880 11 35 –
70 0.67–0.75 440 780–900 10 30 –
75 0.72–0.80 900 1100 8 30 –
80 0.77–0.85 950 1100 8 30 –
85 0.82–0.90 1000 1150 7 30 –

Quality carbon steels are smelted in basic open-hearth furnaces.

Low-carbon steel grade 08кп (or 05кп) has a ferrite structure and
high ductility. To acquire high ductility, the steel, in addition to minimum
carbon content, must contain the least possible amount of silicon which
intensively hardens the ferrite. Therefore, these steels are of the rimming
type and contain no nore than 0.03 % Si. To obtain a smooth surface, the
ferrite grain of steel 08кп (or 05кп) should be as fine as possible. Steel
08кп (or 05кп) is designed for a great variety of cold formed parts of
sheet steel (headlights, fenders and bodies of cars, etc.).
Steels 10 and 15, haying comparatively low strength, high ductility,
and good weldability, are applied for machine parts carrying relatively
small loads and in whose manufacture welding, forming, and other similar
processes are required. They are also used as carburising steels. Steels 10
and 15, for example, are used for machine parts that are to undergo carbu-
rising or cyaniding (guide bushings, screws, gears for light loads, spline

shafts, stops, pins, etc.), as well as for welded structures, in the form of
plates or pipe. Steel Grades 20, 25, 30, 40, 45, and 50 have a higher ten-
sile strength but lower fabrication properties, weldability, and ductility (in
comparison to low-carbon steels). Their machinability, however, is higher.
Steels 20 and 25 have found application for small machine parts
which are not subject to high stresses (bolts, nuts, screws, washers, etc.)
while steels 30 and 35 are for machine parts subject to moderate stresses
(cross-members, tie-rods, bolts, nuts, etc.). Steels 20, 25, 30, and 35 are
ordinarily used in the as-normalized condition.
Steels 40 and 45 are used in making machine parts requiring high
strength (connecting rods, gears, pivots, shafts, bushings, tractor crawler
pins, etc.). They are used either as-normalized or after hardening followed
by high-temperature tempering (structural improvement). Steels grade 50,
55, and 60 are used, after hardening and tempering or after normalizing,
for manufacturing such machine parts as gears, tyres, flat springs, snap
rings, various other springs, leaf springs, mill rolls, etc. Coil and leaf
springs of steel 65 and 70 are hardened by heating to 830–840 °C, quench-
ing in oil, and tempering at 350–450 °C.
Due to the low stability of the supercooled austenite, carbon steels
have a high critical cooling rate and a low hardenability. Carbon steels are
usually quenched in water. However, even in this case, through or full
hardening can be achieved only for parts up to 15 or 20 mm in diameter
(or thickness). Therefore carbon steels possess high mechanical properties
after structural improvement only in small cross-sections and can be rec-
ommended only for small parts.
Carbon steels with a higher manganese content have higher harde-
nability. They include steels grade 15Г, 20Г, ЗОГ, 40Г, 50Г, 60Г, 65Г, and
70Г, which contain from 0.8 to 1.0 % Mn, and steels 10Г2, 30Г2, 35Г2,
40Г2, 45Г2, and 50Г2 which contain from 1.4 to 1.8 % Mn. The addition
of manganese to the steel increases the stability of the supercooled austen-
ite, lowers the critical cooling rate, and increases hardenability.
Oil is the quenching medium used in many cases for hardening
steels alloyed with manganese. Such steels must be carefully heated as
they are very sensitive to overheating.
After high-temperature tempering, steels 50Г to 70Г and 30Г2 to
50Г2 should be quenched in water or oil to prevent temper brittleness.
An increase in the manganese content increases the wear resistance,
tensile strength (σu), yield point (σy) and proportional limit (σp) of steel
while the relative elongation (δ) and the reduction of area (ψ) are de-
creased. In addition to their higher strength in comparison to steels 10, 15,
and 20, steels 16Г and 20Г also have better weldability. Steels with in-
creased manganese content have almost the same applications as steels
with a normal content. Steels 60Г and 65Г are used for coil and leaf
springs. Springs of these steels are ordinarily hardened and then tempered
to troostite. The same steels are employed to make machine parts subject
to highly intensive abrasive wear (teeth of horse and tractor rakes, elevator
chain links, etc.).
Frequently, steels Grade 10, 15, 20, 25, 30, 35, 40, 45, 50, 15Г,
50Г, and 50Г2 are applied in the form of cold drawn bars held within
close tolerances. Their strength is increased by work hardening but the
ductility of the steel is reduced.
Maximum strengthening is obtained for small cross-sections (wire,
sheets, etc.). Cold working is the most highly productive and most eco-
nomical method for obtaining machine parts with smooth accurate sur-
faces. In many cases it can eliminate labor-consuming machining opera-
tions (turning axles, bolts, etc., in lathes).

Carbon steels for castings

Carbon steels for castings are designated by a number indicating the
average carbon content, in hundredths of one per cent, followed by the
Russian letter Л.
Grades 15Л, 20Л, 25Л, 30Л, 35Л, 40Л, 45Л, 50Л, and 55Л are
employed for steel castings. The mechanical properties of some of these
steels, in the normalized or annealed condition, are listed in Table 3.2.
Table 3.2 – Mechanical properties of steels for castings
Steel grade σy, MPa σu, MPa δ, %, minimum ψ, %, minimum a, kg⋅m/sm3
15Л 200 400 24 35 5
25Л 240 450 19 30 4
35Л 280 500 15 25 3.5
45Л 320 550 12 20 3.0
55Л 350 600 10 18 2.5

In respect to quality, cast steels are classed in three groups: normal

quality, improved, and special quality. The difference between these cate-
gories is based on the manufacturing method employed, the amount of
harmful impurities they contain, and requirements as to micro- and mac-
Steel castings are heat treated to obtain the required properties, to
correct the cast structure, to eliminate chemical inhomogeneity, and to
relieve foundry stresses.
Experience shows that the best combination of mechanical proper-
ties is obtained by hardening and tempering. Good results are achieved by
homogenizing with subsequent ordinary annealing and bу normalizing
followed by tempering at 600–700 °C. Somewhat lower but quite satisfac-
tory mechanical properties are obtained by an ordinary annealing opera-
tion with phase recrystallization.

Low-alloy constructional steels

Low-alloy steels find the most extensive applications for all types
of steel structures. They possess comparatively high mechanical proper-
ties without the need of additional heat treatment, good weldability, and
are quite resistant to corrosion.
The addition of alloying elements which strengthen the ferrite in-
creases the mechanical properties of a steel. Nickel and copper, in addi-
tion, lower the temperature of plastic-to-brittle transition. This property is
very essential for construction steels as the loss in impact strength must
not exceed 40 per cent at 50 °C below zero.
Besides strengthening construction steel, which is ordinarily used in
the as-rolled condition, the addition of copper, nickel, chromium, and also
phosphorus increases the corrosion resistance.
Table 3.3 lists the compositions and properties of low-alloy steels,
in the bar, structural shape, plate and sheet form, designed for rivetted and
welded structural members, bridges, industrial and civil buildings, railroad
cars, etc.
Table 3.3 – Compositions and mechanical properties of low-alloy constructional steels
Chemical composition, % Mechanical properties
Steel grade σy, σu,
C Si Mn Cr Ni Cu δ, %
15ХГСНД 0.12–0.18 0.4–0.7 0.4–0.7 0.6–0.9 0.3–0.6 0.2–0.4 520 350 18
18Г2С 0.14–0.23 0.6–0.9 1.2–1.6 ≤0.3 ≤0.3 ≤0.3 600 400 14
12ХГ ≤0.14 0.25–0.50 0.4–0.8 0.4–0.7 ≤0.3 ≤0.3 460 330 15
10ХНДП ≤0.12 0.2–0.4 0.3–0.8 0.5–0.8 0.3–0.6 0.3–0.5 – – –

Steel 15ХГСНД is intended for cranes, all-steel gondola cars, oil

tanks and containers, bridges, etc. Steels 18Г2С and 25Г2С are used for
reinforcing bars in reinforced-concrete construction while steel 12ХГ, for
grooved piles.
Low-alloy constructional steels are very sensitive to stress concen-
tration and, therefore, bar stock and structural shapes must have as clean a
surface as possible.
These steels are supplied in the hot-rolled condition. Specified me-
chanical and fabricating properties are provided for by the chemical com-
position, rolling procedure, and cooling after rolling.
Steel 14ХГС is used for manufacturing large-diameter welded gas
mains. It contains 0.11–0.16 % C, 1.0–1.3 (or 0.75–1.05) % Mn, 0.4–
0.7 % Si, 0.5–0.8 % Cr, < 0.3 % Ni, and < 0.3 % Cu.

Alloy structural steels

Alloy structural steels are widely employed in the engineering in-
dustry for parts that are subject to both static and dynamic loads in opera-
tion. They have a more favorable set of mechanical properties than carbon
steels, especially for articles of large cross-section. The alloying element
strengthen the ferrite, which is the chief constituent in the structure of
these steels; increase the hardenability; refine the grain; and increase the
resistance to softening on heating to moderate temperatures. This pro-
motes an increase in the mechanical properties.
The principal alloying elements in structural steels are chromium,
nickel, and manganese. Tungsten, molybdenum, vanadium, and titanium
are not usually employed as independent additions, they are added in con-
junction with chromium, nickel, and manganese.
Most structural steels, containing from 0.12 to 0.5 % C, are of the
pearlitic class. In the equilibrium state, their structure is that of hypoeutec-
toid steel. Certain high-alloy steels, however, containing a total of 5 or
6 % or even more alloying elements; are of the martensitic class.
Structural steels are manufactured in open-hearth or electric fur-
naces. They are supplied either without heat treatment or with one consist-
ing of annealing or normalizing followed by high tempering (at 640–
660 °C). In accordance with their chemical composition and guaranteed
mechanical properties, they are classified as quality and high-quality steels.
The characteristics of the more widely employed structural steels
are described below.
Chromium and chromium-vanadium steels. Chromium is one of the
cheapest of alloying elements. Its addition increases austenite stability and
reduces the critical cooling rate, thus improving the hardenability. Chro-
mium impedes grain growth in heating to some extent and increases the
resistance to softening at elevated temperatures. The addition of chro-
mium increases the tensile strength and yield point of steel after hardening
and high tempering. The toughness (a) is lowered slightly while the duc-
tility (δ, ψ) is practically unaffected.

Chromium steels are extensively applied in many fields of industry.
Those used contain from 0.15 to 0.5 % carbon and from 0.7 to 1.0 %
chromium (15X, 15ХА, 20Х, 20ХА, 30Х, 30ХА, 35Х, 38ХА, 40Х, 45Х,
45ХА, 50Х, and 50XA).
The most widely used of this series are the carburizing steel 15X
(15XA) and steel 40X (38XA) which usually undergoes structural im-
provement treatment.
Steels 15X and 15XA find application in the automobile and tractor
industry for valve tappets, wrist pins, idler studs, etc. In the machine tool
industry, they are used for gears operating at high speeds and medium
loads, claw clutches, bushings, worms, and other parts subject to carburiz-
ing, cyaniding, etc.
Steels 40X and 38XA are usually subject to structural improvement
(high tempering) and are used for shafts and worms operating under me-
dium loads, bushings, clutches and other components of machine tools
and fixtures. In the automobile and tractor industries, they are applied for
steering knuckles, axle shafts, transmission main shafts and many others.
Heat treating procedures and mechanical properties of steels 15X
and 40X are listed in Table 3.4.
Table 3.4 – Mechanical properties of chromium and chromium-vanadium structural steels
Heat treatment Mechanical properties (minimum)
Steel Hardening temperature, °C
Tempering σu, σy ,
grade First Second δ, % ψ, % a, kg⋅m/sm2
temperature MPa MPa
hardening hardening
15X 860 780 200 700 500 10 45 7
15XA 860 780 200 700 500 11 50 8
38XA 860 – 550 950 800 12 50 9
40X 850 – 500 1000 800 9 45 6
15ХФ 860 780 200 750 550 12 50 8
40ХФА 880 – 650 900 750 10 50 9
50ХФА 860 – 475 1300 1100 10. 45 –

The more chromium in a steel the higher its hardenability will be.
Chromium-carbon steels (30Х, 35Х, 40X, and 45X) are hardened by
quenching in oil. This reduces warping to a great extent. Carburized steels
(15X and 20X) are usually quenched in water. After high tempering, steels
40X and 38XA are quenched in oil or water to prevent temper brittleness.
The addition of 0.1 to 0.2 % V to chromium steels (steels 15ХФ,
40ХФ and 50ХФА) reduces the susceptibility to overheating, improves
the mechanical properties, especially the impact strength, and increases
the resistance to softening on heating. Carburized steel 15ХФ is used for
gears, camshafts, wrist pins, etc. After carburizing, the steel retains a suf-
ficiently fine grain, both in the core and the case. Subsequent heat treat-
ment usually consists of single quenching, directly after carburizing, in
water (after precooling).
The heat treatment of steel 40ХФ consists in quenching in oil fol-
lowed by high tempering (see table 3.4). It is used for crankshafts, bush-
ings, bolts, cross-members, spindles, shafts, etc. Steel 50ХФА is used for
critical springs. Hardening is followed by a medium tempering.
Antifriction bearing steels are a special group of chromium struc-
tural steels. The compositions of these steels are listed in Table 3.5.
Table 3.5 – Chemical compositions and applications of chromium steels
for antifriction bearings
Chemical composition1, %
Steel grade Applications
C Cr
Balls up to 13.5 mm and rollers up
ШХ6 1.05–1.15 0.4–0.7
to 10 mm in diameter
Balls from 13.5 to 22.5 mm and rollers
ШХ9 1.0–1.1 0.9–1.2
from 10 to 15 mm in diameter
Balls over 22.5 mm and rollers
ШХ15 0.95–1.05 1.3–1.65
from 15 to 30 mm in diameter
0.17–0.35 % Si, 0.2–0.4 % Mn, ≤ 0.02 % S, and ≤ 0.027 % P.

In the operation of the antifriction bearings, the material of the rings,

balls and rollers is subject to abrasive and chemical wear, as well as to high
specific loads of an alternating type which may lead to fatigue failure.
The high carbon content in these steels (see table 3.5) enables an
adequately high strength to be obtained after hardening and tempering, as
well as a high fatigue limit, high contact fatigue limit, and ample abrasion
Steel alloyed by chromium has excellent hardenability, and a high
and uniform hardness can be attained. The initial structure of these steels
consists of fine-grained pearlite. There should be no nonmetallic inclu-
sions, porosity, gas holes and other metallurgical defects. The steel must
be free of carbide segregations which may cause breaking out and surface
cracks that will lead to failure of the bearing.
The hardening procedures for the bearing components depend upon
their form and size. Small balls are oil-quenched; large ones are water-
quenched (3.5–5 % aqueous soda solution). Hardening is followed by low
tempering at 140–160 °C. After heat treatment, the steel will have a fine
crystalline martensite structure with dispersed carbides. The hardness will
be HRC 61–65.
Good results are obtained from martempering, comprising heating
to 830–850 °C and quenching in a bath at the temperature 140–190 °C.
The hardness after this procedure should be HRC 61–63.
The dimensions of precision bearings are stabilized by a subzero
treatment (+10 to –30 °C) with subsequent tempering or ageing at 125–
130 °C for 5 to 25 hours.
Silicon and chromium-silicon steels. Silicon steels are applied to a
great extent for making leaf and coil springs which require high elastic
and fatigue limits. Steels 55C2 and 60C2 are most often specified for leaf
springs (Table 3.6).
Table 3.6 – Chemical composition of steels for leaf and coil springs
Chemical composition, %
Steel grade
C Mn Si Crmax Nimax Smax Pmax
55С2 0.50–0.60 0.6–0.9 1.5–2.0 0.3 0.5 0.05 0.05
60С2А 0.56–0.64 0.6–0.9 1.6–2.0 0.3 0.5 0.04 0.04

Spring leaves of steels 55C2 and 60C2 are quenched in oil. This is
followed by tempering at 460 °C (420–520 °C). After this treatment, the
steel will acquire the mechanical properties listed in Table 3.7.
Table 3.7 – Mechanical properties of steels for leaf and coil springs after heat treatment
Mechanical properties (minimum)
Steel grade
σu, MPa σy, MPa δ, % ψ, %
55С2 1300 1200 25 30
60С2А 1600 1400 20 20

An excellent heat-treating procedure for coil and leaf springs of

steel 55C2 is austempering which consists in heating to 850–880 °C and
then holding for 5 to 7 minutes in the quenching medium at 340–380 °C.
The hardness will be HRC 38–41. Austempering increases the elastic
properties, ductility, and fatigue limit of steel 55C2.
Provision must be made for effective measures to prevent decar-
burization in hardening springs since this substantially reduces the fatigue
limit. Leaf springs are frequently shot-peened (surface cold working) to
increase their endurance limit.
The chromium-silicon steels 33XC and 40XC (1.2 to 1.6 % Si and
1.3 to 1.6 % Cr) find numerous applications in industry. After hardening
(heating temperature 900 °C and oil quenching) and a high tempering
(33XC at 630 °C and 40XC at 540 °C), these steels acquire the mechani-
cal properties listed in Table 3.8.

Table 3.8 – Mechanical properties of chromium-silicon
Mechanical properties (minimum)
Steel grade
σu, MPa σy, MPa δ, % ψ, % a, kg⋅m/sm2
ЗЗХС 950 800 13 50 8
40XC 1250 1100 12 40 5

Tempered at 250–270 °C, steel 33XC will acquire a tensile strength

of 1500 MPa as compared to 1700 MPa for steel 40XC after the same
treatment. Chromium-silicon steels possess high strength, moderate
toughness, and low hardenability. They are susceptible to overheating and
temper brittleness and are used for small machine parts subject to high
stresses and wear. Steel 33XC is also used to make thin-walled tubes.
Molybdenum and chromium-molybdenum steels. The properties of
steel are improved by the addition of molybdenum. It increases the harde-
nability, reduces the tendency for overheating, excludes temper brittle-
ness, and eliminates the danger of graphitization after long service at ele-
vated temperatures of such articles as boilers, fire-box components, etc.
Molybdenum steel is used for heat-resistant applications. Steel containing
0.15 % С and 0.5 % Mo, having a tensile strength at room temperature
near that of carbon steel σu ≈ 400–450 MPa), has a strength of 350–
380 MPa at 550 °C. Upon prolonged load application (1000 h), it can
carry up to 190 MPa. Under the same conditions the tensile strength of
carbon steel will be about 200 MPa (at 550 °C) and 80 or 90 MPa (for
1000 h). The low-carbon steels 16M, 12MX, and 15XM (0.8 to 1.1 % Cr
and 0.4 to 0.55 % Mo), as well as 12Х1МФ (0.2–0.35 % Mo) have good
welding properties and find applications in the manufacture of boiler
drums, surface-heated boiler tubes, superheaters and steam piping. Me-
dium-carbon steel 30XM (0.8-1.1 % Cr and 0.15–0.25 % Mo) is used for
fastening parts, high-pressure fittings, and for critical components of tur-
bines, such as rotors, shafts, and discs. Steel 25X2MA (1.5–1.8 % Cr and
0.25–0.35 % Mo) finds use as fastening parts for high-pressure boilers and
steam lines.
Chromium-molybdenum steel, type 30XMA, has exceptionally
high mechanical properties and weldability. It is applied for high-stressed
weldments. Steel 30XM, a modification, is also used for axles, rotors,
gears, perforators, etc. This steel will through-harden in cross-sections up
to 50 mm. The high cost of molybdenum restricts the application of these
steels at the present time.
Chromium-manganese-titanium steels. Manganese, added to the
chromium steel, increases the austenite stability, reduces the critical cool-
ing rate and improves the hardenability. At the same time, however, it in-
creases the sensitivity to overheating. If the hot working of chromium-
manganese steel (inherently fine-grained type No. 6–8) is carefully per-
formed and if titanium is added, which forms stable carbides and reduces
the susceptibility to overheating, together with the chromium, a set of very
high mechanical properties may be obtained.
The most successful of this series are the carburized grades 18ХГТ
and 30ХГТ.
Steel 18ХГТ has the following composition: 0.16–0.24 % C, 0.8–
1.1 % Mn, 1.0–1.3 % Cr, and 0.08–0.15 % Ti.
The carburized steel 18ХГТ is a high-quality substitute for expen-
sive chromium-nickel steels and finds wide application in the automobile
and tractor industries for such components as shafts, gears, large-size anti-
friction bearings, and others.
Steel 18ХГТ is carburized at 920±10 °C and is hardened at 870 °C
by quenching in oil which is followed by tempering at 200 °C. If carburiz-
ing is done in through-type retort furnaces, the work is quenched direct
from the carburizing furnace after precooling to 840±10 °C.
Machine parts made of steel 18ХГТ are deformed to a minimum
extent by heat treatment. Due to the small amount of austenite retained
after carburizing, quenching, and tempering, the case will have a hardness
of HRC 58–61 while the hardness of the core will be HRC 28–36. This
steel machines easily. A disadvantage is its relatively low hardenability
(50–40 mm). At the present time, steel 30ХГТ, having a higher carbon
content, is replacing other steels. This newer grade can be recommended
both for carburizing and for structural improvement (high tempering).
Steel 30ХГТ has better hardenability than steel 18ХГТ and a stronger
core (hardness HRC 32–46). This enables the case depth to be reduced
without danger of its failure due to plastic deformation of the core. This
reduction in case depth substantially decreases the time required for car-
burization. Case-hardened steer 30ХГТ has a higher fatigue limit than
steel 18ХГТ.
In addition to steel 18ХГТ, steel 18ХГМ is also used for gear
manufacture. The latter contains from 0.2 to 0.3 % Mo.
Chromium-silicon-manganese steel (cromansil type). Steels of this
group have found very extensive application in Soviet engineering indus-
tries and have replaced expensive chromium-molybdenum steels to a great
extent. This steel contains from 0.17 to 0.39 % С (average content), 0.9 to
1.2 % Si, 0.8 to 1.1 % Mn, and 0.8 to 1.0 % Cr. Table 3.9 lists the heat
treatment procedure and mechanical properties of cromansil-type steel.
Table 3.9 – Heat treatment and mechanical properties of chromium-silicon-manganese
structural steels
Heat treatment Mechanical properties (minimum)
Hardening Tempering σu, σy ,
grade δ, % ψ, % a, kg⋅m/sm2
temperature, °C temperature, °C MPa MPa
20ХГС 880 500 80 64 12 45 7
30ХГС 880 550 110 95 10 45 5
Austempering: heat to 880°C, quench
in mixture of potassium (KNOS) and
30ХГСА sodium (NaNOg) nitrates or molten 165 140 10 40 5
sodium (NaOH) and potassium
(KOH) hydroxides at 280–310 °C

Alloying with a combination of chromium and manganese provides

an austenite stability sufficient for oil-quenching. The hardenability of
cromansil-type steel is not very high but this is not an essential factor since
the steel is applied chiefly for relatively thin weld-ments. It must be noted
that this steel should be quenched in oil or water after high tempering (at
500–550 °C) to prevent the development of recurring temper brittleness.
The application of austempering enables high mechanical properties
to be obtained (Table 3.9). It also reduces the notch sensitivity and in-
creases the resistance to combined stresses. It is for this reason that bright
austempering is most frequently applied. The work is heated in molten
chlorides and quenched in molten alkalis (NaOH–KOH).
Steel of this type with a low carbon content (20ХГСА) has good
weldability and high ductility; it can be easily formed and bent. This steel
is used in the form of plates and pipe for heat-treated weld-ments.
Steel 30ХГСА with a higher carbon content is available as plates,
pipe, bars, shapes, and forgings. This steel has satisfactory weldability and
An advantageous property of cromansil-type steels is their fair ma-
chinability notwithstanding the high hardness.
Steel 30ХГСНА is finding extensive application at the present time.
It contains 0.28–0.33 % C, 0.9–1.2 % Cr, 1.0–1.3 % Mn, 1.5–1.8 % Ni,
and 0.9–1.2 % Si.
This steel acquires the following mechanical properties after hard-
ening and low tempering: σu = 1700 MPa, σy = 1450 MPa, δ = 11 %,
ψ =45 %, and a =7 kg⋅m/cm2. These properties considerably exceed those
of steel 30ХГСА. The addition of nickel increases the toughness and re-
sistance to brittle failure because the nickel dissolves in the solid solution
and increases the dispersity of the obtained structure.

Steels with boron. The amount of boron which may be added to
structural steels ranges from 0.001 to 0.005 %.
Boron increases the density of the ingot and improves the ability to
undergo hot plastic deformation.
Boron increases the hardenability of structural steel. The strength
(σu, σy) is also increased while ample ductility and toughness (δ, ψ, a) are
The favorable effect of small boron additions on the properties en-
ables steels with high mechanical properties to be developed on the basis
of low-alloy chromium-manganese and chromium grades.
The boron steels 20ХГР (0.17–0.24 % C, 0.8–1.15 % Cr, 0.7–
1.0 % Mn, and 0.002–0.005 % B) and 35XPA (0.3–0.38 % C, 0.8–
1.1 % Cr and 0.002–0.005 % B) have been recommended in the Russia to
replace the molybdenum steels 20XHM and 35XMA which are employed
in the automobile industry for various types of gears.
After carburizing, quenching from 860–880 °C in oil, and temper-
ing at 200 °C, steel 20ХГР will acquire the following mechanical proper-
ties: σu = 1000 MPa; σy = 800 MPa; δ = 9 %; ψ = 50 %; a = 8 kg⋅m/sm2.
Steel 40XP is used after hardening at 860 °C followed by tempering at
540 °C. After structural improvement, this steel develops the following prop-
erties: σu = 1000 MPa; σy = 800 MPa; δ = 12 %; ψ = 50 %; a = 9 kg⋅m/sm2.
The carburized steels 20ХГНР and 15X2H2TPA have found appli-
cations in the manufacture of large heavily-loaded machine parts.
It is necessary to note that boron increases the susceptibility of a
steel to overheating, as well as the deformation (warping) in hardening.
As a rule, carburized boron steels must be inherently fine-grained.
The use of coarse-grained steels often leads to the forming of hardening
Another feature to be noted is that boron is effective only when
added to fully killed steels.
Nickel steels. Nickel, when added to a steel, substantially increases
the austenite stability and thereby reduces the critical cooling rate and in-
creases the hardenability. Nickel increases the strength and hardness char-
acteristics without any appreciable reduction in ductility and toughness.
The improvement in mechanical properties is based on the strengthening
of the ferrite and the formation of more dispersed structures after harden-
ing. Due to the high cost of nickel and the more effective application of
chromium-nickel steels, straight nickel steels find limited use.
Chromium-nickel steels have a high hardenability (through-
hardening can be achieved in parts over 75–100 mm in diameter or thick-
ness) and very high mechanical properties as well. They are extensively
employed for large machine parts of complex form operating under vibrat-
ing and dynamic loads.
The most widely used in the Soviet Union are the carburized steels
12XH3A and 12X2H4A (Table 3.10) which are found in such heavily
loaded components as gears, shafts, ball-joint pins, spindles of precision
machine tools, axles, rollers, etc.
Table 3.10 – Heat treatment and mechanical properties of chromium nickel steels
Mechanical properties
Heat treatment
Steel grade Hardening Tempering
σu, σy, δ, ψ, a,
Temperature, °C Quenching Tempera- Cooling
MPa MPa % % kg⋅m/sm2
First Second medium ture, °C medium
12XH3A 860 780 Oil 180 1000 850 12 55 10
or oil
12X2H4A 860 780 Oil 180 Air 1200 1000 10 50 9
37XH3A 820 – Oil 530 1100 900 10 50 6
or oil
18X2H4BA 950 850 Air 160 Air 1200 1050 12 55 11

Due to the high stability of the austenite after carburizing followed

by cooling the carburising boxes in air, steels 12XH3A and 12X2H4A
acquire a partially hardened structure that makes subsequent machining
quite difficult. To improve machinability after carburizing, it is advisable
to apply a high tempering at 650 °C with several hours of holding time.
The carburized case of chromium-nickel steels retains a large
amount of austenite which lowers the hardness even below HRC 55–56.
The retained austenite may be decomposed by a high tempering or by sub-
zero treatment.
Instead of double hardening (see table 3.10), a single hardening op-
eration, at 780–850 °C, is frequently applied after carburising. This also
provides quite satisfactory mechanical properties.
A typical chromium-nickel steel, usually subject to structural im-
provement, is grade 37XH3A. It contains 0.37 % C, 1.4 % Cr, and
3.25 % Ni. The high austenite stability of this steel imparts high hardena-
bility so that this steel may be recommended for making large components
(heavy forgings, shafts, piston rods, high-strength crankshafts, etc.).
Chromium-nickel steels are most susceptible to temper brittleness after
high tempering (500–550 °C) followed by slow cooling. Therefore, in
such cases, the steel is cooled (quenched) in oil or water (for large arti-
cles). Steel 37XH3A is used, not only in the structurally-improved condi-
tion, but after hardening followed by low tempering at 220 °C. As a result
of the latter treatment, the steel acquires a high tensile strength in conjunc-
tion with ample ductility and toughness.
When tungsten and molybdenum are added to chromium-nickel
structural steel, they increase the tensile strength and yield point without
affecting the toughness. They also increase the resistance to softening on
heating and reduce the susceptibility to temper brittleness. Steels of this
group possess practically full hardenability in the largest cross-sections.
The most extensively used is the chromium-nickel-tungsten steel,
grade 18X2H4BA (18X2H4MA). This steel is used either after carburiz-
ing or after hardening and tempering. Steel 18X2H4BA contains, on an
average, 0.18 % C, 1.5 % Cr, 4.25 % Ni, and 1 % W. It is a typical steel of
the martensitic class.
Steels 18X2H4BA and 18X2H4MA are characterised by the excep-
tionally high stability of the austenite in the pearlite range. This provides
excellent hardenability, excludes annealing, and enables these steels to be
air-hardened (see table 3.10).
Instead of annealing, these steels are tempered at a high temperature
(640 °C) to reduce their hardness (to HB 269–217). Large forgings un-
dergo normalizing (cooling in air from 950 °C) and tempering at 640 °C
to refine the structure and to reduce the hardness.
Grade 18X2H4BA is chiefly used as the carburized steel. It is nec-
essary to point out that in carburizing this steel, the use of an excessively
active carburizer or raising the hardening temperature will cause a large
amount (up to 60–80 %) of austenite to be retained in the case, thereby
sharply reducing its hardness. Retained austenite in the case may be re-
duced by hardening after high tempering (640–650 °C). A sub-zero treat-
ment after quenching will achieve the same results. Steel 18X2H4BA is
also applied extensively after quenching followed by low tempering and
after structural improvement.
This steel acquires a most expedient combination of properties after
hardening in air followed by low-tempering (see table 3.10).
The fatigue limit of steel 18X2H4BA is frequently elevated by ni-
triding (for example, crankshafts of powerful engines).
Steel 18X2H4BA is applied for the most critical parts of machines,
such as various gears, pins, axles, crankshafts, and others which require
high dynamic and fatigue strengths.
This steel is distinguished for its high strength, and its ample ductility
and toughness. It is one of the best structural steels. Its capabilities for air-
hardening and through-hardening in large cross-sections render it indispen-
sable in the manufacture of massive heavily-loaded complex machine parts.
Industries also employ steels 25XHBA, 45ХНМФА (0.45 % C,
1 % Cr, 1.5 % Ni, 0.15 % V and 0.25 % Mo), and 40XHMA (0.4 % C,
0.75 % Cr, 1.5 % Ni, and 0.20 % Mo).
Steel 45ХНМФА provides good service under high torsional
stresses. It is used for making large shafts and heavily loaded springs.
The addition of 0.2 % Mo to steel 40XHMA decreases its suscepti-
bility to temper brittleness. This steel has deep hardness penetration after
quenching and is used as a substitute for steel 37XH3A in the manufacture
of large machine parts for which increased strength (σu ≥ 1000 MPa), in
conjunction with ample ductility and toughness (ψ ≥ 55 % and
a = 10 kg⋅m/cm2), is specified. The final heat treatment for this grade of
steel consists in hardening (heating temperature 850 °C, quenching in oil)
and tempering at 570–640 °C, followed by oil-quenching.

3.1.4 Tool Steels

Carbon and alloy steels designed for the manufacture of cutting and
measuring tools and dies are known as tool steels.
High-speed steels comprise a special group in the general class of
tool steels.
Tool steels are smelted in open-hearth and electric furnaces and be-
long to high-quality classes. In their as-received condition, carbon and
alloy tool steels are usually annealed to a granular pearlite (cementite) and
they thereby possess low hardness, good machinability and ductility, and
decreased susceptibility to overheating.

Steels for cutting tools

This group includes high-carbon eutectoid and hypereutectoid
After hardening and low tempering, the cutting edges of tools must
have a high hardness (HRC 60–65), considerably exceeding that of the
material to be machined. They must have a wear resistance sufficient to
retain the size and form of the cutting edges in metal cutting and ample
strength combined with a certain degree of toughness to prevent tool
breakage in operation. Table 3.11 indicates the composition of the steels
for cutting tools while Table 3.12 indicates heat treating procedures and
applications of these steels.
Carbon tool steels are characterised by the low stability of the su-
percooled austenite. Therefore, they have a high critical cooling rate and a
low hardenability. Through hardening can be achieved only in parts up to
12 or 15 mm in thickness or diameter. Consequently, these steels (У7А,
У8А, У10А, У11А, and У12А) may be recommended for small-size
tools, which are hardened in oil or molten salts, and for comparatively
large tools (15 to 30 mm in diameter) in which the cutting section is only
on the surface layer (files, core drills, short reamers, etc.).
Table 3.11 – Chemical compositions of the most extensively used steels for cutting tools
Chemical composition, %
Steels grade
С Mn Si Cr W
Carbon steels
У8А 0.75–0.84 0.15–0.30 0.15–0.30 0.2 –
У10 (У10А) 0.95–1.04 0.15–0.30 0.15–0.30 0.2 –
У12 (У12А) 1.15–1.24 0.15–0.30 0.15–0.30 0.2 –
Alloy steels
X 0.95–1.1 ≤ 0.4 ≤ 0.35 1.3–1.6 –
9XC 0.85–0.95 0.3–0.6 1.2–1.6 0.95–1.25 –
ХВГ 0.9–1.05 0.8–1.1 0.15–0.35 0.9–1.2 1.2–1.6

When larger tools are hardened (of a diameter over 30 mm) the
layer with a high hardness is so thin, even upon quenching in water, that
the tools are not fit for cutting purposes.
Great disadvantages of carbon steels are their narrow range of hard-
ening temperatures and the necessity for rapid quenching in water or
aqueous alkali solutions (salts).
This leads to considerable deformation and warping and even to the
formation of cracks.
Therefore, tools of complex form with sharp changes in section and
with a large length-to-diameter ratio should not be made of carbon steels.
Warping and crack forming may be reduced somewhat by quenching in
water only to 200–250 °C with subsequent retarded cooling in oil. Stepped
quenching (martempering) is advisable for small-size tools (see ta-
ble 3.12). Good results may be obtained in applying induction hardening
to certain types of tools.
Advantages of carbon steels are their cheapness, low hardness
(HB 170–180), good machinability and formability in the annealed state,
and the fact that they retain a tough unhardened core due to their low
hardenability. This last factor improves their resistance to breakage under
vibration and impacts.
Carbon steels are only applicable for tools operating at low cutting
speeds since their hardness is substantially reduced at temperatures above
190–200 °C.

Table 3.12 – Heat treating procedures for cutting-tool steels
Steel Annealing Hardening Hardness after Tempering Hardness after
grade temperature, °C Temperature, °C Quenching medium quenching HRC temperature, °C tempering HRC
Aqueous solution Lathe and planer
770–790 of alkalis (salts) 63–65 single-point tools,
750–770 or water followed by oil 150–160 61–62 drills, files,
780–800 Molten salt (150–160 °C) 61–63 round threading
790–810 Oil 61–62 dies, taps, etc.
835–850 Oil 62–65 Drills, reamers,
X 770–790 160–180 61–63
840–850 Molten salt (150–160 °C) 61–64 taps, etc.
850–870 Oil 62–64 Milling cutters,
reamers, core
9XC 780–800 170–190 61–62
860–875 Molten salt (150–200 °C) 61–63 drills, broaches,
threading dies, etc.
820–840 Oil 63–65 Milling cutters,
ХВГ 770–790 170–185 62–63 reamers, taps,
830–850 Molten salt (150–160 °C) 62–64
broaches, etc.
Note. Holding time at 150–160 °C, oil, for tools up to 5–12 mm in diameter – 1–2 minutes.
Alloy steels for cutting tools are of the eutectoid and hypereutectoid
type. After hardening, they acquire high hardness and wear resistance but
not particularly high resistance to self-tempering. They are applicable only
in cases when the cutting edges are not heated above 200–225 °C.
Compared to carbon steels, the alloyed grades have somewhat more
wear resistance and less susceptibility to overheating. They harden deeper
and their tendency to deform and warp is less since they are quenched in
oil and retain a larger amount of austenite.
Alloy steels may be recommended for tools having a tendency to
warp in hardening (long thin taps, reamers, etc.) and for large-size tools
(for example, taps and reamers) over 35 or 40 mm in diameter. Such large
tools made of carbon steels would have a hardened layer too thin for prac-
tical purposes.
The most extensively used of these steels for cutting tools are
grades 9XC and ХВГ. Steel 9XC has an increased hardenability. Either
stepped or isothermal quenching (martempering or austempering) may be
applied, and this steel retains its high hardness at temperatures up to 225–
250 °C. Thus its cutting properties are 10 to 15 per cent higher than those
of carbon steel. No appreciable deformation is observed after martemper-
ing or austempering tools made of steel 9XC.
Disadvantages of this grade are the tendency to decarburize in heat-
ing and the high hardness (HB 217–247) in the annealed state which im-
pairs machinability.
Due to the presence of tungsten carbides in steel ХВГ, it has higher
hardness and wear resistance than steels X and 9XC but is more expensive.
High-speed steels are distinguished for their high red-hardness, i.e.,
their capability to retain their structure (martensite), hardness, and wear
resistance at the high temperatures generated on the cutting edges when
machining at high cutting speeds. High-speed steels are designed for the
manufacture of high-production tools with high wear resistance which
must retain their cutting properties at temperatures up to 600–620 °C. The
composition of some high-speed steels is given in Table 3.13.
Table 3.13 – Compositions of high-speed steels
Chemical composition, %
Steel grade
С Cr W V
P18 0.7–0.8 3.8–4.4 17.5–19.0 1.0–1.4
P9 0.85–0.95 3.8–4.4 8.5–10.0 2.0–2.6

In regard to its structure in the annealed condition, high-speed steel

is of the cementitic type. After normalizing, it acquires a martensitic struc-
ture and may be classed as a martensitic steel.
The phase composition of the P18 steel consists of a solution, com-
plex carbides (Fe2W2C, Cr23C6, and VC), and carbides of the cementite
type. About 50 to 70 % of the chromium in the steel is in a solution. The
rest of the chromium and almost all of the tungsten and vanadium are in
the carbides. The structure of cast high-speed steel includes a complex
eutectic, resembling ledeburite, located along the grain boundaries. Hot
working breaks up the eutectic and refines the carbides.
In worked and annealed high-speed steel, both the eutectic and sec-
ondary carbides are uniformly distributed in a matrix of sorbitic pearlite
(Figure 3.2, b).

Figure 3.2 – Microstructure of high-speed steel P18, × 250:

a – eutectic (ledeburite) in cast high-speed steel; b – annealed steel;
c – hardened steel; d – hardened and tempered steel

If the steel has not been hot worked to a sufficient degree, the pri-
mary (eutectic) and secondary carbides will not be uniformly distributed.
Neither will they be uniform in size and in the form in which they precipi-
tate from the matrix (network, bands or stringers along the direction of
working, etc.).
The mechanical properties of the steel are lowered to a great extent by
this lack of carbide uniformity. It also affects the red-hardness and makes it
difficult to obtain uniform high hardness after quenching. Large blanks usu-
ally undergo additional working to eliminate carbide nonuniformity.
When high-speed steel is heated above point A1 (820–860 °C), pear-
lite is transformed into austenite. Further heating above the A1 range will
dissolve the secondary carbides in the austenite to increase its degree of
alloyage. Due to the presence of a considerable amount of excess (pri-
mary) carbides, not dissolved in the austenite, high-speed steel retains its
fine grained structure even when heated to high temperatures (to 1200–
1240 °C for steel P18 and to 1190–1220 °C for steel P9). Heating to
higher temperatures will cause appreciable grain growth. The transforma-
tion of the austenite in cooling depends upon the amount of alloying ele-
ments dissolved in the austenite in heating and upon the cooling rate. The
higher the heating temperature, i.e., the larger the amount of alloying ele-
ments in the austenite, the higher its stability will be in the pearlite range,
as well as in the range of intermediate (bainite) transformation.
High-speed steel usually undergoes isothermal annealing, after
forging, to improve its machinability, reduce the hardness, and to prepare
the structure for hardening. For annealing, the steel is heated to 830–
850 °C, after it is cooled to a temperature of 720–750 °C, which corre-
sponds to the range of pearlite decomposition, and held at this temperature
from 4 to 6 hours to complete the austenite decomposition. After this, the
steel is cooled with the furnace to 600 °C at a rate of 40 to 50 °C per hour
and then, finally, it is cooled in air.
The steel must be hardened and tempered to impart red-hardness to
the tool (Figure 3.3).

Figure 3.3 – Temperature vs time diagram for hardening

and tempering high-speed steel

High temperatures are required to dissolve the carbides to the great-

est possible extent and to obtain an austenite highly alloyed with tungsten
and vanadium. This enables a martensite to be obtained, after hardening,
that has a high resistance to self-tempering, i.e., ample red-hardness.
These indicated temperatures should not be exceeded as this will consid-
erably lower the mechanical properties.
In heating for hardening, the tools are preheated to 800–875 °C, as
a rule, to prevent cracks (see fig. 3.3).
Holding time at the hardening temperature should be sufficient for
dissolving that part of the carbides which can pass into solution at this
temperature. Experience shows that the holding time in a molten salt bath
for tools from 10 to 15 mm in diameter (or thickness) may be taken as 8 or
9 seconds per millimetre of cross-section.
Oil is most frequently the quenching medium for high-speed steels.
Martempering in molten salts (usually KNO,), having a temperature of
450–550 °C, is also applied to reduce warping. The tools are held in the
bath for 2 to 5 minutes.
Austempering by holding in a bath at a temperature of 200–300 °C
for 30 to 60 minutes (according to the bath temperature) will reduce de-
formation in quenching large tools.
Directly after quenching, high-speed steel has a hardness of
HRC 61–63. Its structure comprises martensite, carbides that remained
undissolved in heating, and retained austenite. The higher the hardening
temperature, the lower the martensite points (M and Mf) will be and the
larger the amount of austenite retained. Ordinarily, no more than 20 to
30 % austenite is retained in steel P18. Since the retained austenite lowers
the cutting properties, a sub-zero treatment, at a temperature of 70 to
80 °C below zero, is often applied directly after hardening (not later than
30 to 60 minutes) to prevent stabilisation of the austenite. This treatment
should be immediately followed by tempering. The main purpose of the
last operation is to relieve internal stresses and to transform retained aus-
tenite into martensite.
Tools are subject to several (three) tempering operations at 550–
570 °C (holding time from 45 to 60 minutes); tools undergoing sub-zero
treatment after quenching are tempered once or, less frequently, twice.
Two or three tempering operations are applied after martempering
or after oil quenching to transform the maximum possible amount of re-
tained austenite. Tempering is repeated three or four times after austem-
pering or for tools of large cross-section (see fig. 3.3).
The curve in Figure 3.4 shows the influence of the tempering tem-
perature on the hardness of quenched high-speed steel. The drop in hard-
ness in the temperature range from 150 to 400 °C is probably due to the
decomposition of the martensite and coagulation of carbides of the cemen-
tite type which precipitate from the martensite.
The increase in hardness at higher tempering temperatures (known
as secondary hardness) is due to the transformation of the retained austen-
ite into martensite and to the precipitation of carbides from the martensite
(precipitation hardening of the martensite). The high hardness (HRC 63–
65) obtained by tempering in the range from 550 to 570 °C is retained
when the steel is subsequently heated to 600–620 °C. This constitutes the
high red-hardness of high-speed steel tools.

Figure 3.4 – Effect of the tempering temperature on hardness

of high-speed steel

Low-temperature liquid cyaniding or carbonitriding (at 550–570 °C

is frequently applied after heat treatment to improve the cutting efficiency
of tool.

Brief summary about tool steels

The alloying elements in the low-alloy steels dissolve in the ferrite to
form a substitutional solid solution. This solution strengthens the steel and
gives useful additional strength. The tool steels contain large amounts of
dissolved tungsten (W) and cobalt (Co) as well, to give the maximum feasi-
ble solution strengthening. Because the alloying elements have large solu-
bilities in both ferrite and austenite, no special heat treatments are needed to
produce good levels of solution hardening. In addition, the solution-
hardening component of the strength is not upset by overheating the steel.

For this reason, low-alloy steels can be welded, and cutting tools can be run
hot without affecting the solution-hardening contribution to their strength.
The tool steels are an excellent example of how metals can be
strengthened by precipitation hardening. Traditionally, cutting tools have
been made from 1 % carbon steel with about 0.3 % of silicon (Si) and
manganese (Mn). Used in the quenched and tempered state they are hard
enough to cut mild steel and tough enough to stand up to the shocks of
intermittent cutting. But they have one serious drawback. When cutting
tools are in use they become hot: woodworking tools become warm to the
touch, but metalworking tools can burn you. It is easy to “run the temper”
of plain carbon metalworking tools and the resulting drop in hardness will
destroy the cutting edge. The problem can be overcome by using low cut-
ting speeds and spraying the tool with cutting fluid. But this is an expen-
sive solution – slow cutting speeds mean low production rates and expen-
sive products. A better answer is to make the cutting tools out of high-
speed steel. This is an alloy tool steel containing typically 1 % C,
0.4 % Si, 0.4 % Mn, 4 % Cr, 5 % Mo, 6 % W, 2 % V and 5 % Co. The
steel is used in the quenched and tempered state (the Mo, Mn and Cr give
good hardenability) and owes its strength to two main factors: the fine
dispersion of Fe3C that forms during tempering, and the solution harden-
ing that the dissolved alloying elements give.
Interesting things happen when this high-speed steel is heated to
500–600 °C. The Fe3C precipitates dissolve and the carbon that they re-
lease combines with some of the dissolved Mo, W and V to give a fine
dispersion of Mo2C, W2C and VC precipitates. This happens because Mo,
W and V are strong carbide formers. If the steel is now cooled back down
to room temperature, we will find that this secondary hardening has made
it even stronger than it was in the quenched and tempered state. In other
words, “running the temper” of a high-speed steel makes it harder, not
softer; and tools made out of high-speed steel can be run at much higher
cutting speeds (hence the name).

Steels for measuring tools

Steels for measuring tools should possess high hardness and wear
resistance and should retain their size over a long period of service.
Such tools are most often manufactured of steels X and ХГ (1.3–
1.5 % C, 0.45–0.7 % Mn, and 1.3–1.6 % Cr), which acquire a high hard-
ness after quenching and low tempering.

Snap gauges, scales, rules and other flat, long measuring tools are
usually made of steel plate grades 15 and 15X. They are carburized and
quenched to obtain a working surface of the required high hardness and
wear resistance.
Die steels. Dies for blanking, heading, and drawing operations,
draw plates for wire drawing, and other tools for the cold working of met-
als are made of steels which possess sufficient toughness in addition to
their high hardness and wear resistance (Table 3.14).
Small dies of simple form are frequently made of the carbon steels
У10А and У12А (see table 3.11) which are easily machinable in the an-
nealed condition.
This facilitates die manufacture. Larger and more complex dies are
made of the alloy steels grade X, 9XC, and others (see table 3.11) which
have increased hardenability and are oil-quenched.
The application of carburizing, cyaniding, and chromizing will pro-
vide satisfactory results in increasing the wear resistance of alloy steel
dies which operate in difficult conditions but are not subject to large im-
pact loads.
Cold cutting and working dies, requiring high wear resistance and
increased toughness (complex punches and dies, deep drawing lower dies
for sheet metal, profiled rolls of complex form, thread rolling dies, etc.)
are made of high-chromium steel of the carbidic (cementitic) class,
Grades X12, X12M, Х12Ф1, and Х12Ф. High-chromium steels possess a
very high hardenability (up to 300–400 mm), exceptional wear resistance,
and undergo almost no dimensional changes after quenching.
Cast high-chromium steel has a eutectic structure. This is broken up
in subsequent hot working to form separate carbides. Even with low re-
duction in forging these carbides become arranged in bands or lines along
the direction of metal flow and this impairs the quality of the steel.
In annealed steels of the X12M and Х12Ф type, the carbide phase
comprises complex carbides such as (Cr, Fe)7C3. Steel X12, with an in-
creased carbon content, also contains carbides of the cementite type
(Fe, Cr)3C. Annealed steel X12M may contain up to 13–16 % carbides.
When the steel is heated above the eutectoid temperature, the car-
bides dissolve and alloy the austenite with chromium. Upon heating to
1100–1200 °C most of the vanadium (for steel Х12Ф) and molybdenum
(for steel X12M) is also dissolved. The stability of the supercooled austen-
ite depends to a considerable extent upon the heating temperature in hard-
ening. The higher the heating temperature, the more the austenite will be-
come alloyed and the higher its stability will be.
Table 3.14 – Chemical compositions of certain die steels
Chemical composition, %
Steel grade
С Mn Si Cr W V Ni Other elements
Steels for cold-working dies
Х12Ф1 1.2–1.4 ≤0.35 ≤0.4 11–12.5 – 0.7–0.9 – –
Х12Ф 1.4–1.6 ≤0.35 ≤0.4 11–12.5 – 0.2–0.4 – –
X12M 1.45–1.7 ≤0.35 ≤0.4 11–12.5 – 0.15–0.3 – Mo 0.4–0.6
X12 2.0–2.3 ≤0.35 ≤0.4 11.5–13 – – – –
4XB2C 0.35–0.44 0.2–0.4 0.6–0.9 1.0–1.3 2.0–2.5 – – –
6XB2C 0.55–0.65 0.2–0.4 0.5–0.8 1.0–1.3 2.2–2.7 – – –
5ХВГ 0.55–0.7 0.9–1.2 0.15–0.35 0.5–0.9 0.5–0.8 – – –
Steels for hammer dies
5XHB 0.5–0.6 0.5–0.8 0.15–0.35 0.5–0.8 0.4–0.6 – 1.4–1.8 –
5XHCB 0.5–0.6 0.3–0.6 0.6–0.9 1.3–1.6 0.4–0.6 – 0.8–1.2 –
5XHC 0.5–0.6 0.3–0.6 0.6–0.9 1.3–1.6 – – – Mo 0.15–0.3
5XHT 0.5–0.6 0.5–0.8 ≤ 0.35 0.9–1.25 – – 1.4-1.8 Ti 0.08–0.15
Steels for press-moulds
3X2B8 0.3–0.4 0.2–0.4 ≤ 0.35 2.2–2.7 7.5–9.0 – ≤ 0.3 Ti 0.2–0.5
4X8B2 0.35–0.45 0.2–0.4 ≤ 0.35 7–9 2.0–3.0 – ≤ 0.3 –
There are two main procedures for heat treating dies of high-
chromium steels:
1 Hardening (heating temperatures: steel X12Ф1 – 1070–1090 °C,
X12M – 1020–1040 °C, and X12 – 1000–l040 °C; quenching in oil) fol-
lowed by low tempering (150–170 °C). The hardness after this treatment
will be HRC 61–63.
Steel acquires high hardness (HRC 61–63) and wear resistance after
this treatment. If increased toughness is required at the expense of hard-
ness (decreased to HRC 57–59) and if the dimensions of the part are to be
reduced in comparison to those obtained after quenching, the tempering
temperature is raised to 200–275 °C.
Dies operating under very heavy impact loads are tempered at 400–
425 °C.
2 Hardening (heating temperatures: steel X12M – 1115–1130 °C and
Х12Ф1 – 1150–1170 °C; quenching in oil or molten salt at a temperature of
400–500 °C) followed by repeated tempering (2 or 3 operations) at 510–
520 °C. The steel is not very hard (HRC 44–52), directly after quenching,
due to the large amount of retained austenite. During repeated tempering,
retained austenite is transformed into martensite and the hardness is in-
creased to HRC 60–61. A single tempering at 510–520 °C will prove suffi-
cient if a sub-zero treatment at 80 °C below zero is applied after quenching.
The first procedure will provide high mechanical properties and
minimum deformation.
Steel, to which the second procedure has been applied, acquires
high red-hardness and wear resistance. This procedure is recommended
for dies operating at 400–500 °C and also for cutting tools (milling cut-
ters). It must be noted, however, that quenching from high temperatures is
associated with large dimensional changes and does not ensure high me-
chanical properties.
Disadvantages of high-chromium steels are their poor machinability
in the annealed state (hardness HB 207–269) and their lowered mechani-
cal properties in instances when carbide inhomogeneity is sharply defined.
Steels with increased toughness, such as Grades 4XB2C, 6XB2C,
5ХВГ, and others, are used in the manufacture of air-hammer chisels,
trimming dies, cold-cutting shear blades, and other tools with thin work-
ing edges. These steels acquire a hardness of HRC 50–55 after quenching
with subsequent tempering at 250–270 °C.
Steels for hammer dies used in the hot working of metals must have
sufficiently high mechanical properties at elevated temperatures; they must
resist softening after repeated heating, have a high hardenability, minimum
deformation upon heat treatment, and no susceptibility to temper brittleness.
Steel 5XHCB is usually employed for making large dies (length of
a side > 400 mm), steels 5XHB and 5XHCB for medium-sized dies, and
steel 5XHC for small dies (with the shortest side up to 300 mm in length).
These dies are hardened by heating to 840–870 °C and quenching in oil
(or in air). This is followed by tempering at 520–570 °C to relieve the
stresses and to obtain a homogeneous sorbite structure.
After tempering, the hardness will be within HRC 35–45 (HB 321–430).
The components of die-casting dies for copper and aluminium al-
loys are made of steel 3X2B8; steel 4X8B2 is used for these components
in die casting zinc alloys. These two grades of steel have good thermal
endurance, high mechanical properties at elevated temperatures, and they
are corrosion and erosion resistant. They are employed for heading dies
and for moulds that operate for prolonged periods, at high temperatures.

3.1.5 Steels with special physical properties

Wear-resistant (austenitic) steels

Parts subject to wear under conditions of high contact pressures and
impacts (crawler links, stone-crusher jaws, excavator buckets, frogs of
railroad and street car tracks, etc.) are made of high-manganese austenitic
steels Г13Л and Г13 containing 1.0–1.5 % С and 11.0–15.0 % Mn. As
cast, these steels have an austenitic structure with excess carbides (Mn3C)
along the grain boundaries. This decreases the strength and toughness of
the steel. Therefore, parts made of steel Г13Л are hardened by heating to
l100 °C and quenching in water. Cracks are prevented by heating first to
650 °C at a rate of 70 to 100° per hour for small castings and 50° per hour
for large complex castings. The holding time at 650 °C is from one to four
hours. At the hardening temperature, the holding time is 0.5 to 4 hours, in
accordance with the size of the casting.
These steels are very sensitive to decarburization which causes the
formation of martensite. Therefore, thin articles must be protected against
decarburization by metal coating or special pastes.
After properly conducted heat treatment, these steels have an aus-
tenitic structure and the following mechanical properties:
Tensile strength σu, MPa .................. 630–700
Yield point σy, MPa.......................... 310–350
Unit elongation δ, %......................... 25–15
Reduction of area ψ, % .................... 30–20
Hardness HB..................................... 170–228
The steel is readily work-hardened in the course of operation and
acquires high wear resistance. If the wear is purely abrasive and the steel
is not subject to impact loads which would cause work hardening, the
wear resistance will not be very high.
Corrosion-resistant steels
Corrosion resistance. The fracture of metals and alloys by the ac-
tion of the surrounding medium is called corrosion. Corrosive destruction
begins at the surface, i.e., at the interface between the metal and the me-
dium, and gradually penetrates into the metal. At this, metals and alloys
lose their metallic lustre and become coated with the corrosion products
(rust). The mechanical properties of corroded steel are sharply reduced
even when no visible change is observed on the metal surface.
Plain carbon steels rust in wet environments and oxidise if heated in
air. One effective method of protecting steel against corrosion is to add
elements that will form a stable protective film on the surface of the metal
to prevent destruction of the base metal.
If chromium is added to steel, a hard, compact film of Cr2O3 will
form on the surface and this will help to protect the underlying metal. The
minimum amount of chromium needed to protect steel is about 13 %, but
up to 26 % may be needed if the environment is particularly hostile. The
iron-chromium system is the basis for a wide range of stainless steels.
Stainless steels. Steels resistant to atmospheric corrosion are called
stainless steels.
The simplest stainless alloy contains just iron and chromium (it is
actually called stainless iron, because it contains virtually no carbon). The
interesting thing about Fe–Cr phase diagram is that alloys containing
≥ 13 % Cr have a b.c.c. structure all the way from 0 К to the melting
point. They do not enter the f.c.c. phase field and cannot be quenched to
form martensite. Stainless irons containing ≥ 13 % Cr are therefore always
Hardenable stainless steels usually contain up to 0.6 % carbon.
This is added in order to change the Fe–Cr phase diagram. Carbon ex-
pands the γ-field so that an alloy of Fe–15 % Cr, 0.6 % С lies inside the γ-
field at 1000 °C. This steel can be quenched to give martensite; and the
martensite can be tempered to give a fine dispersion of alloy carbides.
These quenched and tempered stainless steels are ideal for things like non-
rusting ball-bearings, surgical scalpels and kitchen knives.
Many stainless steels, however, are austenitic ( at room tem-
perature. The most common austenitic stainless, “18/8”, has a composition
Fe–0.1 % C, 1% Mn, 18 % Cr, 8 % Ni. The chromium is added, as before,
to give corrosion resistance. But nickel is added as well because it stabi-
lises austenite. The Fe–Ni phase diagram shows why. Adding nickel low-
ers the temperature of the f.c.c.–b.c.c. transformation from 914 °C for
pure iron to 720 °C for Fe–8 % Ni. In addition, the Mn, Cr and Ni slow
the diffusive f.c.c.–b.c.c. transformation down by orders of magnitude.
18/8 stainless steel can therefore be cooled in air from 800 °С to room
temperature without transforming to b.c.c. The austenite is, of course, un-
stable at room temperature. However, diffusion is far too slow for the me-
tastable austenite to transform to ferrite by a diffusive mechanism. It is, of
course, possible for the austenite to transform displacively to give marten-
site. But the large amounts of Cr and Ni lower the Ms temperature to
≈ 0 °C. This means that we would have to cool the steel well below 0 °C
in order to lose much austenite.
Austenitic steels have a number of advantages over their ferritic
cousins. They are tougher and more ductile. They can be formed more
easily by stretching or deep drawing. Because diffusion is slower in f.c.c.
iron than in b.c.c. iron, they have better creep properties. And they are
non-magnetic, which makes them ideal for instruments like electron mi-
croscopes and mass spectrometers.
But one drawback is that austemtic steels work harden very rapidly,
which makes them rather difficult to machine.
Extensively used of stainless steels are the chromium steels 1X13,
2X13, 3X13, and 4X13 containing from 0.1 to 0.4 % С and 12–14 % Cr
(Table 3.15) and the high-carbon steel X18 (0.9–1.0 % С and 17–
19 % Cr) (Table 3.15).
Grades 1X13 and 2X13 are hypoeutectoid steels while grades 3X13
and 4X13 are eutectoid and hypereutectoid steels, respectively. The aus-
tenite formed in heating these steels is transformed into martensite upon
quenching. High-carbon steel X18 is of the cementitic class. Heat treating
procedures and applications of chromium stainless steels are listed in Ta-
ble 3.15.
Chromium stainless steels have a high corrosion resistance in air,
water, and alkali or acid solutions. This resistance is due to the formation
of a thin oxide film (Cr2O3) on the surface of the steel. The corrosion re-
sistance is most pronounced when all of the chromium is in the solid solu-
tion and the steel has a single-phase structure. The formation of carbides
is associated with the depletion of chromium from the solid solution. This
lowers the corrosion resistance.

Table 3.15 – Heat treatment and applications of chromium stainless steels
Heat treatment Applications
Parts requiring high ductility and subject
Hardening: heat to 1000–1050 °C, to impact loads (turbine blades, valves
1X13 quench in water (or oil), of hydraulic presses, cracking plant fittings,
and temper at 700–790 °C bolts, nuts, household appliances, etc.).
Good resistance in air, in water and in steam
Hardening: heat to 1000–1050 °C
The same parts but with higher hardness
2X13 quench in water (or oil),
(3X13 has a hardness of 48 HRC)
and temper at 660–770 °C
Hardening: heat to 1000–1050 °C,
Higher resistance in air, as well
3X13 quench in oil, and temper
as in water and steam
at 200–300°C
Cutting and measuring tools, surgical
Hardening: heat to 1050–1100 °C,
instruments, carburettor needles, and ball
4X13 quench in oil, and temper
bearings (hardness – 50 HRC). Corrosion
at 200–300 °C
resistance tower than for preceding grades
Ball bearings, high-quality knives, bushings,
X18 Ditto globe valves, and other parts subject to wear
and requiring increased corrosion resistance

Consequently, the higher the carbon content, the lower the corro-
sion resistance will be. The corrosion resistance of a part is increased by
heat treatment and polishing. Machine parts made on automatic screw
machines and thereby requiring improved machinability (screws, nuts, and
other threaded parts) are made of steel X14 ≤ 0.15 % C, 13–15 % Cr, and
0.2–0.4 % S) which contains more sulphur than the other grades. Its resis-
tance to atmospheric corrosion is quite satisfactory.
Acid-resistant steels are highly resistant to corrosion under the ac-
tion of various aggressive media. They include ferritic steels with a high
chromium content (X17, X25, and X28) and austenitic chromium-nickel
steels (Table 3.16).
The most important of the acid-resistant steel series are the austen-
itic chromium-nickel steels listed in Table 3.16. Their structure in the
equilibrium state consists of two phases ((austenite + carbides (Cr,
Fe)23C6). Only alloys containing less than 0.04 % С have structures free of
excess carbides. After hardening (heating to 1100–1150 °C and quenching
in water), these steels acquire a single-phase structure.
Due to the low rate of chromium diffusion at temperatures below
500 °C, the austenite is stable and no carbides precipitate from it. The
hardened steel has a comparatively low strength but high ductility. Its me-
chanical properties are: σu = 550–580 MPa, σy = 200–220 MPa, δ = 45–
40 % and ψ = 60–55 %.
Table 3.16 – Compositions and applications of chromium-nickel acid-resistant steels
Steels Chemical composition, %
Certain applications and characteristics
grade С Mn Cr N1 Ti
1X18H9 ≤ 0.14 ≤ 2 17–20 8–11 –Floats and fairing of seaplanes, fuel tanks,
nonmagnetic parts of ship navigation
instruments, devices exposed to sea-water
and weak alkalis. Heat treatment
2X18H9 0.15–0.25 ≤ 2 17–20 8–11 –
is applied, after welding and other
fabricating processes involving heating,
to prevent intercrystalline corrosion
Used in the following industries: nitrogen
(acid tanks, absorption towers, etc.), paint
1X18H9T ≤ 0.12 ≤ 2 17–20 8–11 < 0.8 and varnish (autoclaves, mixers, etc.),
coal (pumps and devices operating in
acidic waters), and dairy (cans and flasks)

The strength of this steel may be increased by work hardening. The

hardened steel is easily formed and has good weldability.
One notable factor is that improper heat treatment (slow cooling in
the range from 500 to 700 °C or heating to 500–700 °C), for instance in
welding, will cause the austenite to decompose and chromium carbide will
be precipitated along the grain boundaries. Due to the low rate of chro-
mium diffusion, carbides are formed only by the chromium at the grain
surfaces. This impoverishes the chromium content of the austenite adja-
cent to the grain boundaries below 12 % and the corrosion resistance of
the steel is sharply reduced. The steel acquires a susceptibility to corrosion
on the grain boundaries (intercrystalline corrosion).
Upon a further development of intercrystalline corrosion, steel loses
its metallic sound and tears will appear in flexure tests along the grain
boundaries at places where the metal is destroyed by corrosion. Metal
with well developed corrosion will fail at small stresses.
Two methods are employed to prevent intercrystalline corrosion: 1)
the carbon content is reduced; this correspondingly reduces the amount of
precipitated carbides (as in steel 0X18H9) and 2) alloying elements are
added to prevent the precipitation of chromium carbide. Such elements are
titanium (steel 1X18H9T) and niobium (steel Х18Н9Б).
The addition of these elements combines all of the carbon in the sta-
ble carbides TiC and NbC thus preventing the possibility of chromium car-
bide precipitation. The amount of titanium or niobium is specified so that all
of the carbon is combined in the carbides of these elements. The amount of
titanium should be ≤ 0.8 % (0.03 % С); niobium is added up to 1.5 (experi-
mental data). Steel 1X18H9T is not susceptible to intercrystalline corrosion.
As a substitute for steel 1X18H9, steel 1Х13Н4Г9 is used to some
extent. Here a large part of the nickel has been replaced by manganese.
This substitute has lower corrosion resistance and is susceptible to inter-
crystalline corrosion.

Scale-resistant and heat-resistant steels and alloys

The scale-resistant steel is one not susceptible to scale formation (gas
corrosion) at high temperatures. This property of a steel is improved, chiefly,
by the addition of chromium, aluminium, and silicon which, upon heating,
form dense oxide films, such as (Cr, Fe)2O3, (Al, Fe)2O3 and others, that pro-
tect the base metal against oxidation. An addition of 5 to 8 % Cr raises the
scale resistance to 700–750 °C, increasing the content to 15–17 % will pre-
vent scaling up to 950–l000 °C, and 25 % Cr will prevent scaling up to
l100 °C. It is necessary to point out that scale resistance depends only on the
composition and not on the structure of the steel. Ferritic and austenitic steels
with equal chromium content have practically the same scale resistance.
Table 3.17 lists the compositions and properties of certain scale-
resistant and heat-proof steels.
Table 3.17 – Compositions, properties, and applications of scale-resistant and heat-proof steels
Grade Composition Chief properties Typical applications
≤ 0.15 % C, Scale-resistant Boiler components operating under
X6C 1.5–2.0 % Si, up to 750 °C increased loads at temperatures up
5.0–6.5 % Cr to 750 °C, valves, etc.
0.35–0.5 % C, Scale-resistant Valves and components operating under
X9C2 2.0–3.0 % Si, up to 800 °C low loads at temperatures up to 800 °C
8.0–10 % Cr
0.07–0.12 % C, Scale-resistant Components operating under low loads
1.2–2 % Si, up to 900 °C at temperatures up to 900 °C
11.5–14 % Cr,
1.0–1.8 % Al
≤ 0.15%C, Heat-resistant Pipe used in cracking processes under
1.5–2 % Si, up to 650 °C sulphur corrosion conditions, pump
5–6.5 % Cr, and scale-resistant components, gate valves, piston rods, etc.
0.45–0.6 % Mo
0.3–0.4 % C, Scale-resistant Heavily loaded components: furnace
2–3 % Si, up to l100 °C, conveyers and fastening parts subject
17–20 % Cr, heat-and acid- to high temperatures and pressures
23–26 % Ni resistant
≤ 0.2 % C,
2–3 % Si,
23–27 % Cr,
18–21 % Ni

Steels X6C, X9C2, X6CM, and X10C2M contain large amounts of
chromium and silicon. They are known as silchrome steels. These steels
are of the martensitic class and require a preliminary heat treatment to im-
prove their machinability (high tempering at 820–850 °C). Final heat
treatment consists of hardening by heating to 1040–1100 °C, quenching,
and then tempering at 750–850 °C to stabilise the structure which is a fer-
rite-carbide mixture.
Silchromes are susceptible to temper brittleness and grain growth.
Molybdenum may be added, as in steel X6CM, to eliminate temper brit-
Heat-resistant steels retain sufficient strength and scale resistance
at high temperatures. Such steels must resist creep well and possess high
short-time and rupture strengths.
Heat resistance depends upon the strength of the interatomic bonds
and upon the structure of an alloy. Alloying increases the heat resistance
of alloys having the same base by increasing the intensity of interatomic
attraction and raising the recrystallisation temperature.
These changes impede the development of plastic deformation at
elevated temperatures. This resistance to plastic deformation is also in-
creased by the formation of a heterogeneous structure and by breaking up
the blocks of mosaic structure.
The heat resistance of steels and alloys is improved by alloying
elements that increase their tendency to age, and thereby they are hard-
ened by the precipitation of microscopic particles that impede plastic de-
formation at high temperatures.
Austenitic steels are more heat resistant than ferritic grades be-
cause, in the former, such processes as recovery and recrystallisation take
place at higher temperatures. Larger grain size also increases heat resis-
Nickel-base alloys are extensively used as heat-resistant materials.
Iron-base heat-resistant alloys may be classified into the following
two groups:
1 Austenitic-carbidic steels (ЭИ69, ЭИ388, ЭИ481, ЭИ590, and
ЭИ374) whose high heat resistance is associated with the strengthening of
the austenite by the formation of highly dispersed carbide phases in age-
ing. In steel ЭИ69 the carbide phase consists principally of Cr23C3; in
steels ЭИ388 and ЭИ481 it consists of Cr23C6 and VC.
2 Austenitic-intermetallic steels (ЭИ696) achieve their high heat re-
sistance by the formation of intermetallic phases Fe2Ti and Ni3Ti or the
intermediate α’-phase during ageing or operation at high temperatures.
Heat-resistant alloys ЭИ437, ЭИ617, and others are of the nickel-
base type. The nickel forms a solid solution with the chromium, tungsten,
and molybdenum. During ageing, aluminium, titanium, and boron form
excess intermetallic phases. In the initial period of solid solution decom-
position, an intermediate α'-phase is formed on the basis of the compound
Ni3(Al, Ti, Cr). It has a face-centred cubic lattice, coherently bound to the
solid solution. The stable phases, nickel titanide (Ni3Ti) with a hexagonal
lattice and Ni3Al are formed after prolonged ageing at high temperatures
(800–900 °C). Carbides and borides may also be formed in addition to
these phases. Nickel-base alloys acquire their high heat resistance due to
precipitation hardening, associated with the formation of intermetallic
phases, and by the alloying of the solid solution with molybdenum and
tungsten. These alloying elements increase the intensity of interatomic
attraction and impede the softening of the alloy at high temperatures. The
heat treating procedures given in Table 3.18 provide for maximum heat
resistance of these alloys.
Table 3.18 – Heat treatment of heat-resistant alloys
Grade Heat treatment
ЭИ388 Quenching from 180 °С in water and ageing at 800 °С for 8–10 hours
ЭИ481 Quenching from 1140-1200 °С and double ageing: 1) at 690 °С for 16 hours
and 2) at 780–800 °С for 10 to 16 hours
ЭИ696 Quenching from 1100–1200°С and ageing at 700–800 °С
ЭИ437 Quenching from 1080 °С in air and ageing at 700 °С
ЭИ617 Double quenching: 1) from 1200 °С (holding time 2 hours) in air, and 2) from
1050 °С (holding time 4 hours) in air, followed by ageing at 800 °С for 16 hours

The alloys are heated to high temperatures in hardening to put the

excess phases (carbides and intermetallides of the Ni3Ti type and others)
into the solid solution. Rapid cooling (quenching) fixes the supersaturated
solid solution. This saturation with alloying elements considerably distorts
the crystal lattice. It also promotes stress increase and breaks up the mo-
saic structure. The result is increased resistance to plastic deformation.
During ageing, the excess phases precipitate from the solid solution.
Proper ageing procedure will increase the hardness and heat resistance of
the alloy.
The precipitation of excess phases improves the heat resistance only if
the solid solution remains alloyed to a sufficient extent. When the solid solu-
tion is considerably impoverished and the grain of the excess phase is coars-
ened, the heat resistance will be lowered. Ageing must be carried out at tem-
peratures that exceed the working temperatures of subsequent operation.

Heat-resistant alloys are very difficult to work and to machine.
The first group of certain heat-resistant alloys includes iron-base al-
loys hardened by heat treatment and “hot-cold working”.
Timken alloys 16–25–6 and 19-9DL are used in the manufacture of
turbo-superchargers and gas turbine rotors after a reduction of 22–25 % at
650–760 °C (hot-cold working). This procedure increases the tensile
strength and the yield point. No additional heat treatment is required by
these alloys.
Alloys A-286 and Discaloy-24 which contain small additions of ti-
tanium and aluminium, are hardened by heat treatment.
Cobalt-nickel-chromium-iron alloys are intended for service at
higher temperatures. They are usually employed in the solution treated or
solution treated and aged condition. The strength properties of these alloys
are improved by the addition of molybdenum, tungsten, and niobium. Ti-
tanium and aluminium, which form intermetallic compounds, may be
added to provide greater response to heat treatment.
Development of the high-strength heat treatable nickel-base alloys
such as Inconel X, 550; Inco 700; M-252; and Waspaloy was based on the
necessity for new alloys to replace, in part, the cobalt-base alloys (such as
S-816) since cobalt is very expensive.
The fourth group lists the cast and wrought cobalt-base alloys
which have been extensively used for aircraft gas turbine guide vanes and
Cast alloys, which contain more carbon and have coarser grain
structure, are more heat resistant than the wrought alloys.
Alloys of molybdenum with titanium, niobium, vanadium, and co-
balt have exceptionally high heat-resisting properties. When reliable
methods are found for protecting molybdenum alloys from gas corrosion,
they will lead among materials designed for high-temperature service.
Scale-resistant alloys with high electrical resistivity. This group in-
cludes scale-resistant alloys and steels with high electrical resistivity in-
tended for making round and flat heating elements of electrical appliances
(for example, furnaces). Ferritic steels, containing chromium and alumin-
ium which sharply increase the scale resistance and electrical resistivity
(X13Ю4-fechral, 1Х17Ю5, 0Х17Ю5, 0Х25Ю5, and 1Х25Ю5), are ex-
tensively employed for this purpose, as well as chromium-nickel alloys
(of the nichrome type) such as X15H60 and X20H80. The resistivity of
these alloys ranges from 1.15 (for X20H80) to 1.3 (for Х13Ю4) ohm-
sq mm per m and they have a scale-resistance from 850 °C (for Х13Ю4)
to 1100–1150 °C (for 1Х25Ю5 and X20H80).
3.2 Non-ferrous alloys
The term non-ferrous alloy is used for all alloys where iron is not
the main constituent, e.g. alloys of aluminium, of copper, of magnesium,
etc. The following are some of the general properties and uses of non-
ferrous alloys in common use in engineering.
Aluminium alloys – Low density, good electrical and thermal con-
ductivity, high corrosion resistance. Tensile strengths of the order of 150
to 400 MPa, tensile modulus about 70 GPa. Used for metal boxes, cook-
ing utensils, aircraft bodywork and parts.
Copper alloys – Good electrical and thermal conductivity, high cor-
rosion resistance. Tensile strengths of the order of 180 to 300 MPa, tensile
modulus about 20 to 28 GPa. Used for pump and valve parts, coins, in-
strument parts, springs, screws.
Magnesium alloys – Low density, good electrical and thermal con-
ductivity. Tensile strengths of the order of 250 MPa and tensile modulus
about 40 GPa. Used as castings and forgings in the aircraft industry where
weight is an important consideration.
Nickel alloys – Good electrical and thermal conductivity, high corro-
sion resistance, can be used at high temperatures. Tensile strengths between
about 350 and 1400 MPa, tensile modulus about 220 GPa. Used for pipes
and containers in the chemical industry where high resistance to corrosive
atmospheres is required, food processing equipment, gas turbine parts.
Titanium alloys – Low density, high strength, high corrosion resis-
tance, can be used at high temperatures. Tensile strengths of the order of
1000 MPa, tensile modulus about 110 GPa. Used in aircraft for compres-
sor discs, blades and casings, in chemical plant where high resistance to
corrosive atmospheres is required.
Zinc alloys – Low melting points, good electrical and thermal con-
ductivities, high corrosion resistance, tensile strength about 300 MPa, ten-
sile modulus about 100 GPa. Used for car door handles, toys, car carburet-
tor bodies – components that in general are produced by pouring the liquid
metal into dies.
As an example of a non-ferrous alloy, consider copper alloys. Pure
copper is a soft material with low tensile strength. For many engineering
purposes it is alloyed with other metals. The exception is where high elec-
trical conductivity is required. Pure copper has a better conductivity than
the alloys. The following indicates the names given to the various types of
copper alloys:
- Copper with zinc – Brasses;

- Copper with tin – Bronzes;
- Copper with tin and phosphorus – Phosphor bronzes;
- Copper with tin and zinc – Gunmetals;
- Copper with aluminium – Aluminium bronzes;
- Copper with nickel – Cupro-nickels;
- Copper with zinc and nickel – Nickel silvers;
- Copper and silicon – Silicon bronze;
- Copper and beryllium – Beryllium bronze.
Brasses with between 5 and 20 % zinc are called gilding metals
and, as the name implies, are used for architectural and decorative items to
give a “gilded” or golden color. Cartridge brass is copper with 30 % zinc.
One of its main uses is for cartridge cases, items which require high duc-
tility for the deep drawing process used to make them. The term common
or basis brass is used for copper with 37 % zinc. This is a good alloy for
general use with cold-working processes and is used for fasteners and
electrical connectors. It does not have the high ductility of those brasses
with less zinc. Copper with 40 % zinc is called Muntz metal.
The changes in the properties of brasses when the amount of zinc is
changed arises from changes in the structure. Brasses with between 0 and
35 % zinc form one type of structure, termed alpha, 5 % and 45 % there is
a mixture of this alpha structure and another structure termed beta. It is
this change in structure, i.e. the way the atoms of copper and zinc are
packed together, that is responsible for the abrupt changes in properties of
brass at 35 % zinc.

Rare metals and their alloys

Among the 102 chemical elements, about 55 are included in the
group of rare metals.
Due to the enormous progress in electrovacuum engineering, semi-
conductor electronics, production of atomic energy, as well as the devel-
opment of new heat- and acid-resistant alloys and cemented carbides, rare
elements find ever-increasing applications in industry. Many of them have
ceased to be rare (Mo, W, and others) and have entered the ranks of ordi-
nary metals.
It is accepted practice to divide all rare metals into the following
groups, in accordance with their physicochemical properties and method
of extraction:
1) Light rare metals, having a low specific gravity – lithium (0.53),
rubidium (1.55), caesium (1.87), and beryllium (1.85).

2) Refractory rare metals – titanium, zirconium, hafnium, vana-
dium, niobium, tantalum, tungsten, molybdenum, and rhenium.
3) Scattered metals – indium, thallium, gallium, germanium, haf-
nium, selenium, tellurium, and rhenium. The metals of this group are of
prime importance in semiconductor engineering. According to their prop-
erties, selenium and tellurium should be considered to be metalloids.
4) Rare-earth metals (Lanthanide series) which include lanthanum
and fourteen elements from cerium to lutetium, inclusively.
5) Radioactive rare metals. These include the naturally occurring
radioactive elements (polonium, francium, radium, actinium, protoactin-
ium, thorium, and uranium), artificially produced metals (technetium,
promethium, and astatine), and the transuranium elements (neptunium,
plutonium, americium, and others). The radioactive properties of these
metals determine their fields of application (nuclear fuel, flaw detection,
etc.) and, especially, their processing techniques.
General characteristics of certain rare metals. Brief characteristics
follow for the more extensively used rare metals in industry.
Most interesting of these are the refractory group which are becom-
ing more and more important in modern engineering and especially in its
newest fields-nuclear and jet aircraft engineering.
Tungsten (W) – [Specific gravity – 19.3, Melting point –
3400 ± 50 °C] – Wire, sheets, and forgings, used in the production of in-
candescent lamps, X-ray apparatus, high-vacuum amplifiers, radio appara-
tus, etc.; also used as an alloying element in steel.
Characteristic features of tungsten are its high melting point, low
thermal coefficient of expansion (4.44 · 10–6), high electrical resistivity
(0.055 ohm-mm2/mm), and good corrosion resistance upon being exposed
to the atmosphere and certain acids (hydrochloric, sulphuric, nitric, and
others). Tungsten easily oxidises when heated above 400–500 °C. The
mechanical properties depend on preceding treatment. Tungsten wire has
the following properties: σu = 1100–4150 MPa, δ = 0–4 % and
E = 350 000–380 000 MPa.
Tungsten alloys have found certain applications. An alloy, for in-
stance, containing 20–25 % Mo has a linear coefficient of thermal expansion
near to that of refractory glass and this alloy is used in electrical high-
vacuum engineering in cases when a high electrical resistivity is required.
Alloys of tungsten with copper or with silver (5 to 40 %) obtained by pow-
der metallurgy techniques, display high resistance to wear and electrical ero-
sion in conjunction with good electrical and thermal conductivity. They are
widely used for electrical contacts which operate under severe conditions.
Molybdenum (Mo) – [Specific gravity – 10.2, Melting point –
2622 ± 10 °C] – Wire and strip used in the production of incandescent
lamps, in electrical high-vacuum engineering, for making electrical heat-
ing elements of high-temperature furnaces, and also as an alloying ele-
ment in steel.
Molybdenum has a high melting point, low thermal coefficient of
expansion (6.35 · 10–6) and a high electrical resistivity (0.0517 ohm-
Depending on the purity, extraction method, and preceding working
and heat treatment, the mechanical properties of molybdenum may vary in
wide ranges.
The mechanical properties listed in Table 3.19 refer to molybdenum
after undergoing cold reduction (35–90 %) and recrystallization. The duc-
tility increases with the degree of reduction.
Table 3.19 – Mechanical properties of molybdenum
Mechanical properties
Production method
σu, MPa σy, MPa ψ, %
Smelting 490–540 340–370 8–72
Powder-metallurgy 530–540 380–450 8–72

Additions of nickel, cobalt, titanium, chromium, vanadium, and zir-

conium increase the strength of molybdenum.
Molybdenum and its alloys have a high melting point and a high re-
crystallization temperature. Their recent wide applications as heat-
resistant materials are due to these features.
A disadvantage of molybdenum is its lack of oxidation resistance at
temperatures above 500–600 °C. By alloying molybdenum with nickel
(~ 15 %), and cobalt (~ 2.5 %), or with chromium (~ 25 %) and titanium
(~ 5 %), its resistance to gas corrosion can be increased to a considerable
extent. Recently great progress has been made towards solving the prob-
lem of protecting molybdenum against gas corrosion by means of special
Wrought molybdenum and tungsten are produced by the powder-
metallurgy method and, to a lesser extent, by a vacuum arc-melting proc-
ess with consumable electrodes.
Zirconium (Zr) – [Specific gravity – 6.507, Melting point –
1845 °C] – Zirconium and its alloys are used in the manufacture of
chemical apparatus; for cooling tubes, pumps, envelopes, fuel elements,
and other devices of nuclear-power reactors; as a getter in electrical high-

vacuum engineering, as well as for the production of copper- and magne-
sium-base alloys; deoxidant and alloying element in steel, etc.
Pure zirconium is soft; it has high ductility and good resistance to
attack in a number of corrosive media. Impurities and absorbed gases
make zirconium brittle. Zirconium can be worked, both hot and cold, with
ease, its machinability is fair, and it can be arc-welded in an argon atmos-
phere. The mechanical properties may vary in wide ranges depending on
the method of production and previous treatment (σu = 200–450 MPa and
δ = 20–30 %). Zirconium alloys with Co, Nb, Si, Ti, V, and others have
higher strengths (σu up to 600–1000 MPa and δ = 1–24 %).
Zirconium and its alloys have a low thermal-neutron-capture cross-
section. This, in conjunction with their high corrosion resistance to radio-
active coolants, is the reason why they have become recognised as a valu-
able structural material for atomic-power reactors.
Additions of Sn, Ni, Fe, and Cr, which have no practical influence
on its thermal-neutron-capture cross-section, increase the corrosion resis-
tance of zirconium in water at high temperatures and pressures.
Tantalum (Та) – [Specific gravity – 16.6, Melting point –
2996 ± 50 °C] – Electrodes of vacuum tube amplifiers; anodes, grids, and
other parts of electronic tubes; for chemical apparatus; in the production
of cemented carbides, etc.
Tantalum resembles niobium very closely in its physicochemical,
mechanical, and fabrication properties.
Ductile niobium and tantalum are produced either by powder metal-
lurgy or by an arc-melting process. They are worked cold with intermedi-
ate annealing in a high vacuum.
Niobium (Nb) – [Specific gravity – 8.55, Melting point –
2415 ± l5 °C] – Parts of nuclear reactors (jackets of fuel elements, cooling
tubes); sheets, strips, and tubing for apparatus in the chemical industries.
Because of its high melting point, good corrosion resistance and low
thermal-neutron-capture cross-section (1.1–1.2 Barns), niobium has found
extensive applications as a structural material in nuclear engineering.
Niobium has the following mechanical properties (for as-annealed
sheet): σu =340–420 MPa, δ = 20–30 % and E = 160 000 MPa.
Beryllium (Be) – [Specific gravity – 1.84, Melting point –
1284 °C] – X-ray windows of X-ray tubes as a material penetrable to X-
rays; sound-conducting components of acoustic apparatus; in nuclear reac-
tors as a neutron moderator; the production of copper-base, magnesium-
base, and other alloys.

Germanium (Ge) – [Specific gravity – 5.323, Melting point –
980 °C] – Semiconductor electronics.
The most important, so far, of the scattered rare metals are germa-
nium, with semiconductor properties, and rhenium. The letter's high elec-
trical resistivity, high corrosion resistance, and good thermoelectric prop-
erties provide opportunities for extensive applications of rhenium in elec-
trical engineering and radio electronics.
In all probability, the rare-earth metals of the cerium group will
soon be widely employed for alloying (modifying) steel to increase its
strength, ductility, hardenability, etc. For this purpose ferrocerium is
added to steel (0.3 to 2.5 kg per ton) or a mixture of rare earths, called
Misch metal, is introduced.
Cerium is used together with magnesium for inoculating cast iron.
Data are available on the improvement of the heat-resistant properties of
nichrome and magnesium alloys by the addition of rare-earth metals.
Recently, there has been a trend towards the application of rare-
earth metals as special-property materials. Cerium alloys, for example, are
used as a pyrophoric material for cigarette-lighter flints while the rare-
earth elements gadolinium, samarium, promethium, europium, and dys-
prosium are used as control rods of nuclear reactors due to their high
thermal-neutron-capture cross-sections.

Michael F. Ashby, David R.H. Jones. Engineering materials 1:
An introduction to properties, applications and design. Third edition.
2005. 0-7506-6380-4.
Michael F. Ashby, David R.H. Jones. Engineering materials 2:
An introduction to microstructures, processing and design. Third edition.
2006. 0-7506-6381-2.
Rajiv Asthana, Ashok Kumar, Narendra B. Dahotre. Materials
processing and manufacturing science. 2006. 0-7506-7716-3.
W. Bolton. Materials for engineering. Second edition. 2000. 0-7506-
William F. Hosford, Robert M. Caddell. Metal forming: Mechanics
and metallurgy. 1983. 0-13-577700-3.
Lakhtin Y.M. Engineering physical metallurgy. English translation.
Mir Publishers, 1975. 444 pp.
Novikov I. Theory of heat treatment of metals. English translation.
Mir Publishers, 1978. 435 pp.

Учебное издание

Турилина Вероника Юрьевна


Механические свойства металлов. Термическая обработка

металлов. Специальные стали и сплавы

Учебное пособие

Редактор И.Е. Оратовская

Компьютерная верстка Л.Е. Чистяковой

Подписано в печать 14.03.13 Бумага офсетная

Формат 60 × 90 1/16 Печать офсетная Уч.-изд. л. 9,625
Рег. № 394 Тираж 100 экз. Заказ 3869

Национальный исследовательский
технологический университет «МИСиС»,
119049, Москва, Ленинский пр-т, 4
Издательский Дом МИСиС,
119049, Москва, Ленинский пр-т, 4
Тел. (495) 638-45-22
Отпечатано в типографии Издательского Дома МИСиС
119049, Москва, Ленинский пр-т, 4
Тел. (499) 236-76-17, тел./факс (499) 236-76-35