Sie sind auf Seite 1von 9

Action of Rust-Preventive Oils*

By E. -R. Barnum,* R. G. Larsen" and A. Wachter*

USTING of iron and ferrous Basic Function of Protective Oils
alloys has been a problem for Since water must be present to
centuries. Paints have been long enable the reactions of atmospheric
employed as protective coatings corrosion to occur, it follows that
while grease and petrolatum have rusting can be prevented if moisture
been used extensively to provide _which may be on the steel surface
protection when removable coatings is removed and no further access
were required. Oil films, which are permitted. To achieve protection of
much easier to apply and -remove, steel by exclusion of oxygen is im-
were inadequate for this purpose practical because of the relatively
until it was found that their rust- high solubility of oxygen in hydro-
proofing efficacy could be greatly carbons.
increased by addition of certain Although hydrocarbon oils do not
types of polar organic compounds.1 provide effective rust-preventive
World War II gave strong impe- films, this property is acquired by
tus to development of such com- addition of certain polar organic
pounded oils for a wide range of compounds (polarity meaning that
service requirements. These oils are a separation of electric charges, or
used extensively now to provide dipole, is present in the molecule).
temporary protection for ferrous When steel is in contact with a rust-
metal articles during manufacturing preventive oil, the polar compounds
operations, storage and shipment, as are adsorbed in oriented layers at
well as for protection and lubrica- the steel-oil interface to form a semi-
tion of machinery, guns, turbines rigid lattice containing oil. The com-
and hydraulic systems.1 bination of oil and polar compound
The urgency of military needs hinders the ingress of moisture to
during World War II was largely the steel surface. Evidence for this
responsible for the adoption of em- basic explanation is presented ~elow.
pirical methods in development of
protective oils. This article presents Adsorption of Polar Compounds
research undertaken to reveal the on Steel
mechanism by which polar com- Other investigators have shown
pounds impart rust-preventing qual- that, when a homogeneous oil solu-
ities to oils. tion of polar additives contacts a
metal surface, the additives are con-
*A paper presented at the Annual Meeting centrated at the metal-oil inter-
of NACE in St. Louis, Mo., April 5-8, 1948.
* Shell De vel oprnen t Co., Emeryville, Calif. Iace.v" It therefore seems reason-


able that adsorption 5

would play an impor- 45-
tant part in determin-
ing the behavior on ~4q (f)
steel of oils contain- . ~ 35
ing polar compounds. b Figure 1-Effect of additive concentration
Hence quantitative ad- .z30
Q on interfacial tension between distilled
sorption studies were water and oil solution of additive.
made of different _J
<t 20 -
types of polar com- 0~
pounds, some of which ~w 15-
had proved to be ef- ~ 10I-

fective additives in oil.

Adsorption was de-
termined by stirring a 0
I _i___::_.:L __ .-:,
01 02 03 04 05 10
iron powder with a
solution of organic compound in iron powder was determined by ad-
a highly refined mineral oil for one sorption of nitrogen on the powder
hour at 30 C. and then filtering off the according to the method described
iron powder. The amount of additive by Brunauer, Emmett and Teller.'
removed by adsorption on the iron was From the surface area, it was calcu-
calculated from the decrease in concen- lated that compounds of Type III
tration in the oil. This was measured are adsorbed to a maximum thick-
by observing the change in inter- ness of 1 or 2 molecular layers,
facial tension (between oil and wa- while the more highly adsorbed
ter), which is very sensitive to con- compounds represented by Curves
centration of polar additives in the I and II gave layers several mole-
oil, and comparing the observed cules thick. Although the exact num-
value with a calibration curve (Fig- ber of the latter could not be deter
ure 1). Interfacial tension was meas- mined because the maximum degree
ured with a du N oiiy interfacial ten- of adsorption could .not be estab-
siometer. Since significant increase lished by the method used, it is ob-
in interfacial tension with time oc- viously considerably greater than 1
curred for some oil solutions, varia- or 2 molecular layers.
tions due to this factor were mini- All of the additives that showed
mized by aging all oil solutions for a high degree of adsorption, illus-
at least 16 hours at 30 C. before trated by isotherms of Types I and
measurement. II, were good rust inhibitors, while
By measuring adsorption at vari- the additives showing low adsorp-
ous additive concentrations and tion, exemplified by Type III, were
plotting the adsorbed quantity as a poor. Since only one additive was
function of additive concentration, examined that gave the Type II
three general types of adsorption curve, no correlation can be made
curves (isotherms) were obtained concerning the shape of the adsorp-
(Figure 2). The surface area of the tion curves and rust-proofing effi-

cacy of the additive involved. How- ated from the monomolecular film
ever, the relationship between the by a waxed thread. By observing
extent of adsorption and rust pro- the motion of the thread, the num-
tection indicates that a multilayer, ber of monomolecular layers trans-
rather than a monomolecular layer, ferred to the steel in repeated im-
-of rust-preventive additive on the mersions could be determined. The
steel surface is required for good number of monomolecular layers
rust protection. and orientation of the additives 111
Multimolecular Surface Films these films are summarized in
Further evidence on the impor- Table I.
tance of multimolecular adsorption Humidity cabinet corrosion tests
was found in a study of "built-up" on films applied in this manner re-
films on steel surfaces. By means of vealed that, with one exception,*
a technique similar to that described they were ineffective in preventing
by Blodgett,5 films of a known num- rusting regardless of the number of
ber of molecular layers of polar molecular layers present (Table I).
compounds on steel were prepared However, some of these polar com-
as follows: An oriented monomolec- pounds were known to be effective
ular film of the compound was first rust inhibitors when applied in oil
spread on the surface of distilled solutions. Therefore, humidity cab-
water in a trough and then trans- inet tests were made with steel
ferred to a polished steel panel by which had been dipped in oil after
lowering the latter (in a vertical po- first coating with comparable multi-
sition) beneath the surface of the layer films. In the presence of oil
water and then withdrawing it (Fig- improved performance was found
ure 3). As the panel moved through for certain types of compounds ap-
the surface, constant pressure was plied as a multimolecular film. It
maintained on the molecular film by may be noted that even additive E,
means of a film of castor oil separ- which provided some protection
alone, was much more
. ffi effective with oil. No
0 protection was observed
for films one or two
z 012.
Q;: molecules thick regard-
<( less of the type of com-
~ .009- pound involved. It ap-
w Figure 2-Types of adsorp- pears that a minimum
a.. tion isotherms for various
w additives on iron at 30 C. of about six molecular
0006- layers was required to
TYPE Il impart rust protection.
w It is recognized that
~ .003
"built-up" films of ad-
c::t ditives do not neces-
~ TYPE m
<[ * Additive E in Table I,
er which showed much closer
"' 0 I 2 3 4 5 6
on. steel than any

to substantiating the theory that

multimolecular adsorption is an im-
portant aspect of rust prevention,
these results show that oil does not
merely provide a medium for plac-
ing the additives on the steel but
also becomes a part of the protective

Chemical Structure of
Rust-Preventive Additives
Several types of organic com-
pounds mel;y serve as 'rust -inhibitors
in mineral oils. All have one charac-
Metal through A is coated on inside with wax .. Mono-
molecular layer of polar compound is designated area teristic in common, namely, a large
B, while area occupied by castor oil is marked C. Curved hydrocarbon group attached to one
line D is waxed thread which separates the two. Polished
steel panel E is raised and lowered through water sur- or more smaller polar groups. If
face by a windlass in order to get uniform motion.
only one polar group is present, it
must have sufficient dipole moment
Figure 3-Apparatus for preparing "built-up"
surface films. to be highly adsorbed at the steel-
oil interface. Compounds such as
0 0
sarily represent the same type of orien- II II
tation or degree of adsorption as that R-S-0-Na and R-O-P-0-:R
produced by direct contact of oil so- II I
0 OH
lutions. Hence the same additives Sodium petroleum
sulfonat e''
were deposited on steel from solu-
tions in a volatile hydrocarbon and are representative.
after the solvent had evaporated The necessary adsorption may
were subjected to humidity cabinet also be provided by two or more
tests. These, like the "built-up" weaker dipoles in the same molecule
films, were ineffective in preventing as represented by the following com-
rusting (Table' I). Thus, in addition pounds:

H H 0 0
R-'C.COOH I !I HH/ -,
R-C-COOH and R-C-OCC C-H.,
I HI I -
H _l_
/ -. """ c
"""- _/ - COOH

a~ Mercaptostearic a-(2-carboxy phenoxy)- Sorbitan rnonooleate"

acid" stearic .acid"

However, a molecule
having only one group - - - ---------------------------- -------------- --------- ---------

with weak attractive

forces, such as R-OH,
0 - Ywo _
R-C=N and R-C-NH2
i's incapable of impart-
ing significant protec- - --------WATER .:_ _

tive action to the oil.

Displacement of
Water by Oils D wo = r so - r sw = r wo Gos 8

Although rust-pre- The receding angle, (), is the stable angle obtained by rotating the steel strip to
give a continuous horizontal interface at A. Interface at 8 would not be a
ventives are usually continuous horizontal line unless () is 90.
applied to dry sur- Figure 4-lnterfacial contact angle between oil, water and steel.
faces, the protection
of previously wetted steel parts, applying the preservative, it can also
such as those from grinding, mill- be provided in one operation by use
ing, and quenching operations, must of water-displacing preservative oils.
be considered on occasion. While . Although straight mineral oil will
protection here can be accom- not displace water, this desirable ac-
plished first by drying and then tion can be made to occur by adding

Properties of Films on Polished Steel Surfaces

---------- ------------------,---------------,-------
Transferred from Water Surface Deposited from
Transferred from Water Surface and Dipped in Oil Volatile
No. of No. of
Monomolecular Humidity Monomolecular Humidity Humidity
Layers and Cahinett Layers and Cabinett Cabinet
Additive Type* Life (Hrs.) Type* Life (Hrs.) Life (Hrs.)
None ..
----- <24 <24 24
A .... 2Y <24 2Y <24
A ... 18Y + 22X <24 :32y 14X+ 430 24
B ...... 2Y <24 2Y <24
B ...... :32Y <24 32Y 264
C ....... 2Y <24 2Y <24
c ..... . 24Y + l9X <24 42Y + 2X 196
D ..... 6Y <24 6Y
216 24
E .... 6Y 88 216 24

*Y films consist of "alternating" layers of polar compounds o o o 0 o
X films consist of "nonalternating"' layers of polar compounds l l Jl l
t Humidity cabinet of the type and operated under conditions described in U. S. Army Specification 2-120
March, 1944. The number of hours represents the time for 10 percent or more of panel surface to rust,

The sand-bliisted steel panel on left is covered with SAE 30 mineral oil, that on the right bears the same oil film
containing an additive; both subjected to falling drops. of dilute (0.05%) sodium chloride solution for 15 minutes.
Figure 5-Effect of polar compounds on rust-proofing efficacy of mineral oil.

to the oil certain polar compounds steel interface if the energy of the
which have a greater tendency than system decreases as a result of this
water to be adsorbed on steel. How- action. The difference in surface
ever, water displacing oils are not energy (Dw between a steel-oil and

necessarily preservative oils, nor is a steel-water interface (interfacial

the converse true, as discussed later. tensions Y. and Y sw respectively) can

The tendency of an oil to displace be calculated from measured values

water from steel can be estimated by of the contact angle (0) and the
considering the relative surface water-oil interfacial tension (Ywo)
energies at all the interfaces in- according to the equation
volved (Figure 4). A water-steel Dwo = Yso -Ysw = Ywo COS 0.
interface will be replaced by an oil- It follows that displacement of

Sand-blasted steel panel on left is covered with a film of SAE 30 mineral oil, that on right bears the same oil
containing an additive; both subjected to the humidity cabinet at 100 F. and 100% relative humidity for 24 hours.
Figure 6-Effect of polar compounds on rust-proofing efficacy of mineral . oil.
water by oil should be expected angles are seldom obtained experi-
when e is between 90 and 180; on mentally;" the water-advancing
the other hand, displacement of oil angle usually being larger than the
by water would be expected when e water-receding angle. Since measure-
is between 0 and 90. ment of the water-receding angle
Since the above equation repre- involves the movement of an oil
sents an equilibrium of forces, the interface over an area previously
contact angle should be an equili- occupied by water, it might be ex-
brium value, i.e., one that does not pected that displacement energies
change with time and is independent of calculated from such measurements
the manner in which it is formed. would be a better indication of the
However, equilibrium values of contact water-displacing tendency of oils.
Comparison of Displacement Energy ( Dwo) and Water-Displacing Action

Oil-Water lnterfacial Observed Water*

Interfacial Contact Dwo Displacing
OIL Tension (dynes/crn.) Angle (e) erga/cm.s Action
Mineral Oil , .
Mineral Oil +Additive A . 23.6 50 23.5 None
Mineral Oil + Additive B . 8.9 50 8.8 None
Mineral Oil + Additive C . 8.9 140 -6.3 Good
Mineral Oil +Additive D . 4.6 150 -4.0 Good
Mineral Oil +Additive E . 20.3 115 -8.6 Good

*See Figure 7 for description of wat cr-displaclng test.


Comparison of displacement ener-

gies, calculated from water-receding
angles and oil-water interfacial ten-
sions,* with results of water-dis-
placement tests shows good correla-
tion between the sign of the dis-
placement energy and water-displac-
ing action (Table II). All the oils of
negative displacement energy are
effective in displacing water from
steel, while those with a positive
displacement energy are poor in this
The fact that some additives act
as water-displacing as well as rust-
preventive agents, while others ex-
hibit only one of these effects, prob-
ably is a function of the relative
distribution of the additive at the
various interfaces. While combina-
tion of adsorption at both the oil-
water and steel-oil interfaces would
remove water from steel, subsequent
rust protection depends upon the
magnitude of adsorption and chemi-
cal structure of the compound at the
steel-oil interface as well as the
thickness of the oil film. Hence,
water displacing additives are not
always good rust-preventive addi-
tives. Conversely, some additives
may be sufficiently adsorbed on dry
steel to improve rust protection but In this test, sand-blasted steel rods were dipped momen-
tarily in 0.05% sodium chloride solution, then immersed
may not lower the oil-water inter- in the designated oil and the water-displacing action
observed. After . one minute, they were removed and
facial tension enough to displace allowed to hang in the laboratory 24 hours. The rod
on right was immersed in. an SAE 30 mineral oil, that
water from previously wetted steel on the left was immersed in the same oil containing a
polar additive; both photographed after 24 hours.
Figure 7-Effect of polar compounds on the water-
displacing action of mineral oil.
Other Factors
While the results of these experi- expected to influence the behavior
ments provide a basic explanation of of such oils. Piltz and Farley1 have
the action of rust-preventive oils, shown that low solubility of the ad-
. there are other factors which maybe ditives in water favors rust-preven-
* In these expertment.s interfacial contact an-
gles were measured by a modification of the
tive efficacy. They also indicated
method of Wenzel12 and the oil-water inter- that increasing temperature and de-
facial tensions with a du Notiy interfacial ten-
siometer. creasing molar concentration of the

additive would decrease protective Summary

action. The following mechanism by
Furthermore, qualitative observa- which oil solutions of polar com-
tions have indicated that the thick- pounds protect steel is proposed:
ness of the oil film, the manner in
Polar molecules are adsorbed as
which it is applied, formation of
oriented multimolecular layers at the
micelles, composition of metal sur-
steel-oil interface.
face, surface finish, and nature of
the exposure environment all may These layers retain oil in their
affect the protection afforded by. the interstices, forming a mixed film
oils. While adsorption plays an im- which is more impervious to water
portant role in rust protection, it is than either the additive or oil alone.
also possible that some additives Additives with weak attractive
(especially acids) react with the forces forming only mono- or bi-
iron or its corrosion products to molecular layers are not effective
form salts which are insoluble in rust inhibitors. A minimum of about
both oil and water and hence re- six molecular layers is required to
main on thee steel surface as a pro- form an effective barrier against
tective barrier. moisture.
These and probably other factors Compounded oils displace water
influence the protective character- from steel if the adsorptive forces
istics of compounded mineral oils. of the additive for steel are greater
Considerably more research would than those of water for steel and
be necessary to evaluate properly consequently form a more stable
the importance of each. interface (one of lower energy).


1. G. P. Piltz and F. F. Farley, Rust crs on a Solid- Surface. J. Am. Chem ..

Preventive Oils. Use of Contact An- Soc., 57, 1007 (1935).
gles to Study the Action of Mineral 6. R. E. Watkin, U. S. Patent No.
Oil Films. Ind. Eng. Chem., 38; 601 1,630,101 (May 24, 1927).
(1946). 7. H. F. Denning, L. G. Berton, and R.
2. H. Karplus, Der Aufbau der Schmier- G. Clarkson, U. S. Patent 2,080,299
schicht und die Kolloidgraphitsch- (May 11, 1937).
mierung. Petroleum Z., 25, 375 (1929). 8. E. R. White, U. S. Patent 2,368,604
3. J. J. Trillat, Recherches sur les phe- (January 30, 1945).
nomenes de lubrifaction au moyen 9. E. R. Barnum, U. S. Patent 2,369,640
de la spectrographie par les rayons (February 20, 1945).
X. Compt. Rend., 182, 843 (1926); 10. T. E. Sharp, U. S. Patent 2,398,193
Ibid, Sur une nouvelle methode de (April 9, 1946).
spectographie par les rayons X. Ap- 11. F. E. Bartell and P. H. Cardwell,
plication a l'etude de l'orientation des Reproducible Contact Angles on Re-
acides gras par le mercure. 187, 168 producible .Metal Surfaces. II Inter-
(1929). facial Contact Angles Between Water
4. S. Brunauer, P. H. Emmett, and E. and Organic Liquids on Surfaces. of
Teller, Adsorption of Gases in Multi- Silver and Gold .. J. Am. Chem. Soc.,
molecular Layers. J. Am. Chem. Soc., 64, 1530 (1942).
60, 309 (1938). 12. R. N. Wenzel, Resistance of Solid
5. K. J. Blodgett, Films Built by Depos- Surfaces to Wetting by Water. Ind.
iting Successive Monomolecular Lay- Eng. Chem., 28, 988 (1936)