United States
Environmental Protection .
Agency
Office of
Drinking Water
State Programs Division
WH 550 E
Washington, DC 20460
EPA 570/9-87-002
August 1987
Technical Assistance Document:
Corrosion, Its Detection and
Control in Injection Wells
-------
TECHNICAL ASSISTANCE DOCUMENT:
CORROSION, ITS DETECTION AND CONTROL IN INJECTION WELLS
Prepared by:
SMC Martin Inc.
Valley Forge, Pennsylvania
and
The Underground Injection Control-Quality Assurance Work Group
for the
U.S. Environmental Protection Agency
Office of Drinking Water, State Programs Division
Underground Injection Control Branch
Project Manager
Mr. Mario Salazar
State Programs Division, ODW(WH-550E)
EPA Headquarters
401 M Street, S.W.
Washington, D.C. 20460
Editing of Text by:
Drew Dawn Enterprises, Inc.
P.O. Box 41126
Washington, D.C. 20018-0526
August 1987
-------
ACKNOWLEDGEMENTS
The preliminary draft of this document was prepared by Mr.
Michael Edelman, Mr. John Mentz, and Mr. David Weiss of SMC Martin Inc.,
Valley Forge, Pennsylvania, under EPA Contract Number 68-01-6288. It was
extensively revised by the UIC-QA work group. Mr. Bert Moniz of DuPont and
Mr. Mario Salazar of EPA contributed extensively to the final product.
The authors acknowledge the valuable input provided by AMF Tuboscope,
Houston, Texas, and Mr. Peter Lassovszky of the Office of Drinking Water,
EPA, Washington, D.C.
The EPA-appointed UIC Quality Assurance Work Group members, who
contributed their time and efforts and were instrumental in shaping the
final document include:
Philip Baca *
New Mexico Oil Conservation Division
P.O. Box 2088
Santa Fe, NM 87501
Gene Coker
U.S. EPA Region IV, GWS, WSB
345 Courtland Street, NE
Atlanta, GA 30365 (404) 881-3866
Jeff van Ee
U.S. EPA, EMSL
P.O. Box 15027
Las Vegas, NV 89114 (702) 798-2367
Richard Ginn *
Railroad Commission of Texas
P.O. Drawer 12967
Capital Station
Austin, TX 78711
Fred Hille
Bureau of Pollution Control
Mississippi Department of Natural Resources
P.O. Box 10385
Jackson, MS 39209 (601) 961-5171
Juanita Hillman
U.S. EPA Region VIII, 8ES
999 18th Street
1 Denver Place, Suite 1300
Denver, CO 80202-2413 (303) 236-5065
*No longer a workgroup member
-------
ACKNOWLEDGEMENTS
(continued)
Linda Kirkland (6E-Q)
QA Office
U.S. EPA Region VI
1445 Ross Avenue
Dallas, TX 75202 (214) 767-9791
Charles A. Koch
NDIC - Oil and Gas Division
900 East Boulevard
Bismarck, ND 58505 (701) 224-2969
B.J. (Bert) Moniz »
E.S.D. Regional Office
E.I. DuPont De Nemours & Company (Inc.)
Engineering Department
P.O. Box 3269 Beaumont, TX 77704
Bernie Orenstein »
U.S. EPA Region V, 5WD
230 South Dearborn Street
Chicago, IL 60604
Paul Osborne *
8WM-DW
1860 Lincoln Street
Denver, CO 80295
Erwin Pomerantz (VH-550E) *
U.S. EPA, Headquarters
401 M Street, SW
Washington, D.C. 20460
Joseph Roesler
U.S. EPA, EMSL
26 W. St. Clair
Cincinnati, OH 45268 (513) 569-7286
Mario Salazar (WH-550E)
U.S. EPA, Headquarters
401 M Street, SW
Washington, D.C. 20460 (202) 382-5561
Ron Van Wyk *
U.S. EPA Region VI, WSB (6W-SG)
1445 Ross Avenue
Dallas, TX 75202
*no longer a workgroup member
-------
DISCLAIMER
This report was prepared for informational purposes. Mention of
specific companies, trade names, or commercial products does not
constitute endorsement or recommendation by the U.S.
Environmental Protection Agency (EPA).
-------
TABLE OF CONTENTS
LIST OF FIGURES i
LIST OF TABLES ii
INTRODUCTION 1
I. TYPES OF CORROSION 2
An Electrochemical Background to Corrosion 2
Factors Which Influence Corrosiveness of
Injection Well Environments 4
Oxygen Corrosion 6
Carbon-Dioxide Corrosion 7
Hydrogen-Sulfide Corrosion 8
Microbiological!? Influenced Corrosion 8
Galvanic Corrosion 9
Non-metallic Degradation 9
II. INORGANIC INCRUSTATION 16
Chemical Stability 16
III. FAILURE DUE TO CORROSION - REPRESENTATIVE EXAMPLES 22
Brine Disposal 22
Hazardous Waste Disposal 22
Inorganic Incrustation in an Injection Well 23
Dilute Organic Acid and Stainless Steel 23
IT. DETECTION AND MEASUREMENT OF CORROSION 24
Weight-Loss Coupons and Corrosion Loops 24
Electrical-Resistance Probes 26
- Polarization-Resistance Probes 26
Well-Logging Methods 26
Detection of Microbiologically Influenced Corrosion 27
V. CORROSION CONTROL 28
Protective Coatings 28
Protection of the Packer and other
Bottom-Hole Components 30
Preinjection Treatment 30
Chemical Inhibitors 31
Cathodic Protection 37
Corrosion and Hazardous Injection Fluids 39
APPENDIX A. References 42
APPENDIX B. Glossary of Corrosion Related Terms 45
APPENDIX C. Check List, Inspection and Operator 47
-------
LIST OF FIGURES
Figures
1 Corrosion Reaction: Carbon Steel in a Reducing Acid 3
2 Oxidation-Reduction Potential (Eh) and pH Diagram
Showing Stability of Iron" Species and Phases At STP 5
3 The Effect of Solution Conductivity on Galvanic
Corrosion, (a) Low Solution Conductivity
(b) High Solution Conductivity 11
4 New Pipe Replacing a Section in an Older Pipeline
Often is Anodic and Corrodes Faster than the
Old Pipe which is Partly Protected by Previously
Formed Coatings of Rust 12
5 The Effects of Pressure and Temperature on Carbon
Dioxide Solubility (Modified after Persons and
Hart, 1980) 17
6 Values of K at Various Ionic Strengths (from
Warner and Lehr, 1977; after Stiff and Davis,1952) 18
7 Graph for Converting Concentrations of Calcium and
Alkalinity (in ppm) to pCa and pAlk (ibid. Figure
5) 19
8 Example of Cathodic Protection Scheme for Well
Casing (EPA, 1982) 38
-------
LIST OF TABLES
Table
1 Galvanic Series for Metals in Sea ¥ater
(Jellinek, 1958) f0
2 Chemical Resistance of Some Commercial Plastics 13
3 Water Analysis Used in Sample Calculation of
Calcium Carbonate Saturation Using Stiff and
Davis Index (Ostroff, 1965) 20
4 Illustration of Corrosion Rates Obtained from
Weight Loss Coupon Testing in "Sour" (Hydrogen
Sulfide Containing) Saltwater (Gulf Oil
Corporation, 1948) 25
5 Suitability of Casing and Tubular Goods to
Various Corrosion Environments (Allen and
Roberts, 1978) 29
6 Common Chemicals Used for Injection Fluid
Neutralization (Warner and Lehr, 1977) 32
7 Alkali and Acid Requirements for pH
Neutralization (Warner and Lehr, 1977) 33
8 Corrosion Inhibitors (Gatos, 1956) 35
9 Class I Injection Chemicals 40
ii
-------
INTRODUCTION
In order to protect underground sources of drinking water (USDWs), injection wells must not
allow fluids to escape into unauthorized zones. Any escape of liquids may cause contamination
of USDWs directly, or by forcing lower quality fluids to move. If a well protects USDWs by not
allowing substances to escape or migrate, it is said to have mechanical integrity. One part of
mechanical integrity is to assure that the well does not develop leaks; and, one possible cause of a
leak in a well is corrosion.
Most injection wells are constructed with metallic materials for structural reasons. Non-
metallic materials may be used in specific areas where metals are not adequate. Corrosion of the
metallic materials and degradation of the non-metallic materials are the chief causes of premature
failure in injection wells.
Although there are several forms of corrosion, they may be grouped into two main forms,
general and localized. General corrosion is the uniform or near uniform thinning of metal. If
the rate of general corrosion is tolerable, an adequate lifespan can be built into the disposal well
materials by adding a corrosion allowance to the design thickness. If the general corrosion rate is
too high, the material should not be used. Localized corrosion consists of several forms of attack
that lead to failure of the equipment before the corrosion allowance is used up. Failure may arise
from the development of a leak, from mechanical failure caused by localized thinning, or from
crack propagation. Examples of localized thinning and crack propagation are pitting and crevice
corrosion.
The degradation of non-metallic materials may exhibit a variety of forms, all of which lead to
loss of structural properties and possible failures.
The purpose of this document is to summarize information on the occurrence, detection and
control of corrosion. It is not intended to establish any regulatory requirements for injection
wells. Section I provides a description of the types of corrosion; section II explains inorganic
incrustation, whereas III gives some representative examples on the effect of corrosion on injection
wells. Section IV, of this manual, discusses well corrosion with respect to typical waste fluids and
section V describes the various corrosion control techniques. Corrosion of the well components by
the underground environment is explained throughout this manual (as applicable). The reader is
directed to Barnes and Clarke (1969) for more detail on corrosion of well components by natural
ground water. Appendix C. has check lists that can be used by the regulatory inspector and the
well operator to detemine the effectiveness of corrosion control measures.
-------
I. TYPES OF CORROSION
Corrosion is a term which is widely used in describing the degradation of construction materials
used in all phases of injection practices. A majority of corrosion reactions directly alter the well
components. For example, rust results from the direct alteration of a metallic casing. However,
incrustation reactions affect well performance by clogging well screen openings and by decreasing
pore space in the injection formation. Failure of a well through incrustation occurs indirectly. For
this reason, incrustation will be discussed in a section of its own. Of special interest are the effects
of hazardous wastes on injection wells, which result because well injection is the most widely used
technique for disposal of these wastes. The appendices contain information on the corrosive effects
of hazardous waste.
An Electrochemical Background to Corrosion.
Corrosion, strictly speaking, is the electrochemical dissolution of metal in an electrolyte. Ex-
amples of electrolytes are acids, alkalies, salt solutions, and the soil. An electrolyte contains positive
and negative ions. When the metal corrodes, it gives up electrons and the metal ions enter the
electrolyte solution.
At the corroding metal surface, two types of reactions are occurring simultaneously:
° The anodic reaction, in which metal atoms are dissolved to form positively charged ions and
electrons.
° The cathodic reaction, in which specific ions in the electrolyte accept the electrons. See Figure
1.
Thus, there is a passage of positive charge from the metal to the electrolyte at the anode,
balanced by a passage of negative charge from the electrolyte to the metal at the cathode. When
corrosion occurs, a current flows from the anode to the cathode. Since the anode and cathode
currents are equal and opposite, no electric shock is experienced from a corroding system.
If the corrosion is general, the anodic and cathodic sites are switching continuously, resulting in
relatively uniform metal loss. When corrosion is localized, such as crevice or pitting corrosion, the
anodic and cathodic sites are located on different parts of the same metal surface. With galvanic
corrosion, the anode and cathode are two different metal surfaces.
The anodic reaction does not change, i.e., metal atoms giving up electrons to form ions.
However, the cathodic reaction will vary, depending on the type of electrolyte. The examples of
steel in aerated salt solution and in aerated and deaerated sulfuric acid will be used to illustrate
three different cathodic reactions.
In all cases, the anodic reaction is the same. At anode sites, the steel corrodes to form ions
and electrons:
Fe (metal] —>• Fe^ + 2e~ (in electrolyte) (l)
For aerated salt solution, the balancing cathodic reaction is:
i<92 + H20 + 2e~ —> 20H- (2)
£t
For aerated hydrochloric acid, it is:
(3)
-------
ACID
(CONTAINS
Fe
2e--
GAS
STEEL
(CONTAINS
Fe°)
Fec
2e~
Figure 1 . CORROSION REACTION
Carbon Steel in a Reducing Acid
-------
And for deaerated hydrochloric acid, it is:
2H+ + 2e~ —> #2 (gas evolved) (4)
The rate of corrosion (the anodic reaction) is often controlled by the type of cathodic reaction.
Disposal well environments exhibit a wide variety of possible cathodic reactions which must be
evaluated. The anodic and cathodic reaction products will often combine to form solid corrosion
products, such as rust. Unless mechanically removed or naturally swept away, the formation of
solid corrosion products often inhibits further attack and reduces the corrosion rate. Figure 2
shows the thermodynamic states of iron and its compounds depending on the oxidation-reduction
potential (Eh) and hydrogen ion concentration (pH) conditions.
Factors Which Influence Corrosiveness of Injection Well Environments.
The corrosion potential of an injection well will depend on the materials of construction, the
nature of the hydrologic and geologic environments and the operating conditions. For example,
concentrated sulfuric acid is not corrosive towards carbon steel at ambient temperature because
it is strongly oxidizing, which causes passivation (to cause the formation of a protective film).
However, if the temperature is raised or the velocity of flow is increased, concentrated sulfuric acid
becomes extremely corrosive towards carbon steel for two reasons. First, increased temperatures
cause the protective film to dissolve; and second, increased velocities cause the protective film to
be mechanically removed.
Alloys which easily passivate, such as stainless steels and titanium, act in an opposite manner
to carbon steel and have better corrosion resistance under aerated or flowing conditions. They are
more likely to be attacked when oxygen concentration is low, such as in joints or cracks.
The presence of aggressive species will alter corrosion behavior. The chloride ions, for example,
may easily penetrate the passive film on stainless steel and cause a deep localized form of attack
known as pitting.
The pH, which is a measure of acidity or alkalinity, is a useful indicator of corrosiveness for
certain alloy systems. With carbon steel, a pH of less than 4 indicates the presence of hydrogen
ion (free acid), which is usually corrosive. If aggressive species are present and the environment
does not promote scale formation, carbon steel may be susceptible to corrosion at pH 5.5. Even
at pH 12-14 (high alkalinity), corrosion of certain metals such as zinc or aluminum may occur.
Increasing the temperature usually increases the corrosion rate; the amount depends on what
factors control the cathodic reaction.
If the cathodic reaction is controlled by the availability of oxygen molecules at the metal
surface, which occurs through diffusion, the increase in corrosion rate with temperature is relatively
slow. If the cathodic reaction is controlled by ionic reduction, such as hydrogen ions to hydrogen
gas the increase in corrosion rate with temperature is relatively rapid. [Increasing temperature
also increases the opportunity for forms of localized corrosion such as pitting or stress corrosion
cracking.]
The synergistic effect of corrosive mixtures must also be considered. Combinations of chemi-
cals, which by themselves are relatively non corrosive, may be extremely aggressive towards specific
alloys. For example, injection streams containing dilute nitric acid or dilute flowing sodium chlo-
ride are usually not corrosive towards stainless steel. However, if the two streams are combined,
severe pitting of the stainless steel may result.
-------
L2
Eh
0.0-1
-i.O
Fe
,*3
CM
-f-
X
o
«£
Fe(OH),
Fe
Fe(OH)2
8
12
pH
Figure 2. Oxidation-Reduction Potential (Eh) and pH Diagram
Showing Stability of Iron Species and Phases at STP
-------
To summarize, a variety of factors will affect the corrosiveness of an injection well environment.
These include the characteristics of the alloy, the presence of aggressive species, the pH, the tem-
perature, oxygen concentration, and velocity or turbulence of flowing streams. It is also important
to know whether chemicals present in the injection stream increase or decrease corrosion.
Oxygen Corrosion (modified after EPA, 1982).
Oxygen dissolved in water causes rapid corrosion of certain metals (Allen and Roberts, 1978).
This effect is particularly pronounced with carbon steel. The effect of dissolved oxygen is realized
at the anodic area of the metal where the insoluble metallic hydroxide (rust) is precipitated. The
oxygen corrosion reaction for iron in the absence of other influencing constituents proceeds as
foUows (Ostroff, 1965):
Fe + 2H+ —> Fe+2 + H2 (5)
#2 + \02 —* H20 (6)
2Fe+2 + -O2 + H2O —> 2Fe+3 + 2OH~ (7)
£
The extent of corrosion is limited by the rate at which oxygen is delivered to the anodic area. In
a closed injection system (not in contact with air), the reaction will continue only until the dissolved
oxygen in the injection fluid is consumed (Warner and Lehr, 1977). In open systems, where air can
enter the injection fluid, corrosion continues as the oxygen supply is replenished. In general, with
increasing dissolved oxygen levels in the injection fluid the corrosion rate progresses until a point
is reached where the flow of oxygen becomes limited by the barrier of metallic hydroxide developed
on the anodic surface.
Oxygen corrosion is enhanced by the presence of dissolved chloride and sulfate ions. Generally,
corrosiveness increases with increasing salt concentration until a maximum is reached after which
corrosiveness decreases. The initial increase is due to the electrolyte conductivity increase, and
the subsequent decrease results from the decreased solubility of oxygen in the electrolyte (Uhlig,
1962).
Carbonate ions inhibit oxygen corrosion; it acts to counter the acceleration effect of chloride
and sufate ions in water containing dissolved oxygen. The degree of inhibition is dependent on the
relative concentrations of the chloride and sulfate salts and carbonate alkalinity. When calcium is
associated with the carbonates, there is further capacity for protective action (AWWA, 1971).
Again, the rate of corrosion because of dissolved oxygen generally increases with increasing
temperature. Higher temperatures allow more oxygen to diffuse to the steel surface. However,
when the temperature becomes higher than 90°C, oxygen solubility decreases, and the corrosion
rate can be expected to decrease even in open systems. In closed systems, the oxygen cannot
escape and the corrosion rate continues to increase with increasing temperature until all oxygen
has been depleted (Ostroff, 1965).
The velocity of the injection fluid can also affect the rate of corrosion. As the velocity increases,
the transport of oxygen to the metal surface becomes faster and consequently the corrosion rate
increases. Also, increases in the velocity of the injection fluid can cause mechanical scouring of
corrosion products, thus removing the protective film. Localized velocity increases can occur at
elbows or at other protuberances. Injection-fluid pH also affects the rate of corrosion of solutions
containing oxygen. For carbon steel, the corrosiveness of a fluid generally increases as pH decreases.
In the pH range of 4 to 9.5, the steel surface is coated by protective corrosion reaction products,
and corrosion progresses at a slower rate as oxygen diffuses through this layer. Below a pH of 4,
the corrosion products dissolve, and consequently, more rapid corrosion ensues (Ostroff, 1965).
-------
Oxygen corrosion causes pits in the metal surface. The pit develops at a localized anodic
point and continues by virtue of a large cathodic area surrounding the anode (AWWA, 1971). Pits
may be either sharp and deep, or shallow and broad. Additionally, a corrosion product may form
over these pitted areas. A common form of oxygen corrosion in wells is the formation of nodules
of altered material called tubercules. These are caused by the aggregation of iron bacteria, mixed
carbonates and hydrated metal oxides.
A special type of oxygen corrosion is caused by differential aeration cells, a result of differences
in oxygen concentration between two parts of a system. Differences in dissolved oxygen concentra-
tions cause differences in the solution potential of the same metal. For these cases corrosion occurs
at the area of the metal where oxygen concentrations are low. An adjacent area of relatively higher
oxygen concentration serves as the cathode in the reaction.
Corrosion products, chemical precipitates, or other debris on a metal surface hinder oxygen
diffusion by covering the metal in local areas. These circumstances can result in high localized
oxygen concentrations with corrosion taking place under the deposit.
The growth of microorganisms in injection wells can also result in the formation of localized
oxygen concentration on parts of the metal surface (Ostroff, 1965). Corrosion can therefore occur
around areas of tubing or casing covered by slimes or masses of bacterial growth (See Microbiolog-
ically Influenced Corrosion).
Carbon-Dioxide Corrosion (modified from EPA, 1982).
Carbon dioxide dissolved in water can contribute to the corrosion of steel, but at equal con-
centrations it is much less corrosive than oxygen (Ostroff, 1965). Carbon-dioxide corrosion of well
components is of particular concern in enhanced-oil-recovery (EOR) operations involving carbon
dioxide miscible injection systems. In these operations, carbon dioxide is injected before, after, or
alternately with water.
When dissolved in water, carbon dioxide forms carbonic acid:
CO2 + H2O —> H-iCOz (Carbonic Acid] (8)
This carbonic acid causes a reduction in the pH of the water which makes it quite corrosive to
steel (API, 1958):
Fe + H2CO3 —> HZ + FeCOz (Iron Carbonate Corrosion Product) (9)
The acidity of the solution, and therefore, the corrosion rate, is influenced by the partial
pressure of carbon dioxide. The partial pressure of a gas refers to the fraction it contributes to
the overall pressure of all the gases in the mixture. At higher pressures, more carbon dioxide will
dissolve in water creating a stronger acid. If the partial pressure values are above 30 psi, or in
metric units 2.1 X 10s Newtons per square meter (-^T) - the well stream is probably corrosive;
seven to 30 psi (4.8 x 104 to 2.1 x 105 ^-) may be corrosive, and 0 to 7 psi (0 to 4.8 x 104 ^) is
noncorrosive (API, 1958). In most injection well environments, carbon dioxide is miscible with the
fluid since pressures rarely fall below 1,200 psi (8.3 x 106 ^y) [critical pressure of carbon dioxide
is 1180 psi] (Allen and Roberts, 1978).
The rate and amount of corrosion caused by dissolved carbon dioxide are also dependent on
the oxygen content, the salts dissolved in the water, the temperatures, and the fluid velocities.
Water containing both dissolved oxygen and carbon dioxide is more corrosive to steel than water
which contains only an equal concentration of one of these gases (Ostroff, 1965). In waters con-
taining magnesium and calcium bicarbonates, an increase in temperature can cause the evolution
-------
of carbon dioxide that results in increased corrosion. At the same time, carbonates of these salts
can precipitate out on the metal surface, resulting in the formation of a protective coating, which
may reduce corrosion rates. As with other types of corrosion, higher than normal fluid velocities
can cause erosion of corrosion products (that normally retard the corrosion reaction), allowing
corrosion to continue unabated.
Carbon-dioxide corrosion may appear as a uniformly thinned metal surface or as rounded
non-uniform pits. Surfaces constantly bathed in dissolved carbon dioxide solutions will .tend to
exhibit uniform thinning, whereas pitting is caused by carbon dioxide dissolved in droplets of
water condensed on the injection-tubing wall (API, 1958).
Hydrogen-Sulfide Corrosion (modified after EPA, 1982).
Hydrogen sulfide gas dissolved in water, even in small amounts, can create a very corrosive
environment (Allen and Roberts, 1978). Dissolved hydrogen sulfide forms a weak acid and in the
absence of oxygen will attack iron and non-acid resistant alloys (Warner and Lehr, 1977); moreover,
it becomes severely corrosive to acid-resistant alloys when oxygen is present. Hydrogen sulfide is
often present in oil field production brines that are subsequently disposed by well injection. This
practice has resulted in instances of severe corrosion in injection tubing, especially when the brines
become contaminated with oxygen during surface handling (API, 1958). The general mechanism
of this type of corrosion as it affects iron and steel is stated as follows (API, 1958):
Fe° + xH2S = FeSx (Iron Sulfides) + xH2 (10)
Other metals react in the same manner to produce metallic sulfides. The corrosion rate in water
containing hydrogen sulfide is also influenced by the presence of dissolved salts and dissolved
carbon dioxide (Ostroff, 1965); when these substances are present, hydrogen sulfide corrosion rates
increase.
Hydrogen-sulfide corrosion of steel or iron results in the deposition of black scale (iron sulfide)
on the metal surface. The scale tends to cause a local acceleration of corrosion because steel is
anodic to the iron sulfide (Allen and Roberts, 1978). This reaction results in deep pits in the metal.
Cracking, in high strength steels, is due to embrittlement caused by atomic hydrogen formed in the
corrosion process which interferes with the ability of the steel to yield under stress. In low strength
steels, atomic hydrogen diffuses into the steel where it combines to form molecular hydrogen. These
trapped molecules cause the steel to blister which may lead to premature failure.
Microbiologically Influenced Corrosion.
Microbiologically influenced corrosion is an insidious form of localized corrosion in natural
waters, which occurs in crevices formed under deposits that harbor microorganisms. Failure occurs
by wall penetration. Since many injection streams will kill or inhibit microorganism growth,
microbiologically influenced corrosion problems are unlikely to be internal to the injection string.
Microorganisms are generally not corrosive until they settle on the metal surface. Microbio-
logically produced hydrogen sulfide is an exception to this rule, since it can cause corrosion remote
from the deposit. Settlement of microorganisms can be aided by tubercule or slime-forming bacte-
ria and also by surface irregularities or roughness on the metal. Once settled, the microorganisms
(and also tubercule or slime-forming bacteria) may accelerate corrosion in several ways, as shown
below:
° the metabolism and growth of the microorganism colony may modify or accelerate the cathodic
reaction, which in turn accelerates the anodic reaction of metal loss
-------
° the deposit itself creates an oxygen concentration cell, which increases the likelihood of crevice
corrosion
0 the deposit tends to concentrate ions such as chloride or manganese from the environment,
which increases the likelihood of localized corrosion
Microorganisms are usually corrosive if they are allowed to settle. Two examples of such microor-
ganisms are the anaerobic sulfate-reducing types and the aerobic iron-oxidizing types, which are
extremely corrosive towards carbon steel and stainless steel, respectively.
Biocides are useful in killing planktonic (floating) microorganisms in contaminated waters.
However, the effectiveness of biocides in killing sessile (attached) microorganisms is unproven.
The tubercule or slime that harbors the microorganism forms a protective layer that inhibits the
effectiveness of the biocide.
Galvanic Corrosion (modified after EPA, 1982 and E.E. Johnson, Inc., 1975).
Galvanic corrosion occurs where two different metals or alloys come in contact under specific
conditions. Almost all metals have different solution potentials, so that when the two metals come
in contact, this difference in potential results in current flow through the electrolyte. This is an
important consideration for brine disposal wells since saltwater is an excellent electrolyte. The
severity of galvanic corrosion depends upon the difference in potential between the two metals,
and the relative size of the cathode and anode areas. The galvanic series for metals and alloys
in sea water are shown in Table 1. Overall, this table is applicable to high TDS fluids but not
necessarily to acidic fluids. This table is a useful guide to indicate whether corrosion can occur
between two metals; however, other factors are equally important, so that separation in the series
is not the prime indicator of whether corrosion is eminent. The more active metals are found at the
top of the series. When two metals are electrically coupled, the more active metal is susceptible to
corrosion and the less active will be protected.
If the area of the active metal is significantly smaller than the area of the noble metal, sub-
stantial corrosion of the active metal may occur. In some cases the noble metal may be enbrittled
by the formation of hydrogen at the cathodic end.
On the other hand, if the area of the active metal is very large compared with the area of the
less active metal, corrosion will not be so severe.
Most of the galvanic corrosion is confined to the interface between the two metals. The
conductivity of the solution determines how the corrosion is spread over the active metal. Figure 3
shows how conductivity affects the extent of galvanic corrosion. In some cases, similar metals may
develop a potential difference because of their different metallurgical histories. Figure 4 shows one
such case.
Non-metallic Degradation.
Non-metallic construction materials include cementitious materials such as ceramics and plas-
tics. The discussion will be confined to plastics. Plastics are of two types: thermoplastics and
thermosets. Both types may be reinforced or filled with materials such as glass or graphite to
enhance their strength, lubricity, etc.
When non-metallic materials are exposed to a hostile environment, they may degrade. The
degradation of non-metallic materials can take several forms, which include blistering, crazing,
swelling, softening, and delamination. In most cases, the result of degradation is the loss of
mechanical properties leading to well failure. Table 2 gives a qualitative evaluation of the chemical
effects of certain organic solvents and mineral acids on specific plastics.
-------
TABLE 1
GALVANIC SERIES FOR METALS IN SEA WATER
(Jellinek, 1958)
Active or Anodic End
Magnesium
Zinc
Alclad 3S
Aluminum 3S
Aluminum 61S
Aluminum 63S
Aluminum 52
Carbon Steel
Alloy Steel
Cast Iron
Stainless Steels (active)
Type 4-10
Type 430
Type 304
Type 316
Ni-Resist
Muntz Metal
Yellow Brass
Admiralty Brass
Aluminum Brass
Red Brass
Copper
Aluminum Bronze
Compostition G Bronze
90/10 Copper-Nickel
70 + 30 Copper-Nickel-Low Iron
Nickel
Iconel
Silver
Stainless Steels (passive)
Type 410
Type 430
Type 304
Type 316
Monel
Hastelloy C.
Titanium
Noble or Cathodic End
10
-------
METAL
A. LOW CONDUCTIVITY SOLUTION
MORE NOBLE
LESS NOBLE
METAL
B, HIGH CONDUCTIVITY SOLUTION
MORE NOBLE
LESS NOBLE
Figure 3. The Effect of Solution Conductivity on
Galvanic Corrosion
a) Low Solution Conductivity
b) High Solution Conductivity
11
-------
M
)r
M Old pipe
'(Cathode) — ~
New pipe
— (Anodic) — -
'
Old pine J
tHI~ (Cathodic) -^
\}r)
Figure 4. New Pipe Replacing a Section in an Older Pipeline
Often is Anodic and Corrodes Faster than the Old Pipe
which is Partly Protected by Previously Formed
Coatings of Rust
12
-------
TABLE 2
CHEMICAL RESISTANCE OF SOME COMMERCIAL PLASTICS
Chemical Class
Cellulose Acetate Butyrate
Nylon
Polycarbonates
Cellulose Acetate
Resin Type
Subclass or Modification
Chemical Resistance
Mineral acids, weak
Mineral acids, strong
Oxidizing acids, concentrated
Alkalines, weak
Alkalies, strong
Alcohols
Ketones
Esters
Hydrocarbons, aliphatic
Hydrocarbons, aromatic
Oils, vegetable, animal, mineral
Chemical Class
Resin Type
Subclass or Modification
Chemical Resistance
Mineral acids, weak
Mineral acids, strong
Oxidizing acids, concentrated
Alkalies, weak
Alkalines, strong
Alcohols
Ketones
Esters
Hydrocarbons, aliphatic
Hydrocarbons, aromatic
Oils, vegetable, animal, mineral
Thermoplastic
Hard
Good
Fair to Good
Good
Poor
Poor
Poor
Poor
Fair to Good
Poor
Good
Polyethylene
Thermoplastic
Lou Density
Good
Good
Good to Poor
Good
Good
Excellent to Poor
Excellent to Poor
Excellent to Poor
Fair
Fair
Good
Thermoplastic
6/6
Very Good
Poor
Poor
No effect
No effect
Good
Good
Good
Very Good
Fair to good
Good
Polyethylene
Thermoplastic
Medium Density
Excellent
Excellent
Good to Poor
Excellent
Excellent
Excellent to Poor
Excellent to Poor
Excellent to Poor
Fair
Good
Excellent
Thermoplastic
Unfilled
Excellent
Fair
Poor
Poor
Poor
Poor
Poor
Poor
Poor
Poor
Polyethylene
Thermoplastic
High Density
Excellent
Excellent
Good to Poor
Excellent
Excellent
Excellent to Poor
Excellent to Poor
Excellent to Poor
Fair
Fair
Good
Thermoplastic
Soft
Fair to Good
Poor
Very Poor
Poor
Very Poor
Poor
Poor
Poor
Fair to Poor
Poor to fair
Fair to good
Methylmethacrylate
Thermoplastic
Unmodified
Good
Fair to Poor
Attacked
Good
Poor
Dissolves
Dissolves
Good
Softens
Good
13
-------
Chemical Class Cellulose Acetate Sutyrate
Resin Type
Subclass or Modification
Chemical Resistance
Mineral acids, weak
Mineral acids, strong
Oxidizing acids, concentrated
Alkalines, weak
Alkalies, strong
Alcohols
Ketones
Esters
Hydrocarbons, aliphatic
Hydrocarbons, aromatic
Oils, vegetable, animal, mineral
Chemical Class
Resin Type
Subclass or Modification
Chemical Resistance
Mineral acids, weak
Mineral acids, strong
Oxidizing acids, concentrated
Alkalies, weak
Alkalines, strong
Alcohols
Ketones
Esters
Hydrocarbons, aliphatic
Hydrocarbons aromatic
Oils, vegetable, animal, mineral
Thermoplastic
Hard
Fair to Good
Poor
Very Poor
Poor
Very Poor
Poor
Poor
Poor
Fair to Good
Poor to Fair
Fair to good
Polysterene
Thermoplastic
Unmodified
Excellent
Excellent
Poor
Excellent
Excellent
Excellent
Dissolves
Poor
Poor
Dissolves
Fair to Poor
TABLE 2
(continued)
Cellulose Acetate
Thermoplastic
Soft
Good
Fair to Good
Good
Poor
Poor
Poor
Poor
Fair to Good
Poor
Good
Polystyrene
acrylonitrile
Thermoplastic
Unmodified
Excellent
Good to Excellent
Poor
Excellent
Good to Excellent
Good to Excellent
Dissolves
Dissolves
Good
Fair to Good
Good to Excellent
Thermoplastic
Unmodified
Excellent
Excellent
Good to Poor
Excellent to Good
Excellent to Good
Excellent to Good
Excellent to Good
Excellent to Good
Good to Fair
Good to Fair
Good
Polyethylene
chloroethylene
Thermoplastic
Unmodified
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Good
Thermoplastic
Copolymer
Excellent
Excellent
Excellent
Excellent
Good
Good below 80°C
Good below 80°C
Good below 80°C
Good below 80°C
Good below 80°C
Polytrifluoro-
ethylene
Thermoplastic
Unmodified
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Chemical Class
Resin Type
Subclass or Modification
Cellulose Acetate Butyrate
Thermoplastic
Unmodified, Rigid
Cellulose Acetate
Thermoplastic
Plasticized (nonrigid)
Chemical Resistance
Mineral acids, weak
Mineral acids, strong
Oxidizing acids, concentrated
Alkalines, weak
Alkalies, strong
Alcohols
Ketones
Esters
Hydrocarbons, aliphatic
Hydrocarbons, aromatic
Oils, vegetables, animal, mineral
(After: The Chemical Rubber Company, 1978)
Excellent
Good to Excellent
Fair to Good
Excellent
Good
Excellent
Poor
Poor
Excellent
Poor
Excellent
Fair to Good
Fair to Good
Poor to Fair
Fair to Good
Fair to Good
Fair
Poor
Poor
Poor
Poor
Poor
14
-------
One method of checking the applicability of non-metallic materials is to remove samples at
regular intervals during exposure and to measure the loss of mechanical properties, such as flexural
strength. If the loss levels off to an acceptable value, the material is usually considered suitable.
15
-------
II. INORGANIC INCRUSTATION
(SCALE-FORMING TENDENCY)
Inorganic scaling corrosion can occur as a result of incompatibility between injection fluids
and the naturally occurring fluids in an injection zone. Persons and Hart (1980) have discussed
calcium carbonate scaling in Class V, heat pump wells. This discussion can also be applied to
other types of injection wells. Calcium carbonate scaling occurs as a result of lowering carbon
dioxide (COg) solubility in solutions containing dissolved calcium carbonate (CaCOz}', this is
because CO^ is in equilibrium with CaCOs and a change in pressure or temperature results in a
shift in the equilibrium. In readjusting to the new equilibrium, CaCOz may be precipitated when
the temperature is raised and/or when the pressure is reduced. Figure 5 shows the relationship
between CO2 solubility and pressure and temperature in water. Other inorganic compounds show
similar behavior to calcium carbonate. Warner and Lehr (1977) discussed methods of predicting the
solubilities of some chemical constituents of injected wastewater through calculation of a stability
index. A discussion of the work of Warner and Lehr (1977), relative to chemical stability, is
presented below.
Chemical Stability.
The following text has been extracted and modified from Warner and Lehr (1977).
Stability of the chemical compounds in the injected wastewater is desirable. An unstable
compound may precipitate during or after injection and cause plugging. The influence of pressure,
temperature and pH change in initiating instability have been individually pointed out, but have
not been quantified. Also, these factors can act simultaneously and/or synergistically - making
interpretation difficult.
A means of anticipating instability in a system affected by more than one variable is through
use of a saturation or stability index. Several such indices have been developed including those
by Langelier (1936), Ryznar (1944), Larson and Buswell (1942), and Stiff and Davis (1952). The
first three indices are applicable to waters of low ionic strengths, while the Stiff and Davis index is
intended for use with concentrated solutions, such as highly saline ground waters. As an example
of the use of such indices, the Stiff and Davis (1952) stability index for calcium carbonate is:
S I = pH - K - pCa - pAlk (11)
In equation 11, K is an empirical constant used to compensate for various ionic strengths and
temperatures. The values of K, pCa, and pAlk are taken from graphs (Figures 6 and 7). A positive
index indicates scale formation and a negative index indicates corrosiveness by the injection fluids.
The following example of the use of the stability index was taken from Ostroff (1965), and
modified.
From the water analysis in the first column of Table 3, the concentration C of each ion in
moles per 1,000 grams of water (molality) was calculated using the relationship:
where epm = concentration of the ion, equivalents per million
Z = valence of the ion
SpGr = specific gravity of the brine = 1.06
TDS = total dissolved solids, ppm = 152, 474 —
16
-------
The effect of temperature on
o
o
f\J
o
o
to
_>
15
05
DC.
Temperature
The effect of pressure on CQ2
CO
J3
Is:
Pressure
Figure 5. The Effects of Pressure and Temperature on
Carbon Dioxide Solubility (Modified after
Persons and Hart, 1980)
17
-------
40-i
IONIC STRENGTH,
Figure 6. Values of K at Various Ionic Strengths (from Warner and
Lehr, 1977; after Stiff and Davis, 1952)
18
-------
Ca HCOJ
10
15 20
pCa, pAlk
25
30
35
:e 7. Graph for Converting Concentrations of Calcium
and Alkalinity (in ppm) to pCa and pAlk.
pCa = -log [Ca], pAlk = -log[Alk],
Alk = [OH~] + [MC03~] 4- 2[C032-] - [H~]
(Ibid.Figure 5 )
19
-------
TABLE 3
¥ATER ANALYSIS USED IN SAMPLE CALCULATION OF CALCIUM CARBONATE SATURATION
USING STIFF AND DAVIS INDEX
(Ostroff, 1965)
Component mg/1
Carbonate (C03)-2
Bicarbonate (HCOj)- 4-6
Sulfate (S04.)~2 7,530
Chloride Cl~ 88,300
Iron Fe+2 14
Calcium Ca+2 8,570
Magnesium Mg+2 2,819
Sodium Na+ 45,195
TOTAL 152,474
epm
0.8
157
2,487
0.5
428
232
1 ,965
Molality (C)
0.001
0.086
2.740
0 .000
0.236
0.128
2.165
20
-------
This gives the results shown in the third column of Table 3. Using these molalities, the ionic
strength u is calculated.
«= ^C1(Z1)2 + C2(Z2)2.-.Cn(^)2] (13)
= |[(0.001)(1) + (0.086)(4) + (2.740)(1) + (0.236)(4) + (0.128)(4) + (2.165)(1)]
= 3.35
Then, from Figure 6, at u = 3.35 and T — 77°F(25°C} (average temperature of the fluid), the
value of K is found to be 2.96.
The next step is to enter the concentrations from Table 3 of Ca2+ (8,570 ppm) and
(46 ppm) as the ordinate of Figure 7. Reading the abscissa, the pCa is 0.67 and the pAlk is 3.12
(pH is given as 6.1).
Substituting in Equation 11, the stability index is calculated as follows:
SI-pH-K- pCa - pAlk = 6.1 - 2.96 - 0.67 - 3.12 = -0.65 (14)
This indicates that the fluid is corrosive and undersaturated with respect to calcium carbonate,
meaning that no scale formation is expected.
OstrofF (1965) discusses the stability of magnesium carbonate, magnesium hydroxide, calcium
sulfate, barium sulfate, iron, and also silica. Barnes (1969) suggests a thermodynamic approach to
predicting the stability of inorganic compounds in solution. Dissolved organic compounds may or
may not be unstable. Selm and Hulse (1960) list polymerization of organic chemicals as a source
of precipitates which can cause plugging. In addition to the predictive methods of analyzing for
chemical stability, empirical methods may also be used. The fluid in question should be subjected
to the pressure and temperature conditions encountered in the subsurface, and then observed to
evaluate its tendency to form precipitates over an extended period of time.
21
-------
III. FAILURE DUE TO CORROSION - REPRESENTATIVE EXAMPLES
Brine Disposal.
Early in the history of brine disposal wells, it became apparent that steel casing would readily
corrode in the presence of strong electrolyte solutions. One of the methods adopted to prevent
corrosion in brine disposal wells was to use wood stave pipe. Although the wood pipe was resistant
to internal corrosion, in many cases, the steel bindings would corrode due to constituents in forma-
tion fluids (Holloway and McSpadden, 1961). Asbestos-cement pipe was used at one time because
of its resistance to electrolytic corrosion. However, Holloway and McSpadden (1961) caution that
this type of pipe is subject to massive scaling problems.
Case 1. In the state of North Dakota, there was a corrosion problem in brine disposal wells
— and, also one in injection wells that used produced water. The produced water contained
sodium chloride concentrations that ranged from 100,000 ppm to over 200,000 ppm which were
very corrosive to the steel tubing. The brine would eventually corrode through the tubing; where
this occurred the tubing, sometimes, had to be replaced hi only several months. Rarely, was the
tubing effective for a period of two years.
The development of a new tubing has aided operators in all but eliminating the problem. The
operators, presently, use fiberglass tubing or internally coated steel tubing. The latter tubing is
used when high temperatures and pressures are encountered.
Case 2. In a Great Lake's state, an oil field contained very corrosive brine (the injection wells
are of carbon steel). The oil company which operated the field experienced 76 well failures over
the course of one year. Of these, 19 failures occurred in wells that had already experienced one
failure; and, three failures happened in wells which had already experienced two failures during
the year. In most cases, the problem was a tubing leak that developed because the brine was so
corrosive. The company tried mixing an additive to the injection fluid, but failures still took place
with the same frequency. The company continues to replace tubing joints as necessary.
However, one possible solution to the problem could be to replace the materials used in the
tubing for fiberglass, or if this is not structurally feasible — then a ferric type tubing with a
protective synthetic coating could be employed. Another possible solution, though expensive,
would be pretreatment of the waste.
Hazardous Waste Disposal.
Different types of wastes are injected hi commercial facilities. Because of this, the potential
for operational problems is higher than for on-site or non-commercial facilities which usually inject
wastes within a certain type range.
Probably due to the problems associated with the diversity of wastes injected and to the
inadequate training of operators at one commercial injection site, up to 45 million gallons of
hazardous wastes were released into an unauthorized zone. The wells in question had fluid seals in
lieu of mechanical packers in the casing tubing annulus. These systems are complex to operate as
the injection pressure, pressure in the annulus, temperature, specific gravity, injection occurrences
and other parameters determine where the fluid seal (interface) is located. The operator had to
add significant amounts of fluid to the annulus to maintain the desired operating conditions. The
fluid was probably flowing out of the casing through holes created by the corrosive action of the
waste. Along with the annulus fluid, large volumes of hazardous waste were also released through
the holes caused by corrosion.
A logical scenario can be reconstructed from the plant record as follows:
22
-------
0 Inadequate monitoring caused the interface of the annulus and injection fluid to rise to a
location in the annulus which was succeptible to corrosion.
0 One or a series of long episodes of subjecting the unprotected casing to the effects of the
corrosive waste caused perforations which allowed the fluid to escape.
0 Lack of corrosion monitoring and indiscriminate injection of many types of wastes made it
difficult for the operator to know that there was a problem.
0 The operator started adding small volumes of annulus fluid to keep the well within the range
of operational parameters. As time passed, the holes in the casing became larger. However,
due to indequate instrumentation, the gradual change in annulus fluid requirements, and
inadequate operator training — the changes went unnoticed and/or the warnings were not
heeded.
° The fluids interface stabilized as the operator found a pressure at which annulus fluid addition
was minimized. Unfortunately, the interface was located at a place which was vulnerable to
further corrosion.
0 Mechanical integrity tests were carried out and damage was discovered.
It is obvious that human error played a large role in this mishap. However, a corrosion
monitoring system would have provided advance warning as to the effects of the injection fluids.
Periodic evaluation keyed by the corrosion monitoring system, would also have minimized the
magnitude of the release.
Inorganic Incrustation in an Injection Well.
Keech (1982) reported an occurrence of inorganic iron and calcium carbonate incrustation
on a well screen in a ground-water heat pump return well which was so intense that backed up
water in the well flowed 70 feet above static water level. After cleaning the well, it was kept under
constant pressure so that the water in the well was constantly moving. This apparently prevented
scale precipitation on the screen from stagnant water.
Dilute Organic Acid and Stainless Steel (Moniz, 1986).
Type 304L stainless steel injection tubing had been in service for two years handling a dilute
organic waste, containing 1% nitric acid. The downhole temperature was 140°F. The annular
space between the injection string and the tubing contained 10% inhibited sodium chloride brine.
The injection tubing developed leaks at the couplings (type 304L stainless steel). Disassembly
of the injection tubing revealed that the joint surfaces exhibited various degrees of localized corro-
sion or pitting in the threads. In severe cases, the localized corrosion had caused wall penetration.
The pitting was caused by chloride ions from the brine and the nitric acid from the waste
mixing that formed a highly corrosive environment for the stainless steel. Pitting was augmented
by temperature and the geometry of the joint, which favored the development of crevice conditions
and prevented the corrosion from being washed out. Pipe dope in the threads did not provide an
adequate seal to prevent the mixing of the two chemicals. A remedy was to replace the annular
space sodium chloride brine with sodium nitrite, which does not cause similar corrosion problems
when mixed with the waste stream.
23
-------
IV. DETECTION AND MEASUREMENT OF CORROSION
(MODIFIED AFTER EPA, 1982)
Tubing and casing materials should be compatible with the injection operation, as well as the
fluid to be injected, and the environment in which the well is constructed. To determine proper
construction materials, it may be desirable to first test the corrosiveness of the injection fluid in
the laboratory. Despite the consideration given to corrosion control during well design, there often
is a need to recognize corrosive environments during well construction and to detect and measure
corrosion in an operating injection well.
Before initiating a corrosion-prevention program, it is necessary to determine if corrosion will
occur, the cause of corrosion, and the rate and severity of corrosion. To determine the effectiveness
of a corrosion-prevention program, the rate and effects of corrosion should be measured before and
after application of prevention measures. The following are descriptions of five commonly used
methods of detecting and measuring corrosion:
Weight-loss Coupons and. Corrosion Loops.
The most common of all corrosion rate measurement tests involves exposing pieces of metal
similar to those in the injection system to the corrosive environment. Small, preweighed and
measured coupons made of different metals are exposed to well fluids for a defined period of time,
removed, cleaned, and weighed to determine the corrosion rate (Allen and Roberts, 1978). Down-
hole coupon installation can be made by using standard wireline equipment. Corrosion rates are
usually measured in mils per year (mpy) penetration or metal loss. A low corrosion rate may
not be acceptable if localized corrosion (such as pitting) is occurring, whereas, a high rate with a
general area type of metal loss may be, in certain cases a relatively insignificant problem.
The visual appearance of the coupon after exposure may indicate the type and cause of
corrosion (Ostroff, 1965). For example, a black sulfide coating shows the presence of hydrogen
sulfide in the system. Ferric oxide indicates oxygen is present, and carbon dioxide corrosion can
be detected by ferrous carbonate deposits. An example of how a coupon test can be used to
evaluate corrosion rates of various metals in a saltwater injection well where hydrogen sulfide
containing brines are being disposed is provided in Table 4. It should be noted that the list of
alloy compositions is not comprehensive and the condition rates are specific to the conditions of
the exposure described. With the exception of a few metals and alloys of changeless performance,
any indicated usage should be correlated in detail with all related corrosion data.
Weight loss coupon tests are only comparative. The difference in the size and thermome-
chanical history of a coupon compared with actual items of equipment means that the corrosion
rate measured on a coupon rarely matches what is obtained in equipment. Nevertheless, coupons
provide the most useful guide to corrosion, particularly localized corrosion effects. [When suitably
fabricated and exposed, coupons predict general corrosion, crevice corrosion, pitting, stress corro-
sion cracking, embrittlement, galvanic corrosion and metallurgical structure related corrosion.]
At one time, coupons were widely used and considered to be the best method for estimating
internal corrosion, especially in oil field production and injection well operations. The principal
disadvantages of the coupon test method are the time required to obtain results and the limited
area tested, i.e., only at the point of installation (Allen and Roberts, 1978).
Another method of determining the corrosion potential of injection fluids is the use of a
"corrosion loop". A corrosion loop is a section of casing which is valved so that some of the
injection stream is passed through a small pipe running parallel to the injection pipe at the surface
of the well. The composition of the pipe is the same as the well casing. The only differences are
24
-------
TABLE 4
ILLUSTRATION OF CORROSION RATES OBTAINED FROM
WEIGHT LOSS COUPON TESTING IN
"SOUR" (HYDROGEN SULFIDE CONTAINING) SALTWATER
(Gulf Oil Corporation, 1948)
Metal or Alloy
Corrosion Rate*
(Mils per year)
Nickel
K Monel
Nickel Plated Steel
Antimonial Admiralty
Type 516 18-8 (Mo)
Aluminum 6061 -T6
Type 304 18-8
Type 347 18-8 (Cb)
70-30 Copper Nickel (70£ Cu, 30* Ni )
Carbon Steel J-55
Carbon Steel N-80
Alclad
Croloy 2-1 /4
Galvanized Steel
Croloy 5
9* Nickel
Copper Steel (0.26 Cu)
Yoloy (2 Ni, 1 Cu) •
5% Nickel
12 Chrome Cast
3# Nickel
0-40 Carbon Cast
Croloy 9
Carbon Steel H-40
Croloy 12
Corten (0.48 Ni , 1.04 Cr , 0.41 Cu)
Ampco Grade 8 (88 Cu, 10 Al , 1 Fe )
Cr-Mo-Si Steel (2.09 Cr , 0.56 Mo, 1.17 Si)
Everdur 1010
Copper Plated Steel
Red Brass Alloy 24 (85 Cu, 15 Zn)
Copper
1 .0
" 1.1
2.8
3.2
5-5
6.9
10.2
10.8
14.0
15-6
16.0
16.2
17.8
23.3
23.4
25.4
25-8
25-9
27.3
28.7
29.0
29.7
30.1
32.6
33.5
35.6
36.0
38.1
62.2
64.6
67.1
107.8
Corrosion rates of insulated coupons 4.5 by 1.5 inches in SWD
line, Darst Creek Field, Texas, 60 days exposure.
Corrosion rates are average of 4 coupons. Saltwater tests:
pH-7; H2S-200 mg/1; total solids-26,000 mg/1;
temperature-120°F; velocity-2 ft/sec.
25
-------
that the corrosion loop pipe has a smaller diameter than the well casing and the temperature is
lower. The corrosion loop is essentially a small scale version of the well. The corrosion loop can
be removed from the system upon closing the appropriate valves. The pipe can then be analyzed
for corrosion. If corrosion is observed in the small scale casing, corrosion is probably occurring in
the well casing.
One well known limitation of the loop is that the environmental conditions inside the well
are different than in the corrosion loop, and it may be more or less subject to corrosion than the
well. For example, the temperature in the corrosion loop is usually lower than inside the injection
string since temperature usually increases with depth. This may lead to incorrect predictions for
corrosion loop monitoring. Also, the effect of corrosion-accelerating features, such as joints, may
not be adequately represented in a corrosion loop.
Electrical-Resistance Probes.
Electrical-resistance corrosion probes are based on an adaption of the Wheatstone bridge, and
measure changes in electrical resistance of a metal speciman as it corrodes (Ostroff, 1965). The
Wheatstone bridge circuit provides a simple but accurate means of measuring electrical resistance.
The basic components of a Wheatstone bridge are a power source and three resistors which are
calibrated before the test. The resistance of the system is introduced into the circuit. By measuring
an induced current, the unknown resistance can be calculated. Probes are available in a variety of
sizes, thicknesses, metals, and alloys.
The probe is attached to a portable galvanometer. Several probes can be monitored at con-
venient time intervals with one instrument; this is particularly valuable when it is necessary to
measure corrosion rates at different points within the system at the same time.
The electrical-resistance probe has found its principal applications in injection systems involv-
ing gas streams because the probe does not have to be submerged in water to function. One of
the disadvantages is that it is usually limited to the measurement of uniform corrosion. The probe
can give misleading results if a deposit forms on it (Allen and Roberts, 1978). It is a sensitive and
delicate instrument not easily repaired, and it is difficult to operate by the untrained.
Polarization-Resistance Probes.
A polarization-resistance probe can be used to measure corrosion current and corrosion rate
because metal loss is directly proportional to current flowing from the test electrode (Allen and
Roberts, 1978). This instantaneous corrosion-rate meter has the capacity to detect very low rates
of uniform corrosion, and it can record data for multi-test points on a continuous basis. In addition,
some progress has been made in using corrosion-rate meters to predict pitting-type corrosion.
The polarization-resistance probe is particularly useful in studying the changes throughout an
injection well system caused by the introduction of corrosion inhibitors, air leaks, or other changes.
The test probes must be submerged in liquid and positioning must be done with care in a flowing
stream to avoid shadowing one electrode by another (Allen and Roberts, 1978). The electrodes
may experience short-circuiting resulting from corrosion products or solids in the injection fluid.
Well-logging methods.
Caliper surveys, electromagnetic pipe analysis survey (PAS) logs, casing potential logs, and
ultrasonic/radioactive-measurement logs are techniques commonly used for evaluating active cor-
rosion. Brief descriptions of these methods are provided below.
Caliper surveys are run to inspect the internal surface of tubing or casing. Mechanical feelers
contact the inside metal surface and will measure the diameter of the pipe and can detect the
metal loss due to pitting and thinning. A baseline caliper log should be run to provide the basis for
26
-------
comparison. Caution should be exercised when running calipers through coated tubing to prevent
pipe coating damage. In addition, caliper feelers may remove protective scales and allow corrosion
to occur in the feeler tracks.
Casing-thickness logs can be developed by using an instrument called a Pipe Analysis Survey
(PAS) tool, which relies on an electromagnetic field to measure the thickness of metal at any point
in the casing. This type of log can be used to calculate external metal loss when the loss of metal
on the inside of the casing has been measured with an internal caliper.
The following discussion of PAS tools has been taken from Nielsen and Aller (1984). The
PAS tool detects casing defects such as holes, gouges or cracks, by measuring fluctuations in an
induced magnetic field with coils mounted in small pads. The PAS is able to discriminate between
defects on the inside and outside of the casing wall by means of a high-frequency eddy current
test, which detects flaws on only the inner surface, and a magnetic flux leakage test, which inspects
the full casing thickness. An evaluation of the various nondestructive mechanical integrity testing
techniques indicates that a combination of high-frequency eddy current and magnetic flux leakage
tests provide an optimum approach for in-place inspection of well casings to detect small, isolated
defects or corroded areas and to determine whether they are located on the internal or external
casing wall.
Current flow in the well casing can be measured with a logging tool with two sets of contactor
knives. Polarity of the voltage reading between the two contacts indicates at any given point
whether current is flowing from the casing. Corrosion is indicated where current is leaving the
pipe. A casing potential log is the best approach to find active corrosion on the outside of the
casing and to show effectiveness of cathodic protection (Allen and Roberts, 1978).
Finally, ultrasonic or radioactive devices can be used to measure wall thickness and detect
thinning of metal. The principle limitations are that small pits may not be detected, and the
measurement is made only at one point.
Detection of Microbiologically Influenced Corrosion (Horacek, 1986).
Although it is easy to detect planktonic microorganism populations in water, the detection of
sessile microorganisms is an uncertain science. Furthermore, the population size of either type is not
necessarily an indication of their potential to induce corrosion. The diagnosis of microbiologically
influenced corrosion is extremely difficult, because the evidence of microorganism contamination
may have disappeared by the time the failure analysis is performed.
27
-------
V. CORROSION CONTROL
Corrosion can be minimized by the application of a number of different design considerations
and operating techniques which include: use of construction materials resistant to potential cor-
rosion by the designated injection fluid, protective coatings to the tubing and casing, provision in
the design of the well to allow for future application of corrosion inhibitors to the casing/tubing
annulus if the need arises and pretreatment of the injection stream.
Metals resistant to corrosion are available for virtually all corrosive environments encoun-
tered in injection operations. The problem with many of these corrosion-resistant metals is their
prohibitive cost. Iron and steel may corrode, but their lower cost, ease of manufacturing, and
strength have helped them become the most commonly used metals in injection operations (Allen
and Roberts, 1978). Other more resistant metals and alloys provide resistance to corrosion in
specific applications, but they cost significantly more than carbon steel.
The choice of metals to resist corrosion in a specific application is affected by the corrosive en-
vironment, as well as the physical requirements for the material. In a hydrogen sulfide environment,
the effect of hydrogen embrittlement on strength and durability of a metal is the primary concern
(Allen and Roberts, 1978). Carbon steels are resistant to sulfide-stress cracking as are other more
exotic alloys. Low-alloy steels, heat treated to high strength levels are not recommended because
of their greater tendency toward sulfide cracking (API, 1958). For carbon dioxide and oxygen
environments, where embrittlement is not a concern, metals should be selected based on control
of metal loss. These metals are typically more expensive alloys, such as stainless steel nickel-base
alloys or titanium; consequently, economics might dictate that other methods of corrosion control
be used. For corrosion resistance to most acids, compatible stainless steel is usually employed for
injection tubing strings. Tables of suitable metals and alloys for hydrogen sulfide, carbon dioxide,
and oxygen corrosion are provided in Table 5.
Downhole applications of nonmetallic corrosion-resistant materials are limited to certain types
of plastics. Other nonmetallic materials, such as asbestos-cement and ceramics, do not possess the
temperature resistance or toughness necessary for injection tubing. The most extensively used
plastic pipes and tubings are fiberglass pipes reinforced with epoxy resins (reinforced thermostatic
plastic, RTF) (Donaldson, 1972). The material is highly resistant to corrosive fluids, and it also
affords relatively good resistance to corrosive attack by acids and alkalies (Ostroff, 1965). PVC and
other plastic pipe also offer this corrosion-resistant capability, but have lower strength and tem-
perature ratings than reinforced fiberglass materials. A disadvantage of epoxy-reinforced fiberglass
and other plastic tubing materials is their relatively poor resistance to attack by organic solvents
and dissolved chlorine.
Protective Coatings.
Coatings prevent corrosion by removing or separating the corrosive environment from the
metal. Paints, plastics, cement, rubber, and ceramics have been used to provide such barriers
(EPA, 1982). In addition, some metal coatings, like zinc on steel, cathodically protect the base
metals.
Organic, inorganic, or metallic coatings are selected on the basis of the temperature, pressure,
and corrosiveness of the environment. A major problem with protective coatings is that a break in
the coating exposes the base metal to corrosion (Ostroff, 1965). These breaks are called "holidays."
Inorganic cement linings are also used extensively for tubing in wells handling brines (API,
1958; EPA, 1982). Cement linings are not recommended for use with highly acidic solutions.
Moreover, cement linings are permeable to water and corrosion products tend to form between the
28
-------
TABLE 5
SUITABILITY OF CASING AND TUBULAR GOODS
TO VARIOUS CORROSION ENVIRONMENTS
(Allen and Roberts, 1978)
ACCEPTABLE FOR HYDROGEN SULFIDE
1. Low and medium alloy carbon steels, <1 percent nickel, not finished
2. J-55- C075, N-80, SOO-90
3. 300 series stainless steel Annealed
4. Incoloy 800 (Ni-Cr-Fe)
5. Inconloy 825 (Ni, Fe, Cr, Mo)
6. Iconel 600 (Ni, Cr)
7. Inconel X-750 (Ni-Cr-Al)
8. Mibek 4-00 (Ni-Cu) Annealed
9- K-monel 500 (Ni-Cu-Mo)
10. Hastelloy C (Ni-Cr-Mo)
11. MP35N (Co-Ni, Cr, Mo)
12. Stellites (Co-Cr-V)
13- Colomonoys (Ni-Cr-B)
14. Cemented carbides (Tungsten Carbide)
UNACCEPTABLE FOR HYDROGEN SULFIDE
1. Low and medium alloy steels, >1 percent nickel or cold finished
2. Free machining steels, >0.08 percent sulfur
3. Stainless steel, cold finished or precipitation hardened
4. K-monel, cold finished
ACCEPTABLE FOR CARBON DIOXIDE
1. Stainless steels, except free machining.
2. Monels (Ni-Cu)
3. Nickel-iron (Ni-resist)
4. Al-bronze (Cu-Al)
ACCEPTABLE FOR OXYGEN
1. Stainless Steel
2. Monels
3- Nickel-iron
4. Al-bronze
29
-------
lining and the subsurface metal. The buildup of corrosion products can eventually lead to cracking
and sloughing of the lining (API, 1958).
For corrosion control in injection well operations, the only metallic coatings of importance are
zinc and aluminum on steel. They may be used on buried steel components where oxygen corrosion
is moderate, but their best application is for atmospheric exposure of surface equipment (Allen
and Roberts, 1978).
Protection of the Packer and Other Bottom-Hole Components.
The packer and other bottom-hole components may be especially vulnerable to corrosion.
Injecting of highly acidic solutions, such as pickle liquor, can cause operational and maintenance
problems to the operator.
In some cases coated metallic packers and plastic liners may reduce the corrosion of these
components. In other cases, operators have chosen to use "fluid seals" instead of packers to
minimize maintenance costs. These fluid seals are designed to isolate the tubing-casing annulus
by creating an equal-pressure boundary at or near the bottom of the tubing above the injection
perforations or by continuously pumping a corrosion inhibitor through the annulus. The liquid
used in the annulus is immiscible with the injection fluid in the former situation. In general,
operation of these wells is complex due to the temperature and pressure deformation of the tubing
that take place during operation and their impact on the annulus fluid pressure and the location
of the immiscible boundary.
Another solution is the placing of oil or other non-ionic liquids in the space between the
bottom of the tubing and the packer and/or in the whole upper portion of the injection cavity
above the perforations. The lower specific gravities of the non-ionic liquids and the geometry of the
bottom-hole components retain liquids between the injection fluid and the vulnerable components.
Protective liquid losses are replaced by the injection of 20 - 100 gallons of it every two or three
months. This solution has been proven very cost effective.
Preinjection Treatment.
Frequently, the removal of corrosive agents from injection fluid by preinjectipn treatment
methods can be the best means of corrosion control. The most common preinjection treatment
used involves degasification and/or neutralization.
Degasification
Degasification involves the complete removal of corrosive dissolved gases from water. The most
common method of degasifying water in preinjection treatment is to selectively remove dissolved
oxygen. Degasification is usually not cost effective for the prevention of corrosion due to acid
gases, carbon dioxide and hydrogen sulfide (Allen and Roberts, 1978). Oxygen degasification
can be accomplished by the use of chemical scavengers, vacuum deaeration, or counter-flow gas
stripping.
Chemical scavengers for oxygen removal are based on a chemical reaction between oxygen and
another chemical. A commonly used chemical is sodium sulfite which is particularly useful for
removing small amounts of oxygen from large volumes of water.
Dissolved oxygen is removed in the oxidation of sulfite to sulfate:
Na2SO3 + ^02 —> Na2S04 (15)
In practice, a concentration of 10 ppm of sodium sulfite is used to remove 1 ppm of dissolved
oxygen (Ostroff, 1965). Catalysts, such as cobalt chloride, are used to increase the rate of the
30
-------
reaction. The presence of hydrogen sulfide in the water reduces the effectiveness of sodium sulfite
to scavenge oxygen. Sulfate-reducing bacteria should also be inhibited from growing in water-
handling systems in which sulfite is used to scavenge oxygen.
The sulfite ion can also be formed in water by adding sulfur dioxide gas. Sulfur dioxide can
be added from bottled liquid containers or on-site gas generators. The bottled liquid is economical
for treating small systems with low oxygen concentrations. The use of sulfur dioxide for oxygen
removal has the potential disadvantage of producing corrosive acids in solution and also of creating
barium or calcium scales that may plug the injection formation.
Dunlop and others (1969) recommend the use of ODASA, the oleyl diamine adduct of SO"2,
as a 40 percent solution in methanol, as a substitute for sodium sulfite when mixing and freezing
difficulties are encountered. A concentration of 25 ppm ODASA is necessary to scavenge 1 ppm of
dissolved oxygen.
Oxygen can be removed from water by running it through a vacuum in a packed tower. The
low pressure and the small amount of oxygen in vapor contacting the water cause the dissolved
oxygen to bubble out of solution (Nielsen and Aller, 1984). The vacuum can be produced by
pumps or steam injectors. Several passes through the vacuum deaeration column are necessary
to reduce oxygen to less than 0.1 ppm (Nielsen and Aller, 1984). Any free carbon dioxide will
also be removed, which may result in scale deposition from the accompanying pH change. If
further oxygen reduction is needed, chemical scavengers can be added after vacuum deaeration.
Vacuum deaeration is applicable where chemical treatment is uneconomical, or where the addition
of scavengers would form barium or calcium scale.
A counter-flow gas stripping column can be used to cause dissolved oxygen to escape from
water to a natural gas stream. Either a packed column or a tray-type column can be used, although
the tray-type is preferred (Nielsen and Aller, 1984). Efficient removal of oxygen has been reported
for a vacuum deaeration system supplemented by hydrocarbon gas stripping (Nielsen and Aller,
1984). Oxygen was reduced from 4.7 ppm to 0.05 ppm which reduced the corrosion of steel by
almost 90 percent.
Injection Fluid Neutralization (after EPA, 1982)
Neutralization of an acidic or basic fluid prior to injection can be an effective way to control
corrosion. Common chemicals that may be considered for neutralization are listed in Table 6.
A potential problem with adding chemicals to neutralize pH is that insoluble precipitates may
form and have to be removed as these solids can cause the physical plugging of the injection zone.
Recommended dosage rates for acid and alkali neutralization are shown in Table 7.
Caustic soda, although the most expensive of the alkali sources for acid neutralization, is
usually preferred because it reacts instantaneously and creates less sludge. For neutralizing alkalies,
sulfuric acid is most often used (Warner and Lehr, 1977.)
Chemical Inhibitors.
The addition of chemical inhibitors may be a cost effective means of preventing or reducing
corrosion in wells; however, some may exhibit toxic characteristics and if allowed to escape, could
contaminate underground sources of drinking waters (USDWs). For this reason, wells should
be tested for chemical integrity before inhibitors are added. Film forming, water dispensable
inhibitors, according to Holloway and McSpadden (1961), are generally more effective and less
expensive than other types. Generally, inhibitors are used to supplement other corrosion prevention
measures. Primary inhibitors are used to chemically combat corrosion, whereas secondary
31
-------
TABLE 6
COMMON CHEMICALS USED FOR INJECTION FLUID NEUTRALIZATION
(Warner and Lehr, 1977)
Injection Fluid Characteristic
Neutralizing Chemical
Acid
Alkaline
Lime Slurries
Limestone
Soda Ash
Caustic Soda
Ammonia
Waste Alkali
Sulfuric Acid
Hydrochloric Acid
Carbon Dioxide
Flue Gas
Sulfur
Waste Acid
32
-------
TABLE 7
ALKALI AND ACID REQUIREMENTS FOR pH NEUTRALIZATION
(¥arner and Lehr, 1977)
Approximate Dosage
Alkali (Ib/lb of H2S04)
Dolomitic Limestone 0.95
High Calcium Limestone 1.06
Dolomite Lime, Unslaked 0.53
High Calcium Limestone, Unslaked 0.60
Dolomitic Lime, Hydrated 0.65
High Calcium Lime, Hydrated 0.80
Anhydrous Ammonia 0.35
Soda Ash 1.10
Caustic Soda 0.80
Approximate Dosage
Acid (Ib/lb CaC05)
66°Be ** 1.0
HC1, 20°Be 2.0
Flue Gas, 15# C02 3.0
Sulfur * 0.3
Would produce a reducing condition which might require additional
treatment to produce an oxygen-containing effluent.
°Be = Degrees on the Baume scale (See Lang's Handbook of
Chemistry, 12th Edition).
33
-------
inhibitors are chemicals that are used as bactericides to kill microorganisms which cause corrosion.
A third type of inhibitor is a bactericide which also is a weak corrosion inhibitor.
Corrosion Inhibitors
The conditions of the environment and type of corrosion govern the choice of corrosion inhibitor
(Ostroff, 1965). Usually, the choice of inhibitor is based on the experience of the corrosion engineer
along with trial and error testing. Some typical corrosion inhibitors are given in Table 8. Generally,
organic inhibitors form films on the metal surface, protecting the metal from attack. Some of the
inorganic compounds, like the chromates, are anodic inhibitors (Ostroff, 1965). Many of these
organic and inorganic inhibitors are considered toxic substances, and therefore, must be used with
caution to prevent contamination of potential potable water sources.
The simple addition of corrosion inhibitors alone may not be sufficient to solve all corrosion
problems. Corrosion caused by oxygen associated with salts cannot be effectively treated with
inhibitors (API, 1958), and for these systems it may be necessary to remove the oxygen as well.
Inhibitors are added to the well tubing as well as to the casing tubing annulus. There are
several methods available for adding corrosion inhibitors during well operation. One method
involves filling the annulus during well completion with a solution containing either an oil or
water-soluble corrosion inhibitor (Donaldson, 1972). Inhibitors can also be slug injected while
operations are shut down for a short time, or placed in fluid circulated above the cement in the
casing-borehole annulus.
Bactericides (Horacek, 1986)
The nature of injected wastes often precludes the need for bactericides to control bacterial
populations. For example, some inorganic wastes contain chemicals such as chlorine, chromate,
and compounds of mercury or silver. These have historically been used as bactericides (Ostroff,
1965). While they are no longer approved for such use in the United States, they effectively control
bacterial contamination of fluids that contain them as part of the waste stream. Nonaqueous waste
streams, it is felt, do not require bactericide treatment.
In some cases, the injected aqueous waste does not contain compounds that are toxic to
microorganisms. In such cases, a bactericide may be added to prevent bacterial growth hi the
injection well. In most cases the bactericides are used against organisms that attach themselves
to the well components. It must be understood that there is no correlation between planktonic
(free-floating) and sessile (attached) bacterial populations (Costerton, Irvin, and Cheng, 1981).
Furthermore, once bacteria attach to the wellbore equipment (i.e., become sessile), there is no
method to predict whether they will become corrosive or not (Pope, Duquette, Wagner, and
Johannes, 1984); nor is there any proof that the added bactericide will prevent microbic-influenced
corrosion from occurring (Ruseska, Robbins, Lashen, and Costerton, 1982). Therefore, bactericides
should not be used solely in hopes of preventing bio-corrosion from occurring. Cathodic protection
may be the most cost-effective method of preventing microbic-influenced corrosion (Postgate, 1979).
Such a system will be most effective if it is designed into the system and installed before the well
is used to inject wastewater.
However, bactericides may be useful in preventing bacterial plugging of the injection formation
(Updegraff, 1982). A plugged formation leads to increased injection pressure, which could fracture
the formation, with possible loss of control of where the injected wastes go.
Bactericides may be added by continuous feed or slug treatments. Service companies that
supply bactericides for oil field use should be consulted when selecting a methodology for microbial
34
-------
TABLE 8
CORROSION INHIBITORS
(Gatos, 1956)
Inhibitor
Approximate
Concentrations
(percent)
Corrosive
Environment
Metallic
System
INORGANIC
Calgon
Disodium hydrogen phosphate
Potassium dichromate
Potassium dihydrogen phos-
phate & sodium nitrate
Potassium permanganate
Sodium benzoate
Sodium carbonate
Sodium chromate
Sodium dichromate +
sodium nitrate
Sodium metaphosphate
Sodium nitrite
Sodium orthophosphate
Sodium silicate
Small amount
0.5
0.55-0.2
Small amount
+ 5
0.1
0.5
Small amount
0.07
0.1 + 0. 05
Small amount
0.005
1
0.01
Water systems
Citric Acid
Tap Water, 68-1 94°F
Sea Water - brine
0.3N NaOH solution
0.03£ NaCl solution
Gas condensate wells
CaCl2 brine
Water
Ammonia
Water
Water, pH = 7.25
Oil field brine
Steel
Steel
Iron-brass
Steel
Aluminum
Mild steel
Iron
Cu, brass
Heat-exchange
device
Mild Steel
condensers
Mild Steel
Iron
Steel pipe
35
-------
TABLE 8
(continued)
Inhibitor
Approximate
Concentrations
(percent)
Corrosive
Environment
Metallic
System
ORGANIC
Formalin
Erthritol
Ethylaniline
Mercaptobenzothiazole
Oleic acid
Phenyl aridine
Pyridine + phenylhydrazone
Quinoline ethiodide
Rosin amine-ethylene
oxide
Tetramethylammonium
Thiourea
Smal1 amount
Small amount
0.5
1
Smal1 amount
0.5
0.5 + 0.5
0.1
0.2
0.5
Oil Wells
K2SO^. solutions
HC1 solutions
HC1 solutions
Polyhydric alcohols
H2S04 solutions
HC1 solutions
1M H2S04
MCI* solutions
Aqueous solutions
or organic solvents
Acids
Oil well equipment
Mild steel
Ferrous metals
Iron and steel
Iron
Iron
Ferrous metals
Steel
Mild Steel
Iron and steel
Iron and steel
MCI stands for metallic chloride salts; the M represents a metal.
36
-------
control in an injection well. Appropriate field tests should be done to show that the selected
chemical bactericide is effective against the microbes present in the waste. However, once the
fluid is injected, it is unwise to assume that the bactericide will continue to be effective; therefore,
bacterial eradication must be done prior to injection to insure that bacterial control has been
achieved.
It should be noted that most bactericides exhibit varying amounts of toxicity to humans and
other animals. Injection of these chemicals should be carefully monitored to protect the integrity of
all aquifers penetrated by the injection wellbore. All casing leaks must be corrected and appropriate
reports made to all local, State, and Federal agencies involved in groundwater protection.
Cathodic Protection.
Cathodic protection consists of applying an electric current to the surface of the protected
metal to overpower the voltage of the corrosion cell and to prevent the resulting discharge of
electrical currents from the metal into the ground (Figure 8). The result is that all areas of the
metal become cathodic and corrosion stops. All previous anodic areas are suppressed as long as
adequate current is applied. In injection well operations, cathodic protection is used primarily
for the external protection of well casings. It is typically applied to supplement other corrosion
prevention techniques, such as cementing casing through potentially corrosive saltwater zones.
Generally, the inside of the casing is protected by corrosion resistant coatings, sprayed onto the
surface to be protected (Chemical Week, 1967).
Cathodic protection requires a direct current which may be generated by using an external
source of alternating current and a rectifier for converting to direct current. Thermoelectric gener-
ators may also be used to produce direct current. Current is discharged into the soil from a group
of anodes called a ground bed. Required current for a cathodic protection will vary from 0.5 to 1
amperes for 1,500 feet (457 m) of 6-inch (15.2 cm) casing to as much as 20 amperes for multicased,
deep wells. The voltage can usually be adjusted as required from 6 to about 24 volts, depending
on the needed current and the resistance of the ground bed (EPA, 1982).
A ground bed can be installed so that electrical resistance between the anode and the sur-
rounding soil are at a minimum (API, 1958). To optimize current distribution on the casing, the
ground bed can be placed about 100 feet (30.5 m) from the wellhead and moved as far as possible
from other pipe lines. Where possible, placement of the ground bed in areas of low soil resistance
is desirable; a low-resistance material is usually packed around the anodes to serve as backfill.
Figure 8 depicts a typical cathodic protection installation for a well casing. A horizontal
ground bed is shown. Vertically oriented ground beds, called anode wells, are also used to protect
injection wells. These anode wells, which are drilled to about 300 feet (91.4 m), provide better
vertical distribution of current, and operate with less power than horizontal ground beds (Allen
and Roberts, 1978).
Much of the previous discussions have been taken from the following EPA reports:
U.S. Environmental Protection Agency, 1982:
Injection Well Construction Practices and Technology
(Contract # 68-01-5971)
Prepared by Geraghty and Miller, Inc., and Booz, Allen, and Hamilton, Inc.
Warner, D.L. and Lehr, J.H., 1977:
An Introduction to the Technology of Subsurface Wastewater Injection
EPA-600/2-77-240 Grant Number R-803889.
37
-------
INSULATING JOINT
\ RECTIFIER
GROUND BED
ANODES
SURFACE
CASING
INJECTION^
CASING
CEMENT-
CASING
Figure 8. Example of Cathodic Protection Scheme for Well Casing (EPA, 1982)
38
-------
Corrosion and Hazardous Injection Fluids.
EPA has recently conducted an inventory of Class I hazardous waste wells in the United States
(Environmental Protection Agency, 1985). Class I wells are those wells that inject hazardous,
industrial and municipal wastes that can affect underground sources of drinking water. The data
and/or inventory have provided a data base for determining the composition of the most generally
injected waste fluids. Table 9 gives a list of the most commonly injected fluids as well as a
description of the type of corrosion that can be caused by those fluids. For most Class I injection
wells, pH neutralizers, cathodic, and protective coatings are probably the most effective methods
for preventing corrosion.
39
-------
TABLE 9
CLASS I INJECTION CHEMICALS
(Most commonly injected chemical given first)
INJECTED CHEMICAL
EFFECT ON CORROSION
ACIDS
Pickle Liquor
(HC1, H2S04., FeCl5,
FeClj (Ferric Chloride)
HC1 (Hydrochloric Acid)
H2S04 (Sulfuric Acid)
HF (Hydrofluoric Acid)
Nonspecified Acids
Strong oxidizers and they
enhance electro-
chemical corrosion.
BASES AND CAUSTICS
NH^ (Ammonia)
Nonspecified Alkalines
Nonspecified Caustics
NaOH (Sodium Hydroxide)
Mostly enhance
chemical and electro-
chemical corrosion.
DISSOLVED SPECIES
NaCl
Sulfates
Nitrates
Carbonates
Sulfides
Nonspecified Salts
Electrolyte - enhances
electrochemical corrosion
Can react to form minor
amounts of acid -
nutrient for bacterial
growth
Can react to form minor
amounts of acid -
nutrient for bacterial
growth
Can raise TDS increasing
electrolyte content of
solution - enhancing
electrochemical
corrosion
Nutrient for bacterial
growth and can react
to form acid
Can raise TDS increasing
electrolyte content of
solution - enhancing
electrochemical
corrosion
40
-------
TABLE 9
(continued)
INJECTED CHEMICAL
EFFECT ON1CORROSION
DISSOLVED SPECIES (continued)
Phosphates
Calcium
Magnesium
Iron
Fluorine
Sodium
Chlorine
Can react! to form minor
amounts of acid -
nutrient for bacterial
growth
Can raise'IDS, increasing
electrolyte content of
solution - enhancing
electrochemical
corrosion
ORGANIC COMPOUNDS
Phenols
Isopropyl Alcohol
Formates
Carbon Tetrachloride
Organic Cyanides
Nonspecified Herbicides
Nonspecified Pesticides
May cause decay of
plastic well casing
and rubber tubing
Nonspecified Organic ¥astes
May cause lack of oxygen
allowing for growth of
anaerobes
41
-------
APPENDIX A. REFERENCES
Ahrens, T.D., "Corrosion in Water Wells," Water Well Journal,l977. p. 29f.
Allen, T.O. and A.P. Roberts, Production Operations,Volume 2, Oil and Gas Consultants, Inc. ,
Tulsa, Oklahoma, 1978.
American Petroleum Institute (API), "Corrosion of Oil and Gas-Well Equipment," Dallas, Texas,
1958.
American Petroleum Institute (API) "Recommended Practice for Biological Analysis of Subsurface
Injection Waters," API Publication # RP38, 1975. 7pp.
American Water Works Association, Inc. (AWWA), Water Quality and Treatment: A Handbook of
Public Water Supplies, McGraw-Hill Book Co., Inc., New York, 1971.
Bakhavalov, G.T. and A.V. Turkovskaya, Corrosion and Protection of Metals,1965. Ch. 1, p. 3f.
Barnes, I., "Water-Mineral Reactions Related to Potential Fluid Injection Problems," Underground
Waste Management and Environmental Implications,T.D. Cook (editor) American Association of
Petroleum Geologists, Memoir 18, Tulsa, Oklahoma, 1972.
Barnes, I. and F.E. Clarke, "Chemical Properties of Ground Water and Their Corrosion and
Incrustation Effects on Wells," USGS Professional Paper, Washington Geological Survey (USGS),
498-D, 1969. 58pp.
Baumgartner, A.V., "Microbiological Corrosion," Proceedings of the Fifth Biennial Secondary
Recovery Symposium, Society of Petroleum Engineers, 1962.
Chemical Rubber Company, "Properties of Commercial Plastics." R.C. Weast (editor), CRC
Handbook of Chemistry and Physics,58th edition, 1978. pp. C791-C800.
Chemical Week,"Coater.s Go to the Well," Chemical Week,1967. 100:6, pp. 79-80.
Costerton, J.W., R.T. Irvin, and K.J. Cheng, " The Bacterial Glycocalyx in Nature and Disease,"
Annual Reviews of Microbiology,!^!, pp. 35, 99-324.
Costerton, J.W., and E.S. Lashen, "The Inherent Biocide Resistance of Corrosion-Causing Biofilm
Bacteria," Corrosion 55;Paper No. 246, National Association of Corrosion engineers, 1983. pp.
1-11.
DiTommaso, A., and G.H. Elkan, "Role of Bacteria in Decomposition of Injected Liquid Waste
at Wilmington, North Carolina," Jules Braunstein (editor), Underground Waste Management and
Artificial Rech.arge,Americaji Association of Petroleum Geologists, Tulsa, Oklahoma, 1973. pp.
585-602.
Donaldson, E.G., "Injection Wells and Operations Today," Underground Waste Management and
Environmental Implications,,U.S. Geological Survey and American Association of Petroleum Geol-
ogists, Washington, D.C., 1972.
Dunlop, A.K., R.L. Howard, and P.J. Raifsnider, "ODASA: Psxygen Scavenger and Inhibitor,"
Materials Protection,1969. 8: pp. 3, 27-36.
Edelman, Michael J., "Description and Implications of Three Early Precambrian Paleoweather-
ing Profiles from South Africa," unpublished masters thesis, Temple University,, Philadelphia,
Pennsylvania, 1985.
Ehrlich, G.G., "Role of Biota in Underground Waste Injection and Storage," Underground Waste
Management and Environmental Implications,,U.S. Geological Survey and American Association
of Petroleum Geologists, Washington, D.C., 1972.
Environmental Protection Agency (EPA), Injection Well Construction Practices and Technol-
o<79,Geraghty and Miller Inc. and Booz, Allen and Hamilton, Inc., 1982. pp. 306.
42
-------
APPENDIX A. REFERENCES
(CONTINUED)
Environmental Protection Agency (EPA), Report to Congress on Injection of Hazardous Waste,
EPA 570/9-85-003, Washington, D.C., May 1985.
Evans, U.R., "The Corrosion and Oxidation of Metals," St. Martins Press, Inc., New York, 1960.
Gatos, H.C., "Inhibition of Metallic Corrosion in Aqueous Media," Corrosion, 1956. Volume 12.
Grainage, J.W. and E. Lund, "Quick Culturing and Control of Iron Bacteria," Journal American
Water Works Association,l969. 61:242-245.
Gulf Oil Corporation, Production Department, Houston, Texas, 1948.
Hart, D. and J. Persons, "What Experience Teaches Us About Scaling Incrustation and Corrosion,"
Ground Water Heat Pump J0urna/,Summer, 1980. pp. 12-16.
Hasselbarth, U. and D. Ludemann, "Biological Incrustation of Wells due to Mass Development of
Iron and Manganese Bacteria," Water Treatment and Examination,1972. 21:20-29.
Holloway, H.D. and T.W. McSpadden, "How We Are Fighting Corrosion," Oil and Gas Jour-
naZ,1961. 59:22, 146-150.
Holdren, R.L., "Why Downhole Casing Sometimes Fails," Northeast Oil Reporter,1983. 3:6, 41-43.
Horacek, G.L., Conoco Production Research and Development, Personal Communication, 1986.
Jellinek, "How Oxidation Occurs," Chemical Engineering,1958. 65(17):1,5,130.
Johnson, E.E., Inc. , Groundwater and Wells,Johnson Division, UOP Inc. , Saint Paul, Minnesota,
1975. pp. 440.
Keech, D.K., "Design of Heat Pump Return Wells," Water Well Journal,1982, June. pp. 32-33.
Krauskopf, K.G., Introduction to Geoc&emisfry,McGraw-Hill, New York, 1979. 617 pp.
Langelier, W.F., "The Analytical Control of Anti-Corrosion Water Treatment," Journal American
Water Works Association,1936, 28:1500.
Larrabee, C.P., "Effect of Composition and Environment on Corrosion of Iron and Steel," C.W.
Bergman et. al. (editor), Corrosion of Metals,194Q. p. 31f.
Larson, T.E. and A.M. Buswell, "Calcium Carbonate Saturation Index and Alkalinity Interpreta-
tions," Journal American Water Works Association,1942. 34:1667f.
Moniz, B.J., E.I. DuPont De Nemours et. cie., Beaumont, Texas, Personal Communication, 1986.
Nielsen, D.M.A. Aller, Methods for Determining the Mechanical Integrity of Class II Injection
We/Zs,Robert S. Kerr Environmental Research Laboratory, Ada, Oklahoma, 1984. 263 pp.
Ostroff, A.G., "Introduction to Oil Field Water Technology," Prentice Hall, Inc. , Englewood Cliff,
New Jersey, 1965.
Perry, R.H. et. al. , Chemical Engineers _flan<£6oo£,McGraw-Hill Chemical Engineers Series, 1963.
Persons, J. and D. Hart, "What Experience Teaches Us About Sealing, Incrustation and Corro-
sion," Ground Water Heat Pump Journa/,Summer, 1980. pp. 12-16.
Pope, D.H., D. Duquette, P.C. Wagner, Jr. and A.H Johannes, "Microbiologically Influenced Cor-
rosion: A State-Of-The-Art Review," Material Technology Institute Publication Number 13,Rens-
selear Polytechnic Institute, Troy, New York, 1984.
Postgate, J.R., The Sulphate-Reducing Bacteria,Cambridge University Press, Cambridge, United
Kingdom, 1979.
Pourbaix, M., Lectures on Electrochemical Corrosion,Plenum Press, New York-London, 1973.
43
-------
APPENDIX A. REFERENCES
(CONTINUED)
Ruseska, J., J. Robbins, E.S. Lashen and J.W. Costerton, "Biocide Testing Against Corrosion-
Causing Oilfield Bacteria Helps Control Plugging," Oil and Gas Journal,Ma.Tch 8, 1982.
Ryznar, J.W., "A New Index for Determining the Amount of Calcium Carbonate Scale Formed by
Water," Journal American Water Works Association,!^^, 36:42.
Selm, R.P., and B.T. Hulse, "Deep Well Disposal of Industrial Wastes," Chemical Engineering
Progress, 1960, 56:5, 138f.
Stiff, H.A., and L.E. Davis, "A Method of Predicting the Tendency of Oil Field Waters to Deposit
Calcium Carbonate," American Institute Mining Metallurgy Engineers Transcripts, Petroleum
Division, 1952. 195: pp. 213-216.
Uhlig, H.H., Corrosion and Corrosion Control,John Wiley and Sons, Inc. , New York, 1962; UIC
Construction Practices and Technology p. 165.
Updegraff, D.A., "Plugging and Penetration of Petroleum Reservoir Rock by Microorganisms,"
International Conference Microbial Enhancement of Oil Recovery, Proceedings, Department of
Energy (DOE) Publication Congress 8205140, Washington, D.C., 1982.
van Beek, C.G. E.M. and D. van der Kooij, "Sulfate-Reducing Bacteria in Ground Water from
Clogging and Nonclogging Shallow Wells in the Netherlands River Region," Ground Water, 1981,
20:3.
Warner, D.L. and J.H. Lehr, "An Introduction to the Technology of Subsurface Wastewater Injec-
tion," EPA-600/2-77-240, U.S. Environmental Protection Agency, Washington, D.C., 1977.
Weber, W.J., Physiochemical Processes for Water Quality <7oniro/,Wiley-Interscience, New York,
1972.
44
-------
APPENDIX B. Glossary of Corrosion Related Terms
Aerobic: presence of unreacted or free oxygen
Anaerobic: an absence of unreacted or free oxygen [oxygen as
or Na2SO^ (reacted) is not "free"].
Caliper Survey: the process, by which a tool that is composed of
mechanical "feelers", measures the inside diameter of a
well casing when it is lowered into a well. The caliper tool
is sensitive enough to detect some well diameter changes due
to corrosion.
Casing Thickness Log or Pipe Analysis Survey (PAS): uses an
electromagnetic field to measure the thickness of metal at
any point in the casing. This tool can detect metal loss on
the inside or outside of the casing.
Cathodic Protection: reduction or elimination of corrosion by
making the metal a cathode by means of an impressed d.c-
current or attachment to a sacrificial anode (usually Mg, Al
or Zn).
Chemical Stability: the ability of a compound to remain in
solution.
Corrosion: the destruction of a substance, usually a metal, or
its properties because of a reaction with its (environment)
surroundings.
Coupon: a thin metal foil which is used to test the
corrosiveness of an injection fluid. The corrosion rate is
ascertained by measuring the amount of weight loss by corrosion
over a prescribed period of time.
Differential Aeration Cells: differences in oxygen concentration
between two points of a system. This results in corrosion
due to differences in the solution potential of the same
metal at the two points of the aeration cell.
Electrical Resistance Probes: used to measure the changes in
resistance of a material as it corrodes. A Wheatstone Bridge
circuit is generally employed to measure electrical resistance.
Electrochemical Corrosion: degradation of materials caused by an
exchange of electrons in an electrolyte solution.
Galvanic Corrosion: Corrosion that is increased because of the
current caused by a galvanic cell (sometimes called "couple
action" ) .
45
-------
APPENDIX B. Glossary
(continued)
Inorganic Incrustation: decrease in injection well performance
caused by precipitation of inorganic substances. .
Nonmetallic Corrosion: degradation of nonmetallic well
components due to chemical action by formation or injection
fluids.
Packed Tower: an air-stripper used to remove oxygen from
solution.
Planktonic: free swimming or free floating microorganisms.
Sessile: Microorganisms that physically attach to a surface,
and they form a continuous film.
Symbiosis: the living together of two dissimilar organisms; the
relationship may be mutualistic (co-dependent), commensal
(benefits only one), amensal (one suppresses the other), or
parasitic (one lives at the expense of the other).
Synergistic: the magnified effect of the combination of two or
more phenomena, which is more than the addition of its
components.
Thermomechanical history: the mechanical working and heat
treatment operations used to make a specific metal product
form, such as tubing.
USDVs: Underground Source of Drinking Water: an aquifer or its
portion: (1.) (i) which supplies or may supply water for
human consumption; or (ii) in which the ground-water
contains fewer than 10,000 mg/1 "total dissolved solids;"
and (2.) which is not an exempted aquifer in accordance with
the Federal regulations.
Vheatstone Bridge Circuit: used to measure the resistance of a
system.
46
-------
APPENDIX C. Corrosion Inspection Check List
Name of Inspector: Date:_
Facility: Veil #:.
/ / Date of Inspection
/ / Date of Last Inspection
Prepared by:
DESCRIPTION OF CORROSION PREVENTION/MONITORING SYSTEM:
.[ ] Corrosion Loop
[ ] Veight Loss Coupons
[ ] Electrical Resistance Probes
[ ] Polarization Resistance Probes
[ ] Logs-Type
[ ] Cathodic Protection
[ ] Soil Potential Survey
[ ] Veil Design (Vhat components aid in the control of corrosion)
[ ] Other (Please describe)
DATE OF LAST CORROSION EVALUATION BY OPERATOR:
TYPE
[ ] Visual
[ ] Other (Describe briefly)
47
-------
APPENDIX C. Corrosion Inspection Check List
(continued)
RESULTS OF EVALUATION
[ ] No Significant Corrosion Detected
[ ] Corrosion Taking Place in the:
[ ] Casing; depth
[ ] Tubing; depth
[ ] Packer
[ ] Other (Indicate component)
Injection Fluid Released? YES [ ] NO [ ]
Potential Contamination of USDWs? YES [ ] NO [ ]
CASING MATERIAL
[ ] Steel
[ ] Stainless Steel
[ ] Monel
[ ] Titanium
[ ] Other; Specify
TUBING MATERIAL
[ ] Steel
[ ] Stainless Steel
[ ] Fibercast
[ ] Fiberglass
[ ] Other; Specify
PACKER TYPE AND MATERIAL
[ ] Tension
[ ] Compression
[ ] Material: Steel [ ] ; Other [ ] , Specify
[ ] Special Protection (Please indicate. Note some packers,
especially tension packers, have rubber pads or special coatings
to prevent contact with injection fluids).
48
-------
APPENDIX C. Corrosion Inspection Check List
(continued)
INJECTION FLUID CHARACTERISTICS
[ ] Unknown
C 1 PH =
[ ] Dissolved Oxygen (concentration) mg/1
[ ] Hydrogen Sulfide, HgS (concentration) mg/1
[ ] Carbon Dioxide, C02 (concentration) mg/1
[ ] Amenable to Biological Degration
[ ] Other Corrosive Characteristics. Describe:
[ ] Most Recent Sample Analysis (attached) Indicates No Significant Changes
Were All the Chemical Test Done in Accordance With the Quality
Assurance Project Plan for Chemical Analysis for the State?
YES [ ] NO [ ] . Unknown [ ]
Date(s):_
Lab(s):
EVALUATION OF THE CASING/TUBING/PACKER MATERIALS TO RESIST CORROSION
(By consulting the tables on page of the manual
a preliminary evaluation can be made. The inspector may
use also different criteria for evaluation; however he/she
should indicate the reason for the decision.)
[ ] Adequate
[ ] Inadequate
Criteria Used:
49
-------
APPENDIX C. Corrosion Operator's Check List
Prepared by: Date:
Facility: Well #:.
/ / Date of Corrosion Evaluation
/ / Date of Last Evaluation of the Wells
I. Is (Are) the Well(s) Designed to Resist Chemical and Physical
Attack by the Waste and the Environment?
YES [ ] NO [ ]
Is Waste Treated Prior to Injection: YES [ ] NO [ ]
Briefly Describe Treatment:
If You Checked "Yes" in the First Part of this Section Please
Indicate Briefly Design Characteristics to Prevent Corrosion
(Address: waste type, tubing, packer and casings):
II. Type of Evaluation Done (Please check)
[ ] Corrosion Loop Inspection
[ ] Weight Loss Coupons
[ ] Electrical Resistance Probes
[ ] Polarization Resistance Probes
[ ] We11-Logging Type
[ ] Soil Potential Survey
[ ] Well Monitoring Only
50
-------
APPENDIX C. Corrosion Operator's Check List
(continued)
[ ] Pulled Tubing (Please elaborate)
[ ] Other
III. Corrosion Detected (Please check one):
[ ] No (If checked, do not complete the rest of the form)
[ ] Yes (If checked, corrosion taking place in the):
[ ] Tubing; depth
[ ] Packer
[ ] Other (Indicate component)
[ ] Casing:
IV. Action Taken:
[ ] None
[ ] Replaced Component
[ ] Changed Operations (Indicate changes)
[ ] Provided Additional Preventive Measures (Indicate briefly)
[ ] In the Process of Providing Additional Preventive Measures
(Indicate briefly)
[ ] Further Description of Problem and Action Taken in Attached Page(s)
51
------- |