RESEARCH REPORT
TECHNICAL-ECONOMIC EVALUATION
OF AIR-POLLUTION CORROSION COSTS
ON METALS IN THE U.S.
to
AIR POLLUTION CONTROL OFFICE
ENVIRONMENTAL PROTECTION AGENCY
BATTELLE MEMORIAL INSTITUTE
-------
FINAL REPORT
on
TECHNICAL-ECONOMIC EVALUATION
OF AIR-POLLUTION CORROSION COSTS
ON METALS IN THE U.S.
to
AIR POLLUTION CONTROL OFFICE
ENVIRONMENTAL PROTECTION AGENCY
by
F. W. Fink, F. H. Buttner, and W. K. Boyd
February 19, 1971
BATTELLE MEMORIAL INSTITUTE
Columbus Laboratories
505 King Avenue
-------
TABLE OF CONTENTS
SUMMARY ,
INTRODUCTION 5
STUDY STRATEGY 6
Economic Framework 7
Conventional Corrosion Cost Evaluations 8
Applicability to Air-Pollution Problem 9
Derivation of Evaluation Formula .... 10
Evaluation Strategy 17
Technical Framework 20
Metal-Corrosion Mechanisms 20
Man-Made Pollutants 20
Susceptibility of Metals to Air-Pollution Corrosion .... 24
Metal-Protection Practices 29
Air Pollution Corrosion Costs 30
COMPONENT-SYSTEM SELECTION 39
Component/Systems Framework , 41
CONTEMPORARY AIR-POLLUTION COSTS, 1960-1970 46
Evaluation Procedures 46
Marginal Maintenance Costs 47
Shortened Lifetime Costs 48
Alternate Materials 49
Evaluation of Surviving Steel Component Systems 50
Steel Storage Tanks • 52
Highway and Railroad Bridges 56
Power Transformers 59
-------
TABLE OF CONTENTS continue.d
Evaluation of Surviving Steel Component Systems continued
Outdoor Steel Metal Work 61
Pole Line Hardware 64
Chain-Link Fencing 66
Galvanized Wire Rope and Cable 68
Power Line Transmission Towers 71
Air Pollution Damage Costs for Alternate Materials 73
Roofing Metals 74
Aluminum Siding, 1969 76
Self Weathering Steel 77
Stainless Steel 78
Summation of Annual Extra Corrosion Losses Caused by Air Pollution 79
ANALYSIS OF COST OF CORROSION DAMAGE BY AIR POLLUTION, 1970 to 1980 . 81
Economic Trends 81
Changes in the Amount of External Structures 83
Increased Use of Aternate Materials 84
Change in Corrosivity of the Atmosphere 87
Extra Annual Corrosion Damage Costs versus Pollution
Levels for 1975 and 1980 90
RECOMMENDATIONS 95
REFERENCES . 97
BIBLIOGRAPHY 101
APPENDIX A A-l
APPENDIX B B-l
-------
TABLE OF CONTENTS continued
LIST OF TABLES
Table
1 Sources in Millions of Tons of Sulfur Oxides for
1965 22
2 Seven Year Corrosion Results for Aluminum Alloys
Exposed to Industrial Atmospheres 27
. 3 Paint Costs for Protecting Steel 36
4 Annual Extra Painting Cost Factors for Protecting
Steel Exposed to Polluted Atmospheres 38
5 Interrelationship or Elements of a City 41
6 System Elements of Construction 43
7 Interrelation Between Building Systems and Components . 45
8 Component Systems which Survived Final Screening and
Were Selected for Detailed Study 51
9 Summation of Annual Extra Losses Due to Corrosion
Damage by Air Pollution to External Structures for
1970 80
10 Economic and Pollution Factors Used in Assessing
Probable Cost of Corrosion Damage by Air Pollution,
1970 to 1980 91
11 Summary of Estimated Annual Air Pollution Corrosion
Damage to Metals for 1975 and 1980 93
LIST OF FIGURES
Figure
1 Projected Growth in Population and In Power Generation
and Increase in Sulfur Dioxide Emissions with no
Regulation 82
2 Growth of Structural Steel and Plate Production and
Estimated Total Change in External Steel Expressed
-------
TABLE OF CONTENTS continued
LIST OF FIGURES
Figure Page
3 Trends in Sulfur Dioxide Pollution Expressed as
Percent 89
4 Cost of Air Pollution Damage to Metals Based on Changes
-------
SUMMARY
Two approaches were undertaken to develop a more realistic
assessment of the added cost of corrosion damage to the Nation resulting
from the exposure of metallic systems and structures to polluted atmos-
pheres, both presently and by 1980. The first approach was to correlate
marginal costs with air pollution levels in four metropolitan districts,
each having a different average level of atmospheric pollution, covering
the range from slightly contaminated to severe industrial. Because repre-
/
sentative regional statistics for assessing corrosion damage do not exist,
and the time required to develop them was found to be far beyond the limita-
tions of the project, the first approach was dropped.
A second approach was then devised, employing applicable national
shipment/value data from the U.S. Department of Commerce to compute average
pollution costs on a national basis. Data developed by the Census Bureau
in the Standard Industrial Classification (SIC) were cross checked where
possible with industry-association statistics. The results were then
applied to a formula, which relates value with air pollution damage to
metals in use, and which makes it possible to compute corrosion damage
accurately on a national scale.
The Department of Commerce statistics are broken down into
thousands of metal components of varying degrees of interest, from many
of no interest to some of prime interest to this study. Accordingly,
the full array was put through a series of qualitative and finally
quantitative screening steps to select only those systems having a high
potential of economic loss due to their numbers and their in-service
-------
Two calculations were considered for summing up the economic
marginal cost of corrosion caused by air pollution over the typical cost
in clean air service. The first method compares the extra amount of
protection and maintenance expense required in polluted atmospheres to
prevent serious corrosion attack. The second is the cost involved in the
shortened life of the system resulting from corrosion attack in polluted
air.
Individual corrosion costs were calculated for nine major
categories that survived the screening. These are regarded as most
sensitive to and most damaged by air pollution corrosion. The grand
total of all these categories was found to be $1.45 billion. Converted
to a per-capita basis, this comes to approximately $7.10 per person per
year.
To estimate the probable cost to the Nation for 1980, Battelle
reviewed the various factors presently influencing the total cost of
air pollution corrosion damage, predicted the trends to be taken by these
factors during the next decade, and integrated these effects into an
estimate for 10 years hence. Changes in the amount of external metal
structures subject to attack by alterations in the corrosivity of the
atmosphere were taken into consideration.
The major factors affecting the corrosivity of the atmosphere
are the level of pollution and the degree of moisture present. While
pollution is expected to change in the next decade, the moisture level is
not. From a corrosion viewpoint, the most important pollutants are sulfur
dioxide. Four estimates are presented for the sulfur dioxide levels in
-------
I. If there were no regulation of pollution, the projected
consumption of sulfur-bearing fuels will result in an increase in the
rate of sulfur oxide pollution of about 55 percent by 1980.
II. If today's state-of-the-art in pollution control is combined
with gradually increasing enforcement of air pollution regulations, the
percent increase in average pollution will be only 15 percent by 1975 and
dropping to 10 percent by 1980. This case takes into account the difficulties
of enforcing regulations with space heating and older power plants.
III. If the present strong public demand for cleaner air continues
to encourage the government authorities to accelerate both the research
in the control of pollution and the enforcement of the regulations, a
40 percent average reduction in sulfur dioxide pollution over present
levels could result by 1980. This case allows for some slippage in the
enforcement of the regulations where offenders are not able to respond
quickly because of local circumstances.
IV. If current legislation and that about to be enacted is
applied without exception to all users of fossil fuels, the sulfur
dioxide emissions will be reduced 60 percent by 1975 and continue at
that level through 1980.
For the four cases of pollution levels discussed above, the
corresponding changes in the cost of corrosion damage are:
I. Assuming a 55 percent increase in pollution, the annual
marginal loss on a per capita basis would increase about 30 percent,
namely from a current level of $7.10 to $9.22 for 1980. The corresponding
annual loss would increase from the present $1.45 billion to $2.1 billion
by 1980.
II. In the case of a 10 percent increase in pollution, the per
-------
loss for the Nation would increase during the decade by approximately
$0.3 billion to $1.73 billion by 1980.
III. With a reduction of 40 percent in pollution, the per capita
cost would show a significant drop from today's level of $7.10 to $4.36 by
1980.
IV. With a reduction of 60 percent in pollution, the per capita
cost will drop from the present value of $7.10 to $2.20 in 1980. The
annual loss for 1970 of $1.45 would be reduced to $0.5 billion or a
savings of about one billion dollars ($4.30 per capita).
Much depends on how the interacting factors, which affect the
costs of air pollution corrosion damage, follow predicted trends. The
most likely impact of pollution to the Nation's economy will range be-
tween an increase of $0.3 billion to a savings of $0.5 billion over
-------
INTRODUCTION
The Air Pollution Control Office (APCO) of the Environmental Pro-
tection Agency has been cognizant for some time of the damaging effects of
pollution not only to health, to the ecology, to the climate and to air navi-
gation (i.e., low visibility), but also to the increased corrosion of metals
and to the deterioration of materials in general. APCO has been developing
information on the burden to the Nation's economy of these many adverse
effects of pollution. To provide a more direct estimate of the cost to
the Nation's economy, specifically resulting from the increased corrosion
damage to metals by exposure to polluted air, APCO sponsored this investiga-
tion to develop information on the magnitude of such loss. This information
is needed as a part of the overall picture of the damage costs of air
pollution to guide both those responsible for establishing pollution controls
and to demonstrate to those in a position to originate legislation the
need for appropriating funds to provide for developing solutions to the
problems. In other words, the reduction of the corrosivity of the atmosphere
will involve expenditures for developing the technology of control and for
the enforcement of such controls. It must be shown that this expense will
be partially recoverable in terms of reduced losses from corrosion damage
to external metal structures.
It should be mentioned that atmospheric pollution may play a role
-------
cracking. This item is excluded from the investigation. Also indoor atmos-
pheres naturally reflect the pollution outside. Electrical contacts and elec-
tronic components may deteriorate and cause malfunctioning of equipment. This
situation is recognized but in order to confine the scope to more manageable
proportions is excluded from this study but is the subject of two other APCO
. A' (1>2)
studies.
Limited to those metals most subject to air-pollution corrosion
damage in the atmosphere, the overall objective of this investigation is to
develop a realistic assessment of extra service costs due to the corrosive
attack of air pollutants. More specifically, the major objectives can
be stated as:
(1) Assess the total economic loss to the Nation in dollars
resulting from the increased corrosion damage of externally
exposed structures or systems caused by the presence of
man-made pollutants in the atmosphere.
(2) Establish from predicted changes in population, industrial
activity, technology of pollution control, corrosion preven-
tion, external structures, and air pollution regulation the
probable size of this economic loss in 1980.
STUDY STRATEGY
Both economic and technical sources of information were reviewed
in developing the plan of investigation.
(1 2)
-------
Economic Framework
The economic framework of this study derives from conventional
corrosion cost studies of systems in general service. Air pollution is
only one of many corrosive elements in the "general service environment",
which may include wind borne mists, fumes, erosive dusts, fungi, and salt
particles, etc. This study narrows down to only the cost of the man-
made air pollution component of the array of damaging environmental com-
ponents in the service environment.
Although defined in this respect, the study expands beyond the
usual by including all service conditions wherein air-pollution attack
-------
Conventional Corrosion-Cost Evaluations
(3)
Corrosion-cost estimating was pioneered by Uhlig who based his
calculations directly on material lost in the corrosion process, and indirectly
on the cost consequences of such material losses in systems in service. In-
direct costs, he concluded, could not be estimated and were dropped. Direct
costs could be accounted for in several ways, namely marginal costs of
over design, coatings, water conditioning, maintenance, and parts replacement.
Subsequent authors essentially deal with elaborations on parts of
(4)
Uhlig s comprehensive calculations. For example, Keynes develops an
equation on the economic tradeoff, the indifference curve, between main-
tenance and replacement, based upon the "corrosion prone ness" of parts in
a given corrosive service environment. Harvey , in a similar consideration,
concludes that more economical protection can be provided with an "exterior
corrosion allowance", or extra surface material as a substitute for paint.
In an attempt to assess the relative damages caused by air pollution
(4)
on all materials, Salmon computed the cost of material lost in the
corrosion process, by a formula that assigns value to the lost metal. The
metal lost is computed in weight, valued at mill product prices plus the
added value (labor, expense, etc.) of converting the mill product into a
finished shape and installed.
Although Salmon follows an intriguing rationale to separate out
only air pollution corrosion, he tacitly accords equal value to all exposd
surfaces and all metal lost by corrosion. Even though he separates out the
cost of cleaning a smudged surface discolored by air pollution constituents,
his calculations appear to overstate the costs.
Uhlig appears to tacitly assume that if it weren't for corrosion, un-
coated plain carbon steels would do. Thus the cost of upgrading into alloy steel,
-------
corrosion. Because many of these extra measures are taken for extra corrosion
reasons, Uhlig's calculations appear to quite overstate the costs in terms
of this study.
Applicability to Air Pollution Problem
In this study, the objective is to calculate the real cost of air pollution
corrosion to metals in existing systems. Here the real cost is equivalent to the
decline in real value of such systems due primarily to air pollution corrosion, and
not to other forces of deterioration. This is a more limiting objective, chosen
to produce results comparable to real property valuation.
The value of real property as conventionally established or appraised by
professional assessors, is accepted by the courts, the tax collectors, and in the
market place. Such wide acceptance gives reality to professional appraisals.
Therefore if one is to measure a "cost to society" of air pollution, it should,
to be realistic, be acceptable to assessors and consistent with their appraisal
formulas.
It is evident that none of the previous formula for measuring corrosion
costs have premises compatible with appraisal formula. Thus, while previous estimates may
have meaning to engineers and others, they have little significances in the practical
world of real property value. For these reasons this study derived its own formula
-------
10
Derivation of an Evaluation Formula
Air pollution corrosion affects a part of the assessed value of real
property. Normally real property has two components, its capital value plus its
inherent ability to create x^ealth over its remaining lifetime. Thus, for example,
the value of a hotel is the sum of its discounted productive income, for the rest
of its useful lifetime, less maintenance and operating costs for the same period,
plus its capital value of the moment. In this case air pollution could affect
productive income, most visibly if it were a resort hotel suddenly beneighbored by
an oil refinery. The income of a residential hotel in a city might also decline as
air pollution increases. However, income for a commercial hotel might increase
with air pollution, as it signals increased commercial activity in the vicinity.
There is still other real property with wealth-creating capacity totally
insensitive to air pollution. For example, warehouses, service stations, office
buildings, and many other commercial establishments, particularly those serving
industry, are quite free of income delimitation due to local air pollution.
From all appearances, the affect of air pollution on real property income,
or utility, or another essential benefit can be positive or negative, but mostly
neither in the nation as a whole. Thus it is assumed here that this effect balances out
to zero, an assumption nonetheless worthy of further study on a regional basis.
The capital value of real property appears affected by air pollution in
two ways. First, it increases maintenance costs, by impelling the owner of the
real property to paint certain components more often, to regalvanize some more
frequently, or to replace them entirely. Second, it may shorten the assessed life
of the property or system, by causing either an earlier overall disfunctioning or
exhibiting evidence of excessive maintenance on components,or both. In the former
case, it is replaced earlier, and in the latter, written off earlier. In either
case, the margin of shortened economic life is a real cost of air pollution to the
-------
11
A third possibility .is to avoid the extra maintenance and shortened life
of real property resulting from pollution by substituting resistant materials for
i
the susceptible ones. This practice usually increases the original capital cost which
can be assessed to the cost of pollution.
In any case, air pollution can cause a loss in value of a real property
system, such as a bridge or warehouse, by promoting corrosion at some surface in
components of the system, a vital structural member or a downspout. Thus the problem
of deriving an evaluation formula is in dealing with extremes, i.e., to measure a
corrosion phenomenon at a surface (a microcosm) and relate it to an overall system
value (a macrocosm) where appraisal formulas and assessors' judgments operate. An
additional problem is to recognize that all surfaces of the same metal are not of
equal value in different systems, or in identical systems of different ages.
For example, in .the first case, the steel surface of a I-beam on a bridge
might have less influence on the bridge's value than the same amount of steel surface
on a cable or bolts on the same bridge, as far as the functional integrity of the bridge
is concerned. Similar comparisons can be drawn across systems, such as an eaves trough
on a house compared to a transformer casing. Corroding surfaces vary in value loss
depending on where they are located.
For example in the second case, the steel surface on a new, original
water drainage subsystem on a new building is more valuable than the same surface
on a replacement subsystem on a fully depreciated building. Because a new downspout
on a condemned building is as worthless as the building, its corroding surface repre-
sents no value loss whatsoever, except possibly in salvage value if recovered.
Still another condition affects the value loss of a corroding surface. If the
surface is on.a system component that is subject to such additional effects as erosion,
acid spills, denting and distortion forces, the value loss due to air pollution corrosion
is masked by such other deteriorating forces encountered in service.
Finally, there is the superimposed situation of rapid systems obsolescence.
If the rate of surface loss is slow compared to the rate of system devaluation,
-------
12
systems in this category suffer instant total depreciation upon first use, such
as tin cans, and garbage cans, in which case no value loss can be charged to
surface corrosion.
Thus the value of a surface and accordingly its corrosion cost varies
widely with the component upon which it appears, and the system in which that coirmonent
appears. Any value formula for systems, therefore, must make provision for
these differences. This is about the limit one can go starting with a deteriorating
surface and rationalizing its impact on a system's value. To develop a formulated
rationale one must start with the conventional assessor value formula for real
property systems and work back toward a surface.
Fundamentally, there are three general approaches to the appraisal process,
which in general are pursued as cross checks, and in specific are taken as basis
for judgment or points of departure in assigning value. The first approach is the
market data approach in which the value of a specific property is compared with the
value received in a recent sale of similar property in the same locale.
The second approach is the income approach in which the value is related
to revenue minus upkeep plus depreciated capital value of the property. The third
approach is the depreciated replacement cost approach, which is a variant of the
second approach, i.e. the same except it depreciates from present replacement costs
rather than original capital cost some years back.
In all cases average depreciation, increased by rough service or decreased
by benign service, plays a fundamental role in assessing value. Average depreciation
is taken by year to be equal to the original cost divided by useful average lifetime,
i.e., so-called straight line depreciation. There are three variants around this
base, namely, 150 percent or 200 percent declining balance method, and sum of the
years-digits method. The variants either extend or shorten the effective lifetime
-------
13
line depreciation is usually the starting point. It is represented by the assessor's
formula (1) for real property systems as follows.
where '<4- = value at present time, t.
V$t0 = original value at time of purchase, t
t = depreciation factor for the system at time, t.
= annual rate of depreciation of the system
= constant < \ I > \ — + *~-
' ' (NV
The annual depreciation rate is determined by dividing the original value
of the system by its lifetime years, according to standard tables (IRS procedure
tf)
62-21). The constant k is the judgment factor effectively used by the assessor to
factor in departures in the condition of real property at time, t, from the average
expected condition at that time. For example, evidence of heavy use of the property
raises k to a point above unity so that Vst falls below average. Evidence of light
use lowers k, so that V . rises above average, if a piece of real property becomes
St.
suddenly obsolete, suffers catastrophic failure, or total destruction, k approaches
f ... \ and V . approaches salvage value, or zero value in its intended purpose.
I*
Of course, the deleterious effect of air pollution would raise k. In
no case in the memory of assessors contacted in this study has an effective increase
in k been directly attributed to air pollution corrosion, as it might have been through
excessive wear, obsolescence, declining neighborhood etc. The reason is that real
property suffers clean air corrosion and any added effect of air pollution corrosion
is missed by charging up the observed condition to poor maintenance. Thus, among the
-------
14
air pollution corrosion of metals is lost among the secondary effects, and not
identified as such.
Because it is less important in the value accounting of equation (1),
it does not mean that air pollution corrosion is not finite. It means that air
pollution corrosion cost is seldom identified as significantly shortening the useful
life of real property. But if the effect is finite, it must appear somewhere else,
and it does — in the maintenance costs charged against the systems benefits and/or
wealth creating capacity over its lifetime.
Thus, if one is to quantify rigorously -air pollution corrosion losses, a
more precise concept and equation is needed.
Conceptually, a real property system can be viewed as an assembly of
components, each with its own susceptibility to corrosion, its own intrinsic value,
its own useful lifetime, etc. It is the component that is affected directly by
air pollution. Its separate decline accordingly in functional and capital value
indirectly affects that of the real property system. This concept leads to the
more precise general equation (2) as follows.
Equation
where V^-t = value °f the real property system at time, t,
= depreciation factor for the system at time, t.
= value of system components
and
V.*- LV
a Ct°
The summing up of the value for all components would lead to the value
of the system at t , or when the system is new. Despite the fact that components
per se might have a different depreciation rate than the system does, they are
normalized to the rate of depreciation of the system the components are in. That
-------
15
that of the system it is in. Components that inherently depreciate faster require
maintenance to keep them up, or periodic replacement. Maintenance outlays compensate
for the difference. For example, the original galvanized eaves and downspouts on
a house require repeated painting during the lifetime of the house to keep these
faster depreciating components up to the slower depreciation of the house. A
replacement component, however, takes an initial depreciation upon installation.
Components that inherently depreciate slower, such as copper eaves and
dox^nspouts, nonetheless carry the faster depreciation of the system. Over design
loss makes up the difference, as these still useful components ultimately come tumbling
down with the others under the wrecker's ball.
Therefore, a system is comprised of over designed components and under
designed components. The value of an over designed component is according to
equation (3).
\/
Equation (3). V
'ct
where V^J. = value of the component at time, t.
I
= value of the component new
= depreciation rate of the system in which the
component appears
u .^.
fe. = constant, \ "
The constant k is unity, where components, face an average service environ-
ment. It is less than unity for more benign service environments, and greater for
malign ones. Approaching — »-. for catastrophic failure conditions, it is noted
(cW\ TV
that the ratio \' Ji-'J is the limiting depreciation rate when that for the component
-------
16
The value of an under designed component is according to equation (4)
Equation (4).
\f
I -
where all factors are the same as in equation (3), except that
I,,.
the component's depreciation rate is a number greater than fG-i J
VcUr'S
To keep the component functioning at the value level of the system it is
in requires maintenance or protective coatings, such as paint, galvanized, terne
plate, etc. If maintenance restores the functional value of the component over
the lifetime of the system, a compensating term must be added to equation (4),
as follows, assuming maintenance cost is equivalent to the functional value of
the maintenance.
Equation (5). M = fl ^
where M = functional value of the maintenance coating
n = number of coatings required over the lifetime
of the system
p = the cost of coating.
Thus, the value at any time of an under designed component is according to
equation (6) as follows.
Equation (6)
-V,.-v
ill ^
fclci
HP
\
where [-7—J because of the effect of the coating, is a presumably
lower depreciation rate than Qj •) ,without the prime,
but variable in time depending on the rigour with
-------
17
, ,
of the owner. It may be zero, if V^j. is low and (jzL,\ is high, i
cbl/ c
The number of maintenance recoatings, n, is according to the judgment
in which
case other replacement is cheaper than maintenance or maintenance value is less
than maintenance cost.
Finally, the value of a system in terms of the integrated value of its
components is as shown in equation (7) as follows.
where C = number of overdesigned components
Cun = number of underdesigned components.
The effect of air pollution corrosion cost shows up in equation (7) by
increasing k's which diminishes the lifetime of components, and by increasing n
in the nP terms which increases maintenance.
Evaluation Strategy
Because the bulk of real property value in building systems is tied up
in unexposed metal and exposed nonmetal components, the effect of even high k's
and n's, for exposed metal components in equation (7), on the systems value is
relatively small. Thus assessors should be expected to relegate atmospheric
corrosion to secondary importance. However, the bulk of real property value in
many other construction systems is tied up in exposed metal components, in which
cases the effect of high k's and n's on these systems value is relatively large.
All systems exposed to the atmosphere fall somewhere between these
-------
18
estimate the effect of air pollution on the r's and n's of components comprising each
system, and then integrate their impact on the system's value is the procedure of the
project.
As it turns out, the k's are less important than the n's. The reason
is that if air pollution increases the k of an equal or overdesigned component
sufficiently to convert it into an effectively underdesigned component, it
immediately becomes a maintenance problem wherein n becomes the concern.
Thus the added cost of air.pollution corrosion is primarily the added
cost of maintenance, and most air pollution problems are maintena problems. How-
ever, there is a small components of cost which arises from the use of alternate,
more costly materials which are relatively immune to pollution effects. This intro-
duces a small simplicfication to the stretegy. It still remains that all exposed
components on all out-of-doors systems require identification, and an assessment
screening, followed by an evaluation of each components system that survives the screens.
The number of component systems is astronomical, particularly when
distinguishing similar component systems of different materials, e.g. copper
eaves and galvanized eaves. It is strategically impossible to evaluate them all.
Thus, a set of screens were conceived whereby broad categories of
component systems could first be examined, rejects made, and survivors kept. In
the second screen, narrower categories were considered, which meant the broad
category survivors were translated into many more narrower categories, and screened
again.
The procedure repeats through three progressively narrower screens,
until finally individual component systems of specific metals survive. They
then fall under a quantitative assessment of air pollution corrosion cost and are
finally integrated into a total corrosion-loss impact on metal containing systems,
-------
19
The Census Bureau, U.S. Department of Commerce, provides a convenient
identification of component systems and a useful categorization of them at eight
levels of specificity, from division categories, through six digited categories,
from 2-digit to 8-digit. In addition, its tabulations provide values and units
shipped by categories. These data provide basic values of component systems
shipped by year.
The "Depreciation Guidelines and Rules", provide average service
lifetimes for systems and components, by SIC classifications. Thus average
depreciation rate can be determined from in-situ value of component systems,
computed from SIC data, divided by average lifetime, taken from IRS Publication
No. 456. The computation is described in more detail in a subsequent section.
The determination of the k's and n's, by component system, is a technical
calculation based on known corrosion characteristics of metals as a function of
air pollutants and their concentrations. For these determinations, the following
-------
20
Technical Framework
Out of the vast data and theory of corrosion, only that pertaining
to the air-pollution problem is applicable to this study. Many of the
numerous metal corrosion mechanisms studied, and metal-protection practices
developed deal with more severe corrosion environments than would be
normally expected in the average living atmospheric environment.
Even metals exposed to clean air environments may suffer
sufficient damage to require protective measures notably painting,
galvanizing, or otherwise providing a protective coating.
Metal-Corrosion Mechanisms
A moist climate is essential for the atmospheric corrosion of
(8,9)
metals, such as iron and zinc. Corrosion attack on iron begins at a
threshold humidity of above 80 percent and increases proportionately
thereover. Once a rust coat has formed, however, the attack will continue
at some lower threshold humidity. Added to humidity, air pollution
synergistically increases the rate of attack, and further reduces the
threshold humidity.
Man-Made Pollutants
The most commonly found primary air pollutants are carbon monoxide,
sulfur oxides, hydrocarbons, nitrogen oxides, and particulate matter. Ozone
and other oxidants resulting from photochemical reactions are common secondary
pollutants. Of the pollutants, sulfur oxides have the most pronounced
accelerating effect on corrosion. For example, sulfur dioxide will rust iron
(10)
and attack zinc at relative humidities as low as about 50 percent. As
its concentration level is increased, sulfur dioxide will take over a greater
-------
21
Oxides of nitrogen have not been studied as extensively for their
influence on corrosion. It is known that a small portion of the oxides of
nitrogen, in the presence of water vapor, will form nitric acid. This may
react with traces of ammonia and be absorbed by hygroscopic particles which,
as discussed below, are corrosive.
High levels of sulfur oxides often are accompanied by high levels
of particulate matter. SSmples of fine solids taken from contaminated air
are found to contain particles of ash unburned fuel, tarry soot, grit,
road dusts, incinerator residues, etc. A typical analysis will show
organics, sulfates, nitrates, iron, lead, and traces of other metals.
The hygroscopic salts present in particulate matter, particularly in the
presence of sulfur dioxide, causes accelerated attack on iron, zinc, and
other commercial metals. The effect of particulate matter often is minor,
when the air is free of pollution.
As will be seen from a study of Table 1, the major source of
sulfur dioxide pollution is from the burning of high sulfur coal and oil
by power stations and industrial plants. These are the fuels that are least
expensive and their continued use has been a matter of public concern for
some time. Pollution from the incineration of waste and from the exhaust
of internal combustion engines adds to the total, but is of far less
-------
22
TABLE 1. SOURCES IN MILLIONS OF TONS OF SULFUR
SOURCES IN MILLIONS
OXIDES FOR 1965 (-13-)
Source
Transportation
Industry
Electric Power Generation
Space Heating
Refuse Burning
Tons x 106
0.5
8.7
10.2
3.4
0.2
23.
Percent
2.2
37.8
44.4
14.8
0.8
-------
23
The amount of man-made pollutants which are emitted into a metro-
politan district can be controlled to large extent. However, the dis-
persion and dillution of such pollutants is more difficult to control,
since the natural ventilation provided by wind, etc., is one of the major
(14)
factors. During a temperature inversion or weather condition resulting
in trapped air masses near ground levels, the concentration of pollutant
will increase markedly. Usually these periods, which may only occur 1 or
2 percent of the time, are accompanied by high humidity.
Based on general experience, the corrosivity of the atmosphere,
where sulfur oxides are the chief contaminant, can be expected to increase
several fold in such a period. However, actual data to show the day-to-day
correlation of short periods of high pollution levels with the resulting
acceleration of corrosion do not appear to be available. Abnormally high rates
of metal corrosion at areas just downwind from an emission source have been
observed. Here the stack height was insufficient and the diluted flue gas
descended down to ground level during stagnant weather periods.
The condensate depositing on metal surfaces such as iron and zinc
will be much more acidic than that corresponding to average pollution.
The amount of sulfate and other contamination in the rust coat on steel
will be increased and tend to affect adversely the corrosion rates during
-------
24
Susceptibility of Metals to Air-Pollution Corrosion
For some metals, particularly steel and zinc, there is a decided
increase in the rate of attack when they are exposed to industrially con-
taminated moist atmospheres as compared with that experienced in rural
clean air. The increased tendency to corrosion when the air is polluted
can result in a significant reduction in the service life of an external
metal structure or system or it can lead to higher costs of maintenance
and repair to insure the same performance as the structure would have in
an uncontaminated atmosphere.
These considerations were used in the selection of each metal for
inclusion in this study. More precisely, the candidate metal must satisfy
each of the following conditions:
(1) Air pollution causes the candidate metal to corrode at
much higher rates as compared to rural clean air.
(2) The life of a structure or system made out of the candidate
metal is significantly shortened by the increased corrosion
in polluted air or it requires greatly increased protection
and maintenance expense to prevent early corrosion failure
of the item in question.
(3) The metal systems or structures, in which the candidate
metal is the major component exposed to the atmosphere,
has high in-place value.
For some metals, Item 1 is satisfied, namely that the rate of
-------
25
some cases, significantly affect the life of the structure or system of
which it is a part as required by Item 2. Even when both Items 1 and 2
are satisfied, some of the applications of the metal may not be significant
from an economic standpoint.
Copper. An important characteristic of copper and its alloys
is its excellent resistance to the atmosphere. Typical corrosion rates
are of the order of 1.3//m/yr (0.05 mpy) in industrial atmospheres as compared
with O.S^m/yr (0.02 mpy) in rural air. ' ' ' Thus while there is some
acceleration in attack resulting from pollution, the higher rate is not signi-
ficant in a damage sense. The higher attack in polluted air usually results in
the more rapid formation of a patina on copper which in turn tends to slow down
the attack. Typically, a copper roof or other component outlasts the building
or structure of which it forms a part, whether the air is polluted or not.
In summary, one concludes that while the corrosion rate for copper
is increased by pollution, and the in-place value in exterior service is
high, neither the maintenance cost nor the life is seriously affected in most
cases by air pollution. Accordingly, the value of copper components are
negligibly sensitive to air pollution. Thus, copper was excluded from the
survey.
Lead. Lead is very resistant to atmospheric corrosion whether
the atmosphere is polluted or not. In a badly polluted atmosphere the
rate of attack may vary from 0.3 to 0.8/im/yr (0.01 to 0.03 mpy). Rural rates
(9)
are about 0.3y
-------
26
service. Service life of lead cable, used in overhead telephone distribu-
tion, is excellent. Lead was not selected for inclusion in the study
because of its excellent resistance to polluted air.
Stainless Steel. The high chromium and chromium-nickel stainless
steels are very resistant to atmospheric corrosion. In polluted atmospheres,
they may become stained and soiled. However, by cleaning and polishing
they can be readily restored to their original luster. Stainless steel
architectural components, installed in the early '30s on the Chrysler
building in New York are virtually unattacked in spite of the severe
industrial pollution in the vicinity. In view of this and other experience,
stainless steels were eliminated from the list of candidate metals.
Stainless may be chosen over aluminum because the latter has a
tendency to pit in severe industrial atmospheres.
Aluminum. About one-fourth of all the fabricated aluminum
produced is used in external applications such as in architectural. In
general, aluminum and many of its alloys have excellent resistance to the
atmosphere.
If the atmosphere is polluted, the surface becomes covered with
(13
dirt and soot and the aluminum tends to become mottled and pitted. In time,
the surface will appear roughened although there is no general thinning.
After an exposure of seven years in a polluted atmosphere, as depicted in
Table 2, pits as deep as 0,35nm(13. 8mils) were obtained. Other results shox* that
most of this pitting occurs during the first two years after which maximum
penetration increases only slightly. After 20 years' exposure, the depth
-------
TABLE 2. SEVEN YEAR CORROSION RESULTS FOR ALUMINUM ALLOYS EXPOSED TO INDUSTRIAL ATMOSPHERES
(17)
Corrosion Rate
Depth of Pitting
Alloy State College ' Richmond^ ' Chicago^ ' Widnes ' Richmond
Chicago Widnes
1 1 QQ
1 1 O c TT 1 /
JL JLJ J — tilt
3003-H14
ono/i wzf.
juu4— ruo
•3 n o/i —H "? A
cl SA-fiHA
cnn^—w?/.
6061-T6
AOA^—TA
//m/yr
"
0 O£A
0.071
-~
0 0££
0 0£A
0.069
mpy
Onno t;
0.0028
0 0096
OOO9 ^
0.0027
/l/ m/yr
/ ' "'
OO9 ^
0.48
n m
U. J J.
n AI
0.38
OO c:
. Z.)
mpy
n 001
0.019
O 090
0017
0.015
Om n
x/m/yr
0 SQ
1.12
1 77
1. 31
1 AO
1.70
i *3n
1. jU
mpy
0 09 T
U.UZ3
0.044
On^/.
0 OS"i
0.067
On^ i
. Uj 1
(hut re
1 9?
3.83
9 71
^ • J 1
— — __
9 70
2.62
i ^n
1. JU
mpy
0 OA8
0.151
o noi
0 1 Ofi
0.103
On^Q
. u.?y
mm
0 071
u . u / i
0.094
O OQ1
0099
0.10
01 1
• Lj
mils
" 8
^•5
3.7
Q f.
J . D
Q 7
4.1
c 1
J . 1
mm
0 90
0.18
01 o
. lil
0 9A
0.14
01 T
. 1J
mils
7 «
6.9
A 7
"t . 1
Q -3
5.6
R T
J. J
mm
Qc;
.26
9 ^
. / J
00
.27
o fi
. /o
!
mils
1 -3 Q
10.3
Q 7
.7 . /
10 Q
10.6
in Q
J.U. y
(1) Rural atmosphere, (ASTM STP 435, see Bibliography).
(2) Mild industrial. (0.01 ppm SO , avg.).
(3) Industrial (0.14 ppm SO , avg., RH, 80%).
-------
28
In rural clean air, the attack is very slight and pitting is
negligible. A rural rate of attack as presented in Table 2 is 0.071yL/m/yr
(0.028 tnpy) compared to an industrial rate as high as 3.83 X/Wyear (0.15 mpy) .
In architectural and in structural applications aluminum alloys,
if properly selected, will give years of service in polluted atmospheres.
Many examples can be cited, such as the Empire State Building and the
Washington Monument, where aluminum is still in service after 30, 40, 50
or more years.
It is concluded that increased corrosion attack due to air pollution
is not a significnat factor in the life of externally exposed aluminum
structures and components. Therefore, pollution costs, based on increased
maintenance of aluminum, were not included in this study. However, the effect
on air pollution corrosion costs of employing maintenance-free aluminum to
replace steel in highly contaminated atmospheres is considered.
Zinc. The corrosion rate of zinc is markedly increased by air
pollution. For example, the rate of attack, for 8-year exposure periods,
was found to be 1.04l/m/yr (0.041 mpy) in rural State College, Pennsylvania,
(19)
compared to 5.79,ym/yr (0.267 mpy) in New York City. Since zinc is not a
structural metal coating on steel, the main influence of pollution is to reduce
the life of the coating. Galvanized steel sheet with a 2 oz/sq.ft. coating on
each face has a 0.0864 mm thick coating (3.4 mils). Theoretically this would
last 83 years in State College and 12.7 years in New York. Partly since the
coating is not uniform in thickness, rust spots are typically observed in
practice at 35 to 50 years in rural atmospheres and 6 to 10 years in severe
industrially polluted areas.
Even in industrial areas, with the much shorter life of the coating,
galvanized steel is widely used because it is a relatively inexpensive
-------
29
Zinc in the form of galvanized steel, has been included in'the
investigation.
Steel. Steel is used in numerous external structures and systems
because of (1) its strength and economy, and (2) its ready availability in
a variety of shapes and forms. A large portion of the Nation's external
steelwork is exposed to industrially polluted atmospheres. As compared
with clean rural air, the corrosion rates in contaminated air are
accelerated greatly. At rural State College, Pa., for example, steel in
7-year exposures corroded at 0.013 mm/yr (0.52 mpy), whereas in New York
(19)
corrosion rates were as high as 0.065 mm/yr (1.60 mpy). .These corrosion
rates, determined over longer periods of time, will be lowered slightly
but not significantly, especially if the residual copper content of the
steel is low.
In view of the fact that steel is the major metal of construction
for external structures and since it has a propensity for more rapid
corrosion when the air is polluted, it was selected for detailed study
in this program.
Metal-Protection Practices
Some metals are highly resistant to the atmosphere, whether it
is polluted or not. Aluminum and stainless steels develop thin continuous
oxide films which confer protection to the underlying metal. Copper and
self weathering steels develop corrosion product films which retard further
(16) (20)
attack.
Ordinary carbon steels, on the other hand, require some form of
(21) (22) (23) (24)
protection. This can be, a metal coating such as zinc or aluminum
which is anodic to the steel, a paint coating, such as an epoxy. or vinyl
-------
30
rust preventive over a phosphated surface. Many other systems could be
mentioned. The most common methods of protection for external structures
are hot dipped galvanizing and painting or a combination of both.
Some aluminum alloys are not resistant to the atmosphere, but
their corrosion performance may be upgraded by cladding with a layer of
pure aluminum or an anodic alloy. Aluminum is sometimes anodized and dyed
for esthetic reasons. This also improves its corrosion behavior.
A recent development is electrocoating, which is used on both
aluminum and steel. High density, pore-free coats of f luor opolymers or
acrylics are available with excellent resistance to the atmosphere.
Air Pollution Corrosion Costs
Air-pollution costs are centered around the extra cost of
maintaining a protective coating, which may be paint or zinc, on steel
components. In a corrosive atmosphere, the major function
of the protective coating is to provide protection although there is some
cosmetic benefit as well. However, it is usually not feasible to assign
a portion of the extra maintenance cost to this latter factor. Thus the
extra painting costs on steel or galvanizing, as a function of air pollution,
are most relevant. To avoid either maintenance costs or early replacement
resulting from pollution damage, alternate materials resistant to
contaminated atmospheres may be employed.
Protection of Steel Surfaces by Painting. Steel surfaces are
most commonly protected from the atmosphere, whether it is polluted or not,
by applying a suitable paint system. Considerable skill must be exercised
in the selection, application and maintenance of an organic protective
coating to effectively control corrosion by polluted atmospheres. Because
-------
31
time, actual maintenance costs are often higher than need be. For example some
companies only paint when they have monies in the budget. At this time costs
may be greater than if painted earlier because of the additional amount of surface
preparation required to remove corrosion products. In addition early failure may
result from the use of low cost paints having less protective power. All these
factors give rise to aggravated costs, however, they can be credited to pollution
since they might not have occurred if pollutants had not been present.
High quality paint formulations are required to obtain satisfactory
(21 27 29)
protection of steel structures exposed to badly polluted atmospheres. ' '
Some measure of the aggressiveness of the atmosphere can be obtained by noting
the behavior of bare steel, whose corrosion rate, as discussed elsewhere, largely
(28)
is a function of the moisture, sulfur dioxide and particulate matter present.
Similarly, the life of a paint coating on steel is adversely affected by these
environmental conditions. In addition, ozone, oxides of nitrogen, and solar
radiation tend to promote weathering of the paint film.
There are two common mechanisms by which the life of a paint film is
shortened in polluted air. One is the local corrosion at holidays in the paint
film. The adjacent painted steel can serve as the local cathode and attack
and undercutting of the paint is accelerated at the holiday. A similar mechanism
involves ion transport through the paint film, and subsequent rusting at the
paint-to-steel interface. A major effort in the development of protective paints
for steel surfaces exposed to atmospheric pollution is concerned with delaying
(29.5)
the onset and spread of this underfilm attack. The second form of paint
failure results from the weathering of the exposed paint surface and its
soiling by particulate matter.
Unlike the rusting of steel, which can be expressed quantitatively
in terms of metal loss or depth of attack, there is no precise method of
recording paint failure. The appearance of rust streaks, blisters, cracks,
or crazing, loss of the top coating and loss of color or surface character are
-------
32
appearance of rust streaks, underfilm attack, or the loss of the top coat is a
sign that repainting is in order. Where maintenance is delayed, excessive
resurfacing costs are reauired.
The annual cost of corrosion protection of a steel system by
paint is greater x^hen the atmosphere is polluted. There are several
approaches to the problem of arriving at the difference between this cost
and that of pollution-free, or clean rural air. Procedures given considera-
tion for use in this investigation are enumerated below:
(1) Establish the original cost of painting a steel structure or system
and convert it to an annual cost, using the corresponding paint
life, in years, for (2) rural and (b) polluted conditions,
respecitvely. The difference between these two figures is the
annual cost to be charged to pollution.
(2) Calculate the cost of painting and repainting the steel structure
to provide underfilm corrosion protection for its normal life
span in (a) rural and (b) polluted atmospheres. The difference,
converted to an annual cost, is charged to pollution.
(3) Establish the cost of increasing the thickness of the original
paint sufficiently to delay underfilm attack in the polluted
atmosphere equivalent to that experienced with a nominal paint
film in rural air. This extra cost divided by the life in years
will be charged to pollution.
(4) Compare the cost of maintaining a protective paint system
on aluminum and on steel in a polluted atmosphere. The
difference in the two annual costs can be ascribed to the
-------
33
The first procedure has merit and is a practical method of
comparing costs in clean and polluted atmospheres. It was not chosen
since it is more usual to repaint and maintain a system or structure to
avoid early failure where the atmosphere is polluted.
The third procedure assumes that the thickness of the paint
is the only factor in the failure of the steel surface. Actually the
initial preparation of the steel surface, prior to painting, also is
an important factor in the tendency for steel to corrode more rapidly
at the paint-to-metal interface where there is pollution. Accordingly,
this method is considered only partially applicable.
There is some merit in the fourth method, namely comparing the
effectiveness of the same system applied to steel and to aluminum. However
the data required to apply this procedure do not appear to be available.
The second method was chosen for the purposes of this
investigation. The comparison of maintenance painting costs in clean and
polluted air is in accord with procedures employed in the real world of
engineering. This follows from the fact that the useful economic lifetime
of external structures has been established on the basis of other criteria
such as amortization for business, tax, financing, and other purposes.
To avoid the problems involved in premature failure of such investments as
are caused by pollution, maintenance procedures are established to insure
normal life.
In some cases maintenance cannot be provided. For this situation,
the replacement costs involved in early failure in polluted air were
-------
34
computed on an annual basis.
In order to arrive at the extra annual maintenance expense
caused by pollution, as outlined in the second procedure it was necessary
to follow a series of steps.
First, the total amount of each item in use was established.
Second^ this amount in tons or other units was converted into exposed surface
area. Third, the portion of this total area exposed to polluted air was
estimated. Fourth, the annual extra cost of protection by paint, per unit
area, was calculated for each system. Fifth, the area (3) and annual cost
figure (4) were combined to obtain the total loss for each item to the
Nation.
It is possible to estimate from past shipment records the total units
and/or weight of any particular system presently in use, the procedures
used in these calculations are described in Appendix A. Sometimes the
total amount of the system in use was available in the number of units or
could be derived from the total capacity, as in the case of tanks. In most
cases, the amount in use was expressed in tons.
For each type of structure the thickness range of the steel used
in typical designs was established. ; Knowing the thickness, one
can convert tons to area. For steel plate as used in tanks, °r ibr sheet in roof ing
and siding, the exposed area on one side only is used. For structural
steel, as in bridges, the total surface area is used.
Each system was considered separately, in arriving at an estimate.
The typical manner in which the system was employed also was taken into
consideration. For example, above ground external storage tanks in petroleum
-------
35
transmission towers are mostly in rural areas and only a small portion
are exposed to pollution.
The total cost of applying a paint system to a steel surface can
be estimated by combining the cost of surface preparation, materials, and
labor. A large number of paint references were searched, but there were
only a few which reported actual costs. The range of values found for
high quality metal protective paints are given in Table 3, Part A. Some of these
have been taken from recent sources and are not too different from current
costs. During the last decade, even with more frequent use of labor-saving
procedures, the main portion of the expense that has increased is labor. The cost
of protective paint systems has not increased markedly, and some systems such
as epoxies are actually lower in cost than when they were first marketed
some years ago.
The paint cost selected for making the calculations are presented
in Table 3, Part B. The original painting usually is higher in cost
because it includes surface preparation, such as blast cleaning. For
most systems, the first cost was taken as $0.50/sq ft and the repaint cost
at $0.40 sq ft. While such a high quality system will protect steel for
12 years in rural air, the life in polluted air will vary with the applica-
tion. Elevated water tanks, on the average, are not exposed to the highest
level of pollution. For a good quality coating, 8 years is taken as typical.
In petroleum service, tanks are normally exposed to a high level of pollution
and the life may be shortened to 5 years. In chemical and other industrial
service, a higher quality coating job is usually specified than in the
petroleum field. The choice of $0.62 per sq. ft. reflects this difference.
-------
36
(21,26,31,32,33,34)
TABLE 3. PAINT COSTS FOR PROTECTING STEEL
A. Range of Published Values
Range of Cost
Description Dollars/Square Foot
Sand blasting 0.10 to'0.20
Primer 0.09 to 0.18
Top coat, each 0.11 to 0.21
Complete coating - 4 to 6 mils 0.30 to 0.75
B. Application Costs in Dollars Per Square Foot
Chosen for Structural Steel and Metal Work
Life Years
System
Water Tanks
Petroleum Tanks
Chemical, etc. Tanks
Bridges
Outdoor Metal Work
Power Transformers
Outdoor Lighting Fixtures
Original
0.50
0.50
0.62
0.65
0.50
0.50
0.50
Repaint
0.40
0.40
0.46
0.65
0.40
0.40
0.40
Rural
12
12
15
15
12
12
12
Polluted
8
5
7
8
8
8
-------
37
repainting is not always done at the most favorable time. Thus considerable
surface preparation and spot priming is necessary. In view of this, the
repaint cost is taken as equal to the original.
The annual extra cost per unit area to be charged to air pollution
is calculated in the following manner. First the typical useful life of the
structure is established through the use of the IRS table. Next the cost of
providing paint protection for this period of time under rural conditions
is computed. Similarly the higher cost required for the same life period under
polluted conditions is computed. The difference between these two costs is
converted to an annual basis by dividing the normal life of the structure.
These factors, reported in dollars per quare foot per year, are shown in
Table 4.
Protection of Steel Surfaces by Galvanizing. Galvanized steel
structures and galvanized roofing and siding often are painted to extend
the life in polluted atmospheres. The extra costs charged to air pollution
can be developed in a manner similar to that already described for painted
steel. Some examples are given in Table 4.
Some types of galvanized steel components are normally not main-
tained by painting. For example, galvanized fencing is usually replaced. Wire
rope and cable, after the zinc coating is gone, rapidly loses strengh as a
result of the steel corrosion. In other words, its useful life has been
shortened by pollution and it also must be replaced. Similarly pole-line
hardware in polluted areas tends to fail earlier than the wooden pole
itself. This involves an extra replacement cost that is not experienced
-------
TABLE 4. ANNUAL EXTRA PAINTING COST FOR PROTECTING STEEL EXPOSED TO POLLUTED ATMOSPHERE
Life in Years
Paint
i
Level of Pollution'
Zinc Coating
Steel System System Clean Polluted Clean Polluted j Degree
Tanks
Water 50
Petroleum . 11
Chemical 11
Bridges ! 30
Power Transformers 30
Street Lighting 20
Outdoor Metal Work
Doors and Frames 45
Window Sash, Frame 45
Structural Steel 45
Roofing, Siding '.. 45
Galvanized Roofing^) 45
and Siding (B) 45
Coil Coated Steel
Coated Galv. Steel ;
Transmission Towers 30
Pole Line Hardware . 30
Chain-Link Fence ; 20
Chain-Link Posts 1 30
i
12 8
12 5
15 7
15 8
12 8
12 8
12 6
12 8
15 7
15 10
10
10
6
10
:> ! Average
i [High
! |High
(Average
| Average
SAverage
; !
i 1
| High
•Average
i
35 10 High
35 15 : Average
35 15 .Average
35 15 'Average
20 8 'Average
35 15 'Average
Range in ppm
of Avg.
SO? (35, 39} dean
.03-. 06 ; 1.77
.08-. 18 i .46
.08-. 12 : .45
.03-. 06 , 1.30
.03-. 06 1.10
.03-. 06 ; .77
•
.08-. 12 ' 1.60
.03-. 06 i 1.60
.08-. 12 . .34
.03-. 08 .34
!
.03-. 06 j
.03-. 06 -;
.03-. 12 ;
.03-. 12 :
Paint Cost
$/So.Ft.
Extra Annual Main-
tenance, Cost in $/
Differ-; Sq,.Ft.
Polluted ence
2.60
.98
.88
2.64
1.60
1.10
3.10
2.35
2.10
1.30
.70
.70
.1074
.70
.83
.52
.43
1.34
.50
.33
1.50
.75
1.76
.96
.70
.70
.107^
.70
/Yr.
Steel" Galvanized
0.0167**
0.0473
0.0391
0.0447
0.0167**
0.0167**
0.0333
0.0167**
4
0.0392
0.0213
0.0233
0.0233
0.0054
0.0233
CO
00
* Difference on extra cost above rural divided by system life.
-------
39
For those cases where painting was employed to extend the life of
a galvanized structure, the same procedures were used as those described
for painted steel. However, it was assumed that the first painting was
not required until the rust started to appear at weathered spots in the
zinc coating.
In the case of wire rope, the cost of corrosion damage by
pollution was established by dividing the cost of the wire rope by the
years of service in polluted and in clean air respectively. The difference
in the annual cost for polluted environments compared with that for rural
service was charged to pollution.
For pole-line hardware, the usual practice is to regalvanize the
items and return them to service. This procedure was made the basis for
the calculation. In general where there are variations in the application
of the basic method, these are explained in the individual sections
where the specific system is discussed.
COMPONENT-SYSTEM SELECTION
Up to this point, considerations have centered on the behavior
of metal surfaces, protected and unprotected, as exposed to outdoor air
environments. It shifts now to the systems, subsystems and components
in which metal surfaces appear.
According to the evaluation formula, the cost to society of
deterioration of a surface depends on the value of the system in.which it
-------
40
corrosive environments. At one extreme, such as the outside surface of a gaso-
line can, the cost of surface deterioration is'effectively negligible. It is a
low-value surface, the corrosion of which does not diminish either the utility of
the container, nor its expected life.
At the other extreme, such as a contact point in an electrical device,
the cost of surface deterioration is effectively great. It is a high-value
surface, the corrosion of which compromises the utility of the device, invokes
maintenance costs, and shortens.the assessed life of the device. Between these
extremes are numerous systems, subsystems, and components of greater or lesser
values linked to the integrity of contained metal surfaces; and that integrity
linked to their corrosion proneness to air pollutants.
The problem therefore is to identify important surfaces, to place
value on them, and limit the quantification of corrosion loss to only the impor-
tant, high-value ones. Placing an accurate value on surfaces, according to thn
foregoing examples and formulae, would require a tenuous evaluation process that
considers (a) the surface's function, (b) its location on a component, (c) the
'intrinsic value of the component and its effect on the economic operation of the
system it is in, and (d) the total value of the systems in operation in the
Nation at the time in question. Because there is no such evaluation process in
existence to our knox^ledge, the value of surfaces have to be qualitatively
estimated and compared by inferences drawn from published data on hard systems,
published primarily by the Census of Manufactures.
Hard systems originate under two broad census categories; Division C--
contract construction, and Division D--manufacturing. Cities are the
product of contract construction xvhere pollution is mainly concentrated and where
most manufactured products (components and systems) reside. Other divisions in
-------
41
service environments for them. That is, all divisions, agriculture, mining,
transportation, etc., all use systems originating in Divisions C and D--SO do
Divisions C and D, for that matter. The point is that component/system defini-
tion is found only in statistics for construction and manufacturing.
Component/Systems Framework
Cities contain two broad categories of constructed structural systems,
infrastructure and to coin a term, "economic structures". Both are built by
the contract construction industry. "Infrastructure" includes general construc-
tion for transportation and communications, power facilities and other public-
service buildings, and capital equipment. "Economic structure" is here defined
as containing all evident buildings used by industry, commerce and households.
All elements of infrastructure and economic structure comprise the real estate
of the city; Division C of the SIC anatomy.
In, on and around the real estate of a city resides manufactured
products or systems; Division D of the SIC anatomy. Thus, there are three
'varieties of systems of concern in this study, related to the city, the statis-
tics and each other as matrixed in Table 5.
TABLE 5. INTERRELATIONSHIP OR ELEMENTS 'OF A CITY
Economic
Structure
Infrastructure
Division C -
Contract Construction
A.
Building
Construction
B.
General
Construction
Division D -
Manu fa c t u ring
C.
Manufactured
Products
C.
Manufactured
-------
42
All constructional and many manufactured systems are exposed to some
extent to air pollution. More specifically, some of the components comprising
constructional and manufactured systems are more constantly exposed than others.
Therefore, it is the component level that is meaningful, and where surfaces
begin to become identified, along with their exposure, service behavior and their
maintenance requirements.
Building construction and general construction breakdown into eight
system elements as shown in Table 6. Manufactured systems components, as produced
by manufacturing industries, are given by SIC 25, 33, 34, 35, 36, 37, 38, 39.
All reside on and around the two types of construction as capital equipment.
Although it is possible to be quite specific in ranking these products, according
to value of volume produced, and tonnage of metal consumed, their distribution
into capital equipment by industry is specific, in only a relatively few cases
in the SIC data breakdowns.
At this point, there appears to be two separate somewhat independent
areas of concern, each to be reduced to sufficient specificity to apply the value
assessment equations. However, skipstepping ahead and taking advantage of
hindsight, it turns out that the most important systems and components in the
Division D--manufacturing areas--were related to construction systems. The manu-
factured items unrelated to construction systems turned out to be relatively
siuMll-volume; or in service in protected environments where air pollution is
absent, such as steel furniture (SIC 25) and computers (SIC 36); or in service in
aggressive environments where air pollution is the least damaging environmental
element, such as automobiles on salted roads (SIC 37), and chemical equipment
(SIC 35). Many more examples can be cited.
Thus, the two major areas become more and more related, as the manu-
-------
43
TABLE 6. SYSTEM ELEMENTS OF CONSTRUCTION
Building Construction
(1) Residential Building
i
Single family
Multifamily
Other
(2) Industrial
Chemical plants
Metallurgical plants
Food processing plants,
etc., (SIC major groups 19-39)
(2) Institutional Building
Religious
Educational
Hospital
Social and recreation
(4) Commercial
Offices
Warehouses
Stores
Restaurants
Garages
General Construction
(5) Transportation Structures
. . Railroads
Highways
Bridges
Overpasses
(6) Utilities
Electric power
Gas
Telephone and telegraph
Water
(7) Public Airports
(8) Military Construction
Note: Military construction is not exactly infrastructure, but is related in the
-------
44
a better matrix representation is as indicated in Table 7. Each cell represents
a component/system combination. For example, (a) steel tanks in a chemical
installation, (b) roof siding and drainage in single family dwellings, (c)
power distribution transformers in electric power utilities, etc.
For practical reasons that after all gives dimension to the problem,
census data is not neatly kept in terms of all of these cells. For large-item
components, it is. But for most, the cells are in effect grouped; in many
cases, it is left to the investigator to deduce breakdowns of component
consumption by system, an intellectual exercise that all too often, for
lack of reasonable access or precision information, degenerates into mean-
ingless speculation.
For these reasons the later computations on corrosion loss are
oriented around components, with the service environment of the various
systems coming in to establish technical judgments on the probable corrosion
behavior of component surfaces accordingly. The details of the procedure
-------
45
TABLE 7. INTERRELATION BETWEEN BUILDING SYSTEMS
AND COMPONENTS
^N. Components
^"^V.
Construction ^v
Systems ^s.
Industrialized Installations
26 Paper
28 Chemicals
29 Petroleum
30 Rubber
33 Primary metals
34 Fabricated metals
35 Machinery, excluding
electrical
36 Electrical machinery
37 Transportation equipment
Other Installations
Single family room
Mult if ami ly
Commercial office buildings
Commercial warehouses
Railroad yards
Highways
Bridges
Overpasses
Utilities
Electrical Power
Gas
Telephone and telegraph
Water
Airports
Military Construction
ill U>
60 &
-V
•-4 V) r-t CO 60 -r-4 v-( "° C
O to 0) -i rj ts
I) JJW fe^-lC V-itJ ^3^QfM
60 CO 0) 1 O (A >i-< COi T-IO) i-l
ra eowH T3OiW60J-»>-igt)On3w
>-l •-''OK r-lr-)CJ •i-ia-t-'>-lC-UV4
o TO-H rac ro-r-i d wo -H 01 o
•u i-it-j ^o ^3 01 c e-HiHTia o
cow SWH-H pfqpn-H na,Qw-r-i Q
J>il-> WJJ t-lO-H CW43
r-lq; U V4 1-4 (0 O M r-l O3 !-ltO COr-l
OJCD 3O O)-r-( DO 0) 0) tJCT1 OIM 4-1 M-l CO
flJH »-iM-l 0)E J-iM-l Ol i-l -PW I5H O 0) J-l
•U 4J 4J 4J 4JO3 O OMO)
CO CO CO CO COJXiO PJ pip-iS
(a)
(b)
-------
CONTEMPORARY AIR-POLLUTION COSTS, 1960-1970
Contemporary air pollution corrosion costs concern existing
component-systems in outdoor use during the decade of the sixities,
under generally prevailing air pollution conditions of that time; it
ignores regional, and/or neighborhood perturbations, as discussed else-
where. Some of the local variations in air pollution would tend to increase
the corrosion damage substantially and thus have an effect on the total
cost of air pollution. The arbitrary constraints of the study made it
necessary to deal strictly in national averages. The following evaluation
procedure was used to reach the air pollution corrosion costs of the
foregoing nine systems.
Evaluation Procedures
In the economic assessment, the corrosion cost of air pollution
divides into two terms. The first is the extra cost of maintenance,
mainly painting. The second is the cost of shortened appraisal life of
systems, due to evidence of pollution, i.e., excessive painting or replace-
-------
Marginal Maintenance Costs
The marginal maintenance cost over a system lifetime is equivalent
to the extra painting costs necessitated by the presence of air pollution
as compared with that experienced in clean, rural air. Other reasons for
extra painting, such as marring, are excluded. Because the painting cost is
proportional to the exposed area, an input estimate of the total exposed
metal area is computed for components attached to pollution-vulnerable
systems. The extra annual cost of repainting this area is calculated for the
economic life of the system exposed to air polltuion.
The annual census statistics on components, (standard industrial
classification), are recorded in either numbers of units shipped, or in
total weight shipped. These data are converted into equivalent exposed
area by computations that vary with the component considered. Straight-
forwardly, for example, the typical dimensions for transformer cases of
various size capacities are found in design references. Numbers of units
shipped by size capacity are found in Census data. To determine total
area, by average unit, it is necessary to calculate the area of average
units in each size range, and multiply by the number of such units shipped.
Perhaps the least straightforward, however, is to compute that area for,
say, pole-line hardware, which is reported as "tonnage shipped" and comes
in a polyglot of sizes and shapes. Most other components, however fall
somewhere in between these extremes. Nevertheless, the best reasonable
-------
48
These total areas are for all systems exposed to all kinds of air
throughout the country. They accordingly relate to an average frequency
of painting. Where pollution imposes a greater than-average repainting
frequency, the extra cost of painting over a system's lifetime, as a function
of air-pollution density, is computed.
Shortened-Lifetime Costs
The average economic lifetimes of all kinds of systems are given
by the depreciation guidelines and rules, procedure 62-21 of the Internal
Revenuse Service for average service, including average air pollution
conditions in the U.S. When considering individual assessments, assessors
use these guidelines, primarily as a point of departure. For example,
the guidelines give galvanized steel fences an average economic lifetime
of 30 years. Upon assessing a fence of some given age in a rural area,
an assessor would probably find little or no corrosion and fewer-than-
expected coats of paint. He accordingly would allow less depreciation
by the formula.
, „ .. , Service Age . . . . ,
Assessed Value = 1- :— . " .— x initial cost.
Economic Lifetime
Oppositely, if he sees an overcorroded or excessively repainted fence for
its age, he will accord it a less-than-average lifetime, using a lower-than-
average value in economic lifetime, in the formula and accordingly computing
a lower assessed value.
Because the annual census statistics on components are recorded
in value shipped it is a straightforward computation to integrate in-place
-------
49
Depreciated value is half of that, and yearly depreciation rate is 1007o T
years lifetime.
As the assessor lowers the economic lifetime of the system as a
function of pollution damage, the higher the depreciation rate. In a given
year, that margin applied to the total in-place value of the component
in a region (assessed on per capita) computes that year's value loss due
to pollution damage.
The integrated in-place value of a component was computed from
data points for as many years as the Census Department has reported
figures during the years of component lifetime. In-between points were
straight-line interpolations; and outside points, straight-line extrap-
olations.
Alternate Materials
The extra expense and inconvenience of maintaining steel con-
struction or systems in polluted atmospheres can be avoided by sub-
stituting an alternate, maintenance-free, pollution-resistant material of
construction. In the typical case, this substitution involves a higher
cost of installation than for the equivalent in steel. This premium,
divided by the life of the system in years, can be taken as an annual
charge to pollution. A more rigorous analysis is to take the annual
expense involved in installing and maintaining a steel and a substitute
material system, respectively, in polluted air, and to compare these costs
-------
50
either painted or galvanized steel, whichever is the most appropriate.
For the alternate material, the extra annual cost in polluted air above
that for steel in clean, air would be a charge to pollution. The saving,
if any, by the use of the alternate material instead of steel can be
established by comparing the annual extra corrosion cost due to pollution
for each material.
Evaluation of Surviving Steel Component Systems
The component systems that survived the final screening are
listed in Table 8. In most cases the tonnage shipped each year was high
in the SIC list. Other cases, where the rank was not so high, were in-
cluded because of the high potential economic loss for corrosion damage
by air pollution. A good example is chain-link fence which is typically used
in urban areas where it is attacked by polluted atmospheres.
Each of the nine component systems were analyzed individually.
The order of presentation is determined not by the economic importance,
but by the similarity of the calculation. For example, systems involving
galvanized steel were grouped together at the end of the list. Under
"Outdoor Metal Work" an assortment of external structures and components
were combined for calculation purposes.
For each of the steel systems, the calculations and the sources
of information used will be found in the corresponding tables in
-------
51
TABLE 8. COMPONENT SYSTEMS WHICH SURVIVED FINAL SCREENING
AND WERE SELECTED FOR DETAILED STUDY
SIC Code
34-435-9
34412
36122
36425
34411
34392
34492
34441-11
34441-16
34460
34421,2
34413
36441
33156
34816
Portion of Metal
Content Subject Potential Economic
Metal to Atmospheric Loss vs. Air Pollution
System Description Rank Corrosion Corrosion Damage
Steel Storage Tanks -
Fabricated from Plate
Highway and Railroad Bridges , ,-
of Structural Steel
Power Transformers 38
Street Lighting Fixtures 35
Outdoor Metal Work
Structural Steel for Buildings 5
Prefabricated Buildings
Portable Buildings
Roofing and Siding
Industrial Siding
Outdoor Gratings, Grills
Fire Escapes, Metal Doors,
Window Sash
Power Line Transformers Towers «
of Galvanized Steel
Pole-Line Hardware 28
Chain Link Fencing 33
and Steel Gates
large
most all
small
most all
small
large
large
large
large
large
all
all
most all
high
high
medium
medium
high
high
high
high
high
high
medium
high
high
33151
Galvanized Wire Rope
30
large
-------
52
Steel Storage Tanks
Steel storage tanks are used for water, petroleum products,
various chemicals, etc., in liquid and gaseous form. Certain food products,
detergents, and industrial raw materials are stored in powdered form in
tanks. The bulk of the storage tanks are externally mounted in regions
where the atmosphere is polluted.
Economic Importance. It is estimated that the tonnage of steel plate
in steel storage tanks of all types is 13,800,000 with a depreciated value, based
on original shipping cost, of $23,289 billion for the Nation. The procedures
for establishing these figures from SIC statistics is given in Appendix $„
From these figures and from other evidence, storage tanks are found to be an
important part of the total external steel work.
Control of Corrosion. Storage tanks exposed externally and above
ground usually are protected from atmospheric corrosion by painting.
Sometimes it is necessary to paint them more frequently than for other
types of structures. Water tanks, for example, are subject to severe
condensation particularly in the spring when humid air comes into contact
with steel chilled below the dew point, by the cold water in the tank.
Bulk storage tanks, containing other liquids, also show the effects of
the precipitation of dew. In the season when pollution is high, this
condensation is acidic and increases the rate of deterioration of the paint
as well as rusting under the paint.
Assessment of Losses. External steel storage tanks were divided
into three categories each being treated separately; namely, (1) elevated
-------
53
and industrial storage tanks (chemical, rubber^ paper and pulp, etc.).
The calculations for each category are shown in Table C-l in Appendix C.
Water Tanks. A one million gallon elevated water tank (1 MG) is
a typical size for an urban community. From a reference listing, all the
water facilities for communities of 25,000 population and over , it
was estimated that the total elevated steel tank storage capacity for the
Nation was equivalent to 11,000 typical sized 1 MG tanks. A second estimate,
as explained in Table C-l, was based on the total tonnage of steel in
place. These calculations indicated that the total number of typically
sized tanks was about 14,300. The average of the two estimates is 12,700.
The annual extra cost of maintenance painting for IMG water tanks is
calculated to be $288/year. Since some water tanks are located upwind
from the pollution sources in the typical metropolitan district, the total
exposed to average pollution was considered to be about 80 percent.
-------
54
Petroleum Tanks. The total bulk storage capacity of the petroleum
industry, according to the American Petroleum Institute figures for 1962,
is 11,981,306,000 gallons. If one projects this to 1970, using growth
in refining capacity as a guide, one finds that the Nation now has
20,000,000,000 gallons or 477,000,000 barrels storage capacity.
On the basis of discussion with engineers familiar with refinery
operations, it was established that bulk storage tanks are typically 100,000
barrels. Oil companies use smaller tanks, such as 10,000 barrels at
distribution depots. Approximately 2/3 of the total storage was judged
to be at the refinery and the rest are largely at the distribution
centers.
Calculations for petroleum product tanks are presented in Table
C-l. On a volume basis, allowing 10 percent for ullage, there are 3,465
large and 17,325 small tanks. The corresponding areas are found to be
31,450 and 6,530 square feet, respectively. Based on the fact that pollution
is relatively high in the vicinity of petroleum product tanks, maintenance factor
derived in Table C-l is relatively high namely $0.0473/sq.ft./yr.
The annual corrosion loss for the total of the two tank sizes is estimated
to be $10,520,000. On a tonnage basis, the loss is $11,100,000. Averaging
-------
55
Chemical and Industrial Tanks. External storage tanks used for
industrial products in gas, liquid, or powdered form are usually subjected
to air pollution. Paint maintenance costs tend to be higher, than say
for water tanks. Appraisers use an 11-year life for steel tanks in chemical
and related industries. Tank life often will be determined by internal
corrosion, in many cases, rather than by external attack resulting from
air pollution. Nevertheless, the external tank must be protected for the
duration of its normal service.
There are some 4,830,000 tons of such industrial tanks in use.
The extra cost of maintaining a protective coating on industrial type
tanks is about 0.0391 sq.ft./yr. A typical area factor for plate going
into steel storage tanks is 200 square feet/ton. If one considers that 85
percent of the tonnage in use is in polluted air, then the annual loss
arising from the extra cost of painting is found to be $32,100,000. These
calculations also are given in Table C-l, Appendix C.
These tanks are in such diversified service that no statistical
data were found that would enable an alternative route for readily checking
the magnitude of this loss. No doubt each of the major industries could
be surveyed and their totals combined.
Summary of Annual Tank Losses. The total extra annual corrosion
loss charged to pollution for externally mounted steel storage tanks
exposed to air pollution is summarized below:
Elevated Water Storage $ 3,400,000
Petroleum Product Storage $10,810,000
Chemical and IndustriaITSnks $32,100,000
-------
56
Highway and Railroad Bridges
Most of the Nation's important bridges are made of steel. A
bridge is defined by highway engineers as a crossing with a span of 20 feet
or more.
Economic Importance. Structural steel bridges are high in dollar
value. It is estimated by the Federal Highway Administration that there
are 563,000 highway bridges in use. Some 236,000 are in the Federal-Aid
Highway Systems. In addition., there are some 94,000 railroad bridges.
By comparing road mileage figures for the Nation for rural, municipal,
state, and federal roads, and by making certain assumptions, it is estimated
that about 30 percent of the total bridges are in metropolitan areas.
This estimate may be compared with that suggested by Erickson and Morgan
of the U.S. Bureau of Public Roads. They estimated in 1955 that 25 percent
(2X>
of the bridges are in urban locations.
Using the Department of Commerce statistics for 1963, the
estimated total tonnage of structural steel for bridges currently in use,
based on a 30-year average life is 18,134,000 tons for the Nation. The
total value of the steel, as shipped, is estimated to be $4,093 x 10 .
Applying the .30 percent estimate to these figures gives .5,450,000 tons and
$1,230 x 10 respectively.
Factors Determining Life of Bridges. Although appraisers use
30 years, the actual life of a bridge is more often between 50 and 100 years.
Corrosion damage is the main reason for reducing the load limit of a bridge
-------
for bridges located in metropolitan areas. One is corrosion of the super-
structure by polluted air, and the other is damage to the bridge deck by
salt. In northern cities where deicing salt is used in winter months,
bridge deck and guard rail damage is fairly common. At marine locations,
ocean spray may be conveyed by the wind to a highway or railroad coastal
bridge and cause both deck and superstructure damage.
The steel bridge structure, especially older designs, has many
vulnerable sites where corrosion damage can arise when the air is polluted.
Girders tend to show corrosion around rivet or bolt heads and at contact
surfaces where moisture may enter between flange plates at bearings and
at connections. Cracks may show up at welds or at a sudden change in
section where there are stress concentrations.
Steel decks must be covered to prevent salt damage. Seals
must be provided over expansion joints to keep out salt, and
foreign matter. Corrosion damaged decks are seldom the controlling factor
for determining the useful life of bridges in polluted atmospheres because
they can be repaired and replaced. A rough deck, however, adds to the
impact load from traffic.
The weakened elements are usually in the superstructure where
loss of section by excessive corrosion in polluted air or the development
of cracks has reduced the load limit.
Co.ntrol of Corrosion. The most common method of controlling
the corrosion of steel bridges is by paint coatings. As with other painted
-------
58
underfilm attack and (b) the weathering of the paint itself. Much research
has been carried out to control underfilm attack. Moisture will in time
penetrate the paint film and cause, in the case of steel, underfilm
rusting.
Bridges occasionally are protected by galvanizing the steel before
erection. If galvanized steel is used, it eventually has to be painted.
To properly maintain a bridge the paint system employed should
be specially selected for the environment. It is recognized that the
corrosivity caused by air pollution may be higher at one end of a large
bridge than at the other. The paint system should be chosen to meet the
worst condition one expects in service.
Since the labor cost of application may be as much as ten times
the cost of the paint materials, it does pay, in most situations to use
the best available coating for the particular environment. Paints which
can resist the acidity developed when moisture deposits on the surface
in a sulfur contaminated atmosphere may cost more than those satisfactory
in rural clean air.
The life of a good quality paint system on steel in rural
clean air may be from 10 to 15 years. In urban polluted a?.r, the life
may be 5 to 8 years on highway bridges.
Assessment of Losses. Both the first cost of the original paint
system and the maintenance of the coating during the life of the bridge
will be higher in an industrial atmosphere than for a similar bridge in
rural clean air. The calculation of the maintenance factor, $0.0447 sq.ft./year
shown in Table 4 is based on the higher costs shown in Table 3 for bridges,
-------
59
The calculations for assessing the extra cost of maintaining
bridges in polluted areas is depicted in Table C-2, Appendix C. The
calculations as previously discussed are based on a conservative estimate
that 30 percent of the bridges are in polluted areas. From the total of
18,134,000 tons in use, an area factor of 125, and a maintenance factor
of $0.0447/sq.ft./year, the annual loss is estimated to be $30,400,000.
Power Transformers
The bulk of the power transformers are boldly exposed to the
atmosphere. Transformers mounted in underground chambers lose capacity
during hot weather, because of overheating. For this reason, transformers
in underground distribution areas are also mounted above ground where
conditions permit. The sizes range from 15 KVA units on power poles to
massive units of 10,000 KVA or larger in substations and at power plants.
Economic Importance. Power transformers have a high inventory
value, namely $12,316 x 10 . However, only a small percentage of this
value is for the external housing. The bulk of the installed value of the
steel is for the silicon core. Maintenance of the housing, however, is
essential to protect this big investment. From personal observation,
externally mounted power transformers are known to be a significant portion
-------
60
Control of Corrosion. Paint is the normal method of protecting
the external surfaces of power transformers from the atmosphere. Utilities
tend to favor lead suboxide paint formulations for this application.
Whether the air is polluted, or not, good results usually are obtained
in this service.
Assessment of Pollution Costs. By combining information obtained
from a study of the design of various sized transformers, with statistics
giving the number of each size shipped each year, it was possible to
develop an estimate of the total area of the housings for all the different
sized transformers in service. The maintenance factor of $0.0167 sq.ft.
for transformers was developed as shown in Table 4, and is based on service
experience with average pollution. The calculations are based on 80 percent
of the installed area of 556,630,000 square feet being exposed to pollution
and indicate that the total extra cost of annual maintenance if $7,450,000.
Additional details are given in Table C-3 of Appendix C.
Street Lighting Fixtures
Outdoor steel lighting fixtures are most commonly installed in
populated districts. Many are now being made of aluminum and require very
little maintenance. Poles are often made of wood, aluminum, reinforced
concrete, or fiber reinforced plastic instead of steel to reduce maintenance
costs. A discussion of the advantages of using maintenance-free materials in
polluted areas is presented elsewhere.
Economic Importance. Although the total tonnage of steel in
lighting fixtures can be estimated from available statistics, the dollar
-------
61
of the electrical fittings but excludes the lamps. The inventory value
of $1,221,438,000 does not represent the mill cost of the steel components.
Lighting fixtures were included since the tonnage in service was significant,
namely 984,574 and since most of the steel used is light gauge and thus
the area factor will be large.
Control of Corrosion. Lighting fixtures made of steel are protected
by galvanizing or aluminizing. Some fixtures are painted for esthetic value.
Painting is the most common form of repairing damaged surfaces and fixtures
that already have seen years of service are maintained by painting.
Assessment of Corrosion Losses. Most of the Nation's lighting fixtures
are in heavily populated areas. About 85 percent of the tonnage going into
light fixtures is estimated to be exposed to air pollution. In some designs
part of the metal is enclosed or buried and not subject to atmospheric corrosion.
For lighting fixtures, the maintenance factor of $0.0167 sq.fti./
year was taken from Table 4. This extra cost is for maintaining external
metal work by painting in an environment with average air pollution. Using
20 percent of the total tonnage and an area factor of 400 sq.ft./ton, it can
be shown that the annual losses for light poles if $1,110,000. Similarly
for the fixtures, the annual cost is $10,800,000 giving a total of $11,910,000.
The details of these calculations together x^ith footnotes concerning the
values selected are presented in Table C-4, Appendix C.
Outdoor Steel Metal Work
Under this heading, a variety of external steel items have been
grouped. This includes roofing, siding, downspouting, portable and pre-
-------
62
(e.g., garage-type) in residential, commercial and industrial buildings,
window sash and frames, and structural steel. Total tonnage in service
figures have been developed for each of these items.
Most of the items have large areas of externally exposed surface
per ton of installed metal. The one exception is structural steel. In
commercial applications only a small percentage of the structural steel
is externally exposed. In existing construction, structural steelwork
is enclosed for the most part.
Economic Importance. From the data developed from the national
metal statistics on outdoor steel work of a variety of types, it is
estimated that 26.6 x 10 tons are exposed to industrial environments.
Large tonnages are also exposed to other urban environments. These tonnage
figures and the fact that much of the steel is light gauge with a much higher
area factor than the material going into, say, bridges or tanks, indicate
the total losses will be extremely high for this category.
Control of Corrosion. Some of the items listed above are
protected by galvanizing, others are protected by aluminizing. The most
common method of controlling the attack is to provide a good coating of a
resistant paint. Painting can be delayed where there is a zinc or other metal
coating on the steel, until weathering has progressed to where rust is
just starting to appear.
Economic Assessment of Losses. Each item, for which a tonnage-in-
use figure was developed was considered separately for purposes of calcula-
-------
63
and roof drainage such as gutters and downspouting is estimated to be 17,200,000
tons. About 20 percent of this tonnage is used in rural applications leaving
13,750,000 tons. As outlined in Table C-5, Appendix C, part of this latter
tonnage, namely 5,500,000 is in industrial, part in commercial, namely 2,060,000
and the .rest is in residential service and not included. Tonnages also were
developed for galvanized prefabricated and portable steel buildings. Com-
bining these tonnages for (a) industrial and (b) commercial environments, gives
10,335,000 and 4,820,000 tons respectively. Since the .material used in
industrial environments is 22 galvanized sheet gauge, an area factor of 1400
sq.ft./ton was used (one side). For commercial service, an area factor of
2500 corresponds to 28 sheet gauge. The extra maintenance cost figures of
o
0.0392 and 0.0213 per ft /yr are taken from Table 4. The calculation shows
$567,000,000 for industrial and $257,000,000 for commercial atmospheres,
resulting in a total cost of $824,000,000/year.
To this figure must be added the cost of painted steelwork (not
galvanized) in polluted atmospheres. These calculations are also presented
in Table C-5, Appendix C. Where the exposure is industrial, a maintenance
factor of $0.0333/sq.ft./year was employed and for commercial exposures,
$0.0167/sq.ft./year. In the case of window sash an area factor of only 100
sq.ft. per ton was used, since only part of the steel is exposed. The main-
tenance factor of $0.06/sq.ft./year is high mostly because of the extra
labor cost involved in preparing and painting metal sash and frames.
For the last item, it was estimated that 15 percent of the
industrial tonnage and 15 percent of the commercial tonnage is in external
use. On the other hand, it was considered that 65 percent of the steelwork
in utility service is external, in view of the practice of externally mounting
-------
64
In summary, the total annual extra cost involved in outdoor metal
work is assessed at $914,015,000. For a list of the items that make up this
total, refer to the end of Table C-5, Appendix C.
Pole Line Hardware
Galvanized steel is the most common material of construction
for overhead pole line hardware. Clevises, eye nuts, J-bolts, rope
clips, and insulator caps are made of copper-bearing forged steel.
Some items such as strain clamp bodies, hooks, and sockets are made of
galvanized malleable iron. These can be lumped with the steel for
purposes of estimating pollution costs. The main factor in pole line
hardware, is the shortened life of the zinc coating in industrially polluted
air. This necessitates its replacement or repair before the wooden pole
itself requires replacing. In rural air, pole-life determines the time
when the pole and the attached hardware is to be replaced. Creosoted
wooden poles typically last 20 to 25 years in either rural or urban
exposures.
Economic Importance. Although the amount of pole-line hardware
per pole is not very significant, the total number of poles is in the
hundred millions. Many metropolitan areas use poles extensively in all
but the civic center itself for the overhead distribution of power and
telephone services. Since the bulk of the hardware is galvanized steel
or galvanized iron, and since this hardware must be replaced at least
once during the life of the pole in industrial regions, the cost that can
be charged to air pollution is a significant part of the total cost for
the Nation.
Corrosion Control. For pole line hardware, the main reliance
-------
65
square foot is specified. Utilities prefer to remove corroded hardware,
which, if still serviceable, is sent to a local galvanizer for reprocessing.
Painting of corroded hardware in situ does not seem to be widely practiced.
The ideal hardware should last at least as long as the pole.
Under rural conditions, this aim is achieved easily. On the other hand,
pole hardware in industrial locations tends to fail in 15 years or less.
Thus, it must be replaced before the deterioration of the pole itself
requires replacement of the whole installation.
Assessing Pollution Costs. Two sources of metal data are avail-
able for pole line hardware. The first estimate is based on steel statistics.
The total amount of pole line hardware and related products is 4,062,735
tons. It is estimated that 75 percent, or 3,040,000 tons is pole-line
hardware. By combining the area factor of 450 square feet per ton with
an estimated average area per pole of 2 square feet, one finds there are
342 x 10 poles.
The second estimate is based on the consumption of zinc for pole
line hardware. It is estimated, from the amount of zinc slabs going into
baths being used for pole-line hardware, that there are 82,500 tons of zinc
coating in use for this purpose. A bath efficiency of 50 percent was
assumed. Each pole uses 8 oz. of zinc (4 sq.ft. x 2 oz./sq.ft.). This
figure, combined with the zinc tonnage figure results in 330 x 10 poles.
-------
66
For service in polluted air, pole-line hardware must
be replaced once during the lifetime of the pole. The charge against
pollution then is the cost of replacing this hardware, see Table C-6,
Appendix C. Since it is common to regalvanize pole-line hardware and
return it to service, this cost, taken at $0.55 per square foot com-
bined with a labor cost of $25 per pole, results in an annual extra cost of
$1.20/pole, based on a pole life of 22.5 years. It was estimated that 40
percent of the poles are exposed to polluted air. Combining these figures
with the 336 x 10 poles gives a total of $161,000,000/year.
Chain Link Fencing
One of the most common metallic structures found in metropolitan
areas is chain-link fence. Nearly all industrial plants use some fencing.
Most of the chain-link fencing in existence today is galvanized. Plastic
coated and aluminized chain-link fence are being used, but represent only
a few percent of the total installed. Aluminum cha.in-link fencing also is
available.
All the metal that goes into a chain-link fence is exposed to
the atmosphere. A wire gauge size of nine (0.148 inch diameter) is most
commonly used for chain-link fence because of the economy involved. This size is
heavier than required for security or for corrosion. If one specifies an
even heavier gauge, the extra material as expected raises the cost of the
product. On the other hand, by specifying a lighter gauge wire, the higher
-------
67
Economic Importance. The amount of chain-link fence in predominantly
industrial service is 4,018,980 tons. In addition, some 6,028,471 tons of
posts and fittings are employed to support this fence. The total in-
ventory value is $1,444,013,000. On the basis of these data it is
evident that chain-link fencing is one of the more important systems
involved in accessing corrosion damage caused by air pollution.
Control of Corrosion. The zinc coating on galvanized fencing is
2
typically 1.25 oz./ft . It may last about 20 years in rural air and 8
years more or less in industrial air. Corrosion control, as with other
galvanized surfaces, sometimes involves painting. To obtain a life
equivalent to that in rural air, chain-link fence will require several
paintings. More often, chain-link fencing is replaced. It is most common
to paint the posts and replace the wire.
Assessment of Losses. Chain-link fence comes in heights varying
from 36 to 144 inch. For the computation, a 72-inch high fence is chosen as
typical. All computations in Table C-7 of Appendix C are based on a 100-foot
length of fence as a unit. (Contractors commonly use this size for estimating,
then make adjustments for other sizes.)
The weight of 100 feet of fence, 6 feet high is calculated to be
415 pounds, which compares well with an estimate of 400 pounds received from
a fence supplier. By taking the total tonnage of chain-link fence in
service of 4,018,980 tons, it can be shown this is equivalent to 19.35 x
10 100-foot lengths, six feet high. Since about 80 percent of such fence
is in industrial and urban service, the total 100-foot lengths exposed to
-------
68
For rural conditions, 20 years for the fence and 30 years for the
posts is a typical life. For polluted atmospheres, a paint coating is
required after 8 and 14 years to obtain the same life as in clear air.
This extra cost, as shown in Table CJ7, Appendix C, amounts to $0.0054
per square foot per year based on the area of the fence, and not on the sur-
face area of the wire itself. A maintenance cost of $0.0233 per square
foot per year is used for the galvanized posts, based on a 15-year life
for the original galvanized coating.
If 25 percent of the owners maintain fencing by painting, the extra
annual cost is shown to be $20,400,000. If another 25 percent replace
the entire fencing after allowing it to rust for a few years beyond
failure of the zinc the extra cost per year, above equivalent rural costs
on this basis comes to $38,000,000. This calculation is based on an in-
stallation cost of $440 per 100 feet of chain link fence. The rural life
was 20 years plus 10 years rusty and the industrial life was 8 years plus
2 years rusty for the original and for each of two replacements. The most
common method of maintaining fence in industrial environments is to replace
the wire only and maintain the posts by painting. The remaining half
of the owners, using this basis for calculation experience on extra annual
cost of $13.65 per 100 feet length of 6-foot high fence as shown by the
calculations in Table C-7. This results in a loss to the Nation of $106,000,000.
A summary of the annual total extra cost for galvanized steel
chain link fencing is provided at the end of Table C-7. The grand total
for chain-link fencing is $165,800,000.
Galvanized Wire Rope and Cable
There are many external uses for wire rope and cable. Overhead lead
telephone cables are supported by galvanized steel messenger wire. Guy wires
are used extensively to brace wooden telephone poles. Traffic lights and signs
are often hung from galvanized steel wire cables. It is estimated that 40
-------
69
Economic Importance. The total weight of galvanized steel rope
and cable shipped each year is of the order of 200,000 tons. The individual
strand of galvanized wire itself has a mill price of about $221 a ton
including $40 a ton premium for the zinc coating. The forming of galvanized
wire into rope increases the cost about three times. The inventory value,
as shipped, of the galvanized wire rope now in service is estima.ted to be
$2,007,500,000.
Corrosion Control. The zinc coating, which typically is thinner
on wire than on sheet, usually will provide protection by isolating the
steel surface from the atmosphere. Once the zinc has weathered away to
where steel is beginning to be exposed, the remaining zinc will provide
sacrificial protection to the adjacent steel surface. Once the zinc is
gone, the corrosion attack is accelerated, the wire loses strength and
requires replacing.
In rural clean air, wire rope maylast typically about 20 years
whereas in industrial atmospheres, wire rope may lose its coating in 5 to
f
8 years. While the heaviest coatings obtainable should be used in .industrial
atmosphere on sheet, this is not practical on wire because of flaking in
service. However, in practice, there is considerable loss of zinc.
Assessment of Corrosion Losses. From steel statistics, it is
estimated that there are 6,301,000 tons in service at a shipped value of
$4,015,000,000. The average cost per ton is about $635. By comparing the
rural life cost per year with that in industrial atmosphere one obtains a
figure of $47.7/ton/year for the extra cost of wire shipped to replace
wire corroded by air pollution. On the basis that 40 percent of the total
tonnage estimated to be in use is exposed to polluted air, the total loss
-------
70
From zinc statistics, the total tons of slab zinc consumed each year
to galvanize wire is known. Since galvanized wire lasts from 5 to 20
years or more in the atmosphere, an average life of 15 years was used to
obtain an estimate of the total tons of zinc coating on the steel wire in
service. The total zinc consumed in wire baths is equal to 572,213 tons
since 1955. The typical zinc bath has an overall efficiency of around 50
percent. Some zinc is lost and some wire receives more zinc than the mini-
mum required for the grade.
For the calculation, a typical wire size was chosen, namely
0.120 inch diamter. Seven strands of this wire are used on 3/8 inch wire
rope, a very commonly used size. A thousand feet of 0.120 inch wire size
after double galvanizing has 26.6 oz. of zinc (0.85 oz./sq.ft.). When made
into 3/8 inch 7-strand rope this becomes 186 oz. of zinc/1,000 feet. If
one assumes all the wire rope in existence was made into this size, one can
show that there will be 49.3 x 10 lengths of 1,000 foot 3/8-inch wire
cable.
The alternate computation also depicted in Table C-8 is based on
the replacement cost of 1000 foot lengths of 3/8-inch stranded wire, and using
the same expected life in rural and industrial life as before. The results
show a total for the Nation of $103,900,000 per year. This is a reasonably
close estimate and was arrived at independently of the steel data above.
Averaging the two values gives an annual loss of $111,800,000.
As originally anticipated, the cost of the damage to wire rope
resulting from air pollution is high. This cost does not take into
consideration the expense of shipping, installation, and replacement.
These charges, if they could be established and included, would increase
-------
71
Power Line Transmission Towers
Approximately 98% of the power transmission towers in service today
M
are constructued from galvanized structural steel. This material also is
used for radio and TV transmitter towers, telephone microwave systems,
and for flood lighting towers (e.g., in sports fields, industrial plants
or rail yards).
For service in industrially polluted areas, there is some interest
in alternate materials which are maintenance free. In the late 50's,
structural aluminum alloys were introduced. Their corrosion resistance to
polluted atmospheres is excellent. More recently self weathering steels
have been tried for power line transmission towers. Because of the
higher strength, less low alloy steel is required in the structure than
when carbon steel is used. Since the corrosion rate is about one-fifth
that of carbon steel, only a small corrosion allowance, if any, is needed.
If the dark rust coat is not acceptable from an esthetic standpoint, the
structure may be painted. Service experience has shown that paint coatings
are more durable on self weathering steels than on carbon steel because
any rust that forms at breaks or holidays is less voluminous and there is
less rupturing of the paint film.
For suitably chosen aluminum alloys, or for boldly exposed self
-------
losses can be ascribed to air pollution. For galvanized steel structures,
on the other hand, polluted air sharply reduces Che life of the zinc coating
and the protective paint system that must be applied after the zinc has
weathered to expose the steel base.
Economic Importance. Power line transmission towers and similar
structures used for other purposes have high in-place value and are also
high in the amount of exposed area per total weight of metal. The present
inventory value is estimated at $1,165,000,000. Only about 12.5 percent
of the towers in existence are in polluted areas.
Control of Corrosion. The heavy zinc coating usually provided
on structural steel intended for tower use will provide excellent protection
in rural atmospheres. For control of corrosion after the zinc has weathered
to the point where the steel is just becoming exposed, a paint system
usually is applied. In rural atmospheres, this point is reached between
25 and 40 years, whereas in industrial and in commercial regions, the life
of the zinc coating is shortened severely to 5 to 15 years.
Assessment of Corrosion Costs. Although it is usual to expect
greater than the assessor's life for a transmission tower of 50 years, for the
purposes of this investigation, this value is used. It is well established
that both the zinc coating on galvanized steel and the subsequent paint
coating that must be used to protect the steel base after the zinc has failed
both have a shorter life in industrially polluted air. A paint system that
gives 15 years' protection over weathered zinc in rural air may only give 6
to 10 years' service in polluted air. The computation is based on a 15-year
-------
73
Actual costs for painting transmission towers vary with the size
and location of the tower. The range is $375 to $1000. It is considered
more costly to paint in an urban area, since air pollution increases the
cleaning and application cost. On the other had, transport of painters and
equipment to a rural site can add to the cost, especially when the line is
not readily accessible by road.
A typical tower weights 7.5 tons and has a surface area of about
2500 square feet. The total weight of galvanized structural steel employed
in towers is estimated to be 1,449,000 tons. Since a typical tower weighs
7.5 tons, it can be calculated that there are about 194,000 "average"
towers in use. This number may be compared with the results of a survey
of 40 utilities conducted in 1963. A total of 124,020 towers and substation
structures were found in their systems. About 12.5 percent were located
in contaminated air.
Of the 194,000 towers in use, using the same 12.5 percent, only
24,300 are exposed to pollution. Using the typical area for a tower of
2500 sq.ft. and the maintenance factor developed in Table 4, the annual
loss as shown in Table C-9, Appendix C, becomes $1,480,000.
Air Pollution Damage'Costs for Alternate Materials
Pollution can be considered a benefit, in the sense that it tends
to promote the use of more resistant materials in place of steel for some
systems. This, in turn, often results in a saving to the economy in the
form of reduced wastage of materials and increased reliability. An
assessment has been made of the cost of corrosion control using selected
alternate materials in polluted environments as compared with the cost of
painted steel or galvanized steel. Several metals were considered in this
-------
74
Roofing Metals
An analysis of the corrosion economies involved in well known
roofing metals is presented in Table Oil, Appendix C. In polluted air,
such as is typically experienced in industrial areas, galvanized steel roofing
may last 10 years or less before rust appears and paint maintenance is re-
quired to save the remainder of the investment. Aluminum will last at
least 50 years without maintenance. Copper will last well beyond 100 years.
Terne plate, even with the heaviest lead-tin alloy coating available for roofing,
namely 0.5 oz. per square foot, must be painted at the time of installation.
In this condition, it will last 50 years, although repainting in this
period may be desirable for esthetic reasons.
The gauge (special galvanized sheet gauge) for galvanized steel
chosen for the study is 22, which is heavier than typical for the roofing
in general use. However, for industrial service, a heavier sheet normally
is preferred. Aluminum is usually about 33 percent thicker for the same
service than its equivalent in steel. In the present example, 18 gauge
(Brown & Sharp) aluminum is chosen for the comparison. Both galvanized
steel and aluminum are commonly corrugated to increase stiffness and strength.
Copper and terne plate are used in thin gauges, and 28 and 26, respectively,
were chosen for the comparison. Each of these two materials requires extra
support and these costs are included in Table C-ll. Mill sheet prices were
used for the comparison.
A galvanized roof when compared on an equivalent installed basis,
is about 2.5 times as heavy as aluminum. It must be overlapped somewhat
more than aluminum. A higher labor cost applies to the heavier galvanized
steel than for the aluminum. For copper and terne plate, there are extra
installation steps, such as soldering or crimping of seams which is not
-------
75
The installed charged as listed in Table C-ll rate copper as the
highest and galvanized steel the least expensive. However, the cost per year
in polluted air is lowest for copper and highest for galvanized steel.
If one uses the galvanized steel roofing cost in polluted air per
year as the basis of reference, it is found that the savings if alternate
*
materials were used would amount to 73.5 percent for aluminum, 77.5 percent
for copper, and 61.3 percent for terne plate. If this basis were used as
the amount to be charged to pollution, and if all the galvanized roofing
were replaced, the annual pollution charges would be reduced from $824 million
for galvanized to $218 with aluminum, or $202 million for copper or $326
for terne plate.
If, on the other hand, one uses as a reference the annual cost
of galvanized steel roofing in rural clean air and compares this expense
with that for each of the four materials in polluted air, one finds that
for copper and aluminum, the annual costs are less, even in polluted air,
than that for galvanized steel in clean air. As shown in Table C-ll, the
aluminum roofing on this basis costs $0.001/sq.ft./year less and the
copper roofing $0.003/sq.ft./year less than rural galvanizing. Thus for
these two materials, it is concluded there is no charge to air pollution.
For terne plate in polluted service, the cost above rural galvanized is
$0.005/sq.ft./year. This may be compared with the cost of galvanized
steel of $0.035/sq.ft./year on the same basis.
About 25,000 tons of terne plate go into roofing each year.
Almost all of this is for residences and institutions. The actual tonnage
used in industrial and commercial applications is not known, but probably
is less than 1,000 tons a year. If one considers that as much as 50,000
tons of terne plate roofing are currently exposed to contaminated atmospheres,
-------
76
In view of the lack of data and the small size of the estimated annual loss,
terne plate was not included in the overall assessment of annual losses.
Aluminum Siding, 1969
Most of the aluminum siding goes into residential and mobile homes.
Industrial consumption is not reported by the Department of Commerce nor
by the Aluminum Association. The latter, however, have privately indicated
their estimate for 1969, namely a total of 53,000,000 pounds of aluminum
(42)
went into industrial siding.
The calculations shown in Table C-12 in Appendix C have made use
of this statistic. A popular thickness for industrial siding is 20 gauge
or 0.032 inch thick. This weighs 55 pounds per 100 square feet. An
installed cost for the aluminum siding is calculated to be $0.59/sq.ft.
versus $0.43/sq.ft. for the equivalent size in galvanized, 0.275-inch
thick.
The 53,000,000 pounds of aluminum, if all of 20 gauge, would have
an area of 96 x 10 sq/ft. Item 5 in Table C-12 shows the installed cost
for this amount of aluminum to be $56,700,000 compared with galvanized steel
of $41,300,000. The premium.paid above galvanized steel, for this area of
aluminum siding is $15,400,000. This amounts to $308,000 per year over the
50-year life. If the production figures for other years were available,
one could estimate the tonnage in use and then calculate the total annual
premium to be charged to pollution.
A more rigorous comparison is to use the annual cost of galvanized
steel in rural conditions as the reference in establishing the annual loss
to pollution. The calculations, as listed in Item 6 of Table C-12 shows
the annual cost of 96 x 10 sq.ft. of siding to be $1,133,000 for aluminum
-------
77
difference in these estimates, it is concluded that there is no ^charge
to pollution for the use of aluminum siding in industrial areas on this basis.
(For this same area, as might be expected, the annual charge to pollution
for galvanized steel is almost $3 million; see Item 7 in Table C-12).
Self Weathering Steel
The cost of self weathering steel is about 40 to 45 percent higher,
on a weight basis, than carbon, steel. These steels are bought on strength
and other physical properties and not on composition. Atmospheric corrosion
rates in polluted air are typically 1/4 to 1/6 that of carbon steel.
On an installed basis, there are two factors which reduce the
cost-differential between these steels and carbon steels. First little
or no corrosion allowance is required, and second the higher strength often
will permit a reduction in the totalweight required for the design in
typical cases. Installation costs tend to run some 10 percent higher because
of the higher skills required for fabrication.
(43)
Schmitt and Mathay provide cost data for comparing carbon
steel with self weathering steel. Their data forms the basis for the cal-
»
culations recorded in Table C-13 in Appendix C. A comparison is made on
the basis of substituting bare self weathering steel for painted carbon
steel in bridge construction.
The installed cost, as shown in Item 2 of the table, is $371
for carbon steel versus $445 for self weathering steel. Since about 10
percent less weathering steel would be used than for an equivalent structure
in carbon steel, this last figure is reduced to $403 per ton. On this
basis, as shown in Item 3 of the table, the annual premium, corresponding
to a bridge life of 30 years, would be $1.07/ton of carbon steel replaced.
This premium, in a sense, could be charged as an annual cost to pollution.
-------
78
as the basis for establishing the extra annual pollution cost, one finds
that instead of a loss, there is a net saving of $4.33/ton/year in the use
of self weathering steel in polluted air; see Items 4 and 5 in Table C-13.
In view of this situation, no annual cost to pollution was charged to the
use of self weathering steel.
As a matter of interest, additional comparisons were made between
painted carbon steel and self weathering steel. For example, if one cal-
culates the expense of maintaining an average carbon steel bridge in a
polluted atmosphere, the annual cost is $11.00/ton. As shown by Items
6 and 7 in the table, the substitution of self weathering steel results
in an annual saving of $9.93/ton.
It also has been determined, that if the self weathering steel
is painted, and one allows a 25 percent longer life in polluted air for
the protective coating than the same coating on steel, there is no essential
difference in the annual cost between the two systems. The real savings
come, as one might expect, by designing the bridge or structure to use
bare self weathering steel.
Stainless Steels
Most of the AISI 200 and 300 series of stainless steels have
excellent resistance to polluted atmopsheres. Although the material
cost is 5 to 9 times that of carbon steel, no corrosion allowance is needed
and the maintenance cost is negligible. Although no calculations were made,
the cost of pollution analysis would be very similar to copper. Compared
to the cost of painted steel in rural atmospheres, the annual cost for
stainless steel in polluted atmospheres would be less. No charge was made
-------
79
Summation of Annual Extra Corrosion Losses Caused by Air Pollution
Nine major classifications of external metal structures have been
analyzed in the assessment of corrosion damage caused by pollution. These
systems are listed in Table 9, together with the estimated annual corrosion
loss in dollars. The total loss, for the Nation, was found to be $1.45
billion.
For painted steel structures, the annual loss was based on the
extra maintenance expense required to insure normal system life. The
extra maintenance expense per unit beyond clean air costs were applied
to the total area of surface exposed to pollution. For galvanized steel,
several procedures were used. For roofin.g and siding, the extra cost of
maintenance painting, after the zinc coating failed, to obtain normal
system life was compared with similar costs pertaining to clean air
environments. This method also was used to compute part of the costs for
chain link fencing. For fencing, 75 percent of the calulation was based
on the costs of replacement. For galvanized pole-line hardware and for
galvanized wire rope, the extra cost of replacement, based on shorter
service life in polluted air was converted to an annual basis, and used as
the cost.
In the study of alternate materials, particularly aluminum, copper,
and self weathering steel, the annual premium above carbon steel paid
for these materials was compared with the annual corrosion cost of gal-
vanized or painted steel in rural clean air. In most instances, the use
of these alternate materials in polluted air cost less per year than steel
in clean air. In view of this situation, no charges against pollution
-------
TABLE 9. SUMMATION OF ANNUAL EXTRA LOSSES DUE TO CORROSION DAMAGE BY AIR POLLUTION
TO EXTERNAL METAL STRUCTURES FOR 1970
Steel System or Structure
Basis for Calculation
A nnua1
Loss in $1000
For Calculation
See Table
Steel Storage Tanks
Highway and Rail Bridges
Power Transformers
Street Lighting Fixtures
Outdoor Metal Work
Pole-Line Hardware
Chain Link Fencing
Galvanized Wire and Rope
Transmission Tox^ers
extra cost of maintenance $ 46,310
extra cost of maintenance 30,400
extra cost of maintenance 7,450
extra cost of maintenance- 11,910
extra cost of maintenance 914,015
extra cost of replacement 161,000
extra cost of maintenance 165,800
and cost of replacement
extra cost of replacement 111,800
extra cost of maintenance 1, 480
$1,450,165
C-l
C-2
C-3
C-4
C-5
C-6
C-7
C-8
-------
81
ANALYSES OF COST OF CORROSION
DAMAGE BY AIR POLLUTION, 1970 to 1980
The total cost of corrosion damage to the Nation ten years hence
will be affected by changes in the corrosivity of the atmosphere and by
the total amount of external metal structures susceptible to attack. The
effects of changes in population, density of population in metropolitan
districts, energy production, industrial activity, control of corrosion
technology, labor cost of maintenance, and level of pollution regulation
must be examined and the individual and combined trends with time estimated.
Economic Trends
Power generation and industrial activity are affected by economic
growth. One of the variables;for which records are available^is popula-
tion growth. A recent report indicates the total population for USA in 1970
(44)
to be about 204 million. As shown in Figure 1 this is expected to be about
227 million in 1980 corresponding to a growth of 11 percent. The metropolitan
areas are expected to accommodate most of this growth and increase at an
average of 12 percent. By 1980, it is expected that 80 percent of the population
will be crowded into metropolitan districts occupying 1.5 percent of the land.
The population growth, together with a higher standard of living
for a greater portion of the population, is expected to increase the demand
for energy and also for products and services. Energy production is increasing
at the rate of 7 to 8 percent each year, and is predicted to increase by
78 percent in 1980. The various sectors which determine industry growth,
judging from the recent past, probably will not increase uniformly. Iron
and steel production has increased 26 percent during the past decade.
-------
!<:. 10 X 10 to tho inch.
' 210 600
. .,
-, ;-rrn "i ' !/• I '
' •-".•;/"::
-200 500
30 190 400
.l J Li.i-'_ L
No. 1 Electric Generation Capacity
.: "i "':.'. ..:.!. : .
No. 2. Population ; / *\
No. 3 Sulfur Dioxide Emission
"(I) "Ele'ctr'ic'PbwerT Edisoti Electric
LTD. jinx xuxj~tx!j!
160 100
~l™ Institute
(2) U. S. Bureau of Census
(3) See Reference 45
J.J-.-J4.:..... 1950 I - I960
1970--- -i-; 1980 ; : Year
• ! M
__...._. r
-rlrh-rh-
FIGURE 1. PROJECTED GROWTH IN POPULATION AND IN POWER GENERATION AND
INCREASE IN SULFUR DIOXIDE EMISSIONS WITH NO RESULATION.
i-' i i !-n-
•' f-f: rpi >
'f~i"'~n~i~rr
a:
KEUFFEl. A ES'>ER CO.
-------
83
at 8.6 percent per annum. Plastic production is growing at a phenominal rate
and will double in the next five or six years. However, the overall growth
of all industry is conservatively estimated to be about 4 percent per year or
(47)
roughly 48 percent by 1980. '
Changes in the Amount of External Structures
The increase in industrial activity can be expected to result in
an increase in some types of external structures. Storage tanks will in-
crease in total capacity as a result of the expansion in the chemical; petrole-
um^ and other such industries. Highway bridges and overpasses will increase
in number with the growth of the highway system needed to accommodate the
11 percent increase in population. On the other hand, railroad bridges
and other such structures may not increase in number, since no major ex-
pansion of the rail system is anticipated.
Power and telephone distribution is showing a tendency to go under-
ground in at least a few of the most modern suburban developments. However,
the high expense involved, and the great demand for low cost homes in these
new areas, probably will delay the general acceptance of this technique
for many years. Only a very small portion of the low and the high voltage
transmission networks will be underground by 1980.
Security measures around public and private buildings will be
intensified. Thus, the use of chain-link fencing, street lighting, store-
front protective grill work, special metal doors and gates, can be expected
to increase markedly in the next ten years. Security measures also may
necessitate moving some types of external structures underground, where
they are less vulnerable to damage.
Also expected to increase with growth in population and the
trend to multi-family dwellings are metal roof-top facilities, such as
-------
84
The total of the new external carbon steel structures will
probably be of the order of 30 percent. This does not seem unreasonable
when one considers the growth trend for structural steel and plate shown
in Figure 2. It is predicted that 80 percent of the existing structures
will still be in use in 1980. Thus about 20 percent of the new carbon
steel structures represent replacement, leaving only 10 percent net growth
in carbon steel structures. The total amount of all new external structures
probably will increase by 30 percent^ of which 20 percent will be made of
materials not subject to corrosion damage by air pollution.
Increased Use of Alternate Materials
The ever increasing cost of maintaining steel systems in polluted
areas is a major reason for considering alternate, maintenance-free materials.
The expense for recoating a steel structure involves the cost of the paint
which has not altered greatly in the past ten years, and the cost of the
labor for its application. In the last decade, the hourly wage has in-
creased about 2.5 times and if one includes the overhead expense, the increase
is about 3 times. For example, Cincinnati contractors are currently charging
about $12 an hour (with overhead) as compared to about $4 an hour ten years
ago. The projected increase would be about $20 per hour by 1980,
which is a greater increase than the predicted growth in the economy.
Thus maintenance-free materials, in spite of their higher cost
in typical situations, are becoming more attractive with time. In fact,
avoiding the cost and inconvenience of maintenance, whether the environment
is polluted or not, is the most important reason for choosing alternate
materials instead of steel. In addition, there often are design economies,
lower labor costs for installation, and esthetic considerations that may
-------
50
40
30
20
o
to
10
c
o
o
S-I
(U
10
20
30
Production of Structural
Steel and Plate
(84)
Estimated Change in
External Steel
co
Ul
1960
1970
1980
FIGURE 2. GROWTH OF STRUCTURAL STEEL AND PLATE PRODUCTION AND ESTIMATED TOTAL CHANGE IN EXTERNAL STEEL
-------
For structural steel applications and for galvanized sheet, both
aluminum and self weathering steel are strong contenders as alternate
materials and should find much wider application by 1980. Statistical data
allowing one to predict growth trends for industrial and commercial appli-
cations of these materials do not appear to exist. For aluminum, the amount
of sheet products used in all forms of construction have increased during
the 60's from 0.58 to 1.18 billion pounds or over 100 percent. Aluminum
siding for residential service has grown at an even faster rate of 225 per-
cent during the past decade. In 1969, aluminum siding for industrial use
was 53 million pounds, or 15.5 percent of the residential market. This
application will probably double or triple by 1980.
Statistics were not found for self weathering steels and from a
discussion with a steel company representative one gained the impression
that only a few percent of the total structural steel production is of
this grade. However, the metals maintenance-free characteristics have
become widely known and the trend to its use in external structures such
as bridges and other exterior structures is expanding. It can be expected
to capture a greater portion of the market by 1980.
Aluminum coated steel sheet and fiber reinforced plastics are
currently increasing in their use and probably will acquire a higher
percentage of the market for painted and galvanized steel by 1980.
Alternate materials are being offered in new combinations to
reduce the price and increase the ease of installation. Roofing is now
available in thin aluminum or copper sheet bonded to plywood panels. To
compete with such products precoated galvanized roofing is being offered
for service in polluted areas.
For siding, coil coated steel or aluminum is now available for
-------
87
the use of all such newer materials will be to lower or eliminate main-
tenance costs that can be charged to pollution.
Changes in Corrosivity of the Atmosphere
The two major factors affecting the corrosivity of the atmosphere
to steel, and to the protective coatings used to retard or control the
attack on steel are (a) the level and type of pollution and (b) the amount
of moisture.
Atmospheric pollution by fuels containing sulfur is the most
important factor in determining its corrosivity to steel. The predicted
potential emission of sulfur dioxide into the atmosphere from all sources
is shown in Figure 1. This prediction, indicating a 55 percent increase in
the discharge of SO into the atmosphere by 1980, does not take into account
the steps, now underway, to find practical methods of reducing the emissions
from the burning of high sulfur coal and oil.
Switching to low sulfur fuels will make an immediate improvement
in the level of sulfur dioxide in a high pollution area. Unfortunately
the supply of low sulfur fuels close to the regions where they are most
needed are limited. Shipping costs combined with high demand has resulted
in a several-fold increase in the cost of these desirable fuels in typical
areas.
The use of low sulfur fuels may only be necessary during periods
of limited ventilation. Power plants, for example, according to recent
studies only make a major contribution to the sulfur dioxide level during
certain stagnant weather periods such as temperature inversions. By the
use of low sulfur fuels during these periods, which may be only 1 or 2
percent of the time, much of the contamination from the large users of
fossil fuels can be avoided. The regulation of the space heating-type
-------
88
Sulfur dioxide levels in 1980 will be determined by the inter-
action of regulatory, economic, and technical aspects of the problem. At
this time, four possibilities are seen:
Case A. If there is no regulation of pollution, the projected
consumption of sulfur bearing fuels will result in an increase in the rate
of sulfur oxide emissions by about 55 percent in 1980. This trend is shown
by curve A_ of Figure 3.
Case B. This case is based essentially on today's technology and
takes into consideration the availability of low sulfur fuels, the technical
difficulties of enforcing pollution regulations especially with space heating
and older power plants, and the general reluctance on the part of some
owners to accept the higher costs involved in converting over to pollution-
free methods. In this case, the normalized prediction is for an increase of
15 percent pollution in 1975, dropping to 10 percent by 1980 as is depicted
by curve B in Figure 3.
Case C. This case is based on the present strong public demand
for cleaner air and the efforts of federal and local authorities to respond
to this pressure. It appears that research in air pollution control will
be greatly expanded in the immediate future. This will increase the prob-
ability of major breakthroughs in the technology of pollution control. The
results of research combined with strict enforcement will lead to a 40
percent reduction in atmospheric contamination by 1980 as shown in Curve C
of Figure 3. Some slippage is allowed for in the enforcement of the new
regulations for older plants where limited space and lack of finances may
delay the implementation of suitable corrective measures.
Case D. If current legislation, plus that about to be enacted,
is applied without exception to all users of fossil fuels, the amount of
sulfur dioxide emission will be reduced 60 percent by 1975 and continue at
-------
50
4C
30
20
10
-30
-40
-50
-60
(A)
Pollution level based on total
sulfur dioxide emissions with no
regulation.
(B) Normalized estimate with strict
enforcement and best use of current
technology.
oo
VO
(C) Sulfur dioxide level involving
strict enforcement, pollution-free
new plants and major breakthroughs
in technology of control.
(D) Pollution level if current legis-
lation is completely enforced.
[Based on proposed ambient air
quality of 0.03 ppm SO ]
1960
1965
1970
1975
1980 Year
-------
90
The other factor affecting corrosivity of the atmosphere is a change
in the humidity or moisture. From ASTM panel studies, and from an examination
of the corrosion literature, there does not seem to be any evidence that the
corrosivity of the atmosphere has been altered significantly in any recent
ten-year period by changes in the availability of moisture and related
meteorological factors. In fact, ten years is a rather short period in time
to expect any major trend in the climate. Where the corrosion rate has
changed significantly, there is almost always evidence that a change in
pollution was involved. Usually there is a correlation with increased
corrosion and a higher rate of consumption of sulfur-bearing fuels in the
area.
Extra Annual Corrosion Damage Costs versus Pollution
Levels for lv>75 and 1980
The extra annual corrosion damage costs, chargeable to pollution
are a function of the corrosivity of the polluted atmosphere and the total
amount cf susceptible external structures exposed to contaminated air. In
assessing the corrosion costs for 1975 and 1980 the effect of the items
listed in Table 10 on the above two factors have been taken into consider-
ation. The changes in pollution levels for different situations were shown
in Figure 3. Considerations in regard to the amount of external steelwork
in 1980 also have been presented. The newly constructed steel structures
in 1980 are predicted to be 30 percent of today's external inventory, of
which 20 percent would be replacement of obsolete steel systems, the
rest being an increase of ten percent. In addition, another 20 percent
-------
91
TABLE 10. ECONOMIC AND POLLUTION FACTORS USED IN ASSESSING
PROBABLE COST OF CORROSION DAMAGE BY AIR POLLUTION,
1970 to 1980
Percent Change,
1970-1980
1. Population Increase
For Nation 11.
For metropolitan districts 12.
2. Energy Production
Increase in power plant capacity 78.
Space heating plant capacity 15.
3. Sulfur Oxide Pollution
Case A. No regulation 55.
Case B. With regulation and improved technology <10.
Case C. With strictly enforced regulation and major -40.
technological breakthroughs
Case D. Complete enforcement of current legislation -60*
4. External Structures
Old steel structures (decrease) -20.
Replacement steel 20.
New steel 10.
New other materials 20.
Total change 30.
Total increase in steel 10.
-------
92
The cost of maintaining the older structure, among the 80 percent
of the 1970 steel systems still in use in 1980, is likely to increase even
with improved technology in corrosion control and increased use of labor
saving techniques. Offsetting this will be the lower costs, in 1970 dollars,
of maintaining new structure incorporating design and other features to
reduce maintenance expense.
These costs are expected to balance out somewhat. Thus, allowing
for some increase in both steelwork and in population, it is predicted
that if there were no change in pollution, the per capita annual corrosion
cost will be essentially the same in 1980 as it is today.
On this basis and taking into account the four pollution cases
depicted in Figure 3, the per capita extra annual maintenance costs were
established and then converted to a national basis. These results, summarized
in Table 11, are discussed below:
Case A. With no regulation and a corresponding pollution increase
of 55 percent by 1980, the per capita cost will increase from $7.10 to
$9.22. The corresponding annual loss for the Nation will increase from
$1.45 to $2.10 billion.
Case B. With regulation applied where feasible, and with full
use of today's technology of control, the predicted pollution level will
be 10 percent higher in 1980. This will result in a per capita increase
from $7.10 to $7.63. The annual loss for the Nation will increase from
-------
93
TABLE 11. SUMMARY OF ESTIMATED ANNUAL AIR POLLUTION CORROSION DAMAGE
TO METALS FOR 1975 and I960.*
Percent Change in
Average Pollution
Annual Loss
in Billion
Dollars
Per Capita
Loss in
Dollars
A. No Regulation and Projected Increase in Sulfur Bearing Fuel
1970
1975
1980
0
25
55
1.45
1.78
2.10
7.10
8.26
9.22
B. Regulation Compatible with Full Use of Current Technology
1970
1975
1980
0
15
10
1.45
1.68
1.73
7.10
7.80
7.63
C. Strict Regulation Plus Major Breakthroughs in Control Technology
1970
1975
1980
0
0
-40
1.45
1.53
.99
7.10
7.10
4.36
D. Complete Enforcement of Current Legislation
1970
1975
1980
0
-60
-60
1.45
.47
.50
7.10
2.20
2.20
Based on changes in pollution shown in Figure 3, changes in population shown
-------
CO
M
TJ
o
n
co
C
O
•H
•H
CQ
d
••-I
CO
co
O
2.2
2.0
1.8
1.6
1.4
1.2
1.0
.8
.6
.4
A. No regulation.
B. Regulation with current technology.
C. Strict regulation plus major breakthroughs
in control technology.
Complete enforcement of present legislation
(a) The slightly higher cost in 1980 results from the predicted
increase in total amount of external susceptible systems in
use.
.2
70
75
80
Year
FIGURE 4.
COST OF AIR POLLUTION CORROSION DAMAGE TO METALS BASED ON CHANGES IN AIR POLLUTION
-------
95
Case C. With strict regulation and major breakthroughs in
the technology of pollution control, the pollution is predicted to decrease
40 percent by 1980. This will lower the per capita cost from $7.10 to $4.36.
The annual loss will be reduced from $1.45 to $0.99 billion, a saving of
almost $0.5 billion.
Case D, With complete enforcement and a reduction of 60
percent in pollution, the per capita cost will decrease from the current
$7.10 to $2.20 by 1980. The annual loss of $1.45 for 1970 will be reduced
almost one billion to 0.5 billion.
The total costs to the Nation for each of the four cases as
listed in Table 14 for 1970 to 1980, have been plotted in Figure 4. The
cross hatched area in the figure depicts the most likely range of probability,
namely from an increase of $0.3 billion to an actual saving of as much as
$0.5 billion over today's cost of $1.45 billion.
RECOMMENDATIONS
Numerous engineers, supervisors, and maintenance specialists
faced with the problem of keeping external steel structures and systems
exposed to air pollution in good repair, were interviexred during the course
of this project. In many cases, corrosion damage by air pollution is con-
sidered a normal cost of operation.
By cooperative arrangements, it is recommended that utilities,
industries, and local governments be encouraged to develop cost records
relating to the corrosion damage resulting from pollution. This^ information,
once developed, could be incorporated into a state-of-art report. Sections
in the report would deal with labor-saving maintenance procedures, savings
-------
96
of different types of corrosion control in polluted atmospheres. This
report, if widely disseminated, would assist those responsible for design
or for maintenance in reducing costs.
In particular there is a need for better cost records of the
performance of the more popular generic paint systems in protecting steel
in clean and polluted areas on actual structures and at test stations.
As a result of APCO's own research, as well as that conducted
by ASTM and other agencies interested in atmospheric corrosion, the rate
of attack for iron and for zinc in air of known humidity and sulfur oxide
content,can be predicted with considerable accuracy. No such correlation
appears to be available for the performance of steel in polluted air when
protected by each of the generic paint systems now recommended for such
services. It is suggested that such a program be inauguarated. One phase
would consist of the evaluation of paint panels at test sites of known
industrial pollution and the other would be a comparison of actual
structures in industrial and in rural environments. Both performance
in polluted air and cost comparisons would be developed.
Another aspect of external corrosion that was frequently mentioned
in discussions with experts is the cost of damage by deicing salts. For
fixed structures like guard rails, traffic signs, mail deposit boxes,
metal posts, bridge decks, etc., and for mobile equipment especially
automotive transport, the total cost of salt damage per year is likely
to be higher than that obtained for air pollution. In view of the serious
effect deicing salt has on the economy, it is recommended that an investi-
-------
97
REFERENCES
1. Robbins, Robert C., "Inquiry into Economic Effects of Air Pollution
on Electrical Contacts", Final Report, Contract PH-22-68-35 by Standford
Research Institute, Menlo Park, California for APCO, April 1970.
2. Survey and Economic Assessment of the Effects of Air Pollution on
Electrical Components, Contract CPA-70-72 by International Telephone
and Telegraph Corporation, Electro Physics Laboratory Incorporated
(In Progress).
3. Uhlig, H. H., "The Cost of Corrosion to the United States", Corrosion, _6,
No. 1, 29-33 (Jan. '50).
4. Keynes, A., "Economic Aspects of Corrosion", Chemistry and Industry
No. JL4, 398-400 (April 5, 1958).
5. Harvey, C. C., "Economic Factors of Atmospheric Corrosion Protection",
API Proceedings, 33, Section III (refining), 286-290 (1953).
6. Salmon, R. L., "Systems Analysis of the Effects of Air Pollution on
Materials", Final Report, MRI Project No. 3323-D, January, 1970.
7. Revenue Procedure 62-21, U.S. Treasury Department, IRS Publication No.
456, August, 1964.
8. Hudson, J. C.,& Stanners,JF, "The Effect of Climate and Atmospheric
Pollution on Corrosion, J. Appl. Chem., _3> P 86-96 (Feb. '53).
9. Schrier, L. L., (editor), "Atmospheric Corrosion", Vol. I, Corrosion
of Metals and Alloys, p 2.3 to 2.12 (1963).
10. Haynie, F. H., and Upham, J. B., "Effects of Atmospheric Sulfur on the
Corrosion of Zinc", Materials Protection and Performance, j), No. 8,
35-40 (1970).
11. Ludwig, J. H., Morgan, G. B. , and McMullen, T. B., "Trends in Urban
Air Quality", Trans. Amer. Geophys, Union, 51, No. 5, 468-475 (May '70).
12. Stephens, E. R., "Chemistry of Atmospheric Oxidants", Jrnl. of Air
Poll. Contr. Asso. JL9, No. 3, 181-185 (Mar., '69).
13. Stern, A. C., "The Changing Pattern of Air Pollution in the United States",
American Industrial Hygiene Association Journal, 28, No. 2, 161-5,
(Mar.-Apr. '67).
14. Baulch, D. W., "Relation of Gustiness to Sulfur Dioxide Concentration",
-------
98
REFERENCES continued
15. Boyer, A. E., and Kaylor, F. B., "Can The Cost of S02 Pollution Control
be Reduced?", Power, 111. 81 (April, 1969).
16. Thompson, D. H., "Atmospheric Corrosion of Copper Alloys", p. 129-140.
cf. Ailor and Coburn in Bibliography.
16.5. Scholes, I. R., and Jacob, W. R., "Atmospheric Corrosion of Copper
and Copper-Base Alloys during Twenty Years' Exposure in a Marine and
an Industrial Environment", Jrnl. Inst. Metals, .98, 272-80, (1970).
17. Ailor, W. H., "Aluminum Corrosion at Urban and Industrial Locations",
Jrnl. of the Structural Div., ASCE, j)5, No. ST10, 2141-2160 [Proc.
Paper 6828, October, 1969].
18. "Guide to Better Anodized Finishes", Metal Progress 93, No. 6,
60-66 (June, 1968).
18.5. McGeary, F. L., Englehart, E. T. , and Ging, P. J., "Weathering of
Aluminum", Presented at NE Regional NACE, Pittsburgh, Pennsylvania,
October, 1965 (Alcoa Reprint).
19. Larrabee, C. P., and Ellis, 0. B., "Corrosiveness of Various Test
Sites as Measured by Specimens of Steel and Zinc", Proc. ASTM 59,
183-201 (1959).
20. Larrabee, C. P., "Corrosion Resistance of High Strength Low-Alloy
Steels as Influenced by Composition and Environment", Corrosion, _9,
No. 8, 259-271 (August, 1953).
21. Bigos, J., Good Painting Practice, Vol. 1, Steel Structures Painting
Council, Pittsburgh, Pennsylvania, 1954 [Chapters 1, 9, 10, and 13].
22. Davies, C., and Hanford, N. J., "Metallic Coatings for the Protection
of Metal Steel From Corrosion at Ambient Temperature", Anti-Corrosion,
JL7, No. 9, 17-22 (September, 1970).
23. Horvick, Ernest W., "What to Consider in Specifying Zinc Coatings,
Metal Progress, 97, No. 6.
24. Porter, F. C., "The Hot Dip Galvanizing Industry", Metallurgia, 75,
241-247 (June, 1967).
25. Wachter, Aaron, "Relation of Corrosion to Business Costs", Corrosion,
JLO, No. 9, 273-78, September, 1954.
26. Charlton, F. R., "The Economics of Corrosion Control by Protective
Coatings", Corrosion Technology _12_, 11-16 (January, 1965).
27. Fancutt, F., and Hudson, J. C., Protective Painting of Structural
Steel", The MacMillan Company, New York, 1957.
28. Holbrow, G. L., "Atmospheric Pollution: Its Measurement and Some
Effects on Paint", Jornl. Oil Col. Chem. A., 45, 701-719 (October, 1962).
29. Keane, John D., "Protective Coatings for Highway Structural Steel",
Highway Research Board, Washington, B.C., Report 74, Results of Survey
(1969), Report 74A, Literature References (1012) and Abstracts (1970),
-------
99
REFERENCES continued
29.5. Hunger, C. G. , "Underfilm Problems That Lead to Coating Failure", Corrosion
Coiitrol'Reporter (Amercoa): Corp.) 8_, No. 1, 2-3,7.
30. Keane, J. D., Wettach, W., and Bosch, W., "Minimum Paint Film Thickness
for Economical Protection of Hot-Rolled Steel Against Corrosion",
.41, No. 533, 372-82 (June, 1969).
30.5. Mills, H. E., "Costs of Process Equipment", Chemical Engineering, 71,
133-156, (March, 1964).
31. Brouillette, C. V., "Cost Comparison of Protective Coatings for Steel", NCEL
Supplement, Technical Report R-501-Y-F020-03-01-001, (November, 1967). NFEC.
32. Gran, J. A., and Park, Ian M., "Bridging with Steel", Engineering
Journal, 50, No. 2, 18-24 (February, 1967).
33. Radecke, F. T., "Selection of Paints and Application in Combating
Atmospheric Corrosion", API Proc. 34, Sect. Ill, 37-49 (1956).
34. Pierce, R. R., "Key to Savings in Painting Costs", Chemical Engineering,
59, 149-53 (May, 1952).
35. Lynn, D. A.,and McMullen, T. B., "Air Pollution in Six Major U. S.
Cities as Measured by the Continuous Air Monitoring System", Jrnl.
A. Poll. Contr. Asso. JU>, No. 4, 186-190 (April, 1966).
36. Larsen, R.I., "Relating Air Pollution Effects to Concentration and
Control", Jrnl. Air Poll. Contr. Asso., 20, No. 4, 214-225 (April, 1970).
37. Thomas, M. D., "Review of Recent Studies of Sulfur Oxides as Air
Pollutants", Jrnl. of the Air Poll. Control Asso., 14, No. 12, 517-520
(December, 1964).
38. Tebbens, B. D. , "Gaseous Pollutants in the Air", Vol. I, Chapter 2,
P. 23-46, in Air Pollution: Air Pollution and Its Effects, A. C.
Stern, editor, 1968, Academic Press.
39. Blade, E., and Ferrand, E. F., "Sulfur Dioxide Pollution in New York
City", Jrnl. A. Poll. Control Asso., II, No. 11, 873-878 (November, 1969).
40. "Municipal Water Facilities--Communites of 25,000 Population and
Over, United States and Possessions", A Report, U. S. Dept. of Health,
Education,- and Welfare, Division of Water Supply and Pollution Control,
Washington, D.C. 20201, January, 1964.
41. News Release, October 13, 1970, by Department of Transportation,
Federal Highway Administration, Washington, D.C. FHWA-516.
42. Private communication from Paul Mara, Aluminum Association, New York
City, January, 1971.
43. Schmitt, R. J., and Mathay, W. L., "Tests Show Performance of Low
Alloy Steels in Chemical Plant Environments", Material Protection,
6, No. 9, 37-41.
44. Bureau of Census, Current Population Reports, Series P-25.
45. Rohrman, F. A., Steigerwald, B. J., and Ludwig, J. H., "S02 Pollution:
-------
100
REFERENCES continued
46. Statistical Year Book of the Electric Utility Industry for 1969,
Edison Electric Institute, New York, New York, September, 1970.
47. From staff specialists in economic trends; the range predicted is
4-6 percent per year.
48. American Iron and Steel Institute, "Annual Statistical Report", (1968).
49. Year Book of the American Bureau of Metal Statistics for the Year 1969,
(June, 1970).
50. Private discussion with H. J. Eppihimer, Porter Paint Company, Cincinnati,
January, 1971.
51. Petroleum Facts and Figures - 1967 Ed., American Petroleum Institute,
New York, New York.
52. Erickson, E. L., and Morgan, N. W., "Painting of Highway Bridges and
Structures", Chapter 10 of Reference 21, p 237.
53. "Aluminum Electrical Structures by Alcoa", Aluminum Company of America,
1961.
54. Pohlman, J. C., "What Price Maintenance Free Structures", Electrical
World, 159, No. 12, 72-74 (March 25, 1963).
55. "Performance and Maintenace of Industrial Fences in Corrosive Environments",
-------
101
BIBLIOGRAPHY
Ailor, W. H., and Coburn, S. K., (Editors), "Metal Corrosion in the
Atmosphere", ASTM STP 435, 1968 (June, 1967, meeting).
Ailor, W. H., "World-Wide Atmospheric Test Program", Corrosion Technology,
12, 12-17 (November, 1965).
Ailor, W. H., and Coburn, S. K., (Co-chairman), Metal Corrosion in the
Atmosphere, ASTM STP- 435, 1968 (Papers from 70th Annual Meeting,
Amercian Society of Testing and Materials, June, 1967.
Bigos, J., Greene, H. H., and Hoover, G. R., "Five-Year Test Results:
AISI Research Project on Paintability of Galvanized Steel", Jrnl. Paint
Technology, 39, No. 508, 316-27 (May, 1967).
Barton, V. K., "Der Einfluss von Staub Auf Die Atmospharische Korrosion
von Mettallen" (The Effects of Dust on the Atmospheric Corrosion of
Metals), Werkstoff und Korrosion, _9, No. 10, 547-9, 1958.
Covy, C. J., "Effect of Atmospheric Corrosion on Maintenance and Economics
of Overhead Line Hardware and Guy Strand", Corrosion, j4, No. 4, 133-140,
No. 5, 207-218; No. 5, 287-303.
Coburn, S. K., "A Low-Cost Maintenance-Free Structural Steel for Highway
Applications", Highway Research Record, 110 (1966), U. S. Steel Corporation,
Reprint.
Christofferson, D. W., "Steel Tank Maintenance", Water and Wastes Engineering,
.4, No. 3, 52-55 (March, 1967).
Downing, D. F., and Marsh, D. E., "Bridge Coating-Paint Maintenance
Program Costs Over a Million Dollars Per Year--", Corrosion, 17, No. 4,
26-33 (April, 1961).
Eberhardt, T. J., "Maintenance of Galvanized Steel Towers and Substation
Structures", District Conference Paper No. DP 62519, Am. Inst. of Elect.
Engrs., Memphis, Tennessee (April, 1962), see also Am. Painter and
Decorator, .38, No. 9, 28-30, 32 (1961).
Ewing, R. C., "Planned Maintenance Means Better Mileage From Storage Tanks",
Oil and Gas Journal, 80-88 (January 23, 1967).
Faith, W. L., "Economics of Air Pollution Effects versus Cost,of Control",
Jrnl. of Air Poll. Assn., 13, No. 8, 363-4, August, 1963.
Frazier, J. W., "A Kansas County Replaced Old Bridges Economically1', Civil
-------
102
BIBLIOGRAPHY continued
Godard, H. P., et.a., The Corrosion of Light Metals, p 92-104, John
Wiley & Sons, Inc.. 1967.
Golden, J., and Morgan, T. R. , "Sulfur Dixoide Emissions from Power Plants:
Their Effect on Air Quality", Science, 171, 381-383 (January 29, 1971).
Greenblatt, J. H., and Pearlman, R., "The Influence of Atmospheric Con-
taminants on the Corrosion of Steel", Chem. Canad 14, 212-5, November, 1962.
Harrison, J. B., "Prediction of Perfomrance of Primers and Its Relationship
to the Performance of a Full Paint System in Practice", British Corrosion
Journal, 4, 58-65 (March, 1969).
Hall, J. R., "The Protection of Bridges by Galvanizing", Corrosion Prevention
and Control, _17, No. 2, 12-15 (April, 1970).
Helms, F. P., "New Concepts of Zinc[paint] Coatings Reduce Maintenance Costs",
Amercoat Corporation, Corrosion Control Rep., 8, No. 1, 4-6 (1958).
Kemp, W. E., "Product Fallout--A Serious Corrosion Problem", Ind. Eng.
Chem., 51, 75A-76A, (July, 1959).
Long, J. B., "The Abilities of Terne Plate", Tin and Its Uses, No. 79,
5-6, (1968).
Mead, J. W., (Editor), The Encyclopedia of Chemcal Process Equipment, p. 941
(1964).
Messhem, R. B., "The Protection of Galvanized Steel", Corrosion Prevention
and Control, 16, No. 3, 22.
Melvin, J. S., "Evaluation of Some Materials and Coatings for Utility
Pole-Line Hardware", Corrosion, 17, No. 4., 14-15, 18-19, (April, 1961).
Morgan, N. W., "Corrosion Problems in Highway Maintenance", see extract in
Corrosion, _17, No. 4, pp 22-23 (April, 1961).
Maurin, P. G., and Jonakin, J., "Removing Sulfur Oxides From Stacks",
Chemical Engineering 77, No. 9, 173-180 (April 27, 1970).
Nylen, P., and Tragardh, K. F., "Exposure Tests of Paints in Sweden",
Corrosion Technology 2, 182-85 (June, 1955).-
Oliver, James, "Developments in Pretreatment and Finishing of Zinc-Coated
Products", 31., No. 8, 48-53, (May, 1967).
Porter, T. C., "Painting of Zinc Surfaces", Br. Corr. Jrnl., U_, 179-186
(July 1969).
Popper, H., Modern Cost Engineering Techniques, McGraw Hill Book Company,
-------
103
BIBLIOGRAPHY continued
Reid, W.T., "What About Air Pollution by Power Plants?", Battelle Research
Outlook,, 2, No. 3, 21-23 (1970).
Rohrman, F. A., Steigerwald, B. J., and Ludwig, J. H., "SO Pollution: The
Next 30 Years", Power, 113, 82-83 (May, 1967).
Rugger, George R., "Weathering Resistance of Plastics", Materials in
Design Engineering, 59, 69-84 (January, 1964).
Robinson, E., and Robbins, R. C., "Gaseous Sulfur Pollutants from Urban
and Natural Sources", Jrnl. A. Poll. Control, Asso. 2C), No. 4, 233-235
(April, 1970). See also American Petroleum Institute Report "Sources,
Abundance, and Fate of Gaseous Atmospheric Pollutants", February, 1968,
by same authors (SRI PR-6755).
Shepard, D. S., "A Load Shifting Model for Air Pollution Control in the
Electric Power Industry", Jrnl. APCA, 20, No. 11 p 756-761.
Sareda, R. J., "Atmospheric Corrosion of Steel", Ind. Eng. Chem., 51,
No. 9, pt. 1, 79A-80A (September, 1959), and "Atmospheric Factors Affecting
the Corrosion of Steel", Ind. Eng. Chem 52, No. 2, 157-160 (February, 1960).
Still, J. M., Jr., iWheels are Turning for Terne Products", Iron Age, 94^
28, (October 15, 1964).
Stanners, J. F., "Use of Environmental Data in Atmospheric Corrosion Studies",
British Corrosion Journal, j>, 118-121 (May 1970).
Smith, D. W., and Day, K. J., "Protection of Steel Bridges from Corrosion",
British Corrosion Journal, _5, 151-158 (July, 1970).
Squires, A. M., "Keeping Sulfur Out of the Stack", Chemical Engineering,
77., No. 9, 181-89 (April 27, 1970).
Spirtos, R., and Levin, H. J., "Characteristics of Particulate Patterns,"
1957-1966, AP-61, HEW.
Tice, E. A., "Effects of Air Pollution on the Atmospheric Corrosion Behavior
of Some Metals and Alloys", Jrnl of The Air Pollution Control Association,
JL2, No. 12, 553-559 (December, 1962).
Tropp, F. E., "Evaluation of Zinc Surfaces for Coating Industrial Production
Line Products", .39, No. 507, 225-254.
Urone, Paul and Schroeder, W. H., "SO in the Atmosphere--" Environmental
-------
104
BIBLIOGRAPHY continued
Yocom, J. E., and McCaldin, R. 0., "Effect' of Air Pollution on Materials
and the Economy", Vol. I, Chapter 15, 617-651, Air Pollution,:
Air Pollution and Its Effects , AC Stern, editor, 1968, Academic Press.
"Alcan Handbook", Aluminum Company of Canda, 1970.
"Atmospheric Effects Can Be Correlated", Canadian Chemical Processing,
51, No. 8, 40-43 (August, 1967).
Aluminum Statistical Review, The Aluminum Association of New York, (1969),
"Air Pollution versus Materials Costs", Materials Protection, _6, 47-54,
May, 1967.
"Long Terne Sheets Long on Processability", Metal Progress, 97, No. 6,
7-8, (June, 1970).
Steel Products Manual "Carbon Sheet Steel, Coils, and Cut Lengths",
American Iron and Steel Institute, May, 1970.
Metal Statistics, 1969, The American Metal Market Company (Also daily
issues of The American Metal Market).
Bureau of Census-Pocket Data Book, USA-1969.
-------
APPENDIX A
COMPUTATION OF UNITS AND/OR WEIGHT OF SYSTEMS
-------
APPENDIX A
COMPUTATION OF UNITS AND/OR WEIGHT OF SYSTEMS
IN USE AT CURRENT TIME
Air pollution corrosion costs are found in this study to be equal to
the cost of maintaining components over the lifetime of the system in which they
appear. Specifically, lifetime maintenance cost is the total number of
paintings, n, time the cost of one paint job, p, over the useful and depreciable
lifetime of the system, according to equation 7, page 17.
^
Beginning with a number for the lifetime of any given system, the
number of paintings, n, is the lifetime of the system divided by the lifetime of
a single paint job. While the numerator is constant, the denominator decreases
with increased air pollution and n increases with increased air pollution.
The cost of a paint job is the cost of paint and application per
unit area, times the total area of the component. Therefore total area of
all certain components in certain systems in use at the moment needs to be
determined. That total area of all certain components is the weight of such
components in use divided by the area per average sized component; or it is the
number of such components in use times the area per average sized component.
The purpose of Appendix A is to describe the procedure used to deter-
mine the weight of specific systems in useful service as of 1970.
Consonant with real-value theory each specifically defined system has
an acceptable useful and depreciable lifetime, given in years. Many tanks,
for example, are in use long after their depreciable lifetime of 50 years.is
-------
A-2
The useful and depreciable lifetime of systems are defined roughly at
the 4-digit level of the Standard Industrial Code, and have been statistically
established by the U.S. Department of Internal Revenue, and published in
"Depreciation Guidlines and Rules", Revenue Procedure 62-21. For example, it
sets the statistical useful lifetime of 13 various buildings, such as apartments
at 40 years, dwellings at 50 years, warehouses at 60 years, etc. Also, it sets
agricultural machinery and equipment at 10 years, recreation and amusement
systems at 10, logging equipment at 6 years, and fences, bridges, shop
machinery, etc., at 30 years, and many others. These lifetimes are accepted as
tax write-off guidelines, and are used as starting points for calculating
assessed values of individual properties or systems by assessors. They are used
here for what they are; the statistically average useful lifetime of all systems
so designated by 4-digit terminology.
System lifetimes are used in this report to compute the number, or weight
of specific systems in useful service by integrating shipment data from 1970 back
over its lifetime of useful and presently depreciable years as depicted in Figure A-l.
Data from the Census of Manufactures normally provides points for the years
shown, 1967, 1963, 1958, 1954, and 1947. For some systems still earlier
figures are available. For more recent years, annual statistics may for some
systems fill in points between 1967 and 1963. Where in-between data points
are missing, interpolated points are computed on a straight line. Points
between 1967 and 1970 are extrapolated either on a line representing average
industrial growth for the system in question, or on a slope which by inspection
appears to be reasonable.
The integrated total, if in units, is converted to total weight on the
basis of average size and weight of this average unit as indicated in manufac-
turers design literature. From data in these same sources, the tonnage can be
-------
4-1 CO
C iJ
O -r-l
S C
O. 3
•1-1
.c n
c/i o
<0 in
3 C
c o
X
I I ' 40
45
Total Units in Useful Depreciable Service
/
50
55 i i
60
j 65 ( (
— Lifetime of System A
Years
>
-------
A-4
An example is given in Tables A-l,and A-2 for 36122-power distri-
bution transformers. In Table A-l, two years are shown with the product mix
breakdown for 1963 and 1958. Wherever such breakdowns were given, the year
totals were built on the indicated product mix. Only data on latter years
provided such refinement, however.
In Table A-2 the underlined years are data points, and the others are
-------
TABLE A-l. 36122 - POWER DISTRIBUTION TRANSFORMERS
1963
3612 - Pwr.,Distr., To t a J KVA Surf. Ar.
& Spec'ty Txfmrs #Units 10J $103 sq.ft.
1958
Tot KVA Surf. Ar.
#Units 103 $103 sq. ft.
36 12 2 - Pur .Uistr. Txfmrs.
05-
08-
12-
14-
16-
17-
21
23
00-
<15 KVa < 15,000 V 419
(Pole type, 11. 2 ft^/unit).
15-50 KVA, < 15,000 V 32°
(Pole type, 21.1 ft^/unit)
51-167 KVA, < 15,000 V 74
0?ole type, A = 30 ft2/Unit)
16G-500 KVA, < 15,000 V 11
500 KVA>15,000 V
(Station type, =138 ft2/unit)
501-5000 KVA, all E 8
(Station Type,A=175 ft2 /unit)
5001-10,000 KVA, all E
(Station type,A=175 ft2/unit)
Over 10,000 KVA, all E
(Primary sub sta,
A = 208 ft2/unit)
Pwr, distr txfmrs , n.s.k. 2
(208 ft2/unit)
TOTAL UNITS, $103 VALUE ' 837
(7T2) TOTAL 17,546
(YRS) LIFETIME
,400 4,889 80,155 4,697,280
,135 9,894 91,967 6,754,638
,282 7,864 60,748 2,228,460
,435 3,992 24,041 1,577,616
,705 12,196 60,128 1,523,375
697 5,240 18,795 121,975
860 59,858 97,501 178,880
,231 277 1,159 464,058
,745 104,206 434,494 17,546,272
,272
30
314,394 3,552 76,093 . 3,521,213
1
137,844 7,188 87,315 2,908,508
33,337 3,781 30,481- 1,000,110
7,281 2,338 23,580 1,004,778
5,319 7,771 66,417 1,359,925
661 5,172 23,491 115,675
677 34,210 108,279 ' 140,816
2,060 255 1,070 428,480
501,573 64,267 421,726 10,479,505
(FT2) LIFETIME IN-
-------
TABLE A-2. INTEGRATIONS OF POWER DISTRIBUTION
TRNASFORMERS IN VARIOUS UNITS
(Average lifetime = 30 years)
Year
1970
9
8
1967
6
65
4
1963
2
1
60
9
1958
7
6
55
1954
3
2
1
50
9
8
1947
6
45
4
3
2
1
40
1939
i
Number of Units
1,600
1,542
1,494
1,443,947
1,272
1,157
964
837,745
769
700
634
566
501,573
531
560
591
623,565
773
894
979
1,032
1,051
1,042
994,386
903
807
715
618
526
435
338
246,275
Total Total
3 2
KVA x 10 Area Ft
33,440
32,228
31,225
180,017 30,242,884
26,585
24,181
20,147
104,206 17,546,272
16,072
14,630
13,251
11,829
64,267 10,479,505
11,098
11,704
12,352
78,947 13,026,273
16,156
18,685
20,461
21,569
21,966
21,778
125,895 20,072,924
18,873
16,866
14,944
12,916
10,993
9,092
7,064
32,231 5,157,000
Total
Value $10
830
800
775
748,900
660
600
500
434,494
432
429
427
424
421,726
427
431
436
441,089
407
374
340
307
273
240
205,943
187
167
148
128
109
90
70
51,005
-------
APPENDIX B
SCREENING PROCEDURE FOR DEFINING
-------
APPENDIX B
SCREENING PROCEDURE FOR DEFINING
SYSTEMS AND COMPONENTS
If the cost of air pollution corrosion is to be limited to real costs,
or losses in the value of systems, then a determination of the most vulnerable,
valuable, and expensive component systems is the first order of business. This
determination began with component/system definition, with the aid of the SIC
array, and continued through a series of qualitative screenings, evaluations, and
simplifications.
The first step was to identify the two broadest categories, Construction,
Division C, and iManufacturing, Division D, as the SIC groupings in which all the
component systems would appear. Other divisions represent services and/or
utilizers of the component/systems contained in Divisions C and D.
Construction contains the largest exposure of metal components. Manu-
factured products also contain a great deal of metal, but mostly serving either in
a protected environment, or in a hostile environment made malignant by excessive
heat, stress, corrosion and other conditions that really mask the less aggressive
air pollution sector of the out-of-doors environment. These generalizations were
less clear at the outset of the study. To realize these observations, however,
makes it easier to follow the extensive, complicated screening process pursued
by this study.
In the first stages of the project, however, this was not realized.
What the team saw were two large bodies of SIC statistics on systems of con-
struction and systems of manufacturing. That some manufactured products went
into construction was clear enough, but their preponderant importance was obscure,
and perhaps not even credible at that moment. In addition, the philosophical
starting point was at the surface vulnerable to air pollution corrosion, while
-------
B-2
the two extremes would have to be linked in a practical, meaningful
way.
Since it is clearly impractical to tabulate every surface in every
component in every subsystem in every system in existence, the study pursued a
course of discrimination to eliminate low-value surfaces, surfaces in protected
systems, and otherwise ones insensitive to the marginal deterioration of air
pollution. Accordingly the study developed a system of screens for sorting out
only corrosion sensitive component-systems of value. The screens were used to
sort out and winnow down the systems and components originating in Division D--
Manufacturing, a group of some 20 manufacturing industries. First the industries
that manufacture metal containing systems were sorted out of these 21 Standard
I-ndustrial Classification, 2-digit categories, as follows:
25 - Furniture and Fixtures
33 - Primary Metal Industries
34 - Fabricated Metal Products
35 - Machinery, Ex.Electrical
36 " Electrical Machinery
37 - Transportation Equipment
38 - Professional, Scientific, etc.
39 - Miscellaneous Manufacturing
All others were ignored, as not generating metal containing components and systems.
An exception to this is 19 - Ordnance, which does produce metal containing
components and systems. That industry was eliminated early because of the general
expendability of its products, the excessive aggressiveness of the service environ-
ment encountered by ordnance systems, and the fatuous projection of the storage
-------
B-3
Procedural Summary
The screening, winnowing and combining procedure applied to Division D—
Manufacturing statistics involved five steps, as summarized in Table B-l. This
procedure accomplished two things. It first showed that exposed metal components
of any importance either resided on or comprised structural systems. All
others were at least an order of magnitude less important according
to in-situ value and tonnage. The scope of this report does not allow for
elaborations on this point, interesting as it may be. The other thing the
screening accomplished was to cut through an almost infinite environment of
statistics and pull out the important salient portion. Fortunately, it reduced
the^number to a quantity that could be meaningfully handled, but barely within the
resources.of the project, as summarized in Table B-l.
Qualitative Screens
All the components in outdoor systems used by society suffer
s.ome deterioration from pollution. But more deterioration may occur from other
extraneous service conditions, such as erosion, deicing salt, deformation,
oxidation, and possibly other effects. Often these other service conditions
are considerably more severe than air-pollution deterioration, and effectively mask
it. Thus of prime concern is to choose from the many in existence those components
and systems that are solely or mainly affected by air pollution at a cost to
society. Also of prime concern is to choose those components and systems that
.fit value sensitive criteria, i.e., those that at once require a large amount
of pollution-sensitive metal in their manufacture, are associated with systems
of higher shipped value, and involve large metal requirements and/or large total
-------
B-4
TABLE B-l. STATISTICS OF SCREENING, WINNOWING,
AND COMBINING
Screen
Description
Input
Survivals Rejects
First: qualitative
Second: qualitative
Third: eliminate Cu
and Al components
Fourth: high profile
items
Fifth: combine
200-4-digit
•a""""*"""*
244-5-digit
P"I
101-6/7-digit
43-6/7-digit
19-6/7-digit
60 140-4-digit
51 193-5-digit
43 58-6/7-digit
19
24-6/7-digit
-------
B-5
The first qualitative screen dealt in four-digit SIC categories of
metal containing components that have metal exposed to the service environments.
There were 200 such categories entered into the first screen, .
Within these categories reside about 1600 specific components,
which when thought of in terms of alternate metals of construction multiply by
that number into 4800 separate considerations. The first screen separated out
for rejection those categories that appeared (a) to be insensitive to pollution, (b) to
perform its function in a protected inside environment, and (c) to be
subject to other more damaging service conditions (erosion, deformation,
scoring, etc.) than could be occasioned by pollution corrosion. Judgments
were actually exercised by a screening task force membered by corrosion
engineers and by design specialists in the manufacturing industries and
by materials consultants. Of the 200 categories (4-digit) which entered
the first screen; 140 were rejected by task-force judgment.
A typical working format is reproduced in Table B-2. The Committee
tockeach 4-digit category, and exercised judgment about the probable functional
sensitivity of contained systems and components to atmospheric corrosion; notated
their probable service sites as protected or unprotected; contemplated likely
masking effects in the service environment; and finally disposed of it as an
"overriding keep" for subsequent screening, "provisional keep", or "reject".
If a category had even one item in it that represented a questionable reject, it
was kept provisionally; or one that might contain an exceptional situation, it
was overridingly kept in.
The second screen was again qualitative. First, the 60 four-digit-
category survivors were expanded into 244, five-digit-category items, which
vere then entered into the second screening assessment. Here again, the
same qualitative criteria as in the first screen were applied. One hundred nine (109)
-------
TABLE B-2. SURVIVAL COMPONENTS FROM THE QUALITATIVE SCREENS
System Destination
Sic Code
25
j43
33
* 151
* 6
7
51
16701
801
000
17601
34
* 411
* 2
* 3
System Description
Furniture
Metal, Porch, Lawn Outdoor
Primary Metals Industry
Noninsulated Wire Rope
Fencing & Fence Gates
Ferrous Wire Cloth & Other Woven
Barbed & Twisted Wire
C.R. Sheet & Strip
C.F. Bars & Shapes
C.R. Sheet & Strip, NSK
Steel Pipe & Tube
Fabricated Metal Products
Fabricated Structural Iron Steel for
Buildings and Around Buildings
Fabricated Structural Iron Steel for
Bridges
Other Fabricated Structural Iron Steel
(Transmission Towers, Substations,
Radio Antennae Towers, Aluminum for
Metal
Employed
Steel
Aluminum
Steel (G)
Steel (G)
Steel
Steel (G)
Steel
Steel
Steel
Steel
Steel
Aluminum
Steel
Aluminum
Steel
Copper
Aluminum
'Metal
Rank
41
19
30
33
45
39
2
8
1
4
5
9
15
15
9
12
18
M
• • QJ •
M Vi E — I M
1 4J 4J 6 O 4J
C to a) to o C to
o c EC o o e
12 O O O i-l O
O 33 0 . XJ 0
4J • • • 4-1 4-1 •
to T) •« "O to i-l T3
3 i— I i-4 i-l 3 -U .— I
•O i-l to iH T3 to i-l
C 3 j • p.
3 O. i-l
.4-1 i-l 3
ij o 3 a1
4J 3 0- W
to I-l H
C 4J O
O w 1-1 1-1
U RJ C3 4J
I-l 4J' tO
c c c. e
S) tH C3 O
~B~ CD CD
X
X
X
X
X
X
X
X
X
X
X
X
transit
c
i-i
~
X
X
X
X
-------
TABLE B-2. (Continued)
Sic Code
0
* 421
* 422
NSK
Metal
Metal
•
System Description
Doors & Frames, Exterior Storm
Window Sash, Frames, Exterior
Storm
* 423
424
425
420
432
Metal
Metal
Sash
Metal
Ditto
Molding, Trim, Store Fronts
Combination Screen & Storm
and Doors
Window and Door Screens
, NSK
Fabricated Steel Plate, Including
Metal
Employed
Steel
Steel
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Aluminum
Steel
Aluminum
Steel
Copper
Aluminum
Steel
Steel
Metal
Rank
25
23
22
3
28
25
4
22
6
42
8
44
21
22
36
12
i u u
C in a) to
0 C 6 C
20 O 0
o as o
4J . . .
«> -O -O T3
3 t-4 1-1 ft
•O -H Bl -i-l
C 3 0) 3
A ~A~~
X X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
System Destination
>J O
u e 4-i u ••-<
• 3 Of) i-l —1 to
*j 4-> • o nj ra *j c
•J) -H t) ^1 4J M O
3 4J —J • U-l i-4 o !-"
T3W-r< Cd 0. E U
CC3 OM !3 O C
Hh-ICQ O^-* O Q l-l
A B CD CD ~
X X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X X
-------
TABLE B-2. (Continued)
Sic Code System Description
* 435-9 Metal Tanks
* 441-0 Sheet Metal Work (Roofing, Siding,
Draining)
460 Architectural, Ornamental Metal Work
(Fences, Gates, Stairs, Fire Escapes,
Railings, Open Flooring Grates, Grills)
* 492 Prefabricated & Portable Metal
Buildings
493 Miscellaneous Metal Building Material
(Curtain Walls)
814-5 Wire Cloth, Ferrous, Nonferrous
* 816 Fencing, Fence Gates, Exterior that
made in 3315
819 Other Fabricated Wire Products (Wire
Chain, Barbed & Twisted Wire not made
in 33159, Wire Baling
Metal
Employed
Steel
Steel (G)
Copper
Aluminum
Steel (G)
Copper
Aluminum
Steel
Copper
Aluminum
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Aluminum
Steel (G)
Copper
Aluminum
'Metal
Rank
7
6
3
1
19
15
8
7
13
12
18
5
26
17
25
33
15
10
9
28
vJ
1 4-1
e w
o c
25 O
CJ
•
4J •
CO *O
3 f-l
•O 1-4
C 3
t-4 C3
A
X
X
X
X
X
X
X
X
X
X
X .
X
u
4J
a) co
E C
o o
x u
• .
T3 "O
*r4 i— 1
W -H
(U 3
pi co
A
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
System Destination
u cu
o t-i • a.
§ -4 M 3 O. -H
g CJ 4-1 -4-1 1-4 3
o C w (jo 3 cr
u o e aj 3 cr fcj
•ri o -4l-l!S) O^—" O Q 1-4
A ~B~ CD CD ~
X X
X
X
X
X X
X X
X X
X
X
X
X
X
X
X
X
X X
X X
W
I -
-------
TABLE B-2. (Continued)
System Destination
Sic Code System Description
35 Machinery, Exterior Electrical
351 Conveyors for Conveying Equipment
2 Conveying Parts & Attachments
36 Electrical Machinery
* 122 Power & Distributor Transformers
3 Power Regulators, Boosters, Reactors
131 Switchgear, Including Power Saw,
Gear Assembly
2 Circuit Breakers
Metal
Employed
i
Steel
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Copper
Aluminum
•Metal
Rank
26
11
23
37
26
30
38
10
31
18
1
21
21
4
17
32
8
26
M
• • QJ •
M M g -< )J
i 4J u gnu
co> a) to o e to
o a EC u o c
*Z O O O •-< O
O B5 W - JJ O
• 3
4J . . . 4J JJ .
w "O "T3 "O o) •»-( *O
3 t-l -^4 r-l 3 iJ i-<
•X3 v-l 0) -H "tj CO •«-)
C 3 -4
-
I
-------
TABLE B-2. (Continued)
System Destination
Sic Code System Description
220 General Industry Power Circuit Devices
and Controls
* 425 Outdoor Lighting Equipment
6 Other Electric and Nonelectric Lighting
Equipment
430 Current Carrying Devices, Including
Lightning Rods
* 441 Pole Line & Transmission Line Hardware
2 Electric Conduit & Fittings
Metal
Employed
Steel
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Copper
Aluminum
'Metal
Rank
24
6
16
35
14
20
40
19
27
31
2
13
28
10
24
27
23
10
n
e to
0 C
"Z. 0
o
U) TJ
3 r-l
*O *ft
C 3
M 03
A
X
X
X
X
X
X
X
X
X
X
X
X
J
JJ
o
-------
TABLE B-2. (Continued)
System Destination
Sic Code
37
121
910
39
931
2
System Description
Transportat ion
Auto Passenger Cars Assembled &
Knocked Down
Trailer Coaches, Housing Types
Miscellaneous Manufacture
Luminous Tube & Bulb Signs
Nonelectric Signs and Advertising
Display
Metal
Employed
Steel
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Copper
Aluminum
Steel
Copper
Aluminum
• • CU •
r jj jj 1 a jj
C W QJ CO O C V)
o e EC u o c
"Z. 0 00 -r4 0
"O »H W «H *O CO 'H
C 3
-------
B-12
The third screen was semiquantitative, in which components were screened
in terms of availability vs metal construction. First, the 51 5-digit-item
survivors were expanded into 101, 6 and 7-digit component considerations, as
each component was recognized as a steel item separate from copper, and aluminum,
and from galvanized and vice versa. Specifically, these components included
such things as aluminum storm doors, curtain wall, etc.; also copper downspouts,
flashing, etc.; and galvanized wire fencing, downspouts, and flashings, etc.
Aluminum and copper components appeared not to require painting,
except for cosmetic appeal, nor did they appear to have shortened service life
under the influence of purely air pollution. They were, therefore, eliminated from
further consideration as metals having higher costs in polluted air than in
clean air. Some items were retained, however, to study the extra expense involved
if aluminum or copper were substituted for high-maintenance steel systems. This
elimination left 43 vulnerable steel components.
At this point, it was observed that the surviving specific components, Table B-2,
are primarily found in systems of construction, i.e., buildings of all kinds,
utilities installations, and infrastructure, embodying such components as bridges,
towers, tanks, fencing, etc. Other systems comprising a region were elimina-
ted as being masked by other effects. For example, mobile transportation
equipment, and roadside fixtures were eliminated because they are more
attacked by de-icing salts, impact of small stones, and other destructive
forces than by air pollutants. All that is left is that tonnage of 19
component categories associated with construction systems. Later, the structure of
the data sources made it convenient to group these together into 14 component
-------
B-13
There are some 41 categories of outdoor-construction systems
in which components appear which on a 41 by 14 matrix turns up 574 component-
system calls. Some of these systems operate in rural atmospheres, however,
such as social and recreational complexes and structures, while some are
relatively small users of components, such as hospital and church buildings.
Eliminating these narrows the number 41 down to 21, which about halves
the number of cells. Then, by grouping similar items again, the number 21
reduces to nine. Thus, with 14 categories of components operating in these
nine systems, there turns up statistically 126 cells, or unique component-system
pairs to examine.
Among these cells are many absurdities, such as structural steel in
bridges opposite electrical machinery systems and some relatively trivial, such
as steel storage tanks across from residential buildings. Other pairs were
reasonable, but small enough to be grouped with other similar pairs, such as
steel transmission towers across from a grouping of utilities, including electric
power, telephone and telegraph, airports and military construction. Happily
this last grouping was essentially done for us by the manner in which statistics
'on components are published.
Finally, we ended up with 23 pairs of component-system and component-
system groupings. Data on those pairs were developed around nine major components,
-------
APPENDIX C
CALCULATION OF CORROSION COSTS FOR STEEL STRUCTURES AND
-------
APPENDIX C
CALCULATION OF CORROSION COSTS FOR STEEL STRUCTURES AND
SYSTEMS IN POLLUTED ATMOSPHERES
C-l Calculation of Corrosion Maintenance for Water Storage Tanks,
Petroleum Tanks, and Chemical and Industrial Tanks
C-2 Calculation of the Extra Maintenance Painting Expense, Chargeable
to Pollution, for Structural Steel Bridges
C-3 Calculation of Air Pollution Corrosion Costs for Externally Mounted
Power Transformers
C-4 Calculation
-------
C-2
TABLE C-l. CALCULATION OF CORROSION MAINTENANCE FOR WATER STORAGE TANKS,
PETROLEUM TANKS, AND CHEMICAL AND INDUSTRIAL TANKS
1. Elevated Water Storage Tanks
First Estimate of Tankage Based on Volume Data
1. Total estimated storage in elevated steel tanks is 11,000 x 10 gallons.
2. Typically elevated storage tank = 1 x 10 gallons (1 MG).
3. Equivalent No. of 1 MG Tanks = 11,000.
Second Estimate Based on Steel Plate Tonnage
4. Total tonnage of steel plate in all municipal water systems is 7,100,000.
Of this, some 70 percent is estimated to be in elevated storage tanks.
5. A 1 MG tank is about 350 tons.
6. Number of 1 MG tanks in service is .70 x 7,000,000 -f 350 = 14,300
7. Average of Item No. 3 and No. 6 = 12,700
Calculation of Pollution Cost
8. Area of hemisphere + roof = 10,050 + 5,020 = 15,070 sq.ft. Other designs
range from 12,000 to 30,000 sq.ft. Use 20,000 sq.ft. for calculation.
Annual extra maintenance cost = 0.0167 (see Table 4).
20,000 x 0.0167 = $334/yr for 1 MG tank.
9. Based on observation, about 80 percent of the elevated tanks are exposed
to contaminated atmospheres.
Total annual extra cost caused by pollution is
.80 x 12,700 x 334 = $3,400,000.
2. Petroleum Product Storage Tanks
First Estimate Based on Capacity
1. Bulk storage capacity (API Data) in 1962 - 17,981,306,000 gallons. Pro-
jected to 1970, using groxvth in petroleum production to gage capacity
increase. 20,000,000,000 gallons.
-------
C-3
TABLE C-l (continued)
2. Petroleum Product Storage Tanks
First Estimate Based on Capacity
3. A. Estimate 2/3 of refinery size, 100,000 barrels.
B. Estimate 1/3 of distribution size, 10,000 barrels.
9
A- °'64 2 X°10610 = 3>150 tanks> 100; 000 barrel size. Add 10% to allow
for ullage.
9
B- °'34 2 x°10510 = 15>750 tanks; 10,000 barrel size. Add 10% to allow
for ullage.
2
4. Area of tanks C = 7 -it-o07 C = capacity in barrels
A. For 100 foot diameter, height becomes 75 feet.
Area of cylinder plus roof A =H^ D and 7?p2 = 31,450 sq.ft.
4
B. For 55 foot diameter, height becomes 24 feet.
Area of cylinder plus roof is equal to 6,530 sq.ft.
5. Annual extra maintenance cost per tank (see Table 4).
A. 31,450 x 0.0473 =$1,490
B. 6,530 x 0.0473 =$ 309
6. Total annual maintenance.
A. 3,464 x 1,490 = $ 5,160,000
B. 17,325x309 = $ 5,360,000
Total $10,520,000
Second Estimate Based on Tonnage
7. Total tons in use 1,950,000. Area Factor (one side) for Heavy Tanks is
120 sq.ft./ton
8. Loss = 1,950,000 x 120 x 0.0473 = 11,100,000
Summary for petroleum storage, annual loss per year:
First method 10,520,000
Second method 11,100,000
Average 10,810,000
3. Chemical and Industrial Storage Tanks
1. Total tons in external use 4,830,000.
2. Typical area factor for chemical tanks—200 sq.ft./ton.
-------
C-4
TABLE C-l (continued)
3. Chemical and Industrial Storage Tanks
4. Estimate 85 percent of industrial tanks in polluted atmospheres.
5. Annual cost of controlling corrosion damage by air pollution.
4,830,000 x 200 x 0.0391 x 0.85 = $32,100,000.
Note: No alternate statistical source was found to enable an alternate
calculation. In view of the good agreement in petroleum tanks,
this figure should also be of the right order.
Summary:
Water $ 3,400,000
Petroleum 10,810,000
Chemical 32,100,000
Total annual $46,310,000
loss for ex-
ternal tanks
Footnotes for Table C-l
Item 1. Elevated Storage Tanks
1. The total gallons of water in elevated storage for communities of 25,000
people or more (reference 40) was obtained by adding up the gallons of
water listed on each of 20% of pages, selected at random, and multiplying
the reaiIt by 5. Where elevated reservoirs instead of steel tanks are
used, the gallons listed were omitted.
2. The data were supplied by Gerry Francis, engineering department, Columbus
Waterworks. The literature study also indicated the 1 MG tank to be
typical and 20,000 sq.ft. well in the r?nge of representative tanks for
designs ranging from pedestal to cylinder types.
4. The tonnage was derived from SIC statistics. From inspection of water
plants it was ascertained that not all steel plate received goes into
elevated storage tanks. The figure "70 percent" was a value judgment.
5. See Note No. 2.
8. Design data from both Columbus and Cincinnati water works and from a study of
-------
C-5
Footnotes for Table C-l Continued)
2. Petroleum Product Storage Tanks
Since it is common experience to find some of the worst pollution around
refineries, all the tonnage for storage tanks was considered to be exposed
to polluted atmospheres .
1. Bulk storage for 1967 taken from reference ( 51 ). The storage capacity
for each major product was totalled to give the 1967 figure. This was
converted to the 1970 figure by assuming the growth in tankage would be
the same as the rate of petroleum refinery growth for which information
was available.
3. It was established by discussion with refinery engineers that 100,000
barrel storage tanks are typical for bulk storage, although some bulk
tanks are several times this size. The distribution tank sizes vary
widely, but the 10,000 barrel size was picked as being in the range.
Although it is known that more than half of the storage capacity is in
the large bulk tanks at the refinery, the 2/3 figure is purely an estimate.
This judgement will not markedly affect the overall result.
A. This is a standard handbook formula which can readily be derived.
7. Derived, as outlined in Appendix A, from the SIC statistics.
8. Heavy bulk storage tanks vary widely in the thickness of plate used with
the heaviest at the bottom. Plate 0.250 thick has an area factor of
196 (one side) and 0.75 inch thick, a factor of 65. The value of 120
was chosen as intermediate.
3. Chemical and Industrial Tanks
1. Derived from SIC statistics.
2. Chemical and industrial tanks vary in thickness, but in general, are much
smaller than those in the petroleum industry.
The area factor :"200 sq.ft./ton" is close to that for 0.250 inch
plate, one side only.
4. This value judgment is based on the observation that most of the industrial
-------
C-6
TABLE C-2. EXTRA MAINTENANCE PAINTING EXPENSE FOR STRUCTURAL
STEEL BRIDGES IN POLLUTED AIR
1. Estimate of bridge distribution.
For State of Ohio 650 out of 1910 bridges are in cities = 34
percent*- '
U. S. Bureau of Public Roads = 25 percent
(c)
Total road mileage municipal 532,000
rura1 3,152,000
Total $3,684,000
No highway bridges = 563,000 or one every 6.5 miles
In urban areas it is estimated there is one bridge per 2.5 miles
Percent = 532^Q°° ^- 563,000 = 38 percent
Based on these estimates, about 30 percent of bridges are in polluted air.
2. Total tonnage 18,134,000 tons in use
Area factor for bridge steel 125 sq.ft./ton
3. Annual loss - see Table 4 for maintenance factor -
.30 x 18,134,000 x 125 x 0.0447 = $30,400,000
(a) Discussion with Fred Ray, State of Ohio, Highway Department.
(b) Estimate made in 1955 by Erickson and Morgan, Reference 52.
(c) Road mileage for 1968.
(d) News release, Federal Highway Administration, October 13, 1970. (FHWA-516),
1. The notes at the bottom of the table explain the sources of information.
If one studies road maps, one finds it is readily apparent that there
are many more bridges in and around cities than in the open country.
-------
C-7
TABLE C-3. CALCULATION OF AIR POLLUTION CORROSION COSTS FOR EXTERNALLY
MOUNTED POWER TRANSFORMERS
1. Total area of transformers, ranging from pole size to substation sizes
556,630,000 sq.ft.
2. It is estimated that 80 percent are in polluted atmospheres.
3. For average pollution use extra maintenance cost of $0.0167 sq.ft./year,
see Table 4.
0.80 x 556.6 x 106 x 0.0167 =$7,450,000 per year.
1. SIC data gives numbers of each class. Designs of typical units allows one
to convert to area exposed to external atmosphere.
2. Most transformers are .(a) at power stations where air is often polluted,
(b) substations in metropolitan districts, or on poles in urban centers
-------
TABLE C-4. CALCULATION OF LOSSES, BASED ON EXTRA MAINTENANCE COSTS IN
POLLUTED ATMOSPHERES FOR STREET LIGHTING FIXTURES
1. Total tons in use - 984,574.
2. About 85 percent of external lighting in polluted areas.
3. About 15 percent of tonnage not exposed.
4. Area factor for remaining 85 percent:
A. 20 percent heavy gage steel - 400 sq.ft./ton
B. 65 percent light gage steel - 1300 sq.ftt/ton
5. Maintenance factor, see Table 4, is 0.0167 sq.ft.
6. A. 0.20 x 0.85 x 400 x 985,000 x 0.0167 = 1,110,000
B. 0.65 x 0.85 x 1300 x 985,000 x 0.0167 = 10,800,000
$11,910,000, total annual loss.
1. Derived from SIC statistical data.
2. Value judgment. Most street lighting is where there is industry,
commercial activity, or masses of people on the move.
3. Examination of various street lighting designs, indicates that a portion
of the steel pole may be buried and that some of the construction in
the lamp is not exposed. The "15 percent" figure is an educated guess.
4. Based on general observation, many lamps are mounted on wooden or other
-------
TABLE C-5. CALCULATION OF THE ANNUAL POLLUTION COSTS INVOLVED IN PAINTED OUTDOOR STEEL
(METAL WORK) AND GALVANIZED STEEL
Galvanized Roofing, Siding, Roof Drainage
Maintenance factor (see Table 4) $0.0392/sq.ft. for industrial and $0.0213 for commercial districts.
1. Total tonnage of roofing, siding and drainage = 17,212,518 tons.
2. About 20 percent is used in rural areas leaving 13,750,000 tons.
3. Tonnage in industrial service = 0.40 x 13,750,000 = 5,500,000
Tonnage in commercial service = 0. 15 x 13,750,000 = 2,060,000
4. Industrial siding tonnage = 495,823
5. Prefabricated and portable steel buildings
Total tons 12,124,871
65 percent in roofing and siding 7,890,000
55 percent of above industrial pollution 4,340,000
35 percent of above commercial pollution 2,760,000
6. Tonnage totals
A. Industrial 5,500,000 B. Commercial 2,060,000
495,000 2,760,000
4,340,000 4,820,000
10,335,000
7. Area factor = 1400 sq.ft./ton (one side) for 22 gauge used in industrial and 2500 sq.ft./ton for 28
gauge used in commercial applications.
8. A. 10,335,000 x 1400 x 0.0392 = $567,000,000
B. 4,820,000 x 2500 x '.0213 = 257,000,000
-------
TABLE C-5 Continued)
Outdoor Gratings, Fire Escapes, and Grill Work (Industrial and Commercial)
9. Total outdoor tons = 3,107,000.
10. Estimated distribution: 10% railings 250 sq.ft./ton
50% 1/4-inch steel work 400 sq.ft./ton
40% 1/2-inch steel work 200 sq.ft./ton
.. A - 10 x 250 + 50 x 400 + 40 x 200 ,_, ,_ ,
11. Area factor = — = 305 sq.ft./ton.
12. Maintenance factor 0.0167 from Table 4.
13. Loss/year 3,107,000 x 305 x 0.0167 =$15,850,000.
14. Total area in industrial service = 64,091,720 sq.ft. o
i-1
15. 64,091,720 x 0.0333 = $2,130,000 per year. °
16. Total in commercial service = 41,485,382 sq.ft.
17. Use maintenance factor of 0.0167 sq.ft./yr.
18. 41,485,382 x 0.0167 = $692,000 per year.
Metal Window Sash and Frame
19. Total tons = 432,044.
20. .Area factor at 100 sq.ft./ton (one side only, part of steel covered by glass and putty).
21. Maintenance factor high, estimated at $0.06 sq.ft./year.
22. 75 percent of sash in average polluted air.
-------
TABLE C-5 (continued)
Structural Steel—External Use
Industrial Use: 62,255,900 tons 15%, area factor 125 sq.ft./ton
Commercial Use: 41,815,433 tons 15%, area factor 125 sq.ft./ton
62.3 x 106 x 0.15 x 125 x 0.0333 =$39,000,000 per year.
41.8 x 106 x 0.15 x 125 x 0.0167 = $13,100,000 per year.
Public Utilities: 6,390,787 tons 65% external 125 sq.ft./ton.
0.65 x 6.39 x 106 x 125 x 0.0333 = $17,300,000 per year.
extra maintenance $0.0333 sq.ft.
extra maintenance $0.0167 sq.ft.
Summa ry
Galvanized roofing, siding, roof drainage, prefabricated and portable buildings
Outdoor gratings, fire escapes, and grill work
Metal doors, frames A
Metal doors, frames B
Metal window, sash, and frame
Structural Steel - External Use
Industrial
Commercial
Public Utilities
Total
$824,000,000
15,850,000
2,130,000
692,000
1,943,000
39,000,000
13,100,000
17,300,000
$914,015,000
1. Tonnage was derived from SIC statistical data.
2. For galvanized roofing and siding a considerable portion is used in rural shelters of all kinds. The
20 percent figure is a value judgment.
3. These breakdowns also are value judgments. Much more of the tonnage is known to be in industrial use.
Galvanized roofing and siding also is used, to some extent, in less affluent residential areas.
4. SIC sources.
5. Part of the metal used in portable steel buildings is not exposed to the atmosphere. It was judged
that about 65 percent is. More of these buildings are in industrial service than in commercial
service. The distribution is a value judgment based on discussions with fellow staff specialists
in housing.
7. Area factors of 1400 and 2500 correspond to 22 and 28 gauge, respectively.
-------
TABLE C-5 (continued)
10. Educated guess based on a study of fire escapes and outdoor grill work.
14,16. Calculated from SIC data.
15,17. Maintenance factors from Table 4.
19. Calculated from SIC data.
20. Educated guess based on observation of steel sash and frame construction in industrial buildings.
21. This estimate is considered conservative, in view of the extremely high labor costs in preparing
the surface and painting steel sash.
22. Since most of the steel sash is used in industrial and commercial applications, and not in residential
this estimate is a value judgment.
o
-------
C-13
TABLE C-6. CALCULATION OF AIR POLLUTION COSTS FOR GALVANIZED POLE-LINE
HARDWARE
1. Total tons of pole-line hardware products, transmission hardware and
related products. 4,062,735 tons.
2. About 75 percent of total is estimated to be pole-line hardware.
0.75 x 4,062,735 = 3,040,000 tons.
3. Area factor for pole-line hardware = 450 sq.ft./ton.
For typical pole with 4 sq.ft. of hardware, number of poles in use is
3,040,000-x 450 = 342 x 106
4
4. From zinc statistical data, if it estimated that there is 165,000 tons
of zinc consumed in galvanizing pole-line hardware still in use. At a
bath efficiency of 50 percent, this is 82,500 tons of zinc in use. At
a coating weight of 2 oz./sq.ft. this equals 8 oz./pole
87,500 x 2,000 x 16 = 330 x 106 poles
8
5. Average no. of typical poles = 336 x 10 .
6. Some 40 percent of poles are exposed to urban conditions (see footnote).
7. In average polluted air, hardware lasts typically 15 years whereas the
wooden pole last about 22.5 years. Thus for service in polluted air,
pole-line hardware is replaced once.
8. Cost of regalvanizing, per pole at $0.55/sq.ft.
4 x 0.55 = $2.10
Cost on weight basis, at $0.11/lb.
2,000 x 4 x 0.11 = $L96
450
Cost of regalvanizing is about $2.00 per pole.
Labor cost of installation and removal $25.00 per pole.
27.00 4- 22.5 = $1.20/year/pole
9. Annual loss based on replacement and regalvanizing is .40 x 335 x 1.20 x 10
$161,000,000
1. Calculated from SIC statistics.
2. No breakdown in the SIC is given for related products: (see item No. 1).
Some of this hardware may be items exposed on the pole, although only
partially exposed, such as bolts and nuts. Brackets, on the other hand,
are almost entirely exposed. The figure "75 percent" is a value judgment.
3. Pole-line hardware with a thickness of 3/15-inch would have an area factor
of 524 and for 1/4 inch it would be 392.sq. ft. The factor, 450, was chosen
-------
C-14
6. This figure is an educated guess. Personal observation suggests that there
may be five times as many poles per rural customer as are required for
urban customers. On this basis, if 20' percent of the population accounts
for 100 parts of the total poles, and 80 percent, 80 parts, the percent
poles in urban areas would be 80 -f 180 or 44. The figure 40 percent
was used in the calculation.
7. According to a fellow staff member who has specialized in wood preservation,
the life of impregnated wooden poles is from 20 to 25 years.
8 The cost of regaIvanizing is given in Reference 32. Note that the cost on
a weight basis, obtained from Brown Steel Galvanizing Company of Columbus,
Ohio, is almost the same.
Labor costs include set-up time, removal of old hardware, and installation
of new. Two men are ffiquired and the labor estimate of $25 includes
-------
015
TABLE C-7 . CALCULATION FOR CHAIN-LINK FENCE
1. Total tonnage of galvanized chain-link fence in service-4,018,980 tons.
2. Sizes vary from 3 to 12 feet in height. Typical height is 6 feet.
Typical wire gauge is No. 9.
3. 80 percent of chain-link fencing is used in industrial and commercial
areas.
4. The weight of steel in a 100-foot length of 6-foot high fence wire
diameter = 0.148^ weight = 0.0578 Ibs/ft. One square foot of fence with
2-inch squares is equal to 12 feet of wire.
600 x 12 x 0.0578 = 415 pounds of fencing/100 feet
5. Total number of 100-foot lengths of 6-foot high fence
4,018,980 x 2,000 = 19.35 x 106
6. Taken at 80 percent in polluted air = 15.5 x 10 100 ft. lengths.
7. System life of posts is 30 years, but fencing is 20 years. Rural zinc
coating life is 20 years. Polluted life varies, 8 years is typical.
8. , If system is painted, roller coating costs would be
First coat 275 sq.ft.,, labor 4.72, paint 3.96 = 8.68 or .0316/sq. ft.
Second coat 375 sq.ft., labor 4.72, paint 4.66 = 9.38 or .022l/sq.ft.
Cost per sq.ft. = .0537 Rural life of galvanized wire - 20 years.
Life in polluted air = 8 years.
Rural exposure - no maintenance for 20 years.
Polluted exposure two coats at 8, 14 years.
Extra cost/sq.ft. for 20 years = 2x .0537
Extra annual cost/sq.ft./yr = .0054
9. Posts and Fittings - per 100-foot length, 6-foot high fence
Amount needed size Area Calculation Sq.ft.
1 3.5 in. diameter x 3.5 x 6 5.5
corner post 12
8 2.5 in. diameter 8 x x2.5 x 6 31.3
fence post 12
10 1.6 in. diameter 10 x 10 x x 1.625 42.5
rail post 12
Add 10% for special fittings 8.0
-------
C-16
TABLE C-7 (continued)
10. Annual extra cost per 600 sq.ft. fence wire and 87 sq.ft. posts.
Wire 600 x .0054 = $3.24 (see Item No. 8)
Posts 87 x .0233 = 2.03 (see Table 4)
11. About 25 percent of owners maintain fencing by painting.
6
.25 x 15.5 x 10 x 5.27 = $20,400,000.
12. Replacement cost of entire fence. No maintenance, last ten years rusty.
A. Rural life 20 years plus 10 years rusty = 30 years
B. Polluted life 8 years plus 10 years rusty = 18 years
13. Cost of fence and installation per 100 feet is about $440.
A. Annual cost 440 T 30 = $14.65/100 ft./yr.
B. Annual Cost 440 -r 18 = $24.45/100 ft./yr.
14. Pollution cost per year is equal to difference $9.80. If 25 percent of
owners replace,the annual extra cost is .25 x 15.5 x 10^ x 9.80 =
$38,000,000.
15. Replacement of galvanized wire and painting posts.
Rural Cost = no wire replacement, last 10 years rusty
Pollution cost = two wire replacement (each wire rusty for two years
before replaced).
Extra cost for 30 years = two wire replacement + two paintings of posts.
Cost of 100 feet wire = $ 54.00
Labor and delivery = 72.00
Total $126.00
Total cost of two extra replacements $252.
Cost of two paintings for 87 sq.ft. of posts.
2 x 87 x (.50 x .40) = $158
16. Total extra cost, 30 years = 252 + 158 = $410
Total annual extra cost per 100 feet of 6 foot fence = 410 -r 30 = $13.65.
If 50 percent replace galvanized wire, then annual pollution cost is:
.50 x 15.5 x 106 x 13.65 = $106,000,000.
17. Summary of annual galvanized fence costs.
Percent Annual Extra Cost Basis Item No. Annual Pollution Loss
25 maintenance by painting 11 $ 21,800,000
25 let fence rust, then re- 14 38,000,000
place
50 Replace wireC1),paint posts 16 106,000,000
-------
C-17
Although wire fencing can be regaIvanized, current handling costs, set-up
time, etc., usually does not make it pay. Present practice is to buy new
wire.
Table C-7
1. Data derived from SIC statistics.
(54)
2. See Reference NA-1. Other details on chain-link fencing were obtained
from catalogs, from dealers in Columbus and Cincinnati, and from the
trade magazine "Fence Industry".
3,6. According to Columbus and Cincinnati dealers, only a small portion of
the total tonnage goes into residential use and almost none into rural
service. This is a value judgment, based on the above discussions.
4. These calculations are for 9 gauge wire, the most common size used by
industrial plant and institutions for general security purposes.
7. System life is provided by the IRS, see Reference 7.
8. These costs were derived using older data in Reference 32 as a guide.
If overhead were included the costs would be higher.
9. These are the sizes of posts and railings recommended for 9 gauge chain-
link fencing in catalogs.
11,12,13,14,15,16. For purposes of computation, it was assumed that 25 percent
of owners prolonged life by maintenance painting, 25 percent let fence
rust then replaced entire fence and 50 percent replaced thewire only and
maintained the posts and rails. A study of fencing practice indicates
all three methods are followed, but the proportions chosen are an
educated guess.
15. Cost of 100 feet of wire provided by a local fence manufacturers.
-------
C-18
TABLE C-8. CALCULATION OF AIR POLLUTION CORROSION COSTS FOR
GALVANIZED WIRE ROPE
FjLrst estimate based on steel statistics
1. Total tons of wire rope in service - 6,301,000.
2. As shipped value - $,015,000,000.
3. Cost per ton - $653 (about $0.32/lb.).
4. Rural life = 20 years 635/20 = $31.7 ton/yr.
Industrial life = 8 years 635/8 = $79.4 ton/yr.
$79.4-$31.7 = $47.7 extra cost per year per ton charged to pollution.:
5. About 40 percent of wire exposed to pollution.
6,301,000 x 0.40 x 47.7 = $120,000,000 per year.
Second estimate based on zinc statistics
6. Total zinc consumed in wire galvanizing baths - 1955 to 1969 = 572,213
tons.
7. At bath efficiency of 50% leaves 286,107 tons.
8. Typical wire size - 0.120 inch diamter.
Area per 1,000 feet of length = x 0.120 x 1000 = 31.4 sq.ft.
12
2
9. Zinc wire coating, Class A, double galvanized = 0.85 oz/ft .
31.4 x 0.85 = 26.6 oz/1000 feet.
10. A 3/8th stranded wire rope has 7 strands.
7 x 26.6 oz. = 186 oz./lOOO ft. wire rope.
11. Total number of 1,000 foot lengths in service:
286,107 x 2,000 x 16 = 49.3 x 10 1,000 foot lengths of wire cable.
186
12. 1968 price of 3/8th 7-wire strand = $62/1,000.
1970 price (estimated) $70/1,000.
13. Rural life at 20 years 70/20 = $3.50/yr.
Industrial life at 8 years 70/8 = $8.75/yr.
14. $8.75 - $3.50 = $5.25/yr/l,000 feet extra cost because 6f pollution.
15. 49.3 x 106 x 0.40 x 5.75 = $103,500,000 per year.
Summary: Both values are of same order. Taking an average of the two
-------
C-19
Table C-8
1,2,3 Derived from SIC data
5. Much of the galvanized wire rope is used in rural andmarine applications,
where man-made pollution is not the major factor. The selection of
"40 percent" is an educated guess.
(49)
-------
C-20
TABLE C-9. CALCULATION OF THE EXTRA MAINTENANCE
Expense for Galvanized Steel Power Line Transmission Towers Exposed to
Air Pollution.
1. System life 30 years.
2. Total tons in use 1,449,000.
3. Typical weight 7.5 tons.
No. of typical towers = 1,449,000 -r 7.5 = 194,000
4. According to a survey 12.5 percent are in contaminated air.
0.125 x 194,000 = 24,300 typical towers subject to pollution.
5. Area of typical tower = 2500 square feet. Maintenance factor for
galvanized towers (Table 4) is $0.0233/sq.ft./year.
Annual loss is estimated to be:
24,300 x 2500 x 0.0233 = $1,480,000.
1. Reference 7.
2. Determined from SIC statistics.
3. See Reference 53.
4. See Reference 54.
-------
TABLE C-10. SUMMARY OF PRINCIPAL DATA USED IN TABLES C-l to C-9
CO
jj**
d
CO
H
i j_,
: Cl)
J->
CO
^
Metal Economics
Tons Shipped (I/ 14^:
InventoryC2) , 7,100 1
Dep. Value^5* 10,659
Depreciation % 2
Life, years 50
Capacity 11x40 gal
Area Factor 125
Pollution-7, or 80
exposure''/ •
\
Cost,$/sq.ft./yr 0.0167
Calculations C-l
in Table
Total Annual loss 3.4
$ x 106* .
'
,— t
0 CO
D co T-I en
CL) ^ M Jsi
1-1 d w d
O CO CO CO
M H 3 H
•u : 'b
CL) Cl
P-i H
201.4 511.4
,950 4,830
695 1,935
9 9
11 11
477xl06
bar.
120 200
100 : 85
0.0473 0.0391
C-l C-l
10.81 32.1
j
j
i
•H >-.
CO CO
Pi ?
jZ
-. 60
CO -i-l
OJ M
M
•o -a
•H d
M CO
M
717.6
18, 134
5,043
3.3
30
657,000(4)
125
30
0.0447
C-2
30.4
i
i
co
d
CO
M co
H M
0)
M 6
a M
£ 0
O M-l
P-l
12,316
12,316
3.3
30
556.6®
L
80
0.0167
C-3
7.45
CO
4-1 4-1
CL) 43
60
>-l «r-l
•U fJ
c/a
54.3
984.6
1,221
5.0
20
tOO/BOO
85
0.0167
C-4
11.91
13
Jsi CD J4
MM M
O -i-l O T3
|3 d |3 CL)
CO 4J
i-( > i-l C
co i— < CO -H
4J CO 4J CO
CD O 0) P-i
s s
1
10, 335 '
2.2: 2.2
45 ! 45
;64,091/
(8)
•41,485^
1400/ rvarious
2500 j
80 ; 40/15
,
0.0392/ 0.0333/
0.0213; 0.0167
C-5 i C-5
i
824. ! 90.02
[
}
1
i
i
cu cu
d M
•r-l CO
,-4 5
1 T3
CD S-i
i— 1 CO
o a
P4
157.6
4,063
1,262
3.3
22.5(30)
30
C-6
161.
d
0
JsJ ! -H
d , cu co
•H • a, co
>J? . O -i-l CO
i cu pi EM
d O : co cu
•H d i cu • d 5
CO CD }-i co O
U ' S H H
4,019 6,301 209.7
577,6 2,008 1,165
3.33
20/30 ; 20 : 30
} 49.3°^ i 194,00^
2,900 ; r
,
80 i 40 ; 12.5
; 0.0233
C-7 C-8 C-9
165. 8 : 111.8 1.48
-------
22
Footnotes for Table C-10
1. Tons, in thousands, as derived from SIC statistics, shipped in 1963.
2. Tonnage of tankage in situ that remains in service, i.e., not yet
depreciated using life of system as determined by IRS.
3. Area in millions of square feet.
4. Total number of highway and railroad bridges in service, see Table C-2.
5. Depreciated value, in thousands of dollars, of tankage still in service.
6. Area factor is given in square feet per ton. In some cases, e.g., steel
plate for tanks, it is for one side only.
7. This figure is the estimated percent of total area exposed to pollution.
8. Area in thosands of square feet for metal doors and frames in (a)
industrial and (b) commercial service. Other items are given in
Table C-5.
9. Life as determined by IRS tables, see Reference (7).
10. Millions of 1,000-foot lengths of equivalent 3/8ths inch 7 strand cable
as calculated in Table C-8.
-------
TABLE C-ll. COMPARISON OF METAL ROOFING FOR SERVICE IN HIGHLY POLLUTED ATMOSPHERE.
(a)
Material
Galvanized
Steel
Aluminum
Copper
Terne Plate
Years Life
Polluted •
10
50
100+
50 j
j
Gauge (c)
22
18
28
26
Thickness
33.6
40.3
12.6
15.9 \
•Ibs/
cu. in.
0.292
0.098
0.322
0.293
: Mill ,
Ibs/ ; Price $/:
sq. ft. pounds
1.406
.10
.57 i .48
.58
.67
••
.92 ;
• 12 \
$ Cost/
sq, ft.
.14
.37
.53
.08
Extras/
sq. ft.
Inst. Labor
__(d)
__(d)
.20(d)
.35
.30
.40
.45(d) .40
i
Installed : $ Cost/
cost, $/ yr. /
sq. f t. ' sq. ft.
.49
.67
1.11
.93
.049
.013
.011
.019
Matgrial
Galvanized Steel
Aluminum
Copper
Terne Plate
$ Cost/yr/sq.ft.
Above Rura1 GaIvanized
0.035
-0.001
-0.003
0.005
(e)
Annual Charge to Pollution
$ per Square Foot
0.035
none (saving)
none (saving)
0.005
Percent Saving Using Cost of
Galvanized Steel in Polluted
Air as Reference
73.5
77.5
61.3
o
(a) Annual costs are based on original installation expense, life of system, and no added maintenance.
(b) Life in years as originally installed.
(c) Brown and Sharp gauge, except for galvanized steel where special sheet gauge is used.
(d) Both galvanized steel and aluminum are considered self supporting, whereas extra expense is allowed for
copper and terne plate to provide backing for support. Terne plate is painted at the time of installation.
(e) Life of galvanizing in rural environment is 35 yenrs. Rural cost is installed cost divided by year =
-------
TABLE C-12. COMPARISON OF ALUMINUM WITH GALVANIZED STEEL SIDING, 1969
1. Aluminum siding varies in thickness from 0.024 to 0.040 inch.
Typical thickness is equal to 0.0320 inch (20 gauge) and weights 55 lbs/100 sq.ft.
2. Total production of aluminum siding for 1969 is 53,000,000 Ibs.
3. The equivalent area for this weight is 53 x 10 -r 0.55 = 96 x 10 sq.ft.
4. Using roofing data in Table 9 as a guide, the installed cost of 0.320 inch thick aluminum would
be $0.59/sq.ft. and for 0.276 inch thick galvanized $0.43/sq.ft.
5. The installed cost for 96 x 10 sq.ft. would be:
Aluminum $56,700,000 Life = 50 years, industrial or rural
Galvanized $41,300,000 Life = 10 years, industrial, 35 years rural
Premium=Difference$15,400,000
6. The annual costs would be:
Aluminum, Industrial $1,133,000
Galvanized, Industrial $4,130,000
Galvanized, Rural $1,180,000
7. Annual charge to pollution using rural galvanized as reference.
Aluminum - none (small saving)
-------
C-25
TABLE C-13. COMPARISON OF THE COSTS OF USING BARE SELF WEATHERING
STEEL AND PAINTED CARBON STEEL FOR BRIDGES EXPOSED TO
POLLUTION
1. Annual maintenance cost for painted carbon steel, 30 year life (as in
bridges), area factor 125, and paint costs from Table 4.
Carbon Steel. Rural Atm., Cost/Ton/Yr = 125 x 1.30 = $5.40
30
Polluted Atm., Cost/Ton/Yr = 125 x 2.64 = $11.00
30
2. Comparison of carbon steel and self weathering steel on an installed
basis.(43) Costs are per ton.
wii T, • Carbon Steel
Mill Price ^TJI Self Weathering Steel
Erection 184 184
Total $371 264
448 (less allowance for
- 45 higher strength)
Total $403 (per ton of carbon steel
replaced).
3. Annual premium for self weathering steel per ton of carbon steel replaced.
403 - 371 = $1.07/ton/year
30
4. Comparison between painted carbon steel in rura1 environment and
unpainted self weathering steel in polluted environment.
(a) self weathering steel - annual premium/ton = $1.07.
(b) painted carbon steel - annual maintenance/ton = $5.40.
5. Annual saving per ton of carbon steel replaced in using bare weathering
steel and costs in Item 4.
5.40 - 1.07 = $4.33/ton/yr.
6. Comparison between painted carbon steel in polluted environment with
bare self weathering steel in polluted environment.
Carbon Steel (item No. 1) Cost/ton/yr = $11.00
Self weathering steel (item No. 3) cost/ton/yr = $1.07
7. Annual saving per ton of carbon steel replaced using costs in Item 6.
------- |