Lecture 07 PhysMet Akdogan Fracture in Metals & Alloys F19

Lecture 11
Fracture of
Engineering Metals & Alloys
“Metallurgical Aspects”
Chapter 8 - 1
Mechanical Failure
Some of the questions we will address...
 What governs fracture in engineering solids?
 What are the modes of fracture?
 How do the fracture resistances of the different
material classes compare?
 How do we estimate the stress to fracture?
 Why metals are so effective in resisting fracture?
 Importance of “impurities” in fracture?
 What are the crack propagation modes?
 Countermeasures against crack propagation.
Chapter 8 -
Examples of Mechanical
Mechanical FailureFailure
“Ship -cyclic loading from waves & stress-corrosion &”
ISSUES TO ADDRESS...
ductile-brittle transition @ low temperatures
 4700 ships were built by 1946.

 1250 of these had suffered brittle fractures by 1953.
 230 of these fractures were classed as serious.
 12 of the ships broke in two.
Adapted from chapter-opening photograph, Adapted from Fig. 22.30(b), Callister 7e.
Chapter 8, Callister & Rethwisch 8e. (by (Fig. 22.30(b) is courtesy of National
Neil Boenzi,
Adapted from chapter-opening
The New York photograph, Semiconductor
Times.) Chapter 8, Callister Corporation.)
& Rethwisch 8e. (by Neil Boenzi, The New York Times.)
Chapter 8 -
“Computer chip -cyclic thermal loading &”
ISSUES TOvacancy
ADDRESS...
clustering at junctions
• How do cracks that lead to failure form?
• How is fracture resistance quantified? How do the fracture
resistances
Fracture of the of the different material classes compare?
wiring is
• How
typically due
do we estimate the stress to fracture?
to • stress-free
How do loading rate, loading history, and temperature
strainaffect
arising
the failure behavior of materials?
from thermal
loading or
vacancy
clustering
which leads to
crack
nucleation
and growth.
Computer chip-cyclic
thermal loading.
Adapted from Fig. 22.30(b), Callister 7e.
(Fig. 22.30(b) is courtesy of National
Semiconductor
Adapted from Fig. 22.30(b), Callister 7e. (Fig. 22.30(b) is courtesy of NationalCorporation.)
Semiconductor Corporation.)
Chapter 8 - 4
4
“Hip implant -cyclic loading from walking &”
chemical attack by bodily fluids-stress corrosion cracking
• How do cracks that lead to failure form?
• How is fracture resistance quantified? How do the fracture
resistances of the different material classes compare?
• How do loading rate, loading history, and temperature

affect the failure behavior of materials?
Early implants were stainless

steel which corroded in the body
over time, ultimately fracturing
due to stress-corrosion cracking!
Hip implant-cyclic
loading from walking.
Adapted from Fig. 22.26(b),
Callister 7e.
Adapted from Fig. 22.26(b), Callister 7e.
Chapter 8 - 5
5
Examples of Mechanical Failure
“Airplane -cyclic loading leading to fatigue failure”
Airplanes need to be periodically check by Non-Destructive Testing
(NDT) to assess the size & distribution of cracks in all relevant parts of
the structure. If you skip the periodic testing and try pushing beyond
the predicted service life limits this is what will most likely happen!
 An airplane has millions of cracks when it rolls off of the assembly line.
The idea is to control the crack growth throughout the service life!
Chapter 8 -
I.
Fundamentals of Fracture
Mechanics
Chapter 8 - 7
Mechanics #1
theoretical Eγ s Eγ s
σ =σ max = σ Griffith =σ Fracture =
strength
o a
 maxE/10
o
 But in reality max>>>Fracture.
 Note that a>>>o.
max
 Fracture is governed by the
flaw size (a) and not the bond
length (o).
National Academyc of Sciences,
Materials Advisory Board’s report on
“Theoretical Strength of Materials? (1966)
Chapter 8 -
Mechanics #1 (Cont’d)

ao
f<E/10max
theoretical Eγ s
σ strength =σ max =
ao

The theoretical strength measures the strength of the bond
essentially, which, of course, is a very large value as it
represents what is holding the atoms together!!!
Chapter 8 -
Mechanics #1 (Cont’d)
 What really matters is the stress concentration at the tip of
the crack which may lead to fracture if it is set into motion…

Notch
Crack tip
The strain energy density @ the
crack tip is very high!
stress field
tableun_08_p276b
Chapter 8 -
Mechanics #2
yy>0
dc2 dc1 dc1 dc2 Incremental crack
propagation in
ductile solids
2c
 yy  d (1)
yy  d yy
(2)

Elliptical crack
yy>0  yy  d  (1)
yy 
yy  2  c  dc(1) 
 yy  d (1)
 yy  d (1)
yy   
 d (yy2)  2  c  dc(1)  dc(2)  etc ...
Cracks open & propagate

under tensile stresses!!!!!!
Chapter 8 -
Mechanics #3
Cracks tend to close &

NOT propagate under
compressive stresses!!!
 Design of built-in stresses in metals & alloys is a major method

of attack in preventing failure in all of its modalities (fracture,
creep rupture and fatigue failure); e.g. shot peening, nitriding,
carburization, ion exchange…
Chapter 8 - 12
Mechanics #4
 DaVinci (500 yrs ago!) observed
that the longer the wire, the
smaller the load for failure.
 Reasons:
Flaws cause premature failure.
 Larger samples contain longer
flaws!
 Flaws are randomly distributed in the
material. As such, failure has a
statistical nature.
Flaw density  Volume of Materials

 fracture  Volume of Materials 
1
Reprinted w/ permission from R.W. Hertzberg, "Deformation and Fracture Mechanics of Engineering Materials", (4th ed.) Fig. 7.4. John Wiley and Sons, Inc., 1996.
Chapter 8 -
Mechanics #5
Assume the rod and the Max. crack size:
rope are of the same is cmax=2rfiber
diameter and length. And
both steels have the same
composition &
microstructure…
Max. crack size:

2r
is cmax=2r
Courtesy of Aeromax Metals Courtesy of Redaelli Tecna
i. (r>>rfiber)[(cmax)fiber<< (cmax)rod]  Manipulation of probabilities

ii. Probability of finding a flaw in fiber via the specimen volume is
is << rod since Vfiber<<Vrod. another major line of attack
iii. Probability of “n” fibers failing due in preventing fracture.
to cmax at the same time
n
p
i 1
i  p1  p2  p3  ........  p( n  2)  p( n1)  pn
Chapter 8 -
Mechanics #6
 Avoid sharp corners!...they are stress concentrators.
σmax σ0
σ0
2.5 w
σmax
r, h
2.0 fillet
radius
increasing w/h
1.5
1.0 r/h
0 0.5 1.0
Adapted from Fig. 8.2W(c), Callister 6e.
sharper fillet radius
Chapter 8 - 15
Mechanics #7
“The Weakest Link (Brittle Failure)”
Prof. Waloddi Weibull, KTH Sweden (1887-1979)
 Weibull’s weakest link model

assumes global failure as soon as
some stress based failure criterion is
fulfilled for one structural element
or in one point.
 Very important for brittle fracture.
A good approach is to distribute the
risk as shown on the left 
Courtesy of P. J. Gustafsson; Lund University, Sweden
Chapter 8 -
II.
Fracture Modes & Their
Morphology
“A Fractographic Survey”
Chapter 8 - 17
Modes of Fracture
(based on macroscopic observations)
Ductile fracture
– Accompanied by significant plastic
deformation
 Brittle fracture
– Little or no plastic deformation
– Catastrophic
Chapter 8 -
Ductile vs. Brittle Fracture
Fracture Extremely
behavior Ductile Brittle
Ductile
Ductile:
Warning before
fracture
Brittle:
No warning!
Plastic def’n
Large Moderate Small
Prior to fracture
 Brittle fracture MUST be prevented at all cost!

 Ductile fracture is not desired either. But it is the preferred mode
of failure. Hence, proper implementation of materials with
optimum strength and toughness is necessary.
Adapted from Fig. 8.1, Callister & Rethwisch 8e.
Chapter 8 -
Example: Pipe Failures
 Ductile failure:
- one piece
- large deformation
- you see it coming!
 Brittle failure:
- many pieces
- small deformat’ns
-catastrophic!!!
Figures from V.J. Colangelo and F.A. Heiser, Analysis of Metallurgical Failures (2nd ed.), Fig. 4.1(a) and (b), p. 66 John Wiley and Sons, Inc., 1987. Used with permission.
Chapter 8 -
Ductile Failure Mediated by
dislocations!
Necking 1
Cavity formation 2
Cavity coalescence 3
(crack forms)
Crack grows 90o applied
Stress
Necking
CUP
Fracture 5
CONE Strain
max Crack propagation 4

@ 45o
Chapter 8 -
Ductile Failure
Nucleation of Fracture surface
voids in the neck has the cup-&-cone
@ inclusion sites! geometry
Inclusions from Fracture surface of tire cord wire loaded in tension

processing
50 mm 100 mm
P. Thornton, J. Mater. Sci., Vol. 6, 1971) Courtesy of F. Roehrig, CC Technologies, Dublin, OH.
Chapter 8 -
Ductile Fracture Video 01
Chapter 8 -
Courtesy of J.D. Begg

Chapter 8 -
Courtesy of Shimadzu Corp.

Chapter 8 -
Ductile vs. Brittle Failure
Cup & Cone Well defined, flat
fracture surface fracture surface
Ductile Fracture Brittle Fracture

Chapter 8 - 26
Brittle Fracture
Arrows indicate point at which failure originated
Cleavage planes
Adapted from Fig. 8.5(a), Callister & Rethwisch 8e.

Chapter 8 - 27
Brittle Fracture
Initiation of the crack
Fan-shaped ridges
Chapter 8 -
Brittle Fracture Video 01
Chapter 8 -
Brittle Fracture Video 02
Chapter 8 -
Effect of Grain Boundaries
 Ductile Cast IronTransgranular FractureGrain
Boundaries have high enough strength+toughness!
Chapter 8 -
Effect of Grain Boundaries
 Intergranular FractureGrain Boundaries Weak (brittle),
e.g. Cementite in hypereutectoid steels!
Crack propagating along GBs
Avoid at all cost!!!

Chapter 8 -
Inter- vs. Transgranular Fracture
Intergranular Transgranular
cracks cracks
4 mm 160 mm
J.R. Keiser & A.R. Olsen, ORNL, in "Metals Handbook", 9th Ed., ASM International (1985). D.R. Diercks, ANL, in Metals Handbook", 9th Ed., ASM International (1985).
304 Stainless Steel 316 Stainless Steel
0.08% C, 17.5-20% Cr, 8-11% Ni, 0.03% C, 16-18.5% Cr, 10-14% Ni,
2% Mn, <1% Si, <0.045% P, <0.03% S 2-3% Mo, 2% Mn, <1% Si, <0.045% P,
<0.03% S
Chapter 8 -
Methods to Measure Crack Size
Extensively used in metallurgical assessments in the

automotive, naval and aerospace industries!
Chapter 8 -
III.
The Stress Concentrator
Concept
&
Fracture Toughness
“Definition of Terms”
Chapter 8 - 35
Concentration of Stress at Crack Tip
“Definition of Terms”
2a<<<L a<<<L
Volume crack Surface crack
Chapter 8 -
Flaws are Stress Concentrators!
o
 The Griffith crack:
a
 m  2 o
t
a
Kt  2
t
where
t = radius of curvature
o = applied stress
m = stress at crack tip
o Kt= stress concentrator
Adapted from Fig. 8.8(a), Callister & Rethwisch 8e.
Chapter 8 - 37
Flaws are Stress Concentrators!
a
 m  2 o m a
t 2
o t
a
Kt  2
t a  t
 m   o
Even if o<ys,
it is possible that
m>UTS because
Kt>>1 as a>>t
Chapter 8 - 38
Fracture Toughness of a Materials
Material property like elastic
modulus or Poisson’s ratio
Fracture
toughness K c  Y c a
Typically Y~1 (depends on geometry of crack and specimen)

Chapter 8 - 39
Mode I Mode II Mode III
K Ic  Y c a
Chapter 8 -
Fracture Toughness of Various Materials
Chapter 8 -
K Ic  Y c a
“The Geometric Factor Y”
Y will be given
in a problem
tableun_08_p276b
Chapter 8 -
Example 1: Crack Size Computation
Chapter 8 -
Example 2: Designing against
Fracture
Chapter 8 - 44
Fracture
Chapter 8 - 45
Fracture
Chapter 8 - 46
Fracture
Chapter 8 - 47
Fracture
Chapter 8 - 48
Example 3: Design Against Crack Growth
 Crack growth condition:
K ≥ Kc = Y a
 Largest, most highly stressed cracks grow first!
 Scenario 1: Max. flaw  Scenario 2: Design
size dictates design stress dictates max. flaw
stress. size.
2
design 
Kc 1  K c 

amax 
Y amax   Ydesign 

 amax
fracture fracture
no no
fracture amax fracture

Chapter 8 - 49
Example 4: Design of Aircraft Wing
 Material has KIc = 26 MPa-m0.5
 Two designs to consider...
Design A Design B
--largest flaw is 9 mm --use same material
--failure stress = 112 MPa --largest flaw is 4 mm
K Ic --failure stress = ?
 Use... c 
Y amax
 Key point: Y and KIc are the same for both designs.
KIc
=  a = constant
Y 
Result:
112 MPa 9 mm 4 mm
 c amax   
A c amax  B
Answer: ( c )B  168 MPa
Chapter 8 -
Chapter 8 - 51
IV.
Griffith’s Theory of Brittle
Fracture
&
Orowan-Irwin Modification
for Ductile Fracture
&
The Joffe Effect
Chapter 8 - 52
Griffith Theory of Brittle Fracture
 The theoretical cohesive strength (~chemical bond energy) is
unrealistically higher than the experimentally measured
strength of materials.
 The discrepancy between cohesive and fracture strength is

attributed to internal flaws or defects in the material which
reduce the fracture strength in materials.
 Griffith proposed, which is based on the concept of strain

and surface energy balance, to explain the said discrepancy.
Chapter 8 -
(Cont’d)
 A decrease in strain energy results from the formation of a crack.
The elastic strain energy per unit of plate thickness is equal to
a 2  2
Ue  - 0
E
where  is the tensile strength acting normal to the crack of length
2a. The negative sign is used because crack growth cause elastic
strain energy release.
 The surface energy due to the presence of the crack is
US  4a s  0
where the c> and S>0 and, hence, Us>0. That is so because energy
is needed to create new surface, i.e. to grow the crack.
Chapter 8 -
(Cont’d)
 The total change in the energy U resulting from the creation of a
crack is given below.
 According to Griffith, the crack will propagate under a constant
stress  if an incremental increase in crack length produces no
change in the total energy of the system, i.e. the increase in surface
energy is compensated by a decrease in elastic strain energy.
U(a,  )  U e  US
a 2  2
U(a,  )  4a s -
E
If U(a,  )  0  crack growth
Chapter 8 -
(Cont’d)
 The stress ()-crack size (a) relation is obtained by evaluating the
first variation of U with c as follows:
U(a,  )   a 2  2 
0  4a s - 0
a a  E 
and solving for a, and rearranging, leads to the Griffith equation:
2E S
c  Plane stress
a
 If applied>c then the crack will grow in an unstable fashion
leading to fracture. Likewise, any crack with a>a* will grow
unstably and result in fracture. Here, c and a* are the critical stress
and crack size, respectively.
Chapter 8 -
(Cont’d)
The Griffith Equation
Plane Stress Plane Strain

(zz=0) (zz=0)
2E S 2E S
c  c 
a (1   2 )a
Chapter 8 -
(Cont’d)
1
c   E  S Intrinsic/Extrinsic
a property
Extrinsic property Intrinsic property
 As shown above, the critical stress c is very sensitive

to crack size a* due to the inverse square root
relationship. The square root product (ES)1/2 has less
of an impact unless one deal with a ductile metal ().
 The fracture strength of the material is ~ entirely

controlled by the flaw (crack) size.
Chapter 8 -
(Cont’d)
US  4a s
200
Equilibrium
100 Energy barrier (+)

U x 10-3 (Joules)
Plane stress in
tension (zz=0)
o 10 30
a (mm)
20
-100
U  U e  U S
a  2 2
Ue = Driving Force (-)
E
-200
Chapter 8 -
Orowan-Irwin Modification of Griffith’s
Theory of Brittle Fracture
 Cracks having sharp tips propagate easier than cracks having
blunt tips.
 A plastic material deforms at a crack tip, which “blunts” the
crack.
No plastic zone,
Crack No blunting! 
t
a a
Ductile Solids Brittle Solids

(Metals & Alloys) (Glass & Ceramics)
Chapter 8 -
Orowan-Irwin Modification of Griffith’s
Theory of Brittle Fracture (Cont’d)
 The Griffith eqn. is strongly dependent on crack size a, and strictly
satisfies only ideally brittle materials such as glass.
 However, metals and alloys exhibit large plastic deformation prior
to fracture, as a result of which the fracture stress is increased due
to crack tip blunting.
 As proposed by Orowan & Irwin, the surface term in Griffith’s eqn.
needs to be augmented by the surface energy of the dislocation
(a.k.a. plastic work). Accordingly, the modified Griffith eqn. reads
c 

2E  S   p   
2E p
 c
c c
as  S   p , where
 S  1 J / m 2 and  p  100  1000 J/m 2
Chapter 8 -
The Joffe Effect
“Effect of Surface Energy on Fracture”
 The importance of surface energy has been
demonstrated by carrying out experiment in solutions
of surface active materials. The sensitivity of fracture
stress in (brittle) materials to surface conditions is
called the Joffe effect.
 For instance, addition of 0.5% Sb to Cu reduces the
surface energy from 1.80 to 1.00 J/m2. Since solute
concentration builds up in grain boundaries,
intergranular embrittlement is a major concern in
engineering metals and alloys.
If  S =0.8 J/m 2 then  c 

by 26% where  c   S !!!
Chapter 8 -
The Joffe Effect (Cont’d)
“Effect of Surface Energy on Fracture”
Diffusion Flux
Surface of
PolyXtal Solid
Grain
Grain #01 Grain #02 Boundary
Diffusion
Grain Bulk
Boundary Diffusion
% Conc’n  c   S !!!
Grain
Boundary Bulk
Concentration
Segregation The “wrong” species
will  S drastically,
x decreasing F thereby;
e.g. Sulfur in Steel.
Chapter 8 -
A Historical Note
“HOT Shortness (Sulphur)”
 Iron & Steel When heated to > 460 oC), it glows in red color.
Normally, iron or steel at and above this temperature
becomes increasingly malleable and plastic.
 If the Sulphur%>>0.05wt%, then iron or steel becomes
crumbly and brittle.
 Sulphur forms FeS which is a low melting point grain
boundary phase causing grain separation.
 Today’s steels have <<0.03 wt% S and in addition Mn is
added to form MnS which has a higher melting temperature
and it does not form a grain boundary phase.
 Hence, hot working of Mn-alloyed low Sulphur steel is no
problem anymore. It used to be called Krupp krankheit as
Krupp steel used to exhibit hot brittleness due to poor
sulphur control in steel back then.
tableun_08_p276b
Chapter 8 -
“Effect of Sulfur on Steel”
 A EXTREMELY harmful element. The S concentration is
typically kept below 0.015% but it may be as high as 0.6%
in cheap steel.
 S forms FeS which is a very brittle substance. This forms as
a thin film separating the Pearlite or Ferrite grains, thus
greatly reducing the strength of the steel.
 MnS is stronger than FeS and provided that there is enough
Mn present the S will all combine with it in preference to
the Fe.
 It is common practice to add eight times the Mn required to
combine with the S.
 FeS melts below the working temperature of Iron.
Keep the Grain Boundaries clean!

Chapter 8 -
“Hydrogen Embrittlement”
What is it really?
 Hydrogen embrittlement is a metal’s loss of ductility
and reduction of load bearing capability due to the
absorption of hydrogen atoms or molecules by the
metal. The result of hydrogen embrittlement is that
components crack and fracture at stresses less than the
yield strength of the metal.
seblog.strongtie.com
© CEphoto, Uwe Aranas
tableun_08_p276b
Chapter 8 -
X 2 (g)  2H+ (%)......Go (T)  Sievert’s law states that the
a H2 2 amount of a diatomic gas is
K=  éêë H+ ùúû = KPH2 dissolved in a metal is
a H2
proportional to the square
æ -  Go ö root of its pressure.
K = exp çç ÷÷
çè 2RT ÷ø
 Dissolved H+ diffuses very

Sievert 's Law: fast in the metal and forms
æ -  Go ö grain boundary phases
éH
êë
+ ù = PH
úû exp çç ÷÷
2 çè 2RT ÷ø which trigger the Joffe
effect! Etc etc…
tableun_08_p276b
Chapter 8 -
H+ is very small…jump frequency 2-3 orders of magnitude larger than

anything else that diffuses in a metallic lattice. Hence, it can be anywhere if
introduced into the metal! Consider hydrogen damage as the cancer of the
Inorganic world. It damages virtually every feature of the metal over 6 orders
of magnitude in length 1 nm – 1mm as depicted in the schematic above!
tableun_08_p276b
Chapter 8 -
tableun_08_p276b
Chapter 8 -
“Hydrogen Embrittlement & Blister Formation”
2H++2e-=H2(g)
tableun_08_p276b
Chapter 8 -
Countermeasures Against Hydrogen
Embrittlement
tableun_08_p276b
Chapter 8 -
V.
Theories of Crack Nucleation
(Zener, Stroh, Smith)
&
Modes of Crack Propagation
Chapter 8 - 72
Zener’s Theory of Crack Nucleation @ Pile-ups
“Filling the gaps of Griffith’s Theory”
Barrier (e.g. GB)  The Griffith theory only shows
the stress required for crack
propagation of an existing crack
but does not address crack
nucleation.
 Zener & Stroh showed that crack
2c
nucleation occurs when the shear
stress S caused by n dislocation
in a pile-up at a grain boundary
reaches the following value (here
i is the lattice friction stress in
the slip plane):
 Per Zener, the dislocations that are
piled-up against the GB under the  2 
action of the shear force will  S   i   S 
coalesce and form the nucleus of a nb 
crack when the said shear stress  
reaches S. Chapter 8 -
Stroh’s Theory of Crack Nucleation @ Pile-ups
“Filling the gaps of Griffith’s Theory”
Dislocation pile-up E

 eff   y   i
4  1  2  d
forming a microcrack
Grain  eff : Effective shear stress

Boundaries
 y : Shear yield stress
d/2  i : Lattice friction
d : grain size
 Stroh included the effect of the grain size d in a model, in which he suggested
that the shear stress created by the dislocation pile-up of length d/2 nucleates
a microcrack.
 Per Stroh’s model, fracture depends on the shear stress acting on the slip
plane.
Chapter 8 -
Smith Model for GB Microcrack Form’n
“Role of Brittle GB Phases such as Fe3C in Steel”
 Smith proposed a model which includes
the role of brittle second phases in the
nucleation of cracks via microcracking at
the GB carbide.
 Per the Smith model, a microcrack is
initiated when sufficiently high applied
stress causes local plastic strain within
the ferrite grains to nucleate a microcrack
in the brittle grain boundary phase (Fe3C)
of thickness Co in hypereutectoid steels.
 Further propagation through the carbide
phase follows the Griffith’s theory.
2
 co  2  4 Co  i   4 E p
     eff
2
1  
      D
   
f 2
 D D  eff   1
Chapter 8 -
Effect of 2nd Phase on Tensile Ductility
2  2nd-phase particle, which
are readily cut by
True strain (%) @ fracture
80 dislocations produce
Reduction of area (%)

Spherical planar slips, leading to
carbide
70 large pile-ups. This
1
promotes brittle fracture.
Elongated 60
Sulphides Pearlitic
carbide 50  2nd-phase particle, which
are impenetrable by
Plate 30 dislocations, greatly
Sulphides 10
reduce the slip distance.
0 5 10
Hence, the number is
Volume of second phase (%) dislocations in pile-up ,
contributing to toughness.
 Small spherical 2nd-phase particles (<1 m) are more resistant to cracking.
 A soft and ductile phase is also beneficial to impart ductility to a brittle
matrix.
Chapter 8 -
Crack Propagation Modes (Cont’d)
(1) The Go-No-Go ModeBrittle materials
 If =F then a
a unstable crack
 propagation occurs,
i.e. fracture occurs
Unstable without  in stress.
Crack
growth
 This mode is
ao characteristic of brittle
No-Go Go fracture as typically
(<F) (>F) seen in ceramics and
glass, i.e. low
F  toughness materials.
Chapter 8 -
(2) The Slow-Growth ModeDuctile/Tough Materials
a  From AB, c with

Controlled by crack ing . As  F , ao
tip plasticity  aF . Crack propagation
Unstable occurs with increasing
crack growth
aF B stress which delays
fracture (a).
Stable
crack growth
 This mode is
A
ao characteristic of ductile
No-Go Slow-Go Go materials exhibiting
(<F) (ys<<F) (>F)
high toughness such as
ys F  engineering metals &
alloys
Chapter 8 -
(3) Static Growth ModeEnvironmental Attack
a  Here <F &  is constant.

 Yet, the crack grows in time
Go Unstable
(acF)
crack slowly. When aaF, then
growth
aF a.
Slow  In this mode, the crack does
unstable
a=f(t) crack
not grow under the action of
growth the applied stress. Instead,
ao a as time  due to external
factors such as chemical
No-Go attack (corrosion) at the
(<F) crack tip by, for instance,
 salt water.
<F F
Chapter 8 -
Effect of Hydrostatic Pressure on
Fracture
 If the hydrostatic component of a triaxial state of stress is
compressive, then it is very helpful in resisting fracture by inducing
crack closure, thereby contributing to toughness. In addition, such
compressive hydrostatic components contribute to the ductility of
the metal/alloys in many metal shaping processes such as wire
drawing and extrusion.
 In such processes, very large plastic deformations are obtained
which would not be possible in the absence of large hydrostatic
compressive component of the state of stress imposed on the
metal/alloy.
Chapter 8 -
Fracture
 Since a hydrostatic component of stress exerts no shear stress, it
cannot increase the number of dislocations in a pile-up or squeeze
them closer together. This is the very reason why hydrostatic stress
affect crack propagation but not crack initiation.
 Compressive stresses act to close up small pores or separations at
phase boundaries, and generally impede crack propagation.
 In steel, hydrostatic pressure (compressive hydrostatic stress) does
not prevent the fracture of cementites (and other carbides if
present) but it does reduce void/crack growth in the ferrite matrix
which contributes to the steel’s toughness.
Pearlite
Hypoeutectoid (ferrite+cementite)
Steel
Proeutectoid
ferrite
Chapter 8 -
Strain @ fracture Fracture
Normal Ductile Solid

Brittle solids are
affected the most by
hydrostatic stress
Low Ductility Solids
because such stress is
the only mechanism
Brittle Solids impeding crack
propagation
Hydrostatic stress
Chapter 8 -
VI.
Crack Tip Plasticity
in
Ductile Metals & Alloys
&
Toughness
Chapter 8 - 83
Ductile Fracture
“A Microplasticity Phenomena in Metals”
Chapter 8 -
Stress Concentration at Crack Tip
“Localization of Elastic Strain Energy”
if r<<c
Near Field
Stress
intensity
Far field factor
Chapter 8 -
A Microplasticity at the Crack Tip
“The Rough Model”
Cracks in soft metals have large

plastic zones; those in ceramics
and glasses have small zones
or none at all.
fig_08_16
Chapter 8 -
Crack Lengths Prior to Fracture
Chapter 8 -
Plastic Zone at the Crack Tip
“A Much Closer Look”
Crack
62ys
Chapter 8 -
Crack Tip Plasticity
Chapter 8 -
Crack Lengths Prior to Fracture
Chapter 8 -
The Stress Field at the Crack Tip (Irwin Model)
Chapter 8 -
Plastic Zone at the Crack Tip (Cont’d)
Chapter 8 -
How does the Plastic Deformation
Occur at the Crack Tip?
Chapter 8 -
“A Closer Look”
fig_08_16
Chapter 8 -
Plastic Zone at the Crack Tip (Cont’d)
fig_08_16
Chapter 8 -
!
g
n n
o d i
i
st m
Chapter 8 -
yc r
a o
> re w c

yr o r
t
o t o
s c p h

e n e s
r i t
o w s s
i
n Ir a th
g e
i s g
l h i i
e t n
d r pt
c
o r p2 e
<
m r r
r

t o
a r pc
(Cont’d)
rp
Sample Problem Plastic Zone at Crack Tip
Chapter 8 -
VII.
Ductile to Brittle Transition (DBT)
in
Steel
a.ka.
“Cold Shortness”
Chapter 8 - 98
A Historical Note #1
“COLD Shortness”
 Cold shortness was a big problem for a long
time in sword making. It means that things
get brittle when it's cold. The word "short" is
old English, meaning "having a tendency to
break or crumble" as in shortcake or
shortening. It doesn't mean that things get
smaller when it's cold.
 Nowadays we call the more general effect
where properties change from ductile to
brittle with decreasing temperature the
"ductile to brittle transition" or DBT.
tableun_08_p276b
Chapter 8 -
Ductile to Brittle Transition (DBT) in
Ferrous Alloys
 Pre-WWII: The Titanic  WWII: Liberty Ships
Reprinted w/ permission from R.W. Hertzberg, Reprinted w/ permission from R.W. Hertzberg,
"Deformation and Fracture Mechanics of Engineering "Deformation and Fracture Mechanics of Engineering
Materials", (4th ed.) Fig. 7.1(a), p. 262, John Wiley and Materials", (4th ed.) Fig. 7.1(b), p. 262, John Wiley and
Sons, Inc., 1996. (Orig. source: Dr. Robert D. Ballard, Sons, Inc., 1996. (Orig. source: Earl R. Parker,
The Discovery of the Titanic.) "Behavior of Engineering Structures", Nat. Acad. Sci.,
Nat. Res. Council, John Wiley and Sons, Inc., NY,
1957.)
 Problem: Steels were used with DBT temperature just

below room temperature or even higher!!!
Chapter 8 -
Ductile-to-Brittle Transition (DBT)
in
Ferrous Alloys
 The steel used for the Titanic hull was not suited for service
at low temperatures as shown in the table given above.
History Titanic Steel Modern Mild Steel

101 Longitudinal Transverse Longitudinal Transverse
Impact
Energy @ 4J 4J 325 J 100 J
-2 oC
DBT
30 oC -42 oC -42 oC -42 oC
Temperature
H.P. Leighly, B.L. Bramfit, and S.J. Lawrence in “Practical Failure Analysis”.
tableun_08_p276b
Chapter 8 -
Impact Testing for DBT Assessment
(Charpy)
Charpy
 Impact loading:
-- severe testing case
-- makes material more brittle
-- decreases toughness
Izod
Adapted from Fig. 8.12(b),

Callister & Rethwisch 8e. (Fig.
8.12(b) is adapted from H.W.
Hayden, W.G. Moffatt, and J.
Wulff, The Structure and final height initial height
Properties of Materials, Vol. III,
Mechanical Behavior, John Wiley
and Sons, Inc. (1965) p. 13.)
Chapter 8 -
Impact Testing for DBT Assessment
Chapter 8 -
DBT in Ferrous Alloys
“Effect of temperature of fracture surface”
Flat surfaces Cup & Cone surfaces
Chapter 8 - 104
Ductile-to-Brittle (DBT)Transition in
Metals & Alloys
 Influence of Temperature on Impact Energy
FCC metals (e.g., Austenite, Cu, Ni, most HCP metals)

Impact Energy
BCC metals (e.g., iron at T < 914ºC)
Brittle More Ductile
High strength steels,

aluminum alloys, titanium
Temperature
Ductile-to-brittle
transition temperature
Chapter 8 -
Influence of Carbon Content in
Steel on DBT
The lower the carbon

concentration, the lower
the DBT temperature!
Mild steel
(Low C-Steel)
Chapter 8 -
Metallurgical Factors Affecting DBT
“Countermeasures against DBT”
 For steels…as %C  y, TS, Hardness, Ductility and
TDBT 
- Counter by adding Manganese
- (Mn : C) should be 3:1
 Phosphorous increases TDBT …keep %P<0.05 wt%
 As % oxygen in steel  the TDBT :

- Use semi-killed by adding Si
- Use fully-killed by adding Si +Al
to remove oxygen;
-Vanadium and Niobium also effective de-oxidizers
 As grain size  toughness ……….. and TDBT.

- Alloying with Niobium and Vanadium keeps
grain size small (GB pinners!)
tableun_08_p276b
Chapter 8 -
By the way…what is a Killed Steel?
“No oxygen bubbles”
 A killed steel has no expulsion of oxygen in the form of
bubbles during ingot casting!
Typical ingot structures observed in industry…

tableun_08_p276b
Chapter 8 -
By the way…what is a Killed Steel?
“Importance of Ingot Mold Design”
https://pmpaspeakingofprecision.com/2010/07/20/killed-steel/ tableun_08_p276b
Chapter 8 -
How to produce Killed-Steel
“Ladle Metallurgy 101”
Chapter 8 -
How to produce Killed-Steel (Cont’d)
 React dissolved oxygen with selected alloying

elements in the ladle.
The resultant oxide will diffuse to the slag due to its
low density in metal oxide form.
Chapter 8 -
How to produce Killed-Steel
Slag-liquid metal
reactions in the ladle
will determine the final
composition of the steel
and hence its DBT
characteristics!
Chapter 8 -
“Effect of Texture on TDBT”
 Rolled & forged

products have varying
impact behavior due to
grain orientation (texture).
 Note that the difference

is not as large at lower
temperature.
tableun_08_p276b
Chapter 8 -
Theory of DBT in Metals & Alloys
“Cottrell Theory”
 In DBT there are two important material parameters that need to be
considered, i.e. the yield stress at which plastic deformation starts and
the fracture stress à la Griffith’s which designates the stress necessary
to induce unstable crack growth in a brittle medium. Per Cottrell, DBT
is the temperature (to be determined experimentally) at which the
following condition is met
( i )
D + k ' k ' ³ G S
where,
i : Lattice resistance to dislocation movement
D : Grain diameter governing slip length
Think of
k’ : Parameter related to dislocation pile-up
Peierls-Nabarro!!! G : Shear modulus
S: Surface energy
 : A constant related to the stress system
Chapter 8 -
 Per Cottrel’s eqn., the DBT temperature is high in materials with
large grain size, high lattice friction stress i (think Peierls-
Nabarro!) and high k’.
 The lattice friction stress i in BCC materials is strongly
dependent on Temperature.
 Materials such as Fe and Mild steel
Mo , which have high k’, -196 oC
are more susceptible to
Stress (GPa)
brittle fracture upon DBT.
 Smaller grain size favors
lower DBT temperature
and hence, suppresses
DBT.
 In mild steel, Ni lowers
DBT temperature, while C,
(Grain diameter)1/2
P, N, S and Mo increase
the DBT temperature
Chapter 8 -
Influence of Carbon & Phosphor on
DBT in Steel
Keep %P<0.3% Low C steel is better !

tableun_08_p276b
Chapter 8 -
Influence of Carbon & Phosphor on
DBT in Steel
Keep %S<0.005% Mn is always good!

Keep %Mn>2%
tableun_08_p276b
Chapter 8 -
Keep the Steel clean against DBT!
Keep it clean…
Literally! Especially the
Grain boundaries need to
be kept clean; i.e.
segregation of impurities
needs to be kept under
check at all times!
tableun_08_p276b
Chapter 8 -
VIII.
Designing Compressive
Stresses
Chapter 8 - 119
Some Methods for Designing
Compressive Surface Stresses
1. Shot PeeningCold work, plastic def’n of surface &
Subsurface; cheap, fast and very effective.
2. Carburization of Steel Austenite to Martensite trans’n

Accompanied by 4% lattice expansion
3. Nitriding of Metals & AlloysFe-N-C compound form’n

on surface and/or subsurface
4. Ion Exchange in GlassExchange of Na+ with K+

in SiO2 glass…chemical strengthening of glass.
tableun_08_p276b
Chapter 8 -
Shot Peening
This can be measured by

X-ray diffractometry 
tableun_08_p276b Chapter 8 -
Shot Peened Surface
300 m
Courtesy of Saint Gobain Inc. tableun_08_p276b

Chapter 8 -
Shot Peening
Chapter 8 -
Shot Peening Mechanics
<0 <0
>0 >0 <0 <0
Deformed volume
Due to clamping
by the much thicker
undeformed part
Unaffected
volume
Restoring forces and the
Free body diagram deformed layer does not fly off!!!
Compression layer
Tension layer
(E,)f(z) No interfaces!!!
Neutral Layer
z tableun_08_p276b
Chapter 8 -
Shot Peening
 Compressive stresses created
in the surface/subsurface are
very beneficial against
fracture, fatigue failure and
even creep resistance!!!
 Compressive stresses created

increase the hardness of the
surface/subsurface, thereby
contributing to abrasion
resistance as well.
tableun_08_p276b
Chapter 8 -
Chemical Strengthening of Glass
“Foundation of Gorilla Glass”
r = rK+ - rNa+  r > 0

\ Causes hydrostatic
compression @ K + sites!
Darken’s
Interdiffusion
Coefficient!
æ ¶CK+ (x, t) ö
JSiO 2
= -D SiO2 çç ÷÷
K+ K+ / Na+ ç ÷ø
è ¶x
Courtesy of Cat-i Glass Companny

Chapter 8 -
Before Ion Exchange After Ion Exchange
The matrix’s reactionary force to the larger K+ is compressive as

K+ is larger than the ion it replaces!
tableun_08_p276b
Chapter 8 -
Chapter 8 -
VIII.
Concluding Remarks
“Engineering over 6 order of magnitude”
Makes Materials Science 6 Engineering
Unique!!!
1 nanometer-1millimeter
Chapter 8 - 129
Metallurgical Aspect of Fracture
From 1 nm to 1 mm
Substitutional Atom Coherent High Melting Point

Precipitation Second Phase
Interstitial Atom
Edge Dislocation Slip Lines
Incoherent Precipitation
Elementary GB
Unit cell Ppt’n
Screw Dislocation
Intergranular
Fracture Transgranular Fracture
Layers of Grain
Boundary Precipitates
tableun_08_p276b
Chapter 8 -
Metallurgical Aspect of Fracture
 Most effective crack path
“torturers” are:
(i) Second phase
(ii) Particles & ppt’n
(iii) Grain size
(iv) Fibering & texturing
 High strength materials usually possess several

microstructural features in order to optimize mechanical
properties by influencing deformation behavior and crack
propagation paths.
 Metals provide the required complexity to design high

strength and toughness concomitantly
tableun_08_p276b
by MS&E!
Chapter 8 -
Fracture Toughness vs. Young’s Modulus
Chapter 8 -
Fracture Toughness vs. Strength
Chapter 8 -

Lecture 07 PhysMet Akdogan Fracture in Metals & Alloys F19

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture 07 PhysMet Akdogan Fracture in Metals & Alloys F19

Uploaded by

Copyright:

Available Formats

Lecture 11

 4700 ships were built by 1946.

• How do loading rate, loading history, and temperature

Early implants were stainless

Cracks open & propagate

Cracks tend to close &

 Design of built-in stresses in metals & alloys is a major method

Flaw density  Volume of Materials

Max. crack size:

i. (r>>rfiber)[(cmax)fiber<< (cmax)rod]  Manipulation of probabilities

 Weibull’s weakest link model

 Brittle fracture MUST be prevented at all cost!

max Crack propagation 4

Inclusions from Fracture surface of tire cord wire loaded in tension

Courtesy of J.D. Begg

Courtesy of Shimadzu Corp.

Ductile Fracture Brittle Fracture

Adapted from Fig. 8.5(a), Callister & Rethwisch 8e.

Initiation of the crack

Crack propagating along GBs

Avoid at all cost!!!

304 Stainless Steel 316 Stainless Steel

Extensively used in metallurgical assessments in the

Volume crack Surface crack

Typically Y~1 (depends on geometry of crack and specimen)

 The discrepancy between cohesive and fracture strength is

 Griffith proposed, which is based on the concept of strain

 The surface energy due to the presence of the crack is

Plane Stress Plane Strain

Extrinsic property Intrinsic property

 As shown above, the critical stress c is very sensitive

 The fracture strength of the material is ~ entirely

100 Energy barrier (+)

Ductile Solids Brittle Solids

If  S =0.8 J/m 2 then  c 

Keep the Grain Boundaries clean!

© CEphoto, Uwe Aranas

 Dissolved H+ diffuses very

H+ is very small…jump frequency 2-3 orders of magnitude larger than

Dislocation pile-up E

Grain  eff : Effective shear stress

Reduction of area (%)

a  From AB, c with

a  Here <F &  is constant.

Normal Ductile Solid

Cracks in soft metals have large

 Problem: Steels were used with DBT temperature just

History Titanic Steel Modern Mild Steel

Adapted from Fig. 8.12(b),

Flat surfaces Cup & Cone surfaces

FCC metals (e.g., Austenite, Cu, Ni, most HCP metals)

BCC metals (e.g., iron at T < 914ºC)

Brittle More Ductile

High strength steels,

The lower the carbon

 As % oxygen in steel  the TDBT :

 As grain size  toughness ……….. and TDBT.

Typical ingot structures observed in industry…

 React dissolved oxygen with selected alloying

 Rolled & forged

 Note that the difference

Keep %P<0.3% Low C steel is better !

Keep %S<0.005% Mn is always good!