UNIT-II

Material Failure
Introduction
The failure of engineering materials is almost always an undesirable event for
several reasons; these include human lives that are put in jeopardy, economic losses,
and interference with the availability of products and services. Even though the
causes of failure and the behavior of materials may be known, prevention of failures
is difficult to guarantee. The usual causes are improper materials selection and
processing and inadequate design of the component or its misuse. Also, damage can
occur to structural parts during service, and regular inspection and repair or
replacement are critical to safe design. It is the responsibility of the engineer to
anticipate and plan for possible failure and, in the event that failure does occur, to
assess its cause and then take appropriate preventive measures against future
incidents.
The following topics are covered below: simple fracture (both ductile and brittle
modes), fundamentals of fracture mechanics, fracture toughness testing, the ductile-
to-brittle transition, fatigue, and creep. These discussions include failure
mechanisms, testing techniques, and methods by which failure may be prevented or
controlled. Fundamental of Fractures Simple fracture is the separation of a body into
two or more pieces in response to an imposed stress that is static (i.e., constant or
slowly changing with time) and at temperatures that are low relative to the melting
temperature of the material. Fracture can also occur from fatigue (when cyclic
stresses are imposed) and creep (time dependent deformation, normally at elevated
temperatures); the topics of fatigue and creep are covered later in this chapter
(Sections 8.7 through 8.15). Although applied stresses may be tensile, compressive,
shear, or torsional (or combinations of these), the present discussion will be confined
to fractures that result from uniaxial tensile loads. For metals, two fracture modes
are possible: ductile and brittle. Classification is based on the ability of a material
to experience plastic deformation.
Ductile metals typically exhibit substantial plastic deformation with high energy
absorption before fracture. On the other hand, there is normally little or no plastic
deformation with low energy absorption accompanying a brittle fracture. The tensile
stress–strain behaviors of both fracture types may be reviewed in Figure. Ductile and
brittle are relative terms; whether a particular fracture is one mode or the other
depends on the situation. Ductility may be quantified in terms of percent elongation
(Equation 6.11) and percent reduction in area Furthermore, ductility is a function of
temperature of the material, the strain rate, and the stress state.The disposition of
normally ductile materials to fail in a brittle manner. Any fracture process involves
two steps (i) crack formation and propagation (ii) in response to an imposed stress.
The mode of fracture is highly dependent on the mechanism of crack propagation.
Ductile fracture is characterized by extensive plastic deformation in the vicinity of
an advancing crack. Furthermore, the process proceeds relatively slowly as the crack
length is extended. Such a crack is often said to be stable. That is, it resists any
further extension unless there is an increase in the applied stress. In addition, there
will ordinarily be evidence of appreciable gross deformation at the fracture surfaces
(e.g., twisting and tearing). On the other hand, for brittle fracture, cracks may spread
extremely rapidly, with very little accompanying plastic deformation. Such cracks
may be said to be unstable, and crack propagation, once started, will continue
spontaneously without an increase in magnitude of the applied stress. Ductile
fracture is almost always preferred to brittle for two reasons. First, brittle fracture
occurs suddenly and catastrophically without any warning; this is a consequence of
the spontaneous and rapid crack propagation. On the other hand, for ductile fracture,
the presence of plastic deformation gives warning that failure is imminent, allowing
preventive measures to be taken. Second, more strain energy is required to induce
ductile fracture inasmuch as these materials are generally tougher. Under the action
of an applied tensile stress, many metal alloys are ductile, whereas ceramics are
typically brittle, and polymers may exhibit a range of behaviors.
Ductile Fracture
Ductile fracture surfaces will have their own distinctive features on both
macroscopic and microscopic levels. Figure 8.1 shows schematic representations for
two characteristic macroscopic fracture profiles.

Fig.1 (a) Highly ductile fracture in which the specimen necks down to a point.
2

b) Moderately ductile fracture after some necking. (c) Brittle fracture without any plastic deformation.
Page
The configuration shown in Fig. is found for extremely soft metals, such as pure
gold and lead at room temperature, and other metals, polymers, and inorganic glasses
at elevated temperatures. These highly ductile materials neck down to a point
fracture, showing virtually 100% reduction in area.
The most common type of tensile fracture profile for ductile metals is that
represented in Figure, where fracture is preceded by only a moderate amount of
necking.The fracture process normally occurs in several stages (Figure 8.2). First,
after necking begins, small cavities, or microvoids, form in the interior of the cross
section, as indicated in Figure 8.2b. Next, as deformation continues, these
microvoids enlarge, come together, and coalesce to form an elliptical crack, which
has its long axis perpendicular to the stress direction. The crack continues to grow
in a direction parallel to its major axis by this microvoid coalescence process (Figure
8.2c). Finally, fracture ensues by the rapid propagation of a crack around the outer
perimeter of the neck, by shear deformation at an angle of about 45 degree with the
tensile axis this is the angle at which the shear stress is a maximum. Sometimes a
fracture having this characteristic surface contour is termed a cup-and-cone fracture
because one of the mating surfaces is in the form of a cup, the other like a cone. In
this type of fractured specimen (Figure 8.3a), the central interior region of the
surface has an irregular and fibrous appearance, which is indicative of plastic
deformation.

Brittle Fracture
Brittle fracture takes place without any appreciable deformation and by rapid crack
propagation. The direction of crack motion is very nearly perpendicular to the
direction of the applied tensile stress and yields a relatively flat fracture surface, as
indicated in Figure. Fracture surfaces of materials that failed in a brittle manner will
have their own distinctive patterns; any signs of gross plastic deformation will be
absent. For example, in some steel pieces, a series of V-shaped “chevron” markings
may form near the center of the fracture cross section that point back toward the
crack initiation site (Figure 8.5a). Other brittle fracture surfaces contain lines or
ridges that radiate from the origin of the crack in a fanlike pattern. Often, marking
patterns will be sufficiently coarse to be discerned with the naked eye. For
very hard and fine-grained metals, there will be no discernible fracture pattern.
Brittle fracture in amorphous materials, such as ceramic glasses, yields a relatively
shiny and smooth surface.
For most brittle crystalline materials, crack propagation corresponds to the
successive and repeated breaking of atomic bonds along specific crystallographic
planes (Figure 8.6a); such a process is termed cleavage. This type of fracture is said
3

to be transgranular.
Page
 Fig. (a) V shape brittle fracture (b) fan shaped ridged brittle fracture

(or transcrystalline), because the fracture cracks pass through the grains.
Macroscopically, the fracture surface may have a grainy or faceted texture (Figure
8.3b), as a result of changes in orientation of the cleavage planes from grain to grain.
This cleavage feature is shown at a higher magnification in the scanning electron
micrograph of Figure.
In some alloys, crack propagation is along grain boundaries this fracture is termed
intergranular. Figure is a scanning electron micrograph showing a typical
intergranular fracture, in which the three-dimensional nature of the grains may be
4
Page
seen. This type of fracture normally results subsequent to the occurrence of
processes that weaken or embrittle grain boundary regions.

Priciples of Fracture Mechanism

Brittle fracture of normally ductile materials, such as that shown in the
chapteropening photograph (of the oil barge) for this chapter, has demonstrated the
need for a better understanding of the mechanisms of fracture. Extensive research
endeavors over the past century have led to the evolution of the field of fracture
mechanics. This subject allows quantification of the relationships between material
properties, stress level, the presence of crack-producing flaws, and crack
propagation mechanisms. Design engineers are now better equipped to anticipate,
and thus prevent, structural failures. The present discussion centers on some of the
fundamental principles of the mechanics of fracture.
Stress Concentration
The measured fracture strengths for most materials are significantly lower than those
predicted by theoretical calculations based on atomic bonding energies. This
discrepancy is explained by the presence of microscopic flaws or cracks that always
exist under normal conditions at the surface and within the interior of a body of
material. These flaws are a detriment to the fracture strength because an applied
stress may be amplified or concentrated at the tip, the magnitude of this amplification
depending on crack orientation and geometry. This phenomenon is demonstrated in
Figure 8.8, a stress profile across a cross section containing an internal crack. As
indicated by this profile, the magnitude of this localized stress diminishes with
distance away from the crack tip. At positions far removed, the stress is just the
nominal stress, or the applied load divided by the specimen cross-sectional area
(perpendicular to this load). Because of their ability to amplify an applied stress in
their locale, these flaws are sometimes called stress raisers.
If it is assumed that a crack is similar to an elliptical hole through a plate and is
oriented perpendicular to the applied stress, the maximum stress, occurs at the crack
tip and may be approximated by
5
Page
 Fig. the geometry of surfaces and internal cracks

Where σ0 is the magnitude of the nominal applied tensile stress, σt is the radius of
curvature of the crack tip, and a represents the length of a surface crack, or half of
the length of an internal crack. For a relatively long micro crack that has a small tip
radius of curvature, the factor (a/_t)1/2 may be very large. This will yield a value of
_m that is many times the value of σ0.
Sometimes the ratio _m/_0 is denoted as the stress concentration factor Kt:

which is simply a measure of the degree to which an external stress is amplified at

the tip of a crack. By way of comment, it should be said that stress amplification is
not restricted to these microscopic defects; it may occur at macroscopic internal
discontinuities (e.g., voids or inclusions), at sharp corners, scratches, and notches.
Furthermore, the effect of a stress raiser is more significant in brittle than in ductile
materials. For a ductile metal, plastic deformation ensues when the maximum stress
exceeds the yield strength. This leads to a more uniform distribution of stress in the
vicinity of the stress raiser and to the development of a maximum stress
concentration factor less than the theoretical value. Such yielding and stress
redistribution do not occur to any appreciable extent around flaws and
discontinuities in brittle materials; therefore, essentially the theoretical stress
6
Page

concentration will result.

Using principles of fracture mechanics, it is possible to show that the critical stress
σc required for crack propagation in a brittle material is described by the expression

All brittle materials contain a population of small cracks and flaws that have a variety
of sizes, geometries, and orientations. When the magnitude of a tensile stress at the
tip of one of these flaws exceeds the value of this critical stress, a crack forms and
then propagates, which results in fracture. Very small and virtually defect-free
metallic and ceramic whiskers have been grown with fracture strengths that
approach their theoretical values.

Tresca Criterion

All mechanics of material have sections on strength and failure criteria. Usually the
Mises and Tresca criteria are presented jointly with little discrimination or
recommendation between them. What is more, often little else in the way of failure
criteria are presented and the non-expert reader can be left with the impression that
these two criteria cover all or nearly all materials, and it doesn’t greatly matter which
one is used. Some books even grant these two criteria the status of being classical
results. A general question to be posed here is this, can either or both of these criteria
be considered to be classical results or are they merely long standing, comfortable
forms but of no special distinction.

This crterion has made several references to the Mises criterion while the Tresca
criterion has barely been mentioned. In this side section a careful look is taken of the
two criteria, mainly in comparison with each other to see if there should be some
inherent preference, even at this level. Concurrently, some observations will be
given of how each fits into the larger picture involving generality beyond just that
of the usual ductile metals.

The main interpretation of the Mises criterion is that it represents a critical value of
the distortional energy stored in the isotropic material while the Tresca criterion is
7

that of a critical value of the maximum shear stress in the isotropic material.
Page
Historically, the Tresca form was considered to be the more fundamental of the two,
but the Mises form was seen as an appealing, mathematically convenient
approximation to it. Now, both are usually stated side by side with little or no
preference.

The two criteria are specified below in principal stress space. Both are one parameter
forms, specified by either the uniaxial tensile strength, T, or the shear strength, S.

Mises Criterion, Critical Distortional Energy

where

Tresca Criterion, Critical Shear Stress

For the principal stresses ordered as σ1 ≥ σ2 ≥ σ3 then

For the principal stresses not ordered

8
Page
where

The three separate forms in (3) are for the maximum shear stresses in the three
principal planes.

Both of these single parameter criteria can be calibrated on either T or S. The figures
below, for biaxial stress states, calibrates both criteria on T and then on S.

Case I, Calibration on T

9
Page
Case II, Calibration on S

The maximum difference between the Mises and Tresca forms for both Cases I and
II is 14.4%. The essential and striking differences between the two are the corners
that occur in the Tresca form and their complete absence in the Mises form.

Two basic questions are immediately apparent:

10

Which procedure is better grounded, calibrating on T or calibrating on S?

Page
 Which criterion is more physically relevant, Mises or Tresca?

The steps of reasoning used here to answer these two questions are as follows:

1. Calibrating on S might seem more logical because yielding is caused by

distortional not dilatational states in ductile materials and distortion more
nearly and easily associates with shear stress rather than with uniaxial tensile
stress.
2. However, shear stress involves only one parameter and that does not naturally
generalize to a two parameter form which is necessary for non-perfectly
ductile materials. On the other hand, T=C naturally and automatically
generalizes to T and C (T≠C) in the more general cases. So T and C, not S, is
the natural combination that converges to T=C for a one parameter form. Case
II, calibrating on S, is accordingly rejected.
3. With calibration by T, giving Case I, there is still the choice between the Mises
and Tresca forms. Mises is smooth, while Tresca has corners. At the crystal
level (single grain) yielding does associate with dislocation movement on slip
planes. This is caused by shear stress on the slip system (resolved shear stress).
It would be tempting to say that this justifies and validates the Tresca criterion.
But that is only part of the “story”. The condition of isotropy implies and applies
to polycrystalline aggregates with the individual crystals taking all possible
orientations. Dislocation induced plastic flow occurs over many slip systems.
Furthermore, dislocation pile-up occurs at the grain boundaries. The vastly more
complex behavior at the aggregate level compared with the crystal level must
involve averaging over a wide variety of physical conditions and effects. This
averaging has a smoothing effect that is much more supportive of the smooth
Mises criterion than of the non-smooth Tresca form.

4. Conclusions 2 and 3 support and justify using the Mises (not Tresca) criterion
for very ductile materials and calibrating on T (not S).

Further relevant background considerations are as follows. Tresca did do testing of

metals that at the time seemed to support the maximum shear stress criterion. But
that testing was superseded by the later excruciatingly careful testing performed by
Taylor and Quinney and as shown in Section VI. The Taylor, Quinney results
support the Mises criterion.

All T=C materials are metallic, polycrystalline aggregates. It is necessary to do

orientation averaging to get the effective isotropic moduli properties. It is a parallel
11

and consistent circumstance that orientational averaging must also be done to get the
Page

strength for polycrystalline aggregates. Mises is consistent with this while Tresca is
not. Infact, Mises is a composition or a type of average of the three separate criteria
in, Tresca.

The non-smooth behavior evinced by the Tresca criterion usually associates with the
competition of failure modes such as with a ductile flow mode and a brittle fracture
mode. But those competitive effects are not present with ductile materials. Now
consider a true 2-D type of continuum to see how it behaves. This is not plane stress
or plane strain which are still 3-D behaviors. In this truly 2-D case it is found that a
maximum shear stress criterion (Tresca) and a maximum distortional energy
criterion (Mises) are identical, both giving smooth behaviors with continuous first
derivatives Then in going to 3-D the Mises form continues this smooth behavior but
the Tresca form brings in corners. The Tresca behavior in 3-D is an artifact of
describing the maximum shear stresses in the three principal coordinate planes. This
ignores the effects that occur at smaller scales in polycrystalline aggregates, and the
averaging necessary to reach macroscopic behavior.

Even though the maximum difference between the Mises and Tresca criteria is only
about 15% this difference represents a systemic error (divergence) on the part of the
Tresca criterion and it should not be used for any isotropic materials, even for ductile
metals. It is inappropriate to place the Tresca criterion on the same level as the Mises
criterion, as is done in most tutorial works. The Tresca criterion only merits a
historical reference. This is consistent with the fact that the Tresca criterion is the
limiting case of the two parameter Coulomb-Mohr criterion which itself is only of
historical interest. As discussed in Section VI the Mises criterion is the limiting case
of a viable, completely general, modern failure criterion. Despite its restrictions, the
Mises criterion is indeed a classical result.

Mohr coulomb law

When the soil sample has failed, the shear stress on the failure plane defines the
shear strength of the soil. Thus, it is necessary to identify the failure plane. Is it the
plane on which the maximum shear stress acts, or is it the plane where the ratio of
shear stress to normal stress is the maximum?

For the present, it can be assumed that a failure plane exists and it is possible to apply
principal stresses and measure them in the laboratory by conducting a triaxial test.
Then, the Mohr circle of stress at failure for the sample can be drawn using the
known values of the principal stresses.
12
Page
If data from several tests, carried out on different samples upto failure is available,
a series of Mohr circles can be plotted. It is convenient to show only the upper half
of the Mohr circle. A line tangential to the Mohr circles can be drawn, and is called
the Mohr-Coulomb failure envelope.

Fig. mohr circle

If the stress condition for any other soil sample is represented by a Mohr circle that
lies below the failure envelope, every plane within the sample experiences a shear
stress which is smaller than the shear strength of the sample. Thus, the point of
tangency of the envelope to the Mohr circle at failure gives a clue to the
determination of the inclination of the failure plane. The orientation of the failure
plane can be finally determined by the pole method.

The Mohr-Coulomb failure criterion can be written as the equation for the line that
represents the failure envelope. The general equation is
13
Page
Where =shear stress on the failure plane
c= apparent cohesion
= normal stress on the failure plane
f= angle of internal friction

The failure criterion can be expressed in terms of the relationship between the
principal stresses. From the geometry of the Mohr circle,

Rearranging,

where

Grifith Criterion
Griffith attributed the discrepancy between the observed fracture strength of crystals
and the theoretical cohesive strength to the presences of flaws in brittle materials.
This theory is applicable only to perfectly brittle material such as glass, and cannot
be used directly to metals. However, Griffith’s ideas formed a base to understand
the fracture in metals.
Griffith proposed that a brittle material contains a large number of fine cracks. He
postulated a criterion for the propagation of such a crack in a brittle material. During
propagation, there is a release of what is called the elastic strain energy, some of the
energy that is stored in the material as it is elastically deformed.
Further-more, during crack propagation, new free surfaces are created as the faces
of a crack. This requires energy to overcome the cohesive force of the atoms, that is,
it requires an increase in surface energy. Griffith developed a criterion for crack
propagation by performing an energy balance using these two energies as- a crack
propagates when the decrease in elastic strain energy is atleast equal to the energy
required to create the new crack surface. The thermodynamic relationship between
these two energies determines the magnitude of the tensile stress needed to propagate
14

a crack of a certain size to produce a brittle fracture.

Page
Consider a wide sheet of unit thickness of a brittle material. It has a lens-shaped
crack of 2 c length which runs from the front to the back face as illustrated in Fig.
15.12. When a longitudinal tensile stress, r is applied, the crack tends to increase its
length in the transverse direction (i.e., in a perpendicular direction to the tensile
stress).
When the crack spreads, the surface area of the crack increases, but the elastic train
energy stored in the thin sheet decreases because the elastic strains are not
continuous across the cracked region. The elastic strain energy released per unit of
plate thickness is given by-

where, E is Young’s modulus. The negative sign indicates the release of elastic strain
energy as the crack grows. As the crack propagates, two surfaces are created. If γ is
the surface energy per unit area of the material and as 2 c is the crack length, the
surface energy of the crack is

Us = 4 c γ
15

Griffith postulated that the crack propagates under a constant applied stress, σ, if an
incremental increase in crack length produces no change in the total energy of the
Page
system, that is, the Griffith condition for fracture is obtained if the rate at which
strain energy is released balances the rate at which energy is required to create the
new surfaces, i.e., the critical value is obtained by setting-

As the applied tensile stress is the external variable for a given material having a
crack of length 2c, it is more appropriate to express the critical condition as a critical
fracture stress, σc. Here, σc is the critical value of stress required for the propagation
of the flaw of length 2c. This is the Griffith equation, a corner stone of modern
fracture theory.
It is a go/no-go condition; that is, the flaw does not propagate until the critical value
of stress is reached. Once the critical stress is applied to a brittle material, the pre-
existing crack propagates spontaneously with a decrease in energy, or rather its rate
of growth accelerates, since the strain-energy released as it lengthens is increasingly
in excess of that required for the creation of new surfaces. This acceleration
continues until the crack acquires a terminal velocity of the order of one-half the
velocity of the longitudinal sound wave in the material.
The Griffith equation (15.17) produces an important result as it gives the value of
critical stress required to propagate a crack in a brittle material as a function of the
size of the micro-crack. It indicates that the fracture stress is inversely proportional
to the square root of the crack length. For example, if the crack length increases by
a factor of 9 decreases the fracture stress by one-third.
The SN Curve
As with other mechanical characteristics, the fatigue properties of materials can be
determined from laboratory simulation tests.7 A test apparatus should be designed
to duplicate as nearly as possible the service stress conditions (stress level, time
frequency, stress pattern, etc.). A schematic diagram of a rotating-bending test
16

apparatus, commonly used for fatigue testing, is shown in Figure 8.18; the
Page

compression and tensile stresses are imposed on the specimen as it is simultaneously

bent and rotated. Tests are also frequently conducted using an alternating uniaxial
tension–compression stress cycle.
A series of tests are commenced by subjecting a specimen to the stress cycling at a
relatively large maximum stress amplitude (_max), usually on the order of two-thirds
of the static tensile strength; the number of cycles to failure is counted. This
procedure is repeated on other specimens at progressively decreasing maximum
stress amplitudes. Data are plotted as stress S versus the logarithm of the number N
of cycles to failure for each of the specimens. The values of S are normally taken as
stress amplitudes values may be used.
Two distinct types of S–N behavior are observed, which are represented
schematically in Figure 8.19. As these plots indicate, the higher the magnitude of the
stress, the smaller the number of cycles the material is capable of sustaining before
failure. For some ferrous (iron base) and titanium alloys, the S–N curve becomes
horizontal at higher N values; or there is a limiting stress level, called the fatigue
limit (also sometimes the endurance limit), below which fatigue failure will not
occur. This fatigue limit represents the largest value of fluctuating stress that will
not cause failure for essentially an infinite number of cycles. For many steels, fatigue
limits range between 35% and 60% of the tensile strength.
Most nonferrous alloys (e.g., aluminum, copper, magnesium) do not have a fatigue
limit, in that the S–N curve continues its downward trend at increasingly greater N
values (Figure 8.19b).Thus, fatigue will ultimately occur regardless of the magnitude
of the stress. For these materials, the fatigue response is specified as fatigue
strength, which is defined as the stress level at which failure will occur for
some specified number of cycles (e.g., 107 cycles). The determination of fatigue
strength is also demonstrated in Figure

17

Fig. Stress Amplitude (S) vs number of cycle to fatigue failure (N) graph
Page
Another important parameter that characterizes a material’s fatigue behavior is
fatigue life Nf. It is the number of cycles to cause failure at a specified stress level,
as taken from the S–N plot.
Unfortunately, there always exists considerable scatter in fatigue data—that is, a
variation in the measured N value for a number of specimens tested at the same stress
level.This variation may lead to significant design uncertainties when fatigue life
and/or fatigue limit (or strength) are being considered. The scatter in results is a
consequence of the fatigue sensitivity to a number of test and material parameters
that are impossible to control precisely. These parameters include specimen
fabrication and surface preparation, metallurgical variables, specimen alignment in
the apparatus, mean stress, and test frequency.
Fatigue S–N curves similar to those shown in Figure 8.19 represent “best fit” curves
that have been drawn through average-value data points. It is a little unsettling to
realize that approximately one-half of the specimens tested actually failed at stress
levels lying nearly 25% below the curve (as determined on the basis of statistical
treatments).
18
Page