I28 T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y .
April, 1910
BACTERIAL ACTIVITY AS A CORROSIVE INFLU-
ENCE IN THE SOIL.
H. GAINES.
B y RICHARD
Received January 7, 1910.
One of the many subjects presented for study in
connection with the construction of the Catskill
Aqueduct was the corrosion of iron and steel.
I n a communication to the American Electrochem-
ical Society, in 1908,' the writer discussed a t some
length two inherent causes for corrosion of steel,
namely, irregularity of structure and segregation of
impurity. Experience has shown that metals which
are non-homogeneous in character are far more sub-
ject to deterioration than those which have the same
composition and structure throughout.
Conclusions reached from the study referred to
were that although the cumulative decomposition of
materials like steel may be conditioned by impurity
the operation of this cause would be negligible if the
process were not vastly hastened by some external
impulse. Moreover, since i t is probably impossible
to modify greatly the composition of steel and retain
the physical properties that render i t useful, the
solution of the corrosion problem would be found in
controlling the ascertained accelerating influences.
A great deal has been published relating to the
corrosion of iron and steel during the last few years,
with much importance attached to the electrochemical
theory of Whitney, on this side the Atlantic, while
in Europe, attention has been chiefly concentrated
on the effect of the presence or absence of carbon
dioxide.
On taking up the present work, it was deemed best
to direct attention to actual phenomena rather than
to be influenced b y preconceived opinion and to base
theoretical conception only on observed facts. Due
examination had led to the conclusion that none of
the existing theories of corrosion account sufficiently
for some of the observed phenomena. Only a phase,
of the research a t the laboratory of the Board of
. Water Supply will be touched upon here. I n con-
nection with the series of studies in progress, analyses
were recently made of about twenty-five samples of
rust, collected from pits in the outside, and tubercles
on the inside of steel conduits located in different
parts of the country. The analyses showed, within
certain limits, a remarkable difference in the character
of the samples of rust, varying according to the under-
ground conditions of exposure to which the iron or
steel may have been subjected. The results appeared
to indicate the operation of hitherto unconsidered
influences accelerating the corrosion process.
For the purpose in view here, i t will suffice to men-
tion b u t two of the ingredients in the samples of rust.
Attention was early attracted to the high percentages
1 Transaclaons of the Amerzcan Electrochemacal Soczely. 13, 5 5 et
sepuzzur.
of sulphur in combination, and the invariable presence
of more or less organic matter in all the materials
examined. Through the courtesy of Mr. T. Merriman,
at whose instance the analyses were made, permission
was granted to cite as examples results obtained from
some of the samples of rust collected from different
steel conduits (names must be omitted) in which
combined sulphur calculated as sulphuric anhydride
was found as follows:
SOs, per cent.
Tubercles from interior of Conduit No. 1 . .
,, '<
. .. .
. . .... . 1.41
" " I<
.
No. 2 . .... .. . . . . . 2.25
'l < I C
I ,I '# . . ..
NO. 3 . . . . . . . . . 2.98
(, / <
..
Rust from "pits" in exterior of Conduit No. 4 . . . . , . . 1.51
I' ' I /' d <
',
"
No. 5 ......... 3.95
" ,# I' 'I 'I "
No. 6 ........ . 6.50
The very high sulphur content shown in the last
of these examples was perhaps in part due to the
peculiar soil conditions surrounding the corroded
conduit, and was, no doubt, a n exceptional case.
The soil backfilled about this pipe gave a loss on
ignition of 85 per cent., and consisted largely of de-
composed organic matter. Whereas the samples
of rust collected from several steel conduits gave a n
average of nearly 3 per cent. of sulphuric anhydride,
analyses of pieces of steel cut from the same conduits
showed only 0.05 per cent. or less of sulphur. I n
contrast with these figures, tubercles artificially
formed (by connecting with a gravity cell, steel plates
immersed in a solution of sodium chloride) gave on
analysis less than 0.15 per cent. sulphuric anhydride.
As is well known, finished steel of the character used
in conduits neyer contains more than a minute fraction
of I per cent. of sulphur, which is present as man-
ganese sulphide, ferrous sulphide only appearing
when the quantity of manganese is insufficient. The
proportion of manganese in the samples of rust did
not correspond with what would be expected if man-
ganese sulphate were the first oxidation product.
On the contrary, the percentage of manganese found
in the rust was no greater than in the steel, and gen-
erally less than 0.35 per cent.
We know that iron has a tendency to form com-
pounds with sulphur just as i t has with oxygen. It
is not, therefore, the mere presence of sulphur, but
the occurrence of its compounds in such quantity in
connection with the rusting of iron that is significant.
It was believed that a satisfactory explanation of
this phenomenon would shed new light on the rapid
deterioration of underground iron and steel structures
by corrosion.
In casting around for a theory to account for the
singular presence of sulphur or its compounds in the
quantities found associated with the rust materials,
early consideration was given to the possible r61e
played by bacteria. Many decompositions hitherto
unsuspected, and chemical changes chiefly destructive
in character, are no doubt accomplished by these
organisms. I n certain prototrophic forms a process
 G A I N E S ON B A C T E R I A L A C T I V I T Y A S A CORROSIVE I N F L U E N C E . 129

goes on comparable to respiration, b u t consisting in as their source of energy much as higher organisms
the oxidation of inorganic compounds. Belonging employ carbohydrates-instead of liberating energy
to this class, some of the thiobacteria possess the as heat by the combustion of sugars, they do i t by
unique power in organic creation of breaking up oxidizing hydrogen sulphide. One of the anaerobic
sulphur compounds and assimilating into the cells forms ( S p i r i l l u m dzsulphuricans) attacks and reduces
of their protoplasm pure sulphur, which is subse- sulphates, thus undoing the work of sulphur bacteria
sequently oxidized into sulphuric acid. Leaving just as certain denitrifying bacteria reverse the opera-
aside tempting speculation concerning the mode in tions of nitro-bacteria. Again, we have sulphur
which life interacts or is associated with matter, taken into the higher plants as sulphates, built up into
the d a t a at hand seemed to warrant the belief that a n proteins, decomposed by putrefactive bacteria, and
important connection existed between bacterial activ- yielding sulpliuretted hydrogen which the sulphur
i t y and underground corrosion. Whatever pro- bacteria oxidize, the resulting sulphur is then oxidized
duces acid conditions in the soil must contribute in to sulphuric acid and again combined with calcium
no small manner to the corrosive influences present. to gypsum, the cycle being thus complete.
T h a t acid compounds are formed in abundance as The well-known part played by iron bacteria aEords
the result of vital activity is easily demonstrated. a n analogy to the circulation and chemical changes
Important recent advances in our knowledge of just described wrought by the life processes of sulphur
bacteria are those having reference to the agency of bacteria in the widely distributed sulphur compounds
these organisms in the circulation of certain elements in nature. Pools and marshes near iron mines abound
in nature. The functions of nitrogen and iron bacteria in bacteria, some of which belong to the remarkable
in this connection have long been known. Definite genera crenothrix, cladothrix and heptothrix and
anaerobic forms also exist which h a r e the power of contain ferric oxide in *their cell walls. This iron de-
fixing carbon and sulphur as well as nitrogen from posit is not merely mechanical, b u t is due to the phys-
inorganic sources. Owing to the peculiarities in the iological activity of the organism which liberates
modes of nutrition and respiration of these bacteria energy by oxidizing ferrous oxide in its protoplasm.
(vzde qnfra) they require neither free oxygen nor The iron must be in certain soluble conditions, and
organic food materials for vital activity. the soluble carbonate of the protoxide seems most
I t is now well known that a whole series of sulphur favorable. The hydrocarbonate absorbed by the
bacteria exist of the genera thiothrix, chromatium, cells is oxidized probably thus:
spirillum and monas, which play important parts
both in the circulation and concentration of sulphur
zFeCO, -k 3H,O + 0 = Fe,(OH), zC0,.
in nature. There exist, moreover, in the mud of The ferric hydroxide accumulates in the sheath and
marshes anaerobic bacteria which decompose cellulose, gradually passes into the more insoluble ferric oxide.
probably hydrolyzing i t first and then splitting the These actions are of great importance in nature, as
products into carbon dioxide and marsh gas. When their continuation results in the enormous deposits
calcium sulphate is present, the nascent methane set of bog iron ore.
free b y the cellulose bacteria reduces the sulphate The writer has recently been informed of a serious
with the formation of calcium carbonate, sulphuretted case of corrosion in the West, in which the evidence
hydrogen and water. This is the explanation of the is apparently conclusive that the damage was due to
occurrence of marsh gas and sulphuretted hydrogen srecific bacteria. During repair work on the founda-
in bogs, and such conditions afiord fax-orable media tion structure of a bridge across Lake Hauser, in
for sulphur bacteria which multiply by oxidizing the Montana, the attention of the engineer in charge
sulphuretted hydrogen and storing the sulphur in was attracted to protuberances which occurred with
their own protoplasm. Processes resulting from the more or less regularity on the steel work of the support-
bacterial decomposition of yegetable mud only take ing tubes. The steel construction men designated
place a t certain depths, and the zone of physiological this “shell rust” and said that i t was often seen on
actiJ-ity rises and falls with the variations of partial steel which had been under water for some time.
pressures of gases due to the rate of evolution of ilk. \Tilton G. Brown, a Montana naturalist, made a
sulphuretted hydrogen. I n the deeper parts of this careful examination of the protuberances and found
zone the partially anaerobic bacteria absorb the sul- that this was not a case of common rust action, but
phuretted hydrogen and as this rises and meets at- under the center of each individual a n irregular pit
mospheric oxygen other bacteria oxidize it and store was eaten into the steel. An inspection of some
up the sulphur; then ascending into planes more highly of the rust material under the microscope soon led
oxygenated other bacteria further oxidize the sulphur to the conviction that some organism had played a part
to sulphuric acid, which combines with any calcium in the destructive corrosion. The phenomena were
carbonate present to form sulphate again. These reported to Professor Beck, of Freiburg, Germany,
bacteria, therefore, employ sulphuretted hydrogen who suggested that such vigdrous action was probably
 T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y . April, 1910
due to bacteria. Specimens of the rust were then trary is a complex and highly skilful procedure,
submitted to the eminent bacteriologist, Dr. Shorler requiring expert labor and involving the application
of Dresden, who identified Gall ionella jerrugi9zea in of many coatings.
great abundance in the 'material. The method of The object of car painting is both for protection
attack of this organism is first the elimination of a n and for decoration, although the latter consideration
acid product b y which the iron is dissolved, and has exerted the greater influence on the modern
then each individual forms a pit by eating out the practice of car finishing. I t is possible to preserve
metal. The structural damage was reported to have the woodwork of a car body just as efficiently by
been enormous for the short period of less than a year frequent painting with suitable oil paints as by cover-
that the bridge had been in service. ing i t with the ten to fifteen coats of paint and varnish
The foregoing considerations apparently warrant customarily applied. This is well illustrated by the
the conclusion that the corrosion of underground iron practice of the Interborough Rapid Transit Company
and steel structures is in part due to bacterial activity: of Xew I'ork, who paint the bodies of the subway
( I ) Directly, or by the attack of a specific microbe. cars with a dark red house paint. Not a vestige of
( 2 ) Indirectly, or by creating acid conditions in the ornamentation is used and the cars are not even
soil. Certain bacteria present, owing to peculiar varnished. This is an extreme case, but the practice
modes of absorption and elimination of sulphur and has been adopted by an experienced management,
its compounds convert harmless substances into and indicates the simple method of car finishing which
corrosive acids. can be used where the appearance of the car is of no
The remedies proposed are: importance. The steam or electric car operated on
( I ) Free drainage quickly carrying off the acid the surface, however, occupies a prominent position,
solutions formed should arrest' or prevent destructive and the public justly demands that i t present a well-
action. kept exterior. The appearance and condition of the
( 2 ) I n localities or situations where drainage is cars is frequently considered a n indication of the
impracticable, slack lime should be packed about liberality of the management, and as a result many
the metal, to neutralize acids formed as a result of companies hare carried this idea to an extent which
bacterial activity. brings an unnecessary burden on the earning capacity
147 VARICK ST., NEW Y O R K . of the system. The cost of painting the same type
of car varies on different roads from $30.00 to $60.00,
and in certain cases a n even larger amount, while
[CONTRIBUTION FROM ARTHURD. LITTLE,INC., O F ENGINEER-
LABORATORY
ING CHEMISTRY.] some roads are forced to repaint their cars every two
VARIATIONS IN CAR PAINTING PRACTICE. years, and others with the aid of one coat of varnish
By CARL F. WOODS. each year are able to operate for ten to fifteen years
Received January 10, 1910. before complete refinishing becomes necessary. I t
The past few years have witnessed a wide-spread is particularly significant that those c9rs which have
movement in this country towards the standardiza- had the most expensive finishing are not of necessity
tion of paint products. Public attention has been the longest-lived. It is obvious, therefore, that there
called to the prevalence of adulteration and to the are certain underlying principles upon which the
practice, particularly in mixed paints, of selling goods durability of the finish depends.
of short weight, while certain investigators ably sup- Car paints as a rule are mixtures of liquids and
ported b y the reputable paint manufacturers have solids having widely diri'erent chemical and physical
revealed many such frauds and have taken decisive properties. Between the priming coat and the final
steps toward the education of the consumer and the varnish, there are not less than twenty elements
establishment of definite standards. Important data intimately in contact, some of which have a n affinity
are being accumulated on the relative value of the for each other, while others may tend to dissociate.
various pigments and vehicles, but comparatively This is particularly true of varnishes, for two different
little attention is being given to the proper application brands of equal durability may so react as to mate-
of the paints themselves. The responsibility for rially injure the wearing qualities of either one. While
poor results in painting rests both with the painter each succeeding treatment has its own specific demands,
and the paint manufacturer, for although inferior the entire paint coating must amalgamate and act as
materials cannot be made to give long life even in a unit to prevent separation of the various films
the hands of the most skilful painter, i t is equally under the physical stresses of service, produced by
true that good paint improperly applied affords no the expansion and contraction of the car under changes
better results than poor paint. There is no class of of temperature, and the wrenching and twisting in-
painting in which this is more clearly illustrated than cidental to operation. The importance of selecting
in that of car finishing, for this is not a comparatively the right composition for the successive coats is self-
simple operation like house painting, but on the con- evident, but it is equally essential to employ a method