Frank C. Porter Corrosion Resistance of Zinc and Zinc Alloys 1994

Corrosion
Resistance
OF
Zinc
AND
Zinc Alloys
FRANK C, PORTER
Prepared for the International Lead Zinc Research Organization, lnc.
Research Wangle Park, North Carolina
Marcel Dekker, Inc. New York*BasebHong Kong
Library of Congress Cataloging-in-Publication Data
Porter, Frank C.
Corrosion resistance of zinc and zinc alloys / Frank C. Porter.
Includes bibliographical references and index.
ISBN 0-8247-9213-0 (acid-free paper)
1. Zinc-Corrosion. 2. Zinc alloys-Corrosion. 3. Corrosion and
p. cm. --(Corrosion technology; 6)
anti-corrosives. I. Title. 11. Series: Corrosion technology (New
York, N.Y.); 6.
TA480.Z6P63 1994
620.1'84223-d~20 94-14915
CIP
The publisher offers discounts on this book when ordered in bulk quantities.
For more information, write to Special Sales/Professional Marketing at the
address below.
This book is printed on acid-free paper.
Copyright 1994 by Marcel Dekker, Inc. All Rights Reserved.
Neither this book nor any part may be reproduced or transmitted in any form
or by any means, electronic or mechanical, including photocopying, micro-
filming, and recording, or by any information storage and retrieval system,
without permission in writing from the publisher.
Marcel Dekker, Inc.
270 Madison Avenue, New York, New York 10016
Current printing (last digit):
l 0 9 8 7 6 5 4 3 2 1
PRINTED IN THE UNITED STATES AMERICA
Preface
The corrosion resistance of zinc is so important that about half the worlds
annual consumption of zinc is used to protect steel from rust. A vast amount
of information now exists on the resistance of zinc (including its alloys and
their use as a coating) in a wide variety of corrosive conditions and is
summarized herein. The present volume draws upon these earlier works, but
also contains much valuable data from previously unpublished and untrans-
lated works. Zinc: Ifs Corrosion Resistance was originally compiled in 1970
by the Battelle Memorial Institute, Columbus, Ohio, which had been commis-
sioned by the International Lead Zinc Research Organization, Inc. (ILZRO)
to collate the available information on the corrosion resistance of zinc and
zinc-coated steel. The edition included information that had become
available since 1970 plus one new chapter dealing with the performance of
zinc as a sacrificial anode.
This book aims to present relevant information in a form that corrosion
engineers and other interested readers can easily use. Material from Zinc: Ifs
Corrosion Resistance has been improved with additions, revisions, and mod-
ifications of format where appropriate. Some data is re-presented in metric
units and in forms more suited to the practicing engineer. After a section on
general considerations, the book is divided into four chapters, dealing with
the corrosion resistance of zinc in atmospheric, aqueous, soil, chemical, and
miscellaneous environments.
Information has been presented in tabular or graphic form where appro-
priate, along with details of the exposure conditions and metal compositions,
iii
Preface
so that the effects of variables can be compared. In some cases, the full details
of test conditions were not available, while in others such conditions varied
widely. Thus, in any comparison of these data, the many factors that affect
corrosion rate must be considered. In particular, great care should be exer-
cised in the application of these data to other situations.
The corrosion behavior of brass and of zinc in dry cell batteries is
excluded. The corrosion resistance of zinc dust paints is a special situation
and is referred to only briefly. The corrosion resistance of zinc-coated steel in
concrete is only summarized, as it is adequately covered in my Zinc Hand-
book: Properties, Processing, and Use in Design (Marcel Dekker, Inc.) and
in the references cited in the text.
I acknowledge the great assistance of Jan van Eijnsbergen, who has
summarized information on duplex coatings and material previously available
only in Dutch and German. With the special help of Dr. E. Proskurkin from
Ukraine, I have also been able to include information previously available
only in German, Russian, and other European languages.
Data from published sources are fully documented and acknowledged.
Finally, I acknowledge the help of various metal and chemical manufacturers
in providing unpublished information.
Frank C. Porter
Contents
...
Preface .......................................................
1 . General Considerations ...................................... 1
I . Subject Range ....................................... . l
II . Materials and Techniques Considered General Behavior .... 3
I11 . Types of Corrosion .................................. 61
IV . How Zinc Can Protect Steel ........................... 83
V . Corrosion Tests and Their Significance .................. 87
VI . Other Relevant Properties of Zinc ...................... 96
VI1 . Practical Experience: Case Histories .................... 98
2 . Resistance to Atmospheric Corrosion ......................... 101
I . Introduction ........................................ 101
I1 . Influence Atmospheric Variables .................... 121
I11 . Corrosion Data by Location .......................... 172
IV . Corrosion Data by Alloy ............................. 211
V . Comparative Performance ............................ 241
VI . Case Histories ...................................... 248
3 . Resistance to Aqueous Corrosion ............................ 263
I . Introduction ........................................ 263
I1 . Influence of Types of Water .......................... 264
I11 . Corrosion Data ..................................... 281
IV . Zinc Anodes in Water ............................... 333
V . Case Histories ...................................... 337
Contents
. Resistance to Corrosion by Soil .............................
I . General ...........................................
I1 . Influence of Types of Soil ...........................
111 . Corrosion Data .....................................
IV . Zinc Anodes in Soil .................................
V . Reinforced Earth ...................................
VI . Case Histories ......................................
5 . Organic and Inorganic Chemicals, Building Materials
and Coatings. and Fuels ...................................
I . General ...........................................
I1 . Cleaning of Zinc Surfaces ............................
I11 . Inorganic Chemicals .................................
IV . Organic Chemicals and Fuels .........................
V . Concrete. Cement. and Plasters .......................
VI . Bitumen ...........................................
VI1 . Other Building Materials ............................. 444
References ...................................................
Index ...................................................... S11
Corrosion
Resis t a m
OF
Zinc
AND
Zinc Alloys
This Page Intentionally Left Blank
1
General Considerations
SUBJECT
The book replaces the second edition of Zinc: Its Corrosion Resistance,
published in 1983 as a partial revision of the 1971 edition, but mainly
referring to pre-1967 data. In addition to including later data, the present book
covers earlier work in foreign languages-notably German and Russian-not
covered by the original authors.
Zinc coatings represent the biggest single use of zinc. The application is
increasing steadily as a result of the need to make the most efficient use of
steel and to conserve materials and energy. The alternative methods by which
coatings are applied-notably hot dipping, electroplating, thermal spraying,
mechanical coating techniques, and painting-are briefly described, together
with aftertreatments. Duplex coatings of zinc followed by paint, plastics, or
powder coatings form a major section. Cathodic protection using sacrificial
zinc anodes instead of a coating is also included.
Zinc coating alloys with improved corrosion resistance have developed
rapidly, particularly those with 5 or 55% aluminum additions and applied by
hot dipping, those with 15% aluminum for thermal spraying, and those with
nickel or small amounts of a wide range of other elements applied by plating.
The special properties of each group, including improved corrosion resis-
tance, are identified.
Zinc-aluminum castings with up to about 27% aluminum have been
prominent for versatility in design possibilities including thin walls (achieved
2 Chapter
most readily with the lower aluminum contents), which make the improve-
ments in corrosion resistance identified here also of significance.
Zinc alloys for sheet with small additions of copper and titanium are
used for reasons other than corrosion resistance, but their corrosion perfor-
mance is slightly enhanced.
The corrosion resistance of brass, a zinc-copper alloy, is outside the
scope of this book.
For convenience, many research programs use zinc sheet rather than
zinc-coated steel and because, in general, the results are applicable to both
types of use, the data are usually grouped together in this book. Only when
special types of corrosion are discussed, or results are expressed in terms of
loss of mechanical properties (e.g., with castings), is it necessary to make
major distinctions. Zinc coatings applied by different methods have for
practical purposes similar corrosion resistance.
Very careful distinction is needed, however, between the results of
accelerated tests (either in natural or artificial environments) and long-term
tests (i.e., 1 l year). Zinc often reacts much faster and forms visible
unprotective corrosion products in accelerated tests, which are intended to
simulate real-life conditions but do not achieve this. Salt spray tests, although
widespread in specifications and in the literature, are particularly unreliable
when assessing corrosion resistance of zinc coatings and duplex systems. The
presence of an overdose of chloride ions influences cathodic depolarization,
generally leading to pitting corrosion. Recently, more reliable accelerated
tests have been introduced, the proposed Swedish Standard SS 11 72 1 1 ,
related to the General Motors Scab Corrosion Test (GM95 1 lP), showing a
much improved correlation with outdoor weathering tests for automotive
enamels on continuously hot dip galvanized sheet (HDGS). The book concen-
trates on corrosion data of use to designers and contractors but also includes
material that is primarily of use to research workers.
This chapter covers information applicable to zinc corrosion behavior in
general. Chapter 2 covers corrosion in the atmosphere-which is the most
important group of environments in which zinc is used. Attack is usually
approximately linear with time, but often with some reduction of rate as
protective films form. Many results are available, and tables have been
prepared for the guidance of designers. Water corrosion follows in Chapter 3,
with distinctions between hard and soft tap water (hot and cold), temperate
and tropical seawater, and tidal and splash zones. Buried structures-together
with a section on earth reinforcement-follow in Chapter 4, and conditions
appropriate for zinc sacrificial anodes are included in both Chapters 3 and 4.
General Considerations 3
Chapter 5 provides a checklist for the designer giving chemicals that
have been used with zinc or zinc-coated steel and tells whether they are,
broadly speaking, suitable or unsuitable. It is emphasized that these lists are
merely a starting point and do not replace the need for trials in large or critical
projects, unless there is directly applicable case history experience. Also,
attention must be gives to concentration, aeration, and temperature of solu-
tions of chemicals in contact with zinc surfaces. Often, alternating attack is
more severe than permanent immersion. Chapters 2-4 of the book conclude
with case histories as additional guidelines.
II. MATERIALS AND TECHNIQUES CONSIDERED:
GENERAL BEHAVIOR
A. Wrought Zinc
Rolled zinc is produced as sheet, strip, plate, rod, and wire in numerous
compositions and alloys, depending on the ultimate uses of the rolled prod-
ucts. When desired, strip is slit into narrow widths. Most sheet zinc is now
produced in a single casting and rolling operation, replacing the discontinuous
pack rolling method used formerly.
The unalloyed zinc used for roofing up to about the was soft and
had a tendency to creep, with the result that if the bays of fully supported
roofing were too wide, they would eventually sag in the middle.
The tendency today is to use specified compositions with small alloy
additions aimed to improve the grain size, give work hardening, and improve
properties such as creep resistance. A number of proprietary compositions are
on the market with additions of copper, manganese, magnesium, aluminum,
chromium, and titanium, but the most widely used alloy contains copper and
titanium. American and British standard specifications for rolled zinc for
building uses are given in Table The minor additions of copper and
titanium appear to slightly beneficial but have little practical effect on
corrosion resistance considerations for designers of roofs. special product is
superplastic zinc containing 22% aluminum; when produced under particular
conditions, this material gives maximum scope for deformation. Its corrosion
behavior has been studied, for example, by Williams et al. The
aluminum addition is generally beneficial except in alkaline conditions.
Higher aluminum alloys (40-50%) have still better resistance to boiling water
or steam (Melton and Edington,
&
Table 1.1 Typical Compositions of Selected Rolled-Zinc Alloys (ASTM B69 and BS 6561)
Element (%)
Alloys (UNS) c u Pb Cd Fe, max AI, max Other, max
ASTM B69
Zn-0.08 Pb (221210)
Zn-0.06 Pb-0.06 Cd
Zn-0.3 Pb-0.3 Cd
Zn-1 Cu (244330)
Zn-1 Cu-0.010 Mg
Zn-0.8 Cu-0.15 Ti
Zn-0.8 Cu (240330)
Zn-Cu-Ti
(Z2 1220)
(Z2 1540)
(245330)
(Z41320)
BS 6561
0.001 max
0.005 max
0.10 max
0.05-0.10
0.005 max
0.05-0.08
0.012
0.012
0.001
0.001
0.001 Sn
0.001 Sn
0.005 max 0.25-0.50 0.25-0.45 0.002 0.001 0.001 Sn
0.85-1.25
0.85-1.25
0.10 max
0.15 max
0.005 max
0.04 max
0.012
0.015
0.001
0.001
0.001 Sn
0.0064.016 Mg
0.001 Sn
0.12-0.50 Ti
0.001 Sn
0.02 Ti
0.50-1 S O 0.10 max 0.05 max 0.012 0.001
0.70-0.90 0.02 max 0.02 max 0.01 0.005
0.0&0.18 0.003 max 0.005 max 0.002 0.010 0.03-0.15 Ti
0.001 Sn
0.001 Sn Zn 0.3-1 .o 0.001 max 0.005 0.002
The Unified Numbering System of the United States.
For the manufacture of cans that form the container and one of the
electrodes of the dry cell (the flashlight battery), the battery manufacturers
have special requirements for the purity and for small alloying additions, both
of which can improve the efficiency of the cell. Historically, the manufacture
of photoengraving plates required a very uniform metallurgical structure and
lack of directional effects, but the market for these plates has largely van-
ished. This book also does not cover the deliberate corrosion reactions
involved in batteries and photoengraving plates.
B. Cast Zinc-Based Materials
Zinc alloy castings offer advantages to designers and production engineers
because they can be produced to near net shape and with useful mechanical
and physical properties. The family of available zinc-based casting alloys,
which usually contain about 4% aluminum, has been increased in the past 15
years to provide stronger, creep-resistant alloys. These properties have been
achieved primarily by alloying with more aluminum. Commercial designa-
tions (notably ZA-8,ZA-12, and ZA-27) are based on the aluminum contents
of 8, 12, or 27% originally proposed. Current compositional specifications of
these and of the traditional 4% aluminum alloys are given in Table 1.2. Fuller
details of zinc-based castings are given in Zinc Handbook (Porter, 1991).
The ZA alloys have better general corrosion resistance than zinc or the
zinc4% aluminum alloys in many environments, while alloying and techno-
logical improvements increase the resistance to intergranular attack, which
was historically a problem with zinc-4% aluminum castings produced from
material in which impurities were inadequately controlled. These advances
also allow more tolerance of impurities in specifications.
Cast alloys must made from at least 99.99% purity zinc ingot to
comply with the standards of the American Society for Testing and Materials
(ASTM B6) and the International Organization for Standardization (IS0 752);
a standard is in preparation by the ComitC EuropCen de Normalisation (CEN:
European Committee for Standardization). The harmful impurities, which
occur naturally in zinc or as contaminants in the other alloying ingredients,
are lead, tin, cadmium, indium, and thallium. These impurities are limited by
specification; higher contents than specified may make the alloys susceptible
to intergranular corrosion, particularly when exposed to warm, moist environ-
ments.
On the higher aluminum content alloys, ZA-8, ZA-12, and especially
ZA-27, the normal protective layer of zinc corrosion products is enhanced by
the formation and inclusion of aluminum oxides and compounds.
Table 1.2 Typical Compositions of Selected Cast Zinc Alloys (ASTM B86 and B791)
Alloy
Element 2 3 5 7 ZA-8 ZA- 12 ZA-27
~ ~
Aluminum 3.5-4.3 3.54.3 3.5-4.3 3.5-4.3 8.0-8.8 10.5-1 1.5 25.0-28 .O
Copper 2.5-3.0 0.25 max 0.75-1.25 0.25 max 0.8-1.3 0.5-1.2 2.0-2.5
Magnesium 0.02-0.05 0.020-0.050 0.03-0.08 0.005-0.020 0.015-0.030 0.015-0.030 O.OI0-0.020
Iron (max) 0.100 0.100 0.100 0.075 0.10 0.075 0.10
Lead (max) 0.005 0.005 0.005 0.0030 0.005 0.005 0.005
Cadmium (rnax) 0.004 0.004 0.004 0.0020 0.004 0.004 0.004
Tin ( ma ) 0.003 0.003 0.003 0.0010 0.003 0.003 0.003
Nickel - - - 0.005-0.020 - - -
Notes: In alloys 3, 5, and 7, nickel, chromium, silicon, and manganese may be present up to 0.02, 0.02, 0.35. and 0.06% respectively.
See ASTM B86 for further information. Other national specifications differ in detail.
Alloy 2 is also known as the German DIN 1743 Alloy 2430 and ASTM AC43A. UNS 33541.
Alloy 3 is also IS0 301 Alloy ZnA14. BSI Alloy A, ASTM AG40A. UNS 233520.
Alloy 5 is also IS0 301 Alloy ZnAlrlCul, BSI Alloy B, ASTM AC41A, UNS 233531.
Alloy 7 is also known as ASTM AGNB, UNS 233523.
ZA-8,ZA-12, and ZA-27 are also UNS 35636,35631, and 35841, respectively, and are also in DD 139 (DD = BSI Draft for Development).
The presence of impurity elements other than iron, lead, cadmium, and tin at the levels likely to be present in the alloys has not been found to
be detrimental to ZA-8, ZA-12, and ZA-27.
Ingot for foundry and pressure die casting may contain chromium, manganese, or nickel in amounts up to 0.01% each or 0.03% total (see
ASTM 8791-91). 3
p1
2
4
-L
C. Zinc and Zinc Alloy Coatings
General
The ability to select particular alloys or particular thicknesses of coating
depends on the type of coating process used. Thus some details of the coating
processes and their major attributes are given here, and Table 1.3 summarizes
aspects of major interest to the designer, contractor, or user. The corrosion
resistance per unit of mass (Fig. 1.1) is largely independent of the method by
which a zinc coating is formed, provided the exposed surface per unit of mass
is the same and nothing extra is incorporated in the coating.
1.
2.
4.
5.
6.
Six methods of coating iron and steel are in general use. These are:
Hot dipping: cleaned steel is immersed in molten zinc or zinc alloy, and
reaction takes place to form a metallurgically bonded coating.
Zinc electroplating (sometimes erroneously referred to as electro-
galvanizing): zinc salt solutions are used in the electrolytic deposition of
a layer of zinc on a cleaned steel surface.
Mechanical coating: suitably prepared small steel parts are tumbled in
zinc dust at room temperature. Deposition from vapor is also used to give
a mechanically or physically adherent coating on steel sheet.
Sherardizing: suitably prepared small steel parts are tumbled in zinc dust
at just below the melting point of zinc so that a metallurgically bonded
coating forms.
Thermal spraying: grit-blasted surfaces are covered with droplets of
semimolten zinc sprayed from a special gun that is fed by either wire or
powder.
Coatings incorporating zinc dust or flake: proprietary developments are
important and can involve inorganic as well as organic media in air-
drying and stoved formulations. Zinc-rich paints are closest in behavior
to other zinc coatings: they contain sufficient zinc powder to give a film
capable of conducting electricity.
Techniques for applying molten metal directly to steel surfaces other than by
dipping are of considerable interest for the future if they can be used in a cost-
effective manner. Microsections of the main types of coating are shown in
Fig. 1.2. simplified diagram (Fig. 1.3) shows the thickness of the main
coatings used on steel products, with sheet galvanizing included for compari-
son. In the following pages, each method of applying zinc to steel is briefly
considered in relation to the way in which the coating protects steel from
corrosion.
Coatings may sometimes be removed in fabrication (e.g., in machining
or drilling operations). Then touch-up procedures are needed, frequently on
Table 1.3 Comparison of Zinc Coatings
Characteristics Electroplating (electro-
of the coating Hot dip galvanizing galvanizing)
Adhesion
Continuity and
uniformity
Thickness
Formability and
mechanical
properties
Extra treatments
Other
considerations
Coating is integral with the
steel because the formation
process produces zinc-iron
alloy layers overcoated
with zinc.
Good-any discontinuities are
readily visible as black
spots. Some excess zinc
at drainage points on
products.
Normally about 50-125 pm on
tube and products; thicker
coatings up to 250 pm ob-
tained by grit-blasting be-
fore galvanizing. Coatings
10-30 pm applied to
continuous sheet and wire.
Conventional coatings applied
to finished articles, not
formable: alloy layer is
abrasion resistant but brittle
on bending. Special coatings
with little no alloy layer
readily formed (e.g., on
sheet) and resistance-welded.
Conversion coating-hro-
mates prevent wet storage
stain; phosphates good
new sheet as a base for
paints. Weathered coatings
often painted (after 10-30
years) for longer service.
Size of bath available. Parts
up to about 25 ym long
can be dipped at special-
ized works. Care required
at design stage for best
results. Continuous wire
and sheet available.
Good, comparable with other
electroplated coatings.
Uniform within limitations of
throwing power of
bath. Pores not a problem,
as exposed steel protected
by adjacent zinc.
Thickness variable at will;
generally 2.5-15 pm.
Thicker layers are possible
but generally unecono-
mical.
Electroplated steel has
excellent formability and
can be spot-welded. Small
components are usually
finished before plating.
Conversion coatings (e.g.,
chromates used to prevent
wet storage stain. Fre-
quently used as a base for
paints.
Size of bath available.
Process normally used for
simple, fairly small com-
ponents suitable for barrel
plating or for continuous
sheet and wire. No heating
except for hydrogen
embrittlement relief on
high-strength steels.
Mechanical
coatings Sherardizing Zinc spraying Zinc dust painting
Good, compa-
rable with
electroplated
coatings.
Thin at comers
-the opposite
of electroplated
coatings.
Variable at
will, usually
10-80 pm.
Good. Does not
cause
permanent
embrittlement
of high
strength steels.
Can have
conversion
coatings
applied.
Best for small
parts (e.g., UP
to 15 cm or
250 g). Access
difficulties
(e.g., inside
tubes).
Good-the dif-
fused coating
provides a
metallurgical
bond.
Continuous and
very uniform,
even on
threaded and
irregular pacts.
Usually 15 30
p m closely
controlled.
Applied to fin-
ished articles;
forming not
required.
Excellent
abrasion
resistance.
Can be painted if
required or
given a black
finish by the
processor.
Generally used
for fairly
small, complex
components;
useful when
close control
of tolerances
important.
Good mechanical
interlocking, provided
the abrasive grit-
blasting pretreatment is
done correctly.
Depends on operator
skill. Coatings are
porous, but the pores
soon fill with zinc cor-
rosion products and are
thereafter impermeable.
will, generally
150 pm but coatings
of up to pm can
be applied.
Thickness variable at
When applied to finished
articles, forming not
required. Can weld
through thin coating if
necessary, but prefer-
able to mask edges to
be welded and spray
these afterwards.
Coating with sealants
that can provide a
base for paints to give
long-life structures.
size shape limita-
tions. Very economical
for work with high
weight-to-area ratio.
Uneconomical on open
mesh. Access difficul-
ties may limit its ap-
plication (e.g., inside
tubes). Best method of
applying very thick
coatings. Little heating
of the steel.
~ ~~ ~~
Good-abrasive grit-
blasting of the steel
gives best results.
Good-any pores fill
with reaction
products.
Up to 40 pm of paint
(and more with
special formula-
tions) can be
applied in one coat.
Abrasion resistance
better than
conventional paints.
Painted sheet can
be formed and
resistance welded
with little damage.
Can be used alone or
as primer under
conventional paints.
Suitable for anything
that can be painted,
though there may be
difficulties of access
in narrow tubular
structures. Can be
brush, spray, dip
applied. heating
involved. Perfor-
mance varies with
media used and per-
centage of zinc dust.
10 Chapter 1
Coating mass (g/m')
0 1 2 3 4 6
Coating weight (oJsq It)
Fig. 1 .I
at Sheffield, England. Coating process: 0 = sherardizing, 0 = zinc plating,
hot dipping, 0 = zinc spraying, (Hudson, 1959).
Effect of zinc coating weight on corrosion rate (for different zinc coatings)
=
site. Adams et al. (1991) dealt with the main techniques: the use of zinc-rich
paints, application of low melting point alloys of zinc (solders) and spraying
molten zinc (thermal spraying). The relative performances of the three touch-
up techniques as determined in this study are given in Table 1.4. When
overpainting zinc coatings, special attention is needed to ensure that the paint
is compatible with the touch-up coating.
Hot Dip Coatings
Hot dip galvanizing is the most important zinc coating process. More than 2
million tons of zinc are used in this method to coat about 40 million tons of
steel each year. All mild steels and cast irons can be coated, but the thickness
and structure of the coating will depend on the alloying elements. Nearly half
of the steel coated is in the form of sheet and a quarter is fabricated work; the
remainder is tube or wire. Each type of product forms a separate industry with
rather different types of plant. Metallurgically, the processes used for fabri-
cated work and tubes are similar, but the process used for sheet has small
additions to the zinc to give a coating that has less iron-zinc alloy (Fig. 1.2a,
middle) and, consequently, is very flexible. The wire galvanizing industry is
more comparable to sheet galvanizing in engineering practices, but it tradi-
tionally uses zinc to give coatings of the same types as on tubes. Gradually,
the distinctions between the four types of industry are becoming less sharp,
particularly with the introduction of zinc-5% aluminum type alloys for coat-
ing each type of product and the zinc-55% aluminum coating for sheet.
The zinc bath is normally maintained around At this tem-
perature, the reaction between zinc and iron or steel is at first rapid and then
usually slows down, the rate of coating formation being related to type of
steel. A fairly constant thickness of coating can be obtained by varying the
immersion time in the bath. The reaction produces several specific zinc/iron
compounds, which form layers in the coating. A layer of pure zinc solidifies
on the outside of the article as it is taken out of the bath. A slightly thicker
coating can be obtained if the speed of withdrawal is increased, but this factor
is secondary to the influence of alloying elements in the ferrous metals; for
example, silicon and phosphorus can give increased zinc-iron alloying and a
linear rate of attack with time, hence thicker coatings. A recent review of the
effect of silicon (the Sandelin effect) is given by Vazquez and Damborenea
(1991), who also discuss the corrosion resistance of the alloy layers formed
and append an extensive bibliography.
Most differences in surface appearances are of little or no significance
with regard to corrosion resistance. The most common were discussed by the
Galvanizers Association (GA-based in the United Kingdom) in General
Galvanizing Guide (1986), and Horstmann (1983) illustrates a wider range.
In the galvanizing of fabricated products, some variations in surface
appearance can arise as a result of the differing optical reflectivity of the zinc
crystals when their atomic structure is at different angles to the light. Rough
surfaces of hot dip galvanized surfaces not only affect aesthetics but may also
increase the retention of pollutants from the atmosphere. Horstmann (1983)
has presented 33 examples of failures caused by faulty pretreatment on
galvanizing. Wesolowski and Jagiello (1989) investigated galvanized wire
and sheet and found that roughness is caused by the three-component phase
FeAlo,,Zn,, phase), which is drawn out of the bath with the work.-A
reduction of the ammonium chloride content in the flux to between 12 and
27% of the amount of zinc chloride leads toward a reduced concentration of
iron in the flux solution. Heating to 130-160C causes the removal of water,
which evaporates during dipping in the molten zinc and, thus, forms a
protective atmosphere over the liquid zinc surface and prevents the burn-
ing of the flux. Galvanized surfaces, after fluxing and drying, should not
show a greenish-yellow color, since this indicates too high a concentration of
12 Chapter
.. ." ~
0.001- continuously hot 0alvJnlr.d coating on mheet
5 6clnssra.s of c o a l ~d shFQl nvnlloble wllh ttltcknesses
. 0. M1 ' 100 002. rach sidc
, Nota: 0.001" 25pm
galvanized coating on killed steel
Compcsl1ton and hlpher Slhcon levels)
Conllna IhlchnesaO 004 10 0 Ol o' accordl no IC stwI
0. 0" zinc sprayed coating on grit blasted stori
Av, ~l l ob! c 0.004. on or Bnrl 01 surface
1.2 (a) Microsections of the ma-
1 jor types of zinc coatings (ZDA). (b) Mi-
crostructure of a Galfan coating cooled
from the processing temperature of
at approximately 200"C/sec. (c) Micro-
zlncrlch paint onahot blasted steel
f ornn~l , ?l ~on and sIw1 surfow oremrallon
l nor ~nni c btnders. PwformancPdcprnds on paint
. . .., . -
, .
O.ooO2'brlght zinc electro plated coating
Dull c o a l ~ n q ~ nrr sl qhl l v rouphnr
sectional photomicrograph of Galvalume
sheet (amyl-nital etch). (e) Schematic of
the mechanisms of corrosion of Gal-
valume coatings (Ayoub et al, 1988).
Top, formation of cracks in the coating.
Bottom, osmotic pressure generation.
Key: 1 = steel substrate; 2 = inter-
metallic (Fe, AI, Si, Zn); = inter-
metallic [AI, Si, Zn (Si highly concen-
trated)]; 4 = Galvalume coating; 5 =
Zn) phase; 7 = pore; 8 = corroded
zinc-rich phase; 9 = cracks generated by
osmotic pressure and propagating into the
coating.
" - - - - " . - 7 zinc-rich phase; 6 = intermetallic
14 Chapter 1
1.2 (Continued)
Zn-ric h AI-ric h
Si
Chapter
a
4 Anodic zone : hi gh concentration
of soluble salts
Zone 7 b
(e)
1.2 (Continued)
ferrous chloride. Keeping the iron content of the zinc bath well below 0.1 %
also assists in obtaining smooth galvanized surfaces.
Van Eijnsbergen has surveyed the protection edges by paint
by galvanizing. Since the number convex edges on profiles is rather high,
this characteristic is important in practice. Because of increased Zn/Fe alloy
growth on comers, a thicker coating is realized, and thus enhanced corrosion
protection. The nature of the drying of paint systems involves contractions in
the areas close to convex edges, resulting in thinner coatings in that area. He
reports that Tomashev showed that an increase of the radius profiles would
17
Zlnc coatlngs on steel
Steel surface Coating weight: g/mz - 460 810
b :o h L3 L '50 : o l k85Ao 1L 1l O ,&I 14
Coating thickness: *m measured horn steel surface
Hot diD aaivanized to 85729
Thick hot din nalvanized coatinn
Cenlrifuaal aalvanizina to BS729
Zinc spraying to BS2569:Zn4
Continuous aalvanized sheet lo BS2989:G275
I'
Zinc coating for many car bodies
Zinc plating lo Zn2 of BS1706
n
U
Sherardizino lo Grade 1 of BS4921
Faints and coatings incorporating zinc dust (for cars)
1
0 x l x2 x3 x4 x5 x6 x 7 x8 x9 x10
Re l a h life R x D B C ~ ~
Key:
Dispersed zinc Pure
zinc oiament alloy layers zinc
Fig. 1.3 Typical thicknesses of zinc coatings (ZDA).
result in 20-3096 better protection by paint coatings when these radii were 2-5
mm larger. When flame or laser cutting steel parts, the zinc coatings on the
cut sides generally are 15-30 pm thinner than on adjacent areas.
Coating structure can also be greatly influenced by the composition of
the molten zinc bath: for example, small aluminum additions reduce or
Chapter
Table 1.4 Performance Touch-up Materials on Bare Areas on Galvanized
Steel
Ease of application
Thickness control
Appearance
Adhesion
Abrasion resistance
Corrosion resistance
barrier protection
cathodic protection
Form application
Paints Solders Zinc spraying
High
LOW Medium
Medium Low High
High
High High
LOW Highb Medium
LOW LowC
High
High LOW
High
LOW Medium High
Compatibility with subsequent paint coatings must be assessed by paint suppliedapplicator.
bHighly variable.
Higher than paints.
Source: Adam et al. ( 1991) .
suppress the reaction between iron and zinc. In the continuous galvanizing
processes for sheet, 0.1-0.3% aluminum is used in the bath, and this produces
a thinner coating with virtually no alloy layer, which can be severely de-
formed without cracking or peeling of the coating. However, the iron-zinc
alloy layers in hot dip galvanized and sherardized coatings may have a
significantly higher corrosion resistance than the same mass of zinc when
exposed to neutral or slightly acid conditions (Salt et al., 1965; Sjoukes,
1976, 1991). Van Ooij et al. (1991) have claimed, based on measuring
potential/current ratios in an electrolytic cell, that the 6-1 (Fe/Zn) alloy,
obtained when hot dip galvanizing at 550-560C, has significantly superior
corrosion resistance. [At 530-54OoC, some alloy is interspersed with the 6-1
and the coating may not be so smooth (Van Eijnsbergen, personal communi-
cation).] No advantage appears to exist in alkaline conditions, and Sjoukes
(1976) reports that some workers found the iron-zinc alloy less suitable in
concrete than zinc. Also, heat treatment of galvanized steel at 650C for 60
minutes, giving 20% iron in the coating, generally increases life in atmo-
spheric exposure, but only in the proportion to the increase in the total coating
thickness, and with only 10% iron the corrosion resistance can be less
(Campbell et al., 1965). Russian work (see Chapter 2, Section 111, B)
suggests rather better performance.
Other minor additions to galvanizing baths, hence to the galvanized
coating, have little effect. In atmosphere tests of 5 8 years duration,
19
depending on location, Frappe et al. (1985) confirmed this for lead, tin,
copper, and aluminum, although Radeker et al. (1961) found copper benefi-
cial (especially around 0.2-0.4%). The former report also contains a valuable
summary of work up to that date on the effects of alloy additions in galvanized
coatings. Cominco (1985), in reporting 15-year results for samples prepared
in galvanizing baths with minor additions, said that vanadium in the coating
gave both lowest coating weights and lowest corrosion rates. Both vanadium
and nickel made it possible to avoid dark gray appearances.
Galvannealing (the conversion of a conventional galvanized coating to
an all iron-zinc alloy coating by heating while in-line immediately after
coating) is probably the favorite coated steel for the automobile industry
today. One widespread problem, the powdering of galvannealed coatings
when deformed, can affect subsequent painting and general appearance.
Close control of the iron content of the coating is important: below 8%.
welding becomes less easy; &lo% iron is best to minimize corrosion from
edges under paint when the galvannealed coating is thin (Jagannathan, 1992),
while powdering during forming increases with iron content up to about 15%
(Mercer, 1992), with some users regarding 12% iron as the maximum accept-
able. The coating structure is also important: fast heating with prolonged
holding at moderate temperatures was shown to limit powdering (Wilmotte et
al, 1992).
Diffusion annealing of galvanized coatings received great attention in
the Soviet Union, and the English translation of Proskurkin and Gorbunovs
book Galvanizing, Sherardizing, and Other Zinc Diffusion Coatings (1975)
gives details.
The major developments of zinc-aluminum alloy coatings, notably 5%
aluminum and 55% aluminum, are discussed mainly in Chapter 2, since
atmospheric corrosion resistance is the application for which they are most
widely used. Microsections of fast- and slow-cooled zinc-5% aluminum alloy
are shown in Fig. 1.2b and Fig. 1 . 2 ~ respectively and a schematic view of the
55% aluminum alloy together with an indication of its mechanism of corro-
sion and corrosion protection in Figs. l .2c-e (Ayoub et al., 1988). Figure 1.4
shows the equilibrium diagram for zinc-aluminum alloys, the significance of
the eutectic at just over 5% aluminum (relevant to the coating alloy composi-
tion), and the solidification patterns of the higher aluminum casting alloys. A
30% aluminum alloy has also been developed for sheet coating (Bonaretti et
al., 1988) and 15% aluminum for thermally sprayed coatings. In zinc-
aluminum-silicon alloys, Memmi et al. (1988) have shown-at least in
accelerated tests-that magnesium additions increase the corrosion resistance
by segregating at grain boundaries and stabilizing the coating against inter-
20 Chapter
700
600
W
iK
500
3 400
t-
iK
a
I
200
0 20 50 60 70 80 90
ZINC CONTENT, Wt. %
Fig. 1.4 Equilibrium (phase) diagram for zinc-aluminum alloys.
granular attack, making it possible to obtain, with lower aluminum contents,
the same corrosion resistance as in the 55% aluminum alloys. Also Nakamura
et al. (1984) report a new zinc-aluminum-tin ternary alloy, typically by
containing zinc plus 15% AI and 0.5% Sn, has been developed and suc-
cessfully adopted in Japan as a hot dip coating material for steel wire,
conferring excellent corrosion resistance and formability.
Hot dipping sometimes is done in two successive stages (e.g., with
wire). This is one of the possible commercial techniques in Galfan coating of
wire. Ochai and Ohba (1989) have studied the microstructure and the corro-
sion behavior of the intermetallic layer obtained by immersing conventionally
galvanized steel wire in a Zn-AI alloy bath containing 3-10% AI. This double
hot dip process forms duplex coating layers. The outer coating layer has the
same chemical composition as the alloy bath, and the intermetallic layer,
which is formed by the diffusion of aluminum from the alloy bath into the Zn-
Fe intermetallic layers of the galvanized wire, has about 30% Al. The
microstructure of the intermetallic layer the Zn-AI alloy coating consists of
the Zn-AI alloy matrix of the quasi-eutectoid phase and the fine, platelike
precipitates of Fe4AIl3 containing some zinc. This intermetallic layer exhibits
such good drawability that the total reduction in area by drawing can be
increased to more than 80% without any crack occurrence. The time to first
21
rust of Zn-AI alloy coated steel wire in the salt spray test exceeds by more
than four times that of galvanized wire, and the high corrosion resistance is
unchanged after drawing because of the lack of coating cracks.
In further work, Ochai and Ohba varied the aluminum content of the
coating, the cooling conditions after coating, the reduction of areas by
drawing, and the blueing temperature after drawing. The eutectoid phase,
with 22% AI and comprising a fine complex mixture of a-AI phase and p-Zn
phase, is dominant in the corrosion behavior of the Zn-AI coatings. Preferen-
tial corrosion occurs along the boundaries, such as p-Zn primary crystals/
eutectic phase, p-Zn matrideutectoid phase, and eutectic colony boundaries.
The corrosion of the p-Zn phase is preceded by that of the eutectoid phase.
The corrosion rate of the coating increases with the increase in the cooling rate
after coating. Either the eutectic structure or the primary dendrite structure
grows along the direction of solidification of coating as the cooling rate
increases. These structures of uniform direction from the surface to the basis
metal are considered to accelerate the corrosion along the preferential corro-
sion paths, resulting in some deterioration of the corrosion resistance.
Work by Motohashi and Shibata (1986) showed that similar 22%
aluminum coatings from a bath to which was added approximately
Si showed good formability, coating adhesion, and flaking behavior, as well
as corrosion resistance and damping capacity. The coating had a fine, equiax-
ial structure, a property that has a favorable effect on superplasticity and
damping of vibration.
Hot dip galvanizing has also been used on calorized steel. Mokhamed
and Kukharev (1987) said that the best corrosion performance in a simulated
atmospheric test was on samples calorized at for 4 hours in the mixture
of ferroaluminum containing 40% AI followed by galvanizing at for 4
hours(!).
The corrosion resistance of galvanized steel forms a large part of this
book, and a useful German-language booklet is issued by Stahl-Informations-
ant r um (1990) titled Corrosion Resistance of Galvanized Steel. A similar
publication in Dutch has been prepared by Van Eijnsbergen [issued by the
Stichting Doelmatig Verzinken (SDV), the Dutch Galvanizing Institute,
19591.
Electroplated Coatings
Zinc electroplated onto steel gives a bright and smooth finish. It is used for
decorative effect and to protect delicate objects where rough or uneven
finishes cannot be tolerated (in instrument parts) and for articles that cannot
withstand the pretreatment or temperatures required in the other coating
processes. Electroplated zinc is also very ductile, and the process is widely
22 Chapter 1
used for the continuous plating of strip and wire where severe deformation
may be required. The electroplating process is sometimes referred to as
electrogalvanizing. This term is misleading, since the chief characteristic
of galvanizing is the formation of a metallurgical bond at the zinc-iron
interface and this does not occur in electroplating.
Four limitations of zinc plating are common to all processes. First, the
article to be coated can be no bigger than the plating bath available. Second,
the articles shape is limited by the ability (throwing power) of the particular
bath composition to reach into recesses and other parts distant from the anode.
The third limitation is economic; above about 15 pm (0.015 mm), other
methods of applying zinc are cheaper. Alloy coatings, such as zinc-nickel,
which have a higher unit corrosion resistance, are of particular commercial
interest. Finally, with steels above about 1050 N/mm, heat treatment is
needed to avoid hydrogen embrittlement of the steel.
Four generic types of zinc plating solution are available: cyanide, acid,
alkaline, and neutral electrolytes. Special bath compositions are used in the
continuous plating processes for strip and wire. Unalloyed coatings produced
by any of these techniques should have the same corrosion resistance, mass
for mass; some research workers, however, report differences that are likely
to be due to surface variations, structure of coating, or entrapment of corro-
sive materials. Alloy coatings are now being used increasingly: iron, nickel,
and cobalt additions are currently the most popular. All give increased
corrosion resistance, but the degree of increase depends on the environment
(see later). In particular, most accelerated tests on alloys, even in environ-
ments designed to simulate natural conditions, show greater improvement
than is eventually obtained in practice, compared with unalloyed zinc. Also,
ternary alloy coatings may not behave as expected from the performance of
the related binary alloys. Thus, Kotegawa et al. (1986) report that the
corrosion resistance and powdering property of Fe-containing Zn-Ni alloy
coatings on steel sheets are poor when the Fe content is more than 4.6%.
Nickel contents of up to about 15% are incorporated in a zinc coating,
usually by electrodeposition but also by mechanical coating. (The addition of
0.06-0.11 96 nickel in hot dip galvanizing, to suppress excessive Zn alloy
formation, does not influence the corrosion resistance: Van Eijnsbergen:
unpublished report).
Currently, more than 100,000 tons of steel is plated weekly in zinc-
nickel alloy baths, almost all of it continuous strip plating for car production
(Geduld, 1988). Coated strip is used under paint to give higher corrosion
resistance than pure zinc, but the degree of improvement may be less in
practice than is shown by accelerated tests. The coating is also applied to
fabricated articles to give improved corrosion resistance, without additional
painting.
A recent finding in zinc-nickel alloy deposition that may simulate its
use in conventional plating has been the report of the effectiveness of the
deposit in eliminating whisker growth (Baker and Holden, 1985). Whisker
growth is a phenomenon whereby plated deposits of pure metals tend to grow
spontaneously tiny filaments of metal on their surfaces. Whiskering may
occur months or years after the plated parts have been put into service and,
with the continued expansion of miniaturized, low voltage electronic applica-
tions for plating, can become a significant problem, causing momentary or
permanent shorts and failure of electrical equipment. Whiskering is a particu-
lar problem in the communications industry. The metal usually associated
with whiskering is tin, but the phenomenon also occurs in zinc and cadmium.
AT & T Bell Laboratories has investigated the use of zinc-nickel alloy baths
for this application, and it has been found that no whisker growth occurs after
5.5 years of use (Geduld. 1988). A modified ammoniated sulfate bath to
deposit zinc-nickel alloys on rack parts has been developed for this purpose.
Iron contents up to 50% are used in some electroplated coatings,
especially in Japan. In the United States, 12 and 18% iron alloys are used.
Work by Goodwin (1988) showed that the as-plated 18% alloy is composed of
a metastable phase. This transforms to equilibrium y phase when heated at
500C for 2 minutes.
Only one study on zinc-cobalt alloy baths was reported by Brenner
(1963), but today Zn-Co is the main alloy bath of the iron family, being
actively marketed on a worldwide scale for the bright plating of fabricated
parts. Essentially, it is believed that the formation of cobalt hydroxide delays
corrosion. Verberne and Hadley (1984) have a patent based on zinc chloride
with cobalt added as sulfate or chloride. Their preferred amount of cobalt in
the deposit is stated as 0.15-0.35%, but elsewhere 0.5-1.5% cobalt is used
(e.g., Eilers, 1987). At the lower cobalt concentration the alloy, when
passivated by conventional dichromating baths, is claimed to have the maxi-
mum corrosion resistance. Chromate receptivity, always a problem on coat-
ings from acid baths, is said to be good. More recently, alkaline zincate baths
have been introduced, designed to give 0.5-1.0% cobalt in the coating: this
process is claimed to be cheaper and easier to operate, but slow. Work at
Aston University (Shears and Hadley, 1989) indicates that corrosion re-
sistance is improved if more than about 1% cobalt is present, in which case
the coating can equal the 12-13% nickel alloys. Trials with substantially
higher cobalt contents ( 5 20%) do not seem to show enough increase in
corrosion resistance to justify industrial use.
24 Chapter
Zinclcobalt electroplating baths started in Germany in 1985. For zinc/
nickel baths, a new type-Zin-by-was introduced with improved precipita-
tion properties. The corrosion protection in accelerated tests was summarized
by Siegert and Hadley (1989), who used German national standards (DIN).
Salt spray test, DIN
50021 (hr to 5% rust) .
Kestemich test,
DIN 50018
Coating White rust Red rust (cycles to rust formation)
Zn/Co; 7.5 pm yellow
Zn/Ni; 7.5 pm yellow
Zn; 7.5 pm yellow
~~~
chromate coating 140-160 400600 6-8
chromate coating 2-3
chromate coating 96-120 4-5
In these tests, Zn/Ni and ZdCo show better corrosion protection than zinc
(thickness for thickness), but the salt spray test is not very reliable for
comparison of corrosion protection. The authors indeed indicate the need for
methods yielding a better correlation with atmospheric corrosion. Thinner
chromate layers result in poor corrosion protection.
Adaniya (1978) reported that cobalt plus chromium improved the corro-
sion resistance of electroplated zinc coatings. Small concentrations of cobalt
and chromium salts were added to zinc plating baths, and thin coatings
containing approximately 0.2% cobalt and 0.05% chromium were produced.
The chromium-cobalt alloyed zinc coatings resisted red rusting in salt spray
more than twice as long as conventional electrogalvanized coatings of the
same thickness. Although extrapolation of salt spray data to natural exposures
is at best risky, these tests do indicate that cobalt and chromium, in small
amounts, significantly improve the corrosion resistance of a zinc coating, and
the use of cobalt at 0. 143% in conjunction with similar quantities of other
elements, notably chromium, has been developed especially for continuous
plating lines. At a recent World Materials Congress, Kawasaki Steel appar-
ently described a Zn-CO-Cr coating containing particles of alumina. Corrosion
data on such combinations are awaited.
The main industrially applied coating of zinc containing some chromi-
um is the zinc-chromium-chromium oxide electroplate (a composite of Zn,
Cr, CrO, applied to steel in Italy for car production and is known by the trade
name Zincrox (Zinkroksid SPA, Italy).
Electroplated layers that incorporate other materials to give specific
surface properties represent a stimulating field of development. A major
example has arisen from the desire to replace cadmium plating of nuts and
bolts; most existing coatings are unsuitable because they do not have the
antigalling properties (i.e., low coefficient of friction) of cadmium. Co-
deposition of zinc and finely divided colloidal graphite is a possible alterna-
tive.
The zinclgraphite composite bath, which is based on a modification of a
conventional bright acid chloride was developed by a plating process supplier,
working with the Ford Motor Company. The deposit resembles a black
phosphate coating and is specified by some car manufacturers for black
fasteners used under the hood.
Combinations of zinc (usually electroplated) with inorganic or organic
materials that form a surface layer (e.g., the Zn-Ni based Durasteel) not
surprisingly perform well in many accelerated tests but, because such surfaces
are often likely to deteriorate with time, it is necessary to check these coatings
in the field (see, e.g., Goodwin and Piessen, 1988). Typically, Rizzi et al.
(1986) report a new electrodeposited zinc coating, Kolsyl Zn, containing
phosphates and silicon, which is claimed to provide superior paint adhesion.
Corrosion tests on coated 1010 and 304 steels in salt spray show Kolsyl to be
superior to regular zinc plating and cadmium coating. Bend tests show good
resistance to coating cracks and spalling. Cracking of type 304 stainless steel
in boiling 40%CaCI2 was prevented by Kolsyl coatings. American Iron and
Steel Institute (AISI) steels of types 316, A286, and AIS1 4140 all galled less
than uncoated material when given the Kolsyl treatment, and it is such other
properties that often merit the use of special coatings, especially on bolts,
rather than corrosion resistance, upon which judgment must be reserved until
long-term outdoor test results are available.
The range of electroplated coatings developed and used for car body
panels is both extensive and changing rapidly. This can usefully be illustrated
(Fig. 1.5) in relation to Sumitomo Metal Industries (SMI) products (1991).
The development continues. A recent survey showed 62 zinc (or zinc-nickel)
electroplating lines in 24 countries (vs. 161 hot dip galvanizing, including
galvannealing lines). The two groups of products are likely to maintain their
strong position with the auto industry. A paper by Soepenberg (1992) re-
viewed developments, including less commercialized systems. Table 1.5
summarizes automotive precoat systems, and many new installations now
exist to produce these coated steel sheets.
Shibuya et al. (1992) used accelerated corrosion and electrochemical
tests to show that a 2-5 g/m2 AI-25Mn laboratory-prepared coating over
26 Chapter 1
I
Durabi I I t y
Code
Trend
of
coated
steel
sheets
SY I
CCL
for
Autonobi le
SY I
ECL
for
Automobile
1975 1980 1985 Present - future
1 1 1 1 1 1 1 1 1 1 1 1 I I I I
I ; ; h: d; !: .
I ,.Code 1 Nordic . New Cuidel ine
5-1.5 Code 6-3 10-6-2-1
I ZHl541 I -. I Tough Coat I 7. 541
I
1977 CA.CI
I968 EC iSBO(Res tar t)EC.ZnN i
pzzG-t-I(Yakaro.a)ECLI
Modified for ZnNi 1984 EC.ZnNi .ZnFe
I (KashinaNlECL I
1983 21 1988 2nNI.TCD etc.
piLGi+-l
Modified for
Tough Coatm
*F
ZnFe
ZnNi
Tough Coat (or TC) = Organic Coniwsite Coating
= Iron rich zinc flash plating (Iron flash)
= Electrolytic zinc-iron alloy coated
= Electrolytic zinc-nickel alloy coated
GA = Galvannealed
GI = Hot Dip Galvanized
EG = Electro-galvanized
ZM .- Zincromelal R
Fig. 1.5
Fukui, 1991).
Trends of coated steel sheets for auto body production in Japan (Nomi and
Table 1.5 Main Automotive Precoat Systems
Coating system
Coated weight (g/m2)
Name Type Interior Exterior
Hot dip galvanized
Galvannealed
Galvannealed dual flash
Refinished
Electrogalvanized
Alloy plated
Alloy dual flash
Alloy plated
Alloy plated + organic
Triple plated
Prefinished
Hot Dip Galvanized
Zn
ZnFe"
ZnFetZnFeb
Zn + pm org
Electrogalvanized
Zn
ZnFe
ZnFelZnFelFeZn
ZnNi'
ZnNi + p,m org
Zn-Cr-CrOx
Zn + pm org
1
+
W- l 1% Fe.
bzn-(l0-20)% Fe.
'Zn-12% Ni.
Source: Resented at Galvetech 92 by Hoogoveas.
galvannealed sheet could give improved corrosion resistance and paintability,
formability, and weldability.
Laboratory work by Kanamaru et al. (1992) on (20 g/m2) acid electro-
plated Zn-(lO-lS)Cr coatings showed similar corrosion resistance to 70 g/m2
zinc in salt spray tests, but none of the more significant tests had been used.
An Fe-(15-20)Mn(wt-%) g/m2 electroplated top coat (which forms
hardly any powdery hydroxides as corrosion products) is proposed by Ahn et
al. (1992) to replace Fe-20Sn, for the outer layer over Fe-Zn in two-layer
coatings on painting, which is a problem with the standard two-layer coating,
i.e., in either case there is no zinc in the top coat.
of interest (because of their possible relation to performance
rolled zinc alloys) are tests on zinc-titanium Ti) and Zn-Cd-Ti
(0.09-0.18% Ti) coatings on St 10 steel substrates. However, the Cd-con-
taining alloy is not now used: it is interesting to note that upon weathering, the
alloy containing cadmium is white rather than dark gray. Results (Dylyaeva et
al., 1984) indicate that the alloying of zinc and cadmium coatings with
titanium increases their corrosion stability and changes the nature of corrosion
failure in comparison with unalloyed coatings. In atmospheric corrosion tests,
28 Chapter
the corrosion resistance of Zn-Ti coatings is almost as good as the more
expensive Cd-Ti coatings, which are less suitable for environmental reasons.
Zinc-plated coatings are usually passivated to prevent wet storage
staining, an unsightly form of mild corrosion that occurs where zinc surfaces
encounter heavy condensation. Historically, a simple chromate dip has been
widely used, but other treatments are being developed. Phosphates are often
favored when the dipped surface is to be painted.
Mechanical Coatings
Mechanical coating involves agitating the parts to be coated with a suitable
mixture of nonmetallic impactors (e.g., glass beads), zinc powder, a chemical
promoter, and water. All types of steel can be coated, but the process is less
suitable for parts heavier than about 250 g because the tumbling process
reduces coating thickness at edges. Mixed powders of zinc and other metals
are also used for coating. Thickness can be adjusted by choice of operating
variables and can match either electroplated coatings or some galvanized
coatings. Coating thickness varying from 5 pm to more than 70 pm can be
obtained. The coating is not, however, alloyed with the steel, nor does it have
the hard, abrasion-resistant iron-zinc alloy layers of galvanized or sherardized
coatings. useful survey is by Schroder (1975).
Zinc foil with an electrically conductive adhesive can also provide a
mechanical coating. It is particularly suitable for wrapping pipes and exposed
steel bars (Yamaguchi et al., 1988) or for touch-up purposes. Vapor deposi-
tion is another mechanical deposition technique (Jones and Laughlin Steel
Corporation, 1967; Butler, 1970; Helwig and Murray, 1975; Maeda et al.,
1988) and may be a duplex layer (e.g., zinc over aluminum: Jones and
Laughlin, 1972; Middlemiss and Hailwood, 1973). Fuller details of mechani-
cal coatings are given in Zinc Handbook (Porter, 1991).
Sherardized Coatings
The articles to be coated are tumbled in a barrel containing zinc dust at a
temperature just below its melting point-typically around 380C, but lower
in the case of spring steels. The zinc bonds to the steel by a diffusion process,
which forms a hard, even coating of zinc-iron compounds. In appearance, a
sherardized coating is dull gray and can readily be painted if necessary. Fuller
details, including an inspection guide, are given in Zinc Handbook (Porter,
1991).
The uniformity of a sherardized coating over complicated shapes
together with the abrasion resistanc-is useful for nuts and bolts which, with
proper allowance for thickness of coating, can be sherardized after manufac-
ture and used without retapping the threads.
The need to ensure even temperature within the barrels imposes a size
limitation on the process, which is generally used for small and fairly intricate
components such as fasteners, which require a uniform but relatively thin
coating. The thickness of the coating can be controlled by changing the
operating conditions. typical coating contains about 200 g/m2 of zinc,
which corresponds to a buildup on each surface of 30 pm (0.030 mm).
Coatings of half this thickness may be specified for indoor use or for short
periods out of doors. Reference to zinc diffusion coatings in Russian and other
work sometimes relates to a sherardized type of coating (i.e., starting from
zinc powder) and at other times relates to hot dip-galvanized coatings that
have been heat-treated-typically for 30 minutes at 650C. Upon weathering,
sherardized coatings often develop a dark grayish to brownish color, which is
not caused by rusting of the steel base.
Thermally Sprayed Coatings
Gas or electric heating is used to melt zinc wire or powder, which is ejected
from a gun into the workpiece. Wesnick (1991) reports on the lightweight
(1.2 kg) spraying pistols now available. new self-regulating nozzles
give efficient application. Today, arc spraying has gained great importance.
Typical characteristics of flame and arc spraying are summarized by Wesnick
as follows.
Characteristic Flame spraying Arc spraying
1. Liquefication of zinc, kgih 18 35
2. Energy costs for liquefying zinc, $/h Approx. 28 Approx. 2.5
3. Effective use, min/h 40 40
4. Effective use, kg/h 12 23
5. Energy costs, $/m2 2.33 0.11
($/kg equals $/m2
at 1 0 0 pm zinc thickness)
zinc
6. Spraying speed, m2/h for pm 12 23
7. Wage costs, $/m2 at 60 $/h 4.17 2.17
8. Total costs, $/mZ (no. 5 + no. 7) 6.50 2.28
The higher investment costs for arc spraying apparatus are rapidly assimilated
because of the higher performance, lower operator costs, and lower energy
costs associated with this equipment.
The semimolten zinc droplets coalesce, and some zinc oxide is present
at each interface between droplets. The zinc oxide forms only a small
30 Chapter
percentage of the coating, and electrical continuity is maintained both
throughout the coating and with the ferrous substrate, that full cathodic
protection can obtained. Fuller details of the coating, including an inspec-
tion guide, are given in Zinc Handbook (Porter, 1991). The Zinc Metallizers
Task Group (Greenwich, CT 06836) maintains an extensive library.
Adhesion to the basis metal is mechanical, being achieved by prior grit
blasting to Sa2Y5-3 of the substrate, i.e., to a near-white or white
surface on the scale originally developed in Sweden and now widely used
The zinc alloys sprayed most commercially are those with 15% alumi-
num (Leclerq and Bensimon, 1976), which have some of the merits of
aluminum coatings added to the merits of zinc. These alloy coatings, now
widely used in North America, Japan, Belgium, and France, were developed
after many trials of different mixed metal and prealloyed compositions, which
are mainly discussed in Chapters 2 and on atmospheric and water corrosion,
respectively.
Mixed metal coatings can deposited (e.g., using electric arc pistols)
by having two different metal wires between which the arc is formed. There is
also some use of zinc coatings over aluminum coatings (antisparking or
special corrosion needs) or aluminum over zinc coatings (more assured
adhesion to steel base, better combination of long-term corrosion resistance
and sacrificial protection), provided the A1 coating is sprayed very soon after
the spraying of the zinc coating. More than 150 pm of sealed aluminum is
desirable for long-term protection.
The sprayed coating contains voids (typically 10% but can up to 20%
by volume) between coalesced particles. These are of little significance as far
as protection is concerned, but the use of a sealer to fill the voids improves
appearance in service and adds to life expectancy: most important, it gives a
better surface for subsequent application of more sealing coats or of paints-
unsealed coatings can cause segregation of binder (by absorption in pores)
from pigment in some paints.
Zinc spraying is also used for local repair-both on site or in works-of
zinc coatings that have been damaged and when local removal has occurred
through machining.
Zinc spraying is also used to protect high strength aluminum alloys
(DEF 03, 1970-a UK Government Defence Specification).
(IS0 8501-1, 1988).
Coatings Containing Zinc Dust or Flake
Zinc paints consist of very fine dust and/or flake suspended in an organic or
inorganic medium. To get the benefits of cathodic protection from the zinc in
such coatings, the dry paint film must be capable of conducting electricity and
in electrical contact with the basis metal. To ensure good contact, steel should
be shot- or grit-blasted before application of zinc dust paints. Other things
being equal, the protection improves with increase in metallic zinc content in
the dried film up to approximately 94% by weight of zinc calculated on the
dried film; good formulation is critical, however.
The second stage of protection by zinc-rich paints is largely a function
of zinc corrosion products providing a barrier effect by blocking the pores of
the paint (Faidi et al., 1991), but it is also postulated by Evans and Mayne
944) and by Fawcett (1 985) that the deposit of zinc corrosion products acts
as a chemical inhibitor.
Paints based on silicate media, especially partially hydrolyzed silicates,
can also withstand high temperatures, sometimes up to 500C. Painting with
zinc-rich paints is sometimes referred to as cold galvanizing. This terminol-
ogy is misleading, since the chief characteristic of hot dip galvanizing, the
reaction of steel to form a chemical bond at the zinc-steel interface, is
lacking. The bonding of zinc-rich paints is, essentially, a physical not a
chemical process.
Zinc dust paints may be used alone for protection or as a primer
followed by conventional top coats. Zinc dust paints are also used as prefab-
rication primers: steel sheet or plate is grit-blasted and given a thin coating of
zinc-rich primer, then cut and fabricated with minimal damage to the paint,
which protects the steel during the entire construction period. Further paints,
either zinc dust or conventional, are applied when the structure is complete to
form the permanent protective scheme. To improve weldability and machin-
ing procedures, an addition of ferrous phosphide (Ferrophos) to the zinc dust
has shown good results in practice, without diminishing the corrosion re-
sistance of such weld primers.
Combinations of zinc dust (particle diameter 12-20 pm) and extrafine
zinc dust (particle diameter pm) yield optimal corrosion resistance. Small
amounts of aluminum paste or powder, added to obtain a more silvery
appearance, will however decrease corrosion resistance-as found in compar-
ative weathering tests by Van Eijnsbergen (personal communication). Gener-
ally, nonsaponifiable binders, such as vinyl copolymers, chlorinated rubber,
methacrylic resins, and epoxy resins, are used. For welding zinc-rich primers,
epoxy-polyamide binders are widely used. They allow a wider range of
sealers and top coatings.
Coatings that require baking to develop their properties form a separate
and important group, arising from early uses of baked zinc dustkodium
silicate mixtures. In the last 25 years, proprietary zinc dusVzinc chromate
combination coatings with water or solvent bases have been developed for and
widely used in the vehicle industry, while in the last 20 years zinc flake/zinc
32 Chapter
chromate combination coatings have also been widely marketed. Other coat-
ings are based solely on zinc dust or zinc flake pigmentation in organic resins,
often with lower baking temperatures (e.g., 185-210C), and in some cases
the heat can be introduced by preheating the work.
Zinc dust paints are also used to repair damage caused by corrosion,
welding, or mechanical damage to other types of zinc coating.
Zinc dust paints for air-drying applications are formulated either in
synthetic organic media, such as epoxy resins and chlorinated rubber, or in
silicate media.
Design for Coating
The design of any steel article should take into account not only the function
of the article and its method of manufacture, but also the limitations imposed
in the application of the desired rust prevention system. BS (British
Standards Institution, in nine separate sections covers the very different
requirements for coatings of different types.
Steel sheet, strip, wire, and tube are all galvanized in the semifabricated
state. Fabrication practice is the same as with ungalvanized material, and the
only additional requirement is to restore the coating (see later) to areas from
which it has been removed by, for example, machining during fabrication.
Similar comments apply to continuously electroplated sheet, strip, and wire.
The other zinc coatings are normally applied after fabrication.
BS Part gives design advice for fabricated parts that are hot dip
coated, including galvanizing. Immersion in molten metal calls for special
design considerations. The main recommendations can be summarized as
follows:
If sealed compartments cannot be avoided, they must be vented during
galvanizing; otherwise, there may be a serious risk of explosion.
The provision of holes for venting and draining tubular fabrications will also
allow a coating to be obtained on the inside surfaces and, therefore, will
ensure better protection for the fabrication.
The design should incorporate features that assist the access and drainage of
molten metal and ensure that air locks and remnants of flux are avoided.
The structure must be designed so that no narrow crevices are created.
The objects should be provided with lifting holes lifting eyes where tools
Inwardly projecting pipe connection stubs should be avoided in containers.
Pipe connection stubs are to be located in comers and diagonally; otherwise
the container cannot be completely emptied of zinc.
Materials of widely differing sectional thickness must not be combined in the
same structure; otherwise heating in the zinc bath will take place
can be applied.
unevenly and the object may warp. Long, slim structures should be
avoided.
Where the metal is thinner than mm, large, flat sheet surfaces need to be
stiffened.
Galvanizing after manufacture is a factory process, and the maximum
length that can be treated in most countries is about 15-20 m in a single dip-
possibly more than 20 m if one end is dipped first, then the other. It is
preferable to design a structure in easy-to-handle parts that are assembled after
hot dip galvanizing, normally by bolting. For example: the galvanizing costs
of a lookout mast and nest for a trawler were reduced by 50% when the
lookout nest and the supporting tubular pole were galvanized separately and
then bolted together. Alternatively, welding may take place after galvanizing,
but some restoration of the zinc coating (e.g., by use of zinc-rich paint,
special solder sticks, or metal spraying) on the weld is desirable after
fusion welding (although not necessary with resistance welding) (Adams et
al., In general terms, to give adequate performance, the coating
thickness applied by the special solder sticks and/or zinc-rich paint should be
about 50% greater than the original thickness of galvanizing.
Thermal (or metal) spraying is used for structures too large to
galvanize, and design advice is in BS Part 7. Metal spraying requires a
design that allows access of the metal spraying gun (which is about 300 mm
long), its supply tubes, and the hand holding it, so that the nozzle of the gun is
about 150 mm away from-and roughly at right angles to-all surfaces being
coated. Grit-blasting is an essential pretreatment; care is necessary to avoid
warping on steel less than 5 mm thick, and the normal process is not suitable
on parts less than mm thick.
Painting with zinc dust paints entails the same design considerations
applicable to all painting projects (BS Part 4). With zinc-rich paints,
adhesion is best achieved by blasting to at least Sa 2 !h, and limitations on thin
steel apply as with the metal spraying.
For zinc plating, the design guidelines are the same as for all plating
operations and are set out in BS Part 2. Not only must the electrolyte
reach all surfaces to be coated, but an adequate electrical current has to pass if
a substantial deposit is to form. Special attention is drawn to the risk of
embrittlement of any high strength steels that are electroplated; there may be a
need to heat-treat the assembly before and/or after plating. Mechanical plating
avoids this problem, provided acid pretreatment is not used.
Sherardizing and mechanical coating both involve a tumbling action
within a container and are not suitable for fragile parts very thin, unsup-
ported parts. Flat surfaces can give problems if two parts stick together and,
34 Chapter
consequently, do not get coated. Parts that tend to interlock accidentally when
tumbled together may be uneconomical to treat by these processes. Both
processes are used mainly for small parts. Mechanical coating is less satisfac-
tory on parts longer than about 15 cm and weighing more than 250 g.
Threaded parts are preferably hot dip galvanized and centrifuged to
maintain a good thread profile and give maximum coating thickness. Finer
threads .(e.g., on bolts 6 mm diameter) or thinner coating requirements are
best obtained by high temperature galvanizing at 55O-56O0C, sherardizing, or
mechanical coating. Electroplating is also acceptable but the coating applied
often is only a few micrometers thick (which is insufficient for corrosive
environments). Therefore, an adequate thickness must be both specified and
checked: the coating is also less abrasion resistant than galvanizing or sherard-
izing. Threaded fasteners cannot function well if the coating on the threads is
too thick. Typically 50 pm galvanizing is desired, whereas twice that thick-
ness may be present on the adjacent structure. The slip value of bolts is
important for allowing correct strength calculations of steel structures. Hydro-
gen embrittlement, which may be pronounced with electroplated bolts, stems
mainly from pickling and phosphating procedures. It can be avoided by
subsequent heating, observing time and temperature limits in relation to
strength and hardness of the metal.
Exposed parts of bolts require supplementary protection. Goodwin and
Piessen (1988) in 10-year tests in an industrial atmosphere showed that the
best duplex systems were galvanized plus either zinc dust/zinc oxide
paint, polyurethane paint, or anodizing.
In choosing a rust prevention system for a steel object or a steel
structure, many technical properties must be considered. Optimum corrosion
protection on the drawing board! The ultimate choice will depend on
many factors: size and shape, design, and economic considerations, as well as
the severity of the corrosion conditions. The environmental conditions exist-
ing on the site for the object or the structure must be ascertained. The
reliability of obtaining consistent coatings and the scope for checking their
quality must be explored, as must mechanical and environmental conditions
during transport, storage, and assembly, and the ease of maintaining the
corrosion protection in service.
The reliability of coatings, especially that of galvanizing over modern
paint systems and a classical paint system on steel, has been calculated by
Van Eijnsbergen (1978), who expressed the reliability of a coating by the
formula:
in which is the total sum of failures possible during coating selection and
application, surface preparation, and degradation of the coating during out-
door exposure. For hot dip galvanizing, an R value of 2.0-2.2 is obtained,
whereas a modem zinc-rich paint system on Sa 2Yi-blasted steel has R = 1.4-
1.6. The classical paint system on handlbrushed steel has only R = 0.8-1 .O.
Duplex systems generally have R values of 2.5-2.8
In addition, an economic valuation of various alternatives is important.
Then the choice of coating should not be made on the basis of initial costs
alone; future costs for maintenance should also be considered. The total costs,
of which initial costs are only a part, may reveal an entirely different picture
(see Section V. B in Chapter 2).
Since the publication in 1971 of the report by T.P. Hoar on corrosion
costs, other reports on this subject have been published (e.g., Bennett et al.,
1978). During the last decade also, attention has been focused on the recyc-
ling significances of coatings and materials. During the 1991 annual Corro-
sion Days, organized by the Dutch Corrosion Center, Soepenberg (un-
published work) showed the interdependence of the use of materials and
coatings and their recycling in the following formula:
c =
NM(1 - R)
D
C = quantity of material (kg/year)
N = number of products
= quantity of material per product (kglproduct)
R = recycling factor (0 < R < 1)
D = durability (years)
Regarding the durability, the reliability factor of a product plays an
important role:
R =
R = degree of reliability (0 < R < 1)
T = tempo of failure
t = time
Through anticorrosive measures, factor f increases and, therefore, lower
values for R result. Considering protective coatings, the important parameters
are the relatively large energy content and the limited recycling possibility of
the coating.
Zinc coatings, especially hot dip galvanizing, occupy a very important
position in the automotive industry, together with electroplated zinc coatings.
It is expected that in Europe, North America, and Japan the production of
zinc-coated steel will approach 15 million tons, which means that per region,
36 Chapter
550,000 tons of zinc will be brought into circulation annually. By the
increased use of zinc-coated steel, the durability expectance for 15 makes of
European cars increased from 8.6 years in 1965 to 13.9 years in 1980 (Elliott,
1991). Continuous hot dip galvanized steel can be relatively easily recycled,
unlike galvannealed sheet and custom-galvanized articles, as well as organic
coatings.
In comparing the technical characteristics of different types of coatings,
Fig. I .6 can provide some guidance (based on Thomas, 1980). Additionally,
paint coatings are normally thinner at sharp comers and powder coatings
usually are slightly thinner, whereas galvanized coatings tend to be thicker at
comers (Van Eijnsbergen, 1986). Then closer evaluations of apparently
suitable types of coating must be made. Paints, plastics, and enamels are
available in countless types with different properties in different respects,
which can alter the evaluation in some cases. Similarly, there are variations in
the suitability of the different zinc coatings, as shown in Fig. 1.7 where,
again, the letter A indicates greatest suitability.
Although zinc and paint are applied to steel for the same purpose-to
protect the steel against rust-they behave in completely different ways. Zinc
coatings corrode from the outer surface in toward the steel and provide
cathodic protection, which ensures that no rust forms between the zinc coating
and the steel (see Fig. 1.8). Conventional paint coatings often destroyed
by the formation of rust at the boundary layer between the paint coat and the
Rusl Proleclion
Electrochemical Prolection
DurabilRy in Atmosphere
Durabdiy in Waler
Resistance Against Damage
Resistance Io Abrasion
LmnalMns in Piece Sizes
of Debrmalion
Inspeclion Posslbililies
Aeslhelic Elfect
Maintenance Costs
lnilial Cos&
AdheSlOn
Vitreous
Zinc Pain it m n Enam I
11
111
I l l
1 1
I I
I I
121
121
1 1
1 1
131
l 2
None Very Low
Litlle LOW
Very High
High
Very High
High
Fig. 1.6 Properties of zinc compared with other coatings. (Based on
1980.)
Thomas,
Alloying Wdh The Steel
Durability 01 Coaling
Electrochemical Protection
Resistance Againsl
Mechanical Damage
Resistance to Abrasion
Limitations in Piece Sizes
Risk of Delormation
Ease of Inspection
Initial Costs
Maintenawe Cons
Suitability as a Substrate
lor Painling
Zinc Mechan-
Hm-Dip Zinc- Etecrro- Sherar Rich ical
Galvanii& Sprayinq platinq dizinq Paints c0arir-q
I t 1 A I D l D l A l 0 1 0
I I I A I A I c I R I r I R
I t 1 A 8 C A C c ,
Ill A B C A C C
C C A C
1 2 3
Mne very LOW
Little l ow
Hiqh High
Very High Very High
Fig. 1.7 Properties of different zinc coatings compared. (Based on Thomas, 1980.)
steel, and because rust occupies a volume approximately twice that of the
steel from which it has been formed, paint coatings are lifted from their
substrate-thus accelerating the spread of rust. The paint coat does not
provide cathodic protection, and after initial damage has occurred, rust can
penetrate further underneath the coat.
Aluminum and, to a lesser extent, high aluminum-zinc sprayed coat-
ings tend in many environments to retain corrosion products within the
coating matrix, and the coating as a whole eventually disintegrates over a
relatively short time, rather than by surface loss of metal.
~~
++
Anodic react i on: Zn -C Zn + 2e -
Cathodic r e ac t i on: 2e-+ H20 + 1/ 2 O2 - 2( OH) -
Fig. 1.8 Schematic view of the cathodic protection of zinc coatings on iron.
38 Chapter 1
Guidance on life to first maintenance is given in BS (British
Standards Institution, a Code of Practice for protective coating of iron
and steel structures against corrosion. The concept is to consider first the time
required before maintenance and then choose a system from the alternatives
offered for that service environment; one looks next at the notes relative to
each component of a protective system and then at the product specifications.
The full document (BS gives more details. The tables prepared from
this document by AM & S Europe are particularly useful and important and
are reproduced as Figs. and
D. Special Coatings on Zinc to Add Corrosion Resistance
Special coatings to improve the corrosion resistance of zinc fall into two
groups: (a) where the zinc surface is converted to another coating, notably in
anodizing and in chemical conversion coatings, and (b) where a coating is
applied that has synergistic properties with zinc-in the case zinc coatings,
the combination of paint and zinc is known as a duplex coating. However, the
term duplex coating has also been used with sprayed metal coatings (e.g.,
where aluminum spray on top of zinc spray can improve both heat resistance
and atmospheric corrosion resistance).
Anodizing
Four basically similar zinc anodizing processes were developed around
The first used an ammonium phosphatechromate-fluoride electrolyte (the
APCF process), gave an olive green coating, and is covered by U.S. Depart-
ment of Defense Specification MIL-A-S Three other anodizing
processes, based on the use of silicate electrolytes, were subsequently intro-
duced; these were designated SSC (light gray), SSCV (charcoal gray), and
SSCSMn (gray-brown) because of the use of chromium, vanadium, or man-
ganese compounds in the solution. Voltages up to 1W200 V are needed.
Applicable articles include those by Glayman and Weisenberger
The anodized coating is hard and abrasion resistant, with a moderate
degree of corrosion resistance and limited color. It is mainly used for specific
applications requiring a combination of frictional wear and corrosion-resistant
properties. The outer coating has pores and can retain lubricants. The silicate-
type coatings can resist cycles in a Taber abraser with wheels and
500 g loading, but the APCF coating resists only about 1000 cycles.
Castings containing aluminum, brass, copper, nickel, and steel inserts
can be anodized successfully. The ratio of dissimilar metal surface area to
zinc area should be a minimum, but experience with castings such as a
washing machine center post suggests that relatively large areas of aluminum,
brass, or copper can be tolerated.
Green-anodized, sand-cast ZA-12 and ZA-27 meet U.S. Military Speci-
fication MIL-A-81801 (176 hours in acetic acidhalt spray test) with a time
margin of 25%; die-cast ZA-27 also meets the test.
Conversion Coatings
Chromate conversion coatings are produced by chemical treatments which-
as the name indicatesconvert the zinc surface to form a complex surface
layer usually km thick, containing chromium hydroxide, zinc hydroxy-
chromate, and zinc chromate. Films range from thin with a clear and bright
appearance to thicker, yellow-iridescent, brown, or drab appearances. Ap-
pearance varies with bath formulation, process parameters, film thickness,
and substrates. Corrosion resistance increases mainly with thickness. Film
characteristics are given in Tables 1.6A and B.
Chromate coatings add corrosion resistance and also act as a bond for
paint and powder coatings mainly because a thin, tightly adhering layer on
the metal surface. In accelerated corrosion testing, the first sign of failure of
the conversion coating is often a blackish appearance locally. ,
The increase in corrosion resistance is significant when applied to thin
coatings, notably electroplated zinc and cadmium, and the benefits have been
discussed widely. Thus Royuela (1969) reports favorable results in exposures
in Spanish towns.
The choice of proprietary conversion coatings should be made in con-
junction with the supplier of the finish to used; modifications may also be
needed to achieve the best results on high aluminum alloys, where thinner
films tend to form (e.g., on ZA-27 vs. ZA-8 or IS0 Alloy 3).
Conversion coatings based on trivalent chromium were successfully
developed in response to the desire to avoid hexavalent chromium solutions
for health reasons; now, however, there is a desire to avoid chromates
completely. Hence, after galvanizing, quench chromating can no longer be
recommended. Phosphate conversion coatings are good as a pretreatment for
painting but because they are hygroscopic, on their own they have less
corrosion resistance than those based on chromates.
Treatments with varying amounts of surface reaction that have been
proposed include molybdates, tannin, titanium salts, and silicate solutions,
and also emulsion coatings of l ,2,6-hexanetriol trithioglycolate (ITH), which
has promising results regarding corrosion resistance. Chelating components
are particularly promising as inhibitors notably (Le Roy, 1978)
dimercapto-l ,3,4-thiadiazole and the alkyl esters of thioglycolic acid. There
Table 1.6A Comparative Characteristics of Paints and Paint Films on Hot Dip Galvanized Steel
Number of paints in Table 1.6B
Paint (film)
characteris tics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Application"
Dryingb
Hardening throughb
Hardness
Flexibility
Impact resistance
Gloss retention'
Color retentiond
Can stability"
Thermal resistance'
Weather resistanceg
Nld
marine
industrial
Resistanceh
acid solutions
alkaline solutions
B B B S s SIB
Lo Lo Mi Sh Sh Mi
Lo Lo Mi Mi Mi Lo
F F F G G F
E E E F I G G G
E E E G E G
G E E P F F
P F G P F E
G G E E E G
P P F P P F
E E E E E E
G G G E E F
G F G G E F
F P F E E P
P P P E E P
SIB
Mi
Lo
FIG
G
GiE
F
E
G
F
E
E
F
F
P
S s s SI B
Mi Sh Sh Mi
Mi Mi Mi Lo
GIE G G F
G G G G
E G E G
E P F F
E P F F
P E E G
G P F F
E E E E
E E E G
G G E G
G G G F
F G G P
S
Mi
Mi
G
FIG
G
P
F
P
F
E
E
E
E
G
S
Mi
Mi
FIG
FIG
G
F
F
G
F
E
G
F
F
P
s s
Mi Mi
Lo Mi
F GIE
G G
G E
P E
E
P P
P G
-
P E
G E
E G
E G
E E
5
0
-3
d
41
42 Chapter
Table Adhesion After More Than 3 Years of Weathering in a Clean Atmosphere
Adhesion
Cold or Sweep-
Freshly Weathered hot blasted
Type of paint (vehicle base) galvanized" galvanizedb phosphate galvanized
1. Alkyd-tung oil-phenolic resin F G E E
2. KO-alkyd-calcium plumbated
E E E E
Alkyd-acrylic combinations
G G E E
4. Chlorinated rubber F-G F-G G E
5. Chlorinated rubber-acrylic G G E E
6. Acrylate dispersions"
F-G F-G G G
7. Acrylic-styrene dispersions" G G G E
8. Acrylicldiisocyanate (2 comp.)
G F G E
9. Vinyl copolymers F-G F-G G G
PVClacrylic combinations G G E E
PVC-dispersions"
F F-G G F-G
12. Epoxy resin (2 comp.)' G G G E
Epoxy esteIg P F G G
Epoxy-tar comp.) P F F G
15. Polyurethane (2 comp.)h P F F-G G
Note: Variations in film properties may occur with variations in formulation.
"Up to about 4 hours after galvanizing.
bWeathered in an unpolluted or mildly polluted climate for months only.
'Precooked tung oil-alkylphenolic resin combinations, chilled with drying oil modified alkyd resins.
dDehydrated castor oil (KO)-modified alkyd resin, pigmented with calcium orthoplumbate (COP) as main
pigment in the priming coat.
'Finely dispersed polymers in water.
'With polyamide hardener.
*Epoxy resin-dehydrated castor oil ester.
hWith encapsulated disocyanate hardener.
Source: Van Eijnsbergen.
combinations'
combinations
are also a vast range of proprietary compounds, some with emphasis on wet
storage stain prevention and others designed as a pretreatment for painting.
Duplex Coatings
Organic Coatings Over Zinc Coatings. Today, there is an increasing
tendency to apply additional coatings on hot dip galvanized sheet and on hot
dip galvanized structures and articles. Various reasons can be given for such
combined organic/metallic coating systems, named duplex systems by Van
Eijnsbergen in the early Both aesthetic reasons (e.g., color, camou-
flage, visibility) and technical reasons (e.g., increased corrosion resistance)
lead to duplex systems. With the correct choice of materials and processing
techniques, the underlying zinc can extend the life of the paint, powder, or
plastics coating on top, while the coatings in turn can prevent the zinc from
corroding away-except where it does so to prevent rust when gaps appear in
the paint, plastic, or powder coating, either through wear or cutting through
the coating during the course of fabrication.
The selection of primers and subsequent coats is critical. Fujisawa et al.
(1992) report unusual 10-50 mm diameter corrosion patches under snow on
profiled roofs in Japan after 5 years. Attack on the polyvinyl difluoride (25
pm) over epoxy-primed pm) galvanized steel was worse when the primer
contained potassium chromate than when the less soluble strontium chromate
was present. Excess dissolved chromate at the primer-phosphate interface
produced excessive acidity, anodic dissolution of zinc, and evolution of
hydrogen. If was also found that when the top coat was too thick and rigid,
hydrogen could not diffuse out.
The synergistic effect of duplex systems is defined as the factor by
which the durability (protection period) of the Zn and ZdFe alloy layer
system plus the durability of the paint system has to be multiplied to obtain
this value for the combination of the two coatings. This factor varies with the
climate. The synergistic effect is estimated to improve the life by 100-150%
in a rural atmosphere and approximately 50% in polluted atmospheres com-
pared with the lives expected from the two coatings taken individually and
added together. The synergistic effect has been proven in environments with a
high degree of pollution, where zinc coatings alone often cannot offer suffi-
ciently long durability. Also, there is a trend to enhance not only total life but
also the appearance of objects. In practice, provided permanent good adhe-
sion exists, such systems will weather in the atmosphere as follows.
The synergistic effect is based on several developments. First of all, the
paint film will weather away almost to its last micrometer. Then the zinc and
zinc alloy layers start to corrode. Finally, local rust spots will appear on the
underlying steel surface, signaling the start of attack of the steel. As an
additional advantage, the slow deterioration of duplex systems can easily be
followed during their period of exposure, and sudden degradation does not
occur. Unsaponifiable paints, based on chlorinated rubber, vinyl copolymers,
polyurethane resins, epoxy resins, and acrylate resins are used. Both solvent
types and water-dispersible types are used. Also the solventless powder
coatings, based on polyesters, epoxy resins, polyurethane resins, vinyl co-
polymers, and polyvinylidene fluoride are used nowadays in increasing quan-
44 Chapter 1
tities, not only for technical reasons (e.g., increased hardness, erosion re-
sistance, impact resistance) but also for environmental reasons (e.g., no
volatile matter, no pollution). It should be emphasized that the zinc coating
must be at least 15 pm thick to afford the optimum synergistic effect during
outdoor exposure. A survey of duplex systems is provided by Porter (1991) in
Zinc Handbook, and a detailed book published by Van Eijnsbergen (1994) has
provided 1991 relative costs of duplex coatings (Table 1.7). Recent environ-
Table 1.7 Comparative Costs of Galvanizing, Painting,
and Duplex Systems: Netherlands
Parts
Relative cost:
note different units
in each group"
A. Heavy gage
(a) Hot dip galvanizing 100 (standard value)
(b) Grit-sweeping galvanized steel 23-35
80-85 (c) Grit-blasted (SA 2%) steel with epoxy
primer and epoxy MI 0 sealer
epoxy-tar top coat
(d) Same as (c) with epoxy primer and 86-98
B. Medium heavy gage
(a) Hot dip galvanizing
(b) Grit-sweeping galvanized steel 3540
100 (standard value)
(c) Same as A(c) 95-1 15
(d) Same as A(d) 105-122
(b) Grit-sweeping galvanized steel 50-85
C. Light gage
(a) Hot dip galvanizing
(c) Same as A(c) 195-230
(d) Same as A(d) 205-225
100 (standard value)
'Standard value: hot dip galvanizing = 100. Heavy gage material < 17 m2 per ton of steel.
Medium gage material 17-32 mz per ton of steel. Light gage material > 40 mz per ton of steel.
bAll figures calculated from actual quotations in the year 1991. Nor included: transportation
costs, reconditioning costs on building site, value-added tax percentages and environmental costs
on building site (sheathing, transportation of toxic materials; inspectors, etc.), and scaffolding.
All figures relate to relatively simple construction parts (e.g., profiles, tubes, floor grids, balcony
railings).
T h e average values for A, B, and C show that today a good paint system on grit-blasted steel
(SA 2%-3) is more costly than galvanizing if light or medium thickness steel is to be conserved.
Even on heavy gage parts, painting costs are nowadays equal to those for galvanizing. Further-
more, there are always extra costs in painting, as indicated in the introduction.
Source: Van Eijnsbergen (1994).
mental regulations have caused the costs of paints and painting to rise sharply;
this increase has affected multicoat paint systems more than duplex coatings,
which use less paint and have lower paint costs.
Prepainted sheet, using galvanizing, Galfan, or Galvalume as a base
and applying the organic coating at the steel mill or in specially constructed
coating lines, is a fast growing industry. Table 1.8 lists some of the common-
ly used coatings. Further descriptions are given in Zinc Handbook (Porter,
1991). Prepainted hot dip 55% aluminum-zinc coated steel sheet is covered in
a technical brochure prepared by the European producers of 55%Al-Zn coated
sheet (BIEC, 1991); in addition to technical properties, it covers information
on selection, fabrication, tolerances, and test methods for coils up to 1650
mm wide and thicknesses up to mm.
Clean zinc-coated steel is a suitable base for paint or adhesive systems,
but the first treatment may be different from those used on uncoated steel.
Chemical conversion coatings (chromate, phosphate, or oxide types) and
primers have been specially formulated for first treatment of zinc-coated steel.
Some passivated surfaces (e.g., chromated) are not suitable for phosphating
Table 1.8 Commonly Used Coatings Over Zinc-Coated Sheet
Coating thickness
Coating materials (pm per side, primer included)
Liquid coatings
Polyesters
Acrylates, methacrylates
Silicon-modified polyester
Epoxy resins
Polyurethane resins
Polyvinylidene fluoride
Vinyl polymer based plastisols
25
25
25
15
25
25
80400"
Special coatings
Laminates'
Heat-resistant, non stick systems
Acrylate films 50-75
Vinyl films l00-30Od
Fluorocarbon films 40
"Inclusive of any embossing.
bNo details of binder(s) given.
CExcluding adhesive and primer thickness.
dInclusive of any embossing with vinyl plastisols and vinyl laminates.
Source: Based on Eurononn 169 (European Coal and Steel Community, 1985).
46 Chapter
or the application of pretreatment (etch) primer. Chromating solutions must
not contain silicates. Zinc or zindcalcium phosphate, together with an accel-
erator (e.g., nickel and titanium salts) are mostly used in combination with a
chromate rinse to seal remaining boundary areas of the zinc phosphate layer,
at the same time reducing the hygroscopy of the phosphate layer.
Conversion coatings are also widely used to improve paint adhesion. In
the car industry, paint adhesion on galvannealed coatings, although more
tolerant initial surface pretreatment than with conventional galvanizing,
can still be improved. Most treatments involve what is called trication
phosphating (because usually zinc, nickel, and manganese cations are in the
treatment), typically processes are designed to give a 1.5 g/m2 coating. High
nickel tends to improve adhesion, and Renault recommends 2.5 g/L nickel,
0.2g/L manganese, and 0.5g/L zinc. Adhesion is very specific to types of
paint. Thus, Okumura et al. (1992) report that although alkyd-melamine
epoxy adheres well over conventional chromate treatments, the simpler
alkyd-melamine does not, and incorporation colloidal silica in the chro-
mate is needed to assure adhesion.
Environmentally, paint treatments free of nickel, hexavalent chromium,
and nitrite are desirable. Mabuchi et al. (1992) propose 0.2-2.3%Cr in the
Zn-( 12-14)Ni coating so that chromate-free solutions can be used for condi-
tioning before painting. Gehmecker in discussion amplifying the Gehmecker
and Kurzmann (1992) paper, proposed zirconium fluoride to replace chromi-
um (111 or VI) in treatments that can either be rinsed or left unrinsed.
Renault (1992), working with Sollac, has done service trials on vehicle
skin panels based on two-sided hot dip galvanized steel (1G14 pm coating)
produced at similar cost to one-sided electroplated zinc. High nickel phos-
phating gave good paint adhesion and stone chipping resistance.
Tests by Shiota et al. (1992) on organic composite coated steel sheets
(comprising 20-40 g/m2 Zn-Ni specially chromated and with 1 pm organic
coating) understandably demonstrate better corrosion resistance than zinc or
zinc alloy coatings without an organic film.
A sheet coating newly developed by Odashima et al. (1992) and made
by applying emulsified olefin-acrylic acid polymer resin mixed with colloidal
silica (7-8 nm diameter) has improved strength, ,formability, and corrosion
resistance.
A laboratory zinc-polymer coating developed by Hisamoto et al. (1992)
permits direct painting without phosphating. The preferred polymer was
based on acrylamide modified by copolymerization of epoxy groups with
acrylamide derivative monomer; for best corrosion resistance, modification
should exceed 50 mol%.
Paint delamination from scribe marks is less on Galfan (Schwan et al.,
1992) but more on Galvalume (Izumi and Hashizume, 1992) than on zinc-
coated substrates.
Powder coatings are nowadays applied over chromated or sweep-
blasted zinc surfaces. Generally, polyester-based powder coatings are used.
When a high degree of wetness will be encountered, an epoxy resin based
priming coat followed by the polyester top coat is applied. Recently intro-
duced powder coatings based on polyvinylidene fluoride offer very high
chemical resistance after baking at 190-220C.
Powder coatings offer increased hardness and resistance against wear
and impact in comparison to most wet paints. Proper application is critical,
and the pretreatment required to get good and lasting adhesion of the top coat
varies with the types of coating used and the nature of the zinc surface. For
greatest reliability, both zinc coating and painting or powder coating should
be carried out on the same premises, but with a separate, positively clean, and
environmentally controlled room for the painting or powder coating.
Thermoplastic coatings are also extensively used on galvanized steel.
Specially formulated paints for direct application to zinc-coated surfaces are
also available. Compatibility of the nonmetallic coating with the zinc coating
must be assured; many coatings that give good results when applied directly to
steel are less suitable or unsuitable for a zinc surface because of its greater
reactivity. If good pretreatment procedures are adopted and appropriate paints
used, zinc-coated steel is no more difficult to paint than other metal sub-
strates, but the techniques may differ. Table 1.6, which gives comparative
characteristics of paints and paint films on hot dip galvanized steel, also
compares corrosion resistance of the applied coatings.
Van Eijnsbergen (1991) notes that adhesion of epoxy resin based prim-
ers for hot dip galvanized steel can be noticeably improved by additions of
silane- and siloxane-based products. Interreactions take place at the zinc-
paint film interface, resulting in stronger adhesive bonds, which also remain
intact upon exposure to wet atmospheres. The addition of such compounds is
critical and encompasses percentages of (by weight of the total paint
formula). Extensive accelerated weathering and immersion tests show perma-
nent good adhesion of such primers, with or without a polyurethane resin
based top coat.
Environmental restrictions are affecting both the pigments and the
media that can be used. Special interest attaches to water-based paints, often
selected to avoid the problems associated with solvents. In all polymer
dispersions used as a vehicle for water-dilutable paint products, the polymer
(generally an acrylic resin, sometimes copolymerized with styrene; or a vinyl
Chapter
chloride copolymer) is very finely and permanently dispersed in water. After
pigmentation and the addition of a coalescing agent, the paint is thinned to
spraying viscosity with water. The pH of such paints, 7.5-8.5, is especially
favorable for hot dip galvanized surfaces. Van Eijnsbergen (1987) reports
good adhesion and avoidance of blistering when the paints are applied to
transmission towers. The paints are not recommended for immersed use, and
for aggressive atmospheres a compatible solvent-based top coat should be
used.
Paints pigmented with a mixture of 80% zinc dust and 20% zinc oxide
and containing a drying oil-modified alkyd binder or a phenolic resin binder
are recommended by Maass (1987) for application to galvanized steel. These
paints have a long history of use in North America. It is emphasized that good
adhesion is reached only after several weeks. The use of wash primers on zinc
has decreased considerably. Various reasons for poor adhesion are described
and means of reconditioning are mentioned by Maass. The zinc layer, in
addition to its barrier corrosion resistance and its cathodic protection of any
steel that may become exposed, also prevents undermsting of the paint (or
plastics) overcoat. Paint coat undermining can still occur after some time, but
it is less pronounced when the zinc is thicker-as is demonstrated (Fig. 1.9)
by accelerated tests with different thicknesses of zinc under the paint (Gessner
et al., 1988).
A brief survey of paints to be used on Galvalume (Aluzink, Aluzinc,
Zalutite, Zincalit, Algafort) is presented in a BIEC publication (1991) togeth-
er with a short survey on the coil coating process and types of coil, as well as
on test methods for coated Galvalume. Polyester-based coatings have good
flexibility and hardness as well as good gloss and color retention. Acrylate
and methacrylate resins yield flexible, decorative coatings with excellent
gloss and color retention. Silicone-modified, polyester-based coatings have
excellent outdoor performance with good flexibility and high chemical re-
sistance. They are predominantly used as a vehicle for primers and for
backing coatings. Polyurethane-based coatings have good flexibility and
excellent gloss and color retention and excellent weathering characteristics.
Microsection examination and electrochemical impedance spectroscopy
paint resistance determinations by Barreau et al. (1992) have established that
paint heterogeneities and cracks are the major corrosion-initiating factors in
prepainted coated steel sheets. Paints with good forming characteristics also
beneficially fill any cracks in the metallic coating substrate. With Zn-55A1 or
Zn-30A1, corrosion of the substrate (in outdoor tests of up to 3 years) occurs
by selective dissolution of zinc-rich phases. Similar effects can occur with
Zn-SA1 if P-phase zinc develops as a result of an inadequate cooling rate.
49
cn
c
.r
c
.r
E
L
W
V
c

CI
rd
0
V
c,
c
m
0-
.r
U
0 2.5 5 7. 5 10 12.5 15
Zi nc c oa t i ng , w
Fig. 1.9
(Gessner et al., 1988).
Undermining of paint coatings on zinc-plated sheet: VDA test 621-415
Poor edge creep corrosion resistance of prepainted Zn-55A1 coated steel
was shown to be due to the combined effect of aluminum and chloride ions. A
primer is needed to supply sufficient chromate ions and maintain the pH level
by buffer action to reduce the problem.
Zinc coatings are equally important as a substrate to powder coatings or
ceramic coatings. Zinc alloy coatings may also be used. However, it should
be borne in mind that solvent-containing coatings often cause blistering and/or
cratering after the powder coating has been baked at 160-20O0C. An extensive
article on powder coatings on galvanized steel appears in OberJ2iichentechnik
in German (1992). Arsenault et al. (1989) have shown that plasma-sprayed
ceramic coatings such as Cr203 and AI2O3.TiO3 deposited on steel without a
Zn-Ni underlayer were susceptible to corrosive attack and loss of bond
strength with time. They used a Zn-Ni coating underlayer to provide corrosion
Chapter 1
protection to the substrate, which can then resist attack by chemically active
ions passing through pores, such as chloride ions in saltwater. The zinc-
nickel coating provides protection against corrosion by mechanisms such as
differential aeration and galvanic cells that are related to the porosity of the
top coating.
An alternative technique is to use zinc soaps (e.g., at 200C) to impreg-
nate porous coatings such as sintered iron parts (Kovarskii et al., 1988)
which, it is said, gave results comparable to coated steel after one year in a
marine environment.
Organic Coatings on Solid Zinc. Castings are the main form of solid
zinc to receive supplementary treatments: in addition to corrosion resistance,
which is relevant in most cases (except copper plating and vacuum aluminiz-
ing), the choice may be based on wear and abrasion resistance or appearance:
the three factors are considered in Table 1.9. The metallic coatings are all
conventional and do not need to be described here. The chemical finishes are
specifically related to zinc; in particular, anodizing of zinc is substantially
different from anodizing of aluminum (see Zinc Handbook, Porter, 1991, pp.
516517).
The organic finishes used are similar to those used on zinc-coated steel
and the information in Table 1.6 is largely applicable. As with zinc-coated
steel, abrasion resistance is critical to good performance.
Some paints and powder coatings can be applied directly to degreased
castings, notably acrylic-based paints containing an acid etching ingredient.
However, most castings receive some pretreatment, either a chemical coating
or an etch primer- paint containing active ingredients that react with the
underlying zinc. The paint manufacturer should advise the buyer on compati-
bility of a paint with zinc alloy and the preferred pretreatment.
Paint media in common use on zinc die castings include acrylics,
alkyds, alkyd-melamines, epoxies, and phenolics, the choice depending on
the use of the finished components. Epoxy amines and epoxy phenolics
provide good corrosion resistance. Textured and patterned paints can also
be applied where required. Airless, electrostatic, or conventional spraying
and dipping are al l used for painting zinc die castings. Care should also be
taken that any stoving recommended will not damage the pretreatment.
Clear organic coatings have been developed to preserve the attractive,
bright, as-polished, zinc, zinc alloy, or zinc-coated surfaces. The coatings
may be either acrylic or polyurethane lacquers or polyester powders and may
be colorless or tinted, usually to simulate brass. They usually incorporate
ultraviolet absorbers and a complexing agent for zinc (dithiooxamide). They
have only a limited life and, in the atmosphere, usually must be renewed after
3-10 years (depending on the environment) to retain appearance.
52 Chapter
Several Japanese and Korean papers deal with the problem of finger-
marking of panels used in home electric appliances: on chromated 20 g/m2
electroplated zinc, Kawanishi et al. (1992) described the use of 1 pm acrylic-
silicon; Kim (1992) noted a one-coat/one-bake aqueous inorganic plus organic
solution; Tajiri et al. (1992) cured a chromate-polyester at 210OC; while
Yoshimi (1992) has an intense, weldable black-dyed film (1-2 pm), but this
technique has not yet been used in practice.
Another method of applying organic finishes to zinc die castings is
electrostatic powder spraying. Degreased castings, gently heated to about
60C to ensure that they are completely free from moisture after chromating,
are sprayed electrostatically-generally with an epoxy or polyester powder.
The parts are immediately stoved for a few minutes at 190-220C, to flow
the powder particles-producing an even, cured plastic coating. Spraying
conditions can be controlled to give coating thicknesses in the region 50-125
pm. Similar finishes can be achieved in a fluidized bed. These powder
coatings are tough and adherent and have good corrosion resistance. Epoxy
coatings generally have the best corrosion resistance; polyester coatings are
the toughest.
Reactions at the Interface. Inasmuch as corrosion performance is
critically related to good adhesion, it is essential to consider factors affecting
adhesion of the applied coating to zinc, whether it be a coating, a casting, or
sheet. Papers presented by Van Eijnsbergen at the Adhesion Symposium of
the Netherlands Corrosion Center in 1991 dealt with the physical and chemi-
cal aspects of adhesion of paints to galvanized steel surfaces, while Frens,
Rijkhoff, Dekker, and Wijdenes (1991) gave new insights on paint adhesion
and on reactions between the paint film and the substrate (steel, aluminum, or
zinc).
Interfacial reactions between galvanized steel and paints have been
summarized pictorially (Fig. 1.10) by Van Eijnsbergen (1988, 1991), but
many compounds and basic carbonates can form on the surface of zinc and
can react differently with applied coatings. For example, the fluffy, volumi-
nous P-Zn(OH), (a major component wet storage stain) can give problems
if not removed. Particles of lead or alumina in the surface layer adjacent to the
zinc and zinc compounds can produce electrolytic cells or differential reaction
on the surface the zinc; such a reaction is often harmless but, in conjunction
with paint, powder, or plastic coatings inadequately formulated for applica-
tion to zinc, can engender reactions that either affect adhesion or create gases.
The gases, in turn, are harmless unless they are formed beneath an organic
layer that is too impervious to allow them to escape: thus trapped, they result
in blisters or pinholes; in some cases, craters can form in the outer coating.
Depolymerisation Products
I Formation of zinc oxide and zinc hydroxides on alvanized steel.
Reactions of hydroxides, water, traces of flux; f ur t kr reactions with
I pollutants dissolved in water and traces of solvents with decomposition
products and their free radicals. Electrolytic cells between zinc and
aluminum oxide.
I
1 .l0 Interfacial reactions paint and galvanized steel (Van Eijnsbergen, 1988
and 1991).
Crystal size and orientation can affect the type of surface films, hence
adhesion of paints. It is thought (CRM, 1983) that some effects of specific
elements in the zinc coating that have been adduced by various investigations
are only indirect effects, working through their influence on crystal size and
orientation.
Treatment of the zinc surface to give a homogeneous nonpowdery
phosphate, chromate, or oxide coating removes the original variability but
Chapter
introduces new problems, such as overly large zinc phosphate crystals andlor
incomplete reaction of the zinc surface. This coating needs to be only a
fraction of a micrometer thick, but practical operation often gives much
thicker coatings, which can have cohesive weakness and may retain air, gas,
or unreactive material or form in an unbalanced reaction to give a powdery
coating. Washing down to remove powdery material or untreated solution is
beneficial only if repeated with spray and dip and if nothing is deposited from
the wash water. The surface must be thoroughly dry before coating. For the
same reason, initial quenching of galvanized steel in water or in solutions
containing wet storage stain preventives is less satisfactory than air cooling or
a chemical treatment. Deposits of calcium carbonate (from the quenching
bath) or silicates (originating from chromate treatments) must be avoided in
any case. Minor additions to the hot dip galvanizing bath, such as nickel, have
no apparent effect on paintability or choice of paints. Zinc-aluminum coat-
ings benefit from positive pretreatment to deal with the alumina layer, partic-
ularly if it is partly hydrated-as can apparently occur on the 55% aluminum
alloy (Van Eijnsbergen, 1987a, b).
Reactions at the organic-zinc coating interface are also significant.
reaction that is particularly to be avoided because of its adverse effect on
adhesion is that between zinc and formic acid, (formed, e.g., when fatty acids
are depolymerized within a drying oil modified alkyd resin). Epoxy esters
suffer less from this problem; an effective acid scavenger in such products is
magnesium carbonate, but it makes such products less suitable for acid
resistance. It has been observed-particularly on galvanized steel sheet,
because of the relatively high percentage of aluminum in this
coating-that paint adhesion suffers when zinc oxide and aluminum oxide are
present in adjacent positions on the zinc surface. When water diffuses through
the paint film toward the zinc surface, the presence of these two oxides cause
the formation of an electrolytic cell in which the zinc oxide is the anode. Then
interfacial reactions occur, accelerating blistering at the interface. When these
two oxides are on top of each other (mostly the zinc oxide on top of the
aluminum oxide), no detrimental influence is observed.
More generally, acidic depolymerization products can be leached out of
many films of paint, unless suitable pigments are incorporated. Zinc dust plus
zinc oxide was the original acid scavenger. Zinc phosphate, zinc femte, and
calcium borosilicate will each exert an anodic inhibitive function at pH values
exceeding 7 while zinc chromate, zinc tetroxychromate, and strontium chro-
mate act independently of pH value (Van Eijnsbergen, 1988). In general, the
pH of aqueous paint film extracts should be between 7.5 and 8.5 and certainly
not below 6 or above 10.
Outdoor testing of duplex coatings takes too long-new organic prod-
ucts are available before the old are fully assessed. Hence, accelerated tests
are used to give some guidance. Schwenk (1991) has described some of his
tests for duplex coatings.
Pigments such as calcium orthoplumbate will oxidize the zinc surface
and also form calcareous deposits with good bond properties at the interface.
A similar but weaker effect has been observed by Van Eijnsbergen when
primers containing red lead, zinc chromate, and basic lead sulfate are applied
to zinc surfaces.
Since paint films are pervious, water can penetrate in service to the zinc
substrate and react to form hydrogen, which will form pinholes if the pores in
the paint film become blocked. The solid corrosion products have very much
greater volume than the zinc from which they are formed (steel corrosion
products are only about twice the volume of the steel from which they are
formed): while this property can be beneficial (by plugging scratches and gaps
in the organic coating), excessive attack causes peeling or blistering of the
organic coating by pushing it away from the zinc. Over a period of 25 years of
outdoor weathering of painted galvanized test panels, it has become evident
that attack does not creep sideways in and around scratches (made to assess
adhesion), provided the original good adhesion has been maintained. On
panels where the zinc coating is less than 15 km thick (e.g., continuously
galvanized sheet), rust is formed in such scratches but, again, rust spreads
sideways from these scratches not at all, or only very slowly. At scratches,
alkali is generated by the cathodic reaction of the protection process, and this
alkali can cause further loss in paint adhesion. This effect can occur when
painted zinc-coated steel and painted black steel are in direct contact, as was
found in some car bodies. Zinc-iron (galvannealed or sherardized) coatings
under the same conditions give neutral corrosion products at the scribes. The
iron present is much less reactive than zinc and also helps to remove the
unwanted acidic products of applied paints: the 6-1 alloy layer present on
galvannealed sheet reacts only very slowly, either in alkaline or weak acid
attack. There is no need for chemical pretreatments unless on some areas the 5
alloy layer is present.
New developments in Germany of fabrication, passivation, and painting
continuously hot dip galvanized steel sheet are discussed by Sander and Mady
(1989). Zinc flowers (spangle) are present when the lead content is higher
than 0.15%. Below lead, no flowers are formed. Pretreatments of the
sheet in coil form nowadays are as follows: (a) zindmanganese or zinchickel
phosphating and no-rinse chromating by Cr03-Cr203 solutions; (b) alkaline
passivation, based on the precipitation of mixed oxides of the spinel-type
ZnCqO, and CoFe204. Above 1 g/m2 zinc phosphate layer, deformability
decreases. In this respect, Zn/Ni-phosphate layers perform better. Because
56 Chapter
chromium in effluent water is undesirable or forbidden, no-rinse technologies
have been advanced, especially in the United States and Japan. The chromate
solution (15 mum2) is applied to the coil by rubber rolls. After 5 m, at a coil
speed of 200 m/min, the coil enters a drying tunnel and leaves at temperatures
between 55 and 60C (reverse coil coating); Cr-free, no rinse processes have
also been developed-a third type of pretreatment.
Duplex Coatings (two metal layers). A layer of zinc followed by a
layer of another metal (usually aluminum) is occasionally advocated to pro-
vide a barrier or more slowly corroding surface initially or to give specific
properties such as reflectivity.
Republic Steel (1965) patented the application by vacuum deposition of
a thin layer of aluminum to conventionally zinc-coated sheet to enhance
corrosion resistance. This is discussed by Jenkins et al. (1969).
In the thermal spraying industry, a processing benefit of applying zinc
first and then aluminum is that the quality of blasting pretreatment required
for zinc spraying is less critical than for aluminum, especially when gas
pistols are used. Such duplex coatings have been used on Royal Navy ships
decks, for example, and also in hot areas, such as near coke ovens, where
aluminum helps to reflect heat.
Other Coatings
For other coatings, aluminum vapor deposition (2-15 g/m2) was shown by
Shin et al. (1992) to be more protective, thickness for thickness, on galvan-
nealed substrates than on zinc-iron electroplating. At 2 g/m2 aluminum, salt
spray corrosion resistance was five times better; more meaningful tests were
not conducted. The aluminum also resisted limited cycling to 500C.
Vapor deposition remains only a possible commercial alternative for
coated steel for cars. High deposition rates and the ease of producing duplex
coatings (zinc followed by aluminum or vice versa) with improved coating
resistance are potential advantages, but engineering for high speed line use
remains to be developed.
Vacuum-deposited silicon (Jun et al., 1992) gave interference-colored
surfaces on galvanized steel which, like many other coatings, increased salt
spray life (but again, meaningful corrosion results are not available). Good
adherence was obtained with fingermark-resistant polymer coatings.
E. Zinc Sacrificial Anodes
Principles of Cathodic Protection
Iron and steel structures that are buried or immersed may be protected against
corrosion simply by making an electrical connection between the ferrous
material and slabs of zinc-zinc anodes. Unlike other methods of protection
with zinc this process of cathodic protection does not rely on any part of the
iron surface being coated. Morgan (1959) published an extensive treatise on
the subject and BSI Code of Practice 7361, Part 1 (British Standards Institu-
tion, 1991) and booklets are available relating to use of zinc by the Zinc
Development Association (ZDA) (1983a) and by the Centre Technique du
Zinc (1973), with a paper by Heurtaux (1979) also in French. Kurr (1979)
summarizes the knowledge for zinc.
When iron is joined electrically to zinc under immersed conditions, all
the iron becomes cathodic and does not corrode, while the zinc becomes
anodic and corrodes in preference to the iron to which it is joined, hence
cathodic protection.
Sacrificial Anode Specifications
Zinc was the first metal used for cathodic protection, beginning nearly 200
years ago, but the impure and variable qualities of zinc then available gave
inconsistent results. Sometimes, dense adherent layers of corrosion product
formed on the zinc surface, blocking the current, stopping the protective
action, and so allowing corrosion of the iron to continue. Today, high purity
zinc is readily available to overcome these problems: in particular, the iron
content of the zinc must be below 0.0015%. Polarization problems can arise
in the presence of copper, lead, and iron, and Vassilev et al. (1974) studied
the use of aluminum l%), cadmium, and mercury to overcome any
adverse effects of iron. However, Shams El Din et al. (1989), in studying the
reversal of potential of zinc and iron at elevated temperatures, suggested that a
controlled amount of iron can help prevent zinc passivation (and potential
reversal) at elevated temperatures. Lyublinskii (1970) studied the corrosion
resistance of zinc containing 0.19-2.14% aluminum together with calcium,
magnesium, manganese and related it to the amounts of and phase
present. Zinc-indium has been proposed (MHO, 1973) for the inside of tanks
to combat seawater or crude oil. The zinc may, therefore, be alloyed with
small amounts of other metals, particularly aluminum and cadmium and
sometimes silicon (Table 1.10) when slightly more iron is permissible. Such
anodes, however, have susceptibility to intercrystalline corrosion, particularly
above 50C (Haney, 1985), and for these temperatures the purer anodes
(ASTM B418 Type 2 in Table 1.10) are preferable.
Design Information
The design of zinc anode cathodic protection systems is relative straightfor-
ward (Kurr, 1979) once it has been determined that zinc will function
satisfactorily in the intended environment. If zinc is chemically compatible
with the environment (soil, water, industrial effluent, etc.) under the relative
58 Chapter 1
Table 1 . I0 Zinc Anode Compositions
Anode material ASTM B418-88
Aluminum 0.1-0.5 0.0005 max
Cadmium 0.025-0.07 0.003 max
Iron, max 0.005 0.0014
Lead, max 0.006 0.003
Copper, max 0.005 0.002
Others (total), rnax 0.1
Zinc Remainder Remainder
velocity envisaged (if any), the corrosion engineer can make a series of
calculations that will indicate whether zinc can be used in a particular
situation.
The anode current output can be calculated by first determining the
anode resistance for relatively long slender anodes from the following Dwight
(1936) equation:
where R is anode resistance (a); p is electrolyte resistivity (Cbcm), L is anode
length (cm), and A is anode radius (cm)
Using Ohms law, the current output is calculated:
where I is current output (A), E is driving voltage (V), and R is anode
resistance (a).
Dugdale (1973) deals with anodes in parallel as well as with single
anodes.
The difference in potential between the zinc anode, normally - 1.1 V,
and that of the cathode being protected is symbolized by E. A value for
protection of steel is usually -0.85 V and, in this illustration, E = 0.25 V. A
generally accepted value for polarized copper in soil is - 0.4 V, and E in this
couple is 0.7 V. The life of a zinc anode system can be calculated using the
following equation:
W X 0.85
CO X CR
L =
where L is life (years), W is the mass of zinc, CO is current output (A), and
CR is consumption rate, which is a constant, 24.8 Nyear for zinc. Any mass
unit for zinc can be used, but the same unit must be used in both cases. The
constant of 0.85 is a utilization factor given arbitrarily to reflect that the
effective life ends in practice after 85% of the anode mass has been con-
sumed.
When the total current requirements of a system are established from
appropriate criteria for similar systems, proper selection of zinc anode sizes,
weights, and numbers can be made utilizing the equations and details just
stated, as well as appropriate references.
Marine and buried anode systems are calculated similarly, taking into
account the lower resistivity of the environment in marine conditions.
Use in Conjunction with Barrier Coatings
Even the best of paint or polymer coating are in some way porous, so
corrosion of the underlying metals must ultimately occur. This may take
several years with the best coatings, but eventually the corrosion destroys the
adhesion between the metal and coating, allowing rapid deterioration to take
place. The use of zinc anodes in conjunction with a protective coating
enormously improves the situation. The corrosion that undermines the coating
is prevented, and the coating remains firmly attached to give its maximum
life. At the same time, the consumption of zinc is reduced dramaticaily-
typically by a factor of 10-because much less steel is exposed and needs
protection: annual zinc consumption of less than about 0.12 kg per square
meter of surface to be protected is a reasonable estimate. The combination of
coating and cathodic protection can, therefore, prove to be most economical.
Overactive cathodic protection-which can occur particularly with
magnesium anodes--or badly controlled impressed current protection-leads
to the reduction of oxygen to form hydroxyl ion (OH-), which can attack
paints. Also, the placing of the zinc anodes requires experience in connection
with both type and thickness of the paint (system). Since zinc anodes do not
allow the production of excessive amounts of alkali, paint failure or the need
to use alkali-resistant paints is avoided.
Use in Conjunction with Impressed Current Systems
Grounding Cells. The use of zinc anodes for grounding cells is
essentially confined to buried pipelines; see Chapter 4 on soil corrosion.
Essentially, the moderately high resistance cell is used to join cathodically
protected parts of a buried system (e.g., gas piping) to unprotected parts (as
on an industrial user's premises) and by taking a current surge will prevent
sparking across an insulating flange. The pipe at the nonprotected side of the
Chapter
flange also receives current from the zinc anode, minimizing current dis-
charge from that area and potential damage.
Low Resistance Current Paths. An indirect use of zinc for corrosion
protection is to provide a low resistance path for impressed current protection
of concrete bridge structures with uncoated steel rebar. Electrochemical
protection of steel rebar in concrete was developed as a repair technique but is
now being promoted also for new structures. Current generated by impressed
anodes on the outside of the concrete and a positive electical contact with the
rebar stops the latter from rusting. The primary anode is usually brass or
copper.
The secondary anode is composed of (a) conductive paints or polymers,
(b) titanium wire mesh, or (c) zinc thermal spray-although one bridge used
zinc sheet scrap from United States penny blanks! Development has been in
that order with zinc spray tials starting in 1982. The zinc is sprayed directly
onto the outer concrete surface of the bridge. Direct contact between the zinc
spray and the steel rebar must be avoided-therwise the zinc would be
prematurely consumed. However, the sprayed zinc coating must not be
sealed, for this would increase electrical resistance and allow less protective
current to flow. Exceptionally, the Department of Transportation (DOT) in
the State of Florida sometimes uses the zinc as a sacrificial anode (e.g., in
1988 on spalled areas with rebar exposed on the Niles Channel Bridge near
Key West; on the Julia Tuttle Causeway Bridge, Miami, 1989; and on the
McCormick Bridge, Jacksonville) where permanently wet and salt-saturated
concrete is the norm.
The Californian DOT in 1982 evaluated materials including Al, AI-Si,
80%Ni-20% Al, zinc, stainless steel, sprayed Babbitt metal (Sn-Sb-Cu), and
C-filled latex paint. They considered conductivity, consumption, adhesion,
weather resistance aesthetics, cost, availability as wire, and compatibility
with salt-impregnated concrete. Babbitt metal or zinc gave best results. The
initial trials were on selected areas of:
Richmond San Rafael Bridge, San Francisco: two columns 20 m2, 225 pm
zinc- in condition after 7 years.
East Camino Undercarriage, Placerville: eastbound, zinc arc-sprayed on deck
with striped pattern to ensure concrete bond to asphalt overlay 12.5 cm
of 225 pm zinc 12.5 cm gap, 300 m2 zinc; westbound, thermally
sprayed 500 pm zinc on soffit, 150 m2.
Three hundred bridges in the United States now have cathodic protection, and
several more are treated each year. Recently, 100 m2 of a parking garage was
protected using zinc. The economics are reasonably encouraging.
TYPES OF CORROSION
A. Introduction
Consideration of the corrosion of zinc is primarily related to the slow general
dissolution from the surface. Zinc is attacked only slightly by pure air, even if
its moisture content is considerable. Below about 20O0C, the film grows very
slowly and is very adherent. Zinc-coated steel behaves similarly to zinc; a
useful booklet on the mechanism of corrosion of a galvanized coating has
been produced by Daesen (1975). Localized pitting occurs only under special
conditions-for example, in waters where a calcareous scale may crack
locally, exposing a small area of zinc. very wide variety of compounds can
be recognized by spectrographic analysis of zinc corrosion products produced
in tap water. Intergranular corrosion of cast zinc alloys was a problem until it
was found that this defect was due to certain impurities"notab1y lead, tin,
and cadmium. Zinc specifications now control these and other elements and
such intergranular corrosion these alloys no longer occurs. Similar controls
apply with alloys for cathodic protection. However, there is scope for more
work, to elucidate the mechanism fully. No intergranular corrosion problems
arise with galvanized coatings containing these elements. Bimetallic corro-
sion of zinc occurs mainly when zinc or zinc-coated steel is protecting
uncoated steel or other base materials such as copper.
B. Uniform Dissolution
Zinc has the ability to form a protective layer comprising basic carbonates,
oxides, hydrated sulfates, depending on the nature of the environment.
When the protective layers have formed and completely cover the surface of
the metal, corrosion proceeds at a greatly reduced rate. large number of
zinc corrosion products have been identified by Gilbert (1952) and by Biestek
(1974).
Some of the factors affecting film formation and subsequent changes
have been analyzed in experiments in recent years in relation to atmospheric
exposure and are, therefore, discussed in Chapter In dry air, a film of zinc
oxide is initially formed by the influence of the atmospheric oxygen (e.g., at a
speed of approximately 40 n d 2 4 h when a part leaves the galvanizing bath).
The subsequent reactions with the atmosphere are complicated and more often
than not depend on the local climate microclimate. Therefore, the forma-
tion of an insoluble zinc patina is irregular both in place and in time. But when
the zinc surface becomes wet with rain, mist, dew, the atmospheric carbon
62 Chapter
dioxide present means that the visible product is a basic zinc carbonate film.
This tends to inhibit the further corrosion of zinc if the film is not removed or
chemically altered. Under these ideal conditions, the subsequent zinc corro-
sion rate would be controlled by diffusion through the film and would be
inversely proportional to the thickness of the film, giving a parabolic curve of
corrosion rate against time; that is, the corrosion rate would be gradually
reduced to an insignificant figure.
However, the formation and maintenance of the protective layers is
governed largely by the pH of the environment, especially the acidic pollu-
tants; but since zinc forms an amphoteric oxide, strong alkaline conditions
also adversely affect its corrosion behavior by interfering with the formation
of the protective layers. Figure which shows how the corrosion rate of
zinc varies with the pH (Roetheli et al., Belisle and Du Fresne,
indicates that the attack is most severe at pH values below 6 and above
while within this range corrosion is very slow. The actual rates of corrosion
shown in this study are not of direct relevance in practice, where aqueous
solutions are normally much more complex and often contain inhibitors,
while the initial rates shown would be modified by corrosion products-
particularly in atmospheric exposure, where there is the beneficial effect of
periodic drying.
The first layer formed influences the corrosion of the zinc throughout its
life, although the factor is less significant than the nature of the environment
in which it is exposed. Zinc corrosion products occupy a larger volume than
the zinc from which they originate and, consequently, a small loss of zinc can
often give a large volume of corrosion products.
The uniform corrosion rates for zinc are not greatly affected by the
purity of zinc; and zincs behave similarly in many conditions.
This is especially true in open atmospheres, where sufficient oxygen is
present to prevent polarization by hydrogen. Some alloying elements increase
the corrosion resistance of zinc significantly.
Aluminum is particularly beneficial in coatings and forms the major
addition in trade-named products such as Galfan and Galvalume and in the
thermally sprayed zinc-15% aluminum; it is also beneficial in castings, such
as the ZA series. Both with Galfan and Galvalume, a dense intermetallic layer
must formed during the dipping procedure with both zinc-rich phase and
aluminum-rich phase in the upper areas of such coatings.
Iron can be beneficial in coatings. The iron-zinc alloys formed in hot
dipping or sherardizing can be up to more resistant in mildly acidic
conditions, but some workers report lower corrosion resistance with some
galvannealed coatings. The iron-zinc alloy layers, while continuing to protect
50
45
40
35
30
W
25
a
9 20
z
En
0
(I
II:
o 15
0
10
5
0
240
pH 20
I
3 - 27
0 SHG ZINC
SHG ZINC
3O.C; 86'FI
-.- (OXYGENATED SOLUTION
__ 1100 ALUMINIUM
I
I
) i
i
/
Fig. 1.1 1
data).
Variation of corrosion rate with pH (based on Noranda, 1993 and earlier
the steel, can allow the protective surface film to become a dull brown color,
and this is sometimes erroneously regarded as failure of the coating. True
rusting is usually a redder brown and rougher (Bayliss et al., 1965;
Horstmann, 1983). The British Iron and Steel Association (BISRA) in un-
published work found that alloying small amounts of zinc with iron had little
effect, but at 35% zinc a good plated deposit was obtained with good
corrosion resistance and easy to paint.
Gilbert (1953) found that a 2% copper alloy increased atmospheric
corrosion by up to 20%, but Dunbar (1967) found no significant difference
when only 1% copper was present (see also data on copper in galvanized
Chapter
coatings). Any beneficial effect is likely to be dependent on adequate removal
of copper ions dissolved from the alloy: if copper ions remain on the zinc, the
corrosivity of the environment may be increased.
Care is needed in assessing performance of alloyed zincs because no
reliable accelerated tests are available. Many such tests on new electroplating
alloys show a much lower corrosion rate relevant to zinc than is eventually
found in service. The titanium present in rolled zinc has only a minor
influence on corrosion resistance but, after weathering, this alloy develops a
dark shade gray, whereas the previously produced alloy containing cad-
mium developed a whitish appearance; hence it was aesthetically unwise to
mix these alloys in the same roof.
Rain or high relative humidity (which can cause condensation) increases
corrosion of zinc that is roughly proportional to the time wetness in a
particular environment. Factors that determine time of wetness have been
discussed by Grossman (1978). Temperature generally has only an indirect
effect in air: the relative humidity usually drops as temperature increases. At
humidities exceeding 7596, corrosion rates rise perceptibly. A drop in tem-
perature has the opposite effect. In solutions, reaction rates increase with
temperature but-more important-in hot tap water there is a potential rever-
sal between zinc and steel typically between 60 and 90C, which then limits
the benefits of a zinc coating to its barrier effect. In addition, localized pitting
may occur, as noted in Section IV (How Zinc Can Protect Steel).
C. Wet Storage Stain (white rust)
The one form of general corrosion that is not protective is white rust, more
properly called wet storage stain because it occurs in storage, where there is
access for water but limited supply of oxygen and carbon dioxide. The
presence of chlorides and sulfates accelerates wet storage stain formation.
White rust is a white, crumbly, and porous coating consisting of 2Zn
C03.3Zn(OH), together with ZnO and voluminous P-Zn(OH),; the surface
underneath the white products is often dark gray.
The coating is found particularly on newly galvanized, bright surfaces
and especially in crevices between closely packed sheets, angle bars, and so
on, if the surfaces come into contact with condensate or rainwater and the
moisture cannot dry up quickly. Zinc surfaces that already have a normal
protective layer corrosion products are seldom attacked.
When zinc or zinc coatings corrode in open air, zinc oxide and zinc
hydroxide are normally formed. If the supply of air to the surface of the zinc is
restricted, as in a narrow crevice, there is not enough carbon dioxide for the
subsequent formation of a zinc carbonate layer.
The layer of zinc oxide and zinc hydroxide is voluminous and porous
and adheres only loosely to the zinc surface. Consequently, it does not protect
the zinc surface against oxygen in the water. Corrosion can, therefore,
proceed as long as there is moisture left on the surfaces.
When wet storage staining has occurred, the objects should be arranged
to permit their surfaces to dry rapidly. The attack will then cease and, with a
free supply of air to the surfaces, the normal protective layer of corrosion
products will form. The white corrosion products are generally washed off
and the surface of the coating takes on the normal appearance of a hot dip
galvanized exposed object. However, attention must be given to the dark gray
to black spots or areas where corrosion attack has already gone too far and
local coating thickness may no longer be at acceptable levels.
White corrosion products that have already formed can partially
removed by brushing with a stiff bristle brush. They can be removed com-
pletely by washing with 10% acetic acid or with a paste containing polishing
chalk and NaOH, either of which must be followed by very thorough
rinsing with water. The original bright, metallically lustrous surface cannot be
restored by these treatments.
Apart from carefully washing off such corrosion products, it is recom-
mended to use a dilute (0.5-1 aqueous solution of n-butyl thioglycollate,
buffered at pH 5.5-5.7 (Van Eijnsbergen, private communication). Alter-
natively, covering such white-rusted surfaces with a neutral, light, low vis-
cosity mineral oil has proved to be effective. This treatment is especially
recommended when galvanized parts, densely packed in ships holds, are to
be transported through tropical areas.
Short-term protection against wet storage staining is obtained by chro-
mating or phosphating. Painting after galvanizing also provides effective
protection. Wet storage stain is best avoided by preventing new zinc surfaces
from coming into contact with rain or condensate water during storage and
transport. Materials stored outdoors should be arranged that all surfaces are
well ventilated (Fig. 1.12) and so that water can easily run off the surfaces.
Uncoated steel wire should not be used for bundling, and pallets should be
used to avoid direct contact with the soil. Material to be exported must be dry
before dispatch. Use nylon or glass fiber reinforced ropes for bundling.
Plastic sheet will soon be damaged, thus accelerating white rust formation.
When using closed steel containers, no moisture-absorbing compounds are
required, which is the case when wooden crates are used. Applying a thin coat
66 Chapter 1
Wood spacer boards
I 1 L
L bars U beams
H.E or I beams
Fig. 1.12 Stacking of galvanized parts out of doors (Thomas, 1980).
of neutral, low viscosity mineral oil is particularly effective when galvanized
parts are to be exported overseas.
Since the corrosion products are very voluminous, occupying-together
with air-a space about a hundred times greater than the zinc that has
corroded, even slight attack will be prominent and may appear to be serious.
Usually, however, a wet storage stain is of little or no importance to the
durability of the corrosion protection: it is usually removed by rain in service,
allowing a fully protective film to form. In the case of thin zinc coatings, such
as on electroplated objects, however, a severe attack of wet storage stain can
be damaging. Long storage of items tightly packed on ships in tropical
conditions can be particularly harmful even for thicker coatings.
A guideline for users appears in Swedish standard SMS 2950:
The special form of corrosion known as wet storage stain (white rust)
that sometimes appears on hot dip galvanized objects after a short
period of storage shall not be regarded as sufficient cause for not
granting approval, unless a special agreement has been reached that wet
storage stain shall not be allowed.
67
D. Bimetallic Corrosion
When two different metals are in electrical contact and are also bridged by
water containing an electrolyte (e.g., water containing salts, acids, or alka-
lies), current flows through the solution from the anodic or baser metal to the
cathodic or nobler metal. a result, the nobler metal tends to be protected
but the baser metal may suffer greater corrosion. This indeed is the principle
behind the use of zinc as a coating on steel or as a sacrificial anode; when,
however, the connection is fortuituous or unavoidable, such corrosion can be
a problem.
Published schedules of electrode potentials are of value in drawing the
attention of designers to the dangers of bimetallic corrosion. Such schedules
can be misleading, since although the potential difference between metals is
the prime driving force the corrosion current, it is not a reliable guide to the
rate and form of corrosion suffered at any particular contact. Statements
claiming that specific differences of potential are safe or unsafe are unreliable.
Electrode potential measurements have some role in the thermodynamics and
kinetics of corrosion of metals, as discussed by Pourbaix (1969).
There is an apparent exception to the necessity for the two metals to be
in electrical contact for bimetallic corrosion to occur: namely, when a noble
metal corrodes slightly and dissolves in water that subsequently flows over a
less noble metal, the more noble metal may deposit on the less noble metal,
forming a true bimetallic contact. For example, copper can dissolve very
slightly in some natural waters and may then deposit on zinc. Bimetallic
corrosion may also be experienced when two dissimilar metals, not in direct
contact, are nevertheless connected electrically.
The severity of bimetallic corrosion is dependent on the ratio of the
areas of metals in contact, the duration of wetness, and the conductivity of the
electrolyte. For example, other things being equal, seawater-which is a
highly conducting solution-gives rise to more severe bimetallic corrosion
than most fresh waters, which are generally of lower conductivity. film of
moisture condensed from the air or rainwater can dissolve contaminants and
provide conditions conducive to bimetallic corrosion. In addition, contami-
nants already present on the metal surface, such as fingerprints or residues of
processing solutions, may absorb moisture from the atmosphere which, al-
though humid, has not actually reached the dew point.
Bimetallic corrosion is more severe under immersed conditions than in
the atmosphere. In the latter, attack occurs only when the bimetallic contact is
wet; this depends on many factors, such as the presence or retention of
moisture in crevices, the effectiveness of drainage, and the speed of evapora-
tion. The relative size the areas of the metals that remain wet in the vicinity
Chapter
m m - m -
3
69
N N N
"
70 Chapter
of the contact have an important effect. In atmospheric conditions, bimetallic
corrosion is usually localized in the vicinity of the line of contact. Beck and
Kaiser (1975) have studied the zinc-iron couple in relation to car body
construction, and Petzold and Efer (1977) have studied cut areas on organic-
coated zinc-coated sheet in relation to exposed area, zinc thickness, and type
of organic coating.
The recommendations in Table 1. l 1 (PD 6484: 1979) are valid for most
environments, but direct experience of behavior in a specific environment
under consideration is always preferable. Similar tables in PD 6484 show the
effect of zinc on the corrosion of other metals. Among comprehensive
exposure studies are those by Johnson and Abbott (1975) and Kucera et al.
(1987a,b). With zinc (as galvanized steel) in contact with bare steel surfaces,
corrosion losses can be severe, as shown in Table 1.12. The additional attack
on a zinc-zinc couple is due to differential aeration and creative effects.
Generally speaking, galvanized steel surfaces may safely be in contact
with most aluminum alloys, stainless steel 304 and 316F, chrome steel (>
12% Cr), and tin, provided the area ratio of zinc to metal is 1:l or lower and
oxide layers are present on both aluminum alloys and the two stainless steels.
Tests canied out by Stichting Doelmatig Verzinken, the Dutch Galvanizing
Institute, showed that contacts between galvanized steel and stainless steel
(types 304 and 316F) do not accelerate corrosion of zinc independently of the
ratio of both contact surfaces. In general, the galvanic series (Table 1.13)
Table 1 .l 2 Zinc and Steel Corrosion in Bimetallic Couples
in the United Kingdom
Corrosion (pdyear)
Medium
Industrial climate (London)
Rural climate (Leatherhead)
Marine climate
River water
Seawater
(Rye)
dam, Devon)
(Caernarvon)
Steel-Zinc (coupled) Zinc-Zinc (coupled)
Steel Zinc additional attack
28 46
5 45
5 68 6
2 185
8 580 114
Source: Johnson and Abbott (1975).
Table 1 .l 3 Electrode Potentials Metals in Seawater
Material Potential (V)"
Magnesium - 1
Zinc
Aluminum
Cadmium -
Cast iron -
Carbon steel -
Stainless steel: Cr, Ni (active)
Lead
Solder (Pb/Sn, 50: SO) - 0.52
Tin -
Aluminum bronze
'AI1 values with respect to a copper-copper sulfate halfcell.
Source: Zinc Handbook, Porter 1991.
Copper -
indicates only the possibility of corrosion or protection by bimetallic action.
When the electrical resistance is high (e.g., through the presence of insulating
films), the amount of corrosion or protection is greatly reduced. When in
doubt, contact areas should be isolated by strips of Teflon or polyvinyl
chloride, by three-ply tape, or by insulating varnishes or lacquers.
An extensive range of bimetallic couples was studied by Southwell et
al. (1976) in Panama, including zinc in contact with various types of steel in
seawater and fresh water. The numerical results depend of course on the
relative size of the test panels, but it is interesting to note that the zinc in these
bimetallic couples in seawater corroded 25 times as fast as uncoated zinc.
When considering zinc-aluminum alloys, the surface oxide film nor-
mally present is likely to reduce any corrosion current. The risk of bimetallic
corrosion is small in atmospheric exposure; trials by Noranda have been in
progress since 1984 on alloys coupled to other common metals. No visual
effects were noted at the 5-year examination (Barmhurst and Belisle, 1992).
A zinc-25% aluminum-0.05% magnesium alloy coupled to other materials
and exposed on the Noranda Research Center roof showed pitting attack on
the zinc-based material (but only up to 0.38 mm deep in 10 years) when
joined to copper, brass, or steel, but less when joined to stainless steel or lead
and least when joined to aluminum.
Some substances that may be present merely as traces in the environ-
ment can have a profound effect on behavior at bimetallic contacts. The
corrosion at the junction may be accelerated or reduced substantially.
72 Chapter
The risk and intensity of corrosion under immersed conditions at a
bimetallic junction is greatest if the area of the more noble metal is large
compared with the area of the less noble metal. Hence, zinc-coated fasteners
in an uncoated steel assembly will rapidly corrode because of the large area of
steel they are trying to protect. Similarly, a few of galvanized steel sections
used to replace a small part of the immersed support structure of a pier will
rapidly lose their coating. Linder and Mattsson (1975) studied couples in
various tap waters at both 25 and 75"C, while Scholes et al. (1977) developed
information based on current flow in seawater over 100-150 hours.
The special case of the bimetallic effect between a zinc coating and the
substrate that it is protecting is discussed under hot water aqueous corrosion
resistance as the normal bimetallic effect whereby zinc protects steel is
reversed in some waters, usually at Bimetallic corrosion of zinc
occurs mainly when zinc or zinc-coated steel is protecting uncoated steel or
other base metals such as copper. Many of the uses of zinc deliberately invoke
this principle, but in other cases an unwanted effect arises as a result of
constructional requirements, and avoidance of bimetallic corrosion is needed.
The basic principle underlying the prevention of corrosion arising from
dissimilar metals in contact is to prevent the flow of the corrosion currents.
This may be achieved:
by insulating the dissimilar metals from each other (i.e., breaking the metallic
by preventing the formation of a continuous bridge of conductive solution
path) or
between the two metals (i.e., breaking the electrolytic path).
Under immersed conditions, the first method may be possible if insulation is
not precluded by a need for electrical bonding. For example, a zinc-coated
steel nut and bolt may be fitted with an insulating bush and washers where it
passes through a steel plate that cannot itself be zinc coated. Complete
insulation is essential in such a case and, therefore, the bush is vital. The
second method may be effected by the application of paint or plastics coatings
to the immersed parts of the metal. Where it is impracticable to coat both
metals, it is preferable to coat the more noble one; that is, the coating is
normally the other metal, not on the anodic zinc. the application of
insulating three-ply tapes based on polyethylene and covered on both sides by
a butyl rubber laminate (e.g., Densolene S-40) is very effective in preventing
contact corrosion, especially in the offshore industries. Blee et al. (1985) have
shown that if lead and stainless steel are in the electrical bonding circuit of a
manhole, they can accelerate corrosion of galvanized hardware there.
Bimetallic corrosion may be greatly reduced by applying to one of the
metals of the couple a coating of the other, or (more commonly) a coating of a
third metal chosen for its compatibility with both. Thus, brass in contact with
aluminum alloys can be prevented from damaging the latter by coating the
brass with zinc. Nonmetallic coatings are useful in particular cases (e.g.,
when the anodized zinc coating has a high resistance).
Jointing compounds should be able to exclude water, should not dry
crack, and should not be corrosive. If formulated with an adequate corrosion
inhibitor, they may suffice for moderate conditions, but painting is also
advisable.
The application to the assembled joint of an overall paint or powder
coating system, suitable for the environment, is effective. The use of a
jointing compound followed by painting is better than either alone. Thick,
water-resisting plastics or mastic coatings, where they can be used, are useful.
E. Crevice Corrosion
Additional corrosion may occur where the supply of oxygen at interfaces
between joined components is limited. This is not bimetallic corrosion, but its
effect at bimetallic joints can be as detrimental as true bimetallic corrosion.
However, no crevice corrosion occurred in couples of zinc-aluminum alloys
and polyethylene in 10-year atmosphere tests (Noranda, personal communica-
tion) nor around nylon bolt heads. Jointing compounds are useful in prevent-
ing crevice corrosion as well as bimetallic corrosion. Also, stressed parts of
the surface tend to be anodic to unstressed parts, but this effect is not usually
of practical significance with zinc and zinc-coated steel.
Zinc may also be used to prevent crevice corrosion of other materials.
Work by Ito et al. (1989) shows that crevice corrosion resistance of zinc-
plated stainless steels is superior to that of unplated in an NaCl solution. Good
general corrosion resistance was also obtained. It seems that these beneficial
results are due not only to sacrificial dissolution of zinc, but also to suppres-
sion of the cathodic reaction of dissolved oxygen by the corrosion products. A
buffer action by the corrosion product also seems to play an important part in
the prevention of corrosion.
F. Intergranular Corrosion
Fifty years ago, intergranular corrosion of the zinc-4% aluminum casting
alloys was a problem because of high impurity levels, notably of lead, tin, and
cadmium. The subsequent introduction of 99.99% zinc as the base for these
casting alloys eliminated intergranular corrosion under normal service condi-
tions. However, an inherent tendency for intergranular corrosion exists, and
this defect can still occur if pure zinc-aluminum alloys are exposed to
temperatures above about 70C under wet or damp conditions. Research has
74 Chapter 1
shown that tin, lead, indium, cadmium, bismuth, mercury, thorium, sodium,
and iron assist intergranular corrosion. Titanium, silicon, nickel, platinum,
and gallium appear to be innocuous. Copper, magnesium, calcium, and
sodium additions help to prevent intergranular corrosion, and deliberate
additions of copper and magnesium give a product with excellent defense
against intergranular corrosion (Pollard and Edwards, 1976).
The mechanics of intergranular corrosion are not clearly understood.
The electrochemical explanation is usually that the zinc-rich phase corrodes in
preference to the aluminum-rich phase, with impurities such as tin acting as
cathodes; however, it is also postulated that the aluminum precipitates at grain
boundaries are anodic to zinc in the alkaline conditions that develop there.
Behavior of zinc-0.1% aluminum alloys in tap water at 80-95C studied by
Devillers and Niessen (1976a,b) confirm this view. In such dilute alloys the
aluminum is capable of passivating only with the beneficial help of copper
and magnesium. Otherwise, it becomes the anodic corroding region, with tin,
etc., again forming the cathodes. It is possible that the controlling factor will
change with the severity of the corrosion conditions; hence, accelerated tests
cannot be directly related to performance in service.
0 30 60 90 I20 150 180
DURATION OF EXPOSURE, d
Fig. 1.1 3
strength (Noranda, 1986).
Effect of intergranular corrosion of zinc-aluminum alloys on impact
Pollard and Edwards (1977) did work to determine the permissible
limits of lead and tin in the ZA-12 alloys. Borbe et al. (1978) claimed that 12
ppm N2 in 10% or 22% aluminum alloys improved both stain resistance and
acid resistance at room temperature. Recent work by Noranda (personal
communication) on ZA-27 alloys with excess lead (0.0105%), tin l%),
or cadmium (0.008%) subjected to a steam test (95"C, 95% humidity) for up
to 100 days indicated that the bulk of the corrosion was in the zinc-rich phase.
Impact strength can decrease through intergranular corrosion (as well as
by aging). At 60"C, even in high humidity, the loss is modest. At 95"C,
intergranular attack is 10 times greater and loss of impact strength increases
(Fig. 1.13).
In practice, the use of zinc-aluminum casting alloys should be restricted
to below 70C and impurities controlled to within specific limits: recent work
indicates that these limits can be safely extended to 0.006% each for lead and
cadmium and to 0.003% for tin.
G. Corrosion Fatigue
Metallic coatings can stop corrosion fatigue by preventing access of the
corrosive substance to the basis metal, as long as the coating in intact. Since
discontinuities are likely to exist initially or to develop in service, it is clearly
preferable to use a coating such as zinc, which is anodic to the basis metal, to
afford electrochemical protection after mechanical protection has ceased. It
is, in fact, found that anodic coatings are usually beneficial and cathodic
coatings sometimes harmful. Nevertheless, the endurance limit of the coated
material in corrosive conditions may still be higher than that of the bare metal.
Most of the available data on metallic coatings are concerned with the
protection of steel. In a comprehensive survey, Gilbert (1956) reported that
during World War Haigh observed that the life of paravane towing ropes
could be enormously increased by using galvanized wire; this was the first
demonstration of the use of an anodic coating to overcome corrosion fatigue.
Gilbert notes that although galvanizing may reduce the fatigue limit in air, it
increases corrosion fatigue resistance. In other research cited by Gilbert,
Krystof recommended electrodeposited zinc and found tin coatings of little
value, Thum and Ochs found that zinc and cadmium coatings were beneficial,
but that nickel and chromium were not. Gilbert (1956) records that cathodic
protection by zinc was demonstrated as early as 1933 by Behrens and was
confirmed by Gould and Evans and by Stuart and Evans. Whereas Huddle and
Evans demonstrated the beneficial effect of zinc-rich paints, Gough and
Sopwith found that galvanized or sherardized coatings gave very satisfactory
protection to steel exposed to salt spray, while electrodeposited zinc or
76 Chapter
Table 1 .l4 Corrosion Fatigue Tests on Steel Specimens
Corrosion fatigue limit
[MN/m2 (kgf/mm*)] determined
on machine with rotation rate
Steel grade Form of treatment 3000 rpm 45 rpm
40U
~~~~
Without protective 175 (17.5) 120-130 (12-13)
coating
40U Zinc coated 270 (27.0) 260 (26)
20KhN Zinc coated
320 (32.0) 330 (33)
1SNM Zinc coated
325 (32.5) 340 (34)
35G2 type
Zinc coated
340 (34.0) 345 (34.5)
Source: Proskurkin (1975).
cadmium gave fair protection, and Gilbert gives full references to the original
work in each case. Loria and Bush showed a constant fatigue strength
for galvanized steel compared with more than less for ungalvanized steel
(making it inferior the galvanized steel) in salt spray fatigue tests of
automobile-type specimens.
Proskurkin ( 1 reported work by the Azerbaidzhan Scientific Re-
search Institute of Oil Production in the (Table and Fig.
showing the beneficial effect of different types of zinc on corrosion fatigue
cycles, mi l l i ons
Fig. 1 .l 4 Corrosion fatigue curves for uncoated and zinc coated steels on (a) fast
and (b) slow machines: 1 = zinc-coated steel 400; 2 = steel 40U without protective
coating; 3 = zinc-coated steel 20KhN; 4 = zinc coated steel lSNM, 5 = zinc coated
manganese steel of 35G2 type (Proskurkin, 1975).
resistance in groundwater containing 500-900 mg/L hydrogen sulfide; before
fatigue testing, the specimens were immersed for one month with access of
air. Tests at rpm gave a corrosion fatigue limit of 28-34 kg/mm2 for
zinc-coated steel and 20-25 kg/mm2 for uncoated. At lower speeds, the limit
increased for zinc-coated steels but decreased with uncoated (typically 26-35
kg/mm2 zinc coated vs. 12-20 kg/mm2 uncoated). Taraevskii and Kuslitskii
(1988) found zinc-plated high alloy steel (20Ni:2Mo) pump rods operating in
a medium containing hydrogen sulfide had increased corrosion fatigue re-
sistance.
Corrosion fatigue under changing stresses is initially more pronounced
in hot dip galvanized steel constructions compared with steel with scale and
steel that has been grit-blasted. However, upon weathering, the fatigue
strength of galvanized steel remains constant whereas it decreases both for
steel with scale and grit-blasted steel. On steel with scale, pitting occurs when
the change in stress is than about 100 cycles per second.
Stress Corrosion
corrosion does not usually occur in zinc zinc-coated steels and,
indeed, zinc can prevent stress corrosion cracking of other materials. Lunde
and Nyborg (1989) showed that zinc anodes and sprayed zinc both prevented
stress corrosion cracking of carbon steel in ammonia solutions.
Stress corrosion has been observed in zinc-aluminum alloys (2, 8, 29,
or 42% aluminum) tested in aqueous sodium chloride solution under constant
load. Time fracture decreased mainly with increasing stress, but also with
increasing temperature, concentration of NaCI, and percentage of cold work,
and with decreasing aluminum content and pH.
The influence of galvanizing on high strength reinforcing steel in casing
tubes before the injection of the concrete, which may lead hydrogen-
induced stress corrosion cracking, has been studied by Riecke and Johnen
(1986). The absorption of hydrogen during corrosion takes place in weakly
acidic solutions, as well as in neutral or alkaline solutions. With increasing
pH of the electrolyte, the hydrogen absorption by proton discharge decreases.
When the steel is plastically deformed and water reacts with the iron, hydro-
gen can also be absorbed. On galvanized steel surfaces where the zinc coating
has been damaged, hydrogen absorption may result from cathodic passivation
of the free steel (iron) surface by dissolution of zinc. The hydrogen interacts
with the micro steel structure, thus weakening the atomic bonds. Long-time
tests on galvanized high strength steels in which the coating was deliberately
given local damage, and under a tension of 0.2% the yield values, did not
show any rupture during the test period (300 hours). However, when notches
78 Chapter
go through the zinc coating and into the steel, galvanized as well as un-
galvanized specimens have shown embrittlement.
Hydrogen Embrittlement (of steel)
Hydrogen embrittlement of steel is not strictly related to corrosion of zinc, but
it is briefly mentioned here because in hot dip galvanized steel applications it
is often confused by users with corrosive attack.
In addition to hydrogen embrittlement, brittle failure can occur as a
result of stress corrosion, liquid metal attack, or strain-age hardening. The
last mentioned cause is well known and can occur on strained steel of any
strength but seldom actually occurs in modem steels; it is adequately docu-
mented in Appendix E of BS 729 (British Standards Institution, 1971; reaf-
firmed in 1986); users often erroneously refer to this effect as hydrogen
embrittlement. This clearly states that strain-age embrittlement is the only
type embrittlement that can be aggravated by the hot dip galvanizing
process.
ASTM Standard A143-74 (reapproved 1984), Standard Practice for
Safeguarding Against Embrittlement Hot Dip Galvanized Structural Steel
Products and Procedure for Detecting Embrittlement, is related apparently to
any form of hydrogen-assisted embrittlement and is based on work at Battelle
Memorial Institute. It similarly refers to aging phenomena, cold working,
and absorption of hydrogen and states that hydrogen embrittlement of
galvanized steel is usually of concern only if the steel exceeds approximately
1100 MPa in ultimate tensile strength or if it has been severely cold worked
prior to pickling. The last part of this sentence is contentious in definition,
inasmuch as BS 729 would regard the same phenomenon as strain-age
hardening.
There is difficulty in distinguishing between stress corrosion and hydro-
gen embrittlement: normally (see, e.g., Voeltzel et al., 1967), under
corrosion conditions, the zinc coating delays the start of embrittlement; any
hydrogen generated has no effect while zinc remains.
Liquid metal attack of steel is again an entirely separate phenomenon.
The matter is adequately summarized in the latest draft revision of IS0 1461
(1994) document as follows:
Hardened and/or high tensile steels may contain tensile stresses of such
a magnitude that pickling and hot dip galvanizing may cause the steel to
crack in the hot dip galvanizing bath. The risk of cracking may be
reduced by stress relieving before pickling and hot dip galvanizing, but
specialist advice should be sought when hot dip galvanizing such steels.
The risk of hydrogen embrittlement increases with the surface hardness or
strength of steel. Most steels used for fabrication or structures have strengths
low enough to make the risk of hydrogen embrittlement unlikely where they
are pickled and galvanized. More information should be sought about any
steels described as high strength or for any application that appears to
demand such steels (e.g., certain types of prestressed or reinforcing steel and
bolts above strength grade 10.9).
Quantitatively, when processing steels with actual surface hardness
above about a Brinell hardness (HB) of Vickers hardness (HV) or
Rockwell hardness C (HRC)] or actual tensile strength above about 1050 MPa
(about 65 tonf/in2 or kg/mm2) respectively, the risk of embrittlement
requires serious consideration. The risk depends on the many factors that can
either aid the penetration of hydrogen into the steel or, conversely, get rid of
hydrogen or make it less likely to cause cracking.
Some hydrogen can be present from the steel production process itself.
Hydrogen generated during pickling can be harmful and, although the heat of
galvanizing can get rid of some of this gas, research studies indicate that some
residual hydrogen may remain, possibly exerting harmful effects. Electroplat-
ing, conversely, generates hydrogen and is a major source of embrittlement of
steels.
Of these higher steels, which seldom are made into fabricated products,
the steels with surface hardness HB, or with actual tensile strengths
MPa, have not normally been regarded as subject to hydrogen
embrittlement when correct pickling and galvanizing techniques are used.
Steels with tensile strengths exceeding MPa have always been at risk
from hydrogen embrittlement, whether galvanized or not, and require heat
treatment to limit any risk of embrittlement.
It is believed that galvanizers can continue to galvanize all steels with
actual surface hardness exceeding HB, HV, or HRC or actual
tensile strength 1050 MPa in the traditional manner and that existing data
would support the following contentions: that such steels are not normally
subject to hydrogen embrittlement and that the conventional pickling and
galvanizing process would not cause hydrogen embrittlement. It is also
important to ensure that working of the steel (e.g., bending, thread-rolling)
does not take surface hardness above the limit mentioned.
Should embrittlement of pickled and un-heat-treated high strength steels
occur, it is very difficult to show scientifically what is the cause. This is partly
due to the general difficultly of proving a negative (i.e., that the galvanizing
process was not the cause) and partly due the complex technical factors
affecting hydrogen embrittlement. Hence, for galvanized steels above surface
Chapter
hardness HB or tensile strength 1050 MPa (referred to hereafter as high
strength), the user must be made fully aware of, and take all responsibility
for meeting, any contentions regarding hydrogen embrittlement.
A further consideration is the possible pickup of hydrogen by steels in
service (e.g., hydrogen generated by the corrosion reaction of zinc). This is
unlikely to be a significant factor in assisting possible embrittlement of
galvanized steels.
It is generally recognized that hydrogen can be absorbed by steel when
it is pickled. Hydrogen that enters steel during pickling prior to hot dip
galvanizing and remains inside can lead to embrittlement of the steel, with the
result that steel parts treated in this way fracture if they are subjected to tensile
stresses. The degree of pickup and the risk of consequent embrittlement varies
with several other factors and is significant only with the high strength steels
(as already mentioned).
Investigations by Horstmann of hydrogen absorption by steels for high
strength bolts during pickling in hydrochloric acid confirmed that hydrogen
absorption is greater when pickling is in more dilute acids than in more
concentrated acids. Commercial inhibitors added to pickling acids prevent
solution of iron but often inhibit hydrogen absorption for only a short time
and, with a long pickling time, cause greater absorption of hydrogen than
would be the case in pickling in noninhibited acids. It is desirable to ask
suppliers commercial inhibitors for technical evidence the suitability of
these products when pickling (or overpickling) high strength steels.
An alternative procedure for high strength steels is to blast-clean the
steel, blow off iron dust with air, and galvanize (without even a pickle
rinse)-some rolled hollow section tubes are so galvanized on automatic
lines.
Most recommendations for the heat treatment of zinc-coated steel are
based on electroplated zinc-coated steel, where hydrogen is primarily picked
up in the electroplating process. Even here, an unpublished report by SRAMA
(Springs Research) indicates that heat treatment of zinc-coated springs is re-
quired only with steels having tensile strength greater than 1050 MPa.
It is interest to note that ASTM B695 for mechanical plating (which
has an acid dip as part of the pretreatment) states: The mild degree of
embrittlement . . . with cleaning methods . . . normally is self-relieving
within a days time at room temperature.
Consequently, if for high strength steels the heat treatments recom-
mended after pickling and after plating (BS 1706: electroplated steels) are
adopted, it is clear that galvanizers are adopting an ultrasafe policy, which
would reassure users.
The United States generally accepts galvanized A325 bolts (with no
problems in service reported) but not galvanized A490 bolts (largely because
of the difficulty in one major research project of demonstrating whether the
A490 bolts that failed would have done so even without galvanizing). How-
ever, Germany galvanizes the higher strength 12.9 as well as 10.9 (similar to
A490 bolts) and 8.8. Moreover, a recent report shows that such bolts can
galvanized either at 450C or 530-560C.
A useful summary is in the latest draft for the IS0 1461 and CEN 21461
standards (both due to published in 1994):
Structural steels are not normally embrittled by the absorption of hydro-
gen during pickling. With structural steels, any absorbed hydrogen is
discharged during hot dip galvanizing. If steels are harder than approx-
imately 34 HRC, 340 HV, or 325 HB (see I S0 4964) care is necessary
to minimize hydrogen absorption during surface preparation.
Where experience shows that specific steels, pretreatments, ther-
mal and mechanical treatments, pickling and hot dip galvanizing pro-
cedures have been satisfactory, the information serves as an indication
that an embrittlement problem is not to be expected for the same
combination of steels, pretreatments, thermal and mechanical treat-
ments, and galvanizing procedures.
Cavitation Corrosion
Resistance to cavitation corrosion is of interest mainly with structural-type
materials such as the ZA series of alloys. One of the few test results is that
reported by Noranda (personal communication) which showed that the ZA-8
alloy is resistant to cavitation erosion corrosion. A nominal ZA-8 alloy (but
with 9.03% aluminum) engine coolant pump prototype was subjected to a
dynamic 300-hour test conducted (Amax Modified Method BL 3-2) using a
15 vol% Prestone I1 solution at a temperature of 1 18C and a pump speed of
4800 rpm. At the end of the test, the pump was disassembled and was
assigned the highest rating of according to ASTM-2809, which corre-
sponds to no visible damage in spite of bimetallic effects; copper deposition
occurred because the total surface area of the copper tubing was too large for
the amount of totyl-triazole corrosion inhibitor (for copper) present in the
coolant.
Metallographic examination revealed limited intergranular corrosion at
the casting surface. The measured maximum depth of penetration was 100
pm after 300 hours of testing. Intergranular corrosion usually progresses
linearly with time and, therefore, a maximum depth of penetration of only 0.5
mm is expected after a service life of the car engine equivalent to 160,000 km.
82 Chapter 1
Earlier static immersion tests by Noranda on ZA-8 alloy samples in Prestone
I1 engine coolant demonstrated no susceptibility of the alloy to intergranular
corrosion. In addition to agitation, differences between the two tests consisted
of a higher test temperature and lower glycol/water ratio for the dynamic test.
Both factors may have been responsible for the small amount of intergranular
corrosion observed.
K. Contact with Other Materials
Concrete, mortar, and plaster attack zinc, but the attack ceases after the
materials have dried and the amount of attack is just sufficient to give a good
bond. Because zinc is very resistant to mildly alkaline conditions, unlike steel
and aluminum, it is to be preferred for such contacts, notably in its use as a
coating on steel. Caution is needed if the zinc alloys with higher contents of
aluminum are to be used with concrete or other alkaline materials.
Galvanized rebars in contact with mortars form calcium zincate on the
surface. The hydrogen formed during this reaction is oxidized by the very thin
hexavalent chromate layer on the galvanized steel surface or by the Cr6' in
the mortar itself. This calcium zincate layer on top of the Zn/Fe alloy layers
adheres well and by its increased roughness improves adhesion to the con-
crete. This is the case with both low and high loads on rebar. It has been
shown that the adhesion of galvanized rebars to concrete is in the majority of
cases better than that of black bars and never less. The use of galvanized
rebars often prevents cracking of the concrete and unsightly brown rust strips
or patches on the concrete surface. Galvanized reinforcement of concrete is a
large study in itself and is not dealt with in detail in this book. Useful surveys
include a volume published by the International Lead Zinc Research Organi-
zation (ILZRO, 198 l) and a paper by Porter (1985).
Green wood and a number of South American and African hardwoods
containing acidic materials should not come into contact with galvanized
steel, since certain acidic substances in these woods have a corrosive effect on
zinc. Some preservative treatments for wood can have an adverse effect on
zinc-coated steel with which it is in contact. Hot dip galvanized nails should
not be used for joints that will be exposed in water, regardless of whether the
wood is pressure-treated. In such cases, nails of stainless steel are preferable.
Dry or moderately moist wood, pressure-treated or not, can, however,
be nailed with good results. Other dry building materials, such mineral
wool, do not attack zinc.
Some vegetation (e.g., moss) can generate acidity and increase attack
on zinc, but the main effect of contact with plants and greenery is to increase
the time of wetness, hence the annual corrosion rate.
HOW ZINC
Steel is so versatile that it is used in all industries, but it is particularly
vulnerable to corrosion. The wide range of uses has greatly extended, at an
economical cost, the use of zinc to prevent corrosion.
Iron is extracted from its ores by complicated thermochemical processes
to break down the chemical bonds between the iron and the other elements
with which it is combined in nature. The comparatively pure metal so
obtained, and put to such a wide range of applications, is unstable and
immediately tends to react with its environment. This case of entropy-that
is, returning from an enhanced energy level to a lower energy level-is the
more important with steel because rust never has a standard composition.
Depending on the type of steel, the weathering conditions, and the mechanical
aftertreatments, rust manifests itself in seven types. Moreover, the volume of
rust is 2-2.2 times larger than the volume of the steel from which it has been
formed, thus destroying the adhesion of otherwise protective organic coatings
(paints, lacquers, etc.). It is emphasized that once rusting underneath a
coating has started, it is very difficult, costly, and often impractical to remove
it before reconditioning.
Fundamentally, corrosion is an electrochemical process. That is, it is
accompanied and accelerated by the passage of very small electric currents
between the corroding metal and any other metal with which it is in electrical
contact .or between different areas on the surface of the corroding metal. For
these currents to flow, a potential difference must exist, either between the
two pieces of metal or between the different parts of the same piece, and
moisture or other electrolyte must be present on the surface to act as a
conductor for the current. Potential differences sufficient to cause current
flow can arise from very small local variations in the chemical constitution of
the surface: phase differences across a grain boundary (e.g., between ferrite
and pearlite or cementite) are quite sufficient to constitute an electrolytic cell.
Electrolytic cells set up in this way comprise a series of positive anodes
and negative cathodes between which currents flow and at which electro-
chemical reactions take place. The reactions at the cathodes may be regarded
as electron-consuming and those at the anodes as electron-producing. The
actual disintegration and corrosion are associated with loss of electrons and
occur at the positive anodic areas.
The simplest way of hindering the corrosion processes at a metal surface
is to seal it off by an impermeable and adherent barrier coating, which
denies access moisture and oxygen. The barrier coatings used on iron and
steel need to be strongly resistant to the corrosive influence of their environ-
Chapter
ment and sufficiently permeable to prevent oxygen and water vapor from
penetrating to the metal surface. They are usually metals, paints, powder
coatings, bitumen-based coatings, tapes, cement products. They also need
to be tough and abrasion resistant and to adhere strongly to the surface they
are required to protect.
Zinc can fulfill these requirements and can be applied to an iron steel
surface in a variety of ways and also at different stages in the manufacture of
products. Zinc is, however, more than a barrier coating-as discussed later.
The application of a metallic coating to a metal surface considerably
affects the type of electrolytic action that takes place. The direction of the
current that flows when two metals are brought into contact in the presence of
an electrolyte depends on the relative positions of the metals in the galvanic
series, part of which was set out earlier (Table 1.13). The metals in Table
1.13 are listed in diminishing order of nobility, the most electronegative
metals being at the top and the most electropositive at the bottom. Strictly, a
galvanic series is specific to a particular electrolyte and conditions of test or
use. The list given relates to seawater, which is relevant to many actual
conditions. Generally speaking, if two metals are in contact in the presence of
an electrolyte, a current will flow from the metal that is higher in the table to
the one that is lower.
Hence, if zinc and steel are in contact in the presence of an electrolyte, a
current will flow from the steel to the zinc, so that the zinc becomes an anodic
electron-producing area while the steel is cathodic and consumes electrons.
The zinc, therefore, corrodes in preference to the steel and, in doing
protects the steel surface. This type of sacrificial or cathodic protection occurs
with zinc anodes when zinc coatings on steel surfaces are subjected to
mechanical damage (e.g., at cut edges or scratches), whereby the continuity
of the zinc coating is broken and the steel surface is exposed. The cathodic
protection offered by zinc coatings depends largely on (a) the dimensions of
scratches, cut edges, drilled holes, and impact damage and (b) the coating
thickness of the (q) zinc layer. The ZdFe alloy layers do not offer sufficient
cathodic protection. It has been proved in practice that zinc layers thicker than
15 Fm offer widespread cathodic protection. Scratches up to 5 mm wide and
cut edges on galvanized parts more than mm thick can be effectively
protected by the surrounding zinc layer. Moreover, such protection is benefi-
cial in practice because the reconditioning of such bared areas need not to be
done immediately and may be postponed for 3-6 weeks, depending on the
atmospheric conditions.
Field tests for up to 5 years by the British Steel Corporation (BSC)
(Andrews, 1974) and years of practical experience in aggressive atmo-
spheres show protection of cut edges of galvanized sheet. Van Eijnsbergen
(Fig. 1.15) has compared the time to first rust for different coatings, including
Ti me to Rust mm Long JMge of Panel
I
l
l
I
I
I
I
I
I
I
I
I
I
I
l
I
I
l
I
I
. . .
I I I
. .
.
6 12 24 2 3 4 5 10 20 30 40 50 100 200300
Hours Days of Exposure
Zinc Silicate, 80 - 120
Zinc Silicate, 80 - 120 + Epoxy Top Coat, 90 - 120
Hot-Dip Galvanized, 65 - 220
Hot-Dip Galvanized, 65 - 220 + Epoxy Top Coat, 90 - 120
Epoxy, 70 - 150
Epoxy, 70 - 150 pm + Epoxy Top Coat, 90 - 120 pm
Top Coat Only
1 .l 5 Edge protection by various zinc coatings; exposure in 6% NaCl solution.
Unprotected edge 5 cm below surface. Note superior protection by galvanizing (Van
Eijnsbergen, 1991; based on Swedish research work by Lindqvist, Meszaros, and
Svensson).
duplex coatings, based on Swedish work by Lindqvist, Meszaros, and
Svensson in 1983-1984. Tierra et al. (1991) have shown that galvanizing
gives greater protection than aluminizing.
Under special conditions, there can be potential reversal, whereupon
zinc becomes cathodic to steel; the best known example is at temperatures
86 Chapter
between about 60 and 90C in fresh water (but varying with water composi-
tion). Glass and Ashworth (1985) have studied the electrochemical behavior
in 0.01 M NaHC03 solution saturated with oxygen containing 1% CO2 at
65C. Both zinc and steel spontaneously passivated in this environment, with
steel passivation occurring more slowly. This discrepancy caused temporary
polarity reversal in the zinc-steel couple. The zinc corrosion product consist-
ed mainly of basic zinc carbonates, and it is suggested that these prompted the
rapid passivation of zinc, hence the polarity reversal. This poorly conductive
corrosion product also resulted in strongly polarized zinc cathodic kinetics.
An active steel-passive steel couple was shown to be more damaging to the
active steel area than a zinc-steel couple after reversal. Van Eijnsbergen in
discussion with the author comments that in the temperature range of 60-90C
the (zinc) layer becomes cathodic to the steel base, whereas the ZdFe alloy
layers become cathodic to the zinc layer. Dehydration of zinc hydroxide to
zinc oxide occurs and, consequently, the protective gelatinous layer becomes
grained and less adherent. The zinc oxide is semiconductive and-together
with the changes in structure-has a lower protective function, which en-
hances the potential reversal effect mentioned earlier.
With zinc or other anodic coatings, a further protective action occurs at
minor discontinuities in the coating as a result of the selective dissolution of
the coating, since comparatively impervious corrosion products, which accu-
mulate in the damaged area, increase the resistance to passage of electrolytic
currents and tend to limit further electrochemical action there. Thus, a zinc
coating on steel provides three fold protection. It operates initially by provid-
ing a tough and adherent coating, which seals the underlying metal from
contact with its corrosive environment. If the coating is subjected to mechani-
cal damage so that steel is exposed, the zinc corrodes preferentially and leaves
the steel intact. Third, the zinc corrosion products often form a barrier layer
on exposed steel such that the zinc no longer needs to corrode as fast to protect
the steel. Other than zinc, the main anodic coating that has been widely used
is cadmium. Comparison of the properties have been made by the Cadmium
Association (1977), Carter (1972), and David (1969). Detailed corrosion data
are given in appropriate sections of this book.
zinc is dissolving sacrificially to protect the steel, the rate of zinc
corrosion will be affected by (a) the current passing per unit of area, which in
turn is related to the conductivity of the media in which they are immersed,
and (b) the relative areas of zinc and steel exposed (see later material on
bimetallic corrosion); zinc alloys and zinc will have different rates of corro-
sion. Sugimoto and Goton (1989) have looked at the relative effect of the
zinc-aluminum alloys for gap protection.
Sharivker and Tomin (1988) made electrochemical and gravimetric
measurements on the effective range of corrosion protection afforded by
protective layers of the sacrificial anode type so that estimates could be made
of distances between layers of Zn sputtered onto a 0.17-0.25 C steel (St 20).
The range of action, which was 12-14 m, in fresh water at 6"C,
decreased with increase in temperature.
V. CORROSION TESTS AND THEIR SIGNIFICANCE
The many accelerated corrosion tests reported in this book should not be
regarded as a direct guide to performance, even in apparently related service
conditions. They are mainly of value as sorting tests for coatings of the same
type or as guides for research workers, as is the case with electrochemical
tests. Many tests relating to corrosion and corrosion protection are now
standardized, and an advertised book by Fomin-in Russian-aims to identi-
and discuss all relevant tests.
early as 1943, Committee B-8 of confirmed that salt spray
does not reveal initial porosity in zinc coatings and provides only a poor
means of estimating the coating thickness. The test was not considered to be
suitable for sacrificial coatings.
This point was considerably amplified by Carter (1972), who refers to
comments by Sample in a paper on the use and misuse of salt spray testing in
which he also pointed out that a plentiful supply well conducting electrolyte
could lead to erroneous results by comparison with atmospheric exposure
where there is alternate wetting and drying by a poorly conducting electrolyte.
In such a case there is an acceleration of the yield of the voluminous white
corrosion products that appear on zinc and can cause zinc to have a poor rating
from some organizations.
Carter also quotes Le Que on the unsuitability of salt spray tests for
sacrificial coatings and comments on the failure of the latter investigator to
place cadmium and zinc in the same order as in outdoor exposure. He makes
the further point that the comparative performance of zinc and cadmium
varies with the composition of the salt solution used for spraying. For
example, the superiority of cadmium is lost when natural seawater (contain-
ing, e.g., magnesium salts also) is substituted for a pure sodium chloride
solution. It is suggested that variations in relative performance may be
explained by the nature of the corrosion products formed and the extent to
which condensation occurs on the surface of the specimens.
The inadequacy of salt spray tests is well demonstrated by Van Eijns-
bergen (1970) in Table 1.15. He adds that corrosion is a topochemical
88 Chapter
Table 1 .l 5 Comparison of Accelerated and Actual Exposure Tests
on Galvanized Steel
Mass of zinc coating (glm')"
Test method At 600 hours At 1400 hours
3% salt spray 170 >S00
salt spray 300 >lo00
Synthetic seawater spray 15 25
Outdoor weathering:
25 m from North Sea
30 60
m from Sea
8 12
Immersed in stagnant seawater
50
7.2 @m2 = 1 pm.
Source: VM Eijnsbergen (1970).
process, where corrosion reactions and corrosion speed differ on different
metal areas, especially on steel surfaces. But atmospheric attack is also a
topospherical process, depending on the pollutants and their concentration in
the air and rainwater, the oxygen content, the presence of free radicals, and
the duration of wetness. Furthermore, the microclimate around an object
causes differences in corrosion and its speed.
For all these reasons, no accelerated corrosion test can imitate atmo-
spheric corrosion, which in itself is a complicated, unstandardized process.
Therefore, it is rather confusing that salt spray tests are still in use, even
in standard specifications. Even outdoor exposure tests at an angle of 45" must
be considered only in relation to real-life conditions.
Many developments have been made in the salt spray test as is well
documented in the literature, but they are not given here because experience
has shown that they are unreliable in evaluating zinc coatings against other
materials, or even against some other zinc coatings. They have a very limited
value as production control tests, since the time to red rust is quite long.
Similar comments relate to humidity tests. Other tests, or the 56-day abrasion/
corrosion tests adopted by some car manufacturers, are more useful but fail to
show the f ul l benefit: of zinc coatings-the test is still too short to allow the
undenusting that eventually occurs in service on steel that is only painted.
One of the most widely used is the "scab" test; typically General Motors
recommends the following:
1. Scribe panels.
Corrosion testing cycle:
(a) Salt solution immersion (5% NaCI) at ambient temperature for 15
(b) Exposure at room temperature for 75 minutes
(c) Humidity cabinet (85% RH, 60C) for 22.5 hours; one cycle = 1
(d) On Monday only, after the room exposure:
minutes
week day.
Oven (60C for 1 h)
Freezer ( - 13C for 50 min)
Humidity cabinet
(e) On Saturday and Sunday, panels remain in the humidity cabinet at
85% RH
3. Run 20 cycles (4 weeks).
4. Rinse with warm water, followed by air blowoff. Scribe and Rate.
A recent standardization project is for Swedish Standard SS 11 72 11:
Corrosion Tests in Artificial Atmosphere-Accelerated Outdoor Test. Strom
et al. (1992) at Volvo are devising four indoor tests. These developments are
still in progress but already show that the 12-month outdoor scab tests can be
simulated (in 12 weeks). The outdoor tests show that the distance of blistering
of painted coatings from a scribe mark is inversely proportional to zinc
coating thickness. Johannson and Rendahl (1991) have evaluated a range of
zinc and zinc alloy coatings but find correlation with marine exposure rather
low.
Blekkenhorst et al. (1991) have developed a 10-week cyclic test that has
good accord with practice but some anomalies. Goodwin et al. of British Steel
(1991) have, however, concentrated on developments in outdoor weathering
tests in view of the limitations of cyclic tests.
Vrable (1985) evaluated for zinc-based coatings four corrosion tests
used by the car industry: Ford, APGE; Chrysler, cyclic-humidity; and Volvo,
atmospheric and highway exposure. Highway exposure was assessed as most
realistic and showed that 10-20% iron in zinc coatings on steel was 5&100%
better than pure zinc coatings on steel or galvannealed steel in perforation
testing.
The automobile industry showed at Galvatech 92 that it is now aware of
the limitations of salt spray testing, and evaluation of alternative tests formed
the substance of several papers noted below.
Electrochemical impedance spectroscopy (EIS) was shown by Ami-
rudin et al. (1992) to be effective in monitoring delamination and corrosion
beneath a paint coating on galvanized steel and in giving advance information
on coating breakdown before damage is visible. Periodic measurements on
cataphoretic paint coatings on steel and on zinc or aluminum coated steel by
90 Chapter 1
(nondestructive) EIS were shown by Wrubl et al. (1992) to correlate with
atmospheric, salt spray, and salt immersion test results.
Cherault and Charbonnier (1992), have established that static and dy-
namic tests at their special testing grounds at Renaults Lardy facility can
equate with actual service use of vehicles. A 60-cycle test (taking or 4
months and including salty mud/water and stone chippings while running
along with spray and thermal cycles) is equivalent to 8 years in Amsterdam,
12 years in Brest, or 15 years in Paris.
The benefits of a duplex coating are achieved only if there is good,
long-lasting adhesion of the paint to the zinc coating. The NMP (N-methyl-
pyrrolidone) test has been described by Van Ooij et al. (1982). The test panels
are immersed in the heated solution at 60C (Fig. 1.16) and the time before
the paint lifts is measured. Figure 1.17 shows the use of this test to demon-
strate how silanes increase adhesion of an automotive epoxy-polyurethane
primer to galvanized steel.
One test developed by Vieille Montagne using 10 ppm and a daily
wash has considerable promise for comparing like products-it has been of
use in the evaluation of zinc-aluminum thermally sprayed coatings (Leclerq
and Sempels, 1975). Sulfur dioxide tests can sometimes be of value in
assessing blistering of paint adjacent to a scratch (Fig. 1.18).
Very favorable but unsound comparisons can be made for zinc-rich
paints as related to metallic zinc coatings by use of accelerated tests because
the binder in the paint delays the onset of formation of white corrosion
products, whereas in practice, the performance is related to the amount of zinc
present and to the general formulation rather than to the initially exposed
surface.
Accelerated tests are, however, very useful in evaluating the quality of
chromate films on zinc or zinc-coated surfaces (IS0 4520: 1981), and the
requirements from IS0 3768 (1976) are given in Table 1 Once white
corrosion products have formed, the test should be discontinued. Specifica-
tions that fix a time before red rust forms are largely meaningless and should
be avoided. Simple nondestructive measurement zinc coating thickness is
much more relevant. Abd El Aal et al. (1992) measured steady state potentials
under open-circuit conditions to determine the minimum concentrations of
inhibitive chromate ion that can withstand certain concentrations aggres-
sive chloride ion (as LiCI, NaCl, KCI, MgCI,, or RbCI) and also showed the
tendency to initiate and sustain pitting corrosion in that order, with LiCl
having the greatest tendency.
Wrought or cast zinc is not usually given accelerated tests for corrosion
resistance because slight variations in composition or form are of little
91
Method Sample Failure
Performance Evaluation With Time
<l Minute: Very Poor Adhesion
l- 10 Minutes: Poor to Mediocn Adhesion
>l0 Minutes: Good to Excellent (>so) Adhesion
Benefits
Interfacc Intact No Segregation Ef f m
No Film Degradation Negligible Film Deformation
NoDisSolution
1 .l6 N-Methylpyrrolidone (NMP) nondestructive adhesion test for paints
(Van Eijnsbergen).
importance for corrosion behavior (provided, of course, that the material is
within specification).
Zinc coatings are usually tested by direct exposure to the environment
of use if adequate corrosion data are not already available. IS0 9226 (1992)
deals with the condition for this type of test. The shape of the test piece is of
92 Chapter
Paint (Electroph. Application)
Phosphate Layer + Chromate. Rinse or Silane Dip
- Zinc Coating
t- SteelBase
Curing: U) Minutes at 170' C; or at Temperatures Between 143'C and 248'C.
(a)
' C
a 220'
F 8
c :
5 % 170'
E
.c
200'
0 20 60 80 100 120 140 160
Minutes of N-Methyl Pyrrolidone (NMP) Test to Complete Loosening
Fig. 1.17 NMP test of automotive epoxy-polyurethane primer on hot dip gal-
vanized steel. (a) Test coating. (b) Influence of silanes in reinforcing interface
bonding; secondary bonds are converted to primary covalent bonds (Van Eijnsbergen).
importance. simple rectangle, exposed at or 45" to the vertical and
facing south is most usual. The back is nonnally left exposed to simulate
unwashed areas. The edges are either (a) coated when the test piece is
protected, (b) cut from a larger piece of metal and the edges stopped off, or (c)
left with cut edges exposed to gain information on the protection of edges. If
such information is particularly important, as is the case with paint over zinc
coatings, an X is scribed on the surface to cut through to the basis steel. It is
Average delamination (mm)
4 -
6
Average delamination (mm)
4 -
2 -
+ GALFAN
* Galvanized
0
2 -
+ GALFAN
* Galvanized
0
0 1 2 3 4 5 6 7 8 9 ? U
Exposure time (weeks)
1 .l 8 Delamination paint from scribes on Galfan and on galvanized sub-
strates in an SO2 corrosion test (Lamberigts et al., 1991).
highly recommended to use shaped panels instead of sheet panels to carry out
accelerated and outdoor exposure tests, to allow evaluation of various posi-
tions of the surface as such changes occur in practice. More complex test
panels sometimes desirable (e.g., top-hat section can give a dead-air
space promoting and retaining condensation on the underside). The European
Coil Coaters Association (ECCA, 1989) has developed a test method (T19)
for coil samples. Q-panels, widely used, especially in North America, are
especially suitable for outdoor exposure evaluation of paint systems. The
panels, about 4 mm thick, cm high, and 20 cm wide (and available in
various steel qualities), have a U-shaped piece welded on the front to evaluate
weld, edge, and water retention aspects.
In North America, Davidson et al. from Chrysler (in conjunction with
Nippon Steel) (1992) measured perforation corrosion with test pieces com-
prising inner and outer panels with a wide range of coatings. Higher coating
94 Chapter
weights gave greater perforation resistance. Hot dip and zinc alloy coatings
were better than pure zinc. The outdoor exposure tests were accelerated by
spraying twice weekly with 5% NaCl and appear to provide a fourfold
acceleration of the corrosion rate in South Florida but only 1.3 times greater in
Detroit and Montreal.
This work was complemented by laboratory studies of the hem flanges
of car doors 4.5-10 years old. Uncoated steel in the hem flange was found to
corrode 100 times as fast as zinc-coated steel (compared with 20 times as fast
in a typical exterior environment). Alternate wetting and drying, increases in
environmental pH with time, and the beneficial effect of residual zinc corro-
sion products (after the zinc coating has corroded away) all have an influence.
Many tests based on electrochemical principles are used as an indication
of the relative corrosion resistance of different materials. However, the results
of carefully performed electrochemical corrosion tests will at least give
valuable indications of expectations from the coating system and/or its sub-
strate. The following paragraphs mention some of these techniques and
summarize the results obtained, but caution should be used in interpreting the
results in practical situations where many factors interreact. It is important to
duplicate the electrochemical conditions in practice on a scribed panel with
the same coating system.
For cars, Wilde (1983, unpublished data) has suggested evaluating the
Coulombic capacity of various coatings to protect steel at -0.725 V versus a
saturated calomel electrode (Vs,d when coupled to bare steel under zero-
resistance conditions in an area ratio equal to that obtained in a scratched
panel (e.g., 4: 1, steel to coating).
Dattilo et al. (1985) measured corrosion rates by Tafel line extrapola-
tion and the polarization resistance technique. Values of the Tafel slopes and
the corrosion currents were compared for pure zinc (99.999%) and for lower
purity coatings obtained from brass smelter flue dust. Corrosion rates were
found to be affected by the medium employed, by pH, and by drawing.
Drawn electrogalvanized steel displays higher values of the Tafel slopes than
do as-plated samples. The Tafel slopes are different from those obtained on
pure zinc, presumably because of inhomogeneous surface features obtained
from additive absorption (during plating) and residual lubricants used in the
drawing process and surface structure. However, the significance of such
electrochemical data is doubtful: the authors found the corrosion rates of
electrogalvanized samples plated in waste-derived zinc electrolytes to be
similar to the corrosion rates of samples plated in relatively pure zinc electro-
lytes. Therefore, wastes are a potential source of zinc for electrogalvanizing.
Impurities in the waste electrolyte caused some problems. Copper was unde-
sirable because it passivated the soluble zinc anode. Other impurities such as
cobalt and nickel were more tolerable, but also slowly coated the anode.
Troquet and Pagetti (1983) emphasize that although electrochemical
impedance measurements cannot separate the various steps of inhibition and
corrosion mechanisms, the method is useful for rapid and reliable determina-
tion of the rate of inhibition. The rate of dissolution for zinc in a molar
hydrochloric acid solution accounts for the difference recorded between the
three methods of corrosion rate evaluation. The frequency sweep in the ac
method can passivate the electrode, with an increasing formation of inter-
mediate reaction speeds. It is assumed the observations at various frequencies
are also valid for high frequencies, owing to the high rate of zinc dissolution
in acid environments. The classical monomolecular layer of inhibitor be-
comes a three-dimensional phase, depending on metal surface evolution as
well as on nature and concentration of corrosion and inhibition products. The
presence of this interphase in zinc corrosion systems accounts for the differ-
ences observed between the R, values, determined by the dc and ac
methods. The later methods, well adapted to R, measurements, are useful for
slower corrosion processes; however, they must be used with great care for
fast corrosion processes.
Chang and Wei (1990) used electrochemical and conversion-electron
Mossbauer spectroscopy (CEMS) methods to study the corrosion behavior of
electrodeposited Fe, "x Zn,, where = 0.15-0.85, on 1010 steel immersed in
a 0.1% NaCl solution at room temperature for 2 weeks. The corrosion rate
measured by metal weight loss and electrochemical methods revealed that the
Fe,,.25Zh,75 specimen was more corrosion resistant than the others. CEMS
analysis showed that the corrosion product of the pure 1010 steel and the
higher iron-containing Fe-Zn alloys on steel in 0.10% NaCl solution is
FeOOH .
Thin films and potentiodynamic data are popular for academic studies
and can give clues to the behavior of thicker, more commercial films. Thus,
Nowak et al. (1989) studied AI,Zn, (where 0 < < l ) films 0.1 pm thick
produced by dc ion plating on Fe-26%Cr-l%Mo. Films were smooth, fine-
grained, and columnar. Potentiodynamic data were taken at 20 mV/min in
aerated 3.5% solutions at room temperature. Corrosion (rest) potentials were
approximately - 1.05 V (SCE) for all values of Anodically scanned
specimens had a short passive region, then one to three active-passive peaks,
followed by a long passive region that ended in transpassive behavior due to
crevice corrosion. The total protection range (rest to transpassive potential)
was approximately 2.2 V for all values of Galvalume (55% Al-Zn) and
Galfan (5% Al-Zn) have ranges of 0.4 V, so ion-plated Al-Zn alloys are better
Chapter
in this respect. Auger electron spectroscopic analysis of corroded films
indicated the presence of aluminum and oxygen with only small amounts of
zinc. No evidence of spinel oxide was found by reflection electron diffraction
(RED). RED results best fit aluminum oxides. The role of zinc is unclear; it
may have formed a very thin spinel layer, or it may have facilitated the
formation of the aluminum oxides that are good barriers.
Nowadays, the use of the reflection electron microscope (REM) or,
recently, the tunnel electron microscope, as well as secondary ion mass
spectrometry (SIMS), AES, electron-dispersive X-ray spectrometry, imped-
ance spectroscopy, and so on, are yielding substantial increases in the knowl-
edge of corrosion reactions in coatings and at their interface with metal or
other substrates. As far as zinc or zinc-coated surfaces are concerned, prob-
lems of interfacial and intercoat adhesion, differential diffusion phenomena
and electrolytic cell behavior on the substrate, and interreactions of zinc with
conversion coatings (chromates, phosphates, silanes, silanols, etc.) have been
analyzed, leading toward spectacular improvements in, for example, paint
adhesion, absorption of conversion coatings and, in general, the protective
action inside films as well as on their substrates.
Electron microscopy or diffraction techniques are also used to study the
growth of protective films on galvanized surfaces in solutions containing
phosphate and potassium ions (Radczewski and Schicht, 1969).
VI. OTHER RELEVANT PROPERTIES OF ZINC
As already indicated, the choice of zinc or of particular zinc alloys or coatings
often depends on factors other than corrosion resistance. The main properties
are given in Zinc Handbook (Porter, but particular attention is drawn to
several important or unusual properties.
A. Abrasion Resistance
With respect to abrasion resistance, hot dip galvanized coatings are at least
four to five times as good as pure zinc and most zinc-rich paints. The zinc-rich
paints themselves are, however, more abrasion resistant than most conven-
tional paints.
B. Frictional Characteristics
Zinc, unlike most paints, can left in place on the faying surfaces of a bolted
joint. This property, together with the use of zinc-coated steel fasteners,
ensures that the joint is fully protected. The initial lubricity of the coating is
also useful and can enhanced for metal working operations by a phosphate
coating.
Slip in bolted joints has been investigated, and with dynamic loading it
is shown that lock-up occurs after a few cycles, confirming the suitability
of friction-grip joints in zinc-coated steel structures.
C. Antisparking
Above a certain impact energy, contact between aluminum (and other light
metals such as magnesium or titanium) and rusty steel can cause sparks
capable of igniting combustible mixtures of gas and air such as are found in
coal mines.
Impact between alloys containing enough of these constituents and
rusted steel or iron can similarly create incendiary sparks. Alloys containing
smaller amounts of these metals do not exhibit the same sparking phenome-
non and are deemed safe for use in hazardous environments including those
underground.
British Coal permits the unrestricted underground use of alloys that
contain aluminum, magnesium, and titanium, provided the total content of the
three constituents does not exceed 15% by weight of the alloy and the content
of magnesium and titanium together does not exceed 10% by weight (National
Coal Board Specification 481/1965). The results of tests by the U.S.
Bureau of Mines support the alloy composition limits set by the British
specification. Accordingly, cast alloys 2, 5, 7, 8, and 12 and all sheet
alloys (except the superplastic alloy with 22% aluminum) and all zinc coat-
ings normally used (except those containing 30 or 55% aluminum) may be
considered to nonincendiary and nonsparking. This characteristic means
that these materials will not ignite hazardous gas mixtures, vapors, or particu-
late matter when struck by rusted ferrous materials.
D. Electrostatic and Electrical Properties
Zinc spraying of plastics or phenolic-impregnated asbestos is used to provide
reflecting surfaces, and the large open-air dishes used in electronic applica-
tions are so coated. In impressed current protection of rebar in concrete, the
exterior of the concrete is sometimes zinc sprayed (Morrow, 1991) to give a
uniform current distribution (see subsection on impressed current systems in
Section 11. E); in such cases, it is often policy to avoid direct electrical
connection between zinc and rebar, since the consequent use of zinc as a
sacrificial anode could counterproductive if only a limited amount of zinc
is present to protect large areas of steel.
Chapter
E. Environmental Suitability and Recycling
The soil contains a large number of zinc compounds. The average zinc
content in soils is approximately 70 ppm (= 70 g/ton). Zinc is a most im-
portant trace element, acting in a large number of biological processes. Tap
water typically contains 0.5 ppm zinc, although much higher levels can occur
and tend to be beneficial. In water purification plants, 5 0 4 0 % of this zinc is
separated and removed with the sludge. Zinc must not therefore, be consid-
ered to be an undesirable heavy metal like lead and cadmium. plant and
animal life depend on the presence of zinc. The amount of zinc in the soil
underneath hot dip galvanized structures depends largely on the corrosivity of
the surrounding air. Widespread research shows that this amount of zinc from
galvanized structures not cause problems of health, especially when the
galvanized structure has been painted.
Recycling of zinc from old, weathered structures is done in pickling
baths, where hydrochloric acid dissolves zinc, forming zinc chloride, which is
used for manufacturing many zinc compounds. Scrap from galvanized parts
must be separated from ordinary scrap before being recycled. Because zinc
boils at 906C and steel melts at 14OO-15OO0C, the former will evaporate and
be precipitated on the filter cloth as zinc oxide, which contains zinc.
In Germany, one large zinc smelter handles annually tons of zinc
oxide, which relates to tons of zinc. The zinc requirement in Germany
is 25% covered by recycled zinc. Other sources of secondary zinc are zinc
ashes, dross, zinc die castings, and zinc scrap. One-third of these materials is
made into primary zinc. The rest is turned into zinc dusts, zinc oxide, and zinc
compounds.
De la Cuadra has surveyed the treatment of zinc-coated steel
sheet scrap including the wet processes (Waelz, Contop, St. Joe, Plasmadust,
and Inmeco) and the dry processes (Cebedeau and Cardiff) and outlined the
new CENIM-LNETI process (developed by the Spanish metals research body
(CENIM) with others) involving leaching with ammonium chloride, all of
which demonstrate that zinc-coated steel scrap gives no undue problems in
treatment.
VII. PRACTICAL EXPERIENCE: CASE HISTORIES
Chapters 2-5 end with some lists of case histories, often with details of the
life to date or to first maintenance. These case histories mainly relate to zinc
coatings, where corrosion resistance is the prime requirement, but also to zinc
roofs and to cast zinc in some specific aggressive situations.
a general rule, because of the near-linear or lower corrosion rate of
zinc with time, inspection after years or more is adequate to form a good
idea of the total life or the date when first maintenance will be necessary. The
elimination of much sulphurous pollution in the atmosphere in recent years
means however that the future life of a zinc coating in any specific corrosive
location is likely to be 50% longer than estimated from case histories estab-
lished or more years ago. Only in a few cases can there be even qualitative
direct comparison with other materials, but these are included where possible.
Many other uses in corrosive environments have been documented by zinc,
galvanizing, and thermal spraying organizations over the years, and often
these could form good long-term case histories, but no positive inspection has
taken pl acedal l s for site visits usually arise from unexpected problems
rather than to see a satisfied customer; some of these apparently successful
uses are also documented in the relevant chapter of the book. However, many
galvanized objects can easily be inspected at close quarters, such as lamp
poles, traffic signs, guard rails, and fences.
2
Resistance to Atmospheric Corrosion
INTRODUCTION
Atmospheric corrosion was defined by Schikorr (1965) to include corrosion
by air at temperatures between -20 and 70C, both in the open and in
enclosed spaces of all kinds. Corrosion in the atmosphere is sometimes called
weathering.
Atmospheres are usually classified subjectively (Table 2.1) as indus-
trial, urban, or rural, with the classification usually corresponding to high,
medium, or low sulfur dioxide, respectively. The numerical value relates both
to the geographical area considered and the year; environmental controls
worldwide are leading to diminution in the amount of sulfur dioxide in the air.
A fourth category, marine or coastal, reflects the presence of chlorides from
the sea and should be subdivided into tropical and temperate areas (because of
the difference in seawater composition and temperature) as well as industrial,
urban, or rural.
Zinc and zinc alloys and coatings have good resistance to abrasion by
solid pollutants in the atmosphere. Thus, zinc sheet and galvanized steel have
been widely used for many years. The main practical points are as follows:
1. The corrosion rate is approximately linear with time in most aggressive
atmospheres (Fig. 2.1 and Table hence long-term performance can
predicted if information on behavior over one year-or even less-is
available. Milder atmospheres usually will enable zinc to form a more
protective patina, and if this happens, corrosion rate decreases with time,
101
\Dm
I
103
Specified minimum t hi ckness, pm
Typloal Rves of rlnc coetlng. in seboted envhonments IUK)
Fig. 2.1 User guidelines for zinc corrosion rates: linear corrosion rates in different
types of atmosphere (BS 5493, 1977). Note: The atmospheric corrosion rates shown
here are typically 50% higher than those now prevailing.
with the result that a straight-line relationship has a substantial safety
margin.
2. The corrosion rate increases with an increase in the amount of sulfur
compounds in the atmosphere. Chlorides and nitrogen oxides usually
have a lesser effect but are often very significant in combination with
sulfates.
Corrosion increases with time of wetness. 3.
Chapter 2
d
m - 2
5
b)
m
>
Chapter 2
Tables 2.3-2.6 and Fig. 2.2 give historic data on corrosion resistance of zinc
classified subjectively by environment. The qualitative terms used by authors
clearly have different corrosion significance in different parts of the world.
Some work for which atmospheric pollution data is available is given in Table
2.7A together with a supplement, Table 2.7B. Averages of six l-year tests in
the worldwide ISOCORRAG series, still in progress, have been published,
however (Knotkova, the full data cover steel, copper, and aluminum as
well as zinc. The interpretation of measurements of atmospheric sulfur diox-
ide and chloride is not clear-cut; different measurement techniques can give
substantially different results, and the relationship between corrosion effects
and the particular method of measurement requires further interpretation.
Where data on both zinc and steel corrosion are available, they are
given separately, notably in Tables 2.8-2.11 (ASTM and DIN tests) or as
ratios (Fig. 2.3); however, generally lower corrosion rates are found in recent
tests and reflect the recent lessening of atmospheric sulfur acidity, notably in
the historic industrial counties (Fig. 2.4 and Table 2.7). Most of the results
from work in countries of the former Soviet Union and Eastern Europe are
given in Section 111.2.
Sheltering from direct rainfall may reduce corrosion rates, as is shown
in many of these tests and in Fig. 2.5.
There have been many attempts to classify the rate of corrosion of zinc
in relation to scientific measurements of sulfur dioxide and chlorides, and
time of wetness at specific locations, and this effort has led to I S0 9223
(Table 2.12). However the committee concerned recognizes that available
data do not always fit the classification adequately and that further work is
needed. As just noted, an important series of tests (ISOCORRAG) is still in
progress by a group working within IS0 TC 156/WG4. The published results
are of particular value because four metals have been tested and, in many
cases, some environmental data are available on sulfur dioxide, chloride, and
time of wetness. The published data for zinc and the calculated steekinc
corrosion ratio is summarized in Table 2.7B.
The results are too scattered for close correlation to be made, but
general guidance from the results is that (unless chloride is greater than 120
mg/m2 Cl-), zinc corrosion rate is a basic 0.5-2pm/year plus 1 pm for every
40 pg/m3 SO2 in the atmosphere. Where chloride concentrations exceed 120
mg/m2 Cl-, specific mass loss data must be provided for each site.
The ratios between the losses of helix and flat samples vary considera-
bly; often the helix specimens corrode more than twice as fast-this observa-
tion is in line with some other test results. single feature seems to account
for the differences, but the specimen geometry is an important factor: high
helix/flat corrosion ratios for zinc are often paralleled by similar high ratios
with mild steel, as can be seen from the similarity of steeYzinc corrosion rate
for helices to those for flat specimens.
The ratio between steel corrosion and zinc corrosion is good and in line
with other work. In about 90% of the tests, steel corroded between 11 and 34
times as much as zinc. The interim ISOCORRAG results (Knotkova, 1993)
include calculations of correlation coefficients and regression analysis be-
tween the different metal samples; this material is of particular significance in
the consideration of the general corrosivity of atmospheres.
discussed later, work in the United States has also led to formulas
based largely on sulfur dioxide measurements; U.S. investigators are also
trying to account for the effect of differences in shape (e.g., see Fig. 2.6,
Tables 2.10 and 2.13), while Van Eijnsbergen also comments on the effect of
orientation in solid structures in certain circumstances (Table 2.14). Effects of
time of day (Table 2.15) are not of significance under practical conditions
characterized by continuing exposure.
The acidity of rain is very significant in zinc corrosion and, since the
acidity occurs mainly from sulfur dioxide, further details are in Section ILB.
Essentially, if rain is below about pH 5 (Fig. 2.7) the corrosion rate will be
increased (see Fig. 1.11). Sulfur emissions are discussed by Likens et al.
(1979); zinc is attacked only slightly by pure air, and zinc oxide forms, which
is converted to hydroxide when moisture is present. Even if the moisture
content is considerable, attack remains slight, but the hydroxide films formed
(Schikorr, 1964 a,b) have a relatively minor protective effect. The zinc
hydroxide reacts further with carbon dioxide in the atmosphere, forming a
basic zinc carbonate. This film is very protective and is mainly responsible for
the excellent resistance of zinc to ordinary atmospheres.
Atmospheres change considerably with time, however, as is shown in
Fig. 2.4 for the first 80 years of this century. Environmental controls now
ensure lower sulfur oxides. Present levels in Europe are half those of the peak
period and zinc corrosion is substantially reduced. Recent data, notably from
the ISOCORRAG tests (Table 2.7B, Knotkova, 1993), indicate that current
corrosion rates in Europe are much less than the rates given in Tables 2.3-2.6,
which reflect the high acidic pollution of 30 years ago. Knotkova and Porter
(1994) have documented this in detail and show that zinc coatings now last
3650% longer.
In Section I1 we turn to the effect of climatic variables, including
pollution and initial conditions of exposure. Research to identify the relative
effect of these factors is then covered, leading to the attempts in IS0 to
provide guidance on behavior if some climatic data are available.
Section 111 presents and discusses additional long-term exposure data in
relation to atmospheres. Long-term atmospheric exposures, however, inevita-
Table 2.3 Historic Corrosion Rates of Zinc in Rural Areas
Country Place
Average zinc
Years loss (@year)' Ref.
Czechoslovakia Dry rural atmosphere
Humid rural atmosphere
France Braine
Bray et LQ
Genllany Berlin, Grunewald
Westphalia
Great Britain Banbury
Brixham
Brixham 650 m from
sea
Cardington
Grnialming (Surrey)
Llanwrtyd Wells
Llanwrtyd Wells
Loggerheads (Cheshire)
Abisko
Abisko
Phoenix, AZ
Phoenix, AZ
Sweden
United States
~~ ~
4
12 (?)
0.8
0.9
0.83 - 1.5
1
5.6 2.011 .5/0.4b
5 1. 1 - 1.8
2.0 - 2.5
2.8 - 3.0
1.1
2.5-4.7
5 2.3
2.4
5 0. 3
0.4 - 0.9
10-20 0.1 -0.3
0.17 - 0.20
~ ~
Barton and Beranek (1959)
Dreulle and Dreulle (1973)
Schikorr and Schikorr (1943)
Daeves et al. (1938)
R.M. Evans (1972)
Hudson and Stanners (1953)
Gilbert (1953)
Hudson (1929)
Hudson (1938)
Hudson (1938)
Hudson and Stanners (1953) n
P
Hudson (1938); Hudson and q
S tanners ( 1953) t4
Anderson (1955) r
State College, PA 2 0.8-1.1 ASTM (1956), (1962) A
State College, PA 0.8 Hudson and Stanners (1953) 2.
Hudson and Stanners (1953) ra
Truitsburg, PA 2.5
Type of atmosphere Place B
a
Hudson and Stanners (1953) a-
Aro, Nigeria 5 0.46 - 2.3 Hudson (1938); Hudson and 4
tropical Stanners (1953) 7
n
Sanyal et al. (1959) $
Cristobal, Canal Zone 4-8 0.8 Forgeson et al. (1958) T
a
Nkpoku, Nigeria 5.
(jungle) J
2
n
&
J
South Bend, PA 1.8 Hudson and Stanners (1953)
B
Dry inland tropical Basra, Iraq 5 0.4 - 1.7 Hudson (1938)
Humid inland
U Khartoum, Sudan 5 0.2 - 0.7
Bombay, India 4.2
wet season 2.8 Hudson and Stanners (1953)
1.8 Hudson and Stanners (1953) dry season
whole year 0.8 Hudson and Stanners (1953)
Panama 16 0.9 Southwell et al. (1976)
'1 micrometer (Fm) = 0.00004 in. = 0.001 mm.
bskyward, ground facing, and boxed, respectively.
Table 2.4 Historic Corrosion Rates for Zinc in Urban Areas
Average zinc
Country Place YearS loss (pJn/year)" Ref.
Austria Jenbach (Tyrol)
Czechoslovakia (not named)
France Levallois-Perret
Marseilles
Tonnay-Charente
Tours
Germany Berlin, center
Berlin-Dahlem
Hamburg-Eppendorf
Niedernhausen (Hessen)
Rostock
Sindlfingen (Wurtt)
stuttgart
Great Britain Barking
Birmingham
Birmingham
Bournville
7
1
1
4.4
6
3.0
1 .o
1.4
1.8
5.3-6.8
2.4-3.1
4.0
2.5-3.1
4.3
2.0
1.6-3.6
4.8
6.1
10.0
4.8-5.1
Schikorr and Wassermann
Dreulle and DreulIe (1973)
Brandt (1961)
Schikorr (1962)
9
Friend (1929) w
B
e
Hudson (1929)
Hudson (1929) N
( 1949)
( 1949)
( 1949)
Cambridge
Hornchurch @sex)
Huddersfield
Leicester (Westcotes)
London
Teddington (Middlesex)
Sutton Coldfield
at a height of 9 m
at a height of 190 m
Wakefield 1
India Kanpur 1
Delhi 5
South Africa Pretoria, CSIR~ 5
United States Bayonne, NJ
Keamy, NJ
New York 10-20
New Yo&
New York
Newark, NJ 2
4.5
1.6-3.2
5.4
2.8
3.6-6.6
2.1
4.8
6.1
5.8
5.8
6.6
0.4
0.15
0.3
4.3
3.8
5.3-5.8
5.5-5.8
6.1
3.6
Gilbert (1953) F
2.
v)
Hudson and Stanners (1953),
6
0"
z
CD
n
Hudson (1959)
Hudson and Stanners (1953) $
U
a-
1
A'
n
4
..I
Hudson (1929) 8.
Sanyal and Singhania (1956) 3
Smuts and Callaghan (1978)
Hudson and Stanners (1 953)
Anderson (1955)
ASTM (1956)
ASTM (1962)
~~
'1 micrometer (pm) = 0.00004 in.
bCounci1 for Scientific and Industrial Research.
112
Chapter 2
'9
B
Great Britain Billingham
Coke ovens, boxed
Derby
Dove Holes tunnel
Euston
Euston
Euston
Manchester
Motherwell
Motherwell
Port Talbot, Wales
Salford
Sheffield
Sheffield University
Sheffield University
Sheffield industrial area
Sheffield industrial area
Woolwich
Woolwich
South Africa Pretoria steelworks
United States Altoona, PA
Altoona, PA
2.3
3
5.8
11-13
5.7
5
5
5
10-20
10.6
2.3
7.6
68 - 98
8.1 - 10.4
7.1 -7.9
8. 114.6/2.3b
10.6
4.6
4.9/2.8/0.3b
8.4
4.1 -5.1
10.4- 12.2
7.7 -9.2
5.1
15.2
4.1
3.2
10.9- 19.7
3.4-4.2
4.8-7.8
6.8-7.8
R.M. Evans (1972)
Hudson ( 1959)
Hudson (1938)
Gilbert (1953)
Gilbert (1953)
R.M. Evans (1972)
Hudson (1938)
R.M. Evans (1972)
Shirley and Truman (1948)
Hudson (1938)
Hudson (1938, 1959)
Hudson (1938)
Anderson (1955)
ASTM (1956)
P
r
a
n
B.
CD
8
n
4
8.
2
T
~
micrometer (Fm) = 0.00004 in. = 0.04 mil.
bskyward, ground facing, and boxed, respectively.
4 Table 2.6 Historic Corrosion Rates of Zinc Near the Coast
-L
a
Average zinc
Country Place Years loss (pdyear)" Ref.
Great Britain
United States
On a ship
Marine tropical
countries
Brernerhaven
Dithmarschen, 0.6 m from sea
Dithmarschen, 11 m from sea
Sylt, 80 m from sea
Calshot
Calshot
Morfa Bank; 10 m from sea
Key West, FL,
Key West, FL,
Kure Beach, NC, 25 m
from sea
from sea
from sea
Point Reyes, CA
Sandy Hook, NJ
southport
India
Bombay
B hubanoswar
Cochin
Mandapam: sheltered 450 m
from sea
5
5.6
1
10-20
2
10-20
1
1
1
1
5.5
5.9
6.5
15.1
2.7-3.7
3.3
7.9127.410.3b
5.1 - 5.3
0.5 -0.8
0.55 -0.66
9.9
4.1
1.8
1.3-2.3
1.5- 1.8
6
0.5 - 2.8
0.7
1.8
2.3
Schikorr and Wassermann (1949)
Hudson (1938)
R.M. Evans (1972)
Hudson (1929)
Anderson ( 1955)
ASTM (1956)
ASTM (1962)
ASTM (1967)
Anderson (1955)
Sanyal et al (1959)
Gupta (1 979)
Sanyal et al (1962)
Rajagopalan et al. (1959)
Mandapam: open 450 m
from sea
Singapore
Singapore
Singapore
APaPa
APaPa
Lagos
Lagos: 45 m from sea
Lagos: 180 m from sea
Locomotive sheds
Nkpoku
Port Harcourt: wet season
Port Harcourt: dry season
Port Harcourt: whole year
Port Harcourt: whole year
South Africa
Cape Town docks
Durban
Durban, Salisbury Island
Durban, Bluff
Walvis Bay
Canal Zone
Cristobal
Panama
Nigeria
1
5
5
1
1.1
1.2
5
5
5
5
4-8
16
6.6
0.81 - 1.6
0.8
0.84 - 1.4
0.8
1-9.7
38.9
15.5
7.0
0.8
1.8
2.5
0.5
0.8
3.3
3.8-5.6
2.5
12.1
82.6
3.8 -3.3
2.6
Rajagopalan et al. (1959)
Hudson (1938)
Hudson (1938)
Ambler and Bain (1955)
Ambler and Bain (1 955)
Hudson (1 938)
Forgeson et al. (1958)
Southwell et al. (1976)
'1 micrometer (pm) = 0.00004 in.
2 bSkyward, ground facing, and boxed, respectively. ,
VI
Chapter 2
2.2A Results of tests classified by type of atmosphere (Stahl Informati
1990).
ion,
bly reflect the historic atmosphere, not the future. The trends with time in the
United Kingdom were plotted by Bawden and Ferguson (Fig. 2.8); the
authors comment that when SO2 was reduced, so was the corrosion, but some
increase in corrosion in the early was unexplained. Levels of SO2
remained low (they had fallen typically by since the late and
Bawden and Ferguson query the effect of other pollutants such as nitrogen
oxides and chloride. More recent results, such as those in the Ministry of
Agriculture, Fisheries, and Food (MAW) maps issued. and how-
ever, indicate substantial reductions in corrosion rates of zinc in the United
Kingdom in recent years.
Section IV compares the various zinc alloys in coatings and castings. As
already discussed, the corrosion rates of the various grades of zinc are not
greatly different, especially in open atmospheres, where sufficient oxygen is
present to prevent polarization by hydrogen. Therefore, the results of corro-
sion investigations on zinc, zinc alloys, and zinc coatings reviewed in this
chapter reflect primarily the effect of atmospheric variables and give the
corrosion rates observed during many years of exposure to a variety of natural
l
B
E
B
2
x
-3-3
- 8 M
5
E
d
sf 2
E
d E d
d
-3 l-
m
-3 -3 -3 -3 -3
tz
\o
c c
119
120
Chapter 2
t d d
d d d
'S 'S
s e s
E E E
Resistance to Atmospheric Corrosion 121
atmospheric environments. Usually such data can be used as an initial guide
to the behavior of other zinc alloys.
Section V gives some comparative data and economics, and the chapter
concludes with a selection of case histories in Section VI.
II. INFLUENCE OF ATMOSPHERIC VARIABLES
A. Humidity, Rain, and Temperature
The control of atmospheric corrosion is due to the formation of films com-
prising basic salts, notably carbonate (Tables 2.16 and 2.17). The most
widely accepted formula is 3Zn(OHh 2ZnC03, which may be written
Zn5(OH)6(C03)2. Environmental conditions that interfere with the formation
of such films, or conditions that lead to the formation of soluble films, may
cause quite rapid attack on zinc. One most important factor affecting the
corrosion of zinc in the atmosphere is the duration and frequency of moisture
contact.
The rate of drying is an important factor because a thin film of moisture
with high oxygen concentration promotes reaction. For normal exposure
conditions, the films dry quite rapidly, and it is only in sheltered areas that
drying times are so slow that the attack of zinc is accelerated significantly.
Stiles and Edney (1989) have studied thin acidified films of water to show a
steady rate of dissolution of zinc until the acidity is destroyed.
The effect of atmospheric humidity on the corrosion of zinc is related to
the conditions that may cause condensation of moisture on the metal surface
and to the frequency and duration of the moisture contact. If the air tempera-
ture drops below the dew point, moisture will be deposited. The thickness of
zinc, its surface roughness, and its cleanliness also influence the amount of
dew deposited. Lowering the temperature of a metal surface below the air
temperature in a humid atmosphere will cause moisture to condense on the
metal. If the water evaporates quickly, corrosion usually is not severe and a
protective film is formed on the surface. If water from rain or snow remains in
contact with zinc when access to air is restricted and the humidity is high, the
resulting corrosion can appear to be severe (wet storage stain, popularly
known as white rust), since the formation of a protective basic zinc
carbonate is prevented.
It is believed that the corrosion products that form when items are
exposed to acid rain are more hygroscopic than those formed under more
neutral conditions. This, say Haynie et al. (1990), effectively lowers the
critical relative humidity from 90% to 75%, although this value is higher than
the critical relative humidity for steel corrosion. The effects have been
demonstrated both in the laboratory and by examples of corrosion occumng
Table 2.7B Addendum: Site Atmospheric Characteristics,
to Zinc Comsiona
Zinc corrosion, Steelhinc ratio (mean)
l-year results (pm) (I-year results)
Helix
Test site Code Flat of wire Flat Helix
lguazu
Camel
Buenos Aires
San Juan
Yubany Base
Boucherville
Kasp Hory
Raha-Bechov
Kopisty
Helsinki
Otaniemi
Athari
St. Denis. Paris
Ponteau Mart
Picherande
St. Remy
Salin de Gir
Ostende (Belgium)
Paris
Auby
Biarritz
Bergisch Glad
Choshi
Tokyo
Okinawa
Judgeford, Wellington
Oslo
Bomgaad
Birkenes
Tananger
Bergen
Svanvik
Murmansk
Batum
Vladivostok
Oymyakon
Madrid
El
Lagoas-Vigo
Baracaldo-Vizcaya
Stockholm Vana
Kattesand
Kvarnvik
Stratford
Crowthome
Rye
Fl eet Hall
Kure Beach
Newark
Panama
Research Triangle
Point Reyes
Los Angeles
ARGl
ARGZ
ARC4
ARG5
CND1
CS I
CS2
SF1
SF2
SF3
FI
F2
F3
F4
F5
F6
FI
F8
F9
D1
J I
J2
J3
NZ I
NI
N3
NZ
N4
N5
N6
SUI
su2
s u3
s u4
El
E2
E3
W
SI
S2
S3
UK2
UK I
UK3
UK4
us2
us3
us4
us5
US6
1.6
I .3
I .o
0.2
I .9
I .4
2.8
.9
3.5
1.3
0.9
0.7
2.6
I .5
0.9
I .5
4.6
5. I
3.0
5.6
4.3
I .6
I .4
.5
3.4
0.7
3.8
I .3
2.3
2. I
3.0
0.8
1.1
I .6
2.3
0.4
0.6
0.5
I .2
I .o
0.6
I .5
1.8
I .7
1. 1
2.5
1.3
2.0
17.5
2.0
0.8
I .7
1.1
2.0
2.2
3.3
4.8
2.6
1.8
3.6
I .2
13.4
2.2
4.2
5.7
10.6
2.8
8.5
8.2
2.8
1.8
1.5
8.8
1.2
1.8
5.7
2.0
3.3
2.2
2. I
I .4
2.0
0.6
3.1
1.6
2.5
I .2
2.6
I .5
2.8
3.5
1.5
I .2
2.0
2.3
3.9
2.2
7.6
3.5
I .8
28
4
16
23
19
17
14
17
20
26
28
18
25
28
29
18
16
19
14
19
20
23
31
26
22
29
19
16
20
9
25
13
28
18
I 1
2
46
31
27
37
41
23
23
34
29
23
21
13
28
23
20
22
14
21
23
16
21
13
14
9
IO
23
23
12
18
17
8
29
33
26
12
30
19
16
14
22
15
21
25
14
22
18
3
18
14
22
28
22
19
48
46
25
21
12
39
42
I I
zinc cnrosion results normally the mean of 18 determinations (three replicates for six
l-year exposures) starting spring and autumn for years.
"Wherc no atmospheric data are available. a qualitative description of the site is given.
Zinc One-Year Corrosion Rates, and Ratio of Steel Corrosion
Environmental characteristicsb
Time of wetness
SO2, mean (TOW), mean
CI-, (mglm) (Myear)
(Semiarid, wet, rural )
(Subtropical, marine, wet)
(Subtropical. marine, wet)
(Subtropical. dry. rural )
(Antarctic. desert)
(Rural)
(Urban)
(Industrial)
4
3
(Rural)
28
24 I
378
7
173
(Urban)
16
I93
(Urban)
56
97
4
2
9
(Acid rain)
308
7
I
19
1
18
(Cold)
(Urban)
(Urban)
59
(Rural-marine)
29
25
(Urban)
76
650
(Industly)
(Rural)
(Marine)
(Urban)
I02
(Industrial)
619
(Urban)
(Marine)
(Urban, marine)
16
17
67
90
19
15
50
4
87
30
9
20
24
53
188
18
8
15
I I
44
14
I
4
17
9
26
5
29
5
5
49
32
IO
5
5
20
21
IO
52
20
5680
6088
4645
855
2693
1396
3206
299 I
2444
3578
3256
3 l05
4268
3846
4171
63
331 I
6083
3189
457 I
4267
5704
2173
3852
2641
3339
4138
4583
4439
2605
3227
3216
3920
38 I
2060
3223
2840
4375
5783
4289
7598
4003
CMarine. splash
Source: Knotkova
Table 2.8 Average Loss of Zinc in 2 Years and SteeVZinc Ratio for Rural and Semirural Atmospheres
-L
N
P
Location or source
Calculated life
of 100 pm zinc
2-year test: Steellzinc coating without
zinc lost per loss ratio maintenance
Described as year (w) (by weight) ( ye a ) Ref."
Norman Wells, Northwest Rural 0.2 10.3 500 A
Phoenix, AZ Rural 0.3 17.0 300 A
Saskatoon, Saskatchewan, Rural 0.3 21 .o 300 A
State College, PA Rural 1.2 22.0 80 A
State College, PA Rural 1.2 19.7 85 B
State College, PA, for 8 years Rural 1 .o 12.7 96 B
Morenci, MI Rural 1.2 18.0 80 A
Potter County, PA Rural 1.3 18.3 75 A
Durham, NH Rural 1.6 19.0 60 A
South Bend, IN Semirural 1.8 20.8 55 A
South Bend, IN Semirural 1.9 18.8 54 B
South Bend, IN, for 8 years Semirural 1.9 11.7 52 B
Esquimalt, British Columbia, RuraVmarine 0.5 31.0 200 A
Ottawa, Ontario, Canada
'A, ASTM STP 435 (1968); B, Larrabe-e and Ellis (1959).
Temtories, Canada
Canada
Canada
Urban/rUral 1.1 19.5 90 A 3
m
B
2
N
under actual service conditions (Gilbert and Hadden, 1950a,b; Anderson,
1955; Schikorr, 1964, 1965). (Corrosion in condensate and dew is further
discussed in Chapter 3, Section I.B.)
Several formulas have been reported to represent the composition of the
basic carbonate films, among them being 2ZnC03 - 3Zn(OH)2 (already men-
tioned), 2ZnC0, * Zn(OH)2, and ZnC03 * 3Zn(OHh. Feitknecht (1959) stud-
ied the structure of various basic zinc carbonates. He indicated that the type of
film to provide corrosion protection for zinc crystallizes in a disordered layer
structure, giving dense, adherent films. In his work, the white corrosion
product from a galvanized outdoor structure gave an X-ray pattern that cor-
responded closely to the foregoing composition.
Atmospheric corrosion of zinc is roughly proportional to the time of
wetness in a particular location, a point emphasized by Mikhailovskii et al.
(1986) for areas of the former Soviet Union, provided the nature and quantity
of environmental pollution do not change; a high relative humidity, which can
cause condensation, increases corrosion. Rain obviously increases time of
wetness, but it can have an indirect beneficial effect by removing corrosive
materials. In practice, time of wetness is often taken as the time when relative
humidity (RH) exceeds 80% and the temperature is above 0C. Thin layers of
solutions (except acids) are more corrosive than bulk solutions (Mansfield and
Tsai, 1979). The general consensus is that the significance of atmospheric
humidity in the corrosion of zinc is related to the conditions that may cause
condensation of moisture on the metal surface and to the frequency and
duration of the moisture contact.
Time of wetness is normally secondary to the effect of atmospheric
pollutants-notably sulfur oxides and other acid-forming pollutants, which
both have a direct effect and can also make rain more acid (Fig. 1.1 l).
In less corrosive areas, however, time of wetness assumes greater
proportional significance, as documented by Mikhailovskii et al (1986) and
Panchenko et al. (1985). The effect of humidity of purified air has been
studied by several people, including Vernon (1935) and Rajagopalan and
Ramaseshan (1959). Vernon showed a linear rate of oxidation over 70 days at
24C and 99% RH equivalent to 0.05 pdyear , with rates slightly lower for
lower humidities.
Rajagopalan and Ramaseshan (Fig. 2.9) confirmed linear rates (up to
2.2 pdyear) but showed greater influence of relative humidity when the
temperature was allowed to fluctuate between 28 and 38"C, giving tempera-
ture differences between the metal and surroundings and permitting condensa-
tion; condensation obviously will occur more often if the air is more nearly
saturated with moisture. The chalky white layer that formed in their tests
127
128 Chapter 2
Table 2.10 Average Loss of Zinc in Years and SteellZinc Ratio: Marine Atmospheres
Calculated life of
pm zinc
2-year test: Steellzinc coating without
zinc lost per loss ratio maintenance
Location source year (pm) (by weight) (Years) Ref."
Fort Amidor Pier,
Panama, CZ
Miraflores, Panama, CZ
Manila, Philippine Islands
Point Reyes, CA
Point Reyes, CA
Point Reyes, CA,
for years
Daytona Beach, FL
Daytona Beach, FL
Daytona Beach, FL,
Kure Beach, NC, m lot
Kure Beach, NC, m)
Kure Beach, NC,
Kure Beach, NC, m lot
Pemne, FL
Pemne, FL, for years
Limon Bay, Panama, CZ
Dungeness, United Kingdom
Cape Kennedy,
ground level
Cape Kennedy,
m up
Cape Kennedy, FL,b
m up
Galeta Point Beach,
Panama, CZ
for years
for years
.o
.O
60
A
A
A
A
B
B
A
B
B
B
A
B
B
B
A
A
A
A
A
A
'A, ASTM Task Force, ca. 1960 (ASTM STP 435, 1968); B, tests from 1948 to 1956 ( h b e e and Ellis,
1959).
Tape Kennedy site is 55 m from the Atlantic Ocean.
Table 2.11 Loss of Zinc in Atmospheric Exposure
2-year test: Steellzinc Calculated life of pm
zinc lost per
loss ratio zinc coating without
Described as year (pm)
(by weight) maintenance (years)
Rural 6 -
Urban .O - -
Industrial" 5-25
Marine"
Tropical
"DIN gives pm for industrial; 2-4 pdyear for marine.
Source: Bottcher andlor DIN
2.3 ASTM tests in Tables classified by atmosphere: R = rural, U =
urban, I = industiral, M = marine, S-R = semirural, R-M = rural-marine, S-I =
semi-industrial, I-M = industrial-marine.
130
131
0-
Tinsley
I
Top deck
site
I
1Imdd:S 11#ml yror for zinc)
- 2 Year
Valley
site
. 2 year
M 'baco
1 Open exposure I
Sheltered
specimens
I girder
I specimens Viaduct I
Porlishead
d
W
h,
3
nl
Fig. 2.5 One- and 2-year corrosion rates for openly exposed and sheltered zinc in U.K. sites: 1 mdd = 5.14 pdyear for zinc 3 7
h,
(Hutchins and McKenzie, 1973).
appears to be white rust, and in later work (Rajagopalan and Ramaseshan,
1960) very erratic results were reported for tests made in small stoppered
bottles without apparent condensation. It is likely, however, that insufficient
carbon dioxide was present to form a protective basic zinc carbonate. It
appears that a protective film was allowed to form at 100% RH (unlike 65-
95% RH) at temperatures up to 60C. When white rust formed in other tests,
however, the corrosion rate might increase with time.
Anderson and Fuller (1939) also noted that the moisture content of the
atmosphere determined the rate at which the carbonate film was formed in
open atmospheres. The time required to develop a film that was thick enough
to prevent detection of the zinc substrate by electron diffraction ranged from
98 days in dry air to 14 days at 33% RH and 1-6 days at 75% RH.
It is possible to obtain high corrosion rates over short periods, as
demonstrated by Gilbert and Hadden (1950a), who achieved a loss of nearly
2 pm in one week even under reasonably constant room temperatures in
humidity tank tests with twice-daily spraying of distilled water. They also
showed that when distilled water was spread into many small droplets, attack
in the first week was greater than if the water was in larger drops. Of more
significance is the longer test of this type (Fig. 2.10), which showed that a
protective film then formed. However, the rate of metal loss calculated from
short-time experiments cannot be extrapolated to indicate total loss in metal
thickness for relatively long periods if protective films or other factors tend to
slow or stop corrosion.
Dew affected by industrial or marine pollutants is likely to be corrosive
to zinc because of the depolarizing effects of Cl- ions and the complicated
reactions between Zn, CO2, H20, and NO,. In such circumstances, the
corrosiveness of the contaminant may be more important than the degree of
moisture condensation. Vernon (1935) showed that corrosion in purified air
was very slight at relative humidities up to 99%. However, the presence of
0.1% sulfur dioxide in the air caused a marked increase in corrosion when the
relative humidity exceeded about 60%. Vernons results illustrating this are
shown in Fig. 2.11. Although Vernon also shows a secondary critical humidi-
ty about 90%, this is of little significance because at 99% RH, the linear
corrosion rate is similar to that at 70-80% RH. Barton and Bartonova (1966)
studied the corrosion products formed in 32-day tests at 70-95% RH with
small amounts of hydrogen chloride. The composition was constant and the
corrosion-time relationship was linear. More details of the combined effect of
humidity and pollutants are given in the subsection entitled Other Pollu-
tants in Section 1I.B.
Low temperatures, such as those encountered in a deep-freeze room, do
not accelerate attack on galvanized sherardized coatings (Sjoukes, 1970) in
prolonged cycles of + 40 to - 24C.
134 Chapter 2
Table 2.12 Derivation of Likely Corrosivity of Atmospheres to Zinc
Corrosivity Corrosion rate,
category steady state
(by IS0
9223) glm2/yb P d Y W
c1 <0.7 <0.05
c 2 0.75 0.05-0.5
c 3 5-15 0.5-2
c4 15-30 2-4
c 5 30-60 4-10
<I0 (<O.I%)
Indoors, climate contml
lndushial pollution by SO2 Airborne salinity chloride deposition rate?
mglmzlday
Classification Concentration Deposition
by IS0 9223 (~g/m) (rnglm21day) 3-60 60-300 300-1500
Po <I2 <I0 I I I
PZ 40-90 35-80 1 I 1 or 2
PI 12-40 10-35)
90-250 60-200 1 2 or 3 2
Torrosivity is expressed as the number part of the comsivity category code (e.g.. 1 instead C 1) according
to Table 6 in IS0 9223.
bAccording to Table 5 in IS0 9223.
According to IS0 9224 (steady-state corrosion rate derived from long-term exposures).
dAccording to Table 3 in IS0 9223: the C prefix (as in, e.g., column I ) is omitted.
According to Table I in IS0 9223.
B. Effect of Impurities
Sulfur Compounds
Sulfur dioxide is the most important individual contaminant to be considered
in atmospheric corrosion of zinc. At relative humidities of about 70% or
above, it usually controls the corrosion rate. Figure 2.12 shows the linear
relationship for zinc and contrasts it with the way in which the unit influence
of sulfur dioxide on steel decreases with concentration (although the total
amount of corrosion on steel is of course much more).
The types of corrosion product formed are shown in Table 2.16. Much
research on the corrosion products has been done (e.g., Grauer and Schl-
indler, 1972, studied the solubility of zinc hydroxychloride), while Spence
and Haynie (1990) have summarized equilibrium constants for selected acid-
gas reactions (Table 2.17). The effect of other acid-forming sulfur compounds
such as sulfur trioxide is probably similar to that of sulfur dioxide. Further-
135
Time of wetness (Myear)'
>
Indoors, no climate control Outdoors in dry and cold Outdoors in temperate cli- Outdoors in damp climates,
except in damp climates climate; ventilated sheds mate, unventilated sheds in humid unventilated sheds
in temperate climate temperate climate; ventilated
sheds in damp climates
Airborne salinity chloride Airborne salinity chloride Airborne salinity chloride Airborne salinily chloride
deposition rate: deposition rate: deposition rate: deposition rate:
m@m2/day mglmz/day mglm2/day mglm2/day
1 l o r 2
5
more, in the cold season the sulfur contamination of the air is greater. Barton
(1970) looked at the basic mechanisms of corrosion, including the increase of
corrosion with temperature and sulfur dioxide.
Sulfur oxides and other corrosive species are brought to react with the
zinc surface in two ways: dry deposition and wet deposition. Sulfur dioxide
has been observed to deposit on a dry surface of galvanized steel panels until a
monolayer of formed (Maato, In either case, the sulfur dioxide
that deposits on the surface of the zinc forms sulfurous or other strong acids,
which react with the film of zinc oxide, hydroxide, or basic carbonate to form
zinc sulfate. The conversion of sulfur dioxide to sulfur-based acids may be
catalyzed by nitrogen compounds in the air-usually referred to collectively
as NO, compounds-and it is believed that this factor may affect corrosion
rates in practice. The acids partially destroy the film of corrosion products,
which will then re-form from the underlying metal, so causing continuous
corrosion by an amount equivalent to the film dissolved, hence to the amount
of sulfur dioxide absorbed. Above about 85% RH, corrosion rates increase
further-probably as a result of the formation of basic zinc sulfates.
The absolute sulfur dioxide content of the air is less important than the
amount of sulfur dioxide deposited per unit time on the unit area exposed,
which can be affected by the geometry of the surfaces. example, there is
UNFAsRlCATED FIELO W E D WI N S M N AVER4GE
WIRE FWCE WIRE W K WIRE OF
FENCE STRIM) MI.
PROWCTS
PRODUCTS
CMIPARSON OFQJRROSION RATESTO YITIALRU~~NG (QRI AVERCX~E
OUNCES PER SOn PER YEAR. MUMINUM n ZINC COATED PRODUCTS
LOCATION.WARRINGlON. ENGUM)
COUPARISON OF CORROSION RATES TO NmAL RUSTING (CtR] AVERAGE
OUNCESPERSOFTPERYEARMUMWMnZINCCOATEDPROOUCTS
LOWTWN. WlNT RMES. W F
PRODUCTS
Fig. 2.6
ft2 = 43 pm (Kelley, 1975).
Corrosion rates for zinc-coated (and aluminum-coated) wire products of various shapes; 1 od
d
W
U
138 Chapter 2
Table 2.13
to Various Environments Prior to 1954
Corrosion Rates of Zinc on Galvanized Steel Products Exposed
Mean corrosion rate and estimated
standard deviation (pm/year)
Site
~~
Sheet Wire and fencing
~
Altoona, PA
Pittsburgh, PA
Sandy Hook, NJ
Bridgeport, CT
Lafayette, IN
Ithaca, NY
State College, PA
Ames, IA
College Station, TN
Santa Cruz, CA
Manhattan, KS
Davis, CA
___ ~~ ~~ ~ ~ ~
7.57 f 0.54
5.63 f 0.34 10.86 & 1.02
2.74 f 0.30 4.37 f 0.45
4.25 f 0.44
2.94 f 0.34
2.68 4 0.42
1.27 f 0.29 2.48 f 0.24
1.68 f 0.19
1.22 f 0.43
0.83 2 0.26
0.79 f 0.27
0.76 f 0.42
Source: Haynie and Upham (1970).
usually greater deposition on wires than on flat surfaces. Sulfur in the form of
zinc hydrogen sulfate is less corrosive than sulfur in the form of zinc sulfate.
Sydberger and Vannerberg (1972) suggest that a relatively high SO2 adsorp-
tion rate decreases with time and also that the adsorption rate depends on
humidity. Duncan and Spedding (1973) confirmed this theory and looked at
Table 2.14
of Galvanized Steel: Residual Thickness After 20 Years
Effect of Orientation to Prevailing Wind (West) on Corrosion
Coating thickness (pm)
Minimum Maximum Average
Orientation
of surface Upper Lower Upper Lower Upper Lower
To southwest 80" 100 120 150 90 155
To northwest 60" 90 90 110 75 100
To northeast 100 150 140 175 120 160
To southeast 90 90 130 150 105 120
"Rusted area: pinpoints of rust after 20 years' exposure in an industrial/marine climate.
Source: Van Eijnsbergen (1982b and 1986b).
2 0 J-
1 5 --
I 0 -1
.-
5 --
I I I 1 A
139
- - 2 f l N r
- 1 pmvmlyr
Table 2.15
at Palmerton, Pennsylvania
Effect of Time of Day on Zinc Corrosion Rates
Corrosion rate (prn/year)
Exposure period First panel Second panel
8.15 A. M. to 4.15 P.M. 1.70 1.91
4.15 A. M. to 12.15 P. M. 1.85 2.11
2.84 12.15 A. M. to 8.15 A.M.
Total for 3 periods 6.20 6.86
Average per period 2.06 2.29
Continuous exposure 2.06 2.29
Duplicate panels (10 cm X 15 cm) were exposed vertically.
bBased on cumulative time exposed; 2 years for the periods and 6 years for the
continuous test.
Source: Anderson and Reinhard (1939).
-
2.64
-
the conversion of S(1V) to S(V1) but found difficulty in establishing a reaction
model.
Much research has been done on atmospheres containing sulfur dioxide
either as the only pollutant or with other factors not deliberately varied.
Schikorr (1964-1965) in a 3-day test showed that up to 85% RH, the weight
Average
Corroslon
Rate
gim*/yr
p n
Fig. 2.7
1991).
Relationship between corrosion and pH of simulated rainwater (Goodwin,
Chapter 2
' O
1967 1969 1971 1973 1975 1977 1979 1981 1983
Year of Measurement
2.8 U.K. trends in corrosion of zinc, 1967-1982 (Bawden and Ferguson,
1989).
Table 2.16 Zinc Corrosion Reactions in Different Atmospheres (Schematic)
Corrosion products Relative
Type of Attacking solubility Corrosion
atmosphere substances" Composition in water rate
Rural + H,O + ZnO + Zn(OH), Very low Very
Marine + H20 + ZnO + Zn(OH),
CO, + 2ZnC0, * 3Zn(OH),
CO2 + + 2ZnC0, 3Zn(OH), + Moderate Low
{
ZnC1, 4Zn(OH), + Zn,0C14
ZnCl, 6Zn(OH), [ Zn40C1,
Urban and 0, + H20 + ZnO + Zn(OH),
industrial CO, + + 2ZnC0, 3Zn(OH), + Good High
+ ZnS + ZnSO, + ZnSOl
TO2 has been included as an attacking substance because it participates in the formation of corrosion
products. However, CO2 is also necessary in order for stable films to form.
Source: Thomas (1980).
142 Chapter 2
Duration of test(months)
0 97% relative humidity
o 85 and 76% relative humidity
Fig. 2.9
jagopalan and Rameseshan, 1959).
Changes in the effect of humidity with fluctuating temperature (Ra-
of zinc corroded was approximately equivalent to the sulfur dioxide content of
the corrosion products (Fig. 2.13). He suggested that above 85% RH the
available sulfur dioxide had been used up and a basic zinc sulfate formed.
Barton and Beranek (1959) showed that in sulfurous acid atmospheres (up to
0.50 vol % SO2), corrosion increased linearly with time over 32 days and
approximately linearly with SO, contents up to 0.10%, after which the unit
effect decreased (Fig. 2.14). However, the actual deposition of sulfur oxides
is the determining factor rather than the atmospheric concentrations.
The quantity of sulfur dioxide deposited may be measured by absorption
on an alkaline surface (Liesegang bell jar method) or on a surface coated with
lead peroxide (Fuel Research method).
An example of the correlation between SO2 absorbed from the air by the
Liesegang method, relative humidity, and weight loss of zinc in a 3-year test
at Berlin-Dahlem is given in Fig. 2.15. This plot shows that all three factors
move together in the same direction, varying in accordance with the time of
year. It was pointed out, however, that the quantity of sulfur dioxide absorbed
in the Liesegang test is probably greater than that absorbed by the zinc
Exposure (days)
I I I I I I l l I I I
0 2 4 6 8 10 14 18 22 28 38
Nurnbef of c i m sprayed
Each point repmsents average IOU for 2 sheets
Fig. 2.10
distilled water (Gilbert and Hadden, 1950a).
Galvanized sheets at 100% relative humidity, sprayed twice daily with
surface. A more direct comparison of corrosion versus sulfur dioxide reacting
with the zinc was made by Schikorr (1961) in his Stuttgart tests described in
detail in his 1965 book. The corrosion products (on the zinc surface plus those
washed off by rainfall) were analyzed for sulfur dioxide (as well as for zinc,
as shown in Fig. 2.16), and in most months the sulfur dioxide content used in
forming the corrosion product was proportional to the total amount of corro-
sion product, except in August and September, when the sulfur dioxide
content was low.
Haynie and Upham (1970) compiled data for the atmospheric sulfur
dioxide level, relative humidity, and zinc corrosion rate for eight U S . cities.
Using regression and curve-fitting techniques, they developed the equation:
Y = 0.001028 (RH - 48.8)SOz
144 Chapter 2
I I l I l l
0 10 30 50 60 70 90 99
Relative (%)
2.1 1 Effect of relative humidity in air containing 0.01% SO2 (Vernon, 1935).
where Y is the corrosion rate of zinc (pdyear), RH is the percent average
relative humidity, and SO2 is the atmospheric sulfur dioxide content (pg/m3).
The equation has excellent correlation with observed data, as shown by
Fig. 2.17, which compares predicted and observed corrosion rates. This
equation accounts for 92% of the variability of the corrosion rates (R2 =
0.92). Thus, only the 8% residual variability cannot explained by the
equation. This formula should used with caution, however. It was devel-
oped from corrosion rate data collected at sites with fairly high humidity and
sulfur dioxide pollution levels. Thus, it predicts no corrosion when the
99
0 20 40 60 80
SO, Concentration pg1m-3
90
80
70
60
50
40
30
145
Fig. 2.12 Atmospheric corrosion rate of zinc, copper, and steel related to sulfur
pollution in rainwater; precipitation rain average, 1 m2 year-'; duration of wetness,
0.25-0.32; pH values of rain on small surfaces (Van Eijnsbergen).
average relative humidity is lower than 48% or when atmospheric sulfur
dioxide is not present. Actually, zinc will corrode under such conditions, but
at a low rate compared to the corrosion at the test sites used in this study. The
equation, therefore, cannot be extrapolated reliably to exposures under mild
conditions, such as average humidity less than 60% RH or sulfur dioxide
concentration lower than 25 pg/m3. The equation also should not be used for
predicting corrosion rates for marine environment sites.
In studies on sulfur dioxide, nitrogen dioxide, ozone, and other vari-
ables in specially designed environmental chambers, Haynie et al. (1978)
found that only sulfur dioxide and humidity were statistically significant.
Haynie later (1980) evaluated the corrosion of small specimens of galvanized
steel and developed the following relationship:
146 Chapter 2
Relative humidiry (%)
SO, content of corrosion products
0-0 Weight loss
0-0
Fig. 2.1 3
with SOz (13.5 mglL) added daily (Schikorr, 1964-1965).
Corrosion of zinc related to SOz absorbed by corrosion products in air
C, = 2.32tW + (0.0134V0.78' X SO2 x t, )
where
C, = corrosion (pm)
rw = time of wetness (years)
V = wind velocity ( d s )
and SO2 is in micrograms per cubic meter.
He studied the results of six sets of exposure investigations to relate the
corrosion of zinc and galvanized steel to concentration of sulfur dioxide. All
six investigations were different, and the data were evaluated differently, and
thus no direct comparison of the published results were possible. However,
Time (days)
SO, volume %
. 0.00
0 0.01
A 0.50
2 ::::
Fig. 2.14
zinc at 21C (Barton and Beranek, 1959).
Effect of SOz content in air at 100% relative humidity on corrosion of
Haynie reevaluated the data using techniques that permitted the comparison of
various data sets (Table 2.18).
Spence and Haynie (1990) looked at the theory behind the deposition of
gases on surfaces and the solubility of atmospheric gases in water. They
postulate that sulfur dioxide, which has a solubility of 1.24 mol/L atm in
water at 25"C, will readily deposit during periods of surface wetness and will
-
Chapter 2
Relative humidity
- Zlnc corrosion
SO, absorption
Fig. 2.1 5 Relation between zinc corrosion, absorption, and relative humidity
(Schikorr, 1964-1965).
react with the zinc. Calcium compounds on the zinc surface were found to
increase the adsorption of sulfur dioxide on the surface, but the additional
amount does not react with zinc or the protective corrosion film because it
reacts preferentially with the calcium compounds. Therefore, the net effect is
nil.
From an earlier study of equilibrium constants, Spence and Haynie
(1976) concluded that the zinc corrosion c d due to the deposition of gaseous
during periods of surface wetness and dryness may be expressed as
follows:
where
c d = zinc corrosion (pm) due to the deposition of Soz
v d = deposition velocity ( cds )
= ambient concentration (pg/m3)
z, = time of surface wetness (years)
Ar = ratio of actual to apparent surface
N = number of periods between times of surface wetness
24
22
20
18
16
14
12
10
8
6
a
2
0
1959 1960
Zinc content of rainwale, drained from soecimens and collected monthly
Zinc content of adherent corroston film
0
o
Weight loss of specimens
Fig. 2.16
month in a I-year test at Stuttgart (Schikorr, 1961a).
Zinc in adherent corrosion product and in rainfall washings month-by-
Chapter 2
[(RH -
2.17 Effect of SO2 and relative humidity on the corrosion rate of zinc:
calculated versus observed results in eight U.S. cities (Haynie and Upham, 1970).
The first term 0.045Vd(S02)fw represents the dep,sition of during periods
of surface wetness due to condensation (dew). The coefficient 0.045 repre-
sents the stoichiometric reaction of with the wet zinc. surface to form
ZnS04. The V, component determines the speed of deposition of on the
zinc surface from atmospheric measurements of concentrations (pg/m3).
Since the deposition rate depends on the shape of the article, however, V,
is difficult to determine. In polluted environments, fw, may best be estimat-
ed according to Spence and Haynie by using the equation RH, = 96. 03 -
0.313DP (where DP is the ambient dew point temperature, "C), hence deter-
mining the surface wetness galvanized steel structures from ambient dew
point measurements. The last term, 1.29 10-4ArN, represents the corro-
sion of zinc that is associated with the formation of a monolayer of during
periods of surface dryness. This corrosion is shown as a function of the
exposed surface area and the number of periods between surface wetness. The
coefficient 1.29 represents the micrometers of zinc loss of a clean
surface as a result of the absorption of a monlayer of
The part of the earlier Haynie study that showed the relative corrosion
rate of zinc on galvanized steel products (sheet, wire, fencing) in various
locations around the United States (Table 2.13) indicates a ratio of no more
than 2 for wire compared with sheet. It should be noted that the corrosion
rates expressed in this table are two to four times greater when the galvanized
product is wet. Theoretical calculations indicated that the average sulfur
dioxide level at the Pittsburgh site during the exposure period was between
350 and 700 pm/m3, and the average sulfur dioxide level at the Altoona site
was calculated to be as high as lo00 pg/m3.
The equation of Spence and Haynie (which has been demonstrated for
some sites to be in accord with practical results) would indicate the following:
1. During periods of surface wetness, reaching the surface reacts
2. Rain acidity reacts stoichiometrically with the zinc.
3. The corrosion film of basic zinc carbonate is soluble in clean rain. The
dissolution depends on the residence time of rain on the galvanized steel
surface.
4. Deposition velocity controls the rate of corrosion of galvanized steel
structures by gaseous during periods of wetness.
stoichiometrically with the zinc.
already indicated, one of the main problems associated with the
equation is the determination of the deposition velocity of from normally
available climatic data. The calculated differences for sheet and fencing vary
by up to a factor of 10, which is not in line with practical experience, where
normally the factor is about 2. Possibly, aerodynamic effects and relative
mass effects are not adequately incorporated. Twelve-year exposures of zinc-
and aluminum-coated wires (Kelley, 1975) show the differences between
unfabricated wire and various wire products; the high corrosion of barbed
wire is presumably due partly to sacrificial protection of cut ends (Fig. 2.6).
The pH of the environment in industrial atmospheres is simply related to
its sulfur dioxide content. Goodwin (1991) reiterates that corrosive species
such as sulfur dioxide are brought to react with the zinc surface in two ways:
dry deposition and wet deposition.
The several studies canied out on dry deposition in the last 30 years
(Schikorr, 1964-1965; Knotkova et al. 1984; Kucera et al., 1987a,b) agree
152 Chapter 2
well as to the effect of dry deposition on zinc corrosion rates. Kucera (1987)
has shown that atmospheric weight losses of zinc caused by dry deposition of
sulfur dioxide can be represented by the equation:
K = 0.22 S02 t
where K is the corrosive attack on zinc (g/m2/year), SO2 is the arrival of sulfur
dioxide to an active surface (mg/m2/day), and t is the time of exposure
(YeW.
This means that if one mole of sulfur dioxide is dry deposited on a zinc
surface, 0.62 mole of zinc is transformed to corrosion products.
Goodwin (1991) has however observed that a simple chemical reaction
model overestimates the zinc corrosion rates due to wet deposition. The actual
corrosion rates are only 15-90% of the theoretical values. The difference is
believed to be caused by (a) incomplete dissociation of available hydrogen,
(b) the presence of previously reacted zinc on the surface, which limits the
amount of available fresh zinc, and (c) simple washing away of reactants
before they can react with the zinc surface. Rainfalls with pH 3.5 and higher
believed to decrease zinc corrosion rates because they wash the panel,
removing many of the potential reactants before they have a chance to corrode
the zinc surface.
To determine the contribution of dry versus wet deposition, Goodwin
initiated laboratory tests using synthetic rainfall of different pH values. Dry
deposition was attained by setting panels in a louvered box outdoors. Rain
intensities between 0 and 12 mr dh were used. This technique made it possible
to vary sulfate concentration independently of pH. Rolled pure zinc panels
were used for all tests. Atmospheric corrosion rates of zinc is Scandinavia,
when only dry deposition was allowed, were measured to be 12 g/m2/
year based on an exposure period of 2.5 months. Exposure of these dry panels
to artificial rainfalls with pH values of 4.5 and 3.5 decreased their corrosion
rates. Only when rainfall pH was lowered to 2.5 did zinc corrosion rate
increase over that seen with solely dry deposition (Fig. 2.7).
In the series of 8 year atmospheric exposures by Kucera et al. (1987) the
sulfur dioxide deposition rates were measured for each of the exposure sites.
The exposure sites at Stockholm and Gothenburg, Sweden, are in the same
region investigated by Goodwin. Sulfur dioxide deposition rates at these two
sites were reported be 42 and 44 mg/m2/day. This is lower than the 55 mg/
m2/day deposition rate calculated for the Goodwin work. Corrosion rates after
8 years on the Stockholm site were 7.62 g/m2/year (1.1 prdyear), equivalent
on a straight line basis to 1.59 g/m2 in 2.5 months. At the Gothenburg site, the
zinc corrosion rate, averaged over 8 years, was 12.5 g/m2 annually (1.7
pm/year), equivalent to 2.6 g/m2 in 2.5 months. Goodwins results over 2.5
months represent a much shorter exposure time than the years of Kuceras
experiments. Samples are expected to passivate with time, which would lower
the corrosion rate relative to the sulfur dioxide deposition rate. This could
explain why the correlations between corrosion rate and sulfur dioxide con-
centration are not exactly parallel for these two researches. However, Kucera
reported that the corrosion rate of zinc with time at these two sites was
essentially linear over the 8-year period, with only a slightly higher rate noted
during the first 2 months. This factor is believed to be the source of the
difference between the two researches. The pH value of rainfall at the two
sites was not reported by Kucera. The rainfall in Stockholm would be ex-
pected to be lower than the other values considered in the present work. A
much lower pH value, close to 4.5, would be predicted for the Gothenburg
site, which is in a heavy industrial zone. The Stockholm site on the other
hand, is on top of a water tower near the citys center.
One 3-year atmospheric exposure program (Cramer et al. 1987) in-
cluded 3-month data (Fig. 2.18) that relate to Goodwins work. Environmen-
tal data are available for the New York and Washington, DC, sites (Figs. 2.19
and 2.20). The latter site was located on the roof of a two-story building in the
middle of the urban area. Sulfur deposition rates from both wet and dry
deposition were measured. After 3 months, the corrosion rate of zinc was
determined to be 9.27 g/m2/year. Thus, a 2.5-month corrosion rate of 1.93
g/m2 was calculated. This agrees well with Goodwin. Sulfur wet deposition
rates of 0.89 pmol/cm2 were determined for the 3-month period, equivalent to
6.25 mg/m2/day. Sulfur deposited by dry deposition was determined to be
0.99 pmol/cm2 for the 3-month period, equivalent to 6.95 g/m2/day. Both
these numbers are very low in comparison with the Swedish data given above.
Cramers work shows that the amount of sulfur dioxide available from wet
deposition is almost equal to that available from dry deposition, a different
situation from Goodwin. Even if the sulfur dioxide levels are combined, a
much lower corrosion rate of zinc is predicted. In Washington, DC, the same
order of corrosion rate is achieved with only one-seventh as much It is
believed that the discrepancy lies in the higher levels of nitric and nitrous
acids (six times greater than observed at the Washington, DC, site, since
the presence of these acids favors the production of sulfur acids. Conversely,
at the Swedish sites, sulfur dioxide was believed to be at least 70% of the
cause of acidity in the rainfall.
Hudson and Stanners (1953b) conducted long-term outdoor exposure
tests in the United Kingdom on rolled zinc and correlated their results with the
amount of sulfur trioxide deposited on lead peroxide (Table 2.7A). Their
25
20
\
a
6 15
m

e
10
C
.-
5
0
Chapter 2
Key
1 - North Carolina
2 - New York
3 - New Jersey
4 - Washington. DC
5 - Ohio
1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5
l Monlh Monlh 1 Year Year
Fig. 2.1 8 Zinc rates at five sites for four exposure (Cramer et al.,
1987).
results have been converted for comparison with Goodwins work. The two
least polluted sites, Godalming and Teddington, had equivalent sulfur dioxide
deposition rates of and mg/m2/day, respectively. These corre-
sponded to annual zinc corrosion rates of pdyear ) and g/m2
pdyear). This range of numbers is thought to be representative the
behavior of zinc in mildly polluted environments where acid rain is present
and should fit into the shaded area of Fig. Corrosion data for Madrid
are shown in Figure
155
2.19 Monthly average SOz concentration for exposure in two cities
shown in Fig. 2.18 (Cramer et al., 1987).
Chlorides
Chlorides act as cathodic depolarizing agents but have far less effect on the
corrosion of zinc than on the corrosion of steel except in flowing seawater and
brackish water. Removal of any salt deposit by washing is, however, benefi-
cial and well worthwhile. Chlorides also have far less effect than sulfur
compounds in determining the rate of corrosion of zinc. Usually, the salt
content of the atmosphere decreases rapidly inland away from the coast (Fig.
Corrosion also decreases (Fig. with distance from the coast, but
the change is more gradual and erratic because chloride is not the primary
factor affecting zinc corrosion. Practical experience suggests that chloride is
most harmful when combined with acidity due to sulfur gases, as occurs on
the coast in highly industrial areas. The effect of sea spray is very different in
temperate and tropical areas (in addition to any temperature or humidity
effects). Temperature seawater and spray contain magnesium salts, which
help to inhibit the corrosion of zinc (see Chapter the same amount of
magnesium salts appears to be less effective in tropical conditions.
Results of tests in Nigeria indicated that an approximate linear relation-
ship exists between corrosion rate and salinity. Appreciable corrosion, how-
ever, occurred only at sites very close to the sea. The data for exposures at
distances from 45 m to 50 km from the sea are given in Table (see Sanyel,
156
Chapter 2
. l o-
.16-
"
E
?
?
.12-
U
. l o-
.08-
a
2 .06-
U
X
.04 -
.02-
Key
- DC
"""_
I
O !
82
Jan
83
Jan
a4 8 5
Jr
86
2.20 Monthly hydrogen ion load for exposure periods in two cities shown in
Fig. 2.18 (Cramer et al., 1987).
et al, entry for marine tropical countries, India). The sodium chloride depos-
ited ranged from 300 to 7.5 mg/m2/day according to a standard test method.
The corrosion rates for zinc at the same extremes were 39 and 0.8 Kdyear in
a l-year-test.
Ambler (1960) found that humidity and temperature in the tropics
magnify effects (Table 2.6) but reported no apparent direct correlation be-
tween corrosion and the presence of these three chemical compounds. Air-
borne chlorides responsible for increased corrosion in tropical and desert areas
were reported by Compton (1947) and Machu (1954). Compton noted that
corrosion caused by condensed moisture in some tropical islands is aggra-
vated by chlorides from coral dust. Machu noted a similar effect on zinc in
Egypt about 225 km from the sea. The desert sand and dust at this location
I
contained appreciable amounts of water-soluble salts, mainly chlorides and
sulfates of sodium, calcium, and magnesium. These salts increased the
corrosion of zinc caused by water of condensation.
Other Pollutants
Deposits of soot or dust can be harmful, both because of their specific nature
and because they may increase the risk of condensation onto the surface and
hold more water in position. This occurs mainly on upward-facing surfaces.
Soot (carbon) absorbs large quantities of sulfur, which are released by rain-
water.
Because of overmanuring of agricultural land, the ammonia content of
the air has increased noticeably: for example, by a factor of in rainwater
in the United Kingdom between 1968 and 1978 (Martin and Barber, 1978).
Ammonia in atmospheric quantities does not accelerate zinc corrosion, and
petrochemical plants where ammonium salts are present show no accelerated
attack on galvanized steel. However, ammonia will react with atmospheric
sulfur oxides, yielding ammonium sulfate, which accelerates paint film corro-
sion and also zinc corrosion.
When ammonium reacts with NO; compounds in the atmosphere,
ammonium nitrite and nitrate are formed. Both compounds accelerate zinc
corrosion but much less than and
However, when rain brings ammonium nitrate, nitrite, sulfate, and
sulfite into the soil, soil acidification takes place. An acidified soil will be
much more corrosive to coated steel parts. Nitrogen oxides also increase the
corrosion of zinc, but their unit influence is thought to be less than that of
sulfur oxides. Reactions 5 and 6 of Table 2.17 indicate that when and
HN03 deposit, the formation of sulfates is favored on wet zinc
surfaces. Conversely, nitrogen dioxide (NO,) has a solubility of only 1
lo", mol/L atm at 25C (Spence and Haynie, 1976), and therefore we expect
very little NO, to deposit on zinc during periods of surface wetness. Nitric
acid (HNO,) and formaldehyde (HCON) are soluble, like sulfur dioxide. As
suggested by these equations, a minimum amount of Zn(N03), would be
expected within the zinc corrosion film, as HN03 is emitted back into the
atmosphere as a gas. The calculated equilibrium constants for these reactions
strongly suggest that in polluted environments, will play the major role in
the corrosion of galvanized steel structures, even when HN03 is present.
However, nitric and nitrous acids may, in addition to their direct effect, have
a very significant catalytic influence on the production of sulfuric acid from
SO,. A significant interaction between and NO, in laboratory tests was
reported verbally by L. G. Johansson at the Electrochemical Conference at
40
30
20
10
0
I I
0 0.5 1.0 1.5
Contamination, mg S03/dm2/day
0.35
12
0.68
0.60 1s
0.66
13
0.7 I
0r3
0.86
0r7
2
37
16
14 1.03
26
7
0
1.22
19
18
I
9
40
2s
2.18
2.0
1.5
1.0
0.5
0
Las Vegas in However, few locations have adequate data on nitrogen
oxides present, and there is little research on the conjoint influence of sulfur
and nitrogen compounds.
Other gaseous pollutants have only a secondary effect, except insofar as
they may increase the acidity of moisture on the zinc surfaces. In Germany
(Bottcher, the pH of rainwater in Hamburg fell in the (pH in
pH 4 in but the trend is now reversing as a result of improved
environmental control. Johnson et al. found that isobutane enhanced
the corrosivity of SOz, although it had no effect on its own. Alkaline pol-
lutants often are beneficial, as zinc is particularly resistant to mild alkaline
conditions.
C. Location
Geographical Area
It is usual to differentiate between corrosion in:
rural atmospheres
marine (coastal) atmospheres
urban atmospheres
industrial atmospheres
Several reviewers have given typical corrosion rates for these atmospheres
(Table Vazquez and Damborenea in particular point out the
subjective nature of these descriptions and the different associated numerical
values of corrosion rates; such materials are used in this book because often
that is the only description initially available for a location. In general terms,
Scandinavia and the Iberian peninsula have purer air than central and north-
western Europe; North America aligns more with the less polluted parts of
Europe. This distribution is reflected in the use of the subjective terms; thus
corrosion in an industrial atmosphere in Finland may be little worse than in
some rural atmospheres in England.
In the counties of the former Soviet Union, corrosion rates tend to be
low, but Mikhailovskii et al. differentiate between the degree of wet-
ness at different sites in each of the four categories, a point that is less
significant in Western Europe except Spain.
The air in industrial areas and cities is contaminated with various sulfur
compounds which, together with the moisture in the air, convert the normally
~~ ~ ~~~~
2.21 (a) Higher sulfur, hence higher corrosion, in central districts of Madrid
(Morcillo and Feliu, 1983). Corrosion rates in Madrid related to contamina-
tion.
I I I
I 2 3 4
I
oistance from high water mark (kml
Fig. 2.22
Johnson, undated).
Airborne salinity versus distance from the British coast (Marsh and
U
X
I
I
0 10 20 30 40
Distance from Coast Krn
Fig. 2.23
and Ferguson, 1989).
Zinc corrosion rate as a function of distance from British coasts (Bawden
impervious and corrosion-resistant zinc carbonate and zinc oxide layer into
zinc sulfate and zinc sulfite. These water-soluble zinc compounds have poor
adhesion to the zinc surface. They are, therefore, washed away by rain
relatively easily. The underlying surface is then exposed to attack by the
oxygen in the air, and the corrosion cycle is repeated. Corrosion in sulfurous
air is, therefore, greater than in clean air-for example, out in the country.
Rural areas do not always have clean air. Neighboring industrial regions
or large towns have their polluted air spread over large areas by the winds. In
the Netherlands, it was found that during stormy periods, salt from the North
Sea was deposited on the isolators of transmission towers 40 km from the
coastline. Similar effects have been observed elsewhere in areas far from the
seaside. For example, heavily polluted areas around Antwerp, Belgium, and
the Ruhrgebeit (Germany) also influence corrosion in rural eastern Holland
and in the provinces of Limburg and Noord Brabant.
In a marine atmosphere, zinc corrosion is affected by the salt present in
the air, but the effects are not usually very significant, as just noted. Effects of
contact with other materials are discussed in later sections.
Microclimates
Within the broad types of outdoor atmosphere, it is necessary to consider
microclimates, both natural (e.g., due to prevailing wind effects, proximity of
vegetation, shelter from rain washing) and man-made (e.g., chimney ex-
hausts) and also those due to the shape of objects. The natural factors are
illustrated by Van Eijnsbergen (1982b and 1986b), who shows that the fastest
corrosion is up to four times as great as the slowest at the same general
location (Table 2.14). After pointing out the amounts of pollutants and their
reactions with galvanized steel, he noted the differences in local corrosion on
one and the same galvanized object caused by discrepancies of wind direction
and speed, relative humidity, and duration of wetness, although systematic
variations in corrosion rate do not always occur especially in open struc-
tures. Thus, according to a Dutch survey in 1990-1991 by Van Eijnsbergen
involving 2000 thickness measurements on weathered transmission towers,
there was no clear difference in residual coating between horizontal top and
horizontal bottom or between vertical outside and vertical inside surfaces of
angle bars.
An orientation effect is shown in relation to coating thicknesses of
hollow profiles supporting Dutch red polyester mail boxes. These galvanized
hollow supports had been exposed for 20 years in. an aggressive climate, with
values ranging from 100 to 350 111. 8S02/m3. The higher coating thickness
on the lower areas of these supports is explained by the shielding function of
nearby bushes.
162 Chapter 2
On a galvanized TV tower in the northeastern Netherlands, corrosion
rates on the northeast sides of the angle bars and gusset plates were approx-
imately half those on the southwest side, which was exposed to the prevailing
southwest and west winds, carrying pollutants from large industrial areas near
Rotterdam and IJmuiden.
Another striking example of specific corrosion losses is offered by a
large galvanized fence around the Uithof skating stadium in the Hague. From
more than IO00 thickness measurements, it was found that the zinc coatings
on the profiles of this fence facing east are pm thicker than the zinc
coatings facing southwest. Measurements carried out in have
confirmed these differences in corrosion. An acceleration in corrosion oc-
curred between and because of the surrounding bushes, which had
grown considerably since prolonging the periods of wetness in this
area, in contrast the beneficial effect in the case of the mailbox The
pH values of the rainwater remained low, with an average of and varying
from to during the period For that reason, the use of
durability (protection) values expressed in micrometers loss per year must
be used with extreme care. deviation of about from the mean must
always be considered likely, purely for microclimate effects.
It has already been mentioned that wire samples corrode more than flat
sheet (Table and several research workers have explained this result in
terms of greater SO2 deposition on wire. Extensive practical results are
available from the ISOCORRAG tests, where one interesting calculation from
the data is the high consistency of the relationship between wire helix samples
and flat samples (Table Both from time of wetness and theoretical
calculations regarding access to and deposition of on a wire compared
with a flat surface, higher corrosion rates would be expected on wire includ-
ing helices. Mears first showed this in For 41 (out of ISOCORRAG
sites for which one-year data are published, the corrosion rate for helix
samples compared with flat samples was greater in the one-year tests (Table
The high degree of relative consistency for the ratio of the mass losses
on steel and on zinc confirms the reliability of the ISOCORRAG test results in
general; the overall corrosion rates tend to be lower than those conducted in
similar sites or years ago. Further details some of these test sites in
Europe, which have also been used for other tests, were given by Barton and
Stanners
Downward-facing surfaces tend to corrode less than upward-facing
surfaces (see data of Evans, Tables and but not always (e.g., Morfa
Bank: Table severe marine). Similarly Legault and Pearson found
the corrosion of upward-facing zinc surfaces to be up to twice as much as on
Table 2.1 8 Experimental Regression Coefficients with Estimated Standard
Deviations for Small Zinc and Galvanized Steel Specimens Obtained from Six
Exposure Sites
Study"
Time of wetness
coefficient coefficient Number of
(pdyear)
(@year) (pg/m3)
data sets
CAMP (Haynie and 1.15 0.60 0.081 0.005 37
ISP (Cavender et al, 1971) 1.05 0.96 0.073 f 0.007 173
Guttman, 1968 1.79 0.024 >400
Guttman and Sereda, 2.47 f 0.86 0.027 0.008 136
Chamber study (Haynie
1.53 f 0.39 0.018 f 0.002 96
St. Louis (Mansfield,
2.36 0.13 0.022 153
Upham, 1970)
1968
et al, 1976)
1980)
.References are to literature cited in the source.
Source: Haynie (1980).
downward-facing surfaces (i.e., sheltered from rain and deposits) on panels
exposed at an angle for both an industrial (East Chicago) and a marine (Kure
Beach, NC, 800 ft lot) atmosphere, presumably because deposition and
retention of pollutants were greater. Their general equation is:
W = KP
where W is the weight loss in grams per square meter of exposed surface, t is
the exposure time expressed in years, and K and N are empirically determined
constants. The empirical values for these sites show a gradual reduction of
corrosion rate with time on all surfaces (i.e., N < 1) except the upward-facing
industrial exposure. These results support the generally accepted view that
extrapolation from l - to 2-year results is sound practice for a panel exposed at
an angle and with both upward- and downward-facing surfaces tested. The
prediction will include a factor of safety in many environments.
Height from the ground affects corrosion rate. In countries with low-
lying mist in the mornings, corrosion can be severe near the ground. This is
true even in rural areas if the mist is acidic as a result of sulfur gases carried by
the prevailing wind from industrial areas. Near Huddersfield, England, for
example, the wind comes from the industrial areas of Manchester. Similar
effects can arise with structures that are often in low cloud.
164 Chapter 2
The corrosion of the zinc layer is thus influenced by many factors. No
general formula for the corrosion rate is, therefore, possible. Good general
guidance based on experience can, however, be given, and this is done for
zinc coatings in Tables 2.1-2.11.
The color of the corrosion products can vary slightly, depending both on
the atmosphere in which they are formed and on the alloying elements in the
zinc. Marine atmospheres give slightly whiter corrosion products than rural
and urban atmospheres. The corrosion products are usually darkest in urban
atmospheres.
D. Initial Atmospheric Conditions
According to Schikorr (1965). the month in which corrosion tests are started
has comparatively little effect on the yearly corrosion rates. He arrived at this
conclusion by studying the results of tests at four different locations. How-
ever, in the studies of atmospheric corrosion of zinc in India by Sanyal et al.
(1956, 1959, 1961, 1962), the month in which the tests were started seemed
to have a definite effect on the yearly corrosion rate at some but not all
locations. No effect was noted at Kanpur, where total corrosion was slight. At
Bombay, however, the yearly rate for specimens exposed at bimonthly inter-
vals ranged from 1.5 to 8 prdyear. It is known that corrosion is more severe
in winter than in summer. Differences in SO2 and NO, contents in the various
seasons and changes of humidity and duration of wetness exert noticeable
influences on corrosion and corrosion speed (Bottcher, 1982). It has also been
reported (Hudson and Stanners, 1953a, b) that in the tropics, corrosion rates
are higher during the wet season than during the dry months. Furthermore, at
any one location, corrosion may vary appreciably from year to year. These
effects are probably all related to local meteorological or pollution conditions.
Initial corrosion rates are modified by the subsequent formation of
corrosion products, particularly in atmospheric exposure, where there is the
beneficial effect of periodic drying: periodic drying may, however, have a
detrimental effect when dissolved salts become more concentrated in areas
that remain moist after most of the water has evaporated, especially when
evaporation results in increased acidity of the residual solution.
Anderson and Reinhard (1939) observed definite differences in the
corrosion rates of sheet zinc exposed during different periods of the day.
Specimens were exposed for 8 hours per day and were stored in dry, unheated
moms for the other 16 hours. The test was planned so that a different group
specimens was exposed during each 8-hour period: 08:15 to 16:15, 16:15 to
W1 5 , and 00:15 to 08:15. The test continued for 6 years, and thus each set
was exposed to the atmosphere of Palmerton, Pennsylvania, for a cumulative
total of only 2 years. Another group was continuously exposed for the full 6
years. Two grades of zinc were tested. Weight losses after chemical cleaning
are given in Table 2.15.
The specimens exposed from 00:15 to 08:15 corroded more than the
others. The authors attributed the effect to condensation of dew, probably
contaminated with industrial pollutants, on the surfaces in the early morning
hours. It is interesting to note that the cumulative average corrosion rate of the
three 8-hour periods was exactly the same as that obtained for the specimens
exposed continuously.
Days exposed
Note: Tests scaned on dillerenl dates to provide dilfer;ng rainlall conditions during lirst 5 days
Fig. 2.24
outdoors at Middletown, Ohio (Ellis, 1949).
Effect of initial rate on the final steady corrosion rate of zinc exposed
Chapter 2
Ellis (1949) used a multiple correlation analysis of the weight loss data
on a larger number of zinc specimens and the prevailing weather conditions to
show that weight losses on 8-day exposures in tests at Middletown, Ohio,
could be correlated with the number of hours when it was raining or when
relative humidity was nearly 100% during the first 5 days. Upon continued
exposure, the panels with high initial corrosion rates showed higher final
corrosion rates than those with lower initial rates (Fig. 2.24). Corrosion
proceeded at a linear rate after the first 28 days in both cases. The corrosion
rates calculated form the slope of the weight loss-time line over the last 168
days of the test show that the specimens having high initial corrosion attained
a steady value of 1.9 pdyear , while specimens with a low initial corrosion
continued to corrode at a rate only 1. 2~dyear .
similar effect was noted by Vernon (1927) in a basement room
atmosphere in South Kensington, London, with a window opening onto a
n m w area adjoining the street. The relative humidity was low in winter and
higher in the summer, but still below about 70%. The slope of the time-
0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3
Time (days. from respective origins)
2.25 Relation between weight increments and time for indoor tests started at
four different dates in 1925: A, January 29; B, April 24; C, May 12; D, August 19
(Vernon, 1927).
corrosion plots varied, depending on the time of year when separate tests were
started (Fig. 2.25). Weight gains were determined as the measure of corro-
sion. Initial corrosion rates quickly changed to lower, constant rates. One test
was continued for about three and a half years, and corrosion progressed at the
same linear rate throughout the test. Gilbert (1953) in tests at Euston, London,
noted similar effects but of no practical consequence; there was, for example,
a 20% variation in corrosion rate according to the position on the test frame.
The investigations mentioning the effect of initial corrosion rate in-
cluded a range of atmospheric conditions; in general, the effects of the initial
conditions and corrosion rate on long-term corrosion rates are small compared
with other factors and usually can be ignored in practice.
However, it is desirable (especially for paint systems and duplex sys-
tems) to carry out corrosion tests in the atmosphere on panels on other speci-
mens, which are exposed in spring and another series in autumn.
E. Analysis of the Components of Atmospheric Corrosion
of Zinc
The key factors determining that gradual loss of zinc in atmospheric exposure
has been particularly addressed by Haynie through the 1980s in conjunction
with others (Haynie et al., 1990; Spence et al., 1990; Spence and Haynie,
1990) for the U.S. Environmental Protection Agency, working in close
conjunction with Carter, Cramer, Flinn, and others at the U.S. Bureau of
Mines (BOM) (Flinn et al., 1985, 1986; Cramer et al., 1987), and some of the
work has been mentioned in the preceding pages.
Haynie in 1988 derived a general relationship:
C = B t +
A[1 - exp( - BC/A)]
B
where
C = corrosion (pm)
B = solution rate of the film (pdyear)
A = diffusivity of corrosive species through the film (pm2/year)
After long exposures, the term -BC/A is large: therefore, exp( -BC/A)
approaches zero and Eq. 2.1 becomes a linear function.
t = time (years)
A
C = - + B t
B
Chapter 2
The net zinc corroded (C) after steady state is the sum of the zinc in the film
(AIB) and the zinc dissolved from the film (Et).
In polluted acidic environments the corrosion of zinc would be expected
to linear early in the exposure. The AIE term of Eq. would be the
intercept, the E term representing the slope with a constant corrosion rate. The
intercept would proportional to the steady thickness of the film when the
rate of film dissolution is equal to the rate of its formation. Diffusivity
through the film can be affected by impurities in the film. The solution rate
( E) is expected to be a function of the rate of delivery of acidic pollutants to
the film The B term may be separated into environmental terms involving wet
and dry deposition. The wet and dry delivery mechanisms provide means of
describing the factors that control the rate of corrosion.
Spence and Haynie combine the dry and wet deposition terms
derived for the corrosion of zinc surfaces to propose the following linear
damage function for predicting the long-term corrosion (pm) of galvanized
steel surfaces:
c = F + cd + CRA + CRC
For this function, F represents the zinc remaining in the corrosion film and is
expressed as AIB for steady-state corrosion of zinc, where the diffusivity
and solution ( E) rates of the corrosion film can best be evaluated from long-
term field corrosion data that are being acquired by the Bureau of Mines at
five field exposure studies for the U.S. National Acid Precipitation Assess-
ment Program (NAPAP) (Figs. The other terms are for the
deposition of SO2 during periods of surface wetness and dryness (c& the rain
acidity (CRA), and the dissolution in the rain of the basic zinc carbonate film
due to atmospheric COz (CRc). These terms can best evaluated from field
and laboratory run off studies of galvanized steel that the BOM and EPA have
conducted for NAPAP, as discussed in the original papers.
A term for the deposition of coarse mode particles has not been included
in the proposed damage function. Coarse particles that are likely to deposit on
zinc surfaces may increase the surface area as well as the time of surface
wetness. The incorporation of terms for the effects of particles, as well as for
other atmospheric gases, must await further work.
Practical tests at Washington, DC (Cramer et al., with associated
climatic data, have enabled these coefficients to be determined by Haynie et
al. for one site (Fig. close correlation is found between the sum
of the four components and the actual zinc dissolution. Since, however, the
authors comment on the difficulty in obtaining or estimating environmental
data, and the calculations of gaseous deposition appear to be unreliable, the
2.6 -
PREDICTED ZINC OlSSOLUTlON - 1
ACTUAL ZINC DISSOLUTK)N
BICARBONATE EFFECT
- HYDROGEN ION EFFECT
e 2
'
-
2 - 2
-
2
SO2 OEPOSlllON WEN WET EFFECT ---------. . 5
SO DEPOSITION WHEN DRY EFFECT -6
1.8
1.6
1. 4
1.2
-
-
2
-
-
1 -
0.8 -
2 6 10 14 18 22 26 30 34
EXPOSURE TIME - Months
Fig. 2.26 Contributions to zinc corrosion at Washington, DC (Haynie et al., 1990).
parameter in their equation does not relate well to practice, particularly at high
sulfur dioxide levels. Figure 2.27 shows reasonably good correlation at the
three mild sites examined (0.4-1 pdyear corrosion rates). It would, how-
ever, be wrong to try to extrapolate this work to more corrosive locations; its
main value is as a base of further research.
Extensive research has also been done at the Non Ferrous Metals
Research Institute in Prague (see later). In eastern Siberia and elsewhere in the
far eastern regions of the Soviet Union, Mikhailovskii et al. (1986) applied a
Chapter 2
2.4
2.2
2
1.8
1.6
1 . 4
l . 2
1
0.8
0. 6
0. 4
0. 2
0
- SITE PREDICTED ACTUAL
- - - - - - - -
STU .......-...._...
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 6 10 14 18 22 26 30 3 4
- Months
mathematical equation to corrosion rates of various metals including zinc and
claim 20% error only (Panchenko et al., 1985); the paper also has valuable
climatic and pollution data for 29 sites in Siberia and the Far East.
Much work has still to done to develop formulas of practical signifi-
cance. At present, users are advised to expose samples (IS0 9226 gives
guidance) for at least months. The results extrapolated in a straight line are
likely to be more accurate than calculations based on meteorological and
pollution measurements, even without allowing for seasonal effects.
The fundamentals of atmospheric corrosion were restated by Goodwin
(1991), who said that zinc has good corrosion resistance in neutral and mildly
alkaline environments, but the protective film can be destroyed by acids.
Evaporation of moisture is believed to allow very strong acids to react with
the normally passive surface film for brief periods of time.
F. Standardization of Atmospheric Corrosion Rates
Which comes first, the chicken or the egg? The same question must be asked
about attempts to define atmospheres (in terms of time of wetness and
deposition rates of SO, and chloride) so that corrosion rates of zinc can be
assessed, whereas others have used zinc corrosion rates to assess the cor-
rosivity of the atmosphere. In this connection, the 10-year tests at seven sites
(Fig. 2.28) by Holler, Knotkova, et al. (1986-1987) mostly show straight-line
relationships with time; this analysis of corrosion rate versus sulfur dioxide
plus chloride contents of the atmosphere showed a linear rate for freely
exposed zinc, but in semienclosed conditions with high sulfur dioxide plus
chloride, the corrosion was slightly less than if a straight-line relationship
applied.
Nevertheless, IS0 9223 and 9224 (following on earlier work by Knot-
kova et al., 1975; Barton et al., 1979, Barton, Knotknova, et al., 1980,
Holler, Knotkova, et al., 1986) have tables based on many cooperative tests
and should be given careful consideration by specifiers if the necessary
atmospheric data are available. The danger of relying on such tables is that
other significant factors that vary considerably from place to place (such as
g/mz
VOOF
t n T
3UO -
ZOO -
years
Fig. 2.28 Open exposure of zinc at seven East European sites: A, Zvenigorod; B,
Moscow; C, Prague-Letnany; E, Usti-na-lak, H, Batumi; K, Orgovan; M, Sarafovo
(Holler, Knotkova et al., 1986).
172 Chapter 2
nitrogen oxides) are ignored, and pitting and nonlinear corrosion rates of
some metals other than zinc cause difficulties if the tables are used for long-
term predictions; I S0 9223 and 9224 also cover steel, weathering steels,
copper, and aluminum. Five levels of corrosion rate are covered, and those
identified for carbon steel and zinc are given in Table 2.12 as applicable to
given ranges of time of wetness, sulfur dioxide deposition, and chloride
deposition.
It is desirable have a map of Europe showing general background
pollution levels, on which regional and local conditions can be superimposed.
However, such a consideration requires more results from a large number of
sites; thus, in the United Kingdom, detailed surveys showed that the back-
ground corrosion rate in southeast England (even in rural areas) is much
higher than the background rate in the southwest, even through the Cornwall
region is subject to high airborne chloride.
From a practical viewpoint, recent tests show that zinc has a corrosion
rate of less than pm/year at about two-thirds of the landmass. Sites with
chloride (as NaCI) above about 50 mg/m2/day merit closer evaluation, as do
sites where content is above 50 pg/m3, but other factors and interactions
make on-site tests very desirable in all major projects. Other atmospheric
constituents (e.g., NO2) remain to be evaluated adequately, especially in their
synergistic effects on chemical change, hence on corrosion reactions.
Usually, the quantitative environmental figures available are too impre-
cise to be used in any formulation of corrosivity (such as in I S0 9223 and
9224) that might be interpreted by professional analysts as comparable in
accuracy to other engineering determinations, such as strength.
It is hoped that more data will be available within a few years from
ISOCORRAG, from tests being conducted by the European Community, and
from the United States (in relation to government environmental classifica-
tions) to give much greater insight. For the present, sulfur dioxide figures
form the best environmental guideline, but experience has shown that expo-
sure trials are much to be preferred. As little as 6 months usually can give
guidance to within a factor of 2, which is sufficient for commercial decisions
on types of protection or maintenance intervals. A years trial is preferable,
however.
CORROSION DATA BY LOCATION
A. General
For convenience in investigation, most long-term exposure trials are on zinc
sheets. It was shown by several authors (Hudson, 1929; Bauer and Schikorr,
1934; Schikorr and Schikorr, 1943; Schikorr, 1965) and by Anderson and
Reinhard (1948) that the degree of purity of the zinc is generally an unimpor-
tant factor in atmospheric corrosion. All tests certainly show that the type of
atmosphere has much greater effect on corrosion rate than the type of zinc.
The results of Anderson (1959) in Table 2.7A show however that alloys may
corrode to a significantly different degree. Surface finish of the zinc (e.g.,
pickled, emeried, or polished) has no significant effect. Corrosion is generally
uniform, although some earlier investigators (e.g., Friend, 1929) refer to
pitting, possibly due to deposits causing localized attack. Even if there is
some localized attack on zinc coatings, the cathodic protection effect will
ensure that the long-term performance is mainly related to the mass of zinc
present. Similar corrosion rates are to be expected on zinc-coated steel as on
zinc. Major collaborative programs of long-term tests have been run by
ASTM (STP 435, 1968; STP 646, 1976) and by ISOCORRAG, and both
these bodies have also tested steel, with the result that steel/zinc ratios are
available (Tables 2.7B-2. IO). Mohler (1974) has plotted the zinc losses
versus the steel losses to show how in the ASTM tests, steellzinc ratios are
much higher in marine atmospheres. He has also developed a zinckadmium
ratio (from various tests on electro-plated coatings) to show the higher ratios
in marine conditions (Fig. 2.29).
Cadmium - per year (log scale)
0.5 1 2 4 8
.64
I
.01 I I I I
.01 .02 .04 .08 .l6 .32 .64
Cadmium -mils per year (log scale)
2.29 Relationship of corrosion rates of zinc and cadmium coatings
atmospheres (Mohler, 1974).
in different
174 Chapter 2
In general, tests lasting a year or more (IS0 gives test procedures)
can be extrapolated to give adequate practical information on long-term
performance. Test periods shorter than I year will often give corrosion rates
higher than are found in the long term for mild atmospheres, whereas for more
aggressive acidic atmospheres the rate is nearer to linear from the This
was shown, for example, by Cramer et al. when for four sites the one-
year results were within the same scatter band but the order of performance
after years was exactly the opposite to that after month (Fig. Short-
term corrosion data on zinc, galvanized steel, Galvalume, carbon steel,
Corten A, copper, and aluminum are also covered by Hinn et al.
In earlier investigations it was common practice to expose zinc-coated
sheet or wire and to assess time to first rust and time to 100% rust visually. In
the United Kingdom, the practical observation was the time for of the
surface to be covered by rust. Corrosion rates calculated from such tests have
been included in Tables Such observations were very subjective and,
since they all concluded that the corrosion life was in general linear and
proportional to the amount of zinc present, it is more common today to
measure zinc losses quantitatively over a fixed period. Many of the research
workers were, however, keen to point out the advantages of more uniform
coatings. Daesen showed that in the ASTM tests the time to complete
rusting with some coatings could be more than twice (+ the time to first
rust (apparently because of uneven distribution of zinc), compared with only
with other coatings.
Some trials have shown differences between types of zinc coatings.
Thus, both Biestek and Hippensteel and Borgmann found that
zinc electroplated from cyanide baths corroded faster than zinc deposited
from sulfate baths when measured as time to first rust. Uniformity, porosity,
and possibly grain size may be responsible for these differences. Porosity
(which is an indication that a greater amount of zinc surface per unit of mass
is exposed) will lead to increased corrosion, as was shown by Hudson and
Stanners for unsealed sprayed zinc. However, such differences are
minor-usually under l oye a nd do not occur on sealed zinc spray, which
has much slower corrosion rates (AWS,
Shape, orientation, and mass of the zinc-coated steel have an effect,
mainly because of the effect of time of wetness but also because a differing
facility to receive deposition from the atmosphere (as already discussed in
relation to sulfur dioxide). The effect of diameter of wire was shown by
Richards in 5-year tests in U.K. sites of widely varying corrosivity
(factor of > where mm wire corroded more than mm wire.
The results were generally similar to those of Reinhart at U.S.
locations on wire, wire strand, barbed wire, and fencing. The specific weight
loss results are included in Tables Mears reported mm
wires corroded four times as fast as mm diameter wires. Cony for
wire strand and Cony and Lowe for overhead line fittings report
extensive tests in service for these products; ASTM tests relating to hardware
are separately documented in the report of Subcommittee XIV of Committee
(ASTM,
Another program to supplement the foregoing information on hardware
parts was undertaken by ASTM Subcommittee XVI in (ASTM,
Flat, round, and bent specimens of various steels were coated with
zinc by hot dipping and electroplating to a specified thickness of pm
and exposed at Newark, New Jersey, and at the and m sites at Kure
Beach, North Carolina, representing industrial and marine conditions, respec-
tively. The data can be summarized as follows:
Flat sheet specimens corroded somewhat more slowly than the bent sheet
specimens. Corrosion of round specimens was several years behind that
of the flat ones.
After years, the bent Newark panels were almost completely rusted.
Such bent specimens still showed only trace rusting m from the sea at
Kure Beach, while those at the m site were rusted.
The effect of bending is to cause some cracking of the coating, exposing more
zinc surfaces to the atmosphere. Development of coatings that reduce this
effect, notably Galfan, has been an important aspect of subsequent work.
Urban atmospheres have been widely investigated, and the data also are
the most widely required in practice. Average zinc loss is given in Table
for tests lasting at least one year. Such data may be extrapolated with
reasonable confidence (see Table although in a few atmospheres the
average corrosion rate appears to increase with time. In addition, the effect is
small and often is attributable to changes in the atmosphere itself: indeed,
with some marine and rural atmospheres, the rate usually continues to de-
crease with time. As seen in Table rates do change from year to year, and
care should be taken to allow for any long-term trends.
Industrial atmospheres are mainly a special case of urban atmospheres,
usually with higher sulfur dioxide concentrations. Available data are given in
Table 2.5, and other research programs with data covering some years include
work by Legrand Given present environmental regulations, however,
such heavily polluted industrial atmospheres are unlikely to be widely experi-
enced in the future, and zinc corrosion rates are likely to be reduced in any
particular location. Nitrogen compounds (NO,), though, are a form of pol-
lution that is increasing and can have both direct and indirect effects on the
corrosion of zinc, including a probable synergistic effect with sulfur pollu-
176 Chapter 2
tion. Marine atmospheres in temperate latitudes have much less adverse effect
on zinc than on steel (see ratios in Table 2.10). The precise effect depends
greatly on the distance from the sea (Table 2.6), the direction of the prevailing
wind, and the incidence of sea spray. Performance is relatively good in
temperate latitudes compared with tropical latitudes because the magnesium
salt is a more effective inhibitor in temperate seawater. From service experi-
ence, the combined effect of chlorides and sulfates appears to be much worse
than the sum of the two effects taken separately, although quantitative data on
their synergistic effect appear to be lacking. It is not known whether there is
synergy between chloride and nitrogen oxides.
Truly rural atmospheres (Table 2.3) give very low corrosion rates on
zinc, but at some country locations the air can be polluted by acid compounds
brought by the prevailing wind from polluted areas more than 100 km away.
The adverse effect is particularly felt if there are early morning mists, which
can deposit highly acidic moisture, especially on structures near ground level.
Caution is advised in considering zinc-coated steel if pH measurements of
such deposits are less than 5.
Distinction must also be made between freely exposed outdoor exposure
and sheltered exposure. Work in Soviet Union, discussed earlier (Mik-
hailovksii et al., 1986), includes much data of this type, and Okorafor (1986)
has done similar work in Nigeria. Generally, corrosion rates in sheltered
exposures are less than in open exposures, but corrosion rates can be higher if
acid water is present for long periods on uncoated surfaces.
Most descriptions of atmospheres are subjective, and care is needed in
comparisons when no quantitative data are available. At one extreme, Swed-
ish Standard SMS 2950 gives guidelines for corrosion rates in different types
of Swedish atmospheres that are at even below the broad ranges of
anticipated corrosion rates for the world as a whole. Conversely, for acidic
industrial pollution, areas in northwestern Europe are among the worst among
countries for which reliable data are available. Even as a rough guideline,
it would appear that about 50% of the land surface of the British Isles will give
a corrosion rate for galvanized steel of less than 5 pmlyear, and many of the
rural areas in southwest England, elsewhere where there is no drifting
sulfurous pollution, historically have had corrosion rates under 2 pmlyear.
Conversely, industrial areas such as Lancashire, part of the West Midlands,
and East London, as well as some of the industrial coastal town, have had
corrosion rates as high as 10 or 12 pmlyear. The special case of the misty
valleys near Huddersfield, with acid pollution from Manchester, historically
gave corrosion rates as high as 20 pmlyear, but this condition is changing as
environmental laws more closely control sulfur-based emissions from indus-
trial and domestic sources.
A general guideline for assessing coatings for the United Kingdom is
given in British Standards Institution document BS 5493 (Figs. 2.30 and
2.31). Simpler leaflets with tabular guidance have been provided by industry
for exterior environments (Fig. 2.32), building interiors (Fig. 2.33), and some
more specialized environments (e.g., cavity walls).
The U.K. data contrast with data from other parts of Europe, such as
Finland, where areas locally regarded as industrial have corrosion rates
equivalent to rural areas in England. In the United States, urban areas often
have much lower corrosion rates (e.g., New York, 0.8 Fdyear; Washington,
DC, 1.9 pdyear , already detailed in this section) than apparently similar
urban areas in northwestern Europe. Figure 2.4(b) looks at sulfur emissions in
Europe, discussed in some detail by Van Eijnsbergen (1982).
Because corrosion occurs differently in different parts of the world,
qualitative descriptions of atmospheres urban, industrial, marine,
or rural are of significance only within a specific geographic region, and
multiplication factors need to be determined for any other region. Neverthe-
less, it can be helpful to show the range of zinc corrosion rates in this form
(Tables 2.3-2.6). SO2 content and time of wetness should be obtained
wherever possible (see Tables 2.7A and B).
Corrosion maps are very desirable. T. R. Shaw has initiated these for
the United Kingdom (Shaw, 1976; Ministry of Agriculture 1982, 1986), and
his test technique has been adopted elsewhere. A new edition was published
by the Ministry of Agriculture in 1992. Some results are available for Canada
and New Zealand. Spedding (1971) gives results for Australia related to
content. It is interesting to compare the U.K. map with the pH of acid rain
there (Fig. 2.4c), which clearly indicates that acid rain is only one of the
factors affecting zinc corrosion.
B. Eastern European Data
Data for the former Soviet Union and other parts of Eastern Europe are not
readily available in English. Fuller details follow, based on a selection made
by Dr. E. Proskurkin, who has primarily selected data from contributions in
two books (Rosenfeld, 1960; Berukshtis and Klark, 1971), plus other selected
reports. The Russian Institute of Physical Chemistry, which has been respon-
sible for much of the work of these investigators, has amplified (Table 2.19)
the brief classification of environments given in Table 2.1. Mikhailovskii et
al. (1986) have developed mathematical expressions for the corrosion of
metals. These expressions differ according to whether the atmosphere is
coastal or industrial. (Note: For the convenience of Western readers, the
symbols used here are different from those in the original paper.)
W = a(S02)b t (for industrial atmospheres)
W = [Wo + (Cl, Cl-)] t (for coastal and marine atmospheres)
Interior envlronments
Seawater Polluted Non-polluted Non-saline
Protection Systems Exterior exposed environments
Water
Splash Immersed Coastal Inland Coastal Inland Wet DV
spray
BIGahnnlzlng SB1 SB1
Hd dip gaIvamz6d
Paintad
SClZ SClZ
scp scin SM SCl Z m z
Dlanc dust painta so1
Organffi
inorganic
so sE2 SEZ sE3 sE3
~
1-
SF8 SF8 SF7 SF7 SF51SF2
H. JIOno pack chemical
realatant paints
Chlorinated rubber, vinyl copamr
K/7Luo pack chemlal reals(snl SKZ SK6 S K 5 S K1
Coal la epoxy. Polyurethane SK7 sK7 sK2 SKZ SK7ElCZ SKWSK2
pi nt s
EiG---- SKBVe ' - SU' . - - ~- Sb'.<-* "%
S U
SL4 SLZ SLZ
SK6%K47-'SK6/SK3?
UChmpack mar hro pack paints
Life to first malntenanw of protective coating The inlormation in this chaR IS based on Bnlish
Standarc 885493 1977 and repmduced mth Iheklnd
permission 01 the British Srandards Ins1t1u1m. 1 0 1
Fmlonvllle Road London N19 ND Complete mplesd
the Standard are available lrom BSI. 2 Park Sfreel.
London WIAZBS
How to use the chart:
More than twenty years
Ten to hwnty yean
Fnrelo ten years
1 Select &r for time required
lo first mainlenana,
2 Ckme system from anemathres
dfered (or ennmnmenl of
swat
Fig. 2.30
Fig. 2.31 for system references.
General guidance for selection of protective coatings for steel (AM&S Europe Ltd., 1980). See
0 5
8
7
A'
r
n
4
a
g.
a
d
v
W
Components of systems
d Nominal system 0 System Surface preparation Zinc coating
0 reference thickness pm
-
Zl ncCodngs SB1 See BS729 Galvanize 85
other than
metal spray& SB2
Grit blast &/or pickle Galvanize 140
SB3 Pickle Galvanize (silicon in steel) 21 0
SB4 See 6.5729 Galvanize and centrifuge 43
SBS See BS4921 Sherardize 15
SBS See BS4921 Sherardize 30
25 SB7 See BS1706 or BS3382 Zinc plate
85+30 min SB6 See BS729 Galvanize plus one coat paint
SB9 See BS729 Galvaiize plus two coats paint 85+60 min
-
Sprayed zlnc SClZ See BS2569 Sprayed zinc- unsealed 100
coatings s c2z See BS2569 Sprayed zinc - unsealed 150
SCJZ SeeBS2569 Sprayed zinc - unsealed 250
sc4z SeeBS2569 Sprayed zinc - unsealed 350
scsz See BS2569 Sprayedzinc-sealed (1) 100
SC6Z See 652569 Sprayed zinc- sealed ( 1) 150
scn See BS2569 Sprayedzinc-sealed (1) 250
scaz See 652569 Sprayed zinc-sealed (1) 400
SeeBS2569 Sprayedzinc- painted (1) 100+(30-100) s c9z
sci oz See BS2569 Sprayedzinc-painted (2) 100+(30-100)
-
I
Organlc zlnc- SD1 Sa 21 Zincdust paint 50
rlch paints
--
502 Sa 21 Zinc dust paint 75
SD3 Sa 2f Zinc dust paint 100
SDS Sa 2f Zincdust paint 150
System Surface preparation
reference
Nominal coating thickness prn
Primer Undercoat Finish coat
Nominal system
thickness pm
EP1-EP2 KU1-KU2 Kf l - KF2
lnorganlc SE1 Sa 21 75 75
SE2 Sa 24 100 100
zinc-rich
Dalnts
SE4 Sa 21 75 100 100 275
SE5 Sa 21 75 200 100 375
EP1-EP2 KU1-KU2 HF1-HF2
SE6 Sa 21 75 100 100 275
SE7 Sa2i 75 200 200 475
SEE Sa 21 75 25 100
S P w d
aluminium
coatings
System Surface preparation Coating composition Nominal system
reference thickness urn
SCl A See BS2569 Sprayed aluminium - unsealed 1 00
SC2A See 882569 Sprayed aluminium- unsealed 150
SCBA See BS2569 Sprayed aluminium - unsealed 250
SC5A See BS2569 Sprayed aluminium - sealed 100
SC6A See BS2569 Sprayed aluminium-sealed (1) 150
SC6AH SWBS2569 Sprayed aluminium-sealed ( I ) 175
SC7A See BS2569 Sprayed aluminium-sealed (1) 250
SCSA SeeBS2569 Soraved aluminium-aainted (11 1 00+(3O-1001
Fig. 2.31
SClOA See BS2569 SDraved aluminium - oainted 12) 100+(30-100)
( 1) . (2) Numberofpretreatmentcoats
Components of systems and product specification in Fig. 2.30 (AM&S Europe Ltd., 1980).
System Surface preparation
reference
Nominal coating thickness pm Nominal system
thickness prn
Primer Undercoat Finishcoat
FP1-FPS FU1-FU4 FF1-FF4
SFl st 2 35 35 35 105
-
Drying-oll
paints SF2 St 2 70 25-40 25-40 120-1 50
SF3 St 2 70 50-80 50-80 170-230
SF4 Sa 21 35 35 70
SF5 Sa 23 35 25-35 25-35 85-105
SF6 Sa 21 70 35-40 25-40 1351 50
S R Sa 21 70 70-80 25-40 165-1 90
SF8 Sa 21 70 70-80 50-80 190-230
KP1 KU1-KU2 FF4
Silicone alkyd SO1 Sa21 70 125 50 245
over two-pack so2 Sa21 70 220 50 345
chemicai-
resistant
paints
HP1-HP2 HU1-HU2 HF1-HF2
One-pack SH1 st 2 35 100 25-30 160
SH2 st 2 70 100 50-60 220
SH4 Sa 21 70 100 25-30 200
SH5 Sa 21 70 100 50-60 220
SH6 Sa 21 70 100 100 270
SH7 Sa 2f 100 100 100 300
chemicel-
resistant
paints SH3 Sa 21 35 60 50-60 150
-L
W
w
FP3-FP4 MU1 HF1-HF2
Drying-oil SJ1 st 2 35 1 00 35 170
primer with
one pack
undercoat
and finish
KP1 KU1-KU2 KF1-KF2
Two-pack SK1 Sa 21 35 100 35-45 170-180
SK2 sa21 70 100 70 240
chemical-
resistant
paints SK3 Sa21 70 1 00 100 270
SK4 Sa 21 70 150 100 320
-
KF3
SK5 Sa 21 150 150
SK6 Sa 21 250 250
SK7 Sa 21 350 350
SK8 Sa2i 450 450
KPl KU1-KUP HF1-HF2
Two-pack SL1 Sa 21 35 100 50 185
SL2 Sa 21 35 100 100 235
prlmer and
undercoat
with one- SW Sa 21 70 125 1 00 295
resistant
flnish SL5 Sa 21 140 200 too 440
SL6 Sa 21 50 150 1 OO(1) + 35(2) 335
packchemical- sL4 Sa 21 70 175 100 345
Fig. 2.31 (Continued) (l)N3Travfdcoat (2)HPl Finishmat
! !
l ;
0 ,
I
!
l
l
i
I
I
I
I
I
~i
8
8
L
U
&
a
L
.-
:
a
m
Red leadlwhite lead (BS 2523 A-C) 92-98 72-84 40
75-50 64-70 40
FPlA-C Blend of raw and processed
drying oils
phenolic or phenolic-modified
-. . - __ -
FlDryfng-oil
--
Calcium plumbate (BS 3698 A-B)
Zinc chromate 40 45 35
Metallic lead 50 45 30
35
40
-
weprlmm FP1D-E
.- . .
I .
FP2A Drying oil modified with Zinc phosphate 40 45 35
- - .
.- resin
FPZB
FP2C
-- - _ ~ - -
- - ~
FP3A-8 Alkyd ormodified alkyd As FP2A to FP2B above 40 45
FP4 One pack drying-oil epoxy ester Zinc phosphate or chromate 40 45 35
FP5A-8 One pack polyurethane oil Zinc phosphateorchromate 40 45 35
50 --
60 FP3C - Red lead
F / D ~ ~ & I I FWA Drying oil modified with Tfianium dioxide andlor colwred pigments - 45 35
Mi caceOus iron oxide 80 50 40
phenolic or phenolic-modified
-.resin
we
undefcoats
FWC Micaceous iron oxide (High build) 80 45 75
Aluminium 95 45 25
-
FUlD
FU2A-D Alkyd or modified alkyd AsFUlAloFUlDabove Asabove
FU3A Chlorinated rubber modified Titanium dioxide andlor coloured pigments - 40 35
Micaceous iron oxide 80 40 35
with alkyd in ratio 2:l
FU3B
FU4A-D One nack drvina-oil emxv ester As FUIA lo FUlD above Asabove
Fig. 2.31 (Continued)
In
a
m
8
e
a
v)
m
v)
I
m
B
e
In
L
.-
U
E
0
-
v)
m
v)
m
m
2
e
c
-
m
c
0
L
a
a
m
8
e
8
v)
m
v)
I
m
B
L
In
._
a
0
-
5
a
Ii
4
U U .
m
m
9 .-
n
._
z
m
0
6
0
.-
m
F
m
*
8
.- e
0
5
0
0
: I . .
RodU* Binder Main pigment Weight % Volume of Dry film
reference pigment solids % thickness
in total wn
pigment
Kmvo-pack KFlA Two pack epoxy Titanium dioxide 90 45 40
chemical-
resistant loo .-
KFlB Titanium dioxide (High build) 90 40
_ _ .. finlshes KFlC Coloured pigments and carbon black - 45 40
~_
-
- .
KFlD (High - build) 40 100
KFlE Micaceous iron oxide 80 45 40
KFlF
--.____.. - .
Micaceous iron oxide (High build) 80 45 100
-- .
.-
KFlG Aluminium 95 45 35
--___
KFPA-F Two pack polyurethane AsKFlAtoKFlF As abcve except high build
thickness75pm
.. . - -
KFSA
TWO packepoxy ormodified Silicate extender 100 60 60
epoxy coal tar
. - - . ~~
- -
KFBB Silicate extender (High build) 100 55 100
KF3D Micaceous iron oxide (High build) 100 55
-._-
KF3C Micaceous iron oxide 100 60 60
- __
100
.
. __-- 60-300
MlBltumen
and coal tar
Refer to British Standards BS1070. BS3416.BS4147 8 BS4164
. . - - - . . - - . .
products -
where
Q
189
As polluted Inland with high airborne Sari levels.
of 3 hm from the coast)
I 2 : ; : 1 p i s salt contaminnled zone may extend to a maximum Very hgh
w = corrosion rate (g/m2/year)
Wo = corrosion rate in a nominal pure atmosphere (g/m2/year)
C1- and SO2 = concentration of chloride ion and sulfur dioxide, respec-
Cl, = the acceleration of corrosion by chloride ions (g/m2.h di-
tively
vided by mg/m2.h)
The values of the constants a and b are as follows:
c1;
Q
Not cwered
in this
document
Metal
As polluted Inland with high airborne Sari levels.
p i s salt contaminnled zone may extend to a maximum
of 3 hm from the coast)
Aggressive Industrial envimnmeMs such ea meelwork
edjacent to acid plenls salt storwe depots electmplali
shops, chemical works'etc. ~ u r i ~ or immiked s t e e d
Sea wter splash zones.
Very hgh
Polluted
coastal
Variable
Steel 549.3 0.52 44.0 5.6
Zinc 50.6 0.75 2.8 0.7
Copper 45.2
0.70 2.6 0.5
Aluminum 0.6 0.50 0.11 0.15
Alloy MA2-1 45.6 0.48 6.8 0.1
EXTERIOR ENVIRONMENTS
Environment 1 Environment I
Category' (see note 8)
Description
Most rural and urban areaa with low sulphur dloxlde. acid.
lScmm awmtlv non-Industrial areas mav be miluted fmm
I Normal I alkali and 8811 pollutbn.
Inland
I I distant dufces akording to prevailing wind and lopography) I
Slgnlflcant
Polluted
Inland
Areas of high affborne sulphur dloxide. of other pollutbn
lrom industrial or domestic murces.
As normal inland wkh high alrbome salt levels.
01 3 km from the coast)
p i s salt conlaminaled zone may extend to a maximum 1 High F
Not cwered
in this
document
Aggressive Industrial envimnmeMs such ea meelwork
edjacent to acid plenls salt storwe depots electmplali
shops, chemical works'etc. ~ u r i ~ or immiked s t e e d
Sea wter splash zones.
Variable
'Categories A, 6 and C are interior envlronments covered by the "Steelwork Corrosion PmleCtiOn Guide - InleriOt Environments"
Fig. 2.32
environments (BSC/BCSA/ZDA/PRA, 1986).
Coating recommendations for atmospheric exposure in exterior U.K.
Chapter 2
Data for the corrosion sites are in Table which is taken from an earlier
paper by Panchenko et al. from the same institute and research group.
This material deals specifically with the Asian end of the group of countries
and apparently uses slightly different formulas for the calculated corrosion
losses. These calculated losses are not given here; in almost all cases, they
were within & 30% although in some benign areas predictions were too high
by as much as 70%.
A feature of these mainly benign atmospheres is that the ratio of steel to
zinc corrosion is less than found further west. The Soviet workers had test
sites for which temperature, relative humidity, days with dew and rain, and
particulate deposits were recorded over 4 years, while Table gives
information on chloride ions and sulphur dioxide (Strekalov and Berukshtis,
together with corrosion rates in both open and sheltered exposures.
. . _. .
Sl33dSV Y3HIO NO S310N
I I
I SlN3WNOtllAN3 UOIU31X3
192 Chapter 2
areas of the surface. In industrial Soviet atmospheres, zinc coatings corroded
at 4 pdyear and cadmium coatings at 10 pdyear.
One-year test results on zinc and cadmium are in Table 2.22. The
results of long-term atmospheric corrosion tests on zinc and cadmium coat-
ings (Fig. 2.35) are compared with those of accelerated corrosion tests (Fig.
2.36) on the same coatings in three cabinets: warm and humid,
gas, and sea mist (Strekalov and Berukshtis, 1965).
Strekalov and Berukshtis concluded that sea fog and warmth and
humidity cabinets satisfactorily reproduce the nature of corrosive destruc-
tion of anodic coatings on steel, as observed under natural conditions in
nonindustrial zones. Accelerated tests in gas cabinets were too harsh,
however, and did not reliably reflect the behavior of zinc and cadmium
coatings in a natural industrial atmosphere.
The zinc tested was 99.98% pure (Pb 0.007%, Fe 0.006%, Cu 0.003%,
Cd 0.001%), while the cadmium contained 0.02% Pb, 0.013% Cu, and
0.003% Zn. Five-year test results are shown in Fig. 2.37, together with the
ratio of corrosion rates in the open air to that in sheltered booths. It is not clear
whether the figures represent the situation after a specific exposure time or are
an integration of results. more detailed analysis of the results compares the
ratios for the various sites with the unpolluted agricultural site (Table 2.23).
Figure 2.34 gives the corrosion rates for both open and sheltered exposure and
for periods up to 4 years.
It can be seen that the rate of corrosion of steel in the open atmosphere is
about 1-4 times higher than that in sheltered booths. The corrosion rate of
zinc, cadmium, and copper is higher by not more than 1 times; aluminum
corrodes in the open atmosphere and in the booth at about the same rate in all
climatic regions.
Additional results from Soviet investigators (Berukshtis and Klark,
1971) produced the plots shown in Fig. 2.38 for average thickness loss of zinc
and cadmium and coatings of Zn and Cd, with and without chromating. These
tests at different dates give some indication of the changes in atmospheric
corrosivity over the years.
Rosenfeld (1960) used seven corrosion sites to test three grades of zinc
over 10 and 20 years. The results (Table 2.24) are a little erratic, especially
the tensile results, which would also need to be related to sheet thickness. In a
short article on the use of metallic protective coatings in the Soviet Union,
Trifel (1989), gave Table 2.25 without indicating the original source. These
data confirm the low atmospheric rates but also quote information from
Cuba-a much higher rate.
Work by Proskurkin and others (1978)primarily related to seawater
immersion in the Black Sea-gave a corrosion rate of 0.7-1.0 pdyear in
that marine atmosphere (18 months of tests) for zinc coatings, including
those produced with small additions of aluminum (< 0.14%), magnesium
(< and titanium (< 0.001%), separately or combined Chapter 3:
Table 3.22). Heat-treated coatings used for seawater pipe tests at Odessa
Naval Scientific Research Station (TsNII) lost 5-8 pm from the outside of the
pipes in 10-20 months in conditions entailing seawater spray (Proskurkin et
al., 1968). Tests on sherardized coatings in and above the Caspian Sea
(Proskurkin, 1972) showed (Fig. 2.39) the benefits of zinc in the alternate
wetting zone: the corrosion rate is no higher than for complete immersion or
out of the splash zone (unlike steel); other data are given in the section on
water immersion.
Buzdakov (1 958) had earlier commented favorably on heat-treated zinc
powder coatings in humid compressed air (Table 2.26). Another set of trials
on pipes for compressor pumps in the petroleum industry showed no attack in
6 months on the outer side exposed to air (Proskurkin and Gorbunov, 1967).
Compressed air at 2.6 atm and also caused no problem to sherardized
pipe systems in a refrigerator system (Proskurkin, 1969). Short-term tests in a
high pressure air line in the oil industry showed a corrosion rate of 0.0026
envimnment
category
envlmnment
cormsion
risk
N m l
B
OCZ&OnSl
CoKlensatlon
C
Condensallon
signincant
mmvered
document
Condensation
inlhls with
High
Atmosphere
examples'
mces
Best Yards over seawater
Ma&laclure
FoundneslSmeiters
'Buildings may contain different Conditions IAl will contain kitchen and IC1
2.33 Coating recommendations for atmospheric exposure in U.K. building
interiors (BSCIBCSAEDAPRA, 1986).
196 Chapter 2
Table 2.19 Recommended Reference Data on the Rate of Atmospheric Corrosion
of Zinc in the Soviet Union
Corrosion rate (pdyear) Concentration Concentration
atmosphere Dry Moderately wet Wet (mg/m3) (mg/m*/day)"
Rural < <
Urban 1.0
Industrial
Coastal <
Marine Up to < >
Type of of so* of c1-
The concentration of pollutants in the atmosphere was determined the methods recommended in GOST
Soviet standard 9039-74.
Source: Mikhailovskii et al. (1986).
similar urban atmospheres, the differences are even less: for large cities
(Berlin, London, New York) the range was pdyear and for residen-
tial suburban areas (Berlin-Dahlem, Hamburg/Eppendorf) prdyear;
the difference between these two ranges can be readily attributed to differ-
ences in the sulfur dioxide content of the air. All urban atmospheres are not
equally corrosive. Morcillo and Feliu in a detailed study of Madrid
(Fig. showed that even the central area atmospheres corroded zinc at
only pdyear , with half this rate in suburban areas. Other work
includes that by Wiegand and Kloos on zinc-coated sheet in urban and
industrial atmospheres.
Historically, however, the annual corrosion at the various places fluctu-
ated little over the years; this is seen in particular the results of Hudson and
Stanners (Table Schikorr and Schikorr found also that it mattered
comparatively little in which month the tests were started.
That corrosion depends little on the height above ground was shown by
tests on a television tower at Sutton Coldfield near Birmingham (Table
This result was confirmed by a survey in the late by Van Eijnsbergen on
weathered Dutch galvanized transmission towers, which showed a residual
pm at the top, 10 pm in the middle, and pm at the bottom-all within the
possible limit of random error.
It is often said that corrosion is particularly severe in the immediate
neighborhood of chimneys and houses, where the sulfur dioxide concentration
in the air can be particularly high, but no specific data have been found. It
may be expected that the higher the chimney, the smaller the effect, since the
Table 2.20 Corrosion Zinc and Steel at Far-Eastem Exposure Stations
in the Soviet Union
Corrosion
rates
Surface Sulfur Steellzinc
Exposure station (mg/m*/day) (mg/m3) Zinc Steel (by mass)
chlorides, Cl- dioxide ( P W ~ ~ ) " corrosion ratio
Rural
Bilibino
Oimyakon
Atka
Klyuch
Pobedino
Yakovlevka
Pogranichnyi
Tynda
Susuman
Ust'-Omchug
Aldan
Komsomol'sk-na-
Coastal
Aion
Apapel'khino
Schmidt
Chaplin
Arman
Ust'-Khairyuzovo
Ayan
Chumikan
Baidukov
Nikol'skoe
Lopatka
Gamov
Okhotsk
Okha
Poluostrov-
Nevel'sk
Vladivostok
Amure
Kamchatskii
17.8
10.6
18.9
15.5
3.1
8.6
6.7
9.4
16.9
18.0
23.8
11.6
10.5
5.4
6.0
9.1
10.5
0.23 0.69
0.25 1.04
0.23 1.95
0.28 3.00
0.60 4.68
0.64 5.21
0.60 6.28
0.74 2.72
0.43 2.18
0.40 1.59
0.76 3.17
0.88 8.10
1.60 8.29
0.85 4.76
3.54 17.76
1.17 17.88
0.83 4.17
1.10 11.83
0.94 6.83
0.90 7.26
1.64 14.50
2.61 28.37
3.18 31.71
21.91
0.005 1.17 11.68
0.005 1.11 8.63
0.015 2.12 13.73
0.015 2.79 17.87
0.015 2.00 22.47
3.3
4.5
9.0
11.5
8.5
8.8
11.3
4.0
5.5
4.3
4.5
10.0
5.6
6.1
5.5
16.6
5.4
11.7
7.8
8.7
9.6
11.8
10.8
10.8
8.4
7.0
6.9
12.2
~
"Panchenko et al. gave the corrosion rates in grams per square meter per year. Data were con-
verted to micrometers per year by dividing by 7.2 (zinc) or 7.8 (steel).
Source: Panchenko et al. (1985).
Chapter 2
Years
2.34 Rates of corrosion zinc and cadmium at various locations in open
atmosphere (0) and in shuttered booths in the Soviet Union: a = industrial; =
rural; v = northern marine; g = southern marine (Berukshtis and Klark, 1971).
greatest effect is at a distance of 10-15 times the height of the chimney from
its base, where the concentration of sulfur dioxide is much lower than in the
flue gases. Furthermore, changes in the wind direction will constantly expose
different surfaces to the contaminated air, reducing its effect at any one point.
The corrosion influence of chimney gases on galvanized objects may stretch
over distances as great as 100 km, depending on the direction and force of
prevailing winds.
D. Industrial Atmospheres
Although corrosion is more severe in industrial atmospheres than in urban
atmospheres, it is essentially of the same type; historically, damage in indus-
trial areas amounts to 3-15 pdyear (Table 2.5 and Fig. 2.2). However, more
severe corrosion is found near points where waste gases are emitted. Some-
times corrosion has been surprisingly low in heavily industrialized areas-the
figure for MulheimRuhr, for instance, was 4 pdyear during World War I1
(Schikorr and Schikorr, 1943), and such low figures are likely to be usual in
the future. The AWS 19-year tests (1974) showed complete protection by 225
pm zinc (12 pdyear), but only 75-150 pm zinc was needed if it were sealed.
The corrosion of zinc in industrial atmospheres, being essentially simi-
l ar to that in an urban environment, is relatively more severe because the
concentration of sulfur dioxide in the air is higher. However, the distinction
Table 2.21 Metal Corrosion over a Year in Open and Sheltered Exposures from Two Sources: A and B"
LOSS of metal (g/m2/yr)b
SO2 CI- per day
Aluminum (mg/m3) (mglm'lday)' Steel Zinc Cadmium Copper
Site A B A B Open Open Sheltered Open Sheltered Open Sheltered Open Sheltered
~~ ~ ~
Moscow 0.20 0.19 1.70 1.66 94 17 12.5 49 17 9.0 6.5 1 .o 1 .o
(industrial
region)
(rural
region)
(marine
region)
coastal
region
Zvenigorod 0.01 0.01 0.17 0.33 57 7 6.0 10 10 6.5 2.3 0.3 0.2
Batum 0.05 0.02 1.00 1.00 66 I 1 4.0 14 6 6.6 3.5 0.1 0.05
Northern 0.01 0.012 14.10 13.33 233 19 12.0 12 7 24.0 14.3 7.2 4.3
~ ~~~~~~ ~
'A, Strekalov and Berukshtis (1965); B, Berukshtis and Klark (1965).
bNot converted to thickness loss because not all metals c o d e uniformly. To get loss in micrometers per year for zinc, steel, and cadmium, divide by 7.2,7.8, and
8.6, respectively. Aluminum does not corrode uniformly, and a calculated loss of thickness is meaningless.
'Quantity of chlorides sealing on the surface from the air, determined by the dry cloth method.
200 Chapter 2
Table 2.22 Corrosion (pm) After One Years Exposure of Zinc and Cadmium
in Open Atmospheres in Western and Eastern Countries
Zinc + Zn coatings Cadmium + Cd coatings
United States,
United States,
England, and Soviet England, and Soviet
Locality other countries Union other countries Union
Rural, humid 0.5-1
Industrial 2-6
Marine 4-20 1.8-15
Table prepared by Strekalov and Berukshtis (1965).
between these two types of environment is imprecise; in some industrial areas
corrosion is no more pronounced than in many large towns, and most descrip-
tions are subjective with respect to the country concerned.
The effect of flue gases may, however, be considerable where such
emissions are present in considerable volumes, or if there is a tendency for the
smoke to lie very low. For instance, a zinc loss of 40 pn/year was recorded
Time, ye ar s
lime, years
2.35 Long-term atmospheric tests on zinc and cadmium coatings in the Soviet
Union; corrosion resistance of (a) zinc and (b) cadmium coatings on steel in various
climatic localities: M = Moscow; B = Batumi; C = North sea territory; 3 =
Zvenigorod; solid lines = nonchromated coatings; dashed lines = chromated coatings
(Strekalov and Berukshtis, 1965).
120L
100 ~ -1 5
80 -
60 -
I
-1 0
-5
7 t 1 5
(a) lime, days lime, days
2.36 Accelerated corrosion tests comparison with Fig. on (a) zinc
and cadmium: 1 = SO2 gas; = sea mist; = warmth and humidity; solid lines
= nonchromated coatings; dashed lines = chromated coatings (Strekalov and Be-
rukshtis,
I i l I I
i
2.37 Five-year atmospheric exposure of zinc and cadmium in open atmo-
spheres (solid lines) and in shuttered booths (dashed lines) in the Soviet Union.
Ks,,,,/Kbaoth ratios for the four localities as follows (ZdCd): a = industrial,
b = r u r a l , v = northern marine, g = southern marine,
(Berukshtis and Klark.
201
202 Chapter 2
Table 2.23
on the Rate of Corrosion of Sheet Zinc and Cadmium
Influence of Sulfurous Gas, Chlorides, and Temperature
Duration of testing (years) for
Zinc Cadmium
Conditions
of exposure 1 2 3 5 1 2 3 5
Influence of sulfurous gas (K industriallK agricultural) ratios
open atmosphere 2.8 3.8 3.8 3.8 5 6 6.7 6.0
shuttered booth 3.0 3.5 3.8 3.7 2.0 2.3 2.4 2.6
open atmosphere 2.5 2.5 2.5 3.0 1.0 0.9 1.0 0.9
shuttered booth 2.6 3.0 2.6 3.6 0.7 0.6 0.6 0.6
open atmosphere 1.7 1.7 1.5 1.5 1.4 1.3 1.4 1.2
shuttered booth 0.7 0.6 0.6 0.6 0.6 0.4 0.6 0.6
Influence of chlorides (K industriallK agricultural) ratios
Influence of temperature (K southern industrialiK agricultural) ratios
Under the influence of chlorides, the rate of corrosion of Zn increases 2.5-3.6-fold (in the open
atmosphere and in shuttered booths), but with increase in concentration due to deposition of salt
from seawater on the metal surface, the increase is ten fold.
Source: Berukshtis and Klark (1971).
between the hoods of a steam locomotive shed, while only 4 m away from this
shed the rate was down to 8 pdyear (Table 2. 5) . Proskurkin and Gorbunov
(1975) also document corrosion rates in a coke-chemical plant, which varied
from 0.1 to 70 pdyear according to the specific location (Table 2.27).
The corrosion rates for tests in Hamburg harbor are worth noting. The
sea is about 100 km away, but there is a significant chloride content to sup-
plement the effects of the industrial or urban atmosphere.
Seasonal fluctuations of corrosion are similar in industrial and urban
environments, particularly the more severe corrosion that occurs in winter
(Schikorr and Schikorr, 1943). The difference in the seasonal corrosion rate
can be explained as follows: even for the same sulfur dioxide content of the
air, it can be expected that in summer the absorption of SO2 by the zinc
surface is less than in winter because of the lower relative humidity in the
warmer months. The same values for absorption in the Liesegang bell were
found whether the covering was impregnated with potassium carbonate or
with sodium carbonate (i.e., two salts with very different hygroscopic proper-
ties). This suggests that the bell absorbs sulfur dioxide irrespective of the
actual relative humidity (Schikorr, 1961a).
E. Marine Atmospheres
The corrosion of zinc in temperate marine atmospheres (Table 2.6) is of the
same order as in urban atmospheres, about 1-7 pdyear; in areas where there
is much sea spray, it is of the same order as in industrial atmospheres
pdyear). The AWS (1974) tests showed a corrosion rate of up to 12 pdyear
but only 4-8 pdyear when the sprayed zinc was sealed.
The air clearly has a higher salt content on the coast and for several
miles inland. According to Ambler (1960), this salt content is approximately
inversely proportional to the distance from the open sea (as determined by a
method similar to that of Liesegang for sulfur dioxide). Areas close to the
coast are apparently significantly corrosive to zinc only where much sea spray
occurs; elsewhere, the content of airborne seawater salts is insignificant for
the corrosion of zinc.
In tropical marine atmospheres, corrosion is much greater in sea spray
conditions, primarily because of the higher temperature. It would seem that
magnesium salts present in the tropics (at a level similar to that in temperate
waters) are insufficient to inhibit corrosion. The dependence of zinc corrosion
on the distance from the coast was discussed earlier.
Table 2.2 shows that the corrosion rate on zinc exposed at the coast
decreases during the first few years, but the results of tests run for IO and 20
years (ASTM, 1956) suggest that no further improvement occurs over longer
periods. This increase in protection seems to be connected with the magne-
sium salts in seawater, as is shown directly when alternate dipping tests with
3% sodium chloride solution and with artificial seawater are compared (Fig.
2.40). Other long-term tests (> years), described by Delpeuch (1972),
also cover industrial atmospheres and flowing seawater.
F. Rural Atmospheres
The corrosion of zinc in rural atmospheres (Table 2.3), usually about
pdyear , is related to the relatively small impurity content. Considerably
higher values ,may however be found in areas with high rainfall or with
airborne pollution: the relatively high value at Llanwrtyd Wells is remarkable;
it suggests that the lower the sulfur dioxide content in the air, the greater the
influence of rain. The tests in Llanwrtyd Wells point to a progressive im-
provement in the protective effect of the film; the mean value of the corrosion
rates determined each year separately for 5 years is 3.4 pdyear , whereas
specimens exposed uninterruptedly for 5 years corroded at an average rate of
only 2.5 pdyear . However, the protective effect may not improve any
further with longer times: thus, no decrease in the corrosion rates was found in
.
l
l
204
the tests at State College, Pennsylvania, extending over 10 and years
(ASTM, 1956).
According to Anderson (1955), corrosion of wire fences in rural atmo-
spheres is a little greater than normal in the shade and at points where the
fence is covered in grass, presumably because moisture is retained longer at
these points. Possibly, however, the metabolism of the grass or leaves has an
effect (see below).
Considerable corrosion of zinc in rural atmospheres was observed in the
following special cases:
1 . Guttering installed with an insufficient fall will retain rainwater, and
this may increase corrosion. In a case studied by Schikorr (1965), in the
village of Musberg (12 m southwest of Stuttgart), guttering with no fall, made
from galvanized steel with a zinc coating of only 30 pm, was perforated after
35 years, whereas other guttering on the same house with an adequate fall
showed no rust.
2. Another case involved corrosion due to plant growth. At Luneberg
Heath, in a moist position, zinc sheet roofing that had been invaded by
various plants, in particular moss and houseleek (Sempervivum rectorum),
was perforated after a few years. Probable causes of this corrosion, which has
not been studied in detail, were the products of plant metabolism or the
continuous effects of humidity retained for a particularly long time, either in
the plants themselves or in accumulated soil while, at the same time, access of
air was hindered.
G. Tropical Atmospheres
"The tropics" here means the region between latitudes approximately 30"N
and 30"S, that is, comprising somewhat more territory than the geographical
tropics (23"N and 23"s). A great variety of climatic conditions can be found in
the tropics, and the atmospheric corrosion of zinc varies accordingly. Distinc-
tions are therefore made between dry tropical atmospheres (deserts), moist
tropical atmospheres (tropical rain forests), and tropical marine atmospheres
(near thc coast).
In the tropics, zinc corrodes less than is usually assumed, under 2
pdyear in dry atmospheres, less than 3 pdyear in humid atmospheres, and
2.38 Average thickness of zinc, cadmium, and coatings of those metals
without ( 1 , 1 ', and 1") and with and chromate coatings, in the Soviet Union,
for the following dates: 1, 1949-1951; l', 1955-1957; 1". 1964-1966; 2, 1955-1957;
1964-1966; 1963-1966. Localities a, v, and g as in Fig. 2.34 (Berukshtis
and Klark, 1971).
206 Chapter 2
Table 2.24 Change of Tensile Strength and Rate of Corrosion Rolled Zinc
Plates with Time
Rate of corrosion Change in tensile
Typical
atmosphere Metal IO years 20 years 10 years 20 years
(pmW=) strength (%)
Rural
Rural
Industrial
Industrial
Marine,
coastline
Marine,
subtropical
Marine
BB
HH
BB
HH
BB
HH
BB
HH
BB
HH
BB
HH
BB
HH
0.28
0.19
0.13
1.13
0.92
0.96
5.15
5.23
4.73
5.15
5.38
5.10
1.88
1.69
I .32
0.55
0.48
1.48
1.44
0.17
0.22
0.42
1.08
1.11
0.94
6.65
7.72
6.73
5.45
5.65
5.60
1.69
1.73
2.09
0.65
0.54
1.83
1.68
-1.1
-0.3
- 5.5
- 1.7
- 1.0
- 9.7
- 14.1
- 8.5
- 17.0
- 14.5
-9.2
- 18.8
- 3.6
-0.3
- 8.2
-2.1
-0.7
- 7.6
-2.5
-2.5
- 10.0
-7.8
-2.5
8.9
- 10.0
-4.9
- 11.4
- 30.6
- 29.3
-31.9
-25.8
-20.9
- 29.0
-9.1
-4.3
- 12.7
- 6.4
-4.4
~~
- 7.7
- 15.9
technical punty Zn (0.84% Pb + 0.042% Fe + 0.20% Cd + 0.0002% BB, high
purity Zn (0.049% Pb + 0.19% Fe + 0.002% Cd + 0.006% HH, special high punty Zn
(0.0055% Pb + 0.0015% Fe + 0.001% Cd + 0.0006%
Source: Rosenfeld (1960).
less than 6 pmdyear at the coast. Only in certain regions where there is
under severe condensation conditions, is corrosion more pronounced. The
results of 16-year tests in Panama are reported by Southwell et al. (1976).
On the basis of tests in Bombay and Port Harcourt, Nigeria (Table 2.6),
corrosion in humid tropical industrial atmospheres seems to have values
similar to those in Europe. Biestek (1974b) reports no corrosion on a chro-
mated 15 pm zinc coating after 2 years in the open 5 years under shelter in
Canton, and (1978) gives results for 2 years at Julin-Hainan Island.
Table 2.25 Test of Metallic Coverings in Different Climates and Parts
of the World
Corrosion rates (Fdyear)
Period of
Zone of testing Metallic
Points of testing testingb (Yea@ Steel zinc coating
Baku, Caspian
Sea
Baturnyi, Black Sea
Tide Power Station,
Hydro Power Station,
Arctic Ocean
River Volga
Republic of Cuba,
Gulf of Mexico
A
S
W
A
A
S
A
S
W
A
1 .o
0.4-0.09
150 18
1.2
150
10
corrosion rate changes during the year: maximum in summer; minimum in winter.
bA, air zone; S, splash zone; W, underwater zone.
Source: Trifel (1989).
a b
Corrosion r at e - mm/year
Fig. 2.39 Performance of sherardized coatings in and above Caspian Sea showing
dependence of the corrosion rate on the test zone for (a) zinc-coated and (b) un-
protected carbon steels; the m zone is the alternate wetting zone (Proskurkin,
208 Chapter 2
Table 2.26 Sea Spray Corrosion of Zinc-coated and Bare Steel Near Oil Fields
in the Soviet Union
Position
m above water
m above water
m above water
m above water
At water level
m below water level
Average depth corrosion (pdyear) for
112-day test test
Zinc-coated" Bare Zinc-coateda Bare
.o
The zinccoated samples are diffusion coatings starting from powder. The tests are apparently at
different timeslplaces.
Source: Buzdakov ( 1958).
Table 2.27 Zinc Diffusion-Coated Tubes in a Coke-Chemical Plant
Corrosion rate (pndyear) specimens
Area of plant in which Galvanized
tests were carried out and annealed Sherardized Uncoated
Absorption solution heat
Cooler-condensers (recuperators
Coal tar absorption oil heat
Benzene separation condenser
Practically
Ammonia vapor fractionating
exchangers (preheater no.
no.
exchangers
(top opening) uncorroded
column (in collector at the
entrance to the apparatus)
tubes in block IV)
space between tubes Practically
space between tubes of
Drip coolers (space between
Shell and tube gas coolers
uncorroded
second section
The specimens were tested for 12 months.
Source: Proskurkin and Gorbunov (1975).
14 days: 198 g1m2
0 2 7 14
Duration of test (dap)
o 309 NaCl + 4.49 Na2SOdI
309 NaCl/l
o Synthetic sea water
0 309 NaCl + 5.69 MgClin
Fig. 2.40
(Schikorr, 1940).
Protective effect of magnesium chloride in alternate dipping tests
In the humid tropical atmosphere of Singapore, zinc is corroded as little
as in a European rural atmosphere; in tropical marine atmospheres on the
African coast, the corrosion decreases rapidly a short distance away from the
coast (Fig. 2.41). The relationship between corrosion and airborne seawater
salts is shown in Fig. 2.42; corrosion is very severe in regions where there is
surf from tropical seawater, low in magnesium salts.
Tests made by Lowe (1961) showed that galvanized steel was heavily
corroded on the coast of Hawaii. In Papua-New Guinea (K. G. Martin,
1973), tests showed that galvanized steel would last at least 20 years.
Often the corrosion rates of zinc decrease considerably over the years in
tropical inland atmospheres; this must be attributed to an improvement in the
protective film. The values in Table 2.6 from Port Harcourt and Nkpoku,
21 0
Chapter 2
A
0 100 200 300 40
Distance from surf region (meters) (100 meters = 320 ft)
o Wet season (six months)
o Dry season (six months)
0 At height of 23 meters (74 ft)
A At height of 23 meters (74 ft)
Whole year
Fig. 2.41
Stanners, 1953).
Dependence of zinc corrosion on distance from tropical surf (Hudson and
Nigeria, also suggest such an improvement: the annual average value is less
than the sum of those for the wet and dry seasons separately.
In the Republic of South Africa, very great differences occur. In the
subtropical area on the Indian Ocean (Durban, Port Elisabeth, Richards Bay)
corrosion of galvanized items is considerably higher than on the veldt (Johan-
Resistance to Atmospheric Corrosion 21 1
0
-
c
o In the open
o Shielded from rain
2.42 Relation between corrosion and airborne seawater salts by season in
southern India (Rajagopalan et al., 1959).
nesburg, Pretoria). Extremely severe corrosion occurs around Walvis Bay,
where a unique combination of fog, salt air, and temperature drops accelerates
the corrosion of galvanized steel. An interesting report has been made by
Smuts and Callaghan (1978).
Also, corrosion on galvanized steel poles and fences in the coastal area
of Simonstown and Muizenberg (Indian Ocean, Cape Province) is pro-
nounced. At a distance of 10-20 m from the Ocean spray, severe corrosion
occurs after years. Only duplex systems offer sufficiently long protection
against rust. Unprotected steel is perforated within only a few years, but this
knowledge led to successful trials of galvanized rails for the railway track.
IV. CORROSION DATA BY ALLOY
A. Zinc and Alloy Coatings on Steel: General
Generally, a galvanized steel coating is corroded in a similar way to solid
zinc. At the points where the zinc coating is defective, the bare steel is
cathodically protected under most conditions. The iron-zinc alloy present in
most galvanized coatings, and forming the whole of galvannealed coatings,
has higher corrosion resistance than zinc in neutral and acid conditions.
21 2 Chapter 2
In air, zinc coatings on steel corrode similarly to solid zinc, and there is
an approximate straight-line relationship between weight loss and time. Cor-
rosion of zinc data given in preceding sections can be to describe
generally the corrosion of zinc coatings. The purity of the coating makes little
difference to corrosion rates, particularly in industrial atmospheres (Anderson
and Reinhard, Anderson, a difference of more than in the
semiarid climate of Phoenix, Arizona (Anderson, is of no practical
importance because even the impure zinc corroded at only pdyear (i.e.,
a pm coating would last years). Anderson also shows a difference in
a mild coastal atmosphere (La Jolla, CA), but even at 2 p d y r (vs. 0.8
year for zinc) the significance in practice is small. Since, in individu-
al urban and industrial environments, the weight loss of zinc is proportional to
time, the life of the zinc coating (i.e., the time until the zinc layer is
perforated and the exposed steel begins to rust) is proportional to the thickness
of the zinc layer (Richards, Hudson and Banfield, ASTM,
This has been demonstrated directly. The protective film on zinc increases
with time in rural and marine atmospheres of some types and, under these
conditions, the life of the zinc may increase more than in proportion to
thickness (Table However, this does not always happen, particularly
with thin nonalloyed coatings on sheet-for example, in the ASTM tests
(Figs. and
Where it is necessary limit the thickness of zinc (e.g., on bolt
threads), it is then necessary to provide a supplementary coating on the
exposed parts or to choose a zinc alloy coating. Results of 10-year tests by
Goodwin and Piessen showed that galvanizing with either zinc
dustlzinc oxide paint or polyurethane paint, anodizing, performed best,
followed by galvanizing sherardizing plus zinc-rich paint. However, gal-
vanized coating thicknesses pm with pm on the anodized sample)
were greater than sherardized pm); without a top coat, the perfor-
mance of coatings of different types was directly related to the amount of zinc
present.
Because of the Mears effect (wire corrodes faster per unit of area than
more massive materials), galvanized wire corrodes some faster than
galvanized sheet (see Table also see the series of Haynie et al. papers for
explanations related to deposition rates of sulfur dioxide), and thin wire
corrodes faster than thick wire.
The life of rope made from galvanized steel wires is, however, greater
than that of the individual wires (Schikorr, ASTM, Schi-
korr explains this by pointing out that the parts of the wire that lie on the
outside are corroded more rapidly and, when the zinc film is penetrated in
24
20
16
12
-
c
8
m
0)
r
0,
C
In
-
.-
c
2
L

0
Q In
C
0
Q)
0
4
5
c
0
i !
F
Zinc coating (g/m*) (100 g/m2 = 0.33 oz/ft*)
o State College, Pa. (rural atmosphere) mean values
State College, Pa. (rural atmosphere) maxima and minima
0 Altoona. Pa. (industrial atmosphere) mean values
Altoona, Pa. (industrial atmosphere) maxima and minima
Fig. 2.43
14 pm (ASTM, 1954).
Galvanized steel sheet: coating weight versus life to first rust; 100 s/m2 =
21 4 Chapter 2
Zinc coating (glm') (100 elm' = 0.33 oz/ff*)
0 Santa Cruz. California (rural atmosphere, 5 km from Pacific)
v Ames, Iowa, California (rural atmosphere)
o State College, Pa. (rural atmosphere)
o Bridgeport. Conn. (urban atmosphere)
Fig. 2.44
pm (ASTM, 1954).
Galvanized steel: coating weight versus life to 1% rust; 100 s/m2 = 14
those regions, the uncorroded zinc inside the rope provides cathodic protec-
tion for the outer regions. The Mears effect may also be responsible for the
greater life of ropes.
If thickly zinc-coated sheet or wire is bent, cracks may occur in the zinc
coating down to the steel. Rust is not generally formed on the steel because it
is more noble than zinc and, in the corrosion cells produced, the zinc acts as
anode and the steel is cathodically protected. Pores in the zinc coating at the
base of which the steel is exposed will behave similarly (Richards, 1938;
Hudson and Banfield, 1946). In tests at three places in the United States
(reported by Pellowe, 1952), this protection prevented formation of rust on
galvanized steel window frames at points where the zinc coating was dam-
aged. Protecting galvanized wire and wire netting by a vinyl polymer based
plastisol yields a very great improvement in corrosion resistance.
The cathodic protection effect of zinc has been disputed by several
authors (Bablik, Klas and Steinrath, Elze, There are
certainly cases in which this effect does not occur-for instance, on zinc-free
zones more than 2 mm wide (Fischer and Bh a n n , Spindler,
having a high electrical resistance of the circuit (which often occurs in the
atmosphere when only a thin layer of pure water is present), since an adequate
preventive electric current cannot then flow.
Outdoor exposure tests by the Dutch Galvanizing Institute showed that
on galvanized panels with pm zinc scratched with widths up to 5 mm,
no rusting took place after years of weathering in an industrial-marine
climate. Similar tests were made on cut galvanized rebars (diameter 6-20
mm), exposed for 5 years. On the 6-10 mm diameter bars, no rust appeared.
On the 20 mm diameter bars, the cut surface showed a uniform, light brown
layer of rust. However, the rust had not undercut the zinc coating.
The cathodic protection on damaged galvanized items depends on (a)
the coating thickness, the width of the bared area scratch, and (c) the local
climate. Allowance must be made for any extra loss of zinc due to cathodic
protection effects due to exposure of more zinc surfaces to the environ-
ment. It also has been shown that coatings of AI/Zn alloys alone offer no
only poor cathodic protection. More less the same applies to duplex
systems, where the paint coat covers the zinc layer, thus preventing
retarding the liberation of the necessary Zn2+ ions.
The effectiveness of cathodic protection probably depends on the com-
position of the outer portion of the zinc layer. Pure zinc is more anodic than
iron-zinc alloys, so bare steel can more readily be protected by zinc than by
the alloy. The degree of cathodic protection depends largely on the exposure
conditions.
vast amount of research attention has been given to the effects of
minor impurities additions to a galvanized coating. Such effects are minor
compared with environmental factors. Gilbert and Radeker et al.
carried out extensive trials in industrial atmospheres which showed
differences in corrosion rate. Gilbert's work gave losses of pdyear for
years at Euston, but without any correlation with bath composition (Al, Sn,
Sb, and Cd additions). Radeker et al. found that in a one-year test pm
was lost, with copper additions giving the lowest corrosion rates; an adverse
effect of aluminum was canceled out by copper. A slight detrimental effect of
tin additions was found; New Jersey Zinc had earlier reached the same
unpublished conclusion but did not publish it.
In Harvey indicated in a review that several conclusions could be
drawn with respect to atmospheric exposure:
21 6 Chapter 2
1. Small additions of aluminum slightly decrease the corrosion resistance of
zinc, especially if lead is also present. The maximum decrease occurs at
0.3% aluminum (in an alloy containing 0.2% lead). Aluminum appears
to increase slightly the risk of intergranular attack in zinc alloys.
2. Lead by itself is not particularly harmful. However, in conjunction with
any concentration of aluminum, it appears to accelerate corrosion.
3. Copper improves atmospheric corrosion resistance and retards inter-
granular attack. This effect may not make any great difference with long-
term corrosion. Magnesium improves the corrosion resistance of zinc and
can counteract any bad effects of aluminum.
4. Small amounts of chromium, nickel, or titanium may enhance the corro-
sion resistance of zinc.
Although most corrosion research has been on zinc sheet or hot dip galvanized
panels, the results can be applied to other zinc coatings because differences in
the behavior of coatings are small compared with environmental factors.
Special mention must be made of the AWS 19-year tests (1974) on
thermally sprayed zinc and aluminum coatings in the atmosphere and in
seawater. Unsealed zinc-sprayed coatings 230 pm thick showed no rust at the
Kure Beach 80 ft. site (i.e. 80 ft. from the sea); edge rusting had started with
the 150 pm coated steels. At the milder 800 ft. site even 80 pm of unsealed
zinc had prevented rust; aluminum-vinyl-sealed zinc spray remained in good
condition, especially with thicker zinc coatings which could absorb more
sealer, and on downward-facing surfaces. Brazos River, Texas, gave similar
atmospheric exposure results after 19 years. Point Reyes, California, sim-
ilarly showed no rust on 80 pm sprayed zinc after 19 years of marine
exposure, but 510% paint blistering occurred when an aluminum-vinyl
sealer was applied.
In industrial atmospheres (New York City; Columbus, Ohio; East Chi-
cago, Indiana), 230 pm of unsealed zinc provided 19 years of maintenance-
free protection; 150 pm zinc showed slight rust on the upward-facing surface
in New York and at edges in Columbus. 80 pm sprayed zinc sealed with
aluminum-vinyl was in generally good condition; deterioration the
aluminurnvinyl often was associated with excessive application. A chlori-
nated rubber sealer was generally less effective than the aluminum-vinyl.
The steel industry, which uses large amounts of zinc in the continuous
coating of steel strip and wire, would like to save money by using a thinner
coating offering equivalent performance. The vehicle industry now uses zinc
or zinc alloy coated steel for about half the bodywork of mass-produced cars.
This proportion is rising and has reached 100% in some cars, such as the Audi
series; the manufacturers wish to save weight as well as direct costs.
v)
- .-
E
u;
v)
s
c 0
ul
L
.-
0,
0
V
0)
F
W >
4:
This demand for lighter weight coatings has stimulated extensive re-
search in the past 15 years on zinc alloy coatings: most testing has been by
accelerated laboratory tests and, although these have become more sophisti-
cated in recent years, they do not yet enable quantitative assessments of
service performance to be made. Although useful as sorting tests, particularly
on materials of the same type, they tend to exaggerate the differences in
performance that are found in practice.
The most important series of alloys consists of the zinc-aluminum
combinations. The general performance of such alloys was shown by Zoccola
et al. (1978) in 5-year tests (Fig. 2.45), but this work was based on simple
alternate-dip test pieces without the sophistication of the alloy and processing
modifications of the past 20 years, which have channeled commercial devel-
0,70
0. 60
0. 50
0.10
0. 30
0.20
0.10
0
0 10 20 YO 10 50 60 70 80
Wt. % Al in At-Zn Coatings
Fig. 2.45 Five-year corrosion tests on early zinc-aluminum alloys: effect of alumi-
num content on corrosion performance of aluminum-zinc alloy coatings after exposure
in various atmospheres; 1 mil = 25.4 pm; 1 ft = 0.305 m (Zoccola et al., 1978).
21 8 Chapter 2
opment mainly in the 5 and 55% aluminum areas for hot dip coatings; the 15%
aluminum area has developed for thermally sprayed coatings with their
different structure.
The following remarks, classified by alloy, relate primarily to cases in
which long-term outdoor atmosphere exposure data are available; some refer-
ences are also made to accelerated test results, to assist investigators.
B. Zinc-5% Aluminum Hot Dip Coatings
Most of the available long-term data relate to Galfan (5% aluminum +
mischmetal). The Galfan-type coatings have been exposed in typical atmo-
spheres, and Goodwin (1988) shows the marked improvement compared with
the reference panels of standard galvanized steel (Tables 2.28), both in a
relatively mild urban-industrial atmosphere and in a marine atmosphere. The
latter is particularly significant, both because the corrosion rate appea-
unlike the case of galvanizing-to be slowing after about 4 years (Fig. 2.46)
and because conventional galvanized steel should show rust in 5 years. The
slower rate of corrosion also means that the zinc-5% aluminum coatings
provide full cathodic protection to cut edges over a longer period (Table
2.29). The 55% aluminum alloy (Galvalume) also lost thickness faster over a
2-year period than Galfan at this aggressive site, although it was much better
in the urban atmosphere. Care should be taken in extrapolating all these
results because chromate coatings-frequently present on coated sheet as
produced-tend to slow down initial corrosion, while the aluminum-rich
coatings have a corrosion rate that also tends to diminish after the first year or
two. Test results of up to 22 years are available, plus up to years as part of
Table 2.28A Five-Year Outdoor Exposure Resultsa
Thickness loss (pm)
Atmosphere Galvanized Galfanb improvement
Industrial 15.0 5.2 2.9
Severe marine >20.0 9.5 B2.1
Marine 12.5 7. 5
Rural 10.4 3.0 3.5
'All samples: 275 gm2 coating weight (ambient cooled); steel thickness, 0.65
mm.
bSIow-cooIed coarse structure Galfan.
Source: Goodwin (1988).
Ratio
Table 2.286 Three-Year Outdoor Exposure Resultsa
21 9
Thickness (pm)
Atmosphere Galvanized Galfanb improvement
Ratio of
Industrial
Severe marine 4.0
Marine
Rural
'All samples: 275 glm' coating weight (ambient cooled); steel thickness, 0.65
mm.
bFast-cooled minimized structure Galfan (optimized).
' I - ! - -). Galvanized
/
/"
/
p-:
0
0 24 36 48 60 72 84 96
Exposure time (months)
2.46 Seven-year exposure of Galfan and galvanized steel in a severe marine
atmosphere (Lamberigts et al., 1991).
220 Chapter 2
Table 2.29 Comparison Cathodic Protection for Galfan and
Galvanized Coatings"
Amount (mm) of bare edges
exposed after years
(coating recession from edge)
Environment Galvanized Galfan
Severe marine 1.6 0.1
Marine 0.5 0.06
Industrial 0.5 0.05
Rural 0. l 0
"All samples: 275 @m2 coating weight (ambient cooled); steel thickness, 0.65
mm.
buildings (Townsend et al. 1988). The good results for Galvalume in the
urban-industrial atmosphere are achieved at the expense of a certain amount
of cathodic protection.
Galfan coating development continues and minimized-spangle Galfan
has corrosion resistance up to about three times that of galvanized steel.
Galfan also provides excellent protection at panel bulges because it can be
formed with much smaller cracks than are obtained in conventional gal-
vanized steel. This lesser cracking also means that less zinc is exposed to the
environment, and this increases the relative performance factor compared
with galvanized steel (Fig. 2.47).
Less information is available on the corrosion resistance of Superzinc
(5% aluminum + magnesium), but this alloy, too, is expected also to
better than conventional galvanized steel.
When unchromated or chromated sheets with zinc-5% aluminum type
coatings are stacked in corrosive environments likely to encourage oxygen
concentration cell corrosion (e.g., hot and humid environments), they tend to
become dark gray or black on the surface rather more than is observed on
conventional hot dip galvanized sheets. In such a corrosive environment,
oxygen concentration cell corrosion stimulates the anodic reaction in the zinc-
rich phase surface which, combined with the surface etching condition arising
from the reactivity of the two phases during chromating and with the differ-
ences in formation of the chromate film and in potential between the phases,
leads to formation the optically black corrosion products.
2.47 Corrosion resistance of deformed Galfan compared with galvanized steel
as demonstrated by SO2 test (CRM, 1991).
Similar but less pronounced blackening occurs in atmospheric expo-
sure. The aluminum-rich surface layer on galvanized steel is typically only
(1 nm) thick, whereas on Galfan produced during pilot trials it averaged 60
pm (0.060 mm). The surface of these layers is always aluminum oxide. The
reason for the thicker layer with Galfan is the higher aluminum content. By
cooling the Galfan at rates from 20 to 20O0C/s, various thicknesses of the
aluminum-rich layer can be produced. Coatings with the thinnest layers are
found to darken much like the normal galvanized material. This has been
proved by extended outdoor exposure tests on commercial material. The
darkened layer can grow with time and is controlled by diffusion. Thus, after
a few months' exposure, it will progress with the square root of time in either
case, galvanized or Galfan. A 20C/s cooling rate appears to be sufficient to
avoid excessive darkening. It is not known whether a similar modification
would apply on Superzinc: for that material, a proprietary special treatment
has been developed to prevent staining.
222 Chapter 2
C. Zinc-55% Aluminum Hot Dip Coatings
The primary alloy development is Galvalume, zinc-%% aluminum-lS%
silicon, now known under many tradenames including Zincalume, Aluzinc,
Aluzink, Algafort, Alugalva, and Zalutite.
Galvalume steel panels produced during early pilot line runs have been
on exposure in the United States for almost two decades at four atmospheric
sites representing rural, industrial, marine, and severe marine environments.
Corrosion-time curves from three of these sites (Fig. indicate the good,
long-term corrosion resistance of Galvalume sheet, notably in industrial and
marine atmospheres (see also Johnsson and Kucera, and also reflect the
nature of the corrosion mechanism described above. In considering the course
of corrosion in the atmosphere of Galvalume and similar coatings, attention
should be given to the interdispersion of the zinc-rich and the aluminum-rich
phases. Of prime importance is the presence of a dense, continuous inter-
metallic layer at the interface with steel. Initially, a relatively high corrosion
loss is observed for Galvalume sheet as the zinc-rich portion of the coating
corrodes and provides sacrificial protection at cut edges. This is true in all
environments, whereas aluminum provides adequate galvanic protection only
in marine chloride environments. After about 3 years, the corrosion-time
curves take on a more gradual slope reflecting a change from active, zinclike
behavior to passive, aluminum-like behavior as the interdendritic regions fill
with corrosion products. Jones et al. have looked at short-term electro-
chemical tests that might assist in evaluation of long-term performance.
In reporting the results for Galvalume sheet (Allegra et al., the
corrosion losses were calculated by using the density of the corroding species,
determined from the weight and chemical analyses of the coatings before and
after exposure. Using ratios of these rates, Allegra has predicted that Gal-
valume sheet should outlast galvanized sheet of equivalent thickness by at
least two to four times over a wide range of environments.
To retard weathering discoloration and improve atmospheric corrosion
resistance, in addition to preventing wet storage staining, chromate passiva-
tion procedures have been developed.
Galvalume sheet provides excellent cut-edge protection in very aggres-
sive conditions, where the surface does not remain too passive, but is much
less satisfactory on the thicker sheets in mild rural conditions, where zinc-5%
aluminum coatings are preferred because their general corrosion resistance is
good and, when sheared edges are exposed localized damage to the coating
occurs during fabrication service, the galvanic protection is retained for a
longer
Interest has been generated in hot dip coatings with aluminum contents
between 5% (of, e.g., Galfan and Superzinc) and 55% (Galvalume) in an at-
tempt to increase the general corrosion resistance of the coating without
incurring the disadvantages of the 55% aluminum alloy (limited cathodic
protection and less resistance to some alkaline conditions). Information cur-
rently available suggests that there is no specific level of aluminum between 5
and 55% that gives a new type of coating performance, rather, corrosion-
resistance properties change seem to gradually from one composition to the
other. For hot dip coatings, the mass of experience now available-and still
being developed-n the 5% (Galfan) and 55% (Galvalume) type material,
promulgated primarily by ILZRO on the 5% alloy and BIEC on the 55%
alloy, makes it much easier for a firm to select one these alternatives to
parallel the traditional galvanized product, which is also likely to retain a very
substantial user demand.
D. Zinc-15% Aluminum Thermally Sprayed Alloy Coatings
More than 10 years experience with thermally sprayed coatings of zinc-15%
aluminum is now available. As a result of the two-phase structure (a zinc-rich
and an aluminum-rich phase) of such coatings, the oxidation of the active
zinc-rich phase occurs within the more inert aluminum-rich phrase. Oxidation
products are encapsulated in the porous layer formed by the latter and do not
build up a continuous surface layer as with pure zinc coatings. Consequently,
no thickness or weight loss is observed even after several years exposure in
atmospheric field testing.
A 35% aluminum alloy was introduced in the 1960s and was suc-
cessfully used, for example, on sections of the Severn Road Bridge in
England. It was perhaps no accident that this alloy was promoted by the
powder manufacturers, since this composition gives a wire that is mechan-
ically unsuitable. Test panels covering the whole range of Zn-AI sprayed
coatings that were used by Hoar as the basis of his recommendations for the
Severn Road Bridge have remained on test at Kure Beach and extend the
10.5-year report (Hoar and Radovici, 1964), which covered more sites.
with the work of B. A. Shaw and Moran (1983, there were problems in
preparing a reliable Zn-A1 alloy wire, and mixed powders usually gave better
performance. Table 2.30 has, therefore, been constructed from the results in
Kain and Bakers paper (1 989) to show the performance with the better of the
two types of material normally applied at 80 pm minimum. Similarly, the
type of gun used and the choice of air or gas propulsion are simplified to show
the results from the best practice, since guns and techniques are much
EXPOSURE TIME, years
P
GALVANIZED
P
ALUMINUM-COATED TYPE 2
2 4 6 8 10 12 14
EXPOSURE TIME, years
224
ALUMINUM-COATED TYPE 1
TIME, years
~ ~ ~~
2.48 Thirteen-year exposures Galvalume in marine and industrial atmo-
spheres. Corrosion performance of galvanized, aluminum-coated, and Galvalume
sheet in (a) a marine atmosphere (Kure Beach, NC; 250 m lot), (b) a severe marine
atmosphere (Kure Beach, NC; 25 m lot), and (c) an industrial atmosphere (Bethlehem,
PA) (Allegra et al., 1985).
Table 2.30 Relative Performance of Thermally-Sprayed Zinc-Aluminum
Coatings: 34 Yearsa
Panels exhibiting 0% base metal rust
Standard Gun Gun 2 Gun 2
Coating on mild steel (air conveyed) (air conveyed) (gas conveyed)
90% Zn + 10% AI MP MP
80% Zn + 20% A1 AP
70% Zn + 30% AI AP
60% Zn + 40% AI AP
Zn + 50% AI
APb MP,b AP
40% Zn + 60% AI
APb MP MPb
30% Zn + AI
MP MP, AP MP, APb
20% Zn + 80% AI MP, AP MP, AP MP
10% Zn + 90% AI AP MP MP
90% AI + 10% Mg MP MP MP
80% AI + 20% Mg MP
0.08 mm AI + 0.08
SEP SEP SEP
0.08 mm Zn + 0.08 SEP SEP SEP
0.08 mm Zn
SEPb
0.15 mm Zn SEPb SEP SEP
0.08 mm A1 SEP SEP SEP
0.15 mm AI SEP SEP SEP
Panels exhibiting up to 30% base metal
rust + yellow rust stain
mm Zn
mm AI
90% Zn + 10% AI
80% Zn + 20% AI
70% Zn + 30% AI
60% Zn + 40% AI
50% Zn + AI
40% Zn + 60% A1
30% Zn + AI
20% Zn + 80% A1
10% Zn + 90% AI
0.08 mm Zn
0.13 mm Mn
0.20 mm Mn
90% Zn + 10% Mg
80% Zn + 20% Mg
80% AI + 20% Mg
60% A1 + 20% Zn +
20% Mg
MP, AP
MP, AP
MP, AP
MP
AP
SEP
MP
MP
AP
MP
MP, AP
AP
AP
AP
SEP
SEP
MP
MP
MP
MP
MP, AP
AP
AP
MP
AP
AP
AP
SEP
MP
MP
MP
MP
226
Table 2.30 Continued
227
Panels exhibiting up to 50-100% base
metal rust + yellow rust stain
Standard Gun Gun 2 Gun 2
Coating on mild steel (air conveyed) (air conveyed) (gas conveyed)
90% Zn + AI
80% Zn + 20% AI
70% Zn + AI
50% Zn + 50% AI
0.08 mm Zn
0.08 mm Mn
mm Mn
0.20 mm Mn
22% Zn + 78% Sn
90% Zn + Mg
80% Zn + 20% Mg
70% Zn + Mg
90% A1 + Mg
80% AI + 20% Mg
70% AI + Mg
60% Zn + 20% AI +
60% AI + 20% Zn +
20% Mg
20% Mg
AP
MP
SEP
SEP
SEP
MP, AP
AP
MP, AP
MP, AP
AP
AP
MP, AP
MP, AP
AP
MP
MP
SEP
SEP
MP,
AP
AP
MP, AP
AP
AP
MP, AP
MP, AP
AP
(AP-missing)
MP
MP
SEP
SEP
MP, AP
AP
AP
MP, AP
AP
AP
MP, AP
MP, AP
AP
"Key = MP, mixed powders; AP, alloy powder; SEP, single-element powder.
bMay have up to 5% area covered with yellow rust stain.
Source: Kain and Baker (1989).
improved over years ago. Kain and Baker also use electrochemical tests (in
seawater) to show differences in polarization resistance and potentiodynamic
polarization behavior, which can be indicative of expected performance.
Introduced by Vieille Montagne in Europe in the late the
alloy is near the top end of compositions that can readily be made into wire.
Hundreds of tons a year are used in France and Belgium, and this alloy is the
preferred thermally sprayed material in North America inasmuch as it is
designed as a multipurpose alloy to eliminate the need to choose between
aluminum and zinc.
The important point is that after 14 years of tests in various Japanese
sites (Nagasaka et al., and more than years in Europe, coatings
22% Chapter 2
150 pm thick are still in good visual condition and show no rust. Butler and
Morrow have reviewed all the available data. Assessment of the
performance of the zinc-aluminum alloys at intermediate stages in long-term
atmospheric tests is difficult because-unlike pure zinc-some of the oxida-
tion products that develop are retained within the coating, as mentioned
above, and are fully released only at failure-a situation that applies even
more significantly with sprayed aluminum. Hence change of weight gives
only semiquantitative information. Nevertheless, Figs. and
(Nagasaka et al., data presented by Butler and Morrow, which
give comparative data on weight change over years, are of considerable
interest. Both the milder light industrial atmosphere (Tokyo), where the zinc
corrosion rate is one-third that in the heavy industrial atmosphere, and the
rural atmosphere (Kanagawa) gave less change after years.
In the circumstances, it is therefore necessary to consider tests that do
bring about failure. The commonly used but unreliable 5% NaCl spray test
shows more than twice the time to red rust for Zn-15% AI cornpared to Zn. A
similar ratio applies in the special test developed by Vieille Montagne
(Leclerq, involving 10 ppm and RH: it was found that 50 ppm
+ 1% CO2 and RH at with 30-second water spray every
hours could destroy 100 pm zinc in 10 weeks, whereas 100 pm of zinc-15%
aluminum remained intact after 44 weeks (Sempels, undated).
Some short-term trials should also be mentioned because they include
comparison with other metallic coatings. B. A. Shaw and Moran
found that a zinc-15% aluminum pseudoalloy was best after 6 months of
atmospheric, splash, and spray exposure (compared with the pure metals and
duplex coatings of two metals), but for some unexplained reason there was
flaking and blistering of some of the prealloyed zinc-15% aluminum alloy.
It is normally recommended that thermally sprayed coatings be sealed to
avoid initial rust stains, to improve appearance, and to facilitate maintenance
painting. Sealing is designed to fill pores and give only a thin overall coating,
too thin to be directly measurable. Nagasaka et al. in addition to the
bare metal coatings, tested a wash primer + phthallic acid resin one coat
system and found this very effective. The Belgian tests (Table include
full duplex systems, although it is interesting to note that after years the
aluminum-vinyl sealed coatings were the best.
Zinc-Iron Alloy Coatings
The information on the corrosion resistance of the ZdFe alloy layers, either
alone or in a zinc plus zinc-iron alloy coating system, is slightly conflicting,
but the balance of data suggests increased corrosion resistance in acid atmo-
229
5
0
-5
-10
N
-
- 1 5
E
U
\
2
- 2 0
g
2
W - 2 5
J
,
I 2 3 4 5 6 9 10 I 1 12 Year
, """"
* - "" -.. ,
'L-
2.49 Weight change thermally sprayed coatings in a heavy coastaVindustria1
Japanese atmosphere. Key: 1 = sprayed AI; 2 = painting on sprayed AI; 3 = painting
on sprayed Zn-AI alloy; 4 = painting on galvanizing ; 5 = painting on sprayed Zn; 6
= sprayed Zn-A1 alloy; 7 = galvanizing; 8 = sprayed Zn; 9 = blasting only (Butler
and Morrow, 1990).
230 Chapter 2
I 2 3 4 5 6 7 8 9 IO 1 1 12 14 Year
5
0
E
\
E -5
C
c
-10
1:
-
9
2.50 Weight change of thermally sprayed coatings in a Japanese seacoast
atmosphere. Key: same as in Fig. 2.49 (Butler and MOKOW, 1990).
spheres but slightly reduced corrosion resistance in alkaline atmospheres,
compared with pure zinc. There has been extensive development work in the
Ukraine on such diffusion coatings of either the hot dip galvanized type or the
sherardized type (Proskurkin, and some data are given in Section
II1.B.
Hadden (1952) reported better corrosion resistance for alloy layers with
approximately 22% iron, and this finding was confirmed by Campbell, et al.
(1965) for coatings with more than 15% iron in industrial atmospheres. The
improvement, although significant in relation to the amount of zinc in the
coating, was attributable largely to the increase in coating thickness due to
diffusion during heat treatment, however. Coatings with only 10% iron gave
shorter lives. Hadden also noted that the visual rating of the coatings requires
the investigator to distinguish between the superficial yellow rust that devel-
ops quickly on the zinc-iron alloy and the red rust of the base metal. in
the annealed coatings, the appearance changes to a uniform dark-gray or
black. The early yellow appearance does not indicate rapid corrosion, but
shows only that the outer layer of the coating contains a greater proportion of
iron than is present in the coatings of unannealed wires.
Sjoukes (1976), at the request of the European General Galvanizers
Association, carried a literature survey. Eighteen articles had agreed that
the corrosion resistance of the zinc-iron alloy layer was usually better than
that of the zinc layer, but results in alkaline solutions were conflicting
(probably because differences in coating structure and slight changes in pH
and composition had a major effect on performance of zinc and zinc coatings
at pH 12-12.5). Thus it would be unwise to assume that the performance of
zinc-iron layers will necessarily be as good as or better than pure zinc in
alkaline atmosphereebut alkaline atmospheres rarely occur.
The often stated higher corrosion resistance in acid atmospheres is
mainly based on the formation of a dense 6-1 layer. If this layer is not dense
and is interspersed with other alloy layers, the corrosion resistance may not be
so high (Van Eijnsbergen, unpublished).
Work by Radeker and Friehe ( 1970) showed that unlike zinc, iron-zinc
alloy (hard zinc) initially gained in weight in German industrial atmo-
spheres. The behaviors of these types of coating were similar after some
years, however.
Sedloev (1970), in laboratory tests simulating an industrial atmosphere,
found that coatings annealed a! 500C give optimum homogeneity of the
phase, and iron had twice the corrosion resistance of unannealed
coatings.
In an industrial atmosphere, the best results were obtained with electro-
plated iron-zinc alloy layers with more than iron (Salt et al., 1965) with
a corrosion resistance higher than zinc. Elsewhere, zinc-iron alloy
galvanized coatings were as good as coatings with an outer zinc layer.
Sherardized coatings were superior to electroplated and equal to galvanized
for the same thickness. However, the structure of the alloy layer affects the
corrosion resistance, as does its composition.
Formation of brown areas on galvanized steel surfaces starts only if the
Table 2.31 Evaluation of Anticorrosion Performances Duplex Systems After Years
System Zn/Znl5-A1
number thickness (pm) 1 2 3 Paint coats Totalb
Paint system' Thickness (pm)
5 90 2P-EP 2P EP 2P-EP 20 - 80 - 30 200
6 90 CLR CLR CLR 20-80-40 210
7 90 2P-EP 2P-EP 2P-PU 20-2 X 40-30 200
8 90 2P-EP CLR CLR 20-80-40 210
9 90 2P-EP ST0 GLY 20-50-2X30 200
11 90 Pb-ST0 2P-PU 2P-PU 20 - 30 200
90 Pb-ST0 GLY GLY 40-40-40 190
13 60 AI-VIN 30 80
14 60 WP- 1 GLY 20 -
15 60 W- 1 80120 20 - 1 IO
15 60 80120
16 60 1 WDHPO 20- 125 185
60 WP- 1 SPDUP 20 - 40
18 60 WP- 1 WDACR 20-40
19 60 1 VIND 20-40
19' 60 VIND 40 90
20 60 ESZRP 20 70
"Key to paint systems:
2P-EFCT 2-pack epoxy-coalltar
CLR chlorinated rubber
GLY glycerophthalic
ST0 styrene oil
2P-EP 2-pack epoxy
2P-PU 2-pack polyurethane
Pb-ST0 metal lead-pigmented styrene oil
WP- 1 2-pack wash-primer (butyral, vinylic)
AI-VIN metal aluminum-pigmented vinylic
80/20 80% zinc dust/20% zinc oxide formulation
WDHPO water-dispersed high polymers
SPDUP specific duplex formulation
WDACR water-dispersed acrylic
VIND dispersed vinylic
ESZRP 2-pack ethyl silicate zinc-rich paint
bCalculated aimed total thickness accounts for penetration of first paint coat into surface roughness of
metallization.
232
of Outdoor Exposure
System
number Metal Industrial Marinelindustrial Urbadindustrial
in various applications
5
6
7
8
9
1 1
13
14
15
15'
16
17
18
19
19'
20
Zn ch
Zn 15 ch
Zn 6 ch
Zn 15 6 ch
Zn ch
Zn 15 ch
Zn 6 ch
Zn 15 6 ch
Zn 8 ch-8 cr
Zn 15 8 ch-7 cr
Zn 8 ch-6 b
Zn 15 8 ch
Zn
7 ch-5.5 p
Zn 15 A1
8 ch-9 p
Zn 15 AI
Zn 15 7 ch
Zn 15 6 ch
Zn 15 6 ch
Zn 15 9 ch
Zn 15 9 ch-9 b-9 p
Zn 15 8 ch
Zn 15
8 ch-8 p
Zn 15
8 ch-9.5 p
Zn 15 ch-8 e
3.5 ch
4.5 ch
9 ch-6 e
9 ch
7 ch
8 ch
9 ch-9 e-9 b
9 ch
9 ch-7 b-7.5 p
9 ch-9
0 ch-0 b-0 p
9 ch-9 b-9.5 p
8 ch-8 b- 3 p
9 ch-7 b-7 b
6 ch-8 b
7 ch
7 ch
IO
9 b
9 ch-9 b
9 ch-9 p
9 ch-9.5 e-9.5 p
ch-0 cr
7 ch
7 ch
9 ch
9 ch
9.5 ch
9.5 ch
9.5 ch
9.5 ch
9.5 ch
9.5 ch
9 ch
9.5 ch-5 b-9 p
9 ch-7 b-8 p
ch-7 b
8 ch-8 cr
7 ch
ch
10
9 ch
8 ch
8 ch
ch-0 cr-4.5 p
8 ch-9 p
'Values are cited only when lower than 10; an overall value of 10 indicates that a coating was undamaged
with respect to every criterion.
dKey to forms of damage: b. blistering; ch. chalking; cr. cracking; e, erosion; p. peeling. Evaluations on a
scale of 1 0 0, completely damaged; 10, undamaged.
Source: Leelerg and Sempels, undated.
q (zinc) layer has weathered or the Zn/Fe alloy layer ( 6 layer) forms the top
layer after galvanizing. Brown staining can occur on sherardized or hot dip
galvanized coatings (particularly those on silicon-killed steels) in atmospheric
corrosion through the oxidation of iron from the zinc-iron alloy layers or from
the substrate. Such staining usually is a dull brown, as distinct from the
brighter red-brown of uncontrolled rust. Often there is a substantial intact
galvanized layer underneath (e.g., see CEGB work: Bayliss et al., and
there is no need for any remedial action unless the aesthetic appearance is
unacceptable: the life of the coating is unchanged. One German article
(translated into English EGGA TECWRES 289) takes the view that because
233
234 Chapter 2
the brown staining occurs only after the phase (pure zinc) and the pure zinc
between the l, (zinc-iron) crystals had been consumed, and because the alloy
layers protect the basis steel, the brown staining was due to corrosion of the S
phase and/or the steel substrate.
The source of the iron that oxidizes to give the brown stain has been
studied by Sjoukes (1991), who concluded it is free iron in the galvanized
layer and arises from (a) the release of iron from the 5 phase during the
conversion, (b) iron present in the liquid zinc bath, and (c) the consump
tion of the zinc outer layer by the 5 phase, thus releasing iron during the
continued growth of the alloy layer while the component is hot.
This conclusion is not inconsistent with the factual German observa-
tions. If due consideration is given to the thermodynamics (A G curves)
(Mackowiak and Short, 1976) and to the different electrochemical situations
that arise, it seems that Sjoukes explanation is indeed plausible.
Van Eijnsbergen (personal communication) has reported on the brown
staining effect on galvanized lamp poles standing on a boulevard in Vlaar-
dingen in an extremely polluted area (Rijnmond). After 8 years, the coating
thickness of the ZdFe alloy layers was still 150-200 pm but, because of the
irregular distribution of free iron particles in the coating as well as local
differences in amount and duration of wetness, brown stains appeared in a
generally random pattern.
Cast Zinc Alloys
Zinc-4% aluminum alloy castings produced according to recognized specifi-
cations behave much like pure zinc. They are resistant to atmospheric corro-
sion, and usually more zinc can lost without affecting performance than is
the case with zinc coatings.
Higher aluminum contents (up to about 55%) will increase the general
corrosion resistance of zinc in test conditions that give continuous, general
attack on the surface, as demonstrated in a salt spray cabinet by Noranda
(1993) for alloy coatings on steel. The structure of the alloy, as well as the
percentage of aluminum, changes the salt spray resistance. Thus, while ZA-8
(zinc-8% aluminum) and ZA-12 (zinc-l2% aluminum) show only modest
improvements over the standard 4% aluminum alloy (with corrosion proceed-
ing through selective attack of the zinc-rich matrix), ZA-27 (zinc-27% alumi-
num), with its significantly larger amount of aluminum-rich phase, is much
superior. The relationship between the corrosion rates of pure zinc and the
zinc-aluminum alloys has been determined by Certified Alloys over 5 years in
an industrial atmosphere. This work (Table 2.32) is particularly interesting
because the firm deliberately included out-of-specification levels of lead,
"m"
8 8 8 8 8 5 8 8 8 8 8
M N - N Z m N N
g g Z 8 8 S 8 8 8 8 8
2"m""
~ ~ s s s s s s s s ?
N
235
236 Chapter 2
cadmium, and tin but found no clear adverse effect. Indeed, all the zinc-
aluminum alloys with 8-29% aluminum and with or without the deliberate
impurities corroded less than 99.99% zinc. Other trials in the humid atmo-
spheres of water purification plants have shown ZA alloys to be marginally
better than pure zinc, although both materials would last the normal life of
such plants.
Many plant atmospheres are contaminated by either acidic or alkaline
fumes. The ZA alloys can be expected to perform better than pure zinc in
acid-contaminated industrial plants. With alkaline contamination, both zinc
and the ZA alloys are more suitable than aluminum casting alloys. Controlled
trials made by Noranda at Kraft process pulp and paper mills (Table 2.33)
showed that all the ZA alloys (and zinc) were suitable for use in each of the
Table 2.33 Trials at Pulp and Paper Mills: Corrosion Rates at the Port
and Powell River Exposure Sites, Macmillan Bloedel, British Columbia: 3-Year
Results on Test Samples
Corrosion rate
Site
g/m2/year CLdYea
Sulfurous acid area: humid SHG
344
434
Bleach plant: humid, presence
SHG
of chloride and chlorate
133
Chemical preparation (bleach
SHG
plant): humid, presence of
4
chloride and chlorate
Recovery boiler building: SHG
outdoors, with flue gases,
precipitation of salts, ash 4
from recovery boiler stack
Paper machine: very humid, SHG 1.1
occasional water splashing 4
3 0.4
0.3
Source: Noranda
five specially contaminated areas selected for test and that the ZA alloys
showed lower weight losses than zinc. In the alkaline bleach plant, ZA-27
was inferior to the lower aluminum alloys. This finding was related to the
poorer performance of aluminum in alkalies (see Fig. 1.1 1) due to lack of
passivation.
At Canadian Electrolytic Zinc, Valleyfield, Quebec, the main pollutant
is sulfur-based gases. Three indoor and two outdoor sites are being tested, and
2-year results are given in Table 2.34 (except for the roaster, which has 5-100
ppm traces of and dust, but is dry and has a known lower corrosion
rate from tests on zinc-coated steels). Qualitatively, the results were compara-
ble to those in the acid areas of the pulp and paper plant.
The effect of copper in zinc alloys is to increase slightly the corrosion
resistance (Zinkberatung, 1975). Kehrer et al. (1982), however, show that
castings have predictable lower corrosion rates (and mechanical properties)
only if the copper content is above 0.7%, which is about the solubility limit of
Table 2.34 Corrosion Rates in the Canadian Electrolytic Zinc Refinery
Atmospheres: 2-Year Tests
Corrosion rate
Site Alloy g/mz/year pdyear
Cell housea indoor: humid,
H2S0, mist
Acid plant, outdoor:
H2S04 mist, SO,,
occasionally CO2 and CO
Cell house, outdoor:
H2S04 mist
Leach house, indoor: humid,
traces H2S04,
and NH,
SHG
ZA-8
ZA- 12
ZA-27
SHG
ZA-8
ZA-12
ZA-27
SHG
ZA-8
ZA- 12
ZA-27
SHG
ZA-8
ZA- 12
ZA-27
254
204
200
161
133
126
I13
68
32
33
31
17
25
17
13
35
28
28
22
18
17
16
9
5
4
2
3
2
2
0.5
"Rate of corrosion calculated from and 4-year tests, which showed a linear rate of' corrosion
with time.
Source: Noranda (1993).
238 Chapter 2
copper in zinc. Outdoor tests on sheet alloys containing 2% copper showed
20% decrease in weight loss (Gilbert, 1953), and this is likely to be the
maximum order of magnitude of the favorable effect.
The bulk of zinc in castings (compared with coatings on steel) makes
general loss of metal of little significance, but any solid corrosion products
that are formed may block valves capillary tubes. Where there is no regular
washing of the surface, attack can be uneven, forming pits. These can reduce
the mechanical properties more than expected from the average loss of
thickness. Some tests such as those by ASTM (1961) have, therefore, mea-
sured loss of strength with time: both AC41A and AG40A were tested, and
between 10 and 20 years of exposure, impact strength decreased rapidly for
total losses of 78 and 69%, respectively, in the outdoor industrial atmosphere,
and 33 and 38% in the outdoor rural atmosphere. Indoors, AC41A lost 52%
(unlike AG40A, which remained unchanged). These decreases in mechanical
properties were probably caused by intergranular corrosion, to which die
casting alloys produced 50 years more ago were often very susceptible.
Intergranular attack can reduce cross-sectional areas and create stress, raising
notches, while not reducing the overall specimen dimensions.
When certain impurities are present and segregate to give phases that
are very different electrochemically, corrosion will proceed rapidly along the
boundaries of the phases. Cadmium, tin, lead, indium, and thallium are
among the impurities that can be present unless controlled and are particularly
harmful, but iron and nickel also must be controlled at low levels. Magnesium
additions were developed early on as a beneficial addition for casting alloys,
although other additions can be used; for example, the 5% aluminum alloy
used for coating steel contains cerium and lanthanum.
Early die casters had great problems in avoiding brittle castings until the
cause of this defect was determined. Nowadays, the regular production of
99.99' % zinc gives die casters a good starting point, but to ensure that there
is no pick up of impurities, they must still practice very good housekeeping
and avoid metal purchases from unknown or less reputable sources. The
current high purities of zinc used to make the alloys have also enabled the
magnesium content of the alloys to be reduced to about 0.04% while still
preventing hot shortness of the alloys.
Inclusions of iron and lead may accelerate corrosion of zinc (Hubner et
al., 1978) but the most damaging impurity is aluminum, especially in con-
junction with certain other elements. Pelzel(l977) examined the relationships
between the presence of aluminum and several other elements and the corro-
sion of zinc. The base material for these tests was zinc with l . l% lead, to
which various elements were added. Tensile specimens were exposed to both
dry and moist air for 10 days at 95C; the resulting decline in tensile strength
indicated the severity of corrosion. This type of test is particularly sensitive to
intergranular attack, which reduces the effective specimen cross-sectional
area. Aluminum increased the corrosion of these specimens when added to the
basic zinc-lead alloy, reducing the strength by 25% in air and causing
complete loss of strength in steam. Copper, on the other hand, improved the
corrosion resistance of an alloy containing 0.03% AI. Titanium also improved
the performance of these alloys. Pelzel (1977) also exposed some zinc alloys
in an outdoor industrial environment for 5 years and measured the resulting
decrease in strength; beneficial effects of copper and titanium on zinc corro-
sion were again found.
Similar effects can be seen in zinc die casting alloys, which are inten-
tionally alloyed with as much as 4% aluminum. Kehrer (1974) investigated
the effects of tin and lead impurities on such alloys. His research was
especially concerned with the then-prevailing DIN- 1743 requirement limiting
the tin content of die casting alloys to 20 ppm and the combined lead plus
cadmium content to 90 ppm. Both tin and lead accelerated the general and
intergranular attack during 16 days of exposure at 95C and 100% relative
humidity.
The general and intergranular corrosion of most of the alloys tested was
reduced by copper additions, with the exception of alloys 13 and 17, which
had very high combined tin and lead contents. Outdoor, 2 year exposures
showed very similar results. In conclusion, Kehrer suggested reducing the
allowable tin content of die castings to ppm and the combined lead and
cadmium level to 60 ppm.
Roberts (1961) also investigated the effects of various elements on the
corrosion of zinc that contained aluminum. His data, based on exposures to
95C steam, indicate that magnesium reduces the intergranular attack on zinc-
low aluminum alloys. In contrast, lead, tin, cadmium, and bismuth all in-
crease the attack on zinc-aluminum alloys. Roberts notes that alloys of pure
zinc and magnesium, together with small amounts of lead or tin, suffer rapid
intergranular attack. He attributes this effect to the formation of magnesium-
lead and magnesium-tin precipitates, which are rapidly corroded.
These foregoing studies have shown that small quantities of alloying
elements can have considerable effects, both good and bad, on the corrosion
of zinc. The effect on zinc of combinations of foreign elements is, however,
quite complex and by no means completely understood. An excellent study of
the mechanisms by which impurity elements influence zinc corrosion was
carried out by Devillers and Niessen (1976a,b). Their work dealt with the
effects of additional elements in zinc-aluminum alloys. Although this report
centered on aqueous corrosion, its findings are relevant to atmospheric corro-
sion as well. The model treats the corrosion of grain boundary phases and the
240 Chapter 2
surrounding matrix in electrochemical terms. The precipitates and the matrix
are assumed to be in good electrical contact and to have approximately equal
areas exposed to the electrolyte that has collected on the exposed, corroding
grain boundary. This electrolyte is presumed to be very alkaline because of
the processes of hydrogen reduction and/or hydroxyl ion formation that occur
in the crevice between grains. The buildup of corrosion products restricts the
flow of electrons into the corroding region. Being conductors, the precipitates
and the zinc matrix all must assume the same corrosion potential at this
corroding grain boundary. In all likelihood, some phases will be anodic to
others at this equilibrium voltage. Aluminum precipitates would be anodic to
nearly any other phase present at a zinc grain boundary. Thus, aluminum
would be dissolving while reduction is occurring at the other phases. Conser-
vation of charge requires that the oxidation and reduction rates in this region
be equal. The authors present polarization curves that show the anodic
dissolution of aluminum and the cathodic reduction that occurs on other
metals in an alkaline electrolyte of pH 10.5.
The polarization results of Devillers and Niessen are used to explain the
observations that elements such as tin, lead, and cadmium accelerate grain
boundary attack in aluminum-containing zinc alloys and that copper mitigates
these harmful effects. The performance of magnesium is somewhat more
puzzling, but the authors explain the beneficial effect this element as
follows. Although magnesium has practically no room temperature solubility
in zinc, in aluminum it can dissolve to approximately 1.5%. Thus, small
amounts of magnesium may combine with the aluminum grain boundary
precipitates. Aluminum that contains 1.5% magnesium has lower anodic
currents than pure aluminum at the same potentials and would corrode less in
contact with other metals. Thus, the magnesium would make the aluminum
precipitates more resistant to corrosion.
When zinc castings are to be installed permanently out of doors, a
protective coating (e.g., paint or powder coating) should be applied to
maintain appearance. Chromating prevents white corrosion products from
forming in storage or in mild condensation conditions, but is not sufficient to
prevent loss of good appearance due to corrosion in long-term exposure.
Anodizing is very effective in marine conditions, including seawater immer-
sion.
G. Sheet Zinc
Comparatively few published data refer to the different grades of zinc sheet
used in roofing, but Pelzel(l977) has tabulated long-term data for coastal and
industrial exposures of zinc-0.058% lead and zinc-0.79% copper-0.093%
lead alloys, together with test results on titanium, aluminum, copper, and lead
additives at various levels. Radeker and Friehe (1970b) have reported that up
to 0.13% titanium can reduce corrosion by 40% in a 15-month exposure.
V. COMPARATIVE PERFORMANCE
A. General
The mechanism of degradation differs for different types of material, and it is
outside the scope of this book to try to compare zinc with plastics, concrete,
wood, or other nonmetals. Some comparisons with other metals can, how-
ever, usefully be made.
Other nonferrous metals are generally more expensive than zinc and are
used mainly when zinc does not give adequate performance because, for
example, its strength is too low or its appearance unacceptable, or because it
is permanently exposed to acids and acidic materials.
Aluminum costs more per ton than zinc but is of the same order of cost
per unit volume; hence the comparative corrosion resistance is often a critical
factor affecting the choice between these two metals. In general terms, zinc is
better in alkaline conditions and aluminum is better in acid conditions. Zinc
more readily provides sacrificial protection when used as a coating, and it
weathers more evenly than aluminum. Thus it can produce a rough layer of
corrosion product that more readily retains dirt because, unlike zinc, alumi-
num surfaces often cease to corrode after a few years. The continued dissolu-
tion of zinc helps to prevent build up of dirt but means that zinc coatings
normally have a finite life. Moreover, aluminum coatings often become inert
but may fail at the steel-aluminum interface in aggressive atmospheres as was
demonstrated by the British Iron and Steel Research Association (BISRA)
after about 20 years exposure at Sheffield. The AWS 19-year tests (1974)
and the results of Baumann 98 1) enable comparisons to be made between
zinc and aluminum, both sealed and unsealed; in general, either metal can be
used, but unsealed zinc is more readily corroded in acid environments.
Results in Sweden (Bottcher, 1990), given in Table 2.35, demonstrate the low
corrosivity of sites in the Nordic countries. Several comparisons between
steel, zinc, and other nonferrous metals have already been given. Table 2.36
gives a comparison for a wide range of South African atmospheres.
In mines that may contain gases liable to be ignited, causing an explo-
sion, aluminum and alloys containing more than 15% aluminum, which can
produce sparks on impact, are not suitable. ZA-12,ZA-8, alloys 3 and 5, and
pure zinc are all approved for use in mines.
242 Chapter 2
Table 2.35 Corrosion Rates in Sweden: Annual Rate Based on 4-Year Tests
Corrosion loss (pdyear)
Material on coating Rural Marine Urban Industrial
Hot dip aluminized, type 1 0.03 0.8 0.2 0.5
Hot dip aluminized, type 1 0.1 0.3
Hot dip aluminized, type 2 0.02 0.7 0.2 0.9
Hot dip aluminized, type 2 0.1 0.3
Aluminum sheet, type AllMn 1 0.04 0.2 0.2
Aluminum sheet, type AI/Mn I 0.3 0.3
Galvalume 0.1 0.4 0.2 0.5
Galvalume (shielded from rain)
0.3 0.2
Hot dip galvanized, G90 0.3 1.3 1 .o
Hot dip galvanized, G90 0.4 0.4
Hot dip galvanized, G165 0.4 1.9 1.1 4.6
Source: Bottcher (1990).
(shielded from rain)
Claims that weathering steels are equal to galvanized steel have proved
to be unjustified. Even when coated with paint systems, such steels do not
offer the supreme corrosion resistance duplex (i . e. , galvanized and over-
coated) materials.
Table 2.36 Atmospheric Corrosion Rates of Various Metals at Sites in South Africa
Corrosion (pdyear)
Durban, Cape
Metal' CSIR Island Bluff Docks Military Camp Steelworks
Mild steel 5.8 46.2 256.8 34.0 106.7 25.9
Zinc 0.330 2.464 12.141 3.26 82.575 3.175
Aluminum (3s) 0.025 0.686 2.184 0.432 0.457 0.737
Bronze 0.254 2.845 1.295 2.108
Pretoria Salisbury Durban Town Walvis Bay Pretoria
Copper 0.813 1.270 2.464 0.864 4.343 4.013
The corrosion rates of mild steel. zinc, aluminum, and copper are based on the results of 5 years of
exposure. The rates for bronze are based on 4 years of exposure at the same sites, except for Cape Town,
where the bronze metal was exposed at Ysterplaat, near Cape Town, not at the docks.
Source: Smuts and Callaghan, 1978.
Steel is cheaper than zinc but rusts readily in most atmospheres, causing
loss of strength and performance, while rust stains also affect surrounding
materials. Hence, zinc and aluminum are used as coatings anodes) to
protect steel. Some of the factors that affect the choice of zinc as a coating are
given in Figs. 1.6 and 1.7. Comparisons of zinc, mild steel, weathering steel,
copper, and aluminum are also made in IS0 9223-4, based on cooperative
tests (see Table 2.12). This work and the American work (Tables 2.8-2.10)
confirm that steel usually corrodes times as fast as zinc, the ratio
tending to be highest in high chloride atmospheres and lowest in acid indus-
trial atmospheres.
Within the family of zinc coatings (except zinc-rich paints), corrosion
rates are similar at all exposed surfaces, and the choice of coating is usually
made for other reasons (Fig. 1.7). For long-term exposure, the metal coating
choice for fabricated work is normally between hot dip galvanizing and
thermal spraying (with zinc, zinc-15% aluminum, or aluminum). Figures
2.30 and 2.31 show the choices in the British Standards Institution Code of
Practice BS 5493: 1977 as depicted by AM & S Europe (198 1).
As a guide, it is recommended to consider first the time required before
first maintenance and then choose a system from the alternatives offered for
that service environment. Next look at the notes relative to each component of
a protective system, and then review the product specifications. The full
document BS 5493 gives more details. This document is complemented by
practical experience. Particularly with paints that are not standardized, choice
of the generic group does not mean that all paints in that group are good
that they are necessarily compatible with other products that may individually
meet the requirements of another part of the protection system.
B. Economic Data for Coatings
Many protective coatings of varying degrees of sophistication and durability
can be used to protect steel. Design engineers must select a system that gives
the best practical protection at the lowest overall cost for the expected service
life. All too often attention has been given only to first costs, with the result
that the tremendous advantages in maintenance costs of metal coating systems
compared with traditional painting systems have largely been ignored. Also,
attention must be given to consistency in application and to the performance
reliability of the coating system, including surface treatment, circumstances
during application, environmental laws, and application methods.
Most of the detailed information relates to surveys of about 20 leading
protective coating firms in England every few years since 1962, covering
galvanizing, metal spraying, and painting of products. The protective
schemes considered are summarized in Table 2.37. The references are to
244 Chapter 2
Table 2.37 Description and Type of Protective Scheme
Protective
scheme
Type
BS 5493
description
(see Fig. 2.31)
l
2
4
5
6
7
8
9
10
11
12
13
Pickle and hot dip galvanize (85 pm minimum)
Grit-blast to Sa2%, pickle and hot dip galvanize
Silicon steel; pickle and hot dip galvanize to 210
Pickle and hot dip galvanize (85 pm minimum),
to 140 pm nominal
pm nominal
followed by phosphating and high build
chlorinated rubber to 60 pm minimum
Grit-blast to Sa2%, aluminum hand spray to BS
2569 Part 1 Grade A1 4 (100 pm aluminum)
Grit-blast to Sa2%, zinc hand-spray to BS 2569
Part 1 Grade Zn 4 (100 pm nominal)
Grit-blast to Sa2%, aluminum hand-spray to BS
2569 Grade A1 6 (150 pm nominal)
Grit-blast to Sa2%, aluminum hand-spray to BS
2569 Grade AI 6 (150 pm nominal)
Grit-blast to Sa2%, zinc hand-spray to BS 2569
Part 1 Grade Zn 4 (100 pm nominal thickness),
followed by one coat of vinyl sealer
Grit-blast to Sa2%, apply drying oil paint system;
two coats zinc phosphate primer applied by
brush followed by two coats incorporating MI 0
pigments by airless spraying (total nominal
thickness 165-190 pm)
Grit-blast to Sa2%; apply organic zinc-rich system
by airless spraying three coats (total nominal
thickness of 300 pm)
Grit-blast to apply three coats of high build
chlorinated rubber by airless spraying (total
nominal thickness 300 pm)
Grit-blast to Sa2%; apply three coats two-pack
chemical-resistant paint by airless spraying
(total nominal thickness 270 pm)
SB 1
SB2
SB3
SB9
SClA
SClZ
SC2A
s c 3 z
s c 5 z
SF7
SD5
SH7
SK3
Resistance Atmospheric Corrosion 245
system designations in BS (see Fig. Duplex systems were
not included because in only a few cases did the same firm apply both metal
coating and paint. Table indexes the average costs per square meter for
several different types of steelwork. Cost data are also given in Figs. and
for the United Kingdom. The relative costs elsewhere are likely to be
similar, except that labor-intensive schemes such as painting will be relatively
cheaper where wages are low.
The general cost patterns in this and earlier surveys are similar. Hot dip
galvanizing is the least expensive of the protective systems considered when
the steel's ratio of surface area to weight is greater than m2/ton, and a
standard sprayed zinc or aluminum coating remains less expensive than a
multicoat paint system. The surveys, therefore, invalidate a popular miscon-
ception, namely, that metal coatings are always expensive. It is also shown
that zinc spraying remains marginally less expensive than aluminum spraying.
Table 2.38 Relative of Protective Schemes for Different Types of Steelwork
Small Box girder
assorted Heavier Plate steelwork
sheet metal Structural structural girder (outside
Protective Average
scheme thickness,
(see Table mm: 5 9 20 33 30
2.37) m'lton: 55 30 14 8 9
items steelwork steelwork steelwork only)
1
2
3
4
5
6
7
8
9
11
12
13
4.11
5.85
6.24b
7.25
7.87
7.53
9.25
14.36
9.18
9.33
8.05
8.45
8.09
3.90 7.29 11.00
5.93 11.43 17.63
4.66 9.Wb 14.00
8.03 10.07 16.3gb
8.00 8.07 8.38
7.60 7.64 8.00
9.97 9.86 10.75
14.30 13.79 13.88
9.07 9.00 10.13
9.43 9.93 11.13
8.07 8.64 9.75
7.81 9.00 10.25
8.03 8.50 9.88
7.44
7.22
9.44
13.67
10.00
11.22
9.89
10.33
9.78
'In pounds sterling per square meter in the United Kingdom in 1984-1985.
%nly three firms quoted, otherwise, figures are for an average of six firms.
246 Chapter 2
The cost of a four-coat traditional paint system now exceeds the costs of
the organic zinc-rich system, the chlorinated rubber two-pack, and the
chemical-resistant three-coat paint systems for each type of steelwork consid-
ered. The reason for change in the relative costs of the drying-oil-type, four-
coat paint system is not certain, but the results may reflect the very high
increase in labor charges over recent years. A high percentage of painting
costs goes to labor charges, which increase with the number of paint coats.
The increase in relative costs painting compared with galvanizing
over the years is due to labor costs rising faster than material costs: painting is
highly labor intensive.
New comparative calculations by Van Eijnsbergen in discussion show
that costs of galvanizing per square meter is roughly half the cost for a three-
coat paint system on grit-blasted steel when the steel construction has more
than m2/ton. For heavy constructions (< m2/ton) galvanizing costs are
more or less equal with the costs of a good paint system (e. g., grit-blasting to
Sa 2% epoxy resin based primer and sealer, and a polyurethane top coat). It
is emphasized that galvanizing costs do not include extra work before or after
galvanizing, nor transportation costs. Paint costs exclude scaffolding, work
delays, repairs on the building site, transportation, and environmental mea-
sures on and around the object to be painted.
Duplex systems cost approximately to more per square meter
than galvanizing (see Table depending on the number of paint coats to
be applied: sometimes one coat, mostly two coats; rarely three coats). Of
course, such initial costs are dwarfed when one considers costs per square
meter per year use andlor durability. Initial costs of protective systems are
also included in publications initially issued jointly by British Steel Corpora-
tion (BSc), British Constructional Steelwork Association (BCSA), Paint
Research Association (PRA), and Zinc Development Association (ZDA) (see
BSc, and with later issues by BSc, which include some simpler
paint systems and duplex systems. Because simple one- or two-coat paint
systems can be used over galvanized steel, the combined cost of galvanizing
and painting is often no more initially than that of sophisticated paint systems,
and freedom from rusting is more assured. Figure is from the publication
relating to exterior environments; Fig. relates to building interiors. Other
leaflets apply to perimeter walls (where two coats of heavy-duty bitumen may
be needed over the galvanizing on the outer leaf in contact with fill material
at an extra cost about 50%) and to refurbishment (mainly paints pigmented
with zinc or zinc phosphate).
In other industrial countries, the relative first costs are qualitatively
similar but, since paint coating involves the greatest percentage of labor costs,
painting will be more expensive in countries with high wage rates but lower in
countries such as India, where wages are low. Material costs vary much less
from country to country.
Maintenance costs for coatings comprise (a) the direct cost of surface
preparation, painting materials, and labor, (b) the cost of access or of return-
ing items to plants for treatment, and (c) the indirect costs due to interruption
of service (to close a typical power transmission line to paint the top cost at
least $18,000/day in 1993).
Surveys made by Zinc Development Association between 1964 and
1972 included an analytical buildup of the direct costs. These show that
provided maintenance is undertaken at the correct time, the direct cost of
repainting is similar over all initial treatments, varying only slightly to reflect
the different costs of the paints involved. Maintenance costs per year are,
however, very different because with galvanizing and metal spraying (unlike
paint systems) the time to first maintenance is usually much longer than with
paint systems. Moreover, intervals between maintenance painting also can be
extended, since lifting of the whole coating through rust spreading sideways
from edges scratches does not normally occur.
The effect of the foregoing properties is to increase still further the
economic advantages of metal coatings compared with paint coatings. No
figures are given because these direct cost advantages are small compared
with the costs of access and of interruption of services-both of which vary
with each example and translate to a greater economic advantage for systems
involving the fewest maintenance treatments during the life of the product or
structure.
The traditional approach has been first to obtain the comparative esti-
mated costs of applying a protective scheme and (if possible) to add to this
amount the expected costs of maintenance over the life of the structure. These
costs are then usually divided by the number of years of expected life to give
an average annual cost. Such an approach ignores basic economic principles
(Brace and Porter, 1968). The essential fact to be grasped is that money has a
time value-a given sum of money available now is worth more than an equal
l sum at some future date (even ignoring possible inflation) because of the use
~ that can be made of the money in the intervening period. In formal economic
terms, a sum of money P now is considered to have the same value as the sum
P( 1 + r) one year from now, where r is the interest (or discount rate) for a
given category of money expressed in percent per annum. Then, after n years,
the value of P is P(l + r)rt.
For most countries other than Britain, surveys have been more limited
in scope, but particular mention must be made of those in Germany [Blohm
248 Chapter 2
(Ed.), 1978, and particularly the papers by Kleingarn and Schliebaum], in the
United States (Kinstler, 1975, Roebuck and Brevoort, 1986; Morrow, 1987),
and in Australia (Galvanizers Association of Australia 1985). The Zinc
Institute reports a 16-year exposure program on ungalvanized and galvanized
guard rail in Florida, where savings per mile over the total period of $64,152
were obtained by using galvanized steel, which needed no maintenance.
Cost comparisons involving galvanizing plus painting are not readily
available because of the long life of the combined coating, but Seils (1969)
and Seiter (1969) have provided useful data, particularly related to the use of
such duplex coatings on transmission towers. Calculations made in 1991 (Van
Einjsbergen, personal communication) for duplex systems in the Netherlands
consisting of sweepblasting, followed by an epoxy primer (30 pm) and a
micaceous iron pigmented polyurethane top coat (50 pm), shows prices of
$20-$30/m2, including galvanizing (by the same company). Galvanizing plus
one coat of polyester powder costs $22-$26/m2. If an epoxy powder coat is
applied as a primer underneath the top coat, costs amount to $27-$32/m2.
Qualitatively, similar arguments apply to the use of steel precoated with zinc:
longer time to first maintenance and greater times/tolerances in periods
between maintenance.
Gogler (1987) compared zinc coatings alone (see Table 2.1) with
duplex systems and showed that-calculated in costs per square meter per
year of average durability-the latter are by far the most economical. With
light gage steel, galvanizing is nowadays even more economic than good
paint systems on steel, without taking durability into account.
Cost comparisons have also been made between galvanized steel and
other construction materials. A survey by Steven Winter Associates (1980)
showed that it was up to 19% cheaper to use galvanized steel framing instead
of wood for house units (depending on location) in a year in which wood was
cheap-its cost fluctuates far more than that of galvanized steel. British Steel
Corporation in 1973 showed the advantage of galvanized steel compared with
aluminum for ISO-type containers.
VI. CASE HISTORIES
Evidence from practice is most important in considering the corrosion behav-
ior of zinc and zinc coatings. The most significant data relate to zinc coatings
because they are thin enough to have a finite layer, and satisfactory perfor-
mance, after even a limited number of years, is significant. Table 2.39 lists
some documented case histories for urban, industrial, marine, rural, and
tropical atmospheres.
The following examples illustrate the benefits of zinc or duplex coat-
ings, including such aspects not directly related to corrosion resistance as
Table 2.39 Case Histories Galvanized Steel Atmospheric Exposure
in the United Kingdom: More than 15 Years in Service
Description
Edinburgh Tattoo stands
Imperial War Museum
Holyhead landing pontoon
Big Ben roof
Immingham conveyor
Gloucester Westgate bridge
Walton Bridge
Whitbread Brewery, Samlesbury
Dairies at Appleby and Lockerbie
Bass Charrington brewery effluent, Runcorn
Bletchley Sports Complex structure
Crystal Palace telecommunications tower
Litchfield TV mast
Central Electricity Generating Board transmission towers, west country survey
Shell-Mex building clock hands
Queen Elizabeth grandstand, Ascot
Lords Werner stand seat frames
"GA" A19 Teesside Parkway bridge beams
Cullompton bridge beams
Farnham sewage works
Wall ties, BRE report
Luton parking lot
Gravesend, Tilbury tunnel cable trays
Mansfield brewery
Albert Dock Village, Liverpool
Phillips Seal Sands pipe bridge
Dolphinarium at Woburn Abbey
Swaine Marine steel Gipsy boat
Merry-go-round wagons
avoidance of rust staining, ease of access for coating, inaccessibility (or high
cost of accessibility) for maintenance, volume of corrosion products, spalling,
frictional characteristics, abrasion resistance, appearance, and hygiene.
A. Demountable Grandstands for the Edinburgh
Military Tattoo
Until 1975 the supporting steelwork for the grandstands of the Edinburgh
Military Tattoo had been standard scaffolding done by contractors each year.
250 Chapter 2
decision was then made by the City of Edinburgh District Council to adopt
a specially designed demountable structure, which increased the capacity to
more than spectators and enabled faster erection and dismantling for a
more cost-effective annual operation. Greater speed of erection is obtained
because of the reliable fit of the galvanized surfaces, and no rust occurs to
cause joints to seize up while the stands are in use.
During the annual erection and dismantling, the individual small pieces
of the 350-ton structure inevitably receive rough handling, which would
damage most protective coatings. The galvanized coating, however, is metal-
lurgically bonded to the basis steel and has a high resistance to damage as a
result of its unique layered structure. direct blow is cushioned by the
relatively softer outer layer of zinc but, should the outer layers become
removed, the residual exposed alloy layers, which are harder than the basis
steel, will resist further abrasion and continue to protect the steel.
The outline design was prepared by the city architect and the detail
developed in collaboration with BICC Construction Ltd., based on the Mero
space frame system, which consists of prefabricated tubular steel components
screwed into forged steel spherical connectors. The grade 43C steel tubes,
which vary in diameter from 48 mm to mm for the high load bearing
members, were hot dip galvanized to BS 729 for protection against rusting
during the short festival season and the 9 months of storage. Both internal and
external surfaces can be protected because, unlike protective systems that rely
on brush or spray application, the molten zinc can reach internal surfaces.
Cost estimates indicated that there would be economic benefits if the structure
lasted more than 5 years. Fourteen years later, in the official responsi-
ble for the annual erection and dismantling reported that the tubes were giving
good service and showing no sign of rust. The nodes, which are not hot dip
galvanized, were showing first signs of rust. Maintenance of these compo-
nents will eventually take place to ensure that the benefits of easy assembly
and disassembly, which are a key technical feature of the galvanized surfaces,
are retained for the structure as a whole. No maintenance of the galvanized
steel is envisaged for the foreseeable future.
New Dome for the Imperial War Museum
The cupola of the Imperial War Museum in London was destroyed by
arsonists about 25 years ago, and the reconstruction was commissioned by the
U.K. Department of the Environment. The structure for the cupola would be
inaccessible for maintenance painting, and hot dip galvanizing was chosen to
give the longest possible maintenance-free life at low cost. To make doubly
sure that expensive maintenance was eliminated, a coating about 150 km
thick was attained by grit-blasting the 40 tons of steelwork before galvaniz-
ing.
Galvanized high strength friction grip bolts were specified for all con-
nections, and the contact surfaces were wire brushed after galvanizing to
increase the slip factor. This roughening of galvanized faying surfaces is
necessary only when even a very small amount of slip is intolerable: the initial
coefficient of friction between as-galvanized coatings is low but builds up
rapidly as cold welding or lock-up occurs in the joint.
After nearly 20 years, the official responsible for maintenance at the
War Museum reported that the structure was in perfect condition, just covered
by dust. No need for maintenance in the foreseeable future is envisaged. It
should be noted that the roof covering is copper but, because the contacts
between the galvanized steel and the copper are in the dry and there is no
water present to provide an electrolytic path joining the two metals, bimetallic
corrosion does not occur (see British Standard PD 6484).
C. Whitbread Brewery
When Whitbreads Northern Brewery was constructed at Samlesbury near
Preston in 1972, hot dip galvanizing to BS 729 was specified-not only to
protect the structural steelwork in the brewhouse, bottling, and warehousing
areas, but also to protect the exposed 10 m high box columns, which were an
architectural feature of the production buildings. The use of ribbed galvanized
sheet cladding panels overcoated with a brown organic finish also succeeded
in blending the project into the landscape.
Galvanized steel has a proven record of hygienic acceptability in
dairies, brewhouses, and similar food-processing areas (Table 2.39). It can
also be used in contact with dry nonacid foodstuffs (e.g., as baking trays).
In this Whitbread brewhouse, it is estimated that the coating will have a
life of more than 30 years. A detailed examination after more than 8 years
showed that the spangled appearance was still evident on the galvanized 43A
steel castellated roof beams and universal columns, with more than pm
coating thickness remaining on the webs and more than 150 pm on the
flanges, well exceeding the minimum 85 pm required on new work according
to BS 729.
The featured external galvanized 43B steel hollow section columns had
weathered to a matte gray, and the zinc-iron alloy coating on the bracing
members had a brown-gray appearance, which blended with the cladding
panels. The minimum remaining coating thickness measured was pm
after 8 years.
The exterior was painted after 14 years (in 1986) to freshen the aesthetic
252 Chapter 2
appearance and to obtain the benefits in future (including longer life between
repaintings) of a duplex coating of zinc and paint for structure as well as for
sheet.
The success of the galvanized steel construction has been confirmed by
its choice for other projects within the Whitbread Group, such as the brewery
at Magor, South Wales.
D. Spectral Reliance Building, Greenbridge Industrial Estate,
The Spectral Reliance Building, a single-story, industrial steel structure, was
built in using profiled plastic-coated galvanized steel for the walls. It
received the London Times Industrial Award in The building was
designed by Team 4, which subsequently split up into two firms-Richard
Rogers and Foster Associatectwo of the leading English firms of architects.
The building has appeared in many publications and journals over the years by
way of illustration of a thoughtful and appropriate industrial building form.
Externally, the exposed ungalvanized steel frame is painted every 3
years, but the vertical cladding had received no maintenance when examined
in detail in and was in an excellent state of repair after 15 years of
service. At one or two places where windows or other new inserts have been
placed and new cladding sheets have been installed, there is some difference
in color due to the fading of the original material-but this change is notice-
able only when the old material lies adjacent to the new. In some
external air conditioning units were fixed to the steel cladding. An acetylene
flame cutter was used, starting a fire in the internal polystyrene insulation,
which in turn peeled small sections of the plastic coating. This damage has
not been repaired, and the galvanized substrate has protected the underlying
steel from corrosion.
Bird droppings and odd spots of paint have been cleaned off the steel
cladding with acetone. This has taken the sheen and texture off the plastic
coating but has not destroyed it. The blemishes are, however, quite notice-
able. The resident engineer in felt that the best policy was to use white
spirit or to actually leave spots paint.
It is important to note that the available plastic coatings are now much
improved in all these respects-indeed, the specific plastic coating used in
is no longer offered.
E. Houses of Parliament, London
Sir Benjamin Hall was a wise Victorian who, as Commissioner of Works for
the Houses Parliament, specified a galvanized coating to protect the special
cast iron roof tiles (each weighing more than kg) put on the clock tower in
Swindon
The biggest bell of all in the clock tower, Big Ben, was named after Sir
Benjamin himself.
At the time of the London blitz during World War 11, some tiles were
removed for safety reasons. More than a decade later, in the late the
hot dip galvanized tiles were cleaned and a coat of metallic lead paint applied
in situ over the galvanized coating, to add protection to areas where the
original coating had weathered and also to give a uniform appearance.
A further examination in showed some areas bereft of zinc,
although about of the roof surface had some zinc present. Because roof
maintenance normally occurs only every years, it was decided to renew the
zinc and, to facilitate on-site work, zinc spraying pm) was specified,
followed by sealing with a dark gray vinyl copolymer sealer. Five years later,
the roof was in perfect condition, and the same treatment was proposed for the
Victoria Tower at the other end of the Houses of Parliament, which was
erected in and has a similar history.
Netherlands, Germany, Scandinavia, and Canada
A large collection of case histories of galvanized objects as well as those
protected by a duplex system has been made by Van Eijnsbergen during the
past three decades. In a number of cases, micrographs have been made and in
many cases coating thickness has been measured. A selection of striking
examples of the durability of hot dip galvanized steel construction, both in
the Netherlands and elsewhere follows. Details of these case histories have
been published in the Dutch-German magazine Thermisch Verzinken-
Feuerverzinken.
Galvanized Steel Dike at Delfiijl
Built in the dike at Delfzijl was painted green (epoxy coating) to match
the grass. The wall protects the town against flooding by the sea. An
inspection made in showed the zinc coating to be still pm thick.
No rust had appeared after years.
Suspension Cable of Rhine Bridge near Rodenkirchen, Germany
After years of service over the Rhine River, the hot dip galvanized coating
underneath the paint system protecting the bridges suspension cables showed
only local small traces of rust. The paint system itself had largely weathered
away.
Dutch Tramways
Galvanized tubular sleepers on one of the Rotterdam tramway lines, situated
in a heavily polluted industrial area, were free of rust after years. Also,
254 Chapter 2
these galvanized rods suffer from occasional impact by stones from the
railbed.
Mobility Road Tracks
Galvanized perforated road strips, used by the army and other authorities to
prevent sinking of trucks in weak soils and mud, have been in use for more
than years. They not only are exposed to the weather but also must
withstand rough handling and severe impact. No rust has weakened the steel
during this period.
Pilot Boats
Two pilot vessels of the Dutch Coast Guard launched in (Rog) and
(Bruinvis) were constructed with galvanized hulls and galvanized deckhouse
(Rog), railings, mooring heads, stanchions, and so on, and painted. When
examined in both the Rog and the Bruinvis were in excellent condition
and still in use on the River Schelde to guide vessels to Flushing, Terneuzen,
or Antwerp. Both chlorinated rubber based products and (later) epoxy resin
based paints were applied during the annual maintenance in dry dock, either
as a touch-up or as new coats of paint. On areas like bulwarks and the hull,
especially in the bow area, where heavy mechanical wear occurs, the paint
layers often have worn off. However, the underlying zinc coating system,
alloyed to the steel base, remains completely or almost fully intact. On both
ship hulls, zinc anodes are mounted for extra cathodic protection.
Turnstile
A galvanized turnstile in the dunes around Katwijk, Holland, showed no rust
after years of exposure near the North Sea, including corrosion by flying
sand particles and occasional chipping by bicyclists passing through.
Bayer Cross
The huge galvanized Bayer Cross on a building near the Cologne-Diisseldorf
motorway was galvanized in and, after 16 years in an industrial atmo-
sphere, there had not appeared a spot of rust.
Road Barriers
To block bicyclists on a pedestrian entrance of a Dutch market square in the
Hague, a system of two galvanized fences was erected. After 15 years of use,
constantly exposed to the weather and to bumps and chipping baby
carriages, trolleys, and bicycles, the zinc layer had weathered away. How-
ever, the very hard and erosion-resistant alloy layers were completely intact.
There was still a 50-60 pm alloy layer left (100-150 pm on the fittings),
offering excellent protection without any rust.
Fence - Corrosion + Abrasion Resistance
Excellent corrosion and erosion protection have been obtained from a gal-
vanized fence around a marketplace in the Hague. After 15 years of wear by
bicycles, carriages, shopping carts, and so on, only the zinc layer had suf-
fered. The Zn/Fe alloy layers remained intact.
Gold Prospectors Equipment
Among the Chikoot Trail in Yukon Territory, Canada, T. Jore from Oslo
discovered two galvanized items, a washing tub and a piece of cable, left by
gold prospectors some 90 years ago. Rust was negligible on these remnants of
the Gold Rush. On the tub, even zinc flowers (crystals) were still visible.
In Rudsvoll, Norway, gold was found in the beginning of this century.
Galvanized constructions for ore washing and crushing were delivered but
never used after 1907 because of the dwindling supply of gold ore. The parts
were left, and grass overgrew the heap of material. After 80 years, the zinc
layer was still intact, and it was estimated that the zinc would last for another
80 years in this rural climate.
Ruhr Shed
A completely galvanized large shed in the German Ruhrgebeit showed no
signs of rust after 20 years of exposure in a very aggressive climate. The zinc
coating still was 80 pm thick.
Road Signs
A road sign, 160 km north of Stockholm, Sweden, was galvanized and
erected in 1927. After 55 years of exposure in a rural climate, the zinc coating
was still 15-25 pm thick and no rust had occurred even on the cut edges of the
sign, where the zinc layer is absent but the alloy layers still present.
C. Additional Applications Worldwide
Towers
Galvanizing has seen common use in the power industry for over 50 years.
The galvanized lattice tower or galvanized substation structure has been in use
in the United States by the Tennessee Valley Authority since the mid-1930s.
At the Clinch River substation, completed in 1936, galvanizing has provided
complete corrosion protection of all steel structures. A recent inspection
256 Chapter 2
indicated first maintenance would be required in the year This means
that the Tennessee Valley Authority will have 70 years of service with no
maintenance costs. This galvanized substation has been the standard in the
power industry. Galvanizing has proved to be the maintenance-free coating.
Power companies are now considering the regalvanizing of substations as a
cost-effective means of corrosion protection.
Two televisiodradio towers in the Netherlands were built and gal-
vanized in After years they were painted with a polyurethane-based
system. In when both towers were inspected, there was still no need to
recondition this duplex system. One tower stands near Utrecht in a mildly
industrial atmosphere. The other stands near Groningen, in the northern part
of the country, in a light marine-industrial climate.
Railway Overhead Structures
German State Railways also used galvanized overhead structures. On the
Fulda-Bad Hersfeld line, the zinc layer, which originally exceeded 100 pm,
had a thickness of pm after years of weathering. There was no rust
anywhere on this line.
Telegraph Poles, Central Australia
Probably one of the oldest galvanized items was discovered by Van Eijns-
bergen during one of his initial visits to central Australia. The galvanized
telegraph poles at the telegraph station in Alice Springs, Northern Territory,
are part of the overland Adelaide-Darwin line. The poles are more
than years old and in this arid climate have not weathered at all.
Animal Slaughterhouses
In the municipal slaughterhouse of the Hague, Kraak (Dutch Galvanizing
Institute) discovered a number of hot dip galvanized profiles that had been
galvanized before Between and they were overpainted;
however, the paint had largely disappeared in After some years,
micrographs showed that the zinc coating was still pm thick on the
profiles, and in the gaps between two rivets the zinc still protected these areas
from rust.
Advertising Tower
On the beach of Scheveningen, Holland, a tubular tower was erected in 1964
to serve as a base for advertisements. Originally, galvanizing results in a
coating thickness of pm. After years of weathering on the beach,
the coating was still pm thick. Rust stains from barbed wire, used on
the area near the ground to prevent children from climbing, are the only
visible sign of corrosion.
Zinc Cemetery Markers
On a holiday tour in France in H. Swart discovered on a heap of dirt an
old discarded grave marker made of rolled zinc. On this zinc plate were
inscribed the name of the deceased, his age, and the date of his death:
After more than 100 years there is no sign of corrosion. Zinc obviously can
survive man!
Other Items
The Silver Jubilee issue of Thermisch Verzinken in published eight
striking examples of the protective value of galvanized steel: (a) the overhead
structure of the Schiedam-Hook of Holland line years old), (b) a transmis-
sion tower near Arnhem years old), (c) the Swedish Manor House near
Stockholm years old), (d) a Japanese transmission tower of the Tokyo-
Inawashiro line years old), (e) a floor grid in a Dutch power plant
years old), a roof ornament of a greenhouse in Holland years old), (g) a
connecting strip in a Dutch greenhouse years old), and (h) a railing in a
water purification plant, also in Holland years old).
Dutch Vegetable Market
Many Dutch auction halls for vegetables and flowers are galvanized. One
example reported by Kraak showed no rust after more than two
decades. No maintenance had been required. The coating thickness on this
building, which is exposed to an industrial-marine climate, remained between
140 and pm.
Bridges
The Callender-Hamilton type of multitruss bridge has been specified for
galvanizing since its conception in the In addition to the galvanizing of
the structural steel sections, all the large-diameter, high tensile bolts are
galvanized to produce a high frictional bolting effect. The absence of rust
assists assembly and disassembly and facilitates the long-term storage of these
standard components, to be ready for use in an emergency. Many of the
similar Bailey bridges are also specified for galvanizing.
In the Netherlands, a number of galvanized Callender-Hamilton road
bridges were bought from the British government immediately after World
War 11, to replace the bridges destroyed. After some years, the steel
showed no rust. When the paint, which had been applied in the was
scraped off, the zinc coating looked "like new."
Bridges up to m long have been fully galvanized. The first known
fully galvanized bridge in North America was built in Quebec, Canada: the
m Lizotte Bridge. Galvanizing has provided cost-effective corrosion
258 Chapter 2
protection and maintenance-free service for the Lizotte Bridge since 1963.
Fully galvanized bridges are located throughout Canada, Michigan, Ken-
tucky, Illinois, Connecticut, and Indiana, and in the salt-laden atmosphere of
Bermuda. The Kinki Expressway in Japan, now 15 years old, is a 1.5-mile,
30 m wide bridge utilizing 3000 tons of galvanized steel for long-term cor-
rosion protection.
At the time the economic analysis was conducted, the first cost of
galvanizing the Lizotte Bridge exceeded the first cost of painting (with a
5-year system) by about $1 1,OOO. Overall, galvanizing added about 3% to the
total first cost. However, the expected life of the galvanized coating was at
least years. Project engineers regarded paint as a losing battle. While it
was true that paint provides some corrosion protection for bridges, paint
system failures in difficult-to-reach and hard-to-inspect places had caused
considerable corrosion damage and related repairhehabilitation expenses on
comparable bridges. It was clear that if the galvanized coating performed at or
near expectations, it would save a great deal of money in both real and
nominal dollars compared with painting.
Based on the highly favorable results of the comprehensive engineering
analysis just summarized, the decision was made to galvanize the Lizotte
Bridge. Several different steels were used in the fabrication of bridge compo-
nents. Bethlehem V-45 and Stelco A44 were used for the main members and
ASTM A36 for the secondary members. Shop connections were welded, and
field assemblies were jointed with 1 in. diameter high strength bolts. In
addition to bridge members, all other steel bridge elements were galvanized:
railings, bolts, expansion joints, curb, anchor bolts, anchor straps for side-
walks, and on.
Galvanizing has performed as well as, or better than, originally pre-
dicted. The Lizotte Bridge has been maintenance-free since 1963. As of 1989,
the thickness of the galvanized coating still exceeded the thickness specifica-
tions of the ASTM, the American Association of State Highway and Traffic
Officials, and the Canadian Standards Association (CSA). It is clear that the
life of the galvanized coating on the Lizotte Bridge has met and will exceed
the initial 30-year estimate. All other galvanized bridges constructed in
Canada during this period have exhibited similar excellent performance.
Major bridges such as those over the Forth and Severn estuaries com-
prise steel sections too large for immersion in the available baths of molten
zinc (about 25 m is the maximum length normally galvanized) and these are
zinc sprayed, sealed, and painted. The need for a zinc coating is demonstrated
by experience with the Pierre Laporte Bridge over the St. Lawrence River,
near Quebec City, Canada. The bridge was opened in 1970, but the mainte-
nance cost of the original paint system over 6 years was so great that it was
decided to zinc spray the structural girder work. Grit-blasting down to bare
metal was followed by pm of sprayed zinc and two coats of vinyl resin
sealer. This is probably the worlds largest steel structure sprayed in situ.
There are thousands of other bridge applications throughout North
America. The State of Ohio alone has more than bridges that are fully
galvanized (i.e., all structural and miscellaneous components). The support
cables for the Brooklyn Bridge represent a well-known early application for
galvanizing. For more than years, galvanizing has provided corrosion
protection for the cables in this historic bridge.
Immingham and Hunterston Coal Terminals
When, in Englands east coast coal terminal was built at Immingham,
on the exposed flatlands at the mouth of the Humber River, the National Coal
Board stipulated that the corrosion protection for the structural and ancillary
steelwork guaranteed for 5 years.
The U.K. Galvanizers Association visited the site and, by comparing
conditions there with similar sites for which galvanizing performance figures
were known, concluded that a standard 85 pm galvanized coating could be
expected to provide for not only but years. After years of service, the
work was closely examined and found to be in excellent condition-in
marked contrast to nearby ungalvanized painted steelwork. In fact, coating
measurements suggested that it would be 20 years before any repainting of the
galvanized coating would needed.
The lessons of Immingham convinced the British Steel Corporation that
hot dip galvanizing should their first line of defense against corrosion for
the steelwork at their new deepwater iron ore terminal on the Ayr coast in
southwest Scotland, where, consequently, tons of structural steel was
hot dip galvanized.
Marinas
In in Galveston, Texas, a saltwater marina consisting of four structures,
each m long, was constructed with galvanized steel. After 20 years,
inspection showed that the zinc coating was well above minimum thickness
requirements. The majority of the steel columns are cm above saltwa-
ter, and all are subject to wind-blown salt spray.
Small Windmills
In about in one of the northern Netherlands provinces, Friesland, where
a marinerural climate prevails, about km from the Wedden Sea a wind-
mill was installed near Birdaard, a small village between Leeuwarden and
260 Chapter 2
Dokkum, in the Waterschap De Walden. Small windmills of this type carried
a wind motor, usually of American origin, which started turning when a
certain water level in the canal had been reached; thus, the water was pumped
out of the polder (reclaimed lowland area) until the preselected water level
had been reached again.
Providing a piece of galvanized angle bar to support his views, a Dutch
official said that such wind motors had been installed in large quantities
before World War I. A galvanized angle bar from the windmill near Birdaard,
measuring mm3, retained remnants of the zinc-iron alloy layer
still protecting the steel. The thickness was 25-60 pm. The underlying steel
was free from any trace of rust.
On one surface, the whole zinc coating system was almost completely
intact after years; the crystals in the q layer, as well as the 5 layer and the
S layer (consisting of a S palisade and a S compact layer) were clearly visible.
The coating thickness varies from 100 to 120 Km. Obviously, only a few
micrometers had weathered away during three-quarters of a century.
After German wind motors were imported, which often were not
galvanized and consequently rusted heavily after a few years in the moist
atmosphere of the Frisian polders.
The Cindu Phthalic Anhydride Plant, Vithoorn, Holland
Nearly 300 tons of steel was used in 1961 to construct this facility, which
consists of a distillation plant and an oxidation plant. Phthalic anhydride is
made by catalytic oxidation of naphthalene or orthoxylene. Therefore, the
atmosphere in and around the open steelwork of the plant is polluted by steam
vapors, phenol, naphthalene, and acid phthalic anhydride, together with
sulfur dioxide from oil burners. Some of the resinous residues, however, have
a certain protective action on limited areas (e.g., around valves, vents, and
resin sprayers).
In after years, almost all galvanized parts were still intact
without any rust; 5-10% the total surface area showed slight rusting,
especially on the edges of angle- and I-beams. The bolts and nuts showed
more rust and had caused rust staining on several beams and girders.
Railings, staircases, ladders, floor grids, and pipe supports showed
slight scattered rust here and there. Originally, the zinc coating had been 100-
125 pm. On the main constructional profiles, there was still 70-100 pm zinc
coating left. No maintenance was then necessary.
Cutty Sark, Dry Dock, Greenwich
The Cutty Surk was built in in Dumbarton with timber hull and decks
and iron ribs. The yards were timber but, in the vessel was restored and
the 17 tapered yards were replaced by welded steel spars. The welded steel
yards were grit-blasted and zinc sprayed to pm, then sealed with a vinyl
copolymer sealer. When the Cutty Sark was fitted out at Greenwich, the yards
were painted with two coats of black paint. At approximately 6-year intervals
in this industrial and marine atmosphere, the yards are bristle brushed to
remove any loose paint and given a further coat of black gloss for appearance.
After years, there was no sign of corrosion or blistering of the zinc.
Mild Steel Railings, Park Lane, London
Before erection in the each section of these railings was grit-blasted,
zinc-sprayed to pm, and sealed with a vinyl sealer. The sections were
welded together, and the weld areas were coated with zinc-rich paint. After
erection, two coats of green-pigmented, chlorinated rubber paint were ap-
plied. Loose debris is removed and the railings repainted as required every
years in this atmosphere heavily polluted by vehicle exhaust and subject to
mechanical damage. Thirty samples were taken at the most corrosive and
abrasive locations in After paint had been removed, the zinc-sprayed
coating was found to be intact, with an average thickness of 50 pm. No
undermsting was observed except at the unsprayed weld areas, where small
areas of rust showed; adjacent zinc-sprayed areas by their cathodic action had
restricted rusting, but the cathodic action had reduced the thickness of the zinc
to pm or less locally.
Renovation Decks Barges Operated in the Bristol Channel
by T. R. Brown & Sons Ltd.
The welded decks of barges, heavily pitted and totally perforated in places,
were repaired in sprayed with pm zinc, and sealed with one
coat of aluminum-pigmented unmodified phenolic resin sealer. At year
intervals, one coat of ordinary red oxide paint is applied for decorative
purposes only.
Inspection in (after years) showed the zinc coating to be
intact on all barges. At one spot cm diameter on the deck of one barge,
where the bargees foot always lands and rotates as he steps out of the
wheelhouse, the zinc coating had worn away to the steel, but there was no
active corrosive attack. Other steelwork not zinc-sprayed (e.g., wheelhouse
sides and hatch covers) had rusted through.
The decks of many Icelandic trawlers have also been successfully
protected by metal spraying.
Middle East Structures
The coastal climate in the northwest, west, and east of the Arabian Gulf is
moist-especially near Kuwait and around the AI Qatif area. The temperature
falls rapidly at dusk, leading to much condensation on steel structures over
262 Chapter 2
and near the beach and adjacent areas (up to 5 km inland). Information from
Van Eijnsbergen in 1978 yielded data that included galvanized water mains
(1 m above soil) at AI Ahmadi, Kuwait, which had withstood rusting for 12
years, and an oil refinery near Qatif whose galvanized parts were free from
rust after 7 years. However, many new plants have a duplex system. Gal-
vanized housing structures in and around Riyadh, Jiddah, Abu Dhabi, and
Bahrain have been observed after some 12 years and were free from rust.
Galvanized oil pipeline supports in the Saudi Arabian desert, observed after
some 15-20 years, have shown no rust. In Jiddah, various galvanized fences,
poles, and so on, have remained free from rust after some 15 years.
Galvanized light poles in Dubai with coating thicknesses ranging from
175 to 250 pm were reported by local residents having some experience in the
behavior of galvanized steel to last 20 years before maintenance.
Galvanized parts on islands in the Gulf (Sini, Das, etc.) remained free
of rust after 7 years, whereas paint systems failed within 5 years. However,
the undersides (horizontal) of such structures were slightly rusted by salty
condensation when less than 10 m above the sea level.
Resistance to Aqueous Corrosion
INTRODUCTION
Zinc, being an amphoteric metal, exhibits good corrosion resistance in water
having a pH near neutral. The corrosion rate is lowest in the pH range from
about 5.5 to 12, but both limits for zinc-aluminum alloys are up to half a pH
unit lower. Most natural potable waters have a pH range from 5 to 8.5, but the
corrosion of zinc in such waters is controlled largely by the impurities present
in the water. Rarely are naturally occurring waters pure. Even rainwater,
which is distilled by nature, contains nitrogen, oxygen, carbon dioxide, and
other gases, as well as entrained dust and smoke particles. Water that runs off
over the ground carries with it eroded soil, decaying vegetation, living
microorganisms, dissolved salts, and colloidal and suspended matter. Water
that has seeped through soil contains dissolved carbon dioxide and becomes
acidic. Groundwater also contains salts calcium, magnesium, iron, and
manganese. Seawater contains many of these salts in addition to a high
amount of sodium chloride. All these foreign substances in natural waters
affect the structure and composition of the films and corrosion products that
form on the zinc surface and control corrosion rates. In addition to these
substances, such other factors as pH, time of exposure, temperature, motion,
and fluid agitation all influence the aqueous corrosion zinc. This chapter
discusses the effects of these factors on the corrosion of zinc in water.
Water hardness is an important variable in zinc corrosion. Hard waters
are usually less corrosive toward zinc because they deposit protective scales
263
264 Chapter
on the metallic surface, while softer waters do not. Temperate seawater also
deposits protective scales on zinc and is less corrosive than soft water.
With zinc, oxygen is not harmful in practice unless it is distributed
nonuniformly. This is because the role of oxygen (and also that of carbon
dioxide) is to form a protective film when the zinc is attacked. Access of
oxygen to all surfaces is important, since on zinc and other common metals,
under conditions of limited oxygen supply, the attack may be localized in the
form of pits; in differential aeration areas, the places with lower oxygen
content act as anodes, whereas the generally larger places with higher oxygen
content act as cathodes. This type of attack can be mitigated by the use of
chromate films. Corrosion rates of steel increase with aeration of the water;
oxygen is five to ten times more aggressive than carbonic acid (Todt, 1961).
Raising the temperature increases the dissolution of zinc and the ZA
alloys (Belisle and DuFresne, 1986) in water. A marked increase occurs up to
about followed by a decrease at higher temperatures. Intergranular
corrosion of the zinc-aluminum casting alloys is a risk above about 70C in
wet or humid conditions (e.g., in steam), when no protective layer can form
and selective dissolution of the structure occurs. steam test at 95C is
indeed a standard method of checking whether alloys have been correctly
produced.
At temperatures usually between 60 and 90C (but varying with the
water composition), a reversal of potential may occur in which zinc coatings
become cathodic to the ferrous substrate.
The well-known Pourbaix diagrams (showing specific zones in which
no corrosion and passivation occur), in which pH is plotted against electrode
potential, are valid only in pure water at 25C. A comparison of the behavior
of zinc and other metals offers a deeper insight into corrosion by water,
however.
For research purposes, to identify the nature of protective films or to
study the mechanism of corrosion, many investigators have used distilled
water, which although highly corrosive is relatively pure.
II. INFLUENCE OF TYPES OF WATER
A. Types of Water
"Water" is a very comprehensive term. Four-fifths of the earth's surface is
covered with water, which evaporates and falls back as rainwater. In transit
through the atmosphere, it collects oxygen, carbon dioxide, and nitrogen and
sulfur compounds, besides others not important to corrosion. Upon soaking
Resistance to Aqueous Corrosion 265
into the earth, rainwater becomes groundwater and is enriched by mineral
constituents (carbonates, sulfates, chlorides, etc.). Rainwater collects in lakes
or flows via streams and rivers back to the sea. This surface water varies in
composition according to the rainfall, the local composition of the soil, and
the nature of the bed and banks of the stream; biological factors may also
influence it. Drinking water is usually obtained from groundwater, while
surface water may be used for industrial purposes.
Since groundwater is in contact with the soil, it is usually enriched with
minerals. If it is in contact with hard rock, however, the mineral content is
small. Groundwater comes to the surface in springs; if the water then contains
large quantities of mineral salts, it is known as mineral water. Among the
forms of mineral water, we may list:
Bitter water: with high contents of magnesium salts
Sulfurous water: with hydrogen sulfide
Carbonated water: with high contents of carbonic acid
Ferrogenous water: with high contents of iron salts
Harbor water is a mixture of sea and river water, sometimes polluted.
Brackish water is diluted seawater found elsewhere than in harbors. A special
group is composed of seawater and brine, which is saltwater concentrated by
evaporation. Effluent waters vary considerably in composition and aggres-
siveness. An indication of the composition of various types of water and their
variability is given in Tables and The nature of seawater is docu-
mented by Chandler (1985).
When natural water is unsuitable for domestic or industrial use, it is
treated to soften it or remove salts, depending on the intended use. Additions
are made to hinder the precipitation of scale or to facilitate the formation of
protective films. Iron and manganese salts may be oxidized or precipitated by
aeration. Ion-exchange and electrical treatment is especially effective. Con-
versely, when it is required to use a soft acid water in conjunction with zinc-
as, for example, for hot water systems in Sofia (Iovchev and Mihailov,
1988)-a possible way of limiting this effect is to remineralize the water using
materials containing magnesium ions (e.g., a mixture of MgC03 or CaCO, +
Na2Si0,, or commercial products such as Neutralite).
Steam may be formed by evaporation and condensed to give distilled
water, which is relatively pure; the purity can be increased by repeated
distillation. Condensation and dew, which are relatively low in salt content,
originate when warm, saturated air comes in contact with cold surfaces. Dew
is caused by precipitation of water onto a surface from saturated air when the
air cools.
6
N
\o
0
M
a
-
2
.-
266
Weir steps,
Offen bach ,
April 1954
Offenbach,
7/16/ I957
5/61 1954
Weir steps,
Lock Feudenheim,
Lock Feudenheim,
611811957
Lock Varloch,
912711 954
Loch Varloch,
41161 1957
Lock Dorpen,
9t21l I954
Lock 2,
Oberhausen
RHK, 1/28/1955
Oberhausen
RHK, 4/13/1957
Lock 2
Light yellow,
slightly
cloudy
yellowish
Cloudy,
Clear, color-
less, slightly
cloudy
Yellowish
Clear, color-
less, slightly
cloudy
Yellow
Slightly cloudy,
yellowish
Slightly cloudy,
brownish
Slightly turbed
light gray
Cloudy,
yellowish
6.9
7.0
6.8
7.2
6.0
6. I
5.6
5.4
6.7
7. I
3
14
7
3
6
27
18
2.2
2.2
0
16
15
20"
29
6.6
4.4
24.5
35
99.3
106
121
28
36
28
227
89
81 2
96 3
123 14
126 6
84 2
96 2
32
30 0
56 2
124 16
0.75 13.8
1.8 17.9
16.4
0.2 19.5
0. I 15.0
0. I 15.8
4.6
0 2.1
17.7
0. I 17.4
8.1
13.4
9.0
13.1
10.1
1.1
0.8
0.8
11.2
10.1
'Humic acid, about 50 gdL.
Source: Kleinschmidt and SchrMer (1%2).
N
U
Q,
Table 3.2 Chemical Analysis of Brackish Water at Various Years and Seasons
Average in Cuxhaven
(Alte liefe)
Average of 7 samples (2 and 5 years)
at Bremerhaven Fish Harbor
Constituent or quality Average Minimum Maximum Average Minimum Maximum
Oxygen, mg/L
pH value (electrometric)
Smell
Color
KMnO, demand, mglL
Free H2S, mg/L
Free COz, mg/L
Aggressive C02, mg/L
C02 (by Heyer), mg/L
Electrical conductivity:
Chloride (Cl-), mg/L
Sulfate (SO:-), mg/L
HCO;, mg/L
Cubonate (co:-), mg/L
Nitrate (NO;), mg/L
Nitrite (NO;), mg/L
i 80c x lo3
10
7.8
None
Green
35
0
0.6
2.5
53
23
8,800
1,300
141
I .5
6
0. I
5.2
6.6
13.8
8.0
Cloudy
22.3
0
0
0
50
13.7
4,800
760
134
0
3
0.07
43
0
I .7
6.6
57
28.3
13,000
1,870
146
6
11
0.1
~ ~ ~~
8.5 3.4
7.65 6.85
Fishy
Yellow, green, cloudy
38 14.3
0.65 0.65
8.5 4.3
46 9.5
10 4.1
I10 70
13.2
8 .O
68
0.65
12.8
34
110
13.6
3,400 1,130
480 218
180 167
0 0
2 2
0 0
5,140
730
192
0
5
0.3
2 2
?
2
2
269
270 Chapter
Water is often present in dissolved suspended form in organic liquids
and may form a sludge with any water-soluble salts present. Serious corrosion
can be caused in this way-for example, in gasoline and diesel fuel tanks and
fuel oil containers-especially at the air-liquid interface. Water can behave in
a similar fashion when it arises from the decomposition or reaction of organic
compounds or when it occurs in an organic medium in the course of use (e.g.,
in trichlorethylene used for degreasing wet articles).
It is useful to be able to classify waters from the viewpoint of corrosion.
Simple classifications by salt content, appearance, smell, taste, and biological
properties are not sufficient. Electrical conductivity is useful, especially in
relatively nonionized waters, and information on the hydrogen ion concentra-
tion (pH value) is most valuable. The major survey on zinc by Wiederholt
(1965) has been supplemented by later work [e.g., Burkard and Revelly, 1979
(in French)], and there are many general accounts [e.g., Wagner, 1977;
Morbe and Werner, 1978 (in German)].
B. Total Salts
It is usually sufficient to differentiate only between small medium salt
contents-up to about mg/L, as in drinking water and waters of higher
salt contents. [Note that mg/L = 1 part per million (ppm).] It is important to
know whether the content is due to hardness-forming compounds alone to
other salts, and especially whether the chloride content exceeds about 50 mg/
L. The content of carbonic acid and of oxygen also is very significant.
Microorganisms may accelerate or retard corrosion: suspended solids such as
sand usually are removed from industrial water before use; the abrasion
resistance of the zinc-iron alloy in coatings is helpful if suspended solids are
present.
C. Effect of pH and Hardness in Supply and Natural Waters
In acid water (i.e., water having a low pH), the attack on zinc occurs by
acidic corrosion with hydrogen evolution.
Zn + 2H + Zn2+ + H,
This consists of the anodic partial process
Zn -+ Zn2+ + 2e-
and the cathodic partial process
2H+ + 2e- + H,
Resistance to Corrosion 271
However, in aqueous solutions with pH values higher than 4.5, there are
insufficient hydrogen ions for the attack to proceed at an appreciable speed,
and the rate of corrosion is controlled by the quantity of oxygen available and
its diffusion to the metal surface. This arrangement controls the anodic partial
process given above and usually means lower corrosion rates. The total
available quantity of acid must still be known, since it indicates the total
availability of hydrogen ions.
There is no abrupt transition from acidic corrosion to oxygen-controlled
corrosion; a boundary region exists in which either or both may occur,
depending on the circumstances.
The primary factor affecting long-term corrosion is whether a water
does does not form scales, which in turn depends on hardness and pH. This
property is determined by calculating the Langelier index or the Ryzner
stability index. The simpler form of Langeliers equation is:
pH, = pCa + pAlk + @K2 - pK,) at constant temperature
where
pH, = saturation pH value
pCa = negative logarithm of the calcium concentration expressed as
ppm CaC03
pAlk = negative algorithm of the alkalinity to methyl orange (ppm of
equivalent CaC03)
pK2 = ionization constant of HC03: [H+] [CO:-]
WC0 ;l
pK, = solubility product of CaC03
This simple form does not apply to pH values exceeding 9.0, and high
salinities affect its accuracy. The term @K2 - pK,) is a function of tempera-
ture and ionic strength (dissolved solids). With an analysis of a given water at
a constant temperature, much useful information can be obtained from the
equation.
The saturation index of a water (SI) is defined as
SI = pH - pH,
where pH is the actual pH of the water. If the saturation index is positive, the
water will be supersaturated with calcium carbonate, whereas if it is negative
the water will be aggressive to calcium carbonate. The determination can be
approached graphically (Fig. 3.1). However, Pisigan and Singley (1985) have
PARTS PER MILLION
Fig. 3.1
American Chemical Society).
272
The Langelier saturation index chart (Shreir, 1976, based on data from the
shown that some waters with a positive SI are nevertheless corrosive to zinc
and urge that other factors be taken into account.
The Ryznar index, which provides a more quantitative approach and
also differentiates between waters with the same saturation index but with
highly different properties, is given as follows:
Thus, for two waters A and B, the following tabulation applies.
Langelier Ryznar
Actual Calculated saturation stability
Water PH PH, index index
A 6.5 6.0 +0.5 +5. 5
B 10.5 10.0 +0.5 + 9.5
As can be seen from the saturation index, it might be predicted that both
waters would be equally scale-forming, yet the Ryznar index indicates other-
wise.
Water with a Ryznar stability index less than 6.0 is definitely scale-
forming; indices of 6.0-7.0 are regarded as good for domestic waters: possi-
bly at values above 7.0, and certainly with indices above 7.5, the waters do
not form a protective scale. One of the reasons for the discrepancies noted in
the performance of waters with identical saturation indices lies in the buffer
capacity of a water.
A second factor is the nature of the scale. Gilbert (1952) showed the
very large number of different corrosion products that can form on zinc
corroding in water. Schikorr (Table 3.3) gives analyses of corrosion products
formed in distilled water or condensates. The nature of the corrosion products
in relation to pH and chloride concentration has been shown by Feitknecht
(1955) (Fig. 3.2). At water temperatures above about 55C (Thomas, 1980),
the corrosion products that form have a coarse-grained structure and less
adhesion to the zinc surface. Corrosion of the zinc will still occur locally as a
result of discontinuities in the scale or local electrochemical action. The
chemical and electrochemical processes in the formation of the protective
layer were discussed by Kruse (1976).
Table 3.3 Composition of Corrosion Products Formed on Zinc in Distilled Waters
Corrosion products (%)
Origin
Bound
Zinc HZO co2 Ref.
Spray experiment with
2ZnC0, - 3Zn(OH), theory
ZnCO, 3Zn(OH), theory
Underside of a roofing
distilled water
sheet
Granulated zinc specimens
in distilled water with-
out and with aeration
58.4 ? 14.5 Gilbert and Hadden
58.5 ? 26.6 ( 1950)
59.5 9.8 16.0 Schikorr (1962, 1965)
57.4 15.8 12.9 Moody (1903)
67. I 9.7 1.9
71.0 5.6 3.2
72.0 7.7 1.2
? 8.5 1.5
? 7.8 4.4
57.2-72.2 12.92-5.89 1.71-16.31 Cocosinschi (1931)
Source: Schikorr (1965).
PH
Fig. 3.2
pH values and CI- concentration (Feitknecht, 1955).
Regions of existence of the various corrosion products of zinc in relation to
274
D. Effect of Specific Constituents
Oxygen, Chlorides, Other Anions and Cations
The oxygen content of a water is a controlling factor, and it may vary widely
according to the source of the water. In well waters, it is usually low, but it is
high in spring waters that have been in contact with the air. Most surface
waters have a high oxygen content. At normal pressure (760 mm Hg) water
will dissolve 1 mg/L oxygen at IOOC, 9 mg/L at 20"C, 6 mg/L at 50"C, and 4
mg/L at 70C. Air dissolved in water contains approximately oxygen and
67% nitrogen.
Of all the anions, chloride is most corrosive to zinc in water, especially
if it is present in amounts exceeding 50 mg/L. The softer the water, and the
lower it is in carbonate, the more pronounced is the effect of chloride. Thus, a
chloride content of 80 mg/L in soft water causes quite severe corrosion, while
in hard water no corrosion occurs even with 700 mg/L (GAV, 1961).
Weast et al. (1961), who reviewed the corrosion of zinc in dilute
aqueous solutions and zinc potentials, obtained linear relationships between
corrosion weight loss and time of immersion at various temperatures in the
50-80C range in water containing mg/L potassium chloride and under
pressure. The zinc corrosion products formed at 88C are about lo00 times
more electrically conductive than those produced at 25C. Grubitsch (1969)
has also looked at potential in sodium sulfate. They did not observe the
maximum at 60C found in distilled water (see later). It was proposed that the
maximum is a characteristic of corrosion in dilute aqueous solutions at
atmospheric pressure, where air oxygen can escape from the solution. The
direct dependence of corrosion rate on the oxygen concentration above the
dilute potassium chloride solutions at 51C was shown a half-century ago by
Kenworthy and Smith (Fig. 3.3).
Sulfate, in the usual concentrations, is less active than chlorides toward
zinc. Nitrates are usually low in tap waters; probably about 160 mg/L is
needed for a specific effect. Nitrate is also usually low in rainwater.
Hydrogen sulfide and other sulfides usually are not present in drinking
waters. They sometimes arise by anaerobic decomposition of sulfur-con-
taining organic matter in the water. Zinc is less sensitive to such waters than
are copper and iron.
Salts of metals nobler than zinc are likely to be present only in special
cases: for example, in effluents from electroplating plants. In such cases, the
nobler metal normally deposits on to the zinc, while an equivalent amount of
zinc goes into solution. Small amounts of dissolved salts (e.g., copper) can
set up local cells, producing pits; others may help inhibit such attack by
276
Chapter 3
I
0
Hours
I I I
20
I I I I
60 80
3.3 Corrosion of zinc as a function of oxygen concentration in solution:
solution, 100 ppm KCl; temperature, 51C (Kenworthy and Smith, 1944).
sequestering the corrosive species. However, substances that form complexes
with zinc but do not form protective films, such as ammoniacal salts, may
promote corrosion in water. Oxidizing salts may also stimulate attack under
special conditions.
Organic Compounds and Microorganisms
Among organic materials, humic acids, which are especially present in
swamp waters, intensify the attack on zinc, but it is unusual for such waters to
come into contact with zinc.
Algae, fungi, bacteria, or other organisms may occasionally be present
(e.g., in cooling water), and tend to increase the waters aggressive action,
although some algae are killed by zinc. Hydrogen sulfide may expected
from the action of sulfate-reducing bacteria or an anaerobic decomposition or
organic material (foulness). The effect of this reaction is mentioned above.
Impurities such as oil may cause nonuniform aeration of the surface and
Resistance Corrosion 277
thus facilitate localalized corrosion. Contamination by illegal dumping into
the sea (of oil, chemicals, etc.) sometimes has increased corrosion attack.
Suspended Matter
Suspended matter and other insoluble matter may facilitate pitting corrosion,
particularly if it accumulates on the zinc surface, because in this way it may
lead to varying aeration of the surface. Rust from distant sources and other
corrosion products may promote corrosion in this manner.
Suspended matter may also cause erosion of protective films, thus
allowing more zinc to go into solution; this can occur particularly at elbows in
pipes carrying fast-flowing water.
Inhibitors
Many substances in solution can inhibit the corrosion of zinc. Desai, Rana,
and Gandhi (1968, 1973) compiled literature references. Witt et al. (1985)
have classified inhibitors in water systems (Table and a list of some
inhibitors with references to detailed source information is given in Table 3.5.
Table 3.4 Inhibitor Types Zinc in Water
Group Examples of basis Effect
Passivators (anodic
inhibitors)
Surface filming agent
(adsorption
inhibitors)
Electrochemical
inhibitors (cathode
inhibitors)
Destimulators
Chromates, alkaline
nitrates, phosphate
polymers
Alkali silicates,
phosphates, borates
Ions of antimony,
arsenic, mercury,
magnesium, zinc,
nickel, cadmium
Sodium sulfide,
hydrazine
Formation of tight protective
layer.
With less than a minimum
concentration strong local
pitting can occur; hence
the term dangerous
inhibition.
Precipitation of almost
insoluble compounds.
Precipitation of metal by ion
exchange reduction.
The damaging effect a
stimulator (e.g., oxygen)
in a corrosive medium is
avoided.
Source: Witt et al. (1985).
Table 3.5 References to Inhibitors in Aqueous Solutions
Inhibitor Effect in Ref.
Acrylonitrile
Agar-agar
Alizarin
Amino acids
Amino chromate
Arsines (tertiary)
Benzimidazole
Benzoates
Benzotriazole
Borax
Cerrous chloride
Chromate dichromate
Congo Red/Crystal
Dichromate (potassium)
Dichromate cation
Violet
Dimethylvinylethynyl
carbinol
Egg albumin
Hexametaphosphate
Lanolin
Maleic acid
Maleic hydrazide
Monochromate
N-containing organics
n-Decylamine
P-Napthalene
Nitrate + acetate +
cyanide + cation
Nitrobenzoate
Octanocylamine salts
Chloride
Water
Sulfuric acid
Sulfuric acid
Water
Perchloric acid
Phosphate
Water
Atmosphere
Water
NaCl
Water
Sulfuric acid
NaOH
Chloride
Acidic sulfate
Water
Water
Water
Water
Water
Chloride
Perchlorates
Water
Chloride
Sulfate
Chloride
Mor and Beccaria
Friend et al. Evans
Gaur and Jain
Grigorev and Kuznetsov (1
Irdnov and Klyuchnikov
Rudresh and Mayanna
Kuznetzov and Podgornova
Wormwell El Haleem and
Fattah Gad Allah et al.
(
Zah et al.
Singh and Benerjee
Wormwell Dutrizac and
MacDonald
Hinton and Wilson
Grace and Co.
Gaur and Jain
Shalaby and Abbas
Horruk Hill
Dechema Werk Costa
Kiryakov and Brandt
Evans (
Dutrizac and MacDonald
La Que Dutrizac and
MacDonald
Mikhailov et al.
Hersch
Rotheli and Cox Shams el
Wormwell and Mercer
Singh and Banerjee
Rudresh and Mayanna
Biallozov and Drzewicko
Banister
Fedorova Persinacera
et.al.
Hersch
Sat0 and Kat0
Din et al.
Werkstoffe und Korrosion (
278
Table 3.5 Continued
Inhibitor
Oxalates
Phosphate
Phosphate glasses
Phosphonium salts
Picrate
Phenothiazine
Piperazine
Pyridine (max mg/m3)
Pyrazols
Silicate
Silicate
Sodium benzoate
Sodium dodecamolybdo-
phosphate
Sodium hydrogen
phosphate
Sodium silicate
Sodium tungstate
Stearyl amine salts
Tannic acid
Tetrabutylammonium
bromide
Thiodiglycol
Thioglycolic acid esters
Thio ureas
Tragantgum
Triazoles (see also
benzotriazole)
Zinc salts
Effect in
Water
Water
HCl
Water
NH,CI
Acids
HCI
Chloride
Acetate
Water
NaOH
Water
Water
Chloride
Chloride
Chloride
Chloride
Acids
Damp
Chloride
Acids
Water
Water
Water
Ref.
El Haleem and Fattah Gad
Allah et al.
Raisstrick Werkstoffe und
Korrosion Awad and
Kame1 Kunzmann
Kruse Ehreke
and Stichel
Hatch
Troquet et al.
Hersch
Aal et al.
Jain and Gaur
Gad Allah et al.
Biallozov
Kawai et al.
Hormk Hoxeng
Arpaia (
Shalaby and Abbas
Wormwell and Mercer
Grover et al. Dechema
Werk (
Brasher and Rhoades-Brown
Costa and Molera
Shams el Din et al.
Schweinitz et al. Dechema
Werk
Shams el Din et al.
Sato and Kat0
Jain and Gaur
Joussellin and Wiart
Noranda Mines
Costa and Molera
Jain and Gaur
R.M. Evans
Wippermann et al.
Liddiard et al.
Sources: Witt et al. (1985), Desai (undated), and others.
a See Witt et al. and Desai papers for references not listed in this book.
279
280 Chapter
Since many inhibitors occur naturally in water, corrosion in practice is often
less than in the laboratory- where solutions of pure chemicals are used.
Perhaps the best example is the use of tests in sodium chloride solution or
spray as a guide to behavior in seawater. Seawater contains many other
substances, of which magnesium salts in particular are inhibitors of zinc
corrosion. Additions that can be made to water to reduce its corrosivity are
given in relation to specific waters later.
It is useful to mention here more specifically some of the additions that
inhibit corrosion. Liddiard et al. (1943) reported that 0.005 mg/L of zinc salts
can retard corrosive action of some waters, and this type of self-inhibition
may often be important in practice. Sanyal et al. (1959), studying atmospheric
attack under laboratory conditions with low SO2, observed that corrosion on
steel is reduced in the presence of zinc not in contact with steel. Presumably,
that is, zinc ions are transmitted to the steel surface.
Inhibitors in acids have been studied by Petrov and Savitskaya (1970)
for nitric acid and by Borchers and Krug (1971) for the effect of cations of
nobler metals in various acid media; corrosion was either accelerated
inhibited depending on the hydrogen overvoltage of the metal deposits
formed, a high overvoltage inhibiting corrosion. Mixed acids were researched
by Shams El Din and Khedr (1971), who also determined critical current
densities for the passivation of zinc in solutions of various concentrations of
different anions (1972, 1974).
For caustic soda, Abdul Azim et al. (1974) looked at gelatin, phthalic
acid, sulfide, silicate, and chromate.
Wormwell (1953) observed that at 1 S-2% concentration, borax affords
complete protection to zinc in contact with certain waters. Sodium benzoate
was found to afford partial protection to zinc in water as a result of experi-
ments made to assess its beneficial effect on steel corrosion (Wormwell and
Mercer, 1952; Grover et al., 1961). Sodium nitrate similarly reduces the
corrosion of zinc in water, while sodium hydrogen phosphate and sodium
silicate can afford complete protection. Wormwell and Mercer also showed
that sodium cinnamate affords partial protection to zinc in water; sodium
o-nitrohydrocinnamate, sodium m-nitrocinnamate, and sodium p-nitrocinna-
mate protect it adequately.
Chromates and molybdates are effective inhibitors for zinc, and they are
the basis of many conversion coatings. Since, however, chromates are anodic
inhibitors, their concentration should be maintained above to avoid the
risk of pitting, which arises when inhibitors of the reaction at the anode are
present in insufficient quantity. Chromates may have to be phased out in the
coming decade, however, to comply with more stringent laws relating to
pollution and toxic substances.
281
CORROSION DATA
Distilled Water
The behavior of zinc in distilled water has often been studied in connection
with investigations on the mechanism of corrosion. Bengough et al.
and Bauer and Krohnke (quoted by Wiederholt, found that zinc is not
greatly corroded in pure water provided it is free from oxygen and carbon
dioxide, even when boiling (Traube, Taboury and Gray,
Evans and Davies also noted that the attack of zinc by distilled
water is slow even in the presence of oxygen. Under stagnant conditions,
however, pitting may occur at weak points in the surface film. In stagnant
water, oxygen may not be replenished rapidly enough at such weak points to
permit the re-formation of a protective film. Pitting could be prevented by
rotating the specimen in the water to provide uniform and rapid replenishment
of oxygen over the entire surface. Pitting was also prevented by supplying the
oxygen to the surface under about atm pressure.
Anderson and Reinhard noted that corrosion is reasonably uni-
form when oxygen is freely available, but below a critical oxygen level,
concentration cells may develop that cause pitting attack. The corrosion rate
increases when oxygen is bubbled through the water, but the attack remains
uniform. Anderson and Reinhard showed in 7-day tests that corrosion in-
creases by a factor of about as the temperature is increased from room (0.7
pm total loss in days) to in sealed flasks; slow bubbling of oxygen,
however, increases the room temperature corrosion rate by a factor of
which is reduced to at similar to behavior at under these latter
conditions. Cox had earlier obtained different results with a maximum
corrosion rate p,m in days) at in aerated distilled water. At
the temperatures of maximum corrosion, the film is granular and nonad-
herent, while at lower temperatures it is compact and adherent. If the granular
coating is more permeable to oxygen, increased corrosion may be expected.
Maconachie noted that Cox's results applied only to a 15-day
test and did not define the influence of duration of exposure within this time.
Maconachie immersed zinc specimens in distilled water aerated by
movement of the test assembly for 7 days at temperatures of Weight
losses were determined on specimens removed at daily intervals. At the
distilled water had a pH of largely as a result of dissolved carbon dioxide.
At the higher test temperatures, however, the pH was probably higher as the
carbon dioxide was .expelled from the solution. The total corrosion as a
function of temperature is shown at daily intervals in Fig. The general
shapes of the curves resemble those reported by Cox for 15 days and
282 Chapter 3
Temperature "C
30 40 60 80 70 80
I I I
4-
DaYs
7-
8-
6-
4-
3-
2-
-
I I
I I I
I I I
86 104 122 140 158 176
Temperature "F
Fig. 3.4
(Maconachie, 1934).
Effect of temperature on the corrosion of zinc sheet in distilled water
by Maconachie (1934) for varying periods, with the maximum corrosion
occumng at 60C regardless of the duration of exposure. The same data,
plotted to show the relationship between total weight loss and time of immer-
sion at the various temperatures, reveal wide fluctuations, varying with time,
in total weight losses; indeed, there were periods during which the specimens
apparently gained in weight. Even with variations in the test procedures, there
was no obvious explanation for this behavior, which was reproduced several
times. In one experiment, at over a 2-week period, a third corrosion
maximum occurred at 12 days.
Grubitsch and his associates (1940) studied the relationship between
temperature and corrosion of zinc in distilled water. Their tests showed that
the familiar peak at 60C occurs in waters under both a pure oxygen atmo-
sphere and a C02-free air atmosphere (see Fig. 3 . 3 , and Belisle and Dufresne
(1986) showed a similar peak for the ZA alloys (Fig. 3.6). In contrast, the
peak disappears completely under an oxygen-free nitrogen atmosphere. Fur-
ther experiments showed that as the partial pressure of oxygen over the water
was reduced from that in air, the peak decreased in magnitude and shifted to
lower temperatures. the temperature increases, the reaction rate initially
increases, but the increasing temperature serves to decrease the oxygen
supply, whereupon the depolarizing effect of oxygen is decreased, and the
corrosion rate slows.
This theory is consistent with the observation that decreased oxygen
supplies (caused by reduced oxygen pressure over the water) lower the
corrosion peak temperature. the oxygen partial pressure is lowered, the
equilibrium oxygen content in the water is lowered for any temperature. The
reduced oxygen content in the water thus lowers the temperature at which
depolarization becomes rate limiting, since the critical oxygen level for
depolarization is reached at a lower temperature. Two of the curves in Figure
show the difference in corrosion rates between water with reduced oxygen
content and water with none. That difference is the part of the reaction rate
that can be attributed to free oxygen in the water which decreases with
increasing temperature.
In another report, Grubitsch (1970) considered the effects of increased
pressure on the elevated temperature corrosion zinc. Zinc was corroded in
distilled water under various pressures of nitrogen and of air. The pressure
a pure nitrogen atmosphere had very little effect on the corrosion rate of zinc
up to atm of overpressure. Since nitrogen itself does not participate in zinc
corrosion, an excess of it in the water should have little effect. The effects of
increased air pressure, however, are quite complex. During tests of approx-
imately 10 hours duration, two maxima appeared in the corrosion rates: the
previously reported peak at 60C plus another at The lower temperature
maximum was a short-term effect that became insignificant after a few
hundred hours. Increased pressure of air above the water increased the
corrosion rate at the lower temperature of while lowering the corrosion
rate at 60C. This effect is shown graphically in Fig. 3.7. The reasons for this
284
Chapter 3
Temperature OC
0 10 20 30 40 50 60 70 80 90 100 110
I I I
I
158
I I
176 194
Fig. 3.5 Hot water corrosion of zinc under various gaseous phases, 10-minute test:
under pure nitrogen; 0, under nitrogen containing 0.6 v01 % oxygen; under air,
CO, A, under pure oxygen; H, 0 minus (Grubitsch and IlIi,.1940).
3 5
-
-
I
N
-
2 5
-
2 0
-
c
z 1 5
-
g 10-
cz
0
0 0 5 -
0.0
285
60 I20 180 200
r~ I '
-
12.0
o e
0 Z A - 1 2
A 2 A - 2 7
-
>
-
10.0 7
-
W
- 8.0
-
6.0
c
z
g
(L
0
-
-
2.0
I
IO 20 50 60 80 90 I O 0
- 0
3.6 Effect of temperature on the corrosion rate of alloys in tap water,
45 day tests (Belisle and DuFresne, 1986).
behavior are not clear. The accelerated corrosion rate that accompanies
increased pressure at low temperature appears reasonable. Increased air pres-
sure would add to the oxygen in the water, depolarizing and accelerating the
reaction. At the increase in oxygen might have the opposite effect by
promoting the rapid formation a stable, protective oxide film (see later).
Smrcek and Barton (1955, 1956) reported maxima at different temperatures,
depending on the type of water, but the influence of individual constituents in
the water apparently was not investigated.
Only with zinc powder, which has a very large surface area per unit
weight, or in the presence of impurities that reduce hydrogen overpotential,
does hydrogen evolution occur. Imai (1955) found that in distilled water at
25"C, 3.2 mol of zinc powder was oxidized with hydrogen evolution.
286 Chapter 3
Temperature, ("C)
Fig. 3.7
of zinc in water (Weast et al., 1961).
Effect of temperature and excess pressure (over atmospheric) on corrosion
Formation of hydrogen peroxide can also occur (Bauer, Krohnke, and Mas-
ing, quoted by Wiederholt, 1965).
The presence of detectable quantities of hydrogen peroxide in distilled
water saturated with air and acting on granulated zinc also has been observed
by Barnes and Shearer (1908) among others. After 7 hours, all the hydrogen
peroxide had decomposed, since the further dissolution of zinc was not
prevented by any protective film. This observation has been confirmed by
Delahay, Pourbaix, and van Rysselberghe (1951) and others.
In water free from carbon dioxide, the zinc hydroxide may be formed in
the colloidal state; this compound later precipitates or, at temperatures above
60C, is converted into granular zinc oxide. This reaction explains the cloudi-
ness sometimes seen when aerated, distilled water acts on zinc. Snyders
(1878, quoted by Wiederholt, 1965) showed that as long as carbon dioxide is
excluded, there is no difference in the action of air-saturated and oxygen-
saturated water. Even impure zinc gives the same result. In distilled water
under a nitrogen atmosphere, 0.02 pm per day dissolved at 20C, an amount
that was doubled under pure oxygen; under a gas atmosphere containing 0.6
v01 of oxygen, or under normal air, about 0.3 pm was lost per day (Grubitsch
et al., 1940).
The separate and combined effects of aeration, agitation, and carbon
dioxide have been investigated in the laboratory on a number of occasions.
Bengough and Hudson (1919), Kenworthy and Smith (1944), Bauer and
Schikorr (1934), and Anderson and Reinhard in the Corrosion Handbook
(Uhlig, 1948) have reported on the effect of these factors on the corrosion of
zinc at room temperature. Agitation, whether by movement of the water or the
specimen, or by bubbling air or carbon dioxide through the water, tends to
increase corrosion.
Burgmann and Schwenk (1975) evaluated the corrosion behavior of hot
dip galvanized steel in cold and warm water through which gases (oxygen,
nitrogen, carbon dioxide, and carbon dioxide/oxygen mixtures) were circu-
lated. Some samples had either no lime hardness or two different levels
lime hardness. The tests were at 20 and 70C, with steel strip galvanized to a
coating thickness of 65-75 pm. The test water contained 150 mglL of sodium
chloride. In water free of oxygen and carbon dioxide, zinc corrosion is due to
hydrogen evolution, which is accelerated by chloride content (preferably 50
mg/L). The zinc coating is attacked by oxygen, but the influence of the
chloride ions is still significant. Protective layers, which may be set up in hard
water, inhibit the acid corrosion that occurs in the presence of carbon dioxide.
In warm water, however, the presence of hardness is very damaging to the
zinc surface.
B. Condensation and Dew
Stagnant rainwater, condensation, dew, and similar forms of moisture mostly
act like distilled water on zinc. Condensation at the contact points of zinc with
building materials may leach out salts from the latter, sometimes with serious
results.
In general, rainwater in the country contains much carbon dioxide,
facilitating the formation of a protective film. The composition of rainwater is
mostly controlled by the local environment, and there may be large fluctua-
288 Chapter
tions from time to time or place to place. Corrosion may be markedly
increased by ammonium salts in rainwater. In industrial districts, the sulfite
and sulfate contents of rainwater are particularly unfavorable to zinc and, near
the sea, aggressive chlorides are present. Nitrates also have some signifi-
cance, Although the first formed corrosion products will contain carbonates,
these will, in time, become converted into basic chlorides and/or sulfates,
which are usually less protective than basic carbonates.
Dew and condensation are similar in composition to rainwater, but the
physical form of the thin films or isolated drops facilitates the formation of
differential aeration cells.
Light condensation, in which the condensed water soon reevapo-
rates, causes white corrosion products, which soon become converted by
ready access of the carbon dioxide of the air into protective films of basic zinc
carbonate. Schikorr has pointed out that heavy condensation,
which does not easily reevaporate, can cause severe corrosion because the
circumstances that prevent reevaporation also usually prevent ready access of
carbon dioxide to the surface. A common case is the underside of inade-
quately ventilated roofing sheet or bridge decks. The corrosion products then
consist of zinc oxide hydroxide, low in carbonate (Table Such films
were studied on. the electron microscope by Feitknecht (1958). The film
readily splits off scales, which showed the honeycomb structure of thin films
at various angles. According to X-ray examinations of Sahli (reported by
Feitknecht) these scales contain basic zinc carbonates. If the corrosion prod-
uct is treated with dilute ammonia solution, the basic zinc carbonate is more
quickly dissolved than the zinc oxide, the scales disintegrate, and the zinc
oxide particles remain. This proves the corrosion product to consist of a
codeposit of oxide and basic carbonate, which is generally porous and perme-
able, therefore not protective. Schikorr advances this substance as an example
of how the cause of corrosion can be diagnosed by examination of the
composition of the corrosion product.
The effect of condensation compared with immersion is seen in partially
filled closed vessels of zinc-coated steel: condensation often occurs in the air
space above the water, on the upper wall, a thick, porous, whitish layer
forms, while on the immersed surfaces there is no obvious corrosion product.
Condensation can cause serious corrosion on zinc or zinc-coated parts
packed for storage or transport where, if condensation occurs, the moisture
cannot readily reevaporate and the carbon dioxide present is soon used up. If
the parts are packed in wooden boxes, the problem is aggravated by evapora-
tion of water from the wood.
Dew, in contrast to condensation, forms during short periods of expo-
289
on surfaces that are particularly well ventilated. The prospect of corro-
sion is small.
However, dew contains only 10 mg of oxygen per liter, whereas in the
atmosphere 294 mg of oxygen per liter of distilled water can be present. Thus,
differential aeration effects can occur and, when is also present, it
dissolves readily in dew droplets (Van Eijnsbergen, 1959).
Knotkova-Cermakova. and Vickova (1970) found that with a controlled
atmosphere containing 1-55 ppm the absorbed rapidly oxidized to
sulfate when the electrolyte film contained dissolved particles of corrosion
products. The pH of the electrolyte alters only slightly and does not vary with
the partial pressure of sulfur dioxide.
C. Rainwater
The effect of rainwater on atmospheric corrosion was covered in Chapter 2.
When rainwater is allowed to collect in pools (e.g., if surfaces are flat instead
of being inclined that water runs off), the main effect is due to the acidity.
With hot dip galvanized steel at pH values of 5.8-6.1, Hudson (1953b) found
the weight loss to be only one-fifth that in the water of acid rain, pH 4.3.
Some compositions of rainwater in Holland from the 1950s are repro-
duced in Table (SDV, 1959). The extremely high salt contents of the
rainwater in Lutterade may be attributed to the chemical industry situated
there. The appreciable alteration with time in the chemical composition of
Table 3.6 Chemical Composition of Rainwater (mg/L) in Locations
in the Netherlands
Location
Constituent Lutterade" Scheveningen Amsterdam Weert
Ca2 +
Mg2+
Na+
c1-
NO,-
HCO
~~ ~
N H ~ +
SO,"
~
45
4
11
252
7
71
4
~
4
1
61
3
21
3
4
1
6
2
7
25
2
1
6
1
1 1
2
'In the vicinity of a nitrogen works.
Source: Stichting Doelmatig Verzinken (1959).
290 Chapter 3
rainwater is indicated by the change in the conductivity of the rainwater in
Lutterade which increased by more than a factor of 5 between and
Over or near the sea, rainwater may contain up to mg/L chloride at
times, but the average on coastal regions is mg/L. As one proceeds
inland, less chloride is found. Thus, at Scheveningen, at a distance of km
from the coast, mgL of Cl- was found in rainwater; but at km inland
the amount was mg/L and at km inland there was only mg/L. The
sulfate content is especially high in industrial districts. Over the countryside,
rainwater often contains nitrate in the range 0.5-5.0 mg/L. In the country, one
can assume a gas content in rainwater of cm3& this is analyzed as
oxygen, carbon dioxide, and nitrogen. In recent years,
more nitrogen oxides have been present, but only limited data are available.
h i s s found that the corrosion products freshly formed on zinc in
the country were rich in carbonate, but these in time became enriched in basic
zinc sulfate or chloride that had not been leached out by rain. As long as
carbonate remains present in the coating, it is protective, and inclusion of
sulfate or chloride at first increases adherence and protectivity. At a later
period, the carbonate begins to disappear and the basic sulfate is transformed
into neutral soluble sulfate, which can be washed out by rainwater. The metal
attack then recommences. In country districts the first part of the cycle had not
ended in 50 years, whereas an end point had been reached after an exposure of
years in a large city.
While the falling rain has little effect from the point of view of its
mechanical action, continued retention of rainwater leads to more severe
corrosion of zinc because of the restricted access of air: for example, in rain
gutters without sufficient fall. It has been found that rainwater dripping from
the glass panels of greenhouses increases corrosion in the area of the gal-
vanized gutters. This local corrosion is due to (a) the longer duration of
wetness and (b) the differential aeration in this small wet area. Care must also
be taken that any acidic cleaners used for the glass are fully washed to
ensure that no residues are left on the zinc. If rinsing is not complete, the
protective film of basic carbonate will not properly form and consequently
localized corrosion may occur. Appropriate design precautions should, there-
fore, be taken to prevent retention of acid water or rainwater on zinc or zinc-
coated steel parts in buildings.
In rain water of high electrical resistance, the cathodic protection of
steel by zinc extends only mm (Schafer, As the electrical
conductivity increases, however, so does the distance over which cathodic
protection is effective at scratches and edges.
D. Supply Waters
Cold: The primary factor is whether the water is scale-forming. This property
is roughly determined by calculating the Ryzner stability index (see
Section 1I.C).
Hot: in hot hard waters, the scale that forms at water temperatures above
about 55C has a coarse-grained structure and adheres less well to the
zinc surface. Corrosion of the zinc will still occur locally, as a result of
discontinuities in the scale or local electrochemical action. Above about
6O"C, zinc usually becomes cathodic to steel, and in bimetallic contacts
the steel will then corrode first. Pourbaix has performed success-
ful laboratory simulations of the corrosion of galvanized water distribu-
tion pipes.
Galvanized pipe is widely used in handling domestic hard water supplies, and
the results are satisfactory. Hard waters contain dissolved salts that may affect
the corrosion of zinc. Carbonates and bicarbonates tend to deposit protective
films that stifle corrosion; hence distilled soft waters, especially those col-
lected from rainwater, are more corrosive. Other constituents in natural
waters, such as nitrates, sulfates, and chlorides, may tend to increase corro-
sion. Their effect, however, is usually overcome in the harder waters by the
carbonates that form films of relatively low solubility in close contact with the
zinc surface. Camiglieri has reported up to reduction in corrosion
by Australian soft waters by silicate and carbonate treatments. A low pH
value and a high chloride content encourage corrosion, as does-although to a
lesser extent-a high sulfate content. Carbonates, silicates, alkali and alkaline
earth compounds, and free carbon dioxide tend to inhibit corrosion. Copper
ions, as well as ammonium ions, which complex them, tend to encourage
corrosion by the cathodic action of the elemental copper deposited on the
zinc. Elevated temperatures encourage the deposition of chalky films from
hard water, but if these are precipitated too rapidly, they may be porous and
not uniformly protective. A differential oxygen content may cause localized
corrosion, which in turn may be affected by the flow.
In general, the grade of purity of the zinc metal is not significant, nor is
the purity of zinc coatings. However, some work on minor additions to
galvanized coatings of aluminum and other elements suggests that these can
reduce corrosion in hot water, as can galvannealing the coating-the results
given later are conflicting, however. Nielsen and Yding tested for 15
months tubes of various types for carrying cold and hot water in homes, as
well as two types of fitting and two different connecting methods. In all, 42
pipes (from Germany, Hungary, Sweden, the Soviet Union, the United
292 Chapter 3
Kingdom, India, and Yugoslavia) were used in the cold water test and
pipes from the same countries in the hot water test. The conditions were as
near as possible to those found in practice. The water was fairly hard and had
a fairly high salt content, which can cause pitting in galvanized pipes. Only
pipes with a very thin zinc layer and pipes with a Zn/Fe alloy coating with no
top layer of zinc layer) showed lower corrosion protection. No differences
were observed with the fittings and the two connecting methods (galvanized
after threading and threading after galvanizing. Blanchard reports that
whereas hot dip galvanized steel pipes (galvanized to DIN did not show
differences in corrosion in water with relatively high nitrate contents, this was
not the case with galvanized pipes, air-blown on the inside. Here, the
formation of loose, fine zinc compound particles caused selective corrosion in
water with high nitrate concentrations. Galvannealed pipes showed local
c,orrosion attack, depending on water velocity.
The pipes in contact with hot water showed much more
corrosion than those carrying cold water. Stagnant hot water caused less
corrosion than flowing hot water. Corrosion was often observed on the thread
just at the end of the pipe. Also, the authors state that in such pipe corrosion
tests a fairly large number of long pipes must be selected. Tests on a ,single
pipe lead to erratic results. Earlier extensive investigation had taken place in
German waters (Muller et al.,
Quantitative corrosion rates are not of primary interest, and few data are
found in the literature outside the Russian work reported later in this chapter.
A review of early work in hot and cold water with references has been
made by Blanchard
In normal soft water the corrosion of zinc and zinc alloys is dependent
on the oxygen content, which is generally 6-10 mg/L. Hard waters usually
produce a protective film, but for water of German hardness, chloride
should not exceed mg/L and for water exceeding German hardness, it
should not exceed mg/L if the protective film is to remain; sulfate,
however, can be substantially higher. Ammonium salts are always unfavor-
able and should never exceed 20 mg/L. Copper is especially harmful and
should be as low as possible, substantially under mg/L. At temporary
hardness, zinc is attacked below pH 6.8, but in harder water of and
pH there is no attack if at least 5 mg/L oxygen is present.
The corrosion of zinc induces a more alkaline solution at the metal
surface. Unless the precipitation is so rapid that the film is porous and thus
nonprotective, such alkalinity can help to accelerate precipitation a hard
scale.
As early as 1878, experiments were carried out on zinc in four spring
waters, mainly to determine toxicity. The corrosion resistance was found to
be determined by the ratio of carbonate to chloride and sulfate. Ammonium
compounds were the most aggressive, chlorides next, and sulfates only
slightly Thus, spring water with high hardness was practically without
attack on zinc, whereas a rather pure soft spring water was aggressive
(Snyders, 1878, quoted by Wiederholt, 1965).
Bauer and Wetsel (1916), Todt (1961), and Gilbert (1948) are among
investigators of zinc corrosion in tap waters, but the quantitative losses of zinc
reported are of little significance.
The factors affecting corrosion by distilled water-namely, oxygen and
carbon dioxide content, and agitation-also affect corrosion of zinc by hard
waters. Kenworthy and Smith (1944) determined the effect of free carbon
dioxide in a hard tap water and in mixtures of equal parts of hard water and
low conductivity water. The results are given in Table 3.7. The data are also
plotted in Fig. 3.8. The data for the iron-zinc alloys shown in Table 3.7 are
for the layers under pure zinc in galvanized sheets. No significant differences
in the corrosion of the Zn-Fe alloy are evident, but the authors reported that
the alloy layers did not provide galvanic protection to the steel, despite the
apparent prevention by the zinc of pitting of the iron-zinc layers. The
behavior over longer exposure periods of galvanized steel in hard and mixed
waters containing various amounts of carbon dioxide is shown in Table 3.8.
The time for rusting is considerably reduced at high carbon dioxide contents,
perhaps because the equilibrium favors the dissolution of calcium carbonate
rather than the formation of a protective scale.
In recent years, most work on the behavior of zinc and galvanized steel
has been done in Germany. Riickert (1 979) studied for 28 months the corro-
sion effects of Berlin drinking (tap) water on galvanized steel pipes. The pH
value has a major influence on corrosion and the life of the galvanized pipes.
Below pH 6.9, corrosion attack increased. Higher pH values (7.9) favored the
formation of dense protective layers. Riickert showed that only in the early
stage of corrosion is oxygen content and flow velocity important.
In a statistical evaluation of the relationship between metal losses during
corrosion tests of up to 18 months and the testing parameters, Riickert et al.
(1987) indicate that the greater the initial corrosion rate, the greater is the
decrease in corrosion rate with time; individual parameters tested (pH value,
velocity, and oxygen concentration) do not have any significant influence.
However, tests lasting up to 10 years clearly show that high pH values result
in a significant decrease of corrosion rates with time, despite low initial rates.
294 Chapter 3
Table 3.7 Corrosion Zinc and Zinc-Iron Alloys in Waters Containing
Carbon Dioxide
CO2 in
solution Loss zinc
Pitting depth (mm)
Wateraab (parts/105) Material (mg/dm2/day) Maximum Average
Hard
Zinc
1.4 0.08 0.03
Mixed Zinc
0.8 0.10 0.03
Mixed 4.8% Fe alloy 2.5' <0.01
Mixed 7.6% Fe alloy 1.1
Hard 0.4 Zinc
2.7 0.09
Mixed 0.5 Zinc
2.7 0.12
Mixed 0.5 4.8% Fe alloy 2.6 c0.01
Mixed
0.5 7.6% Fe alloy <0.01
Hard 1.6 Zinc
4.9 0.08
Mixed 1.8 Zinc 5.0 0.07 0.05
Mixed 1.8 4.8% Fe alloy 7.0 <0.01
Mixed 1.8 7.6% Fe alloy 5.4 <0.01
Hard 3.4 Zinc 7.2 0.06 0.05
Mixed 3.7 Zinc 9.5 0.05
Mixed 3.7 4.8% Fe alloy 11.3 c0.01
Mixed 3.7 7.6% Fe alloy 9.1 <0.01
"Hard water is composed as follows (partsll05):
Temporary hardness 22.2
Permanent hardness 4.4
Lime 14.1
Total solids 40.2
Free CO2
Chloride 4.1
Sulfuric anhydride 2.97
bMixed water: equal parts hard and low conductivity water ( CO. 1 I Ow6 0").
'Attack thought to accelerated by an adhering mold growth.
Source: Kenworthy and Smith (1944).
The influence of the protective layers formed at high pH values cannot be
observed until after several years at least. Time dependence of mass losses has
a logarithmic relationship and is not influenced very much by other testing
parameters. These and other tests have led to a view in Germany that
galvanized steel should used only in water with pH exceeding 7 or possibly
7.2.
Nissing et al. (1982) exposed specimens from galvanized and black
tubes to flowing tap water with various rates of flow and oxygen concentra-
Resistance 295
CO, content of water (pant/ l O)
3.8 Effect of free carbon dioxide on the corrosion of zinc in different waters.
= Conductivity water; 0 = Hard water; = Equal mixture of hard and conduc-
tivity water (Kenworthy and Smith, 1944).
tions mg/L at pH 7.2 and 8.0) for 42 months. During this testing
period, formation of protective layers, occurrence rust, and of weight
were observed. During the first weeks, the corrosion rate was nearly constant.
Later on, corrosion followed a potential law of a bilogarithimical equation.
296 Chapter 3
Table 3.8 Corrosion of Galvanized Steel in Waters Containing Carbon
Dioxide: Tests at 18C; Varying Lengths of Exposure
in Duration Time for Total attack during test (g/m2)
Water (parts/105) (days) (days) Specimen Coating Base
Hard 0.0 42 1 >42 1 54 50 4
Hard 0.4 42 1 292 160 86 74
Hard 1.6 224 135 130 73 57
Hard 3.4 224 105 l80 92 88
Hard 20.0 39 15 296 111 185
Hard 79.0 39 16 29 1 89 202
Mixed 0.0 42 1 >42 1 44
41 3
Mixed 0.5 42 1 384 121 96 25
Mixed 1 .S 224 113 136 75 61
Mixed 3.7 224 85 198 93 105
solution test rusting
Source: Kenworthy and Smith (1944).
These two phases of corrosion depend differently on flow rate, pH, and O2
concentrations, and results obtained over a few weeks must not be used to
develop final conclusions. In oxygen-free water, increased corrosion occurs
as a result of the presence of H2C03 and HC03-. There is much less
corrosion when pH value, oxygen concentration, and flow of water are
relatively high. The influence of pH is considerably higher than the rate
flow and the oxygen concentration. Excessively high initial corrosion rates
are caused by cathodic partial reactions with hydrogen development. Oxygen
content of water depends largely on whether open or closed systems are
investigated. For enclosed systems, continuous supplies of CO2 (or HC03-)
are unavailable for corrosion reaction. Protective layers on galvanized tubes
are dense in oxygen-containing water, whereas in galvanized tubes containing
low speed, oxygen-free water, attack takes place earlier because insufficiently
thick protective layers are formed, or none at all.
Both galvanized and black (ungalvanized) tubes show an increased
corrosion attack when flow speeds around 0.5 are continuous than at
intermittent flow.
Ruckert and Sturrbecher (1988), based on 10-year tests of Berlin drink-
ing water (pH 7.5 + O S , flow speed found that protective
layers on the tubes were formed only as long as there was zinc available,
which has been the case only after these years of testing with a few tubes.
When the zinc coating is dissolved before a dense protective layer can be
Resistance Aqueous Corrosion 297
formed, galvanized tubes will show the characteristic type of corrosion of
ungalvanized tubes (i.e., pitting and porous rust). Loss of mass and time can
expressed in the formula
m = a In (bt + 1)
where constants a and b depend on pH value and flow speed, m is the loss of
mass, and t is time.
In these Berlin tests, a small increase in pH value or a decreased flow
rate caused a disproportionate improvement of the protective value of the zinc
coating and, thus, a prolonged duration of use. They also suggested it should
be possible to attain this improvement by a zinc coating slightly thicker than is
normal on tubes. Such thicker coatings normally exist on tanks and cylinders.
Much of the German work has been done by Kruse and his associates.
Tests at temperatures between and were reported in including
phosphate ion additions. In Kruse et al. reported on tests on behavior in
warm water, including the use of phosphates to inhibit the cathodic reaction.
Kruse et al. reported on cold water tests lasting up to five years,
with specimens of galvanized tubes, under conditions like those existing in
domestic installations. Three different kinds of zinc coating were used (viz.,
commercial standard tube coating as specified in DIN 2444, galvannealed
coating, and a foamy coating obtained by blowing out tubes after galvanizing)
in 11 representative Germany drinking water installations. The corrosion by
all types of water is governed by the CO2 content. Influences of zinc coating
and service conditions were less important, possibly reflecting a rather high
flow rate of the water, not considering flow conditions and contacts with
copper alloys at joints. Electrochemical tests show a restricted applicability of
the Stern-Geary equation for the water-galvanized steel system if an appro-
priate constant is used:
Rp.icom = B
where R, is polarization resistance, icon is the corrosion current derived from
the loss of mass after days exposure in flowing water, and B is a constant.
The interdependence of R, and B for 11 types of drinking water is developed
in the paper cited.
Kruse et al. tested zinc tubes and steel pipe specimens with
different zinc coatings. specimens were corroded in Dortmund tap water
in 11 test lines containing hot water at approximately and cold water,
with both discontinuous and continuous flow; in certain tests there were
occasional additions of polyphosphates or orthophosphates. One test line was
operated with warm water from a water heater, with cathodic protection of the
reservoir by the Guldager method. Visual inspection, as well as the deter-
298 Chapter
mination of mass losses and electrochemical measurements, were carried out
on all specimens, with the following results.
Localized corrosion tendency is not determined by the amount of
potential ennoblement but rather by inhibition of the cathodic reaction on the
surface formed, as shown in the cathodic current-potential curve. Cathodic
reactions can be influenced by variations both in the material and in the
quality of the water. The tubes coated with modified coatings showed results
inferior to those for the commercially galvanized tubes (DIN 2444). Zn/Fe
alloy phases do not give good corrosion protection in warm water. Phosphates
induce an inhibition of the cathodic reaction in zinc. Treatment of the water
by the Guldager process ensures strong inhibition of the cathodic reaction
with both Zn- and especially Zn-Fe alloy layers, resulting in an overall
improvement of the corrosion behavior of galvanized steel in warm water.
The presence of aluminum hydroxide in the protective zinc compounds
makes conductivity much lower than it is with the usual protective layers on
galvanized tubes. It is also emphasized that the rate of oxygen consumption
during the formation of protective layers influences corrosion. Pitting corro-
sion also occurred in cracks and under scale deposits.
Kruse et al. (1987) further reported on these tests on the attack of
galvanized piping in 11 different types of water, flowing intermittently or
permanently. The rate of loss of mass Am after 100 and lo00 days were
compared. The quotient
F =
Am after 100 days
Am after lo00 days
Rate of loss per day after 100 days
Rate of loss per day after lo00 days
i.e. ,
is generally above 1 with permanent water flow. When this flow is intermit-
tent, the factor F is approximately 1.
The Spearman rank correlation method has been used for assessing
water parameters. When no phosphates are present, these parameters have no
great influence on the change in corrosion speed with time. These speeds are
largely influenced by the CO2 contents of the 11 types of water.
The corrosivity of these waters after 5 years of testing can be divided
into three groups:
Group I: Mass losses after 100 days are rather high (> 300 g/m2) and g/m
/l00 days after days. Here no protective layers can be formed, and
consequently galvanized pipes should not be used in such cases.
Group Mass losses after 100 days are above 100 @m2; after 10oO days, Am
values increase linearly in accord with Am after days. With increas-
ing zinc losses, layers containing iron oxide are formed; finally, protec-
tive layers with rust are obtained. Generally galvanized pipes can be
used under these conditions. At relatively high mass losses after 100
days, corrosion damage in the installations may occur.
Group 3: The mass losses after 100 days are small (< 100 g/m2). In such
cases, the values for lo00 days are also small (< 100 @m2). The
zinc coating corrodes very slowly, showing the presence of zinc, even
after 5 years, on the pipes. No danger of corrosion damage will be
expected.
Ehreke and Stichel(l989) studied hot tap water at The influence
of phosphate/silicate inhibitors on general, pitting, and galvanic corrosion of
galvanized steel tubes and on the general corrosion of copper in mixed
installations was investigated. Increasing the inhibitor content reduces general
corrosion rate of both galvanized steel and copper. Only high concentrations
(5 mg/L P207 and 30 mg/L SiO,) will reduce pitting corrosion. However, not
even high concentrations of these inhibitors can fully prevent pitting corro-
sion. Galvannealed tubes are more sensitive to pitting corrosion than gal-
vanized tubes.
In hot water, the corrosion rate is not a significant measure of perfor-
mance in service, even though aggression by soft waters or formation of a
protective scale in hard waters will occur more quickly. As already men-
tioned, however, the nature of the scale may not always mean that it forms a
barrier layer. Reversal of potential so that iron corrodes in preference to zinc
is the cause of perforation corrosion.
Low oxygen and high bicarbonate, nitrite, and nitrate contents favor
reversal, while the presence of oxygen, sulfates, sulfites, chlorides, and
chlorates tends to maintain the natural state of the zinc. Reversals at tempera-
tures as low as 30C have been reported in Germany for waters containing
nitrogen compounds. The high chloride content of seawater can eliminate
potential reversal.
Water in most practical applications may be considered to be a dilute
aqueous solution of gases and salts.
Gilbert (1948) first measured the changes in potential of zinc and steel
specimens immersed in a hard London tap water (same composition as shown
in Table 3.7) during a heating-cooling cycle in which the temperature varied
from room temperature to 85C. Air or nitrogen mixed with carbon dioxide
was bubbled through the water. Reversals of polarity were observed at
temperatures above 60C if oxygen was present. In the absence of oxygen,
zinc remained anodic to steel. Currents flowing in zinc-steel couples were
measured, and the direction was also found to reverse under the same
300 Chapter
conditions. change in the zinc electrode potential was chiefly responsible
for the polarity reversals in the couples. Hoxeng and Prutton also
indicated that the reversal of potential was entirely accounted for by the zinc;
the potential of steel remained essentially within the range of V
negative to a saturated calomel electrode in most of the solutions investigated.
Gilbert also studied the reversal effect in distilled water and in synthetic
solutions made up of distilled water and some of the individual constituents
present in the tap water. Some of the results are given in Tables 1.
From the data in Table it appears that potential may on occasion be
reversed even in distilled water heated to The behavior in dilute
potassium chloride is tabulated in Table The chloride concentration is
equal to that present in the tap water used in the investigation. The zinc did
not become cathodic to steel at any time in this series. The presence of
chloride seems to limit the occurrence of potential reversal. Mixtures of
chloride with bicarbonate were also investigated. It is apparent from the data
in Table 1 that reversals in potential can occur even in very dilute sodium
bicarbonate solution when chloride is not also present. In mixtures of chloride
and bicarbonates, the tendency to reversal depends on the relative concentra-
tions of the two constituents. Bianchini et al. also studied reversal in
soft waters and various diluted aqueous solutions. Von Fraunhofer and Lu-
binski investigated tap water and salt solutions, notably bicarbonate/
chloride. Lkdion and Talbot found polarity reversal at in oxy-
Table 3.9 Behavior of Zinc-Steel Couples in Distilled Water 'Heated
to Daily
No. of days Range of
days
on which currents
Average loss
weight. (mg/dm*/day)
heated current reversed (FA) Zinc Steel
Io" 40
5
5
- 55 to
to
to
5 to '
to
"Open to laboratory air; others in CO,-free distilled water.
Source: Gilbert (1948).
Table 3.10 Behavior ZincSteel Couples in 0.00115 M KC1 Heated to 85C
Daily, in Equilibrium with Various Gaseous Atmospheres
Average loss
No. days Range of
weight
(mg/dm2/day)
on which currents
Gas mixture current reversed Zinc Steel
99% air + 1% CO2
16 25-320 49 6
98.5% air + 1.5% CO2
10 15-525 3
95% air + 5% CO2 16 70-680
43 1
Nitrogen
10 0-300 9 2
Solutions heated to daily for 8 hours and allowed to cool to overnight. Fresh water
siphoned in twice weekly.
genated water of pH containing bicarbonate or nitrate ions; Hubbard and
Shanahan also looked at the composition of the corrosion products.
Ashworth and Fairhurst studied a sodium borate-hydrochloric acid
buffer between and 90C and did not obtain the reversal.
Hoxeng and Prutton measured the potentials of zinc in various
synthetic aqueous solutions, based on distilled water and sodium salts of some
constituents present in tap water. They also found that zinc remained anodic
in oxygen-free solutions. In tests made in open beakers with electrolytes
saturated with air, however, the zinc did not always remain anodic to steel.
Therefore, tests were made in open beakers. The effects on the zinc potential
of temperature and bicarbonate, sulfate, nitrate, and chloride contents were
determined in 6-hour tests. The results were presented in numerous plots. A
few of these, showing the effect of some of the variables are presented in
Figs. Curves passing below the range of the potential of steel
indicate a reversal of the zinc potential to the cathodic side. Figure shows
the temperature effect in two different solutions. Although temperature ap-
pears to be an important factor, it is clear that the composition of the solution
may also exert a strong effect on potential reversal. This is borne out in Fig.
which shows the influence of bicarbonate concentration in the presence
of sulfate alone and with sulfate plus nitrate. The nitrate apparently counter-
acts the influence of sulfate and promotes potential reversal. Figures and
show these respective effects for sulfate and nitrate separately in the
presence of carbonate. Figure shows that chlorides also tend to prevent
the reversal.
W
0
N
Table 3.11
Containing KCl and NaHC03
Behavior of ZincSteel Couples Heated to 75 or 85C Daily in Various Solutions
Average loss
of weight
Concentration No. of No. of days (mg/dm*/day)
Concentration of NaHCO, days on which Range of
of KCl (M) (MI heated reversal occurred currenk (PA) Zinc Steel
0.001 15"
0.00115
0.00115p
0.00115
0.001 15
0.01 15
0.001 15
0.001 15*
0.001 15
0.001 15
0.001 15
0
0
0
0.01
0.01
0.005
0.005
0.001
0.001
0.001
0.0005
0.0005
o.oO01
O.ooOo5
0.001
0.0005
0.0001
17
7
17
7
17
7
8
17
7
7
7
8
9
9
1
2
3
I
0
0
0
0
0
0
0
7
3
5
-30 to >250
-100 to 400
- 112 to >250
-20 to 650
10 to >250
10 to 730
50 to >270
2 to >250
50 to 470
70 to 600
70 to 840
210 to 150
-70 to 205
-27 to 115
49 I 1
22 25
35 27
40 23
28
51 11
21 12
35 16
60 18
70 25
21 27
17 26
- ~
'Heated to 75C daily; others heated to
0
N 0.3-
1 1 1 1 1 1 1 1 1 1 1 1 1 1
0 1 2 3 4 5 6
Time (hours)
(4
0.9- f I I I I I I I I I I
I I 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1
h
W
0
v)
0
. 4
0)
m
CD
C
.-
c
0
N
0.8-
0.7-
0.6 -
0.5 -
0.4 -
0.3 -
0.2 - I
1 1 1 1 1 1 1 1 1 1 1 1 1 1
0 1 2 3 4 5 6
Time (hours)
(b)
Fig. 3.9 Effect of temperature on zinc potential in two solutions: (a) 115-140 ppm
HCO,, 10 ppm SO.,, and 10 ppm NO3 (added as sodium salts); (b) 35-45 ppm HC03
and 5 ppm NO3 (added as sodium salts) (Hoxeng and Prutton, 1949).
304 Chapter 3
Time (hours)
1 1 1 1 1 l 1 1 1 1 1 1 1 ~
0 1 2 3 4 6 6
Time (hours)
Fig. 3.1 1 ' Effect of sulfate on zinc potential at 60C in a solution composed of 125-
135 ppm HCO, and SO, as indicated (added as sodium salts) (Hoxeng and Prutton,
1949).
Time (hours)
Fig.. 3.1 2 Effect of nitrate on zinc potential at in a solution consisting of 130-
135 ppm HCO, and nitrate as indicated (added as sodium salts) (Hoxeng and Prutton,
1 949).
306 Chapter
l .o-
0.9 -
"
I l l l l l l l l l l l l l
0 l 2 3 4 5
(hours)
3.1 3 Effect of chloride on zinc potential at in a solution composed of 40
ppm HC03, 5 ppm SO4, 10 ppm NO3, and Cl as indicated (added as sodium salts)
(Hoxeng and Prutton, 1949).
Hoxeng (1950) extended this investigation to determine the effect such
intentional additions to the water as silicate inhibitors might have on the
potential of the zinc measured after 96 hours. Both calcium and silica
additions tended to make the potentials more anodic. shown in Fig. 3.14,
the combination of these two components is more effective than silica alone.
The authors concluded that the final behavior of zinc will depend on the
combined effect of the various components of the water.
Weast et al. (1961) contributed to this subject by measuring electrode
potentials of zinc and steel in aqueous systems under pressure according to the
procedure of Hoxeng and hutton. It was found that atm of oxygen pressure
with saturated CaC03 + sulphate or chloride made zinc more notable (0.50-
0.65 V, vs. 0.88 V at atmospheric pressure with iron usually V).
With hot water tanks, the bottom is more liable to attack than the sides
because deposits on the bottom (such as iron filings from holes drilled
elsewhere in the produce bimetallic or differential aeration cells. Such
cells also can form on the sides in some waters, especially near the waterline,
however. If the water is made more aggressive (e.g., by the presence of
copper salts, possibly dissolved from elsewhere in the water system), corro-
sion cells will form more readily. The presence in water used in domestic
Temperature "C
al
.-
m
al
v)
>
-
.-
-
m
al
a
"
-
al
.-
m
C
40 50 60 70 80
I I I I
1.1 -
1.0-
0-9-
0. 8-
I
0.7 -
0. 6 -
0.5-
0. 4 -
I I I I I
104 122 140 158 176
Temperature OF
Temperature "C
40 50 60 70 80
I I I I
1.1 -
1 .o-
0.9 -
0.8-
0.7 -
0.6 -
0.5-
0. 4 -
I I I I
1.1 -
1 .o-
0.9 -
0.8-
0.7 -
0.6 -
0.5-
0. 4 -
I I I I I
104 122 140 158 176
Temperature O F
307
308 Chapter
heating systems of copper ions at a concentration of mglL Cu2+ may
cause severe pitting of zinc; cold water systems also will experience pitting,
but it will be much less severe. Installing a new tank in an old water system
can give trouble in borderline cases. Practical experience is that if a gal-
vanized steel hot water tank or cylinder lasts more than 4 years it will usually
last at least years, because a protective scale has formed. Areas where
failures occur in a few years are, of course, not suitable for galvanized water
systems. The British Non-Ferrous Metals Research Association reported that
a magnesium anode installed with a galvanized tank or cylinder in areas where
the water is on the borderline of suitability allows time for a protective film to
form (BNFMRA, The Mg2+ ions will precipitate on the inside the
pipes and stabilize the coating of corrosion products. It is not usual to renew
the anode since, by the time it has been consumed, a scale will have formed in
suitable waters. It is also possible to treat soft waters to give them a scale-
forming capacity (Iovchev and Mihailov, as sgown when Sofia's soft,
acid water received Neutralite treatment. The detailed research on effect of
gases in solution and variation in water compositions has been well docu-
mented in the literature (e.g., Wiederholt, and in subsequent annual
surveys of the literature by H.-J. Bottcher.
Work in the Soviet Union concentrated on the need for additional
protection of steel in hot water systems. Work on corrosion behavior of
galvanized and ungalvanized pipes at in three of aggressive
water was reported by Lapotyshkina and Sazonov in their book on
water treatment. They showed in tests of up to months' duration that the
rate of pitting decreased with time to mdyear (depending on the
water-ither just scale-forming or soft and not scale-forming) after years in
one series tests. But results of another test indicated that the depth of pitting
could be greater on galvanized steel than on baie steel (0.25 mdyear vs.
&year) after years, presumably as a result of localization of pitting due
to polarity reversal.
Proskurkin et al. considered six types of supply water and their
corrosivity to bare steel (Table and then tested various coated steels in
waters from four towns (including two of those in Table The results,
summarized in Tables and show marked improvement when
was in the galvanizing bath used in the production of galvarkealed
coatings. Table gives the diffusion annealing conditions Proskurkin
foundmost favorable for different solutions. Laboratory tests at British Non-
Ferrous Metals Research Association in London water, however, ' did not
show the same degree of improvement with aluminum or galvannealed sam-
ples, and Akimov et al. speak of lower corrosion resistance of coatings
Table 3.12 Typical Compositions of Supply Water in the Soviet Union and Their General Aggressivity
Total Total Chemical content (mgkg) Corrosion
Water salt hardness Saturation activity of
index at waters at supply content (mg (mg Alkalinity
Town source equivkg) equiv/kg) (mg equivntg) Chlorides Sulfates Oxygen (carbon arid). pHa pHb T = 60C T = 60C
FW CO?
Moscow River 222.4 3.96 '2.64 15.0 10.1 9.67 13.5 7.3 7.025 +0.275 Weakly
Dnepro- River 230.8 2.95 1. 1 17.0 10.3 16.0 7.17 7.7 -0.57 Aggressive
Leningrad River 65.0 0.85 0.57 7.8 15.5 10.0 10.0 7.0 8.5 -1.5 Strongly
Moscow River 148.3 2.1 1.8 5.0 4.0 11.3 18.8 6.9 7.3 -0.4 Aggressive
Riga Lake 115.0 4.2 1.6 20.0 60.0 6.0 29.0 6.8 7.4 -0.5 Aggressive
Perm River 192.0 0.96 0.5 147.0 50.8 14.0 28.1 6.9 8.0 -1.1 Strongly
'According to analysis of water of given composition.
bCalculated pH at which carbonic acid compounds are in equilibrium. Details in original paper.
Source: Proskurkin et al. (1981).
Moscow aggressive
petrovsk Dnieper
Neva aggressive
Volga
aggressive
Table 3.13 Corrosion Rates for Galvanized and Aluminized Steel in Four Waters of the Soviet Union
Galvanized specimens
Duration of testing (hours)
Of steel Specimens galvanized subsequently heat- Aluminized specimens
Act ual ht he soecimens with additions treated at 500C for 10 (coated in a bath
Water Total given regime) (uncoated) of 0.11-0.15% Al minutes of molten Al)
Dnieprovsk 8,O00 3,000 0.460 0.042 0.011 0.003
Moskovmnts 24,300 3,000 0.207 0.017 0.009 0.016
Volzhsk 24,300 3 ,m 0.280 0.028 0.020 0.180
Nevsk 24,300 3,000 1.01 Pitting corrosjon of the coating 0. lW
Total duration of testing 10,800 hours; actual duration, u)oo hours.
Resistance to Aqueous Corrosion 31 1
Table 3.14 Results from Corrosion Tests in Running hi eper River Mains
Water at 60C for hours on Specimens with Zinc Coatings Obtained
by Various Methods
Corrosion Depth of
Method of deposition of zinc coating rate (g/m2/h) corrosion (pm)
Sherardizing
Hot dip galvanizing with addition
of aluminum
Electrolytic
Galvanizing with treatment at 55O"C,
minutes; addition of aluminum
minutes
Electrolytic and heat treatment at 5OO"C,
with aluminum additions when subsequently annealed, as a result of increase
of porosity, redistribution of alloying elements, and zinc oxide formation.
Proskurkin emphasizes that control at water temperatures of 55 5C
is desirable, to ensure the best performance. He felt, however, that for some
aggressive waters (presumably soft waters), ename1ed"pipes should be used.
Sharivker and Tomin (1988) recommended zinc-coated steel for heat ex-
changer pipes up to
Proskurkin et al. (1983) then tested other additions to the galvanizing
bath. They found (as did the British Non-Ferrous Metals Research Associa-
tion) that alloying with magnesium (0.014-0.085%) tended to localize corro-
sion. A coating with 0.03412% A1 and 0.14.4% Ni showed after 18
months (presumably at 60C) in water that the rate of corrosion was about 15
p-&year-although from other work the rate would be expected to decrease
with time. The q phase had disappeared and the 5 phase was being attacked.
In 1983 no recommendations to use any specific alloying element in galvaniz-
ing tubes for hot water service in the Soviet Union' appeared to be given. A
minimum inner zinc coating thickness of 50 p-m is proposed.
Sukhomlin et al. (1986) recommend diffusion galvanizing (or aluminiz-
ing) for tubes in the circulating water of refrigeration units.
For closed water systems, inhibition can be used, but formation of a
passivating coating is preferred. After research work by Dreulle and Dreulle
31 2 Chapter
Table 3.15 Diffusion Annealing Conditions Securing the Production of the Most
Corrosion-Resistant Zinc Coatings
Method of Annealing Holding
heating or temperature time
Corrosive medium Method of coating annealing"
("C) (min)
Variable medium Bath with 0.04% AI
NaCl
solution-air)
Bath with AI
Electrolytic
Flowing m/s) Bath with 0.04 AI
seawater
Bath with AI
Hot running mains Bath with 0.04% AI
water (60C)
Bath with 0.12% AI
Electrolytic
R
I
R
I
R
R
I
R
I
R
R
R
IO
IO
10
R = radiation heating; I = induction heating.
(1976), ILZRO patented a method for protecting galvanized steel used in
fabricating domestic hot water systems, electric water heaters, and distribu-
tion piping. This method is based on the formation of a passivating coat of
zinc pyrophosphate (Zn2P207) by contact with a solution at 60C based on
polyphosphates and metasilicates containing traces of nickel chloride as
catalyst. This compound, partly formed at the expense of the zinc coating, is
insoluble and a good electrical insulator due to its covalent property. More-
over, it prevents the formation of electrochemical cells, hence prevents attack
on the zinc coating. The pyrophosphate coating can be applied by either a
chemical or an electrochemical method, the former being essential for hollow
parts. Industrial trials with electric water heaters are described by Rosset and
Jardy (1982).
The aggressive non-scale-forming Point Claire tap water (Noranda,
1991) showed only modest differences in the rate of weight loss for the
different ZA alloys at 50-90C. No intergranular attack occurred at 60C or
below, but metallographic examination of the ZA alloys tested at 90C
showed intergranular attack, greatest on ZA27 at 50 pm in 45 days. The
maximum rate of general attack 60C at may be related to the direct
formation of zinc oxide above this temperature and also to the ease of escape
of air or oxygen from the water at higher temperatures. In general, the ZA
alloys are not recommended for water service above about 70C.
E. Natural Waters
Rivers, lakes, and other natural waters are of the same general type as supply
waters, but composition may vary more and the content of dissolved salts can
be greater. Some test results are given in Table 3.16, but no data are available
to indicate whether the corrosion rate was decreasing with time. Figure 3.15
suggests that in some river waters (presumably where no protective scale
forms) the rate falls off only slightly with time.
Short-term tests by Bauer and Schikorr (1934) on a simulated river
water, hardness 7.6" (German), showed a corrosion rate of around 0.7 km in
the first 20 days but only about a quarter of that in the next 40 days. The same
tests showed about 10% lower corrosion rates for 99.98% zinc compared with
99% zinc but, even here, purity of zinc again is of far less significance than
the nature of the water. The effect of live bacteria in reducing pitting
corrosion in fresh water has been shown by Guillaume et al. (1977a, b), but
corrosion increased when dead bacteria were present; bacteria in seawater
were also investigated.
F. Industrial Process Waters
The compositional variations in industrial process waters are too great to
permit any general guidance to be given. An in situ test, preferably for at least
a year, is desirable, although indicative results often can be obtained in a few
months.
Zinc-coated steel is increasingly used in situations that otherwise would
be too aggressive to steel. Five-year trials in the agricultural industry, partic-
ularly in animal waste environments, showed (Kim, 1991) the superior
corrosion resistance of galvanizing after fabrication, compared to epoxy-
coated steel.
Trials in a North American municipal wastewater treatment plant have
shown variations in zinc corrosion rates in the sedimentation tanks ranging
from 28 to more than 140 pdyear , whereas 85 pm zinc coatings in nomi-
Table 3.16 Corrosion of Zinc and Zinc-Coated Steel Immersed in Natural Fresh Water
Conusion rate
h a t i o n and water Type of zinc YearS Agitation (Pm/year) Remarks
Gatun Lake, Canal Zone
Tropical fresh water
Gatun Lake, Canal Zone
Tmpical fresh water
Pedro Miguel locks, Panama
Fresh water
Fresh water
Fresh water
Fresh water
Fresh water
Ampari River, Brazil
Fresh water
Amazon River, Brazil
Fresh water
Intermediate
Intermediate
Special high
High grade
grade
Intermediate
Select
prime western
4
8
5
5
5
5
5
0.4
0.45
stagnant 13
stagnant 10
18
13
11
14
13
2.8 Ws flow I37
2.2 WS flow 198
Deepest pit 0.0222 in.
Deepest pit 0.19 in.
International Nickel
(0.56 mm)
(0.48 mm)
At iron ore mine:
At manganese ore mine:
5-
4-
2-
1-
a-
I I I l I I l I I
a 2 3 4 5
6 7
Years exposed
3.1 5 Corrosion-time curves for freshwater and seawater in the Panama Canal
zone: Gatun Lake fresh water (immersed); 0, Pacific seawater (immersed);
Pacific seawater (mean tide) (Forgeson et al., 1958).
nally similar locations in the United Kingdom were in excellent condition
after 7 years, with substantial residual zinc. The North American experience
did show that the ZA alloys may be better than pure zinc in, for example, the
secondary sedimentation tank, but additional protection should be provided
(Belisle and DuFresne, 1986) unless-as in the United Kingdom-there is
specific satisfactory performance in a comparable water.
Allen (1991), working in South Africa, said that the use mechanized
stoping methods for deep level gold mines in South Africa, developed to
increase productivity and efficiency, has resulted in the introduction a
fundamentally new industry. Machines powered hydraulically, using mine
water fed from the surface, are constructed of new engineering materials,
resistant to both abrasion and corrosion and universally usable for pipework
and components in mining machinery. It has been estimated that the lifetime
of galvanized steel pipework is in excess of 15 years. Allens work showed
that a mine water solution temperature between 20 and has only a
minimal effect on the predicted lifetime. Conversely, turbulent conditions in
mine water lead to a significant increase in corrosion rate. The presence of
adherent corrosion products on the surface of the coating has a major influ-
31 6 Chapter
ence in determining the corrosion rates in the two synthetic solutions investi-
gated. Extrapolating Allens results, the coating lifetimes can be predicted to
be greater than 20 years and the service life of pipes still greater, owing to the
allowable general corrosion of the underlying steel.
However, in coal mine water in Russia (Kudreiko and Agarev, a
diffusion zinc coating containing 5 or aluminum apparently gave high
corrosion rates of 400-500 and pdyear (presumably a soft, acid
water). In these circumstances, increasing the aluminum reduced corrosion by
encouraging the formation of an Fe-AI compound.
G. Seawater and Brackish Water
A broad distinction must be made between seawater in the tropics and that in
more temperate climates. Temperate seawater is much less corrosive because
the Ocean temperature is much lower and the magnesium salts are therefore
sufficient to act as inhibitors. There is a gradual change from one condition to
the other. Harbor waters in India and other tropical areas were compared by
Deshmukh local biofouling affects results. Tchernov and Ilyin
regard calcium carbonate content as more significant than temperature in the
Indian Ocean. Higher corrosion rates are likely in polluted seawater.
Typical corrosion rates (Table are 25 pdyear in the tropics and
half that in temperate seas, such as the North Atlantic. In tidal areas (i.e.,
immersed in seawater twice daily), or areas washed by seawater waves, the
corrosion rate is typically twice as great as for completely immersed areas,
especially if sea spray is present. Sixteen-year results for Panama are reported
by Southwell et al. Seawater flow rates have negligible influence on
corrosion rate in straight tubes (Yandushkin and Kuris, unlike copper-
base tubes. Zinc alloy castings often will give satisfactory service even where
a zinc-coated steel might have insufficient life.
The high chloride content of seawater would normally tend to corrode
zinc, but the presence of magnesium and calcium ions partially inhibits
attack, which is considerably moderated by the protective films (of magne-
sium and calcium hydroxide, carbonate, basic chloride, etc.) that are formed,
particularly in temperate waters. Thus, water in the North Sea is much less
aggressive to zinc than pure chloride solutions; entirely fallacious data can be
obtained if pure chloride solutions are used in tests to simulate the corrosion
resistance of seawater. Schikorr demonstrated the difference between
the corrosive effect of sodium chloride and of several components of sea-
water. The tests were designed to simulate a sea climate involving alternate
wetting and drying at various rates and, therefore, may not be representative
of continuous immersion conditions. The data in Table however, show
d d m
31 7
3
0
U
S
2
a
4
3
2
G
d
8
c
31
5; m
31
320 Chapter 3
Table 3.18 Corrosion of Zinc in Solutions of Various Components of Seawater
Exposure cycle: Immersed 5 minutes every 2 hours; rest of time in open air during 8 hours
daily and enclosed for the remaining 16 hours.
Weight loss (@m2)
Solution 2 days l days days
g/L NaCl
19.0 96 198
g/L NaCl + 10 g/L Na,S04 10H,O 26.0 150
g/L NaCl + 12 g/L MgCl, - 6H,O 1 .o 2
g/L Na2S0, - 10H20 2.0 11 22
12 g/L MgCI2 - 6H20
Artificial North Sea wateP 1 5
Tomposition:
28.4 g NaCl
8.2 g MgSO, 7Hz0
4. 3 g MgClz 6HzO
2.0 g CaCI2 6Hz0
Source:
the strong corrosion-inhibiting effect magnesium in a sodium chloride
solution.
North Sea water typically contains 30 g NaCl, 4 g MgS04, 2 g MgC12,
and 1 g CaC12 per liter. The pH values are between and 8.1. The
accelerating action of Cl- ions on corrosion is retarded by Mg2" ions.
Corrosion losses zinc in seawater (Van Eijnsbergen, SDV, 1959a) are as
follows:
Water containing
Corrosion loss (pm)
(per liter) (after 2 weeks immersion)
g NaCl 28
g NaCl + 10 g Na,S04 44
g NaCl + g MgC1, 0.6
g MgCI,
Artificial seawater
10 g Na,S04
In practice, galvanized parts of locks were free of rust after 11 years in
brackish water. Galvanized pilings in seawater resisted rusting for years.
Resistance 321
On offshore platforms, duplex systems (galvanizing plus a three-coat non-
saponifiable paint system) have protected against rust for 10-20 years. A test
site on Heligoland run by a German steel association provides more data
(Schwenk and Friehe, 1972).
Table 3.19 gives some different views on compositions of artificial
seawater for test purposes (in relation to actual North Sea water), and Table
3.20 gives the percentages of salt in seas and in lakes. Most tests using
synthetic seawater have been of short duration. 14-day test, for example,
gave a loss of about 3.5 pm, which, on a straight-line extrapolation, would
give 90 pm a year. Because corrosion rates in stagnant immersion fall off
sharply with time, however, this test probably would indicate no more than
Table 3.19 Chemical Composition of Artificial Seawater (g/L)
~~
Ocean
Artificial seawater seawater Artificial North
Constituent (DIN 50 (ASTM, Sea water North Sea water
or property November 1960) D1141/52)" (Schikorr, 1940) (SDV, 1959)
PH
NaCl
MgSO, 7H20
CaSO, 2H20
Na2S04
MgCI, 6H,O
KC1
CaCI, 6H,O
NaHCO,
MgSO.4
PH
Solids
Oxygen
CaO
MgO
c1 -
7-8 8.2 8.2 7.6-8.1
7
28 24.53 28.4 30
4.7
4.09
0.70
5 5.20 4.3 2
2.4 1.16 2 1
0.2 0.201
Seawater according to
Bauer and Pacific Ocean
Schikorr
(Forgeson et al. Corr 13
d'Ans Lax ( 1934)
1958) (1957)
8.2
1.6 0.06 cm'/m~
36.594 35.83 26.4
0.5 0.274 0.400
2.1 2.288 1.272
19.3 18.15 17.4 18.98
2.8 2.119 2.4 2.649
'In addition, KBr 0.101, H3B03 0.027, SKI2 0.025, NaF KC1 0.695, and Na2S0, 4.09
322 Chapter
Table 3.20 Salt Content of the Oceans and Some
Salt Lakes"
Sea or lake
Ocean and North Sea
Baltic Sea
Mediterranean Sea
Caspian Sea
Dead Sea
Great Salt Lake, Utah
Borax Lake, California
Salt Lake, Thebes, Egypt
Salt content (%)
and quantity of salts in residue of Ocean seawater:
Ion Percent
- 55.29
Br- 0.19
S@- 7.69
c0:- 0.21
Na+ 30.58
K+ 1. 11
Ca2 + l .2
Mg2 + 3.73
Source: Klas and Steinrath (1956).
about one-seventh this rate over 5 years in a natural seawater as simulated by
the test composition.
Proskurkin et al. showed that heat treatment of hot dip zinc
coatings noticeably improves their corrosion resistance in flowing seawater
(Table and Fig. However, it would seem that the corrosion rate is
too great to permit galvanized steel to be used in this flowing seawater.
Results Proskurkin et al. summarized in Table are qualitatively
similar for Black Sea water tested over year but quantitatively lower,
suggesting that the results may represent over a period of less than
months. The results show good performance of unalloyed coatings and
still better performance low alloy coatings in intermittent immersion
(details not available) in Caspian Sea water.
Earlier work by Proskurkin et al. at the Odessa Naval Scientific
Research Station, where seawater was pumped at a pressure of kg/cm2 at 4
&cm through heat-treated galvanized pipes, gave a corrosion rate of
~ pm, compared with pm on the outside, exposed to the Odessa marine
n
%
a
B
E.
ni
n
m
Table 3.21 influence of Heat Treatment on the Rate of Corrosion in Flowing (3 d s ) Seawater of Zinc Coatings Produced
a
c
from Melts Containing Various Additions of Aluminum
$
No. of melt (%) Method of heating Time (min) ("(3 ( Wyear) g
Rate of v,
Heat treatment
corrosion n 2
A1 content Temperature
0
140.9 a
1 0.04 Without heat treatment
2 0.04 Furnace 10 500 78.6
3 0.04 Fumace 10 550 84.6
4 0.04 Induction 10 550 88.4
5 0.12 Without heat treatment 188.5
7 0.12 Furnace 10 500 65.1
8 0. 12 Furnace 30 500 103.8
9 0.12 Induction 5 550 79.8
10 0.12 Furnace 10 550 88.8
11 0.12 Induction 10 550 84.6
6 0.12 Induction 5 500 77.9
Source: Proskurkin et al. (1973b).
W
N
W
324
200

m Q X
1
Chapter 3
Fig. 3.16 Corrosion in flowing seawater for various types of galvanized coating,
showing the effect of diffusion annealing on the corrosion resistance in flowing
seawater for coatings obtained in a zinc bath with (I, 111, IV) and A1 (11,
V-X). Length of test, 3600 hours. Coatings I and I1 had not been heat-treated.
Coatings IV-X had been heat-treated (1) in a furnace, and (2) in an inductor (Pros-
kurkin, 1973b).
atmosphere. It is presumed these are corrosion rates per year. The diffusion
heat treatment apparently was applied after fabrication.
Akerdida and Lyon (1988) incidentally used continuous immersion
potentiodynamic polarization tests, and weight loss and ac impedence tests, in
intermittent immersion to compare heat-treated (45 min at 450C) and non-
heat-treated coatings in 0.6 M NaCl. They found that the galvannealed
coatings (in 8-day tests) corroded twice as fast in intermittent immersion in
the NaCl solution. Analysis by ac impedence indicated that a passive film
probably consisting of zinc hydroxy compounds was more effective on zinc
than on the iron-zinc alloy.
The behavior of zinc-iron alloys certainly requires clarification: it may
that the particular alloy or the structure affects the performance.
Biological processes such as fouling may play a special role in
Q1
t
325
326 Chapter
corrosion by seawater, but corrosion of zinc is frequently beneficial in
discouraging the deposition of algae, barnacles, mussels, and on (Hudson
and Kenworthy, 1957; Romanovsky, 1962), since zinc salts are poisonous to
these life forms, although less than compounds of mercury, copper, or
arsenic. Sulfides in seawater increase corrosion at pH values exceeding 7.2
but cause partial inhibition at lower pH values (Mor et al., 1974).
Most information available from tests relates to continuous immersion,
and the rate appears to decrease with time (Fig. 3.15). Only limited data are
available on the effect of depth of immersion in the sea. Van Eijnsbergen has
spoken of tests in the North Sea that showed up to twice the corrosion rate
hundreds of meters below the surface. Larsen et al. (1989) tested a number of
common alloys in slowly flowing Hawaiian surface seawater and cold, deep
seawater pumped from a 590 m depth for periods of 1-10 months. The cold
water had relatively low pH and oxygen contents and was much more
aggressive than the warm surface water toward zinc and galvanized steel, but
much less aggressive toward carbon steel (e.g., 1009, 1090), low alloy steel
(e.g., 4130, 4340), and stainless steel (e.g., 304, 316 SS). Corrosion rates
decreased rapidly with time during the early stages of exposure. For the warm
water, the decrease was more rapid than reported from other sites, possibly
because these other locations were not similarly sheltered from waves, cur-
rents, and macrobiofouling.
Reinhart (1976) has reported on the use of zinc alloys and zinc wire
ropes that were exposed in the Pacific Ocean at depths of 720-2070 m for
periods varying from 123 to 1064 days. The zinc alloy compositiotl was
99.9% zinc, 0.9% lead, and 0.1% iron. The wire ropes were galvanized steel
cables of various The data obtained from the study are given in Tables
3.23 and 3.24. From the data shown in Table 3.23, the corrosion rate of zinc
in Pacific Ocean seawater is seen to decrease with the duration of exposure,
except for zinc at the 2400 ft depth, at which the corrosion rate increased with
increasing time of exposure. Also, the corrosion of zinc was greater at depth
than at the surface. In addition, the report indicated that the corrosion of zinc
was not uniformly influenced by changes in the concentration of oxygen in
seawater between the limits of mL/L.
Steel wire ropes with different amounts of zinc coating (galvanizing)
and different types of construction were exposed in Pacific Ocean seawater at
depth to determine their corrosion behavior. Some were stressed in tension to
determine their susceptibility to corrosion or whether increased
their rates of corrosion. The corrosion behavior of these galvanized steel wire
ropes is given in Table 3.24.
The galvanized ropes were numbered 1, 2, 3, and 7-22. The zinc
Table 3.23 Corrosion of Zinc" in Pacific Ocean Seawater Exposed
at the Surface and at Various Depths for up to Yearsb
Corrosion
Exposure Depth Rate
Environment' (days) (m) (PdYe=) Typed.'
W P
S P
W
C
S 5 .
W
90
W 60 CR, E
S
W
60 p (50)
S
W
GASL
W
1 .S
P
W P
99.9% Zn, 0.09% Pb, and Fe.
bFor details the effects salinity, temperature, and oxygen on corrosion rates, see table
source.
'W = totally exposed in seawater on sides structure; S = exposed in base of structure that
the lower portions the specimens were embedded in the bottom sediment.
dSymbols for types of corrosion:
P pitting
C crevice
CR cratering
PR perforated
G general
E etched
GASL general above sediment line
'Numbers in parentheses indicate maximum depth (pm).
Source: Reinhart (1976).
coatings protected the steel wire rope, but it was not possible to correlate the
weight or thickness of the zinc coating with the duration of corrosion protec-
tion. However, except for electrogalvanized coatings, it is possible to general-
ize that the heavier the coating, the longer the period of time before rust
appeared on the ropes. The breaking strength of the ropes was not impaired by
seawater exposure for as long as 1064 days. Interestingly, the ropes numbered
328
Chapter 3
W
+
S
l-
l-
?!
2
m
I
g
W
x
?L
i?
I
$3
l-
W
x
?L
i?
I
W
+
S
l-
W
330 Chapter 3
19, 20, 21, and 22 were not susceptible to stress corrosion when stressed to
20% of their respective breaking strengths.
The wire ropes numbered 4, 5, 6, 23, and 24, in addition to being
galvanized, were jacketed with plastic coating. In all these cases, the seawater
penetrated along the interface between the ropes and the jackets. There was
some light rust on ropes 23 and 24 beneath the polyvinyl chloride jacket after
751 days of exposure.
Bauer and Schikorr (1934) found a mean.weight loss of 3 pdmont h in
a t i al under stagnant conditions for 60 days in North Sea water. This value
was about three times greater than in river water, but only a quarter of that in
distilled water. In alternate immersion tests in North Sea water, the mean
weight loss for 60 days was 25% higher. Alternate immersion without sea
spray effects does not necessarily give higher rates of corrosion (see Table
3.17, Canal Zone). Zielinski and Speichert (1986) found zinc and aluminum
hot dipped coatings similar in performance in stagnant seawater, but zinc
corroded twice as fast as aluminum in flowing seawater. Van Eijnsbergen
(Table 3.25) has compared the performance of various coatings on steel
pilings in the North Sea.
Gibbs and coworkers (1916) found, for fresh still U.S. seawater at
50C, a weight increase up to 5 days exposure and only thereafter a weight
decrease, which after 3 weeks amounted to about 1.4 pdmonth.
Mor and Beccaria (1975) found that for zinc in continually aerated
synthetic seawater (5 mUmin of dry air containing 0.1% of COz; pH 8.2;
stirred with a magnetic bar at 240 rpm) and at 24,40, and 60C, the corrosion
of the zinc decreased with increasing temperature (Fig. 3.17). It should be
noted, however, that the dissolved oxygen content also decreased with tem-
perature.
Nineteen-year tests (American Welding Society, 1974; Longo and Dur-
mann, ASTM 1974) showed complete protection of steel by 300 pm unsealed
sprayed zinc in seawater.
Diegle and Kelley (1981) have compared Zn, Zn-5% Al, Zn-15% Al,
and A1 sprayed coatings in synthetic seawater for 135 days, but all provided
good protection; in salt spray, Zn-15% A1 was the best. With overcoatings of
vinyl or polyurethane, no undercutting or blistering occurred for any of the
paints or substrates: polyurethane was generally more resistant to blistering
than vinyl; only over zinc did polyurethane blister.
Shaw and (1.985) and Shaw et al. (1987) found that zinc-
(nominal) 15%-aluminum pseudoalloy sprayed coatings were better than zinc
or aluminum in splash and spray conditions-and also better than coatings of
the same nominal composition made from prealloyed material. However,
Corrosion 331
Table 3.25 Corrosion Rates of Steel Pilings After Years in Seawater
Corrosion rate (padyear) at depth
Coating system Overall ma mb 5-9 mc md
Galvanized steel
Flame-sprayed aluminum
Polyester-glass flake
Coal tar-epoxy
Aluminum-pigmented
coal tar epoxy
Zinc-rich primer + coal
tar epoxy
Flame-sprayed aluminum
+ vinyl
Zinc-rich primer +
epoxy-polyamide
Bare carbon steel
Phenolic mastic
Sprayed zinc +
Zinc-rich paint + vinyl
polyvinylidene
'Embedded in mud.
bErosion zone.
'Immersion zone.
dAtmospheric zone.
Source: Van Eijnsbergen (personal communication).
analysis showed higher aluminum (up to 60%) in the pseudoalloy coatings
than specified. The pseudoalloy also gave relatively good performance in 6
months' total immersion (showing only small hard nodules containing corro-
sion product), compared with significant biistering and flaking in the coatings
produced from prealloyed material. The report concludes that for good perfor-
mance, more consistent coatings (than those available for use by Shaw et al.)
are needed, with even distribution,of phases and a more coherent aluminum-
rich phase.
basic study by Bohm and Willert (1977) on hot dip galvanized zinc
coatings on steel in flowing seawater showed that the attack the zinc and
time dependence of seawater corrosion of the hot dip galvanized steel was
influenced by the structure of the galvanized'layer. Fanning out of the 5 phase
is to be avoided; even growth of each:layer is desirable.
332
Chapter
d
p
m-
I
100
I
200 300
I
400
I
rime. hours
3.17 Corrosion in synthetic seawater at different temperatures and oxygen
concentrations (Mor and Beccaria, 1975).
When polluted sewer water has pH values between 3.5 and 5.0, a
duplex system is preferable to galvanizing alone (Van Eijnsbergen, 1986b)
because pollutants, acid compounds, and the generally low oxygen content
prevent the formation of the protective zinc patina. Contact between gal-
vanized parts (e.g., railings, fences, staircases, ladders) and ungalvanized
steel must be avoided, since differences in potential of 100-300 mV can
provoke local, often pitting, corrosion. Very resistant coatings on galvanized
Resistance
parts, exposed to very aggressive effluent water, are based on epoxy resin-
coal tar combinations with dry film thicknesses of 800-1200 Km. In other
cases, a paint system with a dry film thickness of 100-150 pm will be
adequate.
Work in Spain (Guillen and Feliu, 1970) on painted, thermally sprayed
zinc immersed in seawater showed that epoxy resin and polyurethane paints
performed especially well.
ZINC ANODES IN WATER
The general principles of protection by the use of zinc anodes were discussed
in Chapter 1, together with the composition specifications, methods of attach-
ment, design calculations, and uses in conjunction with barrier coatings.
Wingaard, of the Dutch Institute for Applied Research ("NO), discusses
(197 1) the choice of galvanic anodes for protection of steel structures in
seawater and shows that zinc has the greatest efficiency. The importance of
adequate electrical continuity was demonstrated by Drisko (1970), who found
that a cable joined periodically ( 15 m maximum gap) to one of the ground legs
was needed to ensure that the specially cast zinc anodes gave cathodic
protection of the ground tackle of a fleet mooring. Subsequently modified
designs were advocated (Drisko, 1974).
Zinc anodes are consumed in seawater at a rate of about 12 kg1A-year.
On bare steel, the average current density is about 0.10 Mm2, to protect
100 m2 of bare steel requires about 120 kglyear of zinc anodes. Heinrich
(1980) refers to a theoretical capacity of 820 AWkg and an efficiency of 95%
in seawater, giving a practical capacity of 780 AWkg (i.e., 0.089 kyearlkg,
and close to the figure quoted earlier. However, in the North Sea, higher
currents (up to 0.13 Mm2) may be needed than in the Gulf of Mexico
(Eliassen and Steensland, 1977).
The potential of zinc of - 1.1 V, though often disadvantageous in high
resistivity environments, is very desirable in low resistivity systems such as
seawater, marshes, and other similar electrolytes. This voltage is below the
generally recognized value of the hydrogen overvoltage of steel of - 1.2 V,
the level at which hydrogen is liberated at the steel cathode. Since zinc cannot
polarize any structure to a more electronegative value than its own potential of
- 1.1 V, there is little or no danger of hydrogen liberation at the cathode. This
is an advantage, since if hydrogen is so liberated, it can damage other coatings
on structures such as pipelines or ship hulls.
Fully or partially immersed structures, including piers, harbors, and oil
rigs, make use of steel pilings and shuttering that in general cannot be
334 Chapter
removed for painting and coating. In many cases, the pilings penetrate deeply
into mud as well as being exposed to seawater, and the useful lifetime of the
structure is determined by the rate of corrosion in those areas. Cathodic
protection is very successful, especially around the waterline, when zinc
anodes can be suspended alongside the structure or fixed to it.
In the case of pipes in abyssal seawater or deep mud, a good painting
scheme is applied in the factory using a chemically resistant paint, and zinc
anodes are placed at intervals along the pipe before construction. These may
be in the form of bracelets or attached by welding to the pipe.
Zinc is preferred for such bracelet anodes, for seabed use, and for all
conditions in which aluminum may become passive. For the main immersed
structures of oil rigs, however, aluminum tends to be preferred because of its
higher ampere efficiency. AI-Zn alloys are often used in the North Sea for
tanks because of the high (95C) oil temperature (Erskine, 1980). Jensen et al.
(1978) examined the various zinc alloys with regard to suitability when buried
in North Sea mud and at elevated temperatures and found that Zn, Zn-0.02%
Mg, and Zn-0.05% Cd are not suitable; US. ML 18001H composition
showed slight intergranular attack and greater dissolution than elsewhere.
Cathodic protection can be successful only under immersed conditions
and, therefore, only the parts of a ship below the waterline (and holds filled
with seawater ballast) can be protected by this method. Ideally, the ship is
painted in dry dock to the required standard, and zinc anodes are fitted below
the waterline to protect the steel and paint. Since, in general, a zinc anode in
seawater will protect all the steel within a radius of about m, anodes are
spaced at distances of about 6 m. Because of the turbulence around the stem
and because bronze propellers often are present, corrosion rates are higher in
this area and more anodes are placed there.
Modem anodes are streamlined to reduce drag, and the amount of zinc
used is sufficient to protect the ships hull for at least the period between dry
dockings, which is determined for other reasons.
Anodes may be installed to protect ships tanks and holds while these
receptacles are filled with seawater ballast. The anodes remain when the tank
is in use, and zinc is particularly safe for such applications because it does not
produce hydrogen gas or thermite sparks when struck with steel.
From about 1945, the largest consumer of zinc galvanic protectors in
the United States was the U.S. Navy, which specified rolled zinc plates bolted
to the stem areas of a great variety vessels to mitigate corrosion of the steel
in the general vicinity of copper-based or other alloy propellers that were
cathodic to steel. Hull surveys often indicated spotty performance of the zinc
plates: a significant percentage failed to function anodically because a hard,
tenacious, coating had formed on the surface of the zinc. In an effort to
Resistance 335
determine the reason for this relatively poor performance, Tee1 and Anderson
found that at Harbor Island, North Carolina, the zinc anode functioned
reasonably well when the iron content was maintained at a very low level in
high purity anodes. recommendation was made to keep the iron content
below in otherwise pure zinc, to avoid the formation of anodic
corrosion product films of very high electrical resistance. To meet this rigid
specification, anode producers used as their starting stock special high grade
zinc (SHG) slab, pure, with an iron content of generally less than
Additionally, hull anode shapes were specified with cast-in gal-
vanized steel straps for direct welding to steel hulls. This eliminated a bolted
connection between a zinc plate and hull, which often developed high electri-
cal resistance.
Laboratory and field studies by Crennell and Wheeler
showed that an iron content level of more than a few parts per million caused
passivation of zinc and that no other common impurities in zinc were of
importance to anode performance. These studies showed that the addition of
1% aluminum was beneficial and that aluminum plus silicon
significantly increased a zinc anodes tolerance for iron and also produced a
finer grain structure than was obtained with SHG zinc with maxi-
mum iron.
Reichard and Lennox conducted full-sized zinc anode field tests
for up to months on a seawater bulkhead, T-2 tanker, tugs, and a U.S.
Navy YO Class tanker. Their data showed that neither aluminum nor cad-
mium, when added alone to zinc, eliminated the need for controlling the iron
content to low levels. When small amounts of aluminum plus cadmium were
added to zinc, however, the tolerance for iron increased substantially.
Subsequent long-term field tests, reported by Waldron and Peterson
and also by Carson utilizing full-sized zinc anodes with
aluminum plus cadmium as alloying additions, essentially confirmed the field
exposure data of Reichard and Lennox. Carson et al. also showed that
the potential of zinc shifted in a very pronounced manner in the noble
direction during seawater exposures of 60 days as the current density or iron
content increased. Addition of aluminum to the zinc alloy retarded this
potential shift; 1.2% aluminum also retarded the shift, but it was not as
effective (Fig. The authors estimated the optimum aluminum content at
approximately
The Russian researcher Lyublinskii also confirmed that approx-
imately aluminum gives the maximum performance in iron-bearing
zinc anodes. Data from Vassilev et al. also confirm the effects of iron
and aluminum on the long-term exposure potential zinc anodes. In addi-
tion, aluminum is known to promote a finer grain size when added to iron-
Chapter 3
(a) 0. 3% (cl 1.2%
3.1 Effect of aluminum on the potential of zinc anodes in seawater (Carson et
al., 1960).
bearing zinc alloys. The role of grain size on zinc scale formation is unclear,
but a finer grain size may lead to a finer size distribution of harmful inclu-
sions, reducing the tendency of zinc oxide nucleation. The U.S. government
has modified the ASTM zinc anode specification on which Table 1.10 is
based to include the addition of aluminum and cadmium to zinc anodes for
fleet use.
The basic zinc anode specification containing aluminum and cadmium
as alloying elements has become the worldwide standard for zinc anodes used
in cathodic protection systems in both seawater and brackish water at ambient
temperatures. Reichard and Lennox (1957) acknowledged that zinc anodes
containing aluminum and cadmium were subject to intergranular corrosion at
unstated elevated temperatures, and some degree of intergranular corrosion
was observed after subjecting Zn-AI-Cd alloys to 288 hours of steam tests.
There was concern that anodes exposed to high temperatures and humidities
in storage might suffer unacceptable intergranular corrosion, but this appar-
ently does not happen.
Perkins and Bornholdt (1977) studied corrosion products normally
formed on the surface of a zinc anode conforming to MIL-A-18001H when
coupled to steel in seawater. The study relied heavily on direct microscopic
examination of corrosion product morphology using the scanning electron
microscope and also potentiostatic techniques. The authors concluded that the
anodic film is a porous three-dimensional network discrete single-crystal
plates of ZnO; individual plates are about 10-100 pm in diameter, with a
diameter-to-thickness ratio of about 50: 1. The very porous nature of the film
permits the Zn-Al-Cd alloy to function successfully as a galvanic anode in
seawater.
ASTM Committee B-2 recommended as early as 1967 (ASTM B
41 8-67) that the Zn-AI-Cd alloy anodes not be used at exposure temperatures
in excess of 60C. The same specification suggested the use of unalloyed, low
iron content zinc anodes, ASTM type I1 (updated zinc anode specification
ASTM B 418-80, now ASTM B 418-88), to avoid intergranular corrosion at
elevated temperatures. When ASTM 418-80 was issued, Committee B-2
stated that intergranular corrosion could occur on the Zn-AI-Cd alloy (Table
1.10) above about 49C (120F). Kurr (1979) had shown similar susceptibility
for the MIL-A-18001H alloy. Ashworth et al. (1979), in laboratory studies,
found that raising the temperature of this alloy to 70C increased the forma-
tion of an aluminum-rich phase in the grain boundaries and, with anodic
polarization in seawater at this temperature, the sensitized alloy will
undergo intergranular dissolution by selective removal of the aluminum-rich
phase. At 25C, the aluminum-rich phase was not observed, nor was inter-
granular attack found.
Zinc bracelets of the Zn-Al-Cd type had been used with great success on
similar offshore undersea pipelines in the Gulf of Mexico since the
mid-l950s, though the oil temperatures within the pipe were generally below
about 32C.
The use of zinc anodes in the Polish mercantile fleet is described by
Gliszewski (1971). Mackay (1974, 1975) tested zinc anodes at 120-180 m
depths and found that they did not passivate; although the anodes were less
efficient at depth than in shallow water, the steel cathode test plates were
polarized to an acceptable level. Reding and Boyce (1974) found no change in
performance at 300 psi pressure.
V. CASE HISTORIES
A. Supply Waters
Cold water systems of galvanized pipes, tanks, and cylinders-together with
galvanized fittingswere standard in many countries for about a century. In
many cases the pipes and fittings lasted the life of the house; tanks and
cylinders, even in scale-forming areas, tended to leak after 20-40 years.
338 Chapter
A survey by the author while at BNF in the showed that in scale-
forming waters, galvanized hot water tanks and cisterns usually lasted 20
years or more, but otherwise leaks would appear in less than 4 years;
intermediate performances were less common.
An extensive Australian evaluation of cost components included static
and dynamic trials with various detergents, disinfectants, and household
cleaners, as well as the soft Risdon tap water. It was concluded that a
of alloy is adequate for drainage fittings and only slightly inferior to
brass.
B. Natural Waters
A new harbor installation, constructed at Bottighofen in used consider-
able quantities of steel. The sheet piling, about m long, and all other steel
components such as railings, enclosures, doors, gates, and the landing stages
with their floating pontoons, were hot dip galvanized. The same applies to the
mooring posts and the heavy tubular mooring rails in the region of the tank
installation. Then in the harbor installation in Arbon was also expanded
after the pattern of Bottighofen (both places on the Swiss side of the
Bodensee). Again, all steel components were hot dip galvanized-including
about tons of sheet piling and guard sections. Upon inspection in
both installations were found to be in perfect condition after and years,
respectively. far, none of the hot galvanized parts has been replaced or
repaired. Even in the wind and water zone-the region with the most
severe corrosive action-no notable attack of the zinc coating could be
discerned. In view of the excellent condition of the galvanizing with 80 pm
zinc remaining on thick sections, many more maintenance-free years can be
expected.
Galvanized gabions erected in in Italy gave more than years
sefvice but no details are available.
Lock gates are critical structures, which are examined regularly. The
lock gates of St. Denis, a suburb of Paris, were zinc sprayed in and
remain in good condition today after two intermediate repaints:
An inspection in revealed only a thin layer of paint in evidence, but the
sprayed metal coating was intact. Repainting was carried out.
In the gates were again lifted for painting and the sprayed zinc was
found to be sound.
The Anglian Water Authority lock gates at Little Paxton were similarly grit-
blasted and sprayed with zinc to pm plus a multicoat paint system.
Resistance 339
Similar early uses for sluices and weirs for which successful reports
were made include sluice gates in the Rhine at Strasbourg, drainage
sluices in Paris, (reported by the French Metallising Society, SNM, in
and sluice gates in the Panama Canal, (Anon.,
In bilges and naval ships, zinc spray plus a chlorinated rubber paint
system is used, for example, in Royal Navy Sheffield Class destroyers based
on years service before maintenance (Hunter, Galvanizing alone
was used in earlier RN ships.
C. Seawater
Boats have, in general, been more closely monitored than structures. A steel
lifeboat, the MS President Jan Lek, built in Holland in (length m),
had all parts hot dip galvanized, based on good results with other lifeboats
built at the end of the nineteenth century. After assembly, all galvanized parts
were painted with the then available marine paints. The President Jan Lefs
served for more than years, first in the Hook of Holland harbor and in later
years at Burghsluis, a small fishing port on one of the isles of Zealand; it was
returned in and now is a museum boat. When the vessel was examined
in there was practically no rust to be observed, although the hull and
other outdoor parts have been constantly exposed to salt and brackish water,
to chafing, bumping, and rubbing against hulls of ships to be rescued, and
against buoys and landing stages. Micrographs clearly indicate that the duplex
system, although partly weathered, is still effective.
The Charolais Express has 25 years service (Beratung Feurverzinkung,
galvanized pipeline off-shore Cyprus, installed in had had
more than 25 years service when last examined. About a quarter of the line
has been replaced. On most of the surface, much of the galvanized coating
remains with an adherent calcareous deposit, probably due to the line becom-
ing buriedkilted up. There are, however, some pits, where up to 4 mm of the
12 mm wall thickness of the pipe has been consumed; 120 pdyear steel
consumption could be expected on steel without galvanizing. Examination
indicated that these thinner areas, although described as pits, could corre-
spond to chain damage during installation; they were several inches long and
up to mm wide. Application cathodic protection was under considera-
tion, but care is needed to avoid dissolution of the calcareous deposit.
4
Resistance to Corrosion
Natural soils vary too widely in composition to permit very specific advice to
be given, but sandy, well-aerated soils with a neutral or slightly alkaline pH
likely to cause only limited corrosion of zinc and its alloys. Zinc coatings
will prevent pitting of steel in the soil and, even where the zinc coating is
destroyed, the coated steel corrodes much less than do bare specimens.
Control of backfill in earth reinforcement ensures that zinc or its alloys can be
used satisfactorily.
A primary source of information on the behavior of metals in the soil is
the extensive work on underground corrosion done in the by the U.S.
National Bureau of Standards (now the National Institute of Standards and
Technology), including two rolled zinc alloys (Table as well as gal-
vanized pipes (Table 4.2). It was concluded that the life of galvanized steel
buried in soil would be greatly dependent on the nature of the soil. A nominal
85 pm coating would provide protection for at least years in inorganic
oxidizing soils. A pm coating appeared to be adequate (for years)
in most inorganic reducing soils but would not afford sufficient protection in
highly reducing organic or inorganic soils, or in cinders. This work has been
further analyzed in connection with earth reinforcement (Figs.
Tests by the British Iron and Steel Research Association (Hudson and
Acock, covered five types of soil in the United Kingdom and exposures
of 5 years. The galvanized samples failed in cinders; they had an expected
total life of about years in the salty marshland and 25 or more years in the
other soils tested (Table
341
W
e
N
Table 4.1 Corrosion of Rolled Zinc in Various Soils in the United States
Average metal loss (FWyear) and maximum pit depth (mm) after burial for years statedb
LOSS Maximum Loss Maximum Loss Maximum Loss Maximum
after 2.1 pit after 4 pit after 9 pit after 12.7 pit
Soil type Metal" years depth years depth Ye= depth Years depth
Inorganic oxidizing acid soils
Cecil clay loam P
Hagerstown loam P
Susquehanna clay P
A
A
A
Inorganic oxidizing alkaline soils
Chino silt loam P
A
Mojave tine P
gravelly loam A
Sharkey clay P
Inorganic reducing acid soils
A
4
7
8
9
12
9
10
26
34
79
10
11
0.25
0.38
0.81
0.89
0.23
0.30
0.76
0.56
0.64
2.41
0.30
0.36
6 0.25
6 0.56
6 0.64
7 0.69
13 0.23
7 0.41
8 0.91
9 0.41
28 0.71
54 P
- 11 0.20
13 0.71
5
5
3
4
6
4
7
10
4
7
5
6
0.33
0.66
0.33
0.40
0.30
0.48
1.42
0.84
1.12
0.91
0.36
0.91
7
10
4
5
6
6
6
9
18
17
7
9
0.43
0.79
0.41
0.5 1
0.23
0.81
1.42
1.29
0.86
P
0.36
0.94
+ +
d
B
343
w
e
P
Table 4.2 Corrosion of Galvanized Steel Pipe in Contact with a Variety of Soils in the United States
Normal mass coating, 915 g/m2 of exposed area, is the mass of coating on one side of the pipe; 915 g/m2 is equivalent to approximately 0.13 mm
thickness of coating.
Weight loss (g/m2)" and maximum pit depth (mm) after burial for years stated
Loss
after 2.1
Soil type Ye*
Cecil da y loam 90
Hagerstown loam 90
Susquehanna clay 300
Inorganic oxidizing acid soils
Inorganic oxidizing alkaline soils
Chino silt loam 330
Mojave fine 480
gravelly loam
Maximum
pit
depth
Loss Maximum
Years depth
after 4.0 pit
~ ~ ~
Loss Maximum
Years depth
after 9.0 pit
Loss
after 12.7
Years
Maximum
pit
depth
0.23
<O. 15
0.23
<O. 15
0.15
430 0.15
300 0.20
700 0.23
700 0.15
lo00 0.20
1 80 K0.15
210 0.15
270 <O. 15
K0.15
490 330 C0.15
180
180
240
330
330
<O. 15
<0.15
C0.15
<O. 15
<0.15
345
346
Chapter 4
I
348 Chapter 4
Table 4.3 Corrosion of Hot Dip Galvanized Pipes and Flats Buried in Several
British Soils for 5 Years
Exposed area: pipes mm diameter X 330 mm; flats 250 mm X 380 Weight
determined after cleaning with acetic acid. Average of three specimens.
Initial Average thickness
thickness (pm) (pm)
Location Type of soil Pipes Flats Pipes Flats
Benfleet Reclaimed salt marsh 55 51
Pitsea London clay
Rothamsted Moist, neutral clay
Gotham Keuper marl
corby Cinders Not tested h
Calculated.
Coatings failed with much rusting of the steel.
Source: Hudson and Acock (1952).
II. INFLUENCE OF TYPES OF SOIL
Soils vary widely in the physical and chemical characteristics that determine
their corrosivity. An extensive discussion of the science soils is beyond the
scope of this book. However, a summary of the characteristics and properties
of soils was given in the National Bureau Standards Circular on Under-
ground Corrosion (Romanoff, 1957).
More than 200 soil series have been categorized in the United States
according to texture, color, and natural drainage. The physical properties that
are of importance in corrosion are those that determine the permeability of the
soil to air and water. Soils with a coarse texture, such as sands and gravels,
permit free circulation of air. The corrosion under such conditions approaches
that occumng in the atmosphere. Clay and silt soils are generally charac-
terized by a fine texture and high water-holding capacity, resulting in poor
aeration (which tends to reduce general corrosion but makes local pitting more
likely) and poor drainage (which increases corrosion by keeping the surface
wet longer). The aeration characteristics of a soil depend on particle size and
distribution and on the apparent specific gravity. The latter is a measure of the
compactness of the soil.
Many chemicals are present in soils, but those relevant to corrosion
studies are soluble in water. Analyses are usually made for base-forming
elements such as sodium, potassium, calcium, and magnesium, and for the
acid-forming groups such as carbonate, bicarbonate, chloride, nitrate, and
Table 4.4 Soil Aggressiveness of Different Soil Resistivities; Suitable
Corrosion Protection
Soil Resistivity
no. Aggressiveness Soil condition - m) Method of protection
1 LOW
Dry
>l00
Hot dip galvanizing:
2 LOW Moist >450
Moderate
Dry
<l00
1500 g/m2
g/m2
1500 g/m2 plus a rust
allowance of 0.5 mm
on each side
4 Moderate Moist 150-450 Same as for soil
5 High Moist 50- 150
1500 g/m and rust
allowance of 1 mm
on each side
6 Very high Moist or in <SO- 150 Same as for soil 5 but
certain rust allowance of 1.5
cases, mm on each side
H W 4
can form
sulfate. The electrical resistivity of the soils is determined by the nature and
concentration of the ions formed by the chemical salts dissolved in the soil
moisture. These ions may also determine the progress of corrosion. If the
primary product of corrosion is relatively insoluble and deposits as a film on
the metal surface, further corrosion may be reduced or completely stifled.
broad guide (see Tragardh, 1958) to the recommended use of gal-
vanized steel in different types of soil was given by the Swedish State Power
Board (1956) summarized in Table 4.4.
Many factors play a part in underground corrosion, and interrelated
effects often occur. Thus results in corrosive soil environments are sometimes
difficult to explain, and often difficult to predict.
CORROSION DATA
The National Bureau of Standards underground corrosion program (Ro-
manoff, 1957) included tests on commercial rolled zinc and a zinc-4%
aluminum-l% copper. Panels were buried in representative soils in 1937, and
350 Chapter
corrosion rates were determined at intervals up to about years burial in the
soils. Weight losses are given in Table Although the Zn-Al-Cu alloy
usually lost more mass (particularly in the short-term tests), this result was not
very significant. The depth of pitting, however, was less on the commercial
zinc sheet than on the zinc-aluminum-copper alloy. In three corrosive soils
(Rifle Peat, Cinders, Merced Silt Loam) both materials were destroyed by
corrosion in relatively short times. The severe corrosion in Rifle Peat and
Merced Silt Loam was attributed to the acidity/alkalinity of these soils (pH
and respectively), but in fact limited alkalinity is not per se a cause of
corrosion of zinc. Poorly aerated soils are corrosive to zinc, although high
average metal loss is not always indicative of deep pitting.
The National Bureau of Standards underground corrosion program also
included tests on galvanized steel sheet and pipe specimens. 10-year test
started in in soils ranging in pH from to 9.5 showed that in most
of the soils, zinc coatings of pm or less were destroyed, while pm
coatings remained intact on at least half the specimens. A new test was started
in 1937, .using mm steel pipe specimens coated with a nominal 130 pm
zinc coating and buried in 15 soils of different characteristics. The weight loss
and maximum pit depth results for duplicate specimens removed from the test
at intervals are also given in Table Examination of unexposed specimens
showed that the thickness varied considerably from the nominal, not only
between specimens but also in the distribution of the coating on individual
specimens; more than half the total thickness was zinc-iron alloy. The
maximum depth of pitting is also given in Table
In some of the more corrosive soils, corrosion of the steel (after the zinc
has been completely destroyed) is considerably less than that observed on
uncoated steel control specimens. Also, it appeared in some soils that the
protection afforded to the steel was not due solely to sacrificial corrosion of
the zinc but was supplemented by some other mechanism. A few tests
indicated that a film, consisting primarily of zinc silicate formed by galvanic
action between the zinc and the underlying zinc-iron alloy or bare steel, was
responsible for the additional protection.
Galvanized pipe and flat specimens were included in the soil corrosion
tests (Table conducted by the Subcommittee on Buried Metals the
Corrosion Committee of the British Iron and Steel Research Association
(Hudson and Acock, 1952). The galvanized test pieces averaged g/m2
coating (about 95 pm). The specimens were buried for 5 years at a depth of
m in five British soils. With the exception of the specimens buried in
cinders at Corby, all the test pieces were rated excellent (no sign of
breakdown) on visual examination. However, the weight loss data show that
Resistance to Corrosion by Soil 351
about 60% of the coating had been corroded away at the Benfleet site,
compared with only 20% at the other sites.
Numberger studied soils containing sand with particle sizes
mainly (2.5-20.4%) below 0.06 mm; both galvanized and ungalvanized steel
panels were exposed, outdoors and in the laboratory. Variations were made in
temperature, composition, and salt (NaCl) content. The rates of corrosion
(including, with the ungalvanized specimen, pitting corrosion) were mea-
sured. Furthermore, the specific soil resistance and the corrosion potential
were observed continuously. The loss of metal of the ungalvanized panels was
much more pronounced than of the galvanized ones, and losses increased with
an increasing percentage of fine sand. Salt increased corrosion of the gal-
vanized panels only in the early phase of the test. Increased loss of metal
started approximately at mol Cl- plus per kilogram in
the soils. On the ungalvanized panels, corrosion losses diminished upon
further exposure; pronounced pitting was observed here, especially in aerated
soils. With the galvanized panels, no attack of the steel was observed after the
zinc coating corroded away; cathodic protection by surrounding zinc pre-
vented rusting. Rusting of steel depends also on temperature. Raising soil
temperature from 4C to 20C doubled corrosion rates on the steel panels.
This effect was not observed on the galvanized panels.
W. ZINC ANODES IN SOIL
A.
Cathodic protection is a very successful method of protecting structures
buried in soil. This protection can be applied to pipelines, storage tanks, and
building foundations.
In certain soils, zinc anodes used in conjunction with coatings have
lives of up to 5 years. After this time, the zinc anodes can be replaced and an
additional period of protection obtained. Because the electrochemical poten-
tial of zinc is relatively low, current flow is moderate and there is no damage
to any coating on the pipelines or structure. Since soils vary in their content of
water and salts, the electrical resistivity may vary from day to day as well as
depending on the type of soil. However, if the resistivity of the soil is greater
than about 5 R-cm, the electrochemical potential will be too low to produce an
adequate current, and protection will be less.
To overcome this problem in soils of high resistivity, the space around
the structure and the zinc anode is backfilled with a more highly conduc-
tive material. Soils containing significantly more dissolved sulfates and chlo-
352 Chapter 4
rides than carbonates, bicarbonates, nitrates, and phosphates are often com-
patible with zinc anodes, but a prepared backfill chosen because it is not very
corrosive toward zinc is preferred. A successful mixture is 70% powdered
gypsum, 25% bentonite clay, and 5% sodium sulfate: this provides relatively
low resistivity and permits high current output levels. In time, however, the
sodium sulfate tends to leach out of the backfill and the resistivity increases.
In areas where there are many pipelines and cables close together, care
is needed to avoid interaction between the cathodic protection systems. Zinc
anodes accomplish this isolation most simply and effectively. As with im-
mersed structures and. ships, cathodic protection buried metals can be most
economical when used in conjunction with a good coating system. In many
buried structures, damage to coatings occurs in handling and by abrasion from
the overlying soil, but, as the coatings remain protective in most places,
leaves a much smaller area of steel relying on the cathodic protection.
A particular hazard to buried structures is the growth of sulfate-reducing
bacteria, which turn sulfates in the soil into particularly aggressive com-
pounds. The slightly alkaline environment produced by cathodic protection
greatly discourages the growth of these bacteria.
On buried structures and pipelines that are initially wrapped either with
or without the use of cathodic protection, the coating will eventually deterio-
rate. When it is inconvenient, uneconomical, or impossible to lift and rewrap,
continued protection for the structure may be had by installing sufficient zinc
anodes. Provided the zinc anodes are replaced when necessary, a relatively
straightforward job, the useful life of the structure may be extended indefi-
nitely.
Specific Data
The pioneer efforts to use zinc as an anode in underground systems protecting
steel pipe and other ferrous structures go back to the Zinc anodes were
installed bare in a wide variety of soils with mixed results. Walquist
reported some cases of the formation of anodic surface films that significantly
increased the circuit resistance and reduced the current output. Typical iron
content of the anodes was about and often values in excess of 0.01%
were encountered. Subsequently, iron content was controlled (Table to
avoid this problem. Mudd observed relatively consistent performance
when zinc anodes were installed in soils containing gypsum (hydrated calcium
sulfate), as opposed to non-gypsum-bearing soils. As a result, he suggested
packing gypsum around the zinc anodes when installations are made in
nongypsum soils.
The principal zinc specification (ASTM B type 11) used for
underground cathodic protection systems shown in Table 1.10 is the un-
alloyed high purity zinc with iron controlled to 0.0014% maximum. The
literature does not show any long-term field tests comparing the unalloyed
type I1 materials with the Zn-AI-Cd composition MIL-A-18001H or with
ASTM B418-80 type I. Although installations the latter alloy anodes
prepackaged in prepared backfill have been made, comparative results are
lacking.
Since zinc has a considerably lower potential than magnesium, its use is
generally limited to lower resistivity environments. In zinc-steel systems,
upper guide limits of soil resistivities of about fkcm (Zastrow, 1973) are
common for economic reasons. However, in zinc-copper systems, such as the
cathodic protection bare copper concentric neutral wires underground
electrical rural distribution (URD) cable, upper resistivity limits fl-cm
are both economical and practical. In the latter instance, the copper is
generally considered to be polarized at a potential about -0.4 V and the
driving voltage of the zinc is about 0.7 V. This voltage is almost three times
that a zinc-steel system polarized to -0.85 V, with a subsequent driving
voltage of 0.25 V.
Kurr (1979) plotted current output versus soil resistivity for a number of
different anode sizes installed singly. The information was developed by
calculating the anode resistance at various soil resistivities using the Dwight
(1936) equations and Ohms law. The life calculations were made in accor-
dance with the equation:
life =
weight zinc
current output (A) weight zinc consumed per ampere year
Typical plots obtained using a zinc-steel system, a zinc-copper system, and a
continuous zinc ribbon-steel system are shown in Figs. 4.44.6.
Woody (1976) has described long-term experience in a city system.
Originally, pipe-to-water measurements were the criteria for protection, but
now pipe-to-soil potential measurements are used.
In recent years, continuous zinc ribbon anodes have been used in a
variety underground applications (Kurr, 1973; Peabody, 1976; OConnell,
1977). This type product has broadened the applications for zinc anodes,
for it provides small increments current continuously along the entire
length a cathode. Its uses are generally considered to lie in specialty
applications, where other methods of cathodic protection are either impracti-
cal or extremely costly (see later section on induced ac on pipelines). Bagnulo
(1973, 1984) has developed a tape with an electrically conducting adhesive as
described in the Mechanical Coatings part of Chapter 1.
Soil
resistivily
ohm- cm
5 to 20 50 100 200 500 1.000
Fig. 4.4
Soil
ohm- cm
resistivity
1o.OOo .
5.ooo -
-
-
500 -
200 -
l o o -
Fig. 4.5
1
Current output milliamperes
Zinc-to-steel current output versus resistivity (Kurr, 1979).
5 10 20 50 l o o 200 500 1, OOo
Current output milliamperes
Zinc-to-copper current output versus resistivity ( K U ~ , 1979).
Resistance Corrosion by Soil 355
Ohm- cm
resistivity
10.000 -
5.000 -
2.000 -
1. ooo -
500-
200 -
-
50 -
25 -
I O -
,
01 02 05 l 2 5 IO 0 5 I O M 50 200 500
Current output-amperes
4.6 Zinc continuous ribbon anode to steel current output versus resistivity
(Kurr,
C. Potential Reversal
potential reversal occurring in a cathodic protection system involving zinc-
based anodes is possible under certain circumstances. However, a reversal is
highly improbable when the zinc anodes are surrounded by a gypsum-based
backfill, as discussed previously.
Haney (1981) has shown that environment, not temperature, governs
the potential of zinc, and the potential is adversely affected in an environment
rich in bicarbonates, carbonates, and nitrates, even at room temperature. The
addition of elevated temperatures to this hostile environment merely acceler-
ates the reaction, in which the potential zinc becomes more noble and is not
in itself the cause. Haney also showed that adding gypsum is more effective
than adding sodium chloride in preventing a reversal. Figure 4.7 (Haney,
1981) shows that the addition of gypsum to a very hostile solution at room
temperature will restore the potential of zinc from - 0.6 V to about - 1 .O V
(vs. SCE) within a very few minutes. This result suggests that bare zinc
anodes underground which have been adversely affected in a hostile environ-
ment may be reactivated by injecting or otherwise placing gypsum around the
anodes.
356 Chapter 4
- 1.10
- 1. 00
-090
-080
-0 70
-060
-050
-0 40
D
4.7 Effect of adding gypsum to reverse the potential (Haney, 1981).
Grounding Cells
Zinc has long been used in the form of cells and electrodes in pipeline systems
subject to severe electrical fault currents as from lightning or from induced ac
from overhead electrical transmission systems. Since gas piping in an indus-
trial complex often does not have cathodic protection, the underground gas
pipe system may have flanges couplings to provide electrical insulation at
the property line of an industrial gas consumer.
When fault current from a lightning strike or other source travels down
the pipeline to an insulated flange, the latter is often shorted out by metallic
arcing owing to the very high energy levels involved. As a result, the
cathodically protected portion of the pipeline can fall below potential values
accepted by the National Association of Corrosion Engineers (RP-01-69:
NACE, 1976) because the shorted-out system now includes the entire un-
protected underground industrial complex. Apart from the cathodic protection
aspects of the problem, the safety of both personnel and the underground plant
itself must be considered in proper perspective by the pipeline and corrosion
engineers when a system is being designed.
...
Zinc grounding cells, described in detail elsewhere (American Zinc
Institute, 1957; Kurr, 1979; Peabody), generally consist of either two or four
zinc anodes (electrodes either 150 cm3 or 5 5 150 cm3, as
shown in Fig. The anodes are usually separated by 2.5 or 5 cm thick
nonmetallic insulating blocks and are assembled into a single unit using
insulating tape. In the two-unit cell, an insulated copper conductor of suitable
size is silver-brazed to each of the nominal 0.6 cm diameter galvanized steel
full-length inserts or cores. The entire unit is usually placed into a cylindrical
paper tube, into which is poured prepared backfill. When the tube is full and
free from void spaces between the anodes, its top is closed and only the two
copper conductors protrude.
In practice, one cable is secured to one side of the insulated flange while
the other is attached to the opposite side, as shown in Fig. When the
prepared backfill of the cell has absorbed groundwater and its resistivity has
stabilized to a value consistent with its specific formulation, the nominal
resistance between the two anodes is 0.2-0.5 a. When a lightning strike or
other fault occurs, the fault current of relatively short peak duration will
generally pass through the cell, bypassing the insulated flange and, in effect,
protecting it from shorting out. It has been reported (NACE, 1977 Peabody)
that the voltage drop across the insulating joint will be limited to 200-500 V
per l o o 0 A of fault current flow. Additionally, a two-unit cell (American Zinc
Institute, 1957) with number 6 American Wire Gauge (AWG) copper connec-
ting cables can rated at up to lo00 A per cell. Where considerably higher
fault currents are anticipated, two or more cells can be used in parallel.
During normal or nonfault periods, cell resistance and cell polarization
alike are adequate to prevent the loss of current from the cathodically pro-
tected side of the insulated flange to the nonprotected side. The pipe at the
nonprotected side of the flange will receive current locally from the zinc
anode grounded to it, minimizing the possibility that current will be dis-
charged from that area with potentially damaging results.
E. Induced Alternating Current on Pipelines
Often in industrialized countries there is a trend to locate various utility
pipelines and electrical power transmission cable systems in the same right-
of-way, as a means of economic utilization of land. In the United States, for
example, gas pipelines and communication cables may be in the same under-
ground right-of-way with a high voltage power transmission system overhead.
The hazards to both workers and equipment created by this arrangement often
are eliminated by suitable grounding through use of zinc electrodes.
358 Chapter 4
Combination insulated
joint and line current
copper leads (such
as No. 6 AWG. min.)
Detail lour anode
where required
cross connected cell
Note: the anodes with
oblained as a
backlltl may be
packaged u n t
Entlre assembly
hole
Installed in augered
4.8 Typical grounding cell arrangement (Peabody).
Induced ac voltages are generally picked up on the pipeline by conduc-
tive interference or inductive effects (Royce, 1971). Stray ac current picked
up by the pipeline from an ac cable grounding system is referred to as
conductive interference. Induction usually results from electrostatic and elec-
tromagnetic fields around the electric cables, referred to as capacitive and
inductive, respectively.
Induced ac on a pipeline can be great enough to cause an unpleasant and
sometimes severe jolt to a human or an animal that contacts the pipe or an
appurtenance while standing on a ground that gives a large difference in
potential. A very significant worker hazard exists during fault current surges
from lightning strikes or other causes (RP-01-77: NACE, 1977).
Backmann and Funk (1982) have shown that zinc sheet or galvanized
sheet can be used as grounding materials for alternating currents. Whereas
Resistance Corrosion by Soil 359
magnesium shows a positive shift of potential, this is not the case with zinc.
With ac-carrying ground contacts, the weight loss of zinc is very small.
However, it should be noted that additional cathodic protection must not lead
to potential values below mV. Because zinc is an amphoteric metal,
formation of NaOH can cause an increased loss of zinc. The tests of Back-
mann and Funk showed that zinc losses lying between and mV
are low. The potential must be regulated carefully because the superposition
of ac may cause the zinc potential to move toward about -1 mV.
Therefore, the potential should be set initially at to -1400 mV.
Copper cable connecting the anode to a structure must be of sufficient
size to accommodate anticipated fault currents. Figure shows the approxi-
mate fault current needed to raise the temperature of stranded, annealed, soft-
drawn copper cable to 685C above ambient temperature.
Zinc grounding cells are often used between a pipeline and metal tower
footings, supporting an ac power transmission system (American Zinc Insti-
tute, A lightning strike on an overhead cable seeks its way to ground
generally through the tower footings and can readily establish an arc between
the footing and the pipeline, with potential danger to the pipeline and workers
in the vicinity. When one terminal of a grounding cell is connected to the
tower footing and the other to the pipeline, fault current is generally dissipated
without damage to the pipe. Figure illustrates a typical application of
zinc grounding cells where a high voltage electric transmission line closely
parallels a coated pipeline. The zinc grounding cells with internal resistance
of C? limit the drain of cathodic protection current between the pipe
and tower grounds, while providing an adequate path for draining induced ac
into the earth.
Zinc ribbon has been used as a grounding electrode to dissipate induced
ac from a pipeline (Kurr, In addition, zinc ribbon is used as a ground
mat to equalize the potential of the earth at a valve, cathodic protection test
station, or other appurtenance of the pipeline. The ribbon is installed about
0.5 m underground in the form of a flat, horizontal, pancake coil with a
nominal diameter of 2.5 m or more. One end of the wire core of the zinc
ribbon protects the workers who are standing on the earth surface, touching
the pipeline valve or test station, by minimizing the potential difference
between the pipeline and earth.
V. REINFORCED EARTH
The relatively new concept reinforced earth makes extensive use of gal-
vanized steel rods or strips. The subject is well covered by the British
360
Chapter 4
Faull duralion (seconds)
4.9 Approximate current to raise the temperature of stranded annealed soft-
drawn copper cable above the ambient (Dwight, 1936).
Resistance to Corrosion by Soil
361
""8"
High vollage electric
lransrnission line
closely paralleling
coated plpeline
belween pipeline and
steel towers of closely
paralleling eleclric
lransrnlssion line.
Seclion of well coaled
protected pipeline
--- and cathodically
Zinc cell
Fig. 4.1 Typical application of zinc anode grounding cells. Zinc cell capacity to
be installed at each location is based on information concerning maximum fault current
and fault duration obtained from operators of electric transmission line. Necessity for
grounding cells between pipeline and electric transmission line to be based on a study
of the length of exposure, electric transmission line voltage, spacing between pipeline
and electric line, soil resistivity, pipeline coating condition, and so on (Downing,
1964).
Standards Institution's PD 6517 (1988). An earlier review (King, 1977) with
86 references covers soil corrosiveness with particular reference to reinforced
earth.
Reinforced earth is a composite of soil and tensile- or shear-resistant
inclusions that function through the agency of direct soil-reinforcement inter-
action to produce a material having a performance superior to that of soil
alone. Reinforced earth may be used to replace conventional structures, to
reduce area of land required, or to build flexible structures over poor or soft
foundations.
Reinforced earth walls are commonly used for some form of soil
retention. For example, a vertical face constructed without reinforcement
would fail in time, depending on the value of the soil parameters. If the soil in
the region of the face is reinforced, however, it will form a stable mass, which
can support the unreinforced soil behind it. The same principle applies to
reinforced slopes constructed to an angle steeper than their angle of internal
friction.
\
362 Chapter
The advantage of a reinforced fill-retaining structure over a reinforced
concrete wall is that the inherent resilience and flexibility of the composite
enables construction to progress with a minimum of treatment of the founda-
tion soils, where subsequent movements are likely. There is some similarity
to the principle of reinforced concrete in that the reinforced mass may be
considered to be a composite material with improved properties, particularly
in tension, compared with the soil or concrete alone.
The form of the early structures comprised a U-shaped sheet metal or
plastic facing panel connected to metal strips, usually made of galvanized
steel. The reinforcement of the soil is the consequence of tensile forces in the
reinforcement being generated by friction between the soil and the strips. This
composite material was called Terre ArmCe (reinforced earth) by Henri Vidal,
who introduced it.
The good experience with galvanized steel in soils of controlled compo-
sition has led to the wide use of galvanized steel in this new industry.
French standard NF A specifies backfill and steel thickness for
different service lives, both with galvanized steel and with ungalvanized
(Table pH must be between and (Jailloux,
As shown in Tables and and Figs. Romanoff
demonstrated that the rate of corrosion is greatest in the first few years after
burial and decreases to a much lower constant rate thereafter. Romanoff
indicated that this damping of corrosion was a more significant parameter than
the initial rate. He proposed quantitative empirical relationships to calculate
average loss of thickness of plain steel as a function of time.
McKittrick referring to earlier work by his colleagues, said that
the burial site data were selected to be more or less consistent with the normal
range of environments for buried reinforcing strips and extended these data in
accordance with Romanoff's proposals. These investigators compared the
results of the extended or extrapolated data with other pertinent studies such
as the performance of sheet piles and culverts. This comparison for gal-
vanized steel samples and metal culverts is made in the paper. This paper and
the later paper by Darbin et al. demonstrate that even in an aggressive
environment (p = 0-cm, pH the galvanized steel reinforcing strips
currently in use would have a service life of years.
The extrapolation of the Romanoff data requires the solution of the
exponential equation:
= kT"
where
= average loss of thickness with time
k = a site characteristic
363
364 Chapter 4
T = time (years)
n = a site-dependent variable, always less than 1.0
Since selection of k and n requires some subjective interpretation, McKittrick
has attempted quantitative conclusions drawn from the Romanoff/NBS data
from the 47 steel burial sites. Figure 4.1, in attempting to present a relation-
ship between metal loss and resistivity, shows that a well-defined relationship
does not exist.
Figure 4.2, an attempt to show a relationship between metal loss and
pH, again demonstrates that no well-defined relationship exists. However, it
does show greater metal losses at sites with pH values less than 5. Using pH
as the only guide, it is difficult to draw a conclusion. However, if only well-
drained sites are plotted, as in Fig. 4.3, it can be concluded that annual metal
losses at such sites will not exceed 45 g/m (6 pdyear).
The abundance and reliability of the foregoing data and the ability to
extrapolate such numbers to the time period normally associated with engi-
neering works led to the selection of galvanized steel as the material most
commonly used for reinforced earth structures. The zinc coating on gal-
vanized steels forms a sacrificial anode that corrodes while protecting the base
metal. In addition, zinc promotes a more uniform corrosion by preventing the
formulation of pits during the highly aggressive initial stages of burial.
The excellent structural performance of the more than 2200 reinforced
earth structures completed since 1968 and in use for more than 12 years
demonstrates more than any other fact that these structures have been de-
signed for safety. Measurements and observations of movements and stresses
confirm that the working stress design procedures, derived on the basis of a
coherent gravity structure analysis, accurately predict performance. im-
provements are made in the technology, such as the recent introduction of
high adherence reinforcements, basic soil mechanics theory supported by
laboratory and field testing can be used to modify design procedures to
anticipate the effects of these improvements.
backfill materials become more fine-grained, caution must be exer-
cised in selecting design parameters and factors of safety to allow for the more
complex shear strength and corrosion characteristics of these finer grained
materials. Even with these finer grained elastic materials, however, adequate
designs can be developed.
Darbin et al. (1988) reported 10-year laboratory trials, gave backfill
recommendations (Table 4.6), and reported container tests in five soil types
(Table 4.7). Zinc thicknesses of 25, 60, and 85 km were tested: the 80 pm
results are in Fig. 4.11. The investigators concluded that an 80 pm zinc
365
Table 4.6
with Galvanized Steel
Chemical and Electrochemical Criteria of Backfill to be Used
Maximum content (ppm)
Saturated state
resistivity p
Site (a * m) PH CI - so,"-
Dry
>10 5-10 200 1000
Submerged >30 5-10 100 500
Source: Darbin et aL (1988).
Table 4.7 Container Soil Characteristics
Content (ppm)
Resistivity p
Soil type (a - m) PH c1- so,'-
A Red shale: burnt 5.05 7.4 <10 3000
colliery shale,
coarse porous
colliery shale
clean river sand
with artificial
seawater
fine sand
sand
B Black shale: 76.45 7.9 <lo 20
C Artificial sea sand: 21 .00 7.0 300 30
D Clayey sand: very 39.00 7.8 <10 30
E Silty sand: fine 43.40 7.8 <lo 30
~
Source: Darbin et al. (1988).
366
Chapter
Zn
I
I I I I
0.1 7 0.5 1 2 5 10
I : year
4.1 1 Container test results for soils given in Table 4.7 (Darbin et al., 1988).
coating has a life span of years in the more corrosive soils and more than
70 years in soils of low corrosivity. Heim (1977) has also described the
corrosion behavior of hot dip galvanized grounding strip.
VI. CASE HISTORIES
Conventional buried galvanized steel structures and pipelines are seldom
examined and case histories determined, unless there is failure-at which
point, often, cathodic protection is applied.
Conversely, reinforced earth applications galvanized steel have been
extensively documented; Tables 4. 8 and 4. 9 list selected projects that have
performed satisfactorily for 12-20 years. By 1982, 2500 projects had been
completed in 37 countries (R. J. H. Smith, 1982), and by 1989 there were
13,500 structures totaling 7. 7 l o6 m* (Jailloux, 1990). One of the first uses
Table 4.8 Site Investigations in 1986-1988 on 40 Structures
Type Thickness
Date of Age of losslside
Project construction (years) stripa (F@
Vigna-AS, Nice-Menton
La Giraude A8, Nice-Menton
Vallons des Grottes
La Croix
Rouen
Strasbourg
Skte
MerlebacLA32
Avignon
Dunkeque
Palaiseau
Orsay
Strasbourg Roc. Quest
Saint Jorioz
Annecy
Givors
Tours
Brest
La Brecque
Saint Fereol
TMPpeS
Le PaillonS2l-A8
Niort
Corbeil
Le PaiIlon-OA76b-A8
Vallon des Bimes
Franconville
EvrY
Lille
Landerneau
Tours-Pt. Mirabeau
Corbeil-Gilardoni
Lava1
Harfleur
Villeneuve Loubet
Toulon B52
Perpignan
Carrefour Pleyel-Paris
La Dtfense-Paris
Tauves
1969
1969
1970
1968
1969
1969
1972
1972
l97 1
1970
197 1
1971
1970
1970
1973
197 1
1972
1972
1972
1972
1972
1973-1 974
1973
1972
1974
1972
1972
1972
1972
1980
1973
1972
1974
I973
1974
1974
1976
1979
1975
1976
IO
13.5
16
17
12
13
11
11
11
12
12
12
13
13
12
12
11
11
8
11
11
IO
12
11
9
9.2
11
10
11
6
IO
10
10
10
11
9
8
6.5
10
5
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
B
A
A
A
A
A
A
A
B
A
IO
8
95
25
5
20
60
75
20
85
20
25
15
IO
24
25
18
35
35
15
20
22
28
15
25
30
66
25
30
35
30
25
59
25
34
17
36
58
30
10
= smooth, 3 mm thick, 30 pm zinc; B = ribbed, 5 mm thick, 80 pm zinc.
Source: Darbin, M. , Jailloux, J.M., and Montuelle, J. (1988).
Table Established (12-20 years old) Examples of Earth Reinforcement
United Kingdom
Oldham bypass retaining wall
M25lM40
Channel Tunnel, U.K. terminal
Mid-Rhondda link road
Cardiff dock
Blaemont Colliery, NCB
Canary Wharf
East Beckton development, Dockland Light Railway
Leeds multiplex cinema
Risca-Rogerstone bypass
Bournemouth International Centre
Bournemouth ring road
Reading inner distributor road
Ipswich inner ring road
West Th u mk development
Connaught Crossing, Docklands
Beith blast walls, Scotland
Blochain ring road, Strathclyde
Middlesbrough bypass
Australia
Epping Road overpass, Sydney
Bondi Junction bypass, Sydney
Reunion
1 I km sea wall
Spain
BarcelonitLe Perthus expressway
Architerra: tiered living units on hillside, near Madrid
South Aji-ica
Truck dump at Silicon Smelters mine
Canada
Gasp6 Peninsula: sea walls including Mont St. Pierre
St Josephs Oratory, Montreal: parking area support
United States
Interstate Route 80N (Second Narrows section of Snake River)
Sprain Brook Parkway (18 km NE of New York City)
Downtown Huntsville, Alabama: bridge abutment
Valdez, Alaska: containing walls for tank farm
Slot structure for coal storage and handling: Ken-McGee East Gillette Mine 16,
Vail Pass, Colorado, Rocky Mountains: 14 retaining walls
France
Pyrenees (first project 1966)
Wyoming
Sources: McKittrick and Darbin (1979) and others.
368
by
of the product was to support mountain roads, such as the A40 Lyon to
Geneva. The longest project up to then was an 11 km wall on the St. Denis
road on Reunion Island in the Indian Ocean. No theoretical size limitations
exist: a dam in Arizona is 170 long with 17 thick reinforced earth; a 40 m
man-made hill in South Africa has a pressure at the base of the structure of 16
tons/m2 absorbed by the reinforced earth.
5
Organic and Inorganic Chemicals,
Building Materials and Coatings,
and Fuels
This chapter commences with advice on the cleaning of zinc surfaces. The
major corrosion interests are then considered in five groups: (a) inorganic
chemicals and their aqueous solutions, (b) organic materials and fuels, (c)
concretes, cements, and plaster, (d) bitumen, and (e) other building materials.
Consideration must be given not only to the effect of corrosion on the
zinc, but also to the possible effects of corrosion products in noncorrosion
aspects; for example, solid corrosion products may prevent a valve from
shutting. corrosion that may be too slight to affect the zinc may
significantly contaminate the chemical with which it is in contact or change
the texture or appearance of a consumer product. This chapter consists largely
of brief references to actual tests and an indication of the apparent suitability
of the material for contact with zinc, plus lists of chemicals known to be
stored satisfactorily in zinc-coated steel containers.
Many foods and drinks are not affected by contact with zinc but, in
some cases, this property may give rise to unacceptably high levels of zinc.
Long contact with zinc should be avoided with most acid foodstuffs. In
general, zinc levels up to 50 ppm (50 mg/L) will not be harmful to human
health, but many regulations set an unnecessarily low level for zinc. This
subject is outside the scope of this book but is covered in Zinc Handbook
(Porter, 1991).
371
372 Chapter 5
Zinc-coated tanks and cylinders are widely used in oil refineries and
other plants for storing oil and petroleum products, chlorine, carbon dioxide,
and other industrial gases. Refrigerating plants and cooling equipment, as
well as degreasing plants, are almost universally protected by a zinc coating.
Zinc-coated steel is also used on structural steelwork around chemical plants,
where it is exposed to high humidity and a variety of chemical fumes.
Galvanized steel is used extensively for roofing and siding on buildings that
house pulp and paper processing operations. Other uses of zinc-coated steel in
the chemical industry include applications in floating-roof-type storage tanks
for volatile liquids, and galvanized wire cloth and mesh belts for the move-
ment of chemicals through various production stages. Galvanized steel con-
tainers are used to store strategic chemicals in outdoor locations. From these
representative examples, it is apparent that zinc and zinc-coated steel are very
useful in the chemical industry.
CLEANING OF ZINC SURFACES
Certain water-based solutions are used to clean zinc surfaces, notably to
remove corrosion products from test specimens without removing uncorroded
zinc. ASTM Standard Recommended Practice G1 calls for the use of ammo-
nium hydroxide solution followed by boiling chromic acidlsilver nitrate for
cleaning zinc test panels after exposure. Noranda trials show the correction
factors to be applied for metal lost in the cleaning procedure (Fig. 5.1).
Swedish Standard (draft):SS 11 72 11, Accelerated Outdoor Test by Intermit-
tent Spraying of a Salt Solution (Scab Test), recommends using 200 g of Cr03
in distilled water to make lo00 mL for 1 minute at and warns that
chloride contamination of the chromic acid can cause zinc corrosion.
When wet storage staining has occurred, the objects should be arranged
so that their surfaces dry rapidly. The attack will then cease and, with a free
supply of air to the surfaces, the normal protective layer of corrosion products
will form, provided the reaction has not consumed too much zinc (i.e., that
not more than 10-15 pm of zinc coating has been consumed). The white
corrosion products are gradually washed off, and the surface of the coating
takes on the. normal appearance of an exposed hot dip galvanized object.
A guideline for users appears in Swedish standard SMS 2950:
The special form of corrosion known as wet storage stain (white rust)
that sometimes appears on hot dip galvanized objects after a short
period of storage shall not regarded as sufficient cause for not
granting approval, unless a special agreement has been reached that wet
storage stain shall not be allowed.
Chemicals, Building Materials, Fuels 373
IMMERSION CONDITIONS 1
N h O H
SOLUTION SOCUTION
CrO,/AgNO,
a SHC ZINC VPRIABCL: CONSTINT (IS
24 -27 VARIPBCE CONSTbNT (IS # l
IMMERSION TIME IN THE OR SOLUTIONS, min
5.1 Base metal loss of zinc and of ZA 27 when cleaned to G1 procedure
(1SOmUL NH,OH s.g. 0.90 followed by boiling 50 g/L CrO, + 10 g/L
(Noranda, 1991).
If desired, white corrosion products that have already formed can be partially
removed by brushing with a stiff bristle brush. They can be removed com-
pletely by washing with 10% acetic acid, which must be followed by very
thorough rinsing with water. Also recommended for removal of white zinc
corrosion products is a paste consisting of parts by weight of "Cham-
pagne chalk" (a fine white powder from northern France) and parts by
weight of a 10% caustic soda solution in water. After using the paste rub off
the corrosion products, rinse in water and dry.
Another method is to use a dilute solution of sodium tartrate and
ammonium carbonate in water, to which chalk is added. With a sponge
soaked in this solution, the white corrosion products can be easily removed.
The original bright, metallically lustrous surface cannot be restored by these
treatments, however.
General
Mechanical methods (wire brushing, steel wool, abrasives) of cleaning rolled
zinc are rarely acceptable. Areas cleaned by such methods tend to contrast
374 Chapter 5
with adjacent untreated zinc and may take a long time to weather to a uniform
color.
B. Water Staining
Mild staining can be removed by swabbing-repeated, if necessary, with
ammonia (strong household ammonia) followed by rinsing with water.
The following solution is more effective; the composition is not critical:
Ingredient Amount
Ammonia (0.880)
Ammonium chloride
Ammonium carbonate
Water (tap)
Swabbing, like the mechanical methods, should used only in the open, or
in well-ventilated conditions, and protective goggles should be used.
C. Rust Staining
Water draining onto zinc from rusting steelwork (e.g., scaffolding) often
causes conspicuous and permanent brown staining.
The remedy is to swab with a cold water solution of commercial quality
oxalic acid solution (or alternatively use a proprietary solution for the removal
scale from pots and pans) followed by rinsing with water. Oxalic acid is
toxic; rubber gloves and goggles are recommended.
D. Cement and Mortar Stains
Portland cement mortar droppings tend to adhere strongly to zinc and when
fully hardened are often difficult to remove completely without damaging the
metal.
After deposits have been removed as far as practicable, use oxalic
acid as for rust stains. Hydrochloric (muriatic) acid and proprietary mixtures
based on it are sometimes used, but this approach is not recommended
because the action on zinc is too vigorous and it is difficult to rinse off
residues.
Test samples of metal that have been embedded in concretes, and so on,
can be divested of much of the adherent material by gentle heating-the
adherent material will crack off.
E. Finger Marking
Polished zinc surfaces, such as organ pipes, finger-mark very easily when
handled; such marks may be hardly visible initially, taking hours or days to
develop, but once evident, they remain conspicuous even after the metal
has tarnished. To avoid finger marking on zinc surfaces, the objects are
sprayed (before handling) with a clear light-fast acrylate or methacrylate
varnish (dry film thickness 10-15 pm).
The marks may be enhanced by clear lacquer finishes. If this happens,
any lacquer finish must be stripped off and the marks treated as under Section
11. B. Local repolishing may be necessary. Finger marking can be avoided by
wearing fabric gloves and can also be suppressed by the sparing application of
white vaseline (laboratory chemical grade), rubbed on with a clean, soft cloth.
F. Bitumen-Based Products (asphalt, etc.)
Contamination from bitumen-based products arises mainly from accidental
spillage on building sites. The remedy is to clean the surface as soon as
possible with wooden scrapers. Then swab repeatedly with clean cloths and a
solvent such as paint thinner (white spirit) or xylene/toluene (solvent
naphtha). Avoid spreading, and do not apply heat.
C. Conventional Paints
If still wet, clean conventional paints with appropriate paint thinners.
Hardened paint should be removed with a nonalkaline organic paint stripper.
Burning off or the use of steel scrapers is not advisable.
H. Staining by Beer, Wines, and Spirits
Zinc countertops are occasionally used in bars and cafeterias. Particularly
when the metal covering is new, it may develop local stain rings under
glasses containing wine or spirits.
The remedy is to treat with a mixture of ammonia, ammonium chloride,
and ammonium carbonate mixture as described in SectionIi. B.; polishing or
the use of abrasives is not recommended.
Usually, a soft gray matte finish is desired, and periodic wiping with a
beer-moistened cloth (in England draft bitter has been found most effective)
helps to achieve this effect by developing a stain-resistant skin.
376 Chapter 5
Detergents and Cleansers
Mildly alkaline solutions of ordinary laundry soaps develop a protective
coating on zinc and its alloys. They are satisfactory for both warm and cold
applications. Detergent solutions are more corrosive; however, the better
grades include inhibitors, which reduce corrosion to allow reasonably satis-
factory service for zinc alloys. Strong alkalies (e.g., hypochlorite bleach
solutions) or acid solutions (e.g., Harpic, which contains trisodium hydrogen
disulfate) are to be avoided.
Table 5.1 gives results for five series of immersion trials designed to
show the short-term and long-term effects of both simple chemicals and
commercial solutions. Although the corrosion rates are calculated per year
(for consistency with other tables), it must be remembered that often the
contact time will be in hours or even minutes per year; hence, quite high
corrosion rates may be acceptable. Some tests were made on zinc sheet and
some on hot dip galvanized material. The results should be of the same order,
this detail is normally omitted from the tables in this chapter. When the
original source shows significant differences in the two materials, it is antici-
pated that the dissimilarities are due to surface films (e.g., chromating) being
left in place unwittingly. Trials in the United Kingdom were made by BNF
and by Imperial Smelting (unpublished) at room temperature, at 70"C, and at
100C on commercial detergents in use in Britain in 1959. Corrosion was
significant at the higher temperatures, suggesting that for long life of die-cast
components in washing machines, a protective coating is desirable.
Trichlorethylene
Normal trichlorethylene degreasing has no effect on zinc or zinc-aluminum
alloys, provided immersion in solution that has become degraded or contami-
nated is avoided.
INORGANIC CHEMICALS
In the presence of moisture such as rain or condensation, or in solution, many
chemicals react with zinc. Corrosion is serious if the reaction of the chemical
with zinc forms acidic products. Mixtures of chemicals can sometimes be
more corrosive than the sum of their effects taken separately but, when such
mixtures come from natural sources, there are often unsuspected inhibitors
present, which give a net beneficial effect.
A short immersion trial is desirable for water-based solutions. Linear
extrapolation of the results will usually indicate the worst possibility, since
the attack often reduces with time.
"
Jzc
m m
.c .e .e
M M M M
"" 9 9 9 8
o o u u
377
"
m m
"d"
I , , , a
, . . .
Bt j Bt i M
.- %l
s g
2
.a
\
3333
c c c c
mv ) v ) m
378
-
: 2
8
9
W
CD
0
Table 5.1 Continued
Letter code for refmnees cited in the table:
I = International Nickel Co. (direct information)
K = Smpcl and Koch (1959)
L = Bukowiecki (1958)
M = Bablik and Belohlavy (1957)
P = Bacon and Nutting (1952)
Time
(days unless Conosion rates
otherwise Temperature
Materials indicated) (0 #m2/5 cycles cun/ycar Remarks Refs.
Tmdsodium pyrophosphate
0.5 .@in 0 g water
0.5 g/L in 21 g water
0.5 g/L in 21 g water + 1 g/L
1.0 g/L in 21 g water + I g/L
2.0 &n in 0 g water
2.0 g/L in 21 g water + 1 g/L
3.0 g/L in 21 g water + I g/L
m e
pe-e
pe-e
Pe-w
6.0
I .2
1.5
11.2
4.7
5.1
25.8
19.0
21.2
34.5
27.2
32:R
K
K
K
K
K
K
boiling + 15 min cooling: K
each cycle K
g/m2/5 cycles x 328 = pn/year K
Aeration and agitation from boiling K
K
K
K
5 cycles, each 30 min at
+ solution renewed after
381
P a a
\ o w W
382
c
0
K
8
383
.e .e .e
% * R %
384
"6
385
TaMe 5.1 Continued
Lencr code for rcfcrrnceJ cited in thc ale:
I ss Intunacional Nickel Co. (dirrn infonnulon)
K s Srupel and Koch (1959)
L s Bukowkki (1958)
M s Bablik and Belohlrvy (1957)
P s Bacon and Nuning (1952)
Time
{days unless Comsion raw
Marerials indicated) (0 glm2/s cycles ~~~ml year Rca~arks Refs.
0.18 %L sodium nitrate 0 g H20
0.3 &/L magnesium lactate 0 g H20
0.3 g/L magnesium lactate 21 g H20
0.3 g/L calcium chloride 0 g H20
0.36 g/L sodium thiosulfate
0.3 g/L sodium aluminate 0 g H20
0.3 g/L sodium perborate 0 g H20
0.3 glL sodium perborate 21 g &,
0 g/L perborate - 0 g H20
2 p/L perborate - 0 g H,O
4 g!L perborate - 0 g H20
I5 g/L perborate - 0 g H20
Syndet 1, 6 g/L + magnesium silicare 0.18 &+
8 dLperborate - 0 g H20
I2 g/L perborate - 0 g H2 0
20 g/L ~r bont e - 0 g H20
24.1
19.3
21.0
16.0
16.8
21.7
56.2
34.7
51.9
55.8
64.8
75.1
79.0
81.2
82.0
81.9
80.4
63.7
K
K
K
K
K
K
K
K
K
K
K
K
K
K
K
K
K
K
387
388 Chapter 5
Zinc is most useful in solutions with pH 5.5-12 and less useful in acidic
or strongly alkaline chemical environments encountered, for example, in the
chemical process industries. Corrosion of zinc increases rapidly in aqueous
chemical solutions on either side of the pH 5.5-12 range. This rule should not
be considered to be fixed, however, because many other factors, such as
agitation, aeration, temperature, polarization, and, in some cases, the pres-
ence of inhibitors, may have considerable influence on the corrosion.
Zinc is widely used in contact with many milder chemical specialties
such as detergents, agricultural chemicals, and similar materials. In most
cases, zinc or zinc-coated steel comes in contact with chemicals during the
handling, packaging, and storage of the commercial products.
Table 5.2 contains corrosion rate data for chemicals and chemical
specialties assembled from various sources. Particular mention must made
of the compilation by Wiederholt (1976), which has been used extensively to
supplement other data and some references quoted here are only given in full
in that report. The chemicals and simple mixtures are listed alphabetically,
with the main chemical of the mixture appearing first. The test conditions
(temperature, duration, etc.) are also tabulated. There was very little unifor-
mity in these values among the various investigations. For example, test
temperatures ranged from room temperature to the boiling point and time of
immersion from minutes to several months. The test procedures also included
exposure to vapors and partial immersion or total immersion in the solutions.
The test conditions adopted in each investigation may have been dictated by
specific objectives but, in comparing results of the tests, the possible influ-
ence of the test variables should be taken into account. Some of the corrosion
rates reported for the chemical specialties given in specialized units
dictated by the test procedures used by the original investigators; usually,
however, rates have been converted into micrometers per year when attack is
generally fairly uniform. In most cases, the short test periods give higher
annualized corrosion rates than would occur with continuous exposure for a
year.
Table 5.3 covers agricultural chemicals (including pesticides, both
inorganic and organic): Table 5.4 covers fire-extinguishing materials.
Some of the trials reported in Tables 5.1-5.4 are on zinc and others are
on zinc-coated steel; normally, similar performance is expected.
The behavior of fertilizers in relation to crop-spraying aircraft is exam-
ined by Marshall and Neubauer (1955), who rate copper or cobalt sulfides
(either on their own or in superphosphates) as unsuitable for use with gal-
vanized steel equipment. Ammonium phosphate (measured corrosion rate 250
pdyear) and nitrolimes (corrosion rate 300 pdyear) can be used only with
care, for short times. Salt (100 pdyear) is better, as is calcium phosphate,
while chalk and bone meal have little effect on galvanized steel.
.S
"-
.-
$ $ $ 2
-
VI
389
Table 5.2 Continued
Some references that occur frequently (and are mainly tabular summaries) are referred to in this table by letter code:
W
b D
0
A = Helwig and Bird (1973)
B = Ritter (1958)
C = R i d (1964)
D = Dechema (1953)
E = Bauer and Schikorr (1934)
F = Zinkberatung (1959)
G = Wiederholt (1976)
H = Friend and Tidmus (1924)
I = International Nickel Co. (direct data)
I = Fuller (1927)
N = New Jersey Zinc Co. (direct data)
R = Clarke and Lonehunt (1961)
Corrosive media
Corrosion rates
Time Temperature General
(days) ("C) g/m2/year Fdye a r suitability" Remarks Ref.
Ammonium acetate
Solution
0.2% solution
5% NH,F - HF
10% NH,F - HF
20% NH,F - HF
Ammonium bisulfate
Ammonium bromide
Ammonium carbamate
Ammonium carbonate
Ammonium chloride
Ammonium acid fluoride
0.1 N
10%
Zn-1% Cd
Zn-1% Pb
Technical grade 10%
60
1 20
1 20
14 20
2 35
20
20
20
20
6,900
13,900
15,330
694
1,390
62 1
621-1.715
d
d
965 d
1,950 d
2,150 d
d
b
b
b
C
97 b-c
195 C
87 k
87-240 b 4
Corrosion products not
removed
Inhibitor in acids
D, Kloetz (1965)
B
D, Kloetz (1%5)
Miskidch'jan (1970)
D
D, Roebeck (1972)
52
W
4 4 8 4 8 8
391
392
"
-l.
"
cia
8
8 8
393
394
P
NN
Table 5.2 Continued
Some references that occur frequently (and are mainly fabular summaries) are referred to in this fable by letter code:
B = Ritter (1958)
C = Rial (1964)
D = Dechema (1953)
F = Zinkberanmg (1959)
G = Wiederholf (1976)
J = Fuller (1927)
R = Clarke and Longhursf (1961)
Corrosion rates
Corrosive media (days) (C) g/m2/year pdyear suitability- Remarks Ref.
Iodides
Iodine
Solution
Motten, dry
Moist
Gas
10% solution
60% solution
Iron chloride
Iron sulfate
100% solution
Lead nitrate
Moist
100% solution
60
80
80
100
80
With sulfate
b
d
d
d
b For dissolving metal coatings from
cast zinc
d
d Forms NO, and O2
Radovici and Kovacs (1968)
D
D
Ginzburg (1969/7 I )
B, D
C
Ruegg (1970). C
B, D
D, Huber (1968)
d C
m
2
U
.-
P
E
397
W
Table 5.2 Continued
Some references that occur frequently (and are mainly tabular summaries) are refemd to in this table by letter code:
B = Ritter (1958)
C = Ried(1964)
D = Dechema (1953)
F = ZinkberaNng (1959)
J = Fuller (1927)
R = Clarke and Longhurst (1961)
Corrosion rates
Corrosive media (days) (C) glm2/year pdyear suitability Remarks Ref.
Manganese dioxide
Manganese phosphate
Dry Pure a D
Solution
a Phosphating solution Dettner and Eke (1964-1%9),
Neuhaus and Gebhardt
Mercury
100%
Mercury salts
Electrolyte
Milk of Magnesia
Molybdenum chloride
Molybdate ions
Nickel acetate
Wet
60
High
d Formsamalgams
b
b Stored in glass
( 1966)
D
Sony (1966). Mallory (1967)
G (many)
b Equipment uses, MnfZn alloy Heumann and Venker (1968)
b Protective layer forms
d D
g M
3
8
E
.-
399
e
8 Table 5.2 Continued
Some references that Occur fnquently (and are mainly tabular summaries) rn referred to in this table by letter code:
B = Rimer (1958)
C = Ried (1964)
D = Decherna (1953)
I = International Nickel Co. (dimt data)
J = Fuller (1927)
Corrosion rates
Corrosive media (days) (C) g/m2/year pdyear suitabilityW Remarks Ref.
Perchlorate
Solution
Pmhloric acid
Solution
Pemranganate
Solution
Peroxide
Moist
Persulfate
Solution
Phosphates
fire, dry
Mechanism studies
b In polishing solutions
b In dye solutions
b
d
d
wood preservatives, polishes,
laundry powders, surface coatings
Schwabe (1957). Gaspar and
Soriano (1971)
Dettner and Elze (1964-1969)
ILZRO Digesr (1970)
Gmelin (1956)
Gmelin (1956)
B
G (many)
.-
g
S
-
I
m
II
401
402
x d n n
n
u'nn
d
n
n n u u w n w w n n n w w n
403
Table 5.2 Continued
Some references that occur frequently (and are mainly tabular summaries) are refened to in this table by letter code:
0
B = Ritter (1958)
C = Ried(19W
D = Dechema (1953)
F = ZinkberaNng (1959)
I = Inrernational Nickel Co. (direct data)
J = Fuller (1927)
R = Clarke and Longhunt (1%1)
Corrosion rates
Corrosive media (days) ("C) g/m2/year pdyear suitability" Remarks Ref.
Silica
Solution
Sodium acetate
Sodium aluminate
Sodium aluminum fluoride
Sodium aluminum sulfate
Sodium arsenate
10% solution (pH 10)
Sodium arsenite
Sodium benzoate
2% solution
Sodium bicarbonate
Saturated solution
5%
@gR.
20
20
80
20
1,970
1,970
b With 0.01 N KOH
b
d
d
d
215 d
275 C
b Steel anodic to zinc
d
C
L o d e k and Petric (1971).
Peujc (1972)
Konnik (1969)
D
D
D
D
D
D
C
B
405
g Table 5.2 Continued
a-
Some references that occur frequently (and are mainly tabular summaries) am r e f e d to in this table by letter code:
B = Riner (1958)
C = Ried (1964)
D = Dechema (1953)
J = Fuller (1927)
N = New Jersey Zinc Co. (dinct data)
Corrosion rates
Corrosive media (days) (C) g/m*/year pdyear suitability- Remarks Ref.
Molten
3% solution 2-14 3,47&5,140
0.58% 2 35 3,336
1.45% 0.25 N 2 35 4,245
2.9% 0.5 N 8.5 25 2,376
20% immersed 24 RT 59
20% spray 21 35 2,628
63 35 1,168
1 I9 35 2,957
21 35 2,701
63 35 730
119 35 803
485-720
470
595
330
8
370
164
415
380
I02
I12
d
d
d
d
b
C
C
C
C
C
C
C
Imperial Smelting Co. (1%9),
B
I
I
I
I
Zinc sheet: the erratic I
results are. not explained I
I
Galvanized steel: it would I
appear that corrosion I
products slow reaction with I
Ishutin (1971)
time; pitting up to 500 pm
in 119 days
P
cl
m m
407
2:1, 10%
2.4:1, 1%
2.4: 1, 5%
2.4:1, 10%
Sodium silicofluoride
Sodium stannate
Sodium sulfate
5-300 gR.
10 g/L
Sodium sulfide
Sodium sulfite
Sodium thiosulfate
Saturated solution
Sulfides
Pure, dry
Sulfur dioxide
Liquid
0.6% vapor
Zinc chloride
Molten
Zinc chromate
Zinc sulfate
Zinc sulfides
Zirconium oxide
60
14
60
60
60
60
12
bP
150
20
14 20
7 40
500-700
800
3,300
400
550
1,460
475-473
584
2,860
13,320
1,095
672
460 C
55 b
75 b
200 C
d
d
65-10 b
80 b
400 C
d
1850 d
152 C
d
d
b
b
b
93
&day test H
14day test B
Galvanized apparatus D
for concentrating
D
pH 7.8 G
Discolored surface Gmelin (1956)
+ 1% H20 + 0.21% O2
89% RH
D, Bollinger (1952)
Sanyal and Bhadwar (1959)
G
Kemdge and Tariq (1967)
Parker Co. (1967)
G
In 5 N NaOH Merkulov and Flerov
Zn with 10% ME reduces ZIO to Zr gar^ (1970)
( 1968-1 969)
'Most dry materials have no significant corrosion reaction with zinc: those that do m specifically mentioned. This general classification usually covers either chemical solution or moist materials.
Where specific corrosion rates am not given, there is a general classifation.
bRatings: d = definitely not suitable (>3000 pdyear): c = corrosive (1OO-3OOO kdyear); b = borderline for zinc; bad for coatings (1&100 pdyear); a = acceptable usually (< 10pdyear).
'It should be noted that these ratings arc not always aligned with results of short-term tests; often the zinc corrosion products formed initially slow down corrosion subsequently.
-
e
S
.-
0 0
z z
a a
0 0
Agricultural lime
Burnt lime
Basic slag
Bone meal
Phosphorized Pollard
Damp granular fertilizers
Type Components Water (96)
11-48-0 NH4HZm4 ' (NH4)2Hm4
1148-0 NH4HzPO4 ' (NH4)2HPO4
12-12-12 11-48-0 (NH4)2SO4 * KCI
12-12-12 11-48-0 (NHJ2S04 * KCI
13-16-10 11-48-0 (NH4),SO4 * KCI
13-16-10 11-48-0 (NH4)zS04 * KCI
16-201) 11-48-0 (NH4)ZS04
16-20-0 11-48-0 (NH4)ZSOd
35.500 NH4NO3
35.5-0.0 NH4NO3
21-0-0 (m4)zso4
21-04 (NH4)zS04
(a) 11-48-0
(a) 11-48-0
(a) 11-48-0
(a) 11-48-0
(b) 16-20-0
(b) 16-20-0
(b) 16-20-0
(b) 16-20-0
(c) 33.5-0-0
(b) 33.5-0-0
(b) 33.5-0-0
(b) 33.5-0-0
Damp granular fertilizers
(d) 21-0-0
5.41
5.41
6.18
6.18
6.68
6.68
4.10
4.10
3.54
3.54
2.50
2.50
5.4
5.4
5.4
5.4
4.1
4.1
4. I
4.1
3.5
3.5
3.5
3.5
2.5
121 Zinc not penetrated
121
121
121
121
33
33
33
33
33
33
33
33
33
33
33
33
11
11
I 1
I 1
11
11
I1
11
11
11
11
11
11
60
60
60
60
60
60
60
60
60
60
60
60
RT
RT
60
60
RT
RT
60
60
RT
RT
60
60
RT
2,920
1,675
36,680
53,290
1,825
2,010
1,280
5,695
83,585
28,760
800
250
1,370
720
1,405
518
600
518
3,047
2,830
9,125
2,400
2 I ,720
6,130
1,350
400
240
5,140
7,460
255
280
I80
800
I 1,700
4,030
112
35
190
100
200
73
84
73
425
400
1,280
335
3,040
860
190
Alternate humidity, weekly
cycle forced air circulation
24 h, 100% RH (1). 24 h
dry air (2); repeat ( I ) and
(2) then 8 h 100% RH, 64 h
dry air
Fresh water once a week; SHG
A G 4 A
SHG
A G 4 A
SHG
A G 4 A
SHG
AG-40A
SHG
A G 4 A
SHG
AG-40A
SHG
A G 4 A
Four changes of
SHG moist fertilizers
SHG
A G 4 A
SHG
S
S
S
S
S
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
Table 5.3 Continued
Letter code for references cited in this table:
d
h,
D = Decherna (1953)
N = New Jersey Zinc (direct data)
0 = Cook and Dickinson (1950)
S = Marshall and Neubauer (1955)
T
= Consolidated Mining and Srnelfing Co. of Canada Ltd. ( dmt data)
Materials
Time (days,
except as Corrosion rates
indicated) ("C) mB/m2/year *&year Remarks Ref.
(d) 21-04
(d) 21-0-0
(d) 21-0-0
(e) 27-14-0 (a), (b)
(e) 27-14-0 (a), (b)
(e) 27-14-0 (a), (b)
(e) 27-14-0 (a), (b)
(0 8-32-16 (a), (d), KCI
(f) 8-32-16 (a), (d), KCI
(f) 8-32-16 (a), (d), KCI
(f) 8-32-16 (a), (d), KCI
(g) 6-24-24 (a), (d), KCI
(g) 6-24-24
(g) 6-24-24 (a), (d), KCI
(g) 6-24-24 (a), (d), KCI
Liquid fertilizers
Type: 27-14-0, 25% solution
Tvoe: 27-14-0. 25% solution
2.5
2.5
2.5
4.5
4.5
4.5
4.5
5.0
5.0
5.0
5.0
5.2
5.2
5.2
5.2
11
11
I I
I 1
11
11
I 1
11
I 1
11
11
I 1
11
11
11
21
14
21
14
RT
60
60
RT
RT
60
60
RT
RT
60
60
RT
RT
60
60
RT
RT
60
60
720
3,085
1.725
3,360
1.090
12,590
2,210
3,360
1,185
4,745
3,650
3,395
1.640
4,930
2,460
1,240
4,180
1,625
3,830
100
430
240
470
152
1,760
310
470
166
665
510
475
230
690
345
174
585
227
540
Ti;: 27-14-01 25% solution 21 RT 237 33
A G 4 A
SHG
A G 4 A
SHG
A G 4 A
SHG
A G 4 A
SHG
AG-40A
SHG
A G 4 A
SHG
A G 4 A
SHG
AG 40A
p Four changes of
moist fertilizers
in all tests
Oxygen added daily
Oxygen added daily
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
s s
41
Table 5.3 Continued
Letter code for references cited in this table:
A
P
d
D = Dechema (1953)
N = New Jersey Zinc (direct data)
0 = Cook and Dickinson (1950)
S = Manhall and Neubauer (1955)
T =
Consolidated Mining and Smelting Co. of Canada Ltd. (direct data)
Time (days,
Materials indicated) ("C) g/m2/year WYex Remarks Ref.
Pesticidedfungicides
Bordeaux mixture: 10 gR. copper sulfate
+ 10 g/L quicklime"
20 RT 5 N
Calcium arsenate'
20 RT Gained weight N
0
N
0
D
Chlordan+water emulsion 56 weeks 25 2,628
Copper sulfate - 10 glL" 20 RT 7,850
DDT, 5% in distilled water 56 weeks 25 840
DDT, 5% solution in saltwater 25 1500 D
DDT, 10% solution in kerosene 25 5 'Solutions prepared from
Dry lime sulfur - 36 g/La 20 RT 166 weight changes as removed
Formalin" 20 RT Gained weight
N
20 RT 88
N
Paris green" 20 RT Gained weight
N
Pyrethrum' 20 RT Gained weight
N
Xyloi 5% Triton X 100
25 985 140 D
20% solution 1 month 30 Trace Solutions changed weekly T
Saturated urea solution
I month 30 31 5 Solutions changed weekly T
DDT emulsion, 95% water and 5%
commercial products,
changed after 3 and 6 days,
mixture of 25% DDT, 65%
N
N
White hellebore - 2.5 g/L' 20 RT Gained weight from test
Lead arsenate: 4 g/La
Urea
1 month 30 369 52 Slight pitting, kept moist daily T
Moist urea salts, 3% water
Chemicals, Building Materials, Fuels 41 5
Table 5.4 Weight Change of Zinc4% Copper-0.2% Aluminum Alloy After
5 Weeks in Fire-Extinguishing Solution at 60C"
Fire-extinguishing liquids
Weight losses
(mg/dmz/day)b
g Al(OH),, g KHC03, g Na,SO,,
g Al(OH),, g KHC03, g NaZSO4,
g K2S04, 10 g foaming agent per liter of solution
88 g KzS04, g glycerin, g glycol, g
foaming agent per liter
g NCO3, g NaHCO,, g foaming agent
per liter
g NaHCO,, g KHCO,, g glycerin,
g glycol, g foaming agent per liter
g NaCl + g NazC03 per liter
500 g commercial KzCO3 per liter
Carbon tetrachloride (water-free)
Carbon tetrachloride (water-freehvapor
Carbon tetrachloride + v/o water
+
+
+
+
+
Gain
Gain
Gain
Tooled overnight.
bWeight changes determined without removing corrosion products.
Source: Beythein (1947).
Fungicides and pesticides were studied by Steinrath (1960), who found
many the proprietary products then available suitable for use with gal-
vanized steel. New Jersey Zinc data are given in Table 5.3, but other work
reports Bordeaux slurry as having a corrosion rate of 165 pdyear , hence not
suitable for continued contact with zinc. New Jersey work on herbicides is
also given in Table 5.3. Some agricultural chemicals are in other tables but
can be located through the index.
Table 5.4 deals specifically with work by Beythein (1947) on the
corrosion of phosphated zinc alloys by the fire-extinguishing liquids then
available.
Since most galvanized and thermally sprayed coatings are around 100
pm thick, continuous contact with any material that gives a corrosion rate in
excess of pdyear is likely to result in unsatisfactory or uneconomic
service unless supplementary coatings are provided. However, zinc and zinc
alloy coatings and wrought zinc may well be suitable in many uses, even with
rather higher rates of attack.
Additional early qualitative information is given by the American Zinc
Institute (1928), and Wiederholt (1976) has additional information of a
qualitative nature.
41 6 Chapter 5
IV. ORGANIC CHEMICALS AND FUELS
Zinc is used in contact with many organic chemicals during the handling,
packaging, and storage of commercial products. Many organic liquids are
inert in contact with zinc. Table 5.5 gives a guide to some that have been
stored successfully in galvanized containers and are even more likely to be
satisfactory in contact with sheet zinc or with castings in zinc or alloys.
Interpretation of test results requires care since, although total corrosion may
be small, either the resultant contamination of the liquid may be unacceptable
or small amounts of solid zinc corrosion products may cause problems (e.g.,
by preventing the free action of valves in a circulating system). Corrosion
may arise when material is moist or water is present as a contaminant in the
liquid.
Table 5.6 summarizes the results of a comprehensive evaluation of the
behavior of galvanized steel in contact with synthetic organic chemicals.
These data, furnished through the cooperation of the Union Carbide Corpora-
tion, Chemicals Division, are the results Union Carbides evaluation of
materials for storage and packaging of the synthetic organic chemicals they
produce or supply.
The tests were on hot dip galvanized steel strips exposed to commer-
cially pure chemicals in stoppered bottles for a minimum of 100 days at
ambient temperature with part of the sample being fully immersed and part
exposed to vapors; the compatibility (i.e., the effect of contact with zinc on
the chemical properties of the test chemical) is observed. In Table 5.6, most
of the chemicals listed are rated as less than 13 Fdyear ( < OS mdyear),
which was regarded as low enough to show that zinc is compatible with the
chemical involved; further investigation may be necessary to determine
whether the rates were in fact much lower and the chemical suitable for use
with relatively thin zinc coatings.
Relatively few organic chemicals in the group are rated as unsuitable for
use in contact with zinc.
Information on organic chemicals from other tests is given in Table 5.7.
The corrosion of zinc in primary alcohols has been studied by Heitz et al.
(1970) in relation to alcohol chain length.
Air entrapped during encapsulation and vapor from polymeric materials
such as epoxy resin can cause corrosion. The corrosive effect of vapors from
plastics, rubber, and wood on zinc and cadmium has been tabulated by
Knotkova-Cermakova and Vickova (1971); other work, including electro-
plated zinc-nickel alloys, has been discussed by Donovan and Stringer
(1971).
Also, at 85C and 95% relative humidity, zinc can deteriorate in contact
Table 5.5
Hydrocarbons Esters Butyl Atnines and amine salts
Toluene (toluole) butyrate isobutyrate Pyrrolidine
Xylene (xylole) caproate caproate Methylpjperazine
Cyclohexene formate propionate
Some Chemicals That Have Been Successfully Stored in Galvanized Containers
3
3
Benzene (benzole) Ally1 butyrate Pyridine ii
EL
Dicarbethoxypiperazine a
5
Petroleum ethers propionate succinate l-Benzhydryl-4- -
p.
Heavy naphtha Ethyl titanate methylpiperazine 3
3
Solvent naphtha butyrate h P Y I
%
fu
VI
oa
2 : 4-Diamino-5-(4-
isobutyrate butyrate chlorpheny l-kthylpyrimidine)
Alcohols
Methyl paralynol (methyl caproate isobutyrate Hydroxyethylmorpholine
caproate (hydroxyethyldiethylenimide 8.
Morpholinoisopropanol propionate formate oxide) . k
pentynol) caprylate
T
Halides Amy1 isobenzoate guanidine 1
Glycerol (glycerin) succinate propionate p- Aminobenzenesulfonyl-
Carbon tetrachloride butyrate Butyl Butylamime oleate
Amy1 bromide isobutyrate butyrate Piperazine hydrochloride
Butyl bromide caproate caproate monohydrate
Butyl chloride caprylate isobenzoate Carbethoxypiperazine
Cyclohexyl bromide Methyl h P Y I hydrochloride (dry)
Ethyl bromide butyrate caproate Amides
Propyl bromide caproate formate
Propyl chloride propionate propionate Dimethylformamide
Trimethylene bromide succinate Cyclohexyl butyrate Miscellaneous
( I .3-dibromopropane) Benzyl Phenols Glucose (liquid)
Bromobenzene butyrate Phenol Benzilideneacetone
Chlorobenzene isobutyrate Cresols (methylphenols) p-Chlorbenzophenone
Aroclors and Pyroclors propionate Xylenols (dimethylphenols) Sodium azobenzenesulfonate
(chlorobiphenyls) succinate Biphenol (dihydroxybiphenyl) Melamine resin solutions
Nitriles (cyanides) Octyl 2.4-dichlorophenol Crude cascara extract
Diphenylacetonitrile butyrate p-chloro-o-cresol
p-Chlombenzolycyanide caproate Chloroxylenols
And other unspecified lilanates. u
VI
Formamide
P
d
Source: Galvanizers Association.
Table 5.6 Corrosion of Galvanized Steel in Contact with Organic Chemicals
Specimens of galvanized steel partially immersed in commercially pure test chemicals at ambient
temperature for a minimum of days; corrosion rate based on total area of specimen.
Corrosion rate Corrosion rate
Corrosion milslyear Corrosion milslyear
media (1 millyear = 25 prdyear) media (1 millyear = 25 pmlyear)
Acids
Acetic
Propionic
Butyric
Valeric
2-Methylpentanoic
2-Ethylhexoic
Isodecanoic
Di (Zethylhexyl) phosphoric
Acrylics
Acrylic acid
Ethyl acrylate
Butyl acrylate
2-Ethylhexyl acrylate
Decyl acrylate
Glycidyl acrylate
Alcohols
Ethanol, 190 proof
Isopropanol (19 v01 %)
Isopropanol (anhyd)
Butanol
2-Butanol
Isobutanol
l-Pentanol
2-Methyl-l butanol (commercial)
3-Methyl-l butanol
Primary amyl alcohol
l-Hexanol
Isohexanol
Methyl amyl alcohol
2-Ethyl hexanol
2,2,4-Trimethyl-l pentanol
Isooctanol
Diisobutyl carbinol
2-Ethylbutyric
2-Ethyl butanol
a
a
a
5.0
a
a
a
a
a
a
< O S
< O S
< O S
< O S
< O S
1 .o
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
<OS
< O S
< OS
< O S
< OS
< O S
Aldehydes
Formaldehyde
Normal pentaldehyde
Paraldehyde
Propionaldehyde
Valeraldehyde
2-Ethylbutyraldehyde
2-Ethylhexaldehyde
Methacrolein
Crotonaldehyde
Acrolein dimer
Glutaraldehyde-25% aq. soh.
2-Hydroxyadipaldehyde-25%
Alkanolamines
Monoethanolamine
Diethanolamine
Monoisopropanolamine
Diisopropanolamine
Triisopropanolamine
Alkyl amines
Ethylamine-(69%)
Diethylamine
Triethylamine
Propylamine
Dipropylamine
Isopropylamine
Diisopropylarnine
Butylamine
Dibutylamine
Amylamine
Diamylamine
Hexylamine
2-Ethylhexylamine
N-Methylbutylamine
N-Ethylbutylamine
aq. soh.
< O S
2.4
2.5
1.5
a
b
< O S
1.5
3.5
1 .o
a
a
a
< O S
< O S
0.5
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
Table 5.6 Continued
Corrosion rate
Corrosion rate
Corrosion mils/year Corrosion
milslyear
media (1 millyear = 25 pdyear) media (1 miYyear = 25 pdyear)
Isodecanol
2,6,8-Trimethyl-4 nonanol
Tridecanol
Alkylene amines
Ethylenediamine (98%)
Diethylenetriamine
Propylenediamine
1,3-Diaminopropane
Dimethylaminopropylamine
Diethylaminopropylamine
Imino-bispropyamine
Tetramethyl-l ,fbutanediamine
N-Methyl-bis (aminopropy1)-
Aryls
Phenol
Butylphenol
Amylphenol
Nonylphenol
Dodecylphenol
Ethylbenzene
Styrene, monomeric
Acetophenone
a-Methylbenzyl ether
Phenyl methyl carbinol
Chlorine compounds
Butyl chloride
2-Butyl chloride
2-Ethylhexyl chloride
Ethylene dichloride
Propylene dichloride
I , 1 ,2-Trichloroethane
l ,2,3-Trichloropropane
Dichloroethyl ether
Dichloroisopropyl ether
Triglycol dichloride
Ethylene chlorhydrin
Epichlorhydrin
amine
< O S
< O S
< O S
a
< O S
1.5
< O S
0.5
0.5
1.5
< O S
0.5
< O S
< O S
< O S
< OS
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
<OS
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
Esters
Ethyl acetate
Propyl acetate
Isopropyl acetate
Butyl acetate
Isobutyl acetate
Primary amyl acetate
Methyl amyl acetate
2-Ethylhexyl acetate
Methylcellosolve acetate
Cellosolve acetate
Butylcellosolve acetate
Carbitol acetate
Butylcarbitol acetate
Glycol diacetate
Diethyl sulfate
Ethyl silicate (condensed)
Tetraethyl orthosilicate
Tetra (2-ethylhexyl) orthosilicate
Ethers
Dimethyl ether
Ethyl ether
Isopropyl ether
Butyl ether
1,4-Dioxane
Glycol and triols
Diethylene glycol
Triethylene glycol
Tetraethylene glycol
1,3-Propanediol
Propylene glycol
Dipropylene glycol
Hexylene glycol
< O S
< O S
< O S
< O S
< OS
< O S
< O S
< O S
a
a 4.0
0.5
< O S
< O S
3.5
cos
0.5
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< OS
< O S
< O S
< O S
< O S
41 9
Table 5.6 Continued
Corrosion rate Corrosion rate
Corrosion milslyear Corrosion milslyear
media (1 millyear = 25 pdyear) media ( I millyear = 25 pdyear)
Glycol ethers
Methylcellosolve
Cellosolve solvent
Butylcellosolve
Isobutylcellosolve
Dibutylcellosolve
Hexylcellosolve
Methylcarbitol
Carbitol solvent
Butylcarbitol
Butoxytriglycol
Methoxytriglycol
Ethoxytriglycol
Ketones
Acetone
Methyl ethyl ketone
Methyl propyl ketone
Methyl isobutyl ketone
Methyl n-amyl ketone
Methyl isoamyl ketone
Ethyl butyl ketone
Diisobutyl ketone
Mesityl oxide
Isophorone
Diacetone alcohol
Latices
Acrylic latex
Styrene-butadiene latex
Polyvinyl acetate latex
Vinyl acetate-acrylic copolymer
latex
Monomers
Acrylonitrile
Styrene, monomeric
Vinyl acetate
< O S
<os
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< OS
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
< O S
a
a
a
< O S
< O S
< O S
Morpholines
Morpholine (tetrahydro-l , 4-
oxazine)
2,6-Dimethylmorpholine
N-Methylmorpholine
N-Ethylmorpholine
Nitriles
Acetonitrile
Butyronitrile
Acrylonitrile
Ethylene cyanohydrin
Oxides
Ethylene oxide
Propylene oxide
Butylene oxide
Styrene oxide
Plasticizers
Dibutyl phthalate
Dioctyl phthalate
Polyethylene glycols
Polyethoxyethylene glycols
Polymethoxyethylene glycols
Polypropylene glycols
Pyridine and piperazines
a-Picoline
y-Picoline
2-Methyl-5-ethylpyridine
N-Hydroxyethylpipine
N-Aminoethylpiperazine
Vinyl ethers
Vinyl ethyl ether
Vinyl butyl ether
Vinyl isobutyl ether
Vinyl 2-chlorethyl ether
< O S
0.5
< OS
<OSd
< O S
<os
< O S
a
<0.5
< O S
< OS
< O S
< O S
<OS
< O S
< O S
< OS d
<OSd
< OS d
0.5
< O S
< OS
<os
< O S
< OS
Toating completely destroyed.
bzinc coating partly removed.
Incomparable-causes polymerization
dProduct contaminated.
Partially dissolved.
Source: Union Carbide, Chemicals Division.
Table 5.7 Corrosion Data for Zinc and Zinc Coatings with Organic Chemicals
Some referrnee$ that occur heqwntly (and are mainly tabulx summaries) are rcfened to in this table by lener code:
A = Hclwig and Bird (1973)
B = Rimr(1958)
C = Ried(1964)
D = Dechrma (1953)
J = Fuller (1927)
N = New Jersey Zinc Co. ( d i m data)
R = Clarke and Longhurst(1961)
Time (days,
'otherwise Temperature General
C m i v e medium indicated) ("C) g/m*/year @year suitability' Remarks Ref.
Abietic add
AlXtates
Molten Li-Na-K
Acetic acid
0.1 g/L vapor
0.005 ppm vapor
0.05 ppm vapor
0.5 ppm vapor
2.0 ppm vapor
3.5 ppm vapor
5.0 ppm vapor
20.0 ppm vapor
35.0 ppm vapor
50.0 ppm vapor
500.0 ppm vapor
1% in water vapw
0.5% in saturated Na
P acetatevapor
3 pH 3.8 vapor
8
21
21
21
21
21
21
21
21
21
21
6
42
8
197-247
30
30
30
30
30
30
30
30
30
30
30
30
30
RT
2,520
33
18
945
4,300
5,250
7,050
2,015
1.305
1,415
3,125
7.960
200
290
350
5
I 1
132
600
735
990
282
183
198
440
1.115
28
40
d
d
d
b
b
d
d
d
d
d
d
d
d
d
C
C
In conifers
100% RH over water
100% RH over water
100% RH over water
100% RH over water
100% RH over water
100% RH over water
100% RH over water
100% RH over water
100% RH over water
100% RH over water
72% RH
100% RH
D
Marassi (1969)
I
R
R
R
R
R
R
R
R
R
R
R
R
Gilbert and Hadden (1950)
" r;
-r;
"(11
422
423
fi Table 5.7 Continued
P
sane derrnces t h1 - v y (and IIC mainly tabhr runmuries) me d d 10 m this table bu 1 - code:
A = Helarig md Bird (1973)
B = Rittcr(1958)
C = Ried(1964)
D = k h c ma (1953)
E = Bauer and Schikon (1934)
G = Wi&olt (1976)
I = lntrmational Nickel Co. (dM data)
J = Fuller (1927)
N = New J a y Z ~ K CO. (direct dam)
R = ClarkeandLonghmt(1%1)
Time (days,
except as Conosion rates
o(henvie Temperahm
General
Ref. %&year pmlyear suitability' Remarks
C m i v e medium indicated) CC)
d Grnelin (1956)
b
Vapor d C, D
100%
Binders
G (many)
60
Butane
Butyl alcohol
100%
Capryk dwhd
Pure, neutral
Carbon tetrachloride
Dry
(GHCI, or Gq )
Castor oil
Cellulose seetyl
Neutral, moist
Cellulose nitrate
Chlorinated phenols
Pure. dry
Moist
80
20
20
13,870 1,940
510 71
b Can use galvanized drums c, D
b Sent in galvanized drums
c
b
b Binder
Only for shorc contact: other
hydrocarbons better
b Stored in galvanized drums
b Sent in galvanized drums
a Sent in galvanized drums
d
D
Bukowiecki (I%8), B, D
Bukawiecki (1968). B, D
Van Eijnsbergen (1967)
D
D
D
Dirkse (1970)
a
Q D Q
SESE SEES
2
425
II:
d
B B B B
426
4
W'
w d
dm" z z z z
427
5
+
U'
z z D D
0
4: Table 5.7 Continued
Some rrferrnccs ltut occur mqWrmy (and are rminly tabular summaries) lllc m M to in this table by luwr code:
A = Helarig .nd Bird (1973)
B = Rim(1958)
C = Ried(IS64)
D = De c h ( 1 9 5 3 )
E = BauerandSehkorr(1934)
G = Wiedemolt (1976)
I 5 International Nickel Co. (dirret dnIa)
J = Fuller (1927)
N = New Jency Zinc Co. ( d i m data)
R = Clarke and Lonshurn f1%1)
T i (days.
otherwise Temperalure General
C m i v e medium indicated) ("C) glm2/year pmlycar suitability' Remarks Ref.
Methyl cbloride
hy. Pure
Methyl ethyl ketone
+ 0.03% acetic and 0.2% wafer
+ trace acetic and 0.1% wafer
+ trace acetic + heprane
Methylamine
Milk
pH 4.03
pH 6.6
Naphtha
Immersed
Vapor
Nitrobenzine
Nitrotoluene
128
30
142
50
10
10
24
20
97
20
75
155
190
37
22
9,930
9,200
22
I 1
1,124
347
27
5
3
1.380
1,280
3
I .5
157
49
d
b
a
a
d
d
d
a
a
But if water-free can be transported D
in galvanized containers
Some agitation
Moderate agitation
I
I
I
D
Unsuitable because of effect
on the milk anyhow
Gmelin (1956)
Gmelin (1956)
Gmelin (1956)
Gmelin (1956)
I
1
B, C, D
D
d
a
-
o r n o o n o n o n
F:
W
"
"
431
Table 5.7 Continued
Some references that occur hquently (and arc mainly tabular summaries) am refcmd to in chis table by Iemr code:
B = Ritter (1958)
C = Ri d (1964)
D = Dechema (1953)
1 = International Nickel Co. (direct data)
J = Fuller (1927)
N - New Jersey Zinc Co. (dimf data)
R = Clarke and Lonehunt (1961)
Time (days.
othewise Temperature Genenl
Corrosive medium indicated) ("C) g/m*/year pm/ yw suicability' Remarks Ref.
soya oil
Tannk acid
50% solution
Tartaric acid
10% solution
Thiourea
Trichloracetic add
+ nitrobenzene
Trichlorethylew
Immersed
Vapor
Vapor
Vapor
With water
Turpentine
Urethanes
Vinyl acrylic resins
Waxes
41 h
41 h
41 h
b
20 2.8 d
60 36,135 5 . m d
b
d
45
84
10,549 1,480
380 53
496 69
13,760 1,925
Very strong attack
In storage tank
In dryer
In exhaust dua
In still
Attack in water layer
Paint vehicles
Zinc dust paint vehicles
White and Deanin ( 1965)
G
G, Gmelin (1956)
G
F, Gmelin (1956)
I
I
I
I
N
Van Oeteren (1967)
Laberenz (1970)
G
'Ratings: a = linle anaek ( 4 0 pndyear), very useful: b = attack ( 1C~l Oo pmlyear). but may bc usrfuk c = heavier attack (~(KI-3ooO pndycar). not usually suitable; d = rapid attack, unsuitable for use.
with polystyrene, polyethylene, phenol formaldehyde, and Perspex (Bora,
1974).
Lechner-Knoblauch and Heitz (1987) have studied the effect of impuri-
ties such as chlorides and organic acids on the corrosion rate in CH30H and
C2H,0H, both at 50C and at room temperature. At acid contamination levels
of 0.1 mol/L, corrosion rates could reach 1 pm/h or more, but at 0.01 mol/L,
0.03 pm/h was more typical (Table 5.8). It has also been shown that the
presence of water may cause a marked increase in the corrosion of zinc in
organic chemicals. Therefore, the possible effect of moisture contamination
should be considered in using the data from Table 5.6. Constantinescu and
Heitz (1976), in tests on formic, acetic, proprionic, and butyric acids, showed
that water has a significant effect on the corrosion by monocarboxylic acids
with three or more carbon atoms.
The suitability of zinc castings, especially ZA-8, ZA-12, and ZA-27,
for use in engines and equipment has led to many tests-largely by Noranda
(1993)on their suitability with fuels, oils, and so on of different types.
Gasoline
alloys as cast, when totally immersed in any of the six fuel mixtures
(except ethanol-based gasohol) given in Table 5.9, did not corrode signifi-
cantly when water-free. In hydrated fuel mixtures, however, the corrosion
products formed both on zinc alloys and on 384 aluminum alloy might, in
some cases, interfere with the functioning of fuel systems; the possible
benefits of chromating are being investigated (Noranda, 1992). In two-phase
solutions, voluminous white corrosion products formed at the fuel-water
phase. interface. With contaminated water additions (giving a two-phase
liquid), steel, aluminum, and zinc alloys performed in static fully immersed
tests as shown in Table 5.9. The unleaded high aromatic fuel (HAF) gasoline
is ASTM reference fuel C, recommended for studies involving automotive
fuels.
The use of zinc die castings with methanol/petrol mixtures was studied
by Zinkberatung (1979), which concluded that chromated castings were
suitable for use with single-phase solutions (even with 0.03% acetic acid
present); if a second phase is present (especially when fuel additions based on
ethanol are used), however, some problems in use may recur.
Protective coatings are, therefore, usually applied over zinc or ZA
alloys when these materials are used in contact with fuels.
In developing a dual-fuel engi net o use alcohol as an alternative to
gasoline-researchers in Brazil have found that nickel-plated zinc is a reliable
material for the carburetor, resisting the ethanol fuel. Knowledge of this
property opens the way for possible use of the ZA alloys.
l
!
!
I
!
i
!
j
l
!
!
; I
!
i
l
9
,
!
l
I
l
l
0
0
0
0
0
0
0
8
S
8
l-
m
0
m
8
8
m
l-
0
'9
p1
8
N
8
8
W
0
".
0
8
2
S
9
II!
9
13
8
-h!h!h!9
cn
Chemicals, Building Materials, Fuels
435
4
436 Chapter 5
B. Diesel and Fuel Oils
Zinc, including ZA alloys, is not corroded by refined oils, but sulfur or water
in ordinary fuel oils may form compounds with zinc which can clog small
orifices. Chromating of castings may be helpful; zinc anodizing or chrome
plating should provide more than adequate protection.
Tests involving galvanized steel were made by Kruse (1984). Small
reservoirs for heating (fuel) oil, made from mm steel sheet, were con-
structed and tested as follows: (a) hot dip galvanized, 50 pm coating thick-
ness; (b) hot dip galvanized, 100 pm coating thickness; (c) ungalvanized steel
with scale; and (d) ungalvanized steel, grit-blasted. After 2 years of testing in
various fuels, the following conclusions were drawn.
The ungalvanized grit-blasted specimens showed pitting corrosion in
various degrees, with occasionally a corrosion rate exceeding 1.5 mdyear.
The ungalvanized specimens with scale did not show much pitting corrosion;
however, after breakdown of the inhibitive action, rust spread extensively and
there was pitting up to a depth of 1 mm. The galvanized test specimen showed
voluminous zinc corrosion products. However, the underlying steel was not
attacked. The 100 pm thick coatings behaved better than the 50 pm ones.
Further testing was carried out on large tanks, as used in practice. To
the fuel in these tanks, l L of sodium chloride solution was added, and
corrosion was tested in tanks with stagnant fuel and with flowing fuel.
After 4 years of use, all insides of the ungalvanized tanks showed large
formations of rust in the lower third of the interior. All galvanized tanks
remained completely free from rust, and there was no blocking of oil filters or
burner nozzles.
C. Miscellaneous Organic Chemicals
Gear Oils
ZA-12 and ZA-27 do not react with SAE 90 gear oil at 82"C, but oil
breakdown and associated corrosion occurs at in a stirred beaker test.
Lubricants and Hydrocarbons
Heavy oil or greases, particularly the extreme pressure type, produce smooth,
light etching and form a natural protective coating. Lubricants should be
stable and free from acidity. Lubricants of mineral origin are completely
satisfactory. Those containing animal or vegetable oils or fats should be
avoided.
Hydraulic Fluids
No apparent reaction occurs on ZA 27 after 2 weeks at 49C in a static test of
three fluids: Shell Tallus T-15, Shell Decol T-23, and Shell Aircraft Hydrau-
lic Fluid 4.
Glycerin
Pure glycerin has a limited smooth etching action on zinc and is quite
satisfactory as hydraulic fluid in door stops. Similarly, glycerin-alcohol
mixtures are practically inert to zinc, provided the mixture is pure and free of
water. Water or low grade glycerin may cause some pitting.
Engine Coolants
After 32 days at 88C in 33% Prestone I1 (a commercial engine coolant
solution that contains silicate inhibitors), ZA-8 and Alloy 3, either as-cast or
polished, showed no corrosion. Dynamic tests confirmed the suitability of
ZA-8 at 118C. ZA-27 displayed intergranular attack when tested as-cast but
not when its surface was polished before testing. Possibly, the soft, zinc-rich
phase had been smeared over the whole surface during polishing.
Dynamic, as well as static, tests with a ZA-8 pump prototype gave a top
ASTM D-2809 rating even though inadequate inhibitor was present and,
consequently, copper (from the tubing) deposited onto the alloy surface.
Intergranular corrosion was minimal, with pits no more than pm deep
after a 300-hour test (Noranda, 1992).
Domestic Uses
Domestic use of chemicals is also significant, but ZA alloys are not recom-
mended for alcohol-water mixtures represented by beverages. Pure alcohols
are considerably less corrosive than water, but mixtures of water and alcohol
are more corrosive than water alone.
Refrigerants
Freon 22 is inert and stable in contact with zinc alloys. Naphtha-based
refrigerant oils are similarly inert and stable.
V. CONCRETE, CEMENT, AND PLASTERS
Unprotected steel reinforcing bars can corrode in certain environments as
moisture penetrates into concrete through cracks and pores. Because the rust
has a larger volume than the steel from which it is formed, the cover layer
over the reinforcing bars can burst and flake
438 Chapter 5
Partially grouted-in, poorly rust-protected steel, such as bolts and edge
guards, can cause unsightly rust staining on a concrete surface. This type of
damage can be avoided if the reinforcing steel is hot dip galvanized (ILZRO,
1981; Porter, 1985). Galvanizing of concrete reinforcement gives increased
structural life, makes the placement of reinforcement less critical, and allows
the use of steel protrusions with much reduced risk of cracking, rust staining,
and spalling. A number of positive advantages arise; for example, for archi-
tectural uses, facades, etc., the thickness of the cover layer over the reinforc-
ing bars can be reduced, resulting in a lighter weight structure, which can then
be made slimmer and more elegant. Many publications are issued by the
market development associations on the subject, and therefore it is not
covered in this book. Yeomans (1991) said that in Australia, the use of
galvanized steel adds about 6.5% to the typical cost of reinforcement of
concrete; this is less than 2% of the building costs of a typical project and,
with selective use of galvanized reinforcement, the extra cost is only OS-1%
of the building costs. Among concerns of users satisfactorily answered by
research on zinc-coated steel (Swamy, 1990) are the following: the effects of
galvanizing on static strength even after deformation (no adverse effect); bond
strength (similar to galvanized and ungalvanized when the usual modest levels
of chromate are present, i.e., greater than about 20 ppm); fatigue strength in
wet conditions (10% higher for galvanized than ungalvanized according to
work done in Kari, Finland, but 25% lower in the less significant dry
conditions); and effects of possible cracks in the galvanizing after bending
(with 40-70 mm concrete cover, galvanized was less pitted after years
outdoors or in accelerated tests). In Holland, galvanized rebars are also used
in concrete sewer water pipes, church (Apeldoorn), lock walls in the Delta
port area (Rotterdam), Andoc offshore platform, facade elements in the walls
of bank offices, cooling water ducts in Delfsyl, swimming pools, support
columns of motorway viaducts, and bridges and balcony floor elements.
Table 5.1 1 lists some case histories of galvanized reinforcement that has been
in place for more than 15 years, Recent Australian examples include the
Intercontinental Hotel in Sydney, the new Parliament House in Canberra, and
the ASER Hotel and Convention Complex in Adelaide, which had followed
such well-publicized long-term examples as the Sydney Opera House. In
Delhi, India, the Bahai Temple-which looks like lotus petals-needed
galvanized reinforcement for the thin concrete sections made necessary by the
design.
Hot dip galvanized reinforcing bars or mesh are valuable in precast
facade units to ensure that rust will not discolor the facade.
Table 5.10 Corrosion Data for Zinc in Contact with Some Building Materials
Temper- Corrosion rates
Corrosion Time ature
medium (days) ("C) g/m2/year PdYear
Data from Ritter
Quartz sand
Fibrolite
0.1
Asbestos cement
White granulated
Red brick
Mineral wood
Asbestos
Portland cement/
Gypsum plaster
Silica brick
Wood from RT
Wood from RT
Wood from RT
Wood from RT
Wood from RT
Wood from RT
slag
sand mortar
Data from Bauer and Schikorr
linden tree
linden tree
pine tree
pine tree
beech tree
beech tree
"Direct contact.
contact under water.
In the case of deformed bar, the stress when 0.1 mm of slip occurs
between the steel and the concrete is approximately:
Untreated steel 150 MPa
Hot dip galvanized steel 160 MPa
Hot dip galvanized and chromated steel 190 MPa
440 Chapter 5
When the concrete is poured, its pH is around 13. In solutions with such a
high pH, zinc is attacked and hydrogen is evolved, which could give rise to
poor adhesion. However, the attack ceases as soon as the concrete has
hardened.
Corrosion tests show that zinc-coated steel has a higher tolerance to
moderate chloride levels in concrete than black steel (Yeomans, 1991).
Research in Germany has shown the benefit of galvanized steel in
mortars even with substantial amounts of chloride present (Hildebrand and
Schwenk, 1986). Both galvanized and ungalvanized (pickled) steel sheet
specimens were embedded in Portland cement mortar with various cement/
water ratios and curing conditions. Then these specimens were dipped in a 0.5
mol sodium chloride solution for 1-5 years. The fully immersed steel samples
have very negative free corrosion potentials and, for that reason, cannot act as
a cathode in corrosion cells, and the steel sheet within the cement does not
corrode. Partially immersed specimens, however, showed very noble poten-
tials. The galvanized steel sheet potentials are also shifted to the same positive
values during exposure time. Localized corrosion generally occurs at the air-
water line. With the ungalvanized specimens, cracking of the cement occurs
as a result of the formation of voluminous rust. Corrosion is retarded signifi-
cantly by galvanizing. Based on these extensive tests, the authors give
detailed information on the use of galvanized steel in mortars, as rebars, and
as coatings for salt-carrying pipelines.
Simm (1984) has studied the conjoint action of carbonation and chloride
ions on the corrosion of zinc in mortar. Highly concentrated, high chloride
mortars at 100% RH and 25C can corrode zinc at pdyear , but high
carbonation or high chloride on its own causes a loss of zinc of only 5
pdyear. With neither chloride nor carbonation, no corrosion occurred. Sergi
et al. (1985) have electrochemically looked at zinc in solutions pH 9.0-14.0
in relation to the use of galvanized steel in concrete.
Most types of cement contain small quantities of chromates. These
chromates passivate the zinc surface, which is then not attacked by fresh
concrete. If the cement contains less chromate than will give at least 20 ppm
in the final concrete mix, the hot dip galvanized bars can be dipped in a
chromate solution or chromates can be added on the basis of twice the
minimum required. Avoid handling the chromates directly, for this can cause
dermatitis: hexavalent chromate is a toxic substance.
Numberger (1986) in a summary report notes that galvanized rebars and
other galvanized steel parts in contact with concrete will initially react with
strongly alkaline mortars. The reaction between zinc and calcium hydroxide
results in the formation of calcium hydroxyzincates: Ca[Zn(OH),]-2H20. This
type of coating protects the zinc layer. Hydrogen development during this
reaction is counteracted by the presence of very small quantities of hexavalent
Cr ions (approximately 0. 020. 012% CrOi-) present on the galvanized
rebar or in the mortar, or both. The corrosion of the zinc layer remains
between 2 and 10 pm during the first weeks or months; after which it stops.
Only when mortar pH exceeds 13.3 is corrosion accelerated and, conse-
quently, no efficient protection occurs. Blast furnace-and molten clay-
based-cements cause even less corrosion than Portland cement because of
lower concentrations of calcium chloride during hydration. Under conditions
of high humidity and wetness and thin concrete covering layers, the corrosion
of zinc in carbonated concrete will be higher than in alkaline concrete. In
practice, an 80-100 pm thick zinc coating will delay corrosion for very long
periods. When high concentrations of chlorides occur, such as in silos for
road salts, galvanized steel has a higher corrosion resistance than un-
galvanized steel, especially at chloride concentrations above 1%. However,
the zinc coating offers protection during shorter periods only. Galvannealed
steel has larger corrosion losses than galvanized steel, where zinc and ZdFe
alloy layers are present. Because zinc corrosion products are much less
voluminous (a factor 2-3 times less) than rust, the danger of spalling concrete
layers is practically nil with galvanized rebars. This property is especially
valuable when concrete cover is (or must be, because of weight or aesthetic
considerations) rather small.
Care must be taken not to allow contact between galvanized steel parts
outside a concrete construction (e.g., underneath moist insulation materials)
with ungalvanized (black) rebars in the concrete. The chance of corrosion of
ungalvanized and galvanized steel parts is summarized as follows:
Contact of Metals
In concrete Outside concrete Chance of corrosion
Ungalvanized
Ungalvanized* Possible
Galvanized Galvanized None
Galvanized Ungalvanized None
Ungalvanized
Galvanized* Possible
Galvanized*/ungalvanized
- Possible
-
Galvanized*/ungalvanized Possible
*Areas especially sensitive to corrosion
Even at elevated temperatures galvanized rebars do not corrode in
concrete. This behavior is in contrast to the corrosion caused by change
potential in hot water between 60 and 90C.
442 Chapter 5
Contact between zinc and gypsum-containing building materials, espe-
cially in moist conditions, results in corrosion of zinc. This often occurs in
practice during the building period [e.g., when new galvanized roofing comes
in contact with gypsum dihydrate crystals (CaS0,.2H20)]. Magnesia cement
causes severe corrosion of galvanized parts in contact therewith.
Galvanized parts in sand do not corrode very much. In unaerated and/or
contaminated sand, corrosion may occur: always, however, very much less
than on ungalvanized steel. Excessive and retained moisture present between
galvanized steel parts and other building materials, such as isolating strips,
bituminous coatings, and asphalt materials, must be avoided. If this is impos-
sible, other building materials should be used.
Franque and Huppatz (1986) studied the corrosion behavior of nonfer-
rous metal structural elements in contact with building materials. They con-
firmed that metals like zinc and aluminum (unlike lead, copper, and tin)
rapidly form a surface layer of oxide or insoluble salts when exposed to moist
air in buildings. Zinc is much more resistant than lead. It also is only
superficially attacked in mortars containing no sodium chloride, because
insoluble zincates are formed. In contact with magnesia cement and stone
wood, a building material made from wood chips and a magnesia cement
binder, zinc is attacked by the very hygroscopic magnesium chloride and
chloride ions. zinc is attacked by gypsum under moist conditions. When
zinc or galvanized steel is present, care should be taken not to use acidic
materials (e.g., from acidifying stone walls).
Laboratory tests on plasters and cements are of limited value but can
indicate trends. Lodder and Baumgarten (1951) looked at corrosion in damp
gypsum plaster at room temperature. With only 0.046% chloride, corrosion
was about 8 pdyear in a 203-day test, even if the partly embedded rolled
sheet was kept dry; the steel was perforated in 52 days if kept damp.
Chromating (Cronak process) virtually stopped attack (only 1.5 pdyear).
When 1.5% sodium chloride was added to the gypsum, significant
pitting occurred at the interface on partially embedded samples that were kept
moist; chromating (Cronak process) virtually prevented attack in tests up to
one year, but phosphating allowed corrosion to continue at the interface.
Heavy attack (around 500 pdyear) also occurred with surface contact
alone when the area was kept damp, but the Cronak treatment again prevented
corrosion (about 5 pm in a year).
Bauer and Schikorr (1934). in 7 1-day tests at room temperatures in
plaster of paris, showed that when the water content of the plaster was 28%,
corrosion on partly embedded samples was about 10 pdyear; if the water
content was only 15%, localized pitting occurred and corrosion rates of
around 100 pdyear were found in specimens kept in a dessicator over water.
Just by keeping the samples damp, pitting attack was avoided, although the
corrosion rate increased up to 200 pdyear. The same investigators also
tested cement in similar sets of conditions and found attack of 1.1-1.5 pm in
days (annualized rate if linear: pdyear).
In general with alkaline building materials, attack on zinc occurs when
the pH is above about As is seen from the tests quoted, more water will
often reduce the alkalinity to levels at which zinc is resistant to attack. In
border line cases, the protective film is broken locally to give pitting; in such
cases, a surface film containing chromate can be very beneficial. At lower pH
values, caution is needed with materials such as fly-ash cement, where it is
easy for substantial amounts of aggressive materials to be leached from the
cement; attack is then similar to that in polluted fresh water.
The ZA alloys, especially ZA may corrode more than zinc when in
contact with fresh concrete, plaster, and other alkaline materials, but then
they are better than aluminum.
VI. BITUMEN
The use of bitumen in contact with zinc roofing sheet led to a comprehensive
study of the corrosion that may occur. The work was written in German by
Witt and Haller as Part 4 of Corrosion Behavior of Zinc. A brief
summary follows (the original book contains references and a list of about
40 other relevant articles).
Bituminous materials will form acidic products when affected by light
and/or heat. Also, degradation products of additives and hydrolysis of cata-
lysts used in flowing bituminous materials will cause the formation of com-
pounds that may be detrimental to zinc. The use of herbicides and fungicides
to prevent growths on roofs covered by bituminous materials results in attack
on zinc; pentachlorphenol is especially implicated in this regard. When zinc is
exposed to moist air (RH > and acidic products from applied bit-
uminous coatings, corrosion is accelerated. To avoid this, it is necessary to (a)
use stable materials that do not release acidic products upon weathering, (b)
cover bituminous roofing with a protective paint, and (c) avoid contact of zinc
with nearby bituminous layers. To select suitable bituminous materials, UV-
radiation tests in moist air are recommended. Bituminous materials degrade
200 times quicker in sunlight than in dark areas.
Riickert et al. note that rainwater dripping from galvanized
gutters had pH values of whereas water from bituminous roof
claddings had pH values of 5.5-6.5. These authors discuss the influences of
intermittent attack and the humidity of the air.
444 Chapter 5
Bitumen-based coatings for cold application in uses involving contact
with water are standardized in BS (potable water) and BS
(other waters).
Russian work on zinc as a filler in bituminous coatings (Tomashov and
Leonov, reflects interest in a type protection not much used else-
where. These investigators tested films with various concentrations of zinc
filler by volume, which corresponds to a presence of 0-80% by mass
in the coating). The coatings were applied by slow withdrawal of the speci-
mens from the corresponding primer solutions. The film thickness (0.5-500
km) was regulated by the viscosity and rate withdrawal of the specimens.
The zinc particles apparently dissolve to some extent but can be seen as
discrete areas in a section. At the higher concentration, the coating has some
degree of electrical continuity. Figure 5.2 gives electrical capacity and re-
sistance measurements for one zinc loading. After testing, a hydrated zinc
oxide forms on the surface. The authors conclude that in thin coatings 10-
20 pm), zinc essentially fulfills the role of protector, but the service life
such a coating is limited by the time for dissolution of the zinc. In thicker
coatings, the zinc filler protects the basis metal initially by its protective
action and, thereafter (for a longer time), by consolidation of the surface layer
of the coating by corrosion-resistant zinc corrosion products.
VII. OTHER BUILDING MATERIALS
Zinc and zinc-aluminum alloys and coatings are suitable for contact with
most building materials. Corrosion problems have been surveyed by Jones
Fig. 5.2 Change of electrical capacity and resistance with time for an iron electrode
insulated by a bituminous primer containing zinc filler for coating thicknesses:
1 = 2 pm; 2 = 5 pm; = 10 pm; 4 = pm (Tomashov and Leonov, 1965).
Table 5.1 1 Galvanized Rebar: Selection of More Than 10 Years
of Satisfactory Performance
Building structure Location Reference
Buildings: in situ construction, cladding panels window frames
National Theatre London
Opera House Sydney
Dome of New Hall Cambridge
Sports Hall Birmingham University
Civic/council buildings Newham, Woking, Exeter
Civic Centre Johannesburg
Examples since 1930 Bermuda
Barclays Bank DC0 Tower London
Civic buildings, police station, library Sittingbourne, Kent
Parliament Buildings Wellington, New Zealand
Bank of Hawaii Waikiki, Honolulu
Congress Theatre Eastbourne
HUD (Housing and Urban
Church roof Tromso, Norway
Flowers Brewery Luton
University buildings York
Collegiate buildings, University
Former GLC offices (south of
Development) offices Washington, D.C.
College London
Westminster Bridge) London
Civil engineering structures
Bridges in Michigan, Vermont, Pennsylvania,
Connecticut, and Massachusetts, including
Boca Chica bridge Florida
Ames bridge Iowa
Hershey bridge Pennsylvania
Tioga bridge Pennsylvania
Athens bridge Pennsylvania
Croaopolis bridge Pennsylvania
Vimont Overpass Canada
Bridges in Ontario and Quebec, including
Power station cooling ducts Netherlands
Andoc oil storage tanks Bed of North Sea
Docks Bermuda
Onomachi Pier
Japan
300-foot chimney
Japan
Toutry Viaduct
CBte dOr, France
Pont dOuche Viaduct CGte dOr, France
St. Nazaire bridge
Loire-Atlantique
Coke ovens-quenching tower Dunkirk
Concrete yachts United Kingdom
GA
GAA
GA
GA
GA
SAHDGA
GA
GA
GA
GAA
FCP/JvanE
GA
AGA
JvanE
GA
GA
GA
GA
ILZRO
ILZRO
ILZRO
ILZRO
ILZRO
ILZRO
SDV
SDV
GA
JGA
JGA
Galvazinc
Galvazinc
Galvazinc
Galvazinc
GA
%A = Galvanizers Association; GAA = Galvanizers Association of Australia; SAHDGA = South
African Hot Dip Galvanizers Association; FCP = F. C. Porter; JvanE = J. F. H. Van Eijnsbergen; AGA =
American Galvanizers Association; ILZRO = International Lead Zinc Research Organization; SDV =
Stitching Dwlmatig Verzinken; JGA = Japanese Galvanizers Association.
In addition, information in the text and also Fe + Zn 12, 16, and 21 published by the European General
Galvanizers Association and the various national galvanizing association publications. F. C. Porter and
1. F. H. van Eijnsbergen also have additional information.
445
Chapter 5
and Brinn Care is needed when there is long-term contact with water,
which has leached chlorides, sulfates, and other salts from brickwork and
other materials. Table gives results of some tests, but it must be
remembered that most building materials are very variable in composition and
texture. Protection of the zinc is usually desirable when the metal is in contact
with pervious materials that are wet for long periods; if such materials are
normally dry, however, there is no problem. Thus, the performance of wall
ties has been studied by Moore for locations and up to 50 years of
service. In general, the galvanizing in the space between the inner and outer
walls sustained no apparent attack. The damper outer wall caused most attack,
but the amount of corrosion varied considerably because of different locations
and installation conditions and ranged from less than pm/year to
year.
Zinc can be used in contact with most woods, but some corrosion will
occur with Western red cedar and other acid woods. Generally speaking,
galvanized fasteners (e.g.) are not attacked by wood or by insulation foams
(e.g., polyurethane and polystyrene foams). Heavy corrosion may also arise if
the wood rots and forms a wet poultice against the zinc alloy. The survey by
Jones and Brinn pp. gives details of woods and treatments but
concludes that there are wide differences of view on performance or likely
performance and calls for more in situ investigations.
Wood preservatives containing copper salts, and any that produce acid
solutions, will cause corrosion, particularly on fasteners or connecting pieces
if these have only a thin zinc coating. The Greater London Council in the
United Kingdom have particularly cautioned against copper-chrome
arsenate wood preservatives, and work by Simm and Button showed
local and general attack in 30-month tests; Duncan in his work
confirmed the corrosivity and indicated that in the worst places, powder-
coated galvanized steel would be a cheaper solution than stainless steel or
silicon bronze bolts. If, however, the moisture content of the wood can be
kept Iow-Cross and Bailey suggest below 197-severe corrosion
does not occur.
References
Abd El Aal, E. E., Abd El Aal, A., and Abd El Haleem, S. M. (1992). Initiation of
pitting corrosion on zinc under natural corrosion conditions. (private communica-
tion).
Abd El Haleem, S. M., and Abdel Fattah, A. A. (1986). The role of some organic
anions in promoting inhibiting the corrosion of zinc. S@. Coat. Technol., l ),
4149.
Abdel, Aal M. S. , and El-Saied, A. (1983). Phenothiazines as corrosion inhibitors for
zinc in NH,Cl solution. Br. Corros. J . , 18(2), 102-106.
Abdul Azim, A. A., Shalaby, L. A., and Abbas, H. (1974). Mechanism of the
corrosion inhibition of Zn anode in NaOH by gelatine and some inorganic anions.
Corros. Sci., 14(1), 21-24.
Adams, G. R., Battison, L., and Zervoudis, J. (1991). A comparison of touch-up
techniques for bare spots on hot dip galvanized steel. 16th Int. Galv. Conf., EGGA,
London, Paper GE3/1-19.
Adaniya, T., et al. (1958). Zinc-iron alloy electroplating on strip steel. Plat. Surf.
Finish., August, pp. 52-56.
Adaniya, T. (1978). The development of corrosion-resistant composite zinc coated
steel sheet. Sheet Metal Id., 55(12), 73-78, 82.
Adaniya, T. , et al. (1984). Iron-zinc alloy electroplating on strip. Proc. 4th Cont.
Strip Plating Conf., American Electroplaters Society, Winter Park, FL, 15 pp.
AFNOR. (1990). Hot Dip Galvanized or Not Coated Steels in Contact with Natural
Backfill Materials (Soils). French Standard NF A 05-252. AFNOR, Paris (English
ed.) 14 pp.
AGA. (1990). Galvanizing for Corrosion Protection-A Specijiers Guide. AGA,
Alexandria, VA, 60 pp.
AGA. (1991). Galvanizing for Corrosion Protection. AGA, Alexandria, VA, 44 pp.
AGA. See also American Galvanizers Association.
447
448 References
AGK. (1987). (Testing of duplex systems for corrosion protection of steel structures
by hot dip galvanizing and coating data sheet Bl). Werkst. Korros., 38, 135-136;
English translation by ZDA, 6 pp.
Ahn, D.-S., Shin, J.-C., Chae, H."., Cho, S.-J., and Han, K.-H. (1992). Develop-
ment of Fe-MnlZn-Fe double layer plated steel sheet for auto bodies. Galvatech 92.
Verlag Stahleisen, Diisseldorf, Germany, pp. 127-131.
Ailor, W. H. (1978). ASTM Atmospheric Corrosion Testing: 1906 to 1976. ASTMSTP
646. (See ASTM 1956, 1978, 1990), pp. 129-151.
Akerdida, M. T. (1987). Effect of solution composition on the corrosion properties of
galvanized steel and annealed galvanized steel. Master's thesis, University of
Manchester Institute of Science and Technology.
Akhter, R., Watkins, K. G. , and Steen, W. M. (1990). Electrochemical characteriza-
tion laser welded zinc coated steel. Mater. Lett., 9(12), 550-556.
Akimov, V. I., Berezin, A. V., Marchuk, S. I. , and Lavrent'eva, F. A. (1988).
Corrosion resistance of zinc coating on steel after annealing. Izv. Akud. Nauk SSR,
Met. (3), 160-163 (in Russian).
Allegra, L., Dutton, R. J., and Humayan, A. (1985). Galvalume sheet-New techni-
cal developments. 1st In?. Conf. Zinc-Coated Steel Sheet, ZDA, London, pp. A/1-
A/6.
Allen, C. (1991a). Corrosion of galvanized steel in synthetic mine waters. Br. Corros.
J . , 26(2), 93-101.
Allen, C. (1991b). The corrosion resistance a silicon-killed galvanized steel in two
synthetic mine waters. 16th. Int. Gab. Conf., EGGA, London, Paper GH3/1-12.
Aluminum Co. of America. (1968). Solder flux and methods soldering metallic
elements. French Patent 1,579,340 (in French).
AM&S Europe Ltd. (1978). Cover story. Cover Story (UK), Autumn (l), 1-8.
AM&S Europe Ltd. (1980). Specifier's Guide to BS 5493: 1977. Folder republished
by ZDA, London, in 1981.
Ambler, H. R. (1960). Atmospheric salinity and metal corrosion at various places in
Great Britain. J . Appl. Chem., IO, 213-225.
Ambler, H. R., and Bain, A. A. J. (1955). Corrosion of metals in the tropics. J . Appl.
Chem., 5, 437467; Chem. Ind. (1957), 106-107.
American Galvanizers Association. (1992 list). More than 50 publications, including
performance reports on galvanized steel used in relation to highways (5 reports),
waste treatment ( l ) , utilities (3), recreation (2), pulp and paper (2). petrochemical
(3, and food ( I ) . AGA, Alexandria, VA.
American Galvanizers AssociatiorVAmerican Hot Dip Galvanizers Association. See
Also AGA.
American Hot Dip Galvanizers Association. (1982). Painting Galvanized Steel Struc-
tures. AHDGA, Washington, DC, 11 pp.
American Society for Testing and Materials (1944-1977). Reports of Subcommittee
XIV, Committee A-5: (a) Proc. ASTM, 44, 92 (1944); (b) Proc. ASTM, 52, 113
(1952); (c) Proc. ASTM, 54, 1 IO (1954); (d) Proc. ASTM, 63, 143 (1963); (e) Proc.
ASTM, 65, 126 (1965); (9 Proc. ASTM, 67, 119 (1967); (g) Proc. ASTM, 77, 92
(1977).
References 449
American Society Testing and Materials. (1944-1962). Reports of Subcommittee
XVI, Committee A-5: (a) Proc. ASTM, 44, 95-107 (1944); (b) Proc. ASTM, 52,
118-122 (1952); (c) Proc. ASTM, 62, 176 (1962).
American Society for Testing and Materials. (1950-1977). Reports of Subcommittee
XVI, Committee A-5: (a) Proc. ASTM, 59, 133-155 (1950); (b) Proc. ASTM, 60,
116-123 (1960); (c) Proc. ASTM, 65, 129-136 (1965); (d) Proc. ASTM, 67. 141
(1967); (e) Proc. ASTM, 73, 85 (1973); (9 Proc. ASTM, 77, l04 (1977).
American Society for Testing and Materials. (1956). Symposium on Atmospheric
Corrosion Non-Ferrous Metals, ASTM STP 175. ASTM, Philadelphia, 158 pp.
American Society Testing and Materials. (1957). Report of Subcommittee XV on
Atmospheric Exposure Tests Wire and Wire Products. Proc. ASTM, 34 pp.
American Society Testing and Materials. (1961). Report of Subcommittee V,
Committee B-6, on Atmospheric Exposure Tests Zinc Alloy Die Castings. Proc.
ASTM, 61, 273-28 1.
American Society for Testing Materials. (196211967). Reports of Subcommittee VI,
Committee B-3, on Atmospheric Corrosion 1957 Test Program: (a) Proc. ASTM,
62, 216 (1962); (b) Proc. ASTM, 67, 118 (1967); plus unpublished ASTM data.
American Society for Testing and Materials. (1986). D 2092-86. Practices for prepara-
tion zinc-coated galvanized steel surfaces for paint. ASTM Annual Book of
Standards, Vol. 06.01. ASTM, Philadelphia.
American Society for Testing and Materials. See also ASTM.
American Welding Society. See AWS.
American Zinc Institute. (1928). Zinc and Its Corrosion Resistance. AZI, New York.
American Zinc Institute. (1951). Zinc as a Galvanic Anode, 39 pp. with supplement
(1953); Cathodic Protection with Zinc Anodes. Zinc Institute, New York, 49 pp.
American Zinc Institute/Bureau of Ships. (1955). Navy-Industry Zinc Symposium
Cathodic Protection: Zinc as a Galvanic Anode. U.S. Department of the Navy,
Washington, DC, 139 pp.
American Zinc Institute. (1957). Zinc as a Galvanic Anode . . . Underground. MI ,
New York, 39 pp.
American Zinc Institute. (1960). Zinc as a Galvanic Anode . . . Underwater. AZI,
New York, 16 pp.
Amirudin, A., et al. (1992). Characterization of organic coating breakdown on
galvanized steel during atmospheric exposure by electrochemical impedance spec-
troscopy. Galvatech 92, Verlag Stahleisen, Dusseldorf, Germany, pp. 549-555.
Anderson, E. A., and Fuller, M. L. (1939). Corrosion of zinc. Met. Alloys, 10,282-
287.
Anderson, E. A., and Reinhard, C. E. (1939). Chemical removal of corrosion prod-
ucts in the determination the corrosion rate of zinc. Proc. ASTM, 39, 691-697.
Anderson, E. A. (1947). The corrosion of zinc in the outdoor atmosphere. Symposium
Atmospheric Corrosion of Non-Ferrous Metals, ASTM STP 67. ASTM, Phila-
delphia, pp. 2-15.
Anderson, E. A. and Reinhard, C. E. (1948). In Corrosion Handbook, H. H. Uhlig,
Ed., Wiley, New York, pp. 331-347 (zinc section).
Anderson, E. A. (1955). The atmospheric corrosion of rolled zinc. Symposium on
450 References
Atmospheric Corrosion of Non-Ferrous Metals, ASTM STP 175. ASTM, Phila-
delphia, pp.
Anderson, E. A. Zinc in marine environments. Corrosion, I S ,
Andrews, T. Edge Protection by Zinc. BSC Product Development Techni-
Anon. Sluice gates in the Panama Canal. Welding J . ,
Anon. New uses for zinc alloy strip. Mater. Design Eng. , 60,
Anon. 600 tons of steel galvanized for Union Oil refinery. Am. Hot Dip
Galvanizers Assoc. Newsl., May,
Anon. Zinc helps protect worlds tallest tower from corrosion. ABC-Am.
Roofer Building Improv. Contract. (USA), September, pp. 5-6.
Anon. Hot dip galvanized rescue equipment (Feuerveninktes Rettungs-
material). Veninken, S( l ), (in German).
Anon. Hot dip galvanizing in the mining industry (Feuerverzinken im Berg-
bau). Feuerverzinken. Beratung Feuerveninken, Diisseldorf, Germany. Booklet,
pp (in German).
Anon. C-Sentry Cathodic Protection Anodes. ISC Alloys, Bristol, pp.
Anon. Galvanizing steelwork for power station. Corros. Prev. Control
( U. K. ) , 4.
Anon. Metal sprayed coating systems for corrosion protection. Corros. Prev.
Control ( U. K. ) , April,
Anon. Galvanizing stops the rot and at low cost on Rhyl monorail system.
Design Prod. Appl. ( U. K. ) ,
Anon. Corrosion of metal fasteners used in conjunction with CCA (copper/
chromelarsenic) preservative treated timbers. Development Materials Bulletin,
Greater London Council, December
Anon. Intergalva news-sheets (applications and case histories.),
EGGA, London pp. each.
Anon. Metallized coatings for corrosion control of naval ship structures and
components. National Materials Advisory Board, National Academy Press, Wash-
ington DC Report NMAB-409 pp.
Anon. Acid Deposition in the U. K. Second Report of the U. K.
Review Group on Acid Rain. See Burlin for later report.
Anon. Continuous Hot-Dip 55% Aluminium-Zinc Coated Steel Sheet. Techni-
cal brochure, BIEC International, Siegen, Germany, pp.
Anon. A new look at galvanized bridges. Civil Eng, July. Reprint by AGA,
Alexanderia, VA, 4 pp.
Anon. Prepainted Hot Dip 55% Aluminium-Zinc Coated Steel Sheet. Techni-
cal brochure, BIEC International, Siegen, Germany, 45 pp.
Apostolos, J. A., Parks, D. A., and Carello, R. A. Cathodic protection of
reinforced concrete using metallized zinc. Mater. Performance, December, pp.
Archer, W. L., and Harter, M. K. Reactivity of carbon tetrachloride with a
series of metals. Corrosion, 34(5),
cal Digest, pp.
References 45
Arsenault, B., Champagne, B., Lambert, P., and Daillaire, S. (1989). Zinc-nickel
coatings for improved adherence and corrosion resistance. Thermal Spray Technolo-
gy: New Ideas and Processes. ASM International, Metals Park, OH, pp. 409416.
Ashworth, V. , and Fairhurst, D. (1975). The effect of temperature on the behaviour of
a zinc-mild steel couple in a solution containing chloride ions. Corros. Sci., 15( 11-
12), 669-686.
Ashworth, V., Googan, C. G. , and Scantlebury, J. D. (1979). Intergranular dissolu-
tion of zinc alloy sacrifical anodes in seawater at elevated temperatures. Br. Corros.
J. , 14(1), 46-50.
Association GCnerale des Hygienistes et Techniciens Municipaux. (1991). Guide
pratique de la protection cathodique (Practical guide to cathodic protection).
AGHTM, Paris Cedex 17, 142 pp. (in French).
ASTM. (1956, 1959, 1968, 1975, 1976, 1978, 1987, and 1990). Symposia on
Corrosion Behavior. ASTM STPs 175,290, 435,585, 629, 646, 965, and IOOO,
ASTM, Philadelphia.
ASTM. (1961). Report of Subcommittee V., Committee B6, on Atmospheric Expo-
sure Tests of Zinc Alloy Die Castings. Proc. ASTM, 61, 273-281.
ASTM. (1968). Corrosiveness of various atmospheric test sites as measured by
specimens of steel and zinc. Met. Corrosion in the Atmosphere. American Society
for Testing and Materials, Philadelphia, pp. 36&391.
ASTM. (1981). ASTM Standards for Materials Hot Dip Galvanized After Fabrication.
American Society for Testing and Materials, Philadelphia, 261 pp.
ASTM. (1988). Specification B-418-88, Cast and wrought galvanic zinc anodes.
American Society for Testing and Materials, Philadelphia, 3 pp.
ASTM. See also American Society Testing and Materials.
AWS Committee on Metallizing. (1974). Corrosion Tests of Flame-Sprayed Coated
Steel-19-Year Report. American Welding Society, Miami, FL, 31 pp.
Ayoub, C., Davin, A., Goodwin, F. E., and Coutsouradis, D. (1988). Atmospheric
corrosion behaviour of Galfan coated steels. 2ndInt. Conf. Zinc-CoatedSteel Sheet,
ZDA, London, pp. SA2/1-31.
Babenko, A. M., Bezgudova, Zh. I., and Shevchenko, I. B. (1988). Inhibiting effect
of several halide alkylquinolines. Prot. Met., 24(2), 261-262 (translation of
Zushch. Metall.).
Bablik, H. (1941). Zur Korrosion von Zink-Uberzugen (On the corrosion of zinc
coatings), Korros. Metalfschutz, 17, 250-252 (in German).
Bablik, H. (1950). Galvanizing (Hot Dip), 3rd ed., E. & F. N. Spon, London, 502 pp.
Bablik, H. (1954). Sheet Met. Ind., 31(330), 845-846.
Bablik, H., and Belohlavy, H. (1957). Modem detergents and hot dip galvanizing.
Baboian, R. (1989). Corrosion behavior of painted galvanized auto-body steel in
Werkst. Korros., 8, 742-746 (in German).
contact with exterior automotive trim. Gafvatech 89, iron and Steel Instituie, Japan,
pp. 297-305.
Bacon, L. R., and Nutting, E. G. , Jr. (1952). Polyphosphate detergents in mechanical
dishwashing. Ind. Eng. Chem., 44(1), 146-150.
452 References
Bagnulo, L. Adhesive sacrificial anode for cathodic protection. British Patent
April (May 5 pp.
Bagnulo, L. H., Cold Iaminar galvanizing: A new anti-corrosion concept,
Pipe Line Ind., March, pp.
Bailey, J. C. U.K. experience in protecting iarge structures by metal spraying.
8th Int. Thermal Spraying Conf., American Welding Society, Miami, FL, pp.
Bailey, R. W., and Ridge, H. G. Corrosion of metals in buildings. The
performance of zinc and zinc coatings. Chem. Ind.,
Bakalyuk, Ya. Kh., Proskurkin, E. V., Polyarush, V. A., et al. Coatings for
corrosion protection of pipes in heat exchanger equipment. Koks Khim.,
(in Russian).
Baker, R. G. , and Holden, C. A. Zinc-nickel alloy electrodeposits: Rack
plating. Plat. Surf. Finish., March, pp.
Ballard, W. E. Metal spraying. Met. Rev.,
Barnes, H. T., and Shearer, G. W. Production of hydrogen peroxide from
aluminum and zinc. J . Phys. Chem., 12,
Bamett, W. G. S. , and Brett, M. A. Proc. 4th South African Institution of
Corrosion Conf., Johannesburg.
Bamhurst, R. J. , and Belisle, S. Corrosion Properties of Zamak and Za
alloys, Noranda, Toronto, pp.
Barreau, C., et al. Mechanisms for atmospheric corrosion of prepainted steel
sheets used in the building industry. Galvatech 92, Verlag Stahleisen, Diisseldorf,
Germany, pp.
Barton, K., and Smreck, K. Merhody zkouseni korosni odolnosti materialu
(Corrosion test methods). Statni Nakladelstvi Technike Literatury, Prague, pp.
(in Czech).
Barton, K. Der Einfluss von Staub auf die atmosphi4rische Korrosion von
Metallen (The influence of dust on the atmospheric corrosion of metals). Werkst.
Korros., 9, (in German).
Barton, K., and Beranek, E. Reaction mechanism of the atmospheric corro-
sion of metals in moist air contaminated with sulphur dioxide. Werkst. Korros., IO,
(in German).
Barton, K., and Bartonova, Mechanism of kinetics of corrosion in a moist
atmosphere in the presence of hydrogen chloride vapours. Tr. Mezhdunar. Kongr.
Korroz. Met., 4, (in Russian); Chem. Abstr., Dec.
Barton, K. Atmospheric corrosion of galvanized steel. 9th Int. Galv. Conf.,
Industrial Newspapers, U.K., pp.
Barton, K., and Stanners, J. F. Atmospheric corrosion sites in Europe
available for tests on steel protection. European Federation of Corrosion WG2, Br.
Corros. J . ,
Barton, K., Knotkova, D., Strekalov, P., Kemhadze, V., et al. Atmospheric
corrosion of metallic systems. 11. Analysis of the corrosion aggressiveness of the
media at the atmosphere testing stations of member nations of Comecon, according
453
to the results of five-year tests on steel, zinc, copper and aluminum. Prot. Met.,
July-August, I5(4), 323-329 (translation of Zashch. Metall.).
Bauer, O. , and Wetsel, E. (1916). Versuche iiber das Verhalten von technischem Zink
und Blei gegeniiber Wasser, wasserigen Salzlosungen, Gips, Zement und ver-
schiedenen Mortelgemischen (Tests on the behavior of commercial zinc and lead
with water, aqueous salt solutions, plaster, cement, and different mortar mixes).
Mitt. Mat.-Priif.-Anst., 34, 333-373 (in German).
Bauer, O. , and Schikorr, G. (1934). The corrosion of electrolytic and refined zinc.
Metallkd., 26, 73-80 (in German).
Baumann, K. (1981). Jahrbuch der Oberj7achentechnik (Yearbook of Sur&x Coating
Technology) (in German), Metall-Verlag, Berlin.
Baumann, K. (1988). Corrosion Protection for Metals. Deutsche Verlag fur Grund-
stoffindustrie, Leipzig (in German).
Baumbach, S. (1956). Betrachtungen iiber Temperatur-, Feuchte- und Beliiftungs-
messungen im Laderaum. Seewetteramt, Hamburg (Consideration of temperature,
humidity and ventilation factors in loading rooms). Laderaummeteorologie (9), 22-
30 (in German).
Baiimler, J., and Muller, T. (1959). Stadtluftuntersuchungen in Basel 1957-1959
(Town air investigations in Basel). Praventivmedecin, 4, 257-276 (in German).
Bawden, R. J., and Ferguson, J. M. (1989). Trends in material degradation rates in the
U.K. Ind. Corros., 7(8), 9-15.
Bayliss, D. A., Hollis, R., and MacLennan, A. D. (1965). Hot Dip Galvanizing of
Transmission Tower Structures. CEGB, SE Region, U.K. Service Report 1451165,
12 pp + 19 pp.
Beck, F., and Kaiser, H. (1975). Elektrochemische Untersuchungen an EisedZink-
hkalelementen (An electrochemical study of iron-zinc couples). Metalloberfliiche,
29(6), 281-286 (in German).
Becker. (1978). Appraisal of the hot dip galvanizing of a suspension cable from the
Rodenkirchen Bridge over the River Rhine. Wire, September-October, 28(5), 196-
197.
Belisle, S. , and DuFresne, R. (1986). Corrosion resistance of ZA alloys. Int. Symp.
Zinc-Aluminium Casting Alloys, Toronto, Canadian Institute of Mining and
Metallurgy, Montreal, pp. 109-126.
Belz, H. and Dorken, E. (1989). Organic-inorganic thin film corrosion protection
coating. Metallobeflache, 43(1 l), 524-526 (in German).
Belz, H. W. (1989). Coated and galvanized screws compared from a technical and an
economical point of view. Metalloberfliiche, 43, 81-82 (in German).
Bengough, G. D., and Hudson, F. (1919). Fourth report to the Corrosion Commit-
tee the Institute of Metals, Section 11: The corrosion of zinc. J . Inst. Met., 21,
Bengough, G. D., and May, R. (1924). The corrosion of zinc by seawater. J . Inst.
Bengough, G. D., Stuart, J. M., and Lee, A. R. (1929). Quantitative measurement of
Bennett, L. H. et al. (1978). Economic Effects of Metallic Corrosion and Its Preven-
37-352.
Met., 32, 82-243.
corrosion of metals in water and salt solutions. Trans. A " , 83, 2946.
454 References
tion in the U.S.A. U.S. Department of Commerce/Bureau of Standards, Washing-
ton, DC.
Benninghoff, H. (1970). Erhohte Korrosions- und Verschleissbestandigkeit von Zink
(Increased corrosion and wear resistance of zinc). Oberji'uche, 4, 131-134 (in
German).
Bensimon, R. (1976). Bridges and large structures in France. 8th Int. Thermal
Spraying Conf., American Welding Society, Miami, F%, pp. 232-235.
Bergstroem, U., and Engberg, G. (1988). Aluzink with good coating ductility for
severe forming applications. 2nd Int. Conf. Zinc-Coated Steel Sheet, ZDA, London,
Berkeley, K. G. C., Stuart, M., and Lee, A. R. (1962). Theory of corrosion in light
of quantitative measurements. Proc. R. Soc. London 116, 425-467.
Bernal, N., and Tierra, P. (1988). Hot-dip zincJ% aluminium coatings on fabricated
steel and cast iron components. 15th Int. Galv. Conf., ZDA, London, pp. GF91-8.
Berukshtis, G., and Klark, G. (1965). Atmospheric corrosion of steel, zinc, cadmium,
copper and aluminium in different coastal and continental regions. Korroz. Met.
Splavov Sb., (2), 332-350 (in Russian).
Berukshtis, G. K., and Klark, G. B. (1971). Corrosion Resistance of Metals and
Metallic Coatings Under Atmospheric Conditions. Nauka (Science), Moscow, pp.
50-62 (in Russian).
Beuerlein, P. (1941). Die Korrosion der Metall durch Handschweiss und ihre Ver-
hiitung (The corrosion of metal by hand perspiration and its prevention). Werkstatt
Betrieb, 74, 23-25 (in German).
Beythein, R. (1947). Corrosion of phosphated zinc alloys by fire extinguishing liquid.
Arch. Metallkd., I , 286-288 (in German).
Bharucha, N. R., and Baker, M. T. (1967) Lacquers for zinc alloys. Trans. Inst. Met.
Finish., 45, 141-146.
Bianchi, G. (1954). Passivity of zinc in relation to conditions of aeration. Metall. Ifal.,
46, special issue, 29-30 (in Italian).
Bianchi, G. (1958). Corrosion of zinc by differential aeration. Corrosion, 14(5), 245-
248. See also Metall. Ital., 45(9), 323-327 (1953).
Bianchini, A., Pozzoli, S., and Lanfranco, G. (1968). Anodic behavior of zinc in
water and diluted aqueous solutions. Research Report, Centro Ricerche Metal-
lurgiche SPA, Turin, December, 51 pp.
BIEC InternationaUBHP Steel (1987). 55% AI-Zn. BIEC, Bethlehem, PA, 20 pp.
BIEC International (1991). Continuous hot-dip 55% aluminium-zinc coated steel
sheet. BIEC International, Siegen, Germany. 48 pp.
BIEC International (1991). Prepainted hot-dip 55% aluminium-zinc coated sreel sheet.
BIEC International, Siegen, Germany. 45 pp.
Biestek, T. (1959a). Die Korrosionsbestandigkeit von galvanischen Zink- und Kad-
miumuberziigen nach verschiedenen Schnellpriifverfahren (The corrosion behavior
of plated zinc and cadmium coatings in various accelerated tests). Jahrb. Ober-
Jachentech., 15, 322-333 (in German).
Biestek, T. (1959b). Badania korozji atmosfercznej pasywowanych elektrolytyoznyde
pp. SC5/1-8.
References 455
powlok cynkowych na stali. Wyniki czteroletnick badan W latach 1954-1958
(Research on the atmospheric corrosion of chemically passivated electrodeposited
zinc coatings. Results of four-year tests, 1954-1958). Pr. Inst. Mech. Prec.,
8(28-3), 23-28; 53-60 (in Polish).
Biestek, T. (1961a). Comparative outdoor exposure tests of electrodeposited zinc and
cadmium coatings of steel in a natural industrial atmosphere. 1st. Int Congress on
Metallic Corrosion, London, pp. 269-275; or preprint 455-460.
Biestek, T. (1961b). Badania korozji atmosferycznej powlok elektrolityznyck Druji
lomunikat o przebiegu i nyniky baden W latack 1958-1959. (Research on the
atmospheric corrosion of electrodeposited coatings: 2nd Progress report for period
1958-1959). Pr. Inst. Mech. Prec., 9(3I-l), 39-79 (in Polish).
Biestek, T. (1974a). Produkty korozji cynku tworzqce sie W roznych srodowiskach
korozyjnych (The corrosion products formed in zinc in various corrosive environ-
ments). Powoki Ochr., (6), 24-27 (in Polish).
Biestek, T. (1974b). Corrosion resistance of electroplated coatings. Tropical climate
conditions (China). Met. Finish. 39-41.
Bilfinger, R. (1957). Klimaschutz durch genonnte galvanische Metalluberzuge unter
besonderer Beriicksichtigung des metallischen Tropenschutzes fur electrotechnische
Erzeugnisse (Climatic behavior of plated metal coatings for electrochemical prod-
ucts). Dtsch. Electrotechn., 11, 116-123 (in German).
Blanchard, F. (1977). Corrosion of Galvanized Steel Tubes by Hot and Cold Waters.
Literature Survey. CEBELCOR (Belgium), October, 132 (RT.240) (133 refs.) (in
French and English).
Blee, J. J. , Kelly, S. W., and Williams, J. L. (1985). The galvanic effects of noble
metals in manholes. Corrosion 85, NACE, Houston, TX, Paper 373, 9 pp.
Blekkenhorst, F., Nagel Soepenberg, E., and Schoen. J. P. (1991). Results of a new
cyclic corrosion test for automotive steel products based upon micro-climate mea-
surement in running vehicles, 3rd Int. Zinc Coated Steel Sheet Conf., Preprint S3D,
BNFMRA. (circa 1960). The corrosion of zinc alloy die castings in hot and boiling
solutions of some commercial detergents. Typescript, 3 pp. (Unpublished; copy in
ZDA library, London).
BNFMRA. (1964). Guidance on the Use of Galvanized Tanks and Cisterns.
BNFMRA, London, 8 pp.
BNFMTC. (1988) Unpublished work BNF R508110 Wantage, U.K.
See also British Non-Ferrous Metals Research Association.
Boger, K., and Kubler, E. (1985). Wasser- und Wertstoffeinsparung. Spultechniken
an Tmmmelverzinkungsanlangen. Technishce Grandlagen und Betriebserfahrungen
(Water and materials conservation rinsing techniques in barrel zinc plating plants.
Technical principles and operating experiences). Galvanotechnik, 76( 1 l ) , 1659-
1666 (in German).
Bohm, D., and Willert, H. (1977). Studies of the corrosion behavior of dip galvanized
zinc coatings on steel in seawater. Schigbauforschung, 16, 81-87 (in German).
Bollinger, J. (1952). Uber die Korrosion verschiedener Metalle in verglussigtem
13 PP.
456 References
Schwefeloxid (Corrosion resistance of various metals in liquid sulfur dioxide).
Schweiz. Arch. Angew. Wiss. Tech., 18(10), (in German).
Bonaretti, A., Capoccia, A., Giardetti, G. , and Muccino, U. Process and use
properties of Lavegal. 2nd lnt. Conf. Zinc-Coated Steel Sheet, ZDA, London, pp.
Booker, C. J. L. The Thames Barrier project. A case study in corrosion and
Bora, J. S. Metal-plastic contact corrosion. Corros. Sci., 14(8), 503-506.
Borbe, P. C., Erdmann-Jesnitzer, F., and Jun, E. J. Verbesserung des
Korrosionswiderstandes von Zink-Aluminium-Legierungen durch Stickstoff (Im-
provement of the corrosion resistance of zinc-aluminum alloys by nitrogen). Metall,
(in German).
Borchers, H., and Krug, H. Die Beeinflussung der Korrosion von Zink durch
Metalle, die im Korrosionsmedium als Kationen enthalten sind (Influence of metal
cations in the corrosive medium on the corrosion of zinc). Werkst. Korros., 22(2),
(in German).
Borgmann, C. W., and Evans, U. R. The corrosion of zinc in chloride
solutions. Trans. Electrochem. Soc., 65,
Borgmann, C. W. The initial corrosion rate of mild steel. lnd. Eng. Chem.,
29,
Borzillo, A. R., and Niederstein, K. Galvalume sheet-licensing and applica-
tions. 1st l nt . Conf. Zinc-Coated Steel Sheet, ZDA, London, pp. HI-H8.
Borzillo, A. R., Forand, J. L. , and Chehi, S. E. Worldwide application for
Galvalume sheet steel. 2nd l nt . Conf. Zinc-Coated Steel Sheet, ZDA, London, pp.
SE7t1-8.
Bottcher, H. J. (Yearly). Feuerverzinkung. (A yearly review of hot dip galvanizing
developments with references.) Jahrbuch OberjKchentechnik, including 36,
protection. Bull. Inst. Corros. Sci. Technol., March, pp. 5-8.
45, 46,
(Special reprints from GAV.)
Bottcher, H. J. Zur atmospharischen Korrosion von Zink und Zinkiiberziigen
(Atmospheric corrosion of zinc and zinc coatings). Werkst. Veredll.,
(in German).
Brace, A. W., and Porter, F. C. The Economics ofProtection by Zinc Coat-
ings. ZDA, London, 5 pp. [Reprinted from Met. Mater.
Brandt, J. Private communication.
Brenner, A. Electrodeposition of Alloys: Principles and Practice, vols.
Academic Press, New York.
Bretschnieder, E., and Rabald, E. and later). Dechema-Werkstofle-Tabelle
(Materials Tables). Verlag Chemie, Weinheim-Bergstrasse, pp.
British Non-Ferrous Metals Research Association. Improved method and
means for preventing or hindering corrosion of zinc, zinc alloy zinc coated parts
in water systems. British Patent 554,046.
References 457
See also BNFMRA.
British Standards Institution. (1966/1991). BS 1133: Section 1966 Temporary
Prevention Metal Surfaces Against Corrosion, 120 pp. Subsection 6.1 (Clean-
ing: 1991), 32 pp. and 6.2 (Temporary protectives): 1991, 20 pp. BSI, London.
British Standards Institution. (1977). BS 5493: Code of Practicefor Protective Coat-
ing of Iron and Steel Structures Against Corrosion. BSI, London, 114 pp.
British Standards Institution. (1979). PD 6484: Corrosion of Bimetallic Contacts and
Its Alleviation. BSI, London, 32 pp.
British Standards Institution. (1984). BS 6497 . . . Powder Organic Coatings . . . on
Galvanized Steel. BSI, London, 12 pp.
British Standards Institution. (1990). BS 4479 Design of Articles That Are TO Be
Coated, BSI, London, Parts 1-3, 6, 7, + 16 + 12 + 20 + 12 pp. BSI,
London.
British Standards Institution. (1991). CP Cathodic Protection: Parr 1: Code
Practice for Land and Marine Applications. BSI, London, 116 pp.
British Steel Corporation. (1981). Galvatite Technical Manual. BSC Strip Mill Prod-
ucts, Newport, Gwent, 30 pp.
British Steel Corporation. (1984). Colorcoat and Stelvetite Pre-Finished Steel-
Technical Manual. BSC Strip Mill Products, Newport, Gwent, 36 pp.
British Steel Corporation (1987). Roofing and Cladding in Steel-A Guide to Architec-
tural Practice and BSc Colorcoat Technical Specijkations. BSC Strip Mill Prod-
ucts, Newport, Gwent, 32 pp. + 8 pp.
BSC Strip Mills Products. (1986). Zalutite: Zinc-Aluminium Alloy Coated Steel,
Technical Manual, Newport, Gwent, 22 pp.
BSCIBCSAIPMAIZDA. (1986). Steelwork Corrosion Protection Guide for Steelwork
in Building Refurbishment. British Steel General Steels, Rotherham, U.K., folder, 8
PP.
for Perimeter Walls (issue British Steel General Steels, Rotherham, U.K.,
folder, 8 pp.
in Exterior Environments (issue British Steel General Steels, Rotherham, U.K.,
folder, 8 pp.
in Interior Environments (Issue 3). British Steel General Steels, Rotherham, U.K.,
folder, 8 pp.
Bukowiecki, (1958). Investigation of the attack on metals by detergents. Schweiz.
Arch. Angew. Wiss. Tech., 24, 355-370 (in German).
Bukowiecki, A. (1968). Zur hiifung von Chlorkohlen-wasserstopffen auf ihre metal-
langreifenden Eigenschaften (Chlorinated hydrocarbons attack on metals). Schweiz.
Arch. Wiss. Tech., 34(11), 363-361 (in German).
Bullett, T. R. (1969-1970). Intmduction to zinc dust paints. Symposium on Protecting
Steel with Zinc Dust Paints, ZPDA, London, 4 pp.
Bullig, H. J. (1956). Zur Entstehung eines Laderaumklimas, von Laderaum- und
References
Ladungsschweiss (Loading room climate including perspiration). In Seewetteramt
Hamburg: Laderaummeteorologie. EinzelveroTentlichung, (in German).
Bundesverband der Deutschen Industrie. Recycling gerechte Produktion und
Produktgestaltung (in German).
Burchard, H. Klimabeanspruchung und Oberflachenschutz (Climatic require-
ments and protective coatings). Metall, (in German).
Burgmann, G. , and Schwenk, W. Corrosion behavior of hot dip galvanized
steel in cold and warm waters with gassing by COz and/or as a function of lime
hardness. Werkst. Korros., (in German).
Burkard, P., and Revelly, P. Phenomtnes de corrosion sur les tubes dacier
zinguts au trempe (Corrosion phenomena in hot dip galvanized steel tubes). Con-
struire, March, (in French).
Bums, R. M., and Bradley, W. W. Protective coatings for metals, 2nd ed.
Reinhold, New York, pp.
Bums, R. M. Preservation of electrical equipment. Corros. Technol.,
Butler, J. F. Vapour generation and deposition of zinc at high rates. J . Vac.
Sci. Technol.,
Butler, J. T., and Morrow, H. and Corrosion performance of thermal
sprayed zinc-aluminum alloys (typescript). ZMTG, P.O. Box Greenwich,
CT pp.
Butlin, R. N. The efsects of acid deposition on buildings and building
materials. H.M.S.O., London, pp.
Buzdakov, A. P. Zinc coatings as a means of protecting steel against
corrosion. Proc. Conf. Marine Corros. Met., Baku Azerbaizhan government,
Buzdakov, A. P. Resistance of pipes with thermodiffused zinc coatings in
various media. Collect Sci. Pap. Mfr. Pipes with Anti-Corrosion Coatings, pp.
(in Russian).
Cadmium Association Technical Notes on Cadmium: Cadmium Coatings,
Cadmium Association, London, 8 pp.
Callaghan, B. G. Long-Term Atmospheric Corrosion Programme-Results
Afrer IO Years Exposure. National Building Research Institute of the CSIR,
Pretoria, South Africa. WBOU pp.
Camiglieri, W. Anticorrosion water treatment of soft water supplies. Austral-
asian Corros. Eng. ,
Camp, T. R. Water and I t s Impurities. Reinhold, New York.
Campbell, H. S. The use of non-ferrouus metals in domestic water supply.
Chem. Ind.,
Campbell, H. S. , Stanners, J. F., and Watkins, K. Effect of heat treatment
on the protective properties of zinc coatings on steel. J . Iron Steel Inst.,
1.
Carson, J. A. H., Phillips, W. L. M., and Wellington, J. R. A laboratory
evaluation of zinc anodes in seawater. Corrosion,
pp.
References 459
Carson, J. H. Zinc as a self-regulating galvanic anode for ship hulls. Corro-
sion, 16 (Oct.), 99t-I04t,
Carter, J. P., Linstrom, P. J., Flinn, D. R., and Cramer, S. D. The effects of
sheltering and orientation on the atmospheric corrosion of structural metals. Mater.
Performance,
Carter, V. E. The choice between cadmium and zinc plating for the protection
of steel. Met. Finish.,
Cauchetier, J. (Observations on sprayed zinc samples in the sea). Corros.
Anticorros., 7, see also Corros. Anticorros., 8, (in
French).
Celestin, A. B., Harrison, R. G., Lynch, R.F., and Goodwin, F. E. Galfan
coated steel for upgraded formability and corrosion protection, Metal Bull. Coatings
Conf., Oct. IO, pp.
Centre Technique du Zinc. Resistance d la corrosion atmospherique des
alliages de zinc laminh. (Atmospheric corrosion resistance of zinc alloy sheet).
Centre Technique du Zinc, Levallois-Perret, France. pp (in French).
Centre Technique du Zinc. Project ZM-155, Corrosion Resistant Fasteners
for Galvanized Structures Report 9. ILZRO, Research Triangle Park, NC, pp.
Cerveny, L. Passivierung des Zinks mittels I-Dimethylathan-l-ol-Chromats
in der Gasphase (Passivation of zinc in the gaseous phase by means of 1 , l -
dimethylethane-l-ol-chromate). Werkst. Korros., (in German).
Champion, F. A. Corrosion Testing Procedures. Chapman & Hall, London,
pp.
Chandler, K. A. Marine and Ofisshore Corrosion, Section Marine Environ-
ments. Butterworths, London.
Chandler, K. A., and Bayliss, D. A. Corrosion Protection of Steel Structures.
Elsevier, New York, 1 pp.
Chang, J.-C., and Wei, H.-H. Electrochemical and Mossbauer studies of the
corrosion behavior of electrodeposited Fe-Zn alloys on steel. Corros. Sci.,
Cherault, M., and Charbonnier, G. Ten years of working and testing results on
corrosion proving ground. Gafvatech Verlag Stahleisen, Dusseldorf, Germany,
Chernov, B. B. , and Ilin, A. K. Corrosion of metals in the tropical zone of
the Indian Ocean. Prot. Met., l ) , (translation of Zashch. Metall.).
Childs, J. N. Zinc sprayed coatings in a liquid cargo tanker ship. 8th Int.
Thermal Spraying Conf., American Welding Society, Miami, FL, pp.
Christofferson, D. W. Coatings for steel water storage tanks. J . Am. Water
Works Assoc., 53,
Clarke, S. G. Metallic coatings: Their atmospheric corrosion resistance and
protective value. Met. Ind., 55,
Clarke, S. G., and Bradshaw, W. N. Test of the protective value of metallic
coatings under sheltered conditions (marine atmosphere). J . Appl. Chem.,
pp.
References
Clarke, S. G., and Longhurst, E. E. (1961). The corrosion of metals by acid vapours
from wood. J . Appl. Chem. (1 l ) 4 3 5 4 3 .
Clarke, S. G., and Longhurst, E. E. (1962). The corrosion behaviour of metals and
protective coatings in tropical atmosphere exposure tests. 1st Int. Congress on
Metallic Corrosion, Butterworths, London, pp. 254-268.
Clothier, A. D. (1981). Coatings for galvanized steel in industrial building. J . Oil
Colour Chem. Assoc., 64(8) 294-298.
Collins, K. J. (1957). The corrosion of metal by palm sweat. Br. J . Ind. Med., 14,
Cominco Product Research Centre. (1985). Atmospheric corrosion results of ZM-l13
galvanized samples after fifteen years of exposure. Report ILZRO Project ZM-113.
Cominco Product Research Centre, Sheridan Park, Ontario. Sponsored by Interna-
tional Lead Zinc Research Organization, Inc., 43 pp.
Compton, K. G. (1947). Corrosion in the tropics. Trans. Electrochem. Soc., 91,705-
713.
Compton, K. G., Arnold, S. M., and Mendizza, A. (1951). Effect of packaging on
corrosion of zinc plated equipment. Corrosion, 7, 365-372.
Compton, K. Mendizza, A., and Bradley, W. W. (1955). Atmospheric galvanic
couple corrosion. Corrosion, 11, 383t-392t.
Constantinescu, E., and Heitz, E. (1976). On the influence of the water content on
corrosion of metals in mono-carboxylic acids. Corros. Sci., 16(1 l ) , 857-867.
Cook, A. R., and Lattin, B. C. (1962). Zinc anodes in a crude oil tanker. Mater.
Prof., 1(11), 8-14.
Cook, G. S. , and Dickinson, N. (1950). Corrosion of metals by insecticidal solutions.
Corrosion, 6, 137-139.
Copson, H. R. (1951). Effect of specimen shape on corrosion in the atmosphere.
Cordner, R. J. (1990). Atmospheric corrosion survey of New Zealand. Br. Corros. J . ,
Cotterell, N. H., Noel, R. E. J. , and Allen, C. J. (1990). Pitting behaviour of
corrosion resistant steels in synthetic mine waters. Br. Corros. J . , 25( l ) , 60-72.
Couy, C. J. (1948). Effect of atmospheric corrosion on maintenance and economics of
overhead line hardware and guy strand. Corrosion, 4,
Cox, G. L. (1931). Effect of temperature on the corrosion of zinc. Ind. Eng. Chem.,
Cramer, S. D., Carter, J. P., Linstrom, P. J. , and Flinn, D. R. (1987). Environmental
effects in the atmospheric corrosion of zinc. Symposium on the Degradation
Metals in rhe Atmosphere, ASTM STP 965. American Society for Testing and
Materials, Philadelphia, pp. 229-247.
Crennell, J. T., and Wheeler, W. C. G. (1956). Zinc anodes for use in seawater. J .
Appl. Chem., 415-421.
Crennell, J. T., and Wheeler, W. C. G. (1958). Zinc alloy anodes. J . Appl. Chem.,
191-197.
25(2), 115-1 18.
23, 902-904.
8(9), 571-576.
References 461
CRM. ILZRO project. ZE3-Painted galvanized steel. Progress reports
CRM. ILZRO project. See Lamberigts and Leroy
Cross, J., and Bailey, G. Research unravels mystery of corroded fasteners.
Building Today,
Czakow, J. Spectrochemical analysis of high purity cadmium with precon-
centration of impurities by cementation. Chem. Anal., 12(5), (in Polish).
Daesen, J. R. Corrosion resistance of galvanized coatings. 2nd Int. Congress
on Metallic Corrosion, National Association of Corrosion Engineers, Houston, TX,
pp.
Daesen, J. R. Mechanism of Corrosion of the Coating. Pamphlet, the U.S.
Galvanizing Institute, pp.
Daeves, J. , Piingel, W., and Radeker, W. Die Haltbarkeit von Verzinkungen
gegeniiber Korrosionsangriff, (The durability of galvanizing against corrosion at-
tack). Stahl Eisen, (in German).
Daeves, K., Mewes,.K. F., and Schulz, E. H. Uber die Einflussgrossen und
die Auswertung des Naturrostungsversuchs bei unlegierten und schwach legierten
Stiihlen (Influence and evaluation of natural rust formation with unalloyed and low-
alloy steels). Korros. Metallschutz, (in German).
Darbin, M., Jailloux, J. M., and Montuelle, J. Durability of reinforced earth
structures: The results of a long-term study conducted on galvanized steel. Proc.
Inst. Civil Eng., Part I ,
Dattilo, M. Polarization and corrosion of electrogalvanized steel-Evaluation
of zinc coatings obtained from waste-derived zinc electrolytes. J. Electrochem.
Dattilo, M., Cole, E. R., and O'Keefe, T. J. Recycling of zinc waste for
electrogalvanizing. Conserv. Recycling,
David, G. Comparative study of the protection afforded by zinc and cadmium
coatings. Corrosion, (in French).
Davidson, D. D., Soreide, L. E., and Stephens, M. L. Assessment of the
Chrysler chipping corrosion test using laboratory scab and long-term outdoor
exposure testing. Galvatech Iron and Steel Institute, Japan, pp.
Davidson, D., et al. Corrosion behavior of precoated steel sheets in Volvo
tests. Galvatech Verlag Stahleisen, Dusseldorf, Germany, pp.
Davies, D. E., and Lotlikar, M. M. Passivation and pitting characteristics of
zinc. Br. Corros. J., 1(4),
Davies, J. G. , Doremus, E. P., and Pass, R. B. New aspects of the application
of cathodic protection to offshore facilities. Corrosion Toronto, National Asso-
ciation of Corrosion Engineers, Houston, TX, Paper 15.
Davis, R. L. Cathodic protection of offshore structures and pipe lines. Proc.
Appalachian Underground Corrosion Short Course, University of West Virginia,
Morgantown, p.
Davy, Sir Humphrey. On the corrosion of copper sheathing on ships by sea
from CRM, Belgium. ILZRO.
l),
462 References
water and a method of preventing this effect. . . . Phil. Trans. R. Soc., 114, Part 1,
197.
de Bie, I. R. (1970). Staaldraad en verzinkung (Galvanized wire). Thermisch Ver-
zinken (702), 10-14 (in Dutch).
de Konig, H. (1969). Metro Rotterdam. Bescherming van stalen onderdelen van de
eindvoegconstructies van de afgezonden tunnelstukken (Corrosion problems in the
new Rotterdam Metro). Thermisch Verzinken (692), 3-7 (in Dutch).
de Waard, C., and Posch, W. (1972). Cathodic protection in cargolballast tanks of
crude oil tankers. Werkst. Korros., 23(11), 1008-1012 (in English).
Decherna [Deutsche Gesellschaft fur chemisches Apparatewesen]. (1953). De-
chema-Werkstoffe-Tabells (Materials tables). Verlag Chemie, Weinheim-
Bergstrasse, lo00 pp + .
De la Cuadra, (1991). Recycling of zinc coated scrap. 3rd Int. Zinc Coated Steel
Sheet Conf., Barcelona, Preprint S4L, 14 pp., 58 refs.
Deiss, E. (1936). Uber Beschadigungen an Zinkrinnen und Vorstossblechen von
Biturnenpappdachern (The corrosion of zinc gutters and weatherings by bitumen
roof board). Verfag-Jahrbuch, 1936, 123-135 (in German).
Deiss, E. (1941a). (Atmospheric corrosion-two articles). Verlag-Jahrbuch, 1941,
3145, 46-51. (in German).
Deiss, E. (1941b). Das Verhalten des Zinks an Bauwerken gebeniiber atmo-
spharischen Einfliissen (The behavior of zinc in building against atmospheric
influences). Wiss. Abhl. dtsch. Materia1pruf.-Anst. 2nd ser. (2), 3 145. Zinkkorro-
sion und die konservierende Nachbehandlung von Pappdachern (in German).
Delahay, P., Pourbaix, M., and van Rysselberghe, P. (1951). Potential-pH diagram
of zinc and its application to the study of zinc corrosion. J . Electrochem. Soc., 98,
Delpeuch, Y. (1972). La protection de lacier par des revetements de metallisation au
zinc et h laluminium (Protection of steel by zinc and aluminum metallized coat-
ings). Corrosion, 20(6), 416-420 (in French).
Dempster, N. S. (1965). Cathodic protection of Kort nozzles. Mater. Prot., 4(2), 24,
26-27.
Denison, 1. A. (1947). Contribution of Sir Humphrey Davy to cathodic protection.
Corrosion, 3, 295.
Desai, M. N., Rana, S. S. , andGandhi, M. H. (1968). CorrosionInhibitorsforZinc.
Gujarat University, Ahmedabad, India, 20 pp.
Desai, M. N., Rana, S. S. , and Gandhi, M. H. (1973). Corrosion inhibitors for zinc.
Anticorros., Methods Mater., 20( 12), 3-6.
Deshmukh, M. B. (1987). Corrosivity of harbour waters. Bull. Electrochem., 3(5),
379-384.
Determann, H. (1956). Der Schutz technischer Exportguter durch geeignete Ver-
packungsmethoden. Seewetteramt Hamburg. Laderaummeteorologie, (The protec-
tion of export goods through suitable packaging methods). Special publication
Number 9, 34-39 (in German).
101-105.
References 463
Determann, H. Korrosionsschutz bei Seeversand im Tropenklima (Corrosion
protection for sea transport in tropical climates). Werkst. Korros., 8, (in
German).
Dettner, H. W., and Elze, J. , Eds. Handbuch der Galvanotechnik
(Handbook of Electroplating Technology). Carl Hanser, Munich; Vol. 1, part
Vol. 1, part Vol. Vol. (in German).
Deutsche Institut fur Normung: see DIN for German standards.
Devillers, J. P., and Niessen, P. The mechanism of intergranular corrosion of
dilute zinc-aluminum alloys in hot water. Corros. Sci.,
Devillers, J. P., and Niessen, P. Accelerated grain boundary corrosion of
dilute-aluminum alloys under high applied cathodic currents. Corrosion,
Diegle, R. B. and Kelley, K. l ) . Comparative corrosion evaluation ofzinc-based
and aluminum metallized coatings. Batelle Inst. (for Texasgulf) Report pp.
Dillenberg, H. Bath for the electrolytic stripping of galvanic coatings made of
nickel, chromium gold from base bodies made copper, copper alloys, silver,
zinc titanium. U.S. Patent pp.
DIN Parts 1 to Probability of Corrosion of Metallic Materials When
Subject to Corrosion from the Outside, Beuth Verlag, Berlin, 5 + + 11 pp. Engl.
ed.).
DIN Hot Dip Batch Galvanizing. Beuth Verlag, Berlin, 5 pp. (Engl.
ed.).
DIN Parts 1 to 5. Corrosion of Metallic Materials Under Internal
Corrosion Stress by Water. Beuth Verlag, Berlin (in German: English edition of
Part deals with hot dip galvanized materials), + IO + + IO + pp.
Dirkse, T. P. Passivation Studies on the Zinc Electrode. 7th Int. Power
Sources Symp., Brighton, preprint, 8 pp.
Dirkse, T. P., and Hampson, N. A. Anodic behavior of zinc in aqueous
solution. Part Passivation in mixed KF-KOH solutions. Electrochim. Acta, 17(5),
Donovan, P. D., and Stringer, J. Corrosion of metals and their protection in
atmospheres containing organic acid vapours. Br. Corros. J. ,
Downing, A. G. Frictional sparking of cast magnesium, aluminum and zinc.
Natl. Fire Prot. Assoc., Q.
Dreulle, N., and Dreulle, P. Die Beziehung zwischen Umweltverschmutzung
und Korrosion von Zink und feuerverzinktem Stahl (Relationship between environ-
mental pollution and the corrosion zinc and galvanized steel). Metall,
628 (in German).
Dreulle, N., and Druelle, P. Use of surface treatment for solving a corrosion
problem of galvanized steel by water. 11th Int. Gab. Conf., ZDA, Portcullis Press,
London, pp.
Drisko, R. W. Cathodically protecting mooring chains in seawater. Mater.
Prot. Performance,
464 References
Drisko, R. W. Equipment and procedures for cathodic protection of fleet
moorings. Muter. Prot. Pe~ormunce,
Dugdale, P. J. Calculations of the resistance to earth of vertical galvanic
anodes. Anticorros. Methods Muter.,
Dunbar, S. R. Effect of one per cent copper addition on the atmospheric
corrosion of rolled zinc. Metal Corrosion in the Atmosphere, ASTM STP
ASTM, Philadelphia, pp.
Duncan, J. R., and Spedding, D. J. Initial reactions of SO, after adsorption
onto metals. Corros. Sci., l ),
Duncan, J. R., and Spedding, D. J. The effect of relative humidity on
adsorption of sulphur dioxide onto metal surfaces. Corros. Sci.,
Duncan, J. R. Corrosion of metal fasteners in timber. Construct. Building
Muter.,
Dwight, J. H. Elec. Eng., 55,
Dylyaeva, L. I. , Chertovskikh, N. B., and Usenko, K. B. The corrosion
stability of titanium-alloyed zinc and cadmium coatings. Zushch. Met.,
(in Russian).
Eberhard, K., et al. (Improvements in the corrosion protection of zinc by
codeposition of metals of the eighth subgroup.) MetulloberJliche,
(in German).
ECSS. See European Convention for Structural Steelwork.
EGGA. See European General Galvanizers Association.
Ehlert, I. Beitrage zu Fragen der Korrosion von Eisen und Zink durch
Bakterien (Contribution to the question of the corrosion of iron and zinc due to
bacteria). Muter. Org.,
Ehreke, J. , and Stichel, W. (Influence of increasing phosphate/silicate con-
tents on the pitting and general corrosion of galvanized steel tubing and copper in
warm water mixed installations). Werkst. Korros., (in German).
Eilers, W. (Modem zinc-corrosion protection systems.) Merullobefluche,
(in German).
Eisenstecken, F. Der Wert von Schutzschichten bei Forderwagen (The value
of protective coatings for railway trucks). Gliickuuf, (in
German).
Eisler, S. J. , and Faigen, H. L. Investigation of synthetic fingerprint solutions.
Corrosion, 10,
Eliassen, S. , and Steensland, Cathodic protection criteria for the North
Sea. J . Met.,
Elliott, P. State of the of automobile corrosivity up to Corrosion
National Association of Corrosion Engineers, Cincinnati, Paper
Ellis, B. Effect of weather on the initial corrosion rate of sheet zinc. Proc.
Ellis, B. How zinc coatings slow atmospheric corrosion. Iron Age,
Eke, J. , and Oelsner, G. Die durch Handschweiss verursachte Korrosion
ASTM,
References
metallischer Oberflachen und Wege ihrer Verhutung (How hand perspiration causes
corrosion of metallic coatings and ways of preventing it). Blech, 4, (in
German).
Elze, J. Schaden an oberflachenbehandelten Blechteilen und ihre Verhutung
(Damage to coated sheet parts and its prevention). Mitt. Forsch. Ges. Blechverarb.,
pp. (in German).
Era, A., Takehara, Z., and Yoshizawa, S. Die Wirkungen von Verun-
reinigungen, insbesondere Blei, in Mangandioxyd auf die Korrosion von Zink (The
effect of impurities, particularly lead, in MnO, on the corrosion of zinc). Denki
Kagaku, 35, (in German). See also Electrochim. Acta,
Erskine, D. The North Sea current flowing round platforms to combat
corrosion. Engineer,
ESB Inc. Zinc electrode containing lead sulfide to reduce gassing at the zinc
electrode in heat-sterilized silverhinc alkaline battery. U.S. Patent
Escalante, E., Iverson, W. P., et al. Corrosion andProtection ofSteel Piles in
a Natural Seawater Environment. Institute for Materials Research, National Bureau
of Standards, Nl3S Monograph pp. Available from Government Printing
Office, Washington, DC.
Escalante, E. The effect of soil resistivity and soil temperature on the
corrosion of galvanically coupled metals in soil. Galvanic Corrosion. American
Society for Testing and Materials, Philadelphia, pp.
Espada, L., Sanchez, A., Pena, G. , and Merino, P. (Sheltered and unsheltered
corrosion comparative study. Two year exposure results.) Rev. Metal.,
(in Spanish).
European Coal and Steel Community. Euronorm Continuously Organic
Coated Steel Flat Products, ECSC, Luxembourg. Available from BSI, London, 1
pp. (To be replaced by EN IO
European Coil Coaters Association. ECCA-Eurodes. 3rd Int. Zinc Coated
Steel Sheet Conf., Preprint S3B, pp.
European Convention for Structural Steelwork. Protection of Steel Structures
Against Corrosion by Coatings. ECSS, 50, pp.
European Federation of Corrosion. Eurocorr Includes papers on corrosion
products formed on zinc from Svensson and Johansson and from Odnevell and
Leygrac 15-year tests on metal sprayed steel by Henriksen.
European General Galvanizers Association. The Corrosion Resistance of
Galvanized Steel. EGGA, London, pp.
European General Galvanizers Association Proc. 17th Internutionul Galvaniz-
ing Conference. [Earlier conference volumes also issued] EGGA, London.
European Zinc and Alloy Coated Sheet Committee Proc. 4th Zinc Coated
Sheet Conference [Earlier conference volumes also available] EGGA, London.
European Zinc Institute. Galfan. Brochure, available from ZDA, London,
PP.
Evans, E. LI., and Stroud, G. Prevention of corrosion in packing and storage:
466 References
The control of atmospheric corrosion of metals in storage and transit. Chem. Ind.,
Evans, R. M. An assessment of the atmospheric corrosion resistance of hot
dipped galvanized steel after five years exposure. British Steel Report SM1569/C,
Evans, U. R., and Britton, S. C. Preliminary investigation of laboratory
corrosion tests. First Report of the Corrosion Committee. Iron & Steel Institute,
London, pp.
Evans, U. R. An Introduction to Metallic Corrosion. Edward Arnold & Co.,
London, pp.
Evans, U. R., and Davies, D. E. The pitting of zinc by distilled water and
dilute solutions. J . Chem. Soc., Also in Corrosion, 8,
Evans, U. R. Corrosion and Oxidation of Metals. Edward Arnold, London,
pp.
Evans, U. R. Corrosion and Oxidation Metals: First Supplementary
Volume. Edward Arnold, London, pp.
Evans, U. R. Corrosion and Oxidation Metals: Second Supplementary
Volume. Edward Arnold, London, pp.
Feitknecht, W. Uber den Angriff von Metallen in feuchten Dampfen der
Halogenwasserstoffsauren (On the attack of metals (zinc, cadmium, nickel and iron)
by moist hydrogen halide vapours). Helv. Chim. Acta, (in
German); Zinc Absrr.,
Feitknecht, W. Uber den Zusamrnenbmch der Oxydfilme auf Metallo-
berflachen in sauren Dampfen und den Mechanisms der atmosphirischen Korrosion
(The breakdown of the oxide films on metal surfaces in acid vapours and the
mechanism of atmospheric corrosion). Chimia, (in German); Zinc Abst.,
Feitknecht, W. Die festen Korrosionsprodukte als korrosionslenkende Fak-
toren (Influence of adhering corrosion products on corrosion behavior). Werkst.
Korros., (in German).
Feitknecht. W. Uber die Diagnose von Korrosionsursachen aus den Korro-
sionsprodukten (Diagnosing the causes of corrosion from corrosion products).
Werkst. Korros., 9, (in German).
Feitknecht, W. Studies on the influence of chemical factors on the corrosion of
metals. Chem. Ind.
Fielding, B. C. Two approaches painting new and weathered galvanized
steel. U.K. Corrosion Conf., Institute of Corrosion Science and Technology,
Birmingham, paper, pp.
Finch, G. J. , and Quarrell, A. G. Oxidation of zinc. Proc. Phys. Soc., 46,
et
Fischer, H., and Bh a n n , H. Gemeinsames und Besonderheiten ver-
schiedener galvanischer Zinkiiberziige (General and specific properties of different
types of electrodeposited zinc coatings). Z. Metallkd., I), (in Ger-
man), Zinc Abstr., 1.
PP.
References 467
Fitzgerald, J. H. (1972). Cathodic protection of stationary marine structures. Mater.
Prot. Performance, 11(5), 23-27.
Fitzgerald, J. H., 111. (1981). Fundamentals of corrosion. Appalachian Underground
Corrosion Short Course, University of West Virginia, Morgantown.
Flinn, D. R., Cramer, S. D., Carter, J. P., and Spence, J. W. (1985). Field exposure
study for determining the effects of acid deposition on the corrosion and deteriora-
tion of materials-Description of programme and preliminary results. Durabiliy
Build. Mater., 3(2), 147-175.
Flinn, D. R., Cramer, S. D., Carter, J. P., Hurwitz, D. M., and Linstrom, P. J.
(1986). Environmental effects on metallic corrosion products formed in short-term
atmospheric exposures. Materials Degradation Caused by Acid Rain. American
Chemical Society, Washington DC, pp. 119-151.
Fomin, G. (1992). Corrosion and Corrosion Protection Handbook (in Russian).
GOSSTANDART, VNIISOT, Russia (relates particularly to IS0 TC 156 and other
international tests).
Fontana, M. G. (1987). Corrosion Engineering, 3rd ed., McGraw-Hill, New York, p.
169.
Forgeson, B. W., Southwell, C. R., Alexander, L., Mundt, H. W., and Thomp-
son, L. J. (1958). Corrosion of metals in tropical environments. Corrosion, 14,33t-
41t, 73-81t.
Forgeson, B. W., et al. (1963). Abyssal Corrosion and Its Mitigation. Part I I . Results
a Pilot Test Exposure. U.S. Department of Commerce, OTS, Report AD
296,545, 23 pp.
Foster, G. R., Lee, S. G., and Myers, J. R. (1974). Stress corrosion of eta-phase, hot
dip zinc alloy in potable water. Corrosion, 30(7), 239-241.
Francis, R. C. (1959). Zinc anode cathodic protection system on U.S. submarine
Wahoo; Bureau of Ships J . Corros. Technol., 183-184.
Frappe, L., Limare, A., Piessen, P., and Vacher, J. C. (1985). Investigation of the
influence of impurities or alloying elements on the atmospheric corrosion of gal-
vanized steel. 14th Int. Galv. Conf., ZDA, London, pp. 8/3-8118.
Frasch, J. (1961). Corros. Anticorros., 9, 225 (in French); Chem. Abstr., 1962, 57
8331.
Freeman, D. B., and Brunt, P. J. (1982). Pretreatment'options cold-rolled and
galvanized steels. ECCA Congress, Brussels, 8 pp.
Frens, Rijkhoff, Dekker, and Wijdenes. (1991). Paper presented to Adhesion Sympo-
sium, Netherlands Corrosion Centre.
Friend, J. N., and Tidmus, J . S. (1924). The relative corrosion of zinc and lead in
solutions of inorganic salts. J . Inst. Met., 31, 177-191.
Friend, J. N. (1928). The relative corrodibilities of ferrous and non-ferrous metals and
alloys. Part 1, The results of four years' exposure in Bristol Channel. J . Inst. Met.,
Friend, J. N. (1929). The relative corrodibilities of ferrous and non-ferrous metals.
Part 11, The results of seven years' exposure to air at Birmingham. J . Inst. Met., 42,
149-155.
39, 11 1-143.
References
Friend, J. N. (1932). The relative corrodibilities of ferrous and non-ferrous metals and
alloys. Part 111, Results of three years exposure at Southampton docks. J . I ns t .
Met., 48, 109-120.
Froguet, M., and Pagetti, J. (1989). Contribution of the electrochemical impedence
measurement technique in zinc corrosion inhibition studies in acid media. Werht.
Korros., 40, 634-641 (in English).
Fuel Research. (1957). Measurement ofAir Pollution. HMSO, London, pp. 14-20.
Fujisawa, H., et al. (1992). Anomalous corrosion on fluoro-resin top coated gal-
vanized steel sheet in regions with heavy snow. Galvatech 92, Verlag Stahleisen,
Dusseldorf, Germany, pp. 351-355.
Fukutani, E., and Akiba, T. (1961). The relation between corrosion of propeller shaft
and zinc anode for cathodic protection of propeller. Corros. Eng. , 10, 289-292.
Fukutani, E., Kimura, A., and Mizoghuchi, H. (1961). Zinc anode protection for peak
tanks and double bottom tanks of cargo vessels. Corros. Eng. , 10, 53-55.
Fuller, T. S. (1927). Report of ASTM Committee B-3 on Corrosion of Non Ferrous
Metals and Alloys. Proc. ASTM, 27(1) 281-298.
Gabe, D. R., Ed. (1983). Coatingsfor Protection. Institution of Production Engi-
neers, London, 162 pp.
Gad Allah, A. G. , Hefny, M. M., Salih, S. A., and El-Basiouny, M. S. (1989).
Corrosion inhibition of zinc in HCI solutions by several pyrazole derivatives.
Corrosion, 45(7), 574-578.
Gad Allah, A. G. , Mogoda, S. , and Hefny, M. M. (1990). Corrosion and control
of zinc by some organic acid anions. Corros. Prevent. Control, 37(2), 47-53.
Gallaccio, A., and Comet, I. (1960). Report on Marine Atmosphere Exposure
Galvanic Couples Involving Magnesium, ASTM STP 255. ASTM, Philadelphia, 26
PP.
Galvanizers Association. (1979). Galvanizing Guide. GNZDA, 58 pp.
Galvanizers Association (1986). Thickness of Zinc Coatings. GA, London, 1 p.
Galvanizers Association (1993). Hot Dip Galvanizing, GA, London, 32 pp.
Galvanizers Association of Australia. (1985). Galvanizing Manual and Directory.
GAA, Melbourne, 175 pp.
Galvatech 89. (1989). Int. Conf. Zinc and Zinc Alloy Coated Steel Sheet, Tokyo. Iron
and Steel Institute, Japan, 80 papers, 632 pp. [Note: For Galvatech 92 (1992)
references are given to the main relevant papers.]
Galvazinc Association. (1991). Une ethique des r6alisations (Examples of galvaniz-
ing). Galvazinc, Paris, folder (in French).
Garg, S. P. (1970). Reduction of zirconium oxide with zinc-magnesium alloys in
molten flux media. Trans. Indian Inst. Met., 23(1), 37-42.
Gartner, F. W. (1976). Corrosion protection of iron and steel by thermal spraying with
zinc and aluminum. 8th Int. Thermal Spraying Conf., American Welding Society,
Miami, FL, pp. 214-222.
Gaspar, L., and Soriano, A. (1971). Anodos de cinc en banos cianurados alcalinos
(Zinc anodes in alkaline cyanide baths). Tecn. Met., 26(184), (in Spanish).
Gaur, J. N., and Jain, B. N. (1978). Effect of some surface active substances on
corrosion of zinc. J . Electrochem. Soc. India, 27(2), 117-1 18.
References
GAV (1967). (Cost analysis for galvanizing of fabricated parts.) Translated by ZDA,
London. 3 pp.
Gavranek, V. V., and Levchenko, A. A. (1967). Microscopic study of the initial
stages in the cavitation failure of single crystals. Phys. Metals Metallogr., 24(6),
Gebhardt, E., and Sheghezzi, H. D. (1956). Uber das Korrosionsverhalten von
flammenensgespritzen Zinkschichten (The corrosion behavior of flame-sprayed zinc
coatings). MetalloberJache, IO, 369-376 (in German).
Geduld, H. (1988). Zinc Plating. Finishing Publications, Ltd., Teddington, and ASM
International, Metals Park, OH, 360 pp.
Gehmecker, H. , and Kurzmann, P. (1992). Cooperation between supplier and cus-
tomer in developing new pretreatment processes with higher environmental compat-
ibility. Galvatech 92, Verlag Stahleisen, Dusseldorf, Germany, pp. 227-230.
Gellings, P. T. (1977). Energy analysis of corrosion and corrosion protection. Pro-
cestechniek-W, 32 (in Dutch).
German, G. (1978). Corrosion prevention with thermal-sprayed zinc and aluminum
coatings. Proc. Anniversary Symposium, Atmospheric Factors Affecting the Corro-
sion of Engineering Metals, ASTM STP 646. ASTM, Philadelphia, pp. 74-82.
Gessner, G. , Hiitterbriucker, K.-F., Meyer zu Bexten, J. H., and Mohr, K.-P.
(1988). Electrogalvanized sheet steel-The German scene. 2nd Int. Zinc Coated
Sheet Conf., ZDA, London, Paper SE4/1-18.
Gibbs, W. E., Smith, R. H., and Bengough, G. D. (1916). Corrosion of zinc and zinc-
copper couple in seawater. J . Inst. Met., 15, 108-147.
Gilbert, P. T. (1948). Corrosion of zinc and zinc coated steel in hot water. Sheet Met.
Ind., (BNFMRA RRA 776 reprint,
Gilbert, P. T., and Hadden, S. E. White rust formation on zinc. J . Inst. Met.,
l ), 47-70.
Gilbert, P. T., and Hadden, S. E. (1950b). Corrosion of cadmium and zinc coatings in
electrical equipment, J . Electrodep. Tech. Soc., 25,
Gilbert, P. T. (1952). The nature of zinc corrosion products, J . Elecrrochem. Soc.,
99(1) 16-21.
Gilbert, P. T. (1953). The effect of impurities in the metal on the rate of corrosion of
zinc and galvanized coatings in the atmosphere. J . Appl. Chem. 3, 174-181.
Gilbert, P. T. (1956). Corrosion-fatigue. Metall. Rev., 1(3), 379-417.
Ginzberg, V. I. , and Kabakova, I. (1969). Korrosionsbestandigkeit von Metallen
in Jod und jodhaltigen Medien. (Corrosion behavior of metals in iodine and iodine-
containing media). Metallschutz, 5(6) 627-632 (in German).
Glass, G. K., and Ashworth, V. (1985). The corrosion behavior of the zinc-mild steel
galvanic cell in hot sodium bicarbonate solution. Corros. Sci., 25(1 l), 971-983.
Glayman, J. (1970). Loxydation anodique du zinc (Anodizing of zinc). Galvano,
39(397), 147-149 (in French).
GLC. (1981). Corrosion of metal fasteners used in conjunction with CCA (copper/
chromelamenate) preservative treated timbers. Development Materials Bulletin 135,
Greater London Council, London, pp. 4/ 14/ 3.
102-107.
32 PP.)
470 References
Gliszewski, L. (1971). Aspekty ekonimiczne ochrony katodowej kadlubow statkow
krajowej floty transportowej (Economic aspects of the cathodic protection of ships
hulls in the Polish mercantile fleet). Ochr. Przed. Koroz., 14(5), 15-19 (in Polish).
Gmelin-Institut fur Anorganische Chemie. (1956). Gmelins Handbuch der Anorgan-
ischen Chemie, Zink Erganzungsband (Grnelins Handbook of Inorganic Chemistry,
Zinc Supplement), 8th ed., Verlag Chemie, Weinheim, (in German), 1025 pp.
Gogler, P. (1987). Technology and economics of hot dip galvanizing. Metal-
lobetflache, 41(2), 93-94 (in German).
Goncharevskii, M. S. , Shchesno, L. P., and Khitrik, I. S. (1966). Corrosion re-
sistance of pipes with thermodiffused zinc coatings. Collect. Sci. Papers Mfr. Pipes
with Anti-Corrosion Coatings, pp. 127-132 (in Russian).
Goodwin, F. E. (1985). An up-date on Galfan. 1st Int. Conf. Zinc-Coated Steel Sheet,
ZDA, London, pp. 511-516.
Goodwin, F. E. (1988). The effect of heat treatment and iron content on the structure
and properties of zinc plated steel sheet. 2nd Int. Conf. Zinc-Coated Steel Sheet,
ZDA, London, pp. SD4/1-11.
Goodwin, F. E., and Piessen, P. (1988). Results of a ten-year exposure on fasteners
for structural steel in an industrial atmosphere. 15th In?. Galv. Conf., ZDA,
London, pp. GH2/l-15.
Goodwin, F. E. (1991). The effect of environmental acidification on atmospheric
corrosion of zinc. 16th In?. Galv. Conf., EGGA, London, pp. GHlll-12.
Goodwin, T. J., Isaac, J. S. , and Parsons, T. (1991). Exterior corrosion tests for the
automotive industry. 3rd Int. Zinc Coated Steel Sheet Conf., EGGA, London,
S3Cll-15.
Gorbunov, N. S. , Glovatskii, V. G., Sukhomlin, A. I. , et al. (1977). Corrosion
protection of pipe systems functioning in coke gas purification media. Koks Khim.
(6), 43-45 (in Russian).
Gorling, P. (1957). Uber die Ursachen und Verhutung der Korrosion bei verpackten
Metallgegenstanden (Cause and prevention of corrosion of packaged metal objects).
Verpackungsrundschau, 8, 2-8 (in German).
Grauer, R. (1966). Zinkformiat-dihydrat als Korrosionsprodukt auf Zink in Kontakt
mit Phenol-Formaldehydharzen (Zinc formate dihydrate as the corrosion product on
zinc in contact with phenol formaldehyde resins). Werkst. Korros., f7(9), 766 (in
German).
Grauer, R., and Schlinder, P. (1972). Die loslichkeitskonstanten der Zinkhydroxid-
chlorids-ein Betrag zur Kenntnis der Korrosionsprodukte des Zinks (Solubilities of
zinc hydroxychloride-A contribution to the study of zinc corrosion products).
Cor m. Sci., 12(5), 405414 (in German).
Gray, R. (1986). Paints and stains. Archit. J. , June 23 suppl, 38 pp.
Greater London Council. See GLC.
Greenburg, L., and Jacobs, M. B. (1956). Sulphur dioxide in New York City
Grigorev, V. P. (1967). Increasing the wear resistance and corrosion stability of
atmosphere. J . Ind. Eng. Chem., 48, 1517-1521.
References 471
metals and alloys by saturating the surface layers with halogens. Russian Patent
189,278 (in Russian).
Grigorev, V. P., and Kuznetsov, V. V. (1969). Application of the Taft constants in
studying the inhibiting effects of certain amino-acids. Prot. Metals, 5(4), 356-361
(translation of Zashch. Metaff.).
Gross, H. (1984). Salt deposits on anticorrosive coatings of bridges. fnd. Corr., 2(9),
14-17.
Grossman, P. R. (1978). Investigation of atmospheric exposure factors that determine
time-of-wetness of outdoor structures, Atmospheric Factors AfSecting the Corrosion
Engineering Metals, ASTM STP 646, ASTM, Philadelphia, pp. 5-16.
Grover, T. K., Mukheriji, N. K., and Sangal, B. (1961). J . Sci. fnd. Res., 200, 445.
Grubitsch, H., and Illi, (1940). (Hot water corrosion of zinc: 11.) Korros. Metal-
lschurz, 16, 197-203 (in German).
Grubitsch, H., and Sinigoj, J. (1940). (Hot water corrosion of zinc: I) Korros.
Metallschutz, 16, 194-197 (in German).
Grubitsch, H. (1969). Uber die Heisswasser-Druckkorrosion von Zink (Corrosion of
zinc in hot water, under pressure). Werkst. Korros., 20(1 l), 949-954 (in German).
Grubitsch, H. H., and Huemer, H. (1973). (Scanning electron microscope studies of
the hot water corrosion of zinc.) Werkst. Korros., 24(1), 1-7 (in German).
Guest, R. M. (1956). The potentials of zinc and steel in tap water. Can. J . Technof.,
Guillaume, I. (1970). Role de certaines bactkries dans la corrosion de cuivre, du nickel
et du zinc en milieu doux et marin. (Role of certain bacteria in the corrosion of
copper, nickel and zinc in soft water and seawater). Electrochim. Acta, 15(9),
1445-1454 (in French).
Guillaume, I. , Grimaudeau, J. , and Brisou, J. (1977a). Corrosion et protection du zinc
par un milieu bactkrien doux et marin. (Corrosion and protection of zinc by a fresh
water and sea water bacterial medium). Corros. Sci., 17(9), 765-775 (in French).
Guillaume, I. , Grimaudeau, J. , and Brisou, J. (1977b). Contribution bactkrienne B la
corrosion et B la protection des metaux (Bacterial contribution to corrosion and
protection of metals). Corros. Sci., 17(9), 753-763 (in French).
Guillen, M. A., and Feliu, S. (1970). Behavior of different paints on steel coated with
zinc and aluminum in seawater. Corros. Prot., 1(1), 3542, 45.
Gupta, A. N. (1979). Corrosion of steel and galvanized steel sheets in Bhubaneswar
atmosphere. Indian J . Technol., 17(1 l ), 427-430.
Gusev, N. I. , and Rakcheev, P. V. (1968). Kinetics of anodic processes and polariza-
tion of a rotating zinc electrode in zinc sulfate and sodium perchlorate. Fiz.
Khim., 42(8), 1983-1986 (in Russian).
GUS, H., and Regula, H. (1956). Die flugklimatogischen Verhiiltnisse deutscher
Verkehrsflughdfen (Proportions of air constituents at German airports). Ber. Dtsch.
Wetterdienstes, 3(21), Bad Kissingen (in German), 48 pp.
Haarmann, R. (1959). (Zinc thickness and the corrosion of galvanized parts.) Meral-
foberfliiche, 13( 1 l ) , 342-345 (in German).
34, 245-247.
472 References
Haarmann, R. (1960a). Grundlagen des Korrosionsschutzes durch Feuerveninken
(Fundamentals of protection against corrosion by hot dip galvanizing). Indus-
trie-Anzeiger, 82(34), 517-521 (in German).
Haarmann, R. (1960b). Das Kreuz im Venn (The Venn cross). MetalloberJliiche, 14,
53-57.
Hadden, S. E. (1952). Effect of annealing on the resistance of galvanized steel to
atmospheric corrosion. J . Iron Steel Inst., 171(2), 121-127.
Hall, D. E. (1983). Electrodeposited zinc-nickel alloy coatings-A review, Plat.
Surf. Finish, 1 1) 59-65.
Halliday, M. D., and Horton, C. A. P. (1967). Etch pit technique for revealing
dislocations intersecting surface parallel to (1 120) planes in zinc. J . Inst. Met.,
Halliday, R. (1978). Corrosion protection aspects of the Thames Barrier project.
Corros. Prev. Control, 25(3), 19-22.
Haney, E. G. (1981). The zinc-steel potential reversal in cathodic protection. Corro-
sion 81 Conf., Toronto, NACE, preprint, 15 pp.
Haney, E. G. (1985). Zinc sacrificial anode behavior at elevated temperatures.
Corrosion 85, NACE, Houston, TX, paper 7 pp.
Hansen, M. (1958). Constitution ofBinary Alloys, 2nd ed., McGraw-Hill, New York,
1305 pp.
Hapke, H.-J., Pallauf, J., and Kampe, W. (undated) Ohne zink kein leben (no life
without zinc) ZinkberatunglBeratung Feuerveninken, Dusseldorf, Germany 8 pp.
Hart, R. G., and Townsend, H. E. (1983). Heat reflectivity of 55% AI-Zn alloy coated
steel sheet. Met. Progr., 124(6), 24-33.
Harvey, J. (1978). Corrosion Resistance of Zinc Coatings with Low Alloy
Content-Literature Review. Research and Technology Report No. 662, John
Lysaght (Sydney, Australia) Ltd.
Hatch, G. B. (1952). Inhibition of galvanic attack of steel with phosphate glasses. Ind.
Eng. Chem., 44, 1780-1786.
Hayashi, K., Ito, Y., Kato, C., and Mytoshi, Y. (1990). Behavior at underfilm
corrosion tips of Zn-Fe alloy coated steel sheets. Tetsu To Hagane, 76(9), 1496-
1503 (in Japanese).
Haynie, F. H., and Upham, J. B. (1970). Effects of atmospheric sulfur dioxide on the
corrosion of zinc. Mater. Prot. Performance, 9(8), 3540.
Haynie, F. H., Spence, J. W. , and Upham, J. B. (1976 and 1978). Effects of air
pollutants on weathering steel and galvanized steel: A chamber study. Atmospheric
Factors Affecting the Corrosion of Engineering Metals. ASTM STP 646, Proc.
Golden Anniversary Symposium. ASTM, Philadelphia, pp. 30-47. Also report
Haynie, F. H. (1980). Theoretical air pollution and climate effects on materials
confirmed by zinc corrosion data. In Durability of Building Materials and Compo-
nents, ASTM STP 691, P. J. Sereda and G.G. Levitan, Eds., ASTM, Philadelphia,
95(12), 381-383.
EPA-600/3-76-015 (1976), 85 pp.
pp. 157-175.
Haynie, F. H. (1987/1988). See ASTM STP 965, p. 451.
References 473
Haynie, F. H., Spence, J. W., et al. Evaluation of an atmospheric damage
function for galvanized steel. Corrosion Testing and Evaluation: Silver Anniversary
volume. ASTM STP 1000. ASTM, Philadelphia, pp.
Heim, G. Korrosionsverhalten von feuerverzinkten Bandstahl-Erdern (Corro-
sion behaviour of hot dip galvanized steel earthing strip). TU Sicherheit
Zuverlassigkeit Wirtsch. Betrieb Verkehr (in German).
Heinrich, G. Corrosion and corrosion protection. Review of the Activities,
22nd ed.. Zinc-The Versatile Metal. Metallgesellschaft, Frankfurt, pp.
Heitman, T. M. Zinc anodes stop pipeline corrosion. Eng. Dig.,
Heitz, E., Hukovic, M., and Maier, L. Grundvorgange der Korrosion von
Metallen in organischen Losungsmitteln (Fundamental processes of corrosion of
metals in organic solvents). Werkst. Korros. (in German).
Helwig, L. E., and Bird, J. E. Protection of galvanized coatings from attack
by organic acid vapors. Met. Finish.,
Helwig, L. E., and Murray, M. V. Vacuum metallizing with zinc. Met.
Finish.,
Hercules, Inc. Zinc-rich coatings containing conductive carbon black. U.S.
Patent
Heumann, T., and Venker, Formation of Alloys Between Low and High
Melting Metals by the Reduction Volatile Metal Halides and Clarijication the
Composition of Alloys Obtained by This Process. Communautk Europ6en de L'En-
ergie Atomique (Euratom), pp.
Heurtaux, J. Cathodic protection by sacrificial anodes. ATB Metallurg.,
18( 34) , (in French).
Hildebrand, H., and Schwenk, W. (Effect of galvanizing on steel corrosion in
concrete immersed in NaCl solution.) Werkst. Korros., (in German).
Hindhaugh, E. W. Coil coated steel sheet. Ind. Finish.,
Hinton, B. R. W., and Wilson, L. The corrosion inhibition of zinc with
cerrous chloride. Corros. Sci., 29(8),
Hippensteel, C. L., and Borgmann, C. W. Outdoor atmosphere corrosion of
zinc and cadmium electrodeposited coatings on iron and steel. Trans. Am. Electro-
chem. Soc., 58,
Hisamoto, J. , et al. The properties of paint adhesion and corrosion resistance
after painting of steel sheet covered with a zinc-polymer composite coating.
Galvatech 92, Verlag Stahleisen, Diisseldorf, Germany, pp.
Hoar, T. P. Report of the Committee on Corrosion and Protection. Depart-
ment of Trade and Industry. HMSO, London, pp.
Hoffmann, J. ( Korrosionssichere Verpackung (Corrosion-preventive packag-
ing), 2nd ed., Verlag fur Fachliteratur, Berlin-Grunewald (in German), pp.
Holler, E. Schadenfalle und ihre meteorologische Deutung (Harmful deposits
and their meteorological interpretation). Seewetteramt Hamburg: Laderaurn-
meteorologie. Einzelveroffentlichung N. 9, (in German).
Holler, P. , Knotkova, D., Cerny, M., Strekalov, P. , et al. Atmospheric
474 References
corrosion of unalloyed steel, zinc, copper and aluminum and its relation with certain
characteristics of the atmosphere. Analysis of the results of ten years of tests camed
out in European countries of the Committee for Mutual Economic Assistance. P rot.
Met., 1987 22(5), 539-546 (translation from Zashch. Metall.).
Horstmann, D. (1964). The influence of steel and of galvanizing conditions on the
properties of galvanized sheet and strip. 7th Int. Galv. Conf., Pergamon Press,
Oxford, pp. 146-166.
Horstmann, D., and Peters, F. K. (1971). The reactions between iron and zinc. 9th
Int. Galv. Conf.., Industrial Newspapers, Redhill, U.K., pp. 75-106.
Horstmann, D. (1983). Faults in Hot Dip Galvanizing. Verlag Stahleisen, Dusseldorf,
Germany, 79 pp.
Horstmann, D. (1985). Formation and growth of iron-zinc alloy layers. 14th Int.
Galv. Conf., Munich, ZDA, London, pp. 6/1-5.
Horvick, E. W. (1961). The use of zinc in corrosion service. Metals Handbook, Vol.
1 , 8th ed. American Society for Metals, Metals Park, OH, pp. 1162-1 169.
Houghton, C. J. , and Ashworth, V. (1981). The performance of commercially
available zinc and aluminum anodes in sea-bed mud at elevated temperatures.
Corrosion/81, Toronto, NACE, preprint, 14 pp.
Hoxeng, R. B., and Prutton, C. F. (1949). Electrochemical behavior of zinc and steel
in aqueous media. Corrosion, 5, 330-338.
Hoxeng, R. B. (1950). Electrochemical behavior of zinc and steel in aqueous media.
Part 11. Corrosion, 6, 308-312.
Hubbard, D. J. , and Shanahan, C. E. A. (1973). Corrosion of zinc and steel in dilute
aqueous solutions. Br. Corros. J . , 8(6), 270-274.
Huber, R. (1968). Trockenbatterien (Dry batteries) 2nd ed. Varta AG, Frankfurt am
Main, (in German), 142 pp.
Hubner, W. , Goetze, W. , and Suwelski, G. (1978). (Accelerated pitting of zinc due to
incorrect heat treatment-A damage case.) Neue Hiitte, 23(8), 309-31 I (in Ger-
man).
Hudson, J. C. (1929). Atmospheric corrosion of metals. Third (experimental) report to
the atmospheric corrosion committee. Trans. Faraday Soc., 25, 177-252.
Hudson, J. C. (1938). The committees field tests on atmospheric corrosion. Fiflh
Report ofthe Corrosion Committee. Iron and Steel Institute, London, pp. 13-68.
Hudson, J. C., and Banfield, T. A. (1946). The protection of iron and steel by metallic
coatings. J . Iron Steel Inst., I I , 229P-264P.
Hudson, J. C., and Acock, G. P. (1952). Special Report, 45th Symposium on the
Protection of Buried Metals. Iron and Steel Institute, London, pp. 1-28, 55-74.
Hudson, J. C., and Stanners, J. F. (1953a). The protection of iron and steel by metallic
and non-metallic coatings-11. J . Iron Steel Inst., 175, 381-390.
Hudson, J. C., and Stanners, J. F. (1953b). The effect of climate and atmospheric
pollution on corrosion. J . Appl. Chem., 3(2), 86-96.
Hudson, J. C., and Kenworthy, C. (1957). Antifouling properties of zinc coatings.
Chem. Ind., 1390-1391.
References 475
Hudson, J. C. Sirth Report of the Corrosion Committee. Special Report
Hudson, J. C. Final report of service trials of steel coal wagons
Hudson, J. C. Corrosion of steel in buildings. Chem. Ind.,
Humayun, A. Roll forming of metal coated sheet steels. National Steel
Corporation (U.S.A.), presented at ECCA, Spring Meeting pp.
Hunter, N. War at sea. Cover Story, No. AM&S Europe, Bristol, pp.
Also nos. 5
Hutchins, J.S., and McKenzie, M. Characterisation of bridge locations by
corrosion and environmental measurement-First year results. Department of the
Environment, Transport Road Research Laboratories, Crowthorne. Berks., TRRL
Report LR550, pp.
Ichiyama, K., Kobayashi. J., and Sugimoto, S. A new zinc-aluminium
coating for general hot dip galvanizing. 14th Int. Galv. Conf., ZDA, London,
Ikeda, K., and Satoh, H. An analysis of corrosion cracking phenomena of
electrodeposited Zn-Ni alloy layers. Galvatech 89, Iron and Steel Institute, Japan,
pp.
ILZRO Garfan Manual, 3rd ed. ILZRO, Research Triangle Park, NC, pp.
ILZRO I ) . Galvanized Reinforcement for Concrete-II. ILZRO, Research Trian-
Imai, H. Reaction of zinc dust with water and sodium dichromate. J . Chem.
Imhoff, W. G. Keeping white rust from zinc coated products. Iron Age, 21( l ) ,
Imperial Smelting Corp. (circa Mazak 3"Tests in solutions of detergents.
Unpublished typescript, pp. (Copy in ZDA library, London.)
Imperial Smelting Corp. Ltd. Metal fusion and purification. French Patent
International Standards Organization IS0 Metallic coatings-Neutral
salt-spray tests (NSS test), pp.
International Standards Organization IS0 Chromate conversion coat-
ings on electroplated zinc and cadmium coatings, pp.
International Standards Organization. IS0 Corrosion of Metals and
Alloys-Classijication Corrosivity of Atmospheres, pp.
International Standards Organization. IS0 Corrosion of Metals and
Alloys-Values for the Corrosivity Categories of Atmospheres, 5 pp.
International Standards Organization. IS0 Corrosiviry of At-
mospheres-Methods of Determination of Corrosion Rate of Standard Specimens
for the Evaluation of Corrosivity, 8 pp.
International Standards Organization. IS0 [Revision to edition on Hot-dip
galvanized coatings to be published
Iron and Steel Institute, London, pp.
J . Iron and Steel Inst. 194 (Jan)
pp.
gle Park, NC, pp. and appendices (includes bibliography of refs).
Soc. Japan, Ind. Chem. Sect., 58,
476 References
Ishutin, A. V. (1971). Anodic dissolution of Zn-Mn alloys in molten mixture of
sodium and potassium chlorides. Izv. Vyssh. Ucheb. Zaved. Tsvetn. Met. (6), 49-51
(in Russian).
Ito, K., Oohashi, H., and Yoshii, T. (1989). Corrosion behavior of zinc plated
stainless steel. Galvatech 89, Iron and Steel Institute of Japan, pp. 569-576.
Ivanov, E. S., and Klyuchnikov, N. G. (1969). Adsorption kinetics of amino chro-
mates from aqueous solutions onto steel, zinc, cadmium, copper and aluminum.
Zashch. Metall., 5(5), 547-550 (in Russian).
Ivochev, M., and Mihailov, G. (1988). Restraining the corrosion of galvanized steel
by the use of Neutralite-A case study. 15th Int. Galv. Conf., ZDA, London, Paper
Izeki, T., et al. (1992). Development of a new hot dip process for Zn-5% AI alloy
coatings on steel structures and evaluation of its corrosion resistivity. Proc. 1st
Asian Pacific Gen. Galv. Cong., pp. 158-166.
Izumi, M., and Hashizume, Y. (1992). Edge creep of prepainted Zn-AI alloy coated
steel sheet. Galvatech 92. Preprint, 6 pp.
Izumo, S., Sueyoshi, H., Kitamura, K., and Ohzuno, Y. (1990). Corrosion of metals
in volcanic atmospheres: The effects of volcanic ash. Corros. Eng. , 271-
282, 247-253 (in English).
Jagannathan, V. (1992). The influence of coating characteristics on the powdering and
corrosion resistance galvannealed coated steel sheet. Galvatech 92, Verlag
Stahleisen, Dusseldorf, Germany, pp. 127-131.
Jagner, D., and Kryger, L. (1975). Computerized electroanalysis. Part 111. Multiple
scanning, anodic stripping, and its application to seawater. Anal. Chim. Acta,
Jailloux, J. M. (1990). Galvanized steel reinforcement in soils: Experience of Terre
ArmCe. EGGA General Assembly, Lausanne, EGGA, London, p. 2.1.
Jain, B. L., and Gaur, J. N. (1978). Corrosion inhibitors for zinc in organic acids.
Indian J . Technol., 439-440.
Jasovsky, F., Lacova, E., and Alexy, M. (1988). Steel plate with aluminum-zinc
coating: Properties and uses. Hutnik, 38(10), 382-385 (in English).
Jenkins, W. M., Bevan, J. R., and OConnor, B. (1969). Looking to the future in the
processing of thin steel strip. J . Iron Steel. Inst., 207(6), 916-920.
Jensen, F. O., Rygg, A., and Setre, (1978). Testing of zinc anodes for offshore
buried pipelines at elevated temperatures. Mat. Performance, 17(9), 9-15.
Johansson, E., and Rendahl, B. (1991). Field exposure and outdoor scab-testing of
different pre-coated steels. 3rd Int. Zinc-Coated Steel Sheet Conf., S3All-9.
Johnen, H. J., Ed. (1984). Zinc fur Stahl (Zinc for steel). Zinkberatung, Dusseldorf,
Germany, 216 pp., 183 refs. (in German).
Johnson, J. G. B., Skerry, B. S., and Wood, G. C. (1983). The influence of smoke
and isobutane on corrosion in sulphur dioxide-polluted environment. J . Electro-
chem. Soc., 130(8), 1650-1656.
GH3/ 1-5.
80(2), 255-266.
References 477
Johnson, K. E., and Abbott, J. S. (1975). Bimetallic Corrosion Effects on Mild Steel
in Natural Environments. Report CELICC15175. Corporate Engineering Laboratory,
British Steel Corporation, London, 12 pp.
Johnson, W. R. , and Pfkter, L. E. (1984). High speed continuous strip plating of zinc
alloys. Proc. 4th Cont. Strip Plating Conf., Chicago, American Electroplaters
Society, pp. 14.
Johnson, T., and Kucera, V. (1988). Eight years field exposure of coatings of zinc,
aluminium and their alloys. 2nd Int. Conf. Zinc-Coated Steel Sheet, London,
Paper 611-1 1.
Jonassen, H. B. (1958). Zinc as a cathodic inhibitor. Corrosion, 14, 37%-376t.
Jones, D., and Brinn, D. (1983). Zinc Coated Steels in Building. BSC Welsh
Laboratory. Report WL/RSC/202/83/D(R), 210 pp.
Jones, and Wilde, B. E. (1977). Corrosion performance of some metals and
alloys in liquid ammonia. Corrosion, 33(2), 46-52.
Jones, F. E. (1957). Corrosion of metals in buildings. The resistance of non-ferrous
metals to corrosion in buildings. Chem. Ind. (31), 1050-1062.
Jones, R. D., et al. (1988). Short-term electrochemical evaluation of coated steel strip.
2nd Int. Conf. Zinc-Coated Steel Sheet, ZDA, London, Paper 5/ 14.
Jones and Laughlin Steel Corp. (1967). Coating a metal by vapour deposition. British
Patent 1,183,971, 4 May.
Jones and Laughlin Steel Corp. (1972). Vacuumhapor deposited zinc coatings. U.S.
Patent 3,700,485.
Joussellin, M., and Wiart, R. (1979). Inhibition of anodic dissolution of zinc by
tetrabutylammonium bromide in ZnCI,-NH.,CI electrolyte. Electrochim. Acta,
24(8), 891-893.
Juchniewicz, R., Jankowski, J. , and Szukalski, J. (1991). Electrochemical protection
of domestic hot and cold water systems. Mater. Performance, January, pp. 22-26.
Jun, J. H., et al. (1992). Corrosion resistance and coloring of galvanized steel sheets
by vacuum deposition of silicon top layers. Galvatech 92, Verlag Stahleisen,
Dusseldorf, Germany, pp. 346-350.
Kain, R. M., and Baker, E. (1989). Marine atmospheric corrosion museum report
on the performance of thermal sprayed coatings on steel. ASTM STP ASTM,
Philadelphia, pp. 21 1-234.
Kanamam, T., et al. (1992). Corrosion resistance of Zn-Cr alloy electrocoated steel
sheets. Galvatech Verlag Stahleisen, Diisseldorf, Germany, pp. 331-335.
Katz, W. (1976). Seawater Dechema. Data sheets 2-5, Deutsche Gesellschaft fir
Chemisches Apparatewesen. FrankfurVMain, Germany.
Kawai, S. , Kato, H., et al. (1975). Influence of pyridine derivatives on the corrosion
of zinc. Denki Kagaku Kogyo Butsuri Kagaku, 43(3), 127-130 (in English).
Kawanishi, Y. , et al. (1992). Performance of organic composite-coated steel sheet for
home electric appliances. Galvatech 92, Verlag Stahleisen, Dusseldorf, Germany,
pp. 402406.
References
Kayser, F. (1963). Uber die Kennzeichnung der Auflagenstirke von Phos-
phatschichten (Phosphate coatings). Galvanotechnik, 63( l), 11-23 (in German).
Kayser, F. (1972). Ubor die Kennzeichung der Auflagenstkke von Phospharschichten
(Behavior of phosphate films) Galvanotechnik l), 11-23.
Kehrer, H. P. (1974). The effect of small impurities on grain boundary corrosion of
zinc die castings. Metall, 28(9), 883-889 (in German).
Kehrer, H. P., et al. (1982). Influence of the copper content on the properties of zinc
die castings, Metall, 36(11), 1216-1219; 36(12), 1310-1312 (English translation,
ZDA, London, 14 pp.
Kelley, V. I. (1975). Atmospheric Corrosion Investigation of Aluminium-Coated,
Zinc-Coated and Copper-Bearing Steel Wire and Wire Products. A Twelve-Year
Report. ASTM STP 585, ASTM, Philadelphia, 89 pp.
Kenworthy, L., and Smith, M. D. (1944). Corrosion of galvanized coatings and zinc
by waters containing free carbon dioxide. J . Inst. Met., 70, 463489.
Kemdge, D. H. , and Tariq, S. A. (1967). Solution of zinc in fused zinc chloride.
J . Chem. Soc. A, Inorg. Phys. Theoret., (7), 122-1 125.
Ketchin, V. A., and Lyublinskii, E. Ya. (1986). Zinc Alloys. Mettalurgiya, Moscow
(in Russian). Chapter 3, pp. 89-100, covers corrosion of zinc.
Khanlarova, A. G., Mamedov, M. I. , and Aliev, A. G. (1974). Corrosion protection
of marine oil platform supports by thermodiffused zinc coatings in the periodic
wetting zone. Korrs. Zashch., (l), 30-33 (in Russian).
Khanlarova, A., Nagiyeva, Ya., and Aliyev, A. (1991). Investigation of corrosion
resistance of metal sprayed coatings for the supports of offshore structures. Alum-
ozinc Symposium, Dnepropetrovsk, typescript (in Russian), 5 pp.
Kim, M. M. (1991). Use of zinc-coated steel as building panels and roofing materials
in agricultural applications. 16th f nt . Galv. Conf., EGGA, London, Paper GBll-6.
Kimberley, J. L. (1957). Control of corrosion with zinc coatings, Corrosion, 13,
King, C. V., and Mayer, N. (1953). The rate of dissolution of zinc and cadmium in
chromic chloride solutions. J . Electrochem. Soc., 100( 1 l), 473479.
King, R. A. (1977). A review of soil corrosiveness with particular reference to
reinforced earth. TRRL Supplementary Report 316. Transport and Road Research
Laboratory, Crowthorne, Berks., 27 pp. (87 refs.).
Kinstler, T. J. (1975). Galvanized steel-Added cost or capital investment. Mater.
Performance, (10), 3440.
Kinstler, T. J. (1989). Research update and survey of hot dip galvanized steel in the
pulp and paper industry. TAPPI J. , 72(10), 139-142.
Kinstler, T. J. , Bennett, D. C., and Allen, C. (1989). Problem solving forum. J . Prot.
Coat. Linings, 6(8), 9, 10, 12, 13, 80, 82, 83.
Kiryakov, G. Z., and Brandt, A. G. (1969). Influence of temperature on the inhibiting
effect of some acetylenic derivatives during the corrosion of zinc in acidic sulfate
solutions. Tr. Inst. Khim. Nauk Akad. Nauk Kaz. SSR, 24, 41-49 (in Russian).
Klas, H. , and Steinrath, H. (1956). Die Korrosion des Eisens und ihre Verhiitung,
Verlag Stahleisen, Dusseldorf (in German), 504 pp. This book includes discussion
of the corrosion of galvanized steel.
385t-391t.
References 479
Kleingam, J.-P. Costs and duration of protection of the principal corrosion
protection schemes for steel structures. EGGA General Assembly, Florence, pp.
Kleingarn, J.-P. Case history: Charolais Express. Feuerverzinkung.
Beratung Feuerverzinken, Germany, (in GermadDutch).
Kleinschmidt, and Schroder, Auswertung der Naturversuchtie iiber das Ver-
hulten von Metall- und Kunstoffiberziigen sowie Rostschutzenstrichen fur den Stahl-
Wasserbau (Evaluation of natural tests on the performance of metallic and decora-
tive coatings for steel use in the water industry), Information sheets of Bundesanstalt
fur Wasserbau (in German), pp.
Klinge, R. Protection of road bridges in Norway by metal spraying. Sympo-
sium on the Protection of Steel Structures by Metal Spraying, Association of Metal
Sprayers, U.K. Paper 5, 5 pp.
Klinge, R. Metal spraying on bridge work in Norway. 7th h r . Metal Spraying
Conf., London, Welding Institute, Abingdon, Cambridge, Paper pp.
Klinge, R. Sprayed zinc and aluminium coatings for the protection
tural steel in Scandinavia. 8th Int. Thermal Spraying Conf., American Welding
Society, Miami, FL, pp.
Kloetz, G. Verhalten von hachbehandelten Zink- und Cadmiumiibeniigen auf
Stahl in Gegenwart von Pressmassen auf Phenolharzbasis (Behavior of post treated
zinc and cadmium coatings on steel in the presence of phenolic-resin based mold-
ings). Galvunotechnik, 54( IO), (in German).
Kloetz, G. Das Korrosionsverhalten von Cadmium- und Zinkiiberziigen bei
Einwirkung von Ammoniak, Essigsaure, Phenol und Formaldehyd (Corrosion be-
havior of cadmium and zinc coatings as affected by ammonia, acetic acid, phenol
and formaldehyde). Galvanotechnik, (in German).
Knotkova-Cermakova, D., and Vickova, J. Eingenschaften von Elektrolyt-
Filmen bei der atmosphiirischen Korrosion (Properties of electrolyte films formed
during atmospheric corrosion). Werkst. Korros, 21(1), (in German).
Knotkova-Cermakova, D., and Vickova, J. Corrosive effect of plastics, rub-
ber, and wood on metals in confined spaces. Br. Corros. J . ,
Knotkova, D., Hasil, F., and Barton, K. Racionalni ochrana ocelovych
konstnikci proti atmosfericke korozi (Efficient protection of steel constructions from
atmospheric corrosion). Strojirenstvi, 25( IO), (in Czech).
Knotkova, D., Gullman, J . , Holler, P., and Kucera, V. Assessment of
corrosivity by short-term atmospheric field tests of technically important materials.
9th In?. Corrosion Conf., Toronto, 3,
Knotkova, D. Atmospheric corrosivity classification. Results of the interna-
tional testing programme ISOCORRAG. Paper presented to International Corro-
sion Conf., Houston, TX, typescript, pp.
Knuckley, P. J. , and Dempster, N. S. Development of zinc alloy anodes for
cathodic protection of marine structures. Australasian Corros. Eng., 7,
Kohashi, T., and Fujii, T. Weathering tests on defective area of hot dip
galvanized panel. 1st Asian Pacific Gen. Galv. Conf., pp.
Kohl, F. W. (Galvalume: surface-refined steel sheeting with aluminum-zinc
References
coating-A new material with outstanding characteristics.) Blech Rohre Profile,
30(5), 1 (in German).
Kohler, W. (The chemical corrosion behavior of non-ferrous metals towards
alkali mono- and alkali hydrogen fluorides.) Werkst. Korros., (in
German).
Kohnen, J. Langlebiger Korrosionsschutz mit anorganichen Zinksilikaten
(Long-term corrosion protection with inorganic silicates). Verfahrenstechnik,
(in German); Anti-Corros. Methods Muter.,
Kokorin, G. A. Studying the diffusion layer of galvanized sheet metal. Sb. Tr.
Tsent. Nauch Issled. Inst. Chern. Met. (in Russian).
Kopczynski, C. Korozja stopow miedzi, aluminium i cynku W wodach
podskornych kopalni siarki W machowie (Corrosion of the alloys of copper, alumi-
num and zinc in the subsurface waters at the Machow sulfur mine). Ochr. Prsed
Koroz., 15(10), (in Polish).
Kopczynski, C. Korozja stopow miedzi, aluminium i cynku W atmosferze
przemyslu siatkowego (Corrosion of the alloys of copper, aluminum and zinc in the
atmospheres of the sulfur industry). Ochr. Przed Koroz., (in
Polish).
Korovin, N. V. Alloy deposition. Theory, practice and current applications.
Part Alloys for protective and/or decorative applications. Electroplut. Mer.
Finish.,
Koster, W. The ternary system iron-silicon-zinc. Metalfurgiu 90,
Kotegawa, J. , Watanabe, K., Miura, I., Fukuda, S. , and Irie, T. Effects of
iron2+ or iron3+ in the bath on Zn-Ni alloy coating. Nisshin Steel Tech. Rep.
(MA (in Japanese).
Kovarskii, N. Ya., Tolstokonev, A. P., and Bogdanovich, V. B. Zushch.
Metuff., 24(5), (MA (in Russian).
Kraak, L. A. Introduction of hot dip galvanizing the steel dyke construc-
tion of the Netherlands Water Associations. Feuer-verzinken (Germany) or Ther-
misch Verzinken (Netherlands), 5(1), (in DutchGerman).
Kraak, L. A. Cathodic protection with galvanized objects. Thermisch
VerzinkenlFeuerverzinken, 13( (in DutchGerman).
Kraak, L. A. Galvanized auction market building in use for years.
Thermisch VerzinkenlFeuerverzinken, (in DutchGerman).
Krochmal, F., and Puacz, A. Influence of anions on the anodic behaviour of
zinc in m KCI. Zeszyry Nuuk, Uniw. Pozmuniu. Mat. Fiz. Chem.,
(in Russian); Chem. Abstr., d
Kronshorst, B. (The corrosion behavior of non-ferrous metals in powder
coating.) Merull, (in German).
Kruse, C.-L. Uber das Korrosionsverhalten von Zink und feuerverzinktem
Stahl in erwbntem Wasser. (On the corrosion resistance of zinc and galvanized
steel in warm water.) Werkst. Korros., (in German).
Kruse, C.-L., Schmitt-Thomas, K. G., and Grafen, H. Korrosionsverhalten
von feuerverzinktem Stahl in erwbntem Wasser (Corrosion behavior of hot gal-
References 481
vanized steel in warm water). Werkst. Korros., 34(1 l), 539-546 (in German). Also
as GAV Report 87, 12 pp.
Kruse, C.-L. (1984). Komsionsverhalten von ungeschutztem und feuerveninktem
Stahl bei dei Lagerung von Heizol (Corrosion behavior of unprotected and hot dip
galvanized steel for fuel oil storage). Werkst. Korros., 35(4), 15CL156 (in German).
Also as GAV Report 95.
Kruse, C.-L., Friehe, W., and Schwenk, W. (1986). Field tests on galvanized steel
tubes in drinking waters. Werkst. Korros., 37, 12-23 (in German).
Kruse, C.-L., Friehe, W., Schulze, M., and Schwenk, W. (1987). Field corrosion
tests on hot dip galvanized steel pipes. Assessment of long-term data. Werkst.
Korros., 38, 229-233 (in German).
Kucera, V., Haagenrud, S., Atteraas, L., and Gullman, J. (1987a). Corrosion steel
and zinc in Scandinavia with respect to the classification of the comsivity of
atmospheres. ASTM STP 965. ASTM, Philadelphia, pp. 264-281.
Kucera, V., Knotkova, D., Gullman, J. , and Holler, P. (1987b). Corrosion of
structural metals in atmospheres with different corrosivity at eight years exposure in
Sweden and Czechoslovakia. 10th Cong. Metallic Corrosion, Madras, Vol. 1.
Session 1, 167-17.
Kudreiko, N. A., and Agarev, Yu. A. (1985). Influence of aluminum additions on the
corrosion resistance of a zinc diffusion coating. Sov. Mater. Sci., 21(3), 289-290
(in English).
Kudreiko, N., and Agarev, Yu. (1991). Protective properties of Zn-AI coatings in a
mine environment. Alumozinc Symposium, Dnepropetrovsk, typescript (in Rus-
sian), 6 pp.
Kuhn, N., Humayun, and Fisher, T. W. (1984). The New Prepainted Galvalume
Sheet. BIEC International, Bethlethem, PA, 25 pp.
Kunitake, H. (1985). Continuously Coated Steel Sheet-The Challenge the Automo-
tive Industry. International Iron and Steel Institute, Brussels, 13 pp.
Kurimoto, T., et al. (1983). (Painted zinc-nickel trials.) 2nd Automotive Corrosion
Prevention Conf., pp. 237-244.
Kurimoto, T. (1988). Improving the performance of aqueous cutting fluids by
galvanic+athodic protection of cutting tools. Wear, 127(3), 241-251.
Kurr, G. W. (1967). Sacrificial zinc ribbon anodes. Mater. Prot.. 6(4). 30-33.
Kurr, G. W. (1969). Sacrificial anode applications in ships. Mater. Prot., 8(1 l),
, ,
19-23.
Kurr, G. W. (1973). Zinc ribbon anodes underground. Mater. Prot., 12(2), 17-22.
Kurr, G. W. (1979). Zinc anodes-Underground uses for cathodic protection and
grounding. Mater. Performance, 18(4), 34-41.
Kutzelnigg, A. (1957). Massnahmen zur Verhinderung der Korrosion durch Auf-
bereitung der Atmosphare (Reducing corrosion by purifying the atmosphere).
Werkst. Korros., 8, 492498 (in German).
Kuznetsov, Yu. I. , and Podgornova, L. P. (1985). Protection of zinc in alkaline
phosphate solutions by means of benzimidazoles. Prot. Met., 21(3), 408-410
(translation of Zashch. Metall.).
482 References
Laberenz, A. (1965). Reaktionen au Schicht-Defekten in Zinkstaubsystemen (Reac-
tions at defects in zinc dust coatings). Farbe Lack, 71(10), 81 1-822; 71(1 l), 887-
893; 71(12), 1019-1026 (in German).
Laberenz, A. (1970). Zinkstaub-Anstrichmittel (Zinc dust paint coatings). Meral-
lobe$uche, 24(7), 238-242 (in German).
Lainer, V. I. , and Sedlow, I. A. (1971). Russ. Eng. J. , U, 77-79.
Lamberigts, M., Leroy, V., and Goodwin, F. E. (1991). Corrosion resistance and
ductility: Extent and origins of the advantage of Galfan over classical galvanizing.
3rd Int. Zinc Coated Sheet Conf., EGGNATEG, Paper SlD17-13.
Lamberigts, M., and Leroy, V. (1992). Influence of substrate-coating interactions on
the microstructure and properties of Galfan products. Galvatech Verlag
Langelier, W. F. (1946). J . Am. Water Assoc., 38, 169.
Lapotyshkina, N. P., and Sazonov, R. P. (1982). Water Treatment and Hydrochemi-
cal Regime of Thermal Networks. Energoizdat, Moscow, pp. 110-1 17.
LaQue, F. L., and Copson, H. R., Fds. (1963). Corrosion Resistance of Metals and
Alloys, 2nd ed. Reinhold, New York, Chapter 9, pp. 223-247.
Larrabee, C. P., and Ellis, B. (1959). Corrosiveness of various atmospheric test
sites as measured by specimens of steel and zinc. Proc. ASTM, 59, 183-201.
Larrabee, C. P., and Ellis, B. (1961). Corrosiveness of various atmospheres.
Mater. Res. Stand., I, 977.
Lechner-Knoblauch, U., and Heitz, E. (1987). Corrosion of zinc, copper and iron in
contaminated non-aqueous alcohols. Electrochim. Acta, 32(6), 901-907.
Leclerq, M., and Bensimon, R. (1976). Combined paper: New zinc-based alloy for
metallizing. 8th Int. Thermal Spraying Conf., American Welding Society, Miami,
FL, pp. 417-429.
Leclerq, M. (1978). The electrochemical determination of the proportions of metal and
metal oxide in zinc-base coatings obtained by hot spraying. Anti-Corros. Methods
Mar., 25(a), 5-11.
Leclerq, M., and Sempels, R. E. (undated) Long term protection of steel by painted
zinc-base coatings obtained by hot spraying. Vieille Montagne, Angleur, Belgium
Report, 18 pp.
Ledion, J. , and Talbot, J. (1969). Condition for reversal the direction of the
galvanic current in iron-zinc corrosion piles. Compr. Rend. C., lo), 916-918
(in French).
Legault, R. A., and Pearson, V. P. (1978). Kinetics of the atmospheric corrosion of
galvanized steels. Atmospheric Factors Affecting the Corrosion of Engineering
Metals. ASTM STP 646. ASTM, Philadelphia, pp. 83-96.
Legault, R. A., and Pearson, V. P. (1978). Atmospheric corrosion in marine environ-
ments. Corrosion, 34( 12), 433437.
Legrand, J. (1970). Essais de corrosion en ambience naturelles de revetements dacier
par metallisation au pistolet et applications de peintures (Corrosion of painted, metal
sprayed steel by natural environments). Corrosion, 18(5), 301-312 (in French).
References 483
Leroy, R. L. Chelate inhibitors for zinc and galvanized products. Corrosion,
Li, Y., Liu, C., Lu, J. , andChen, J. Effectsof aluminium andRE(rareearths)
on the corrosion resistance and structure of the hot dip Zn-5% AI-RE alloy coated
steel sheet. Galvatech Iron and Steel Institute of Japan, pp.
Liddell, R. W. Inhibition of the corrosion zinc in acid solutions by
substituted amides. U.S. Patent
Liddiard, E. A. G., Kenworthy, L., and Frapwell, S. A. British Patent
Liesegang, W. Uber die Verteilung schwefelhaltiger Abgase in freier Luft (On
the distribution of sulfurous gases in the open air). Gesundh.-lng., (in
German).
Liesegang, W. Uber den Nachweis von Verunreinigungen durch Industrie-
Abgase in der freien atmospharischen Luft (On the proof of pollution by industrial
gases of the open air). K1. Mitteil Preuss. Lundesanst. Boden-, Wasser-, und
Lufthygiene, 8, (in German).
Liesegang, W. Uber einige gesundheitlich wichtige gas- und dampfformige
Verunreinigungen der Luft. 111-Schwefel-dioxyd (About some important health
aspects gas and vapor impurities in the air. 111: Sulfur dioxide). K/. Mitteil
Preuss. Landesanst. Boden-, Wasser-, und Lufhygiene, (in German).
Likens, G. E., Wright, R. F., Galloway, J. N., and Butler, T. J. Acid rain.
Sci. Am. ,
Linder, M., and Mattsson, E. Galvanic corrosion of combinations of metals in
freshwater. 7th Scandinavian Corrosion Congr., Trondheim. Royal Norwegian
Council for Scientific and Industrial Research, Oslo, pp.
Linder, M. The Influence of Environmental Acidification on the Atmospheric
Corrosion of Zinc. Swedish Corrosion Institute, Stockholm, Report to ILZRO
Lodder, L. J., and Baumgarten, S. Corrosion tests with zinc alloy die
castings in damp gypsum plaster. Metallurgia,
Liihberg, K. Series of articles: Investigation of corrosion by steandmoist
airlwater on zinclaluminum alloys containing lead. in Metal/ forschung.
Liihberg, K. Zur Frage der interkristallinen Korrosion von Zinklegierungen
(The intercrystalline corrosion of zinc alloys). Z. Metallkd.,
(in German).
Lahberg, K. Beitrage zur Frage der Wasserdampfkorrosion von
Zinklegierungen. V. Eine Beobachtung iiber die Bildung von Wasserstoff. VI. Uber
die Temperaturabhangigkeit der Wasserdampfkorrosion (Investigations of corrosion
by stean-dmoist aidwater on zinc/aluminium alloys containing lead-A series of
articles). Metallforschung, (in German).
Liihberg, K. Beitrage zur Frage der Wasserdampfkorrosion aluminiumhaltiger
Zinklegierungen. IV: Uber den Einfluss von Blei. Metallforschung,
(in German).
PP.
484 References
Longo, F. N., and Durmann, G. J. (for AWS Committee on Thermal Spraying).
(1974). Corrosion prevention with thermal sprayed zinc and aluminum coatings.
Welding J . , 53(6), 362-370.
Longo, F. N., and Durmann, G. J. (1976/1978). Corrosion prevention with thermal-
sprayed zinc and aluminum coatings. Atmospheric Factors Affecting the Corrosion
Engineering Metals. ASTM STP 646. ASTM, Philadelphia, pp. 97-1 14.
Loria, E. A., and Bush, G. W. (1974). Fatigue properties of galvanized steel and hot
rolled steel before and after exposure to salt spray. Sociefy of Automotive Engineers,
Automotive Engineering Congress, Detroit, Paper 740034, 11 pp.
Lorking, J. F., and Mayne, J. E. (1961). Some observations on the corrosion of
zinc. 1st Int. Congress Metallic Corrosion. Butterworths, London, pp. 144-146.
LovreEek, B. and Petric, N. (1971). Die passivierende Wirkung von SiO, auf Zink
(Passivating effect of SiO, on zinc). Werkst. Korros., 22(12), 1019-1022 (in
German).
Lowe, T. A. (1961). Aluminum, aluminized steel and galvanized steel in severe
coastal environments. Corrosion, 17, I77t-180t.
Lucas, R. W. (1971). Surface oxidation of zinc in the range 300 to 400C (573 to 673
K). J . Inst. Met., 69-76.
Ludwig, N. (1956). Priifung und bewertung von gasgeschweissten Verbindungen aus
Zink und Zinklegierungen (Testing and assessment of gas welded zinc and zinc
alloys). Metallkd., 47, 473477 (in German).
Lunde, L., and Nyborg, R. (1989). Stress-corrosion cracking of carbon steel in
ammonia. Mater. Performance, 12), 29-32.
Lyman, J. , and Fleming, R. H. (1940). J . Mar. Res., 3, 134.
Lyublinskii, E. Y. (1970). Phase composition and anode behaviour of zinc protective
alloys used in seawater. Prof. Met., 6(4), 399400 (translation of Zashch. Metall.).
Maass, W. B. (1987). Hot dip galvanizing: Preferably protecting also the zinc coating.
Metalloberfliiche, 41(4), 166-168 (in German).
Mabuchi, M., et al. (1992). Properties of Zn-N-Cr coated steel sheets. Galvutech
Verlag Stahleisen, Diisseldorf, Germany, pp. 265-270.
MacDougall, J. (1969). Microbiological effects in the corrosion of metals. Australian
Corros. Eng. , 13, 13-16.
Machu, W. (1954). Corrosion of metals and metal coatings in tropical and subtropical
climates. Werkst. Korros., 5, 395-398 (in German).
Mackay, W. B. (1974). Deep water testing of sacrificial anodes. Mater. Performance,
13(8),
Mackay, W. B. (1975). Deep water testing sacrificial anodes. Part II. Corrosion
75, Toronto, National Association of Corrosion Engineers, Houston, TX, Paper 18,
7 PP.
Mackowiak, J. , and Short, N. R. (1976). Corros. Sci., 16, 519-528.
Mackowiak, J. , and Short, N. R. (1977). Effect of chemical potential of zinc on
composition and nature of interaction layers formed during reaction between solid
iron and zinc vapor at 793 K. Met. Sci., November, pp. 517-522.
References 485
MacLennan, A. D. (1961). Materials of construction for chemical plant-Zinc coat-
Mackod, G. D. (1966). Alkyl silicate based coatings. Mater. Prot., 5(8), 4547.
Maconachie, J. E. (1934). Anomalies observed in the rate of corrosion of zinc. Trans.
Electrochem. Soc., 66, 213-221; Met. Ind., 45, 418-420.
Maeda, M., et al. (1988). Use of a vacuum sealing system in a continuous line for
vapor deposition of zinc. Rev. Metall., 85(4), 341-349 (in Spanish).
Makimattila, S. M., and Ristolainen, R. (1985). On the structure of a Zn-5% hot
dip galvanized coating. 1st Int. Conf. Zinc Coated Steel Sheet, ZDA, London, pp.
Mallory Battery Co. Ltd. (1967). Mixture and method for the manufacture of porous
electrodes for electric cells similar structures. French Patent 1,523,090.
Mallory Batteries Ltd. (1968). Improvements in or relating to primary cells. British
Patent 1,188,611, 4 pp.
Manning, M. I. (1988). Effects of air pollution on structural materials. Met. Mater.,
4(4), 213-217.
Mansfield, F., and Kenkel, J. V. (1975). Laboratory studies of galvanic corrosion. I:
Two-metal couples. Corrosion, 298-302.
Mansfield, F., and Tsai, S. (1979). Laboratory Studies of Atmospheric Corrosion of
Steel and Zinc. Corrosion 79, Atlanta, NACE, Katy, TX, Paper 208, 1 1 pp.
Marassi, R. (1969). Standard electrode potentials of Cd/Cd(II), InlIn(III), Pb/Pb(II),
Tfli(I), Zn/Zn(II) in molten alkali acetates. J . Electroanul. Chem. Interfucial
Electrochem., 22(2), 215-219.
Markovic, T., Dugi, Z., and Rubinic, L. (1958). Influence of inhibitors on the
dissolution of zinc. Werkst. Korros., 9(1), 17-20 (in German).
Marshall, T., and Neubauer, L. G. (1955). Corrosion of aircraft structural materials by
agricultural chemicals. Corrosion, 11(2), 44-52, 84-92.
Martin, H., and Rauen, A. (1975). Studies on the Behavior of Galvanized Reinforcing
Steel in Concrete. Gemeinschaftsausschuss Verzinken eV, Dusseldorf, Report 68,
pp. 61-77 (in German).
Martin, H. C. (1982). Acid rain: Impacts on the natural and human environment.
Mater. Performance, 21( l), 36-39.
Martin, K. G. (1973). Atmospheric corrosion of metal cladding in Papua-New Guinea
and Melbourne. Australasian Corros. Eng. , 17(5), 19-28.
Masuhara, K., and Kumon, F. (1989). Corrosion resistance of prepainted Zn4%
AI-0.1 % Mg mischmetal alloy coated steel sheets. Galvatech 89, Iron and Steel
Institute, Japan, pp. 596-602.
Mathewson, C. H. , Ed. (1959). Zinc, Reinhold, New York. Zinc coatings (J. L.
Kimberley et al.), pp. 453-523.
Matsuda, S. T., Tanaka, T., and Tanikawa, K. (1970). Pinhole test for electro-
galvanized steel sheet. Plating, 57(6), 629-630.
Matsui, E. S. (1974). Performance of fencing materials in a marine atmospheric
environment. Mater. Performance, l), 17-19.
ings. Chem. Proc. Eng. , 42, 535.
Bll-B16.
486 References
McKittrick, D. P. Reinforced earth: Application of theory and research to
practice. Reprint from Ground Eng. , 12 January, pp.
McKittrick, D. P., and Darbin, M. World-wide development and use of
reinforced earth structures. Reprint from Ground Engineering by Reinforced Earth
Co., Telford, U.K., pp.
Mead, W. J., Ed. The Encyclopedia of Chemical Process Equipment. Rein-
hold, New York. Zinc (E. W. Horvick), pp.
Mears, R. B. Combating rust with metallic finishes. BellLab. Rec., 11,
144.
Meetham, A. R. Atmospheric Pollution. Its Origins and Prevention. Per-
gamon Press, London and New York, pp.
Melton, K. N., and Edington, J. W. Effect of increasing aluminum content on
the boiling water corrosion resistance of superplastic Zn-AI alloys. Corrosion,
Memmi, M., Mucchino, U., and Giardetti, G. Properties of prepainted
Lavegal. 2nd Int. Conf. Zinc-Coated Steel Sheet, ZDA, London, Paper SC7/1-11.
Mercer, P. D. Factors which affect the alloy growth rate in galvannealed
coatings. Galvatech Verlag Stahleisen, Diisseldorf, Germany, pp.
Merkulov, A. V., and Flerov, V. N. Effect of sulfide-containing additions on
the corrosion of zinc in alkaline solutions. Zushch. Metall., 5(1), (in
Russian); Zinc Abstr.,
Metallurgic Hoboken Overfelt (MHO) Zinc-indium alloy for cathodic protec-
tion of inner walls of tanks against corrosion and a method for its manufacture.
German Patent pp (in German).
Middlemiss, A., and Hailwood, J. Vacuum deposition of metallic coatings
onto steel wire. Wire Id.,
Mikhailovskii, Yu. N., Panchenko, Yu. M., and Sokolov, N. A. Recom-
mended reference data on rates of atmospheric corrosion of metals in various
climatic regions of the USSR. Prot. Met., 21(5), refs.) (translation of
Zashch. Metall.).
Ministry of Agriculture, Food, and Fisheries and Geographical Map of
Great Britain Depicting Corrosion Rates of Zinc, 1 sheet each. New edition,
Relative Values of Acid Deposition in the United Kingdom 1986-1 991, ADAS Farm
Buildings Research Team, Coley Park, Reading.
Ministry of Defence Defence Guide for the Preservation from Corrosion of
Cadmium and Zinc Plating by vapours from Organic Materials. HMSO, London,
Minnesota Mining and Manufacturing Co. Zinc-containing algicidal surfacing,
Miskidchjan, S. P. Inhibiting effects with some halogen salts with substantial
Mitsui Mining and Smelting Co. Supergalva (SG) Technical Report, pp.
Miura, N. (Zinc-nickel for car panel tests.) Trans. Iron Steel Inst. Japan,
Mohler, J. B. The corrosivity of the atmosphere. Plating, 61(1),
6 PP-
method and granules. U.S. Patent pp.
ammonia. Saschtschita Metall., (in Russian).
(in Japanese).
References 487
Mokhamed, K. E., and Kukharev, B. S. (1987). Investigation of corrosion properties
of alumozinc coatings of samples of steel. Metallurgiya; (21), 99-103 (in
Russian).
Moore, J. F. A. (1981). The performance of cavity wall-ties. Building Research
Station, Report CP3/81, 26 pp.
Mor, E. D., and Beccaria, A. M. (1973). Inhibitory action of acrylonitrile on the
corrosion of zinc in seawater. Br. Corros. J . , 25-27.
Mor, E. D., Beccaria, A. M., and Poggi, G. (1974). Behavior of zinc in seawater in
the presence of sulfides. Br. Corros. J . , 9(1), 53-56.
Mor, E. D., and Beccaria, A. M. (1975). Effect of temperature on the corrodibility of
copper and zinc in synthetic seawater. Corrosion, 8, 275-279.
Morbe, K., and Werner, H. (1978). Korrosionsschutz fur wassefuhrende Anlagen in
Haushault und Indusrrie (Corrosion protection of domestic and industrial water
supply systems). Zentralstelle fur Korrosionsschutz, Dresden, Germany, (in Ger-
man), pp.
Morcillo, M., and Feliu, S. (1983). La corrosion de metales en las grandes
ciudades. Efecto de la contaminacion atmosferica (Corrosion of metals in large
cities. Effect of atmospheric pollution). Rev. Metall., CENIM, 19(2), 61-66 (in
Spanish).
Morcillo, M., Ruiz, J. L., and Matilla, M. G. (1989). Environmental factors responsi-
ble for the unusual corrosion of galvanized steel. Werkst. Korros., 40(11), 668-
673.
Morcillo, M., and Ruiz, J. L. (1991). Severe corrosion of galvanized steelwork in a
marine atmosphere: Possible control through duplex systems. 16th Int. Galv. Conf,
EGGA, London, GH2/1-12.
Morgan, J. (1975). Monitoring and control of cathodic protection of offshore struc-
tures, Corrosion 75, Toronto, NACE, Paper 20, 8 pp.
Morgan, J. H. (1959). Cathodic Protection. Leonard Hill (Books) Ltd., London,
325 pp.
Morrow, H. (1987). Economics of thermal spraying for the long-term corrosion
protection of steel structures. J . Prot. Coat. Linings, 4(1), 39-47.
Morrow, H. (1989). Zinc and zinc-aluminum alloy thermal spraying for the corrosion
protection of steel bridges, American Welding Society Annual Meeting, Washing-
ton, DC.
Morrow, H. (1991). Impressed current protection of rebar in concrete using zinc. Talk
to U.K. Association of Metal Sprayers, typescript, 3 pp.
Motohashi, Y., and Shibata, T. (1986). Hot dipping of steel sheet into AI-Zn eutecoid-
base superplastic alloy and the properties of the hot dip coated steel. J . Met. Finish.
Soc. Japan, 625-630 (in Japanese).
Mottram, K. G., and Petchell, F. M. (1986). Surface preparation for painting new
galvanized structural steel. J . Oil Colour Chem. Assoc., (4), 89-94.
Mudd, C. (1943). Experience with zinc anodes. Petroleum Industries Electrical
Association Annual Meeting, Houston, TX.
Muller, D., Dautel, R., et al. (1971). Untersuchungen uber die Korrosion veminkter
Stahlrohre durch Leitungswasser zwecks Verbesserung ihrer Haltbarkeit: Teil 3
References
(Investigation of the corrosion of zinc-coated steel pipes by tap water in order to
improve their stability: Part Chemisch-Technisches Priifamt Report (in
German), Stuttgart, pp.
Muller, D., Dautel, R., et al. Zur Korrosion verzinkter Eisenwerkstoffe durch
naturliche Brauchwasser (Corrosion of zinc-coated iron materials by natural indus-
trial waters). Werkst. Korros. (in German).
NACE. Control of External Corrosion on Underground or Submerged Metal-
lic Piping Systems. Recommended Practice RP-01-69. NACE, Houston, TX.
NACE. Mitigation of Alternating Current and Lightning Effects on Metallic
Structures and Corrosion Control Systems. Recommended Practice RP-01
NACE, Houston, TX.
Nagasaka, H., Uchida, S. , et al. Weathering test on zinc, aluminum and zinc-
aluminum alloy sprayed coatings for days. Int. Metal Spraying Conf., poster
Nagel Soepenberg, E. Hot dip or electrodeposited coatings for automotive
application. Galvatech 92, Verlag Stahleisen, Diisseldorf, Germany, pp.
Nakamura, Y., Kawakami, H., Satoh, K., Fukinhara, H., and Urai, M.
TRIGALVA steel wire with enhanced corrosion resistance and formability. Kobe
Res. Dev., (MA (in Japanese).
paper, P.
National Association of Corrosion Engineers (U.S.). See NACE.
Neuhaus, A., and Gebhart, M. Kristalline Korrosionsschutz und Korro-
sionsschutzschichten auf Metallen und ihre Beziehung zur Epitazie (Crystalline
corrosion preventives and coatings on metals and their relationship with epitaxy).
Werkst. Korros., 17, (in German).
Nicholls, J. H. The hot dipped galvanized coating and corrosion resistance.
Corros. Technol., 6,
Nielsen, K., and Yding, F. Influence of tube quality on the corrosion of
galvanized water-carrying pipes in homes. Werkst. Korros., (in Ger-
man).
Nihon Parkerizing Co. Ltd. Erzeugung von Chromatschichten (Method for
production of chromate layers). European Patent Application EP
November (October (in German), pp.
Nissing, W. , Friehe, W., and Schwenk, W. Influence of oxygen concentra-
tion, pH value and speed of water on the corrosion of hot dip galvanized and
ungalvanized unalloyed steel tubes in potable water. Werkst. Korros.
(in German).
Nomi, R., and Fukii, S. (1991). Galvanneal and organic composite coatings. Steel
Times, 219(9),
Noranda Personal communication.
Noranda See Barnhurst, R. J., and Belisle, S.
Noranda Mines, Ltd. Thioglycolic acid polyesters. British Patent
Noranda Designing Zinc Castings for Corrosion Resistance. Noranda, Toron-
August (June pp.
to, pp.
References 489
Nowak, W. B., Bums, L. E., and Harris, V. G. (1989). Corrosion resistant ion plated
aluminum-zinc films. J . Vac. Sci. Technol., 7(3-II), 2350-2354.
Noyce, R. W. , Ostrowski, R. W., and Ritchie, J. M. (1975). Corrosion of galvanized
metal culverts. Presented at 54th Annual Meeting, Transportation Research Board,
National Research Council, Washington, DC. Available as PB 245,568 from
National Technical Information Service, Springfield, VA.
Numberger, U. (1986). Corrosion behavior of galvanized steel in contact with build-
ing materials. Werkst. Korros., 37, 302-309 (in German).
Numberger, U. (1989). Corrosion behavior of hot dip galvanized construction steels in
sand-containing soils. Werkst. Korros., 40(1), 7-16 (in German).
OConnell, J. M. (1977). Cathodic protection of hot pipeline in frozen earth. Muter.
Performance, 16(5), 13-21.
Ochai, I., and Ohba, H. (1989). Microstructure and corrosion behavior of inter-
metallic layer of coating in Zn-Alloy coated steel wire manufactured by the double-
dip process. Tetsu-To-Hugune ( J. Iron Steel Inst. Japan), 75(2), 298-305 (in
Japanese).
Ochai, I. , and Ohba, H. (1989). Influence of microstructure of coating on corrosion
resistance of Zn-al alloy coated steel wire manufactured by the double-dip process.
Tetsu-To-Hugune ( J. Iron Steel Inst. Japan), 75(2), 290-297 (in Japanese).
Odashima, H., et al. (1992). Corrosion resistance mechanism of a newly developed
organic composite coated sheet steel for automotive use. Gulvutech 92, Verlag
Stahleisen, Diisseldorf, Germany, pp. 377-382.
Oelsner, G. (1957). Korrosion metallischer Oberflachen durch Handschweiss (Corro-
sion of metallic surfaces by hand perspiration). Juhrb. Obefluchentechn., 13, 49-
58 (in German).
Oelsner, G. (1959). Die Handschweisskorrosion metallischer Oberflachen (Hand
perspiration corrosion of metallic surfaces). Gulvunotechnik, 50, 77-84 (in Ger-
man).
Okorafor, E. (1986). (Corrosion under sheltered and unsheltered sites in Awka.)
Corros. Prevent. Control, 33(3), 64-67.
Okumura, K., et al. (1992). Aging mechanism of paint adhesion of electrodemsited
chromate films. Gulvutech 92, Verlag Stahleisen, Dusseldorf, Germany, pp. 255-
259.
M. J. (1976). Einstatz feuerverzinkten Bewehrungsstahls bei der Olgewinnung
in der Nordsee (Introduction of hot dip galvanized reinforcing steel for oil extraction
in the North Sea). Verzinken 5(1), 5-7 (in German).
Panchenko, Yu. M., Shuvakhina, L. A., and Mikhailovskii, Yu. N. (1985). Depen-
dence of rates of atmospheric corrosion of metals on climatic considerations in the
Far-Eastern USSR. Prot. Met., 20(6), 641-651 (translation of Zashch. Metull.).
Parker, A. K. (1960). Zinc Coatings: Survey, ZDA, London, 51 pp. (Reprint from
Met. Finish. J. , June 1959-0ctober 1960.)
Parker Co. (1967). Method and composition for the treatment of metals. French Patent
1,538,327.
Paul, N. (1981). Zinc anode versus aluminum anode for cathodic protection. Lead,
490 References
Zinc and Cadmium: Retrospect and Prospect, International seminar, New Delhi,
Indian Lead Zinc Information Centre, pp.
Paul, W. (1956). Systematische Versuche zur Priifung der Korrosionsverhindernden
Wirkung von VPI auf eine Reihe von gebeuchlichen Metallen (Systematic re-
searches into the corrosion-inhibiting properties of VPI on a number of common
metals). Werkst. Korros. 7(4), 189-198.
Peabody, A. W. (1955). Zinc anodes perform satisfactorily in sea water piling
installations. Corrosion, 11(6), 92.
Peabody, A. W. (1976). Special challenge: Cathodic protection on trans-Alaska
pipeline. Pipeline Gas J. , December.
Peabody, A. W. (undated). Control of pipe line corrosion. NACE, Houston, TX.
Peart, J . F. and Steinicke, A. R. (1973). Some applications of cathodic protection to
structures immersed in water and sewage. Australasian Corros. Eng. , I7(8), 17-25.
Pellowe, E. F. (1952). The corrosion resistance of galvanized coatings with special
reference to galvanized window frames. 1st Int. Galv. Conf., Copenhagen, July
1950, ZDA, London, pp. 121-124.
Pelzel, E. (1954). (Intercrystalline corrosion of Zn-AI-Cu alloy.) Metall, 8, 169-173
(in German).
Pelzel, E., and Schneider, H. (1961). Uber die Walzwerkstoffe auf Feinzinkbasis und
die Legierung ZnCulTi. (Rolling materials based on pure zinc and the Zn-Cul-Ti
alloy). Metall, 14, 782-784 (in German).
Pelzel, E. (1977). Das Korrosionsverhalten von Zinkwalslegierungen in Freigelande
und im Damptest (The corrosion behavior of rolled zinc alloys in the open air and in
the steam test). Metall, 31(9), 978-981 (in German).
Perkins, J. , and Bornholdt, R. A. (1977). The corrosion product morphology found on
sacrificial zinc anodes. Corros. Sci., 17, 377-384.
Petric, C. (1971). Interactions between zinc and gold in liquid tin. Phys. Chem.
Liquids, 2(3), 181-191.
Petrov, L. (1968). Effect of the etching of zinc alloys in nitric acid on their strength
properties. Fiz. Khim. Mekh. Mater., 4(6), 737 (in Russian).
Petrov, L., and Savitskaya, 0. P. (1970). Inhibition of the dissolution of steel and zinc
in nitric acid. Fiz. Khim. Mekh. Mater., 6(1), 113-114 (in Russian).
Petrow, C., Martschewska, M., and Entschewa, M. (1970). Oxydation von Zink in
alkalischer Liisung durch Wechselstmm von 50Hz. 2: Liisungen von
Natriumhydroxyd und Natriumchlorit (Oxidation of zinc in alkaline solution with an
alternating current of 50 Hz. Part 2 Solutions of sodium hydroxide and sodium
chlorite). Metalloberfuche, 24(7), 243-246 (in German).
Petzold, I . , and Efer, K. (1977). Korrosionsverhalten der Schnittflachen von feuer-
verzinktem organisch beschichtetern Stahlband (Corrosion behavior of cut surfaces
of hot dip galvanized organically coated steel strip). Neue Hiitte, 22( lo), 584-585
(in German).
Pisigan, R. A., and Singley, J. E. (1985). Effect of water quality parameters on the
corrosion of galvanized steel. J . Water Works Assoc., 111(6), 76-82.
Plucinski, J. (1972). Badania korozjne powlok metalowych W naturalnych warunkach
References 491
dolowych kopain wegla kamiennego (Corrosion tests on metallic coatings under the
natural underground conditions encountered in bituminous coal mines). Ochr. Pned
Koroz., 15(5), 128-133 (in Polish).
Pollard, W. A.. and Edwards, J. 0. (1976). Effects of Cu, Mg, Pb and Sn on the
corrosion of sand-cast Zn- 12% AI alloy in wet steam. Physical Metallurgy Research
Laboratories, CANMET, Department of Energy, Mines and Resources, Ottawa,
Canada, Report MRPIPMRL-76-1 l(J), 31 pp.
Pollard, W. A., and Edwards, J. 0. (1977). Combined effects of Pb and Sn additions
on the corrosion of Zn-12% AI-l % Cu-0.02% Mg sand-cast alloy in wet steam.
Metall. Trans. B. 8B(4), 597-599.
Porter, F. C., and Walden, G. K. (1967). Inspection of sprayed zinc coatings. 2nd
Conf. Institute of Welding, Spraying and Plastic Coatings Div., London. Reprint,
ZDA, London, 7 pp.
Porter, F. C. (1969). Painting of Zinc Surfaces. ZDA, London. 8 pp. (Reprinted from
Er. Corros. J . , 4, 179-186.)
Porter, F. C. (1978). Comparative Costs of Protecting Steel. ZDA, London, IO pp.
Porter, F. C. (1985). Galvanized reinforcement for concrete. 1st Int. Conf. Deteriora-
tion and Repair of Reinforced Concrete in the Arabian Gulf. Construction Industry
Research and Information Association (U.K.), pp. 193-201.
Porter, F. C. (1991). Zinc Handbook-Properties, Processing and Use in Design.
Dekker, New York, 629 pp.
Pourbaix, M. (1948). Corrosion et protection des toitures en zinc (Corrosion and
protection of zinc roofs). Met. Corros., 24, 21-25 (in French).
Pourbaix, M. (1969). Recent applications of electrode potential measurements in the
thermodynamics and kinetics of corrosion of metals. Corrosion, 25(6), 267-28 1.
Pourbaix, M. (1970). Corrosion de tubes en acier galvanis6 utilis6s pour la distribution
deau (Corrosion of galvanized steel pipes used in water distribution systems).
CEBELCOR Rapp. Tech., 114(180) (in French), IO pp.
Powell, S. T., Bacon, H. E., and Lill, J. R. (1945). Ind. Eng. Chem., 37, 842.
Powell, S. T., Bacon, H. E., and Lill, J. R. (1946). Recent developments in corrosion
control. J . Water Works Assoc., 38(7), 808-825.
Price, C. G. (1963). Sherardizing for corrosion prevention. Corros. Prev. Control,
IO( l ) , 25-29.
Proskurkin, E. V. , and Gorbunov, N. S. (1967). Protection of pipes in the petroleum
industry. Zashch. Metall. (special issue No. 3). 372-373 (in Russian).
Proskurkin, E. V. , Gorbunov, N. S., and Lavrenko, N. A. (1968). Corrosion re-
sistance of pipes with a diffusion-coated zinc layer in welded and bent zones.
Zashch. Metall. (special issue No. 6), 730-732 (in Russian).
Proskurkin. E. V. (1969). Zinc diffusion coating of pipes refrigerators functioning
in aggressive media. Teklinol. Org. Proiz Vodstva (6), 68-69 (in Russian).
Proskurkin, E. V., Zholudev, M. D., and Gorbunov, N. S. (1970). Effect of the
method of zinc coating on the corrosion resistance of zinc coatings. Zashch.
Metall., 6(1), 54-56 (in Russian).
Proskurkin, E. V. , Gorbunov, N. S., Myasoed, S. M.. et al. (1973a). Production of
492 References
zinc diffusion coatings of uniform corrosion resistant structure. Zashch. Metall.,
9(2), 171-175 (in Russian).
Proskurkin, E. V., Gorbunov, N. S., Sukhomlin, A. E., et al. (1973b). Corrosion
resistance of heat-treated zinc coatings in a stream of seawater. Zaschch. Metall.
(special issue No. 5), 606-607 (in Russian).
Proskurkin, E. V., and Gorbunov, N. S. (1975). Galvanizing, Sherardizing and Other
Zinc Dzfision Coatings. Technicopy, Stonehouse, U.K., 261 pp.
Proskurkin, E. V., Mitnikov, I. E., Norvillo, N. Yu. et al. (1978a). Corrosion
resistance of alloyed zinc coatings in marine conditions. Zashch. Metall. (special
issue No. l ) , 4647 (in Russian).
Proskurkin, E. Mitnikov, I. E., et al. (1978). Corrosion resistance of alloyed zinc
coatings under marine conditions. Prot. Met., 14(12), 35-36 (translation ofZashch.
Metall.).
Proskurkin, E. Bakalyuk, Ya. K. L., Zholudev, M. D., et al. (1981). Problems
associated with increasing the corrosion resistance of pipes of hot water supply
systems. Water Supply Sanit. Technol. (lo), 16-18 (in Russian).
Proskurkin, E. V., Mitnikov, I. E., Norvillo, N. Yu. et al. (1983). Long life of
galvanized hot water pipes. Water Supply Sanit. Technol. (12), 13-15 (in Russian).
Putilova, S. T., Balezin, S. and Barannik. V. A. (1960). Metallic Corrosion
Inhibitors. Pergamon Press, Oxford, 196 pp.
Rabald, E. (1954). Zeitweiliger Rostschutz (Long-term rust prevention). Werkst.
Radczewski, 0. E., and Schicht, R. F. (1969). Uber die Bildung von Schutzschicten
auf verzinkten Eisen-blechen. (Formation of protective layers on galvanized iron
sheets). Elektrizitswirtscha~, 68( 17), 602-604 (in German).
Radeker, W. (1958). Das Korrosionsverhalten kiinstlich erzeugter Hartzinkschichten
(The corrosion resistance of artificially produced hard zinc layers). Metal-
IoberJuche, 12(4), 102-104 (in German).
Radeker, W. , Peters, F. K., and Friehe, W. (1961). The effect of alloy additions on
the properties of hot galvanized coatings. 6th Int. Galv. Conf., ZDA, London, pp.
238-264.
Radeker W. , and Friehe, W. (1970a). Effect of the composition and temperature of
corrosive waters on the corrosion of zinc coats with different alloying additions
deposited on steel. Stahl Eisen, l ) , 567-571 (in German).
Radeker, W., and Friehe, W. (1970b). Beobachtungen bei der atmosphirischen
Korrosion feuerverzinkter Oberflachen (Observations on the atmospheric corrosion
of hot dip galvanized surfaces). Werkst. Korros. 21(4), 263-266 (in German).
Radovici, O., and Kovacs, G. (1968). Influence of specific adsorption of halide anions
on anodic dissolution of zinc in neutral solutions. Rev. Roum. Chim., 13, 1409-
1414 (in Rumanian).
Radtke, S. F. (1986). State-of-the-art: Processing of sheet steel. 1st h r . Conf.
Galvanizing into the Next Century, SAHDGNNIMR, Pretoria, preprint, 15 pp.
References 493
Rajagopalan, K. S. , and Ramaseshan, G. (1959). Relative corrodibility of zinc and
steel in unpolluted atmospheres. J. Sci. Res. 18B, 87-89.
Rajagopalan, K. S. , Sundaram, M., and Anamalai, P. L. (1959). Corrosion of metals
at Mandapam Camp, India, Corrosion, 15, 631t-634t.
Rajagopalan, K. S. , and Ramaseshan, G. (1960). Relatve corrodibility of zinc and
steel in unpolluted atmospheres. J. Appl. Chem., 10, 493-496.
Rajagopalan, K. S. , Annamalai, P. L., and Sundaram, M. (1961). Atmospheric
corrosion on the southeast coast (of India). J. Sci. Ind. Res. 200, 226-232.
Rajagopalan, K. S. , et al. (1971). Atmospheric corrosion of metals at Mandapam
Camp, India. Br. Corros. J., 6(4), 175-183.
Rance, V. E., and Cole, H. G. (1958). Corrosion of metals by vapours from organic
materials. HMSO, London, 25 pp.
Reding, J. T., and Boyce, T. D. (1974). Cathodic protection performance of offshore
anodes at a simulated depth of 650 feet. Mater. Peflormance, 13(9), 37-40.
Reichard, E. C., and Lennox, T. J. (1957). Shipboard evaluation of zinc galvanic
anodes, Corrosion, 13(6), 68t-74t, 41Ot416t.
Reinhart, F. M. (1961). Twenty-Year Atmospheric Corrosion Investigation of Zinc-
Coated and Uncoated Wire and Wire Products. ASTM, STP 290. ASTM, Phila-
delphia, 141 pp.
Reinhart, F. M. (1976). Corrosion of Metals and Alloys in the Deep Ocean. Civil
Engineering Laboratory, Naval Construction Battalion Center, Report TR-834.
Republic Steel Corp. (1965). Zinc coated steel with vapor-deposited aluminum over-
lay and method of producing same. U.S. Patent 3,438,754, February 18, 5 pp.
Rhodes, F. H., Riedel, P. A., and Hendricks, V. K. (1934). Corrosion of metals by
phenols. Ind. Eng. Chem., 26, 533-534.
Richards, C. E. (1938). The atmospheric corrosion galvanized iron wire and its
bearing on specification testing. J. Iron Steel Inst., 137, 127P-164P.
Riecke, E., and Johnen, B. (1986). Hydrogen-induced corrosion cracking of
ungalvanized and galvanized construction steels. Werkst. Korros. 37, 310-317 (in
German).
Ried, G. (1964). Ubersicht iiber Werkstoffe fur Absperr-Organe, die bei angreifenden
Medien zu empfehlen sind. (Review of organic coating materials recommended for
aggressive media). Werkst. Korros., 15(1), 19-25 (in German).
Ritter, F. (1958). Korrosionstabellen metdischer Werkstofle (Corrosion tables for
metallic materials), 4th ed., Springer-Verlag, Vienna (in German), 290 pp.
Rizzi, K. W., Spiliotic, N. J., and Blurton, K. F. (1986). New zinc electroplate fights
both wear and corrosion. Met. Progr., 129(2), 51-54.
Robeck, H. (1972). Erfahrungen bei der Anwendung von Fluorkohlenwasserstoffen
zum Reinigen von Oberflachen (Experience with fluorinated hydrocarbons and
coatings). Jahrb. Oberfliichentech., 28, 42-50 (in German).
Roberts, C. W. (1953). Intercrystalline corrosion in cast zinc-aluminum alloys. J .
Inst. Met., 81, 301-309.
494 References
Roberts, C. W. (1961). An investigation into the causes of intercrystalline corrosion in
zinc-a1 uminum alloys. Metnllurgin. 64( 382), 57-66.
Rocha, L. A. . and Barbosa, M. A. (1991). Microstructure, growth kinetics and
corrosion resistance of hot dip galvanized Zn-5% A1 coatings. Corrosion. 47(7),
536-54 1.
Rodionova, V. A. (1957). Loslichkeit von Zink in Alkalien (Solubility of zinc in
alkalies). Uchen. Zupiski Moskaii . Gos. Ped. Irzst. V . I . Leniiiu, 99, 22 1-226 (in
Russian), Chem. Abstr., 5 4 I ) , 48b (1960).
Roebuck, A. H.. and Brevoort. G. H. (1986). Coating works costs: Computer
application and inspection. Presentation at Corrosiorr 86. NACE, Houston, TX.
Roetheli, B. E., and Cox. G. L. (1931). Effects of temperature and pH. Ind. Eng.
Chent., 23, 902-904, 1084-1090.
Roetheli, B. E.. Cox, G. L., and Littreal. W. B. ( 1 932). Effect of pH on the corrosion
products and corrosion rate of zinc in oxygenated aqueous solutions. Met. Alloys, 3.
Roikh, J . L. (1958). Chemical detection of hydrogen peroxide in the atmospheric
corrosion of metals. Zhr. Fi z. Klzim. 32. 1136-1 147 (in Russian).
Roman, M. P., and Lynch, R. F. (1989). Current automotive applications for Galfan
coated steel. Society of Automotive Engineers, Warrendale, PA, Technical Paper
890709. 12 pp.
Romanoff, M. (1957). Underground Corrosion. Government Printing Office, for
National Bureau of Standards, Washington. DC, Circular 579, 227 pp. (407 refs.).
Romanovsky. V. ( I 962). Les propriCtCs antisallissures du zinc (The antifouling
properties of zinc). Corros. Anticorros., 10, (7-8), 267-269 (in French).
Rosenfeld, I. L. (1960). Atinospheric Corrosion of Metals (in Russian). Akademiya
Nauk, SSSR. Moscow, pp. 301-304. 312, 325-326.
Ross, T. K. , Sunith, T. E., and Mansford. R. E. (1958). Korrosionsschutz von Stahl
durch gespritzte Zn- und Al-Uberzuge (Corrosion protection of steel by sprayed Zn
and A1 coatings). Werkst. Korros. 9, 755-760 (in German).
Rosset, R., and Jardy. A. (1982). Protection of galvanized steel against cold and hot
water corrosion by a pyrophosphate coating. /3th Int. Galv. Coif., ZDA, London,
Paper 12/1-3.
Royce, C. A. (1971). Alternating current problems of pipelines. AGA Transinissioiz
Coizf., Houston, TX.
Royuela, J . J. (1969). Corrosion de recubrimientos de cinc y cadmio. Effect0 del
cromatado (Corrosion of zinc and cadmium coatings: Effect of chromate treatment).
Rev. Metull., 5( 5) , 535-542, 88 refs. (in Spanish).
Ruckert, J. (1979). Influence of pH value, oxygen content and flow velocity of cold
drinking water on corrosion behavior and surface layer formation on galvanized
steel tubes. Werkst. Korros.. 30. 9-34 (in German).
Ruckert, J. , Neubauer, F., and Zietelmann, C. (1983). Infirerice of Biturninous
Roofing Materials on the Corrosion of Zinc and Galvanized Steel Gutters and Pipes.
Gemeinschaftsausschuss Verzinken eV. Diisseldorf, Report 91 (in German), 1 I pp.
73-76.
References 495
Ruckert, J.. Schulze, M.. and Schwenk, W. (1987). Discussion on time dependence
of corrosion rates in flowing waters of hot dip galvanized steel pipes. Werksr.
Ruckert, J., and Sturrbecher, D. (1988). Long term performance of galvanized steel
tubes in drinking water with different pH-value and flow velocity. Werkst. Korros. ,
39, 7-17 (in German).
Rudresh, H. B., and Mayanna, S. M. (1977). Diphenylethylarsine as an interfacial
corrosion inhibitor for zinc in perchloric acid solution. Surf. Techizol.. 6(2), 139-
144.
Rudresh, H. B., and Mayanna, S. M. ( 1979a). The synergistic effect of halide ions on
the corrosion inhibition of zinc by ii-decylamine. Corros. Sci., 19(6). 361-370.
Rudresh, H. B., and Mayanna, S. M. (1979b). Role of the thiocyanate ion in the
corrosion inhibition of zinc in perchloric acid by n-decylamine. Swf. Technol.,
Rudresh, H. B., and Mayanna. S. M. (1980). Tertiary arsines as corrosion inhibitors
for zinc in acid solutions. Werksr. Korros., 3/(4), 286-290 (in German).
Riiegg , W. ( 1970). Obei~ci'cheribehandlirrig von Zinkdruckgziss (Surface treatment of
zinc castings) (in German). Juris Druck Verlag, Zurich, 493 pp.
Ruzicka, J., and Svoboda, M. ( 1959). Konservosani kosoi:vclz ~y- obku (Protection of
metal products) (in Czech). Statni Nakladelstvi Technike Literatury, Prague. 255
PP.
Sakauchi, T., and Kunimi, H. (1989). Evaluation of the corrosion resistance of zinc-
alloy plated steel sheet applied to automobile body. Gnli~ntech 89, Iron and Steel
Institute, Japan, pp. 528-535.
Salt, F. N. , Stanners, J. F., and Watkins, K. 0. (1965). Atmospheric exposure trials
of electrodeposited iron-zinc alloy coatings. Br. Corros. J . , 1(1), 5-8.
Sander, J., and Mady, R. (1989). Corrosion protection by hot dip galvanizing.
Metalloberflliche, 43(7), 329-334 (in German).
Sanyal, B., and Singhania, G. K. (1956). Atmospheric corrosion of metals. Part 1. J .
Sci. Ind. Res., 15B, 448455.
Sanyal. B., and Bhadwar, D. V. (1959). The corrosion of metals in synthetic
atmospheres containing sulfur dioxide. J . Sci. Inn. Res., 18A, 69-74.
Sanyal, B. , Nandi, A. N. , Natarajan, A., and Bhadwar, D. (1959). Atmospheric
corrosion of metals. Part 11: Corrosion of metals in Bombay. J . Sci. Ii2d. Res. , 18A.
Sanyal, B.. Das Gupta, B. K. . Krishnamurty, P. S. V., and Singhania, G. (1961).
Atmospheric corrosion of metals. Part 111: Corrosion of metals in Calcutta. J . Sci.
Ind. Res., 200. 27-30.
Sanyal, B. , Balakrishnan, A. , Singhania, G. K. , and Menon, U. G. K. (1962).
Atmospheric corrosion of metals. Part IV: Corrosion of metals in Cochin. J . Sci.
Ind. Res. , 210, 185-188.
Schafer, G. J. (1962). Cathodic protection of metals immersed in water. Corros. Eng.
(Australia). 6(9), 33-37.
8(3), 185-194.
127-131.
References
Schikorr, G. (194311944). Die Witterungsbesttindigkeit verzinkter Stahldahte und
Drahtseile (The behavior of galvanized steel wire and wire rope). Arch. Eisen-
huttenwes., 17, 147-150 (in German).
Schikorr, G. (1939). The cathodic behavior of zinc versus iron in hot tap water. Trans.
Electrochem. Soc., 247.
Schikorr, G. (1940). Das Verhalten von Zink gegen Natriumchloridlosung und gegen
Meerwasser unter atmosphkischen Bedingungen (The behavior of zinc in sodium
chloride solution and in seawater under atmospheric conditions). Z . Metalk., 32,
314-317 (in German).
Schikorr, G. (1941a). Das atmosphiirische Rosten des Eisens (Atmospheric rusting of
iron). Korros. Metullschutz, 17, 305-313 (in German).
Schikorr, G. (1941b). Einige Zerstorungserscheinungen an Aluminium, Eisen, und
Zink in Mauerwerk (Deterioration of Al, Fe, and Zn in medieval stonework).
Wissenschafl. Abhandl. Dtsch. Mat.-Priif.-Anst, I I , Part 2, 51-54 (in German).
Schikorr, G. (1943). Die Zersetzungserscheinungen der Metalle (Degradation of
Metals) (in German). Verlag J. Barth, Leipzig, 146 pp.
Schikorr, G., and Schikorr, I. (1943). Uber die Witterungsbestandigkeit des Zinks
(The resistance of zinc to atmospheric corrosion). Z. Metallkd., 35(9), 175-181 (in
German).
Schikorr, G. (1974). die Abhangigkeit der atmospharischen Korrosion der
Metalle von den schwefelhaltigen Verunreinigungen der Luft (The dependence of
the atmospheric corrosion of metals on the sulfur-containing impurities in the air).
Metallobefliiche, 1( 5) , 115-1 16 (in German).
Schikorr, G. (1949). The attack on zinc in hot tap water. Z . Metallkd., 41,344-348 (in
German).
Schikorr, G., and Wassermann, G. (1949). iiber den Einfluss der natiirlichen Wit-
terung auf die Spannungskorrosion von Aluminiumlegierungen (The influence of
weather conditions on the stress-corrosion of aluminum alloys. Note: The corrosion
of zinc is here used to measure the aggressiveness of the atmosphere). Metallkd.,
40, 201-205 (in German).
Schikorr, G. (1951). Uber die Korrosion von Kadmium- und Zinkiiberzugen (The
corrosion of cadmium and zinc coatings). Metallobeflache 5(12), 177-185 (in
German).
Schikorr, G. (1953). Uber die Schwitzwasserbildung aus der natiirlichen Feuchtigkeit
bei verpacktem Eisen (Effect of natural moisture within packaged iron). Werkst.
Schikorr, G. (1961a). Relation between the atmospheric corrosion of zinc and nickel
and the sulfur dioxide content of the air. Metall, 15, 981-987 (in German).
Schikorr, G. (1961b). Uber die Korrosion von Metallen in holzemen Gehausen ( me
corrosion of metals in wooden houses). Werkst. Korros., 12, 1-10 (in German).
Schikorr, G. (1964). Corrosion behavior of zinc and zinc coatings in the atmosphere.
Werkst. Korros., 15, 537-543 (in German).
Schikorr, G. (1965). Corrosion Behavior of Zinc, vol. 1, Behavior Of zinc in the
References 497
Atmosphere. Metall-Verlag GmbH, Berlin, English edition published by the Ameri-
can Zinc Institute, New York, and the Zinc Development Association, London, 44
pp. (156 refs.).
Schlapfer, P., and Bukowiecki, A. (1948). Contribution to the study of the corrosion
of metals by motor fuels. Met. Corro., 23, 267-277 (in French).
Schliebaum, R. (1976). Auch im Briickenbau ist die Feuerverzinkung wirtschaftlich
(Hot dip galvanizing is also effective in bridge construction). Verzinken, 5(1), 21-
22 (in German).
Schliebaum, R. (1978). (Calculations of the economic effectiveness of corrosion
protection systems) Wirtschaftlichkeit des Korrosionsschutzes. Ed. Lack und Chem-
ie, Elvira Moeller, Filderstadt, Germany. (English text prepared by ZDA, London).
Schmid, E. V. (1986). Painting of Zinc Surfaces and Zinc-Containing Anticorrosive
Primers. Oil and Colour Chemists Association. Monograph 3. OCCA, Wembley,
U.K., 11 pp.
Schmitt, E. J. A. (1970). Metaalbescherming voor bouwwerken in of nabij indus-
triegebeiden (Steel conservation of buildings in or near industrial areas). Thermisch
Verzinken, (702), 3-7 (in Dutch).
Schneider, H. (1953). Uber die Lebensdauer von Zinkblechen fur Bedachungen (The
durability of sheet zinc roofs). Metall, 7(21/22), 846-848 (in German).
Scholes, I. R., Astley, D. J. , and Rowlands, J. C. (1977). Bimetallic corrosion in
seawater. Eurocorr 77: 6th European Congress on Metallic Corrosion, London.
Society of the Chemical Industry, London, pp. 161-169.
Schaer , K. F. (1975a). Mechanisches Verzinken, Teil 1: Chemische Grundlagen,
Technologie. (Mechanical zinc plating, Part 1: Chemical principles, technology).
Metallobefliiche, 29(9), 429434 (in German).
Schriider, K. F. (1975b). Mechanisches Verzinken, Teil 11: Einfluss der Obefflachen-
behandlung auf die Wasserstoffverspriidung; Wirtschaftlichkeitsberechnung (Me-
chanical zinc plating, Part 11: The effect of surface treatment on hydrogen embrittle-
ment; economics). Metallobefliiche, 29(1 l ) , 544-551 (in German).
Schroder, K. F. (1976). Zunkftssicherer Korrosionsschutz im Klanverk Bonn-Bad
Godesberg (Assured long-term corrosion protection in the Bonn-Bad Godesberg
clarification works). Verzinken, 5(1), 18-20 (in German).
Schulze, W. M. H. (1943). Klima und Werkstoffebeansprung in den Tropen (Clima-
tology and materials use in the tropics). Z. Fernemeldetech, Werk und Geriitebau,
7, 21-28, 3540 (in German).
Schulze, W. M. H. (1956). Technoklimatologie. Die Technik (Climatology-
Technological aspects), 11, 112-1 15 (in German).
Schwabe, K. (1957). (On the periodic variations in current intensities during the
anodic dissolution of zinc.) Metallobefliiche, 11(1), 1-6 (in German).
Schwabe, K., and Sonntag, H. (1958). iiber die periodischen Stromstarke-
schwankungen bei der anodischen Auflosung des Zinkes (The periodic fluctuations
in current amplitude in the anodic solution of zinc). Z. Elektrochem., 62(3), 344-
348 (in German).
References
Schwarz, W. , et al. Hot dip galvanized steel sheet with ZdAl and AVZn alloy
coating for building and appliance industry. Galvatech 92, Verlag Stahleisen,
Dusseldorf, Germany, pp. 1.
Schwenk, W. , and Friehe, W. Korrosionverhalten feuer der sprintzverzinkter
Stalhbleche mit und ohne Schutzanstrich auf dem seewasserversuchsstand des
vereins Deutscher Eisenhuttenleute in Heligoland. (Corrosion behavior of hot dip
galvanized and zinc sprayed steel sheets with without a protective paint coating at
the seawater test site of the Verein Deutscher Eisenhuttenleute on Heligoland). Stahl
Eisen, (in German).
Schwenk, W. Korrosions schutzwirkung von EP-Beschichrungen auf fever-
verzinkten Stahl (Corrosion behavior of coatings on galvanized steel.) Farbe Luck,
97(5), Reprint GAV 4 pp., GAV Dusseldorf (in German).
SDV. See Stichting Doelmatig Verzinken.
Sedloev, I. Effect of diffusion annealing on the corrosion resistance of hot dip
galvanized coatings in media containing sulfur dioxide. Con$ Protection of Steel
Structures Against Atmospheric Corrosion, Prague, Czech. State Res. Inst. Protect.
Mater., pp.
Seils, A. (Optimum corrosion protection with metal + paint . . .) Werkst.
Korros., (in German).
Seiter, W. Feuerverzinkung plus Anstrich-ein uberagender Korrosionsschutz
fur Stahlkonstruktionen der EVV (galvanizing + painting). Elektrizitiitswirtschaft,
(in German). 9th Int. Galv. Conf., Users Day, ZDA, London,
pp. 11-18 (in English).
Sempels, R. E. (undated). Comparative Study of Zinc and Zinc-Aluminum Alloy
Metallized Coatings. Vieille Montagne, Liege, 11 pp.
Seo, M. Cathodic protection of ship hulls. Report I. Results obtained with zinc
anodes. Corros. Eng. , 5,
Sereda, P. J. Characteristics of moisture deposition on corrosion specimens.
Mater. Res. Stand.,
Sergi. G. , Short, N. R., and Page, C. L. Corrosion of galvanized and
galvannealed steel in solutions of pH to Corrosion, 41( 1 l ) ,
Shalaby, L. A., and Abbas, L. Behavior of zinc and zinc-silver alloys in the
presence of some inhibitors. Corros. Sci.,
Shams El Din, M., and Khedr, M. G. A. Die Auflosung von Zink in
Mischungen aus korrodierenden, inhibierended und passivierended Sauren (Disso-
lution of zinc in mixtures corroding, inhibiting and passivating acids). Metal-
loberfache, (in German).
Shams El Din, A. M., and Khedr, M. G. A. Role of anions in the corrosion
and corrosion inhibition of Zn in aqueous solutions. Part 2 Behavior in de-aerated
solutions and the determination of critical currents for passivation. Corros. Sci.,
12(5),
Shams El Din, A. M., and El Haleem, S. M. Studies of the pitting
corrosion of zinc in aqueous solutions. Part Variations of the potential of the zinc
References 499
electrode in relation to the concentration of the aggressive anion. Werkst. Korros.,
25(8), 571-577 (in English).
Shams El Din, A. M., El Hosani, H., and Mohammed, R. A. (1989). Peculiarities in
potential reversal of the zinc/steel galvanic couple. 9th European Congr. Corrosion,
Jaarbeurs Congress Centre, Utrecht, the Netherlands, Paper FU-087, 8 pp.
Sharivker, S. Yu., and Tomin, V. P. (1988). Corrosion of 20 steel and its protection
by zinc coating in water in heat exchangers. Prot. Met., 24(2), 235-237 (translation
of Zushch. Metall.).
Shastry, C. R., and Townsend, H. E. (1989). Evaluation of the cosmetic corrosion
resistance of zinc coated steel sheets-Using the Chrysler chipping corrosion test
and long-term outdoor exposure testing. Galvatech 89, Iron and Steel Institute of
Japan, pp. 519-527.
Shaw, B. A., and Moran, P. J. (1985). Characterization of the corrosion behavior of
zinc-aluminum thermal spray coatings. Corrosion 85, Boston, NACE, Houston,
Paper 212, 26 pp.
Shaw, B. Leimkuhler, A. M., and Moran, P. J. (1987). Corrosion performance of
aluminum and zinc-aluminum thermal spray coatings in marine environments.
Testing of Metallic and Inorganic Coatings. ASTM STP 947. ASTM, Philadelphia,
pp. 246-264.
Shaw, H., and Mantell, C. L. (1963). The corrosion of zinc coupled to stainless steel
304 in flowing seawater. Report. Znd. Eng. Chem., Research Results Service, 27
PP.
Shaw, T. R. (1976). Corrosion map of the British Isles. ASTM, STP 646, ASTM,
Philadelphia, pp. 204-215. (See also Ministry of Agriculture, Food, and Fisheries.)
Shears, P., and Hadley, J. (1989). Zinc-cobalt deposits from an acid chloride
electrolyte. Institute of Metal Finishing Conference, Brighton, pp. 81-92.
Shibuya, A., et al. (1992). Corrosion behavior of galvannealed steel sheet flash-coated
with AI-Mn alloy layer electroplated from molten salt. Galvatech 92, Verlag
Stahleisen, Diisseldorf, Germany, pp. 336-340.
Shida, K. (1955). Nippon Shi h Zairyo Kikui Gakhei Zusshi 2, 42; Chem. Abstr., 52,
1014 (1958).
Shimogori, K., Satoh, H., and Miki, K. (1988). Corrosion behavior of cold-rolled and
zinc alloy electroplated steel sheets in various cyclic corrosion environments. Kobe
Res. Dev., 38(2), 57-60 (in English).
Shin, J. C., et al. (1992). Corrosion resistance of zinc and zinc alloy coated steel
sheets with AI vapor deposition in the upper layers. Galvatech 92, Verlag
Shiota, T., et al. (1992). The corrosion resistance of organic composite-coated steel
sheets. Galvatech Verlag Stahleisen, Diisseldorf, Germany, pp. 372-376.
Shiozowa, L. (1950). Corrosion of Zinc. Case Institute of Technology, Cleveland,
OH.
Shirley, H. E., and Truman, J. E. (1948). Study of the corrosion resistance high
alloy steels to an industrial atmosphere. J . Iron Steel Inst., 160, 367-375.
500 References
Shreir, L. L., Ed. (1976). Corrosion, Vols. 1, 2. Newnes-Butterworth, Boston/
London.
Shuldener, H. L., and Lehrmann, L. (1958). Effect of increasing the area of iron to
zinc on the rate of reversal of potential in a zinc-iron system. Corrosion, 14(12),
54%.
Shulman, S. L., and Weast, R. C. (1962). Surface Characteristics of Zinc in Aqueous
Media. Corrosion of Zinc in Pressurized Dilute Aqueous Systems. Case Institute of
Technology, Cleveland, OH, 55 pp.
Siegert, W., and Hadley, J. (1989). Alloy coatings: Zinchickel and zinckobalt.
Corrosion and deposition behavior. MetalloberJltiche, 43(2), 78-80 (in German).
Simm, D. W. (1984). The corrosion of zinc in mortar and its relevance to wall tie
corrosion. Corros. Prev. Control, 31(6), 6-10.
Simm, D. W., and Button, H. E. (1985a). Electrochemical measurements on metals in
CCA-treated wood. Corros. Prev. Control, 32(4), 67-74.
Simm, D. W., and Button, H. E. (1985b). The corrosion behavior of certain metals in
CCA-treated timber. Environmental tests at 100% relative humidity. Corros. Prev.
Control, 32(2), 25-35.
Sjoukes, F. (1970). Het gedrag van zinklagen en geandiseerd aluminium bij lage
temperaturen (Behavior of zinc coatings and anodized aluminum at low tempera-
tures). Thermisch Verzinken (701), 16-17 (in Dutch).
Sjoukes, F. (1973). Low temperature behavior of zinc coatings and anodized alumi-
nium. Anticorros. Methods Mater., 20(3), 13-14.
Sjoukes, F. (1974). The structure and determination of the structure of zinc coatings
applied by hot dip galvanizing. Electroplat. Met. Finish., 27(11), 17-21.
Sjoukes, F. (1976). The corrosion resistance of iron-zinc layers. EGGA General
Assembly, pp. 14-16, and T/R Report 111 (1978), 9 pp.
Sjoukes, F. (1978). Sherardizing. Finish. Ind., 2(4), 26, 28-29.
Sjoukes, F. (1984). General Hot Dip Galvanizing. Typescript (unpublished) copy with
ZDA, London, 375 pp.
Sjoukes, F. (1987). The role of iron in the hot dip galvanizing process. Mefall, 41(3),
Sjoukes, F. (1991). Brown staining of hot dip galvanized coatings in atmospheric
corrosion. Br. Corros. J. , 26(2), 103-107.
Skenazi, A. F., Davin, A., Coutsouradis, D., and Goodwin, F. E. (1985). The
influence of alloying elements in zinc-aluminium baths for continuous hot dip
galvanizing. 1st Int. Conf. on Zinc Coated Steel Sheet, ZDA, London, pp. OA/1-5.
Smith, C. A. (1980). Designing a cathodic protection system. Elecfr. Rev. 207(18),
35, 39, 41.
Smith, D. J. , and Van der Schijff, 0. (1989). Corrosion of galvanized steel and
carbon steel in deaerated aqueous solutions of industrial fertilizer chemicals. Br.
Corros. J . , 24(3), 189-191.
Smith, D. W. (1969). Protection of steel bridges from corrosion. African Highway
Conference, Addis Ababa, Ethiopia, paper, l1 pp.
References 501
Smith, R. J. H. Construction of reinforced earth structures. Con$ Construc-
tion Practices Engineering, Surat. India, pp.
Smith, W. Action of certain solutions upon aluminium and zinc. J . Soc.
Chem. Ind. London, 23,
Smrcek, K. , and Barton, K. Electrochemical and corrosion studies of
zinc in various waters. Chem. Listy, 49, (in Czech); Collect.
Czech. Chem. Commun., (in German).
Smuts, J. , and Callaghan, B. G. The Use of Thermal Sprayed Coatings to
Combat Atmospheric Corrosion in the Building and Construction Industry. Avail-
able from National Technical Information Service, Springfield, VA, as PS
pp.
Soepenberg, E. Nagel Levensdum en hergebruik van materialen; tussen
Scyllaen Charibdis van een tweezijdige milieu-invloed (Durability and reuse of
materials: Between Scylla and Charybdis of a duo-sided influence of the environ-
ment). Paper presented at the Corrosion Days of the Nederlands Corrosie Centrum
(NCC). pp. (Also in an NCC syllabus in
Sony Group. Automobile battery charger. British Patent pp.
Sotskov, N., and Olshanskaya, Corrosion protection of fasteners by
diffusion zinc coatings. Alumozinc Symposium, Dnepropetrovsk. Typescript, 10 pp.
(in Russian).
Souske, R. (Use of zinc anodes for cathodic protection.) Corros. Anticorros.
I , (in French).
Souske, R. Les toitures en zinc et les t2les galvanisbes en atmosphCre marin
(Zinc roofing and galvanized sheets in a marine atmosphere). Corros. Anticorros.
(in French).
Southin, R. T. Some observations on the microstructure of zinc alloy anodes.
J . Inst. Met., 93(
Southwell, C. R., Bultman, J. D., and Alexander, L. A. Corrosion of metals
in tropical environmentsFina1 report of 16-year exposures. Mater. Pegormance,
Sparkes, H. L. Construction and operational aspects of the Ekofisk gas
pipeline. Pipes Pipelines h r . ,
Spedding, P. L. Corrosion by atmospheric sulfur dioxide. Australasian Cor-
ros. Eng. ,
Spence, J. W. , and Haynie, F. H. GalvanicandPitting Corrosion-Field and
Laboratory. ASTM, Philadelphia, pp.
Spence, J. W. , and Haynie, F. H. Derivation of a damage function for
galvanized steel structures: Corrosion kinetics and thermodynamic considerations.
Corrosion Testing and Evaluation: Silver Anniversary volume. ASTM STP 1000.
ASTM, Philadelphia, pp.
Spence, J. W. , Edney, E. O., Haynie, F. H., et al. Advanced laboratory and
field exposure systems for testing materials. Corrosion Testing and Evaluation:
Silver Anniversary volume. ASTM STP 1000. ASTM, Philadelphia, pp.
502
References
Spindler, H. (1958). Femschutzwirkung aufgespritzer Metallschichten auf Stahl
(Thermally sprayed coatings on steel.) Werkst. Korros., 9, 278-283 (in German).
Sri Krishna, T. L., Dayananda, M. A., and Grace, R. E. (1971). Kinetics of
dissolution of aluminum in bismuth and zinc in mercury. Met. Trans., 2(12), 3355-
3359.
Stalker, W. W., Dickerson, R. C., and Kramer, G. D. (1963). Atmospheric sulfur
dioxide and particular matter-A comparison of methods of measurement. Ind.
Hygiene J. , January-February, pp. 68-79.
Stanners, J. F. (1960). Comparison of corrosiveness in indoor atmospheres. J . Appl.
Chem., 10, 461470.
Steel Structures Painting Council. (1986). (Includes papers by Beitelman, Chason, and
Chandler on thermal spraying for U.S. Army, U.S. Navy and construction industry,
respectively). Preprints. SSPC Annual Symposium.
Steinhauser, S., and Wiinsch, R. (1967). Fachber. OberJliichentech. July-August,
pp. 1 1 1-116 (in German).
Steinrath, H. (1960). Die chemische Untersuchung der biiden zur Beurteilung ihres
korrosionschemischen Verhaltens (Assessment chemical corrosion behavior).
Gas- Wasserfach, 101, 145-149 (in German).
Stem, M., and Geary, A. L. (1957). Electrochemical polarisation. I: A theoretical
analysis the shape of polarisation curves. J . Electrochem. Soc., 104( l), 56-63.
Steven Winter Associates. (1980). Cost Comparison Study of Galvanized Steel vs.
Wood Framing Systems. Zinc Institute, New York, 12 pp.
Stichting Doelmatig Verzinken. (1962). (Steel protection by hot dip galvanizing.) (in
Dutch), SDV, The Hague, Netherlands, 76 pp.
Stiefel, R. (1987). Halogenierte Kohlenwasserstoffe im Betrieb und in der Umwelt
(Halogenated hydrocarbons in tests and in service) MetellobelJIiiche 41 (3), 109-
113.
Stiles, D. C., and Edney, E. (1989). Dissolution of zinc into thin aqueous films as a
function of residence time, acidic species and pH. Corrosion, 45(1 l ), 896-901.
Stoneman, A. M. (1986). Protective Coatings for Steel: Initial Costs in Perspective.
ZDA, London, 6 pp.
Stratmann, H. (1957). Olfeuerung und Lufthygiene (Oil heating and air hygiene).
Mitte V.G.B. (47), 12-13 (in German).
Stratmann, H. (1959). Schwefeldioxyd-Immissionen eines Heizkraftwerks in
Miinchen (Sulfur dioxide emissions at a works in Munich). Staub, 19, 352-360 (in
German).
Strekalov, P., and Berukshtis, G. (1965). Atmospheric corrosion of zinc and cadmium
coatings on steel and the changeover from accelerated to actual testing conditions.
Korroz. Met. Splavov. Sb. (2), 264-278 (in Russian).
Strekalov, P. V., and Wui, W. D. (1985). Mathematical model of probable rates of
atmospheric corrosion of metals in tropical and temperate climates. Pror. Met.,
21(4), 429-437 (translation of Zashch. Metall.).
Strom, M., et al. (1992). Mechanism of underfilm corrosion propagation in painted
automotive galvanized steels during exposure in the new Volvo high performance
chamber. Galvatech 92, Verlag Stahleisen, Diisseldorf, Germany, pp. 521-527.
References 503
Stroud, D. (1992). Galvanizing with zinc-nickel alloys. Proc. 1st Asian Gen. Galv.
Conf., pp. 169-175.
Stupel, H. , and Koch, F. (1959). The corrosion of zinc by modem detergents. Werkst.
Korros., 11, 33-39 (in German); English version in Soap, Perfum. Cosmet., 32,
Sugimoto, S. , Tanaka, A., and Gotoh, H. (1988). Supergalva-Recent marketing
developments. 15th Int. Gab. Conf., ZDA, London, pp. GF7/1-6.
Sugimoto, S. , and Gotov, H. (1989). Prevention of gap corrosion by zinc-aluminum
alloy. Bosei Kanri, 33(1), 21-33 (in Japanese).
Sulyok, J. (1975). Indirekte Korrosionseinwirkung von Isoliermaterialien (Indirect
corrosion by insulating materials). Werkst. Korros. 26(1 l), 858-862 (in German).
Summerville, 0. (1970). Some aspects of corrosion testing at the Point Cook marine
test site. Aust. OCCA Process News, 7(9), 7.
Suzuki, I. (1985). The behavior of corrosion products on zinc in sodium chloride
solution. Corros. Sci., 25(1 l), 1029-1034.
Swamy, R. N. (1990). Answering problems from users of galvanized rebar. EGGA
General Assembly, Lausanne, EGGA, London, pp. 2.2-2.6.
Swedish Standard. (1991). SS 11 72 11 (draft). Corrosion Tests in Artificial
Atmospheres-Accelerated Outdoor Tests by Intermittent Spraying of a Salt Solu-
tion (Scab Test), SIS, Stockholm, 5 pp.
Sydberger, T., and Vannerberg, N. G. (1972). Influence of the relative humidity and
corrosion products on the adsorption of sulfur dioxide on metal surfaces. Corros.
Sci., 12(10), 775-784.
Taboury, F., and Gray, F. (1939). Korrosion von Zink (Corrosion of zinc). Bull. Soc.
Chim., 368-382 (in German).
Tai-Sam, K. Y., and Tung-mou, Tsai. (1992) R + D of galvanizing 5% AI-Zn alloy
by batch type process. Proc. 1st Asian Pacific Gen. Galv. Cong., pp. 167-168.
Tajiri, Y., Shimada, S. , Yamaji, T., and Adaniya, T. (1989). Effect of coating
structure on corrosion resistance of hot dipped Zn-5% AI + Mg alloy coated steel
sheet. Galvatech 89, Iron and Steel Institute of Japan, pp. 553-559.
Tanaka, N., Aoki, Y., and Yamada, A. (1969). Determination the electrode
reaction of hydrated zinc ions by polarographic and potentiostatic methods. Electro-
chim. Acta, 14( 1 l), 1155-1 165.
Tano, K., and Higuchi, S. (1985). Development and properties of zinc-aluminum
alloy coated sheet with high corrosion resistance. Nippon Steel Technical Report 25,
pp. 29-37.
Taraevskii, S. I. , and Kuslitskii, A. B. (1988). Lifetime increase of pump rods
operating in media containing hydrogen sulfide. Zashch. Pukrytiya Met., 22,92-94
(in Russian).
Tattersall, H. (1971). Value analysis and the user of die castings. Metallurgia,
83(495), 9-12.
Teel, R. B., and Anderson, D. B. (1956). The effect of iron in galvanic zinc anodes in
seawater. Corrosion, 12, 343-349.
Teeple, H. 0. (1952). Corrosion by some organic acids and related compounds.
1111-1118.
504 References
Tefankjan, D. A. (1972). Application of cathodic protection. Mater. Prof., IJ(1 l),
Thiem Corp. (1971). Galvanizing preflux wash composition. U.S. Patent 3,740,275.
Thomas, R. (1980). Rust Prevention by Hot Dip Galvanizing. Nordic Galv. Assoc.;
English edition available from ZDA, London, 32 pp.
Thomas, R. (1989). Hot Dip Galvanizing, 2nd ed., 37 pp.; Swedish edition (ed. by T.
Wallin), (English edition 1992).
Tierra, P., Bernal, M., Molera, P., and Fernandez, L. (1991). Comparative study of
the cathodic protection supplied to steel by hot dip coatings. 3rd Int. Zinc Coated
Steel Sheet Conf., EGGA, SlEA-6 pp.
Tiidt, F. (1958). Metallkorrosion, Allgemeines, Messung, Verhiiung (Metal corrosion;
general, data, prevention), 2nd ed., de Gruyter, Berlin, 122 pp.
Todt, F., Ed. (1961). Korrosion und Korrosionsschutz (Corrosion and corrosion
protection), 2nd ed., de Gruyter, Berlin, 1422 pp.
Todd, J. M., and Perkins, J. (1976). Corrosion of zinc anodes in seawater. Naval Eng.
Tomashov, N., and Leonov, V. (1965). Effect of Zinc as a Filling Agent on P rofective
Properties of Bituminous Coatings. Reprint. 12 pp.
Townsend, H. E., Borzillo, A. R., and Barber, W. D. (1988). Performance of AI-Zn
alloy coated sheet after 32 years of atmospheric corrosion testing, 2nd Int. Conf. on
Zinc-Coated Steel Sheet, pp. SA1 .l-1 1.
Tragardh, K. F. (1958). Protection of steel against underground corrosion. Corros.
Anticorros., 6(9), 292-297 (in French).
Traube, M. (1885). Uber die Mitwirkung des Wassers bei der langsamen Verbrennung
des Zinks, Bleis, Eisens und Palladiumwusserstoffs (The involvement of water in
the slow burning of zinc, lead, iron and palladium hydrides). Deutsch. Chem.
Industry Report 18, 1877-1887.
Trautmann, B. (1960). Das Zinkdach im Schulbau (Zinc roofs for school buildings).
Zink (4), 1-5 (in German).
Trifel, M. (1989). The use of metallic protective coatings in the Soviet Union. J . Prot.
Coat. Linings, 6(12), 29-31.
Tropeano, A. (1987). Lavegal: A new Zn-AI-Mg coating for sheet steel. National
Assoc. Coil Coaters Meeting, Report, pp. 13-19.
Troquet, M., Labbe, J. P., and Pagetti, J. (1981). The mechanism of the inhibition of
zinc corrosion in 1 N HCI solution by tetraphenyl-phosphonium bromide. J .
Corros. Sci., 21(2), 101-117.
Troquet, M., and Pagetti, J. (1983). Inhibition of metallic corrosion in an acid medium
by means phosphonium salts: Zinc and iron. Werkst. Korros., 34(1 l), 557-562
(in German).
Trusdale, C. E., et al. (1935). Technical Publication 651-E, American Institute of
Mining, Metallurgical, and Petroleum Engineers, New York, 3 pp.
Uhlig, H. H. Ed. (1948). The Corrosion Handbook. Wiley, New York; Chapman &
Hall, London, 1188 pp. Section on Zinc (E. A. Anderson, and C. E. Reinhard),
pp. 331-347; Zinc coatings on steel (W. BIum and A. Brenner), pp. 803-816.
J 88(4), 65-72.
References 505
Uhlig, H. H. (1963). Corrosion and Corrosion Control. Wiley, New York/London,
372 pp.
Ulanovsky, I. (1965). Crevice corrosion of zinc, cadmium and lead. Korroz. Met.
Splavov. Sb., (2), 359-365 (in Russian).
Underwood, J. D., et al. (1971). Fingerprint corrosion on telephone switching equip-
ment. Trans. Inst . Met. Finish., 49(3), 123-126.
UN1 68.00.003.0 (1993). Evaluation of the Corrosivity of Soils for Low Alloyed and
Unalloyed Metallic Materials (English translation by Di Biase), Italian Standards
Organization, Milan, 25 pp.
Union Carbide Corporation. (1984). Single package solvent-resistant, zinc-rich ther-
moplastic polyhydroxyether coatings. U.S. Patent 4 529 756, October 19. (July 16,
1985), 7 pp.
Upton, M. (1987). Field experience with galvanized steel and practical aspects of its
overpainting. U. K. Corrosion Conf., Institute of Corrosion Science and Technol-
ogy, Birmingham, paper, 9 pp.
U.S. Treasury Department, U.S. Coast Guard (1964). Navigation and vessel circular
3-64, and 6-64, excerpt from Fed. Regis., May 16, Title 46, Shipping.
Van Eijnsbergen, J. F. H. (1959a). The Corrosion Resistance of Hot Dip Galvanized .
Steel, 3rd ed. (in Dutch). SDV, the Hague, Netherlands, 100 pp.
Van Eijnsbergen, J. F. H. (1959b). Die Korrosionbestandigkeit feuerverzinkter
Stahlkonstruktionen (The corrosion resistance of hot dip galvanized steel construc-
tions). Jahrb. Oberfliichentech. 15, 334-343.
Van Eijnsbergen, J. F. H. (1967). Feuerverzinkter Armierungsstahl in Beton (Hot dip
galvanized reinforcing steel in concrete). Jahrb. Obefluchentech., 23, 133-139;
Metall, 19, 1263-1267 (1965) (in German).
Van Eijnsbergen, J. F. H. (1976a). Feuerzinkte Lotsenboote (Hot dip galvanized pilot
boats). Verzinken, 5(1), 14-15 (in German).
Van Eijnsbergen, J. F. H. (1976b). Twenty years of duplex systems. 11th lnt. Galv.
Conf., ZDA, London, pp. 123-129. See also: Metall, 29(6) 585-591, (1973, with
Eng. transl. (BISI Trans. 14052).
Van Eijnsbergen, J. F. H. (1978). The reliability of galvanizing versus two-coat paint
systems and a duplex system (in Dutch). SDV, the Hague, Netherlands.
Van Eijnsbergen, J. F. H. (1981). Hydrogen blistering and its influence on the
adhesion of paint films on galvanized steel. Report to ZDA, London, 2 pp.
Van Eijnsbergen, J. F. H. (1982a). Comparison between galvanizing and zinc-rich
paints. Members Letter 1138, Galvanizers Association, London, 5 pp.
Van Eijnsbergen, J. F. H. (1982b). Increasing air, soil and water pollution in W and
NW Europe and its impact on the corrosion resistance of galvanized steel. 13th Int.
Galv. Conf., ZDA, Portcullis Press, London, pp. 7/1-8.
Van Eijnsbergen, J. F. H. (1983). Edge protection on galvanized and on painted
profiles. Thermisch VerzinkenlFeuerverzinken, 12(2/3), 50-52 (in Dutch and Ger-
man).
Van Eijnsbergen, J. F. H. (1985a). Thirty years of duplex systems. 4th Int. Conf.
South African Corrosion Institute, Johannesburg, 27 pp.
506 References
Van Eijnsbergen, J. F. H. (1985b). Contact corrosion. Thermisch Verzinkenl
Feuerverzinken, 14(2/3), 44-45 (in DutchGerman).
Van Eijnsbergen, J. F. H. (1986a). Duplex coatings: A review of recent develop-
ments. EGGA General Assembly, Monaco, ZDA, London, pp. 3.3-3.5.
Van Eijnsbergen, J. F. H. (1986b). Topochemical corrosion phenomena on a hot dip
galvanized fence in an industriallmarine climate. Thermisch Verzinkenl
Feuerverzinken, 15(2), 30-32 (in DutchGerman).
Van Eijnsbergen, J. F. H. (1986~). Protecting steel by hot dip galvanizing in water
purification plants. Thermisch VerzinkenlFeuerverzinken, 15(4), 55-56 (in Dutch
and German).
Van Eijnsbergen, J. F. H. (1987a). Aqueous dispersion paints for hot dip galvanizing
steel. EGGA General Assembly, Dublin, ZDA, London, pp. 3.6-3.7.
Van Eijnsbergen, J. F. H. (1987b). Duplex systems in practice. Fincorr Corrosion
Conf., Turku, Finland, paper, 22 pp.
Van Eijnsbergen, J. F. H. (1988). The interface of duplex systems. 15th. Int. Galv.
Conf., ZDA, London, pp. G12/1-G12/8.
Van Eijnsbergen, J. F. H. (1991). New insights into the interfacial zone of duplex
systems. 16th Int . Galv. Conf., EGGA, London, pp. GIl/l-7; Adhesion of duplex
systems. Thermisch Verzinken/Feuerveninken/Hot dip Galvanizing, September (in
Dutch, German, and English).
Van Eijnsbergen, J. F. H. (1994). Duplex Systems; in press (Elsevier).
Van Oeteren, K.-A. (1967). Haftungsschwierigkeiten von Anstrichen und
Lackierungen auf verzinktem Stahl durch 01- und Fettschichten (Problems with
paint and lacquer coatings through oil and grease layers). Werkst. Korros., 18(2),
Van Oeteren, K.-A. (1984). Filmafbradk en levensduur bij deklagen op Staal (Film
breakdown and life of top coatings on steel). Oppervlaktetechnieken, 28(7), 203-
205 (in Dutch).
Van Oeteren, K.-A. (1987). New zinc coating for protecting sheet and wire. Metal-
loberjlache, 41(2), 89-90; Jahrb. Oberjlachentech., 43, 275-278 (in German).
Van Ooij, W. J. , Leijenaar, S. R., and van den Burgh, B. (1982). Study of interfacial
interactions between paint systems and various metals using surface analytical
techniques. XVI FATIPEC Congr. 56 pp.
Van Ooij, W. J. , Chidambaram, P. R., and Rangarajan, V. (1991). Characterization
of high temperature hot dip galvanized coatings. Surf. Coat. Technol., March, 20
PP.
Varady, E. (1971). Comparison of the corrosion resistance of hot galvanized steel,
zinc-sprayed steel and zinc plated (electrogalvanized) steel in urban industrial
atmospheres. EGGA General Meeting, Cannes. Paper, available from EGGA.
Vassilev, K. K. , Petrov, M. D., and Krasteva, V. Z. (1974). The influence of the
structure of zinc alloys on the protective capacity of zinc anodes in seawater.
Werkst. Korros., 25(8), 587-592 (in German).
Vazquez Vaamonde, A. J. , and Damborenea Gonzalez, J. J. de (1991a). The Sandelin
effect and continuously cast steels. Int. J . Mater. Prod. Technol., 175-216.
References 507
Vazquez, A. J., and Damborenea, J. J. de (1991b). Corrosion resistance of galvanized
coatings: Retrospective evaluation. 16th f nt . Gab. Conf., EGGA, London, GB4,
1-12.
Verberne, W. J. C., and Hadley, J. S. (1984). Zinc-cobalt alloy plating. U.S. Patent
4,439,283.
Vernon, W. H. J. (1923-1924). First report to the Atmospheric Corrosion Committee.
Trans. Faraday Soc., 19, 839-936.
Vernon, W. H. J. (1927). Second report to the Atmospheric Corrosion Committee.
Vernon, W. H. J. (1935). A laboratory study of the atmospheric corrosion of metals.
Vernon, W. H. J., Akeroyd, E. J. , and Stroud, E. G. (1939). The direct oxidation of
zinc. J . Inst. Met, 65(2), 301-343.
Vignier, J. (1961). (Sprayed zinc on steel constructions.) Zinc, Cadmium Alliages,
(26), 35-39 (in French).
Voeltzel, J. , Murray, G. , and Constant, A. (1967). The effect of a galvanized coating
on the resistance to stress corrosion of high tensile wire in the presence of hydrogen.
8th Int. Galv. Conf., Ind. Newspapers, U.K., pp. 271-280.
Von Backmann, W. G., and Funk, D. (1982). Corrosion losses of zinc with direct and
alternating currents. Werkst. Korros., 33, (in German).
Von Fraunhofer, J. A., and Lubinski, A. T. (1974). Polarity reversal in the zinc-mild
steel couple. Corros. Sci., 14(3), 225-232.
Von Hofe, H. (1953). Rostschutz von Stahlbauten durch Zinkspritzen (Zinc spraying
for protecting structural steelwork against corrosion). Stahlbau, 22( 12), 265-274 (in
German).
Vrable, J. B. (1985). Comparison of corrosion performance of zinc and iron-zinc
alloy coatings in accelerated, atmospheric and highway automotive tests. Interna-
tional Congress and Exposition, Society of Automotive Engineers. SAE, Warren-
dale, PA, Paper 850002, pp.
Vui, V. D., Chi, N. K., and Ki, N. T. (1988). Atmospheric salinity in different
climatic regions of Vietnam and its effect on the corrosion rate of steel, zinc and
cadmium. Prot. Met., 24(2), 171-177. [Translation ofzasasltch. Metall. 24(2), 227-
233.1
Wagner, I. (1977). Influence of water quality and water treatment on corrosion and
coatings in steel and galvanized steel tubes. Eurocorr '77. 6th European Congr.
Metallic Corrosion. Society of the Chemical Industry, London, pp. 413419.
Wagner, J., and Dannohl, R. (1985). Influence of different nitrate contents of drinking
water on the corrosion behavior of galvanized steel pipes. Werkst. Korros., 36,
(in German).
Waldron, L. J., and Peterson, M. H. (1960). Effect of iron, aluminum and cadmium
additions on performance of zinc anodes in seawater. Corrosion, 16, 375t-379t.
Walker, A. G. , and van Eijnsbergen, J. F. H. (1969). Calcium Plumbate Paintsfor
Galvanized Steel. ZDA, London, 8 pp.
Wall, N. J., Spittle, J. A., and Jones, R. D. (1985). Thedeformationcharacteristics of
References
zinc coatings on hot dip galvanized mild steel strip. Ist Int. Conf. Zinc-Coated Steel
Sheet. ZDA, London, pp. FI-FIO.
Wallington, R. H. (1974). Painting of Flame Sprayed Zinc Coatings. ZDA, London
12 pp. [Reprinted from Br. Corros. J . 9(2) 72-83).]
Walquist, H. W. (1945). Corrosion, I , 119.
Watanabe, T. (1992). Microstructure and corrosion resistance of hot dip zinc coatings
on castings. Ist Asian Pacific Gen. Galv. Conf., pp. 188-194.
Weast, R. C. (1954). study of some variables which influence the corrosion rate of
galvanized steel hot water storage tanks. Proc. Am. Gas. Assoc. 34, 863-867.
Weast, R. C., Kotnik, L. J., and Geehan, D. M. (1961). Corrosion of zinc in dilute
aqueous solutions. 6th Int. Galv. Conf., ZDA, London, pp. 155-176.
Weast, R. C., and Shulman, S. L. (1962). Corrosion of zinc in pressurized dilute
aqueous solution systems. Corrosion, 18, 417t420t.
Wegener, K. (1942). Haus und Klima (Houses and climatic variations). Angew.
Meteorol. 59, 1-6 (in German).
Weins, W. N. (1991). First hot dip galvanized bridge beats corrosion. Reprint from
Roads and Bridges (Nov.) 3 pp., AGA.
Weisenberger, L. M. (1975). Zinc anodizing comes of age. Prod. Finish., 39(5), 51-
58.
Wellington, J. R. (1961). The low potential zinc anode in theory and application.
Corrosion, I7(1 l ) , 550t-556t.
Wesnick, R. (1991). Economic aspects regarding the use of flame spraying or arc
spraying. Korrosion, 22(5), 292-293 (in German).
Wesolowski, J., and Jagiello, W. (1989). Rough surfaces of hot dip galvanized steel.
MetalloberJliiche, 43(7), 325-327 (in German).
White, R. A., and Deanin, R. D. (1965). Clearfinishesfor polished zinc. ILZRO,
New York. Project ZE-82, Final Report, May, 63 pp.
Whitecavage, J. B. (1985). Coupon evaluations of manhole hardware metals consider-
ing corrosive ions in electrolyte. Corrosion 85, Paper 374, reprint, 7 pp.
Whitney, W. R. (1947). The corrosion of iron. J . Am. Chem. Soc., 22, 394.
Wiederholt, W., Ed. (1945). Korrosionsprufierfahren (Corrosion test procedures).
Verlag Chemie, Berlin, 210 pp.
Wiederholt, W. (1960). Das Verhalten von Metall- und Kunststoffuberzugen und von
Rostschutzanstrichen in Hafen-, Meer- und Kalanwasser (The behavior of metal,
decorative and anti-rust coatings in harbour, sea and canal water). SchifHafen, 12,
special issue, 105-115 (in German).
Wiederholt, W., and Sonntag, J. (1963). Das Verhalten von spritzmetallisierten
Blechen ohne und mit Rostschutzanstrichen oder Kunststoffuberzugen bei
Laboratoriumsversuchen. (Laboratory investigation of the behavior of metal-
sprayed steel sheets with and without corrosion-protective or synthetic coatings).
Schweiss Schneid., 15, 113-123, 163-168 (in German).
Wiederholt, W. (1965). Behaviour of Zinc in Wafer. Metall-Verlag, Berlin, 162 pp.,
English typescript translation available in ZDA library, London.
References 509
Wiederholt, W. (Behaviour of zinc with chemicals.) Korrosionverhalten von
Zinc, Band (in German but identifiable data), pp.
Wijngaard, B. H. Cathodic Protection of Steel Constructions in Seawater.
Economic Choice of Galvanic Anodes. Dutch Institute for Applied Scientific Re-
search (TNO), Metaalinst, Delft, Netherlands, Publication pp.
Williams, D. N., et al. Corrosion behavior of Zn-22 AI alloy steel. J .
Electrochem. Soc. (U.S.A.),
Williams, L. F. G. The effect of chromate content on the corrosion of
chromated zinc electroplate on steel. Surf. Technol.,
Wilmotte, S. , et al. Control of the galvannealing process. Galvatech
Verlag Stahleisen, Diisseldorf, Germany, pp.
Wippermann, K. Schultze, J. W. Kesel, R., and Penninger, J. The inhibition
of zinc corrosion by bisaminotriazole and other triazole derivatives. Corros. Sci.,
Witt, C. A., and Haller, D. Korrosionverhulten von Zink. Band 4: Verhaften
von Zink in Verbindungen mit bituminosen Baustoffen (Corrosion behavior of zinc.
Part Behavior of zinc in combination with bituminous building materials). Metall-
Verlag, Berlin. Booklet (in German), pp. Bibliography, items.
Witt, C. A., Dnisga, I. , and Kola, W. Korrosionsinhibitoren fur Zinksysteme
(Corrosion inhibitors for zinc systems). Metall, (in German).
Wolf, W. Verhalten von Feinzinklegierungen in den Tropen (The behavior of
high purity zinc alloys in the tropics). Metall, (in German).
Wood, R. B. Metallization of the Forth Road Bridge. Symp. Protection of
Steel Structures by Metal Spraying, London, Association of Metal Sprayers, U.K.
Paper pp.
Woody, C. L. Long-term experience with galvanic anodes in a city distribu-
tion system. Corrosion '76, Houston, TX, NACE, Paper pp.
Wormwell, F., and Mercer, A. D. Sodium benzoate and other metal benzoates
as corrosion inhibitors in water and in aqueous solutions, J . Appl. Chem.,
Wormwell, F. Chem. Ind.,
Wrubl, C., et al. Traditional testing methods for painted metal substrates
supported by electrochemical impedance measurements. Galvatech Verlag
Stahleisen, Diisseldorf, Germany, pp.
Yamaguchi, H., Hirakata, Y., Kita, D., Moriya, M., and Kamata. F.
Corrosion prevention of exposed iron bars in concrete by zinc tape on the joins.
Bosei Kanri (Rust Prev. Control), (in Japanese).
Yandushkin, K. N., and Kuris, V. P. Corrosion of zinc-coated tubes in
flowing seawater. Zashch. Metall., (in English).
Yangzong, L., Chum, L., Jintang, L. and Jinhon, C. Effects of AI and RE on
the corrosion resistance and the structure of the hot dip Zn-5% AI-RE alloy coated
steel sheet. Galvatech Iron and Steel Institute of Japan, pp.
Yeomans, S. R. Investigations of galvanized and epoxy coated steel reinforce-
ment in concrete. 16th Int. Galv. Conf., EGGA, London, GBlll-5.
51 0 References
ZALAS (1980). Galvanizing Manual. Zinc and Lead Asian Service (ZALAS),
Zarubitsky, G. , and Salabai, A. (1969). Behavior of low melting point metals in
Zastrow, W. (1973). Corrosion problems in underground electric distribution
ZDA. (1983a). Technical Notes on Zinc: Cathodic Protection. ZDA, London, 8 pp.
ZDA. (1983b). Technical Notes on Zinc: Zinc Coatings. ZDA, London, 8 pp. (Also
separate leaflets on each coating.)
ZDNBSC. (1971). Detailing Coated Steel Sheet l : Unframed Non-Load-bearing
Skin. ZDA, London, 32 pp.
Zielinski, A., and Speichert, B. (1986). Corrosion tests of zinc and aluminum hot
dipped coatings in marine environments. Ochr. Przed Koroz., (in Polish).
Zinc and Lead Asian Service. See ZALAS.
Zinc Institute. (1981). Galvanized Steel vs. Wood Framing Systems. Leaflet. Zinc
Zinc Institute (undated). How Does Galfan Compare? Zinc Institute, New York, 2 pp.
Zinc Metallizers Task Group. ZMTG Library List. ZMTG, Greenwich, C T , 384 items.
Zinkberatung eV, Diisseldorf. (1959). Zink-Taschenbuch (Zinc Pocketbook). Metall-
Verlag, Berlin-Grunewald, 380 pp. (in German). Also Zink-Taschenbuch, Metall-
Verlag, Berlin, 1981, H. Johnen, Ed. (in German), pp.
Zinkberatung eV. (1975). Differences and Similarities Between Alloys 2410 and 2400,
English translation by ZDA, London, 4 pp.
Zinkberatung eV. (1979). Zinc Die Castings Stand Up to Methanol. Zink NO. 14.
Translation by ZDA, London, pp.
Zinkberatung eV. (1990). Zink fur Stahl (Zinc for steel). Zinkberatung, 216 pp. (183
refs.) (in German).
Zoccola, J. C., Townsend, H. E., Borzillo, A. R., and Horton, J. B. (1978).
Atmosphere Corrosion Behavior of Aluminum-Zinc Alloy Coated Steel. ASTM STP
646. ASTM, Philadelphia, pp. 165-184.
Melbourne, 222 pp.
molten caustic soda. Ukr. Khim. Zh., 35( I l ), 1209-1210 (in Russian).
systems. IEEE Rural Electric Power Conf., Minneapolis, MN.
Institute, New York, 12 pp.
Index
NOTE: The material used for each test is not indexed as most information on
corrosion of zinc applies to zinc-coated steel and zinc alloys. Individual
geographic locations of atmospheric corrosion data is not indexed nor are
individual chemicals. Reference should made to the relevant tables in the
indexed section.
Abrasion resistance,
Accelerated tests (see Testing)
Acids, (see also Corrosion, pH
effect)
Adhesion,
Adhesion test (NMP) for paints,
Adhesive zinc foil,
Agricultural chemicals,
Alcohol stains, removal
Algafort (see Alloys, with aluminum
Alkalis, (see also Corrosion, pH
Alloy designations, 6,
inhibitive additions,
and silicon)
effect)
[Alloy designations, continued]
ZA-8, 5-6,
corrosion resistance,
aluminum,
5-6,
Alloying elements, effect on
cadmium,
calcium,
cerium, (see also Alloys with
chromium,
cobalt,
aluminum and mischmetal)
51 1
51 2
Index
[Alloying elements, continued]
copper,
gallium,
indium,
iron,
lanthenum, (see also Alloys
with aluminum and
mischmetal)
lead,
magnesium,
manganese,
mercury,
nickel,
platinum,
silicon,
sodium,
tin,
titanium,
vanadium,
castings,
coatings,
corrosion effect,
with aluminum, 3-6,
with aluminum and cadmium,
with aluminum and copper,
with aluminum and magnesium,
with aluminum and mischmetal,
with aluminum and silicon,
Alloys of zinc
with aluminum and tin,
with chromiumlchromium oxide,
[Alloys of zinc, continued]
with cobalt,
with cobalt and chromium,
with copper,
with copper and titanium,
with iron,
with lead,
with lead and cadmium,
with magnesium,
with manganese,
with nickel,
with tin,
Alugalva (see Alloys of zinc, with
aluminum and silicon)
Aluminized steel compared to zinc,
Aluminum coatings compared to zinc,
Aluzinc or Aluzink (see Alloys of
zinc, with aluminum and
silicon)
Ammonia in air,
Annealing (see Galvannealed)
Anodes of zinc,
Anodes to protect zinc,
Anodizing of zinc,
Anti-sparking,
Appearance (see Surface appearance)
Applications (see Case histones)
ASTM specifications,
reactivation,
ASTM tests,
Atmospheres, classification, 101,
pollutant levels,
Atmospheric corrosion,
(see also Corrosion, types of
atmosphere; Corrosion rates)
Index 51
Auto body codes,
Auto durability,
Auto precoat systems,
Backfill,
Bacteria, (see
Battery cans,
Bending (see Fabrication)
Bimetallic contacts,
Bitumen coatings on zinc,
comparison with other coatings,
178-188
removal from zinc,
Boats,
Bolted joints,
Brass,
Breweries, corrosion resistance in,
Brick,
Bridges,
British Standards,
444
British Standards, guides
Corrosion, microorganisms)
underside corrosion,
BS on design for coatings,
BS on coating selection,
PD on bimetallic contacts,
PD on earth reinforcement,
1
BS on cathodic protection,
Bronze compared with zinc,
Building materials,
Buried structures, cathodic protection,
Case histories,
in air,
in concrete,
as earth reinforcement,
in water,
intergranular corrosion, 81
non-detrimental additions,
painting,
Castings, 5-6,
Cathodic protection,
anodes, 56-60,
with zinc paints,
Caustic soda, inhibitive additions to,
Cavitation corrosion,
Cement (see Concrete)
CEN Standard,
Chemical industry uses,
Chemicals, corrosion of zinc,
Chlorides in air,
cement stains, removal,
Chromate inhibition, 90,
Chromating,
accelerated tests, 90
replacements for,
Cleansers,
Coal terminals, applications,
Coatings of zinc, (see also
Corrosion; Corrosion rates)
economics,
heat-treatment,
life to first maintenance,
microsections,
Cadmium coatings compared with performance,
zinc, practical limitations,
selection guidance,
Calorized steel, thickness,
514
Index
Coatings on zinc surfaces, 8-9, 21,
34, 38-52 (see also Painting
on zinc surfaces)
Coke-chemical plant, 56, 208
Cold water systems, 337
Comparative performance, 178-195,
Concrete, 18, 60, 82, 97, 43743
zinc-aluminum alloy contact, 82
Condensation, 133, 274, 287-289
Conversion coatings (see Chromating;
Copper compared with zinc, 145,
Coral dust, 156
Corrosion features (see also Alloying
24 1-248
Phosphating)
189, 199, 242-243
elements, effect on corrosion
resistance; Corrosion
resistance; Impurities; Water
and aqueous solutions)
acid conditions, 63, 326
alkaline conditions, 18, 62-63, 82,
limitations for high-aluminum
appearance, 1 l , 27, 29, 36-37, 51,
atmospheric corrosion components
bending of coatings effect, 175
blackening, 220-221
brown staining, 233-234
carbon dioxide effect, 62, 140,
293, 295-296
cavitation, 8 1
chloride in air effect, 103,
159, 236
alloys, 82
55, 164
analysis, 167-170
134-135, 155-157, 172,
196197, 199, 202
chloride in water effect, 275
contact with other materials,
copper in solution effect, 67, 291,
67-73, 82
446
[Corrosion features, continued]
crack effects, 48, 175, 214,
crevices, 73-75
edges, 10, 16-17, 47, 49, 84-85,
218, 220, 222, 290
electroplating methods, effect on
corrosion behavior, 174
fatigue, 75-76
hardness effects in water, 263,
heat-treatment effect, 18, 80-8 1
height from ground effect, 163,
humidity effect, 121-125,
220, 438
270-274, 291-292, 299
196
133-135, 142-144, 147-148,
220
hydrogen embrittlement, 34, 78-81
hydrogen evolution, 79, 283
hydrogen induced cracking, 77
impurities in atmosphere effect,
inhibitors, 31, 39, 54, 80, 90, 155,
134-159
175, 209, 277-280, 299, 316,
320-321
initial conditions, 61, 164-167
maps, 116, 177
mass or thickness effect, 6
microclimate effects, 161-164
microorganisms, 276, 3
nitrogen in alloys, 75
nitrogen oxides in air, 103
orientation effects, 108, 114, 138,
oxygen effect, 63, 264, 275-276,
pervious materials, 446
pH effect, 18, 62-63, 77, 139,
161-174
283-285, 292-293, 296, 332
152, 263, 270-274, 292-296,
326, 347, 388, 440-441, 443
pitting, 6, 195, 292-294, 297,
pore effect, 214
299, 308, 317-318, 342-345
Index
515
products,
volume,
rain acidity effect,
rate with time,
rates,
(see also Corrosion rates,
individual environments)
reaction equations,
regression equations,
resistivity effects in soil,
shape effects,
samples,
sheltered conditions,
shuttered booths compared with
open atmosphere,
effect,
helix compared with flat
wire compared with sheet,
sprayed coatings extra surface
structure related,
sulfate-reducing bacteria,
sulfates in water,
sulfur oxides in air,
surface finish effect,
temperature effects,
testing,
thermodynamics,
time of day effect,
time of wehess effect,
time of year effect,
uniform dissolution,
vegetation effects,
wind effects,
Corrosion rateslresistance (see also
aluminum compared with zinc,
aqueous,
atmospheric,
Of,
Corrosion features)
East European data,
formula for atmospheres,
bronze compared with zinc,
cadmium compared with zinc,
chemicals,
copper compared with zinc,
design for,
detergents and cleaners,
film or scale formation, 5,
Galfan compared with zinc,
indoor atmospheres,
industrial atmospheres,
inorganic chemicals,
iron-zinc alloy layer,
location,
marine atmospheres,
6
Index
26220, 224-225, 229-230,
325
tropical, 176, 209
263-3 16
196-199
years, 106, 107
temperate compared with
non-saline water, 74, 178-188,
pollutant-related, 117-120,
reduced atmospheric rates in recent
rural atmospheres, 102-109,
116-118, 122-124, 129, 140,
178-192, 203-206, 212-214,
2 17-220
seawater, 71, 103, 178-188,
268-269, 312, 315-333
temperate compared with
tropical, 155
soil, 157, 341-369
splash zone, 157, 193, 207-211
steel compared with zinc, 76, 107,
189, 192, 196-197, 199,
122-124, 126-129, 145, 173,
242-243
tropical atmospheres, 102-104,
tropical surf, 210
urban atmospheres, 102-107,
122-123, 129, 203, 205-211
1 10-1 1 l , 116-120, 122-124,
126-127, 129, 175, 178-192,
194-198, 214-215
solutions)
162, 212
water (see Water and aqueous
wire compared with sheet, 138,
wire rope compared with wire, 212
wood, 82, 288, 439, 446
zinc dust particle size effect, 31
zinc purity effect, 62, 102, 172,
173, 212, 291
zinc, 51
Corrosion resistance, coatings on
Corrosion testing, 87-96
Corrosivity, 134-135
Costs (see Economics)
Cronak process (see Chromating)
Cutty Sark, 261
Damage resistance, 36-37
Deformation in processing, 36-37
Design for coating, 32-34
Design avoid wet storage stain, 65
Detergents and cleansers, 376, 387
Dew, 133 (see also Condensation and
Diecastings (see Castings)
Diesel oils, 436
Diffusion annealed, 19, 29 (see also
Diffusion coatings, 29
DIN Specifications, 6, 24, 239,
Drinks, zinc in, 371
Duplex coatings
Water, distilled)
Galvannealed)
297-298, 321
metal on metal, 28, 30, 38, 56
paint on zinc surfaces, 40-56,
synergistic effect, 43-44
178-195, 232-233, 246, 248
Durability, 36-37 (see also Corrosion
Dust, 157
resistance)
Earth reinforcement, 359-369
case histories, 366-369
Economics of coatings, 35-37, 44,
Edge protection, 10, 16-17, 47, 49,
84-85, 218, 220, 222, 290
Edinburgh Military Tattoo, 249-250
Electrochemical properties, 36-37,
189-195, 243-248
57-59, 67-70, 74, 87, 90,
94-96, 239-240, 299-306
Electrode potentials, 67, 71
Electrogalvanizing, 22 (see also
Electroplating with zinc)
Electromotive series, 71, 83, 84
Index 51 7
Electroplating on zinc, 50-51
Electroplating with zinc, 8-10, 13,
of high-strength steels, 80-81
zinc-chromiudchromium oxide,
zinc-cobalt alloy, 23-24
zinc-iron alloy, 23, 2627, 89,
zinc-nickel alloy, 22-23, 2627,
17, 21-28, 33-34, 37, 51
24, 27
230-23 1
46,49-50
Embrittlement, 78-81
Engine coolants, 81-82, 437
Environmental considerations, 4647,
Environmental control effects, 101
Equilibrium constants, acidic gas
reactions, 141, 157
Equilibrium diagram, zinc-aluminum,
20
98
Fabrication zinc-coated steel, 7-9,
flame or laser-cutting effect, 17
18, 20, 36-37, 175
Fertilizers, 388, 41W14
Finger-marking, 52, 375
Finishes on zinc (see Coatings on
Fire-extinguishing materials, 415
Flame spraying (see Spraying of zinc)
Food, zinc in, 371
Forming (see Fabrication of zinc-
Frictional characteristics, 9697
Fuel oils, 436
Fuel Research Method, 142
Fungicides, 410415, 443
zinc surfaces)
coated steel)
Galfan (see Alloys with aluminum
Galvalume (see Alloys with
and mischmetal)
Galvanic corrosion (see Cathodic
Galvanic series, 71, 83-84
Galvanized steel
choice compared with other
protection)
coatings, 178-195, 243-246,
33 1
contact with ungalvanized, 332
economics, 33, 244-246
heat-treatment, 18, 79-81 (see also
Galvannealed)
iron-zinc alloy layer, 18, 19, 26,
46, 62, 208, 211,
228-234, 292-294, 308-3 12,
322-324, 441
paint adhesion, 42
painting (see Painting of zinc
surfaces)
for water use, 291-312
wire rope, 326-330
Galvanizing, 8-21, 32-33, 37, 178
alloys, 15-22, 218-223
aluminum addition, 17-18, 54, 291
bath additions, 17-18
blast cleaning pretreatment, 80
design for, 32-33
high temperature, 18
hydrogen from pickling
pretreatment, 80
nickel additions, 22
process, 11, 80
quenching prior to painting, 54
Galvannealed. 19, 26, 46, 208,
308-312, 316, 322-325, 441
(see also Galvanized steel,
iron-zinc alloy layer)
Gases, 389409
52
generation under organic coatings,
Gasoline, 433-436
Gear oils, 436
General Motors Scab Test GM951 lP,
2,88-89
Index
Geographical locations, (Not
indexed individually. See
Tables 2.2-2.10, 2.20 and
2.21 for main atmospheric
sites.)
Glycerin,
Grain boundary corrosion (see
Grease,
Grounding cells, (see
magnesium compared with zinc,
Intercrystalline corrosion)
also Cathodic protection)
Guides (in Standards) (see British
Standards Guides)
Gypsum-containing materials,
Hard zinc (see Galvanizing, iron-zinc
Health limits for zinc,
Heat-treatment of zinc-coated steel,
Herbicides (see Fungicides)
High temperature galvanized coatings,
Hot dip galvanizing (see Galvanizing)
Hot shortness,
Hot water systems,
Houses of Parliament,
Humidity,
Hydraulic fluids,
Hydrocarbons (see Organic chemicals)
Hydrogen embrittlement,
Hydrogen sulphide in water,
alloy)
Imperial War Museum,
Impregnation of porous coatings, 50
Impurities in atmosphere,
Impurities in zinc, effect on corrosion
resistance
harmful, 5,
individual elements
[Impurities in zinc, continued]
aluminum,
bismuth,
cadmium, 5-6,
copper,
indium, 5 ,
iron,
lead, 5,
mercury,
nickel,
sodium,
thorium, 5,
tin,
Indoor environments,
Industrial atmospheres (see Corrosion
rates, industrial atmospheres)
Inhibitors,
list,
magnesium salts in sea water,
Inorganic chemicals, corrosion
effects,
inhibition by,
Inspection ease,
Insulating tapes,
Intercrystalline or intergranular
corrosion, 5,
Iron-zinc alloy in galvanizing,
ISOCORRAG,
IS0 Standards, 81, 90,
IS0 and (corrosivity
guides),
Jointing compounds,
Index 519
Lacquering,
Langeliers Index,
Lavegal (zinc-30% aluminum),
Liesegang bell jar,
Life of coatings (see Corrosion
Liquid metal attack (by zinc),
Lubricants,
resistance and Corrosion rates)
Madrid survey,
Magnesia cement,
Magnesium anodes (to protect zinc),
Magnesium salts, inhibition by,
Maintenance costs,
Marine atmospheres (see Corrosion
Mean effect,
Mechanical cleaning,
Mechanical coating,
rates, marine atmospheres)
definition,
Mechanical properties after corrosion,
Metallurgical structure (see Corrosion,
structure related)
Metal spraying (metallizing) (see
Spraying with zinc)
Methanollpetrol mixtures,
Microclimate effect,
MIL Specifications,
Mines,
Mist,
Mixed metal coatings,
Mortar,
Moss,
NACE Specification,
Nails,
NAPAP program,
NBS program,
Neutralite,
Nickel-plated zinc,
Nitrogen oxides in air,
Organic chemicals, corrosion by,
inhibition by,
Organic coatings (see Paints; Painting
of zinc surfaces)
Organic liquids, water in,
Ozone,
Paint, comparison with other
coatings,
characteristics and properties,
Paintability,
Painting on zinc surfaces,
adhesion to galvanized steel,
coil coating, 55
delamination, alloy effect,
economics,
interfacial reactions,
pH of suitable paints,
pretreatment, 53-55
removal,
synergistic effect,
undermsting limited, 55,
after weathering,
Paints, zinc dust,
for high temperature use,
zinc dust/zinc oxide,
zinc flake,
Passivating (see Chromating and
Phosphating)
Peen plating (see Mechanical coating)
Pesticides,
520 Index
Phase diagrams (see Equilibrium
Phosphate coatings,
Phosphates, corrosion behavior,
Pipeline protection,
Plasma sprayed coatings, zinc
underlay,
Plasters (see Concrete)
Plastics, comparison with other
Polluted atmospheres (see Corrosion
Potential reversal,
Pourbaix diagrams,
Powder coatings,
Powdering of coatings,
Pretreatment for painting (see
Painting, pretreatment)
Prices (see Coatings of zinc,
economics)
Properties (see individual properties)
Protective coatings (see Coatings of
zinc and special types)
Protective mechanisms,
Pulp and paper mills,
Pyrophosphate coating,
diagrams)
coatings,
rates, industrial or urban)
Q panels,
Rain, also
Water, distilled)
Rebar, galvanized,
Rebar, zinc for impressed current
Recycling,
Refrigerants,
Reinforcement for earth walls,
Relative humidity (see Humidity)
Reliability of coatings,
zinc content,
protection, 60
Repair of zinc coatings,
Ribbon anodes,
Roughness on wire,
Rural atmospheres (see Corrosion
Rust color,
Rust staining, removal,
Ryznar Index,
rates, rural atmospheres)
Sandelin,
Scratches, protection at, (see
also Edge protection)
Sealing of sprayed zinc,
Seawater (see Water; Corrosion rates,
seawater)
Severn Road Bridge,
Sheet,
Sherardizing
definition,
Ships, cathodic protection,
Size limitations for protective coating,
Soaps,
Soil corrosion,
Soot,
Sparking hazard,
Spraying with zinc,
arc compared with flame,
on concrete, 60
duplex metal coatings,
mechanism of coating failure,
mixed metal coatings,
powder compared to wire,
sealing,
zinc-aluminum alloy,
Index
521
Stains, removal
Steel compared with zinc in air,
Steel, protection methods using zinc,
risk of hydrogen embrittlement,
Storage, good practice,
Storage of organic chemicals in
Strain age embrittlement,
Stress corrosion,
Structure (see Corrosion, structure
related)
Sulfate-reducing bacteria,
Sulfur oxides in air,
galvanized containers,
catalytic effect of nitrogen oxides,
determination,
dry versus wet deposition,
isobutane effect,
reduction in recent years,
Sulfur trioxide,
Superplastic zinc,
Superzinc,
Surface appearance, 11,
Swedish Standards,
SMS 66,
11
Syndets,
Technigalva,
Testing,
accelerated, 64,
compared with natural,
cleaning solutions,
[Testing, continued]
corrosion fatigue,
Electrochemical Impedance
General Motors scab test,
Kesternich test,
long-term,
paints on zinc,
salt-spray test,
sodium chloride compared with
artificial seawater,
sulfur dioxide test,
Thermal spraying (see Spraying with
zinc)
Thickness of coatings,
Threaded components,
Towers,
Trichlorethylene,
Types of corrosion,
Spectroscope,
Uniform dissolution of zinc,
United Kingdom corrosion maps,
UNS Specification,
Urban atmospheres (see Corrosion
Uses (see Case histories)
rates, urban atmospheres)
Vapor deposition,
51,
Vegetation,
Vitreous enamel, comparison with
of aluminum (vacuum aluminizing),
of silicon on zinc,
other coatings,
Washington, DC,
Waste water treatment plants,
Water and aqueous solutions,
corrosion of zinc (see also
522
Index
[Water and aqueous solutions, cont.]
Corrosion of zinc surfaces;
Corrosion rates)
aeration, 287
agitation, 287, 293
alternate wetting, 207-208
anions and cations, 275-280, 292,
299-308, 326
artificial seawater, 321
bicarbonate, 302-307
brackish water (see seawater)
calcium (see hardness effect)
carbon dioxide effect, 293,
chemical solutions, 277-280, 376,
388-415
chloride effect, 275
condensation and dew, 133, 174,
copper in solution, 67
corrosion products, 274
distilled, 133, 274, 281-290
hardness effect, 263, 270-274,
295-296
287-289
291-292, 299
hot, 3, 73-74, 291-292, 299-302,
306-308, 338
hydrogen peroxide formation, 286
immersed compared with
alternating, 2, 62
industrial process waters, 3 13-316
inhibitors, 277-280, 299
microorganisms, 276, 3 I3
mineral waters (see natural waters)
natural waters, 266-267, 308-309,
nitrates in water, 275, 280,
Ocean depth effect, 326-331
organic constituents, 276
oxygen effect, 63, 264, 275-276,
pH, 18, 62-63, 263, 270-274,
313-316, 338
303-306
283-285, 292-293, 296, 332
,293-296, 320, 326, 388
[Water and aqueous solutions, cont.]
potable water, 285, 291-313,
potential measurements, 299-307,
potential reversal, 64, 85-86,
pressure, 283, 286, 337
rainwater, 289-290
rates, limited significance, 291
remineralising, 265
river waters, 266-267, 313-316,
338
scale-forming, determination,
270-274
seawater, 71, 178-188, 207-208,
268-269, 312, 315-333
silica in water, 307
steam, 3, 75, 265, 336
sulfates in water, 275, 303-304,
supply or tap waters, 285,
surface waters (see natural waters)
suspended matter, 277
temperature, 73, 264, 282-287,
337-338
336
299-306, 355-356
307
291-313, 337-338
291-292, 297, 299-308,
336-337
turbulance, 315
types, 263-270
Water stains, removal, 65, 374
Wear resistance (see Abrasion
resistance)
Weathering (see Atmospheric
corrosion)
Weathering steels, 242
Weld protection, 191, 195
Wet storage stain, 28, 52, 64-66,
121, 372-373
effect on painting, 52
removal, 65
Whisker growth, 23
White rust (see Wet storage stain)
Wire ropes, 75
Index
523
Wood,
Wrought zinc materials,
ZA alloys (see Alloys of zinc)
Zalutite (see Alloys of zinc with
Zincalit (see Alloys of zinc with
Zincalume (see Alloys of zinc with
Zinc coating methods, (see also
individual coating methods)
Zinc dust or flake in painting,
Zinc filler for bitumen, 444
Zinc flake,
Zinc foil (as a coating),
Zinc-iron alloy coatings (see
Zinc powder,
Zinc purity effect,
Zinc refining atmospheres,
Zinc-rich paints (see Paints, zinc
Zincrox,
Zinc salts, inhibitive effect,
Zinc soaps for castings impregnation,
Zinloy, 24
Galvanizing; Electroplating)
dust)

Frank C. Porter Corrosion Resistance of Zinc and Zinc Alloys 1994

Enviado por

Dados do documento

Descrição original:

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Frank C. Porter Corrosion Resistance of Zinc and Zinc Alloys 1994

Enviado por

Direitos autorais:

Formatos disponíveis

Corrosion

Você também pode gostar