I’(|A

 

 

   

....‘r«~~'

 

__',..‘.....~n

  
   

' -'~ .f-_VR

 

 

mvu» 9“. v.

 

;‘
hum. 4 ~

 

IHESIS

 

 

men‘f
Michigan State
University |

 

 

7——

This is to certify that the

thesis entitled

CoRRoSmN IN AIRCKHFT - THE EFfEeTs
0F Ewntwmavr nun MHINTEIVHNCE oPERnnow

presented by

KENNETH B. LENZ

has been accepted towards fulfillment
of the requirements for

M. 5- degree in MHTERmI-S SCIENCE.

 

Major professor
Robert Sunmitt
Date

0-7639 MSU is an Affirmative Action/Equal Opportunity Institution

 

 

 

MSU

LIBRARIES
m

 

 

BEIURNING MAIQELfib§:
Place in book drop to
remove this checkout from
your record. fjfl§§ will
be charged if book is
returned after the date

stamped below.

 

 

 

 

 

 

 

 

Tim-

 

CORROSION IN AIRCRAFT - THE EFFECTS OF
ENVIRONMENT AND MAINTENANCE OPERATIONS

by

Kenneth B. Lenz

A THESIS

Submitted to

Michigan State University

in partial fulfillment of the

requirements for the degree of

MASTER OF SCIENCE

Department of Metallurgy, Mechanics, and

Materials Science

1985

ABSTRACT
CORROSION IN AIRCRAFT - THE EFFECTS OF
ENVIRONMENT AND MAINTENANCE OPERATIONS

by

Kenneth B. Lenz

Ambient environmental factors are taken into consideration in the
development of three corrosion severity algorithms that can be useful
for the planning and implementation of corrosion preventative mainten-
ance schedules for United States Air Force aircraft. These algorithms
include Hashing Intervals, Complete Strip - Repaint Intervals and Ex-
pected Corrosion Damage Evaluations. Washing Intervals and Expected
Corrosion Damage results are determined for non-CONUS (overseas) USAF
airbases.

A thorough investigation of the automatic aircraft rinse facility
located at MacDill Air Force Base, Tampa, Florida, is documented.
Through this study, specific recommendations are suggested in order to
improve its effectiveness against aircraft corrosion.

Four major categories of corrosion inhibitors are identified based
on their mechanisms of operation. Such a classification system can be

useful in the selection of corrosion inhibitors for specific applications

and also in future inhibitor research.

DEDICATION

The author would like to dedicate this work to his loving parents,
Reinhart J. and Elizabeth J. Lenz for their unwavering support, en-
couragement, and sacrifice throughout his life from which he has bene-

fitted greatly.

ii

ACKNOWLEDGEMENT

The author would like to express his deep graditude to his academic
advisor, Dr. Robert Summitt, whose guidance, wisdom, and assistance were
invaluable throughout the author‘s graduate program and especially during
the coarse of this investigation.

Thanks are also extended to the U.S. Air Force, Wright Aeronautical
Laboratories, the Department of Metallurgy, Mechanics, and Materials
Science, and the Division of Engineering Research at Michigan State Univ-

ersity for the financial assistance during the Masters Program.

TABLE OF CONTENTS

L392

LIST OF TABLES.................................................... v
LIST OF FIGURES................................................... vi
I. INTRODUCTION........................................... 1

II. THE NON-CONUS PACER LIME PROGRAM....................... 2

III. MACDILL AIR FORCE BASE RINSE FACILITY.................. 42

IV. INHIBITORS............................................. 66
REFERENCES........................................................ 92

iv

LIST OF TABLES

Table Page

1. Ranges of Environmental Ambient Parameters, Continental U.S.... 14
2. Working Environmental Corrosion Standards (NECS)............... 30

3. Non-CONUS Airbase Data and Environmental Corrosivity Ratings
for Aircraft Washing and Corrosion Damage...................... 39

4. Daily Tampa Water Analysis Results - July 1984................. 57
5. Daily Tampa Water Analysis Results - January 1985.............. 58
6. Corrosion Inhibitors for Steel in Strong Acids................. 70
7. Adsorbed Layer Forming Inhibitors for Nonacid Systems.......... 71
8. Electrochemical Potential of Some Oxidizing Inhibitors......... 78
9. Conversion Layer Formers....................................... 79
10. Neutralizing Inhibitors........................................ 82

11. Classification System of Inhibitors............................ 91

LIST OF FIGURES

£12222 Page

I. One Year Weight Loss for Rimmed Carbon Steel vs.
Atmspheric SUIfur DiOXideOOOOOOOOOOIOOOOOOOOOOOOOOOOO..0.0.... 16

2. One Year Weight Loss for Aluminum vs. Atmospheric Sulfur

DiOXideCCOOCOOOOOOOOOOO0.0...00.00.00.000...OOOOOOOOOOOOCOOOOOO17

3. Relative Humidity vs. Corrosion Rate of Steel.................. 18

4. Dew Formation Conditions - Relative Humidity vs.
Temperature Difference for Condensation........................ 19

S. Chloride Concentration in Rainwater vs. Distance From The Sea.. 23
6. Salt Content in Air vs. Distance From Sea...................... 25
7. Corrosion Rate of Iron vs. Distance From The Sea............... 26
8. Marine Corrosivity Index vs. Distance From The Sea............. 28
9. Aircraft Hashing Interval Algorithm............................ 33
10. Aircraft Complete Repaint Interval Algorithm................... 34
11. Aircraft Corrosion Damage Algorithm............................ 35
12. Vicinity Map - MacDill Air Force Base.......................... 43
13. Project Location - MacDill Air Force Base...................... 44
14. MacDill Rinse Facility - Schematic................... ......... 45
15. MacDill Rinse Facility - Schematic............................. 46

MacDill Rinse Facility Schematic........ 47

16. Water Storage Tank

MacDill Rinse Facility Schematic........ 48

17. Water Storage Tank

Schematic........ 49

18. Water Storage Tank MacDill Rinse Facility

MacDill Rinse Facility Schematic........ 50

19. Water Storage Tank

20. Ten Month Chloride Values for Tampa City Water................. 59
21. Ten Month Specific Conductivity Values for Tampa City Water.... 60
22. Ten Month Total Hardness Values for Tampa City Water........... 61

23. Ten Month pH Values for Tampa City Water....................... 62

vi

Figure Page

24. Schematic Polarization Diagram Illustrating Cathodic
Inhibitor in an ACid salutionOOOOOOOOOOO0.0.0.0..0..OOOOOIOOOOO 72

25. Schematic Polarization Diagram Illustrating Anodic Inhibitor
in an ACid SOIUtiOHOOOOOOOIOOOOOOOOO...OOOOOIOOOIOOOOO000...... 73

26. Schematic Polarization Diagram Illustrating a Mixed Inhibitor
in an ACid so1ution000000000000000000.00000000000000000.0.00... 74

27. Schematic Polarization Diagram for Nitrite Inhibition of
SteEI - a PaSSivating InhibitorOOOOOOOOOOOOO0.0.0.000...0...... 76

vii

INTRODUCTION

A recent study conducted by the National Bureau of Standardsg8 in-
dicated that the cost of corrosion in the United States is approximately
$70 billion annually, with the United States Air Force bearing more than
$1 billion of the burden. Furthermore, Air Force costs for both operation
and maintenance have increased, while procurement costs remain fairly con-
stant. Hence a major effort to reduce the costs of maintenance operations
began in the early 1970's and continues today.

This thesis, which deals with corrosion maintenance of USAF aircraft,
is divided into three main parts:

Part 1 - The Non - CONUS Pacer Lime Program, quantifies the corrosion
risk factors in ambient atmospheres and then recommends corresponding main-
tenance schedules for U.S. Air Force aircraft stationed at overseas (non-
CDNUS) airbases.

Part 2 - The MacDill Air Base Rinse Facility, is a thorough investiga-
tion of the automatic aircraft rinse facility located at MacDill Air Force
Base, Tampa, Florida. Through this investigation, specific recommendations
are given in order to improve its effectiveness against aircraft corrosion.

Part 3 - Inhibitors, identifies four major categories of corrosion in-
hibitors based on their mechanisms of operation. Such a classification can
be useful in corrosion inhibitor research currently being conducted at

Michigan State University for USAF aircraft applications.

THE NON-CONUS PACER LIME PROGRAM

I. Introduction

The Pacer Lime Program is an evaluation of the corrosivity of ambient
atmospheres at domestic (CONUS) United States Air Force bases. This study
was completed in 1979 by Robert Summitt and Fred T. Fink of Michigan State
University. The goals of this study were two-fold: to quantify the cor-
rosion risk factors in ambient atmospheres and then to recommend corres-
ponding maintenance schedules for the aircraft to reduce corrosion damage.

The Air Force since has requested a similar evaluation for its over-
seas (non-CONUS) airbases in EurOpe, the Far East and Central America.

This report, which is modelled closely after the Pacer Lime Program, deals

with these bases.

II. The Environmental Severity Classification System
1. Environmental Variability

It has been shown through atmospheric testing programsz’ 3’ 4 that
atmospheric corrosion severity varies significantly from one site to
another. Usually, relative severity has been classified as "rural",
"industrial“, “urban", "marine", or a combination of these terms. It

2’ 4 that atmospheres containing moisture, salt and pol-

has been shown
lutants tend to accelerate corrosion. In order to devel0p a more accurate
environmental rating system, it might be more useful to incorporate all

of these factors when considering the corrosivity of an environment.

2. Atmospheric Corrosion in Aircraft

2 describes three types of atmospheric corrosion:

Tomashov

(1) “Net atmospheric corrosion“ caused by visible droplets of con-
densed moisture on surfaces. Such moisture may originate from dew, rain,
frost or snow.

(2) "Moist atmospheric corrosion", occurs at a relative humidity
less than 100% and takes place under a very thin, invisible layer of
electrolyte formed on the surface by capillary action, physical or
chemical adsorption.

(3) "Dry atmospheric corrosion", occurs when no moisture is present.

In aircraft, both wet and moist corrosion occur. Hater accumulates
on metal surfaces as condensation (fog, dew, from humid air on post-flight
surfaces), rainfall on exterior surfaces and through open hatches and
accidental spills. Dry atmospheric corrosion is considered unimportant
since aircraft alloys generally do not corrode without moisture.

The range of problems in aircraft can be characterized as followslz

(1) Net and moist corrosion of unprotected surfaces.

4

(2) Net and moist corrosion of protected metal surfaces subsequent
to failure of protective coatings. Protective coatings can fail because of
solar radiation, atmospheric contaminants (such as ozone and other oxidants,
particulates, fuels, and exhaust gases), high speed air ablation and mech-
anical abrasion and flexure.

(3) Corrosion caused by contaminants of human origin such as human
waste, spilled beverages, hydraulic fluids and battery acids.

(1) and (2) can be directly attributable to ambient atmospheric factors
which accelerate corrosion of metals or protective coatings. (3) is a pro-
blem which should be easily controlled, but is in fact a serious problem
at many USAF bases.

3. Factors Affecting the Rate of Corrosion

The factors that contribute to the rate of corrosion largely involve
weather conditions, especially those related to moisture; atm05pheric
pollutants, both natural and man-made; and the nature of the metal.

a. Weather

Weather parameters include temperature, precipitation, solar radiation,
wind direction, wind speed, relative humidity, dew point, cloud cover and
fogs. Of these, water has the largest effect on the corrosion rate of
metals. According to Vernon6’ 7, a given metal corrodes rapidly when the
relative humidity exceeds a critical value, but corrodes slowly when at a
lower humidity. Critical humidity can vary from one metal to another and
the presence of various pollutants can change the value. They can also
change the corrosion rate. In the presence of $02, the critical value of
aluminum is about 70%.

8

Grossman states that a film of moisture deposits from humid air

on metal surfaces of aircraft if the metal is colder than the air (follow-

5

ing high altitude flight), if hygroscopic salts (corrosion products or

pollutants) are present or through adsorption. The thickness of the
layer is determined by the humidity along with the mechanism of the
adsorption processzo

Rain is thought to promote corrosion by providing moisture and
washing away soluble corrosion products. It is believed to retard
corrosion by washing away pollutants. Therefore light rain is harmful
but heavy rain would be beneficiall.

The beneficial effects appear to be negligible in aircraft corrosion
because paint protects most exposed surfaces. Corrosion occurs under-
neath the cracks of the paint where the washing effects of the rain are
ineffective. Interior surfaces carelessly exposed to rain are wetted
but not washed. This water is harmful to these lesser protected areas.
It must be concluded that rain should be considered a harmful source of
moisture.

Air temperature, humidity, solar radiation, cloud cover, and wind
speed affect the rate of evaporation. Since temperature affects the
rate of corrosion reactions, the corrosion rate is expected to increase
as the temperature rises. It is important to point out, however,
that dissolved oxygen in water is necessary for most corrosion reactions
and the solubility of oxygen decreases with increasing temperature.

Rozenfeld9 discusses the interaction of temperature and moisture.
The time of wetness will vary with temperature. Consequently the cor-
rosion rate is greater in northern regions, where the temperatures are
low, than in warmer southern regions because moisture remains on the
metal surface longer. A combination of higher temperatures and pro-

longed moisture, however, will result in severe corrosion. A good

example of this marine pilings in the summer where they are continually
wetted.

b. Pollutants

Atmospheric pollutants are natural and man-made airborne pollutants
present at harmful concentrations. These substances are described as

followslot 11

, including only those known to contribute to corrosion:
(1) Particulates

Particulates are both solid and liquid material in particle sizes
ranging from 0.11 to 100 pm in diameter. Dust, grit, fly ash, and visible
smoke particulates settle to the ground quite quickly. Smaller par-
ticulates remain su5pended much longer and may travel very long dis-
tances. Larger particulates may cause corrosion problems close to their
source, whereas small particles can be a factor at great distances from
their source.

The chemical composition of particulates are classified by their
sourcelz: (1) Salts from sea spray and salt flats. (2) Dust from ag-
ricultural lands. (3) Soots from the incineration of agricultural wastes
and the burning of fuels. (4) Agricultural and industrial dusts. Ninety
percent1 of airborne particulates originate from natural sources. Very
few air pollution monitoring stations report the chemical composition
of particulates, only their concentrations. Although the corrosivity
of the particulates varies widely, there is no way of classifying them
since the data are not available. An exception to this is proximity to
salt sources.

The presence of salt greatly increases the corrosion rate in nearly

2, 4

all metals . Environments where airborne salt exists will be high

risk environments. When soluble salts such as sodium chloride and

ammonium sulfate are present, the corrosion products are usually water
soluble and easily removable. Corrosion products that form in the pre-
sence of water are weakly soluble and are not easily removable. These
products usually serve as a protective layer for the underlying metal.

There is a synergistic effect between salt deposits and the atmos-
pheric water content. The deliquescent salts will undergo a phase trans-
formation from dry crystal to a solution droplet when the ambient water
vapor pressure exceeds that of a saturated solution of the highest
bydratez. The relative humidities at which this transformation occurs
for ammonium sulfate, sodium sulfate, sodium chloride, and ammonium
nitrate are 80, 86, 75, and 62 per cent, respectivelyl. Therefore salt
deposits both attract moisture to the surfaces and provide an electrolyte
solution required for corrosion.
(2) Sulfur Compounds

Sulfur is another important atmospheric pollutant. It enters the
atmOSphere in several forms, including sulfur dioxide, hydrogen sulfide,
and sulfate salt particulatesl3. About two-thirds of all atmospheric
sulfur comes from natural sources, mainly as H25 from bacterial action
which later is converted to sulfur dioxide. Other sources of sulfur
dioxide are the result of combustion of sulfur containing fuels.

Sulfur dioxide initially is oxidized photochemically to sulfur
trioxide, which then combines with water to form sulfuric acid.

Hydrogen sulfide emmisions play a large part in the total sulfur
concentration in the atmosphere. It is produced naturally by decaying
matter and by volcanoes.~ It also is produced by industrial operations

and in catalytic converters. Hydrogen sulfide also is oxidized in the

air to form sulfur dioxide and subsequently sulfuric acid and sulfate

salts.

Another problem associated with sulfur pollution is acid rain.
As the concentration of sulfur compounds in the atmosphere increases, so
the acidity of the rainwater will also rise.

Vanderberg and Syndberger14

examined the reactions of adsorbed $02
on wet surfaces. They observed a quantitative relationship between the
number of 502 molecules adsorbed on an iron surface and the number of
protons generated, indicating that almost every adsorbed $02 molecule
is oxidized to sulfuric acid. This was attributed to the catalytic

action of the ferric oxide film. The overall reaction was reported to

be:

1 _ + I -
$02 + H20 4" 702 - H + H504

The catalytic action on the oxides of aluminum as measured by the
rate of proton generation was observed to be slight, and $02 is con-
sumed in the formation of sulfur containing corrosion products. 502
is thought to cause corrosion by being taken into solution where it is
oxidized to sulfuric acid. The anodic solution then attacks the passive
pxide layers. Spedding15 studied 502 adsorption onto aluminum surfaces
at varying humidities and reached a similar conclusion.

(3) Hydrocarbons

9 mostly come from natural decomposition of organic

Hydrocarbons
matter. AnthrOpogenic sources are important, however, because they may
be highly concentrated geographically where they are not rapidly dispersed.
The most notable example is in the Los Angeles basin, where the sources
are primarily gasoline automobile engines. The fate of the hydrocarbon

pollutants involves their reaction with oxides of nitrogen to form photo-

chemical smog, which includes a variety of secondary pollutants such as
ozone, nitrogen dioxide, and peroxyacetyl nitrates. Hydrocarbons are
not damaging either to metals or protective coatings, but photochemical
oxidants are harmful to both23.
(4) Oxides of Nitrogen

Three oxides of nitrogen, nitrous oxide (N20), nitric oxide (NO),
and nitrogen dioxide (N02), are found in the atmosphere in appreciable
quantitieslS. Nitrous oxide is present in concentrations of about 450
ug/m3 and results primarily from biological processes in soil. Because
of its low reactivity, the pollutant effects of N20 have largely been
ignored16.

AtmOSpheric upper limit values for NO and N02 are estimated as

86.1 ug/m3 and 135.4 ug/m3 respectively16

. Usually NO and N02 are
analyzed together and reported as NOX.
The effects of nitrates on atmospheric corrosion have not been

17, 18

studied extensively. Hernance, McKinney and co-workers examined

the effects of nitrate on nickel-brass springs and have shown a corre-
lation between stress-corrosion cracking, relative humidity, and nitrate
concentration. Gerhard and Haynie19 have shown the presence of ammoniun
nitrate in the atmosphere contributed to bridge failure. Alternatively,
Haynie and Upham20 found nitrate in suspended particulates to be a
relatively unimportant factor in the atmOSpheric corrosion of steel.
Summitt and Fink1 postulate that N0x may decompose protective finishes
on aircraft resulting in premature failure of the coating and the metal.
(5) Establishing Environmental Quality Standards for Corrosion
Summitt and Fink1 state that corrosion accelerates when the

following environmental factors are present:

IO

(1) Humidity, rainfall and solar radiation

(2) Proximity to the sea and other salt sources

(3) Pollutants, mainly sulfur oxides, particulates, photochemical
oxidants, and nitrogen dioxide.

As these factors increase, the environmental corrosivity will also
increase. At low values the effects become negligible. Consequently,
it is necessary to establish threshold values for each factor, either
alone or in combinations, which will establish severity. The critical
value may sharply divide slow or rapid corrosion such as iron or aluminum
in the presence of sulfur dioxide vs. humidity (cf. Rozenfeld4, pp 106
and 109). The variation of damage due to the environment also can be
gradual, such as the repainting of houses vs. particulate concentration

21, p. 98), thus the critical value is less pre-

(cf. Stoker and Seager
cisely defined. Where such critical values are known, they can be
utilized directly as environmental quality standards.

It is unfortunate that such data is non-existent for all environmental
factors except possibly humidity. Most laboratory studies on the effects
of pollutants on corrosion use much higher concentrations than even the
most polluted atmospheres so it is difficult to establish their relation
to real environments. Although much effortlz’ 13’ 22'24 has been devoted
to establishing critical concentration levels with respect to human health,
plant and animal welfare, critical concentrations as related to material
damage may be higher or lower than these. It can be seen, then, that the
problem of establishing critical concentration standards for corrosion is
neither simple nor straightforward.

A set of working environmental corrosion standards (WECS) might be

developed by consideration of the following:

11

(1) The range of values for the ambient parameters, which establishes
the limits for environmental exposure, if not the damage to be expected.
Such data include maxima, minima, medians and percentiles for the measured
parameter. Since the actual environments are known to vary in actual
severity, it follows that critical concentrations for practical use must
be within the range of ambient levels, perhaps near the median or higher.

(2) Ambient air quality standards established by the Environmental
Protection Agency are concerned primarily with human health. It is
assumed, however, that they do summarize careful consideration of all
available evidence by a host of scholars and bureaucrats. The values
represent the highest levels believed to be safe for human health and
comfort. Although materials may endure higher concentrations or may
experience damage from long term exposure, these values represent
criteria for damage to something.

(3) Experimental studies which relate corrosion damage with pol-
lant concentrations and weather may provide information for establishing
WECS. Several studies, using both real and simulated studies have been
published.

a. Ranges of Ambient Parameters

Throughout the world, a number of weather and air quality parameters
are measured by national and international agencies. Weather is measured
most commonly at aerodromes because it is a critical factor in aircraft
operational safety. The USAF Environmental Applications Center (ETAC)50
reports weather data from virtually every significant airport in the world.
Air quality data measurements of a limited number of pollutants are col-
lected by federal, state, municipal, and private air monitoring stations

in the United States and international, national, and private stations

12

throughout the world. The purpose of most of these programs is to
evaluate air quality primarily in the most densely populated areas.

Thus the results are representative of the population distribution
rather than geography. They would not necessarily represent the
environments in which aircraft are exposed and may not be directly
relevant to aircraft corrosion. Moreover, many monitoring stations-
especially private ones- were established to track specific pollution
sources, e.g., certain manufacturing operations, thus their data nay
reflect highly localized conditions. It should also be noted that

some data is not always available, especially in lesser developed
countries. Because of these limitations, the data compiled by the World
Health Organization (WHO)53’ 54 and the World Meteorological Organization
(WMO)55. 35 31, are the only data available to determine the range of
exposure.

25 analyzed ambient atmospheric conditions

Graedel and Schwartz
and quality based on National Weather Service and EPA data. Weather
data spanned 30 years from over 200 measuring sites. Air quality
data for several pollutants, mostly from 1973, represented as few as
82 to as many as 3760 measuring sites. The objective was to determine
the range of environmental parameters to which materials are exposed
in the U.S. and thus establish bench marks for laboratory and field
testing. Weather data analyzed were mean annual temperature and
mean annual absolute humidity. Pollutant data were the annual median
of hourly averaged continuous data for each measuring site.

Three results of Graedel and Schwartz should be noted: the median

of the 50th percentiles, the median of the 99th percentiles, and the

highest value reported (Table 1). The 50th percentile median represents

13

the "average of averages" reported, whereas the 99th percentile median
was exceeded only a 1% of all air quality sites. Graedel and Schwartz
define the 99th percentile medians as the Atmospheric Upper Limit Values
(AULV) which any be used for design purposes with the expectation that
99% of the applications will encounter levels below the AULV. The
maximum value was the highest mean reported.

The distribution of means as shown by Graedel and Schwartz is
essentially Poisson-like distributions for all factors except ozone
and $02. For ozone, a large number of sites reported values below
20 ug/m3 and a substantial number were grouped between 30 and 60
ug/m3. Nevertheless, the median, 36 pg/m3, probably is a valid
demarcation between high and low concentration sites. Sulfur dioxide
data from 447 measuring sites, however, were slightly skewed toward
low values. The maximum number of sites reported values at the median
and mean value of 43 pg/m3 and only 17% of the monitoring stations
reported means greater than 53 ug/m3. Because of this, the significance
of the median value for $02 is rather dubious when compared to other
parameters. This is unfortunate because of the important role 502
plays in corrosion.

Critical levels of atmospheric factors probably are somewhere
between the median values and the worst-case maxima or even the AULV's.
The AULV's represent the most hostile environments and this worst 10%
level would be inapprOpriate to use in a practical environmental
rating scale. Graedel and Schwartz suggest that the engineer may wish
to plan for all but the most hostile environments but Summitt26’ 27
shows that this has not been the case in the past. He submits that a

list of monitoring stations which exceed the AULV's include San

14

TABLE 1. RANGES OF ENVIRONMENTAL AMBIENT

PARAMETERS, CONTINENTAL 0.5.25

Total suspended particulates
(us/m3)

Sulfur Dioxide
(us/m3)

Photochemical oxidants

as ozone (pg/m3)

Nitrogen oxides
as NO (pg/m3)

as N02 (pg/m3)
Temperature (°C)

Humidity, absolute
(9/m3)

50th
Percentile

61

43

36

25

72

11.8

7.1

99th
Percentile

185

186

90

88

135

23.3

16.5

Maximum

Reported

500

410

110

98

150

25.7

18.3

M

15

Bernardino, CA only once (for nitrate ion particulates), and Travis, CA
and Charleston, SC are not mentioned. All three of these have been
shown to be severe environments, the first for paint degradation and
the latter two for metallic corrosionlo

It is possible to compare some of the demarcation points derived
from Graedel and Schwartz to actual material damage data.

Atteras and Haagenrud40 conducted weight loss studies of various
metals and give yearly averages of meteorological and pollution data.
Figure I shows a plot of one year weight loss values of rimmed carbon
steel vs. atmospheric 502 measurements for various sites in Norway.

A best fit line is drawn through the points. The halfway point on

the line denotes the value between the most severe and least severe
values. This value corresponds to a $02 value of about 42 ug/m3 of $02
which almost exactly matches the 43 ug/m3 SO2 demarcation value as given
by Graedel and Schwartz.

Figure 2 shows a plot of atmospheric $02 values vs. weight loss
for aluminum. The data points for this test are much more scattered
(presumably because of experimental error) and it is more difficult to
choose a best fit line. Nonetheless, a halfway point is chosen and
this corresponds to approximately 36 pg/m3 $02. This value is still
very close to the demarcation value. Therefore, we can conclude that
43 ug/m3 302 is a valid demarcation point based on both statistical and
actual materials damage data.

Knotkova-Cermakova and Barton41 plotted the effects of 502, relative
humidity and temperature vs. the corrosion rate of steel samples. Figure
3 shows a plot of corrosion rate vs. relative humidity at a constant SO2

uptake value of 30 mg/mzd and at a temperature of 15°C. The $02 concen-

16

.oefi «gamed

.- .=u\m.aeooeu_.. oe_xo_o apefinm

aw so or on o

_
.—
.—
.—
>—
T...
i”

 

—--—-—---q

 

 

Aw:

 

 

uex=omd2wnzmmg§tc.mim9m
zagafiut=_m:mm:.xzw.sm»w8

Q3

ssod ; no 1. on

(w'bS/B)

17

255-2023: «2.85 .523

cc

N mesmwd

 

 

 

.3
.3
.2
.to
.2
.2
.2
.3
I...
L
fifi~
3

 

 

8:88 m5.“ SEE .8
£55.: 8... 83 :95 «m» as

$501 a tab 1 on

(m-bs/é)

18

o2

.va mesmwd

8 >233. 32.3“

9 8 R

)—
—
W

 

 

 

8

 

.85 8
as. 33.58 .8 535. 3:59.

a ; Ira no 1 504.403

( °4q_uu:> °bs/6H)

92 S a. 2.23. 3:23.
8

.me mesmwd

3

 

19

 

1

( 3 90-46OQ )

L

 

 

222398 8... H2353 .59 .8
E25: Jua.mzo_:eao 2258... :3

- sung- ,5 p 1 a -qn ; '4-dmol

20

tration is considered low and the temperature moderate. Figure 3 shows
a dramatic increase in the corrosion rate from 75% to 85% relative
humidity. If a demarcation point of 75% relative humidity and 15°C
temperature is taken, these values correspond to an absolute humidity
of 9.6 g/m3. This value is considerably above the 7.1 g/m3 demarcation
point as given by Graedel and Schwartz.

It would seem wise, however, to retain the 7.1 g/m3 absolute
humidity demarcation point for aircraft because there will usually be
a large temperature difference between the surface of the aircraft
after high altitude flights and air temperature near the ground. Such
a difference will cause condensation on the aircraft surface. Even at
lower humidities and temperatures, thus condensation will occur. There-
fore we will allow a "safety factor" of about 2.5 g/m3 absolute humidity.

Grossman9

gives dew formation conditions of relative humidity vs. tem-
perature difference of surfaces and the surrounding air to cause con-
densation. This data is shown graphically in Figure 4.
b. Proximity to the Sea and Other Salt Sources

Several studies3’ 28’ 29’ 30’ 31 have shown that accelerated atmos-
pheric corrosion near the sea shore is correlated with airborne sea salt.
Establishing a critical distance from the shore is difficult because
there is little quantitative information relating corrosion to atmospheric
salt concentrations, or even relating salt concentrations to the distance
from the shore. Sea salt is a primary concern because there are few
other sources of airborne salt. Coastal salt flats, however, have been

shown to contribute atmospheric chloride downwind32.

21

A suspension of solid and liquid particulate matter in a gaseous
medium is defined as an aerosol46. Typical constituents of the atmos-
pheric aerosol are: silicates; salts such as NaCl, MgClz’ M9504, Na2504,

NaNO3, (NH4)2 504, NH Cl, NH4N03; acids such as H2504; metal oxides;

4
organic combustion products; biological material; volcanic material; and

extraterrestrial material of volcanic and stony compounds33. The concen-
tration of aerosol particles decreases with increasing size and over land
is typically 103 cm'3, 1 cm'l, 1 liter’l, 1 m"3 for particles larger than

4 3 2, cm in radius, respectively33. In air over the

10'5, 10' , 10' , 10'
ocean, with winds just strong enough to produce white caps, the concen-
tration of all hygrosc0pic particles including sea salt is typically 102
cm'3, 10 cm'3, 1 cm'3, 1 liter"1 for particles with masses larger than

14 -12 47

10‘16, 10' , 10 , and 10'9 g, respectively . Particles larger than

20 um in radius remain airborne for only a short time and their occurance
is restricted to the vicinity of their source33. The lower end of the
size spectrum is controlled by coagulation causing particles smaller
than about 5x10"7 cm in radius to become rapidly attached to larger
particles or water dr0plets. Thus, small ions may only exist for only
a few minutes in relatively clean air and only a few seconds in heavily
polluted air. The lifetime of particles of intermediate sizes is con-
trolled in the tr0posphere by the combined action of sedementation,
coagulation with other aerosol particles and with other cloud drops,
rainout and washout. In the lower troposphere, the residence time is
33, 48

typically between two and four weeks . Chloride in rainwater is

correlated with small particles, whereas direct settling of larger

22

sodium chloride particles occurs near the shore. Thus measurements of
sodium chloride in rainwater and of atmospheric sodium chloride par-
ticulates vs. distance from the sea may suggest values for the critical
distance.
(1) Salt in Rainwater

Atteras and Haagenrud40 collected rainwater at various distances
from the sea and analyzed for chloride concentration. These results
are shown in Figure 5. This plot shows a decrease in chloride concen-
tration at about 50 to 100 meters from the sea. Thus, the concentration
of sodium chloride in rainwater is high over and near the ocean but
diminishes inland32. In the U.S., the concentration decreases loga-
rithmically with distance from the sea up to 500 km and is constant at
greater distances. In EurOpe, the concentration decreases logarith-
mically up to 300 km , but increases slightly beyond that apparently
because of the influence of the Baltic Sea.

It is unlikely, however, that chloride in rainwater is relevant
to aircraft corrosion. The exterior surfaces of aircraft exposed to
rain are protected by paint, whereas most interior surfaces are not
exposed to rain. Moreover, the decrease of chloride in rainwater occurs
over large distances, whereas the decrease in corrosion is quite
abrupts’ 9. Corrosion rates 10 km from the shore are approximately the
same as corrosion rates far inland. Consequently, the critical proximity
should not be determined from rainwater chloride concentrations.
(2) Particulate Sodium Chloride

Duce gt_al;§4 have measured the concentration of particulate sodium
chloride and other ions in the air at various elevations and distances

from the sea shore on Hawaii Island, HI. All measuring sites were down-

23

.. mass 3m .3: 3535 ca
33385..
we as is we
a pi, e _ e e

m oe=m_d

 

3m mi 5.: magma .8
mmgazzm E EEcmEBzS 83.3

__._ll_ .3,
a. 'S

G.
‘.

__‘._._____ .___1__..”__

tr".

“.1-.. _ __._ w
l“!

._-1__
CS-
:—

 

_l

1'3'1'4

.20 to I

’32.

E0

LICOV ‘t; \j ‘|l

-‘ F) '

—.

|‘ U] ' I‘I

24

wind from offshore trade winds. Their results show chloride concen-
trations at all sites varying widely with ambient weather conditions.
The results show a consistent, monotonic decrease in chloride concen-
tration with increasing distance from the shore.

Hudson and Stanner31 found in Nigeria that sodium chloride con-
centration in the air varies within wide limits and depends strongly
on the distance from the shore. The sodium chloride content in the
air is about .22 mg/m3. The amount of salt that settles out on the
surface under these conditions reaches values from 10 to 1000 milli-
grams per square meter per year. Corrosion tests were conducted at
various distances from the shore with simultaneous determination of
airborne salt concentration. The corrosion rate of iron samples and
collected salt virtually coincide vs. the distance from the sea. The
corrosion rate and the amount of salt collected falls sharply from 0
to 0.4 km from the sea and then levels off to an almost constant value
for values greater than 0.4 km.

McCaul and Goldspeil42 give data on airborne NaCl vs. distance
from the sea. A plot of this data is shown in Figure 6. A dramatic
increase in chloride concentration occurs at approximately 370 meters
from the sea. The corrosion rate of iron vs. distance from the sea

is also given. Figure 7 shows the corrosion rate of iron increasing
with higher salt concentrations, with a dramatic increase also occuring
at about 370 meters from the sea.

Junge33 (p.176) has drawn together the available data on giant
salt particles (particles greater than lium radius) vs. distance from

the sea. Concentration values of 5 ug/m3 correspond to near the shore

and approach 0.5 ug/m3 at distant points inland.

25

. o me=m_d

 

 

 

 

 

{mm scam mochmHo
.w: mac z~ Hzmhzou 54cm

.m,,..$u...._._. 3m so: 3535 Ne
AmvcamDOLPH

; mg 05 :5 Na 0
L. b e — p p b a
a“
.N
.n
T:
.m
.o
.N
.m
.e
.9“
. ccsoum crane noon mp “A fififi

f. unsoum oeona «can he “a

2

trove/IQEN J0 6w

£1

6

26

. n weaned

 

 

 

 

 

 

 

 

$335 new 5.: SEE me
3338;:
m; H mu... 0.0 2.. Wm a
h _ e e a a b P L p e a. a
T M
_.
a .
m 1.:
w
_
w an
_
m
,
I”
2395 cream «can t "a .
m
ear

.3 m: :2: 82520
.8 29: do HE §8¢m8

34‘323

l' .’-.-j “I g

27

The available data on atmospheric corrosion near marine environ-
ments suggest that the decrease in corrosion rate parallels this
decrease in giant salt particules, and marine atmospheres are aggres-
sive in direct pr0portion to the concentration of airborne sodium
chloride particules.

Most studies suggest a critical distance of less than 1.5 km
for sites where strong off-shore winds are not prevalent. Summitt and
Fink1 allow for the variability of weather and extend this to 4.5 km.

43 classify various locations around the world on

Doyle and Wright
the basis of their Marine Corrosivity Index (M.C.I.) that is based on
the CLIMAT test which was first develOped by Bell Laboratories. Figure
8 shows a plot of M.C.I. vs. distance from the sea at Durban, South
Africa. This curve shows a marked increase at about 3.5 kilometers
from the sea. It can be seen, then, that allowance for various off-
shore weather conditions is definitely necessary and acceptance of the
4.5 km critical distance as proposed by Summitt and Fink1 is wise.

c. Working Environmental Corrosion Standards (WECS)

After considering the existing literature on materials degradation
and environmental factors, it can be seen that there are no firm guide-
lines for setting WECS, with the exception of humidity. Metallic
corrosion is definitely accelerated in the presence of $02 and high
humidity, and probably accelerated by N02, oxidants, and many par-
ticulates. Organic protective finishes are deteriorated by solar
radiation, oxidants, some particles, and possibly NOx and $02, Un-
fortunately, published reports do not tell us at what levels these
factors become significantly damaging.

We can, however, ad0pt the view that critical values lie within

28

. mezmp
new . .._
_EL_ «mm 502% mucm~m_o

I“~A

H

00
:1"
O

 

I... l... .2

ram

-l-l ‘

 

l
l
l

 

3
. .. o
. E. N
.f O
, s
F. .r .n
.J 1 G t,
__ a...
.. e I.
a. .. ..
r an t.
_ _ me
’ _ a
: ....
_fe
_
g...

cum “IF zone wuchmHo
.m: mezH ehmsfimommcu wz~m¢z

29

the range of ambient values, because accelerated corrosion has been
observed in existing environments. Summitt adopts two sets of WECS based
on the analysis of Graedel and Schwartzzs. The first set are the

50th percentile values and the second set are the 50th percentile

values plus twenty percent of the difference between the 99th and

50th percentiles. These are listed in Table 2. The values for prox-
imity to the sea are based on the analysis presented earlier. The

solar radiation data values are based on the mean (July) values for

the continental U.S.

These WECS have been used in the Corrosion Severity Index Algorithms
(described in the subsequent section) and the results compared with
experimental environmental ratings. The agreement is sufficiently good
that the values of Table 2 together with the Algorithms may be used to
compute accurate relative environmental serverity for corrosion in air-
craft.

5. Corrosion Severity Algorithms

Summitt and Fink1 proposed a set of three algorithms based on
locally measured factors which rely in part on maintenance experience
as recorded in U.S. Air Force records (AFM 66-1). A particular feature
of this approach is that the authority to set intervals is left in the
hands of local management. These decisions would be based on locally
measured meteorologic and pollutant conditions and would be subject to
changes dictated by local experience.

(1) Corrosion Maintenance in Aircraft
Excluding housekeeping, corrosion maintenance involves:
(1) washing of exterior surfaces

(2) repair or replacement of protective coatings and sealants

30

 

TABLE 2. WORKING ENVIRONMENTAL CORROSION
STANDARDS (WECS)
Annual mean
I II
Suspended particulates 61 86
lug/m3)
Sulfur dioxide 43 72
(lg/m3)
Ozone 36 47
(Hg/m3)
Nitrogen dioxide 64 78
(ug/m3)
Absolute humidity* 7.1 9.0
(9/m3)
Proximity to sea
or salt source
(km) 4.5 2.0
Solar radiation, July
(Langleys) 600 650
Rainfall, total
(cm) 125 150

*Absolute humidity is the product of relative humidity and the mass of

water in one cubic meter of water-saturated air

at a given temperature.

31
(3) treatment and repair of corroded components

Environmental elements which corrode metal are not necessarily the same
as those which deteriorate paint and sealants. Humidity, $02, and
certain other contaminants corrode bare metalll, whereas paint films
deteriorate under the action of sunlight, photochemical oxidants, and

9, 23, 24.

other pollutants Soil deposits which are also harmful to paint

films, are related to su5pended particulates, and their damaging effects
are accelerated by contaminants such as $0235.

No single algorithm can classify an environment with respect to
all three corrosion problems. Instead three decision algorithms are
required to determine intervals for:

-aircraft washing

~complete repainting

-corrosion inspection/maintenance
Each algorithm would assess the level of local contaminants and via
a decision map, lead to recommended intervals for each maintenance
cycle.
(2) Aircraft Washing

Aircraft are washed both to maintain appearance and to remove soil
deposits which may damage the paint. There are several sources of soil:
engine exhausts, fuels, and lubricants; airborne particulates; and the
workers' shoe soles during maintenance and servicing operations. Soil
deposits will attract and retain moisture from humid air and gaseous
pollutants, particularly 502. Thus, the damaging effects of soil are
compounded by high humidity and pollutant concentrations. It is not
likely that surface soils accelerate paint degradation by sunlight or

gaseous oxidants, but there is no evidence to support this view. Thus,

32
aircraft washing intervals selected to protect the paint and exposed

metal should be related to particulates (and proximity to the sea),
$02, (possibly) N02, and humidity. It is likely that cosmetic purposes
will be served by the same intervals. USAF recommended washing intervals,
for several years, have been 45, 60, and 90 days, depending on local
conditions. At many airbases, where indoor washing facilities are not
available and winters are severe, even the 90 day wash interval is
impractical. Other airbases plan 30 day intervals. Practical washing
intervals, which are consistent with environmental risk factors and
rigorous climates, are 30, 60, and 120 days. They are designated as A,
B, and C, respectively.

The Washing Algorithm (Figure 9) first determines if the distance
to the sea is less than the WECS distance. If it is, washing interval
A is recommended. If it is not, particulate concentrations are compared
with WECS. If the ambient exceeds the standard, then the ambient SO2
concentration is taken into account. If the S02 is higher than WECS,
interval A is recommended; if lower, interval 8.

If particulates are below the standard, 302 concentration again
is checked: If high, interval 8 is recommended; if low, moisture
factors are considered. High moisture values - either AH or rainfall
greater than WECS - lead to interval 8 recommendation; low values lead
to interval C.
(3) Painting

Aircraft are painted to protect metal surfaces, although operational
and cosmetic factors are significant. Protective maintenance for the
paint finish is effected at three levels: (a) minor touch-up; (b) major
touch-up; (c) complete strip-repaint. Minor and major touch-ups are

done in the field or intermediate level maintenance, whereas complete

33

>61 >43 0

 

{43

WASH I NC Il‘i'l'l‘invfilii

A in day:
II (in (IOVS
(‘ l/l (lays

Figure 91. Aircraft Washing Interval Algorithm. Working Environmental
Corrosion Standard I (see Table 2) are used. Units for TSP,
and $02 are pg/m3, for AH g/m3, and for Rainfall Annual

Total cm.

34

>600 <36 >43
502

1600 :43

 

 

>43

 

SO

 

Repaint Intervals

A 36 months
8 72 months
:43 C 120 months

Figure 101. Aircraft Complete Repaint Interval Algorithm. Working En-
vironmental Corrosion Standard I (see Table 2) are used.

hv, are Langleys (July), for Ozone and $02, and ug/m3.

35

4 . 'ikm

 

 

 

Expected Corrosion Damage

very severe
severe
moderate
mild

061);

Figure 111. Aircraft Corrosion Damage Algorithm. Working Environmental
Corrosion Standard (see Table 2) are used. Units for AH
are g/m3, for Rainfall, Total Annual cm, for $02, 03, and
TSP, ug/m3-

36

repaint is authorized only at depot level for large aircraft36. Minor
touch-up is done to repair damage caused by ablation and the like.

Major touch-up is applied to fasteners, runway damaged lower surfaces
and solar damaged upper surfaces. The need for touch-up painting must
be determined at field-level inspections. An environmentally-based
algorithm should not be used. The following paint interval algorithm
refers to complete strip/repaint maintenance.

As before, three intervals, A, B, and C, are recommended. Paint
systems currently in use - epoxy or polysulfide primers and polyurethane
finish coat - should provide a service life of 10+ years in the mildest
environments37. Consequently, the A, B, and C intervals may be equated
to 36, 72, and 120 months, respectively.

Environmental factors which deteriorate paint are, in order of
severity, solar radiation, oxidants, and sulfur dioxide absorbed on
soil deposits. Soil deposits themselves might be included, but there
is insufficient information to relate repaint schemes with the nature
of the soils. Thus, only sunlight, oxidants, and $02 are considered.
The repaint algorithm (Figure 10) compares the solar radiation level,
ozone, and sulfur dioxide concentrations with WECS values. High values
for all three lead to the C interval. Various combinations of high and
low values lead to the B interval.

(4) Corrosion Damage

The Corrosion Damage Algorithm (CDA) is of a different nature than
those for washing and repainting, which recommend intervals appropriate
to the environment. Although CDA might be used in the same way, such

use is unlikely. Corrosion repairs routinely are effected simultaneously

with phased and isochronal maintenance efforts, and would be both un-

37
desirable and difficult to impact their scheduling.

The CDA is intended as guide for anticipating the extent of corrosion
damage and for planning the personnel complement and the time required to
effect their repairs. The guidelines at this point are general. Event-‘
ually they should be incorporated into the Reliability-Centered Main-
tenance phase schedules for specific aircraft systems.

The algorithm (Figure 11) considers first the distance to salt water
(or salt flats), leading either to the very severe (AA) rating or a con-
sideration of moisture factors. After moisture factors, pollutant con-
centrations are compared with WECS either for $02, TSP, or 03. High
values for any one of the three pollutants together with a high moisture
factor leads to the A rating, but if they are low, together with high
moisture, the moderate 8 rating results. Low moisture factors with a
high pollutant value also result in a 8 rating. If all are low, the
mild C rating results.

III. Non-CONUS USAF Base Evaluations

The Corrosion Severity Algorithms have been established and it is
now possible to apply them to the non-CONUS (overseas) Air Force bases.
It is important to note that pollutants are not monitored in the world
with the same ways and with the same enthusiasm as they are in the United
States. Consequently, application of these algorithms to non-CONUS
locations is neither simple nor straight forward.

It is not suggested that non-US scientists and engineers do not
study atmospheric pollutants. The literature contains much data, and
very detailed studies have been published, especially for Europe and

51, 56, 57.

the Soviet Union The problem stems from the fact that no

single agency, like the U.S. Environmental Protection Agency, functions

38

as a central office for compiling and publishing data in a standard
format. Under sponsorship of the World Health Organization53’ 54’ 56’ 57
programs are in progress to correct this situation. Currently, however,
data are simply “hard to come by".

In some cases, estimations of pollutant levels were based on
population distributions and geography. In some other cases, the only
pollution data available was from many miles from the actual air base
site. Weather factors were taken from the USAF Environmental Technical
Applications Center (ETAC)50. ETAC data span the entire globe, hence
may be used for non-CONUS bases.

Weather and pollution data, along with the resulting Aircraft
Washing Interval and Aircraft Corrosion Damage evaluations for the non-
CONUS airbases are shown in Table 3. Repaint evaluations have not been
computed because of the lack of available ozone (oxidant) data. Ozone
analysis is difficult and expensive to perform and most European and
Far Eastern air monitoring stations do not record such data49.

IV. Results and Conclusions

Summitt and Fink39 report that these algorithms have been used to
compute ratings at nearly 200 sites. These ratings have been compared
with actual corrosion measurements as a part of the Pacer Lime Program
and USAF corrosion maintenance experience with large aircraft. These
experimental results provide excellent support for the environmental
ratings. As we have seen in this report, they also compare favorably
to exposure tests reported in the literature. It is with high con-

fidence, then, that we can expect good results when applying these

guidelines to overseas Air Force facilities.

39

 

<< < o.m 0"" m.w m.- on “Know” zvmmmo emcox .m cemczx

<< < o.~ m- m.m o.m ov :onwwo zommco ecu—wu_ ease—e—og
so. so—

<< < m4 Zn mg: m.- noes»: unseen: $12 538 :32. 2:63.

<< < e Nm o.o a.o~ ufiono zwfiwmo aoxeap L_e~_
xo_ zo—

m a me we ~.o~ ~.o~ nasamma cesammo mowmmo zoo~mo xmxeah x__c_oc_

m a o~ vmw ~.m~ m.o~ aw N" me 3cmo~o zmmmo ecoN .oceu tease:

< m m ov ~.¢ ~.- on mHN uvemmo zvmsmc mucosa cox_:m__e:

m < mwm «N o.o m.~ mm ON wm_~o szocc acescou .3 ago:

< < mw n- m.m~ o.o~ Nm we unmow— z-m~o mu:_a.—._se ace—u

< < N” KN n.~ v.o "a mo unomo zofimmo mace—Logan: Eeoeoume<

3w: gang

m < SN 3 as n.» m... S :25 Emma c8528 .3 953:.

< a o.~ mm e.~ ~.m m~ go go wwwgo zwommo x: «coaoxucom

< < mm cmd o.m m.- mm~ MN wom-o zmocvo »_au_ oce_><

a < mm mm o.~ v.o m~ as w“ xmfioo z-~mo x: meancoo_<

wmaseo postage. as .eom ca hw_u_e=:
co_moccou gnu: oucaum_o o .pouepom u nus—omo< u ocsuocoaeo» a oz a wan 5 am» a uu:u_ocoa a ova».ua4 agucaou amen ew<
.omoEoo co.moccou uco m=.:ma3 uuoeucp< so. mmc_u~¢ hu.>.moecou .eucueeoc_>cu ecu ouao amene_< mzzou-coz .m o_oep

4i)

 

a < men we o.” ~.a co em ueweo zmeavo seesaeo .3 eexueee_exN
< a n2 3 N.» v.3 we an :38 20:3 .3on «SEES
a a me am; m.o ~.: 3 mm 332 zmvmno :32. 33.:
< m 2 3 mg s6 ms 8 3 News zmcmmo x: owe—Lats:
< < no on ~.~ ~.o m~ 2 as :38 235: V5 vemwhmu
< < can 3 o.~ «.2 2 m3 338 zmwcco 53m saute»
< < oo ow" ~.m 5.2 .: 332 23mg 2:3. .m .608.
a < m3 3 m6 o6 mm 2 3.8 22.33 Ace—Eng .3 30—53:an
a < mom on m6 n.» 8 on wwmmc 2°33 nan-tow .3 523.3
a < mom 3 mi ¢.m m3 m2 33¢ :83: 9.353 .3 5.2-505.
m < 2n No To «6 on on moms .833 531.89 .3 5333.
a a 2 m2 «.5 fl: 8 ~85 .323 3.32 .m :25
a < 8 3. 1s .5 ms 2 as was z-~ma g: 22:02 E
a < on em 73 Ta 3 ms 2. ~38 53.8 x: 535.33
.5— 33 .8—

<< < 9? v3.3. @953. van-fine 32:3 553 3.32 30.... man:

Ilommeoa .2535 5. an 3 5.3%: . x m

5.3....3 :3: 3:32: o :35... a 3:32: o 9:59.093» 4 oz cm .3» a 093.93.. a 333: >353 38 e:

..e.u .n e.g.e

41

Footnotes:

a.

b.

C.

d.

e.

Latitude and longitude are in degrees and minutes, e.g., 05222N is
52° 22' N, where the last two digits are minutes.

Atmospheric contaminants are mean annual values in micrograms per

cubic meter.

Mean annual temperature, degrees Celcius.

Absolute humidity is the product of mean annual relative humidity

times the mass of water, grams per cubic meter, in water-saturated
air at the mean annual temperature.

Mean annual rainfall, centimeters.

MACDILL AIR FORCE BASE RINSE FACILITY

I. Introduction

Corrosive and unsightly deposits from atmospheric aerosols, in-
dustrial and agricultural dusts, and engine exhausts accumulate directly
on aircraft surfaces, open wheel wells, service bays and cockpits. It
is the policy of the USAF to wash aircraft at 30-60-120 day intervals
depending upon the ambient concentrations of these airborne pollutants.
This practice serves to remove corrosive soils and improves appearance.

Figure 12 shows a vicinity map for MacDill Air Force Base, Tampa,
Florida. Salt water surrounds the base on three sides, and is as close
as 1 km from aircraft flight lines. Moreover, nitric acid, sulfuric
acid, and potash fertilizer plants are located about 7 km east of the
base. While easterly winds occur only about 37% of the time61, air-
borne ocean-spray salt could be expected to deposit as much as 100 mg/mz-

68, 69

day on open surfaces within 1 km of the shore. An F-16 fighter/

interceptor airplane has an estimated upper surface area of approximately

2, and thus can accumulate as much as 5 g of NaCl per day67. Hence

50 m
the corrosion environment at MacDill is considered to be especially
severe.

U.S. Navy carrier-based aircraft operate in a similar environment,
and are rinsed periodically with fresh water as a part of their corrosion
control program70’ 71. In 1976, USAF began work on the installation of
an automatic rinse facility at MacDill. The facility is located on a
short taxiway near the main NE-Sw runway (Figure 13). Initial operation
of the facility began in 1979. Pilots were instructed to taxi through
the rinse at the end of each daily flight schedule, thus removing soluble

72

salt deposits .

Because of several taxi-way construction projects implemented at

42

43

g.

 

 

 

 

'3
i
5'.
g . . .. I. .
. , \\:f'll'l d; ‘ . qfifi ’ a
. ,. ‘ a“ " 5......
vucxmrvl . l 4:" t“ ;
MAP ' Q. ' ' '7 / '
' V suwswmc A '
g ‘1 ._ rum . .
l l 5' ' \ n . _ V‘_.__'_'______. '

Figure 1269. Vicinity Map. MacDill AFB and Environs.

44

rpaoue‘cr. LOCATION

, f’akse' courmcrmo
.. omcen ,

3
(

 

CA HO.

,g-aLoo 9.3.2 .

. ‘ . ‘ .
3., ' ‘
. . . . a
“‘ I ‘. ' n
.. e... ~_ .‘ u ‘

 

 

 

 

 

 

‘ "ease cML ENGINEERS
. ' '.‘ . ,OFFIG '.

 

apps. :0

 

 

 

 

 

\
“ s
. ’9 ' - use ‘
Q mam.

 

 

 

 

 

 

 

 

 

Figure 1369. MacDill AFB. Location of Rinse Facility.

45

. mic-ran. (At!
lg .
TOR CABLE m \
” WWN‘WQN
'3 . FORM mu
I "
(I

runway 6k

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l .
. 4'. - . «l
’ '- \ Mam-tem-

\\_QF ‘“fi$£§fi$de.o
" germuoaoans.

I

 

I )LI'IL
.‘ ?’

 

 

 

 

 

Figure 1469. MacDill Rinse Facility-Schematic.

46

 

 

 

In: ggwous
/1:m.u ' gig
Sta-aw m max:

_ RT“!

 

 

 

 

 

-/”.

M 6 msuo'u
«ms-ac mun
\ ma 3: unto" my -
mam: nevu\ . I Z
tfi ‘
\
'9 m we w‘w‘ . m~
‘n8 -
L A
l' ‘\
ours on TNNCH/ l—ll‘i “BL. 125-fl \
MM

Figure 1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

569.

 

 

MacDill Rinse Facility-Schematic.

47

MZCLOW HOLE

 

 

 

 

If. 950‘ _ "'l '
I
2x . l
D
.‘(gy

 

 

 

 

 

 

 

..
3‘ -

 

 

 

‘-
.

 

 

 

 

"" A. I.

 

 

t‘?‘\- 1.. 1 I
4"‘31’E€L OIPE

WATER. SUPPLY)
GLEEVE ANO
CAULKE WATfiR-
TIGHT

 

 

 

 

 

 

 

 

 

Figure 1669. Side View of MacDill Rinse Facility Water Storage

Tank-Schematic.

48

Moron JUNCTION Box

2.0-LEVEL «$1:me
okaFLow no Le
't
60! AROUND FLOAT “-

(soup Slot $-
“momma 901mm)

 

 

 

 

 

 

 

 

 

 

 

 

 

‘0. ‘£‘
froc'mcnzAssR:// '3
Buf’vc‘kaL "t, '

Figure 1769. MacDill Rinse Facility Water Storage Tank. Cut-Away

fr

 

 

View-Schematic.

49

 

' 3:532:03. “firm
«5 DIRECT 5:25105 Mtg

22mm WATER mo—1 (nu, are) 10 mm: FtEhP
on. SEPARATOR ASTE Sump

~$NELL :- - “AMI. Ogsa-wzea

' '::l ' :: ”I I.- h
3"?“ 1.39““ - ‘ r! I mm a HOSE
_ 1‘ I ”.1 __ _I_] I

 

—-_
~—.___ ~-———~~

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

L ;M
flea . I
.J ,L____j______JT "‘-’ -

3 . :
ILsunoA'r BRACKET

 

 

 

Figure 1859. MacDill AFB Rinse Facility Water Storage Tank. Cut-

Away View-Schematic.

so 9 tour out“ 1'
Felt “a“ we

50 n? Pump
/ e 6166
AL ,

50

(0‘00“!wa 5060R-

 

 

w—mss u»: l /I
- 5
van: MN. I
To muauY- (it! . .
A—Id-D

III:

—i film—am: |.-‘-' ‘QII-

Housman l.
M nos-r .

5 W \ILvt t MIMTOR
u LII! '

 

 

 

 

ll‘

 

‘1\

 

OWN

 

 

 

Q
I

 

 

...,....../ 3,

Figure 1959.

View-Schematic.

mum” m
u. '5 -
\W-Ve "T m STIUUIONS)
é
\- ”you BflCKET fat.

I!“ “GMT“! WE
LII-to um. {mm-l

/ 3" «alumna» LI»:

SUVPORT am not.
Wouufluo VALvt

 

MacDill AFB Rinse Facility Water Storage Tank.

ORA woof

naught mo 3 MYWVE
MoTo OHM "it'UC‘VIW
SW50. CONYROLL‘

 

Top

51
MacDill since 1979, the rinse facility has been used only sporadically
since its construction. It was not in operation in 1977-78 and 1983-84,
but records are not available that document when it was available for
use as well as used regularly.

Moreover, the efficiency of a fresh water rinse for the purpose of
controlling corrosion in aircraft has not been demonstrated. No system-
atic experimental test has been designed to address the question.

It has been suggested that a fresh water rinse may do more harm
than good because salt-laden water may accumulate in “nooks and crannies"
where accelerated corrosion may result. Such an occurance would be
worse than no rinse at all. However, if the rinse water contained cor-
rosion inhibitors, such undesirable effects might be avoided. Moreover,
the collection of inhibitor-containing rinse water in crevices would
have beneficial effects by providing a measure of residual corrosion
protection. It was proposed63 that the addition of a borax-nitrite
inhibitor to the rinse water would be beneficial. The proposal was
accepted and a pre-mixed inhibitor formulation, corresponding to the

63

"Optimum" formulation of Khobaib was supplied to MacDill from a local

chemical supply company73. Inhibitors where first added to the rinse

facility in 1979.

Some believe that rinsing and/or washing creates more problems
than it solves. The nuisance argument is recongized but cannot be con-
sidered seriously. Summitt72 lists major charges against rinsing at
MacDill:

\a) water “drying spots" on the airplane canopy,

(b) corrosion problems on landing-gear-well electrical connectors,

and

52

(c) brake-disc fractures, caused presumable, by cold water sprayed
on hot wheels after landing.

Summitt72 addresses the canopy "spotting" problem by suggesting the
use of a surfactant additive to the rinse water or by “stripping" by a
high velocity air stream much like those used in automatic car-wash
operations. He also suggests that if the airplane were rinsed pre-
flight instead of post-flight, not only would the canopy “spotting"
problem be solved, but the brake disc fracture problem as well.

Summitt continues to make a strong argument in favor of a pre-
flight rinse. Some believe that salt and other contaminants accumu-
late on the airplane during low-altitude flight over salt water. At-
mospheric salt concentrations decrease rapidly with altitude, with con-
centrations of all size classes, large through small, ranging from 0.8
to 13 ug/m3 at 100 m over calm ocean seas. Sea level measurements 50 m
from the shore are about 100 pg/m3. This value is also comparable to

68’ 69. Hence atmospheric salt

sea level values 1 km from the shore
concentrations near the sea are much greater at ground level than at
the lower altitudes. It is also very doubtful that an airplane flying
at speeds of 150-200 kt through such an atmosphere will accumulate
enough salt deposits to warrant an immediate post-flight rinse.

Ocean spray aerosols can deposit as much as 100 mg/mz-day of sea
salt on open surfaces within 1 km from the shore62. Hence, most of the
sea salt on an airplane will accumulate while it is on the ground.
Consequently, a pre-flight rinse would appear to have more merit than
a post-flight rinse.

II. The MacDill Automatic Rinse

The facility is located in a short taxiway (Figures 2, 5, 7, and 8).

53
Automatic "on-off" sensors embedded in the pavement control the 600-
gal/min pump, delivering water through 31 spray nozzles set below the
taxiway. The sensors are spaced 178 feet apart, hence at an estimated
taxi speed of 5 mph, the system will operate for approximately 24 sec-
onds, delivering 240 gallons of water68. A portion of the rinse water
is collected by catch basins embedded in the runway and returned to the
water storage tank.

When the system is fully operational, all aircraft are required to
pass through the rinse facility after each mission72- The rinse facility
is used exclusively by smaller (primarily fighter) aircraft.

The underground water storage tank for the facility is shown schemat-
ically in Figures 3, 4, and 668. The tank is divided into four compart-
ments: (1) the valve pit, (2) the fresh water compartment, (3) the re-
turn water and oil separater compartment, and (4) the waste sump com-
partment.

The valve pit holds the shut-off valve for the fresh water supply.
The fresh water and return and oil separater compartments supply the
rinse facility with the water that rinses the aircraft. The water from
the two compartments is mixed continuously by a recirculating pump.

Oil and other contaminants which accumulate in the return water usually
float and are removed through a weir that spills into the waste sump.
Waste water is removed from the sump by a manually operated pump that
pumps the water directly to the drain field. The sump also handles

any excess water (rain) that enters the system. The capacity of the
system is 2990 gallons68.

III. Rinse Solution Constituents

54

As a result of recommendations by Khobaib and co-workers

corrosion inhibiting chemicals are added to the rinse facility water
to reduce the effects of trapped hard water in crevices and dry bay

areas. The chemicals and their recommended optimum concentrations are

given below:

0.35%
0.10%
0.05%
0.01%

Sodium Borate (% by weight)
Sodium Nitrate
Sodium Nitrite

Metasilicate Pentahydrate

20 ppm Sodium Meta-hexa Phosphate

20 ppm Sodium 2-Mercaptobenzothiazole (MBT)

These inhibitor materials were pre-blended for convenience by a

private contractor (Erny Supply 00., Tampa, Florida73). The batch

formulation for the inhibitor material is given below:

Sodium Borate...........................535
Sodium Nitrate..........................335
Sodium Nitrite..........................165
Sodium Metasilicate Pentahydrate.........16.5
Sodium Hexa-meta Phosphate................3.5

MBTOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO00.0.0105

TotaIOOOOOOOOO...00....0.0.0.0000000000105605

It is important to hold the constituents to optimum concentration
levels to insure pr0per inhibition characteristics because they can
become very corrosive at both insufficient and excessive concentrations.

The MacDill facility automatically replaces unrecovered rinse water with

lbs.
lbs.
lbs.
lbs.
lbs.
lbs.
lbs.

55

fresh water. Hence, the concentration of additives will decrease with
each rinse. Concentrations after each rinse can be calculated if the
amount of rinse water lost per rinse is known, although rainfall would
complicate the matter somewhat.
IV. Additive Depletion

Rinse water may not be completely recovered because of evaporation,
carry through, and run-off. Estimates of the fraction recovered ranges
from 25 to 75%; an accurate value could be obtained from daily water
meter readings and frequency of use data.

The additive concentration, C", after a number of rinses, n, can
be calculated from the initial concentration Co and the amount of water
lost per rinse. Lost water is the fraction x of water used. Since the
system delivers 600 gal per minute, an airplane taxiing at 5 mph through

the 178 foot rinse area would draw a=240 bal in 24 sec. Thus,
_ n
Cn - Co[(b'xa)/bJ 9

where b is the water capacity of the system, (2990 gal or 11317 kg);
units are appropriate to the concentration expression, i.e., mass per
volume or weight per cent.

As an example, if the fraction of water lost is x=0.50, then,
after twenty rinses, C20=0.441 Co, and after 100 rinses, C100=0.017 Co.
The amount of additive lost after n rinses may also be calculated from
(Cn-Co)b. When the inhibitor mixture was added to the MacDill facility,
the initial charge was 120 lb563. A typical MaCDill week constitutes
about 100 rinses75, thus the "optimum“ inhibitor concentration63 would
have been depleted by 118 lb assuming a 50% water loss.

Chemical and electrical analysis methods are recommended to monitor

additive concentrations until an accurate determination of rinse water

56
recovery is performed. Analyses should be performed on a daily or
weekly basis. A conductivity bridge was built into the MacDill system
at the time of construction and was used to measure the concentration
of corrosion additives63. However, surfactant concentrations are usually
much lower, and not all surfactants are ionic in nature. Hence, a con-
ductivity bridge may not be an appropriate measuring device. It may be
possible to install in the system different probes for electrically con-
ductive additives and recalibrate it for lower concentrations. At last
report, (June 1984), the conductivity bridge was in a state of disrepair
and nonfunctional66.

Moreover, conductivity and other constituents may change in the
fresh water which is delivered by the Tampa Water Department. This
water is drawn mainly from the Hillsborough River, but, when the river
is low in the winter, the river water is supplimented by ground-fed
springs. The spring water is high in chloride leading to January-

March chloride concentrations which are an order of magnitude greater
than for the rest of the year76. Sample analyses of Tampa Water and
graphical plots of chloride, conductivity, total hardness, and pH are
shown in Tables 4 and 5 and in Figures 20, 21, 22, and 23, respectively.
Hence, a regularly scheduled chemical analysis of rinse water additives

is recommended.

V. Program

66

Summitt pr0poses an eXperimental program which will determine

whether fresh water rinsing is useful in corrosion control for aircraft
in severe environments. It will also determine whether the addition of
corrosion inhibitors to such rinse water is an effective corrosion con-

trol measure. The corrosion damage to selected airplanes at three air-

57

Table 497. Daily Tampa Hater Analysis Results for July 1984.

TAMPA HATER DEPARTMENT
DAILY QUALITY CONTROL LABORATORY ANALYSIS
LABORATORY I.D. NUMBER 5 54096
LOCATION ! HILLSBOROUBH RIVER TREATHENT PLANT
HATER SOURCE. TYPE 5 FINISHED/SURFACE
ANALYSIS COVERING THE HONTH OF JULY 1984

RESLLTS EXPRESSED IN HILLISRANS PER LITER UNLESS OTHERWISE INDICATED
ND I NOT DETERHINED

can mum MI. '5 “-055 51511. mm OIOIKS 55551515 75' m 5M?“ 5515 51.
'1“ 51553 1M5 .115“ “5 H55 5551. 5M ‘55 III! U715“. 1E5.

u.-N.-u-uN—-

3833§3u283

757“.

55553 “53 55553 55553 ' ”141515

4 55 162 7.75 63 144 15 23 365 fl .31 5 .1 3.3
5 155 165 7.75 65 145 25 21 355 27 .2 5 1.2 2.6
5 54 155 7.7 56 132 15 21 355 25.5 .15 5 .51 3.1
5 53 147 7.5 54 135 17 25 255 25 .41 5 .51 3.3
5 57 145 7.7 51 132 16 25 365 25.5 .17 5 .11 3.1
5 52 146 7.7 54 135 16 15 355 25 .15 5 5 3.3
5 I 145 7.7 5'7 135 15 21 325 26 .42 6 5 2.5
5 52 143 7.7 61 127 16 25 325 25.5 .33 5 .5 3.1
6 54 143 7.7 55 125 15 15 325 27 . 13 5 .2 2.5
6 52 147 7.5 65 132 15 21 335 27.5 .13 5 5 3.7
6 75 151 7.7 73 134 17 23 335 25 .11 5 .77 3.5
6 72 147 7.3 75 134 13 23 345 27.5 .1 5 .55 3.3
7 75 147 7.55 72 132 15 22 355 25 .1 5 .75 3.2
6 73 145 7.5 75 135 15 22 345 26 .44 5 .22 2.5
6 72 145 7.55 76 132 16 22 335 26.5 .35 5 .15 2.5
6 71 145 7.5 77 134 14 21 355 25 .14 5 .42 2.1
6 75 153 7.7 75 135 15 21 355 27.5 .55 5 5 3

6 77 152 7.5 75 135 14 21 375 27.5 .57 5 5 , 2.5
6 N 152 7.5 72 135 14 25 365 27.5 .57 5 .55 3.1
6 74 151 7.75 77 135 13 21 365 27 .55 5 .16 2.6
5 75 154 7.65 75 135 15 21 335 27 .35 5 .55 3.1
5 76 155 7.65 52 142 16 25 335 27 .1 5 5 2.7
5 51 161 7.65 U 146 15 15 355 27 .55 5 .25 2.2
6 75 155 7.7 51 142 17 25 365 27.5 .1 5 1.6 2.6
5 75 155 7.6 53 142 16 25 325 27 .1 5 .11 2.2
5 74 154 7.6 I 145 14 ' n 365 27.5 .55 5 .2 2.5
5 75 157 7.6 52 143 14 22 375 27 .N 5 5 2.5
5 72 145 7.6 77 135 14 23 355 26 .22 5 5 3.2
5 73 152 7.6 75 135 14 2'3 35 26 .32 5 1.43 3.3
5 75 153 7.65 74 135 I5 22 355 26.5 .55 5 .15 3

6 72 145 7.5 76 133 15 22 35 26.5 .11 5 .13 2.5

15.52

7 155 162 7.55 53 146 25 23 375 25.5 .44 5 1.6 3.7
4 71 143 7.3 51 127 13 15 2‘55 26 .07 5 5 2.1
5 55 151 7.67 71 136 15 21 344 27.3 .15 5 .32 3

ERNEST T .

WILDER 8 TECHNICAL SERVICES SUPERVISOR

5523

597. Daily Tampa Water Analysis Results for January 1985.

TAMPA WATER DEPARTMENT
DAILY QUALITY CONTROL LABORATORY ANALYSIS
LABORATORY I.D. NUMBER 3 54096
LOCATION 3 HILLSBOROUGH RIVER TREATMENT PLANT
WATER SOURCE. TYPE 3 SURFACE/FINISHED WATER
ANALYSIS COVERING THE MONTH OF JANUARY 1985

Table

RESULTS EXPRESSED IN MILLIGRAMS PER LITER UNLESS OTHERWISE INDICATED
NO I NOT DETERMINED

55! C155 Will" TOTE '41 “-055 CALCIIII M51011 OlOHIDES SPECIFIC TEN? 11155 1251170511 NIH C1.
UIIIS COC53 HORD UIIIS H555 HARD HARD 45 CL COH5 555 C II“ HFIIOOHL INCH HES

c0003 cIc03 :0c03 700:03 ,.00urs
I 3 I30 2I0 7.03 00 I70 40 I9 330 22 .23 0 0 2.0
2 I I31 200 7.9 I9 I72 20 73 370 2 .II o 0 3
3 I I30 203 7.0 73 I73 30 51 300 22.3 .I3 0 .7 3
I I I3: 200 7.7 77 I7I 34 90 390 22 .II 0 .02 2.0
3 3 I30 204 7.73 7I I72 32 00 3I0 II .3 o o , 2.9
I 3 I30 155 7.73 I0 I70 20 73 3I0 15.5 .14 o 0 2.0
7 I I32 155 7.0 II II0 30 I3 300 II.3 .I o 0 3.2
0 3 I3I 202 7.7 7I I7I 3I I0 3I0 15 .I 0 0 3.I
9 3 I30 I93 7.I3 I3 II7 20 I3 3Io 15 .12 0 0 3
I0 3 I33 III 7.03 I3 III 30 73 400 II .II 0 o 3.3
II I III 20: 7.7 70 III 32 03 320 I0 .09 o 0 3
I2 3 I29 213 7.73 54 175 33 93 I20 I7 .II 0 0 2.0
I3 3 I2I 2II 7.03 02 I77 34 I00 I30 II.3 .23 o o 2.0
II 3 I30 200 7.0 70 I72 3I 03 3I0 II.3 .2I 0 I2 3
I3 4 I32 242 7.73 II0 I92 30 140 730 II.3 .07 0 0 3.3
II I I32 216 7.7 54 III 40 I00 II0 I3.3 .07 0 0 3
I7 I I29 2I2 7.7 03 I7I 3a 90 3I0 II .0I 0 .3 3.2
II 3 I30 23I 7.7 I0I I92 II II0 700 17.5 .1 0 .I 2.0
II I I30 220 7.73 I0 I00 40 I00 030 I7 .27 0 0 2.7
20 ID sanPIE IIIIIIILE
2I 3 I33 231 7.73 I03 I00 40 II0 II0 I3.3 .09 o 0 3.2
22 I I3I 214 0.I3 03 150 34 II3 I00 II .2I 0 0 3.0
23 I I3I 220 7.0 92 I00 40 I20 730 II .II 0 0 3.3
24 3 I30 224 7.03 94 175 II IIO IIO I3.3 .I 0 0 2.9
23 3 I33 234 7.03 III I07 I7 I70 515 II.3 .II 0 .22 3.I
20 3 I30 200 7.9 70 II3 I3 II0 II0 II.3 .I9 0 0 2.7
27 3 I2I 202 7.9 73 II0 42 I20 I30 I3 .2 o 0 2.0
2! I I29 204 0.0I 73 IIO I4 I20 390 I3.3 .0I 0 .I3 3
29 3 I2I 227 7.7 90 152 Is II0 I00 II .I3 0 0 3.0
30 I 124 II2 7.3 I0 III 3I 00 330 II .II 0 0 3.3
II I I20 222 7.7 II I00 I2 I23 II0 I7.3 I 0 0 2.0
10151 ‘ 1.51
no: I I3I 234 I.I3 III I92 I7 I70 II0 22.3 .3 o .7 3.0
III 3 I24 I92 7.3 I3 II0 20 I3 400 II .56 o o 2.
0v; 4 I30 2I3 7.79 02 175 30 I02 I0I I7.3 .II 0 .56 3.I
I

ERNEST T. WILDER 3 TECHNICAL SERVICES SUPERVISOR

59

.cmuu: xu_u qumP cow mm:_m> mu_copgu gucoz cm» .Nmom mc:m_m

82.8 5% 3 r83 5:.

 

 

 

 

8 g 8

 

 

 

 

 

w. a w .r
as am_LTa wmrIaw .
[TF‘L b
h... a f
a I I am.
8. a_ a. I... as.“
a a a a
a. an ._na
mew
a a
a mu

 

 

 

 

 

«ES. :8 $2.: S... 8.33 83.3

8

8"

oz

9:

8—

30 I 80-35-40

( T/OW)

60

.coua: have ensue com mospm> pr>Puusu=ou uww_umam gaze: cop. .NmHN wc3m_m

naaq .om Shame o— raa_.~ L42»
6 o z o m < w

 

 

 

 

 

 

 

 

 

 

 

 

 

mBs_
ph~° cmzch sou »h_=~hu=azoo cum—ummm

AL I n I LOOONOO O I :5 I OBdS

(sqomouLaim)

61

.Lopm: pru mass» Lo$ mmzpm> mmwccgm: Pouch gucoz cab

82 .8 .55 8 ES ._ 3:.
z n. a o z o

,.I .; 5;. III; ‘. .

 

 

 

 

 

 

 

N

mNN mgamwu

 

 

 

 

 

 

 

 

 

 

 

 

 

 

L
o
L
U
.1
H
U
a
a
N
3
s
S
a __7 ff” 2 m
a 6
P I “a g x
a a a ...
E. E ..
E

q oz

__ on

an

“ES Eu $5 .3 8mg .52

62

.gmumz auwo «mam» Lop mm=_m> In cucoz amp .mmmm mg:m_g

82.3 .55 2 $2.— 5:.

< z h. . a

I’ll‘ll? Ill. . b

o .

z »

?

 

 

2

 

 

S
A;

 

 

 

 

 

.2

 

 

 

 

 

 

mg :3 is s“. was a

uIIIpbrPLrLubeIIhPIPLrbnulpPLrIII-a .0;
38-2. in .88: a
.¢I¢|IIIT u Iw‘ AVIw‘ leITIIIIIIIk¢g
can
IquJu : c o
a 3
a
3

SJ. INF! Hd

63
bases will be monitored for one year. Monitoring will include inspec-
tions, maintenance records, operational records, and, of course, the
MacDill rinse facility. Except for inspections, the program requires
virtually no participation of operational or field commands.

Three domestic air bases are chosen as sites for comparison:

- MacDill AFB, having a humid, salt water environment and a spor-
adically-used rinse facility.

- Hill AFB, Utah, also having a salt water environment, less severe,
but having no rinse facility and presumably no practice of fresh water
rinsing.

- Shaw AFB, South Carolina, having a less humid environment, no
nearby salt sources, no rinse facility, and no particular reason to rinse
airplanes.

While these environments differ in a variety of factors, it should
be possible to separate the effects of such factors and demonstrate
whether rinsing has any impact on corrosion, whether bad or good.

Selected airplanes, four at each airbase, are to be inspected at
the start, middle, and end of the program. Inspections will be made at
ZOO-hour phase inspections or at regularly scheduled USAF phase inspec-
tions. Four inspections are required. The inspection points on each
aircraft will number ten or less, pre-selected from analysis of main-
tenance data, previous corrosion control inspections, and materials
properties.

The MacDill facility program requires continuous on-site monitoring
of a variety of details, which include

- water meter readings

- chemical analysis of the rinse water

64
- additive replenishment as required
- usage of the facility by aircraft
- repair whenever inevitable breakdowns occur.
Summitt has carefully tailored the program to counter most of the
known experimental hazards:
(1) It will span the shortest feasible time, viz., one year.
(2) Personnel complications are minimized.
(3) Suitable controls are provided, including
(a) flight schedule monitoring via weekly and monthly schedules,
(b) maintenance monitoring via standard USAF monthly maintenance
reports,
(c) comparison with other bases,
(d) careful tracking of selected airplane serial numbers which
will not be revealed to airbase personnel,
(e) inspection of selected airplane serial numbers by an appro-
priate 4-person team at ZOO-hour Phase inspections.
VI. Discussion
Since no sound experimental evidence exists that supports or refutes
the merits of rinsing of aircraft as a corrosion control tool, the subject
continues to be controversial. Obviously, advocates believe that it is
effective and well worth the cost, while detractors argue that it does
more harm than good.

66 will demonstrate the effects of

The program outlined by Summitt
rinsing, whether positive or negative, as well as the value of adding
corrosion inhibitors to the rinse. The costs and time span of the pro-
gram are minimal, and it will have no impact on MacDill operations. If

the program is a success, Air Force Logistivs Command and others will

65
have the information needed in order to make decision of whether the
rinsing of aircraft is an effective way to reduce corrosion damage in

terms of cost and reliability.

INHIBITORS

I. Introduction

There are many systems in which the use of corrosion inhibitors is
the most economical and effective approach to controlling corrosion. A
substantial number of different types of corrosion inhibitors are used
in various applications. In selecting corrosion inhibitors for a spec-
ific problem, the formulator should have a basic understanding of the
mechanism of the corrosion process as well as some qualitative concepts
as to how corrosion inhibitors might function in a particular applica-
tion.

In some cases, the mechanism of inhibition is poorly understood,
but in many cases, we have at least a qualitative understanding of how
the inhibitor functions.

In considering the inhibition of the corrosion of aluminum and
aluminum alloys by aggressive ions, the starting point is to examine
the steps by which aggressive ions act on aluminum. These steps have
been established83:

(1) Adsorption of halide (e.g. Cl”) on the aluminum oxide surface,

(2) Complexing of aluminum cation in the oxide lattice with halide
to form a soluble AlCl; species,

(3) Soluble species diffuse away from the surface resulting in a
thinning of the protective oxide film,

(4) At sufficiently thinned sites, the aluminum acts directly
with the electrolyte.

A compound pr0posed as an inhibitor may be involved in the mechanism
in either of the first two steps. The compound may compete for adsorp-
tion sites and retard the formation of the soluble halide species.

66

67

Secondly, the compound, as its anion, can compete with Cl' in the Al+3
+ 4CI' = AICIZ equilibrium reaction again preventing the formation of
the soluble halide species. But, in the latter, if an organic anion
forms a stable soluble complex ion with the aluminum cation, then dis-
solution will proceed just as it would with the formation of the alumi-
num halide species.

The purpose of this paper is to provide a broad overview of many
different types of corrosion inhibitors used commercially. These
materials are classified by their mechanism of their inhibitive action.
This approach is problematic with regards to materials with mechanisms
not completely understood. However, such materials are relatively few,
and should not cause serious problems.

Four general catagories of corrosion inhibitors have been identi-
fied; barrier layer formers, neutralizers, scavengers, and miscellaneous.
II. Barrier Layer Formers

Corrosion inhibitors, which form barrier layers on the corroding
metal surface compose the largest category of corrosion inhibitors.
These materials deposit on the metal surface and interfere with the
corrosion reaction; hence lowering the rate to an acceptable value.

This type of corrosion inhibitor has received the most attention from
the technical community and is generally the most widely used. There
are a number of corrosion inhibitors in this category.

(a) Adsorbed Layer Formers

The corrosion reaction can be inhibited by materials which adsorb
strongly the metal surface and interfere with either the cathodic and/or
anodic reactions occuring at the adsorption site. If the adsorption

site is relatively complete and the surface coverage is total, then the

68
corrosion reaction can be reduced by many orders of magnitude. Materials
such as the acetylenic alcohols have been shown to function this way on
d84.

steel in hydrochloric aci Other organic nitrogen compounds, such as

quinoline, some aromatic amines, and some quaternary salts, are also ef-

85' 86. Organic sulfur compounds. SUCh

fective in strong acid solutions
as thiourea are also effective, as are other organic compounds containing
periodic table group 5A or 6A elements. Inorganic compounds, such as
sulfides, arsenic, antomony, and halides also function as adsorbed layer
inhibitors in strong acid solutions.

Most strong acid inhibitors interfere with the rate limiting step
in the dissolution process. Usually the most effective inhibitors inter-
fere with both anodic and cathodic reactions, so there is not a marked
shift in corrosion potential accompanying the process. Acid inhibition
is in a number of areas, including oil well acidizing, metal pickling,
and acid cleaning. Table 6 provides a listing of some strong acid in-
hibitors.

In neutral pH solutions, adsorbed layer inhibitors are also avail-
able, although their use is not widespread. Under these conditions, the
primary cathodic reaction usually is the reduction of dissolved oxygen,
rather than reduction of hydrogen ions, as is the case for strong acids.
The potential driving force for this reduction of dissolved oxygen re-
action is much greater than that for the reduction of hydrogen ions.
Hence, inhibition through simple adsorption is more difficult. Again,
nitrogen containing organics, such as benzotriazole and related com-
pounds, are effective for c0pper alloys and, to a limited extent, steel.

Other nitrogen, sulfur, and oxygen-containing organics, such as sodium

benzoate, sodium acetate, sodium oxalate, azelaic acid, benzotriazole,

69

87, 93

sodium mercaptobenzothiazole, and others are also effective.

There are a few inorganic cations such as zinc, and to a lesser extent
cadmium, which are effective as cathodic inhibitors in neutral and
slightly acid solution588. Examples of adsorbed layer inhibitors are
given in Table 7.

The effectiveness of these adsorbed layer inhibitors is usually
evaluated in terms of how much they reduce the overall corrosion re-

actions. This reduction is expressed as percent inhibition, with 100%

being no inhibition. The percent inhibition is calculated as follows:

% inhibition = 100 (R - R'),
R

 

Where R corrosion rate without inhibitor and R' corrosion rate with in-
hibitor.

These inhibitors are also classified according to how they affect
the component electrochemical reactions which make up the overall cor-
rosion process. Figures 24-26 show electrochemical polarization diagrams
for cathodic, anodic, and mixed inhibitors. In general, if the inhibi-
tion process is accompanied by a shift in the corrosion potential in the
negative direction, then the inhibitor is a cathodic inhibitor. If the
shift in potential is in a positive direction, then the inhibitor is
anodic. If there is no or very little shift, both reactions are being
inhibited, then the inhibitor is mixed.

(b) Oxidizing Inhibitors-Passivators

Passivators are another type of barrier layer former. They are use-
ful in aqueous solutions within the neutral range. These inhibitors fun-
ction by shifting the electrochemical potential of the corroding metal

into a region where a stable, insoluble oxide or hydroxide forms which

70

Table 6. Corrosion Inhibitors for Steel in Strong Acids

CLASS

Acetylenic Alcohol

Heterocyclic amine
Fatty Amine

Sulfur Compound

Arsenic Compound

2W
Propargyl
Hexynol
Hexynol

Ethyl Octynol
Quinoline
Rosin Amine
Thiourea
Sulfonated Oil

Arsenic Trioxide

FORMULA
HC = C'CHon
HC - C-CHOCHC3H7

HC

C-CHOH-C7H15
C6H4N = CHCH=CH
(NH2)2cs

AsO

REFERENCE

84

84

84

89

85

85

89

85

Table 7.

METAL

Copper Alloys

Steel

Aluminum

71

Adsorbed Layer Forming Inhibitors for Nonacid Systems

ACTIVE ATOM

 

Nitrogen

Nitrogen

Nitrogen

EXAMPLE OF INHIBITOR

 

Benzotriazole

Tolytriazole

Sodium Mercaptobenzathiazole

Sodium Mercaptobenzathiazole
Imidazoline

Fatty Amines

Benzotriazole
Cupferron
Oxamide
Rubeanic Acid
Quinaldic Acid

Sodium Mercaptobenzathiazole

72

\S‘Fco -° F.“ f 20

L E

 

 

 

 

 

 

2 ‘\\~ “’
Eéx.\
\
\ {loft-i. *XH;
\
\
\
\
\
\
Q
. .u’ ~xw§ mum
'éx ‘ax
bgl

Figure 2494. Schematic polarization diagram illustrating cathodic

inhibitor in an acid solution. Note the negative

corrosion potential shift.

73

 

 

 

 

 

 

. l
./
/‘$F.° -0 F0” 0 20 Inhibited
/'
650' / Foo ~ F.” f 20
/ 6
71
E” /
/'
5 /
/'
/’
/
o . H’ .. XH;
_ I.oor |cor
log l
Figure 2594. Schematic polarization diagram illustrating anodic

inhibitor in an acid solution. Note the positive

potential shift.

74

I
I F°° " F0“ ‘ 2o mauled
/‘ )

 

 

‘\
\ I
\
5&3.- / \ ' l'-'o°'°FoH +2.
I \
/ \ é
Ear [1 \
/
O E //’ \S. (:2
/ \\ . § H *KHZ
\

 

 

 

 

bal
Figure 2694. Schematic polarization diagram illustrating a mixed
inhibitor in an acid solution. Note the minimal

potential shift relative to the degree of inhibition.

75
protects the metal surface. This type of inhibitor is especially effec-
tive on steels, although it is also effective on aluminum and copper
alloys and certain other alloy systems.

The major advantages of this type of inhibitor are that they are
relatively inexpensive, effective at low concentrations, and reduce the
corrosion rate to very low values, (e.g. less than 411m/y for steel87).
Figure 27 is a polarization diagram for a steel in an aqueous system
showing the anodic polarization curve with the devel0pment of passivity.
An oxidizing inhibitor such as sodium nitrite shifts the cathodic polar-
ization curve in the positive direction, thus stabilizing the corrosion
potential in the passive region. Chromates have become relatively un-
available as corrosion inhibitors because of environmental problems.
Nitrites also have become less p0pular due to the formation of nitro-
samines, which are environmentally unacceptable. Consequently, there
has been renewed interest in other types of oxidizing inhibitors such
as molybdates and tungstates. These materials are not as strongly
oxidizing and are of substantially higher cost. Table 8 shows a com-
pilation of the electrochemical reduction potential of various potential
passivating type inhibitors.

(c) Conversion Layer Formers

The passivating inhibitors rely on the development of an insoluble
metal oxide of hydroxide which forms on the metal surface. This layer
forms a barrier to the corrosion process. Another approach which can
be used is the addition of materials to the environment which form in-
soluble compounds on the metal surface without relying on the oxidation
requirement of the passivating inhibitors.

Phosphates are probably the most widely used type of material in

76

e :3
3.420 . ZFCO

 

 

" F0203 . 6H. 9 6.
\
\
£50! \ \
\\
\ °‘ ”"20 mo: 4.: m-
inhibitor
reaction
E
\
I
I
i
Eoor H 0 20

 

 

‘0 . 02 ’ 2H20 *‘OH.

 

 

- It» .0.

knl

 

Figure 2794. Schematic polarization diagram for nitrite inhibition

of steel - a passivating inhibitor.

77
this category. They are used extensively in cooling water inhibitors.
The principle of their operation is that they form an insoluble iron
phosphate on the metal surface. This layer builds up to sufficient
thickness to provide an apparent passivity. Because ferrous phosphate
does have some solubility in water, the corrosion reaction is not
totally eliminated, rather it is reduced to a manageable degree.

Silicates are also used in some applications, eSpecially for
aluminum alloys. Ferrous sulfate has been used to inhibit copper alloys
in heat exchangers using natural water as a cooling medium. In this
case, the ferrous sulfate reacts with oxygen on the copper alloy sur-
face to form a film of FeOOH which in turn provides resistance to cor-
rosiongo. Other examples of conversion layer formers are given in
Table 9.

It should be noted that the conversion type inhibitors are aided
frequently by the presence of divalent cations, such as zinc, calcium,
and magnesium. These cations tend to supress the solubility of the
iron compounds as well as interfering with the anodic reduction of
oxygen on the metal surface.

In neutral solutions, the presence of calcium and magnesium ions
also inhibit corrosion by the formation of an insoluble calcareous

scale on the metal surface. The controlling reactions are:
4e + 02 + 2H20 = 40H" cathodic reaction
Ca+2 + HCOS + OH‘ = CaCO3 + H20 scaling reaction

The build up of calcareous scale effectively blocks diffusion of

oxygen to the cathode areas and gradually reduces the corrosion rate.

78

Table 812. Electrochemical Potential of Some Oxidizing Inhibitors

Type of Inhibitor

 

Chromate

Nitrite

Nitrate

Molybdate

Tungstate

Pertechnate

R

m

10H+ + 2Cr;2 = Cr203-6e

ZNO' + 8H+

2

2N0; + 12H

-2
M004 + 4H

NO'2 + 4H+

4

TcO4 + 2H2

= N2 + 4H20'69

+

N2 + 6H20 -10e
T = MoO + 2H 0 2

2 2 ' e
= ”02 + 2H2-2e

o = Tco2 + 4H+ - 3e

5° vs SHE

1.3-0.0985 pH
1.52-0.079 pH
1.25-0.071 pH
O.606-O.l18 pH
O.386-O.1182 pH

0.738-0.019 pH

METAL

Steel

Copper

Aluminum

Zinc

Table 95. Conversion Layer Formers

CLASS

PhOSphate
Silicate
Zinc PhOSphate
Chromate

Ferricyanide

Chromate
Phosphate

Ferrous Sulfate

Silicate

Phosphate

Chromate

Phosphate

79

W

N32P04
Na25103

ZnHPO4

Nazcr04

NazFe(CN)6

NaZCrO4
Na3PO4

FeSO4
Na25i03

NaZHPO4

NaZCrO4

NazHPO4

.LALEB.

FeHPO4
FESIO3
(ZnFe)HPO4
Cr203 Fe203

Fe Fe(CN)

3( 6)2
CuO, Cr203
CuO, CuHPO4

FeOOH

Al2(SiO3)3

AIPO4

ZnCrO4

ZnHPO4

80
This mechanism occurs frequently in hard well water and sea water.
Hence other inhibitors are sometimes not required.

In general, the conversion type inhibitors are less effective and
more prone to control problems than the passivating inhibitors. How-
ever, they are becoming more attractive because of their nontoxic char-
acteristics.

III. Neutralizing Inhibitors
In many systems the corrosion process is driven by the reduction

of hydrogen ions to hydrogen gas.
H+ + e = 1/2 H2 (9).

In neutral water at ambient temperatures, the concentration of
hydrogen ions is so low, especially compared with the concentration
of dissolved oxygen, that the reaction is not usually considered sig-
nificant. However, chemical processes can occur locally which liberate
hydrogen ions and these can contribute to the corrosion processgl. At
higher temperatures, the diffusivity and frequently the concentration
of hydrogen ions increase. This can become a significant factor. As
a result, a class of inhibitors has been developed whose primary function
is to reduce the hydrogen ion concentration in the environment. A list
of different types of neutralizing inhibitors is given in Table 10.

This approach has been especially useful in treatment of boiler
waters. In this case, a weak acid such as carbonic acid, which is not
normally a problem at ambient temperatures, becomes extremely corrosive.

As a result, materials such as morpholene, cyclohexylamine, and ammonia

are added to boiler waters in order to suppress the free hydrogen ions

81

according to the reaction similar to the one shown below:

R3 + H+ = R3NH+.
These materials form weak bases, but do not form high concentrations of
hydroxyl ions which can be corrosive and subject steel and other metals
to caustic cracking. In addition, these materials are volatile and can
be carried in the steam to prevent carbonic acid attack in the conden-
sate.

Neutralizing inhibitors are used mainly in oil field applications.
Crude petroleum frequently contains brine and hydrogen sulfide. The
brine is extremely corrosive to carbon steel. Above ground disposal
of brine usually is not possible because of environmental restrictions.
Its rejection into the producting reservoir serves to stimulate further
production. In order to control the corrosion which accompanies hand-
ling this material, it has been found that amines are effective. These
materials react with the hydrogen sulfide to remove the hydrogen ion
content as well as acting as an adsorbed layer inhibitor, as discussed
earlier.

Materials such as sodium carbonate, ammonia, and sodium hydroxide
are commonly used in both the chemical process and the petroleum pro-
cess industries to control corrosion. Small quantities of acidic
materials such as hydrogen chloride, carboxylic acid, carbon dioxide,
and acidic phenols and related compounds are present in many process
streams. However, in separation processes such as distillation, these
acidic materials can concentrate in specific areas and cause severe
localized corrosion. The addition of materials such as soda ash and

ammonia tend to remove the hydrogen ions which cause the corrosion

82

Table 10. Neutralizing Inhibitors

 

 

APPLICATION INHIBITOR CHARACTERISTIC
Boiler Water Ammonia Volatile
Morpholene Volatile
Cyclohexylamine Volatile
Sour Hater Polyamines
Crude Oil Production Alkyl Amines

Fatty Amines

Petroleum Refining Na2C03 Nonvolatile
NH3 Volatile
NaOH Strong Base

Ethylene Glycol Coolant NazB4O7°10HZO Buffer

83
process as well as changing the volatility of the acidic materials,
thereby preventing localized accumulation within the separation unit.

Hashing with a dilute sodium carbonate solution is also used to
remove sulfur compounds in petroleum refinery equipment during shut-
down589. These sulfur compounds would otherwise react with oxygen and
fonn polythionic acid, which leads to rapid intergranular cracking of
sensitized stainless steel equipment.

Neutralizing inhibitors are also used in ethylene glycol cooling
systems to prevent the develOpment of acidic conditions within such
systems. Ethyene glycol can be oxidized to oxalic acid which is ex-
tremely corrosive to many alloys. The use of buffers such as sodium
borate, sodium phosphate, and to some extent, sodium silicate can
delay the onset of corrosion from oxalic acid by reacting with the
hydrogen ions and thereby acting as a buffer.

IV. Scavengers

The neutralizing inhibitors described earlier are added to re-
move or react with hydrogen ions. However, in many systems, there
are other corrosive materials present in small concentrations which
cause equally severe problems. Inhibitors which remove such chemicals
must be tailored to the particular corrosion process which is occuring
in the system of interest.

Perhaps the most widely used scavenger system is employed in boilers
to remove oxygen from the feedwater. Typical techniques such as steam
stripping can remove the bulk of dissolved oxygen from water. However,
such methods become increasingly costly when the last traces of oxygen
must be removed. In these cases, chemical techniques for oxygen re-

moval become much more attractive. The two most widely used scavengers

84
in boiler systems are hydrazine and sodium sulfite. Chemical reactions

which occur in both these cases are shown below:
2(H2NNH2) + 1/202 = 2NH3 + H20 + N2
-2 _ -2
$03 + 1/202 - SO4 .

The hydrazine process is the cleaner reaction. The reaction pro-
ducts are volatile and useful in terms of controlling the pH; however,
hydrazine is more expensive than sulfite and its use is restricted to
high pressure boiler applications. Hydrazine is also considered to be
a carcinogen. Both hydrazine and sulfite require the addition of cata-
lysts to the system in order to make these chemicals effective, in the
framework of residence times, within most commercial boilers. As a
result, pr0prietary formulations employing both neutralizing inhibitors
and oxygen scavengers are used to treat boiler waters.

Another class of scavenging inhibitors is used to handle organic
compounds which tend to break down to form acidic decomposition pro-
ducts. Chlorinated compounds such as 1-1-1 trichloroethane have been
widely used as cleaning solvents in both the dry cleaning and metal
finishing industries. These compounds can decompose in water and high
temperatures to form hydrochloric acid, which is corrosive to steel,
aluminum, and other structural materials. This problem is handled by
adding small concentrations of inhibitors which either react with hy-
drogen chloride or which interfere with the decomposition process.
Volatile amines have been used to react with hydrochloric acid.
Chemicals such as dioxane and methyl butynol have been found to be
effective in preventing the decomposition from occuring.

Another approach which has been used to scavenge small concen-

85
trations of hydrochloric acid in nonaqueous systems has been to add
ethylene oxide. This chemical reacts with hydrochloric acid as shown

below:
/0\ _ - -

The ethyene chlorohydrin resulting from this reaction is rela-
tively nonvolitile and is separated from the process. A variety of
other scavengers are used frequently in petroleum products, such as
high performance lubricating oils, greases, and related compounds.
Materials such as hydroquinone are effective in blocking the peroxide-
free radical mechanism which, if left uninhibited, would result in the
formation of carboxylic acids which are corrosive.

V. Miscellaneous

There are a variety of other inhibitors which have been developed
to interfere with processes that ultimately result in corrosion. In
many cases, these materials act not only as corrosion inhibitors, but
provide other desirable benefits as well. An example of this is the
biocides used in cooling water applications. There are a variety of
bacteria that grow in aqueous systems which promote localized corrosion
of materials. The sulfate-reducing bacteria are probably the most
widely known but there are many others as well. The use of biocides,

92

such as quaternary ammonium compounds , can control both the cor-

rosion and the fouling which result from excessive biological growth.
Another type of inhibitor which is used in cooling water applica-
tions is the scale inhibitorgz. These materials are added to the system

to prevent growth of deposits on heat transfer surfaces caused by pre-

cipitation of insoluble species such as calcium carbonate. These

86
materials function by interfering with the normal crystal growth and
thereby form a soft, nonadherent precipitate in solution rather than
on the metal surface.

The prevention of deposits on the metal surfaces minimizes the
problems from deposit corrosion or poultice corrosion. This type of
corrosion is a form of crevice corrosion which can result in pitting
and perforation. Deposits can also cause localized overheating re-
sulting in corrosion. Materials which are used for this purpose are
typically phosphonates, gluconates, and polyacrylic acids.

Chelating agents such as ethylene diamine tetra-acetic acid (EOTA)
are used for preventing scale formation. It should be noted, however,
that chelating agents can increase the corrosion rate of many alloys
by increasing the solubility of the protective corrosion products.
These materials are generally more expensive to use than the crystal
growth modifiers.

There is a final group of inhibitors whose detailed mechanism is
not completely understood, however, they are used in specific applica-
tions and are very effective. For example, water is added to commer-
cial liquid ammonia to inhibit the stress corrosion cracking of carbon
steel containment vesselsgB.

It has been found that 0.2% by weight water addition to liquid
ammonia practically eliminates the risk of stress corrosion cracking
in structural steels. The function of the water in this case is not
well understood; however, it appears to act as a passivating inhibitor
and encourages the formation of a more protective corrosion product

on the steel. Similarly, it has been found that anhydrous ethylene

glycol is corrosive to aluminum under certain conditions; the addition

CI)

7
of a small quantity of water (e.g., 1%) inhibits the corrosion. In
this case, the water seems to interfere with the tendency of aluminum
to react directly with ethylene glycol to form a somewhat soluble
Grignard-type compound. The water may also encourage develOpment of
a more protective corrosion product film on the aluminum.
VI. MacDill AFB Rinse Facility Inhibitors

The bulk inhibitor mixture used for the MacDill rinse facility as

prepared by Erny Supply Co.73 is:

50.63% Sodium Borate (% by weight)
31.5% Sodium Nitrate

15.6% Sodium Nitrite

1.60% Sodium Metasilicate Pentahydrate
0.33% Sodium Meta-hexa Phosphate

0.14% Sodium 2-Mercaptobenzothiazole (MBT)

60 62, 63

In immersion tests and polarization measurements , this mixture
has been shown to be an effective inhibitor for aluminum alloys in
aqueous solutions containing sodium chloride.
In this section each component of this inhibitor mixture is reviewed.
(a) Sodium Borate - The borate ion is considered to be a passivator,
that is, it tends to promote the formation of a protective oxide layer on
the surface of the metal. The interesting aspect that makes borax some-
what unique is the fact that it is a weak base, hence a non-oxidizer.
Uhlig88 states that in order for borax and other non-oxidizing passivators
to be effective, dissolved oxygen must be present in the media. Being

alkaline, the borate ion will tend to displace H adsorbed on the surface

of the metal, thus decreasing the probability that dissolved 02 will re-

88
act with adsorbed H. Such an occurence will increase the concentration
of 02 in solution while the surface H concentration decreases as it re-
acts with the borate ion. As a result, the excess 02 replaces the H at
the cathodic sites and is itself reduced and adsorbed on the metal sur-
face. This displacement reaction then gives rise to the formation of a
passive metal oxide layer.

(b) Sodium Nitrate and Sodium Nitrite - Both of these compounds are
considered to be oxidizing passivators, which are very effective in
aqueous systems that contain chloride88. These compounds, which are
readily reduced, tend to shift the cathodic polarization curve in the
positive direction, thus promoting the formation of a stable, insoluble
metal oxide on the metal surface (Figure 27).

Uhlig88 questions whether nitrate will be of much use as a passiv-
ator in the presence of nitrite, since nitrite reduces much more rapidly
than the nitrate ion. This point is well taken and we can only speculate
that, since the nitrate is present in higher concentrations, the aggres-
sivity of the more oxidizing species (nitrite) must be tempered to allow
the formation of conversion layers.

(c) Sodium Metasilicate Pentahydrate - This material tends to form
insoluble layers on the metal surface, which will cause passivity when a
sufficient thickness is achieved. Such a compound is classified as a
"barrier layer former."

Silicates are known to be eSpecially effective on aluminum alloys.
The insoluble layer formed on an aluminum surface is Al2(SiO3)3 (Table 9).

(d) Sodium Hexa-meta Phosphate - This material is another example of
a barrier layer former. The phosphate ion reacts with the aluminum at the

surface to form an insoluble A1P04 conversion coating (Table 9).

89

(e) Sodium Z-Mercaptobenzothiazole (MBT) - This compound tends to
adsorb onto the metal surface and interfere with the cathodic reaction.
Hence this compound can be described as an adsorbed layer former which
inhibits the reduction of H+ at the cathode.

MBT is considered to be a surfactant because it is organic in nature
and surface active. Research is currently being conducted at Michigan
State University to investigate the effects of other organic surfactants
on the inhibition of aluminum.

In summary, we have seen that the inhibitor mixture used in the Mac-
Dill rinse facility contains 4 different types of inhibitors:

l. Non-oxidizing Passivator - Sodium Borate.

2. Oxidizing Passivators - Sodium Nitrate and Sodium Nitrite.

3. Barrier Layer Formers - Sodium Metasilicate Pentahydrate and
Sodium Hexa-meta Phosphate.

4. Adsorbed Layer Former - MBT.

VII. Discussion

Corrosion inhibitors fall into two general categories: first,
those which act in the solution to remove small quantities of cor-
rosion constituents; secondly, those which act on the metal surface
to improve its corrosion resistance. Both types are used in specific
appications and are effective. Tailoring a corrosion inhibitor to
combat a specific corrosion problem usually requires a detailed know-
ledge of the corrosion process and creative thinking on the part of
corrosion engineers to devel0p a cost effective solution. In general
the use of neutralizing and scavenging type inhibitors seem to be best
suited to closed systems where chemicals are not lost in the systems.

In open systems where the process equipment sees large thruputs on a

90
once-through basis, the inhibitor must be effective at low concentra-
tions and inexpensive.

The classification system preposed in this section for inhibitors
is summarized in Table 11. The classification of inhibitor types by
function gives a fairly simple and concise approach, although it has
limitations in cases where mechanisms are not known. Others have
classified inhibitors as organic vs. inorganic, solution vs. vapor
phase, anodic vs. cathodic, and functional groups.

VIII. Summary

1. Corrosion inhibitors may be classified as either barrier
layer formers, neutralizing inhibitors, scavengers, or miscellaneous.

2. The barrier layer formers are the largest class of corrosion
inhibitors. They include the adsorbed layer inhibitors, oxidizing
inhibitors, and conversion layer formers.

3. The neutralizing and scavenger type inhibitors operate on
the environment rather than on the metal surface by removing the cor-
rosive species from the environment. As such, they represent a dif-

ferent approach to the problem of corrosion inhibition.

l.

2.

3.

91

Table 11. Classification System of Inhibitors

DESCRIPTION

 

Barrier Layer Formers

1a.

1b.

1c.

Adsorbed Layer Formers
Iai. Cathodic Inhibitors
laii. Anodic Inhibitors

laiii. Mixed Reactors

Oxidizing Inhibitors-Passivators

Conversion Layer Formers
1ci. Insoluble Corrosion PPOdUCtS'

1cii. Cathodic Deposits

Neutralizing Inhibitors

2a. Volatile Neutralizers
2b. Nonvolatile

Scavengers

3a. Oxygen Scavengers

3b. Decomposition Inhibitors

Miscellaneous

4a.

4b.

4c.

Biological Growth Inhibitors
Scale Inhibitors

Other

EXAMPLES

Acetylenic Alcohol in HCl

NaNOz in Water

Phosphate
C3C03

Cyclohexylamine in Boilers

Amines in Brine

Na2503 in Boilers

Dioxane in CH3CCl3

Quaternary Amines
Phosphonate

H20 in NH3

92

References

1.

2.

3.

5.

7.

8.

9.

IO.
11.

12.

13.

R. Summitt and R. T. Fink, "PACER LIME: An Environmental Corrosion
Severity Classification System,“ Publication AFWAL-TR-80-4102 Part

I Air Force Wright Aeronautical Laboratories, Wright-Patterson AFB,
OH, 1980.

N. D. Tomashov, “Theory of Corrosion and Protection of Metals,"
Translated and edited by E. H. Tytell, I. Geld, and H. S. Preiser,
H. S. McMillan, New York, 1966.

W. H. Ailor, "ASTM Atmospheric Corrosion Testing: 1906 to 1976,“
"Atmospheric Factors Affecting the Corrosion of Engineering Metals,
ASTM STP 646,“ S. K. Coburn, Ed., American Society for Testing and
Materials, 129-151, 1978.

I. L. Rozenfeld, "Atmospheric Corrosion of Metals,“ B. H. Tytell,
transl., National Association for Corrosion Engineers, Houston, TX,
1972.

P. J. Sereda, "Weather Factors Affecting Corrosion of Metals,“
"Corrosion in Natural Environments, ASTM STP 558," American Society
for Testing and Materials, 7-22, 1974.

W. H. J. Vernon, Trans. Faraday Soc. 21, 265-277, (1931).
W. H. J. Vernon, Trans. Faraday Soc. 31, 1668-1700, (1935).

P. R. Grossman, "Investigation of Atmospheric Exposure Factors That
Determine Time-of-Wetness of Outdoor Structures," "Atmospheric
Factors Affecting the Corrosion of Engineering Metals, ASTM STP
646," S. K. Coburn, Ed., American Society for Testing and Materials,
1978.

U.S. Department of Health, Education and Welfare, "Air Quality
Criteria for Hydrocarbons," March 1970, National Air Pollution Con-
trol Administration Publication AP-64.

T. R. Oke, “Boundary Layer Climates," Methuen, London, 1978.

H. Guttman, and P. J. Sereda, "Measurement of Atmospheric Factors
Affecting the Corrosion of Metals," "Metal Corrosion in the Atmos-
phere, ASTM STP 435," American Society for Testing and Materials,
326-359, 1968.

U.S. Department of Health, Education and Welfare, "Air Quality

Criteria for Particulate Mater," January 1969, National Air Pol-
lution Control Administration Publication AP-49.

U.S. Department of Health, Education and Welfare, "Air Quality
Criteria for Sulfur Oxides," January 1969, National Air Pollution
Control Administration Publication AP-SO.

93

References con't

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

N. G. Vannerberg, T. Syndberger, "Reaction Between $02 and Wet
Metal Surfaces," Corrosion Science, 19, 43, 1970.

D. J. Spedding, "Kinetic Study of the Uptake of Sulfur Dioxide by
Aluminum,“ British Corrosion Journal, 1, 281, 1972.

P. Urone, "The Primary Air Pollutants-Gaseous; Their Occurance,
Sources and Effects," 3rd ed., A. C. Stern, ed., Air Pollution,
Vol. 1, Academic Press, New York, 1977.

N. McKinney, H. W. Hernance, “Stress Corrosion Cracking Rates of a
Nickel-Brass Alloy Under Applied Potential," Stress Corrosion
Testing, ASTM STP 425, p. 274, 1967.

H. W. Hernance, C. A. Russell, E. J. Bauer, T. F. Egan and H. V.
Wadlow, "Relation of Airborne Nitrate to Telephone Equipment
Damage," Environmental Science Technoloby, 5, 9, p. 781, 1971.

J. Gerhard and F. H. Haynie, "Air Pollution Effects on Catastrophic
Failure on Metals,“ EPA Report 650/3-74-009, Nov, 1974.

F. H. Haynie, and J. B. Upham, “Correlation Between Corrosion Behav-
ior of Steel and Atmospheric Data," "Corrosion in Natural Environ-
ments, ASTM STP 558, American Society for Testing and Materials,
33-51, 1974.

H. S. Stoker, and S. L. Seager, "Environmental Chemistry: Air and
Water Pollution," 2nd edn., Scott, Foresman and Co., Glenview, IL,
1976.

U.S. Environmental Protection Agency, "National Air Quality and
Emissions Trend Report," 1976, EPA-450/1-77-002.

U.S. Department of Health, Education and Welfare, "Air Quality
Criteria for Photochemical Oxidants," March 1970, National Air
Pollution Control Administration Publication AP-63.

U.S. Environmental Protection Agency, "Air Quality Criteria for
Nitrogen Oxides," January 1971, National Air Pollution Control
Administration Publication AP-84.

T. E. Graedel, and N. Schwartz, "Air Quality Reference Data for

Corrosion Assessment," Materials Performance 16, 17-25, August,
1977.

R. Summitt, "Corrosion Tracking and Prediction for C-141A Aircraft
Maintenance Scheduling," AFML-TR-78-29, Air Force Materials Labora-
tory, Wright-Patterson AFB, OH, 1978.

R. Summitt, "Forecasting Corrosion Damage and Maintenance Costs for
Systems of Large Aircraft: Environmental Corrosiveness," CORROSION/
80, Chicago, March 6-10, 1980.

94

References con't

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

ASTM, “Corrosiveness of Various Atmospheric Test Sites as Measured
by Specimens of Steel and Zinc,“ “Metal Corrosion in the Atmosphere,
ASTM STP 435," American Society for Testing and Materials, 360-391,
1968.

W. H. Ailor, "Seven-Year Exposure at Point Reyes, California,"
"Corrosion in Natural Environments, ASTM STP 558," American Society
for Testing and Materials, 75-81, 1974.

T. R. Shaw, “Corrosion Maps of the British Isles,“ "Atmospheric
Factors Affecting the Corrosion of Engineering Metals, ASTM STP
646,“ S. K. Coburn, Ed., American Society fOr Testing and Materials,
204-215, 1978.

J. C. Hudson, and J. F. Stanner, J. Iron and Steel Inst. 180, 271
(1955).

C. E. Junge, and Gustafson, “0n the Distribution of Sea Salt Over
the United States and Removal by Precipitation," Tellus g, 164 (1957).

C. E. Junge, "Air Chemistry and Radioactivity,“ Academic Press, New
York, 1963.

R. A. Duce, J. W. Winchester, and T. W. Van Nahl, “Iodine, Bromine,
and Chlorine in the Hawaiian Marine Atmosphere," J. Geophysical Res.
.19, 1775 (1965).

H. H. G. Jellinek, “Chain Scission of Polymers by Small Concentrations
(1 to 5 ppm) of Sulfur Dioxide and Nitrogen Dioxide, Respectively, in
the Presence of Air and Near Ultraviolet Light," J. Air Pollution Con-
trol Assoc. 20, 672 (1970).

United States Air Force, “Exterior Finishes, Insignia, and Markings
Applicable to USAF Aircraft,“ T. 0. 1-1-4, 1973.

B. Cohen, private communication with R. Summitt, 1976. See reference
1.

R. C. Weast, ed., “Handbook of Chemistry and Physics,“ Slst edn., CRC
Press, W. Palm Beach, FL, E-5, 1971.

W. H. Ailor, Jr., "Performance of Aluminum Alloys at Other Test
Sites," "Metal Corrosion in the Atmosphere ASTM STP 435," American
Society for Testing and Materials, 285-307, 1968.

L. Atteras and S. Haagenrud, "Atmospheric Corrosion in Norway,“

Atmos heric Corrosion, W. H. Ailor, ed., J. Wiley and Sons, New
YorE, 873-891, 1982.

95

References con't

 

41.

42.

43.

44.

45.

46.

47.
48.

49.

SO.

51.

52.

53.

54.

55.

D. Knotkova-Cermakova and K. Barton, "Corrosion Aggresivity of
Atmospheres (derivation and classification),“ Atmospheric Corrosion
of Metals, ASTM STP 767, S. W. Dean, Jr., and E. C. Rhea, Eds.,
American Society for Testing and Materials, 1982, 225-249.

C. McCaul and S. Goldspiel, “Atmospheric Corrosion of Malleable and
Cast Irons and Steels," AtmOSpheric Corrosion, W. H. Ailor, ed., J.
Wiley and Sons, New York, 431-452, 1980.

 

D. P. Doyle and T. E. Wright, "Rapid Methods for Determining Atmos-
pheric Corrosivity and Corrosion Resistance,“ Atmos heric Corrosion,
W. H. Ailor, ed., J. Wiley and Sons, New York, 227-543, 1982.

J. H. Payer and W. H. Safranek, “Atmospheric Corrosion of Electro-
plated Zinc Alloy Die Castings," ASTM STP 646, S. K. Coburn, ed.,
American Society for Testing and Materials, 1978.

0. E. Richard and H. J. Snelling, “Weather Data and Its Use in Cor-
rosion Studies," "Proceedings of the 1974 Triservice Corrosion of
Military Equipment Conference, 29-31 October 1974,“ Volume II, 325-
327, 1975.

S. I. Rasool, "Chemistry of the Lower Atmosphere,“ Pleunum Press,
New York, 1973.

B. J. Mason, "The Physics of Clouds," Oxford Press, London, 1971.

E. A. Martell, "Advances in Chemistry No. 93,“ American Chemical
Society, Vol. 9, p. 138, 1970.

H. de Konig, World Health Organization, Geneva, private (written)
correspondence.

U.S. Air Force Environmental Technical Applications Center, "World-
wide Airfield Climatic Data,“ Vols. I-XI, Washington, DC, 1970-1973.

"National Survey of Air Pollution," Volumes 1-5, Warren Spring Lab-
oratory, Dept. of Trade and Industry, Her Majesty's Printing Office,
London, 1972.

Warner-Robins Air Force Base, GA, Air Logistics Command, per Karrie
Jo Robinson.

"Air Quality in Selected Urban Areas 1973-1974,“ World Health Organ-
ization, Geneva, 1976.

"Air Quality in Selected Urban Areas 1977-1978,“ World Health Organ-
ization, Geneva, 1980.

"Technical Note #106, Meteorological Aspects of Air Pollution," Pub-
lication 251.TP.139, World Meteorological Organization, Geneva, 1970.

96

References con't

 

56.

57.

58.
59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.
71.

H. W. de Konig and A. Kohler, "Monitoring Global Air Pollution,"
Env. Sci. and Tech. 12, 884-889, (1978).

"Air Monitoring Programme Design for Urban and Industrial Areas,"
World Health Organization, Geneva, 1977.

F. Chuang, private communication, 1984. (per R. Summitt).

E. F. House, “Automated Aircraft Rinse Facility MacDill AFB, FL,"
Report No. 82-C-039, Warner-Robins ALC, Robins AFB, GA, 25 May 1982.

D. A. Schoener, "Corrosion Inhibition of Aluminum Alloys,“ Senior
Thesis, Michigan State University, 1985.

“Percentage Frequency of Wind Direction and Speed-MacDill AFB, 1967-
70, 1973-78," Tables, Global Climatology Branch, USAFETAC, Air
Weather Service/MAC, per J. Hart, MacDill AFB, 1985.

F. Castellanos, "Effect of Surfactants Upon Corrosion of High-
Strength 7075-T6 Aluminum Alloy," Masters Thesis, Michigan State
University, 1985.

M. Khobaib, "Material Evaluation: Part II-Development of Corrosion
Inhibitors,“ Publication AFML-TR-79-4127, Air Force Wright Aeronau-
tical Laboratories, Wright-Patterson AFB, OH, 1979.

L. W. Masters and W. C. Wolfe, “The Use of Weather and Climatological
Data in Evaluating the Durability of Building Components and Mate-
rials,” Publication NBS-TN-838, National Bureau of Standards,
Washington, DC, 1974.

E. W. Turns, private communication, 1985, (per R. Summitt).

R. Summitt, “The Efficacy of Fresh Water Rinsing in Corrosion Control
of Aircraft,“ AFWAL/MLS - Systems Support Division, Wright-Patterson
AFB, OH.

Illustrated Parts Breakdown, “Publication T.O. 1F-IGA-4-23, General
Dymanics Fort Worth Division, 1983.

U.S. Environmental Protection Agency, "National Air Quality and
Emissions Report, 1976," Publication EPA-450/1-77-002, Research
Triangle Park, NC, 1977.

Construction Drawing, Aircraft Rinse Facility, Civil Engineering
Office, MacDill AFB, 1975, et. seq.

1. Schaffer, private communication, 1985, (per R. Summitt).

G. T. Browne, private communication, 1985, (per R. Summitt).

97

References con't

 

72.
73.

74.

75.

76.
77.

78.
79.

80.

81.

82.

83.

84.

85.

86.

87.

R. Summitt, private communication, 1984.

M. Eades, Erny Supply Corp., Tampa, private communication, 1984,
(per R. Summitt).

M. Khobaib, "Accelerated Corrosion Testing," Publication AFWAL-TR-
82-4186, Air Force Wright Aeronautical Laboratories, Wright-Patterson
AFB, OH, 1982.

M. Cawley, private communication, 1984, (per R. Summitt).
E. T. Wilder, private communication, 1985, (per R. Summitt).

R. Summitt, R. T. Fink, R. Suter, and N. Samsami, "Extending the
Lifetime of Operational Systems through Corrosion Tracking and
Prediction,“ Publication AFWAL-TR-83-4090, Air Force Wright Aero-
nautical Laboratories, Wright-Patterson AFB, OH, 1983.

T. Kozan, private communication, see also reference 77.

U.S. General Accounting Office, "The Air Force Can Improve Its Main-
tenance Information System," Report FA 1.13 GAO 83-20, January 1983.

American Public Health Association, "Standard Methods for the Exam-
ination of Water and Wastewater," 14th edition, 1976.

“Water and Environmental Technology," 1984, Annual Book of ASTM
Standards, Sec. II, Vol. 11.01, American Society for Testing and
Materials, 1984.

Hach Company, “Procedures for Water and Wastewater Analysis,“ Love-
land, CO, 1984.

Trung Hung Nguyen and R. T. Foley, "On the Mechanism of Pitting of
Aluminum,“ J. Electrochem. Soc., Vol. 126, p. 1855, (1979).

S. W. Dean, H. J. Delzer, C. A. Cooper, “Strong Acid Inhibitors with
Acetylenic Alcohols,“ NACE National Conference, 1979, Houston, TX.

M. G. Fontana, N. D. Greene, Corrosion Engineering, pp. 198-202,
McGraw-Hill, New York, 1967.

E. McCafferty, "Mechanism of Corrosion Control by Inhibiting," H.
Leidheiser, Jr., Ed., "Corrosion Control by Coatings," p. 279 ff.
Science Princeton, 1979.

S. W. Dean, R. Derby, G. T. Von Dem Bussche, "Inhibitor Types,"
Corrosion, December 1979, National Association of Corrosion Engin-
eers, Houston, TX.

98

References con't

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

H. H. Uhlig, "Corrosion and Corrosion Control," first edition, 224-38
John Wiley, New York, 1963.

R. H. Hausler, P. B. Phipps, G. R. Brubaker, Eds., "Corrosion
Chemistry," p. 262, ACS Symposium Series 89, Washington, DC, 1979.

F. L. LaQue, "Behavior of Metals and Alloys in Sea Water," H. H. Uhlig,
ed., The Corrosion Handbook, p. 401, John Wiley, New York, 1948.

I. L. Rosenfeld, "Crevice Corrosion of Metals and Alloys," 8. F. Brown
et. al. Editors, "Localized Corrosion," p. 378, NACE, Houston, 1974.

"Cooling Water Treatment Manual," TPC Publication no. 1, 8-11, NACE,
Houston, 1971.

B. W. Samules, K. Sotoudeh, and R. T. Foley, "Inhibition and Accel-
eration of Aluminum Corrosion," I'Corrosion," February, 1981, Na-
tional Association of Corrosion Engineers, Houston, TX.

M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions,
256-289, et. seq., Pergamon Press, London, 1966.

M. J. Collie, ed., Corrosion Inhibitors-Developments Since 1980,
Chemical Technology Review No. 223, Noyles Data Corp., Park Ridge,
N.J., 1983.

I. M. Kolthoff, E. B. Sandell, E. J. Meehan, S. Bruckenstein,
Quantitative Chemical Analysis, 4th edition, Macmillan Co., London,
1969.

Tampa Water Department, Daily Water Control Laboratory Analysis,
Tampa, Florida, E. T. Wilder.

L. H. Bennett, J. Kruger, R. L. Parker, E. Passaglia, C. Reinmann,
A. W. Ruff, and H. Yakowitz, Economic Effects of Metallic Corrosion
in the United States, Parts I and II, NBS Special Publication 511-
1, 2 (U.S. Department of Commerce, National Bureau of Standards),
Washington, D. C.