orJ\,\ ~,('.·.... . . . :emtO ATIN 1ii~~!P!!!!,~,,1~ FOR OFFtciXC osz eru~K N ..,.,
File D 11.2/ 18 ~ .• ~ cC~Q- FIELD REPORT, SERIAL No. 1636
' • ... • ,. V
AIR SERVICE INFORMATION CIRCULAR
<AVIATION)
PUBLISHED BY THE CHIEF OF AIR SERVICE, WASHINGTON, D. C.
Vol. III October 1 , 1921
REPORT ON THE CORROSION PREVENTION
OF METALLIC AIRCRAFT PARTS
Ralph Brown Draughon
LIBRARY
APR 2 3 2013
Non·Depoitory
Auburn University
(MATERIAL SECTION REPORT NO. 137)
Prepared by Engineering Division, Air Service
McCook Field, Dayton, Ohio
May 24, 1921
WASHINGTON
GOVERNMENT PRINTING omcE
1921
No. 274
)
I
""-' • I
REPORT· ON THE CORROSION PREVENTION OF NrnTALLIC/
AIR CRAFT P ARTS.- P ART I. , ~ ~ ... · ·
PURPOSE.
An investigation of the theory of corrosion prevention,
with particular reference to the corrosion of metal aircraft
parts.
SUMMARY.
The theory generally accepted as best, explaining the
corrosion of iron and steel, is the electrolytic theory,
which states that corrosion is essentially an electrolytic
process in which iron passes into solution in the presence
of water and is subsequently converted into ferric oxide.
GENERAL DISCUSSION.
In the past, the. material used in the construction of
airplanes has largely been wood. Numerous considerations
are leading to a gradually increasing use of metal
in aircraft construction. The 'strength of the metal can
be definitely determined and the uniformity in strength
can be relied upon. :Metal airplanes withstand landing
shocks much better than those made of wood. :Metal adds
the two fundamental prerequisites for long-distance
flights- namely, re\iability and durability. Practically
the only things which limit the life of the metal airplane
are corrosion and fatigue.
In corroding, iron and steel are eaten away, with the
result that the strength of the material is considerably
reduced, and, as in the case of relatively thin gaged aircraft
parts, this reduction in strength may so weaken a
member that failure during sudden stresses or strains
beco'i'.nes imminent. Several instances of such failures
have occurred, and two accidents in particular have been
traced directly to the " ·eakening of aileron horns .by internal
corrosion. Corrosion started on the inside 0f the
aileron horn, was unobserved. and continued to eat away
the metal until failure finally occurred while in flight. A
study of the problem of corrosion and its prevention is,
therefore, extremely pertinent at the present time, especially
in view of the a.II-metal program.
THEORETICAL CONSIDERATION OF THE PROBLEM
OF CORROSION OF IRON AND STEEL.
The process whereby steel and iron are corroded and
eaten away is ·called by the chemist "oxidation ." It has
been found that a metal ,vill corrode much more rapidly
if it is placed in a medium of high-moisture content. A
steel structure, for example, in an arid region will suffer
very little from corrosion, whereas the same structure near
the· ocean will corrode very rapidly . Water is essential to
co1Tos1on. In the absence of moisture, steel does not
corrode. · It has been found that in undergoing this process
of corrosion , iron or steel, when in contact Thith water,
will always pass into solution. In doing this, it always
goes into the ionic form , which means that it carries away
from the metal a positive charge of electricity for each
ion. Thus, a potential clifference is established and we
have at hand, in the study of corrosion of iron and steel,
a study of the effects of various conditions upon the potential
di fference between the metal in question and its
immediate environment. The problem, therefore, is
largely electrochemical and the rate of corrosion is simply
a function of the electromotive force present and the resistance
offered to it by various conditions. The tendency
to pass into solution is a property peculiar to all metals
and is different for each metal. Some go into solution
very rapidly, such as magnesium and potassium, while
others, such as platinum and gold, go into solution very
slowly. This tendency is known as "electrolytic solution
pressure" of the metal, and the rate of corrosion is directly
proportional to this pressure.
The ionization of a metal leaves its surface negatively
charged, while the solution which is inimediately in con·
tact with the metal is positively electrified. Under the
influence of the electrolytic solution pressure, the ions
tend to pass into solution, while the electrostatic attraction
of the electrolytic double layer tends to retard the
action. vVe have then two opposing forces, both of which
are important. The potential difference, produced by
the ionization of the metal, can be measured by applying
Nernst's formula:
F O.OOOl9S3 T l oo-_!'.__volts.
V "p
F=Potential difference.
'!'= Temperature.
P =Osmotic pressure of univalention.
p = Electrolytic solution pressure.
v= Valence of the ion.
The follo"ing list gives the tendency, measured in
atmospheres, of some of the more common metals to go
into solution.
Electrolytic solution
pressure in atmospheres.
:Mg................... . ............. 1044
Zn....... . ....... ............ . ... 1018
Al ..... .. .. .... .... . .. .... .. .... .
Cd .. . ... . ....................... .
i'e . -. - - - - . - - - -.. - - - - . -. . - . - - - - . -..
Co .... ....................... . .. .
li ........... . .................. .
Pb . . .. .. .. .. . ... . .. .. . ... ... .... . .
Hg .................. · ... . . . --· ..... .
Ag .... · ···· · ·· · ·· ··· · · ···· · ······
Cu . .. .... ....... . . .. ......... ... .
10'"
3Xl06
10'
2x10°
1 x 10°
103
lQ1 6
1011
1020
It will be noted that iron stands approximately at the
middle of the list, copper at the bottom, and zinc and mag-
66533- 21 (3)
f
4
nesium at the top. Iron tends to dissolve or oxidize with
a force which is approximately equivalent to 10,000 atmospheres
at normal temperatures. In other words, iron,
in contact with water, will tend to dissolve until the concentration
of the iron ions becomes of such a magnitude
that their osmotic pressure is equal to the value given
above. When this pressure counterbalances the electrolytic
solution pressure, a condition of equilibrium is
reached and no more iron goes in solution. This means
that sufficient iron must have gone into solution so that
the iron solution is 450 times normal, which, according to
Whitney, is a concentration practically unobtainable.
Furthermore, such a condition of equilibrium is reached
only under certain conditions--constant temperature,
constant pressure, and no change in the quantity of
solvent or metal. Such a condition obviously does not
exist in the air.
In the case of a metal in contact with the solution of one
of its salts, as occurs when a metal is going into solution,
the osmotic pressure -usually differs from the electrolytic
solution pressure. The osmotic pressure may either be
greater or less than the latter. If the osmotic pressure
is less than the electrolytic solution pressure, the metallic
ions pass into solution until an equilibrium is established.
LeBlanc has shown that this action is not great in pure
water, but if the osmotic pressure of the ions already in
solution, in the case of impurn water, oppose the electrolytic
solution pressure, or if the concentration of the ions
in solution is so great that the osmotic pressure exceeds
the electrolytic pressure, the ions of the solution will
separate out and precipitate upon the metal, giving up
to it their positive charges. The metal then becomes
positively charged anp. tlie solution, which before was
electrically neutral, becomes negatively charged, and the
electrolytic double layer thus produced now counteracts
the electrolytic solution pressure until equilibrium is
established. There are then the three conditions:
p=p
p=Solution pressure.
p=Osmotic pressure.
Equilibrium. No potential difference.
P>P
Metal is negatively charged. Solution is positively
charged. Electrostatic attraction opposes solution pressure.
P<P
Metal is positively charged. Solution is negative.
Electrostatic attraction superposed on the solution pressure.
Here, then, we have one of the fundamental points in
the study of corrosion. Osmotic pressure and electrolytic
solution pressure are functions of the corrosion of
iron and steel.
It has been observed in the case of two differently
concentrated amalgams of the same metal, as, for example,
two amalgams of zinc of different concentrations in a
solution of one of its salts, that the zinc passes from the
more concentrated amalgam into the solution and from
the solution to the weaker amalgam. Further, if two
metals form a mechanical mixture, the potential will be
that of the metal having the greatest electrolytic solution
pressure. If metallic iron, for example, is mixed with
metallic zinc and is used as a negative electrode, practi-cally
zinc ·ions only pass into solution. Even if the
electrolytic solution pressure of the zinc is repressed by
the addition of zinc ions to the solution, only a small
number of iron ions will dissolve, and if, on the contrary,
iron ions are put into the solution, a secondary reaction
takes place in which iron ions are deposited and replaced
in the solution by zinc ions until the potential difference
between zinc and zinc ions equals the potential difference
between iron and iron ions. This observation is of great
importance, as it is the basis of the increased corrosion of
nonhomogeneous steel, which is discussed below.
STIMULATION AND INHIBITION OF CORROSION.
The solution of the problem of corrosion prevention is
considerably facilitated by a study of the stimulation and
inhibition of corrosion. If it is possible to determine the
factors which influence the rate of corrosion, the retarding
of corrosion is subsequently made a matter of much greater
simplicity. It has been found that the corrosion of iron
and steel may be stimulated from two distinct angles.
The first of these is the nature of the environment or
medium in which the iron and steel in question is located;
the second is the physical, metallurgical, and chemical
characteristics of the metal itself.
It has already been pointed out that moisture is an
essential prerequisite of corrosion, and that in the absence
of moisture corrosion does not occur. Hence, in the
medium in which iron and steel are to be found the first
factor determining the rate of corrosion is the percentage of
moisture which is present. Whitney has found that iron
will corrode even in pure· water. Certain substances,
if present in the water, will increase the rapidity of corrosion.
Therefore, a second factor entering into the
stimulation of corrosion is the presence of these substances,
which are known as "electrolytes." These electrolytes
can be divided, for the sake of convenience, into two main
classes- acids or solutions of gases in water, and salts.
The rate of corrosion of a piece of steel is, in general,
directly proportional to the percentage of acid present
in the ionic state. A piece of metal which is placed into
water containing sulphuric acid will rust with an acceleration
considerably greater than a piece of steel which
has been placed in pure water.
Considered from the point of view of aeronautics, these
acids come in contact with metal in the following way:
Coal is burned in large quantities in densely populated
districts. One of the products of combustion of coal is
sulphur dioxide. The moisture of the air combines with
the sulphur dioxide to form sulphurous acid-H20+S02=
H2S03• This acid is a fairly strong electrolyte and stimulates,
to a remarkable extent, the corrosion of any metal
with which it comes in contact. Acids of any kind will
accomplish the same results to a greater or lesser extent,
depending upon their relative ionization values. It is
to be expected, in view of the above fact, that iron and
steel will corrode more rapidly in districts which are
densely populated, or in large manufacturing districts,
than in regions sparsely populated. This has been found
to be the case for iron and steel structures in the locality
of Pittsburgh, for example, corroded more rapidly than
iron or steel in Ohio or in the Dakotas. The acid, which
probably is the most common corrosive agent, is carbonic
acid. Carbon dioxide is a product of the breathing of
...
plants and animals as well as the product of combustion
of any sort.
The carbon dioxide content of the atmosphere is dependent
upon four factors:
(a) Location.-Land or sea, height above ground,
country or city, height above sea level, latitude, etc.
(b) Time.-Day or night, season of year, etc.
(c) Weather.-Rain, fog, or clear, sunny or cloudy,
windy or calm, hot or cold, etc.
(il') Chance.-Factory fumes, forest fires, etc.
Variations in carbon dioxide content are, however,
slight. The average value usually given is 3.53 parts of
carbon dioxide in 10,000 by volume. This gas is soluble
in water. Its solubility has been determined by Bohr
and Bock.
Tempera- I Absorption
ture. coefficient.
Sol. in g.
mol./lit.
I
o• c. I 1. 713 0.0764
18° C.
I
. 928 • 0414
25° C. . 759 . 0338
Now, the magnitude of the corrosive action of water
containing dissolved carbon dioxide is determined largely
by the specific conductivity of water in equilibrium with
carbon dioxide of the atmosphere. Such a solution
possesses considerable electrical conductivity, due to the
fact that the solute combines with the solvent to form
cabonic acid.
COigas)~C02( dissolved)+ H20~H2C03~H + + HCO;
A secondary dissociation occurs in which the carbonic
acid is split up. This dissociation, however, is very
slight.
Carbonic acid has an ionization constant for the above
temperature as follows :
7c=2.24Xl0-7 at T=0° C.
7c=3.12Xl0-7 at T=l8° C.
7c=3.50Xl0-7 at T=25° C.
From these figures, Kendell had determined the following
specific conductivities of a saturated solution at the
above temperatures:
Specific conductivity (reciprocal olms X 10- 6).
T - 0° C. T-18° C. T- 25° C.
Average specific con-ductivity..............
0.60--0.67 0.75--0.80 0.80--0.85
It is this relatively high specific conductivity of carbon
dioxide saturated moisture which aids the corrosion of
iron and steel by facilitating the flow of local currents
from nodes of higher solution pre3sure to those of lower
solution pressure, and also by directly attacking the metal
itself.
It is important that each step of this process be
thoroughly understood before an attempt is made to
inhibit it. The chemical changes which take place are
as follows:
The iron, which goes into solution under the influence
of the electrolytic solution pressure, reacts with carbonic
5
acid to form either the acid ferrous carbonate or the monohydroxy
ferrous carbonate. The acid ferrous carbonate is
changed to the monohydroxy ferrous carbonate, the monohydroxy
ferrous carbonate is changed to the dihydroxy
ferrous carbonate, both of which subsequently break down
to · form ferric hydroxide and carbon dioxide. In the
final step, two molecules of ferric hydroxide combine· to
form a molecule of ferric oxide, which is iron rust. The
chemical equations for these reactions are as follows:
1. Fe+2 H2C03=Fe (HC03) 2+H2,
or
Fe+~C03+0=Fe (OH) (HC03).
2. 2 Fe (HC03) 2+3 H20=2 Fe (OH) HC03+2 H20,
or
2 Fe (OH) HC03+3 H,0=2 Fe (OH)2HC03+H20+
H2.
3. Fe (OH) (HC03) 2=Fe (OH)3+2 CO2 ,
or
Fe (OH)i(HC03)=Fe (OH)a+C02 .
4. 2 Fe (OH)3=Fe20 3 • 2 H20.
The final product is a molecule of ferric oxide, which is
extremely hydroscopic and will hold in suspension varying
amounts of moisture, depending upon the temperature and
condition of humidity of the medium surrounding the piece
of steel. This moisture, being present in the layers of iron
hydroxide adjacent to the unaffected steel, will allow more
of the steel to go into solution. It is because of this hydroscopic
nature of the ferric oxide molecule that the corrosion
of a piece of steel is continuous and stops only when there
is no more metallic iron left to go into solution; that is,
when the whole piece of metal has been eaten up.
Walker, Cederholm, and Bent, as well as several other
investigators, have shown that oxygen has an important
bearing on the corrosion of iron and steel. They state that
the primary function of oxygen is in depolarizing the
cathode portions of the metal upon which hydrogen tends
to precipitate during the corrosion process, and that the
secondary function of oxygen is the oxidation of ferrous
iron to the ferric condition with its subsequent precipitation
as ferric hydroxide. They determined that the rapidity
of corrosion of a metal in water is a linear function of the
partial pressure of the oxygen in the atmosphere. Thus,
it is seen that acids and gases tend to increase the speed
of corrosion of a piece of iron or steel.
A third class of electrolytes which stimulates corrosion
is salts. This class consists largely of sodium chloride,
which is the chief electrolyte present in sea water. The
presence of this electrolyte explains the increased corrosion
that occurs on metallic structmes near the sea as compared
with those farther inland. It is the action of this
strong electrolyte which causes the difficulty which the
Navy has been having with the corrosion of metal parts on
sea planes.
An extremely important factor in determining the
rapidity with which steel corrodes is the nature of the
metal with which it is in contact. A piece of steel in
contact with a dissimilar metal will corroa.e more rapidly
than will a piece of steel in contact with another piece of
steel of approximately the same chemical composition.
The corrosion of any piece of metal may be stimulated by
bringing itin contact with a metal which is below itin the
electropotentiiiJ seril:)s. Fpr exiimple, a piece of steel in
6
contact with copper or tin will corrode more rapidly than
if the same piece of steel is in contact with a metal electropositive
toit, such as aluminum and zinc. The reason for
this becomes perfectly apparent when considered from the
point of view of difference in electropotential, as pointed
out in the theoretical discussion given on page 3. This is
an extremely important item from the point of view of airplane
design. Two dissimilar metals in contact will
always have a difference in potential, and are bound to
corrode more rapidly than if they were not in contact.
The ideal to be striven for in design is the avoidance of
contact of dissimilar metals. Wherever possible aluminum
should be in contact with aluminum and steel in contact
with steel, for wherever aluminum and steel or copper
and steel are in contact with each other, all other conditions
being equal, marked corrosion will inevitably result.
This problem. is one which merits further investigation
because of its increasing importance to metal aircraft
construction.
Corrosion is stimulated to a considerable extent by the
nature of the metal itself. The presence of impurities in
iron and steel will increase the rate with which they corrode,
as has been explained on pages 3 and 4. Minute local
currents are set up on the surface of the metal, running
from particles of greater solution pressure to those of lower
solution pressure, through which the former are gradually
eaten away. Such corrosion is particularly marked if the
impurities present are nonhomogeneously distributed
through the metal, for then nodes of solution pressure are
established between such zones of impurities and adjacent
metal, which tend to magnify the action just described.
In the preparation of steel the rolling which is done subjects
the metal to marked stresses and strains. The distortion
which the crystalline structure of a piece of steel
undergoes during rolling has a great influence on the electropotential
differences of these crystals. Accordingly,
mechanically induced strains are extremely injurious to
the metal and aid in a very remarkable degree in stimulating
the rate with which steel corrodes. The improper
heat treatment of the steel, in which the metal is not left
as a eutectoid, is believed also to stimulate corrosion,
inasmuch as it is believed that noneutectoid steels corrode
more rapidly than steels which are in the eutectoid condition.
Annealing removes the strains and reduces the
tendency of the metal to corrode.
The porosity of the metal is another function of its
corrosion ratio, for a metal which has been burned or is in
a porous condition will absorb considerably more moisture
than normal steel, and therefore will corrode more rapidly.
Occluded gases, blowholes, and scratches are also factors
which accelerate the rate of corrosion of metal.
Thus it is seen that corrosion may be accelerated in the
medium itself by the addition of moisture and electrolytes
in the form of acids, gases, or salts, and also by the electrolytic
solution pressure of the metal with which the steel in
question is in contact. Furthermore, the condition of the
metal is a determining factor, for a piece of steel which is
high in impurities, nonhomogeneous in structure, full of
mechanically induced strains, improperly heat-treated or
annealed, saturated with occluded gases, porous and full
-of blowholes or scratches, will corrode very rapidly.
already been pointed out that steel will not corrode in the
absence of moisture. Its rate of corrosion will corresponding!
y decrease as the percentage of moisture in the
medium decreases. The same holds true for the reduction
in percentages of electrolytes present, whether they be
acids, gases, or salts. In the design of airplanes, care
should be exercised in order that, wherever possible, di ssimilar
metals may not be in contact with each other.
-In-the composition and treatment of the metal, its corrosion
may be retarded by a decrease in the amount of im-purities
which are present. Pure steel corrodes much
,lower than does steel containing a large amount of impurities.
The homogeneity of the steel, the lack of mechanically
induced·strains, the proper heat-treatment and
subsequent annealing of the metal, the avoidance of porosity,
of blowholes, and of scratches, as well as occluded
gases, all tend to retard corrosion and thus increase the life
of the metal.
PROTEC::TION OF IRON AND STEEL.
The problem of the protection of iron and steel from
corrosion divides itself into three main divisions :
1. The treatment of a metal EO as to get the surface into
a passive condition.
2. The development of a metal which is itself resistant
to corrosion.
3. The protection of iron and steel by means of a coating
of some kind.
The literature shows that attempts to prevent corrosion
by treating the surface of iron or steel in some manner,
usually a chemical treatment with chromic acid or chromates,
have not been particularly successful. .Results
obtained have been irregular, some being very good and
others equally bad. Hittorf and Finkelstein haYe developed
a theory covering this subject which appears to
be given more credence than the others. Briefly, it is
thought iron has two allotropic forms of which the trivalent
is passive and the clivalent active, and it is because of its
state of higher oxidation in the triYalent condition that
steel or iron is made passive by such chemical treatment.
Byers and Langdon have shown that passivity is also produced
by occluded oxygen . The fundamental cause of
passivity does not appear to be clearly understood in spite
of various explanations, and the subject as a whole is quite
problematic. It would, indeed, be highly desirable to
have developed a satisfactory commercial method by
which ferrous metals could be rendered rust resistant.
The inhibition of the corrosion of steel is accomplished
by rev~rsi:p.~ the coμditions just meμtio:p.ecl. It l:ias
The second method of making iron and steel resist.ant
to corrosion is the addition of certain chemical elements
to the steel itself. Silicon is one of these elements and its
addition to cast iron, up to 12 to 15 per cent, produces·a
casting which is practically rust resistant. Such material ,
however, is apt to be quite brittle. Kalmus and Blake
have found that additions in small amounts of copper,
nickel, and cobalt (0.25-3 per cent) to iron in ingot form
produced increased resistance to atmospheric corrosion.
It is generally accepted that corrosion varies with the percentage
of carbon, the high-carbon alloys being the more
corrosion-resistant. Buck has found that additions of
small percentages of copper increase the life of sheet steel..
Numerous attempts have been made to develop alloy
steels whic.:\l are made rust re~sta:p.t bf tb,e a,ddition of
.,
,:
'f
such elements. Some of t hese have withstood atmospheric
exposure remarkably well, others have done very
poorly. The chemical analyses of some of the steels belonging
to this class are as follows :
C. 1fn. P. S. I Ni. Cr. Si.
Steel No.L. ..... . ...... : 0.05 0.30 0~ o.01 l= --9 - 3-
Steel No. 2. ... . . .. . .. . . . . . . 30 . 80 . . . . . . . 01 25 15 4. 30
Steel No.3 ...... . . . ........ 33 .75 ............ ,25 15 4. 50
Steel No. 4. ... . . . . . . .. ... . . 37 .15 . 002 . 01 . 25 13
Steel No. 5 ................. 40 .01 .02 .25 13
The difficulty with such steels, particularly the quatenary
steels, is that it is practically impossible to work them,
and machining is entirely out of question. Obviously,
such material is unsatisfactory for aircraft work.
The considerations outEned above lead to the inevitable
conclusion t hat neither of the above methods--namely,
t he passive condition and the addition of certain chemical
elements-are satisfactory methods ·for the corrosion prevention
of aircraft steels, and it therefore becomes necessary
to turn to a protective coating of some sort. The
protection of steel by means of protective coatings of
various kinds is taken up in Part II of this report.
BIBLIOGRAPHY.
The Corrosion of Iron and Steel. Cushman and Gardner.
Electro-Chemistry. LeB!anc.
Electrodeposition of Metals. Langbein.
7
The Corrosion of Metals--Ferrous and Nonferrous.
Jolll'. Faraday Soc. 11, p. 183.
The Corrosion of Iron. "Whitney, Jolll'. Am. Chem.
Soc., Vol. XXV, No. 4, April, 1903.
Oxygen Prime Factor in Corrosion. Richardson, Chem.
& Met. Eng. , July 7, 1920.
Passive State of Metals. Byers, Jour. Am. Chem. Soc .,
vol. 30, vol. 36.
Symposium of Corrosion of Metals and Alloys. Chem.
& Met. Eng., vol. 24, No. 19, May, 1921.
Protective Metallic Coatings for Rustproofing Iron and
Steel. Bur. of Stds. Circular No. 80.
Corrosion Prevention of Aircraft Metal Parts. Gardner,
Jour. Soc. Auto. Eng., Vol. III, No. 6, December, 1918.
Action of Sulphuric Acid on Alloy Steels. Aitchison,
Jour. Chem. Soc. 109, 288~298, December, 1916.
Corrosion Resistance of Copper Steels. Buck.
Influence of Composition upon Corrosion of Steel.
Aitchison, Trans. Faraday Soc., 1915.
Corrosion of Iron and Steel. Walker, Cederholm &
Bent, Jour. Am. Chem. Soc., September, 1907.
Protection of Iron and Steel. H . Hess, Met. & Chem.
Eng., 16:13, January , 1917.
Tests for Relative Corrosion. B. Fever, Chem. & Met.
Eng., June, 1920.
Corrosion of Ingot Iron Containing Cobalt, Nickel, and
Copper. Kalmus & Blake, Jour. Ind . & Chem. Eng.,
Vol. IX, No. 2, February, 1917.
Passivity of Metals. Byers and Langdon, Jour. Am.
Chem. Soc. 36:10, October, 1914.
REPORT ON THE CORROSION PREVENTION OF METALLIC
AIRCRAFT PARTS-PART II.
PURPOSE.
To investigate the efficiency of various protective coatings
as corrosion preventives for metal aircraft parts.
CONCLUSIONS.
The most efficient corrosion P,reventive is a metallic
coating of zinc. The method of zinc coating most adaptable
to the aircraft industry is the process of electroplating
from a zinc cyanide solution.
MATERIAL.
The materials generally used as protective coatings are
divided into two classes--the one organic coatings, which
consist of varnishes, paints, enamels, oil, lacquers, rustpreventive
compounds, etc.; and the second consists of
metallic coatings, such as terneplate, tin plate, zinc plate,
copper plate, etc. Coatings of both classes were tested to
determine their relative corrosion resistant properties
under atmospheric exposure conditions. The following
list contains the coatings which were examined:
tested. These were given a southern exposure to the
atmosphere. The samples were inclined at an angle of
45°. Periodic examinations were made of the condition
of the surface and the amount of surface corroded was
estimated. This observation was recorded as' 'Percentage
surface conosion."
METHOD OF ZINC PLATING AIRCRAFT PARTS.
The following method is used for cleaning and zinc
plating metal aircraft parts. It will be noted that this
procedure duplicates very closely the commercial practice.
The few changes which have been made have been instituted
in order to avoid certain harmful results whjch
would interfere with the use of this process in the aircraft
industry.
The material is first immersed in commercial benzol, in
order to remove excess of grease, and then placed in a hot
electric alkaline cleaner for about 20 minutes. The
material is suspended from the anode. This serves to
remove all grease and dirt. This is followed by a short
dip in a 10 per cent solution of sulphuric acid to remove
ORGANIC COATINGS.
Name. Manufacturer. Character of coating after drying for
24 hours.
African Slush .................... .. .... .. E. F. Houghton & Co., Philadelphia, Pa ...................... Black, hard.
Hippo Baking Varnish. . . . . . . . . . . . . . . . . . American Chemical Mfg. Co., Norfolk, Va................... . . Transparent, dry.
Burnt Oil Coating .. ..... ... ..... ..... ... McCook Field, Dayton, Ohio .................. . . . ... ..... ..... Black, dry.
Coro-Knot. . ..................... .... . ... Rust-Knot Mfg. Co., Toledo, Ohio ........................ . .... Transparent, dry.
Coro! Compound......................... Coro! Co., Chicago, Ill......................................... Transparent, grease.
Exhaust Pipe Enamel (Spec. 4) .••.••..• Tower Varnish Co., Dayton, Ohio .............. ..... ... .. ..... Black, dry.
Fokker Enamel. .... ... ..... .... .. ..... .. Fokker Airplane .......... ......... ... . . ................ . .. .. . Gray enamel.
No-OX-ID A & C ............ ... ........ Dearborn Chemical Co., Chicago, Ill.. ......... ... ..... ...... .. Semitransparent grease.
Parkerizing .............................. Parker Rust-Proof Co. of America, Detroit, Mich ............. Black, dry.
Red Enamel.. . .. .. ·.-··:················· Empire Metal Aircraft Corr,., Colle~e Point, L. I. ..... . ....... Opaque, dry.
Rust Veto Amber L1qmd ... .... ... .. .... E .. F. Houghton & Co., Philadelphia, Pa ..... : . ... . . ... ....... Transparent grease.
Rust Veto Heavy ... .............. . . . ... ..... . do......................................................... Semitransparent heavy grease.
Rust Veto Soft ............................... do ...................... . ................ . . . ..... . ..... . .. . Moderately heavy transparent grease.
Steel Gloss .................................... do .... .. .. ... ........ . ... .... ............ . .... . .... .. ...... Transparent, dry.
Whitmore Slushing Compound . . .... .... Whitmore Mfg. Co., Cleveland, Ohio ...... . . .. ... . ............ Opaque, grease.
METALLIC COATINGS. scale and rust. It is then rinsed in water and placed in
the plating bath.
Name. Manufacturer.
Cadmium Plate . . . . . . . . .... Louth-Patten Co., Kokomo, Ind.
Cadmium Plate covered Do.
Treatment.
with Tin Plate.
Copper Plate ............. .
:t:r:;n;fa1fJ~:::::::::::::::::
Chemical Laboratories, McCook Field,
Dayton, 'Ohio.
From stock, McCook Field, Dayton, Ohio.
Immersed in commercial benzol. .............. . .
Do.
Cleaned in electric alkaline cleaner ...... .. . .. . . . .
Zinc Plate ... ... ..... .. ... . Chemical Laboratories, McCook Field,
Dayton, Ohio.
METHOD OF PROCEDURE.
Samples of sheet steel, 4 by 6 inches, of the same composition,
were cleaned to remove all scale, rust, and grease,
tli9rou~hly qrjed1 ~q ~9ated wit!:!. the m!),terial to be
(8)
Suspended as anode.
Composition of cleaner:
12.5 g/L-Sodium hydroxide.
25.0 g/L-Sodium carbonate.
10.0 g/L-Sodium cyanide.
Current density, 10 amp., 5 volts,
Temperature.
'C.
20
74
Time.
Min.
I
)
)
Treatment.
Acid dip:
10 ner cent sulphuric acid (conunercial 66°
Baume) ........... . .... .......... ....... .
Plating ............................. .
Composition of solution:
45 g/L-Zinc oxide.
75 g/L-Sodium cyanide.
15 g/L-Sodium hydroxide.
Tempera- Time.
ture.
0 0. !Ir . .!if.
22 10
20 1 30
Current density; 2 amp. /dm .2 (20 amp./sq . ft .).
RESULTS.
The results obtained in each case are noted on the attached
graphs. P ercentages surface corrosion are plotted
along the ordinates and days' exposure to the atmosphere
are plotted on the abscissa.
DISCUSSION OF RESULTS.
9
Or!!anic coatino-s are absolutely unsatisfactory for the
corro: ion preven tion of ~etallic aircraft parts b ecause of
the following considerations:
All organic coatings are difficult to apply in a uniform
manner, so that absolute assurance may be bad that all
parts are adequately coated with the rust preventive.
Any material which is applied by hand or by means of the
spray has this defect. It is likewise diffi cult to get a
continuous coating of the material on sharp edges or on
protruding parts, of which there are a great many on structural
aircraft par ts. Tlie material flows away from the
edges and leaves them bare. All organic coatings have
the further objectionable feature of being extremely
sensitive to temperature and atmospheric changes. They
likewise have little resistance to heat and to vibration.
Upon exposure to the atmosphere, these organic coatings
soon crack, and the cracks break through the coating,
leaving the steel structure directly exposed to the atmosphere.
Changes in humidity, changes in temperature,
stresses, strains, and vibrations rapidly produce these
results. Organic coatings also have the distinctly disadvantageous
featme in that they are very easily removed
by abrasion or scratching, such as they are very likely to
be subjected to by mechanics during tightening and truing
up structural members. It is, therefore, to be expected
that organic coatings do not hold up well under atmospheric
exposure conditions. This expectation is verified by the
results obtained in this investigation.
Metallic coatings have the advantage over organic
coatings in that they can be applied with greater uniformity
, they adhere more firmly to the metal base, and
are not sensitive to changes in atmospheric conditions,
nor do they suffer materially under alternate stress or
under vibration. It is likewise difficult to remove a
metallic coating by scratching or abrading it. The metallic
coatings which have been used on metallic aircraft
parts or in the aircraft industry in general are copper
plate, terneplate, tin plate, and zinc plate. From the
electropotential series, given in Part I of this report, it is
seen that each metal has a definite t endency to go into
solution, and that copper, tin, and lead are electronegative
to iron; that is to say, they corrode less rapidly than does
iron and steel.
It is naturally to be inferred from this that if steels were
coated with one of these metals a very high degree of
protection against corrosion will have been established.
This, however, is not the case. As long as the metallic
coatings of copper and tin, or of any metal electronegative
to iron, is absolutely continuous, it is true that the iron is
protected from corrosion. However, let ever so slight a
scratch be made on this coating, and let ever so small a
portion of the surface of the steel base be exposed to the
atmosphere, then marked corrosion is the result, due to the
fact that the iron will corrode more rapidly than the
copper, and the copper, tin, or t erne coatings serve merely
as accelerators to this action. It is, therefore, not at all
surprising to find that tin plate, copper plate, and terneplate
in particular suffered considerable corrosion during
the atmospheric exposure tests made in this investigation.
If, on the other hand, the steel sub-base is coated with a
metal which is electropositive to iron, such as aluminum
and zinc, the results are changed very remarkably.
Almninum and zinc, being electropositive to iron, have a
greater solution tension pressure than iron. Hence, if the
metallic coating of aluminum or zinc is abraded or in any
way destroyed , so that the steel base is exposed, the corrosive
action will be such that the coating will be sac1ificed,
but the steel will remain practically intact. It has also
been found that zinc and aluminum exert an inhibitive
zone about them so that considerable protection is
obtained, even in those areas where the steel sub-base is
not directly covered. In other words, if a mechanic
scratches through such a coating while working on a
fitting, the loss of this coating will not be as serious as in
the case of copper, tin, and terneplate. It is, therefore,
to be expected that aluminum and zinc are far superior to
copper, tin , and lead as protective coatings. Thus far,
however, it has been impossible to satisfactorily coat iron
and steel with ahuninum, and therefore the natural
alternative is to turn to zinc.
Zinc has long been used as a protective coating for iron
and steel. There are four chief commercial processes for
its application at the present time. The oldest of these,
and probably the most universally used, is the process
called "hot-galvanizing. " This consists of immersing
·the material to be coated in a bath of spelter which has
been heated up to the melting point, drawing the metal
throuoh the bath, and withdrawing it at the other end.
The ~olten zinc, coming in contact with the metal,
combines witli the surface of the metal and if cooled
rapidly by immersing in water, crystallizes in small
crystals, or if cooled at room temperature, crystallizes in
the larger crystalline form so commonly observed on
galvanized roofing. The methods used in this process
are extremely crude and unscientific. A large amount of
material is wasted and the coating is unnecessarily thick
for the purpose to which it is to be applied. The usual
method by which the thickness of the coating is controlled
is wiping, which is unsatisfactory on other than flat pieces
and therefore unapplicable to the aircraft industry where
uniformity in thickness is desired. Ammonium chloride
iS usually used as a fluxing agent and this frequently
ibecomes occluded between the metal and the zinc coating
with the result that galvanic action immediately sets in.
10
A second commercial method is a process called
"sherardizing. " In this process . the steel members to
he coated are placed, in alternate layers with fine
granular zinc, in a steel drum, which is sealed and heated
to 350°-400° C. for 5 to 10 hours, depending upon the
thickness of the coating desired. The process is essentially
a vapor deposition process.
The Schoop spray process consists of spraying molten
zinc onto the material to be coated by means of a compressed
air spray gun.
In each of the above three processes, the zinc penetrates
into the steel for a considerable distance, forming a series
of zinc-iron alloys, which are rich in iron at the steel base
and rich in zinc on the coated side. This series of alloys
does not aid materially in the prevention against corrosion,
and it is, therefore, only the outer layer of pure zinc
which functions properly. Investigations made on the
relative corrosion-resisting properties of each of these
processes as compared with the process described below,
show them to be considerably inferior to the latter. They
have the further disadvantage, from the view of the aircraft
industry, that an unnecessary amount of weight is
added and that the t emperatures required to deposit the
zinc are apt to affect heat-treated steels. The processes
involving the use of heat and producing coatings containing
alloys of zinc and iron also have the disadvantage of
being unable to withstand bending. The zinc-iron alloys
formed are quite brittle, and bending causes cracking and
flaking of such coatings.
The electroplating process consists of thoroughly cleaning
the steel base to be coated, immersing it in a tank containing
a solution of a zinc salt, and depositing the zinc
on the steel by means of an electric current. The solution
originally used, and still being used by the electroplating
industry to a considerable extent , is known as the " zinc
sulfate" or "acid zinc" solution. The chemi cal composition
of such a solution is approximately as follows:
ZnS04 •• ••••• • _ •• •••. • •••• _ _ _ __ • 2 pounds.
Al.(S0,)3 • .•• . • _ ___ ___ __ • ·· - ••• ... 4 ounces.
H20 .. . ...... .... ......... .. _. _. 1 gallon.
It has been found , however, that a zinc sulfate solution
does not deposit zinc very well in depressions. In the
case of threaded parts , for example, the thickness of zinc
deposited on the top of the thread is considerably greater
than that at the bottom. A zinc cyanide solution avoids
this objectionable feature. It is, therefore, much more
adaptable to the aircraft industry because of the numerous
parts of peculiar shapes and i rregular parts which occur
in structural steel parts and in steel fittings. It has been
found by atmospheric exposure tests that material coated
with zinc from a zinc cyanide solution will withstand
corrosion longer than material coated with zinc from a
zinc sulfate solution.
The advantage of the electroplating process for coating
a metal is that the thickness of the coating can be very
carefully regulated and that the heat treatment of steels
is in no way interfered with, as is the case with processes
employing heat. The plating process outlined
in this report in no way impairs the physi cal properties
of the steel. It produces a coating of fine, dense,
crystalline structme, which is homogeneous ductile, and
adherent . It should be noted that of all the coatings
tested, zinc, deposited by the electroplating process from
a zinc cyanide solution, offers .the best resistance to
corrosion.
11
"' ,-,-,-r-
I
I'
'
I --- ··.- '
A
,...
~
I ~ ' ~ '
I'
Cl :~
' I ,,
I ,,
, ..
'" ~
'
I ..... ,~
~
I-~
~ I I
~ .
~ .,-.., ~
,- ., '
I
j
~ - I -
C~r) I • I'
I I I/ ". ·~
~
,- I/
i..::
~ \j
~ :=a I
.~ I
' I/
J I/
i I/
j I
I
I/ I
I/ -- ' ' I / J I
i--:-- . .
i j I
I J ,
I I /
I ,, ,.
I I
I I J
" I
I I
/ ,,
J ' I/ ...
. I I, / .,, I/
/ I/ 1/
'J / I / ---"' - ,.,
i I/ I. /-,
I -~ I/ I I I/ I ' -~ I .. .... I ~ ... -L . I
() /On ·/5(} ;:.z...,
O~YS E/< r'IJ_.5'i/fiE
r-, fIG. 1
I/
II
I I/
I
I/
I
I
' I
I
I ..
,.
I
I
I
J II'
IY
I I/'
1
I
I/
" I
I
J
,,
;
I
I
I
I
I
ufl6A!'I IC'
.CQ11TJ1t6.5.
I
I ; + ==t
' 1"
I I
I I I
I r-- '
300 .37~-
I
j
I
j
I
i
I
I
"' I
J
1,1
j
I
I
:
I
l j
:
I!
I
450
Ii
10
-
l'l/1
AA -
,;n
/n --
7d
':O
:qd
~
~
~
~
~
~
~
~
J./CO
12
-- ,. I I I
I I I
I
;
I -l-
I
,_
' ·1
I
I
I
i
I
I I 1.#
I
. 1
I I
I
:
j ...... ...... 7l I I J
I 'I I
I I
I
I t i
I I
I
I
~
~ I- jj
W)
~
~
~
~ ~ I'.! ~
I
~ .
. ~
l
Cr)
~ : I I
~ "' I I, t J , ..
~
r,
} t l I/ J
'1
Ii I/
ti: :· ;
! .T
+ /,
~ ·
~W-. ,, I
1-1/tJ ,
' 1 ...
V
I/
1 ... , ... t,
I/ ... "" 1, ,. .. ~
I I I, I
I - - 1 ·• ,
""' "" ,
' ~ .. I~
0 75 150 zz6
OllYS t:XFt)8t.lflE
FIG. 2.
r.. I
I I , I I
: 'I
~ I CA
/
I I
'
I/ '
'
'
i
I I ' I
I I j j
I ' I . i
I 'I
i '
' I , I
:
I
i '
I +-h-'-
I
I I I I I
I '
I I
' I I ' i i I I I I '
I
I I '
I
I
I I,
I ,
' I
I
; I I
I
'
, /
I i ,
' I I
I I L '
I I
I
I, I
l/1 I
'\I/
"'
'
J
I
I
I
0/'f'BR'/'/IC ·._
C0/171/f{JS
300 .375 45tJ
0
10
ZO
30
..
' --
~
;
I
:
~
~
R
~
~ '::
~
~
~
~
~
R
~ '-i t I(
'()()
"""
)
I::) ~ 1 I,\ I . , 1 .. 1, ~
',> ;
~ "" l.[7 ~
I\
l ~ll+H~~H-+-,-++-+-1-+-1-+-~ .... ~H-+-+-R"t-1....,1-+-++++-+-IH-+-+-H--H-+-H-++++-H-+-H-++++--HH-t+++-H-+-H-++++-+-1-+-H-++-H-H-+-t+++-+-t-+-i-H
N
., il
" 1'
" ... 1,
"' \
\ I\
.... "
1
' ~ '
~ ~
1 ..
~
I'..
~ " ~ '
~L.. ·, I'..
·1
I
~ ~H
1:5~ j
" " . .... , ....
....
I'..
''
I'-
....
..... ....
•ill
"
~"~
"'"~
r"
r"
"
....
, .....
I .... ... ..
~ t I •
/"Cl'fCEl'/rACJC CO,Ff'.Ff't.'.YIO/i l'fL:S'l.5"Tl9 /'ICC
....
I
I
I
. I
TT
....
)
+,---+--t---+-1c-t-t-:i-+--i--+-1'--lj-f~,-t- :t-_tfB±!Pf"/fCfltTIIBt' ~UHF~ -eo;RRb,?ld/l.-.,.+-·-'+_u.,._ 1,- 1:.-.+-,_L+W'·++;--1+-;--+r,-;f_-if-- =,_- +,;.-.+'·+--::;-~-tr-i1---+-t-++-+<-+-j---t--++'--+_-t-_ ~~-:-=::~::::=:
++-+-l---+-1H-+ ~ -~ -l+j+ 1 ,J • ~ '' ~ ·h) I~
_, __
I:)
! i
~~
~ : I - t r, -
+-+-+-t-+ ++-+->-l-+1-+--h-ti-t-H-tt-HH-t-1--t-l-+-t-++t-t-H-+t-Hc++-t--H-+-A'-..±-+-H-+++-IH-+-t--H-+-+-++-+-+-H-++- ~-+-+-+-H-+-H-++-+-H-++-H -+--H--H
~ \ ..
'" ...
, ...
1--1--~, .... ..
, ... ~ :.,
~ I "" I ' ~
~
... ---
~ ~
~~ I
~ I
0 ~~ !
!'"
~ I ' "
' ~ I " "
I l• '
-f'<'~tt-HH-tt-H-t-H--H-t-H-tt--H-+-tt-H-t-H-tt-HH-tt--H-t-tt-H-t-H-tt--H-t-tt--H-t-H--t-t-t- 1-+-+-t-+-t-+-++-+-+-t-++-f""'-+i-i'-l-W
, I i
'-1 I
1,'.•,;
; to ! "'
~ I
, I I
I I ;- ,..,
.. ; f- ~ ~ '" ~ "" ~ ·H- H-· ~ I ~ ~ --~
~ ri- i-+ ~ L..!.-t--~ ~ I' _+,_'
I '-r:r ll) ' ...
.--; ;. I ; '
4\ 4= r+ . f-++_d >-·fh
II I -++:-::-:: 1--:-·· .
~ -(, ·· - ,_.
~ ,__ + I
I +---_!_~ ; -
1- - I I
.. \ "' ' ~>-+-+-+-+-r-+-++-++-t-+-t-t-+-i-t-+-t-+-t-+-t-11-1-·t- :
•-+-+-+-++-+-t-+-r-+-1-+'0,--+-t-f·· f-- ~~ -: - 1-r ·l-+-t-+-+- r-+-++-H-t-t-++-+-+-+-+-+-+-t--1-t-t-++-+-+·+-+++-+r-1-t-++-t-+-+-+-+-+-+-•·-l-t-+-+-+ e- 1-r-1--
. ~ Tr- - \
..-+-+-<-t-+-+-+-+-+-r-+-++-+-t->-~-+-l-+-l-'--t·-t·. ..,.... .,-t-·-+-+-tr-f- -1--+-+-+-I-+-...... +-H-t-,-++-+-+-t-+++-+· •-'-
l-'--++-+ ,1-t-++-t-+Hl-+-++'+-+H-++-H-+~ r 14~--r-+-++4+-+4-+-1-t-+-+-++ -< 4 -+-r-+-++---t-t~- r-+--+-+-->-++-,---+--t~+-+---+--t-+--- -+-+-tr--+-++---t-t-+-t-<
H-,H ·-t-l-·r-t-++-f-+r-t-+ H-+-+'.t--+ :- · - r -f- ·-· · - ~ ,'<; '- - !-t-++-t-t-t-t--1- _j_ -+-+-++-+-<-+->-+-<-
++-rl"'lct-+-+-t-+-t-t-+-t-!-·~ - -+-+-+-r-+-++-tr---+-+-+--+--t~- +-+--+-+-1- r-+-++-t-+1-t-+-1r-<
~ ~ ~ ~ ~
r'cRCclt7rl6f' C//lf'RtJ.5'/tJ/1 t
l'i'L'"SISTIYICC
Air Service Information Circulars
