15.

Fracture

APSC278 by Warren Poole and Anoush Poursartip

 Questions
• How can we determine if a crack will grow
through a structure?

APSC278 by Warren 2
Poole and Anoush APSC278 by Warren Poole and Anoush Poursartip
 Learning objectives
• Explain how stress concentrations arise in
structures with notched and flaws
• Define and use the stress concentration
factor in practical problems
• Explain and apply the Griffith theory of
brittle fracture
• Use the concept of a stress intensity factor
in crack propagation problem
3
APSC278 by Warren Poole and Anoush Poursartip
 Failure

ALOHA AIRLINES, FLIGHT 243

BOEING 737-200, N7371I,
NEAR MAUI, HAWAII
APRIL 28, 1988

APSC278 by Warren Poole and Anoush Poursartip

 Failure
• Failures can occur at applied stress levels below the yield
stress, e.g. fatigue or brittle fracture
• Understanding how this happens is critical to determining
safe component design

APSC278 by Warren Poole and Anoush Poursartip

 Failure

Highly Ductile Fracture: Moderately Ductile Brittle Fracture:

Specimen necks down to Fracture: No plastic deformation
a point Some necking

APSC278 by Warren Poole and Anoush Poursartip

 Cup and Cone Fracture
Initial Necking Cavity initiation
Cavity coalesce into a crack

Final shear fracture (45o to loading direction)

Cup and Cone Fracture: Brittle Fracture:

Aluminum Steel

Crack Propagation

APSC278 by Warren Poole and Anoush Poursartip

 Ductile Fracture
Scanning electron microscope fractographs

Ductile fracture in uniaxial tensile Ductile fracture under shear loading:

test: spherical dimples parabolic dimples

APSC278 by Warren Poole and Anoush Poursartip

 Crack Origin Brittle Fracture
~15 mm
V-shaped “chevron”
markings

Crack Origin

Radial fan-shaped
ridges ~20 mm

APSC278 by Warren Poole and Anoush Poursartip

 Brittle Fracture
Transgranular fracture Intergranular fracture

APSC278 by Warren Poole and Anoush Poursartip

 Fracture Characteristics
• Ductile – the material exhibits • Brittle – the material exhibits
high toughness low toughness
– Extensive plastic deformation – No (or little) plastic
– To naked eye the fracture deformation
surface is grey and fibrous – Fracture surface is crystalline
– “Dimpled” or cup and cones (shiny) –e.g. intra-granular or
morphology on fracture trans-granular
surface – Flat surface
– Evidence of shear – e.g. failure – Evidence of “Cleavage” of
along paths of maximum shear grains
– Chevron patterns with fracture
surface perpendicular to load
– Fan-shaped pattern on
fracture surface

APSC278 by Warren Poole and Anoush Poursartip

 Fracture Characteristics
• The key phrase in the two previous list relates to
the use of the term toughness.
• All manufactured materials contain defects:
– Porosity
– Inclusions (contamination arising from processing)
– Surface damage accumulated during
manufacturing/during service
– “dislocations”
• Toughness relates to the ability of a material to
tolerate these defects under load

APSC278 by Warren Poole and Anoush Poursartip

APSC278 by Warren Poole and Anoush Poursartip
 The Effect of a Defect (notch) on Fracture

Another way of
looking at the effect
of a defect or notch  WHY
?

APSC278 by Warren Poole and Anoush Poursartip

 Stress Concentration
F

where s0 is the nominal stress away from the notch –

e.g. s0=F/Anominal

APSC278 by Warren Poole and Anoush Poursartip

Stress Concentrations

APSC278 by Warren Poole and Anoush Poursartip

 Stress Concentrations

so = nominal stress
  a 
1

sm = max stress
2

s m  s o 1  2  
  t   t = radius of curvature at
  crack tip

APSC278 by Warren Poole and Anoush Poursartip

 Stress Concentrations
 1
 so = nominal stress
 a  2
sm = max stress

s m  s o 1  2  
  t   t = radius of curvature at
  crack tip

Key concept – Kt is a measure of the amplification of the stress

that arises at the tip of a crack or defect.
APSC278 by Warren Poole and Anoush Poursartip
Theoretical Kt Values

APSC278 by Warren Poole and Anoush Poursartip

Simple but practical example –
Steel Bridge Infrastructure

A common approach applied to retard or arrest

cracks in steel bridge structures is to drill holes at the
end of the detectable cracks – obviously not always
effective.
APSC278 by Warren Poole and Anoush Poursartip
 Simple but practical example

Hole-based crack retardation is greatly improved if the hole can be expanded

and a compressive residual stress around the hole can be induced.

Repairing and Preserving Bridge Structure by Innovative Crack Arrest Repair System
Author: LEN REID, Fatigue Technology, Seattle, Washington

APSC278 by Warren Poole and Anoush Poursartip

 Stress Concentrations
• There are flaws in all materials, often very small
• A. A. Griffith (1920s) proposed that very small, microscopic cracks weaken
brittle materials
– This is why the theoretical cohesive strength of a brittle elastic solid (~E/10) is not
achieved
• Ductile materials are much less sensitive to stress raisers, as they can yield at
the crack tip
– When the maximum stress exceeds the yield strength
• Griffith proposed that all brittle materials contain a population of small cracks
and flaws that have a variety of sizes, shapes, and orientations
– Fracture occurs when the theoretical cohesive strength of the material is exceeded
at the tip of one of these flaws
– This was shown with glass, where very fine and super-strong whiskers were shown
to have strengths close to the theoretical
• However, they very quickly form surface flaws, and the strength drops!

APSC278 by Warren Poole and Anoush Poursartip

 Griffith Theory of Brittle Fracture
• When a crack propagates
– there is a release of elastic strain energy
– There is energy consumed to extend the crack by creating new fracture
surfaces
• Griffith proposed that a crack will propagate when the release of
elastic strain energy > the energy required to form new crack surfaces

sc - nominal stress
E - elastic modulus
s - surface energy
a - 1/2 the length of an internal
crack, length of a surface
crack

APSC278 by Warren Poole and Anoush Poursartip

 An energy justification

When a crack of length a

forms on the surface of
an elastic solid:
1. The area within the
two triangles a-ba is
unloaded (elastic
energy released)
2. Crack surface area per
unit thickness created
is 2a

s
Recall from the MOR 1 𝜎2
𝑈 = 𝜎𝜀 = Units are J/m3
e 2 2𝐸
APSC278 by Warren Poole and Anoush Poursartip
 An energy justification
The elastic energy released per unit thickness when a crack extends from
distance a accounting for stress concentration is:

where b is approximately equal to p for a brittle material – i.e. sharp crack

tip (Inglis)
The surface energy required per unit thickness to extend a crack from
0 to a is:

∆𝐸𝑠𝑢𝑟𝑓𝑎𝑐𝑒 = 2𝛾𝑠 𝑎
Summing these:

APSC278 by Warren Poole and Anoush Poursartip

 Extension to Ductile Materials…
For ductile materials, we must include plastic deformation at the crac

s is replaced by s + p, where p is the plastic work of fracture

In the 1950s, G.R. Irwin, replaced s + p with the term

Gc, the critical strain energy release rate
1
𝐸𝐺𝑐 2
𝜎𝑓 =
𝜋𝑎
In the above equation, E, s, p are material properties, thus fracture occurs
when the applied stress equals the critical stress, sc, for a given crack
length
APSC278 by Warren Poole and Anoush Poursartip
 Example Problem
• A large glass plate is subjected to a tensile stress of 40 MPa
• Determine the maximum length of surface crack that is possible
without failure
– Given E=69 GPa and s = 0.3 J/m2
• Solution
– Solve for ‘a’ in Griffith equation

Substituting

Note: a is the length of a critical surface crack; 2a is the

length of the critical internal crack
APSC278 by Warren Poole and Anoush Poursartip
 Stress Intensity Factor K
• The stress concentration factor Kt includes a
measure of the crack tip radius of curvature

The reality is that many cracks are very sharp; such that t  to 0!
• This implies that the stress at the crack tip is infinite if the crack tip
radius is infinitely sharp!
• In reality, no crack tip is infinitely sharp, and there is always some
finite radius at the crack tip
• However, mathematicians looking at the crack tip stress fields have
realized that we can describe the stress fields very elegantly using the
concept of the stress intensity factor K

APSC278 by Warren Poole and Anoush Poursartip

 Stress Intensity Factor K
• There are three modes of loading and
associated stress intensity factors

KI KII KIII
APSC278 by Warren Poole and Anoush Poursartip
 Stress Intensity Factor K
• K is a measure of the stress field at the crack tip
• K is different from the stress concentration factor Kt. Note: difference
in Units. One is a ratio (no units) and the other has units of MPa-m(1/2)
!!!!
• K is a function of
– the applied stress
– the crack length
– shape of the crack
– the geometry of the structure

𝐹
Simple case 𝐾 = 𝑌𝜎 𝜋𝑎 = 𝑌 𝜋𝑎
𝑊𝐵 Y(a/W) – in this expression,
means a function of a/W, with the
General case 𝐾 = 𝑌 𝑎/𝑊 𝜎 𝜋𝑎 functional relationship dependent
on geometry

APSC278 by Warren Poole and Anoush Poursartip

Solution for K – Single Edge Notch

APSC278 by Warren Poole and Anoush Poursartip

Solution for K – Double Edge Notch

APSC278 by Warren Poole and Anoush Poursartip

Solution for K – Single Edge Notch Bending

APSC278 by Warren Poole and Anoush Poursartip

 Fracture Toughness
• A key concept in fracture mechanics is that two very
different geometries, with very different cracks, have the
same stress field ahead of the crack tip if they have the
same stress intensity factor K
– K uniquely defines the situation around the crack tip
• So, if one is going to fail, so is the other one!
• The value of K at which failure occurs is called Kc, the
critical stress intensity or fracture toughness.
• Kc is a material property

APSC278 by Warren Poole and Anoush Poursartip

Example Fracture Toughness
Values for Common Materials
KIC (MPam0.5)

APSC278 by Warren Poole and Anoush Poursartip

 Solution of crack problem using
fracture toughness approach
1. Determine the length (a) crack geometry and mode of
loading (Y)
2. Calculate the stress intensity factor K
3. Look up the fracture toughness value Kc of the material
from a handbook
4. Check if K >= Kc
• If true, then rapid crack propagation will likely take place

APSC278 by Warren Poole and Anoush Poursartip

APSC278 by Warren Poole and Anoush Poursartip
 Example Problem
• Consider an aluminum plate of 7075-T651, thickness 20
mm, width 100 mm, with a 40 mm edge crack. Will the
plate fail or yield (bulk) under a load of 100 kN?

1. How high are the nominal stresses compared to the yield

stress?
𝐹 100,000
𝜎𝑎𝑝𝑝𝑙𝑖𝑒𝑑 = = = 50 MPa note: unit use
𝑊𝐵 0.100∗0.020

2. How high are the net section stresses compared to the

yield stress?
𝐹 100,000
𝜎𝑛𝑒𝑡−𝑠𝑒𝑐𝑡𝑖𝑜𝑛 = = = 83 MPa
𝑊−𝑎 ∗𝐵 0.060∗0.020
𝜎𝑌𝑆 ~ 495 MPa so no bulk yielding for either
APSC278 by Warren Poole and Anoush Poursartip
 Example Problem
3. Determine if fracture occurs based on KIC – e.g.
crack propagation.
From previous table, Kc = 24 MPa m½

Determine Y, the geometry

correction factor?
Recall, Y is a function of (a/W)

𝑎 0.04
= = 0.4
𝑊 0.100

APSC278 by Warren Poole and Anoush Poursartip

Example Problem, cont’d

APSC278 by Warren Poole and Anoush Poursartip

 Example Problem, cont’d
For this geometry, for a/W = 0.4 gives Y = ~2.1

Remember Kc = 24 MPa m½ , since K>Kc, failure will likely occur.

What is the critical crack length for this material and this loading/crack geometry?

2
1 𝑊𝐵𝐾𝐼𝑐
rearranging 𝑎𝑐 = = 17𝑚𝑚
𝜋 𝑌𝐹

APSC278 by Warren Poole and Anoush Poursartip

 Example Problem, Discussion
• So, a 17mm critical crack is pretty large. The implication
being is that 7075-T651 is pretty tough alloy. 7075 is often
used in transport applications, including marine, automotive and aviation
(high strength-to-density ratio). Other applications include rock climbing
equipment, bicycle components, in-line skating-frames and hang glider
airframes are commonly made from 7075 aluminum alloy. T651 refers to heat
treatment and post heat treatment residual stress relief. Its relatively
expensive…

• The crack length seems outside the range of what maybe expected to
be present due to a manufacturing related defect or an in-service
scrape/scratch. What are we missing here? Are we being too
conservative in our design? Is there another mechanism of crack
initiation and growth in load bearing structures? A crack of this length
is certainly detectable prior to failure (which is desired).

APSC278 by Warren Poole and Anoush Poursartip

 Detecting Cracks
• Detecting cracks before they become critical is
very desirable
– If detected before failure we can remove the
component from service
– Use a variety of methods, visual inspection, non-
destructive testing techniques (NDT), die penetrant
ultrasonic, etc…
– Use concepts such as “leak before break” in pressure
vessels

APSC278 by Warren Poole and Anoush Poursartip

 Questions
• How can we determine if a crack will grow
through a structure?

APSC278 by Warren 44
Poole and Anoush APSC278 by Warren Poole and Anoush Poursartip
 Learning objectives
• Explain how stress concentrations arise in
structures with notched and flaws
• Define and use the stress concentration
factor in practical problems
• Explain and apply the Griffith theory of
brittle fracture
• Use the concept of a stress intensity factor
in crack propagation problem
45
APSC278 by Warren Poole and Anoush Poursartip