Professional Documents
Culture Documents
Corrosion Principles PDF
Corrosion Principles PDF
Refinaria de Sines
CORROSION
PRINCIPLES
Pgina: 1 de 40
IRS-009.00
INSPECO
Refinaria de Sines
Table of Contents
Page No.
1.0
Introduction
2.0
2.1
References
2.2
Definitions
3.0
4.0
5.0
6.0
3.1
Corrosion Defined
3.2
Metal Ores
3.3
4.1
Corrosion by Water
4.2
Voltage Source
4.3
Electrical Circuits
4.4
Polarization
Nature of Metals
5.1
6.1
8.0
Conductivity
6.2
6.3
Dissolved Gases
6.3.1 Oxygen
6.4
7.0
Chloride Ions
10
Physical Variables
7.1
Temperature
10
10
7.2
Pressure
11
7.3
Velocity
11
11
Forms of Corrosion
12
8.1
Uniform Corrosion
12
8.2
Pitting Corrosion
12
8.3
Bimetallic Corrosion
14
8.4
15
8.5
Cavitation Corrosion
15
Pgina: 2 de 40
IRS-009.00
INSPECO
Refinaria de Sines
Table of Contents
Page No.
8.6
Intergranular Corrosion
15
8.7
16
8.8
16
8.9
17
8.10
Corrosion Fatigue
18
8.11
19
8.12
Oxygen Corrosion
19
8.13
20
8.14
21
Attachments:
1.
22
2.
23
3.
24
4a.
Corrosion Cell
29
4b.
29
5.
30
6a.
Polarization Curve
31
6b.
31
7.
32
8.
33
9.
34
10.
35
Pgina: 3 de 40
IRS-009.00
INSPECO
Refinaria de Sines
1.0
INTRODUCTION
This document is intended to provide operations personnel with an overview of
the upstream oil and gas industry corrosion principles.
The reader is
encouraged to read about the subjects covered in this overview in more detail
as required.
2.0
2.1
References
Corrosion and Water Technology for Petroleum Producers, L. W. Jones, 1988
Corrosion Engineering, M. G. Fontana and N. D. Greene,1978
Recommendations for Preventing Hydrogen-Induced
Upstream Operations, EPR EPR.23PS.89
Cracking
(HIC)
in
Definitions
m
- pipe ID (inches)
- potential (volts)
HIC
HRC
- current (amps)
medium
MIC
Pgina: 4 de 40
IRS-009.00
INSPECO
Refinaria de Sines
- pressure
QG
QL
S.G.
SOHIC
SRB
SSC
SWC
- Stepwise Cracking
- temperature
Ve
VG
VL
3.0
3.1
Corrosion Defined
In its broadest sense, corrosion can be defined as the deterioration of a
substance or its properties because of a reaction with its environment.
Primarily in the oilfield, it is more specific to say the concern is with the
destruction of metal by a chemical reaction with a given environment, caused
by existence of an electrochemical mechanism. This overview specifically
addresses metallic corrosion. Deterioration of non-metallic materials is beyond
the scope of this document.
3.2
Metal Ores
Most metals are found in nature as ores which are typically metallic oxides or
sulphides. Metals are refined from ores by reduction, a forcing of the metal
ions to give up their charges and become metallic atoms in a regular lattice
array. For this to take place, energy in the form of heat or electricity must be
supplied. It is the energy stored in the metal during the refining process that
provides the driving force for corrosion.
3.3
Pgina: 5 de 40
IRS-009.00
INSPECO
Refinaria de Sines
source of energy. Attachment 1 shows the calculated free energy change for
the formation of a number of metal oxides. What these numbers mean is that,
from a free energy standpoint, aluminium will tend to form its oxide very
easily, iron almost as easily, copper only to minor extent, and gold only with an
input of energy. Since steel (iron) is our basic structural metal and its free
energy is so negative, we are continually faced with the tendency of the metal
to return to its oxide. It should be noted that the magnitude of G does not
alone determine the rate at which a metal returns to its oxide.
4.0
4.1
Corrosion By Water
Nearly all corrosion problems which occur in oilfield production operations are
due to the presence of water. In order to corrode, the metal surface must be
in contact with a water phase. For example, if a well produces at a high oil-towater ratio, very little corrosion is likely to occur because the water is mixed
with oil as an oil-external emulsion. On the other hand, in low oil-to-water
ratio wells, corrosion occurs because free water contacts the metal surface.
Corrosion in the presence of water depends on electrochemical processes.
Electric current flows and there must be a driving force and a complete
electrical circuit.
4.2
Voltage Source
The source of voltage (driving force) in the corrosion process is the energy
stored during the refining process. Due to the difference of free energy values
of various metals, these voltages are different for each metal. Potential values
are a function of both the metal and the chemical and physical characteristics
of the water. Attachment 2, the Galvanic Series, is a practical table where
common metals and alloys are arranged based on their respective potentials in
sea water. Attachment 3 is a guide that shows the probable effect of coupling
two dissimilar metals in sea water.
4.3
Electrical Circuits
The electrical circuit of the corrosion process consists of an anode, cathode,
electrolyte and metallic return path (see Attachment 4a).
The anode is that portion of the metal surface that is corroded. It is the
point at which metal dissolves, or goes into solution.
When metal
dissolves, the metal atom loses electrons and is oxidised. The reaction for
iron is:
Fe
Fe++ + 2e-
(1)
The iron ion goes into solution and two electrons are left behind in the
metal.
Pgina: 6 de 40
IRS-009.00
INSPECO
Refinaria de Sines
The cathode is that portion of the metal surface where reduction takes
place and does not dissolve. The electrons left behind by the oxidation
reaction at the anode travel through the metallic path to the cathodic
surface where they are consumed by reaction with an oxidising agent that
must be present in the electrolyte. This consumption of electrons is called
a reduction reaction. Three typical reactions are:
2H+ + 2e-
H2
(Acidic Solutions)
(2)
4H+
O2
2H2O
4e-
2H2O
(Acidic Solutions)
(3)
OR
+ 4e- 4OH- (Neutral or Alkaline Solutions) (4)
4.4
Polarization
Attachment 4a shows a typical corrosion cell on the surface of a metal. As was
explained in the previous Section, an electrical circuit is formed between the
cathode and anode, thus allowing a potential difference to be established
between them. It is this shift in potential as the corrosion current flows that is
termed polarization.
The effects of polarization are also important in galvanic (bimetallic) corrosion,
which is more fully discussed in Section 8.3. To illustrate this, consider the
experimental copper-zinc cell shown in Attachment 4b. If the circuit is open
and no current is flowing, then the electrodes each have a characteristic
potential as shown in Attachment 5. Attachment 6a shows a plot of those
potentials at zero current on a potential versus current graph.
When a small amount of current is allowed to flow through the external circuit,
Pgina: 7 de 40
IRS-009.00
INSPECO
Refinaria de Sines
the potential of the electrodes changes slightly, the copper electrode becomes
more anodic, and the zinc electrode more cathodic (as shown in Attachment
6a). As more current is allowed to flow, the potentials continue to change
(polarize) and the potential difference between the two decreases to a value
limited primarily by the resistance of the electrolyte solution (Re).
If polarization is not taking place, then there is no corrosion.
The magnitude of strength of a corrosion cell is determined by the difference in
electrical potential between the anode and the cathode. This voltage difference
is greatly influenced by the level of polarisation which may be achieved in the
corrosion circuit. Several factors may reduce the polarisability of the corrosion
circuit, such as:
increased pressure
increased temperature
IR
(5)
OR
E c - Ea
Where:
Re
IRe +
IRm
(6)
= resistance of electrolyte
Rm
= resistance of metal
Ec
Ea
Pgina: 8 de 40
IRS-009.00
INSPECO
Refinaria de Sines
Corrosion will continue, but the rate will only change if the polarisability is
altered hence producing a change in polarisation. This occurs when hydrogen
is removed from the cathode. Once an E = IR steady state is reached, further
cell action depends upon the rate of hydrogen removal from the cathode. That
rate is largely controlled by the rate at which oxygen reaches the cathode. In
oilfield operations, hydrogen removal from the cathode is usually accomplished
by one of the mechanisms discussed in paragraph 4.3 above.
5.0
NATURE OF METALS
5.1
Pgina: 9 de 40
IRS-009.00
INSPECO
Refinaria de Sines
6.0
6.1
Conductivity
The electrical resistance of typical electrolytes is usually much higher than that
of metal, therefore the resistance of the electrolyte will normally predominate
in the corrosion cell reaction. The more conductive the electrolyte, the easier
current can flow and the faster corrosion will occur. The amount of metal that
dissolves is directly proportional to the amount of current flow between anode
and cathode. For iron, one amp of current flowing for one year will result in
the loss of 20 pounds (9.1 kg) of metal. It is important to remember that
other factors will also have an impact on the corrosivity of the electrolyte, the
conductivity only determining the ease at which corrosion currents are able to
flow from anode to cathode.
6.2
-log (H+)
(7)
The greater the concentration of hydrogen ions, the more acid the solution and
the lower the pH value. Hydrogen ions (H+) make a solution acidic and,
therefore, force the pH towards zero. Hydroxyl ions (OH-) make a solution
basic or alkaline and force the pH towards 14. The following lists the pH
ranges for acidic, neutral, and alkaline conditions:
pH Range
Classification
< 7.0
Acidic
7.0
Neutral
> 7.0
Basic or Alkaline
The corrosion rate of steel usually increases as the pH of the water decreases,
Pgina: 10 de 40
IRS-009.00
INSPECO
Refinaria de Sines
Dissolved Gases
Oxygen, carbon dioxide, or hydrogen sulphide gases, when dissolved in water,
increase its corrosivity. Dissolved gases are the primary cause of most
corrosion problems in oil and gas production. The following paragraphs discuss
each gas independently, but it is important to note that corrosion rates are also
greatly influenced by physical variables such as temperature, pressure and
velocity. Similarly, the Figures referred in these paragraphs are for specific
conditions and are only intended to reflect the relative corrosion tendencies of
each.
6.3.1
Oxygen
Dissolved oxygen can cause severe corrosion at very low concentrations (less
than 100 ppb or 0.1 ppm) and if either or both CO2 and/or H2S are present, it
further increases their corrosivity. Attachment 7 is a composite graph from
results of three different studies showing corrosion rates as a function of
oxygen concentration:
The solubility of oxygen in water is a function of pressure, temperature, and
chloride content. Although it is not usually present in produced water, it is
often introduced into oilfield water handling systems through failures to
maintain oxygen free gas blankets on water handling vessels, vacuums created
by positive displacement pumps or separator dump valves and/or exposure to
the atmosphere. Water from lakes, streams, fresh water aquifers, rain or
oceans usually will be oxygen saturated. Oxygen is more soluble at high
pressures and lower temperatures and is less soluble in salt water than in fresh
water.
Oxygen accelerates corrosion under most circumstances because it is a strong
and rapid oxidising agent in cathodic reactions. It will easily combine with
electrons at the cathode and allow the corrosion reactions to proceed at a rate
limited by the rate at which oxygen can diffuse to the cathode as discussed in
Section 4.3:
O2
4H+
O2
2H2O
4e-
(3)
OR
(4)
4e-
Pgina: 11 de 40
IRS-009.00
INSPECO
Refinaria de Sines
6.3.2
Carbon Dioxide
When carbon dioxide dissolves in water, it forms carbonic acid, decreases the
pH of the water, and increases its corrosivity. Corrosion in the presence of
There are numerous
dissolved CO2 is referred to as sweet corrosion.
intermediate reactions which may be summarised as:
CO2
Fe +
(9)
H2O H2CO3
H2CO3
(Carbonic acid)
FeCO3
H2
(8)
(Iron
Carbonate)
x
(10)
Hydrogen Sulphide
Hydrogen sulphide is soluble in water at pressures and temperatures common
in oilfield operations and, when dissolved, behaves as a weak acid and usually
causes pitting. Attack due to the presence of dissolved hydrogen sulphides is
referred to as sour corrosion.
The general corrosion reaction of steel is:
H2S
Fe
H2O
FeS
H2
(11)
The iron sulphide produced generally adheres to the surface as a black scale
and is cathodic to the steel that, in the presence of water, causes local severe
corrosion in the form of deep pitting. However, in some instances, a thin iron
sulphide scale may be relatively impermeable and actually slow down the
corrosion reaction if erosion or some other mechanism does not remove the
Pgina: 12 de 40
IRS-009.00
INSPECO
Refinaria de Sines
scale.
Hydrogen sulphide can be generated by sulphate reducing bacteria (SRB).
These bacteria contribute to corrosion by their ability to flourish in the absence
of oxygen and their ability to change sulphate ions into hydrogen sulphide.
The anaerobic conditions under a colony constitute a differential aeration cell
with the bulk of the electrolyte, whereas their ability to produce hydrogen
sulphide can cause severe localized corrosion. Refer to Section 8.11.
Under certain pressure conditions, the hydrogen produced by the corrosion
reaction can diffuse into the metallic lattice to cause embrittlement and
subsequent cracking of susceptible metals.
Refer to Section 8.7 and
Attachment 8 for further discussion of this phenomenon.
6.4
Chloride Ions
The most common electrolyte in oil production is water, and one of the most
common ions in solution is the chloride ion.
The chloride ion and its
concentration has a major effect on the corrosion reactions as noted below:
7.0
Chloride anions (negatively charged ions) tend to react very easily with
cations (e.g. Fe+2) going into solution at the corrosion cell anode. These
reactions, therefore, reduce polarization by allowing more cations to come
into solution that increases the conductivity of the electrolyte.
PHYSICAL VARIABLES
The variables of temperature, pressure, and velocity need to be accounted for
when designing and implementing a corrosion control program.
Correct
application inhibitors and cathodic protection as corrosion control methods are
very dependent on these variables. Temperature and pressure are interrelated,
and the corrosivity of a system is further influenced by velocity.
7.1
Temperature
Like most chemical reactions, corrosion rates generally increase with
temperature. For example, in a system open to the atmosphere, the corrosion
rate generally increases with increasing temperature until the concentration of
dissolved gases decreases. In a closed system, this is not necessarily the case.
In addition, many metallic alloys have minimum temperature limitations to
Pgina: 13 de 40
IRS-009.00
INSPECO
Refinaria de Sines
Pressure
Pressure also affects the rates of corrosion reactions.
More gas goes into
solution as the pressure increases, which may, depending on the dissolved gas,
increase the corrosivity of the solution. The partial pressure of CO2 or H2S in a
system is calculated as follows (refer to Section 6.3.2):
partial pressure (psia pp) = Total absolute pressure (psia)
mol % CO2 or H2S
7.3
x
(12)
Velocity
Velocity has a significant effect on corrosion rates. Stagnant or low velocity
fluids usually give low general corrosion rates, but pitting rates may be high.
Corrosion rates generally increase with increasing velocity due to the
depolarising effect on the cathode.
High velocities and the presence of
suspended solids or gas bubbles can lead to erosion corrosion, impingement, or
cavitation. On the other extreme, oil, gas, or multi-phase pipelines operating
at low velocities can result in corrosion along the bottom of a pipeline. The low
flow condition is referred to as stratified or laminar flow which can be modelled
using commercially available computer programs.
7.3.1
Ve =
(13)
(m
1
) 2
=
=
=
C
C
where:
Ve
m
C
SRP
(14)
Pgina: 14 de 40
IRS-009.00
INSPECO
Refinaria de Sines
where:
S.G.
S
R
T
Z
P
=
=
=
=
=
=
= 0.012
QL
(15)
d2
=
=
=
=
=
= 60
Q G TZ
d2P
(16)
8.0
FORMS OF CORROSION
8.1
Uniform Corrosion
Uniform corrosion is a weight loss corrosion where the area of metal loss is
spread over a relatively large area so that the surface is attacked uniformly.
This type of corrosion can be common in atmospheric corrosion of industrial or
marine locations. Strong uninhibited acids will uniformly corrode steel very
rapidly. In general, however, uniform corrosion occurs gradually thus failures
take longer to occur.
Uniform corrosion can be addressed in original design criteria by adding a
corrosion allowance to design metal thickness.
It is very important to
recognize that this is not a corrosion control measure. The extra metal
thickness simply allows more time for detection and mitigation before
equipment capability or operations safety is jeopardized. Refer to Attachment
9 which lists the relative severity of average (uniform) corrosion rates.
Pgina: 15 de 40
IRS-009.00
INSPECO
Refinaria de Sines
8.2
Pitting Corrosion
Pitting is the most common cause of corrosion failures occurring in oil and gas
industry production equipment. It occurs when the metal undergoing corrosion
suffers metal loss at localized areas rather than over a large area or the entire
surface area. The entire driving force of the corrosion reaction is concentrated
at these localized areas. The corrosion rate at these areas will be many times
greater than the average corrosion rate over the entire surface. Pitting is much
more dangerous than uniform corrosion because the pitted area can become
penetrated in a short time. Refer to Attachment 9 which lists the relative
severity of corrosion pitting rates.
8.2.1
8.2.2
There are many causes of pitting in metals and alloys but, generally, they are
caused by localized differences in the metal or in the electrolyte. Localized
differences in electrolyte composition are referred to as concentration cells as
discussed below:
When metal surfaces are uniform and the electrolyte contains varying
amounts of dissolved oxygen, oxygen concentration cells occur.
An
example is a bare or poorly coated pipeline buried under a paved or
compacted road. Under the road, the soil is not well aerated, but on either
side of the road the soil is well aerated. Locally, severe attack can occur on
the pipe at the edge of the road in the transitions from the compacted
nonaerated soil to the loose, aerated soil.
A common cause of pitting in carbon steel is the formation of local cells due to
the partial breakdown or destruction of protective scales. When a corroding
metal becomes covered with a corrosion product that is dense, impermeable
and adherent, the product protects the metal from its environment. Localized
removal of this protective scale may precipitate the onset of preferential
corrosion of the exposed metal surfaces. If the scale is a conducting scale
(e.g. FeS), then damage to it could result in intense attack of the area that lost
its protective scale layer. This is the classic pit scenario. In the case of
detachment of a conducting scale, or removal of a non-conduction scale (e.g.
calcium carbonate), then there should be little attack at the damage scale area,
although the attack may be greater beneath a detached scale.
Pgina: 16 de 40
IRS-009.00
INSPECO
Refinaria de Sines
8.2.3
Stainless steels are the most susceptible ferrous alloys to pitting, primarily
because of the very properties that make them stainless. Stainless steels
derive their corrosion resistance from the formation of a thin iron chromium
oxide film and the ability to maintain that film. If the film becomes destroyed
at local areas, those areas become anodic and pitting results.
8.3
Bimetallic Corrosion
When two dissimilar metals are placed in electrical contact in electrolyte
containing an oxidizing agent, the more reactive one will corrode and the other
may either not corrode at all, or will corrode at reduced rate, depending upon
the extent of polarization of the cathode (see Attachments 3 and 4). This
coupling of dissimilar metals is referred to as a bimetallic couple or a galvanic
cell. It can be destructive, accelerating the corrosion rate of the more reactive
of the two metals.
8.3.1
The galvanic cell is used in a practical way in cathodic protection. When steel
is connected to a more reactive metal, such as magnesium, the steel is the
cathode and is protected; and the magnesium is the anode and corrodes.
8.3.2
The area principle implies that in galvanic corrosion, the cathode to anode ratio
is of great significance in a galvanic cell with a large anode and a small
cathode, corrosion is spread over a large area and very little corrosion damage
will occur. However, in a cell with a small anode and a large cathode, corrosion
is concentrated at the anode and severe corrosion damage will occur. Refer to
Attachment 3.
8.3.3
8.3.4
Pgina: 17 de 40
IRS-009.00
INSPECO
Refinaria de Sines
downhole tubing and usually occurs as sweet corrosion (See Section 8.13).
The heat required in upsetting causes the transition zone in the heated end,
or upset, to have a different grain structure from the rest of the pipe body.
A transition zone in grain structure near the upset runout, typically within
18 of the pin end, can cause a potential difference and set up a local
corrosion cell. Fully normalizing the tubing after the upset process will
return uniformity to the grain structure which will eliminate this corrosion
potential.
8.4
8.4.1
8.5
Cavitation Corrosion
Although corrosion normally plays a minor role in the rate of cavitation
damage, it is important to recognize its effect. Cavitation is the formation and
collapse of vapour bubbles in fluids because of rapid changes in pressure. It
can occur whenever the absolute pressure at a point in the liquid stream is
reduced to the vapor pressure of the fluid so that bubbles form, and this is
followed by a rapid rise in pressure resulting in bubble collapse. Cavitation
damage is the wearing away of metal from repeated impact blows from
collapse of bubbles within a fluid, such as that caused when a pumps suction is
starved.
8.6
Intergranular Corrosion
Intergranular corrosion is preferential attack of a metals grain boundaries.
Intergranular corrosion can occur in the absence of stress. It is caused by
precipitation of impurities at grain boundaries, or changes in the alloying
elements at the grain boundary areas. For example, depletion of chromium in
the grain boundary of stainless steel can be lead to intergranular corrosion.
Pgina: 18 de 40
IRS-009.00
INSPECO
Refinaria de Sines
Proper heat treatment or control of chemistry of the steel can reduce the
susceptible grain boundary constituent and render the alloy more resistant to
intergranular attack.
8.7
8.7.1
Hydrogen entry into low strength steel can result in HIC if there is a
microscopic defect in the steel such as a lamination or inclusion. HIC is
typically manifested as blisters, stepwise cracking (SWC), and/or stress
oriented HIC (SOHIC). A lamination or void in the steel provides a place for
hydrogen atoms to combine, form hydrogen gas, and result in sufficient
pressure to cause blistering. Blisters are visible on a surface when the voids
responsible for the HIC are near the surface. SWC is initiated at inclusions on
different planes through the metal wall. The initial crack propagation occurs
along the axis of the inclusion and then interconnects with cracks on other
planes to form the SWC. The term SOHIC is used to indicate that the
characteristics of HIC have been influenced by applied or residual stresses,
usually from welding.
These stresses will cause cracks to initiate and
propagate perpendicular to existing HIC.
This can result in cracks that
eventually propagate through the steel wall (see Attachment 8).
Solutions to HIC include controlling the metal chemistry to minimise the affects
of inclusions or laminations, and controlling corrosion with coating or inhibitors
to minimise the generation of hydrogen.
8.7.2
Hydrogen entry into high strength steels or steels with a hardness above 22
HRC can result in hydrogen embrittlement. A material can fail in a brittle
manner at stresses well below its nominal yield strength.
Hydrogen
embrittlement is normally limited to high strength materials due to the fact
that these materials reach tensile strengths high enough to initiate the failure
mechanism. It is often very difficult to distinguish between failures caused by
hydrogen embrittlement or stress corrosion cracking. The susceptibility to
hydrogen embrittlement increases with increasing strength and hardness.
Even though a steel contains hydrogen, no permanent damage occurs unless
sufficient stress is applied to cause the steel to crack. Therefore, in many
cases, the hydrogen can be baked out by heat treatments, restoring the
original properties of the metal.
Pgina: 19 de 40
IRS-009.00
INSPECO
Refinaria de Sines
8.8
8.9
Plain carbon steels with maximum yield strengths above 90,000 psi and
hardness above HRC22 are susceptible to sulphide stress cracking at
temperatures below 190oF (88oC). The higher the steel strength, the
shorter the time to failure. A hardness of HRC22 is not an absolute value;
alloying with other materials can cause failures below this level, and for
certain heat treatments, the permissible hardness level can be increased.
The stresses
Pgina: 20 de 40
IRS-009.00
INSPECO
Refinaria de Sines
The most common corrosive which cause stress corrosion cracking in carbon
and low alloy steels are sodium hydroxide and nitrate solutions. Chemical
inhibition and proper stress relieving are widely used to control cracking in
these environments.
8.9.2
Stress Corrosion Cracking (SCC) of high strength steels occurs in salt solutions,
moist atmospheres, and even in tap water if the steel is ultra high strength
steel. Cracking tendency increases with the strength of the steel. Experience
or testing is necessary to determine the corrosive and conditions which will
cause cracking of high strength steels.
8.9.3
The hardenable stainless steels, which include martensitic stainless steel and
precipitation hardening stainless steels, when treated to high hardness levels
have tendencies to stress corrosion crack. To reduce this tendency, the
strength of the steels can be reduced by heat treating the alloys.
8.9.4
8.10
Corrosion Fatigue
When metals are repeatedly stressed in a cyclic manner, they will fail in a
brittle manner at stresses far below the yield or tensile strength of the
material. There exists a limiting stress below which steel may be cyclically
stressed indefinitely without failure. This stress is called the endurance limit
and is always lower than the yield and tensile strengths. The fatigue life of
metal is substantially reduced when the metal is cyclically stressed in a
corrosive environment. The simultaneous occurrence of cyclic stress and
corrosion is called corrosion fatigue, and the steel no longer exhibits an
endurance limit.
8.10.1
Pgina: 21 de 40
IRS-009.00
INSPECO
Refinaria de Sines
8.10.3
Surface conditions and stress history have a large bearing on fatigue life.
Notches imposed in manufacturing or installation can decrease fatigue life.
Also, the time available for corrosion damage to occur is extremely important
in the corrosion fatigue behaviour of materials. Cycle frequency must also be
considered as a major variable in corrosion fatigue performance.
8.11
(17)
+ S-2
or simplified
(11)
H2S
H2O
Fe
FeS
(18)
2H
Pgina: 22 de 40
IRS-009.00
INSPECO
Refinaria de Sines
8.12
Oxygen Corrosion
8.12.1
Water produced with oil seldom contains dissolved oxygen; however, oxygen
corrosion can be found in downhole and surface facility equipment. It is
usually caused by careless operating techniques, faulty equipment or by
oxygen rich wash water used for desalting. The presence of CO2 or H2S
drastically increases the rate of oxygen corrosion. Attachment 7 shows the
relation of dissolved oxygen content to the corrosion rate of steel in brine.
Note that oxygen can be highly corrosive above 20 parts per billion. As
discussed in Section 6.3.1, the cathodic corrosion reaction of steel in an oxygen
environment is as follows:
O2 + 2H2O + 4e- 4OH-
(Acidic Solutions)
(3)
OR
O2 + 4H+ + 4e- 2H2O (Neutral or Alkaline Solutions)
(4)
8.12.2
8.12.3
8.12.4
Corrosion of offshore structures is divided into four areas of attack and is due
to oxygen corrosion in each zone. The mud zone, that portion below the mud
line, may suffer the least attack because of the high resistivity of the mud and
little dissolved oxygen.
This, of course, will not be the case where
microbiologically assisted corrosion can take place. The submerged zone is
always covered by water and the corrosion tends to be uniform, generally
shallow pits. The splash zone, that area which is alternatively wetted by wave
action, tends to be the most corrosive zone. The atmospheric zone may
appear dry but, especially near the water surface and on the underside of
members, the humidity is always close to 100% so the surfaces remain moist.
8.13
Sweet Corrosion
Sweet corrosion results from the presence of water containing dissolved carbon
dioxide. Dissolved carbon dioxide (CO2) in water decreases the pH of the water
and increases its corrosivity. Refer to Attachment 10. The following shows
Pgina: 23 de 40
IRS-009.00
INSPECO
Refinaria de Sines
+
+
H2O H2CO3
H2CO3 FeCO3
(Carbonic Acid)
+
H2 (Iron Carbonate)
It should be noted that the corrosion rates experienced in reality are higher
than indicated by these summary reactions.
8.13.1
Corrosion in the tubing of gas condensate wells usually takes the form of deep
pits with steep, undercut sides. This is sometimes referred to as mesa
corrosion due to the shape of the pitting profile, i.e. areas of unattacked metal
adjacent to pitted areas.
The pits may penetrate the wall
completely in a relatively short period of time. This pitting is caused when the
carbon dioxide dissolves in water droplets that condense on the tubing wall.
The tubing below the point of condensation may relatively corrosion free.
8.13.2
The most serious sweet oil corrosion problem can usually be found in gas lift
wells. They are usually high water producers, and corrosion can be accelerated
if the injected gas lift gas contains carbon dioxide and/or small amounts of
oxygen.
8.14
Sour Corrosion
Corrosion caused by hydrogen sulphide dissolved in water is referred to as sour
corrosion. Refer to Section 8.11.
8.14.1
In oil well tubing, hydrogen sulphide can cause the development of scattered
pits with cone shape sides leading to a rounded bottom. Black iron sulphide
(FeS) scale will usually be present. The cathodic nature of the iron sulphide
scale results in the pitting of the underlying steel.
8.14.2
Equipment in high-pressure sour gas wells may fail due to sulphide stress
cracking, hydrogen embrittlement, or hydrogen induced cracking. Severe
pitting corrosion can also occur in sour gas wells.
8.14.3
Corrosion on the interior of oil or sour water storage tank roofs (fixed roof
design tanks) begins when condensed water droplets are saturated with
hydrogen sulphide. The extensive pitting which results can perforate the tank
deck. The interior bottoms of oil storage tanks are also subject to attack
because of the layer of water which usually forms at the bottom under the oil.
Additionally, corrosion scale falling from the top to the bottom of the tanks may
cause corrosion concentration cells. Corrosion of the shell of oil tanks is
normally negligible. However, sour saltwater tanks will be subject to roof, wall,
and floor internal corrosion.
Pgina: 24 de 40
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 1
(Section 3.3)
FREE ENERGIES (KINETICS) OF FORMATION OF METAL OXIDES
Symbol
Metal Oxide
Calories/Mole
(G)
Al2O3
Aluminium Oxide
-376,700
Fe2O3
Ferric Oxide
-177,100
MgO
Magnesium Oxide
-136,130
ZnO
Zinc Oxide
-76,080
CuO
Cupric Oxide
-30,400
Ag2O
Silver Oxide
+2,600
Au2O3
+18,810
Notes:
1. For example, aluminum will tend to form its oxide very easily, iron almost as easily,
copper only to a minor extent, and gold only with an input of energy.
2. The magnitude of G does not alone determine the rate at which a metal returns to
its oxide.
Pgina: 25 de 40
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 2
(Section 4.2)
GALVANIC SERIES IN SEAWATER
VOLTS: SATURATED CALOMEL HALF-CELL REFERENCE ELECTRODE
+0.2 +0.1
0.0
-0.1
-0.2
-0.3 -0.4
-0.5
-0.6
-0.7
-0.8
-0.9 -1.0
-1.1 - 1.2
-1.3
-1.4
-1.5
-1.6 -1.7
Magnesium
Zinc
Beryllium
Aluminium Alloys
Cadmium
Mild Steel. Cast Iron
Low Alloy Steel
Austenitic Nickel Cast Iron
Aluminium Bronze
Naval Brass, Yellow Brass, Red Brass
Tin
Copper
Pb-Sn Solder (50/50)
Admiralty Brass, Aluminium Brass
Manganese Bronze
Silicon Bronze
Tin Bronzes (G & M)
Stainless Steel - Type 410, 416
Nickel Silver
90-10 Copper-Nickel
80-20 Copper-Nickel
Stainless Steel - Type 430
Lead
70-30 Copper - Nickel
Nickel - Aluminium Bronze
Nickel - Chromium Alloy 600
Silver Brazed Alloys
Nickel 200
Silver
Stainless Steel - Types 302, 304, 321, 347
Nickel - Copper Alloys 400, K-500
Stainless Steel - Types 316,
Alloy "20" Stainless Steels, cast and wrought
Nickel-Iron-Chromium Alloy 825
Ni-Cr-Mo-Cu-Si Alloy B
Titanium
Ni-Cr-Mo Alloy C
Platinum
Graphite
Alloys are listed in the order of the potential they exhibit in flowing sea water. Certain alloys indicated by the
symbol: in low-velocity or poorly aerated water, and at shielded areas may become active and exhibit a
potential near -0.5 volts.
Pgina: 26 de 40
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 3 (Page 1 of 5)
(Section 4.2 and 8.3.3)
CORROSION OF GALVANIC COUPLES IN SEA WATER AT 40- 80F (4-27C)
METAL CONSIDERED
COUPLED
WITH
1 2
Note: These Numbers correspond with the numbers and alloy designations listed in the left column.
3
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
1 Magnesium Alloys
E
L
2 Zinc
x
x
L
S x
3 Beryllium
4 Aluminum Alloys
x
x
S x
5 Cadmium
x
x
L
S
L
S
E
L
x
x
x
x
Pgina: 27 de 39
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 3 (Page 2 of 5)
(Section 4.2 and 8.3.3)
METAL CONSIDERED
COUPLED
WITH
1 2
S
Note: These Numbers correspond with the numbers and alloy designations listed in the left column.
3
S
E
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
x
x
x
x
Si Bronze (CA655)
x
x
x
x
L
S
E
S
11 Tin
E
S
14 Tin-Bronze (G&M)
E
L
x
x
x
x
x
x
x
x
x
x
x
x
Pgina: 28 de 39
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 3 (Page 3 of 5)
(Section 4.2 and 8.3.3)
METAL CONSIDERED
COUPLED
WITH
1 2
Note: These Numbers correspond with the numbers and alloy designations listed in the left column.
3
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
L
S
L
S
E
x
L
S
x
x
x
x
x
x
x
x
x
x
E
L
S
20 Lead
21 70/30 Copper-Nickel
(CA715)
E
L
x
x
x
x
x
x
x
Pgina: 29 de 39
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 3 (Page 4 of 5)
(Section 4.2 and 8.3.3)
METAL CONSIDERED
COUPLED
WITH
1 2
Note: These Numbers correspond with the numbers and alloy designations listed in the left column.
3
22 Nickel-Aluminium-Bronze
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
L
S
x
x
x
L
x
E
S
26 Silver
S
E
E
L
x
x
x
x
x
S
and K-500
x
x
x
x
x
x
x
x
25 Nickel
x
x
x
x
x
x
x
x
Pgina: 30 de 39
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 3 (Page 5 of 5)
(Section 4.2 and 8.3.3)
METAL CONSIDERED
COUPLED
WITH
1 2
S
Note: These Numbers correspond with the numbers and alloy designations listed in the left column.
3
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
x
x
L
S
S
E
C-276, 625
x
x
x
x
x
x
x
x
x
x
x
x
KEY:
x
S
E
L
UNFAVORABLE
UNCERTAIN
COMPATIBLE
- Exposed are of
- Exposed are of
- Exposed are of
Source: Galvanic Corrosion Indicator, International Nickel Co. and NACE Corrosion Engineer's Reference Book, 1983 Edition.
Pgina: 31 de 39
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 4
(Sections 4.3 and 4.4)
4a : Corrosion Cell
(Section 4.3)
Electrolyte
H+
e-
Electrons
Anode
(Corrosion area)
OH-
e-
Flow
Cathode
A
V
Zn
Cu
Electrolyte
Anode
Cathode
Pgina: 33 de 40
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 5
(Section 4.4)
ELECTROMOTIVE FORCE SERIES
Symbol
Metal
Standard Reduction
Potential Volts at 25oC
(vs Normal Hydrogen Electrode)
Potassium
-2.92
Greatest Tendency
Ca
Calcium
-2.87
to
Na
Sodium
-2.71
Corrode
Mg
Magnesium
-2.34
Be
Beryllium
-1.70
Al
Aluminium
-1.67
Mn
Manganese
-1.05
Zn
Zinc
-0.762
Cr
Chromium
-0.71
Fe
Iron
-0.44
Cd
Cadmium
-0.402
In
Indium
-0.346
Co
Cobalt
-0.277
Ni
Nickel
-0.250
Sn
Tin
-0.136
Pb
Lead
-0.126
Hydrogen
0.00
Cu
Copper
0.345
Hg
Mercury
0.799
Ag
Silver
0.800
Pt
Platinum
1.20
to
Au
Gold
1.42
Corrode
Least Tendency
Pgina: 34 de 40
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 6
(Sections 4.4 and 6.2)
6a : Polarisation Curve
(Section 4.4)
Potential
Cu
E
corr
Re
corr
Re =
Electrolyte Resistance
Zn
Current
corr
Corrosion Rate
14
pH
Pgina: 35 de 40
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 7
(Sections 6.3.1 and 8.12)
Relation of Dissolved Oxygen to the
General Corrosion Rate of Steel
Corrosion Rate (mpy)
30
25
20
15
10
5
Note change of scale
O
0
20
50
75
125
1000
3000
4500
6000
8000
Exxon Data
O&G Consultant
Note: Variation in values are due to differences in CO2/H2S content and accuracy of O2 measurement.
Pgina: 36 de 40
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 8
(Sections 6.3.3, 8.7.1 and 8.8)
H2S Service Induced Cracking of Plate Steel and Welds
Vessel ID in Sour Environment
SSC
Laminations or
Inclusions
Blistering
SWC
SOHIC
LEGEND:
Plate Steel
Weld
Pgina: 37 de 40
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 9
(Sections 8.1 and 8.2)
QUALITATIVE CATEGORIZATION OF COUPON CORROSION RATES FOR OIL PRODUCTION SYSTEMS
Pitting Rate
US Customary
(mpy(1))
Metric (SI)
(m/a(2))
US Customary
(mpy)
Metric (SI)
(m/a)
<1.0
<25
<5
<127
Moderate
1.0-4.9
25-126
5-7.9
127-201
High
5.0-10
127-254
8-15
202-381
>10
>254
>15
>381
Low
Severe
(1)
(2)
mpy =
m/a =
Note:
Pgina: 38 de 40
IRS-009.00
INSPECO
Refinaria de Sines
ATTACHMENT 10
(Section 8.13)
Corrosion Rate (mpy)
1000
150oF (65oC)
Flowing
77oF (25oC)
100
10
7
4
10
100
1000
Pgina: 39 de 40
IRS-009.00
INSPECO
Refinaria de Sines
Note: Graph is from a compilation of NACE data and is intended to show only the relationship of CO2 partial pressure to corrosion rate in flowing and nonconditions. Do not use this figure to predict CO2 corrosion rates.
flowing
Pgina: 40 de 40
IRS-009.00