Modes of Material Failure, Fracture , Creep , Fatigue And More
When the load on a ductile material exceeds the elastic limit, it becomes permanently deformed and elastic failure is said to have occurred. The material may still be intact but it is likely that the component from which it is made will no longer be fit for its intended purpose.
fracture
Brittle materials such as cast iron, very often fail in the elastic range with the brittle types of fracture shown in Figure 1 . Brittle fracture, which is also known as cleavage fracture, is more prevalent in materials with BCC and CPH crystal lattice structures. Under certain conditions, ductile materials can
also fail with a brittle type of fracture, as will be explained.
Both kinds of failure are to be avoided by incorporating a suitable factor of safety into the design of engineering components. As a general rule, factor of safety of at least 2 should be employed on static structures. With this in place, the working stress in the material should always be less than half of that which will cause failure. That is
In spite of the best intentions of design engineers, components sometimes fail in service. Static loads can be hard to predict and dynamic loads on the component parts of machinery, motor vehicles and aircraft are very difficult to analyse. Combinations of direct loading, shearing, bending and twisting are very often present. A complex stress system is then said to exist, the resultant of which may exceed the predicted working stress and lead to failure.
An additional danger is the presence of stress concentrations in a component. These can occur at sharp internal corners, holes, fixing points and welds. They are known as stress raisers, where
the stress may exceed that at which failure occurs. Under certain conditions, cracks can spread from these points, which eventually lead to failure. These kinds of failure are usually detected at the
prototype stage and the design modified to prevent them occurring.
Material faults such as the presence of cavities, impurities, large grain size and inappropriate heat treatment can also contribute to failure if not detected by quality control procedures.
Under certain circumstances, materials can fail at comparatively low stress levels that would normally be considered to be quite safe. The main reasons for this are changes in temperature, which can affect the properties of a material, and cyclic loading. Low temperatures can cause brittleness and loss of strength. High temperatures can cause the material to creep, and eventually fail,
under loads that are well below the normal elastic limit. A material is subjected to cyclic loading when it is repeatedly being loaded and unloaded. The loads may be well below that which would be
expected to cause failure, but over a period of time, failure can occur due to metal fatigue. Some of these failure modes will now be described
Materials can fail due to metal fatigue, creep and brittle fracture at stress levels which would normally be considered safe
Brittle fracture
The plastic deformation which precedes a ductile fracture takes a finite amount of time to take place. If a load in excess of that which will cause fracture is suddenly applied, as with an impact
load, there will be insufficient time for plastic deformation to take
place and a brittle form of fracture may occur. This can be observed during an Izod or Charpy impact test where an otherwise ductile material is suddenly fractured by an impact load. Brittle, or cleavage fractures usually have a granular appearance due to the reflection of light from the individual grains figure 3 . Too large a grain size can affect the strength of a material and make it brittle. Grain growth can occur when materials are operating at high temperatures for long periods of time. Here the grains feed off each other in cannibal fashion, reducing the strength of the material and increasing the likelihood of brittle fracture.
Some metals which exhibit ductile behaviour under normal conditions become very brittle at low temperatures. The temperature at which the change occurs is called the transition temperature.
Mild steel becomes brittle at around 0 ⁰C. As can be seen from Figure 4 the transition temperature is judged to be that at which the fracture surface of an Izod or Charpy specimen is 50% granular
and brittle and 50% smooth and ductile. It is generally the
metals with BCC and CPH crystal lattice structures that are affected in this way by low temperatures. The ferrite grains in steel, which are almost pure iron, have a BCC structure at normal temperatures. Chromium, tungsten and molybdenum are also BCC and suffer from low temperature brittleness in the same way as iron.
The cracking and sinking of the all-welded liberty ships in the Second World War was attributed to brittle fracture. The cold north Atlantic temperatures produced brittleness in the steel hulls.
Cracks appeared in places of stress concentration and these quickly spread with disastrous effects. The transition temperature in mild steel is raised by the presence of phosphorous and lowered
by the addition of manganese and nickel. A relatively high level of phosphorous was present in the steel of the liberty ships and all ship hulls are now made from steel which is low in phosphorous.
NOTE :At a transition temperatures of around 0⁰C mild steel with a high phosphorous content becomes brittle. The presence of nickel and manganese lowers the transition temperature.
Creep
Creep is a form of plastic deformation which takes place over a period of time at stress levels which may be well below the yield stress of a material. It is temperature related and as a general rule, there will be little or no creep at temperatures below
where
T = is the melting point of the material measured on the Kelvin scale.
For mild steel, T= 1500 ⁰C which is 1773 K and so there should be very little creep below 0.4×1773=709 K which is 436 ⁰C. It should be stressed that this is only a general rule and that some of the softer low-melting point metals such as lead, will creep under load at normal temperatures
With the more common engineering metals, creep is a problem encountered at sustained high temperatures such as those found in steam and gas turbine plant. Under extreme conditions it can
eventually lead to failure. A typical graph of creep deformation against time is shown in Figure 6
Figure 6 shows the behaviour of a material which is above the threshold temperature at which creep is likely to occur. When it is initially loaded, the elastic extension OA is produced. If the stress in the material is below a level called the limiting creep stress or creep limit at that temperature, there will be no further extension.
If, however, the stress is above this level, primary creep AB, commences. This begins at a rapid rate, as indicated by the slope of the graph, and then decreases as work hardening sets in. It is followed by secondary creep BC, which takes place over a comparatively long period at a steady rate. The final stage is tertiary creep CD, where the deformation rate increases. Necking becomes
apparent at this stage, leading finally to fracture at D. Creep in polymer materials below the glass transition temperature is found to proceed in the same way
Increases in temperature and/or increases in stress have the effect shown in Figure 8. The rate at which the three stages of creep take place is increased and failure occurs in a shorter time.
Study of the nature of creep suggests that the plastic deformation is partly due to slipping of the planes of atoms in the grains and partly due to viscous flow at the grain boundaries. The atoms tend to pile up in an irregular fashion at the grain boundaries which would normally lead to work hardening. The high temperatures, however, have a relieving effect, and the smaller the grains the greater is the viscous flow at the grain boundaries
Creep resistance can be increased in two ways. The first is to introduce alloying elements which reduce slipping within the grains. The second is to have as course a grain structure as possible,
bearing in mind that this can lead to increased brittleness at normal temperatures. Many creep resistant materials have been developed over the last 50 years, in particular the nimonic series of
alloys which have been widely used in gas turbines
Failure due to material creep is likely to occur if a material is loaded for sustained periods above the creep limit stress and above the creep threshold temperature.
Fatigue
Fatigue failure is a phenomena which can occur in components which are subjected to cyclic loading. That is to say that they are repeatedly subjected to fluctuating or alternating stresses. Typical examples are the suspension units on motor vehicles and the connecting rods and crankshafts in internal combustion engines. The forces and vibrations set up by out-of-balance rotating parts can
also produce cyclic loading. The alternating stresses may be well below the elastic limit stress, and the material would be able to carry a static load of the same magnitude indefinitely. Failure
usually starts with a small crack which grows steadily with time. Eventually the remaining cross-sectional area of the component becomes too small to carry the repeated loads and the material
fractures.
It is found that in ferrous metals, there is a certain stress level below which fatigue failure will not occur no matter how many stress reversals take place. This is called the fatigue limit and is given the symbol SD. As a general rule for steels, the fatigue limit is about one half of the UTS of the material. The higher the stress above this value, the fewer will be the number of reversals or stress cycles before failure occurs.A typical graph of stress level against the number of cycles leading to failure for a material such as steel is shown in Figure 10. The graph is often referred to as an S–N curve
Fatigue cracks are observed to spread from points of stress concentration. Cyclic loading at stress levels above SD produces slip in the planes of atoms in the grains of a material. This results in the appearance of small extrusions and intrusions on the surface of an otherwise smooth material, as shown in Figure 11. Although the intrusions are very small, they act as stress raisers from which a fatigue crack can spread. If other stress raisers are present such as sharp internal corners, tool marks, and quench cracks from heat treatment, the process can be accelerated and these should be guarded against
The fracture surfaces of a fatigue failure have a characteristic appearance as shown in Figure 12. As the fatigue crack spreads, its two sides rub together under the action of the cyclic loading.
This gives them a burnished, mother-of-pearl appearance. Eventually, the material can no longer carry the load and fractures. The remainder of the surface, where fracture has occurred, has a
crystalline or granular appearance
There are many non-ferrous metals which do not have a fatigue limit and which will eventually fail even at very low levels of cyclic loading. Some steels, when operating in corrosive conditions,
exhibit these characteristics. The S–N graph for such materials is shown in Figure 13. Instead of a fatigue limit, they are quoted as having an endurance limit which is given the symbol SN. The
endurance limit is defined as the stress which can be sustained for a given number of loading cycles. Components made from these materials should be closely monitored, especially when used
in aircraft, and replaced at a safe time before the specified number of cycles N, has been reached.
Fatigue failure will eventually occur when components are subjected to cyclic loading at stress levels above the fatigue limit. The presence of stress concentrations accelerates the effect.
Degradation
Ferrous metals are affected by two kinds of corrosion. Low temperature or ‘wet’ corrosion is due to the presence of moisture and results in the formation of red rust. This, as you well know, is very
loose and porous. Red rust is an iron oxide formed by electrochemical action, in which the moisture acts as an electrolyte.
Adjacent areas of the metal, which have a different composition, such as the alternate layers of ferrite and cementite in the pearlite grains, become the anodes and cathodes. Corrosion occurs at the anode areas resulting in rust formation
Figure 14 shows that the ferrite layers, which are almost pure iron, become anodes and corrode to form FeOH3 which is red rust. The same kind of electrolytic action can occur between adjacent
areas which have been cold worked to a different extent. Figure 15 shows a fold in a sheet of metal which is more highly stressed than the surrounding areas. In the presence of moisture, the region in
the fold becomes an anode and corrodes. This kind of electrolytic action is called stress corrosion. It is the form from which motor vehicle panels can suffer if they are not properly protected
High temperature or ‘dry’ corrosion occurs due to a direct chemical reaction between the metal and oxygen of the atmosphere. It results in the formation of black millscale when the metal is heated for forging or for heat treatment. Millscale is another form of loose and porous iron oxide whose chemical formula is FeO (Figure 16).
As has been mentioned, the oxide films that form on the surface of non-ferrous metals and alloys are generally quite dense. Polished copper, brass and silver very soon become tarnished but once a thin
oxide film has formed, it protects the metal from further attack. Sometimes the oxide film is artificially thickened by an electrolytic process known as anodising. Aluminium alloys for outside uses such as door and window frames, are treated in this way.
Key points
- Wet corrosion of ferrous metals is an electro-chemical process where moisture acts as an electrolyte and different regions of the material become anodes and cathodes.
- Dry corrosion is a direct chemical reaction between a material and oxygen in the atmosphere.
Solvent attack
Thermosetting plastics tend to have a high resistance to solvents and it is generally thermoplastics and rubbers which are most vulnerable. The action of the solvent is to break down the Van der Waal forces and take the polymers into solution. Industrial solvents used for degreasing and for paint thinners, petrol, fuel oil, lubricating oils and greases can have this effect on some polymers
Radiation damage and ageing
The ultra-violet radiation present in sunlight can have a degrading effect on some thermoplastics and rubbers. It progressively causes oxygen atom cross-links to form between the polymers. These cause the material to become brittle and can also lead to discolouration. Ultra-violet lamps and X-rays used in industrial processes can also cause this kind of degradation. Colouring pigments are often added during the polymer forming process, and this reduces the effect. The darker colours are the most effective, black being the best of all.
The degradation of thermoplastics due to ultra-violet radiation can be slowed down by the introduction of colouring pigments.
Deterioration of ceramics
The ceramic tiles, bricks, cements and natural stone used for building degrade with time due to moisture and pollutant gases in the atmosphere. The absorption of rain-water into the surface
pores can cause deterioration in winter. When the moisture freezes, it expands and over a period of time it can cause cracking and flaking. Sulphur from flue and exhaust gases combines with moisture in the atmosphere to form sulphurous acid which falls as acid rain. This attacks many types of ceramic building material and in particular, natural stone. The refractory ceramics used to
line furnaces, and the ladles for carrying molten metal, can suffer from thermal shock if heated too quickly. Because they are poor conductors of heat, there can be a very large temperature difference
between the heated surface and the material beneath it. As a result, the expansion of the surface layer can cause flaking or spalling.
Refractory linings can also be attacked at high temperatures by the slag which rises to the surface of molten metal. There are two types of slag which form, depending on the impurities present in the metal. One is acidic and the other is known as basic slag. Linings of silica brick are resistant to acidic slag whilst linings of dolomite, which contains calcium and magnesium carbonates, and magnesite, which contains magnesium oxide, are resistant to basic slag.
When the load on a ductile material exceeds the elastic limit, it becomes permanently deformed and elastic failure is said to have occurred. The material may still be intact but it is likely that the component from which it is made will no longer be fit for its intended purpose.
Brittle materials such as cast iron, very often fail in the elastic range with the brittle types of fracture shown in Figure 1 . Brittle fracture, which is also known as cleavage fracture, is more prevalent in materials with BCC and CPH crystal lattice structures. Under certain conditions, ductile materials can
also fail with a brittle type of fracture, as will be explained.
 |
| FIG 1 |
Both kinds of failure are to be avoided by incorporating a suitable factor of safety into the design of engineering components. As a general rule, factor of safety of at least 2 should be employed on static structures. With this in place, the working stress in the material should always be less than half of that which will cause failure. That is
 |
| FIG 2 |
An additional danger is the presence of stress concentrations in a component. These can occur at sharp internal corners, holes, fixing points and welds. They are known as stress raisers, where
the stress may exceed that at which failure occurs. Under certain conditions, cracks can spread from these points, which eventually lead to failure. These kinds of failure are usually detected at the
prototype stage and the design modified to prevent them occurring.
Material faults such as the presence of cavities, impurities, large grain size and inappropriate heat treatment can also contribute to failure if not detected by quality control procedures.
Under certain circumstances, materials can fail at comparatively low stress levels that would normally be considered to be quite safe. The main reasons for this are changes in temperature, which can affect the properties of a material, and cyclic loading. Low temperatures can cause brittleness and loss of strength. High temperatures can cause the material to creep, and eventually fail,
under loads that are well below the normal elastic limit. A material is subjected to cyclic loading when it is repeatedly being loaded and unloaded. The loads may be well below that which would be
expected to cause failure, but over a period of time, failure can occur due to metal fatigue. Some of these failure modes will now be described
Materials can fail due to metal fatigue, creep and brittle fracture at stress levels which would normally be considered safe
Brittle fracture
The plastic deformation which precedes a ductile fracture takes a finite amount of time to take place. If a load in excess of that which will cause fracture is suddenly applied, as with an impact
load, there will be insufficient time for plastic deformation to take
place and a brittle form of fracture may occur. This can be observed during an Izod or Charpy impact test where an otherwise ductile material is suddenly fractured by an impact load. Brittle, or cleavage fractures usually have a granular appearance due to the reflection of light from the individual grains figure 3 . Too large a grain size can affect the strength of a material and make it brittle. Grain growth can occur when materials are operating at high temperatures for long periods of time. Here the grains feed off each other in cannibal fashion, reducing the strength of the material and increasing the likelihood of brittle fracture.
 |
| FIG 3 |
Mild steel becomes brittle at around 0 ⁰C. As can be seen from Figure 4 the transition temperature is judged to be that at which the fracture surface of an Izod or Charpy specimen is 50% granular
and brittle and 50% smooth and ductile. It is generally the
metals with BCC and CPH crystal lattice structures that are affected in this way by low temperatures. The ferrite grains in steel, which are almost pure iron, have a BCC structure at normal temperatures. Chromium, tungsten and molybdenum are also BCC and suffer from low temperature brittleness in the same way as iron.
 |
| FIG 4 |
Cracks appeared in places of stress concentration and these quickly spread with disastrous effects. The transition temperature in mild steel is raised by the presence of phosphorous and lowered
by the addition of manganese and nickel. A relatively high level of phosphorous was present in the steel of the liberty ships and all ship hulls are now made from steel which is low in phosphorous.
NOTE :At a transition temperatures of around 0⁰C mild steel with a high phosphorous content becomes brittle. The presence of nickel and manganese lowers the transition temperature.
 |
| FIG 5 DUCTILE FRACTURE |
Creep
Creep is a form of plastic deformation which takes place over a period of time at stress levels which may be well below the yield stress of a material. It is temperature related and as a general rule, there will be little or no creep at temperatures below
0.4×T
where
T = is the melting point of the material measured on the Kelvin scale.
For mild steel, T= 1500 ⁰C which is 1773 K and so there should be very little creep below 0.4×1773=709 K which is 436 ⁰C. It should be stressed that this is only a general rule and that some of the softer low-melting point metals such as lead, will creep under load at normal temperatures
With the more common engineering metals, creep is a problem encountered at sustained high temperatures such as those found in steam and gas turbine plant. Under extreme conditions it can
eventually lead to failure. A typical graph of creep deformation against time is shown in Figure 6
 |
| FIG 6 |
Figure 6 shows the behaviour of a material which is above the threshold temperature at which creep is likely to occur. When it is initially loaded, the elastic extension OA is produced. If the stress in the material is below a level called the limiting creep stress or creep limit at that temperature, there will be no further extension.
If, however, the stress is above this level, primary creep AB, commences. This begins at a rapid rate, as indicated by the slope of the graph, and then decreases as work hardening sets in. It is followed by secondary creep BC, which takes place over a comparatively long period at a steady rate. The final stage is tertiary creep CD, where the deformation rate increases. Necking becomes
apparent at this stage, leading finally to fracture at D. Creep in polymer materials below the glass transition temperature is found to proceed in the same way
 |
| FIG 7 |
Increases in temperature and/or increases in stress have the effect shown in Figure 8. The rate at which the three stages of creep take place is increased and failure occurs in a shorter time.
Study of the nature of creep suggests that the plastic deformation is partly due to slipping of the planes of atoms in the grains and partly due to viscous flow at the grain boundaries. The atoms tend to pile up in an irregular fashion at the grain boundaries which would normally lead to work hardening. The high temperatures, however, have a relieving effect, and the smaller the grains the greater is the viscous flow at the grain boundaries
 |
| FIG 8 |
bearing in mind that this can lead to increased brittleness at normal temperatures. Many creep resistant materials have been developed over the last 50 years, in particular the nimonic series of
alloys which have been widely used in gas turbines
Failure due to material creep is likely to occur if a material is loaded for sustained periods above the creep limit stress and above the creep threshold temperature.
Fatigue
Fatigue failure is a phenomena which can occur in components which are subjected to cyclic loading. That is to say that they are repeatedly subjected to fluctuating or alternating stresses. Typical examples are the suspension units on motor vehicles and the connecting rods and crankshafts in internal combustion engines. The forces and vibrations set up by out-of-balance rotating parts can
also produce cyclic loading. The alternating stresses may be well below the elastic limit stress, and the material would be able to carry a static load of the same magnitude indefinitely. Failure
usually starts with a small crack which grows steadily with time. Eventually the remaining cross-sectional area of the component becomes too small to carry the repeated loads and the material
fractures.
 |
| FIG 9 |
It is found that in ferrous metals, there is a certain stress level below which fatigue failure will not occur no matter how many stress reversals take place. This is called the fatigue limit and is given the symbol SD. As a general rule for steels, the fatigue limit is about one half of the UTS of the material. The higher the stress above this value, the fewer will be the number of reversals or stress cycles before failure occurs.A typical graph of stress level against the number of cycles leading to failure for a material such as steel is shown in Figure 10. The graph is often referred to as an S–N curve
 |
| FIG 10 |
Fatigue cracks are observed to spread from points of stress concentration. Cyclic loading at stress levels above SD produces slip in the planes of atoms in the grains of a material. This results in the appearance of small extrusions and intrusions on the surface of an otherwise smooth material, as shown in Figure 11. Although the intrusions are very small, they act as stress raisers from which a fatigue crack can spread. If other stress raisers are present such as sharp internal corners, tool marks, and quench cracks from heat treatment, the process can be accelerated and these should be guarded against
 |
| FIG 11 |
This gives them a burnished, mother-of-pearl appearance. Eventually, the material can no longer carry the load and fractures. The remainder of the surface, where fracture has occurred, has a
crystalline or granular appearance
 |
| FIG 12 |
There are many non-ferrous metals which do not have a fatigue limit and which will eventually fail even at very low levels of cyclic loading. Some steels, when operating in corrosive conditions,
exhibit these characteristics. The S–N graph for such materials is shown in Figure 13. Instead of a fatigue limit, they are quoted as having an endurance limit which is given the symbol SN. The
endurance limit is defined as the stress which can be sustained for a given number of loading cycles. Components made from these materials should be closely monitored, especially when used
in aircraft, and replaced at a safe time before the specified number of cycles N, has been reached.
 |
| FIG 13 |
Degradation
Ferrous metals are affected by two kinds of corrosion. Low temperature or ‘wet’ corrosion is due to the presence of moisture and results in the formation of red rust. This, as you well know, is very
loose and porous. Red rust is an iron oxide formed by electrochemical action, in which the moisture acts as an electrolyte.
Adjacent areas of the metal, which have a different composition, such as the alternate layers of ferrite and cementite in the pearlite grains, become the anodes and cathodes. Corrosion occurs at the anode areas resulting in rust formation
 |
| FIG 14 |
Figure 14 shows that the ferrite layers, which are almost pure iron, become anodes and corrode to form FeOH3 which is red rust. The same kind of electrolytic action can occur between adjacent
areas which have been cold worked to a different extent. Figure 15 shows a fold in a sheet of metal which is more highly stressed than the surrounding areas. In the presence of moisture, the region in
the fold becomes an anode and corrodes. This kind of electrolytic action is called stress corrosion. It is the form from which motor vehicle panels can suffer if they are not properly protected
 |
| FIG 15 |
 |
| FIG 16 |
oxide film has formed, it protects the metal from further attack. Sometimes the oxide film is artificially thickened by an electrolytic process known as anodising. Aluminium alloys for outside uses such as door and window frames, are treated in this way.
Key points
- Wet corrosion of ferrous metals is an electro-chemical process where moisture acts as an electrolyte and different regions of the material become anodes and cathodes.
- Dry corrosion is a direct chemical reaction between a material and oxygen in the atmosphere.
 |
| FIG 17 |
Solvent attack
Thermosetting plastics tend to have a high resistance to solvents and it is generally thermoplastics and rubbers which are most vulnerable. The action of the solvent is to break down the Van der Waal forces and take the polymers into solution. Industrial solvents used for degreasing and for paint thinners, petrol, fuel oil, lubricating oils and greases can have this effect on some polymers
 |
| FIG 18 |
Radiation damage and ageing
The ultra-violet radiation present in sunlight can have a degrading effect on some thermoplastics and rubbers. It progressively causes oxygen atom cross-links to form between the polymers. These cause the material to become brittle and can also lead to discolouration. Ultra-violet lamps and X-rays used in industrial processes can also cause this kind of degradation. Colouring pigments are often added during the polymer forming process, and this reduces the effect. The darker colours are the most effective, black being the best of all.
 |
| FIG 19 |
The degradation of thermoplastics due to ultra-violet radiation can be slowed down by the introduction of colouring pigments.
Deterioration of ceramics
The ceramic tiles, bricks, cements and natural stone used for building degrade with time due to moisture and pollutant gases in the atmosphere. The absorption of rain-water into the surface
pores can cause deterioration in winter. When the moisture freezes, it expands and over a period of time it can cause cracking and flaking. Sulphur from flue and exhaust gases combines with moisture in the atmosphere to form sulphurous acid which falls as acid rain. This attacks many types of ceramic building material and in particular, natural stone. The refractory ceramics used to
line furnaces, and the ladles for carrying molten metal, can suffer from thermal shock if heated too quickly. Because they are poor conductors of heat, there can be a very large temperature difference
between the heated surface and the material beneath it. As a result, the expansion of the surface layer can cause flaking or spalling.
Refractory linings can also be attacked at high temperatures by the slag which rises to the surface of molten metal. There are two types of slag which form, depending on the impurities present in the metal. One is acidic and the other is known as basic slag. Linings of silica brick are resistant to acidic slag whilst linings of dolomite, which contains calcium and magnesium carbonates, and magnesite, which contains magnesium oxide, are resistant to basic slag.
 |
| FIG 20 |
manha
ReplyDeletemanha