Fracture Mechanics

Ratna Kumar Annabattala me II Tm

OriginalSlidesfromProf
 Background

❖ Increased use of metals in 19th century resulted in

increased accidents and casualties
❖ Approximately 200 people/year were killed in train
accidents in Britain during 1860-1870
❖ Mostly due to derailment of wheels caused by
fracture
 News items on fracture failures
❖ In Britain
❖ 19th March 1830: Montrose suspension bridge collapsed when one of its chains gave
away with 700 people assembled on it witnessing a boat ride
❖ 20th April 1887: Most serious train accident due to breakage of a draw bar
❖ 27th May 1887: Six people killed due to bursting of a wheel of a train
❖ 23rd June 1887: Broken Rail
❖ 2nd July 1887: Broken axle
❖ Liberty ship failures
❖ Out of 2500 Liberty ships made by USA at least 1500 instances of brittle fracture
❖ 3 of them broken into two pieces
❖ Welding as opposed to riveting increased the risk
 Presence of Cracks

❖ Several cracks may be observed during periodic

inspection of commercial aircrafts
❖ Cracks or crack like defects are found in ship structures,
bridges, pressure vessels, oil and gas pipe lines, heavy
machinery, vehicles and other machine components
❖ Cracks or crack like defects are of primary concern in
the design of nuclear reactors.
 Introduction
❖
r I
Theoretical stress concentration factor (Kt): indicates the average load required for the
onset of plastic deformation
❖

❖
O
Kt is limited to structures with known dimensions
Kt approaches infinity when there is a crack/flaw/inclusion where root radius tends to
zero
❖ For a very small r, local plastic deformation begins rendering elastic SCF useless
afterwards
❖ Failure criteria studied so far can’t be applied if there are flaws/cracks present in the
material
❖ Yield strength, ultimate strength and buckling load based design can’t be used
❖ Hence, a fracture mechanics based design is essential for the safety of
components/structures
❖ In this class, we will only focus on brittle fracture involving only linear elastic fields
around the crack tip
 A crack in a structure

Source: David Broek, Elementary Engineering Fracture Mechanics

❖ The pre-existing crack grows with the application of load environmental

condition
❖ Crack length increases with time
❖ Residual strength reduces with increase in crack size
❖ Due to the presence of crack, residual strength reduces
❖ Hence, the part is prone to failure from the expected high service load
 What are the key questions?
❖ What is the residual strength as a function fo crack size?
❖ What size of crack can be tolerated at the expected
service load? (Critical crack size)
❖ How long does it take for a crack to grow from its initial
size to the critical size?
❖ What is the maximum size of the flaw/crack that can be
permitted when the structure starts its service life?
❖ How often should the structure be inspected for cracks?
 What is the way out?

❖ Fracture Mechanics (FM) can actually answer some of

the questions.
❖ FM can show a way forward for some of the questions
which it can’t answer directly.
 Fracture Mechanics across
Scales

Source: David Broek, Elementary Engineering Fracture Mechanics

❖ Successful design based on FM requires knowledge

across the domains shown above
 Three modes of fracture
Crack in solid can be stressed in three differentmodes

P P
p results
The three modes of cracking
normal stress out ofplane
give rise to opening shear shear
in sliding
 Griffith’s experiment

9994

𝜆2𝑎

2a

Unstretched Stretched Crack at the centre

tilt
 Griffith’s experiment
❖ Take a sheet, apply a force and then hold its grips
❖ Use a knife to cut a slit normal to the loading direction
❖ The length of the crack is small to begin with and nothing happens
❖ Continue to increase the length of the slit with the knife
❖ At a certain length, without further help from the knife, the crack extends
❖ Now the questions are
❖ What is the critical length at which the crack advances on its own?
❖ Can this critical length be changed for a given material?
❖ What happens if we alter the force applied to begin with?
 Surface Energy
❖ Like Surface tension in liquids, solids also possess surface energy
❖ In a solid, all the atoms in the bulk are surrounded by the similar
environment
❖ However, for the atoms on the free surface and few layers below the
free surface, there is only bonding force from beneath and nothing or
very little from above
❖ As a result, the surface layer of atoms experiences strain. The
deformation of the surface atoms requires some energy called
“Surface Energy”
❖ Griffith hypothesized that the crack in a material advances when
the released energy (i.e, incremental energy released) exceeds the
energy required to create two new surfaces.
 Griffith’s analysis
❖ The total energy released
𝜎2
𝐸𝑟 = (Total volume of the traingles) ×
2𝐸

1 𝜎2 2𝜆𝑎2 𝐵𝜎 2 𝜋𝑎2 𝐵𝜎 2
𝐸𝑟 = 2 (2𝑎)(2𝜆𝑎)𝐵 × = = (with 𝜆
2 2𝐸 𝐸 𝐸
= 𝜋/2)
❖ If γ is the surface energy per unit area of one surface, the
surface energy required (Es) is

𝐸𝑠 = 2(2𝑎)𝐵𝛾 = 4𝑎𝐵𝛾

B is the thickness of the specimen

 Griffith’s analysis
𝑑𝐸𝑟 𝑑𝐸𝑠
≥
𝑑𝑎 𝑑𝑎

2𝜋𝑎𝑐 𝐵𝜎 2
≥ 4𝐵𝛾
𝐸
𝐸𝑟 , 𝐸𝑠

Hence, for a safe crack

𝐸𝑠 𝐸𝑟
2𝐸𝛾
𝑎𝑐 ≤
𝜋𝜎 2

Δ𝐸𝑠 The maximum stress

to advance a crack of
length a
Δ𝐸𝑟

𝑎0 𝑎𝑐 𝑎 2𝐸𝛾
𝜎𝑐 ≥
𝜋𝑎
 Discussion

❖ With an advancing crack,

❖ Strain energy in the component decreases or increases
❖ Stiffness of the component decreases
❖ The points of application of load may or may not move
❖ Work is done on the component by the forces if the
points move.
❖ Energy is consumed to create two new surfaces
 Formulation of Energy Release
Rate
❖ How much energy is released when a crack advances?
❖ Energy release rate (G): Energy release per unit increase in crack area
❖ Consider an incremental increase in the crack area ΔA
❖ An incremental external work ΔWext is done to create the crack
extension
❖ Strain energy of the body increases by ΔU.
❖ The available energy GΔA, provides the energy balance
❖ GΔA = ΔWext - ΔU
❖ Dividing the equation by ΔA and taking the limit ΔA—> 0

potters The energy is available from the system if

𝑑 𝑑Π the potential energy decreases. Note that G
𝐺=− (𝑈 − 𝑊ext ) = − is always positive.
𝑑𝐴 𝑑𝐴
 Formulation of Energy Release
Rate
❖ Assuming a plate of uniform thickness, B

𝐺=−
1 𝑑Π
𝐵 𝑑𝑎

Double Cantilena Beam

A
Example of a DCB specimen with compliance approach

𝑢 = 𝐶𝑃
With constant load: compliance of material in use ofstiffness
1
𝑈 = 𝑃𝑢; 𝑊ext = 𝑃𝑢
2
1 𝑃 𝑑𝑢 𝑃2 𝑑𝐶
Π = 𝑈 − 𝑊ext = − 𝑃𝑢 ∴𝐺= = .
2 2𝐵 𝑑𝑎 2𝐵 𝑑𝑎

Idfa
With fixed grip
P E dig
qt
1 2 2
𝑢 𝑑𝑃 𝑢 𝑑𝐶 𝑃 𝑑𝐶
Π = 𝑃𝑢, ∴𝐺=− = =
2 2𝐵 𝑑𝑎 2𝐵𝐶 2 𝑑𝑎 2𝐵 𝑑𝑎
 Formulation of Energy Release
Rate
❖ For a DCB specimen with one cantilever-end attached to a fixed jaw of a tensile
machine, the deflection of the moving jaw is twice that of one cantilever end.

2 𝑃𝑎3 𝑢 2𝑎3 𝑎3
𝑢= ⟹𝐶= = =8
3 𝐸𝐼 𝑃 3𝐸𝐼 𝐸𝐵ℎ3
𝑑𝐶 𝑎2 Itt
∴ = 24
𝑑𝑎 𝐸𝐵ℎ3

I
𝑃2 𝑑𝐶 12 𝑎2 𝑃2
∴ 𝐺𝐼 = =
2𝐵 𝑑𝑎 𝐸 𝐵 2 ℎ3

Determine the critical energy release rate of a DCB specimen loaded in a

tensile testing machine. The thickness of the DCB specimen is 30 mm, depth
of each cantilever 12 mm and crack length 50 mm. It is made of hardened
steel with a modulus of 207 GPa and the crack is about to propagate at 15405
N pulling load.