9

ATMOSPHERIC CORROSION

9.1 INTRODUCTION

In the absence of moisture, iron exposed to the atmosphere corrodes at a negli-

gible rate. For example, steel parts abandoned in the desert remain bright and
tarnish-free for long periods of time. Also, the corrosion process cannot proceed
without an electrolyte; hence, in climates below the freezing point of water or of
aqueous condensates on the metal surface, rusting is negligible. Ice is a poor
electrolytic conductor. Incidence of corrosion by the atmosphere depends,
however, not only on the moisture content of air, but also on the particulate
matter content and gaseous impurities that favor condensation of moisture on
the metal surface.
Ambient air quality in the United States has improved dramatically since the
Clean Air Act was enacted in 1970. The Environmental Protection Agency (EPA)
monitors air quality and compares the data with the National Ambient Air
Quality Standards (NAAQS), which have been established for ozone (O3), carbon
monoxide (CO), nitrogen dioxide (NO2), sulfur dioxide (SO2), lead (Pb), and
particulate matter (PM, airborne particles of any composition). Improvements in
air quality help to mitigate atmospheric corrosion, as will be discussed in this
chapter.

Corrosion and Corrosion Control, by R. Winston Revie and Herbert H. Uhlig

Copyright © 2008 John Wiley & Sons, Inc.

191
192 ATMOSPHERIC CORROSION

9.2 TYPES OF ATMOSPHERES

Atmospheres vary considerably with respect to moisture, temperature, and con-

taminants; hence, atmospheric corrosion rates vary markedly around the world.
Approaching the seacoast, air is laden with increasing amounts of sea salt, in
particular NaCl. At industrial areas, SO2, H2S, NH3, NO2, and various suspended
salts are encountered. Approximate concentration ranges of corrosive gases in
urban areas are presented in Table 9.1 [1]. Acids that can form from these gases
include sulfuric acid (H2SO4), nitric acid (HNO3), and organic acids, such as
formic acid (HCOOH) and acetic acid (CH3COOH).
A metal that resists corrosion in one atmosphere may lack effective corrosion
resistance elsewhere; hence, relative corrosion behavior of metals changes with
location; for example, galvanized iron performs well in rural atmospheres, but is
relatively less resistant to industrial atmospheres. On the other hand, lead per-
forms in an industrial atmosphere at least as well as, or better than, elsewhere
because a protective film of lead sulfate forms on the surface.
Recognition of marked differences in corrosivity has made it convenient to
divide atmospheres into types. The major types are marine, industrial, tropical,
arctic, urban, and rural. There are also subdivisions, such as wet and dry tropical,
with large differences in corrosivity. Also, specimens exposed to a marine atmo-
sphere corrode at greatly differing rates depending on proximity to the ocean.
At Kure Beach, North Carolina, specimens of steel located 24 m (80 ft) from the
ocean, where salt water spray is frequent, corroded about 12 times more rapidly
than similar specimens located 240 m (800 ft) from the ocean [2].

9.3 CORROSION-PRODUCT FILMS

Corrosion-product films formed in the atmosphere tend to be protective; that is,

the corrosion rate decreases with time (Fig. 9.1) [3]. This is true to a lesser extent
of pure iron, for which the rate is relatively high, compared to the copper-bearing
or low-alloy steels, which are more resistant. Rust films on the latter steels tend

TABLE 9.1. Atmospheric Corrosive Gases in Outdoor

Urban Environments [1]
Gas Approximate Concentration Range (ppbv)

H2S 0.2–700
SO2 3–1000
NH3 1–90
HCl 0.5–100
NO2 0.5–300
O3 0.9–600
RCOOH 0.5–30
CORROSION-PRODUC T FILMS 193

Figure 9.1. Atmospheric corrosion of steels as a function of time in an industrial environment

(C. Larrabee, in Corrosion Handbook, H. H. Uhlig, editor, Wiley, New York, 1948, p. 124)

to be compact and adherent, whereas on pure iron they are a powdery loose
product. The corrosion rate eventually reaches steady state and usually changes
very little on further exposure. This is characteristic of other metals as well, as
can be seen from data obtained by the American Society for Testing and Materi-
als for various metals exposed 10 or 20 years to several atmospheres (Table 9.2)
[4]. Within the experimental error of such determinations, the rate for a 20-year
period is about the same as that for a 10-year period. The data in Table 9.2 also
show the beneficial effect of alloying elements, primarily 1.1% Cr and 0.4% Cu,
in the low-alloy steel compared to the 0.2% C steel.
Data of Fig. 9.1 follow the relation

p = kt n (9.1)

where p can be expressed as specimen weight loss (g/m2), or as specimen penetra-

tion (μm), during time t (years), and k and n are constants that depend on the
metal and on the atmospheric conditions (i.e., climatic and pollution factors) at
the test site. In addition to carbon steels, this relation is also found to apply to
atmospheric test data for galvanized, aluminized, and 55% Al–Zn coatings on
steel [5–7]. Values of n typically range from about 0.5 to 1. The constant, n = 1,
applies for a linear rate law; that is, the corrosion product film is not protective.
In comparison with carbon steels, weathering steels have very low values of n,
194 ATMOSPHERIC CORROSION

TABLE 9.2. Average Atmospheric Corrosion Rates of Various Metals for 10- and 20-Year
Exposure Times, mils/yeara (American Society for Testing and Materials) [4]
Metal Atmosphere

New York City La Jolla, CA State College,

(Urban (Marine) PA (Rural)
Industrial)

10 20 10 20 10 20
Years Years Years Years Years Years

Aluminum 0.032 0.029 0.028 0.025 0.001 0.003

Copper 0.047 0.054 0.052 0.050 0.023 0.017
Lead 0.017 0.015 0.016 0.021 0.019 0.013
Tin 0.047 0.052 0.091 0.112 0.018 —
Nickel 0.128 0.144 0.004 0.006 0.006 0.009
65% Ni, 32% Cu, 2% Fe, 1% Mn 0.053 0.062 0.007 0.006 0.005 0.007
(Monel)
Zinc (99.9%) 0.202 0.226 0.063 0.069 0.034 0.044
Zinc (99.0%) 0.193 0.218 0.069 0.068 0.042 0.043
0.2% C steelb (0.02% P, 0.05% S, 0.48 — — — — —
0.05% Cu, 0.02% Ni, 0.02% Cr)
Low-alloy steelb (0.1% C, 0.2% P, 0.09 — — — — —
0.04% S, 0.03% Ni, 1.1% Cr,
0.4% Cu)
a
1 mil/year = 0.001 in./year = 0.0254 mm/year = 25.4 μm/year.
b
Kearney, New Jersey (near New York City); values cited are one-half reduction of specimen thickness
[C. P. Larrabee, Corrosion 9, 259 (1953)].

usually less than 21 [8]. A value of n = 0.5 corresponds to a parabolic

rate law.
From (9.1), the linear bilogarithmic law may be expressed as

log p = A + B log t (9.2)

where A = log k and B = n. The atmospheric behavior of a specific metal at a

specific location can be described using the two parameters A and B. This biloga-
rithmic law can be very useful in predicting long-term atmospheric corrosion
damage based on tests of shorter duration. Extrapolation up to 20–30 years from
tests of 4 years is possible with reasonable confidence [9]. Over the longer term,
changes in the environment are likely to be more significant than deviations from
the model.
The mean corrosion rate, p/t, can be calculated as

log ( pt ) = A + B − 1 log t
( ) (9.3)
FAC TORS INFLUENCING CORROSIVIT Y OF THE ATMOSPHERE 195

and the instantaneous corrosion rate, dp/dt, can be expressed as

log ( dpdt ) = A + log B + (B − 1) log t (9.4)

When there is a linear relationship between log p and log t, there are also linear
relationships between log p/t and log t, and between log dp/dt and log t.
Metal surfaces located where they become wet or retain moisture, but where
rain cannot wash the surface, may corrode more rapidly than specimens fully
exposed. The reason for this is that sulfuric acid, for example, absorbed by rust
will continue to accelerate corrosion, perhaps by means of the cycle
1 1 1 1
H 2SO4 + O2 O2 + H 2SO4 1 H 2O 1 3
Fe ⎯⎯⎯⎯⎯ 2
→ FeSO4 ⎯4⎯⎯⎯⎯⎯
2
→ Fe2 (SO4)3 ⎯2⎯⎯→ Fe2O3 + H 2SO4
2 2 2
(9.5)
Intermediate formation of ferric sulfate has not been demonstrated, and so
FeSO4 may oxidize directly to Fe2O3. Nevertheless, rust contaminated in this way
catalyzes the formation of more rust. Direct exposure of a metal to rain may,
therefore, be beneficial, compared to a partially sheltered exposure. This advan-
tage presumably would not extend to uncontaminated atmospheres.

9.4 FACTORS INFLUENCING CORROSIVITY OF THE ATMOSPHERE

In all except the most corrosive atmospheres, the average corrosion rates of
metals are generally lower when exposed to air than when exposed to natural
waters or to soils. This fact is illustrated by the data of Table 9.3 for steel, zinc,
and copper in three atmospheres compared to average rates in seawater and a

TABLE 9.3. Comparison of Atmospheric Corrosion Rates with Average Rates in Seawater
and in Soilsa
Environment Corrosion Rate (gmd)

Steel Zinc Copper

Rural atmosphere — 0.017 0.014

Marine atmosphere 0.29 0.031 0.032
Industrial atmosphere 0.15 0.10 0.029
Seawater 2.5 1.0 0.8
Soil 0.5 0.3 0.07
a 1
Atmospheric tests on 0.3% copper steel, 7 2-year exposure, from C. Larrabee, Corrosion 9, 259 (1953).
Atmospheric rates for zinc and copper, 10-year exposure, from Symposium on Atmospheric Exposure
Tests on Non-Ferrous Metals, ASTM, 1946. Seawater data from Corrosion Handbook, H. H. Uhlig,
editor, Wiley, New York, 1948. Soil data for steel are averaged for 44 soils, 12-year exposure; for zinc,
12 soils, 11-year exposure; for copper, 29 soils, 8-year exposure–from Underground Corrosion, M.
Romanoff, Circ. 579, National Bureau of Standards, Washington, D.C. 1957.
196 ATMOSPHERIC CORROSION

variety of soils. In addition, atmospheric corrosion of passive metals (e.g., alumi-

num and stainless steels) tends to be more uniform and with less marked pitting
than corrosion in waters or soils.
Specific factors influencing the corrosivity of atmospheres are particulate
matter, gases in the atmosphere, and moisture (critical humidity). In the United
States, the Environmental Protection Agency (EPA) compiles data on ambient
outdoor-air quality from information obtained at air-monitoring stations, where
concentrations of the more common gases found in trace amounts in the atmo-
sphere are measured continuously. Because of the importance of trace constitu-
ents on atmospheric corrosion behavior, the engineer or architect would be well
advised to review air-quality data for a corrosion assessment at the design stage,
to avoid disappointment at the operational stage. Microclimatic conditions (e.g.,
east versus west exposure, sunlight, wind direction, proximity to a highway where
deicing salts are used) can be different from macroclimates and should be
considered.

9.4.1 Particulate Matter

Particulate matter (PM) is the term for a mixture of solid particles and liquid
droplets in air. PM consists of several components, including acids, organic chemi-
cals, metals, and soil or dust particles. The importance of atmospheric dust was
established in the early experiments of Vernon [11], who exposed specimens of
iron to an indoor atmosphere, some specimens being entirely enclosed by a cage
of single-thickness muslin measuring several inches larger in size than the speci-
men. After several months, the unscreened specimens showed rust and appre-
ciable gain in weight, whereas the muslin-screened specimens showed no rust
whatsoever and had gained weight only slightly.
Particulate matter is found near roads and some industries, in smoke and
haze; it can be directly emitted from sources such as forest fires, and it can be
formed when gases emitted from power plants and automobiles react in the air.
The NAAQSs for particulate matter, revised in 2006, are set at 150 μg/m3 as the
24-h average for coarse particulates (2.5- to 10-μm diameter) and are set at
35 μg/m3 as the 24-h average for fine particulates (≤2.5-μm diameter) [12]. Particle
pollution is controlled by reducing directly emitted particles and by reducing
emissions of pollutants that are gases when emitted, but form particles in the
atmosphere. In the United States in 2006, the average concentration of fine par-
ticulates amounted to about 13 μg/m3 [13].
Particulate matter can be a primary contaminant of many atmospheres. In
the 1950s, it was estimated that the average city air contained about 2 mg/m3, with
higher values for an industrial atmosphere, reaching 1000 mg/m3 or more [14]. It
was estimated that more than 35,000 kg of dust per km2 (100 tons per square
mile) settled every month over an industrial city [15].
In contact with metallic surfaces, particulate matter influences the corrosion
rate in an important way. Industrial atmospheres carry suspended particles of
carbon and carbon compounds, metal oxides, H2SO4, (NH4)2SO4, NaCl, and other
FAC TORS INFLUENCING CORROSIVIT Y OF THE ATMOSPHERE 197

salts. Marine atmospheres contain salt particles that may be carried many miles
inland, depending on magnitude and direction of the prevailing winds. These
substances, combined with moisture, initiate corrosion by forming galvanic or
differential aeration cells; or, because of their hygroscopic nature, they form an
electrolyte on the metal surface. Dust-free air, therefore, is less apt to cause cor-
rosion than is air heavily laden with dust, particularly if the dust consists of
water-soluble particles or of particles on which H2SO4 is adsorbed.

9.4.2 Gases in the Atmosphere

The carbon dioxide normally present in air neither initiates nor accelerates cor-
rosion. Steel specimens rust in a carbon-dioxide-free atmosphere as readily as in
the normal atmosphere. Early experiments by Vernon showed that the normal
carbon dioxide content of air actually decreases corrosion [16], probably by
favoring a more protective rust film.
A trace amount of hydrogen sulfide in contaminated atmospheres causes the
observed tarnish of silver and may also cause tarnish of copper. The tarnish films
are composed of Ag2S and a mixture of Cu2S + CuS + Cu2O, respectively.
Important atmospheric pollutants are sulfur dioxide (SO2), nitrogen oxides
(NOx), ozone (O3), and chloride ions (Cl−). Sulfur dioxide and nitrogen oxides
form corrosive acids, and ozone is a powerful oxidizing agent.
Sulfur dioxide originates predominantly from the burning of coal, oil, and
gasoline. In New York City in the 1950s, it was estimated that about 1.5 million
tons of sulfur dioxide were produced every year from burning coal and oil [17].
This amount was equivalent to burdening the atmosphere with an average of
6300 tons of H2SO4 every day and has been reduced by limiting the allowable
sulfur content of fuels burned within city limits. Since fuel consumption is
higher in winter, sulfur dioxide contamination is also higher (Fig. 9.2). It is also
obvious from this cause that the average sulfur dioxide content of the air (and

Figure 9.2. Variation of average sulfur dioxide content of New York City air with time of
year [17].
198 ATMOSPHERIC CORROSION

TABLE 9.4. Variation of SO2 Content of Air with Distance from Center of City (H. Meller, J.
Alley, and J. Sherrick, quoted in Ref. 17)
City Parts per Million

Distance 0–5 5–10 10–15 15–20 20–25 25–30

Miles Kilometers 0–8 8–16 16–24 24–32 32–40 40–48

Detroit 0.023 0.012 0.006 0.004 0.004 0.005

Philadelphia–Camden 0.030 0.018 0.016 0.021 0.012 0.012
Pittsburgh 0.060 0.030 0.015 0.018 0.009 0.010
St. Louis 0.111 0.048 0.029 0.020 0.018 0.014
Washington, D.C. 0.003 0.001 0.001 0.001 0.001 0.002

corresponding corrosivity) falls off with distance from the center of an industrial
city, and it is clear that this effect is not as pronounced in the case of a residential
city, such as Washington, D.C. (Table 9.4).
Although the SO2 levels in the air of many major centers of population
around the world may be similar to the data in Fig. 9.2, the engineering and
industrial responses to the U.S. Clean Air Act have resulted in a major decline
in the levels of air pollutants. About 10% of SO2 emissions result from industrial
processes, and transportation sources contribute most of the remainder. The EPA
estimates that ambient SO2 levels have decreased by more than 50% since 1983.
In comparison with the data in Fig. 9.2, the annual average concentration of SO2
in New York City air in 2006 was about 0.01 ppm, and the NAAQS for SO2 is
0.03 ppm (annual arithmetic mean) [18].
A high sulfuric acid content of industrial and urban atmospheres shortens
the life of metal structures (see Tables 9.2 and 9.3). The effect is most pronounced
for metals that are not particularly resistant to sulfuric acid, such as zinc, cadmium,
nickel, and iron. It is less pronounced for metals that are more resistant to dilute
sulfuric acid, such as lead, aluminum, and stainless steels. Copper, forming a pro-
tective basic copper sulfate film, is more resistant than nickel or 70% Ni–Cu alloy,
on which the corresponding films are less protective. In the industrial atmosphere
of Altoona, Pennsylvania, galvanized steel sheets [0.381 kg zinc per m2, 0.028 mm
thick (1.25 oz zinc per ft2, 1.1 mil thick)] began to rust after 2.4 years, whereas in
the rural atmosphere of State College, Pennsylvania, rust appeared only after
14.6 years [19].
Copper exposed to industrial atmospheres forms a protective green-colored
corrosion product called a patina, composed mostly of basic copper sulfate,
CuSO4·3Cu(OH)2. A copper-covered church steeple on the outskirts of a town
may develop such a green patina on the side facing prevailing winds from the
city, but remain reddish-brown on the opposite side, where sulfuric acid is less
readily available. Near the seacoast, a similar patina forms, composed in part of
basic copper chloride.
FAC TORS INFLUENCING CORROSIVIT Y OF THE ATMOSPHERE 199

Industrial atmospheres can cause S.C.C of copper-base alloys, mostly

accounted for by presence of nitrogen oxides (see Section 20.2.3). Copper did
not fail, but copper alloys containing >20% Zn failed on exposures for up to 8
years [20].
Nickel is quite resistant to marine atmospheres, but is sensitive to sulfuric
acid of industrial atmospheres (Table 9.2), forming a surface tarnish composed
of basic nickel sulfate. Corrosion in the industrial atmosphere of New York City
is about 30 times higher than in the marine atmosphere of La Jolla, California,
and about 20 times higher than in the rural atmosphere of State College,
Pennsylvania (Table 9.2).

9.4.3 Moisture (Critical Humidity)

From previous discussions, it is apparent that, in an uncontaminated atmo-
sphere at constant temperature, appreciable corrosion of a pure metal surface
would not be expected at any value of relative humidity below 100%. Practi-
cally, however, because of normal temperature fluctuations (relative humidity
increases on decrease of temperature) and because of hygroscopic impurities
in the atmosphere or in the metal itself, the relative humidity must be reduced
to values much lower than 100% in order to ensure that no water condenses
on the surface. In very early studies, Vernon discovered that a critical relative
humidity exists below which corrosion is negligible [21]. Experimental
values for the critical relative humidity are found to fall, in general, between
50% and 70% for steel, copper, nickel, and zinc. Typical corrosion behavior of
iron as a function of relative humidity of the atmosphere is shown in Fig. 9.3.
In a complex or severely polluted atmosphere, a critical humidity may not
exist [22].
An important factor determining susceptibility to atmospheric corrosion of
a metal in a particular environment is the percentage of time that the critical
humidity is exceeded [23]. This period of time is called the “time of wetness.” It
is determined by measuring the potential between a corroding metal specimen
and a platinum electrode [23, 24]. Surface moisture from either precipitation or
condensation is the cell electrolyte. To estimate atmospheric corrosion rates,
Kucera et al. designed the device shown in Fig. 9.4 [25, 26]. The cell B is placed
about 1 m above ground level with the surface inclined at 45 °. An electronic
integrator automatically integrates cell currents over extended periods of time.
Calibration using data from weight-loss experiments at the same site shows that
the electrochemical technique is suitable for estimating short-term variations in
corrosion rates [26].
Atmospheric corrosion testing is important to the suppliers of metals and to
the engineers and architects who use metals under atmospheric conditions.
Reviews have been prepared summarizing the atmospheric corrosion standards
and testing procedures of the American Society for Testing and Materials
[27–30].
200 ATMOSPHERIC CORROSION

Figure 9.3. Corrosion of iron in air

containing 0.01% SO2, 55 days’ exposure,
showing critical humidity [21].

Figure 9.4. General arrangement of electrochemical device for measurement of atmospheric

corrosion: A is a zero resistance ammeter, the circuit of which is shown on the right; B is an
electrochemical cell with (a) electrodes and (b) insulators; C is an external emf. [25, 26]. (Copy-
right ASTM INTERNATIONAL. Reprinted with permission.)
REMEDIAL MEASURES 201

9.5 REMEDIAL MEASURES

1. Use of Organic, Inorganic, or Metallic Coatings. Coatings are discussed in

Chapters 14, 15, and 16.
2. Reduction of Relative Humidity. Heating air—or, better still, reducing the
moisture content—can reduce relative humidity. Lowering the relative
humidity to 50% suffices in many cases. If the presence of unusually
hygroscopic dust or other surface impurities is suspected, the value should
be reduced still further. This protective measure is effective except perhaps
when corrosion is caused by acid vapors from nearby unseasoned wood
or by certain volatile constituents of adjacent plastics, or paints.
3. Use of Vapor-Phase Inhibitors and Slushing Compounds. These are dis-
cussed in Chapter 17.
4. Use of Alloys. When alloyed with steel in small concentrations, copper,
phosphorus, nickel, and chromium are particularly effective in reducing
atmospheric corrosion. Copper additions are more effective in temperate
climates than in tropical marine regions; chromium and nickel additions
combined with copper and phosphorus are effective in both locations
(Table 9.5). Corrosion rates of structural steels in tropical atmospheres
(e.g., Panama) were found, in general, to be about two or more times
higher than in temperate atmospheres (e.g., Kure Beach, North Carolina)
mainly because of the higher relative humidity and higher average
temperatures.

The usefulness of low-alloy steels to resist atmospheric corrosion through

formation of protective rust films has resulted in the development of weathering
steels. These are used for construction of buildings and bridges and for architec-
tural trim, without the necessity of painting, thereby saving appreciable amounts
in maintenance costs over the life of the structure. A typical commercial composi-
tion is as follows: 0.09% C, 0.4% Mn, 0.4% Cu, 0.8% Cr, 0.3% Ni, 0.09% P. Such
steels do not have any advantage when buried in soil or totally immersed in water,
because the corresponding rust films formed under continuously wet conditions
are no more protective than those formed on carbon steels. Their more noble
corrosion potentials compared to carbon steels may make them useful in certain
galvanic couples (see Section 7.3.2.1). It is only by a process of alternate wetting
and drying that the rust continues to be protective.
Misawa et al. [32] proposed that an inner cohesive protective rust film is
formed on low-alloy steels after long atmospheric exposure (industrial or urban).
It consists of amorphous δ-FeOOH, the formation of which is catalyzed by copper
and phosphorus on the steel surface; alternate drying and wetting favors its pro-
tective qualities. Keiser et al. [33] confirmed that the typical inner adherent rust
layer consists mostly of δ-FeOOH.
Stainless steels and aluminum resist tarnish in industrial, urban, and rural
atmospheres, as is apparent from their satisfactory use over many years as
202 ATMOSPHERIC CORROSION

TABLE 9.5. Effect of Low-Alloy Components on Atmospheric Corrosion of Commercial

Steel Sheet (Eight-Year Exposure)
Steel Composition, % Loss of
Thickness

C P Cu Other mm mils

Industrial Atmosphere (Kearney, NJ) [19]

Carbon 0.2 0.02 0.03 0.20 8.0
Copper bearing 0.2 0.02 0.3 0.11 4.4
Low chromium 0.09 0.2 0.4 1 Cr 0.048 1.9
Low nickel 0.2 0.1 0.7 1.5 Ni 0.051 2.0

Temperate Marine Atmospherea(Kure Beach, NC) [19]

Carbon 0.2 0.02 0.03 0.24 9.5
Copper bearing 0.2 0.01 0.2 0.15 5.8
Low chromium 0.1 0.14 0.4 1 Cr 0.069 2.7
Low nickel 0.1 0.1 0.7 1.5 Ni 0.076 3.0

Tropical Marine Atmosphere (Panama Canal Zone) [31]

Carbon 0.25 0.08 0.02 0.52 20.4
Copper bearing 0.2 0.004 0.24 0.45 17.6
Low chromium 0.07 0.008 0.1 3.2 Cr 0.23 9.1
Low nickel 0.2 0.04 0.6 2.1 Ni 0.19 7.5
a
7.5-year exposure.

architectural trim of buildings. Hastelloy C (54% Ni, 17% Mo, 5% Fe, 15% Cr,
4% W) is very resistant to tarnish in marine atmospheres, making it a useful alloy
for reflectors on board ship.

REFERENCES

1. C. Leygraf and T. Graedel, Atmospheric Corrosion, Wiley, New York, 2000, p. 41.
2. F. L. LaQue, Am. Soc. Testing Mater. Proc. 51, 495 (1951).
3. C. P. Larrabee and S. K. Coburn, in Proceedings, First International Congress On Metallic
Corrosion, Butterworths, London, 1962, pp. 276–284.
4. Symposium on Atmospheric Corrosion of Non-Ferrous Metals, STP 175, American
Society for Testing Materials, Philadelphia, PA, 1955. Also previous Symposium,
1946.
5. M. Tullmin and P. R. Roberge, Atmospheric corrosion, in Uhlig’s Corrosion Handbook,
2nd edition, R. W. Revie, editor, Wiley, New York, 2000, pp. 315–317.
6. R. Legault and V. Pearson, Corrosion 34, 433 (1978).
7. H. Townsend and J. Zoccola, Mater. Perf. 18(10), 13 (1979).
8. L. Veleva and R. D. Kane, in ASM Handbook, Vol. 13A, Corrosion: Fundamentals,
Testing, and Protection, ASM International, Materials Park, OH, 2003, pp. 203–204.
GENER AL REFERENCES 203

9. T. Kodama, in ASM Handbook, Vol. 13B, Corrosion: Materials, ASM International,

Materials Park, OH, 2003, p. 7.
10. M. Pourbaix, The linear bilogarithmic law for atmospheric corrosion, in Atmospheric
Corrosion, W. H. Ailor, editor, Wiley, New York, 1982, pp. 107–121.
11. W. Vernon, Trans. Faraday Soc. 23, 113 (1927).
12. http://www.epa.gov/oar/particlepollution/naaqsrev2006.html
13. http://www.epa.gov/air/airtrends/pm.html
14. A. Kutzelnigg, Werks. Korros. 8, 492 (1957).
15. Chem. Eng. News, January 4, 1971, p. 29.
16. W. Vernon, Trans. Faraday Soc. 31, 1668 (1935).
17. L. Greenburg and M. Jacobs, Eng. Chem. 48, 1517 (1956).
18. http://www.epa.gov/air/airtrends/sulfur.html
19. C. P. Larrabee, Corrosion 9, 259 (1953).
20. J. Popplewell and T. Gearing, Corrosion 31, 279 (1975).
21. W. Vernon, Trans. Faraday Soc. 23, 113 (1927); ibid., 27, 255 (1931) and 31, 1668 (1935);
Trans. Electrochem. Soc. 64, 31 (1933).
22. D. Rice, P. Phipps, and R. Tremoureux, J. Electrochem. Soc. 126, 1459 (1979); 127, 563
(1980).
23. P. Sereda, in Corrosion In Natural Environments, STP 558, American Society for
Testing and Materials, Philadelphia, 1974, p. 7.
24. F. Mansfeld, in Atmospheric Corrosion, W. Ailor, editor, Wiley, New York, 1982, p. 139.
25. V. Kucera and E. Mattsson, in Ref. 23, p. 239.
26. V. Kucera and M. Collin, In Eurocorr ’77, Sixth European Congress On Metallic Cor-
rosion, London, 1977, Society of Chemical Industry, London, 1977, p. 189.
27. W. Ailor, in Atmospheric Factors Affecting the Corrosion of Engineering Metals, S.
Coburn, editor, STP 646, American Society for Testing and Materials, Philadelphia,
1978, p. 129.
28. S. Dean, in Ref. 24, p. 195.
29. H. H. Lawson, Outdoor atmospheres, in Corrosion Tests and Standards: Application
and Interpretation, 2nd edition, R. Baboian, editor, ASTM International, West
Conshohocken, 2005, pp. 343–348.
30. J. D. Sinclair, Indoor atmospheres, in Corrosion Tests and Standards: Application and
Interpretation, 2nd edition, R. Baboian, editor, ASTM International, West
Conshohocken, 2005, pp. 349–361.
31. C. Southwell, B. Forgeson, and A. Alexander, Corrosion 14, 435t (1958).
32. T. Misawa, T. Kyuno, W. Suetaka, and S. Shimodaira, Corros. Sci. 11, 35 (1971).
33. J. Keiser, C. Brown, and R. Heidersbach, Corros. Sci. 23, 251 (1983).

GENERAL REFERENCES

W. Ailor, editor, Atmospheric Corrosion, Wiley, New York, 1982.

R. B. Griffin, Corrosion in marine atmospheres, in ASM Handbook, Vol. 13C, Corrosion:
Environments and Industries, ASM International, Materials Park, OH, 2006, pp.
42–60.
204 ATMOSPHERIC CORROSION

C. Leygraf, Atmospheric corrosion, in Corrosion Mechanisms in Theory and Practice, 2nd

edition, P. Marcus, editor, Marcel Dekker, New York, 2002.
C. Leygraf and T. Graedel, Atmospheric Corrosion, Wiley-Interscience, New York, 2000.
C. Leygraf, J. Hedberg, P. Qiu, H. Gil, J. Henriquez, and C. M. Johnson, Molecular in situ
studies of atmospheric corrosion, Corrosion 63(8), 715 (2007).
P. R. Roberge, Corrosion inspection and Monitoring, Wiley Series In Corrosion, R. W. Revie,
series editor, Wiley, New York, 2007, pp. 301–311.
L. S. Selwyn and P. R. Roberge, Corrosion of metal artifacts displayed in outdoor environ-
ments, in ASM Handbook, Vol. 13C, Corrosion: Environments and Industries, ASM
International, Materials Park, OH, 2006, pp. 289–305.
M. Tullmin and P. R. Roberge, Atmospheric corrosion, in Uhlig’s Corrosion Handbook, 2nd
edition, R. W. Revie, editor, Wiley, New York, 2000, pp. 305–321.
L. Veleva and R. D. Kane, Atmospheric corrosion, in ASM Handbook, Vol. 13A, Corrosion:
Fundamentals, Testing, and Protection, ASM International, Materials Park, OH, 2003,
pp. 196–209.

PROBLEMS

1. A carbon steel is exposed to a highly polluted, industrial-marine environment.

After one year of exposure, the penetration is 130 μm, and after 4 years, the
penetration is 190 μm. Assuming the bilogarithmic law applies, calculate the
following:
(a) The constants A and B in Eq. (9.2).
(b) The average corrosion rate, in terms of penetration per year, over a 40-year period,
assuming that environmental conditions do not change significantly during this
time period.
(c) The instantaneous corrosion rate, in terms of penetration per year, at 5, 25, and 40
years of exposure.

2. A weathering steel is exposed to the same environment as the carbon steel in

Problem 1. After one year of exposure the penetration is 25 μm, and after four
years the penetration is 28 μm. Assuming the bilogarithmic law applies, calcu-
late the following:
(a) The constants A and B in Eq. (9.2).
(b) The average corrosion rate, in terms of penetration per year, over a 40-year period,
assuming that environmental conditions do not change significantly during this
time period.
(c) The instantaneous corrosion rate, in terms of penetration per year, at 5, 25, and 40
years of exposure.