General Practice in Failure Analysis

Stages of Failure Analysis:

Although the sequence is subjected to variation, depending on the nature

of a specific failure, the principle stages that comprise the investigation
and analysis of failure are:

1. Collection of background data and selection of samples.

2. Preliminary examination of the failed part.
3. Nondestructive testing.
4. Mechanical testing.
5. Selection, identification, preservation, and / or cleaning or all
specimens.
6. Macroscopic examination and analysis.
7. Microscopic examination and analysis.
8. Selection and preparation of metallographic sections.
9. Examination and analysis of metallographic sections.
10.Chemical Analysis (bulk, local, surface corrosion products,
deposits or coatings,… etc.).
11.Analysis of Fracture Mechanics.
12.Testing under simulated service conditions (special tests).
13.Analysis of all the evidence, formulation of conclusions, and
writing the report (including recommendations).

1. Collection of background data and selection of samples:

 Collecting available information regarding:

o Manufacturing (drawings, specifications, and design aspects).
o Service history of the failed part (depending on record keeping).
 Photographic records of the failed component.
 Selection of samples should be done properly before examination.
 It is often necessary to compare with similar components that didn't
fail.
 Abnormal conditions during service, recent repairs …etc.

2. Preliminary examination of the failed part:

 Thorough preliminary visual examination before any cleaning.

 Lens or low power microscope (about 6 to 25x) can be used.
 General photography of the fractured part using direct lighting and
various angles of oblique lighting.

3. Nondestructive Testing:

 Liquid-penetrant inspection.
 Magnetic particle inspection of ferrous metals.
 Ultrasonic inspection.
 Eddy-current inspection of conductive materials.
 Radiography for internal examination.
 Acoustic emission inspection.
 Experimental stress analysis: strain gages, stress-coating
(using brittle laquer), photoelastic coatings …etc.

4. Mechanical Testing:

Hardness testing is the simplest and most versatile for failure analyst
(for evaluating heat treatment, approximate tensile strength of steel,
work hardening or softening caused by overheating, decarburization,
nitrogen pickup …).

Hardness techniques: Brinnel (HB), Rockwell (HRA, HRB, HRC),

Vickers (HV), Knoop (HK), Rebound (Scleroscope HS), Leeb rebound
(HL), Mohs Hardness
Tensile, fatigue, or impact tests where appropriate.
Determination of plane-strain fracture-toughness values may be
justifiable.

5. Selection, preservation, and cleaning of fracture surfaces:

Fracture surfaces may suffer mechanical or chemical damage that

destroy or obscure vital evidence.
Fracture surface should be protected by cloth or cotton during shipment
without being touched or rubbed with the fingers.
Chemical (corrosion) damage to a fracture specimen can be prevented
using desiccators or packing with a suitable desiccant.
Specimens contaminated with seawater or with fire-extinguishing fluids
require thorough washing with water followed by rinsing with acetone
or alcohol before storage in a desiccator.

Cleaning: of fracture surfaces should be done only when absolutely

necessary. Dry air blast or soft-hair brush, immersion or jet of
alkaline or acidic solvent to remove deposits, ultrasonic cleaning, and
use of replicas are common cleaning procedures.
Sectioning: of the fracture surface may be necessary for electron
microscopes. Records (sketches of photographs) to show the location of
sectioning should be kept. Saws or abrasive cut-off wheels are commonly
used.
Opening secondary cracks: when the primary fracture has been
damaged or corroded to such a degree that most of the information about
the cause of fracture is obliterated, it is desirable to open any secondary
cracks to expose their fracture surfaces for examination. These cracks
may provide more information than the primary fracture.

6. Macroscopic examination:

Detailed examination of fractured surfaces at magnifications from 1 to

25X, using a hand lens or a low power optical microscope.
For too large or too heavy specimens, replicas may be used.
It usually detects the direction of cracks and the origin of failure in the
form of Macrographs (Small resolution – High depth of field).

Fig.Macrostructure of as-cast aluminum ingot. 1.5×

7. Microscopic Examination:
 Can be carried out using optical microscopes, scanning electron
microscopes (SEM), or transmission electron microscope (TEM).
 These microscopes are specified according to:
◦ Magnification, e.g. 500X
◦ Resolution, the power to clearly show details.
◦ Depth of field: is defined as the amount of vertical
displacement of the features of a specimen that can be tolerated
without loss of focus. The depth of field in an optical
microscope is very short and is in the range of few microns.

 Optical Microscopes: They have limited resolution (1 micron).

Fig. Principle of optical microscope

There are two types of optical microscopes:

◦ Stereo Microscope: uses two separate light paths to get a true
stereo image of the specimen.

◦ Inverted Microscope: has one beam path but may split into
two paths, one for each eye

 SEM: is extensively used for investigating the micro-structure of

metallic materials. It allows resolutions of 10 to 30 nm. Useful
magnification thus extends beyond 10,000x up to 100,000x.
 Compared to optical microscopy, the depth of focus is larger by
more than two orders of magnitude.

Fig. Principle of SEM

 TEM: Can magnify objects up to 200,000x, and projects the image

onto a screen or a photographic plate. It requires thin specimens to
permit transmission of electron beams. Therefore replicas of the
fracture surface are used.

Fig. Effect of etching on microstructure

8. Selection and preparation of metallographic sections:
 Metallographic examination of polished and polished-and-etched
sections by optical microscopy and SEM is vital in failure
investigation.
 It provides indication of whether the material has the desired
structure, or abnormalities that may cause early failure (e.g. aging
of mild steel).
 The microscope may provide information regarding the method of
manufacture or heat treatment (either intentionally during
manufacture or accidentally during service).
 Revealing other service effects such as corrosion, oxidation, and
severe work hardening of surfaces.
 Revealing factors causing initiation and modes of propagation of
cracks.

9. Examination and analysis of metallographic sections:

 Microscope examination of metallographic sections reveals
inclusions, micro-structural segregation, decarburization,
carbon pickup, improper heat treatment, un-tempered white
martensite, and intergranular corrosion.
 Metallographic examination also measures some parameters as
case depth, coating thickness, grain size, and HAZ, all of which
may have a bearing on the cause of failure.

10. Determination of fracture type:

 Fractures should be classified in terms of their growth mechanism,
while crack initiation is not considered. These should be classified
to Ductile and Brittle fracture.
Ductile Fracture: extensive plastic deformation and toughness before
fracture.
Brittle Fracture: little plastic deformation and low energy absorption
before fracture.
a. Ductile Fracture:
 Involves Plastic extension, initially without necking, during which
included particles crack or de-cohesion of particle/matrix interfaces
occurs creating microvoids.
 When necking begins, triaxial stresses are set up that cause lateral
extension of the microvoids which coalesce to form a central
crack.
 Fracture of the remaining section to produce an annulus slant
fracture (shear at 45o).
 Fractography of ductile fractures reveals equiaxed dimples. Slant
fracture involving shear stress components (such as in torsion)
generates elongated dimples.

a) Necking, b) Micro-void formation, c) Micro-void coalescence

to form a crack, d) Shear of the edges, f) Fracture surfaces

Cup and cone fracture in Al. SEM Fractographs

b. Transgranular Brittle Fracture:
 Transgranular cleavage is the most commonly encountered
process of brittle fracture. It usually occurs in BCC metals and
their alloys as tungsten, molybdenum, chromium, ...etc. and
some HCP metals as zinc, magnesium, and beryllium. FCC
metals and alloys are usually immune from this mechanism.
 Iron and low carbon steel show a ductile-to-brittle transition
with decreasing temperature. Transition temperature is not a
physical constant but depends on shape and size of the specimen
and the strain rate. The catastrophic brittle fracture of ships in
heavy seas and the failure of bridges on unusually cold days are
examples.

 Metallurgical changes, especially strain aging, may cause brittle

fracture of crane hooks, and chain links after long periods of
satisfactory operation.
 Cleavage fracture produces a pattern of brightly reflecting crystal
facets, and such fractures are often described as crystalline. The
plane of fracture is normal to the axis of maximum tensile
stress.

c. Intergranular Brittle Fracture:

It may have three causes:
 The absence of sufficient deformation systems. A good example is
the fracture of polycrystalline ceramics at low temperature.
 The presence of a large area of second phase particles at the grain
boundary (carbides in iron-nickel-chromium alloys).
 The segregation of a specific element or compound to grain
boundary where a layer of few atoms thick is sufficient to cause
embrittelment (oxygen in iron or nickel, or antimony in
copper).
A. Transgranular fracture: fracture cracks pass through grains.
Fracture surface have faceted texture because of different
orientation of cleavage planes in grains.
B. Intergranular fracture: fracture crack propagation is along grain
boundaries, which are weakened or embritteled by impurities
segregation.
 Fig. V-shaped “chevron” markings Characteristic of brittle fracture

Fig. Sketch of a fractured case-hardened shaft showing chevron marks

pointing back toward the fracture origin. The interior, or core, is likely to
have a ductile fracture with dimpled rupture on the microscale.

d. Fatigue Fracture:
 Results from the application of repeated or cyclic stresses
substantially below the nominal yield strength. This is one of the
main causes of service failure.
 The most noticeable macroscopic features of fatigue fracture
surfaces are the progression of beach marks, which indicates
successive textured near the origins of advancing crack front.
Surfaces are smooth textured near the origins and generally show
slight roughening as the crack grows.
  Microscopically, surfaces of fatigue fracture are characterized by
the presence of striations.

10. Chemical Analysis:

 In specific investigations where corrosion and stress-corrosion are
involved, chemical analysis of any deposit, scale, or corrosion
products is required to assist in establishing the primary cause of
failure.
 Where analysis shows that the content of particular element is
greater than that of the specifications, it is often doubtful whether
such deviation has played a contributory role in failure.
 All techniques of chemical analysis used to determine the
elemental composition of a given material are based upon
analyzing the electron energy levels within the atom by a process
called excitation. It involves the transition of electrons between
allowed energy levels of electrons, which is accompanied by
emission of electromagnetic radiations with wavelengths
characteristic of these elements. Resolving the emitted radiation
into its constituent wavelengths gives the chemical composition.
This process is called ''Spectroscopy'', and the exciting instruments
are called Spectrometers.

 Spectroscopes are classified according to the type of excitation to:

1. Atomic absorption spectroscopes.
2. Inductively coupled plasma emission microscopes.
3. X-ray fluorescence.
4. X-ray diffraction, etc.

11. Application of Fracture Mechanics:

The concepts of fracture mechanics are useful in measuring fracture
toughness and providing a quantitative framework for evaluating
structural reliability.
12. Simulated Service Testing:
 During the concluding stages of the investigation, it may be
necessary to conduct tests that simulate he condition under which
failure is believed to have occurred.
 Simulated testing of the effects of certain selected variables may be
helpful in planning corrective action that will avoid similar failure,
or at least extended service life (evaluation of the efficiency of
special additives to lubricants to reduce wear, wind tunnels in
aircraft industries, and tank tests to evaluate hull modifications in
naval architecture).
 Most of the metallurgical phenomena involved in failures can be
satisfactory reproduced on a laboratory scale, and the information
derived from such experiments can be helpful to the investigator.

13. Analyzing The Evidence, Formulating Conclusions, and Writing

The Report:
 During investigation, it is to be expected that some of the work will
not directly assist in determining the cause of failure; nevertheless,
some negative evidence may be helpful in dismissing some causes
of failure from consideration.
 If, after investigations, the cause of failure is illusive, a search
through published reports or similar instances may be required to
suggest possible clues.
 The failure analysis report should be written clearly, concisely, and
logically. It is proposed that the report should include the following
sections:

1. Description of the failed component.

2. Service conditions at time of failure.
3. Prior service history.
4. Manufacturing and processing history of the component.
5. Mechanical and metallurgical study of failure.
6. Metallurgical evaluation of quality.
7. Summary of mechanisms that caused failure.
8. Recommendations for prevention of similar failure or for
correction of similar components in service.

 Lengthy reports should begin with abstract, and a glossary of terms

may also be helpful. Appendices serve to keep the body of the
report clear and uncluttered.