J. S. A . Inst. Min. Meta11., vol. 87, no. 12.Dec. 1 87. pp. 425-437.

M crobial corrosion of iron-based alloysby c. RINGAS. and F.P.A. ROBINSONt

SYNOPSISVarious theories proposed to account for microbial corrosion are reviewed as a background to a better understanding

of this type of corrosion in iron-based alloys. Experimental evidence is presented to show that even stainless-steelalloys are prone to microbial attack by sulphate-reducing bacteria. The corrosive attack by the bacteria is due primarilyto the hydrogen sulphide that is produced as part of their metabolic processes.

SAMEV A TTINGV~rskillende teoriee wat aan die hand gedoen is om mikrobiese korrosie te verklaar, word hersien as agtergrond

vir 'n beter begrip van hierdie soort korrosie in legerings met 'n ysterbasis. Daar word eksperimentele bewyseaangevoer om te toon dat selts vtekvryestaallegerings onderhewig is san mikrobiese aanvalle deur sulfaatreduserendebakteriee. Die korroderende aanval deur die bakteriee is in die eerste plek toe te skryf san die waterstofsulfiedwat as deal van hul metaboliese prosesse voortgebring word.

INTRODUCTION

Co siderable attention has been devoted to the micro-bial c rrosion of ferrous metals, and, in particular, themicro ial corrosion of mild steel has been the subject ofintens investigation since 1934. The microbial corrosionof alu inium and its alloys has also been studied exten-sively since this first manifested itself in the form of in-ternal corrosion of aircraft fuel tanks. Less attention hasbeen aid to the effect of microbes on stainless steel,perha s because these alloys are regarded as corrosionresist n1. However, over the past ten years, the incidencesof mi robial corrosion of stainless steel have been on theincre se.

Th paper discusses the theories that have been pro-posed to account for microbial corrosion. Various failures

.. of st inless-steel components are discussed in order toemph size that stainless steels are not immune to micro-bial tack. Experimental evidence is presented to showthat tainless steels are attacked in pure cultures ofsulph te-reducing bacteria (SRB).

Fig 1 is a schematic representation of a pipe, showingthat any metals and alloys are attacked by microbes,inclu ing stainless steel, copper alloys, and low-nickel-cont °ning alloys. Only titanium and some high-nickelalloy seem to be immune to microbial attack. Copperalloy, although resistant to biofouling, can still be cor-rode in the presence of SRB. It is interesting to note thatmate ials such as plastics, concrete, and hessian are alsodegr ed by microbes.

It ° apparent from Fig. 1 that various organisms areresp sible for the microbial-corrosion problem, but theSRB re responsible for the bulk of these. Consequent-ly, re attention has been paid to the SRB.

S cause problems in the offshore oil industry by cor-rodi pipelines, reducing the value of the oil by raising

. R earcher.t Pr fessor of Corrosion Science and Engineering.

B h the above of the Department of Metallurgy and MaterialsE ineering, University of the Witwatersrand, I Jan Smuts Avenue,Jo annesburg 2000.

@ T South African Institute of Mining and Metallurgy, 1987. SAIS N 0038-223X/$3.00 + 0.00. Paper received 18th October, 1986.

its sulphide content, and posing a threat to workers onthe platform through the production of toxic hydrogensulphide. The failure of high-strength steels used in theoil industry due to hydrogen sulphide (sulphide stresscracking) is another area in which the SRB have been im-plicated. Hydrogen-induced cracking of pipeline steels ex-posed to sour media is a well-known problem. The sourceof the hydrogen sulphide is immaterial. Furthermore, anymetal immersed in seawater is liable to become colonizedby a wide range of organisms from bacteria to algae!.This fouling can lead to a number of problems, such ashydrodynamic loading, hindrance of corrosion inspec-tion, and severe corrosion beneath the fouling layer.

In the metal-working industry, there have been nu-merous cases of microbial contamination of the emul-sions, lubricants, and coolants used in machining, wire-drawing, rolling, and deep-drawing operations. Thisusually results in the corrosion of drawing dies, wires,and sheets, and in the staining or deterioration of sur-face finishes2.

SULPHUR CYCLEFig. 2 is a schematic representation of the sulphur

cycle3. Sulphur and its compounds are an essential partof the metabolic processes of the most important bacteriainvolved in microbial corrosion. Together with animalsand plants, these micro-organisms play an important rolein the progressive large-scale transformation of sulphurin nature.

The main reservoirs of sulphur and sulphur compoundsin nature are deposits of sulphidic minerals and elementalsulphur. Dissolved sulphates occur mainly in marine en-vironments. Hydrogen sulphide is produced in both soiland aquatic environments by the decay of sulphur-con-taining biological materials and by SRB. The hydrogensulphide is either precipitated as metal sulphide, or itenters the water, soil, or air, and may then be oxidizedspontaneously (chemically) or biologically.

Bacteria may oxidize sulphide or hydrogen sulphide tosulphur, and further microbial oxidation may yieldsulphate. To complete the full cycle, plants or bacteria,

JOU NAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY DECEMBER 1987 425

SOIl.~" ~~ -,>'~

'"\\\'?

(o,\,c-;;

""

d "\J~\\"~I /-/10 ASI

'JI\L "'\\\\

'/(°1""

.," .",Co"

I:CT/VE COJ\1\

Fig. 1. Diagram to Illustrate the principal methqds of microbialcorrosion of metals and protective coatlngs2

a. Tubercle (Inorganic + Iron bacteria) causing corrosion by dif-ferential aeration and providing environment for b

b. Anaerobic sulphate-reducing bacteriac. Sulphur-oxldlzlng bacteria, producing sulphates and sulphuric

acidd. Hydrocarbon utilizers, breaking down allphatlc and bitu-

minous coatings and allowing access of b to metal underneathe. Various microbes producing organic acids as end-products

of growth, attacking mainly non-ferrous metals and certaincoatings

f. Bacteria and moulds able to break down polymersg. Mainlyalgae, forming sllmes on above-ground damp surfacesh. Slime-forming moulds and bacteria (some of which may pro-

duce organic acids or utilize hydrocarbons), which provideconditions for the setting up of differential cells and for thegrowth of b

I. Mud on river bottoms, etc., providing matrix for heavy growthof microbes (Including oxygen-free conditions for b)

j. Sludge-Inorganic debris, scale, corrosion products, etc.,-providing matrix for heavy growth and setting up differentialaeration cells; and organic debris providing nutrients forgrowth

k. Debris (mainly organic) on metal above ground, providing con-ditions for growth of organic acid-producing microbes

or both, reduce the sulphate to the original sta~tingmaterial, viz sulphide. It is important to bear in mind thatthe bacterial corrosion processes involving steel are partof this general sulphur cycle. Corrosion occurs eitherbecause of attack initiated by bacterial end-products orduring the metabolism of the bacteria.

The oxidation of elemental sulphur to sulphates is ef-fected by a group of aerobic bacteria of the genus Thio-bacillus. These are also the main organisms concernedwith aerobic microbial corrosion, often with some pre-liminary assistance from a group of organisms capableof converting sulphides to sulphur via polythionates andthiosulphate.

The important organisms in anaerobic conditions arethe SRB of the genera Desu/fovibrio and Desu/foto-maculum. In the presence of reducing conditions, thesebacteria successfully reduce sulphate to sulphide.

The reactions shown in the bottom half of Fig. 2 arecarried out by the sulphur-oxidizing bacteria. The SRBate able to short-circuit the system by reducing SO~- toS2- directly.

I t is important to note that all sulphur species between

4?6 DECEMBER 1987 JOURNAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY

the oxidation numbers of - 2 (sulphide) and + 6 (sul-phate) are thermodynamically metastable, with the ex-ception of elemental sulphur.

SULPHATE-REDUCING BACTERIAThe SRB were discovered by Beijerinck4 in 1895. He

described a sulphate-reducing organism called Spirilliumdesu/furicans, which is now known as Desu/fovibriodesu/furicans. In 1904, Van Delden5 reported marine,salt-tolerant varieties and, in 1925, Elion6 described thethermophilic types. The two main types of SRB involvedin corrosion are the genus Desu/fovibrio and Desu/foto-maculum. Other species such as Desu/fobulbus, Desul-fococcus, Desu/fobacter, Desu/fonema, and Desu/fosar-cina are involved but to a lesser extent. A photomicro-graph of Desu/fovibrio is shown in Fig. 3.

Baars7 published an extensive study of these bacteria.In 1931, Stephenson and Stickland8 demonstrated astoichiometric uptake of gaseous hydrogen by non-growing bacteria supplied with a small quantity ofsulphate according to the following equation:

S~- + 4 H2 - S2- + 4 H2O.

The SRB are anaerobic heterotrophs9, Le. they arenon-phototrophic and derive their carbon and energyfrom organic nutrients. They are obligate anaerobes anduse sulphate, as an alternative to oxygen, as the terminalelectron acceptor, with the resultant production of sul-phide. However, they occasionally use sulphur to acceptthe electrons released by the oxidation of nutrientslO.

SRB can be found in an extremely wide range ofanaerobic environments including fresh and salt waters(sea water and brines with sodium chloride concentrationsof up to 30 per cent), at temperatures from 0 to 100°C,and at hydrostatic pressures up to between 700 and1000atm. 11.

The economic losses through the deterioration ofmaterials as a result of SRB are fairly severe. It is unfor-tunate that no study has looked at the direct cost of theiractivities as a percentage of the total cost of corrosionin anyone year. The discussion below will highlight someof the diverse environments in which SRB have beenfound to be activel2.

Pollution of Waters. Canals, harbours, estuaries, andstagnant waters at or near industrialized regions are proneto become anaerobic as the detritus of human activity in-creases their biological demand for oxygen. All suchwaters contain sulphate and, particularly in warm climatesor seasons, sulphate reduction ensues, and a smell ofhydrogen sulphide develops, with consequent nuisanceto those living in the region and damage to metal andpaintwork.

The effects of SRB per se in such situations can be ofconsiderable economic importance in terms of damageby hydrogen sulphide to paintwork, corrosion of shipsand metal installations, tarnishing of domestic im-plements and decorations, damage to health and tourism,and so on. In this regard, one thinks of the complaintson the Witwatersrand about the 'rotten egg' smellsemanating from the installations at Sasol 2 and 3.

Infection of Sands and Soils. Very often, sands and mudscontain black products that impart a dark colour to them.

assimilatory

sulphatereduction(plants & microorganisms)

mineralisation

processes

(spontaneous/

microbiological)

dissimilatory

sulphate reductio(e.g. Desulfovibrio

biological oxidation

with°2

or N032-

(e.g. Thiobacillus spp.)

anaerobic biologicalphotosynthetic oxidation(e.g. green and purple

sulphur bacteria)

Fig. 2-Schematlc representation of the sulphur cycle3

Fig. 3 Transmission electron micrograph of Desulfovibrio desul-

furicans (magnification 10 OOOX)

If this type of formation is exposed to air, it turns brownafter a certain period of time. In certain instances, thechange in colour is accompanied by an odour of hydrogensulphide. The iiUerp:n~tationis that black ferrous sulphide,formed in the sand during anaerobic conditions, becomesautoxidized to brown ferric oxides and sulphates on beingexposed to air. Such a colour change serves as a warningthat metal and stone installations in such environmentsare threatened by microbiological corrosion. This isreflected by the damage suffered by marble and stonestatues in Europe and Asia. Soils totally free of SRBoccur very rarely, but the SRB are inactive if the soilsare aerobic. Soils showing pronounced blackening orclays showing 'greying' are normally anaerobic and oftenheavily infected. These are potential initiators of metalliccorrosion.

Corrosion and Deterioration. The anaerobic corrosionof buried ferrous metals, enhanced by the activities ofSRB, is one of the best-known activities of this group ofbacteria. It is also worth mentioning that the sulphur-oxidizing bacteria are also able to corrode ferrous metals,but to a lesser extent.

JOU NAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY DECEMBER 1987 427

Involvement in Oil Technology. SRB are normal in-habitants of the watery phase of oil-bearing strata andshales!3, and they are also normally present in the injec-tion waters used in the secondary recovery of oil. Noquantitative estimates of their cost to the oil industry havebeen made, but in terms of the corrosion of machinery,the plugging of itljection systems, and the contaminationof drilling systems and products, the cost must be sub-stantial.

Paper Industry. Blackening of paper and pulp can occuras a result of the activity of SRB. The presence of SRBleads to disposal problems, highlighted by the exampleof a plant that disposed paper-pulp wastes into a river,which led to sulphide pollution unless about 4,5 t ofnitrate was added per day at the same time to control thepollution.

Gas Industry. Town gas in gas holders is always underlainby a layer of water. The growth of SRB in the gas-holderwater leads to the formation of hydrogen sulphide con-taminating the gas and causing serious corrosion of thecontainment vessel.

CORROSION MECHANISMSAlthough SRB are the mor~important micro-organisms

involved in corrosion, other micro-organisms from a widerange of genera and species are also associated with cor-rosion. However, it must be emphasized that microbio-logically induced corrosion does not involve any newform of corrosion process.

It is well recognized that the corrosion of metals in anaqueous environment is an electrochemical phenomenonin which part of the metal substrate is oxidized andtransferred into solution (anodic reaction). The balancingcathodic reaction is the simultaneous reduction of somecomponent in the corrosive environment to preserve theoverall neutrality of the electrolytic cell. The metal ionsin solution may subsequently be precipitated as insolubleproducts, which may be loose and bulky, or become firm-ly attached to the surface of the metal, to which they con-fer corrosion protection to a greater or lesser extent.

The mechanisms of microbial corrosion can be broadlydefined as those taking place in the presence of oxygen(aerobic) and those occurring in the absence of oxygen(anaerobic). In most situations, aerobic bacteria colonizea metal substrate first. As they grow, they create an idealanaerobic environment beneath them in which anaerobic

bacteria can survive and multiply. Therefore, it is pos-sible to have a bulk solution that is to all intents and pur-poses aerobic, but in which both aerobic and anaerobicbacteria are active at the same time.

Microbial Corrosion under Aerobic ConditionsMicrobial corrosion under aerobic conditions involves

the colonization of the material by microbial growths anddeposits, or the formation of tubercles. Organisms thatcolonize a surface may lead to the formation of anoxygen-concentration cell. As the organisms,grow, theyconsume oxygen until eventually the region below the col-ony becomes depleted of oxygen. However, the peripheryof the colony has a higher concentration of oxygen, anda differential oxygen concentration is thus set up. Thedifference in oxygen content leads to the formation ofan anode in the centre with the corresponding oxygenatedperiphery functioning as a cathode. Such a differentialaeration cell is illustrated in Fig. 4.

These conditions can be created through colonizationon the metal by microbial growth such as slime-formingbacteria, or through the production of tubercles by theiron bacteria.

Where tubercles are formed, the outer shell consists ofa:hard layer of magnetite (F~O4) overlying a mixture ofhydrated ferrous oxides and iron sulphides. The latter areproduced by the activity of sulphate-reducing bacteria.The outer skin of the tubercle in contact with the wateris covered by a thin layer of goethite (FeO.OH) mixedwith compounds that have been absorbed by the tuberclefrom the water, particularly silica, manganese, andorganic matter. Siderite (FeC°3) has also been iden-tified, as well as up to 5 per cent of the sulphur assulphide. According to Schaschl!4, the bacterial popula-tion of sulphate reducers is higher on the outer surfaceof the tubercle than in the interior. This is considered tobe due to the inhibitive properties of the sulphide. Hence,under these conditions it is suggested that sulphate reduc-tion takes place at or near the surface of the tubercle,and the sulphide produced diffuses back into the noduleto react with ferrous ions to precipitate more iron sul-phide. Fig. 5 shows a schematic cross-section through atypical corrosion tubercle.

It is well established that variations in the oxygen orion concentration on a metal surface can set up electro-lytic currents with resultant corrosion of many metalsls.Miller et al.16 measured potential differences of up to60 mY between insulated segmented aluminium surfaces

/Ub.,010

l

Solution

f:~.Fig. 4-Dlfferentl.1 .eretlon cell

428 DECEMBER 1987 JOURNAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY

Iron BacteriaT e iron bacteria are a miscellaneous group of bacteria

that re associated with the oxidation of ferrous to ferricions The end result of their metabolism is therefore thefor ation of ferric hydroxide as a coating round the cellsacc ding to the following equation:

NAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY

Fig. 5 Cross-section through atubercle

Loose sediment

00"'ln1n& ..""

coverd by a fungal mass (anodes) and uncovered seg-ment (cathodes). Metal goes into solution at the anodicareas of low oxygen concentration under the mass, andthe e ctrons combine with water and oxygen to formhydr xyl ions at the more oxygenated cathodic areas. Asecti of buried pipe in an anaerobic soil can thereforebe a dic to another section of the same pipeline thatis in n aerobic soil.

Sulp ur-oxidizing Bacteria .W en corrosion can be attributed to the formation of

inorg nic and/ or organic acids, it is nearly always theresul of the activity of sulphur-oxidizing bacteria (SOB)such as Thiobacillus thio-oxidans or the presence offung I growth.

S are found in environments with high concentra-tions of inorganic reduced sulphur compounds and lowwate activity, e.g. conditions existing in coal mines, acidsoils, and sulphur deposits. The corrosive effects of SOBcan e attributed mainly to their ability to producesulp ric acid from the oxidation of inorganic reducedsulp r compounds or elemental sulphur, and occurespe ally in environments characterized by low pH.

S comprise two groups: the colourless SOB, and thecolo red SOB. The first group oxidizes sulphur com-pou s aerobically with oxygen, 01" sometimes withnitra e as the terminal electron acceptor. The secondgrou are photosynthetic and oxidize su~phur compou~dsanae obically3. Thiobacillus ferro-oxldans can derivetheir energy for growth from the oxidation of reducedsulp ur compounds as well as from the oxidation of fer-rous 0 ferric iron. Oxidizing ferric ions are potent cor-rosiv agents. Acidophilic colourless SOB are found inenvi nments where there are high concentrations ofredu ed inorganic sulphur compounds and the amountof w ter is limited. The resultant metallic corrosion thatoccu s can be attributed to the fall off in pH as a resultof th production of sulphuric acid. Even in the absenceof s lphur, microbial corrosion can still occur. Kino etal.17 emonstrated rust removal from iron surfaces incult es of Thiobacillus ferro-oxidans, which recycles theferri ions required to initiate the corrosive attack.

Thin hard black shell

mainly magnetite (Fe304)

Layer of goethite (~FeO.OH)

. in conjunction with silicon,.: . manganese and aluminium.

Hydrated ferrous ".:

/oxides and.'sulphides ".".

4 FeCO] + O2 + 6 H2O - 4 Fe(OH)] + 4 CO2,

The coating and the slime masses associated with it setup oxygen-concentration cells that cause corrosio~ and,as previously mentioned, may set up an anaerobic en-vironment for SRB. Several of the iron bacteria have veryunusual morphologies, and this has led to confusionregarding the classification of these organisms. So~e ofthe iron bacteria can also cause sulphur transformations.Iron bacteria are aerobic but they can grow in environ-ments with less than 0,5 p. p. m. of oxygen, in which theycontribute heavily to slime formation. Some iron bacteriaare autotrophic, obtaining their energy directly from theoxidation of ferrous ions.

The genus Gallionella comprises bacteria in which theferric oxide stalks are twisted or straight bands resemblinga ribbon or a row of beads. The bacteria are curved rodsat the top of the stalks and are often called stalkedbacteria. This genus is often found in association withSRB, and has been implicated in the pitting of austeniticstainless-steel alloys in service. This genus, one of themost common of the iron bacteria, can concentratechlorides thereby forming ferric and manganic chlorides,which m~y lead to the stress-corrosion cracking (SCC)of stainless steels.

The genus Sphaerotilus can oxidize dissolved ferro~sions to insoluble ferric hydrate. In this genus, the bacterialcells are rod-shaped to oval, and contained in a visiblesheath that may be branched.

Microbial Corrosion under Anaerobic ConditionsUntil 1934, microbial corrosion had been associated

with the presence of oxygen, but the process of anaerobiccorrosion had been encountered several times prior to thatdate. However, it was in 1934 that Von Wolzogen Kilhrand Van der VlugeS confirmed the presence of sulphate-reducing bacteria and their relation to corrosion. Theyobserved the severe corrosion of spun grey cast-iron waterpipes in polder land (land reclaimed from the sea) to !henorth of Amsterdam. The polder lands were low-lYingand contained wet, clayey soil that was rich in organicmatter and had an almost neutral pH. Under these con-ditions, minimal corrosion would normally be expectedsince the cathodic corrosion reaction would becomepolarized by a layer of accumulated hydrogen and con-sequently the corros~on rate would decrease. However,the water pipes had to be replaced every few years as theyunderwent graphitization. From the presence of ironsulphide, which was found as an adherent corrosion prod-uct and also in the vicinity of the pipes, it was deduced

DECEMBER 1987 429

Tubercle Microfouling

or sediment

Macrofouling SRB or organic acid

Pitting of ferrous metals

"'-Iron sulphide

I

I~ \

Pitting of non-ferrous

metals Fig. 6-Summary of cor-

rosion damage caused bySRB1

I'film or dissolved H2S

\

high strength steels

Hydrogen cracking of Hydrogen blistering

of steel with

Graphitization of cast iron

and eventually confirmed that SRB were present.The types of corrosion damage that the SRB can cause

are summarized in Fig. 6. SRB require anaerobic reducingconditions with a source of energy, nutrients, and sul-phate ions for growth, as well as a near-neutral pH. Suchconditions are possible within the fouling layer, as shownin Fig. 7, the anaerobic micro-environments developingas a result of oxygen consumption by aerobic organismsif the conditions are favourable. This layer forms firston metal surfaces in any aqueous environment, and theSRB consequently find their way into the anaerobicregions of the layer and proliferate.

Sulphide reduced from sulphate reacts with availablehydrogen and iron to form hydrogen sulphide and fer-

OXYGENATED SEAWATER

cathodic protection

rous sulphide. The tying up of hydrogen in this fashionleaves an excess of hydroxyl ions, which produce acharacteristic alkaline environment. SRB can thrive inconditions with a pH range of 5 to 9,5.

In 1931, Stephenson and Stickland8 discovered thatthe SRB contained the enzyme hydrogenase. This dis-covery was eventually to lead to the formulation of thecathodic depolarization theory.

Cathodic Depolarization TheoryAs mentioned previously, in 1934 Von WolzogenKiihr

and Van der Vlugt proposed the cathodic depolarizationtheory to account for corrosion in anaerobic environ-ments. This theory is sometimes referred to as the classical

0 % 100 % Aerobic bacteria consume oxygen:

Produce carbon sources

"z........:I

8~

OXYGEN

TENSION IIIII...H2

(, Anaerobic....

-~-~' sources: SRB produce~

H2

bacteria use carbonS2-

sulphides-- ~.S2-

:;J

E-<

~

Fig. 7-Cross-section through a typical foul-

ing layer1

430 DECEMBER 1987 JOURNAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY

theor .Kti rand Vlugt proposed that the bacteria remove

atom' hydrogen that has formed on polarized cathodicareas of steel, and oxidize it to protons and electrons,The cteria then utilize the reducing power obtained toreduc sulphate to sulphide. They used the followingequat ons19 to describe this mechanism:Elect olytic dis-

soci tion ofwat r

Ano ic reactionCath dic reac-

tioCath dic depo-

lari ation bybac eria

Corr sionpro uct

Corr sionpro uct

Over 11reac-tio 4 Fe + S~- + 4 H2O - FeS + 3 Fe(OH)2 +

2 (OH)-.Fr m the final equation, there should be a 4: 1 ratio

betw en the total iron content and the iron as ironsulp ide in the corrosion products. One of the criticismsof t is theory is that this ratio, called the Ktihr ratio(Nk), can vary from between 0,9: 1 to nearly 50: I.

B nkero, in his examination of burst or leaking cast-iron ipes, reported sulphide-sulphur contents in the cor-rosi products of 6 to 9 per cent. Mara21 suggested thatvalu s of Nk greater than 4 would necessarily demon-strat the operation of another corrosion mechanismunre ated to the amount of sulphide reduced, Spruit andWa lyn22 claimed to have demonstrated the existenceof s ch a mechanism. They showed that, under hetero-trop ic conditions, the value of Nk varied between 0,9and ,5 while, under autotrophic conditions, the valuesof ranged from 5 to 9.

H rvath and Solti19 showed that cathodic depolariza-tion ccurred on mild-steel electrodes immersed in cul-ture of SRB. However, they did'11ot characterize theirparti ular organism fully, and therefore their work didnot ave much impact.

Both and Tiller supported the theory of VonW olzo-gen uhr and Van der Vlugt and concluded, after a seriesof e periments, that cathodic depolarization is a func-tion f the ability of an organism to utilize atomic hydro-gen. hey also showed that cathodic depolarization wasa fu ction of the activity of the bacteria.

Si ilar results were obtained by Booth and Tiller3usin two strains of thermophilic spore-forming sulphate-redu ing bacteria, Clostridium nigrificans (called Desul-foto aculum nigrificans). They found that the cathodiccurv s were influenced by the hydrogenase-containingorg ism only, and that both organisms caused anodicpola ization due to the production of partially protectivesulp ide films. Figs. 8 and 9 show the results obtainedwith a divided cell. Depolarization occurred when thecath de compartment was inoculated and the anode com-part ent was sterile (Fig. 8). Conversely, no depolari-zati n occurred when the anode compartment was in-ocul ted and the cathode compartment was sterile(Fig 9).

8 H2O - 80H- + 8H+4 Fe - 4 Fe2+ + 8 e-

8 H+ + 8 e- - 8 H

S~- + 8 H - S2- + 4 H2O

Fe2+ + S2- - FeS

3 Fe2+ + 6 OH- - 3 Fe(OH)2

<U....

'"<JCl)

-0,4Anodic Anodic-0,4

'"<U000...

'<JE -0,6 -0,6

>", Cathodic

....

'",.-<...,

'"<U~ -0,8

""

-0,8

0 100 1005050 0

Current Density ~A/cm2

Fig. 8-Anodic and cathodic polarization scans28

- Initially' " After 1 day - - - After 2 days

-0,2 -0,2

,-.Q)....

'"<JCl)

'"-0,4Q)

000...

'<J;>...c::

----Anodic /,.,."'" Anodic

/'~,-:-. -:-. ~ :-'-0,4,,' (- .'

Cathodic

> -0,6....

'",.-<...,

'"Q)...,0p... -0,8 Cathodic

0 100 0 1005050

Current Density ~A/cm-2

Fig. 9-Anodic and cathodic polarization scans28

- Initially' " After 1 day - - - After 2 days

Booth and Tiller4 concluded that their results lentfurther support to their view that the ability of anorganism to utilize hydrogen is the criterion for microbialcorrosion, and that the hydrogenase and sulphate-reductase systems need not be coupled.

However, later work by Booth et al.25, using nine dif-ferent strains of SRB maintained in a state of high ac-tivity by a semi-continuous culture technique, and withbacteria having varying hydrogenase activity, did not con-firm the direct relationship previously reported by Boothand Wormwelf6 between hydrogenase activity and cor-rosion. However, Miller and Kini7 have shown thatthese organisms often retain iron sulphide attached totheir cell walls. Hence, the cells could probably have beencontaminated, and sulphide could have been assisting inthe depolarizatien process.

Booth et al.28, using actively growing cultures of anumber of different strains of SRB, showed that the rateof corrosion was independent of enzymatic activity, andthat hydrogenase-negative organisms could also producehigh rates of corrosion.

Alternative Depolarization TheoryIn the early sixties, the effect of iron sulphide on

JO RNAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY DECEMBER 1987 431

microbial corrosion was beginning to be studied moreclosely. Booth et al.29 investigated the influence of fer-rous iron on the corrosion rate, using a semi-continuousculture technique. In this study, they found no direct rela-tionship between hydrogenase activity and corrosion rate.They suggested that the high concentration of soluble fer-rous salt in the medium may have influenced the rate ofcationic dissolution to become rate-determining.

Later work by Booth et al.3Oled to the conclusion'that the precipitate of ferrous sulphide associated withthe growth of sulphate-reducers in conventional mediumcontaining sulphates and ferrous iron and in natural con-ditions als031plays a part in cathodic depolarisation'.Meyer et al. observed that, when ferrous sulphide, freesulphur, and ferrous sulphide, or free sulphur were placedon steel coupons during corrosion by hydrogen sulphidegas, the rate of corrosion was slightly enhanced and thenumber of blisters increased.

Klas32 explained pronounced corrosion in pipes carry-ing brine containing iron sulphide as being due to the for-mation of sulphide deposits, which have a rather highelectropositive potential and thus act as excellent cathodicsurfaces. Herzog33 demonstrated that steel belowsulphide deposits behaves like a galvanic cell (ferroussulphide-iron), iron being the anode and ferrous sulphidethe cathode.

. Bo?th, Robb, and WakerlefB indicated that protec-tIve fllms were formed on metal surfaces low in ferrousiron while, at concentrations sufficiently high to precipi-tate all the bacterially produced sulphide as iron sulphideno protective films were formed and the rate of corro~sion increased markedly. Electrochemical studies show-ed that sustained and vigorous cathodic depolarizationoccurred even if the ferrous iron concentration was subse-quently reduced. Booth et al.3Oconcluded that cathodicdepolarization of steel by SRB could be brought aboutby utilization of the hydrogen by the hydrogenase systemof the organism, or by the interaction of the polarizedsurface with tile precipitated ferrous sulphide.

St~dies by King et al.34showed that the corrosivity ofchemically prepared iron sulphide was a function of thesulphur stoichiometry, but that stimulation of corrosionby ~ron.sulphid: decreased with time. High rates of cor-rosi~n m bactena-~ree systems could be obtained only bycont~nuous replemshment of the iron sulphide. Theirstudies showed that there was a linear relationship be-tween the amount ?f chemically prepared iron sulphideadded to the bactena-free system and the additional cur-rent required to maintain a cathode at a particular con-trolled potential. It was also noted that the efficiency ofdepolarization by iron sulphide alone decreased with timein the absence of bacteria. This was attributed to bindingof the atomic hydrogen within the sulphide lattice. Whenthe bacteria were added to the system, renewed depolari-zation took place, indicating that the bacteria utilized ordis~onded the hydrogen contained in the iron sulphidelattice. Thus, Miller and King provided an alternativemechanism to account for the high rates of corrosion thatare observed in the field (Fig. 10).

An ~portant feature associated with Miller and King'stheory is the nature of the sulphide film formed. Thec~emistry of the f?rmation of six stable iron sulphidem1Oerals from studies by Rickard35, and subsequently by

432 DECEMBER 1987 JOURNAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY

Lactate

BACTERIA

Fe2+H2O

OH-

Lactateand C8°42-

Fe2+

Fig. 10-Comparlson of the classical theory (top) with the alter-native, depolarization theory27

Mara and Williams36, confirms that the severity ofmicrobial corrosion can be explained by the nature of thesulphide films formed on the surface. There are sixnaturally occurring iron sulphide minerals as shown inFig. 11. '

Depolarization by Hydrogen SulphideHydrogen sulphide is known to be corrosive as regards

metallic materials. Costell037 investigated the corrosiveaction of SRB on ferrous metals. Using electrochemicaland biochemical theory, he critically evaluated thecathodic depolarizing theory and 'showed theoreticallythat, if depolarization occurred in that way, it was unlike-l~ to affect corrosion rates. His experimental results in-dicated that the cathodic depolarizing phenomenon wasdu~ to the cathodic activity of dissolved hydrogen sul-phide produced by the organisms. He demonstrated thisphenomenon experimentally by showing that growingcells of the genus Desu/jovibrio do lead to stimulationof iron cathodes exposed to them, but that this stimula-tion is retained in cell-free centrifugates of active cultures.Furthermore, the depolarization can be removed if aninert gas is bubbled through the bacterial solutions.Hence, cathodic depolarization is caused by some speciesproduced by the SRB rather than by the organismsthemselves.

Sulphur-concentration CellsSchascl14 proposed that the presence of elemental

sulphur in de-aerated solutions acts in the same manneras dissolved oxygen in aerated solutions in promoting the

PyriteFeSZ

Ftg. 1-N81ura11poccurrillg sulphkfe ...........

SideriteFeco3

02

~

02 Fe-+Fe2+ + 2~-

ArwdXc reaction

AnCMliic re-a.ctim:-

co ion of steel. The suJphur appears to promote 00£-r

. n by a concent:ration-cdl mechanism similar to thatd for differential aenmon cells. i..~ an anodic areadev ops beneath a porous material that slUelds the steelfro dissolved sulphur. the corresponding cathodic area.

the adjacent region where sulphur is readilyavail-able as a cathodic reactant (Fig. 12).

S' ce microbial corrosion is the result of a communityof ganisms. certam species of bacteria can produceele ntal sulphur as an intermediary during their meta-boli processes. Even SRB themselves can. under certaincon itions, produce sulphur. The sulphur so producedcan hen act as a cathodic depolarizer, as previously men-tion d. It would therefore appear that this theory ofmic bial corrosion is more important than was previous-ly r alized.

Ph :phorus-compound TheoryI erson38 conducted some experiments with SRB in

whi h a black precipitate was obtained. X-ray diffrac-tio of the non-magnetic material did not reveal thepres nce of iron or any iron-containing compound. Whenthe recipitate was heated in a vacuum and allowed tocoo to room temperature, X-ray-diffraction analysis in-

EqtmB'ac1:er£.a ~Cbemi.ca 1 rautu'Ef'--+E~ /foon

Mmzc;aorlt e-

h-Sz

0,5"

IronFe 2+ S2-

MackinawitB; .FeSl-X:

2. Eqpn-

52-Stnyt-h i t e'

Fe.3S4

F...~.,.. 8IIIWiI::cmncentratic81calf(-,8Id_IIII88nIItiicCDIIJI..., ,~..t (bAl ,.

~(JI::CJthat dae~II.~.M'" .. inJD~ IftISOn

~ that ~tr iD dtc ~ a:tnId.. beingrednc:ed iDthe ~ of ~1ch~ by the hydrogenaseof the lJesIII;fovibriD CJJPIIkrm: to ~ whichsubsequently reacted with ferrous ioos to produce ironphosphide.

Later work by lverson and 0150'" presented evidencethat the primary cause of anaerobic conosion is thepresence of a volatile. highly corrosive product contain-ing phosphorus. This is produced by the action of bac-terially produced hydrogen sulphide on inorganic phos-phorus compounds in the environment in which the SRBare growing. The authors also suggested that directbacterial formation of the compound may also occur.This theory is presented schematically in Fig. 13.

Hence, the initiating corrosion process for iron woulddepend on which metabolite reached the surface first. Ifthe phosphorus metabolite reached the surface first, cor-rosion would occur. If hydrogen sulphide reached the sur-face first, a protective film of iron sulphide would beformed, and corrosion would occur only once this filmbroke down. The author suggested that this phenomenonwould explain the erratic results that had been obtainedby various workers.

DECEMBER 1987 433JO RNAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY

I

IRONI

PIPE

..-FeS

f.- ~- ;',; -H:)-

'", - _,PO4)- o..! 0-

" c::>--~_H2S~~

H2S-- 0--/DESULFOVIBRIO

C~SH

+

CH)SSCH)

CORROSION

PRODUCTS + ACIDanoxi) H2sj + PH)f

fast slow

Fig. 13-Phosphorus compound theory39

Hence, it appears that corrosion by the SRB is a con-sequence of the production of corrosive metabolites(hydrogen sulphide or phosphide).

Corrosion by Other MechanismsIn addition to the mechanisms described previously,

the SRB have been implicated in a variety of corrosionproblems, such as hydrogen embrittlement.

In the oil industry, the occurrence of very deep drill-ing wells in sour-gas environments (which have highhydrogen sulphide activities) has led steelmakers to lookfor steels exhibiting high mechanical strengths and highresistance to sulphide stress cracking40.

Sources of hydrogen can be hydrogen gas (H2) if dis-sociatipn into atomic hydrogen occurs, or electrolytichydrogen if the recombination of atomic hydrogen toH2 is prevented (as by sulphides or other hydrogen-evolving poisons). Certain substances such as sulphideions, phosphorus-containing ions, and others are knownto be effective poisons of the conversion of atomic tomolecular hydrogen41. In the presence of such com-pounds, uncombined hydrogen atoms increase in numberon metal surfaces, and hence the probability that theywill be absorbed into the metal lattice also increases.

Thus, the production of hydrogen sulphide by SRB inanaerobic environments may stimulate the absorption ofatomic hydrogen in metals by preventing its recombina-tion into hydrogen gas.

RECENT FAILURES OF STAINLESS STEELSKobrin42 reported the pitting of AISI 3O4L and 316L

stainless steels used in storage vessels and piping of 3 mmwall thickness. Potable water (containing about 200 p.p.m.of chlorides) was used for the hydrostatic testing. About

a month after hydrostatic testing, water was seen to bedripping from the butt welds of the 3O4L piping. Analysisyielded high counts of bacteria.

TatnaW5 presented five case histories of microbio-logically-induced corrosion, three of which occurred onstainless-steel components. One case dealt with the crevicecorrosion of AISI 304 stainless steel in a recirculatingcooling-water system. The following mechanism was pro-posed by the author to explain the observed corrosion:(I) free-floating slime-formers become attached at low-

velocity sites, such as gasketed joints and crevices;(2) the growing slime mass entraps suspended solids from

the water, such as iron oxides and free-floating fila-mentous iron bacteria;

(3) an anaerobic environment forms below each depositas the slime-formers and iron bacteria consume theoxygen diffusing into the deposit;

(4) SRB find their way into the deposit and concentratein the anaerobic zone at the metal surface;

(5) the passivity of the stainless steel is broken down bya combination of the oxygen-free condition and thesulphide generated by the SRB.

Piluso, in the same paper by Tatnall, described the cor-rosion problems that occurred from a recovery processinvolving white water in the pulp-and-paper industry. Th/'pH value was 6 to 7, and the temperature varied between40 and 46°C. Corrosion occurred on a large AI SI 304impeller and screen located ahead of the paper machine.Corroded areas were covered by a layer of aerobic slimeunder which anaerobic bacteria were positively identified.

Pope et al.43 studied the microbiologically inducedcorrosion of heat-exchanger systems at the SavannahRiver nuclear facility. The tubing is made of AISI 304Lstainless steel..

Puckorius et al.44 reported the case of an unexpectedmassive failure of a utility power condenser from thewater side. The condenser contained over 28 000 AISI 304tubes. Investigation revealed that 3 to 5 per cent of thetubes were leaking. SRB were detected in the slimy depositsin the tubes.

Tiller46discussed two examples of microbial corrosionof stainless steel. The first failure occurred after eighteenmonths of service, and the second after two years.

Robinson and"'Ringas47reported two failures of 3CRI2corrosion-resisting steel. The failures were due to a com-bination of microbial corrosion and improper picklingand passivation of welds laid down on the alloy. ,

The mechanisms of corrosion discussed previously forcarbon steels apply equally to stainless steels. However,the stainless steels rely on a tightly adherent oxide layerfor their corrosion protection. Therefore, in cases wherethis layer is damaged or its formation impeded, corro-sion may result. The breakdown of this layer often leadsto pitting corrosion. Chloride ions are notorious for theirability to break down the passive layer and cause pittingof stainless steels. However, there are a number of othercompounds that can also break down the passive layerand lead to pitting corrosion. Sulphur species are especial-ly aggressive in this regard. The SRB and the SOB, ashas been discussed, produce a whole range of sulphurcompounds ranging from sulphide through elementalsulphur to sulphate.

It therefore appears that stainless steels are susceptible

434 DECEMBER 1987 JOURNAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY

to mic bial corrosion under certain conditions. The cor-rosion s due mainly to the sulphur species that are pro-duced y the bacteria. SRB are found in diverse environ-ments nd consequently can corrode a large number ofdiffere t kinds of equipment. The fact that AISI 304stainle s steel is particularly prone to microbial attack isdistur ing.

The ractice of replacing mild-steel piping that has fail-ed (ow g to microbial attack) with stainless steel requiresfurthe consideration since it may not be the most cost-effecti e solution.

EXPERIMENTAL RESULTSFig. 14 to 22 show a series of photo micrographs of

stainl So-and mild-steel samples that had been immersedfor 4 eeks in pure cultures of s:RB'8. A semi-continuouscultur technique was used, giving a retention time of 10days.

Fig 14 shows the surface of alloy 304L in sterilemedi . As can be seen, little surface damage is evident.The s ratches are due to the polishing carried out on thealloys prior to testing.

Wh n the alloy was immersed in the bacterial culture,the su face was colonized by the bacteria, as clearly shown

Fig. 5-304L after Immersion in bacterial culture medium48

in Fig. 15. Fig. 16 shows the formation of a pit on thesurface of 304L after it had been immersed in the bacterialculture.

The surface damage on alloy 316L is clearly shown inFig. 17. The attack is intergranular and initiates at thetriple points in the microstructure. The intergranular at-tack was observed in all cases. Fig. 18 shows intergranularattack on AISI Type 409 stainless steel.

Pitting attack on the 3CR12 alloy is shown in Fig. 19.Pits were initiated at Ti (C,N) inclusion sites. The corro-sion on all the alloys was of a dual nature, i.e. inter-granular and pitting.

A large corrosion tubercle on a mild-steel specimen isillustrated in Fig. 20, with the bacteria on the surfaceclearly visible. It is interesting to note that some of thebacteria have actually been incorporated in the corrosionproduct.

Fig. 21 shows a mild-steel surface after it had beencleaned. The outline of the tubercle is clearly demarcatedby the corrosion pattern produced. Fig. 22 shows theintergranular nature of the corrosion and its shiny ap-pearance, and is a close-up of Fig. 21. The shiny ap-pearance of mild-steel surfaces can actually be used asa diagnostic tool for the identification of microbial at-tack by SRB.

Fig. 16-PIt on surface of 304L48

Fig. 17-Surface damage on 316L48

JOU NAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY 435DECEMBER 1987

Fig. 18-lntergranular attack on surface of alloy 43()48

Fig. 19-Plttlng attack on 3CR1248

Fig. 20-Large tubercle on mild steel (note bacterla)48

Fig. 21-Surface of mild steel after cleanlng48

~ , fFig. 22--ctose-up view of corroded area beneath tubercle on mildsteel, the shiny appearance being due to the Intergranutar nature

of the attack48

~38 DECEMBER 1987 JOURNAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY

NAL OF THE SOUTH AFRICAN INSTITUTE OF MINING AND METALLURGY

CONCLUSIONSFro the various theories dealing with microbial cor-

rosion it appears that the primary causes of corrosionare th metabolic end-products of bacteria. These prod-ucts a e either acids or aggressive substances such ashydro en sulphide. Stainless steels are attacked by avariet of sulphur species, but the main species ishydro en sulphide. Stainless steels have been shown notto be mmune to microbial attack, and care should beexerci d in the use of these alloys in environments wheremicro es are active. A better understanding of the cor-rosio mechanisms will lead to the application of properpreve tative techniques.

ACKNOWLEDGEMENTSTh authors are grateful to the University of the Wit-

water and and Middelburg Steel & Alloys (Pty) Ltd forpermi sion to publish this paper.

REFERENCES1. SA DERS, P.F. Biological aspects of marine corrosion. Metals

So iety World, Dec. 1983.2. HI L, E.C. Microbial aspects of metallurgy. American Elsevier

P lishing Co., 1970., P., and KUENEN,1.G. Microbiology of sulphur-oxidizing

ha eria. Proceedings of Conference on Microbial Corrosion, Ted-dn gton, U.K., 8-10 March, 1983. London, The Metals Society,19 3.

4. B JERINCK,M.W. Zent. Bakt. Parasit., vol. 2, no. 1. 1895.5. V NDELDEN,A. Beitrag zur Kenntnis der Sulfatreduktion durch

B terien. Zent. Bakt. Parasit., vol. 2, no. 2. 1904.6. E ON, L. Zent. Bakt. Parasit., vol. 2, no. 63. 1925.7. B RS, 1.K. Over sulfaat reductie door bacterien. Doctoral disser-

ta on, Technical University of Delft, 1930.8. S PHENSON,M., and ST!CKLAND,L.H. Hydrogenase 11. The

re uction of sulphate to sulphide by molecular hydrogen. Biochem.J. vol. 25, no. 215. 1931.

MILTON,W.A. The sulphate reducing bacteria: Their physiologya consequent ecology. Proceedings of Conference on MicrobialC rrosion, Teddington, U.K., 8-10 March, 1983. London, TheM tals Society, 1983.

MILTON,W.A. Sulphate-reducing bacteria and offshore oil in-d stry. Trends in Biotechnology, vol. I, no. 2. 1983.

11. BELL,C.E. Producers Monthly, vol. 22, no. 7. 1958.12. P STGATE,1.R. Economic importance of the sulphur bacteria.

P il. Trans. R. Soc. Lond., vol. B 298. 1982.13. DINS, W.S., OLSON,G.l., McFETERS,G.A., and TuRBAK,S.C.

D ssimilatory bacterial sulfate reduction in Montana groundwaters.G omicrobiol. J., vol. 2. 1980.

14. S HASCL,E. Elemental sulphur as a corrodent in deaerated neutrala ueous environments. Mater. Per/., vol. 19, no. 7. 1980.

15. E ANS, U.R. The corrosion and oxidation of metals. London,E ward Amold, 1960.

LLER,R.N., HERRON,W.C., KRIGENS,A.G., CAMERON,1.L.,a d TERRY,B.M. Micro-organisms cause corrosion in aircraft fuelt ks. Mater. Protect., vol. 3, no. 9. 1964.

17. NO, K., SUGIMOTO,N., KURODA,K., and USAMI,S. Biochemicalr st removal by Thiobacillus ferrooxidans. European J. Appl. Bio-t hnol., vol. 13. 1981.

18. NWOLZOGENKOHRand VANDERVWGT, LS. The graphitiza-ti n of cast iron as an electrobiochemical process in anaerobic soils.

ater (The Hague), vol. 18, no. 16. 1934.RVATH,1., and SaLT!, M. Beitrag zum Mechanismus des anae-

r ben mikrobiologischen Korrosion der Metalle im Boden. Wersk.rros., vol. 10, no. 624. 1959.NKER,H.l. A review of the physiology and biochemistry of the

s Iphur bacteria. London, H.M. Stationary Office, 1936.RA, D.D. On the anaerobic corrosion of steel by microbes.

.D thesis, University of Dundee, 1970.22. RUIT,C.l., and WANKLYN,1.N. Iron sulphide ratios in corro-

s n by sulphate-reducing bacteria. Natun!, vol. 168, no. 151. 1951.

3.

9.

10.

16.

19.

20.

21.

23. TILLER, A.K., and BoOTH, G.H. Polarisation studies of mild steelin cultures of sulphate-reducing bacteria. Part 2: Thermophilicorganisms. Trans. Farad. Soc., vol. 58, no. 1. 1962.

24. BoOTH, G.H., and TrLLER, A.K. Polarisation studies of mild steelin cultures of sulphate-reducing bacteria. Part 3: Halophilic organ-isms. Trans. Farad. Soc., vol. 58,-no. 2. 1962.

25. BoOTH, G.H., SHINN, P.M., and WAKERLEY, D.S. The influenceof various strains of actively growing sulphate-reducing bacteriaon the anaerobic corrosion of mild steel. Congr. into Cor. MarineSalissures Cannes, Paris, CREO, 1964.

26. BooTH, G.H., and WORMWELL, F. Corrosion of mild steel bysulphate-reducing bacteria: Effect of different strains of organisms.Proceedings 1st International Congress on Metal Corrosion. Lon-don, Butterworth, 1961.

27. MILLER, 1.D.A., and KING, R.A. Microbial aspects of thedeterioration of materials. Lovelock, G.W., and Gilbert, R.l. (eds.).Academic, 1975.

28. BOOTH, G.H., ROBB, 1.N., and WAKERLEY, D.S. The influence offerrous ions on the anaerobic corrosion of mild steel by activelygrowing cultures of sulphate-reducing bacteria. 3rd Intermitiona/Congress on Metallic Corrosion. Moscow, 1967, vol. 2, sect. S.

29. BOOTH, G.H., COOPER, D.M., and WAKERLEY, D.S. Corrosion ofmild steel by actively growing cultures of sulphate-reducing bacteria:The influence of ferrous iron. Brit. Corr. J., vol. I, no. 11. 1960.

30. BooTH, G.H., ELFORD, L. andWAKERLEY, D.S. Corrosion of mildsteel by sulphate-reducing bacteria: An alternative mechani$m.Brit. Corr. J., voJ. 3. p. 242. 1968.

31. MEYER, F.H., RIGGS, O.C., McGLAssoN, and SUDBURY. Corro-sion products of mild steel in hydrogen sulphide environments. Cor-rosion, vol. 14, no. 2. 1958.

32. KLAS, H. Contribution to the solution of corrosion problems oc-curring in oil field piping carryjng iron sulphide bearing brine. SixthWorld Petroleum Congress (Preprint), Germany, 1963.

33. HERZOG, E. Developing steel to resist hydrogen sulphide. Ind.Eng. Chem., vol. 53, no. 9. 1961.

34. KING, R.A., MILLER, 1.D.A., and SMITH, 1.S. Corrosion of mUdsteel by iron sulphides. Brit. Corr. J., vol. 8. 1973.

35. RICKARD, D.T. Stockh. Contr. Geo/., vol 13. 1969. p. 49.36. MARA, D.D., and WILLIAMS, D.S.A. The mechanism of sulphide

corrosion by sulphate-reducing bacteria. Biodeterioration ofmaterials. Waiters, A.H., and Hueck-van-der-Plas, E.H. (eds.).Barking, Applied Science, 1972. vol. 2.

37. COSTELLO, 1.A. The mechanism of cathodic depolarization ex-hibited by sulphate-reducing bacteria during metallic corrosion pro-cesses. Ph.D thesis, University of Cape Town, 1975.

38. IVERSON, W.P. Nature, vol. 217, no. 5135. 1968.39. IVERSON, W.P., and OLSON, G.l. Anaerobic corrosion by

sulphate-reducing bacteria due to highly reactive volatile phosphoruscompound. Proceeding,s of Conference on Microbill/ Corrosion,Teddington, U.K., 8-10 March, 1983. London,The Metals Society,1983.

40. CHAVANE, A., HABASHI, M., PRESSOUYRE, G.M., and GALLAND,1. High strength steels with improved sulphide stress crackingresistance. Corrosion, vol. 43, no. 1. 1986.

41. WALCH, M., and MITCHELL, R. The role of microorganisms inhydrogen embrittIement of metals. Corrosion 83, Anaheim, Califor-nia, April 18-22, 1983. Paper no. 249.

42. KOBRIN, G. Corrosion by microbiological organisms in naturalwaters. Mater. Perf., vol. 15, no. 7. 1976.

43. POPE, D.l., SORACCO, RJ., and WILDE, E.W. Studies on biologic-ally induced corrosion in heat exch;mger systeIDli at .the SavannahRiver Plant, Aiken Sc. Mater. Per/., vol. 21, no. 7. 1982.

44. PUCKORlUS, P.R. Massive utility condenser failure caused bysulphide producing bacteria. Mater. perf., vol. 22, no. 12. 1983.

45. TATNALL, R.E. Fundamentals of BIC. Mater. Per/., vol.20, no.9. 1981.

46. TILLER, A.K. Is stainless steel susceptible to microbial corrosion?Proceedings of Conference on Microbial Corrosion, Teddington,U.K., 8-10 March, 1983. London, The Metals Society, 1983.

.47. ROBINSON, F.P.A., and RINGAS, C. Microbial corrosion and its ef-fect on 3CR12 steel. Chromium Review, no. 6. Mar. 1986.

48. RINGAS, C. Microbial corrosion of stainless steel. M.Sc. thesis,University of the Witwatersrand, 1986.

DECEMBER 1987 437