Classification of corrosion according to the appearance of the corroded surface
Not only does corrosion act by forming rust deposits or causing loss of lustre in metal, but it may also cause cracks and diminish the metal’s resistance and ductility.
1) Uniform Corrosion
It is the most common type of corrosion, and the easiest to control. It is shown by a constant thinning on the entire corroded surface. When designing equipment, additional thickness must be taken into account, in allowance of this phenomenon. This type of corrosion is frequent in high-conductivity environments such as sea-water. It is also common in acid environments, and in the case of amphoteric metals, in alkaline solutions.
| Thickness loss may be indicated according to area and time. Some used units are: | |
| ipy: | Inches per year |
| mdd: | Milligrams by square decimetres per day |
| mpy: | Millimetres per year. |
In order to change units, we need to know the density of the metal.
It is important to specify the duration of the study, since corrosion speed will vary along the way, which might lead to errors in time specifics, extrapolating a much longer time frame than that of the study itself.
If the attack is uniform, we may accept the following criterion:
1) 0.13 mpy: Very good corrosion resistance. These metals may be used in elements in which dimensions play a critical role, such as valve heads, pump axes, etc.
2) 0.13 to 1.30 mpy: Satisfactory. Metals apt for receiving pipes, valve bodies, etc.
3) 1,3 mpy: Not Satisfactory.
2) Localized corrosion
2.1) Macroscopic
2.1, a) Galvanic Corrosion
Galvanic corrosion takes place when we place two elements with different galvanic potential in contact with each other. It is purely electrochemical. We may state as examples copper pipes in contact with galvanized tanks, passive stainless steel in contact with active stainless steel, etc. Galvanic corrosion is the thinning of the anode by exclusive action of the circulating flow (Faraday’s law). Local corrosion batteries may also exist in the anode. This type of corrosion is dangerous, since the results of corrosion precipitate far away from their origin, without diminishing the speed of the entire process.
2, 1, b) Erosion Corrosion
We may define erosion as the destruction of a material by the abrasive action of gas or liquid; usually accelerated by the presence of solid particles in suspension. We know the definition of corrosion.
When the fluid that causes erosion is also corrosive to the metal under study, the deterioration speed is higher than when each phenomenon takes place separately. We then speak of erosion corrosion.
It is hard to determine the proportional effect of these two phenomena. We may generally state that mechanical factors (erosion), cause the passive film to rupture, or prevent the appearance of oxide formations which would slower the speed of corrosion.
2.1.c) Crevice Corrosion
The existence of areas of difficult access for a fluid (in sharp re-entrant corners, incomplete weld penetration or overlapping surfaces) imply the possibility of finding totally different ionic concentrations in that specific area than in the rest of the solution.
These conditions might generate an inhibiting ionic force (Cr2O7), or an extremely low ph, or a decrease in O2 concentration. This last case is extremely damaging for stainless steels and the other so-called passivable materials. In these cases, the area becomes anodic in relation to the rest of the surface.
This type of corrosion affects surfaces exposed to corrosive atmospheric conditions, since water remains inside the crevices for extended periods of time.
2.1.d) Exfoliation Corrosion
It is a selective type of corrosion which starts to show on the surface, and then extends beneath it. The final aspect is that of scales, or blisters. It is a common phenomenon in A1 and in alloys of 80/20 and 70/30 Cu-Ni. The alloy 90/10 Cu-Ni is immune to this type of corrosion.
2.1.e) Selective lixiviation or Dealloying
Dealloying occurs when one or more components of an alloy are more susceptible to corrosion than the rest, and are dissolved, leaving a porous residue which retains the original shape. The most important example of dealloying is the removal of zinc from brass, known as dezincification Cu-Zn).
2.1.f) Pitting Corrosion
This type of corrosion may present itself in any type of metal, but it is often found in situations where resistance against general corrosion is conferred by passive surface films.
Alloys most susceptible to pitting corrosion are usually the ones where corrosion resistance is caused by a passivation layer: stainless steels, nickel alloys, aluminum alloys. Metals that are susceptible to uniform corrosion in turn do not tend to suffer from pitting. This kind of corrosion is extremely insidious, as it causes little loss of material with small effect on its surface, while it damages the deep structures of the metal.
Theory of Pitting formation
| Nowadays, this phenomenon is universally accepted to be electrochemical, and divided in two stages: |
| Formation of localized corrosion points. |
| 2) Development of the attack. |
The factors which determine the propensity of certain areas to pitting are not yet fully known. It is accepted, however, that a rupture takes place in the passivation layer which covers the metal surface. This area becomes anodic while the area with excess of oxygen becomes cathodic, protecting contiguous areas and limiting the extent of the attack; leading to very localized galvanic corrosion. The corrosion area tends to bury into the mass of the metal, with limited diffusion of ions, further pronouncing the localized lack of oxygen. The rupture of the passivation layer is produced at a certain potential (pitting potential) value, in which a chemical absorption of ions is produced. From the alloy elements commonly found in stainless steels, those which present greater resistance to pitting are: Cr, Ni, Mo, and Si. The most advisable treatment to reduce pitting risk is water cooling from 1,050 to 1,100ºC.
2) Localized Corrosion
2.2) Microscopic
2.2 a) Intergranular Corrosion
This type of corrosion, to which austenitic stainless steels are particularly predisposed (18/8 and 18/8/Mo), can be detected when the material is porous, fragile and presents great crevices.
This type of corrosion attacks austenitic stainless steels which have been subject for a certain amount of time slow cooling in the 500 to 900ºC (800 to 1650F) temperature range, depending on alloy composition, and are then exposed to mildly aggressive environments.
The steel whose corrosion resistance is reduced in this manner is called sensitized steel. The environments which are capable of producing intergranular corrosion in sensitized steel are sea water, crud oil, chloride organic solvents, ascetic acid, chromic acid, ferric chloride, sulfuric and nitric acids and their combinations, among others.
This phenomenon was attributed to carbide precipitation, and for a long time, this was thought to be the only possible cause of intergranular corrosion. Then it was proven that this type of attack can also take place in non- sensitized steel, in nitrous environments at boiling points with oxidizing ions (Cr6+, Fe3+, V5+, etc.), or in K(OH) to 40-50% at 250°C.
The most widely accepted theory to explain intergranular corrosion is that of chromium depletion on grain boundaries.
When austenitic stainless steels are heated or cooled through the temperature range 1.050-1,100ºC (in which carbon solubility is of 0.2%), chromium tends to combine with carbon to form chromium carbides. The carbides precipitate, according to this decrease in solubility, mainly at grain boundaries depleting chromium from the adjacent areas. As chromium diffusion speed is much lower than that of carbon, many more chromium atoms will be consumed in grain boundaries, lowering the chromium percentage to less than 13%. In these conditions, the steel adopts an active potential compared to the rest of the grain which remains passive. An electrochemical reaction takes place, in which the grain boundary becomes anodic.
If the steel is cooled quickly passing through the sensitization temperature range, carbide precipitation will not occur, and the steel alloy will not be affected by this type of corrosion.
If on the contrary, the steel is kept within the sensitization temperature range for a long enough period of time, the carbides will precipitate, and chromium depletion will take place.
We must mention that, according to the TTT diagram, if the steel is kept within the sensitization range for a long enough period of time, the carbides may re-dissolve. Although this is not a practical or economical solution, it is possible (the time-temperature intersection must occur off the curve).
These effects are illustrated on the TTT (temperature-time-transformation) diagrams, where we may determine the time during which steel is sensitized according to each temperature.
To avoid this situation, carbon content in the alloy is reduced to levels lower than 0.03%, to obtain steel alloys such as L: 304L; 316L; 317L; etc
2. 2, b) Stress Corrosion Cracking
This type of corrosion takes place under tensile stresses and a corrosive environment, causing cracks and breaking of the material. Tensile stress sources may be residual or current (static or dynamic), or both. In the case of dynamic corrosion we shall discuss corrosion fatigue cracking.
Static tensile stresses may be internal (due to cold working, heat treatment, internal structural changes involving volume changes, etc), or external. Generally, the necessary stress values to trigger this reaction are close to the fatigue limit of the material. The environments in which this type of attack takes place are those containing metal haloids, even in low concentrations, in temperatures over 70ºC.
Other possible environments for this reaction to take place are ethylic chloride with water, sea water, acidic sulphite lye, caustic soda, water in combination with SH2 polytonic acids, and water vapor.
As a general rule, the less concentrated the corrosive agent, the lowest susceptibility to breakage.
In the case of ph values over 4.5, the risk in almost null. Austenitic steels are particularly sensitive to stress corrosion when the nickel content is low (8%). Ferritic steels show great resistance to this type of attack, but not to other types of corrosion.
Lately, a Ferritic-austenitic alloy has been developed, which shows excellent resistance to stress corrosion, and has a performance like that of a AISI 316L before other types of attacks. Another alloy which resists stress corrosion very well is austenitic, with high nickel content (more than 25%)
Theories
| The mechanism is justified in the following conditions: |
| 1. There must be a preferential attack area. This area might be grain boundary, (inter-crystalline stress corrosion), a sliding plane, or inter-metallic phases precipitation areas (trans-crystalline stress corrosion). These areas become anodic to the rest of the structure. |
| 2. The material must be subjected to stress conditions close to its yield limit, which tend to break apart the metal in the above mentioned areas. |
| 3. Existence of a specific corrosive environment. |
Under these conditions, a micro-crevice is produced, at the bottom of which stress accumulates, exposing the material to the corrosive environment.
To lower the risk of this type of corrosion, it is preferable to avoid tensile stress, both internal and external. In this sense, we must pay special attention to cooling speed after thermal treatment; avoid building equipment with alloys that combine different thermal dilation coefficient, etc.
We must also prevent the formation of stagnant areas, where aggressive ionic concentration (C1-) may increase.
Cathodic protection and inhibitors such as nitrates, phosphates, etc. are advisable.
Fatigue Corrosion Cracking
A metal’s fatigue limit is the value of the stress force which can be applied alternatively without causing cracking, even in an infinite number of cycles. For a value higher than that, cracking takes place after a certain number of cycles.
If the material is in contact with a corrosive environment, there won’t be a fatigue limit, and cracking will take place after a lower number of cycles (under the same stress) than it would in the absence of such an environment.
The damage caused is generally greater that that caused by fatigue or corrosion separately. The cracks are transgranular, branched, and often start in surface pitting.
Steel suffers this attack in natural fresh water, sea water, concentrated combustion products, etc.
To reduce this risk, cathodic protection and inhibitors are advisable.