The single most important property of stainless steels, and the reason for their existence and widespread use, is their corrosion resistance. Before looking at the properties of the various stainless steels, a short introduction to corrosion phenomena is appropriate. In spite of their image, stainless steels can suffer both "rusting" and corrosion if they are used incorrectly.
PASSIVITY
The reason for the good corrosion resistance of stainless steels is that they form a very thin, invisible surface film in oxidising environments. This film is an oxide that protects the steel from attack in an aggressive environment. As chromium is added to a steel, a rapid reduction in corrosion rate is observed to around 10% because of the formation of this protective layer or passive film. In order to obtain a compact and continuous passive film, a chromium content of at least 11% is required. Passivity increases fairly rapidly with increasing chromium content up to about 17% chromium. This is the reason why many stainless steels contain 17-18% chromium.
The most important alloying element is therefore chromium, but a number of other elements such as molybdenum, nickel and nitrogen also contribute to the corrosion resistance of stainless steels. Other alloying elements may contribute to corrosion resistance in particular environments - for example copper in sulphuric acid or silicon, cerium and aluminium in high temperature corrosion in some gases.
A stainless steel must be oxidised in order to form a passive film; the more aggressive the environment the more oxidising agents are required. The maintenance of passivity consumes oxidising species at the metal surface, so a continuous supply of oxidising agent to the surface is required. Stainless steels have such a strong tendency to passivate that only very small amounts of oxidising species are required for passivation. Even such weakly oxidising
environments as air and water are sufficient to passivate stainless steels. The passive film also has the advantage, compared to for example a paint layer, that it is self-healing. Chemical or mechanical damage to the passive film can heal or repassivate in oxidising environments. It is worth noting that stainless steels are most suitable for use in oxidising neutral or weakly reducing environments. They are not particularly suitable for strongly reducing
environments such as hydrochloric acid. Corrosion can be roughly divided into aqueous corrosion and high temperature corrosion:
· Aqueous corrosion refers to corrosion in liquids or moist environments at temperatures up to 300 oC, usually in
water-based environments.
· High temperature corrosion denotes corrosion in hot gases at temperatures up to 1300 oC.
The following sections contain a brief description of the various forms of aqueous and high temperature corrosion,
the factors which affect the risk for attack and the effect of steel composition on corrosion resistance.
AQUEOUS CORROSION
The term aqueous corrosion refers to corrosion in liquids or moist gases at relatively low temperatures, less than
300 oC. The corrosion process is electrochemical and requires the presence of an electrolyte in the form of a liquid
or a moisture film. The most common liquids are of course water-based solutions.
General corrosion
This type of corrosion is characterised by a more or less even loss of material from the whole surface or relatively
large parts of it. This is similar to the rusting of carbon steels.
General corrosion occurs if the steel does not have sufficiently high levels of the elements which stabilise the
passive film. The surrounding environment is then too aggressive for the steel. The passive film breaks down over
the whole surface and exposes the steel surface to attack from the environment.
General corrosion of stainless steels normally only occurs in acids and hot caustic solutions and corrosion
resistance usually increases with increasing levels of chromium, nickel and molybdenum. There are, however,
some important exceptions to this generalisation. In strongly oxidising environments such as hot concentrated
nitric acid or chromic acid, molybdenum is an undesirable alloying addition.
The aggressivity of an environment normally increases with increasing temperature, while the effect of
concentration is variable. A concentrated acid may be less aggressive than a more dilute solution of the same acid.
A material is generally considered resistant to general corrosion in a specific environment if the corrosion rate is
below 0.1 mm/year. The effect of temperature and concentration on corrosion in a specific environment is usually
presented as isocorrosion diagrams, such as that shown in Figure 5. In this context it is, however, important to
note that impurities can have a marked effect on the aggressivity of the environment (see Figure ).
From the isocorrosion diagram in Figure above it is apparent that the aggressivity of sulphuric acid increases with
increasing temperature, also that the aggressivity is highest for concentrations in the range 40-70%. Concentrated
sulphuric acid is thus less aggressive than more dilute solutions. The grade ‘904L’, with the composition 20Cr-
25Ni-4.5Mo-1.5Cu, exhibits good corrosion resistance even in the intermediate concentration range. This steel
was specifically developed for use in sulphuric acid environments.
The effect of the alloying elements may be demonstrated more clearly in another way. In Figure 6 the limiting
concentrations in sulphuric acid, i.e. the highest concentration that a specific steel grade will withstand without
losing passivity, are shown for various stainless steels. The beneficial effect of high levels of chromium, nickel and
molybdenum is apparent, as is the effect of copper in this environment.
The aggressivity of any environment may be changed appreciably by the presence of impurities. The impurities
may change the environment towards more aggressive or towards more benevolent conditions depending on the
type of impurities or contaminants that are present. This is illustrated in Figure 7 where the effect of two different
contaminants, chlorides and iron, on the isocorrosion diagram of 316L(hMo) in sulphuric acid is shown. As can be
clearly seen from the diagram, even small amounts of another species may be enough to radically change the
environment. In practice there is always some impurities or trace compounds in most industrial environments.
Since much of the data in corrosion tables is be based on tests in pure, uncontaminated chemical and solutions, it
is most important that due consideration is taken of any impurities when the material of construction for a certain
equipment is considered.
Pitting and crevice corrosion
Like all metals and alloys that relay on a passive film for corrosion resistance, stainless steels are susceptible to
localised corrosion. The protective passive film is never completely perfect but always contains microscopic
defects, which usually do not affect the corrosion resistance. However, if there are halogenides such as chlorides
present in the environment, these can break down the passive film locally and prevent the reformation of a new
film. This leads to localised corrosion, i.e. pitting or crevice corrosion. Both these types of corrosion usually occur
in chloride-containing aqueous solutions such as sea water, but can also take place in environments containing
other halogenides.
Pitting is characterised by more or less local points of attack with considerable depth and normally occurs on free
surfaces. Crevice corrosion occurs in narrow, solution-containing crevices in which the passive film is more readily
weakened and destroyed. This may be under washers, flanges, deposits or fouling on the steel surface. Both forms
of corrosion occur in neutral environments, although the risk for attack increases in acidic solutions.
Chromium, molybdenum and nitrogen are the alloying elements that increase the resistance of stainless steels to
both pitting and crevice corrosion. Resistance to localised corrosion in sea water requires 6% molybdenum or
more.
One way of combining the effect of alloying elements is via the so-called Pitting Resistance Equivalent (PRE)
which takes into account the different effects of chromium, molybdenum and nitrogen. There are several equations
for the Pitting Resistance Equivalent, all with slightly different coefficients for molybdenum and nitrogen. One of
the most commonly used formula is the following:
PRE = %Cr + 3.3 x %Mo + 16 x %N
This formula is almost always used for the duplex steels but it is also sometimes applied to austenitic steels.
However, for the latter category the value of the coefficient for nitrogen is also often set to 30, while the other
coefficients are unchanged. This gives the following formula:
PRE = %Cr + 3.3 x %Mo + 30 x %N
The difference between the formulas is generally small but the higher coefficient for nitrogen will give a difference
in the PRE-value for the nitrogen alloyed grades.
The effect of composition can be illustrated by plotting the critical pitting temperature (CPT) in a specific
environment against the PRE-values for a number of steel grade, see Figure 10. The CPT values are the lowest
temperatures at which pitting corrosion attack occurred during testing.
Critical pitting temperature (CPT) in 1 M NaCl as a function of PRE values.
Since the basic corrosion mechanism is the same for both pitting and crevice corrosion, the same elements are
beneficial in combatting both types of corrosion attack. Due to this there is often a direct correlation between the
CPT- and CCT-values for a certain steel grade. Crevice corrosion is the more severe of the two types of corrosion
attack and the CCT-values are lower than the CPT-values for any stainless steel grade. This is illustrated in Figure
11 where the critical pitting temperature (CPT) and the critical crevice corrosion temperature (CCT) in 6% FeCl3
has been plotted against the PRE-values for a number of stainless steels. Again the CPT and CCT values are the
lowest temperatures at which corrosion attack occurred.
Critical pitting temperature (CPT) and critical crevice corrosion temperature (CCT) in ferric chloride for various stainless steels.
As can be seen from the diagrams in Figures 10 and 11 there is a relatively good correlation between the PREvalues
and the CPT and CCT. Consequently the PRE-value can be used to group steel grades and alloys into
rough groups of materials with similar resistance to localised corrosion attack, in steps of 10 units in PRE-value or
so. However, it can not be used to compare or separate steel grades or alloys with almost similar PRE values.
Finally, it must be emphasised that all diagrams of this type show comparisons between steel grades and are only
valid for a given test environment. Note that the steel grades have different CPT’s in NaCl (Figure 10) and FeCl3
(Figure 11). The temperatures in the diagrams cannot therefore be applied to other environments, unless there
exists practical experience that shows the relation between the actual service conditions and the testing conditions.
The relative ranking of localised corrosion resistance is, however, often the same even in other environments. The
closer the test environment is to the “natural” environment, i.e. the closer the test environment simulates the
principal factors of the service environment, the more can the data generated in it be relied on when judging the
suitability of a certain steel grade for a specific service environment. A test in sodium chloride is consequently
better than a test in ferric chloride for judging whether or not a certain grade is suitable for one of the neutral pH,
chloride containing water solutions which are common in many industries.
In order to obtain a good resistance to both pitting and crevice corrosion, it is necessary to choose a highly alloyed
stainless steel with a sufficiently high molybdenum content. However, choosing the appropriate steel grades is not
the only way to minimise the risk for localised corrosion attack. The risk for these types of corrosion attack can be
reduced at the design stage by avoiding stagnant conditions and narrow crevices. The designer can thus minimise
the risk for pitting and crevice corrosion both by choosing the correct steel grade and by appropriate design of the
equipment.
Stress corrosion cracking
This type of corrosion is characterised by the cracking of materials that are subject to both a tensile stress and a
corrosive environment. The environments which most frequently causes stress corrosion cracking in stainless
steels are aqueous solutions containing chlorides. Apart from the presence of chlorides and tensile stresses, an
elevated temperature (>60°C) is normally required for stress corrosion to occur in stainless steels. Temperature is
a very important parameter in the stress corrosion cracking behaviour of stainless steel and cracking is rarely
observed at temperatures below 60 oC. However, chloride-containing solutions are not the only environments that
can cause stress corrosion cracking of stainless steels. Solutions of other halogenides may also cause cracking and
caustic solutions such as sodium and potassium hydroxides can cause stress corrosion cracking at temperatures
above the boiling point. Sensitised 18-8 stainless steels are also susceptible to intergranular stress corrosion
cracking in the steam and water environments in boiling water reactors if the stress level is sufficiently high.
Cracking may also occur in high strength stainless steels, such as martensitic or precipitation hardening steels.
This type of cracking is almost always due to hydrogen embrittlement and can occur in both environments
containing sulphides and environments free of sulphides.
Stress corrosion cracking adjacent to a weld in a stainless pipe exposed to a chloride-containing environment at 100°C.
The risk for stress corrosion cracking is strongly affected by both the nickel content and the microstructure. The
effect of nickel content is apparent from Figure 13. Both high and low nickel contents give a better resistance to
stress corrosion cracking. In the case of the low nickel contents this is due to the structure being either ferritic or
ferritic-austenitic. The ferrite phase in stainless steels with a low nickel content is very resistant to stress corrosion
cracking.
For high strength steels the main factor affecting the resistance to hydrogen embrittlement is the strength. The
susceptible to hydrogen embrittlement will increase with increasing strength of the steel.
In applications in which there is a considerable danger of stress corrosion cracking, steels that either has a low or a
high nickel content should be selected. The choice could be either a ferritic or ferritic-austenitic steel or a highalloyed
austenitic steel or nickel-base alloy. Although about 40% nickel is necessary to achieve immunity to
chloride-induced stress corrosion cracking, the 20-30% nickel in steel grades such as ‘654 SMO’, ‘254 SMO’,
‘904L’ and ‘A 28 (commonly known by the Sandvik tradename SANICRO 28). is often sufficient in practice.
In this context it should, however, be noted that nickel content is not the only factor that governs resistance to
stress corrosion cracking: the entire composition of the alloy is important. Molybdenum has been found to have a
considerable effect on resistance to stress corrosion cracking. However, more than 4% molybdenum is required to
obtain a significant effect, as is apparent from a comparison of ‘904L’ and ‘254 SMO’ in Figure 14. Selecting a
stainless steel for service in an environment that can cause stress corrosion cracking cannot just be done on the
basis of nickel content.
Stress corrosion cracking can only occur in the presence of tensile stresses. The stress to which a stainless steel
may be subjected without cracking is different for different steel grades. An example of the threshold stresses for
different steel grades under severe evaporative conditions is given in Figure 15.
As can be seen in the diagram in Figure 15 high alloy austenitic stainless steels have a very high resistance to
chloride stress corrosion cracking in contrast to the lower alloyed grades of this category.
In this type of diagram the threshold stress level is often given as a percentage of the yield strength at a certain
temperature, here 200 oC, which is related to the testing temperature. Due to the varying strengths of the different
steel grades the actual maximum stress levels will vary. The threshold stress level gives a good indication of the
stress corrosion cracking resistance of a certain grade but an adequate safety margin must also be incorporated in
any design based on these threshold stresses. The reason for this is that the actual service conditions may deviate
from the test conditions in many ways, for example regarding maximum temperatures, chloride levels, the effect of
residual stresses, etc.
Intergranular corrosion
This type of corrosion is also called grain boundary attack and is characterised by attack of a narrow band of
material along the grain boundaries.
Intergranular corrosion is caused by the precipitation of chromium carbides in the grain boundaries. Earlier this
type of corrosion caused large problems in connection with the welding of austenitic stainless steels. If an
austenitic or ferritic-austenitic steel is maintained in the temperature range 550 - 800°C, carbides containing
chromium, iron and carbon are formed in the grain boundaries. The chromium content of the carbides can be up to
70%, while the chromium content in the steel is around 18%. Since chromium is a large atom with a low diffusion
rate, a narrow band of material around the carbides therefore becomes depleted in chromium to such an extent
that the corrosion resistance decreases. If the steel is then exposed to an aggressive environment, the chromiumdepleted
region is attacked, and the material along the grain boundaries is corroded away. The result is that grains
may drop out of the steel surface or in severe cases that the grains are only mechanically locked together as in a
jigsaw puzzle while the stiffness and strength of the material have almost disappeared. Ferritic stainless steels are
also sensitive to intergranular corrosion for the same reason as the austenitic and duplex steels, although the
dangerous temperatures are higher, generally above 900 - 950oC.
Temperatures that can lead to sensitisation, i.e. a sensitivity to intergranular corrosion, occur during welding in an
area 3-5 mm from the weld metal. They can also be reached during hot forming operations or stress relieving heat
treatments.
The risk for intergranular corrosion can be reduced by decreasing the level of free carbon in the steels. This may
be done in either of two ways:
· by decreasing the carbon content.
· by stabilising the steel, i.e. alloying with an element (titanium or niobium) which forms a more stable carbide
than chromium.
The effect of a decrease in the carbon content is most easily illustrated by a TTS-diagram (time- temperaturesensitisation),
an example of which is shown in Figure 17. The curves in the diagram show the longest time an
austenitic steel of type 18Cr-8Ni can be maintained at a given temperature before there is a danger of corrosion.
This means that for standard low-carbon austenitic steels (L-grades) the risk for intergranular corrosion cracking
is, from a practical point of view, eliminated. All high alloyed austenitic and all duplex grades intended for
aqueous corrosion service have carbon contents below 0.03% and are consequently “L-grades”. Due to the low
solubility of carbon in ferrite the carbon content will have to be extremely low in ferritic stainless steels if the risk
of intergranular corrosion is to be eliminated. In ferritic stainless steels stabilising and extra low carbon contents
are often used is to eliminate the risk for intergranular attack after welding or other potentially sensitising
treatments.
Addition of titanium or niobium to the steel, so-called stabilisation, means that the carbon is bound as titanium or
niobium carbides. Since titanium and niobium have a greater tendency to combine with carbon than does
chromium, this means that carbon is not available to form chromium carbides. The risk for intergranular corrosion
is therefore appreciably reduced. There is, however, a disadvantage associated with stabilisation. In the area
closest to a weld, the temperature during welding can be so high that titanium or niobium carbides are dissolved.
There is then a danger that they do not have time to re-precipitate before the material has cooled sufficiently to
allow the formation of chromium carbides in the grain boundaries. This leads to so-called knife line attack in
which a narrow zone of material very close to the weld suffers intergranular corrosion. Since the carbon level in
stabilised steels is often quite high (0.05-0.08%) this can give serious attack.
A sensitised microstructure can be fully restored by adequate heat treatment. In the case of austenitic and ferriticaustenitic
duplex stainless steels a full quench anneal heat treatment is necessary. For ferritic stainless steels an
annealing treatment is normally used.
It should also be mentioned that many high temperature steels, which have high carbon contents to increase the
strength, are sensitive to intergranular corrosion if they are used in aqueous environments or exposed to
aggressive condensates.
Galvanic corrosion
Galvanic corrosion can occur if two dissimilar metals are electrically connected together and exposed to a
corrosive environment. The corrosive attack increases on the less noble metal and is reduced or prevented on the
more noble metal, compared to the situation in which the materials are exposed to the same environment without
galvanic coupling.
The difference in "nobility", the ratio of the area of the noble metal to the area of the less noble metal in the
galvanic couple and the electrical conductivity of the corrosive environment are the factors that have the largest
influence on the risk for galvanic corrosion. An increase in any of these factors increases the risk that corrosion
will occur.
The risk of galvanic corrosion is most severe in sea water applications. One way of assessing whether a certain
combination of materials is likely to suffer galvanic corrosion is to compare the corrosion potentials of the two
materials in the service environment. One such "electrochemical potential series" for various materials in seawater
is given in Figure 19. The larger the difference between the corrosion potentials, the greater the risk for attack of
the less noble component; small differences in corrosion potential have a negligible effect.
Stainless steels are more noble than most of the constructional materials and can therefore cause galvanic
corrosion on both carbon steels and aluminium alloys. The risk for galvanic corrosion between two stainless steel
grades is small as long as there is not a large difference in composition such as that between AISI 410S and AISI
316 or ‘254 SMO’. Galvanic effects to be operative when one of the materials in the galvanic couple is corroding.
This means that galvanic corrosion is rarely seen on alloys that are resistant to the service environment.
HIGH TEMPERATURE CORROSION
In addition to the electrochemically-based aqueous corrosion described in the previous chapter, stainless steels can
suffer attack in gases at high temperatures. At such high temperatures there are not the distinct forms of corrosion
such as occur in solutions, instead corrosion is often divided according to the type of aggressive environment.
Some simpler cases of high temperature corrosion will be described here: oxidation, sulphur attack (sulphidation)
carbon uptake (carburization) and nitrogen uptake (nitridation). Other more complex cases such as corrosion in
exhaust gases, molten salts and chloride/fluoride atmospheres will not be treated here.
Source: Avesta SS hand book
No comments:
Post a Comment