STRUCTURE AND

PROPERTIES OF METALS
Metals are one of the most widely used types of engineering materials.
Some of their properties, e.g. elastic constants, can be directly related to the
nature of the metallic bonds between the atoms. On the other hand, macroand
microstructural features of metals, such as point defects, dislocations,
grain boundaries, and second phase particles, control their yield, flow, and
fracture stress. Images of microstructural elements can be obtained by modern
imaging techniques. Modern computer aided methods can be further used
to obtain a quantitative description of these microstructures. These methods
take advantage of the progress made in recent years in the field of image
processing, mathematical morphology and quantitative stereology. Quantitative
description of the microstructures are used for modeling processes
taking place under the action of applied load at a given temperature and
test (service) environment. These model considerations can be illustrated on
the example of an austenitic stainless steel, which is an important material
for power generating and chemical industry. Reports recently published also
show that properties of materials can be significantly modified by the effect
of free surface. Examples of such situations include environmental effect on
the mechanical properties of materials. Data for an austenitic stainless steel
is used to discuss contribution of the free surface to the mechanical properties
of metals.
PACS numbers: 62.20.—x, 62.20.Fe
 Introduction
Metals are one of the most important types of the materials. They have
relatively high values of elastic constants and can be made strong by alloying and
proper heat treatment. What is particularly important they combine stiffness and
high strength with considerable ductility. As a result, most metals yield before
they fracture and this makes them the most widely used engineering material (see
Table I and for more information [1]).
X-ray diffraction analysis shows that the atoms in metals usually are arranged
in a regular, repeated three-dimensional pattern. The concept of the perfect
lattice of metal crystals can be used to explain the structure insensitive proper- .
ties which are listed in Table II (see also [2]). The values of structure insensitive
properties are primarily controlled by the strength of inter-atomic bonds, which
in turn are influenced by the arrangements of atoms in crystal lattices.
Real metals deviate from the perfect periodicity of their mathematical models
in the form of ideal crystals (see for example [3-6]).Various imperfections in

ideal arrangements of the atoms in crystalline materials have been divided, on the
basis of their dimension, into the following classes:
a) point — 0-dimensional,
b) linear — 1-dimensional,
c) surface — 2-dimensional,
d) volume — 3-dimensional. •
Examples of the defects falling in these respective categories (a) to (d) are given
in Table III. A number of useful properties of metals (see Table IV), are controlled
by these imperfections which in turn are determined by its processing route and
service conditions.
2. Quantitative characterization of structure
Images of the elements of microstructure of metals can be obtained with required
precision by modern imaging techniques such as transmission electron microscopy
(TEM), scanning electron microscopy (SEM), secondary-ion mass spectroscopy
(SIMS), X-ray photoelectron spectroscopy (XPS). Examples of applications
of these techniques are shown in Fig. 1. Modern computer aided methods
can be further used to obtain a quantitative description of the microstructure of
a studied material. These methods take advantage of the progress made in recent
years in the field of image processing, mathematical morphology, and quantitative
stereology.
Characterization of the microstructure of materials involves identification of
the main microstructural elements present and a quantitative description of their
sizes, shapes, numbers and positions within the specimen of the studied material.
In other words this is a process that answers the following questions [7]:
1. what are the elements in the internal structure of a given material that
distinguish it from other materials of that kind (of say similar chemical
composition);
2. where are these elements located and in what quantity;
3. what is their size and shape?
Answers to all these questions provide a comprehensive description of the material
microstructure that can be used to explain its properties and to gain better control
over its technological usage. However, the first of these questions is a domain of
material physics. In fact, microstructural elements such as dislocations and grain
Images of microstructural elements in metals: (a) dislocations and grain boundaries
imaged by transmission electron microscopy; (b) grain and inter-phase boundaries
imaged by light optical microscopy; (c) second phase particles revealed by scanning
electron microscopy.
boundaries by themselves have become subjects of extensive theoretical studies.
In the present text the focus is placed on the two last questions.
Size and shape are the attributes of 1-, 2- and 3-dimensional elements of
the microstructure. The size of microstructural elements can be quantitatively
described by:
• length, 1, for 1-dimensional elements,
• area, A, for 2-dimensional elements,
• volume, V, for 3-dimensional elements.
Direct measurements of the length, area, or volume of microstructural elements
are rarely possible. As a result, the measurements are usually done on sections
or projections of the microstructural elements. The results of such measurements
characterize the size of the elements studied but do not define their volume, area,
or length.

3. Stereology The elements of a microstructure extend into 3 dimensions and are

distributed over the volume of the specimen. This means that characterization of the
microstructural elements should be based on some 3-dimensional model for the
material studied. On the other hand, in an experimental approach they are commonly
studied on 2-dimensional cross-sections or via examination of thin slices. . On the
other hand, a large number of properties are related to microstructural elements
which are distributed over the volume of the material. In this situation, the required
3-dimensional description of the microstructure is inferred from the 2-dimensional
images by means of the methods of quantitative stereology.
Fig. 2. Examples illustrating applications of
stereological methods: (a) estimation of volume
fraction, Vv, for a two-phase system via
measurements of area fraction, AA; (b) estimation
of grain boundary surface are, Sv, via measurements
of the density of intersection points for a system of
test lines with traces ofgrain boundaries.

A quantitative description of the properties of two-

dimensional images of microstructures is a prerequisite
for the quantitative inference of the properties of
the 3-dimensional microstructure. In the past, a
number of methods have been developed that allow
an appropriate quantitative characterization of two-
dimensional images by means of simple counting
methods. Some details of such methods are given in
Fig. 2. In recent years significant progress has been
made in developing automatic computer-aided
procedures (see Fig. 3).
4. The influence of microstructure on the properties of materials
One of the foundations of modern materials science is the recognition of
the fact that the properties of materials are related to their microstructure. For
example, the properties of polycrystalline materials depend on the properties of
grain boundaries and grain interiors, and on the amount of grain boundaries in
a unit volume and, in turn, on the size of grains making up the polycrystalline
aggregate. The grain size can be described and measured in different ways. The
simplest procedure, frequently employed in physical metallurgy, is based on the
concept of intercept measurements, schematically explained in Fig. 3. A system of
test lines is superimposed on the image of the grain boundary network revealed on
a cross-section of the polycrystal. The mean intercept length, 1, is then calculated
from a simple formula

where LT is the total length of the test lines and Na — the total number of
their intersections with the traces of grain boundaries on the section. The value
of 1 decreases with decreasing grain . and the parameter is assumed to be
a desirable measure of the grain size and as such is correlated with the grain
boundary-sensitive properties of polycrystals.
Fig. 3. Live image of a section of polycrystalline metals (a); its binary form revealing
traces of grain boundaries (b) and histogram of grain section area A (c).

Resistance to plastic deformation is one of the grain size-sensitive properties

of polycrystals. One of the simple measures of this resistance is the hardness test.
Some hardness tests, such as Vickers and Brinell, yield a number that defines the .
resistance to plastic deformation in MPa. Figure 4 presents the results of Vickers
Fig. 4. The results of Vickers tests carried out on a series of specimens of a single-phase
fcc alloy with different grain sizes (for details see [10]).
tests carried out on a series of specimens of a single-phase fcc alloy with different
grain sizes. The hardness is plotted against 1 -1 / 2 in accordance with a linear
Relationship

This formula, known as the Hall—Petch relationship [8-10], is one of the most useful
in the field of physical metallurgy. It can be used to predict the properties of a
wide range of materials including Metals, ceramic, intermetallics and polymers.
5. Free surface effect

Reports recently published also show that properties of materials can be

significantly modified by the effect of free surface. Examples of such situations
include environmental effect on the mechanical properties of materials. Data for
an austenitic stainless steel is used to discuss contribution of the free surface to
the mechanical properties of metals.
In the temperature range from 200 to 700°C type 316 austenitic stainless
steels exhibit a weak or a negative dependence of the macroscopic flow stress at
a given total elongation, e, on the test temperature, Ta . In the literature such
properties are related to dynamic strain aging (DSA). Part of this temperature
interval is characterized by the presence of serrations on stress—strain curves. This
phenomenon is termed as the Portevin—le Chatelier effect (PLC).
The PLC effect is generally linked to the interactions of dislocations with
point defects. In the case of serrated flow in austenitic stainless steels interactions
of carbon and chromium atoms, which form chromium carbides, were suggested
to be relevant mechanism [11].
The environmental effect on the dynamic strain aging of 316 austenitic stainless
steels is exemplified by the date presented in Fig. 5. The specimens deformed
in the range of temperatures of the PLC effect in vacuum exhibit a lower frequency
of serrations, higher values of the flow stress and higher ultimate tensile strength
(UTS).
Fig. 5. The environmental effect on the dynamic strain aging of 316 austenitic stainless
steels observed in static tensile straining.
Fig. 6. Characteristic features of the scale formed as a result of annealing and straining
in air and vacuum.
Observations of the outer surface of strained specimens were carried out
using light and electron microscopy. Characteristic features of the scale formed
as a result of annealing and straining in air and vacuum are illustrated by the
micrographs shown in Fig. 6. Direct observations of the scale reveal a system
of ciacks nearly perpendicular to the axis of straining. The cracks, which are
observed on the surface of specimens tested in air, are deeper, more regular and.
• more clearly elongated. Observations of the sections of the oxide scale show that
it is characterized by relatively high roughness. The mean thickness of the scale
depends on the test environment and temperature. In the case of straining at
600°C in air it exceeds 0.01 mm while for straining in air the mean thickness is '
smaller than 0.005 mm.
More detailed examinations show that the oxide
scale contains a stack of
geometrically irregular layers containing a mixture
of FeO, NiO and Cr 2O3 .
Outer layers of the scale embody mainly nickel
oxides while the inner layers — chromium
oxides. This finding is confirmed by the variation
in the concentration of alloying.
elements along lines perpendicular to the surface
of the specimens (Fig. 7). There
is a strong concentration of Ni in the vicinity of
the outer surface and a peak of Cr
close to the oxide–substrate interface. The oxide
scales also exhibit an increased
concentration of Si with two peaks: one at the
outer surface and the other at the •
oxide–substrate interface.
Relatively high roughness and the presence of cracks characterize
microstructures
typical of the near-surface zone. Selective etching reveals also in this zone
the grain boundaries exhibiting a lower resistance to the corrosion. The
analysis of
the concentration of the alloying elements shows that this is related to
chromium
depletion. The results of X-ray microprobe mapping demonstrate that the
grain
boundaries in the substrate close to the oxide scale feature a measurable
depletion
of Cr, Mn, Si and enrichment of Ni. The concentration of Cr in the vicinity
of the
scale drops down to less than 8% — see Fig. 8.
Fig. 8. The concentration of Cr in the vicinity of the scale along the grain
boundaries
emerging on the surface of an annealed specimen of austenitic stainless steel.
The results obtained on 316 steel specimens annealed in vacuum and in air
can be viewed as composed of three major elements [12, 13]:
a) the core which is recrystallized fine-grained austenite with a low density of
carbides — this is by far the largest contributing volume and so it will dominate
the macroscopic features of the stress–strain curves at larger strains;
b) the near surface zone is associated with a region where both the microchemistry
and the microstructure are modified — perhaps a major factor here is
that some of the grain boundaries have become sensitized — which suggests
that the carbide populations and solute chemistry is altered;
c) the oxide scale consists of many layers of different oxides and it is associated
with variable degrees of surface roughness — which in turn will act as stress.
concentrators.
Loading the system to a given small strain may be assumed to strain the
cylindrical core, the near surface zone "tube" and the oxide scale tube to the
same
level. In the elastic range, the measured modulus will be based on some form
of
law of mixtures behavior. One may then assume that the oxide scale will
rupture
(in a brittle manner) presumably from the outside and these cracks will
propagate
towards the near-surface zone. It is then the constraints at the scale/near-
surface
zone interface, which will define how flow occurs and propagates into the
material.
Serrations on the stress–strain curve suggest that the flow is localized
(geometrically
representing something like a Luders band) and one may suggest that this
is because individual oxide cracks (possibly together with sensitized grain
boundaries
in the near-surface zone region) act as stress concentrators which encouraged
localized flow.