.r.
y... . :9va 4D .
¢ . \xiiil. 11.9.‘1‘g
Mmikfitblrer tier-‘9
1 e xvii . .: I!

p.
. ‘ , . .t r
Pt... DH”

{:91 1:. gig“
:zlrtvxlo ct!!.$!1
”rs-)3-
! it...
. 3.

'§,
Kari.
. t1.»
. :2 5.51.:
33?”?!
o, t; :
). ?¢n¢¢:.).€!. .
l..- o. . 4.50.9”

32:
f-§l\3‘l’.\€

9.! l. I.) It‘-

.4. is... U? .

2&2. .R n

.f

. .. v \. ‘ 1.: .o .r...
:2. . :{L‘Zx 2. .P . . y it... ...Zv)$x.§.l.$l
. .I z :7; . I. 3“ ‘ .17.!!! 53.1.3.5 .6...-¢.llc.9
4 ‘ I 3:»... u ‘3...“‘3. . .3... .51313232331‘
752%.... ifsr 5.0:. rt

: L‘rsfiv .e
xi 1...:
:31.
.l. .
ttv. . II; .I.
3-11 " 8’sz‘}f‘ n..-
6; » I!!!)xnrvezi ..

L. r . .|\ I . .
. ‘ in“: , . ‘
, ‘ . . . ‘ ‘ . v V .val {333’
.. , . 535.111 :5. 5k

al.'<!:r i

Ital-i

71:73"!-
.... .}.. 2.14

 

, 3 $4.! “$3.3
.mftsifiuurficwuF....n..-.3 1 i. _

‘II.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ilnllu'l‘l'ilul‘”Infilllfllllll i 7‘ 5 05 9491 ‘l

31293 005741131

 

LIERARY
Michigan State
University

 

'1 his is to certify that the

thesis entitled

Aqueous Corrosion of AA-2024 By Zinc Coated

Steel Fasteners

presented by

Mahesh Rajayya

has been accepted towards fulfillment
of the requirements for

Master's Materials Science

degree in

 

 

Major professor

 

 

Date-Q / N Wew'é‘? /’7 6";

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

"I'-

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

DATE DUE DATE DUE DATE DUE

ll

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_l
_JL44

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution

 

AQUEOUS CORROSION OF AA-2024 BY ZINC COATED
STEELFASTENERS

BY

MAHESH RAJAYYA
A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of MetallurgY. Mechanics 8: Materials Science
1989

i ..P

V'-

\t"
m

.
.
\j

ABSTRACT

Aoueous CORROSION OF Alt-2024 FASTENED WITH Zn COATED
STEEL FASTENERS

Zinc is more widely used as a protective coating for steel in
automotive and aircraft industries, but cadmium has been preferred in aircraft
structures for various reasons. However. cadmium has several disadvantages
Over zinc coating which are environment related. The parts which are made Of
AA-2024 fastened with zinc coated steel fasteners have been found to suffer
from serious corrosion attack. The corrosion behavior of AA-2024 fastened
with zinc coated fasteners should be understood properly before choosing the
proper coating material. The zinc coating on the fasteners is to anodically
protect the aluminum alloy from the corrosion. because the coating is consu-
med and rapid corrosion of aluminum alloy results.

A cyclic wet/dry exposure test with sodium chloride soluti-
ons. galvanic and dripping dlectrolyte experiments are conducted to Observe
the corrosion behavior. The parameters Of the experiments are explained, and
the performance Of the exposed metals are reported in terms of weight loss and

galvanic currents Observed.

Dedicated
to
My Father, Mother and Sister

iii

TABLE OF CONTENTS

1 . Introduction

2. Background

3. Experiment

3.1 AA-2024 product classification

3.1.1 Composition of AA-2024

3.1.2 Physical 8: Mechanical properties Of AA-2024
3.1.3 Sampling proceedure

3.1.4 Grain orientation

3.1.5 Location of samples

3.2 Alternate Immersion Test

3.2.1 Summary

3.2.2 Apparatus

3.2.2.a Cycling mechanism

3.2.2.b Rate Of Immersion

3.2.3 Specimens

3.2.3.a Preparation

3.2.3.b Identification of specimens
3.2.3.c Number of replicate samples
3.2.3.d Size and area of samples
3.2.3.a Mounting of samples

Page

10

12

3.2.4 Test solution

3.2.4.a Volume of test solution

3.2.4.b Aeration Of test solution

3.2.4.c Temerature of test solution
3.2.4.d Evaporation loss of test solution
3.2.4.e Air circulation

3.2.4.f Test duration

3.2.4.g Cleaning of samples

3.3 Potentiostatic Anodic Polarization Measurement
3.3.1 Apparatus

3.3.2 Anodic polarization circuit

3.3.3 Experimental poceedure

3.4 Galvanic Experiment
3.5 Dripping Electrolyte Experiment

4. Results 8: Discussions

4.1 Alternate Immersion Experiment
4.2 Galvanic Experiment

4.3 Dripping Electrolyte Experiment
4.4 Potentiostat Experiment

5. Conclusions

5. ROTOI’ODCOS

Page

20

24

24

26

48

50

Table

List of Tables

Chemical Composition of AA-2024 ..,__ ., w .
Physical & Mechanical properties of AA-2024

Results of weight loss experiment from alternate Immersion
experiment” - 1

Reduction factor for sodium chloride solutions with varying
% of ZnCl2 as an additive

Results of the alternate immersion test of AA-2024 fastened
Results of weight loss from galvanic experiment A. .1 _. . . M

Results Of weight loss from dripping water/ electrolyte experiment

Page

II

27

28

4|

4|

List of Figures

Figure Page
1. Experimental set-up for alternate immersion cyclic test 14
2. Identification of samples by using notches.
3. Corrosion cell for potentiostat experiment . 15
4. Anodic polarization wiring diagram a, . , _ . 2,
5. Schematic diagram for dripping electrolyte experiment . ‘_ 23
6. Effect Of concentration Of NaCl with/without ZnCl2 on corrosion
7. Optical micrograph Of an uncorroded sample of AA-2024 etched

with Keller’s reagent . _ r. _, . . 30
8. Scanning electron micrograph of AA-2024 maintained in 0.1 M

NaCI without ZnCl2 .. . , .. _ .. . .. 32
9. Scanning Electron micrograph Of AA-2024 maintained in 0.1 N

NaCIwith0.1°/oZnCI2 . ‘ 34
10. Scanning electron micrograph Of AA-2024 maintained in 0.01 N

NaCl with ZnCl2 ‘ 34
11. Scanning electron micrograph of AA-2024 maintained in 0.01 N

NaCl without 0.1%an2 ~ H - - ~ 36
12. Optical micrograph of a corroded sample Of AA-2024 fastened

with zinc coated steel fasteners kept in 0.01 N NaCl . - 36
13. Optical micrograph of a corroded sample Of AA-2024 fastened

with zinc coated steel fasteners kept in 0.001 N NaCl ~ ~ - ~ 30

Figure Page
14. Time-Voltage plot of Galvanic experiment of AA-2024 and

galvanized sheet maintained at various concentrations of

sodium chloride . . . . , . . . 4o
15. Optical micrograph Of the corroded AA-2024 from Dripping

electrolyte experiment - -- - - , 43
16. Anodic polarization curve of AA-2024 with/ without ZnCI2

maintained at 0.1 N NaCl . ’ . 46

VIII

1. INTRODUCTION:

This thesis describes the corrosion behavior Of aluminum alloy
(AA-2024-T3) with zinc- coated steel fasteners, with or without the presence of
chloride ion. To evaluate their use as an alternative to cadmium on fasteners.
the properties (especially corrosion resistance ) of zinc coatings are examined.
The plating catholically protects steel bolt and prevents galvanic corrosion
at contacts with aluminum alloy. Zinc is the more widely used coating for
steel but cadmium has been preferred in aircraft structures for its excellent cor-
rosion rersistance. However it suffers from three major draw-backs:
1) Its ions are toxic. the body accumulating the metal from very low dose rates.
2) Cadmium can embrittle high strength steels.
3) The coating process is costlier, hence electro-plating is put out of business.
Hence. in order. to find a substitute for cadmium coatings. zinc coatings are
considered. Zinc-coated fasteners used in automotive industries, but
these fasteners are inferior to those of cadmium coated ones. with respect to
corrosion.

In order to investigate the corrosion mechanism of zinc-coated steel faste-
ners with AA-2024, galvanic compatibility tests and alternate immersion cycle
tests have been found useful. because these tests accelerate the corrosion
process and obtain the results in a short period of time. The other advantage
of these tests is the condition used in operation can be reproduced.

When zinc-coated bolts and aluminum alloy plate are immersed in an ele-

2
ctrolyte, and elctrically connected, a galvanic current is produced. Hence
by measuring the galvanic current as a function Of time and calculating the
weight change of aluminum alloy electrode.the galvanic compatibility Of
aluminum alloy with zinc-coated steel fasteners can be assessed. Zinc coa-
ting protects aluminum alloy for the most of the exposure period. When the
zinc coating corrodes, it will eventually expose the underlying steel metal
surface which leads to the galvanic corrosion of aluminum alloy. The results
obtained from the galvanic experiment reveal the same.

By measuring the weight loss of the exposure period, the exact corrosion
damage occured to aluminum alloy can be calculated. The corrosion test on
the bolted assembly of aluminum alloy and fastener is also performed. Alter-
nate immersion cycle test is chosen for this purpose. The results Obtained by
alternate immersion tests are in parallel with the results obtained from the gal-
vanic experiment. TO substantiate the above results from the above experi-
ments, a simple test is conducted. in which the electrolyte is allowed to drip on
the galvanized sheet and the same solution then made to fall on the aluminum
alloy plate. The weight loss results indicate that aluminum alloy is corroded
by ferrous ions which are dissolved from the galvanized steel sheet.

A model is proposed for the corrosion mechanism of the aluminum alloy
with zinc coated steel fasteners. Furthermore. some suggestions have been
made for the future work to minimize the galvanic corrosion Of the fastened as-

sembly.

2. BACKGROUND:
Although the corrosion phenomenon is well understood by aircraft manufa-

cturers and operators , most current aircraft types continue to exhibit many of

of the fundamental corrosion defects shown by earlier designs [1]. For many
years, the aircraft industry has confronted varied corrosion problems, largely

due to the trend toward lighter and higher strength alloys which present increa-

sed corrosion susceptiblity because of their inherent qualities [2].
Aircraft corrosion problems are of three kinds:

1) wet and moist corrosion of unprotected metal,

2) wet and moist corrosion of protected metal subsequent to pro-

tective coating failure, .

3) corrosion caused by human factors. The corrosion rates of the

aircraft are influenced by weather conditions, atmospheric pollu-

tants and the nature Of the metal [3].

During operation, the wing stucture of the aircraft which is partly made of
AA 2024-T3. is subjected to in-flight mechanical stress and also influenced
by ambient conditions such as, atmospheric humidity, sometimes associated
with salinity with posssible stagnation of moisture inside cavities not airtight
toward the exterior. These ambient conditions are conducive to coroslon

phenomena, which starts either from external surfaces or from intemai ones

4
subject to moisture stagnation or to other local negative factors, may then

rapidly propagate themselves to other areas of the structures [4].

Environmental effects both in air and on the ground are very important fac-
tors in determining the ability of aircraft structure tolwithstand corrosion dam-
age. In a dry weather, corrosion damage is low, however. if the same aircraft
is exposed to a wet and saline climate, the corrosion damage is comparatively
higher than the previous situation. During the service. moisture condenses
and airborne salts tend tO accumulate on skin, crevices and in structural en-

closed areas such as flap cavities and undercarriage bays [5].

Corrosion of fasteners and adjacent material has long been a problem
and its forms are well documented. Seizing and gelling make fastener removal
difficult and costly. A 1969 survey shows that airlines are spending 50-100
man-hours per airplane overhaul and 14-20 per maintenance check in drilling
out ”frozen screws". A joint that is completely sealed and protected against

moisture seldom gives problems [6].

Most aircraft are protected with a finish coating which, of course, extends
over the fastener patterns on the outside skin. If this coating could be kept in-
tact. the countersink areas of the skins in which the fasteners are installed
would not be exposed to corrosive media, hence there is no corrosion damage.
This of course. is not the case. Most aircraft finish coatings tend to crack
around the fastener heads and thus expose countersink areas to moisture and

other conceive media [7].

5

A common cause for aicraft corrosion is when the paint is deteriorated
or plating and protective finishes are damaged and base metal is vulnerable
to corrosion. Most common forms of corrosion are electrolytic in nature and
once cOrrosion starts and affects the internal structure Of thematerial it can
continue even through the surface. It is an unfortunate fact that aircraft mat-
erials which require a high strength/ density ratio are generally all susceptible
to corrosion

In general the higher the strength the higher the susceptibility to attack.

The most common forms of corrosion which are encountered by

aircraft are: [1], [a], [9], [10], [11], [121,[131 :

1. Fitting corrosion:

When corrosion is localized in well-defined areas and such areas are relative-
ly small as compared with the whole surface, they are defined as pits. Their
size, depth and number vary tremendously. For pitting corrosion to occur an
electrolyte must be present. Oxygen must also be present to initiate the pitting
corrosion [12]. Pitting corrosion is observed on wing and fuselage skin. air
intakes. and air intake ducts.

2. Intergranular corrosion:

Intergranular or lntercrystalline attack is selective corrosion of the grain bound-
aries or closely adjacent material without appreciable attack on the grains

or crystals themselves. The mechanism probably involves a difference in
potential between the grain boundaries and the grain. in alloys such as alumi-
num-copper, the precipitates such as (CuAlzand Alx Cu; Mg) are cathodic to

the depleted solid solution [13]. This type of corrosion is encountered in wing

and fuselage skin parts of the aircraft.

3. Exfoliation corrosion:

Exfoliation, also called layer corrosion or lamellar corrosion, is a type of selec-
tive subsurface attack that proceeds along multiple narrow paths parallel to
the surface of the metal. The attack is usually along the grain boundaries. but
it has also been seen along striations of insoluble constituents that have strung
out in parallel planes in'the direction of working. Exfoliation is characterized
by leafing, or alternate layers of thin, relatively uncorroded metal and thick
layers of corrosion product that are more bulky than the metal from which they
came [13].

4. Galvanic corrosion

Two dissimilar metals immersed in the same electrolyte will usually differ in
potential. If they are brought into contact, current will flow from the. less noble
or more active metal through the solution into the more noble or less active
metal. The amount of corrosion of less noble metal above its normal rate is
called galvanic corrosion. For each environment. a galvanic series can be
constructed in which metals are arranged in order of their corrosion potential,
with the most active metals at the top, and the most inactive metals at the
bottom. This type of corrosion is observed in the vertical stabilizer at the

tail section and cadmium coated steel jaw bolts under the main wings which

results in damage of aluminum alloy parts [10].

5. Stress corrosion:

This type of corrosion occurs in alloys which are susceptible to cracking when
under tensile stress and exposed to corrosive environment. The failure

may involve an electrochemical mechanism, attack by a molten phase, hydro
gen embrittlement, or some other factor [11]. The stress corrosion cracking

failures of aluminum alloys are always intergranular [13]. When stressed in the

7
longitudinal or short transverse directions, stress corrosion cracking occurs

when stresses in the order of yield stress are present [12] .This type Of corro-
sion is seen at the tailwing which is connected to the fuselage and stablizer
part and fastener assembled areas in the wing part of the aircraft [10].

6. Flllform corrosion:

This is a special case of crevice corrosion that may occur on an aluminum sur-
face under an organic coating. It takes the form of randomly distributed thread-
like filaments, and the growth of these filaments probably involves a differential

aeration cell. The corrosion products formed raise a bulge in the surface coat-
ing, much like molehills in the lawn [13] . This type of corrosion is Observed

underpaint schemes usually adjacent to fastener heads [14].

7. Microbiological corrosion:

Micro organisms which live and breed in water contained in jet fuel can cause
serious corrosion of metal surfaces in integral tanks. The micro-organisms form
dark colored slimy sludge in the fuel tanks and can deteriorate fuel tank seal-
ant, protective treatments and corrode tank pipes and wing structure [14] .

To achieve the most efficient aircraft structure, it is necessary to use
many different metals which are usually in electric contact. Thus there are
more chances of having galvanic or electrolytic corrosion, for instance
corrosion of aluminum alloy in contact with copper rivets. While it is possible
to overcome galvanic corrosion by theoretical approach, it is not possible in
practice. The components which are joined by means of fasteners are highly
susceptible to corrosion in the area of joint, if materials having a dlferent
potential are joined to be electrically conductive [15].

8
For choosing a proper coating for an aircraft fastener application two im-

portant factors are considered,
1) corrosion characteristics of the coating material
2) galvanic compatibility with aluminum alloy [16]. Cadmium and zinc coatings
are used as the coating material on steel bolts for fastener purpose. The plat-
ing catholically protects the steel bolts and prevents galvanic corrosion at
contacts with aluminum alloys. When fasteners are exposed to marine atmos-
phere the corrosion resistance of cadmium coated bolts is better than the zinc
coated bolts, and the galvanic compatibility of cadmium coated bolts with
AA 7075-T6 is better than the zinc coated ones.

However. cadmium coated bolts tend to suffer from two major draw-backs:

cadmium ion is toxic, the body accumulating the metal from very low dose
rates [17]; and cadmium can embrittle high strength steels [16]and titanium

alloys [19]. The results Obtained from the marine atmosphere conflict with the
results obtained from the laboratory, hence to formulate a rapid screening test
for galvanic compatibility, the currents generated and weight loss which
occur in galvanic cells between the bolts and the aluminum alloy electrode
are measured [16]. The other test which is employed is alternate immersion
cycle experiment. This method was adopted by Standard Pressed Steel Co.,

in order to screen fastener materials and coatings for their resistance to corro-

sion and subsequent stress corrosion of loaded bolts. The advantage of the
laboratory tests is that the conditions used can be reproduced precisely and
the tests could be accelerated. but their relevance must be established by

parallel natural environmental tests [20] .

9 i“

 

Monel fasteners [23] are used with AA-2024 in the wing structure of
the aircraft. The purpose of zinc coating is to anodically protect aluminum
alloy from corrosion. On exposure to serious environmental conditions, zinc
coating corrodes and expose the cathodic steel surface to aluminum alloy.
Hence the corrosion of aluminum alloy is accelerated. ln C-130 type aircraft
[21], high strength aluminum alloys such as 7178-T6 contain appreciable
amount of elements other than aluminum. As a result, they develop anode and
cathode areas when in contact with an electrolyte. Corrosion usually origina-
tes in countersink areas of the fastener holes and first attacks the and grains.
The grain boundaries being anodic to the grains. are converted to corrosion
products. These products occupy a greater volume than the original metal and
exert tremendous pressure with subsequent expansion within the corroded
area. The corrosion then follows a laminar pattern along the grain boundary

and destroys the structural integrity of the metal.

There is an evidence to substantiate the galvanic corrosion between
aluminum and galvanized steel. Overhead electrical power transmission con-
ductors used in Great Britain constructed from aluminum wires centrally re-
inforced by galvanized steel strands. in a few regions are found to suffer from
intemai corrosion. which is associated with the presence of chloride ion [22].
The intemai corrosion manifests itself localized attack on the galvanized steel
strands. The attack is accelerated after the initial deterioration has started.
Initially, as the galvanized steel zinc layer is anodic to aluminum, no signifi-
cant corrosion on the current carrying strands occurs. However when the
underlying iron-containing layers are exposed, corrosion of aluminum com-

mences. as it is then anodic to the steel.

3. EXPERIMENTAL SECTION:

3.1 Ali-2024 product classification:

AA-2024 is a mill product. and it is in the form of a sheet. The temper
classification number is T-3. These alloys are solution heat treated, cold
worked, and naturally aged to a substantially stable condition and it also to
applies to products that are cold worked to improve strength after solution heat

treatment, or in which the effect of cold work in flattening is recognized in

mechanical property limits [23].

3.1.1 Composition of AA-2024:
The composition of AA-2024 is given in table 1. [24].

3.1.2 Physical & Mechanical properties of AA-2024 depending on its thickne-

as is given in table 2.

3.1.3 Sampling procedure:
Mill finish flat AA-2024 was sheared to 5cm x 1.5cm, with long edge par-
allel to the rolling direction. Specimens were taken from a bulk sample so that

they were representative of material that was tested.

3.1.4 Grain orientation:
The grain structure tends to be equiaxial in the longitudinal direction al-

ong the sample surface.

10

Table l - Chemical Composition of AA-2024

 

 

 

 

Element Atomic Conc. (%)
Silicon 0.5
Iron 0.5
Copper 3.8-4.9
Manganese 0.3-0.9
Magnesium 1.2—1.8
Chromium 0.1
Zinc 0.25
Others Each 0.05
Others Total 0.15
Aluminum 90.9-93.2

 

 

 

Table 2 - Physical and Mechanical Properties of AA-2024

 

 

 

 

Density Specific Modulus Modulus Minimum Mechanical Properties (Ksi) Endurance Brinell
of of Number
Gravity Elasticity Rigidity Tension . Shear Bearing Limit 500 Kg
10 mm
lb . . .
(71;?) (P31) (P31) Fty 51008 % Flu Fey Fbru Fhry (K31)
0.1 2.77 10x10‘ 4.0x10‘ 42 12-15 40 24 124 76 20 120

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11

 

12
3.1.5 Location Of samples:
Short transverse specimens were taken from the sheet 2 12 plate
thickness away from a side of a plate (The side of the plate is defined as the

edge parallel to the rolling direction).

3.2 Alternate Immersion Test:
3.2.1 Summary Of the test:

Alternate immerSion test is an accelerated corrosion test to predict the
performance of a given material and is a representative of certain natural con-
ditions. This test involved 1 hour cycle, inclusive Of a 10 minute period in eq-
ueous solution of different cocentration of sodium chloride followed by a 50 mi-
nute period out of the solution, during which samples were allowed to dry. The

above cycle was repeated for 7 days i.e., 166 hours.

3.2.2 Apparatus:
3.2.2.a Cycling mechanism:
There were various methods available to achieve the cycling mechan-
ism. In this experiment, the following technique was adopted: The samples
were plaaced in a movable rack, it was periodically lowered to and lifted
from the beakers containing solution, by means of timer controlled motor as

shown in fig.1.

3.2.2 b Rate of immersion:
The rate of immersion and removal of the samples from the solution sh-

ould be as rapid as possible. It was about 15 seconds in this experiment.

I3

Fig.1. Experimental set-up for alternate immersion cyclic test

 

 

 

 

 

15

3.2.3 Specimens:
3.2.3 a Preparation:

The sheared edges of the samples were refinished by wet grinding prior
to testing. Since the sheared edge of the samples were subjected to cold wor-
king and might possibly fracture the edges, they were polished. Mill scale
on the as-received surface should be removed by polishing. All the surfa-
ce of the samples including the edges, was surface finished with 240, 320, 400
and 600 size grit paper. Then they were further surface finished with 1, 0.3 and
0.05 micron size iron-free aluminum oxide abrasive powder in the lapping whe-
els. Care must be taken to avoid excessive heating during surface preparation
because this might include undesirable residual stresses and in some cases it
might cause metallurgical or chemical changes on the surface.

The edge of the samples were polished in the same way as the faces
were. Prior to weighing and measuring, the samples were de-greased with
organic solvent such as acetone, then the dimensions were measured to the
accuracy of 0.01 mm., rinsed with acetone, and weighed to the accuracy of
0.1 mg. After degreasing the samples were handled with gloves till they were
exposed to the solution. Prior to exposure of the samples to the test solution,

samples were rinsed with test solution.

3.2.3.b Identification of samples:

Samples were identified by means of notches made around the edges of
the specimen as shown in fig.2. The notches were made in such a way that
they follow binary code. A reference notch was made either in the lower or
upper corner of the specimen. The value Of the upper comer notch was zero,
whereas the value of lower corner notch was fifteen. The upper or lower edge

notches were assigned the following values based on their respective position

1's

 

 

 

 

Fig. 2. Identification of samples by using notches

I7

the following method was used in identifying value of notches.

The value of the first upper edge notch closest to the upper corner notch
was: 2° :- 1.

The value of the second upper edge notch closest to the upper comer notch
was: 21 =- 2.

The value of the third upper edge notch closest to the upper corner notch
was:22 . 4.

The value Of the fourth upper edge notch closest to the upper corner notch

was: 23 = 8.
The reference point for the upper or lower notch was with respect to the hole
which is made on the top, to hang the samples.

Similar approach was used to figure out the value of a notch which was
made on the lower edge of the sample. Hence with this approach thirty sampl-

es were identified without any ambiguity.

3.2.3.c Number of replicate samples:

The best practice was to use one sample for each container. Considering
the space constraint, experimental design and number Of samples tested, it was
appropriate to use as many as three samples in each beaker. The two replicate
samples which were tested in the same way would give more accurate results
about corrosion rate. It was taken care that the samples tested belong to the
same batch and the specimen support prevents the specimen from touching
each other.

I8

3.2.3.d Size and area Of the samples:
The samples were of rectangular shape and the sample size is 5.0 x 1.5
cm. A hole is made on the sample and the diameter is 3.0 mm. The total surface

area exposed to the sample is given by the following equation,
Aa2(lw+wt+lt-1tr,2)+21rr,t=2(lw+wt+lt)-21tr,(r1-t)

In case of sample having two holes for fasteners and one hole for hanging it,

the total surface area exposed to the solution would be,

A a 2 ( lw 4» wt + If ) -21rr,2 -4nr.‘,2 + 21cm + 41trzt

=2(lw+wt +lt)-2n(r,2+2r22)+2m(r, +2r2)
where,
l- length of the sample,
w - width of the sample,
t - thickness of the sample,

r1- radius of the hole to hang the sample,

rz- radius of the hole to clamp the fasteners.

3.2.3.a Mounting of the samples:

The support device and container should not be affected by or cause any
contamination of the solution. The method of supporting should be designed to
insulate the specimens from touching each other physically or electrically.

The shape and size of the specimen supports and holders should be such that:
1. They avoid , as much as possible, any interference of free contact of the sp-
ecimen with the salt solution.

2. They don't obstruct air flow over the specimen, thereby retarding the drying
rate.

3. Drainage from one sample does not directly contact any other sample.

19

3.2.4 Test Solution:
3.2.4.a Volume of the test solution:

The volUme of the test solution should be large enough to avoid any
appeciable change in its corrosivity during the test by means of accumulation

of corrosion products that might affect further corrosion. The volume of sol-

ution/specimen area was 28 ml./ cm2.
3.2.4.b Aeration of solution:

The solutions which were employed in this test were not de-aerated and
the test simulates the actual service condition in which the material would be
exposed.
3.2.4.c Temperature of the solution:

The temperature of the solution was around 23 +/- 1.
3.2.4.d Evaporation loss of the solution: .

The average evaporation loss was 1 - 1.5% Of total volume of solution.
The loss was taken care by means of adding distilled water NOT the salt solu-
tion, everyday. The simple way of doing this was to refill the solution to a liquid
line and refill to that line everyday.
3.2.4.a Air circulation:

It is an important factor, because it affects both the rate at which the sa-
mples dry and the evaporation loss of water from the solution. Forced air blast
was avoided because of the difficulty in maintaining uniform drying of the sam-
ples. Stagnant air condition was also avoided.

20

3.2.4 f Test duration:

The duration of the test was based on the inherent resistance to corrosion
of the alloy, configuration of the test specimen and the nature Of the test solu-
tion. In general, the duration of the test should be as long as possible,
commensurate with the resistance Of the material under the test. The minimum
duration of the test in hours was approximately 50 divided by the expected
corrosion rate in millimeters per year. The duration of the test was 168 hours.
3.2.4.g Cleaning Of the samples after experiment:

The following solution was employed in cleaning the corroded samples:
Chromic acid: 20 grams
Phosphoric acid: 50 milliliters
Water: 1000 milliliters

Temperature of the solution: 80 ° C

The samples were placed in the solution for about 5 minutes and then
washed with cone. nitric acid for about a minute. Then they were rinsed with
distilled water to remove the loose coating on the sample surface. Then they

were weighed to the accuracy of 0.1 mg.

3.3 Potentiostatlc Anodic Polarization Measurement: [26]
3.3.1 Apparatus:

The test cell was constmcted of following items:
1) Working electrode
2) two auxiliary electrodes
3) reference electrode with a salt bridge connection.

A suitable cell is shown in fig. 3, a round bottom flask has been modified

by the addition of various necks to permit the introduction of electrodes, gas in-
let and outlet tubes and thermometer. The salt bridge separates the bulk solut-

THERMOMET ER

SALT BRIDGE
CONNECTION

PROBE

WORKING
ELECTRODE

 

 

 

 

 

 

 

 

 

GAS OUTLET

GAS INLET

AUXI LIARY
ELECTRODE '
HOLDER

Fig. 3. Corrosion cell for potentiostat experiment [25]

22

ion from the saturated calomel electrode.

Potentiostat:

A potentiostat that will remain as electrode potential within 1mV of a
preset value over a wide range of applied currents should be used. For the
type and size of standard specimen supplied, the potentiostat should have a

potential range from -0.6V to 1.6V and an anodic current output range from 1.0

to 105 uA.
3.3.2 Anodic polarization circuit:

A schematic anodic polarization wiring diagram is illustrated in fig. 4.
The working electrode was a thin strip which was cut from the AA-2024 plate.
The auxiliary electrodes were two high-density graphite rods. At the end Of
every run, the rod was scmbbed with 400 size grit paper. The reference
electrode was a saturated electrode calomel electrode.

3.3.3. Experimental prceedure:

The temperature of the solution was maintained around 30° C. The oxy-
gen level in the solution was reduced by means of bubbling nitrogen gas for 30
minutes. The working electrode was prepared 1 hr before the experiment. The
samples were wet ground 240- grit SIC paperand wet polished with 600 - grit
SiC paper until previous coarse scratches were removed.
The dimensions of the sample were measured to the nearest 0.01 mm.
Then they are degreased. The sample was mounted with the help of a nibber
gasket. Then sample was transferred to the corrosion cell and the salt bridge
probe was adjusted so that it was 2 mm or 2 times the dip diameter.
The open-circuit specimen potential was recorded after 55 min im-
ersion. Start the potential scan 1 hour after specimen immersion. The potenti-
al step of 50 mV was used every 5 min., the current value was recorded at the

end of each 5 minute period. These steps were were repeated until a potential

23

 

 

 

 

POTENTI OSTAT
C—_ '0 won KING
( ¢AUXILIARY REFEKDCE?

 

L_

 

 

POLARIZATION
REFERENCE

 

 

 

 

 

CELL

 

 

 

 

Fig. 4. Anodic polarization wiring diagram [26]

24

+1.6 V SCE was reached.
Then using the data obtained from the experiment, a plot was made.

The plots Obtained from this experiment is given in the results section.

3.4 Galvanic Experiment:

A galvanic current was generated when dissimilar metals immersed in an
electrolyte are electrically connected. The measurements of current
generated and weight loss occur in galvanic cells between zinc coated
steel electrode and aluminum alloy electrode was shown to be a useful and
very rapid screening test for galvanic compatibility.

To assess the compatibility of the zinc coated bolts with AA-2024, galva-
nized steel plate was kept at a fixed distance (about 40 mm) from AA-2024
2024 in aqueous chloride solution, electrical connections were made through
the galvanized steel electrode and AA-2024 electrode to a multimeter. The
voltage generated was continuously recorded for every 8 hours till 48 hours

period and the weight change of AA-2024 electrode was then determined.

3.5 Dripping Electrolyte / Water Experiment:

This was a very simple and efficient method to ascertain the compatibility
of zinc coated steel fasteners with AA-2024. In this experiment the electrolyte
or distilled water was made to fall on the galvanized steel sheet, and then
eventually drips on AA-2024 plate. The flow-rate of the electrolyte was regul-
ated by means of a glass stopper. At the end of the experiment aluminum alloy
plate was cleaned by chemical means, accurately weighed and the corrosion

rate was determined. ( Experimental set-up for the experiment is shown in fig.5)

25

 

 

 

 

 

 

 

 

 

 

 

 

-: : : : :- DISTILLED WATER! ELECTRQYTE
— - "' T '- ‘T I
W
«ng‘s‘flwm
U
[L GALVANIZED“
ii

 

 

 

 

 

 

 

 

 

 

 

 

Ll

Fig. 5. Schematic diagram for dripping electrolyte experiment

4. Results 8. Discussions:

4.1 Alternate Immersion Experiment:
The corrosion rate is calculated from the following equation:
The average corrosion rate (in mpy) - ( K x W ) / (A x T x D )

where,

K - a constant , 8.176 x 106 in millimeters per year unit,

T - time of exposure in hours to the nearest 0.01 h,

A - area in cm2 to the nearest 0.01 cmz,
W - mass loss in grams,
0 - density in g/ cm3 ( 2.77)

The results from the alternate immersion experiment are listed in Table 3 &
Table 4. By Observing the Table 3., it is evident that zinc ions are not respon-
sible for corrosion of aluminum alloy. The effect of zinc ions is well understood

by considering the following equation:

Corrosion rate in mpy, at a given concentration of sodium
chloride solution

Reduction factor =-

 

Corrosion rate in mpy. at a given concentration Of sodium

chloride solution

The reduction factor is large at higher concentration of sodium chloride
solution, i.e., zinc ions are helpful in reducing the corrosion rate. This effect is
not observed at lower concentration of chloride solutions, because of the pre-

sence of chloride ions surpass the inhibition effect.

Table 3. Results of weight loss experiment from the alternate

Immersion experiment

 

 

 

 

 

 

 

 

 

 

Solution Additive to Eli-£233 Corrosion
the Solution res Rate
(4”,) (mpy)
cm
1 N none 2.605 19.31
0.2% 711C]; 0.730 5.416
NaCI
0.5% ZnClz 0.841 6.24
none 1.788 13.25
0.1 N
0.1% ZnC12 0.539 3.996
NaCI
0.5% ZnClz 0.791 5.871
none 0.799 5.923
- 0.01 N 0.1% ZnClz 0.524 3.891
0.2% an12 0.819 6.075
NaCl 0.5% ZnClg 0.854' 6.329
Zn metal 0.700 5.190
0.001 N none 0.253 1.880
NaCI 0.5% ZnClz 0.873 6.480
none 0.083 0.61
Distilled 6.6% ZnClz 1.377 10.21
Water 0.66% ZnClz 0.840 6.230
0.066% ZnClz 0.797 5.71

 

27

 

Table 4 - Reduction Factor for Chloride Solutions With Varying % of ZnCl2

 

 

 

 

 

% of ZnClz in the Concentration of Reduction Factor
Chloride Solution Chloride Solution
0.1 0.1 N 3.317
0.01 N 1.522
0.2 1 N 3.5659
0.01 N 0.975
0.5 l N 3.095
0.1 N 2.258
0.01 N 0.9358
0.001 N 0.2901

 

 

 

is

 

Fig. 6. Effect or concentration of NaCl with/without nor, on corrosion
of AA-2024

29

.230 m3

Fl Ni

-1l1-...illrL.:-.ii ill-211F111 .. Ll Fl .11 L i

to

 

4i

 

1 P

 

3O

 

 

 

 

0.
o

T'T‘TT‘T’W—e—T' “L-

0.
In
,_

 

_..
a
w.
.T -
”ll 1 HTIII' NR -il
ail \\ l-I.
m. \\
r \\.
\
r ..\\
T \.\...
T .\\
T \\R
s. 1
.1 .\..
r \ 02.5.50 o£n RED 5? .500 .002 O
.- w. com-.020 oEN NNd 53 .500 .002 mm
,- .. 8:26 can and 5s. .58 6oz e
02.530 OSN 3055 .500 .002 O
.-l i 2.111.. .. ..-.r_ili--...ritii--..-.p.-..H.i..-...-.. lM-M-WUii-rlfl.-. --

 

 

r"'T"‘-T' T—T j rm T‘TT‘T’

 

 

O
N
‘0

l

O

31

By observing the scanning electron micrographs, it is easy to judge
the extent of damage occurred to the samples. The pitting is extensive at higher
concentration of chloride solution without zinc chloride. The pitting damage is
comparatively very less for a sample maintained in same concentration of sod-
ium chloride solution with zinc chloride in it. This coUld be easily understood
by observing fig 7 to fig 11.

The results of alternate immersion test conducted on assemblies of zinc
coated bolts fastened into aluminum alloy plate are listed in table 5. The results
indicate that as the zinc coating is slowly removed, vigorous corrosion of alum-
inum alloy results. As it can be seen in the fig. 12 8. 13, the region under zinc
coated steel bolt is protected, but the area around the bolt is subjected to gal-
vanic corrosion. The corrosion damage sustained by the fastened assembly
of zinc coated bolts on aluminum alloy is more at higher concentration of

sodium chloride.

4.2 Galvanic experiment:

The results from the galvanic experiment is listed in table 6. As observed
from fig.14, the voltage changes with respect to time. At the initial periods,
when the alloy is cell anode and galvanized steel sheet is cathode, the value
of the voltage is -0.5233V. After 48 hours the voltage reading was -0.498V.
The value of the voltage tend to drop and attains a plateau after some time. As
the voltage tend to drop, aluminum alloy electrode tends to be more anodic,
Whereas the results from the lower concentration indicate that, there is a dr0p
of potential difference, but it is not as steep as it is in the case of higher conce-

ntration.

3‘2

- 6 ' ' | m“); _\i,0'.f';ls{:!. ’-
~. - “w

   

”- h'a- "" I'

l.
’5' ‘v ‘
m'.,'-‘;‘..§‘l_i. 25 “ml

.y
b?!

Fig. 7. Optical micrograph of an uncorroded sample of AA-2024 etched

with Keller’s reagent (450 x)

33

Fig. 8. Scanning electron micrograph of AA-2024 maintained in 0.1 N

NaCI without ZnCl2 ( 250 X)

Fig. 9. Scanning Electron micrograph of AA-2024 maintained in 0.1 N

NaCI with 0.1% ZnCIz ( 250 X )

34

 

35

Fig. 10. Scanning electron micrograph of Alt-2024 maintained in 0.01 N

NaCI without ZnClz (250 X)

Fig. 11. Scanning electron micrograph of AA-2024 maintained in 0.01 N

NaCI without Zrici2 (250 x )

36

 

37

Fig. 12. Optical micrograph of a corroded sample of AA-2024 fastened

with zinc coated steel fasteners kept in 0.01 N NaCI ( 250 " )

Fig. 13. Optical micrograph of a corroded sample of AA-2024 fastened

with zinc coated steel fasteners kept in 0.001 N NaCl (250 x)

' 12
3.4? tax-11:24:» 4'1"." 9;“
~,g _—. 4.01613} :, "Pic-‘6'.
' -.. 1’ ."' ‘ ,9 —:;?"), .' . ‘
4

     
   

39

Fig. 14. Time-Voltage plot of galvanic experiment of AA-2024 and
galvanized steel sheet maintained at various concentrations

of sodium chloride

40

A005 035.

 

 

 

 

 

 

 

 

 

 

we em 0
F L Li H F u tr O©.OII
l .1000]
ill--...
chi-...... .---oi...- tld a tom-.0.-
A
. . mt
..- ...-0.? Of... 11
Q
00
9
a- 100.0!
illii ittlttiiitltiiliMV
....- h. «...-MA“- _. imp-.0...
. mooz z 30.. o «to.
m .002 2 do. 0 GHQ
.- _ -..F W .
.- F! L tl ti 1.----. .-.-tilil ..--...!!! _ .-.-illit- -..iL. tili..l.......L 0” Oil VF

Table 5. Results of alternate immersion test of AA-2024 fastened with

zinc-coated steel fasteners

 

 

 

 

 

Solution conc- Weight loss/ Area Corrosion Rate
(me/cmzl (mpy)
0.01 N NaCI 0.8675 6.21
0.001 N NaCI 0.686 4.915

 

Table 6. Results of weight loss of AA-2024 from galvanic experiment

 

 

 

 

 

, Solution Concentration Weight loss/Area
(mo/cmzl

0.01 N NaCl 7.3

0.001 N NaCI 4.8

 

 

41'

42
The weight loss results indicate the same. There is a considerable amount

of weight loss at aluminum alloy electrode at higher concentration of chloride

solution than in the case of lower concentration.

4.3 Dripping electrolyte :

The results from the experiment reflect the same as those of alternate imm-
ersion and galvanic experiment. By observing the corroded sample, it is evid-
ent that, as the zinc coating is slowly removed, ferrous ions come into contact
with aluminum alloy electrode resulting in serious corrosion of aluminum alloy
(fig. 16) . The experiment conducted with distilled water explains that, after the
subsequent removal of zinc ions from the galvanized sheet, ferrous ions are
dissolved in water and they cause extensive damage to the aluminum alloy.
The weight loss calculations indicate that, a substantial portion of aluminum

is removed by ferrous ions. The weight loss is directly proportional to the
concentration of the solution employed for the experiment. By observing the
table. 7, it is evident that. the corrosion rate increases rapidly with the conc-

centration of the solution.

4.4 Potentiostat experiment:

The experiment is conducted to evaluate, whether zinc ions are responsi-
ble for corrosidn of aluminum alloy or not. The results indicate that, zinc ions
are not responsible for corrosion. By observing fig. 17, it is observed that the
polarization curve for the solution without zinc chloride is more steep compare-
bed to the solution with zinc chloride in it. The solution without zinc chloride is
more corrosive than a solution without zinc chloride.

The slope of the polarization curve is quite same for both solutions during

the initial part of the experiment, at current density value of 140 mA/cmz, the

43

 

Fig. 15. Optical micrograph of the corroded AA-2024 from dripping

distilled water experiment ( 250x)

Table7. Results of weight loss of AA-2024 from dripping electrolye/distilled

 

 

water experiment
Solution cocentration Weight loss
5 (me/cmzi
0.01 N NaCI 7.1069
0.001N NaCI 3.847
NIL 1. 275
1 i

 

 

 

 

45

Fig. 16.Anodic polarization curve of AA-2024 with/ without ZnCl2

maintained at 0.1 N NaCI

46

Nchi-4.5

 

 

 

on or
com . p » 0% Fr if p . b . P . ®.OI|
... itiiitiltl\t.“.ii\ll\\e r
1-i\tHlitli\-i\\\“l\0\ .
.- w
i T.V.O|
._ r A
s 0
TI.
-. 1.
.. e
- T no
.. TN oi a
- A
- s
.. M .s.
- . .3 ...o.o 3
.6 a fl
.. a. . % -.....uow.-woomeE aez zoo o o
.o 0-0 .

 

 

..63 who ES 82 7:

.

 

 

47

tendency of the curves changes. The curve for the solution without zinc chlo-
ride becomes more active, which means that zinc ions are helpful in protecting

the aluminum alloy.

5. Conclusions:

1) The results from the alternate, galvanic and dripping electrolyte experi-
ment reveals the fact that zinc coating catholically protects aluminum alloy for
the most of the exposure period. With some of the coating partially or complete-
ly consumed, exposing steel layer to aluminum alloy, which leads to the corro-

sion of the latter.

2) The extent of damage incurred by the aluminum alloy varies with con-
centration of the electrolyte. The higher the concentration, higher the corro-

sion rate of aluminum alloy.

3) The results of dripping water indicates clearly that, the water which falls
from the galvanized plate is not corrosive during the initial period, that indi-
cates that zinc ions are protective to aluminum alloy. After, the coating is con-

sumed ferrous ions are exposed, and initiate galvanic corrosion.

4) Zinc coating is found to have good corrosion resistance property, and
the application of the coating is limited by the environment conditions. By ob-
serving the advantages and disadvantages of zinc coating, one has to make a
proper choice of coating material depending upon the application. Since
cadmium coatings of sufficient thickness are rarely produced (becuse they are
expensive), galvanized coatings are preferred to cadmium coatings in indutri-

al application.

5) The corrosion problem encountered by A-10 aircrafts, is similar to the

galvanic corrosion of zinc-coated bolts and AA-2024.

48

49

6) The galvanic corrosion of AA-2024 fastened with zinc-coated bolts

could be minimized by using a proper coating to aluminum alloy surface.

6. REFERENCES:

1. R. G. Mitcchell, "An Airline View of the Corrosion Problem,” AGARD
Conference Proceedings No.315 on Aircraft Corrosion, Cesme, Turkey,

510 April 1981. '

2. A. R. Erben, "Evaluation of Corrosion Control on Aircraft Skin Countersink
and Fasteners,” J.Materials Protection, 5,18, (1965) August.

3. R. Summitt and F. T. Fink, "Forecasting Corrosion Damage and Maintena-
nce Costs for Large Aircraft,” AGARD Conference Proceedings No. 315 on
Aircraft Corrosion, Cesme, Turkey, 5-10 April, 1981.

4. Ing. M. Scolaris, "Corrosion Prevention Metods From Direct Experience
With Aerospace Structures,” AGARD Conference Procedings No.315 on
Aircraft Corrosion, Cesme, Turkey, 5-10 April, 1981.

5. Thomas A. Roach, ”Aerospace high performance fasteners resist stress
corrosion cracking,” J.Materials Performance, 23, 42, (1984) Sept-

ember.

6. M. J. Church, ”Aerospace Fastener systems- A proposal for standardiza-
tion”, Proceedings of National SAMPE Technical Co-nference, Vol. 1,Seattle,
Washington, 9-11 September, 1969.

7. Edward Taylor, “Compatible Coatings for Corrosion Resistant Aerospace
Fasteners,“ J.Materia|s Performance, 13, 29, (1974) September.

8. Henri Lajain, "Corrosion Protection Schemes for Aircraft Structures for the
Corrosion Behaviour of Al-Alloys," AGARD Conference Proceedings No. 315
on Aircraft Corrosion, Cesme, Turkey, 5-10 April, 1981.

50

51
9. V. C. R. McLoughlin, "Corrosion control measures for military aircraft prsent

UK requirements and future developments”, AGARDConference Proceedings
No. 315, Cesme, Turkey, 510 April, 1981.

10. M. Domk, ”Some observation on the corrosion of aircraft at the airforce
base in Bandirma, Turkey", AGARD Conference Proceedings No.315,Cesme,
Turkey, 510 April, 1981. '

11. N. D. Tomashov, "Theory of corrosion and protection of metals,“ Transla-
ted and edited by E. H. Tytell, l.,Geld and H. S. Presier, New York, McMillan,
1966.

12. F. L Laque and H. R. Copson, ”Corrosion resistance of metals andalloys",
2nd edition., Reinhold Publishing Corporation, New York (1963).

13. John E. Hatch, ”Aluminum properties and physical metallurgy", American
Society of Metals, Metals Park, Ohio (1984).

14. R. H. Brown, "galvanic Corrosion“ American Society for Testing and Mat-
erials, Bulletin 1261.

15. A. E. Hohman, ”Exterior aircraft finish systems to protectfasteners', J. of
Materials Performance, 13,, 46, (1£74) February.

16. V. C. R. McLoghlin, "An evaluation of coatings for steels and titanium alloy
fasteners for aircraft applications", AGARD Conference Proceedings No. 315,
Cesme, Turkey, 510 April, 1981.

17. E. Taylor, “The mod look in aircraft fastener corrosion control”, Proceed-
ings of National SAMPE Technical Conference, Vol. 1, Seattle, Washington,
9-11 September, 1969.

18. G. Braude, “Cadmium Colloquy', Plating and Metal Finishing, Q, 64,
(1977).

19. D. N. Fager and W. F. Spurr, “Solid cadmium embrittlement: steel alloys”,
Corrosion-MACE, 21, 72, (1971 )-

52

20. D. N. Fager and W. F. Spurr, ”Solid cadmium embrittlement: titanium
alloys”, Corrosion-NACE, 26, 409, (1970).

21. Robert N. Miller, "Aircraft exfloiation corrosion methods for prevention in
fasener holes”, J. Materials Protection, _6_, 55, (1967) February.

22. S. B. Lyon, G. E. Thompson, J. B. Johnson, G. C. Wood, and J. M. Ferg-
uson, "Accelerated atmospheric corrosion testing using a cyclic wet/dry exp-
osure test: Aluminum, Galvanized Steel, and Steel”, Corrosion-NACE, .43, 71,
(1987) December.

23. Reynolds Metal Company, 'The Aluminum Data Book’, Losiville, Ky, 1954.
24 John E. Hatch,’Aluminum properties at physical metallurgy’, American
Society of Metals, Metals Park, Ohio, 1984..

25. Greene, N. 0., Experimental Electrode Kinetics, Renseller Polytechnic
Institute, Troy, NY, 1965.

26. F rance, Jr., W. D., ’Controlled Potential Corrosion Tests, Their Applicat-
ions & Limitations,“ Materials Research And Standards, Vol 9, No. 8, 1969,
p.21.

MICHIGAN STATE UNIV. LIBRARIES
WWWlllHlNWlllllllW|||||H|||NW|H||N1|
31293005741131