by

Morris C. Place Jr.

Published by
CHAMPION TECHNOLOGIES
Houston, Texas

2008
The author thanks the management of Champion Technologies for their
support in writing this document. The author also thanks Stan Moore for
his review of the “Beam Pumping” section of Chapter II. And the author
particularly thanks both Charles Rector for his critique of the entire
document and Buddy Schaefer for his work in finding and documenting
all the corrosion failure examples and for his review of the “Beam
Pumping” section of Chapter II. Finally, the assistance of Kathy
Eddlemon, Business Process Administrator, was essential to the
preparation and publication of this document.

Morris C. Place Jr.

March 1, 2008
Corrosion has serious economic consequences for most industries. In
2002, a U.S. Department of Transportation study concluded that
corrosion costs the U.S. economy $376 billion per year or 3.1% of gross
national product. Corrosion is also a current serious economic problem in
oil and gas production.

Edwin Drake drilled the first well specifically for oil in 1859 to a total
depth of 69.5 feet, although petroleum had been produced for centuries
in Egypt, China, Persia and Mesopotamia. Marco Polo reported oil
springs in Baku in the late 1200s. Salsomaggiore, Italy, was near a
natural gas source and adopted the crest of a burning salamander. In the
U.S., a Franciscan visited oil springs in New York in 1632 and a Russian
traveler visited oil springs in Pennsylvania in 1748. But until about 1850,
when the coal oil process was developed, animal fat was the main
source of machinery lubricants and oil for lighting. It was coal that fueled
the industrial revolution.

World oil production was only a little more than 400,000 B/D (Barrels per
Day) in 1900. By 1930, world oil production had risen to nearly 4 million
B/D. And for the next seven decades, oil and gas have supplied
essentially all the growth in the world energy supply. Coal production
from the 1930s to the 1960s stayed constant at about 15 million B/D of
oil equivalent. Nuclear energy use has continued to expand in the last
two decades, but its share of the total world energy production has
remained about constant.

Corrosion problems grew in the oil field as oil and gas production grew.
Corrosion first manifested itself in the 1930s with low pressure oil
production, but was not of widespread, serious economic concern. In the
1940s, corrosion grew on manifold fronts. First, water production
continuously increased in the low pressure oil production, both in
frequency of occurrence and volume. Corrosion costs were accelerated
by the war demands for increased oil and gas production. Increased
drilling and production of natural gas in the 1940s resulted in the
intensification of corrosion problems on two fronts: sweet gas-
condensation corrosion from CO2 and corrosion from H2S, both in the
form of general corrosion and SSCC (Sulfide Stress Corrosion Cracking,
currently known as SSC for Sulfide Stress Cracking). Tubulars high in
nickel content (9%) were the industries initial attempt to develop a
corrosion-resistant material. These tubulars were not successful as an
economic solution to corrosion; also, unbeknown to the users, high nickel
tubes were not resistant to SSC. As result, the first SSC failure occurred

i
in 1950 in a sour gas well equipped with 9% nickel tubing in the Pincher
Creek field in Alberta, Canada.

“Necessity is the mother of invention,” and modern corrosion alleviation

technology was born in the 1950s and has advanced on all fronts.
Modern chemical corrosion inhibitors were developed beginning in the
1950s as part of this technology growth. Initial corrosion inhibitors, were,
for the most part, chemical industry by-products (waste products)
dissolved in solvents. Some of these early multi-component corrosion
inhibitors were not completely soluble in typical hydrocarbon solvents
and required more aromatic solvents.

Corrosion inhibitors have advanced from oil-soluble to water-

dispersible/water-soluble. Concurrently, corrosion inhibitor applicability
advanced to higher temperatures and to more severe environments.

In the 1950s, the prediction of corrosion and the development of

corrosion alleviation solutions started about the same time. Both the
prediction of corrosion and the development of corrosion inhibition
programs started out as trial and error. Again, sophistication developed
on numerous fronts. Rules of thumb were first developed to predict
corrosiveness and how to inhibit against corrosion when warranted.
Corrosion inhibitor testing was initiated in the 1950s and the first wheel
tester, built in the late fifties, utilized soft drink bottles to hold the
corrosion coupons and corrosive fluids. Flow modeling also was initiated
in this same time frame.

Today, corrosion and its control are not random events; when conditions
are known, both uninhibited corrosion and the results of corrosion
alleviation efforts are predictable. Flow modeling, when calibrated with
field experience, provides reliable predictions of anticipated uninhibited
corrosion rates. Modern computers have enhanced the ability to model
flow, predict uninhibited corrosion under flow conditions, and predict the
results of corrosion alleviation systems. Corrosion predictions in
horizontal flow models are currently more sophisticated than vertical flow
models, mostly because there is much more field data with which to
calibrate the flow models.

Corrosion modeling is obviously a moving target. There is much work

being done in this field and model enhancements are continuous. In the
future, even better models will be available to the industry, models which
better represent actual production environments and which better predict
corrosion, uninhibited and inhibited.

ii
Modern corrosion testing techniques utilize reliable tools to predict the
effectiveness of corrosion alleviation programs. Finally, experience has
shown that old rules of thumb are not always correct, but they can act as
starting points when too little information is known to make predictions
using modern analytical tools. Relevant experience is always valuable,
both when predicting corrosion and developing corrosion alleviations
systems. But with experience and known conditions, such as flow rates,
temperature, pressure, fluid composition, and production system
geometry, current technology that includes flow modeling and laboratory
testing, much of the trial and error is taken out of successful corrosion
alleviation.

Starting with “Rules of Thumb,” this document is written with a view

toward using the best technology to achieve the most economic
corrosion alleviation system. An economic corrosion alleviation system
is a system that combines 1) field experience, with 2) corrosion-resistant
materials, 3) flow modeling, 4) corrosion inhibitors developed and tested
with modern laboratory techniques, and 5) field monitoring, both reactive
and proactive.

iii
 CHAPTER I – METAL LOSS CORROSION IN OIL AND GAS
PRODUCTION

Description Page
Introduction................................................................................................ 1
Corrosion Fundamentals ........................................................................... 2
Carbon Dioxide Corrosion ......................................................................... 5
Carbon Dioxide Corrosion-Erosion ..................................................... 9
Erosion ................................................................................................ 9
Velocity-Related Corrosion ................................................................. 9
Hydrogen Sulfide Corrosion .................................................................... 11
Sulfide Stress Cracking (SSC) ......................................................... 12
Microbiologically Influenced Corrosion (MIC) ......................................... 13
Introduction ....................................................................................... 13
Bacteria ............................................................................................. 13
Acid Producing Bacteria (APB) ......................................................... 14
Sulfate Reducing Bacteria (SRB) ..................................................... 14
SRB Generated H2S ......................................................................... 15
Iron Sulfide Particles ......................................................................... 15
Corrosion from SRB Generated H2S ................................................ 15
Reservoir Souring ............................................................................. 16
SRB Detection .................................................................................. 17
SRB Control ...................................................................................... 17
SRB Monitoring................................................................................. 18
Oxygen Corrosion in Oil and Gas Production
Introduction ....................................................................................... 19
Air Ingress into Production Systems................................................. 20
Beam Pumped Oil Wells ................................................................... 20
Electric Submersible Pumps (ESPs) ................................................ 21
Air Entrainment in Tanks .................................................................. 21
Air Entrainment in Transfer and Injection Pumps ............................. 21
Injection System Leaks ..................................................................... 22
Vapor Recovery Systems ................................................................. 22
Air Content in Water Sources ........................................................... 23
Removal of Oxygen from Injection Waters ....................................... 24
Vacuum Deaeration .......................................................................... 24
Gas Stripping .................................................................................... 25
Catalytic Oxygen Removal ............................................................... 25
Other New Oxygen Removal Techologies ....................................... 25

iv
 CHAPTER II – OIL WELL CORROSION ALLEVIATION

Description Page

Introduction.............................................................................................. 27
Beam Pumping ........................................................................................ 29
Design Bases .................................................................................... 29
Operating Procedures....................................................................... 30
Rod Failures ..................................................................................... 30
Corrosion Failures ............................................................................ 31
Tubing Failures ................................................................................. 32
Pump Failures ................................................................................... 32
Guides for the Estimation of Corrosion............................................. 32
Water Cut and pH ............................................................................. 32
Field Measured pH on Fresh Samples ............................................. 33
Thirty-Day Flow Line Corrosion Coupon Data .................................. 33
Field Detection of Corrosion ............................................................. 33
Design of a Corrosion Inhibition System .......................................... 35
Application of the Corrosion Inhibition System ................................. 37
Initial Treating Frequency ................................................................. 38
Treating Procedure ........................................................................... 39
Gas Lift .................................................................................................... 39
Electrical Submersible Pumps (ESPs) .................................................... 41
Hydraulically Pumped Oil Wells .............................................................. 43
Improved Oil Recovery (IOR) .................................................................. 43
Water Floods ........................................................................................... 44
CO2 Miscible Floods ................................................................................ 44
Steam Floods .......................................................................................... 45
Other Thermal Recovery Methods .......................................................... 47

v
 CHAPTER III – GAS WELL CORROSION

Description Page

Introduction.............................................................................................. 48
Calculating Partial Pressures .................................................................. 48
Predicting CO2 Corrosivity ...................................................................... 49
Designing a Corrosion Inhibition System ................................................ 50
Continuous Corrosion Inhibition .............................................................. 50
Batch Inhibition ........................................................................................ 52
An Effective Corrosion Inhibitor............................................................... 52
Sufficient Corrosion Inhibitor Volume ...................................................... 52
Batch Inhibition Techniques .................................................................... 53
Batch Frequency ..................................................................................... 54

CHAPTER IV – CORROSION ALLEVIATION IN OIL AND GAS

PIPELINES AND FLOW LINES

Description Page

Introduction.............................................................................................. 57
Economics of Corrosion .......................................................................... 57
Safety ...................................................................................................... 57
Environment ............................................................................................ 58
Gas Transmission Pipelines.................................................................... 58
Oil Transmission Pipelines ...................................................................... 59
Multiphase Pipelines and Flow Lines ...................................................... 60
Corrosion Prediction ................................................................................ 61
Corrosion Inhibition ................................................................................. 61
Pigging .................................................................................................... 62
Hydrotesting of Pipelines ........................................................................ 63
Land Operations ...................................................................................... 64
Marine Operations ................................................................................... 64

vi
 CHAPTER V – MONITORING OF CORROSION ALLEVIATION
EFFECTIVENESS

Description Page

Introduction.............................................................................................. 66
Direct Monitoring ..................................................................................... 66
Indirect or Interpretive Monitoring ........................................................... 66
Proactive and Reactive Monitoring ......................................................... 66
Corrosion Coupons ................................................................................. 67
Probes, Devices and Techniques for External Monitoring of Internal
Corrosion ................................................................................................. 68
Electrical-Resistance (ER) Probes .......................................................... 69
Hydrogen Patch ...................................................................................... 69
Multi-End Element Tube.......................................................................... 69
X-Rays ..................................................................................................... 70
Ultrasonics............................................................................................... 70
Wet Chemistry ......................................................................................... 70

CHAPTER VI – CORROSION ALLEVIATION WITHOUT CORROSION

INHIBITION

Description Page

Introduction.............................................................................................. 72
Corrosion-Resistant Alloys (CRA) ........................................................... 72
Non-Metallics........................................................................................... 74
Non-Metallic Coatings ............................................................................. 74
Solid Non-Metallic Materials.................................................................... 75
Galvanic Corrosion .................................................................................. 75

References .............................................................................................. 72
Glossary .................................................................................................. 82

vii
 !
!"#
Introduction
The term “corrosion” derives from the Latin word “corrosio,” the act of
gnawing. There are three basic types of metal degradation: wear, metal
dissolution, and cracking. Corrosion relates to the latter two types of
degradation. Wear is a design issue related to material selection and is
outside of the general context of corrosion; wear will not be addressed in
this handbook.

Dissolution of metal is the usual meaning of corrosion and will be the

major topic of this document. Metal dissolution can occur in many forms
and each form is dependent upon the environment. These forms of
corrosion include general weight loss corrosion, pitting corrosion, or
crevice corrosion. (A crevice is a natural pit initiation site and is treated
the same as pitting corrosion.)

Cracking, with accompanying corrosion, will be dealt with briefly in an

empirical way. In the oil field, cracking is avoided by proper material
selection and never by corrosion inhibition. The reader will be referred to
other standards, which deal with cracking alleviation.

Corrosion alleviation is important to both reputation and profits. The

highest priority is to prevent corrosion-related failures, which jeopardize
the personal safety of both company employees and the general public.
Corrosion failures that cause environmental damage are also a very high
priority. Finally, prevention of other economic loss such as equipment
loss (or malfunction) and production loss is a priority. All of the above
concerns are of importance jointly or severally.

Nearly always, some type of corrosion alleviation strategy is required

before all the oil or gas reserves of any field are depleted. Thus, the real
question is not if a corrosion alleviation system will be needed. The
relevant questions are which corrosion alleviation system to use and
when to apply it.

In order to economically alleviate corrosion in oil and gas production, it is

necessary to have a rudimentary knowledge of corrosion basics and the

1
tools available both to predict and alleviate the corrosion. While
laboratory tests and computer models give insight into reality, it is
important to remember that these tests and models are not reality.
Applications of these laboratory tests and models in the field are reality.

Corrosion occurs because most materials used in oil and gas production
operations are thermodynamically unstable in their environment. The
purpose of this document is to provide a basic knowledge of corrosion
mechanisms, to recognize corrosion mechanisms in operations, and to
design economically reliable corrosion alleviation systems. The principal
corrodents in oil and gas production are CO2, H2S, oxygen, chloride ions,
and bacteria. All of these corrodents use water as a medium.

Corrosion Fundamentals
The term “corrosion” commonly means metal dissolution. (Stress
Corrosion Cracking will be handled in Chapter VI and Sulfide Stress
Cracking, etc. will be cursorily handled later in this chapter.) Corrosion
processes are usually divided into chemical and electrochemical. This
division is arbitrary because chemical corrosion often involves
electrochemical processes within the surface layer. For practical
purposes, all corrosion is electrochemical and takes place in an
electrolyte and is the result of electron transfer between the corrodent
and the metal. In this document all corrosion will be considered
electrochemical.

Since all corrosion is electrochemical there are always two reactions as

shown in Figure 1:
0 ++ -,
The anodic reaction, Fe Fe + 2e and the cathodic
+ -
reaction, 2H + 2e H2

Figure 1

2
The rate of corrosion is affected by many environmental parameters, but
the rate of corrosion can be restricted by either the anodic or cathodic
reaction. Since corrosion seldom proceeds at unrestrained rates >3,000
mpy (mils per year), corrosion is always restricted by either the cathodic
or anodic reaction. The metal is always the anode, but the cathode
usually restricts the corrosion rate.

If oxygen is present, two other reactions are possible:

+ -
O2 + 4H + 4e 2H2O
- -
O2 + 2H2O + 4e 4OH

Again, the metal loss is always at the anode and typical anodic reactions
are as follows:
o ++ -
Fe Fe + 2e

In actual conditions, the cathode and anode are both in the metal
(Figure 2). The electrical potential between the anode and the cathode
in the same metal can be caused by many things such as inclusions,
small variations in compositions, surface roughness, and surface films.

Figure 2

The final corrosion product can precipitate as one of many products such
as iron sulfide, iron carbonate, iron oxide, or iron hydroxide, dependent
upon the water chemistry.

3
Typical gross corrosion reactions are as follows:

Fe + H2S FeS + H2
Fe + S FeS

Fe + H2O + CO2 FeCO3 + H2

4Fe + 3O2 2Fe2O3

For HCl (Hydrochloric acid), ferrous chloride is usually formed as shown

below. When HCl dissolves rust, some ferric chloride (FeCl3) will also be
formed.
+ -
Fe + 2H + 2Cl FeCl2 + H2

Water solutions of both carbon dioxide and hydrogen sulfide are acidic:

H2O + CO2 H2CO3 (carbonic acid)

Hydrogen sulfide in H2O:

+ -
H 2S H + HS
and
- + =
HS H +S

Corrosion products, and thus the corrosion rates, are both controlled by
the dominating corrosion mechanism. As will be explained more fully
later, H2S corrosion is mass transport limited and velocity independent
and the corrosion products are iron sulfide. Carbon dioxide corrosion is
velocity sensitive and the corrosion products are the more soluble iron
carbonate. It has been published that unless the CO2/H2S partial
pressure ratio is greater than 500, the H2S corrosion mechanism will be
the controlling mechanism. The corrosion will be velocity independent
and the corrosion products will be iron sulfide. Although the exact ratio is
dependent on the existing conditions, under all conditions if the CO2/H2S
partial pressure ratio is not above 100, H2S corrosion will control and
probably the ratio has to be above 300-500 before CO2 corrosion will
control. Hydrogen sulfide and CO2 are often produced together and are
probably synergistic, but the interrelationship and resulting corrosion is
not well understood. Some investigators believe that CO2 enhances
pitting in H2S corrosion.

4
 $ %!

Carbon dioxide corrosion was first encountered in Texas gas wells in

1943 and again manifested itself in oil wells in 1944 in Louisiana.
Organic acids discovered in 1944 were determined to be secondary and
ignored for 30 years. Carbon dioxide corrosion increased as the
production matured and water production increased. By 1948, the cost
of CO2 corrosion (lost revenue, replacement costs, etc.) was about 7% of
the value of the gas.

Corrosion from CO2 manifests itself in two forms: mesa corrosion

(shallow pitting), shown in Figures 3 and 4 and corrosion-erosion.

Figure 3 – CO2 Corrosion in Bottom of Flow Line

5
 Figure 4 – CO2 Mesa Corrosion

Mesa corrosion got its name because the area in between the pits, which
have round bottoms, smooth sides, and sharp edges, look like mesas.
Both forms can be severe, with corrosion-erosion being the more severe.
Although as mentioned above, mesa corrosion can be severe, CO2
corrosion is velocity sensitive, with higher velocities more corrosive and
harder to inhibit than lower velocities.

In the early days, CO2 severity was estimated by CO2 partial pressure
and there were various rules of thumb. With CO2 partial pressures
above 30 psia, corrosion problems were expected. With CO2 partial
pressures below 3 psia or 7 psia (dependent upon which rule of thumb)
CO2 was not expected to be a problem. The CO2 partial pressures
below 30 psia and above 7 psia were environments that might be
corrosive.

CO2 corrosion first manifested itself in gas condensate wells, which

produced no connate water and only water of condensation. With fresh
unbuffered water and CO2 at cool temperatures in the well tubing, the pH
was low and mesa corrosion was often severe, even at low velocities.
Pitting corrosion rates greater than 300 mpy were common.

Many scientists have studied CO2 corrosion and numerous papers have
been written on the corrosion mechanism and predicted corrosion rates.
Much of this research can be summarized as follows: CO2 forms
carbonic acid (H2CO3) a weak acid, when it is dissolved in water and
corrosion rates are inversely related to pH. However, the corrosion
mechanism is not similar to strong acid corrosion, i.e. hydrochloric acid
(HCl), and is probably dominated by the cathodic reaction.

6
Carbonic acid, as a weak acid, ionizes very little. According to de
Waard, only about one molecule in 1,000 molecules of carbonic acid
ionizes at normal temperatures as follows:
+ -
H2CO3 H + HCO3
- =
The number of bicarbonate ions (HCO3) that ionize to carbonate (CO3)
+
and a hydrogen ion (H ) is less than one in 1,000.

However, for CO2 corrosion, the carbonic acid does not need to ionize
and the strong acid H+ is not required for corrosion. The cathodic
reaction is believed to be as follows:
- o -
H2CO3 + e H + HCO3
- - o =
HCO3 + e H + CO3
o o
H +H H2

The anodic reaction is as follows:

o ++ -
Fe Fe + 2e

Below 175°F (Fahrenheit), as temperature increases, CO2 corrosion

rates go up for the same pH. Above this temperature, the iron carbonate
film, dependent upon flow regime, starts to offer some protection against
corrosion for two reasons: the film is more tenacious and its solubility in
water goes down as the temperature increases.

Much historical data ignored the water chemistry and only looked at CO2
partial pressure and the calculated resultant pH. Subsequent studies
have shown that water chemistry is important and these studies continue
to refine water chemistry parameters. The dominant reason that CO2
corrosion concerns were questionable over a large range of CO2 partial
pressures (3 psia - 30 psia) was water chemistry (bicarbonates and
organic acids). Carbon dioxide in the reservoir is usually in contact with
calcium carbonate and in equilibrium with calcium bicarbonate in the
water. For moderate CO2 partial pressures, highly buffered water, if
produced in sufficient quantities sometimes alleviated the CO2 corrosion.
Further, for many years, organic acids were titrated and reported as
bicarbonate alkalinity.

Organic acids in produced well fluids have been known since 1944 and
were ignored for more than the next 30 years. Organic acids, such as
acetic, propionic, butyric, and valeric, are too weak to be a corrosion
problem on their own. Carbon dioxide controls the pH. In relatively non-

7
buffered waters, where CO2 can drive down the pH, organic acids can be
very corrosive, even at CO2 partial pressures below 3 psia. Driving this
unexpected corrosion is the relatively high solubility of iron acetate in
comparison to iron carbonate. Produced water rates and chemistry are
an essential part of modeling CO2 corrosion and CO2 corrosion inhibition.

As stated earlier, corrosion-erosion is a form of CO2 corrosion and this

CO2 corrosion is definitely velocity-dominated. For constant CO2 partial
pressure, corrosion increases with increasing velocity. Chlorides also
aggravate CO2 corrosion. As will be discussed later, the higher the wall
shear stress (velocity) the more difficult it is to inhibit CO2 corrosion.

Changes in metallurgy can also affect CO2 corrosion. Weldments,

including the heat affected zone, are nearly always more difficult to
inhibit than the pipe or vessel body (Figure 5). Corrosion tests should
always include a weldment if welding is done on the production system
to be corrosion inhibited. Upset API tubing often suffered ringworm
corrosion at the interface of the tubular steel and the upset joint because
microstructure dissimilarities in the metal set up electrochemical
corrosion cells. Normalizing (a type of heat treatment) the whole tubing
joint after upsetting solved this problem.

Figure 5 – Preferential Corrosion of a Weld Seam

8
Carbon Dioxide Corrosion-Erosion
Corrosion-erosion is not the result of erosion. Corrosion-erosion is really
velocity-related corrosion. And as velocity-related corrosion, it only
applies to CO2 corrosion; H2S corrosion is velocity independent.

To clear the air, a brief discussion of erosion is in order before velocity-

related corrosion is discussed.

Erosion
Solids free oil field fluids, oil, water, and gas, are usually of insufficient
velocity to erode steel. For example, a liquid droplet on cast iron
requires a velocity of more than 300 ft/sec (feet per second) to erode the
cast iron. For stainless steels, the velocity is about half that velocity, 125
ft/sec +/-.

It is common knowledge that solids and gas can be very erosive, as in

sand blasting. When liquids are present with natural gas-suspended
solids, erosion is much less. The liquid reduces the velocity of the
particles hitting the metal. A major oil company found more than 40
years ago that, after oil field liquids and gases are separated, the sand-
laden liquid has much less erosive properties.

“API 14E Recommended Practice for Design and Installation of Offshore

Production Platform Piping Systems,” has a velocity guide for oil and gas
piping. However, it is an approximation, sometimes too conservative,
sometimes not conservative enough, and sometimes probably about
right. Few major oil companies rigorously follow it in flow line design.

Velocity-Related Corrosion
What is so called corrosion-erosion? Corrosion-erosion is more
accurately called velocity-related corrosion, which applies only to CO2
corrosion.

For corrosion to proceed, the iron corrosion product must continue to go

into solution. For some metals, the corrosion product forms a film, which
reduces or eliminates further corrosion. This is the case with aluminum.
Aluminum oxidizes nearly immediately in air, but the aluminum oxide
coating forms a tight barrier and prevents further corrosion. Even with
aluminum cookware, the aluminum oxide film prevents further oxidation
at oven temperatures.

The corrosion product for steel and carbon dioxide is iron carbonate,
although simple thermodynamic calculations suggest that at higher
temperatures the solid stable phase could be magnetite (Fe3O4).

9
Japanese investigators report that the CO2 product of low alloy steel
corrosion in sweet brine is exclusively ferrous carbonate (FeCO3) up to
482° F. Although ferrous carbonate is not particularly soluble in water, at
low temperatures it does not form a very good barrier to further
corrosion; CO2 corrosion proceeds as fast as the ferrous carbonate,
which is not very soluble in water, can go into solution.

There are several ways to increase the dissolution rate of iron carbonate.
Lower pH, which is reflected by higher CO2 partial pressures, can
increase iron carbonate dissolution rates and thus CO2 corrosion rates.
The addition of organic acids to the water also increases the dissolution
of iron. In simple terms, the CO2 lowers the pH which increases the
dissociation of organic acids such as acetic acid and allows iron acetate
to form. Iron acetate is much more soluble in water than iron carbonate.

Increasing the salinity will increase the solubility of iron carbonate and
enhance CO2 corrosion.

There is one thing that will enhance all these mechanisms of increased
CO2 corrosion: a continual supply of new water that contains little or no
corrosion products. The fastest way to bring more fresh liquid to the
corrosion site is with increased velocity, thus the term corrosion-erosion.
The high velocity liquids remove the corrosion products, suspended or
dissolved, as fast as they are formed. These high velocities provide a
fresh metal surface, with no protective film of corrosion products, for
which the CO2 to react (Figure 6).

Figure 6 – CO2 Corrosion-Erosion

There is one small caveat to the above. At low temperatures, the iron
carbonate is not particularly protective against retarding more corrosion,
but its effect is not zero. However at temperatures above about 175° F,
the iron carbonate forms a very protective film that retards further
corrosion. However, this film does not stand up to high velocity
conditions. Thus, at very high velocities, very high CO2 corrosion rates

10
are also possible at high temperatures (>175° F), high salinity, and high
CO2 partial pressures.

Hydrogen Sulfide Corrosion

Hydrogen sulfide causes two types of corrosion in steel; metal loss
corrosion and SSC (Sulfide Stress Cracking). The former corrosion will
be discussed in this section and the cracking issues will be discussed in
a later section.

In regard to H2S corrosion, H2S is seldom produced alone. Carbon

dioxide is usually produced in conjunction with H2S, but as stated above,
unless the CO2 partial pressure is over 500 times greater than the H2S
partial pressure, H2S will usually be the corrodent. Although there may
be more than one corrosion mechanism, the most documented one is
shown below (Figure 7). Chloride is a catalyst and very little chloride ion
is required. The H2S reacts with the chloride to form hydrochloric acid,
which reacts with the steel to form iron chloride, which reacts with the
H2S to form iron sulfide and free the chloride to form more HCl and
repeat the process.

Examination of the corrosion

mechanism in Figure 7 reveals that
the rate of corrosion is dependent
upon how fast H2S can be
transported to the base steel via
process intermediaries. Velocity
plays no part in this mass transport.
Velocity does not carry away
corrosion products and corrosion
occurs at the same rate whether the
H2S is flowing or static. H 2S
corrosion is mass transport
controlled and is velocity
independent (Figure 8). Hydrogen
sulfide corrosion with H2S partial
pressures of more than 6,000 psia
and with velocities in excess of 100
Figure 7
ft/sec have been controlled
successfully with corrosion inhibition.

There is no consensus on the corrosion effect of CO2 in conjunction with

H2S. The above example has approximately 1,000 psia CO2 partial
pressure, and the CO2 did not seem to measurably affect either the
uninhibited corrosion rate or the ability to inhibit corrosion. Often when

11
produced gas contains high concentrations of H2S, the gas precipitates
elemental sulfur as the gas is depressured and cooled. Although there is
a lot of data in the literature in regard to how much elemental sulfur the
sour gas could contain if it is saturated with elemental sulfur, in fact, the
sour gas is seldom saturated with elemental sulfur. Thus, the amount of
elemental sulfur precipitated, if any, is nearly always not known until the
well is produced.

Elemental sulfur greatly increases the corrosivity of the gas and

corrosion inhibition is extremely difficult and probably impractical. The
current successful way of handling the corrosivity of precipitated
elemental sulfur is to add enough oil to dissolve the sulfur and then add a
corrosion inhibitor to the oil.

Figure 8 – H2S Pitting Corrosion

Sulfide Stress Cracking (SSC)

Sulfide Stress Cracking (SSC) was formerly known as SSCC or Sulfide
Stress Corrosion Cracking. NACE (National Association of Corrosion
Engineers International) adopted SSC as the correct description of the
metallic cracking caused by H2S.

The gross corrosion reactions are as follows:

H2S + Fe FeS + H° +H°

H° + H° H2

12
 o
However, the sulfide poisons the recombination of nascent hydrogen (H )
into hydrogen gas (H2), and the nascent hydrogen stays in the steel, and
permeates the metal matrix and causes embrittlement; cracking of the
steel may result. The primary cracking form in the oil field is SSC.
Hydrogen Induced Cracking (HIC), Stress Oriented Hydrogen Induced
Cracking (SOHIC), and hydrogen blistering are also possible. Hydrogen
blistering, HIC, and SOHIC happen when the nascent hydrogen collects
at voids in the steel matrix, usually around inclusions, and forms
hydrogen gas. Hydrogen induced cracking and SOHIC are more
common in controlled rolled steels; these forms of cracking are seldom
experienced in seamless tubulars. Hydrogen blistering usually occurs in
low strength steels and is seldom experienced in the oil field.

Sulfide stress cracking can be rapid and catastrophic. Historically, the

NACE MR0175 standard was the document used to select steels,
including stainless steels, which would resist environmental cracking,
primarily SSC. Recently, MR0175 has been updated. International
Organization for Standardization (ISO) standards ISO 15156-1, ISO
15156-2, and ISO 15156-3 are the international versions of NACE
MR0175. ISO has appointed NACE as custodian and administrator of
ISO 15161-1, ISO 15161-2 and ISO 15161-3. Most U.S. state and
federal regulations require compliance with NACE MR0175.

It is beyond the scope of this document to discuss SSC in detail.

Although SSC is Sulfide Stress (Corrosion) Cracking and thus requires
“Corrosion” to occur, (the corrosion produces the nascent hydrogen)
SSC is avoided by material selection, not by corrosion inhibition.
Corrosion inhibition systems are seldom 100% effective, 100% of the
time and since steel SSC failures can be fast and catastrophic, SSC (as
well as HIC, and SOHIC) are avoided by selecting cracking-resistant
materials.

# $ &# ' (" # ! ) *

Introduction
Microbiological Influenced Corrosion (MIC) is a complex and difficult field
and this document does not comprehensively address this complex
science. This discussion will address the major issues of MIC related to
oil and gas production.

Bacteria
Bacteria are endemic to the oil field. No system is ever sterile; bacteria
are always present. This author has never tested for SRB (Sulfate
Reducing Bacteria) and not found them. The question is not “are

13
bacteria present?” They are present. The questions are “are the
conditions right for the bacteria to flourish and grow and if they grow will
they cause a problem that needs to be addressed?”

All bacteria produce a slimy coating called a capsule or biofilm. This

capsular material is primarily polymeric starches and is called
exopolysaccharide. Some produce more slime than others. Planktonic
bacteria are free-floating and sessile bacteria are attached to a surface.
Science has identified more than 5,000 strains of bacteria and the oil
field has its share. Bacteria are classified many ways: shape,
environment, temperature, nutrient systems, etc. The nutrient system is
most commonly used in the oil field: sulfate reducing bacteria (SRB),
nitrogen reducing bacteria (NRB), acid producing bacteria (APB), and
iron oxidizing bacteria (IOB). Sulfate reducing bacteria are the dominant
bacteria in the oil field, or at least they cause the majority of the
problems. But SRB do not cause all the bacterial corrosion-related
problems. Acid producing bacteria are sometimes found, and in
combination with SRB, their combined effects can be synergistic in
regard to MIC corrosion. It is important to be mindful that in oil field
corrosion problems, although one corrosion mechanism may dominate,
usually several corrosion mechanisms are concurrently at work. The
following discussion is essentially limited to SRB with a few initial
comments about APB; in regard to reservoir souring, NRB is briefly
discussed.

Acid Producing Bacteria (APB)

Acid producing bacteria are often found in old water flood fields,
particularly where the water has been contaminated with polymers.
Although the author knows of no systematic study of APB and polymers,
there are many cases where APB flourish following well fracture
treatments where gels (polymers) are used both to reduce friction and
build sufficient viscosity to carry fracture proppants.

In one field, well pulls per year were reduced more than ten-fold when
the amount of biocide, THPS (Tetrakis Hydroxymethyl Phosphonium
Sulfate) in this case, added to the injected fluids was increased to 8,000
ppm (parts per million) from 1,000 ppm. It is strongly recommended that
the fluid injection tanks be checked for APB and SRB before fracture
treatments, even if the pumping vendors add a biocide package. APB
can cause significant corrosion failures in wells if left unchecked.

Sulfate Reducing Bacteria (SRB)

Sulfate Reducing Bacteria (SRB) are anaerobic bacteria, meaning they
flourish in oxygen-free environments. Oxygen does not kill the SRB;

14
they just become dormant, and when the oxygen is removed they can
flourish again. Sulfate reducing bacteria can live and flourish in a macro-
oxygenated environment in micro-anaerobic niches, such as under
biofilms.

In gross terms, SRB metabolize sulfates and sulfites in combination with

low molecular weight organic acids (formic, acetic, propionic) and emit
H2S. It is the H2S that causes the problem for oil and gas operations.
Sulfate reducing bacteria are very adaptive to their environment,
including temperature, salinity, pressure, pH, etc. Pressures to 8,000
psig (pounds per square inch gauge) have shown no effect on SRB
growth up to temperatures of 208° F. Sulfate reducing bacteria have
been reported to grow at 219° F and 6,000 psig.

SRB Generated H2S

The H2S generated from SRB causes three problems:
• The H2S reacts with the system iron in the presence of water to
cause iron sulfide particles.
• The H2S by its reaction with the iron causes corrosion.
• The H2S causes normally sweet reservoirs to turn sour and may
cause metallurgical failures (SSC) in the production system.

Iron Sulfide Particles

The iron sulfide particles that result from the H2S generated by SRB are
usually the most serious problem caused by SRB. This iron sulfide often
becomes oil wet and this makes their presence more troublesome.
These particles plug injection wells; they cause problems with oil
dehydration and with water de-oiling. The particles cannot be effectively
filtered out, because the planktonic bacteria move and generate more
H2S downstream of the filters which results in more H2S and more iron
sulfide particles.

Corrosion from SRB Generated H2S

Corrosion per se from SRB-generated H2S is not usually a serious
problem. Many studies have indicated that the planktonic SRB cause a
maximum of 3 mpy of uniform corrosion and usually less than 3 mpy
corrosion. Other corrosion concerns often far outweigh the modest 3
mpy or less from SRB-generated H2S. A different type of corrosion
occurs when the SRB become sessile. This MIC corrosion can usually
be identified by shallow circular pits as shown in Figure 9. Again, the
corrosion rate is usually small and is usually only of importance in gas
and oil transmission pipelines with low corrosion allowance and which
are expected to be in service for many years.

15
However, MIC corrosion can vary widely, and a low corrosion rate is not
always the case. MIC corrosion, particularly when combined with APB
and other corrosion mechanisms, can sometimes be severe.

There are cases in sour rod pumped wells that failures are of serious
economic consequence when bacterial growth is not controlled. As
mentioned above, in one field, control of bacterial growth with adequate
biocide treatment reduced well pulls by 90%.

Figure 9 – Microbiologically Induced Corrosion by Sulfate Reducing Bacteria

Reservoir Souring
A final concern is reservoir souring. When sulfate-laden waters such as
seawater are injected into reservoirs for pressure maintenance or for
water floods, they carry with them SRB, which are endemic to our
environment. The reservoir usually provides the other necessary
nutrients and the SRBs thrive and produce H2S. Many times, the
production facilities were not designed to resist H2S cracking (SSC,
discussed earlier) and the facilities or wells require major renovation just
because of reservoir souring caused by SRB.

One of the ways operators try to alleviate reservoir souring is to inject the
nitrate ion, usually as calcium nitrate. The hope is that NRB (Nitrogen
Reducing Bacteria) will consume the same short-chain organic acids that
SRB need to metabolize sulfate and thus starve the SRB. This

16
technique is in its infancy. Obviously, if the short-chain organic acids are
in abundant supply and there is enough food for both the SRB and the
NRB, both may flourish, dependent upon the nitrogen and sulfate
supplies.

SRB Detection
There are many techniques to determine the relative presence of SRB.
Practical detection techniques are not quantitative, but qualitative. In the
oil field, the serial dilution method is the dominant technique used to
measure SRB presence. The technique is described by API RP38.
NACE also has documents on this procedure. The important thing to
remember is that SRB by themselves are not bad; it is the H2S they
generate that causes problems. Potentially, although less common, the
SRB biofilm can provide an environment for under deposit corrosion.

SRB Control
Sulfate reducing bacteria are never eliminated except in rare cases with
Herculean efforts. Thus, since SRB cannot be eliminated, the control
technique will depend on why they are causing a problem to the
operator. The most common biocides are quaternary amines and
aldehydes, formaldehyde, glutaraldehyde and acrolein. Glutaraldehyde
is the most common. Acrolein is a very effective biocide, but because of
its vapor pressure and toxicity, it is seldom used offshore. Its toxicity
prevents its use where the treated water is surface-disposed. But,
because it is an unsaturated aldehyde, it is more reactive and, if
accidentally released, more readily neutralized in receiving waters.
Tetrakis Hydroxymethyl Phosphonium Sulfate is a more environmentally
friendly biocide, but it is more expensive than glutaraldehyde. For this
apparent reason, glutaraldehyde is used more often than THPS.
However, THPS is usually used when the treated solution must be
surface-disposed, which is often the case with hydrotest water.

The first line of defense should be in the design and operation of the
production system. Dead spots should be eliminated wherever possible.
Filters and ion-exchange beds are preferred SRB lodging places.
Velocities should be kept above 3 ft/sec. Pipelines should be pigged to
remove biofilm. Pigging both reduces corrosion by removing the sludge
where SRB can prosper and it facilitates effectiveness of biocides to kill
the bacteria. Another way to control SRB is to reduce the food supply for
them, namely sulfate or sulfite (sulfites are used for oxygen removal).
Sulfites are the preferred food for SRB; for SRBs it is more easily
metabolized and oxygen removal by the sulfites provides the anaerobic
environment the SRB need. In the case of reservoir souring, the
introduction of NRB may also reduce SRB growth.

17
The above cannot always be accomplished for practical reasons and
SRB will have to be controlled with biocides. To be effective, the biocide
must contact the bacteria. This can be difficult and may require some
innovation. In pipelines, pigging is always preferable prior to biocide
treatment to remove biofilm in order to treat the bacteria next to the pipe.
Treating bacteria growing in filters is also challenging. Filter
backwashing usually enhances biocide effectiveness. Multiple injection
points and multiple treating techniques may be required.

There are three basic techniques used to inject biocides: batch, semi-
continuous (slug), and continuous. Batch treatment is a short-term
exposure, one to four hours, in the system with a high concentration of
biocide; 200 ppm to 500 ppm of biocide is a good starting point. Slug
treatment is usually a lower concentration of biocide with a longer
exposure time on a regular interval such as three to five hours once per
day; 100 ppm to 500 ppm biocide is a good starting point. For
continuous treatment, 20 ppm to 50 ppm is a good starting point. Slug or
batch treatments are often used in combination with continuous
treatment.

The treatment technique and concentration will depend on the problem

being addressed. Sessile bacteria corrosion in pipelines may be treated
by pigging and batch treatment. Control of severe planktonic SRB may
require continuous injection. Slug treatments may more economically
control modest SRB problems.

In at least one instance, reservoir souring was averted by treating all

injected seawater with ultraviolet radiation to kill the SRB in the
seawater. This technique was used in combination with biocide
treatment of all fluids that contacted the reservoir (drilling fluids,
completion fluids, stimulation fluids, and occasional biocide treatments of
injection wells).

Obviously, the use of biocides must comply with all legal requirements
(safety and environmental) and must be compatible with all system fluids
and materials.

SRB Monitoring
Monitoring the effectiveness of an SRB control program also depends on
the problem being addressed. The easiest way is to monitor SRB is with
serial dilutions. If SRB are controlled, the problem is alleviated. If iron
sulfide particles are the problem being addressed, then Millipore filters
may be used to monitor the iron sulfide particles in the system. The
control of remote sessile bacteria corrosion in pipelines may be

18
measured by smart pigs. Iron or manganese sulfides are both water
insoluble and determination of their concentrations in produced water in
sour environments is not, in the author’s opinion, a reliable technique to
monitor corrosion. There are varying opinions on this subject in the
literature.

%'& ! !"#

Introduction
Forty years ago, oxygen corrosion was a serious corrosion problem in oil
field operations. Systems were open to the atmosphere, oxygen
entrance was not well understood, and it only takes ppb (part per billion)
of dissolved oxygen to be corrosive. It is generally agreed, that oxygen
content of water must be 20 ppb or less if oxygen corrosion is to be
avoided. In addition, oxygen corrosion is of the pitting type and can
penetrate a steel pipe or vessel in a short time with a small-diameter
deep pit. Since oxygen corrosion is nearly always of the pitting type,
when oxygen is in combination with CO2, the pitting tendency of CO2 is
enhanced.

Oxygen cannot coexist long-term with H2S. The oxygen and H2S react to
form elemental sulfur and water.

H2S + O2 = S° + H2O

As will be discussed later, this can be a very corrosive combination.

However, with current technology and with well-designed and carefully

operated systems, air, the usual source of oxygen, can be eliminated
and the resulting oxygen corrosion can be avoided. Oxygen nearly
always comes from air and it is introduced into the production system
either through air leaks or the use of surface water containing dissolved
oxygen. Subsurface water, except shallow fresh water sources, never
contains dissolved oxygen in the reservoir.

Not very many metals are resistant to oxygen corrosion. The martensitic
SS (stainless steels), such as 410 SS, are not resistant to oxygen
corrosion; for instance, 410 SS valve parts have been known to corrode
on the warehouse shelf. Austenitic stainless steels, in the presence of
both oxygen and chloride ions are subject to pitting corrosion. These
alloys include 304 SS, 316 SS, Incoloy 825, Hastelloy G-3, and Inconel
625. This pitting tendency goes up as the temperature increases and
goes down as the non-ferrous content of the alloy goes up. (Chromium
and molybdenum are elements that help corrosion-resistant alloys avoid

19
pitting.) For example, in the presence of oxygen, Inconel 625 starts to pit
at approximately 4,800 ppm chloride at 212° F, while 316 SS pits with
approximately 250 ppm chloride at 212° F. At 4,800 ppm chloride and
oxygen, 316 SS pits at approximately 64° F.

The copper-nickel alloys are resistant to oxygen corrosion (Monel is an

early member of in this family of alloys.) and are often used in seawater
service; for example, seawater cooling systems for boats or TLP
(Tension Leg Platform) ballast water use copper-nickel. However, these
alloys cannot stand polluted water containing sulfides (100 ppb). These
sulfides coat the surface of copper nickel alloys and act as a catalyst for
oxygen corrosion of the copper nickel. An ocean going vessel, after
cruising through a polluted harbor, suffered a failure of its once-through
copper nickel seawater cooling system and sank because the sulfur
compounds in the harbor water caused the copper nickel to unabatedly
corrode.

Air Ingress into Production Systems

Air ingress into production systems can be from a variety of sources.
Many seem insignificant. In the following example of the reaction of
oxygen and H2S, the results of slight oxygen contamination can be
catastrophic. In one sour-gas operation, stripping gas, as is often the
case, was used to remove the final traces of water from glycol. In this
case, the stripping gas was contaminated with a small amount of air,
which in turn transferred a small amount of oxygen to the glycol, which in
turn contaminated the dehydrated sour gas with oxygen. Sour gas is
non-corrosive when completely dry. In this instance, the oxygen reacted
with the H2S and formed elemental sulfur and water, which in turn
corroded the gathering pipeline. A fatality resulted from the sour gas
pipeline leak. It is this type of problem that causes offshore operators to
design systems that avoid the need for vapor-recovery systems (which
may pull in air occasionally) in sour production operations whenever
possible.

The following review of more common sources of oxygen (air) in oil and
gas production systems is followed by a review of oxygen removal from
oxygen contaminated surface waters.

Beam Pumped Oil Wells

When oil wells are beam pumped with the annulus closed or with a high
annular fluid level, air ingress is unlikely.

A more frequent source of air entrainment is the polished rod stuffing

box, particularly when wells are operated in a “pumped-off” condition. In

20
this condition, pump efficiency is low and a slight vacuum will occur at
the wellhead with each pump stroke. Typical polished rod stuffing box
seals are designed to contain internal pressure, not internal vacuum.
Unless the stuffing box is tight and will hold a vacuum, air will
intermittently enter the well fluids. This type of leakage can also occur in
other packing type seals between the wellhead and flowline check valve.
It is important in stripper well production that all wellhead connections
and packing assemblies be able to hold a vacuum.

Electric Submersible Pumps (ESPs)

Air ingress into the annuli of wells produced via ESPs is usually not a
problem. These wells produce lots of fluid (the reason for the ESP) and
a vacuum is seldom on the annulus. Usually, sufficient gas is produced
with the high volume of liquid to keep the annulus air free. An exception
to this rule is ESPs for water source wells, which have no associated
gas.

Air Entrainment in Tanks

Open water tanks are often a source of air contamination of the
production system. All water tanks should be gas-blanketed. The
normal gas-blanket pressure is several inches of water or a few ounces
of pressure (4 ounces of pressure approximates 7 inches of water).
Tanks are always equipped with vacuum-relief valves to prevent tank
collapse as the tank is emptied. The gas-blanket source must be able to
deliver sufficient gas to the tank as it is emptied to prevent the vacuum
valve from opening and allowing air to enter the tank.

Gas-blanketing the water tank will keep out air and prevent oxygen
corrosion, but the air-free environment is ideal for SRB. If sludge
accumulates in the tank bottom, SRB may thrive. They must be
controlled to prevent MIC because tank bottoms do not usually have any
corrosion allowance. The SRB can be controlled by proper use of
biocides as described in the MIC section of this document.

Air Entrainment in Transfer and Injection Pumps

Pump installations are frequent sources of air contamination of water.
For a given set of pumping conditions, a positive displacement pump will
try to deliver the specified amount of fluid per stroke. When the liquid is
not available in sufficient quantity and at sufficient pressure to fill the
piston cavity, cavitation with accompanying partial vacuum will result.
Unless the packing glands on the pump suction and pump-suction piping
are vacuum-tight, air will be drawn into the system and oxygen will
dissolve in the water. Centrifugal pumps can also draw in air, but they
are usually less of a problem than positive displacement pumps. In

21
either case, this source of air can usually be prevented by proper design
of the liquid-supply system to the pump. The following are a few simple
rules, which will help reduce or eliminate air from being drawn into the
pump suction:

• Streamline suction piping by avoiding as many bends as

possible.
• Place the pump as close to the surge vessel as possible.
• Suction valves should be full-opening and through-ported.
• Follow all pump manufacturer recommendations including:
Supply pipe should be larger than pump-suction fitting.
Supply sufficient fluid head.
• Use suction pulsation dampeners for high speed positive-
displacement pumps.
• All connections and seals must be air-tight.

Injection System Leaks

Normally the pressure side of injection systems is not a problem, except
for leaks. Any leaks result in oxygen and water being mixed together,
which is corrosive. To aggravate the problem, when salt water is
pumped, any leak will usually leave a wet-salt deposit, which is severely
corrosive. Installing and maintaining pressure-tight connections and
seals will avoid all these problems.

Vapor Recovery Systems

Vapor recovery systems are notorious for pulling in air. In the effort to
collect all hydrocarbon vapors and prevent their venting to the
atmosphere, vapor recovery systems nearly always result in partial
vacuums from time to time and the result is air ingress. This
hydrocarbon gas then carries oxygen. If H2S is present, the H2S reacts
with oxygen to form elemental sulfur and water; the consequences of this
reaction can be severe corrosion. Any oxygen pulled into a production
process will eventually react with something. If liquid water is present, it
will react with the steel. If only dry hydrocarbons are present in the gas,
they will eventually react with the oxygen to form CO2. This is a very
slow reaction at ambient conditions.

22
Potential Sources of Air Ingress

Pumped Producing Well

• Well annulus
• Polished rod stuffing box
• Wellhead valves

Production facilities:
• Inadequate gas blanket on the produced water tank
• Water transfer-pump piping
• Water transfer-pump shaft seals

Water injection facilities:

• Open annulus of a water supply well
• Inadequate gas blanket on water tanks
• Any piping with a potential vacuum including:
Water meters
Wellhead valves
• Injection pump manifold suction and discharge components
• Injection pump seals

Air Content in Water Sources

Water is subsurface injected for numerous reasons, and water sources
also vary with the reason for injection. On land, all produced salt water
must be subsurface-disposed.

Water is used for water floods and pressure maintenance. In carbonate

reservoirs, sometimes fresh water is used for injection; in sandstone
reservoirs, only salt water is used to avoid clay swelling and the resulting
permeability reduction.

Water wells completed in deep aquifers far from surface outcrops are
oxygen-free. When these wells are completed and operated to assure
no air ingress, the water will remain oxygen-free. Some shallow wells
may contain oxygen. Oxygen-free sources of water are always
preferred. However, sometimes they are not available. On land, for
carbonate reservoirs, surface fresh waters are sometimes used out of
economic necessity. Offshore, seawater is usually the only abundant
water supply available. Regardless of the type of water available,
oxygen, if present, must be removed and the water must be monitored to
confirm that oxygen has not reentered.

The amount of oxygen water can dissolve at atmospheric conditions

depends primarily on temperature. Dissolved-oxygen capacity of fresh

23
water is maximum at 10 ppm at 32° F and 0 ppm at 212° F. Fresh water
can dissolve 8 ppm oxygen at 60° F, and this is the value usually used if
the actual dissolved oxygen content is not known. Dissolved salt
reduces the capacity of water to dissolve oxygen. Because of its
corrosivity, the oxygen must be removed. Inhibition of oxygen corrosion
is usually not a viable alternative.

Removal of Oxygen from Injection Waters

Oxygen can be mechanically removed or it can be chemically
scavenged. Local conditions will determine which system is the most
economical. For small water volumes, oxygen scavengers are the norm.
For large water volumes some type of mechanical removal followed by
polishing with a chemical scavenger is the norm. Although the various
techniques will theoretically achieve varying levels of oxygen, such as
500 ppb, 700 ppb, 50 ppb, etc., these varying levels have little practical
difference. When an oxygen scavenger is injected, a little over treatment
for safety is the norm, and whether 50 ppb (0.05 ppm) oxygen or 500
ppb (0.5 ppm) oxygen are being removed with chemical scavenger, the
amount of chemical scavenger used is about the same. Thus, the user
should select the more economic, robust, forgiving and reliable system.
The following are brief descriptions of current systems in use. No doubt,
other processes will be developed which will be more economic and
robust.

Vacuum Deaeration
Historically, either vacuum deaeration or gas stripping were the only
choices, other than chemical scavenging, to remove oxygen. Vacuum
deaeration was the early preferred choice because, theoretically, it could
reduce oxygen levels lower than gas strippers. Oil field vacuum
deaeration is no longer the standard for water deaeration for several
reasons. First, with specified levels of oxygen of less than 1 ppm but
greater than 20 ppb, polishing with a chemical scavenger to less than 20
ppb oxygen is still required. Secondly, vacuum leaks are hard to locate
and any vacuum leak both lets in air and reduces efficiency. In addition,
packed towers in an oxygen-free environment are an ideal place for SRB
to thrive (Bubble-cap towers can help alleviate this problem).

In the offshore environment, particularly in deep water, weight and space

considerations have driven development of newer technologies that
weigh less and require less space. This is of primary importance in
floating offshore facilities such as TLPs. To supply buoyancy for
processing equipment, including live loads, the cost is several dollars per
pound. Thus, it is often more economic to purchase lighter, more
expensive equipment rather than provide the buoyancy required for

24
spacious heavy facilities, such as vacuum deaeration towers and their
auxiliary equipment.

Gas Stripping
Theoretically, vacuum deaeration towers can reduce oxygen to lower
levels than gas-stripping towers, but gas-stripping towers are simpler to
operate and do not suffer vacuum leaks. The technology has advanced
but, from a practical standpoint, it will still take 1-2 cubic feet of gas per
barrel of water to remove oxygen to less than 1 ppm with seven
theoretical trays (15 actual trays). Again, theoretically, by increasing the
gas volume to four cubic feet per barrel, oxygen concentrations below 20
ppb can be obtained. The norm is to use two cubic feet of oxygen-free
gas per barrel and use a chemical scavenging agent to get the oxygen to
less than 20 ppb. Like vacuum deaeration towers, lighter, more compact
processes are being used on offshore floating structures in lieu of gas-
stripping towers.

Catalytic Oxygen Removal

This technology was initially developed by Norsk Process, Norway.
Catalytic oxygen removal is one of the new technologies. It is light and
compact and uses a catalyst to react dissolved hydrogen with the
dissolved oxygen, with the result being water.

The system employs a meter to measure the inlet volume of water. This
inlet meter sends a signal to a hydrogen generator, which electrolytically
generates sufficient hydrogen (plus 30% excess +/–) to react with all the
dissolved oxygen in the water being treated. The oxygen concurrently
generated from the electrolysis of water is vented.

The hydrogen is dissolved in the water and passed over a palladium

catalyst. The palladium catalyst causes the dissolved oxygen and
dissolved hydrogen to react to form water. Oxygen content in the
effluent as low as 5 ppb is possible, and any excess hydrogen is vented.

A disadvantage of this process is that the dissolved hydrogen will also

react with any dissolved chlorine to form HCl. A few ppm HCl are of no
consequence. However, either a different biocide must be added or
chlorine will have to be reintroduced downstream of the catalytic oxygen
removal unit. Commercial catalytic oxygen removal units have been
made and operated.

Other New Oxygen Removal Technologies

Another of the newer technologies is a two-stage co-current system
where nitrogen is used to strip the oxygen out of the water and the

25
nitrogen is then regenerated by passing it over a catalyst with methanol
vapor between 356°F and 446° F. The methanol reacts with the oxygen
in the nitrogen to result in nitrogen with 0 ppb oxygen. This nitrogen is
then reused. Although oxygen levels in the effluent water of 10 ppb are
possible, provisions for chemical scavenging are always wise.
Commercial units have been made and operated.

26
Introduction
Oil and gas wells can be defined by reservoir phases or by legal limits on
production. In this document, an oil reservoir is defined as having two
hydrocarbon phases in the reservoir, an oil phase and a gas phase,
although sometimes the gas phase is nearly nonexistent. Oil wells
produce from oil reservoirs. For simplicity of regulation, legal authorities
often define the difference between an oil well and a gas well as the
difference in gas/liquid hydrocarbon production ratio. For example, to be
called an oil well, a well can have a maximum gas volume per barrel of
liquid hydrocarbon. If this ratio is exceeded, the well is a gas well.

A gas reservoir has a single gas phase in the reservoir, usually a critical
phase fluid that separates into gas condensate and gas as the pressure
is reduced. This is called retrograde condensation.

Most oil and gas wells are initially pressured to the normal hydrostatic
gradient of 0.433 psi/ft. (psi per foot) of depth. (Geopressured wells,
usually gas, can be pressured up to about 1.0 psi/ft. of depth.) For a
5,000 ft. deep well, the expected reservoir pressure is 2,165 psi and a
column of 35° API oil results in a liquid-column pressure of 1,840 psi.
Thus, the oil well will initially flow (2,165 psi-1,840 psi = 325 psi); in
addition, gas breaks out of the oil in the reservoir and in the tubing as the
oil is depressurized; this gasification of the oil column in the tubulars
prolongs the flowing life of the well. The oil well will continue to flow until
the hydrostatic head in the tubing exceeds the flowing bottom hole
pressure. Eventually, the well will need artificial lift to deplete the
remaining economically recoverable reserves.

The timing of the need for artificial lift will depend on many factors, for
example, relative water production volumes. Since water is denser than
oil, water and oil in the tubing will impose a greater hydrostatic pressure
on the reservoir than an oil column.

Unless the reservoir pressure is artificially maintained through injection of

fluids, water, gas, steam, solvents, CO2, etc., usually reservoir pressure
declines with depletion of the oil reserves. The rate of reservoir pressure
decline depends on the drive mechanism in the reservoir. The force that

27
drives the fluids to the well-bore, in most reservoirs, is one of the
following or a combination of the following:

• Solution gas drive.

• Gas cap drive.
• Bottom (or edge) water drive

When the oil well no longer flows, or when natural flow needs to be
increased, artificial lift is used. The following are the principal artificial lift
techniques:

• Beam pumping or rod pumping.

• Gas lift.
• ESPs.
• Hydraulic pumping.
• Flowing IOR (Improved Oil Recovery)

This section of the document is about corrosion of oil wells, and in

particular how corrosion alleviation relates to the artificial lift method. As
was discussed earlier, corrosion alleviation is nearly always needed
before all the oil reserves are produced. The key question is not, “Will
corrosion alleviation be needed?” The key questions are always, “When
will a corrosion alleviation system be needed?” and “Which one should
be applied?”

There are six important questions that must be answered in order to

design a successful corrosion alleviation system:

• Will the recommended corrosion alleviation system control

corrosion if implemented as designed?
• Can the recommended corrosion alleviation system be
implemented as designed, or are there technical, legal,
environmental or practical considerations that prevent its
implementation?
• Will the system be implemented as designed?
• What are the consequences of partial or imperfect
implementation?
• Is the recommended corrosion alleviation system compatible
with all field and refining operations?
• Is the recommended corrosion alleviation system tolerant of
changes in producing conditions?

Oil wells that flow unassisted without reservoir pressure maintenance or

some type of IOR (Improved Oil Recovery) usually do not require

28
corrosion alleviation. Corrosion in conventional oil wells not producing
via IOR is usually not a problem until water production reaches 25%-40%
of the liquid volume. By the time a well is producing this much water, it is
usually on artificial lift.

Beam Pumping
Beam pumping units or rod pumping is one of the most common types of
artificial lift. Rod pumping is used on low volume wells, i.e. wells with a
low PI (Productivity Index). In simple terms, PI is the volume of
production per psi of drawdown and may be expressed as barrels per psi
for oil or MCF per psi for gas. The goal of oil and gas operating
companies is to maximize profits. This usually means implementing a
strategy to minimize failures. But it may not be economically attractive to
completely eliminate all failures. An acceptable failure rate is a failure
rate that generates the most profit from the lease or production unit.

Beam pumping systems have many components, but the three principal
subsurface components we will deal with are the tubing, the sucker rods,
usually called the rods, and the pump.

What is a failure? In this document, a failure is a catastrophic event that

requires removing a failed component from the well and repairing or
replacing it. Failures come in many forms. Some failures are related to
design deficiencies; some failures are the result of successful corrosion
alleviation systems (Pumps eventually wear out even in the absence of
corrosion.); and some failures are the result of corrosion.

Failure reduction is the result of the integration of proper design, good

operating procedures, and an effective corrosion alleviation system.

Included are a review of design parameters, good operating practices,

mechanical and corrosion failures and corrosion alleviation systems.

Design Bases
The well and the pumping equipment are irrevocably tied to each other.
Either they work well together or they don’t work very well at all. Any
good beam pumping unit design starts with a well PI. Using the well PI,
the pumping system design is then adapted to the well configuration, the
well depth, casing size and fluid volume to be pumped.

Important design parameters include the sucker rod diameter and

strength, tubing diameter and strength, and the pump size and type,
including stroke length and frequency. Beam pumping system design is
complicated and well beyond the scope of this document. The reader is

29
referred to the latest edition of ANSI/API 11L, “Design Calculations for
Sucker Rod Pumping Systems (Conventional Units),” or the latest edition
of API 11L3, “Sucker Rod Pumping System Design Book.” Note that
both of these industry documents are “Pumping Systems.” This
“System” concept cannot be over emphasized.

Operating Procedures
Operating procedures include both equipment handling and equipment
installation, particularly the sucker rods and associated operations. As
will be discussed later, nearly all failures are fatigue failures as a result of
stress concentrations. Poor sucker rod storage, handling, and running
procedures can cause excessive localized stress concentrations and
result in fatigue-related failures.

A very good initial design may turn into an unsatisfactory design as the
well conditions change over time. These well condition changes require
either a change in design or a change in operations. Sometimes the
economic solution is a pump-off controller and/or other techniques to
help adapt an old design to new well conditions. Dynamometer analysis
is probably essential to successful maximization of profits from the
operation of beam pumped wells. Dynamometer analyses, interpretation
and application are well outside the scope of this document.

Restriction of pump filling can be the cause of sucker rod failures. This
can be caused by many conditions such as gas lock where down hole
gas is not properly separated. The gas interferes with pump efficiency
and may gas lock the pump. Another condition that can affect pump in-
flow and efficiency is dirty, emulsified annular fluids trying to enter the
pump. Such dirty emulsions can be from an accumulation in the annulus
or can occur as a result of conditions of the produced fluids, but it is
usually more the former than the latter.

Rod Failures
Sucker rods usually fail by one of two mechanisms: tensile failures or
fatigue failures. Though rare, tensile failures occur when the load stress
on the rod exceeds the rod’s yield stress. Tensile failures are easy to
identify because the rod is “necked down” at the failure site.

Nearly all rod failures are fatigue failures, which occur when applied
stresses remain below the rod yield stress. Fatigue failures are caused
by stress concentrations and these stress concentrations can be caused
by many things:

30
1) Manufacturing defects can cause stress concentrations.
2) Stress concentrations can be caused by mechanical damage that
occurs during storing, handling, or running the rods. The rods can
be bent, damaged by the elevators, or damaged by improper make-
up or breakout procedures. Proper rod handling procedures are
detailed in the latest editions of ANSI/API RP 11BR “Care and
Handling of Sucker Rods.”
3) Pits caused by corrosion also cause stress concentrations.
Historically, rod failures are about evenly split between mechanically
caused stress concentration and corrosion-initiated stress
concentration. Likewise, rod failures are about evenly split between
connection and body failures. Box failures represent about 60% of
connection failures, while the balance are pin failures.

The time-to-failure has many components: the chemistry of the steel in

the rod, the size of the rod, the yield stress of the steel of the rod, the
toughness of the steel of the rod, the stress applied to the rod, the size
and shape of the stress concentration, etc. In general, the higher the
applied stresses are in relation to the yield stress of the rod steel, the
shorter the time to failure. As a rule of thumb, not a design rule, the
useful stress limit for steels to prevent fatigue failures is about half their
yield strength.

Rod wear from rubbing against the tubing can also cause failures, but
usually the rods are harder than the tubing and tubing failures happen
first.

Corrosion Failures
Corrosion historically accounts for about half of all rod failures and nearly
all of the tubing failures. There are many corrodents in oil wells: H2S,
CO2 and organic acids, MIC and sometimes O2. In addition to the
corrodents, velocity, salinity and abrasive solids also play a role in the life
of the rods and tubing. Scale can initiate crevice corrosion and can also
prevent a corrosion inhibitor from getting to the corroding surface.

Any corrosion alleviation program must be part of a comprehensive

chemical system which includes the following:
• Demulsifiers used to clean annular fluids to improve pump in-fill.
• Corrosion inhibitors used to alleviate corrosion.
• Scale inhibitors used to protect productivity and increase rod life.
• Biocides used to alleviate MIC.

All of these components must work together.

31
Tubing Failures
Tubing can and does fail by corrosion. But usually in corrosive
conditions, the rods will fail first, because the stress to which they are
subjected is so much greater. Also, any corrosion pit will cause a stress
concentration, which will result in a fatigue failure. Tubing failures are
usually the result of rod wear. Rod wear can be alleviated, but not
eliminated. Rod guides (centralizers) help, but there is no cure for rod
wear in a crooked well bore, and no well-bore is perfectly vertical; the
only difference in well-bores is that some have more deviation from
vertical than others and some well-bores have sharper dog legs than
other wells. This problem increases with well depth. The deeper a well
is drilled, the more difficult it is to drill a vertically straight hole. Rod wear
on both the tubing and rods is always an issue in rod pumped wells.
Corrosion inhibition of the rods will usually protect the tubing from
corrosion.

Pump Failures
Pump failures can have manifold reasons: normal wear, corrosion,
abrasion, galling, etc. However, pump problems are nearly always
solved by using the correct materials or by controlling sand production.
Seldom does corrosion inhibition solve pump failure problems.

Guides for the Estimation of Corrosion

Although field history is the best guide to anticipated corrosion and the
need for corrosion inhibition, this is not always available. Likewise, there
are no reliable corrosion prediction models for rod pumped wells. The
following are rules of thumb, based on experience that provides the
corrosion specialist a place to start in evaluating the need for corrosion
alleviation procedures before failures become an economic problem.

Water Cut and pH

Will Steel be pH pH* Corrosion
Water Cut
Water-Wet (0 – 7) (7 – 14) Likelihood
0 – 25% Unlikely Unlikely
25% – 45% Possible X Possible
25% – 40% Possible X Unlikely
>45% Probable X Probable
>45% Probable X Possible
*Scale deposits may be a problem.
Table 1

32
Field Measured pH on Fresh Samples
pH Estimated Corrosivity
>7.0 Corrosion unlikely, scale deposition possible
7 -6.5 Mild corrosion likely
Moderate corrosion likely and
6.5-6.0
Corrosion inhibition Probably needed
Severe corrosion likely and aggressive corrosion inhibition
<6.0
probably needed
Table 2

Thirty-Day Flow Line Corrosion Coupon Data

The measurement of pitting corrosion on coupons is always a significant
challenge, because in relation to the total metal surface of the production
system, the coupon area is relatively small. Also, since pitting is
relatively random, it is never known for sure if pitting on the coupon (or
lack of pitting) is representative of the entire system. This problem is
solved by using a coupon with a crevice; a crevice is a forced site for a
pit to initiate.

Coupon Corrosion Rate (mpy) Probable Corrosivity

0-2 Mild Corrosion
2-5 Moderate Corrosion
>5 Severe Corrosion
Table 3

Field Detection of Corrosion

As stated previously, corrosion inhibition is usually economically
advantageous in all oil and gas production sometime before the reserves
are depleted. The key is to know when to start corrosion inhibition and
which corrosion inhibition system is the most economic. In rod pumped
wells, corrosion usually becomes a problem when the tubulars and rods
become water-wet. Whether the tubulars are water-wet depends on
many variables, but principally depends on the water cut and the crude
oil properties.

Historically, the tubulars may become water-wet when the water cut is as
low as 25%; the tubulars are always water-wet by the time the water cut
has reached 45% or more. It is a safe assumption that if CO2 and/or H2S
are present, the water will be corrosive. Dependent upon water
chemistry, very low CO2 partial pressures can be corrosive.

33
The following chart (Figure 10) helps determine the corrosion rates of
wells. When many wells are involved, typical wells can be evaluated.
The selected wells should always include the wells with the highest water
cut. Whenever a well has more than one rod failure per year, it probably
would be more economical to initiate a corrosion inhibition program. Rod
failures in the first three months after installation are probably the result
of a manufacturing defect or a design deficiency.

WATER CUT
25% OR OVER

CORROSION POSSIBLE

pH LESS pH pH pH
THAN 6.0 6.0 TO 6.5 6.5 TO 7.0 OVER 7.0

CORROSION CORROSION CORROSION CORROSION

PROBABLE LIKELY POSSIBLE UNLIKELY
(1)

INSTALL CORROSION COUPONS

COUPON COUPON
DATA DATA

LESS THAN LESS THAN MORE THAN LESS THAN LESS THAN MORE THAN
5 MPY 5 MPY 5 MPY 5 MPY 5 MPY 5 MPY
NO PITS PITS (3) NO PITS PITS (3)

CORROSION ISOLATED CORROSION CORROSION ISOLATED CORROSION

UNLIKELY CORROSION OCCURRING UNLIKELY CORROSION OCCURRING
(2) (2)

CORROSION INHIBITION CORROSION INHIBITION PROGRAM

PROGRAM DESIRABLE PROBABLY DESIRABLE

1. Corrosion occasionally occurs above a pH of 7.0. Where field experience indicates

possibility of corrosion.
2. When equipment becomes water-wet, corrosion will occur. Maintain a planned
monitoring program.
3. Check systems for air entrainment. If air entrainment is found, eliminate and re-test.
Figure 10

34
Design of a Corrosion Inhibition System
The most important characteristic of a corrosion inhibitor is that it
prevents corrosion. Properly formulated corrosion inhibitors should not
cause any other problems, but regardless of ancillary problems,
corrosion inhibitors first and foremost, must control corrosion.

In addition, successful corrosion inhibitors have the following

characteristics:
• The corrosion inhibitor must be sufficiently dispersible in the
flush fluid, usually lease brine or fresh water, so that it is carried
down the annulus to the pump intake.
• The corrosion inhibitor must not increase friction in the pump or
sticking of the pump because of “gunking” or any other reason.
• The corrosion inhibitor must remain dispersible in the presence
of solids, such as iron sulfide or formation solids, which tend to
accumulate at the oil/water interface.
• The inhibitor must have surfactant/dispersant properties
sufficient to disperse and move any solids from the well-bore
without having an adverse effect on the pump.
• The corrosion inhibitor must form a uniform protective film on
equipment surfaces and enhance oil-wetting of equipment.
• The corrosion inhibitor should not cause emulsions, or have a
negative impact on oil-water separation processes. However, it
is paramount that the corrosion inhibitor alleviates corrosion,
even if emulsion or water treating problems require additional
attention.

In summary, the corrosion inhibitor must not only alleviate corrosion, its
sole purpose, but it must also be part of an integrated production system,
which includes the mechanical components, the chemical treating
components, and good operating procedures.

It is always better to initiate a corrosion inhibition system before

significant corrosion occurs, because it is always easier to protect a
smooth surface than a rough, corroded surface covered with corrosion
products. The metals must be relatively clean for the inhibitor to work.
The inhibitor must be able to contact the metal surface. For this reason,
when a well has been producing for some time, it may be beneficial to
clean the well before the initial inhibition treatment. Paraffin,
asphaltenes, scale, and loose corrosion products should be removed.
The cleanup treatment should be designed for the current well
conditions. Obviously, whenever clean up treatments are conducted, the
return fluid may cause emulsion problems at the surface.

35
Another consideration in the design of corrosion inhibition systems is
certainty versus inconvenience. For example, a corrosion inhibitor
pumped to bottom always gets to bottom, while an inhibitor injected at
the surface will probably eventually fall to the top of the standing liquid
column. It is more inconvenient to displace the corrosion inhibitor to
bottom (or circulate it to bottom), than it is to inject the corrosion inhibitor
at the surface and let it fall. But if the corrosion inhibitor is pumped to
bottom, the operator is sure the corrosion inhibitor gets to bottom.
Monitoring can be used to enhance this certainty, but monitoring in itself
is an inconvenience.

For rod pumped wells, the basic rules for successful inhibition are few
and simple, but their application can be complex.

The mechanical pumping design and operation must fit both the well
productivity and physical conditions:
• No rod stacking.
• No over stressed rods.
• No gas lock of the pump.
• No pump-fill problems because of annular sludge/emulsions.
• No insufficient produced-fluid volume to fill the pump.
• Etc.

At the start of an inhibition program, the sucker rods and tubing should
be thoroughly filmed with corrosion inhibitor before production is initiated.

An effective corrosion inhibitor must be selected and delivered at

effective concentrations. If field history is not available, initial corrosion
inhibitor selection and required concentration can be determined with
laboratory testing and then optimized based upon field experience.

Since rod pumped wells are each a liquid-producing system, continuous

inhibition is probably the best solution. The key is how to deliver
corrosion inhibitor to the well tubing and rods continuously in the most
economical way.

The development of corrosion inhibitors is a dynamic field. Corrosion

inhibitors are continuously being improved in manifold ways: improved
partitioning to the water phase, improved water-soluble inhibitors, and
improved effectiveness is a continuing trend. If field experience is not
available, a corrosion inhibitor will have to be selected based on
laboratory tests and its effectiveness confirmed in the field.

36
Between 10 ppm and 50 ppm of inhibitor based on the total fluid are
usually required: 10-25 ppm for mild corrosion, 25-40 ppm for moderate
corrosion, and 50 ppm or more for severe corrosion.

Preferably, the delivery of corrosion inhibitor is continuous, and this is

often achieved by storing the inhibitor in the annular fluid. However,
many beam pumped wells operate in the pumped-off condition and thus
there are no fluids stored in the annulus. This type of well is usually
batched treated with a concentrated film-forming inhibitor. Batch
frequency may vary from twice per month for low volume wells to twice
per week for higher volume wells. Although corrosion inhibitor chemistry
is a dynamic field, the current trend is to use water-dispersible corrosion
inhibitors.

Determination of the most economic time to initiate a corrosion

alleviation program is difficult. When the water cut reaches 25%, the
operator should be aware that corrosion inhibition might be profitable.
Figure 10 will help in this determination; corrosion coupons will also help
in this decision. In non-H2S systems, iron counts may also help. The
author does not believe the use of manganese instead of iron in sour
systems is reliable, because manganese sulfide is nearly as insoluble as
iron sulfide. Frequent rod failures mean the operator has waited too long
to initiate corrosion alleviation procedures.

Application of the Corrosion Inhibition System

When new rods or tubing (new or used) are run into the well, the newly
installed steel should be coated with corrosion inhibitor. An acceptable
treatment is to pump one gallon of corrosion inhibitor per 1,000 ft. of
tubing into the annulus as a one percent solution. This corrosion
inhibitor solution is circulated once or twice, and then the corrosion
inhibitor solution is left in the annulus. Oil is the preferred solvent, but
dependent upon the corrosion inhibitor, deoxygenated water may be
used.

If warranted, a well should be cleaned up before corrosion inhibition is

initiated to remove any wax, asphaltene, scale, corrosion products,
annular emulsions, etc. The cleanup has to be designed to fit the
conditions. For example, a hydrocarbon solvent may remove wax or
asphaltenes, but will not remove scale or corrosion products.

Once the well is cleaned up, it should be initially filmed the same as
when new rods or tubing are installed.

37
As stated above, the required corrosion inhibitor concentration varies
from 10 ppm (total fluid) for mild corrosion to 50 ppm or more for severe
corrosion.

Gallons of inhibitor required per week can be calculated as follows:

Gallons per week = BF/D (Barrels of Fluid per Day) x 7 days per week x
42 gallons per barrel x ppm corrosion inhibitor/1,000,000

For example, if a well produces 25 BO/D, 80 BW/D (a combined 105

BF/D) and 35 ppm of corrosion inhibitor are needed, the calculation is as
follows:

Gallons of inhibitor/week = 105 BF/D x 7 days x 42 x 35ppm/1,000,000

Gallons of inhibitor/week = 1.1 gallons

Continuous inhibition is always the goal. In rod pumped wells, because

of the standing fluids in the annulus and the small amount of corrosion
inhibitor needed (as illustrated in the above example only 1.1 gallons/
week or 0.6 quarts/day), a continuous flush is required. When a portion
of the produced fluids with corrosion inhibitor added is continuously
applied to the annulus, the rods and tubing are continuously inhibited.

Although, continuous addition of corrosion inhibitor is the preferred

system, batch treatment is the norm. Corrosion inhibitor solutions are
periodically pumped into the annulus where they mix with the standing
annular fluids and are slowly displaced into the producing stream.

Initial Treating Frequency

Treating frequency depends on many variables which are difficult to
quantify such as corrosivity and well-bore deviation from vertical. Field
experience may dictate deviation from the following recommendations.
But without any other data, historical experience indicates that the
following procedures with a good corrosion inhibitor will achieve
acceptable results.
Production Rate Treating Frequency
Up to 150 BF/D Every two weeks
150- 300 BF/D Weekly
300-800 BF/D Twice weekly
Table 4

38
Treating Procedure
Less than 1,000 feet of annular fluid:
• Pump a barrel of produced water down the annulus.
• Pump required inhibitor solution (1% +/- concentration) into the
annulus.
• Pump 0.5 to 1 barrel of produced water per 1,000 ft. of depth.
• Sometimes, corrosion inhibitor is dispersed into the entire flush
water volume.

More than 1,000 ft. of annular fluid:

• Pump required inhibitor solution (1%+/- concentration).
• Circulate the well sufficiently to completely displace the annular
fluids and return the inhibitor solution to the annulus.
• If the well cannot be circulated, flush the annulus with one barrel
of produced water per 1,000 ft. of depth to the pump.

Squeezing any corrosion inhibitor into the reservoir is seldom a good

idea, because the treatment can cause reduced production; the inhibitor
may be hard to remove from the reservoir rocks, and the return rate of
the corrosion inhibitor is usually not well controlled. Corrosion inhibitors
control corrosion by controlling the electrochemical current at the
cathode of the corrosion cell. The rocks of the reservoir will have a
strong tendency to adsorb the corrosion inhibitor and not release it. In
turn, this adsorbed corrosion inhibitor may reduce well productivity.

Gas Lift
When gas is available, gas lift is usually the most economic artificial lift
technique for high volume oil and water wells. Gas lift is most efficient
when the back pressure is the lowest. Thus, gas lift wells are always
directed to the low pressure separator. The design principles of gas lift
are simple. Although the wells are usually corrosive, gas lifted wells are
the most difficult to inhibit effectively for corrosion of any of the artificial
lift methods. Often times, corrosion-resistant materials rather than
corrosion inhibitors are used to alleviate corrosion.

As always when designing an artificial lift system, a PI is needed. The

gas lift system is designed to gasify the liquid column in the tubing, which
in turn lightens the hydrostatic column of liquid. The gasification of the
liquid in the tubing results both in the fluids flowing up the tubing and the
lowering of the bottom hole pressure. The reduced bottom hole pressure
enhances fluid entry into the wellbore.

39
In a typical design, tubing is run with a packer, which is set just above
the perforations. The tubing is equipped with multiple gas lift valves.
The upper valve(s) are used to unload the well and as the fluid column is
lightened, the gas injection automatically moves down the tubing to lower
placed valves. Because of high productivity of the well, gas is seldom, if
ever, injected on bottom. The exception to this rule is for very high
volume wells where the flow is reversed and gas is injected down the
tubing and both the liquid and gas flow up the annulus.

There can be many alternate designs, but the above designs are the
most common.

One of the principal advantages of gas lift is that there is no restriction to

flow in the tubing. The system easily handles small amounts of
produced sand. The gas lift valves can be replaced via wire line.

Gas lift production is usually corrosive because it is high volume, high

velocity, and usually water-wet because of a high water cut and a low oil
cut. The gas is usually recycled, and as a result, the CO2 content
gradually builds up because CO2 is released by bicarbonate
- =
decomposition (2HCO3 2CO2 + CO3 +H2O) of the produced water.
The increased CO2 concentration in the gas lift gas increases corrosivity.

Effective corrosion inhibition is exceedingly difficult. Because of the high

liquid velocities, effective inhibition must be continuous; effective batch
inhibition is not realistic. Injecting inhibitor with the gas is not usually
done for numerous reasons. First, the gas seldom goes to bottom, thus
injecting corrosion inhibitor with the gas would not inhibit the whole
tubing string. The gas lift gas is usually dehydrated to prevent both
hydrate formation and corrosion in the gas lift system.

Thus water-soluble corrosion inhibitors, unless dissolved in glycol, would

be stripped of their water and only the viscous active ingredients would
remain, which would foul the gas lift valves. For oil-soluble corrosion
inhibitors, the problem is basically the same but more complex. For low
temperature, low pressure systems, high boiling point solvents are
available that, when used in sufficient volumes, will remain liquid. But in
many cases, gas probably strips the solvent from the corrosion inhibitor
formulation and, again, only the viscous active ingredients remain.

Effective vapor phase corrosion inhibitors are not available.

There are only two realistic alternatives for corrosion alleviation of gas lift
gas wells: provide a path to inject corrosion inhibitor into the bottom of
the well or use corrosion-resistant materials.

40
Often, internally plastic-coated tubing is used as a compromise. Tubing
coated with non-metallic coatings has a finite life of a few years, usually
two-plus years, for at least two reasons. First, a perfect coating
application of the entire pipe is very hard to achieve. Second, water and
CO2 will eventually penetrate the non-metallic coating. However, by the
time the tubing coating has failed, the gas lift design usually needs
revision. Smart coating technology is in its infancy. In the future, this
technology may have applications alleviating tubing corrosion in gas lift
wells.

Austenitic CRA tubing has too low a yield strength and is too expensive
to be a profitable alternative. With the introduction of martensitic (410
type) stainless steel, the economics of using CRA for gas lifted wells may
be changing.

Electrical Submersible Pumps (ESPs)

Electrical submersible pumps usually are used on high volume wells
producing little or no gas. If wells are not high volume, another more
efficient pumping method would be used, such as beam pumping. If
there is a corrosion problem, the high liquid velocity requires continuous
corrosion inhibition.

An ESP is used when there is insufficient gas to recycle and artificially lift
production with gas lift. This also means there is little gas to handle
down hole. If there is essentially no gas, the well may be completed with
a packer. With low pressures and little or no gas, there will be low acid
gas partial pressures.

An ESP frequently is used when a well produces high liquid volumes,

and high liquid volumes usually mean high water cuts. The water should
be oxygen-free unless oxygen is introduced from the atmosphere via an
open annulus or the well is shallow and near a source outcrop. For the
most part, subterranean waters are oxygen-free.

Corrosivity will depend on the water’s chemistry, acid gas partial

pressure and organic acids, but probably the produced fluids are not too
corrosive.

If corrosion is a problem, based on either corrosion studies, field

measurements, or a history of corrosion problems, either corrosion
inhibition or corrosion-resistant materials will be required.

Again, if corrosion inhibition is to be effective, the corrosion inhibitor must

be delivered to the bottom of the well. The technique used to achieve this

41
goal will depend on the way the well is completed; if corrosion inhibition
is essential to the economic success of the project, the well will be
completed to achieve effective inhibition of the complete well.

An ESP causes high shear, and it is preferable that the corrosion

inhibitor not enhance emulsion problems. However, the universal rule
should be followed; the corrosion inhibitor must alleviate corrosion, even
if a difficult emulsion is the result. Emulsions can always be handled at
the surface and more cheaply than down hole corrosion problems can be
repaired.

Corrosion inhibitor concentrations of 10-50 ppm are usually sufficient

(based on water production), however, in unusual situations, a higher
concentration of corrosion inhibitor may be required.

If the corrosion inhibitor cannot be delivered to the bottom of the tubing

and corrosion is a concern, then corrosion-resistant materials are an
alternative.

Corrosion-resistant materials are probably the only practical, permanent

solution for one specific application of ESP, namely, seawater pumping.
ESPs are often used offshore to pump the seawater for seawater water
floods, for fire water systems, and for wash down water. Aerated
seawater is corrosive to both low alloy steel and many CRA, and oxygen
corrosion cannot be inhibited with traditional oil field corrosion inhibitors.

Non-metallic internal coatings may have an economic application for

ambient seawater pumping for two reasons. First, both the cost of
replacement and the consequences of failure are low. Second, since the
pressure is low and the temperature is low, the tendency of the water to
penetrate the non-metallic coating is low. In this situation, until the
seawater is deoxygenated, either the acceptance of severe oxygen
corrosion or the use of a CRA specifically alloyed for this service (a
permanent solution) or the use of an internal coating are the only real
alternatives.

Solid FRP (Fiberglass Reinforced Pipe) has been used with only limited
success, because it is flammable, mechanically fragile, and has low
fatigue resistance to vibrations, pump pulsations, etc.

A non-corrosion issue is often calcium carbonate scaling caused by the

higher temperature of the pump. ESP motors are cooled by the fluids
they pump. If this carbonate scale is allowed to continue to deposit on
the pump, it will insulate the pump electric motor and prevent the

42
pumped fluids from cooling the motor. The result is premature motor
burnout.

Hydraulically Pumped Oil Wells

Infrequently, hydraulic pumping of oil wells is the most appropriate
artificial lift technique. Hydraulic pumping is used only in special
occasions and corrosion is not usually a severe problem.

Sometimes oil wells are located in environments that are not suitable for
the pumping facilities to be located at the wellhead, such as seasonally
flooded land locations, and gas lift is not economically feasible.
Directional drilling sometimes solves this problem, and sometimes
hydraulic pumping is the lift method of choice.

For high volume wells, an ESP might also work in this environment.

The usual hydraulic-pumping method is to use dehydrated lease

produced oil as the hydraulic power fluid. This power-oil is commingled
down hole with the produced fluids and these produced fluids are
handled in the usual production process. The oil and water are separated
and the oil is dehydrated. A portion of the dehydrated oil is used as
power oil and balance of the oil is sold as lease production.

Corrosion in the power-oil system is minimal and caused by any residual

water in the dehydrated oil. Like residual water in oil pipelines, water-
related corrosion may be handled with a minimal amount of corrosion
inhibitor, which partitions to the water. Sufficient corrosion inhibitor
should be added to give about 100 ppm of inhibitor in the water phase. If
corrosion in the oil well is a problem, the corrosion problem may be
handled just like it is handled in rod pumped wells. Corrosion inhibitor
should be added to the power oil to provide 25-50 ppm of corrosion
inhibitor in the total produced liquid. The lower concentration of
corrosion inhibitor is sufficient for mild corrosion and the higher
concentration is needed for more severe corrosion.

Improved Oil Recovery (IOR)

In 1960, conventional domestic technology left an average of 70% of the
original oil in place at abandonment. Reservoirs with strong water drives
or ones that have been water flooded still may have 20% or more of the
original oil in place at the end of the project.

Many supplemental recovery methods were evaluated/tested under the

heading acronym EOR, meaning (Enhanced Oil Recovery). IOR is
slowly replacing the EOR as the description of all supplemental recovery

43
techniques. These supplemental recovery techniques include varieties
of thermal floods (fire, steam, and electrical induction), polymer floods,
solvent floods, detergent floods, CO2 miscible floods and no doubt
technology will continue to develop new methods to recover existing oil
in place. But, the current mainstays are water floods, CO2 miscible
floods, and steam floods.

Water Floods
Water floods may present two types of corrosion problems: the injected
water and the produced water. The main corrosion problem with injected
water is usually oxygen contamination. (Alleviation of oxygen corrosion
is discussed in Chapter 1 of this handbook, in a section entitled “Oxygen
Corrosion in Oil and Gas Production.”) Oil wells producing via water
floods are treated for corrosion just like any other oil well that produces
water. The corrosion alleviation method, as reviewed above, depends
upon fluid corrosivity and the method of production, natural flow or
artificial lift. Usually, production wells producing via water floods need
artificial lift because the water cut is usually high.

CO2 Miscible Floods

Economically valuable oil remains in the reservoir even after the
reservoir has been water flooded. Carbon dioxide is sometimes injected
into the reservoir to recover some of this remaining oil. The CO2
combines with the remaining oil to form a miscible gas phase. This
residual oil travels as a single phase to the producing well-bore.
Sometimes the production wells flow naturally and sometimes they have
to be artificially lifted. They are nearly always corrosive.

Oil reservoirs usually have residual oil saturations greater than 20% after
either water flooding or being flooded by a natural water drive. CO2
injection offers a technique to recover significant portions of this residual
oil. Often, comprehensive phase-behavior tests are conducted before a
CO2 injection project is initiated. A comprehensive discussion of phase
behavior of CO2 in oil reservoirs is beyond the scope of this document.
In short, under the right conditions, the CO2 and the residual oil combine
to form a gas phase, which in turn releases the immobile oil. This gas
phase, a combination of reservoir oil and gas and injected CO2, is free to
flow toward the well-bore. Dependent on the required conditions for
miscibility, sometimes the reservoir is sufficiently pressurized with CO2
for the wells to flow and sometimes the wells are produced with artificial
lift. As temperature and pressure are reduced at surface conditions, the
oil and CO2 revert back to two phases—liquid oil and gas.

44
Regardless of whether wells using CO2 miscible flooding are artificially
lifted or they flow because of pressure maintenance, they are always
corrosive. The gas phase is predominantly CO2. (Too much methane
makes miscibility between the oil and the CO2 more difficult.) In flowing
wells, the CO2 partial pressure may be several hundred psia and the
waters may be very saline. Iron carbonate is sometimes protective at
higher temperatures with relatively fresh water. With saline water, iron
carbonate is more soluble and less protective. In the cooler regions of
the wells and production systems, corrosion is aggressive.

Continuous corrosion inhibition is always required unless corrosion-

resistant materials are used. Historically, oil-soluble corrosion inhibitors
were the key to successful corrosion inhibition. But corrosion inhibitor
technology is dynamic and currently, water-soluble corrosion inhibitors
sometimes give superior economic performance. Although lower
corrosion inhibitor concentrations have been reported to yield acceptable
performance, usually between 200 ppm to 300 ppm of corrosion inhibitor
are required based on total produced liquids.

In carbonaceous reservoirs, the corrosion may be mitigated by the

neutralization of some of the CO2 by the CaCO3 in the reservoir. As a
result, the produced water may contain high levels of bicarbonate.
These wells may not necessarily be more corrosive nor treated
differently than non-CO2 flooded wells.

Remember that there are major differences between limestone/dolomite

reservoirs and siliceous reservoirs. Siliceous reservoirs, when CO2,
flooded, have a history of severe corrosion problems.

Steam Floods
When oils are too viscous to be produced at economic rates under
reservoir conditions, steam is sometimes injected to make the oil more
mobile. The two initially dominant techniques have been steam soaks
and steam drives. There are many variations of these two techniques.
Corrosion is usually not a severe problem. Steam floods are limited to
relatively shallow reservoirs. The heat delivered to the reservoir comes
principally from heat of condensation of the steam. The amount of heat
of condensation per pound of steam declines as the pressure increases
and the heat of condensation goes to zero at 3,206 psia where the heat
content of the liquid and vapor are the same. Thus, steam flooding is
limited to reservoirs with pressures below 3,200 psia. As a practical
matter, for the heat value of steam to be significantly superior to hot
water, reservoir pressures are usually much below 3,200 psia.

45
There are large accumulations of hydrocarbons which are too viscous
under reservoir conditions to be economically produced using normal
production techniques.

Sometimes mining and surface retorting is the most economical recovery

technique and sometimes in-situ stimulation is the economically
preferred technique. In these cases steam is used to lower the viscosity
of the in-situ oil.

Oil reservoirs containing oil with gravity as high as 19° API have been
economically steam flooded. But usually, steamed reservoirs contain oil
with much lower API gravities than 19° API.

Steam soaks are not used much any more, as this method is mainly an
acceleration technique. The purpose of this method is merely to reduce
the oil viscosity; the reservoir must still provide the drive mechanism to
move the oil to the well-bore. In simple terms, steam is injected; the well
is shut in and then returned to production. This cycle can be repeated
until it is no longer economic.

Steam drives are much more common because they both supply the
heat to reduce the oil viscosity and the energy to push the fluids to the
well-bore. The goal of steam floods is to reduce the mobility ratio
between the oil and the water. They are operated much like a water
flood with injection wells and producing wells. As the reservoir is heated
by the condensing steam, the ratio of oil-viscosity to water-viscosity
decreases, making oil production easier in relation to water production.
There are many variations of steam drives.

Corrosion is not usually a serious problem on either the injection side or

the production side.

On the injection side, proper feed water treatment for the steam
generators alleviates both scale problems and oxygen-corrosion
problems.

Corrosion is not usually a serious problem on the production side either,

because reservoirs suitable for steam flooding usually are shallow and,
thus, not very saline and the condensed steam further reduces salinity.
Production is normally hot and any bicarbonate in the water is
decomposed, which results in high pH of the water (and CO2 in the
steam). In hot temperatures, fresh water, and low velocities of oil
production, the iron carbonate film is protective. Oil-wetting of the steel
is more likely because, as reported by de Waard, the amount of water
held in a water-in-oil emulsion is inversely related to the oil API gravity.

46
Other Thermal Recovery Methods
One of the largest unexploited hydrocarbon reserves in the United States
is contained in shale oils found in the Rocky Mountain region. Newer
technology will be required to recover economically these reserves in an
environmentally acceptable way. No doubt these reserves will be
produced and no doubt new corrosion problems will be associated with
their production, regardless of the heat source.

47
Introduction
Gas well corrosion was first recognized in the 1940s in wells producing
sweet (no H2S) gas and gas condensate. The corrosion was the result of
water of condensation and CO2. The production of sour gas (natural gas
containing H2S) followed shortly thereafter. The key question to answer
to achieve a successful corrosion alleviation systems is not if you will
need corrosion control measures, but when you will need corrosion
alleviation and which corrosion alleviation system to use. Phase
behavior will play a large role in the determination of which corrosion
alleviation system to use.

Calculating Partial Pressures

In the evaluation of gas production systems and the potential need for
corrosion control measures, a few fundamental tools are needed. One of
these tools is calculation of the CO2 partial pressure of the corrosive gas.

Example:

Gas Well Conditions:

Bottom hole pressure 2,000 psi (gauge)
CO2 content 3%

CO2 partial pressure = System absolute pressure x mole fraction of CO2

Absolute pressure (psia) = gauge pressure + atmospheric pressure
(14.7 psia~15psia)

Mole fraction CO2 = %CO2/100

CO2 partial pressure = (2,000 gauge + 15 atmospheric pressure)
x 3/100
= 2015 x 0.03
CO2 partial pressure = 60.45 psia

When pressures are very high, the difference between absolute pressure
and gauge pressure is negligible. If 2,000 psi had been used in the
above example instead of 2,015 psia, the CO2 partial pressure would
have been 60 psia instead of 60.45 psia. This is a negligible difference.

48
However, if the system pressure is low, adding atmospheric pressure to
the measured gauge pressure is important.

Predicting CO2 Corrosivity

Incidentally, this well will be corrosive. The prediction of the corrosivity of
a well can be the most challenging task. Corrosion modeling is a very
dynamic field and any comments or evaluations of the current models
would be out-of-date before the ink of this document is dry.

The only way to know for sure of the condition of well tubing is to run a
caliper or to pull the tubing to inspect it. All other techniques are
estimates or predictions. The most corrosive regions (regions hardest to
inhibit) will be areas of high turbulence, including the well head.

Field history is a good technique to predict both corrosivity and the

effectiveness of a corrosion inhibition system. For a new field with no
production history and with no analog production history, gas analyses
and modeling with unknown gas and water production rates is probably
all the information available. Good water analyses are seldom, if ever,
available. First, water analyses are not important to evaluate the
economics of the project. Reservoir engineers are trained to evaluate
valuable commodities like natural gas and oil, not waste products such
as water. For obvious reasons, gas wells are not completed into the
water, so no connate water sample is available to sample. Secondly,
water production forecasts are historically low.

Thus, for fields with no production history, predicting corrosivity is a

daunting task. As discussed earlier, all one has to go on is the CO2
partial pressure.

For existing wells, the task of predicting corrosivity is easier. The first
information to ascertain is whether there have been any corrosion-
related failures. With no failures, the upper limit of corrosivity can be
established. If there have been corrosion failures, then corrosivity has
already been established. Iron content of the produced water is an
indication of corrosivity in a gas well. Based on the assumption that all
the iron is coming from the tubing, average corrosion rates can be
estimated as follows:

49
 Average MPY Corrosion = (BWPD x ppm iron)/ (0.082 x D x L)
MPY Mils per year
BWPD Barrels of water per day
D Tubing inside diameter in inches
L Tubing length in feet
0.082 Constant which reconciles the various units
Table 5

Iron counts are better for monitoring corrosion than predicting it. If only
water of condensation is produced, then probably all the iron dissolved in
the water came from the well tubulars. If formation water is produced,
then an unknown quantity of this iron came from the formation and not
the tubing. Carbon dioxide corrosion is pitting corrosion and a pit will
penetrate the tubing wall long before the average corrosion rate indicates
a problem. The general corrosivity rules outlined in Chapter 1 are the
best guide. Organic acids will increase corrosion rates, as will high gas
velocities in the tubing. Unlike oil, gas condensate, except in large
quantities, does not reduce CO2 corrosion.

Designing a Corrosion Inhibition System

When the first question, “Is a corrosion inhibition system needed?” has
been answered affirmatively, the next question is “Which corrosion
inhibition system to use?” In the design of the corrosion inhibition
system, it is important to remember that corrosion inhibitors are not
smart. They do not automatically go to the correct location; they must be
properly applied.

For the most part, the choices of inhibition method are batch or
continuous. Continuous inhibition is preferred for numerous reasons, but
phase behavior is critical. Batch inhibition is predominantly for lower
temperatures and lower velocities. These two inhibition techniques will
be reviewed respectively.

Continuous Corrosion Inhibition

Continuous corrosion inhibition is the most reliable way to control
corrosion. The most severe CO2 corrosion to be controlled requires
continuous inhibition. Continuous corrosion inhibition does not interrupt
production. The corrosion inhibitor can be applied down a capillary tube,
down the annulus 1) through a valve, 2) through a hole in the tubing, or
3) with an open annulus. In all cases, the key to success is appropriate
phase behavior. The corrosion inhibitor must stay in the liquid phase to
be effective.

50
At the relatively low temperatures and low pressures of the past (200°F
+/- 2,500 +/- psi Flowing Bottom hole Pressure) or less, phase behavior
was usually not a factor. At the above conditions, the corrosion inhibitor
was usually in the liquid phase. But as temperatures increase and
flowing bottom hole pressures increase, a liquid inhibitor phase is not
assured and the importance of phase behavior becomes paramount.
Everyone is familiar with retrograde condensation where, as gas
pressure is reduced, liquid hydrocarbons condense. The reverse is also
true; if gas and liquid condensate are re-pressured and reheated to
above the dew point, the liquids and gas will revert to a single phase.

Gases that are under-saturated with liquids may also vaporize oil-based
corrosion inhibitors. The smaller molecular weight solvents vaporize first
and then, if conditions are correct, the larger molecular weight active
ingredients also vaporize. The higher the CO2 partial pressure the more
oil-based liquids that can be vaporized per given volume of gas.
Hydrogen sulfide has even higher capacities to vaporize liquid
hydrocarbons. High temperature, in combination with high H2S partial
pressure, can vaporize large volumes of liquid hydrocarbons per given
volume of gas. These liquid hydrocarbons will condense up hole as the
pressure and the temperature are reduced.

A rigorous discussion of phase behavior is well beyond the scope of this

document, but it must be remembered that corrosion inhibitors do not
inhibit as intended if they are in the vapor phase. Whenever small
volumes of corrosion inhibitor are pumped into a hot, high pressure gas
stream, the user must be certain of the phase behavior if the corrosion
inhibition system is to be successful. Obviously, if a liquid hydrocarbon
phase is present throughout the entire tubing string, then phase behavior
is not an issue for oil-soluble corrosion inhibitors.

Water phase behavior is much easier to predict. Often, water is present

throughout the tubing string. Sometimes it condenses up hole. Because
water phase behavior is so much easier to predict and because water-
soluble corrosion inhibitors are now available for high temperatures
(350° F and climbing), water-soluble corrosion inhibitors are often the
best choice for continuous inhibition of hot, high pressure gas wells.

The corrosion inhibitor and initial inhibitor concentration are usually

selected based on previous experience (same field or analog field) or
laboratory tests. Required inhibitor concentration is usually less than
250 ppm based on total liquid production. The actual concentration will
be based on field experience.

51
Batch Inhibition
Batch inhibition is for low velocity, low temperature wells. Above 200° F,
10 ft/sec is a practical upper velocity limit; below 200° F, 20 ft/sec is a
convenient upper velocity limit. Obviously, there is smooth transition
between these rules of thumb; this is not a step function. Carbon dioxide
partial pressure is important because CO2 corrosion is velocity affected.
Hydrogen sulfide corrosion is velocity independent, and the success of
batch inhibition is primarily dependent upon water salinity, liquid
hydrocarbon content, H2S partial pressure, and temperature. Unless
contrary information is available, batch inhibition of sour gas wells should
not be attempted unless condensate production is at least 100 barrels
per MMCF (Million Standard Cubic Feet) of gas production and the oil-
water ratio is at least 10.

The four keys to a successful batch inhibition program are as follows:

1) An effective corrosion inhibitor.
2) Sufficient volume of corrosion inhibitor per treatment.
3) Correct treatment technique.
4) Correct treatment frequency.

An Effective Corrosion Inhibitor

Selection of the correct corrosion inhibitor can be done in many ways.
When no other data are available, laboratory tests must be performed. If
successful corrosion inhibition has previously been experienced in the
same field or in a similar application, then the successful corrosion
inhibitor should initially be used.

Sufficient Corrosion Inhibitor Volume

The volume of the corrosion inhibitor can be calculated based on 100
percent utilization. In the real world, 50 percent utilization would be more
likely. Previous studies by Superior Oil Company have indicated that a
4-mil inhibitor coating is the maximum to be expected. Based on 50%
utilization of the inhibitor, the gallons of inhibitor to lay down a 4-mil
coating can be calculated as follows:
M x D x L
Gallons of Inhibitor = x2
60 x A
M = Corrosion inhibitor film thickness in mils (4)
D = The tubing internal diameter in inches
L = Tubing length in feet
A = The % activity of the corrosion inhibitor
60 = Constant which reconciles various units

52
For example, for a 10,000 ft well with 2” ID tubing and 50% active
corrosion inhibitor, the volume of corrosion inhibitor needed is as follows:
4 x 2 x 10,000
Gallons of Inhibitor = x2 = 53 gallons of inhibitor
60 x 50

Batch Inhibition Techniques

As stated above, corrosion inhibitor is not smart; it will not automatically
go to where it is needed. It is up to the user to apply the inhibitor for
maximum effectiveness. There are many ways to batch inhibit gas wells,
but the two most common are: 1) Tubing displacement and 2) Batch and
fall.

1) Tubing displacement: Tubing displacement is the preferred batch

treatment technique. Oil-soluble corrosion inhibitors usually have the
best film persistency and are most often used. A solution of 10%-50%
+/- concentration of the corrosion inhibitor is mixed with a solvent (diesel
or lease condensate are common). The inhibitor solution is displaced to
the bottom of the tubing with either diesel or lease condensate. Care is
taken not to over displace the inhibitor into the formation. Water cannot
be used as a displacement liquid because it is heavier than the inhibitor
solution; it would fall through the inhibitor solution and the inhibitor would
not be displaced to bottom.

The reasons tubing displacement is the best technique are manifold.

First, there is no doubt the corrosion inhibitor reaches the bottom of the
tubing because it is pumped there. Secondly, the tubing is totally
exposed to the corrosion inhibitor twice, when it is pumped in and when it
flows back. Finally, production interruptions are minimal. There is no
reason to let the well set after the corrosion inhibitor has been displaced
to bottom. The well can be immediately returned to production.

There are only a few principal disadvantages. First, the column of liquid
may kill the well. If this is the case, the inhibitor may be displaced with
nitrogen, which is probably more costly. Secondly, if lease condensate is
not available, then the purchased solvent (diesel) cannot be sold for the
purchase price. Finally, there must be equipment to handle and pump
larger liquid volumes.

2) Batch and Fall: This is the most common batch technique. The
inhibitor is diluted with at least an equal amount of diesel or condensate
to reduce its viscosity. The reduced viscosity increases fall rate and
enhances the filming of the corrosion inhibitor on the tubing. The well is
shut in and the solution is pumped into the well head and allowed to fall

53
to bottom. Since the tubing is not empty, the gas in the tubing has to
pass through the inhibitor (or vice versa) and, thus, coating of the tubing
with corrosion inhibitor is not uniform. Superior Oil published a study
using radioactive tracers that indicated fall rates of 1,500 ft/hr (feet per
hour). A more conservative approach would be to design based on
1,000 ft/hr -1,200 ft/hr. The well remains shut in until the corrosion
inhibitor has time to reach the bottom of the tubing. One hour should be
added to the shut-in time for each down hole restriction.

The fall rate of a batch inhibitor, however, can be complex. Gas density
(pressure and composition), liquid density, and tubing diameter can all
affect fall rates. Sometimes, it is economically warranted to use wire line
measurements to attempt to monitor the liquid fall; if the fall rates are
measured to be faster than predicted, then shut-in times can be reduced.

The principal advantages of the batch and fall are three-fold. First, batch
and fall will not kill the well and it can be easily returned to production.
Second, unless it is a high pressure well, a pump truck (or high volume,
high pressure pump) is not needed to inject the corrosion inhibitor 3),
large volumes of displacement fluids do not have to be handled—
purchased and/or, transported, and pumped.

There are several disadvantages to batch and fall. First, the well is shut
in for several hours, depending upon depth, and this production is lost.
The user is never sure how far the corrosion inhibitor has fallen and,
thus, whether the whole tubing string has been inhibited. The corrosion
inhibitor will not fall through any standing liquid on bottom, so the bottom
of the tubing may never be inhibited. Because the corrosion inhibitor and
the gas in the tubing have to move past each other, the coating of the
tubing with corrosion inhibitor is not uniform. The denser the gas phase
(higher pressure and/or higher CO2 content), the slower the inhibitor fall
rate.

In spite of these disadvantages, in many cases batch and fall is the only
practical way to inhibit some wells and it is often the least expensive.
Thus, it is the most common batch inhibition technique.

Batch Frequency
Batch inhibition frequency is determined through a mixture of both
science and experience. The procedure outlined below is a place to start
when no applicable experience is available. The batch frequency can
then be adjusted based on monitoring.

54
In cases of relatively low bottom hole temperatures (<200°F+/-) and low
velocities, batch inhibition frequency may be required every three
months. Dependent upon producing conditions, the required batch
frequency can be anything in between from every few days to every
three months.

The higher the fluid velocity, the more frequent the batch inhibition.
Likewise, the higher the producing temperatures, the more frequent the
batch inhibition.

“RP14E - API Recommended Practice for Design and Installation of

Offshore Production Platform Piping Systems” has an equation to
calculate erosional velocities for offshore production piping. This is not a
real erosional velocity since the erosional velocity for solids-free liquids is
more than 300 ft/sec. The apparent erosional velocity of API 14E is a
conservative design velocity for offshore mixtures of oil, water, gas and
some suspended particles.

The following equation from API 14E is not a perfect relationship. Past
studies have revealed that few major oil companies rigorously follow it.
However, it is a convenient reference point on which to base experience
and develop a procedure to estimate initial batch treatment frequencies
when no better information is available. Once the batch inhibition
program is initiated, the batch frequency can be adjusted based on
monitoring results.

The erosional velocity from API 14E is as follows:

C
Vc =
(PM)1/2

Vc = Fluid erosional velocity in ft/sec.

C = 100 which an empirical constant. This constant may be
reduced when more than nominal amounts of suspended
solids are present.
3
PM = Gas/liquid density at pipeline conditions in lb/ft (pounds per
cubic foot).

55
Initial batch frequencies can be determined as follows and revised as
warranted based on monitoring:

Actual velocity/ API 14E Erosional Velocity Treating Frequency

< 0.25 3 to 4 months
0.25 to 0.5 2 to 3 months
0.5 to 0.9 1 to 2 months
> 0.9 < 1 month
Table 6

An inspection of the API equation reveals that as pressure goes up (gas

density increases) the acceptable velocity goes down, as it should.
However, as temperature goes up and the gas density goes down, the
acceptable velocity goes up according to the API formula. For batch
inhibition, the period between batches should decrease, not increase, as
temperature goes up. This is another reason why the above equation is
used as an initial approximation only.

Other non-quantifiable relationships also affect batch inhibition

frequencies. Solids, sand or other suspended solids, in the production
fluids will shorten the period between batch treatments. If the condition
of the tubing is poor or if the fluids are extremely corrosive, time between
treatments will also be shorter.

In severe cases where continuous inhibition has been temporarily

suspended because of equipment failure, batch inhibition may be done
every other day (batch and fall for a day and produce for a day).

Finally, the above techniques are initial approximations only and should
be adjusted based on monitoring after the batch treatments have been
initiated.

56
Introduction
As mentioned earlier, weldments are usually harder to inhibit than un-
welded steel. For this reason, pipe with longitudinal welds such as ERW
are usually laid with the weld up 30 degrees off center. The welds are
alternated, with the welds of every other joint being off set 30 degrees in
the opposite direction. First, this avoids having the weld, which is more
difficult to inhibit for corrosion, on bottom where water may accumulate.
Second, in case of some inherent weakness at the weld seam, this
avoids the pipeline from coming “unzipped” down the weld seam. This
practice is usually followed whenever non-seamless pipe is laid, except
when spiral-welded pipe in used.

In regard to corrosion alleviation, one of the most important concerns is

corrosion inhibition economics. This concern dominates all discussions
of gas transmission pipelines, oil transmission pipelines, pipeline pigging,
and pipeline hydrotest water. Following is a respective discussion.

Economics of Corrosion
Safety has always been, and still is, the major concern for maintaining
the integrity of pipelines. However, the historic classic economics of
inhibition costs being weighed against the cost of deferred production
and cost of repair are no longer relevant. The bases of costs have
changed.

Safety
Safety, particularly personal safety of the public or employees, has
always been a driver for maintaining pipeline integrity. But the cost of
injury to either an employee or the public has been magnified several-
fold over the past years by society. If a corrosion failure injures
someone, the cost of compensation will far outweigh the cost to repair
the failure. Failure prevention is more economic than the consequences
of a failure.

57
Environment
The same is true for environmental damage caused by a corrosion
failure. Pipeline failures may result in fines and/or restrictions of future
operations.

The consequences of corrosion failures are far more than the cost of
repairs and deferred production; they include costs of injury to the public
or employees and damage to the environment. These latter costs
usually dwarf the cost of repairs following a corrosion failure. Again,
failure prevention is more economic than the consequences of a failure.

Gas Transmission Pipelines

In this discussion, gas transmission pipelines are pipelines carrying
dehydrated gas from producers to customers, either in the interstate or
intrastate market. By definition, gas transmission lines are intended to
be single phase pipelines. However, in the real world of oil and gas
production, processing and sale, no pipeline is ever always single phase.

Gas transmission pipelines are dehydrated to prevent the formation of

gas hydrates. As a side consequence, with no water, there should be no
corrosion either. Pipeline specifications also limit the amount of CO2 and
H2S that the gas can contain and be carried in a common carrier
pipeline. Without corrosive gases, corrosion, again, is limited.

In a perfect world, with no liquid water and very low acid gas
composition, corrosion should not be a problem. But in the oil field, a
perfect world never exists. First, dehydration systems fail and some
liquid water may be present via condensation. Glycol can both carry
over from the dehydration system and, in some cases, condense from
the gas. Glycol solutions can be corrosive in the presence of H2S or
CO2. Condensing aqueous liquids can also cause TOL (Top of Line)
corrosion. To add to the mix, some condensate inevitably condenses
and potentially accumulates. Finally, MIC can be a problem in gas
transmission pipelines.

Any successful corrosion inhibition system must address all of the above
considerations, and flow regime knowledge is an essential element in
developing a robust corrosion inhibition system. If the pipeline flow
regime is turbulent or the more common annular mist, there will be no
dead spots where liquid can accumulate and continuous inhibition will be
the best choice. Continuous corrosion inhibition with 1-2 pints per
MMCF of gas is the norm. A biocide program is added when evidence of
bacteria exist. (Monitoring and treatment for MIC is outlined in Chapter
I.)

58
The most reliable way to determine flow regime is with one of many
available computer models. If no modeling data is available, the rule of
thumb is that, above 15 ft/sec, flow is annular mist or turbulent.

If the pipeline is in laminar flow (stratified), then corrosion inhibition is

much more complicated. First, all liquids will accumulate in the bottom of
the pipeline, but these will be treated with the above recommended
continuous inhibition. These treatments may need to be adjusted if,
based on total liquid present, the corrosion inhibitor concentration is not
50-100 ppm.

By definition, no liquid water can enter a dehydrated gas line; just glycol
carryover from the glycol dehydration system. Condensation of water
from insufficient dehydration or glycol condensation should be the only
liquids in the pipeline. A possible exception to this assumption might be
a gross upset in a dry desiccant bed dehydration system where free
water not only enters the dry bed through a process upset, but passes
through. This is a very unlikely scenario. However, TOL corrosion is
possible from either water or glycol condensation, and both are corrosive
in the presence of H2S or CO2. Likewise, for any type stratified flow, TOL
will not be treated with continuous injection of a liquid corrosion inhibitor,
which will follow the liquids to the bottom of the line.

The best way to treat a gas pipeline for TOL corrosion is to batch treat
with a corrosion inhibitor between pigs. The volume of the batch can be
calculated the same way as a batch volume is calculated to batch treat a
gas well. Batch frequency should be based on experience, but a batch
treatment every 2-4 weeks is the norm.

Oil Transmission Pipelines

Oil transmission pipelines are common carrier pipelines that carry
dehydrated liquid hydrocarbons from the producers to market. Although
the market is ultimately a refinery, there may be many terminals in
between producer and ultimate user. By contract, the crude oil and
condensate must be dehydrated before the hydrocarbons can enter the
pipeline. Contract maximum water content, dependent upon local
custom and contract, will vary from 1% to 4%.

Regardless of the acceptable residual water content, widespread general

corrosion is usually not a problem. The problem comes in the bottom of
the line in low places where water will settle out in low-flow and no-flow
situations. Pitting corrosion will be spotty in these water accumulations,
but the consequences of failure, as mentioned above, may be immense.

59
A rule of thumb is that 3 ft/sec is the minimum velocity required to
prevent water from settling out of the oil and accumulating in the bottom
of an oil pipeline. However, the prediction of water dropout is difficult
because it depends on many more things than velocity. Corrosion
inhibition of a small water component is preferable to the assumption that
water will never settle out and cause corrosion, or that oil with excess
water will never enter the pipeline.

As in natural gas pipelines, the first and foremost concern raised by a

corrosion-related leak in an oil pipeline is the assurance of the safety of
the general public and employees. In addition, the cost of repair will be
dwarfed by the environmental costs, including cleanup and fines.

Inhibition, if warranted, should be with a water-soluble corrosion inhibitor

in sufficient quantities to yield 250 ppm of corrosion inhibitor in the water
phase.

As discussed later in this section, pigging is an important component of a

successful corrosion alleviation program, both to prevent under deposit
corrosion and to remove the accumulated water.

Multiphase Pipelines and Flow Lines

As can be surmised from the previous two sections, all pipeline
throughput is really multiphase. “Dry” gas transmission pipelines usually
contain some liquid and “oil” transmission pipelines some water. This
section is about the more traditional flow lines/pipelines which contain
unseparated oil, produced water, and natural gas.

These multiphase pipelines are the most difficult to inhibit for corrosion.
The gas, oil, and water can be in any ratio; the gas may contain H2S
and/or CO2 of varying concentrations. The water salinity may be
anywhere from fresh to saturated with salt. The water composition may
be one that deposits scales, such as calcium carbonate, calcium sulfate
and barium sulfate. The multiphase stream also can contain produced
solids such as sand and drilling fluid solids. Organic acids in the water
may also be an important consideration in corrosion alleviation
programs.

A known flow regime is an essential ingredient to success. The flow

regime knowledge is a tool both to predict uninhibited corrosion rates
and to develop a successful corrosion alleviation system.

60
Corrosion Prediction
Hydrogen sulfide and CO2 corrosion have been discussed earlier in this
document, but a few things bear repeating. Carbon dioxide corrosion is
highly velocity dependent with wall shear stress being the operative
variable. Liquids will flow slower than gas in multiphase flow systems.
Weldments will be harder to inhibit than unwelded pipe steel. Deposits
of solids can be an initiation site for crevice corrosion and under deposit
corrosion. These deposits also provide sites for bacterial growth and
MIC. As would be expected, the deposits prevent biocides from
contacting the biomass and/or corrosion inhibitors from reaching the
steel surface.

Corrosion Inhibition
Again, flow regime knowledge is fundamental to success. Which liquid,
oil or water, is the external liquid phase? Do liquids continuously wet the
entire internal pipe surface through turbulent flow or annular mist flow?
Do liquids intermittently wet the whole pipeline internal surface such as in
slug flow? Is flow laminar or stratified or wavy where liquids never or
seldom wet the top of internal pipeline surface?

For turbulent, annular mist, and slug flow, a continuous inhibition system
is preferred; for laminar, stratified, and wavy flow, a batch corrosion
inhibition treatment will be required, in addition to continuous inhibition.

Corrosion occurs in the water phase and, thus, water-soluble corrosion

inhibitors that partition to the water phase are usually preferred and are
usually the most economic. However, oil-soluble, water-dispersible
corrosion inhibitors have more persistent protective films and have been
used successfully for years in water-dominated production systems. A
well-designed corrosion alleviation system with either a good water-
soluble corrosion inhibitor or a good oil-soluble, water-dispersible
corrosion inhibitor will give good results. A well-designed corrosion
alleviation system includes sufficient agitation to disperse the oil-soluble,
water-dispersible corrosion inhibitor.

For a gas condensate-water mixture, a water-soluble corrosion inhibitor

is preferred. The gas condensate can aggravate the corrosivity, and
required inhibitor concentrations can vary from 100 ppm to 500 ppm
based on total fluid, dependent upon the corrosivity of the system.

As mentioned above, for laminar, stratified and wavy flow, both

continuous and batch inhibition are required. Continuous inhibition will
protect the liquid-wetted surfaces. Batch inhibition between pigs will be

61
needed to place corrosion inhibitor on the top of the pipeline to prevent
TOL corrosion.

When flow regime data is not available, a rule of thumb predicts that
when the gas velocity is above 18 ft/sec, continuous application of the
corrosion inhibitor will protect the total pipeline internal surface. When
the gas velocities are less than 18 ft/sec, batch inhibition, in addition to
continuous inhibition, will be required to achieve corrosion alleviation of
the entire internal pipeline surface.

Pigging
Scheduled pigging is an important part of any pipeline corrosion
alleviation system for any pipeline with a diameter of six inches or more.

For low-flow regimes, which accumulate water and/or hydrocarbon

liquids in the bottom of the pipeline, the pig will remove the liquids and
expose a clean surface for the application of corrosion inhibitor. This is
very beneficial for non-liquid packed pipelines.

However, for low-flow gas pipelines or low-flow multiphase pipelines that

accumulate large volumes of liquids, pigging can cause problems. The
pig cleans all the liquid out of the pipeline. Gas flows faster than liquids
in multiphase pipelines and liquids often accumulate. In long, large-
diameter pipelines, a slug of accumulated liquids may be very large, and
the pig delivers all of these accumulated liquids all-at-once to the
receiving facility. These accumulated liquids in front of the pig may
overload pipeline receiving facilities. The ability or inability to
accommodate large slugs of liquid often limits the feasibility for the
operators to pig two-phase or multiphase pipelines.

When pipelines with biomass are pigged, the pig removes the biomass
and exposes the pipe surface to biocide to kill remaining bacteria.
Pigging, combined with biocides, is an integral part of MIC control. In the
same vein, pigging may remove any accumulated solids that prevent
both the corrosion inhibitor from protecting the pipe wall and biocides
from contacting the bacteria.

Finally, for paraffin-prone oil pipelines, continuous application of paraffin

inhibitors is often an economic way to reduce paraffin deposits. But
usually in paraffin-prone pipelines, the continuous application of a
paraffin inhibitor will have to be supplemented with regularly scheduled
pigging. In order to improve the effect of pigging, paraffin dispersants
and solvents are often used in conjunction with pipeline pigging. This

62
pigging also will remove deposits/accumulations that cover corrosion pits
and thus allow the corrosion inhibitor to contact the steel surface.

Hydrotesting of Pipelines
The goal of hydrotesting is four-fold:
• Ensure physical integrity of the pipeline.
• Prevent oxygen corrosion during and subsequent to the
hydrotesting.
• Prevent MIC during and subsequent to the hydrotesting.
• Comply with all environmental and safety laws.

Obviously, a pipeline should be capable of operating as designed. The

hydrotest is commonly performed at 120% to 150% of design working
pressure, dependent upon the applicable design code, and helps assure
that the pipeline was fabricated and installed as designed.

Normally, the pipelines are tested with water, i.e. hydrotesting and
catalyzed oxygen scavenger is usually added to prevent oxygen
corrosion. Oxygen scavengers come in many forms, but for ambient
applications they are usually some form of sulfite, which with the help of
a catalyst, will react with oxygen to form a sulfate. Fresh water holds
about 8 ppm of oxygen at ambient temperatures. If the dissolved oxygen
is not scavenged, it reacts with the pipe. If the corrosion from 8 ppm
oxygen is averaged over the whole pipeline internal surface, the
corrosion damage is quite small. The problem is that the water is usually
pumped from one end and the oxygen reacts quickly; thus, the oxygen
corrosion is concentrated around the water injection point. In addition,
oxygen corrosion is commonly pitting corrosion, so the average corrosion
rate is ordinarily meaningless.

MIC can be a more serious problem, dependent upon on how long the
hydrotest water remains in the pipeline. The bacteria of major concern
are SRB, which metabolize sulfate and sulfite along with other nutrients
and a by-product of their metabolism is H2S. Again, if the H2S corrosion
is uniformly distributed over the entire internal pipeline wall, pipe loss by
corrosion is not significant.

But MIC does not usually occur uniformly; instead, for a number of
reasons, it occurs in the form of pits in selected places, such as under
solids or biomass accumulations. Biocides are usually added to the
hydrotest water to prevent MIC. If done properly, the biocide can prevent
the initiation of MIC biomass sites, which otherwise may have to be dealt
with later in pipeline operations.

63
Finally, hydrotest water has eventually to be disposed. And this must be
done in an environmentally acceptable and legal manner.

Land Operations
Hydrotesting on land presents a minimum of problems. First, fresh water
is usually available. This means that calcium sulfate scale, caused by a
reaction of calcium in the water and the sulfate from the oxygen
scavenger, will not be a problem because the water will usually have
very little calcium. Second, on land, toxicity of hydrotest water is less of
a problem in the disposal of the hydrotest water because it will be
subsurface-disposed.

Thus, it is less likely on land operations than in marine operations that

the pH of the hydrotest water will have to be raised (see Marine
Operations below) to reduce biocide use. In cases where the pH must
be raised, the low level of calcium in the fresh water will result in little or
no calcium carbonate scale.

Marine Operations
Hydrotesting in marine environments is much more complex, because
seawater is normally the hydrotest fluid.

Oxygen reacts with steel more rapidly in salt water than fresh water, thus
without an oxygen scavenger, the oxygen corrosion will be more
concentrated at the point of water injection.

Seawater also has a high sulfate concentration, 2,700 ppm +/-. Without
an effective biocide, this sulfate will provide nutrients for SRB for a long
time and the resultant corrosion can be cause for concern. Biocide
treatment will control the SRB. The problem is that the disposal of
hydrotest water containing biocide is toxic to marine life.

Currently, the two principal biocides of choice are THPS and

glutaraldehyde. Glutaraldehyde is less expensive than THPS, but is
more toxic and thus causes more disposal concerns. Increasing the pH
to 9.5 or higher reduces bacteria growth, but is not fatal to the bacteria.
Thus, the effectiveness of both biocides (which reduces concentration)
can be improved by increasing the water pH to 9.5 or higher. However,
in seawater, an increase of pH will precipitate calcium carbonate scale.

Environmentally, THPS is the biocide of choice in marine hydrotest

operations. But the use of THPS is more complex than the use of
glutaraldehyde.

64
At neutral pH, THPS has a half-life of about 72 days. When the pH is
raised to 9.5, the half-life is seven days. When the pH is raised to pH 10
or above, the half-life is approximately two days. Thus, a user can
formulate THPS to kill bacteria completely and then allow the hydrotest
fluid to remain in the pipeline long enough for it to deactivate (lose
toxicity) before the low toxicity hydrotest liquid is surface disposed.
THPS can also be neutralized with peroxide.

However, an oxygen scavenger reduces the effectiveness of THPS.

Thus, if THPS is used with an oxygen scavenger, it must be a sequential
operation. A solution of caustic, scale inhibitor and oxygen scavenger
can be added. When the oxygen is scavenged and most of the sulfite
has reacted, the THPS can be added. Dependent upon the oxygen
scavenger/caustic scale inhibitor mix, complete oxygen scavenging can
take from one to 20 minutes.

Obviously, the more oxygen scavenger used, the faster the reaction.
This means that for a continuous operation, the faster the scavenging
reaction, the smaller the needed vessel to hold the hydrotest water
before the THPS can be added. Conversely, the more excess
(unreacted) sulfite left after all the oxygen has been scavenged, the more
THPS required.

Glutaraldehyde is also an effective biocide, but until recently it was

believed not to degrade at neutral pH; at pH 9.5, glutaraldehyde will
degenerate over time, but much more slowly than THPS. Thus, the
hydrotest water will be more toxic for a much longer time, therefore
making it more difficult to comply with environmental restrictions on
surface disposal of hydrotest fluids.

Recent studies by Dow indicate that at least in some fresh water

anaerobic environments, glutaraldehyde will degrade over time. THPS,
however, appears still to be more environmentally friendly.

In summary, a successful hydrotest design requires careful consideration

of many complex variables as they apply to each particular situation.

65
 + ,

Introduction
Effective monitoring of a corrosion inhibition system is an integral part of
a successful and economic corrosion alleviation program. There are two
kinds of monitoring: actual (or direct) and inferred (or indirect). Likewise,
there are two application techniques of monitoring: reactive and
proactive. Direct monitoring and indirect monitoring and reactive and
proactive applications are further discussed.

Direct Monitoring
Direct monitoring is the actual measurement of corrosion throughout the
entire piping system. For wells, this can be done with either calipers or
by pulling the tubing and inspecting the pipe while on the surface.

For pipelines, direct monitoring can be accomplished with smart pigs.

For more complex surface piping and pressure vessels, comprehensive

use of ultrasonics and/or x-ray equipment reveals the actual condition of
the steel at the locations tested. To know the condition of all the steel, all
the steel must be tested.

All other monitoring techniques are indirect or inferred monitoring.

Indirect or Interpretive Monitoring

Indirect or interpretive monitoring comes in many forms, including
corrosion coupons (surface and subsurface), water analyses, and failure
history. While direct monitoring scans the whole system, indirect
monitoring looks at a few, very small portions of the system and
extrapolates the measurements acquired to the whole system.

Proactive and Reactive Monitoring

All of the corrosion monitoring techniques described below are reactive
monitoring; measurements are made to see if there has been corrosion
in the past. Proactive monitoring is a two-step process. First, a
corrosion alleviation system is developed and implemented and the

66
performance of this system is monitored reactively to confirm whether
the subject corrosion alleviation program will control corrosion.

Second, most of the subsequent monitoring effort is directed to assure

the corrosion alleviation program is rigorously followed. Because it has
been proven that if the recommended corrosion alleviation program is
followed corrosion will be controlled.

The purpose of proactive monitoring is to assure corrosion is not

occurring. Conversely, the purpose of reactive monitoring is to
determine whether the prescribed corrosion alleviation program was
followed in the past and how much if any corrosion has occurred.

From a practical perspective, it is always wise to do a little reactive

monitoring to be certain that conditions have not changed and that the
corrosion alleviation system doesn’t needs revising. But with limited
resources, it is usually more economically efficient to use the limited
resources to be certain that the corrosion control measures are being
properly followed rather than to discover later through reactive
monitoring that they were not followed.

Corrosion Coupons
Corrosion coupons are steel coupons inserted into the flow stream and
periodically removed and evaluated for corrosion. The corrosion rate of
a few square inches is then extrapolated over the entire system.
Because of the practicality of coupon location, they are nearly always
placed downstream of the well choke, even when monitoring for weight-
loss corrosion. When compared to down hole conditions of the well, this
location is cooler, has lower pressure, experiences a different water
volume, a different water composition and probably a different flow
regime. From a square-inch coupon on the surface, the operator infers
what is going on down hole.

The length of time to leave a coupon in service is another variable.

“Virgin” steel corrodes faster than steel covered with a film of corrosion
product. As a result, coupons exposed 10 days will show higher
annualized corrosion rate (mpy) than for coupons exposed six months.
Thus, there is always a tradeoff between getting timely information and
the most representative information. Coupon exposure of 30-60 days is
the industry norm.

Corrosion coupons come in many shapes and their effect on the flow
pattern in the pipe or vessel affects the measured corrosion rates. For
pipelines, probably the best is a flush mounted coupon which follows the

67
contour of the pipe. Not only does this coupon most closely resemble
flow pattern of the pipeline wall, but this coupon also does not interfere
with the use of pipeline pigs. For vessels, since pigging is not a concern
and because there is normally less flow turbulence, some of the
advantages of the flush mounted coupons over the competitive designs
are lost. The author has also used rod coupons, orifice coupons, strip
coupons, and crevice corrosion coupons.

Pitting corrosion and crevice corrosion are essentially the same thing; a
crevice is just a forced pit. A crevice corrosion coupon can be used to
indicate how effective the corrosion inhibition system is against pitting
corrosion. Sometimes in laboratory testing, crevice corrosion coupons
are used to prove that the testing environment without corrosion
inhibition is a pitting environment and that the tested inhibition system,
when applied, will stop crevice or pitting corrosion.

For pressurized systems, corrosion coupons are usually mounted in

holders that provide for changing the coupon without depressuring the
piping system or vessel.

Corrosion coupons can also be placed down hole in subsurface holders

and retrieved and replaced via wire line.

Finally, radioactive tubular coupons have been placed in tubing strings

and then the produced fluids are monitored or counted for radioactivity.
The radioactivity in the produced liquids that is above background
radioactivity is assumed to come from corrosion of the radioactive
coupon. From the measured radioactivity in the produced fluids and
from the known exposed surface of the radioactive coupon, an average
corrosion rate can be calculated.

The user should always remember that a corrosion coupon only

measures what happened on the coupon. Any other interpretation is
inferred or extrapolated.

Probes, Devices and Techniques for External Monitoring of Internal

Corrosion
There are many techniques that monitor internal corrosion that do not
require the physical insertion and removal of the corrosion coupon.
However, like removable corrosion coupons, they only measure
corrosion in the precise location of the coupon.

68
Electrical-Resistance (ER) Probes
An ER (Electrical-Resistance) probe is a small wire placed in a stream to
be measured for corrosion. The wire’s electrical resistance can be
measured via an external connection. As the wire corrodes, the
diameter gets smaller and the electrical resistance increases. This
increase in electrical resistance is then converted to a calculated metal
loss. Since readings are easily obtained, many measurements can be
made, and comparisons of different corrosion inhibition systems can be
made without changing the probe. One drawback is that for a change in
resistance to be meaningful, the wire must have a relatively small
diameter, and when the wire corrodes in two, the probe must be
replaced. Like many other corrosion-measurement methods, an ER
probe only measures corrosion at its location and position in the pipeline
or vessel. Also, ER probes are not recommended for some sour
systems because some iron sulfide corrosion products may be
conductive.

Hydrogen Patch
This Hydrogen patch device is marketed under many names. In general,
corrosion, (as outlined in Chapter I), generates hydrogen. And since
hydrogen is a very small molecule, some of it may permeate the steel. A
device can be externally placed on the steel that measures hydrogen
permeation. These devices may pick up hydrogen generated by
corrosion, but the readings are difficult to convert to a corrosion rate.
The advantage of hydrogen patches is that they are completely
noninvasive. However, they currently do not enjoy widespread use in oil
and gas operations.

Multi-End Element Tube

A pipe sized tube with “multi-end elements” is on the market and has
experienced limited application. These end elements are placed around
the total circumference of the tube and the tube is welded into the
pipeline. The technology is complex and outside the scope of this
document. But in short, these end elements are reported to be able to
detect and measure very small changes in the full circumference of the
pipe wall. The data from the tube with end elements are loaded into a
computer, which, with proprietary software, converts the data to
measured corrosion rates. However, these devices are both expensive
to purchase, and expensive to maintain. There is very little data
available that independently compares device corrosion measurements
with independent measurements of proven corrosion-detection tools,
such as smart pigs or corrosion coupons.

69
With technology advancement, these devices may become more robust
and less expensive to install, operate, and maintain.

X-Rays
X-raying steel is a time tested method to look at internal steel surfaces.
The principal advantages are that it can be done externally and is
completely non-intrusive. A disadvantage of x-raying is that it looks only
at a very small part of a pipe or vessel wall. Other disadvantages are
that it requires special equipment and specially trained personnel, and as
compared to many other monitoring techniques, x-raying is often more
costly.

Ultrasonics
Ultrasonic technology can look directly through metal by reflecting sound
waves off the metal’s interior wall. The reflection time is converted to
wall thickness. Often in vessels, measurement points are marked on the
exterior and ultrasonic measurements are made periodically at the same
place to monitor internal corrosion. Ultrasonic data are subject to
misinterpretation because the sound waves will also reflect off any
internal defect, such as laminations or inclusions.

Ultrasonic measurements are completely noninvasive and require no

special consideration during construction. However, the technology does
require special equipment and specially trained personnel.

Wet Chemistry
Water analyses have been used for decades to monitor corrosion.

Iron is the usual ion monitored. The industry has developed special
protocols for sample collecting, handling, and analysis. Iron content of
produced water for non-H2S wells is a good indicator of corrosion trends.
Iron measurements are often used to determine when to reapply
corrosion inhibitor in wells that are batch inhibited. Iron content of
produced waters is most valuable in monitoring corrosion trends.

Iron measurements have limited value in determining absolute corrosion

rates. Theoretically, if the iron content of the connate water is known
(and it is seldom if ever known), pounds of iron lost from the pipe could
be calculated by multiplying the water production volumes by the iron
concentration of produced water. The author has never found this
technique to be a reliable method to determine absolute corrosion rates.
Iron measurements of produced water, however, can be an indicator of
corrosion- rate changes.

70
Iron measurements are of no value when H2S is produced. Iron
sulfide is insoluble in water, and iron in water samples is not reflection of
dissolved iron. (There is no dissolved iron in the water.) Iron in water
samples from sour production is a reflection of the sampling technique;
the iron content of the water is a reflection of how much precipitated iron
sulfide was collected.

Manganese is sometimes used as a substitute for iron measurements

when H2S is produced. Most oil field steels contain 2%+/- manganese,
and manganese sulfide is more soluble in water than iron sulfide. The
idea is to measure for manganese since, if the steel corrodes,
manganese will be released and can be measured. There are some
papers in the literature about manganese analysis of produced waters in
sour service to monitor corrosion.

If there is no manganese in the connate water, manganese from

corrosion should only be 2% of expected iron content. Thus, manganese
content will be low compared to iron. In reality, manganese sulfide,
although more soluble in water than iron sulfide, is essentially also
insoluble in water. One hundred times zero is still not very much
solubility. The author has no confidence in manganese measurements
of produced water in sour wells as an indicator of corrosion changes.

71
 + ,

Introduction
The author has performed many economic analyses that compared the
use of corrosion-resistant materials versus the use of corrosion
inhibitors. If the use of corrosion inhibitors is both practical and they will
be reliably and regularly applied, the use of corrosion inhibitors is nearly
always the more economic choice.

There are situations in oil and gas production where corrosion-resistant

materials are more economic than the use of corrosion inhibitors, such
as facilities, wells or pipelines systems with complex geometry or
locations where logistics prevent reliable corrosion inhibition. In these
situations, corrosion-resistant materials are often successfully applied.
These corrosion-resistant materials can be corrosion-resistant metals or
they can be non-metallics.

Whenever dissimilar metals are mixed, galvanic corrosion should also be

addressed.

Corrosion-Resistant Alloys
Corrosion–resistant alloys are metals, initially called stainless steels, that
resist corrosion. Stainless steels are defined as steels with at least
10.5% chromium (Uhlig); some authors set higher minimum chromium
levels such as 12% (Craig) and others use a minimum level of 11%
(Sedricks). The two original stainless steels are 1.) a 410 type
martensitic stainless steel, which is a 12% chromium martensitic
stainless steel initially developed in England in 1912 as a 12.8%
chromium corrosion-resistant gun barrel; and 2.) an 18-8 (18% chromium
and 8% nickel) austenitic stainless steel (current, 304 stainless steel)
developed in Germany in 1912. The development of CRA did not begin
in earnest until the 1970s. Prior to this, molybdenum was added to 304
SS to make 316 SS and the copper-nickel alloys were developed to
resist oxygen corrosion.

The development, alloying, and testing of CRA is well beyond the scope
of this text. In the simplest terms, chromium prevents corrosion, nickel

72
prevents cracking, and molybdenum prevents pitting. For example, 316
SS is 304 SS with 2%-3% molybdenum. Gas processing plants often
use 304 SS because of low chloride water, while oil field operators use
316 SS to avoid chloride pitting.

Pitting resistance of stainless steels and CRA is known as the PREN

(Pitting Resistance Equivalent Number) and is expressed as follows:

PREN = Cr% + 3.3(Mo% + 0.5W %) + 16N%

With the exception of oxygen corrosion, all CRA have sufficient

chromium to resist corrosion. The key to successful CRA use is the
avoidance of cracking, both cathodic cracking, SSC and anodic cracking,
SCC (such as chloride cracking). Usually SSC is a low temperature
event and SCC is a high temperature event. However, H2S can
contribute to both SSC and SCC of CRAs.

Compliance with NACE International MR0175/ISO 15156 is the industry

standard for avoiding environmental cracking in oil field applications of
metals in H2S service. In most locations, compliance with MR0175 is
mandated by law. Again, the key in avoiding environmental cracking is
both proper alloying and correct manufacturing processes. If the CRA is
highly alloyed enough to avoid environmental cracking, it most likely will
have sufficient chromium to avoid corrosion.

Oxygen corrosion is a different situation. For avoidance of oxygen

corrosion, copper-nickel alloys are used with great success. A word of
caution is in order, however. When copper-nickel alloys are
contaminated with sulfide, they will corrode under flow conditions. As
discussed earlier, there is a documented case of a boat sinking that had
copper-nickel piping for seawater systems. When the boat came into a
harbor with dirty water containing sulfide, the seawater system failed and
the ship sank.

Corrosion-resistant alloys, including the martensitic stainless steels (410

SS and 420 SS, etc.) are not resistant to oxygen-pitting corrosion. For
instance, as previously mentioned, 410 SS trim has been known to
corrode while sitting on the warehouse shelf waiting to be used.
Chlorides and oxygen will cause pitting on nearly all CRA. Both
increased chloride concentration and increased temperatures enhance
pitting. By inspection of the PREN calculation, 316 SS has more pitting
resistance than 304 SS. Oxygen corrosion is a concern when subsea
CRA control lines, such as long umbilicals for remote subsea sites, are
used. Shell found that a PREN of >42.5 in super duplex would avoid
oxygen-chloride pitting in ambient seawater.

73
Non-Metallics
Another option when corrosion inhibition is not practical is the use of
non-metallics. These corrosion-resistant materials can be used as
coatings or as solid materials. They all have two characteristics: they are
permeable and they are flammable.

All thermoplastics are permeable. This is why carbonated beverages go

flat in plastic bottles; the CO2 leaks through the plastic. This is why the
annulus behind a 0.5-inch thick Nylon 11 liner in flexible flow line has to
be vented because methane penetrates the Nylon 11 liner. Water
permeates Teflon. Gas permeability is why elastomers explode when
rapidly depressured. Methanol can also leak though polyethylene tubes.

Non-metallic materials are usually hydrocarbons—or at least have a

significant hydrocarbon component—and as such, they are flammable.

Compatibility with the fluids in the intended environment is essential.

These non-metallics are organic and production and treating fluids are
organic, thus compatibility is an important concern. Glycol and methanol
may remove plasticizer from polymers and make them brittle. Some
non-metallics are not compatible with acids or xylene. Treating fluids,
such as corrosion inhibitors and paraffin inhibitors, may not be
compatible with the non-metallics.

Compatibility is particularly important when FRP (Fiberglass Reinforced

Pipe) is used as solid pipe or as liners inside steel pipe. If the incorrect
curing agent of the epoxy resin is used, CO2 will leach the epoxy from
the FRP, leaving only the glass fibers.

Compatibility with the produced and treatment fluids is essential if non-

metallics are to be successfully used as a corrosion alleviation tool.

Non-Metallic Coatings
Internal coatings have been used for years as a corrosion control
measure. External paints have been in use for hundreds of years to
alleviate corrosion. Internal coatings have been part of the oil field for
more than 50 years. When coatings are used on steel, the steel
provides the structural strength and the coating provides the corrosion
alleviation.

The down side of non-metallic coatings is that they are permeable, and
eventually water and CO2 will permeate the coating and reach the steel.
Thus, down hole non-metallic coatings have a finite life, usually less than
a few years. For maximum benefit, the coating must be applied and put

74
in service defect-free, a nearly impossible task. Since the steel is
providing the structural strength, the coating must be more ductile than
the steel. Otherwise, when the steel moves as it is loaded, the non-
metallic coating will crack.

Non-metallic coatings in less severe service (cooler, less saline or fresh

water, little CO2 or H2S, neutral pH, etc.) have a longer expected life.

Solid Non-Metallic Materials

Solid non-metallics have enjoyed widespread success for years in
specialized applications. These non-metallic pipes and tanks have low
mechanical strength, but supreme corrosion resistance. Non-metallics
are commonly used for chemical storage tanks and drums. With no steel
backing, miniscule permeability usually does not matter.

The downside of solid non-metallic tanks, drums, and pipe is the low
mechanical strength and low fatigue properties. In addition, they have
more restrictive operating temperatures than metals. As non-metallic
materials cool, they become less ductile and when they get too warm
they weaken. They do not stand up well in vibrating or pulsating loads
and are more prone to damage from bumping and falling objects. In
spite of these drawbacks, solid non-metallic pipe is widely used for salt
water handling systems on land and are near universal standard for
residential water and sewerage systems.

Galvanic Corrosion
When a metal is immersed in a water solution, its ions tend to pass into
solution; the result is corrosion. The corrosion is resisted by osmotic
pressure of the ions already dissolved in the water. When equilibrium is
reached between the osmotic pressure and the force of the metal ions to
dissolve, this is called the electrode potential. Pure metals can be
arranged in accordance with their electrochemical potential from the
least noble to the most noble. This is called the electrochemical series.
When metals become oxidized on the surface, their electrochemical
potential changes. Arrangement of metals in accordance with their
oxidized surfaces is called the galvanic series. Since most metals
become oxidized on the surface, the galvanic series is the tool used to
avoid galvanic corrosion. A partial galvanic series from least noble to
most noble follows:

75
 Galvanic Series
(In Seawater)

Magnesium
Zinc
Beryllium
Aluminum Alloys
Cadmium
Mild Steel and Cast Iron
Low Alloy Steel
Austenitic Nickel Cast Iron
Aluminum Bronze
Naval Brass, Yellow Brass, Red Brass
Tin
Copper
Lead-Tin Solder 50/50
Admiralty Brass, Aluminum Brass
Magnesium Bronze
Silicon Bronze
Tin Bronze
410,416 Stainless Steel
Nickel Silver
90-10 Copper Nickel
80-20 Copper Nickel
430 Stainless
Lead
70-30 Copper Nickel
Nickel Aluminum Bronze
Nickel Chromium Alloy 600
Silver Braze Alloys
Nickel 200
Silver
302, 304, 321, 347 Stainless Steel
Nickel Alloys (Monel)
316, 317 Stainless Steel
Alloy 20 Stainless Steel
Alloy 825
Titanium
Hastelloy C (276)
Graphite
Gold
Platinum

76
When two metals of different galvanic potentials are in electrical contact,
the less noble will corrode and protect the more noble metal from
corroding. This is the basis of cathodic protection. The more noble steel
is protected from corrosion by sacrificial zinc or aluminum anodes. The
anode corrodes and prevents the steel from corroding. Thus, metal
anodes are consumed and have to be periodically replaced. Galvanizing
with zinc uses the galvanic series to protect the steel. When the zinc is
all consumed, the galvanized item will corrode. Incidentally, the same
galvanic protection can be achieved with an impressed electrical current.

Inspection of the galvanic series has some practical implications in

equipment design. When two dissimilar metals are in contact in an
aqueous environment (moist air can be an aqueous environment) the
less noble metal needs to be of much larger volume than the more noble
metal (Figures 10a & 10b).

For example, for a ship with a steel hull, the rivets should be stainless
steel so the hull corrodes to protect the rivets; the small amount of
galvanic corrosion to protect the rivets is spread over a wide area of the
hull and is of no concern. If the opposite were the case and the rivets
corrode to galvanically protect the large ship hull, the rivets would soon
be consumed and the results catastrophic.

In oil and gas operations where low alloy steel and stainless steel are
mixed, the low alloy steel will galvanically protect the stainless steel. A
stainless valve inserted into a large steel facility will be no problem. The
low alloy steel will galvanically protect the stainless steel, and any
galvanic corrosion to protect the small valve is spread over a large area.
However, if a low alloy steel valve is used in a large stainless-steel
facility, the low alloy valve will galvanically protect the large stainless-
steel facility and any galvanic corrosion will concentrate on the low alloy
steel valve. Figures 10a and 10b show examples of galvanic corrosion.

77
Figure 10a – Galvanic Corrosion – Less Noble Metal was Placed Between Two More
Noble Metals

Figure 10b – Galvanic Corrosion – Less Noble Metal was Placed Between Two More
Noble Metals

78
1. “The Petroleum Handbook,” Royal Dutch Shell, Elsievier,
Amsterdam, 1983.
th
2. “The Petroleum Handbook,” Royal Dutch Shell, 5 Edition, Balsing
and Mansell Publishers, Wisbeck Cambridgeshire, 1966.
3. Revie, R. Winston, Editor, “Ulig' s Corrosion Handbook,” Second
Edition, John Wiley and Sons, New York, 2000.
4. Craig, Bruce D., “Practical Oil field Metallurgy” Bruce Craig publisher,
1993.
5. McGannon, Harold E., Editor, “The Making and Shaping of Steel,
Eight Edition, United States Steel.
nd
6. Sedriks, A. John, “Corrosion of Stainless Steels,” 2 Edition.
7. S.A.E. and ASTM Publishers, “Metals and Alloys in the Unified
th
Numbering System,” 8 Edition.
8. Crolet, J.L., “Which CO2 Corrosion, Hence Which Prediction,” Elf
Aquitaine, Pau, France.
9. de Waard, C. and Lotz, U., “Prediction of CO2 Corrosion in Carbon
Steel,” Paper 69, Corrosion 1993, NACE.
10. Kapusta, Sergio, D. and Place, Morris C. Jr., “Corrosion Inhibition in
Oil and Gas Production “ Corrosion 1994, NACE.
11. Landrum, R. James, “Fundamentals of Designing for Corrosion
Control,” National Association of Engineers,1989.
12. Pots, B.P.M. and Hendriksen, E.L.J.A., “CO2 Corrosion Under
Scaling Conditions-The Special Case of TOL Corrosion in Wet Gas
Pipelines,” Paper 31, Corrosion 2000, NACE.
13. Crolet, J.L., Thevenot, Nicolas, and Dugstad, Anne, “Role of Free
Acetic Acid on the CO2 Corrosion of Steels,” Paper 24, Corrosion
1999, NACE.
14. Kapusta, Sergio D., Personal communication.
15. Dunlop, A.K., Hassell, H.L., and Rhodes, P.R., “Fundamental
Considerations in Sweet Gas Well Corrosion,” Corrosion 1983,
NACE.
16. Place, Morris C. Jr., “Corrosion Control Considerations for High
Pressure Sour Gas Reservoirs,” Sulfur 1984, Calgary.
17. NACE, Technical Practices Committee Publication, “Corrosion
Control in Petroleum,” 1979.
18. NACE, T-1, D-41 Committee Document 2003-3-20.
19. Materials Performance, March 2004, “Expert Consensus on MIC
Prevention and Monitoring,” NACE.
20. Belkin, Shimshon and Jannasch, Holsen W., “A New Extremely
Thermophilic Sulfur Reducing Heterotrophic Marine Bacterium,”
Active Microbiology, Woods Hole Institute.
21. Rodgers and Templeton, Personal Communication.

79
22. Avesta Sandvik Tube, “Chloride Welded Stainless Tubes and Pipe,”
Fagersta, Sweden.
23. Ibidunni, A. O., Bookwalter, J.C., and Guenther, D.A., “Analyses of
Engineering Materials Failure in Marine Environment,”
Microstructural Science, Vol. 15, Editors, Blum, M.E., French, P.M.,
Middleton, R.M., and Vander Voort, G.F.
24. “Betz Handbook of Industrial Water Conditioning,” Ninth Edition,
1991.
25. Hendricks, Clayton T. and Stevens, Russell D., “Managing Sucker
Rod Failures in a Rod Pumped Well,” Norris.
26. Hendricks, Clayton T. and Stevens, Russell D., “Sucker Rod Failure
Analyses,” Norris, 2000.
27. Champion Technologies Inc., “Mechanically Induced Failures,” 1995.
th
28. API, “Specification for Sucker Rods,” API Specification 11B, 25
Edition, 1995.
29. Moore, Keith, “Stop Sucker Rod Failures to Save Money,” Petroleum
Engineer Int., July 1981.
30. Weigel, Francis, Bender, Richard D., and Bell, James Daniel, SPE
Paper 67269, Oklahoma City, March 2001.
31. API, “Care and Handling of Sucker Rods,” ANSI/API 11BR
Specification, 1989.
32. API, “Specification for Sucker Rods,” API 11 B, 1998.
33. API, “Design Calculations for Sucker Rod Pumping Systems,”
ANSI/API Specifications 11L.
34. NACE, “Recommended Practice for Corrosion Control of Sucker
Rods by Chemical Treatment.”
35. Jordan, Ken, Personal communication.
36. Bross, Harold E., Personal communication.
37. API, “Recommended Practice for Design and Installation of Offshore
Production Platform Piping Systems,” API Recommended Practice
th
14E, 5 Edition, October 1991.
38. Kapusta, Sergio D., Pots, B.F. M., and Connell, R. A. “Corrosion
Management of Wet Gas Pipelines,” Paper 45, Corrosion 1999,
NACE.
39. Prasad, Rupi, “Chemical Treatment Options for Hydrotest Water to
Control Corrosion and Bacterial Growth,” Champion Technologies
Inc.
40. Jastrzebski, Bigniew D., “Nature and Properties of Engineering
Materials.” John Wiley and Sons, 1959.
41. Erosion/Corrosion Center-University of Tulsa, “Erosion-Corrosion
Problems-Can an Inhibitor Help?” 1996.
42. Filho, Jose C. Cardoso, and Orazem, Mark E., “Application of a
Submerged Impinging Jet to Investigate the Influence of
Temperature, Dissolved CO2, and Fluid Velocity on Corrosion of
Pipeline-Steel in Brine,” Paper 01058, Corrosion 2001, NACE.

80
43. Rhodes, P.R., “Abstract No. 107,” Electro-Chemical Society Meeting,
Las Vegas, October 1976.
44. Zitter, von H., “Korrosionserscheinungen in Sauergassonden”
Erdoel-Erdgas. Zeitschrift, Vol 89, March 1973.
45. Kapusta, Sergio D., Personal communication on corrosion inhibition.
46. Daniels, W.J., Personal communication on CO2 corrosion.
47. Weeks, S.G., Personal communication on corrosion inhibition.
48. Ragout, R., Personal communication on hydrotest pressure codes.
49. Koch, Gerhardus H., Brongers, Michiel P. H., Thompson, Neil G.,
and Virmani, Y. Paul, “Corrosion Cost and Preventative Strategies in
the United States,” United States Department of Transportation,
Federal Highway Administration, 2002.
50. McCarthy, John, “Frequently Asked Questions About Nuclear
Energy,” Stanford University Professor Emeritus, Web page, Nuclear
Energy Now, 2007.
51. Endean, Howard. J., “Oil Field Corrosion Detection and Control,”
Champion Technologies Inc., 1989.
52. Place, Morris C. Jr., “Corrosion Control-Deep Sour Gas,” SPE 8310,
1979, Las Vegas.
53. Place, Morris C. Jr., “Corrosion Inhibition for Severely Corrosive Gas
Wells,” Corrosion, Volume 48, No. 4, April 1992, NACE.
54. Iken, George N., Personal communication on FRP pipe.
55. Bradley, Bryant W., “Two Oil field Water Systems,” Robert E.
Krieger, Malabar, FL, 1987.

81
APB Acid Producing Bacteria
B/D Barrels per Day
CRA Corrosion-Resistant Alloys
EOR Enhanced Oil Recovery
ER Electrical-Resistance
ESP Electric Submersible Pumps
F Fahrenheit
FRP Fiberglass Reinforced Pipe
ft/hr Feet per Hour
ft/sec Feet per Second
HIC Hydrogen Induced Cracking
ID Internal Diameter
IOB Iron Oxidizing Bacteria
IOR Improved Oil Recovery
MCF 1000 cubic feet
MIC Microbiologically Influenced Corrosion
MMCF Million Standard Cubic Feet
mpy mils per year
NRB Nitrogen Reducing Bacteria
PI Productivity Index
ppb Parts per billion
ppm Parts per million
PREN Pitting Resistance Equivalent Number
psi/ft psi per foot
psig pounds per square inch gauge
SCC Sulfide Stress Cracking
SOHIC Stress Oriented Hydrogen Induced Cracking
SRB Sulfate Reducing Bacteria
SSC Sulfide Stress Cracking
SSCC Sulfide Stress Corrosion Cracking
Tetrakis Hydroxymethyl Phosphonium
THPS Sulfate
TLP Tension Leg Platform
TOL Top of Line

82