CTM T3 Failure

Failure
S. Altmeyer
Escola d'Enginyeria de Telecomunicació I Aeroespacial de Castelldefels
Universitat Politècnica de Catalunya
March 20th 2023

Failure
Objectives of the lesson ...

How does failure occures? How do materials break? Why?
Different kind of failure: static, dynamic, at low / high T, …
(1) Fracture: Brittle versus ductile fracture mode
(2) Fatigue:
• Cyclic stresses
• Crack initiation and propagation
• Factors affecting fatigue
(3) Creep (Fluencia):
• Effects of temperature and external loads
20/03/2023 S. Altmeyer, Failure, CTM 2
Motivation
● Why is failure understanding important? (difficult to predict)
– Matter of fact - things break! (and, normally, no one wants this to occur)
– Estimated costs associated with fracture of components about 4-5% of GDP
– Knowledge on fracture processes has advanced significantly since World
War II and, particularly, since the mid-70s (before: “try-and-error”)
● Why do structures or components fail?
– Negligence during the design phase, construction or operating conditions.
– Application of a new design, of a new manufacturing technique or a new
material that produces an unexpected (and unwanted) result.
– improper material selection and fabrication errors
– improper design of components
– Misuse in operation
→ Engineer: develop contingent strategies, study the

causes, prevent, predict and control them
Motivation – Historical perspective
Ship-cyclic loading from waves.
Stone materials
– risk of breaking in structures under tensile /
bending stress
– solution: to design looking for compressive stress
states → structure & design optimization
– Metals (steel in particular) show a good behavior
under tensile conditions
– nevertheless, they can experience fracture
WWII: Liberty ships
Welded ships
– ships halved spontaneously
CAUSES:
– brittle welds, with the presence of defects.
– Brittle ↔ Ductile transition at low temperature
→ importance of environmental conditions!
Flying accident
ALOHA Airlines B737 (1988)
– Partial detachment of the fuselage during the
flight → explosive decompression
– One of the air hostesses dissapeared.
CAUSES:
– Processes of fatigue leading to fracture in the
junction of the fuselage
MISSING MAINTAINANCE PROGRAM
Prestige accident (2002)
Reason: environment – humid, salty
– 77000 Tm of fuel were spilled.
CAUSES:
– Accumulation of damage due to
thermal and mechanical fatigue that
led to the total fracture of the ship

Challenger Accident (1986)
– Death of the 7 members of the crew. Stopping
of the Space Program.
CAUSES:
– O-rings malfunction due to the low
temperatures at the time of launch.
– The flight had been delayed up to 7 times, due
to the low temperatures.
– Reliable forecast of the NASA were
sometimes up to a factor of one thousand
different of what engineers estimated.
– ‘For a successful technology, reality must take
precedence over public relations, for nature
cannot be fooled’.
– Thermal embrittlement of the O–rings.
→ Brittle – ductile transition

Hyatt Regency (1981)

– Walkway collapse
– Overstressed steel support rods
CAUSE:
underdesigned
Tacoma Narrows Bridge (1940)

(https://www.youtube.com/watch?v=j-
zczJXSxnw)
– Resonances
CAUSE:
Poor design

Goals – Understanding Failure
Objectives:
● Identify: causes for failure/fracture, e.g. design parameters,
environment, material selection, misuse, …
● Destinguish: catasthrophic failure vs slow fracture
● Predict: life of structures; based on fatigue and creep
● Understand: failure mechanism and parameters controlling them
Defenition: Failure is the structural collaps of a material

→ undesirable event as:
• human lives are put to jeopardy
• economic losses
• interfears with availability of products

Fracture
Key ideas of fracture mechanics
“no material
→ no stress can
be transferred”

Fracture
Defenition: Simple fracture - a body subjected to static load (constant or
slowly changing) in temperature conditions much lower than that of
melting separates into two or more pieces.
● Two types of fracture modes: DUCTILE and BRITTLE
● Classification according to the ability to undergo plastic

deformation: in order to absorb deformation energy
● Two stages: crack formation and propagation

Fracture – Ductile vs Brittle
Ductile fracture: Brittle fracture:
- few pieces - many pieces
- extensive plastic deformations - little or no plastic deformation
- slow process - fast process
- dull fracture surface - bright and shiny fracture surface
- stable crack: no progress if σ → cleavage
doesn't increase - unstable crack: once crack starts
- cup-and-cone type fracture propagating → continue to failure
- evidences (visible in advance) - grainy-faced fracture surface
- no evidence (not visible in advance) -
catastrophic & sudden

Fracture – Brittle
There are three main factors of brittle fracture in material:
1) High strain rate, that is, rapid rate deformation
2) Stress concentration
3) Triaxial stress state, which may be introduced by the presence

of a notch.
Fracture Mechanics:
- Discipline that studies the behavior of the material in the presence of defects.

Fracture – Ductile vs Brittle
● Engineering point of view: ductile prefered against brittle fracture:
1) Brittle fracture occurs suddenly (no prior symptoms).

In the ductile fracture there are symptoms of plastic deformation:
one can take preventive measurements
2) Ductile fracture needs a higher deformation energy
→ larger resilience; ability to absorb energy
(ductile materials are generally more tencious
than brittle ones)
● General classification
– metals at not too low temperatures are ductile
– ceramics, brittle
– polymers can have both type of fractures

Ductile – Fracture – Micromechanism
Ductil fracture: formation, growth and coalescense of microvoids
● Most common ductil failure: cup-and-cone type fracture.
Mechanism:
1. Moderate amount of stiffness, initial necking (stretching bounds)
2. Formation of small cavities or microvoids form in interior of crosssection
after beginning the squeezing → grain boundary separation
3. The microvoids/cavities grow, come together, and coalesce in order to form an
elliptical crack with its long axis perpendicular to the applied load.
4. The crack continues to grow/progress parallel its major axis through the
phenomenon of coalescence.
5. At the end, rapid propagation of the crack on the outside of the shear due to
internal shear (approximately 45° with respect to the load; !maximum shear
stress for this angle!).

Ductile – Fracture – Micromechanism
Ductil fracture: formation, growth and coalescense of microvoids
Metallic materials with plastic behavior:

• Favored by T↑, σ Y ↓, dσ/dt ↓
●
Ductile Fracture – Examples/Fractography
Cup-and-cone fracture
Microscopics Macroscopics
- Generation and growth of microvoids. - Large amount of plastic deformation.
- Dimplet fracture - Very rough and matt fracture surface.
- High energy consumption
Brittle Fracture – Examples/Fractography
Grainy-faced fracture Cleavage
• Favored by low temperatures and

high loading rates
transgranular intergranular • Favored in materials with high σY.

Brittle Fracture
Cleavage: successiv and repeated breaking of bonds along specific crystalline planes
Transgranular fracture Intergranular fracture

● fracture crack passes through the
● crack propagation along boundaries
grains
● leaves a 3D surface structure after
● result typically in a very flat, bright the fracture
and shiny surface
● usually appears subsequent to
processes, that weaken boundaries
Brittle Fracture – Examples/Fractography
(potential) visible pattern
V-shaped “chevron”
– cross section that point back
toward the crack initiation site
“fan-shaped” ridge
– lines or ridges that radiate
from the origin of the crack
Note: for brittle fracture – not necessary any pattern is visible!

Fracture – Materials: Theory vs. Reality
● Leonardo DaVinci:
- longer & older cable lengths need lower loads to cause breakage.
● Reasons:
- defects cause premature fatigue
- older cables show more defects
- theoretical values → based on atomic bonding energies
NO material is perfect!
● Resistance to fracture is a function of the cohesion forces:
→ theoretical fracture strength: σth~ E/10
Substantially lower than predicted by theory from interatomic
bonding forces
→ theory of dislocations σTS(real) << σTS(theory)

Fracture – Materials: Theory vs. Reality
Stress state in crack tip

● Fracture is a consequence of the stress concentration produced
by discontinuities in the material (“defects”).
● Example: fracture of a paper sheet with a small crack.
● Inglis (1913) studied the stress concentration effect of an
elliptical drill in an infinite plane plate, in linear elastic regime

Fracture – Mechanics – stress distribution
● Reality: presence of microcracks in the interior and surface
● The inevitable existence of many defects is the major reason for this
important discrepancy. σth~ E/10
● Macrocracks are in the material as a result of the manufacturing process
(e.g. welding) or develop due to service conditions (fatigue, creep).
● Reduce the resistance to fracture of the material.
● Ability to amplify the applied stress
→ stress concentration
stress amplification
Inglis (1913)
stress concentration factor

Fractor toughness
● measure of a material’s resistance to brittle fracture

when a crack is present (unit: MPa m1/2)
● limit to fracture (Y geom. Factor)
● Thin specimen Kc depends on specimen thickness
Plane strain fractor toughness

● Large specimen K independ of thickness Design Using Fracture Mechanics
● Plane strain – no force in other directions
2 parameters prescribed → 3rd one fixed
● K depends on temperature, strain rate, and
IC
microstructure. Max design stress
● K large in ductil
IC
● K small in brittle
● K
IC Max flaw length
IC
decreases with:
– increasing strain rate; decreasing temperature
– solid solution or dispersion additions; strain hardening
● K increases with:
IC
● – reduction in grain size

Exercises:
● A large plate is subjected to a tensile stress of 100 MPa. Calculate the
(amplified) stress (by Inglis) in the tip of an elliptical defect with a
total lengh of 10 mm if the radius of curvature is 10–3 mm.
Solution:
Note: ultimate strength of a high strength steel is that of ‘only’ 2000 MPa.
● Obtain the maximum permissible crack in a semi-infinite plate

(geometric factor Y = 1.12) made out of a steel with toughness
KIc=100 MPa · m1/2, subjected to a stress state (mode I) σ = 100 MPa.
Solution:

Fracture – Mechanics – stress concentration
● Stress amplification is not restricted to microscopic defects
(internal and surface)
● Can occur in macroscopic discontinuities (more comman!):
holes, sharp angles (corners), notches, etc.
→ called stress raiser
→ optimize design!
→ surface defects; starting point for failure

Fracture – Mechanics – stress concentration
● The effect of a stress concentration (stress raiser) is more significant for
brittle than ductile materials
● Sharpest crack tip in brittle materials
→ the sharper the faster the crack propagation
● Tip deformation in the case of ductile; greater distribution of the applied
stresses; rounding of the tip
● Ductile material: ability to absorb energy in the binding bonds
→ plastic deformations
Energy balance on the crack

• Elastic strain energy
• energy stored in material when elastically deformed
• this energy is released when the crack propagates
• creation of new surfaces requires energy

Fracture – Ductil-brittle-transition (DBTT)
Variation in environmental conditions
● Temperature change
– low T: material behaves in a brittle way – small
absorbed energy
– high T material behavior will be ductile – energy
absorbed high. Example: High strength
– Intermediate zone: transition zone – abrupt jump steel alloys exhibit a DBTT
Titanic: steel with a DBTT ~ Room temp.

Fracture – Ductil-brittle-transition (DBTT)
Factors affecting toughness
● Temperature change
● Loading rate.
● Orientation.
● Grain size.
● Neutron irradiation (steel).
● Carbon content (steel)
● Others, ....

Fatigue
Definition: Fatigue is the break of structures due to dynamic stresses and
(periodic, cyclic) fluctuating for tension/stress below the yield strength
or the elastic limit
Fatigue = failure under cyclic stress (at moderate T)
● It occurs over a long periods of time
● Occurs unexpectedly and sudden, as if it were a brittle fracture
(even in ductile materials) dutil-to-brittle transition

● No plasic deformation → NO EVIDENCE impossible to predict
● Occurs in metals, ceramics, polymers, (???composites???)
● It is estimated that up to 90% of material failures in metals is due to
fatigue
● Example: famous cases
- Havilland Comet (1954)
- ALOHA AIRLINES (1988)

Fatigue
Definition:
● Failure mode in materials that involves initiation and propagation
of cracks in elements subjected to varying stresses over time
● It is called ‘subcritical process’ because it is a phenomenon that
takes place before (final) fracture (static at σTS; but lead to it)
Typical morphology of fracture: INITIATION AND PROPAGATION:

Fatigue – How?
● Under cyclic loading a crack can
– Be initiated for loadings with σs < σy < σTS
– Propagate for Ki < KC
Total life approaches

– Unable to account for inherent defects
Example: design of the De Havilland Comet
using total life approach. Defects resulting
from punched riveting of square windows
caused failure of aircrafts
– Unable to predict crack propagation

Fatigue – History – first studies
19th Century: early scientific studies:

• ALBERT (1829), fatigue tests on iron chains.
• PONCELET (1829) AND RANKINE (1843), WÖHLER (1850-1870),
breakage of train rails after many cycles with low stress
20th Century: emergence of Material fatigue (MF) and application of

fatigue
• OROWAN and TAYLOR (1930) develop dislocation concept.
• GRIFFITH (1921) and IRWIN (since 1950) lay the foundations of MF.
• PARIS (1961): application of FM to fatigue processes.
• Other authors: extension to various materials and structural situations

Fatigue – History
De Havilland Comet 1954 (first jet airliner, new designs, swept wings, ...)
came sudden apart in midair! → link to documentary in Atenea
● Example: failure not “just” appear

→ accumulation of many “mistakes”, missing “knowledge”, no mainatainance
● Outcome: → knowledge of fatigue
→ significant change in aviation industy for safty
Fatigue – History
De Havilland Comet 1954 (first jet airliner, new designs, swept wings, ...)
came sudden apart in midair! → link to documentary in Atenea
● Speculations:
sabotage, fire, flutter and
an explosion in a fuel tank
→ grounding & investigation
“no flaw ” with the plane
→ reenter to service
● 2nd similar accident
→ certification revoked
● Truth: Metal fatigue – Tests: fail after 3,000 pressurization cycles

Fatigue – History
Lessons Learned
● windows are no longer designed square → runded
● crack-stoppers are now placed between frame-cutouts
● FAIL-SAFE design philosophy – materials are assumed to contain a finite
initial defect size before entering service that may grow due to fatigue
loading in-service → maintainance program!!!
● (before SAFE-LIFE (TOTAL-LIFE) – structure for whole life and no
accumulation of damage during service e.g. cracking)
● Regular inspections
rounded off windows, for what became the Comet 2
→ Remodified design: rounded off windows, for what became the Comet 2
Fatigue – life time philosophy
General scheme:
The fatigue life evaluation can be done from two perspectives:
● 1) Estimation of the TOTAL LIFE of the element (initiation +

propagation).
It has its origin in the studies of Wöhler, Basquin and Goodman (19th
Century), and it is based on experimental and statistical studies. The
design parameter is the endurance.
● 2) Evaluation of the component life – FAIL SAFE – through

propagation, assuming the existence of certain size cracks. This
perspective starts with MF and Paris work (60’s).

Fatigue – stress cycles
● Tension applied: axial (traction-compression), bending, torsional
● Three modes of time-dependent fluctuation
- Reversed stress cycle: the amplitude is

symmetrical with respect to a stress level
equal to zero
- Repeated stress cycle: the amplitude is

symmetrical with respect to a stress level
other than zero.
- Random stress cycle

Fatigue – stress cycles
● Tension applied: axial (traction-compression), bending, torsional
● Particular case (easy) of oscillating loads.
● Evolution of stresses during a cyclic loading process.
● Shapes
Sine, square, trapezoidal, ...

Fatigue – loads at stress cycles - expressions
● Sign convention:
traction load are positive; compression loads are negative.
● Mean stress:
● Stress range:
● Stress amplitude:
● Stress ratio:
● Frequency:
(imited unless it goes accompanied by
environmental effects, humidity, high
temperatures)
Resonance effects

Fatigue – testing machine
Examples:
● Uniaxial stress test:
● Rotary bending test: tensile-compression:
● Tasks:
- duplicate in-service stress conditions as nearly as possible
(stress level, time frequency, stress pattern, temperature,
environment, …)
● Key points: Fatigue ...
- causes part failure, even though σmax < σy
- causes ~90% of mechanical engineering failures
Fatigue – S-N curve
Stress amplitude (S) vs number of cycles (N): S-N curve
● Realisation of multiple destructive tests are carried out at different
stresses/loads (usually cyclic) until failure/fracture
● It is usually represented the maximum tension or amplitude of tension
against the logarithm of the number of cycles (prop. time)
● Two types of curves (with and without fatigue/endurence limit):
SA SA
VE VE
Fatigue life can be zero!

● Decreasing curves:
higher tension → lower the number of cycles
● Fatigue limit Sfat: highest fluctuating stress/tension

value that does not cuase fracture
(e.g. steel, between 35 and 60% TS)
● If the curve continues to decrease, breakage can

occur at any load value (theoretically goes to zero)
● Fatigue resistance SR: tension level that produces

failure for a given number of cycles.
● Fatigue life Nf: number of cycles required to

produce the break for a specified level of stresses.

● low-cycle fatigue (LCF) ε-N:

< 104 – 105 cycles
oligofatiga (deformations elastic and plastic)
– Higher stresses than the elastic limit.
● high-cycle fatigue (HCF) S-N:

>104 – 105 cycles
(deformations practically elastic)
– Lower stresses than the elastic limit

Fatigue – S-N curve - examples
Metals Polymers
Fatigue affecting parameters

corrosion roughness temperature

Fatigue – S-N curve - Probability values
Problems:
● The inability to recreate identical trials results in a dispersion of the
curves and a consequent uncertainty
● Real data are scattered!
– material variation, defects, positioning, surface, test frequency, ...
→ statistical treatment of the S-N curves.
● Classical: Curves present mean values.
● Improvement/better: S-N curves of constant probability.

Example:
● A material is subject to cyclic stress with mean stress σM=200MPa.
What is the life time that this material will fail with a probability less
than (a) P<0.99 (b) P<0.9 (c) P<0.5
!!!Logaritmic scale!!!
Example:
● Experimental data: Al 6061-T6
!!!Logaritmic scale!!!
Fatigue – Process to rupture - lifetime
● Three stages: (HCF)
● Fatigue life Nf is the sum of the number of cycles required to initiate
Ni and propagate Np the crack:
● 3 distinct steps: Nf = Ni + Np (+ NR )
(1) crack initiation – small crack forms at some point of high stress
concentration (Ni)
(2) crack propagation – crack advances incrementally with each
stress cycle (Np)
(3) final failure/rapture – occurs very rapidly once the advancing
crack has reached a critical size (NR)
(The final break occurs very quickly: the number of cycles NR

necessary in this last stage is negligible)

LCF vs HCF
● The relationship between Ni and Np will depend on the type of the
applied tension/load and material
● Example:
- low number of fatigue: Ni <Np;
- high number fatigue: Ni> Np
● For long life:

→ improve Ni

Fatigue – Basquin's law
S-N curve (HCF)
● BASQUIN’S LAW (1910)
– Valid if: σm = 0; (equivalently, σmax = –σmin or R = –1)
Material constants are obtained from experimental data:

• Typically “a” varies between 1/15 and 1/8 in steel.
S-N procedure (or Basquin’s) doesn’t distinguish between each of these stages.
Fatigue – Multi-cycle-stress
● Cumulative Fatigue: Variable amplitude over time Δσ (t)
SEQUENCE OF ACTIONS
Palmgren-Miner Cyclic ratio summation rule
– Miner's rule (Linear damage Rule)
- where ni is the number of cycle at stress level σi and

- Ni is the number of cycles to failure at stress level σi
Assumption:
The stress sequence does not matter and the rate of damage
accumulation at a particular stress level is independent of the stress
history.
● Other problem: overload or unregular!

Fatigue – Multi-cycle-stress
Cumulative Fatigue:
● SEQUENCE OF ACTIONS – Miner's rule
(Linear damage Rule)
• Correct the data to R = –1 (Soderberg, Goodman, Gerber, etc.).
• Sort the data in i = 1..M blocks of constant amplitude.
• Calculate the associated damage to each “i” block:
• Get the total damage:
• Break condition: D = 1

● Microscopic observations for cycling loading
– I. Crack initiated at stress concentrations (nucleation)
– II. Crack growth resulting into surface striations
– III. Failure of the structure when the crack reaches a critical size Origin
Example of a crank axis
● Development of damage tolerant design

– Assume cracks are present from the beginning of service (FAIL-SAFE)
– Predict crack growth and end of life

Fatigue – Crack initiation – step I
Fatigue failure almost always starts at the surface, in points whith stress
concentrations (surface defects/effects, discontinuities → fracture mechanics)
→ the moving is approx. 45° of the load (max shear stress):
● I. Crack initiation/Nucleation: cracks initiated for K << K
C
- Existence of defects, marks or previous surface damage
- Greater surface stress → stress raiser
- Surface: deformations result from dislocations
motion along slip planes → from interior to exterior
- The external environment can favor these (corrosion)
- extrusions and intrusions can be formed (discontinuities
on microscopic surfaces)
- self-damage while working (misuse)

Fatigue – Crack initiation – step I
Texture of fatigue fracture surface – example

– self-damage while working
→ extrusions and intrusions (SEM) high
Lifetime – number of cycles

low
Internal errors → external/surface errors

Fatigue – Crack growth – step II
● II. Crack growth:
– Stage I fatigue crack growth:
• Along slip planes
multi-parallel process (in each grain)
(at the end: one-win scenario)
– Stage II fatigue crack growth:

• Across several grains
– Along a slip plane in each grain
– Straight ahead macroscopically
• Striation of the failure surface:
corresponds to the cycles

Fatigue – Crack growth – step II
● Crack growth & propagation:
• In which direction will the crack grow?

– For anisotropic & isotropic material
composite – In metals or ceramics
– In composites? Jumping behavior - irregular
• Cracks follow the path of least resistance
– Important to predict this path during the design
– “Fail safe design”

Fatigue – Failure – step III
● III. Structural failure striations
– As the crack growth K tends toward KC
– Increasing crack speed (self-enforcing)
– Change of direction: perpendicular to applied load
– Macroscopically: beachmarks and striations
– Not always visible (one-way observation)
– For critical size of the crack
→ static failure (fracture at 45°)
beachmarks

Fatigue – Fatigue life
Stages of crack growth (for moderate ductil material):
● Stage I: Crack initiates on a surface (scratches, keyways, stress
concentrations) Slow initial propagation of a stable crack (for ductile
material) along crystallographic directions
● Stage II: relatively faster crack propagation rate and ⊥ to the loading
direction . Persistent slip bands (PSBs), beachmarks & fatigue striations
● Stage III: Unstable crack growth & fracture at 45° due to plastic
deformation
● Factors affecting fatigue life

1. Applied Stress; mean stress, stress amplitude, range, ...
2. Surface Effects
3. Design Factors
4. Environment
5. Misuse
Note: for brittle material stage I, II and III basically fall together!
Fatigue – Fatigue life – Affecting factors
1) Nonzero mean stress: σm ≠ 0:

● Empirical models that allow to obtain stresses amplitude that, with σm = 0,
produces the same damage (same number of cycles resisted) than the
applied amplitude (σm ≠ 0) have been developed.

● The various empirical methods available propose simple functions that

depend on the parameter, σa(σm = 0). (which determines Nf in the
conventional S-N curve).


● S-N graphics usually represent values of
mean stress equal to zero. (better probability
curves)
● S-N curve changes for different mean

stresses (increasing the mean stress →
reduces fatigue life).
● Medium compression forces?

→ Positive effect

Fatigue – Fatigue life – Basquin's law
● BASQUIN’S LAW (1910)
Exercise: Given are the following parameters of Basqin’s law:
a = 0.090; C1 = 1320
● Obtain the S-N curves for σm = –200, –100, 0, 100 and 200 MPa,
respectively. Suppose that the elastic limit of the material is σY = 600 MPa.
(Soderberg)

Crack initiates on a surface
2) Surface condition
● Maximum stresses occur on the surface;
Consequently, most of the fatigue cracks
are nucleated on the surface (notch, tip,
sliped plane → stress raiser)
● Fatigue life is especially sensitive to

surface conditions:
– Design
– Surface treatment

3) Design
● Design factors: notches, geometric discontinuities can exerting
of stress concentrators and favoring fatigue.
● Examples: grooves, holes, threads, rivets, sharp edges etc.
→ stress raiser
● Surface defects: during manufacture, under working conditions,
transportation or misuse.
3) Environment
● Two types of influence:
– Thermal fatigue: thermal stresses due to fluctuations in the
temperature (no external load is required)
Thermal stress (coefficient of thermal expansion)
– Corrosion fatigue: facilitates the nucleation and propagation of the
crack (Aloha Airlines 1988)

Fatigue – Fatigue life – Improvements
– General
● Proper material selection → chose the “right” material
– Design
● Simple and “cheap” way
● Avoid any undesirable design factors (avoid following concentrations):
abrupt changes or discontinuities of geometry, surfaces of small
curvature, etc.
– Examples: avoid grooves, notches,
holes, threads, riveting, etc.
– Surface treatment
● Improve the surface finish by polishing
● Coating for protection
→ eleminate or minimize stress raiser
– Surface strenghten: shot peening

● Modifications in the surface material
● Introduce local and superficial compression forces which counteract
tensile stresses.
– Example:
“local” Cold work
perdigonado - shot peening – plastic deformation

– Surface strenghten: Case hardening

● Surface hardening: carburizing and nitriding (diffusion atomic of a
gaseous phase), superficial hardening (delays the entrance in plastic
zone and introduces residual stresses of compression).
– Surface protection/cover
● Against corrosion, choose suitable materials depending on the working
environment (saline environments, extreme temperatures, etc.). By time
anodized or other surface protection.
● coating
– Case hardening - carburization / nitriding

● Carbon diffusion in the surface of the steel
Strengthening
● Example: Interstitial diffusion example in a gear Interstitial
Result:
– more difficult to deform and to break;
– C prevents interplanar landslides
– C hinders motion of dislocations
Substitutional

Fatigue – Recent example
Rolls-Royce crisis– Trent 1000 (high-bypass turbofan) issues (in B787)
● Observation/Problem – WHAT went wrong? WHY?

– failure of different parts mainly caused by fatigue
(chain reaction – connected and not only individual)
● 3 points/challanges to deal with

– Intermediate pressure (IP) compressor blades
– High pressure (HP) blades
– IP rotor resonance
Fatigue –Observation→Analysis→Solution
1) Intermediate pressure (IP) compressor:
Cause of failure:
– Environmental condition
Fact: Force on each fan blade around 100t

→ need to be very strong
→ single-crystal structure (by crafting → see also creep)
eleminate microscopoic defects in grain boundaries
● Problem & What happen?:

- unexpected (NOT CONSIDERED) environmental conditions
(increased level of pollutation)
affected durability of IP blades
→ higher rate of sulfidation (contaminents in contact with high temperature air inside engine)
→ higher potential rate of crack initiation → lower lifespan
● Solution:
- new blade design
- new protective coating
- new base metal
Fatigue –Observation→Analysis→Solution
2) HP blades – Design flaw: (reconstructed/new

interior pa rts; smaller and more efficient)
● Problem:
Nickl alloy in HP blades – melting point 1455°C
Temperatures in the heart of the engine about 1700°C
→ integrity failure - melting
● Solution:
- redesign of HP turbine blades
- improved manufactoring process
- coating with low condactivity ceramic
- cool air (600°C) fed through cooling channels and around the holes by tiny wholes
specific design for each hole in size and directoin, …
(note potential stress raiser → challenging task)
→ metal remains below melting point (although the environment is!)

Fatigue – Observation→Analysis→Solution
3) IP rotor resonancs: – Design flaw:
● Problem:
- Weight reduction/optimization for efficiency: Trent 1000 has fewer number of fan
blade but keeping the same power (as earlier engine types)
→ depending on conditions the rotation of these blates created a resonance frequency
→ this frequency caused wave of excitement further down the IP compresor line
→ result in cracking in rotor 1 and rotor 2 compressor blades
complex three-pronged problem! Not an easy part-swap fix
● Solution:
- redesign of blades and further internals
- new technics & maintainances

Diffusion
Atoms movements in materials
Movement of atoms in solids, liquids and gases is very important
Examples:
Hardening steel, chrome-plating, gas reactions, Si wafers, Phase
diagrams, ... etc.
Overview:
● Atomic mechanisms of diffusion
● Mathematics of diffusion
● Factors affecting diffusion
● Steady state diffusion
● Nonsteady state diffusion

Diffusion
What is diffusion?
Defenition: Diffusion is the mass transport through atomic motion at high
temperature
→ can have self-diffusion or interdiffusion between two materials
– Self-diffusion occurs in pure elements
● Example: Two chambers, each containing a different gas, separated by a
removable barrier; when the barrier is pulled away, interdiffusion or impurity
diffusion occurs
● driving force is the chemical or concentration gradient through

materials or diffusion couple (pure material: self-diffusion)
Diffusion – in Solid Materials
Requirement: temperature must to

be high enough to overcome
energy barriers to atomic motion.
Reason: Heat causes atoms to vibrate

● Vibration amplitude increases
with temperature
● Melting occurs when vibrations
are sufficient to rupture bonds
The process of substitutional diffusion requires the presence of vacancies

(Vacancies give the atoms a place to move)
Diffusion – Arhenius equation
Arhenius equation:
As T increases → number of vacancies increases, and energy increases,

→ diffusion is faster
Diffusion – Mechanisms
Vacancy diffusion
● Atoms are constantly in motion and
vibrating
● change of atomic position requires:

– vacant site
– energy to break atomic bonds
● Two types of diffusion mechanisms

– vacancy diffusion (“slow”)
– interstitial diffusion (“fast”)
● movement of vacancies in one direction

is equivalent to atomic movement in the
opposite direction

Interstitial diffusion
● Interstitial atom moves from one interstitial site to another (empty)
● Energy needed, again to squeeze past atoms.
● Usually much faster because many more empty interstitial sites and
no vacancies are required

Diffusion flux
● Diffusion is a time-dependent process
→ rate of diffusion is important
● diffusion flux (J) is defined as the mass, M, diffusing through
unit area, A, per unit time, t

Diffusion – Steady-State
Fick’s First law: ● Concentration profile
● Concentration gradient
● Fick’s 1st law (1D):
Where D is the diffusion coefficient (diffusivity) or speed of

diffusion (m2/s).
“–“ because atoms diffuse down concentration gradient

(from high to low)
● Example of steady-state diffusion is gas diffusing through a metal

plate (gas pressure constant).
Diffusion
Example: Case Hardening
● Example of interstitial diffusion

is a case hardened gear
● Diffuse carbon atoms into the

host iron atoms at the surface
● Result: The "Case" is

- hard to deform: C atoms "lock"
planes from shearing
- hard to crack: C atoms put the
surface in compression

Diffusion – Non-steady-State
● In most real situations diffusion is not constant
● Flux and concentration gradient vary with time
● The changes of the concentration profile is given

in this case by a differential equation
● If diffusion coefficient is independent of

composition then:
● Solution of this equation is concentration profile

as function of time
Fick’s Second law:
● Solution requires boundary conditions

– A useful solution is for a semi-infinite solid when the surface
concentration remains constant
● What is semi-infinite solid?

– semi infinite bar: If non of the diffusing atoms reaches the bar end
during the time over which the diffusion takes place
● l >10*sqrt(Dt), where l: bar length, D: diffusion coefficient and t:

time

Fick’s Second law: – Application
● Simple boundary condition is where the surface concentration is held constant,
– e.g. gas phase with constant partial pressure at the surface
● Conditions are:
– before diffusion, solute atom have a homogeneous concentration of Co
– x is zero at the surface and increases with distance into the solid
– time is zero just before diffusion begins
● Mathematically, for
t = 0, C = Co at 0 < x < inf
t > 0, C = Cs at x = 0 and C = Co at x = inf
● Cs = constant surface concentration
● applying these boundary conditions gives:



Diffusion
Factors Affecting Solid-State Diffusion
● Diffusing species and host material are important
– smaller atoms can “squeeze” in between host atoms more easily
– in lower packing density host material → easier for atoms to
migrate with fewer bonds to expand
• faster in more open lattice (BCC faster than FCC)
– in lower melting point host material → weaker bonds (easier to
push apart)
● Temperature has a strong effect on diffusion rates:
● Activation energy is the barrier to diffusion and controls the

diffusion coefficient, D
●
Creep – static stress + elevated/high T
Question: Can also material subjected to a static load/tension lower than
tensile strength or elastic limit break? Similar to fatigue for cycle strength;
Answer: → Yes, because of creep phenomenon (fluencia)
Defenition: Creep is the time-dependent and permanent/irreversible
(plastic) deformation of materials when subjected to a constant load or
stress at elevated temperature (>0.4Tm)
● Creep is a plastic deformation!
● Example of creep: chocolate bar that breaks on a hot day due to its own
weight
● Factors involved:
– applied stress/tension (σ)
– temperature (T)
– exposition time (t)
Creep - Appearance
Where and when do we see creep?
● Appears on components subjected to high/elevated 0.4Tm
temperature and static loads
Example: in service (e.g. rotors and turbine
reactors, pipes, heat exchangers, car radiators, …)
● Creep appears in all kind of materials (similar fatigue)
● In metals important for operating temperatures

exceeding 0.4Tm
● Creep rupture tests are usually carried out at

constant stress/load or at constant tension
according to complexity and objective.
(usually at const. load; avoid crack formation and
propagation appearing uner tension)
● For metals, test specimens and machinery are

in many cases those used in tensile tests.
Creep – Stages of creep
The creep process can be split in 4 different steps (1 + 3):
● (0) t = 0: instantaneous deformation (mainly elastic – reversible)
(4)
● (1) Region I: primary or transient creep
(rate of deformation decreasing:
strain-hardening by plastic deformation)
● (2) Region II: secondary or steady-state creep

(strain rate constant: balance between (3)
strain-hardening and recovery) (1) (2)
● (3) Region III: tertiary creep (acceleration
of the process: rupture).
(0)
● (4) Rupture: results from microstructures and/or metallurgical (grain
boundary separation, formation of internal cracks – cracking, holes and
cavities, constriction: reduction of effective straight area that supports the
load, and voids)
Creep – important parameters
Two important design parameters
1) Minimum creep speed or steady-state creep rate dεs/dt
● speed of creep strain rate in the secondary region - approx linear
– long life design parameter where large deformations are not allowed)
– constant variation → predictable!!!
– Failure is “NOT” an option → e.g. Nuclear power plant
2) Rupture lifetime tr
– short life design; on the “edge”
– “highest performence”
e.g. F1 engines, turbine blades,
rocket nozzles
1981 BMW 1.5l 2021 Mercedes 1.6l

Creep – parameter dependence
Factors affecting creep behavior (compare with fatigue):
– Influence of 3 main factors:
(a) load/stress σ
(b) temperature T
(c) time t
●
With the increase in T or σ:
– Increasing instantaneous deformation
– Increasing steady state creep rate dεs/dt
– Decreasing fapture time tr (time to failure)
● At temperatures less than 0.4Tm

→ deformation is time independent
(speed of steady creep equal to zero cf. Fatigue/endurence limit)

Representation of results
● Graph that relates the three factors that intervene in the creep process:
stress σ, temperature T and time t.
● Empirical observation:
- linear relationship between the logarithm of the tension σ and log of fracture
time tr for each temperature T
increasing T → worser creep behavior (note: metals: T< Tm → “no creep”)

Creep – Analytics
Influence of load and temperature (Experimental observation)
1st analytical approximation (simplest - limited)
●
A linear relationship between the logarithm of dε/dt and the logarithm of σ
● K1 and n are parameters that depend on the material properties and T

Creep – Analytics
Influence of load and temperature (Experimental observation)
2nd analytical approximation (improved – “expand K1”)
● Including the influence of temperature T:
Qc is called the activation energy for creep and is constant

K2 is also a material-dependent constant.
● As a consequence of the above, the influence of
diffusion plays a role in creep
Creep is still not fully understood!

● Different proposed mechanisms:
dissemination of vacancies to grain boundaries due stress, diffusion along
grain boundaries, intergranular sliding and dislocation movement
Factors that influence the creep behavior
● Three main parameters:
- Elastic modulus (E)
- Melting temperature (Tm)
- Grain size (d)
The larger E, Tm and d, the LARGER is the creep resistance of the material!
Note! Opposite for strengthening: larger grain size d → weaken the material
Example:
a) polycrystalline
b) directional grains
c) single crystals
Creep = combination of Cold-Work-hardening + Recovery

Creep – Influencing factors
● Various factors can influence creep behavior:

Creep – Data extrapolation
How can we predict creep and it's results?
● Problem:
- impractical to collect from normal
laboratory tests
- long-time appearance - many years!
(especially at low T)
● Solution:
Extrapolate the data
- “speed up” → go to larger T or σ
→ Larson-Miller parameter:
(one option) T[C + log(tr)] :

Creep – Data extrapolation
Example: Larson-Miller parameter

Creep – Responsibel mechanism
Mechanisms proposed for creep according to values of σ and T
● High stress σ:
High T:
– Grain boundary sliding diffusion-controlled (generation and) motion of
dislocations in the matrix (lattice diffusion); (GBS)
“Low” T:
– Dislocation (climb) creep or power-law creep dislocations to overcome
obstacles
– Dislocation gliding creep: dislocations glide through the crystal in slip planes
● Low stress σ:
High T:
– Nabarro-Herring creep – Bulk diffusion: flow of vacancies (and counter
flow of atoms) through the lattice from grain boundaries under tension to those
under compression
“Low” T:
– Coble creep – Grain boundary diffusion flow along grain boundaries
Further proposed mechanisms: solute drag creep, Harper-Dorn creep , sintering, Thermally
activated glide
Creep – Mechanism
Beginning of tertiary stage and final fracture:
● Beginning NOT coincident with damage nucleation but with phenomena like
accumulation of inter-granular damage, (GB) cavities and micro-cracks
● Final fracture due to grain boundary cavitation, neck formation, transgranular
crack development and creep resistance loss due to precipitate coarsening
deformation mechanism map

Creep – failure – Examples
Failure along grain boundaries Diffusion Creep – Flow of vacancies
Nabarro-Herring Coble
through grains along grain boundaries
Grain boundary sliding
Creep – failure – Examples
● World Trade Center Collapse
– due to structural damage + extra weight of planes → larger stress σ

– burning jet fuel → increasing temperature T
→ reduced cross section of the material
FAILURE
Failure – Summary
● Engineering materials don't reach theoretical strength σth~ E/10
● Distingush between 3 main failure types: fracture; fatigue; creep
● Flaws produce stress concentrations that cause premature failure
● Sharp corners produce large stress concentrations and premature failure
→ stress raiser
●
Failure type depends crucial on temperature T and stress σ
(material properties)
Failure Temperature T Load/Stress σ

Fracture low static
Fatigue low cyclic
time-dependent
Creep High (T > 0.4Tm) static

THANK YOU!
GRACIES!

CTM T3 Failure

Uploaded by

Document Information

Original Description:

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

CTM T3 Failure

Uploaded by

Copyright:

Available Formats

Failure

March 20th 2023

Objectives of the lesson ...

→ Engineer: develop contingent strategies, study the

20/03/2023 S. Altmeyer, Failure, CTM 5

20/03/2023 S. Altmeyer, Failure, CTM 6

Hyatt Regency (1981)

Tacoma Narrows Bridge (1940)

20/03/2023 S. Altmeyer, Failure, CTM 7

Defenition: Failure is the structural collaps of a material

20/03/2023 S. Altmeyer, Failure, CTM 8

20/03/2023 S. Altmeyer, Failure, CTM 9

● Classification according to the ability to undergo plastic

20/03/2023 S. Altmeyer, Failure, CTM 10

20/03/2023 S. Altmeyer, Failure, CTM 11

There are three main factors of brittle fracture in material:

1) High strain rate, that is, rapid rate deformation

3) Triaxial stress state, which may be introduced by the presence

20/03/2023 S. Altmeyer, Failure, CTM 12

● Engineering point of view: ductile prefered against brittle fracture:

1) Brittle fracture occurs suddenly (no prior symptoms).

20/03/2023 S. Altmeyer, Failure, CTM 13

20/03/2023 S. Altmeyer, Failure, CTM 14

Metallic materials with plastic behavior:

• Favored by low temperatures and

20/03/2023 S. Altmeyer, Failure, CTM 17

Transgranular fracture Intergranular fracture

Note: for brittle fracture – not necessary any pattern is visible!

20/03/2023 S. Altmeyer, Failure, CTM 20

Stress state in crack tip

20/03/2023 S. Altmeyer, Failure, CTM 21

stress concentration factor

20/03/2023 S. Altmeyer, Failure, CTM 22

● measure of a material’s resistance to brittle fracture

Plane strain fractor toughness

20/03/2023 S. Altmeyer, Failure, CTM 23

● Obtain the maximum permissible crack in a semi-infinite plate

20/03/2023 S. Altmeyer, Failure, CTM 24

20/03/2023 S. Altmeyer, Failure, CTM 25

Energy balance on the crack

20/03/2023 S. Altmeyer, Failure, CTM 26

Titanic: steel with a DBTT ~ Room temp.

20/03/2023 S. Altmeyer, Failure, CTM 27

20/03/2023 S. Altmeyer, Failure, CTM 28

● Occurs unexpectedly and sudden, as if it were a brittle fracture

(even in ductile materials) dutil-to-brittle transition

● Occurs in metals, ceramics, polymers, (???composites???)

● It is estimated that up to 90% of material failures in metals is due to

20/03/2023 S. Altmeyer, Failure, CTM 29

Typical morphology of fracture: INITIATION AND PROPAGATION:

20/03/2023 S. Altmeyer, Failure, CTM 30

Total life approaches

20/03/2023 S. Altmeyer, Failure, CTM 31

19th Century: early scientific studies:

20th Century: emergence of Material fatigue (MF) and application of

20/03/2023 S. Altmeyer, Failure, CTM 32

● Example: failure not “just” appear

● Truth: Metal fatigue – Tests: fail after 3,000 pressurization cycles

The fatigue life evaluation can be done from two perspectives:

● 1) Estimation of the TOTAL LIFE of the element (initiation +

● 2) Evaluation of the component life – FAIL SAFE – through

20/03/2023 S. Altmeyer, Failure, CTM 36