Corrosion Brine

BRI NE CHOI CE MAT T ERS CORROSI ON
Tittel
Tittel
1
Brine choice matters
2
CORROSI ON BRI NE CHOI CE MAT T ERS
Contents
The word corrosion originates from the latin word corrdere which means to gnaw away. A more scientic
denition is the destruction or deterioration of a material by reaction with its environment. The corrosion types
encountered in the oileld are general corrosion, localised corrosion and hydrogen embrittlement. Within the category of
localised corrosion, pitting and stress corrosion cracking are becoming more common with the advent of HTHP drilling
and completion.
General corrosion (rust): Occurs slowly. Metal loss is uniform over the exposed surfaces and is often linear with time.
Relatively easy to keep under control with corrosion inhibitors, protective coatings and cathodic protection, but can
have large economic impact over time.
Localised corrosion: Occurs quickly. Metal loss is concentrated in a small surface area. Creates high risk of unpredict-
able and rapid failure of metal integrity. Pitting corrosion and stress corrosion cracking (SCC) are most common and
frequently occur in well tubulars constructed from so-called corrosion resistant alloys (CRAs). Corrosion inhibitors have
little or no effect and can actually initiate corrosion.
Hydrogen embrittlement: A brittle mechanical fracture of high-strength steels caused when atomic hydrogen dissolves
in the crystal structure of a metal rather than forming H
2
gas. Occurs in corrosive environments under constant tensile
stress, similar to stress corrosion cracking (SCC). The most common form is sulphide stress cracking (SSC), which
occurs when stressed metal is exposed to water containing hydrogen sulphide or other sulphur compounds, generally
under aerobic conditions. Limited to very high-strength alloys and low temperatures.
WHAT I S CORROSI ON?
High stakes, high risks Page 3
No CRA failures from formate brines Page 4
Corrosion prevention why formate brines outperform halide brines Page 5
Corrosive gases and CRAs the gamble you take Page 6
Oxygen can you breathe easy? Page 8
Carbon dioxide eating away your prots Page 12
Hydrogen sulphide a destructive force Page 19
3
High stakes,
high risks
Drilling in High Temperature High Pressure environments has turned clear opportunities into real rewards. But when the stakes
are high, so are the risks.
If traditional oileld metals are used under HTHP conditions, corrosion runs rampant. Operators have shown their
resource fulness by using higher-grade metals, such as Corrosion Resistant Alloys (CRAs), and have increased use of brine-
based drilling, workover and completion uids. Paradoxically, increased use has led to a greater risk of potentially catastrophic
types of corrosion. Of particular concern are two forms of localised corrosion, namely pitting and stress corrosion cracking (SCC).
New research shows that when CRAs are jointly exposed to a common gas, such as oxygen, carbon dioxide or hydrogen
sulphide, and a halide brine, pitting and SCC corrosion are likely to occur. When this happens, costs can be astronomical.
The bottom line is this:
More operators are recognising the hidden costs of using halide brines and are turning to naturally passivating, anti-oxidising
formate brines to prevent corrosion. This is paying dividends, as the following equation shows:
CRA + O
2
/CO
2
/H
2
S + halide brine = High risk of localised corrosion
CRA + O
2
/CO
2
/H
2
S + formate brine = Insignifcant risk of localised corrosion
General corrosion is relatively easy to control. Localised corrosion is a whole different ball game
4
With over 100 HTHP applications and no CRA failures, formate brines from Cabot Specialty Fluids are ideally suited for
challenging drilling, completion, workover and fraccing applications. Formate brines have proven their worth in the worlds most
demanding well environments. At temperatures as high as 216C (420F) and pressures up to 1,172 bar (17,000 PSI) , corrosive
gases, such as carbon dioxide (CO
2
), hydrogen sulphide (H
2
S) and oxygen (O
2
), are commonplace. Its no wonder that high-
performance formate brines are replacing high-risk halides as operators brines-of-choice for extreme wells.
No CRA failures
from formate brines
Corrosion protection for over twelve years
No matter how high the temperature or pressure; no matter whether CO
2
, O
2
or H
2
S are present; no matter which CRA is used, there
have been no recorded cases of material failure caused by formate brines in over twelve years consistent use in drilling, completion
or workover applications. The table below provides a selection of eld cases where formate brines have proved their worth.
BP Rhum
3/29a
Shell
Shearwater
Marathon
Braemar
BP
Devenick
Total
Elgin/Franklin
Statoil
Huldra
No. of wells 3 6 1 1 10 6
Hydrocarbon
Gas
condensate
Gas
condensate
Gas
condensate
Gas
condensate
Gas
condensate
Gas
condensate
Max.
temperature
C 149 182 135 146 204 149
F 300 360 275 295 400 300
Completion
material
CRA S13Cr 25Cr 13Cr 13Cr 25Cr S13Cr
Liner
material
CRA S13Cr 25Cr 22Cr VM110 P110 S13Cr
Formate brine
density
g/cm
3
2.00 2.20 2.05 2.20 1.80 1.85 1.60 1.65 2.10 2.20 1.85 1.95
CO
2
% 5 3 6.5 3.5 4 4
H
2
S ppm 5 10 20 2.5 5 20 - 50 10 14
Exposure
time
days 250 65 7 90 450 45
Application Perforation
Completion
Workover
Well Kill
CT
Workover
Perforation
Workover
Perforation
Drill
Completion
Workover
Completion
CT / Well Kill
Perforation
Drilling
Completion /
Screens
Statoil
Kvitebjorn
Statoil
Kristin
BP
High Island A-5
Devon Energy
West Cameron
165 A-7, A-8
Devon Energy
West Cameron
575 A-3
Walter O&G
Mobile Bay 862
No. of wells 7 to date 6 to date 1 1 1 1
Hydrocarbon Gas condensate Gas condensate Gas Gas condensate Gas Gas
Max.
temperature
C 150 171 163 149 163 216
F 302 340 325 300 325 420
Completion
material
CRA S13Cr S13Cr S13Cr 13Cr 13Cr G-3
Liner
material
CRA 13Cr S13Cr S13Cr 13Cr 13Cr G-3
Formate brine
density
g/cm
3
2.00 2.06 2.09 2.13 2.11 1.03 1.14 2.06
CO
2
% 2 3 3.5 5 3 3 10
H
2
S ppm Max. 10 12 17 12 5 5 100
Exposure
time
days 57 57
4
(+ 3 yrs packer)
240 and 270 120
16
(+ 325 packer)
Application Drilling
Completion /
Screens / Liners
Drilling
Completion
Well Kill
Completion
Packer
Packer Packer Well Kill
Completion
Packer
5
Corrosion prevention
why formate brines outperform halide brines
There are signicant differences between formate and halide brines in their ability to prevent corrosion. Heres why:
Anti-oxidising properties
Oxidants, such as O
2
, are known to cause corrosion. Formate brines are inherently anti-oxidising, which is demonstrated by
their ability to stablise biopolymers, such as xanthan, at higher temperatures than any other brine.
Naturally protective pH
Formate brines are naturally alkaline (pH 9-11). Most forms of corrosion are promoted at a low pH. For comparison, the pH of a
high-density halide brine would normally range from two to six, depending on the type of halide.
pH scale
1 2 3 4 5 6 7 8 9 10 11 12
Halide brines Formate brines
Acidic pH promotes corrosion Alkaline pH protects against corrosion
Compatibility with pH buffer
Inux of CO
2
can cause serious corrosion problems in all brine types. The only effective protection against CO
2
corrosion is the
addition of a carbonate/bicarbonate buffer. In formate brines, this buffer not only prevents the pH from dropping, but it promotes
the build-up of a passivating iron carbonate layer on the metal surface.
In contrast to formate brines, solutions based on divalent halide brines (CaCl
2
, CaBr
2
, ZnBr
2
) are not compatible with even small
amounts of carbonate/bicarbonate buffer. This is due to the precipitation of divalent carbonate salts (CaCO
3
, ZnCO
3
). This
means that divalent halide brines cannot benet from the corrosion protection provided by carbonate/bicarbonate buffers.
Scavenges H
2
S
An additional benet of formate brines is their ability to scavenge large amounts of H
2
S gas, effectively preventing both stress
corrosion cracking (SCC) and sulphide stress cracking (SSC). In an extreme laboratory test, buffered formate brine reduced H
2
S
levels by 78.4%, from 3,200 ppm to 690 ppm.
With the exception of zinc bromide, halide brines do not scavenge H
2
S. This is a clear disadvantage, which makes halides very
aggressive towards cracking.
A buffer is a solution in which the pH remains reasonably constant when acids or alkalis are added to it, i.e. it buffers
against changes in pH. Carbonate/bicarbonate buffers are not compatible with divalent brines, such as calcium and
zinc bromide, as they form insoluble calcium carbonate or zinc carbonate.
WHAT I S A BUFFER?
6
Corrosive gases and CRAs
the gamble you take
If the odds for corrosive gases, such as O
2
, CO
2
and H
2
S, entering the uid are negligible then the brines ability to combat
corrosion is less important. The problem is, however, that the risk is always there. The question is, how great?
With greater concentration on HTHP drilling, the potential for localised corrosion from corrosive gases is growing. The table
below compares halide and formate brines and their propensity to cause corrosion with common gases.
Halide brines vs. formate brines: corrosion susceptibility with common gases
HOW ARE CRAS CHOSEN?
Historically, the well engineer selects the tubular metallurgy based on produced uids and the production conditions in the
wellbore. Emphasis is always placed on the potential for H
2
S in the production stream. The increased use of CRAs in recent years
has raised the costs of tubulars considerably, making it ever more important to maximise life expectancy.
Common CRAs and their properties

Material Speci -
cation
Cr
%
Ni
%
Mo
%
Strength
KPSI
Max. temp.
C
Max. temp.
F
13Cr L80 13.0 0.4 0.09 80 149 300
S13Cr LC80-130M 12.92 5.26 0.77 80 177 350
22Cr EN 1.4462 22 5 3.2 128 232 450
25Cr EN 1.4410 25.39 6.45 3.92 135 232 450
C-steel 13Cr S13Cr 22Cr 25Cr C-steel 13Cr S13Cr 22Cr 25Cr
No gas

1)

1)

O
2
Immediate SCC
_

_

Long term SCC
_

_

Small/medium CO
2
, with or without H
2
S
Pitting

Immediate SCC
_

_

Long term SCC
_

_

General corrosion

High CO
2
, with or without H
2
S
Pitting

Immediate SCC
_

_

Long term SCC
_

_

General corrosion

S containing additive
_

NA NA NA NA NA
High susceptibility
1)
Dependent on type of halide and use of corrosion inhibitor.
Low susceptibility
2)
Dependent on chloride contamination.
Negligible susceptibility N.B.: Corrosion rates increase with temperature.
2) 2)
7
Two of the most common and dangerous forms of localised corrosion seen in the oileld are pitting and stress corrosion
cracking (SCC).
Pitting corrosion a hole lot of trouble
Pitting corrosion can result in perforation of pipes or equipment and subsequent system failure. It develops from defects in the metal
oxide layer (passive lm), which provides CRAs with improved corrosion protection. The local corrosion cell created by pitting
corrosion propagates into the underlying material and forms narrow but deep pits as illustrated below. Acidication of the aqueous
solution trapped in the corroding pit causes self-acceleration of the metal dissolution, resulting in relatively fast corrosion rates
and possible perforation of the affected pipe or equipment. In addition to the presence of an oxidising agent, such as oxygen,
carbon dioxide or hydrogen sulphide, halide ions, like chlorides or bromides, are usually required for pitting corrosion to develop.

The pitting corrosion mechanism Pitting corrosion attack
SCC here today, gone tomorrow
The most frightening aspect of stress corrosion cracking (SCC) is the speed of its development. From a small crack to full material
failure can happen in a matter of hours. The results can be catastrophic.
SCC develops from local defects in the surface oxide lm in areas where pitting corrosion has penetrated the metal surface. The
conditions necessary to create SCC are shown in the diagram below.
The combination of three factors cause SCC Chloride SCC in CRA
The metal dissolution (anode reaction) takes place in the crack tip. This causes the crack to propagate perpendicularly to the
tensile stresses in an attempt to relax, or fade out, the stresses. The internal stresses arise from the original steel manufacturing
and welding processes or from external loads.
O
+ H
O
Halide brine CRA under
tensile stress
O
, CO
or H
S
Fe(OHj
Fe(OHj
Porous gap
Passive film
Cl
-
Cl
-
+

H
O
O
+ H
O
Fe
+
e
-
e
-
lron
Acid chloride
pit electrolyte
Cl
-

+
H
+
+ H
O
Halide brine CRA under
tensile stress
O
, CO
or H
S
Fe(OHj
Fe(OHj
Porous gap
Passive film
Cl
-
Cl
-
+

H
O
O
+ H
O
Fe
+
e
-
e
-
lron
Acid chloride
pit electrolyte
Cl
-

+
H
+
Localised corrosion
a ght against time
8
Oxygen
can you breathe easy?
Dissolved oxygen from the atmosphere is present in all oileld brines. As halide brines do not enjoy natural protection against
this corrosive gas, CRAs weaken and fail through SCC. To prevent this from happening, an oxygen scavenger is normally added.
However, this remedy is far from dependable and cases where an oxygen scavenger fails or is overwhelmed are relatively
frequent. In fact, concentrations of less than 1.0 ppm are all thats needed to initiate SCC.
In comparison, formate brines contain natural anti-oxidants and prevent SCC without additives of any kind.
Calcium bromide
CRA type Without added chloride With 1% added chloride
No. of coupons
cracked
Time
period
No. of coupons
cracked
Time
period
Super 13Cr 3 out of 3 1 month 3 out of 3 1 month
22Cr 3 out of 3 2 months 3 out of 3 2 months
Halide brines vs. formate brines: Testing for SCC with oxygen
9
Potassium/cesium formate
CRA type Without added chloride With 1% added chloride
No. of coupons
cracked
Time period
No. of coupons
cracked
Time period
Super 13Cr 0 out of 3 3 months
2 out of 3 with
crack initiation
3 months
SCC testing on a 1.7 s.g. (14.2 ppg) potassium/cesium formate uid and a 1.7 s.g. (14.2 ppg) calcium bromide uid, with O
2

headspace. Temperature of 160C (320F). P
N
2
of 10 bar (145 PSI). P
O
2
of 0.2 bar (2.9 PSI).
Latest research uncovers new concerns
It has long been accepted that chloride brines contaminated with oxygen cause serious SCC problems. However, a
research study completed by the Hydro Research Centre Porsgrunn, Norway shows that bromide brines also instigate SCC. In
fact, results prove that SCC is already initiated on lower-grade alloyed steel after one months exposure to bromide brine with
minimal or higher levels of chloride contamination (see results on opposite page). It is worth noting that bromide brines, due to
their manufacture, always contain chlorides to a lesser or greater degree.
Formate brines, even with a large contamination of chloride, did not cause any SCC during the rst month of exposure and
limited SCC over the next two months (see results below).
Halide brines vs. formate brines: Testing for SCC with oxygen
10
Cracked after seven days
The Hydro study was taken one step further by CAPCIS Ltd. in the UK during a 14-day test of
calcium bromide without added chloride. The study establishes that SCC occurs after only seven
days in both S13Cr and 22Cr steel. In fact, all three of the S13Cr samples broke in two after
14 days. Based on these results, it is extremely inadvisable to use calcium bromide brine in
short-term applications where S13Cr and 22Cr steels are selected.
Short-term SCC test with O
2
and calcium bromide
CRA type
No. of coupons
cracked
Time
period
No. of coupons
cracked
Time
period
Super 13Cr 1 out of 3 7 days 3 out of 3 14 days
22Cr
3 out of 3 with
crack initiation
1)
7 days 3 out of 3 14 days
25Cr 0 out of 3 7 days 0 out of 3 14 days
1) After seven days several suspect areas were found on these samples. These
worsened after 14 days. Sectioning showed that cracks were present.
SCC testing on a 1.7 s.g. (14.2 ppg) calcium bromide uid, with O
2

headspace. Temperature of 160C (320F). P
N
2
of 10 bar (145 PSI),
P
O
2

of 0.2 bar (2.9 PSI).
Nine days
After only nine days, stress corrosion cracking on S13Cr martensitic stainless steel stopped production on Petronas HTHP Resak
A-6 well in Malaysia. Cracks were found on components such as tubing, blast joints, sliding sleeves and the packer mandrel.
Analysis suggested that cracking was caused by a combination of the CaCl
2
completion uid that contained oxygen and CO
2
and
H
2
S in the well uid. Laboratory tests in the CaCl
2
brine environment with the same eld conditions showed that SCC could be
prevented by the use of oxygen scavenger in the brine alone. However, when H
2
S and CO
2
were combined with CaCl
2
brine the
environment became extremely corrosive and chemical additives could not prevent cracking.
Temperatures varied between 140C (284F) and 156C (313F), with CO
2
partial pressure of 4.4 to 9.6 MPa (638 to 1,392 PSI) and
H
2
S partial pressure of 0.0004 to 0.0014 MPa (0.06 to 0.2 PSI).
Ibrahim, et. al.: NACE paper no. 03097
11
Halide brines have no anti-oxidising properties.
Chloride brines contaminated with oxygen have long
been known to cause serious pitting corrosion and
SCC.
Bromide brines are now proven to promote SCC,
even in the absence of chloride contamination.
S13Cr fails in a week and 22Cr shows signicant
pitting in the same period.
Formate fuids are strong anti-oxidants.
Even when formate brines are signifcantly contami-
nated with both oxygen and halide ions, no SCC
has ever been detected in 22Cr and 25Cr. With
such signicant contamination, SCC initiation has
occurred in S13Cr, although the process is much
slower and less aggressive than SCC in halides.
CRA under tensile stress + O
2
+ halide brine = Pitting and SCC
Staying the distance
If it were the Olympics then this would be the marathon and it would be run over some of the toughest country
imaginable. In other words, with downhole temperatures in excess of 204C (400F) and pressures of 1,105 bar (16,000
PSI), conditions in Totals Elgin wells G1 and G3 couldnt get more extreme.
When Total made the decision to suspend and abandon both wells, it needed to recover its 25Cr Duplex tubing.
The tubing had been exposed to cesium formate in both wells, including a 15-month (450-day) period in G3. As
recommended by Cabot Specialty Fluids, no corrosion inhibitor or oxygen scavenger were used. Mechanical tests
conrmed compliance with the original properties of the tubing material. No localised corrosion was found and it was
certied t for further use in HPHT well conditions.
Furthermore, due to casing cement issues in G3, the well was killed with a combination of SBM and cesium formate.
The mechanical tests on the removed tubing compared hydrogen charging on both sections and concluded that, while
the 25Cr Duplex tubing was unchanged in the section exposed to cesium formate, the tubing exposed to the SBM
showed surprisingly higher levels of hydrogen charging. This proves that formate brines are stable under extreme HTHP
conditions and CRAs, immersed in these brines, do not suffer from hydrogen embrittlement.
To sum up
The dangerous relationship between oxygen and halide brines is summarised by the following equation. The table then compares
halide and formate brines ability to cause or prevent pitting and SCC.
12
Carbon dioxide is present in most reservoirs and can enter a completion, workover, drilling or suspension uid over time.

CO
2
corrosion occurs when CO
2
gas combines with water to form carbonic acid (H
2
CO
3
). Carbonic acid is very corrosive to
C-steel and 13Cr steel. The only efcient protection against CO
2
corrosion is a carbonate/bicarbonate pH buffer. While
formate brines are highly compatible with buffers, halide brines are incompatible with carbonate/bicarbonate buffers as they
form insoluble calcium carbonate or zinc carbonate precipitates.

Consequently, when analysing CO
2
corrosion in completion and packer uids, it is crucial to understand how buffered (formate)
and unbuffered (halide) brines compare under all scenarios of CO
2
inux.
When acid gas (CO
2
) enters an aqueous completion/packer uid, the following reactions take place:
The value for pK
a
1
in equation three is 6.35. This means that at pH 6.35 equal amounts of bicarbonate (HCO
3
2-
) and carbonic acid
(H
2
CO
3
) exist in solution. In uids with pH above this level, the dissolved CO
2
is largely converted to non-corrosive bicarbonate,
while in uids with pH below this level, it is largely converted to very corrosive carbonic acid.
Effect of CO
2
gas inux on brine pH
Carbon dioxide
eating away your prots
1. CO
2
(g) = CO
2
(aq)
2. CO
2
(aq) + H
2
O = H
2
CO
3
(aq)
3. H
2
CO
3
(aq) = HCO
-
3
(aq) + H
+
(aq)
K
a1
0 40 80 120 160 200 240 280 320 360 400
Bbls gas influx per 1,000 bbls of brine (5% CO
, 150`C (302`Fj, 10,000 PSlj

Buffered formate brine
Calcium bromide brine
2
3
4
5
6
7
8
9
10
11
CO
2
converted to b|carbonate
CO
2
converted to carbon|c ac|d
p
H
13
No CO
2
inux
Depending on the brine type or blend, the pH ranges
from 1 to 6. Consequently, uid corrosivity depends
on certain factors, e.g. halide type, use of corrosion
inhibitors, oxygen scavengers and the initial pH of the
brine. Pitting corrosion is frequently seen, particularly
in the very acidic zinc bromide brine. At temperatures
of 107C (225F) and upwards, thiocyanate corrosion
inhibitors and other sulphur-containing additives are
prone to thermal decomposition and subsequent
pitting. In turn, depending on the temperature and
material type, SCC or SSC are likely to occur.
Small/medium CO
2
inux
As soon as CO
2
enters the unbuffered halide brine,
carbonic acid is created. For each dissolved mole of
CO
2
, one mole of carbonic acid forms. As carbonic
acid is very corrosive, CO
2
corrosion commences
immediately. Due to the lack of carbonate/bicarbonate
ions in the halide brine, all carbonate required for the
formation of the layer comes from the CO
2
gas,
slowing the formation process considerably.
Consequently, the protective layer that forms is of
low quality and pitting corrosion is found in halide
brines under these conditions. Small/medium CO
2

inux has the same effect as large CO
2
inux.
Large CO
2
inux
As soon as CO
2
enters the unbuffered halide brine,
carbonic acid is created. For each dissolved mole of
CO
2
, one mole of carbonic acid forms. As carbonic
acid is very corrosive, CO
2
corrosion commences
immediately and an iron carbonate layer begins to
form. Due to the lack of a carbonate/bicarbonate
buffer, all carbonate required for the formation of the
layer comes from the CO
2
gas, slowing the formation
process considerably. Consequently, the protective
layer that forms is of low quality and pitting corrosion
is found in halide brines under these conditions.
Large CO
2
inux has the same effect as small/ medium
CO
2
inux.
No CO
2
inux
Formate brines remain at the initial pH of 9 to 10 and
exhibit lower general corrosion rates than halides.
This is due to their favourable pH and anti-oxidising
properties. No pitting corrosion has ever been
reported in formate brines. The absence of additives,
including corrosion inhibitors containing sulphur,
eliminates the risk of corrosion caused by thermal
decomposition of these materials.
Small/medium CO
2
inux
When carbonic acid forms in formate brines it is
absorbed by the buffer and converted to bicarbonate.
The amount of carbonic acid that exists in
equilibrium with the bicarbonate is insignicant in this
high pH range and CO
2
corrosion does not occur.
Consequently, corrosion rates remain at an extremely
low level.
Large CO
2
inux
If extremely high amounts of CO
2
enter the buffered
formate brine, the buffer is eventually overwhelmed,
i.e. all carbonate present in the buffer converts to
bicarbonate. pH then drops and causes CO
2
corrosion
to commence immediately. Although general CO
2

corrosion rates in formate brines can be just as
high as in halide brines, pitting and localised
corrosion do not occur due to the rapid formation of a
high-quality passivating iron carbonate layer. This layer
forms quickly as it takes bicarbonate ions from the buffer
and provides excellent protection against localised and
generalised corrosion.
Halide brines vs.
formate brines with CO
2
inux
14
Hydro Research Centre Porsgrunn has completed extensive research into CO
2
corrosion and brines. The key nding of these
studies show that buffered formates are superior to halides in building up a protective iron carbonate layer that protects the
metals against pitting and general corrosion.
The graph depicts measured CO
2
corrosion rates in various formate and halide uids as a function of time. The unrealistically
high CO
2
level used in the experiment is high enough to overwhelm immediately the pH buffer in the formate brine. As can be
seen, the initial corrosion rates during formation of the passivating layer are higher but of much shorter duration in the formate
brines than in the bromide brines. This gives a denser, higher-quality passivating layer. The common corrosion inhibitor in the
bromide brine used for this test has no effect.
Halides and formates
the facts speak for themselves
C-steel corrosion coupon (1.5 mm thick) exposed to
a calcium bromide uid in the presence of CO
2
. The
temperature was ramped from 120C (248F) up to 180C
(356F) and down again during the 50-day long exposure.
The C-steel coupon experienced severe pitting corrosion
and perforated in the calcium bromide brine. A separate
test in a calcium bromide uid with a commonly used
corrosion inhibitor gave the same poor performance.
C-steel corrosion coupon (1.5 mm thick) exposed to
a potassium formate brine. The CO
2
level used was
high enough to immediately overwhelm the pH buffer
in the formate brine. Temperature was ramped from
120C (248F) up to 180C (356F) and down again
during the 62-day long exposure. Only limited general
corrosion could be seen in the formate uid.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
jC
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
4.0
4.5
5.0
5.5
6.0
0 100 200 300 400 500 600
Time (hoursj
CaBr
2
1.53 s.g. Unhibited 120C (248Fj
CaBr
2
1.53 s.g. lnhibited 120C (248Fj
K formate 1.53 s.g. 120C (248Fj
KCs formate 1.70 s.g. 150C (302Fj
KCs formate 1.70 s.g. 175C (347Fj Format|on of FeCO
3
protect|ve |ayer
0
50
100
150
200
Initial CO
2
corrosion in formates and bromides
15
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
20 40 60 80 100 120 140 160 180 200 220
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Borderline corrosion rate
Acceptable corrosion rate
Large CO
influx
No/small/medium CO
influx
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
Temperature (Cj
Large CO
influx
No/small/medium CO
influx
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Temperature (Cj
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
Typical Opreating Window
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
Temperature (Cj
0
5
10
15
20
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
Large CO
influx
No/small/medium CO
influx
Large CO
influx
No/small/medium CO
influx
CO2 Corrosion rate of 13Cr inFormate Brines
68 118 168 218 268 318 368 418
Temperature (Fj
Acceptable corrosion rate Typical Opreating Window
CO2 Corrosion Rates of S13Cr in Formate Brines
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Temperature (Cj
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Large CO
influx
No/small/medium CO
influx
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
CO2 Corrosion Rates of 22Cr in Formate Brines
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
Temperature (Cj
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
Large CO
influx
No/small/medium CO
influx
CO
2
corrosion rates of 22Cr in formate brines
CO
2
corrosion rates of 25Cr in formate brines
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
20 40 60 80 100 120 140 160 180 200 220
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Large CO
influx
No/small/medium CO
influx
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
Temperature (Cj
Large CO
influx
No/small/medium CO
influx
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Temperature (Cj
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
Temperature (Cj
0
5
10
15
20
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
Large CO
influx
No/small/medium CO
influx
Large CO
influx
No/small/medium CO
influx
68 118 168 218 268 318 368 418
Temperature (Fj
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Temperature (Cj
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Large CO
influx
No/small/medium CO
influx
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
Temperature (Cj
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
Large CO
influx
No/small/medium CO
influx
CO
2
corrosion rate of C-steel in formate brines
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
20 40 60 80 100 120 140 160 180 200 220
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Large CO
influx
No/small/medium CO
influx
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
Temperature (Cj
Large CO
influx
No/small/medium CO
influx
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Temperature (Cj
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
Temperature (Cj
0
5
10
15
20
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
Large CO
influx
No/small/medium CO
influx
Large CO
influx
No/small/medium CO
influx
68 118 168 218 268 318 368 418
Temperature (Fj
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Temperature (Cj
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Large CO
influx
No/small/medium CO
influx
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
Temperature (Cj
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
Large CO
influx
No/small/medium CO
influx
CO
2
corrosion rate of 13Cr in formate brines
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
20 40 60 80 100 120 140 160 180 200 220
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Large CO
influx
No/small/medium CO
influx
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
Temperature (Cj
Large CO
influx
No/small/medium CO
influx
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Temperature (Cj
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
Temperature (Cj
0
5
10
15
20
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
Large CO
influx
No/small/medium CO
influx
Large CO
influx
No/small/medium CO
influx
68 118 168 218 268 318 368 418
Temperature (Fj
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Temperature (Cj
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Large CO
influx
No/small/medium CO
influx
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
Temperature (Cj
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
Large CO
influx
No/small/medium CO
influx
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
20 40 60 80 100 120 140 160 180 200 220
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Large CO
influx
No/small/medium CO
influx
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
Temperature (Cj
Large CO
influx
No/small/medium CO
influx
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Temperature (Cj
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
Temperature (Cj
0
5
10
15
20
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
Large CO
influx
No/small/medium CO
influx
Large CO
influx
No/small/medium CO
influx
68 118 168 218 268 318 368 418
Temperature (Fj
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Temperature (Cj
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
Large CO
influx
No/small/medium CO
influx
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
0.0
0.1
0.2
0.3
0.4
0.5
0.6
20 40 60 80 100 120 140 160 180 200 220
Temperature (Cj
0
5
10
15
20
68 118 168 218 268 318 368 418
Temperature (Fj
C
o
r
r
o
s
i
o
n

r
a
t
e

(
m
m
/
y
j
C
o
r
r
o
s
i
o
n

r
a
t
e

(
M
P
Y
j
Large CO
influx
No/small/medium CO
influx
CO
2
corrosion rates of S13Cr in formate brines
Graphs show corrosion rates in formate uids as taken from literature. The rates include a combination of CO
2
inux only and
CO
2
/H
2
S inux. Corrosion rates measured by weight loss of less than 30 days are not included for C-Steel and 13Cr steel. Halide
corrosion rates are not plotted as the presence of pitting corrosion makes general corrosion rates meaningless. General corrosion
rates in halides, where pitting is not taken into account, are generally higher than in formates. In fact, just low levels of CO
2
inux
in halide uids have comparable corrosion levels to high levels of CO
2
inux in formates. Similar corrosion rates can only be seen
in formates and halides with a very large CO
2
inux.
16
Possible pitting/localised corrosion
depen ding on pH, i.e. brine type
CO
2
corrosion
Pitting/localised corrosion
CO
2
corrosion
CO
2
corrosion
Insignicant general
corrosion
No CO
2
corrosion
Insignicant general corrosion
No CO
2
corrosion
Insignicant general corrosion
General CO
2
corrosion only at high
temperatures
No pitting/localised corrosion
Halide and formate brines performance under C0
2
inux
Pitting leads to SCC
In many SCC incidents, pitting corrosion helps initiate the crack. Formate brines and halide brines performance against this
devasting form of corrosion varies considerably.
No CO
2

inux
Small CO
2

inux
Medium CO
2

inux
Large CO
2
inux
Over and out
Erskine is a rich gas condensate eld in the North Sea, with a BHT of 177C (350F), 965 bar (14,000 PSI) pressure and produced
uids containing 4% CO
2
and 15-30 ppm H
2
S. A 25%Cr Duplex stainless steel production tubing was chosen for well W4. When
it came to a packer uid, 1.3 s.g. (11.3 ppg) CaCl
2
was used as the base, with biocide (C
5
H
8
O
2
), oxygen scavenger (NH
4
HSO
3
)
and a thiocyanate corrosion inhibitor (NaCNS) added. After 11 months the production tube failed. A metallurgical investigation
concluded that pitting, followed by chloride SCC, was the likely cause of failure, most probably as a result of air O
2
ingress into the
annulus during the bleed-off of annular pressure.
Mowat, et.al., SPE67779
SEM photo of the protective iron carbonate layer as
formed on C-Steel in cesium formate. The CO
2
level
used was high enough to immediately overwhelm the pH
buffer in the formate brine. A very dense surface lm with
thicknesses varying from 5 to 20 m formed in the cesium
formate brine.
No CO
2

inux
Small CO
2

inux
Medium CO
2

inux
Large CO
2
inux
Not advisable to use this brine Proceed with caution Brine safe to use
17
A clear comparison
Research completed at the end of 2005 at the Hydro Research Centre Porsgrunn, clearly compares the effect halide and formate
brines have on commonly used metals in the presence of CO
2
without O
2
or H
2
S

. It conrms eld experience by showing that halide
brines cause SCC, while formate brines prevent it. What was startling, however, is the speed at which SCC develops on Super 13Cr
and Duplex 22Cr. Within a month, SCC had developed on all three coupons for both metals. This shows that there is a risk of SCC
even during short-term operations, such as completions and workovers, with halide brines and these common metals.
SCC testing on a 1.7 s.g. (14.2 ppg) potassium/cesium formate uid and a 1.7 s.g. (14.2 ppg) calcium bromide uid, with CO
2

headspace. Temperature of 160C (320F). PCO
2
of 10 bar (145 PSI). In this worst-case scenario, with high levels of CO
2
and
chloride, the pH buffer in the formate brine was immediately overwhelmed and the pH allowed to drop.
CRA type
No. of coupons
cracked
Time period
Super 13Cr 3 out of 3 1 month
Duplex 22Cr 3 out of 3 1 month
Duplex 25Cr
2 out of 3 with
crack initiation
3 months
CRA type
No. of coupons
cracked
Time period
Super 13Cr 0 out of 3 3 months
Duplex 22Cr 0 out of 3 3 months
Duplex 25Cr 0 out of 3 3 months
Calcium bromide + 1% Cl
Potassium/cesium formate + 1% Cl
Halide brines vs. formate brines: Testing for SCC with carbon dioxide
18
To sum up
The dangerous relationship between carbon dioxide and halide brines is summarised by the following equation. The table then
compares halide and formate brines ability to cause or prevent pitting and SCC.
When CO
2
enters halide brine, it immediately
converts to carbonic acid, creating a corrosive
environment. This means that halide brines may
cause SCC when exposed to tensile stresses. The
speed of SCC depends on material, halide type
and temperature. Only a very small amount of CO
2

is required to cause failure.
No SCC has ever been measured or reported in a
formate brine exposed to carbon dioxide. Formate
brines provide complete protection against SCC
when exposed to CO
2
.
A very signifcant CO
2
inux, or a long exposure
period with a moderate inux, is required for
carbonic acid to form due to the unique carbonate/
bicarbonate buffer. This prevents any adverse
impact on the vast majority of wells with CO
2
inux.
In extreme laboratory tests when the buffer has been
overwhelmed by a massive simulated CO
2
inux over
a very long time period, SCC has never been recorded
in formate brines, even when saturated with chloride
ions. This proves that formate brines protect CRAs
against SCC even in very aggressive chloride/CO
2

environments.
Calcium bromide + 1% Cl
Potassium/cesium formate + 1% Cl
Adding to the problem

A 1.32 s.g. (11 ppg) CaCl
2
brine with an oxygen scavenger and sodium thiocyanate high-temperature corrosion inhibitor
proved devastating in Shells Deep Alex gas well in the Gulf of Mexico. A 22Cr Duplex stainless steel production tubing was
selected based on temperatures up to 188C (370F), pressures of 116MPa (16,800 PSI) and existence of CO
2
and H
2
S.
The well produced for one month before the production tubing failed and it was shut in. A closer look showed that
multiple joints of the tubing had either failed altogether or contained stress corrosion cracking. A failure analysis and
subsequent laboratory programme demonstrated that the problem was caused by environmentally assisted cracking of
the 22Cr Duplex. This was proved to be a direct result of H
2
S production from decomposition of the sodium thiocyanate
corrosion inhibitor and/or ammonium bisulphate oxygen scavenger from the CaCl
2
completion uid in the annulus.
Mack, et. al., NACE paper no. 02067
CRA under tensile stress + CO
2
+ halide brine = Pitting and SCC
19
Hydrogen sulphide is commonly referred to as rotten egg gas and is one of the most destructive gases known to man.
Concentrations of 50 ppm cause highly stressed, high-strength steel to fail in a matter of minutes. Small concentrations can
reduce drill pipe life by a factor of ten. In fact, only 0.1 ppm greatly reduces failure time of highly stressed, high-strength drill pipe.
H
2
S can enter the completion or packer uid either from the reservoir gas, along with CO
2
, or from decomposition of sulphur-
containing additives (for example thiocyanate). It causes pitting and stress corrosion cracking (SCC) at high temperatures and
sulphide stress cracking (SSC) at low temperatures. Many recent failures in halide brines have been caused by H
2
S, which has been
formed by decomposition of the commonly used corrosion inhibitor, thiocyanate.
H
2
S in halides and formates
Laboratory results show formate brines superiority
Both Statoil and Hydro Research Centre Porsgrunn have included H
2
S in CO
2
corrosion testing. The CO
2
corrosion results
shown earlier are unaffected by the presence of H
2
S. Statoil included H
2
S in its SCC testing with CO
2
on S13Cr. H
2
S did not
have any impact on the ability of formate brines to prevent SCC in the presence of H
2
S.

Hydrogen sulphide
a destructive force
Sulphur-based additives commonly used.
These decompose to form H
2
S.
Lack of pH buffer causes H
2
S to react immediately
to iron sulphide and atomic hydrogen.
No additives means no H
2
S from decomposition.
H
2
S inux from the reservoir gas is absorbed by
the pH buffer and by the inherent scavenging
nature of formates.
Halides Formates
Further reading
Leth-Olsen: CO
2
Corrosion in Bromide and Formate
Well-Completion Brines, SPE 95072
Stevens et. al.: Oilfeld Environment-Induced Stress Corro-
sion Cracking of CRAs in Completion Brines, SPE 90188
Mack et al.: Stress Corrosion Cracking of a Cold Worked
22Cr Duplex Stainless Steel Production Tubing in High
Density Clear Brine CaCl
2
Packer Fluid, NACE 02067
Ibrahim et al.: Corrosion Behavior of Super 13Cr Marten-
sitic Stainless Steel in Completion Fluids, NACE 03097
Scoppio et al.: Corrosion and Environmental Cracking
Testing of a High-Density brine for HPHT Field Appli-
cation, NACE 04113
Mowat et al.: Erskine Field HPHT Workover and Tubing
Corrosion Failure Investigation, SPE 67779
Ke and Qu: Thermal Decomposition of Thiocyanate
Corrosion Inhibitors: A Potential Problem for Successful
Well Completions, SPE 98302
Downs and Leth-Olsen: Effect of Environmental Contami-
nation of the Susceptibility of Corrosion Resistant Alloys
to Stress Corrosion Cracking in High-Density Completion
Brines, SPE 100438. To be published May 2006
API SC13/TG06 Corrosion Resistant Alloy Testing
Programme extensive test programme available for
sponsoring companies
Cabot Specialty Fluids Ltd
Cabot House, Hareness Circle
Altens Industrial Estate
Aberdeen. AB12 3LY
UK
Tel: (44) 1224 897229
Fax: (44) 1224 870089
Cabot Specialty Fluids AS
Bygg K8
P.O. Box 4, N-5347 Kystbasen
gotnes, Bergen
Norway
Tel: (47) 56 32 31 56
Fax: (47) 56 32 31 10
Cabot Specialty Fluids Inc.
Waterway Plaza Two
10001 Woodloch Forest Drive, Suite 275
The Woodlands, TX 77380
USA
Tel: (1) 281 298 9955
Fax: (1) 281 298 6190
www.cabot-corp.com/csf

Corrosion Brine

Uploaded by

Copyright:

Available Formats

Corrosion Brine

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Corrosion Brine

Uploaded by

Copyright:

Available Formats

What are the main types of corrosion encountered in the oilfield?

What are pitting corrosion and stress corrosion cracking?

BRI NE CHOI CE MAT T ERS CORROSI ON

, 150`C (302`Fj, 10,000 PSlj

Adding to the problem

You might also like