APPLIED MICROBIOLOGYVol. 12, No. 3, p. 197-200 May, 1964Copyright © 1964 American Society for Microbiology

Printed in U.S.A.

Selection of a Microbiological Corrosion System for StudyingEffects on Structural Aluminum Alloys

H. G. HEDRICK, C. E. MILLER, J. E. HALKIAS, AND J. E. HILDEBRAND

Applied Science and Engineering Laboratories, General Dynamics/Fort Worth, Fort Worth, Texas

Received for publication 18 November 1963

ABSTRACT

HEDRICK, H. G. (General Dynamics/Fort Worth, Fort Worth,Tex.), C. E. MILLER, J. E. HALKIAS, AND J. F. HILDEBRAND.Selection of a microbiological corrosion system for studyingeffects on structural aluminum alloys. Appl. Microbiol. 12:197-200. 1964.- Two laboratory methods, a metal-strip test anda tank test, were evaluated as microbiological corrosion systemsfor producing corroded test specimens on a structural aluminumalloy. The results show that corrosion of the test alloy occurredbest in the metal-strip test in a deionized water-fuel mediuminoculated with a mixture of microorganisms under aeratedconditions. The metal-strip test was more successful for produc-ing large numbers of corroded test specimens and proved moreeconomical than the tank-type test, since less structural materialis needed to obtain a specimen with sufficient corrosion areas, andsince the corrosion can more easily be restricted by maskants tocertain areas for specific test purposes.

Microorganism contamination of integral fuel tanks injet fuel aircraft has been recognized as a serious problemin the past few years. Mod-Maintenance programs haverevealed large deposits of slime or bacterial growth on thebottom and sides of fuel tanks. Widespread corrosion hasbeen detected beneath these deposits in varying degrees,from small isolated pits and areas of exfoliation to exten-sive corroded effects.To study the effects of microbiological corrosion on

structural materials used in aircraft fuel tanks, it is desir-able to produce such attack on a large number of speci-mens under laboratory conditions. Considerable informa-tion is available (Harris, 1962; Baumgartner, 1962; Rogers,1948; Starkey, 1956; Baudon, 1958) on microbial corrosionof metals other than aluminum. Recent reports (Churchill,1963; Ward, 1963) have attributed the corrosion of integralfuel tanks to the presence of microbiological contamination.However, little, if any, information is available on themethods for producing microbiological corrosion on struc-tural aluminum alloys under laboratory conditions.The purpose of this study was to select a microbiological

corrosion system capable of producing corrosion in thelaboratory which duplicates that found on naturally cor-roded integral fuel tank specimens.

MATERIALS AND METHODS

In developing the microbiological system for determin-ing the best system to produce corrosion under laboratory

conditions, two approaches were used. One utilized metal-strip type tests and the second employed tank type tests.

Metal-strip test. The materials for the metal-strip typetest included cultures of Pseudomonas aeruginosa GD/FWB-3, Desulfovibrio desulfuricans Prince, ASD, Cladosporiumresinae QMC 7998, and Aspergillus niger GD/FW F-1,which were used singly and in a mixture, and a naturalmicroflora from a contaminated fuel sample collected atRamey Air Force Base, Puerto Rico. This sample con-tained primarily Pseudomonas type microorganisms. Theinocula were prepared from laboratory stock culturesgrowing in a Bushnell and Haas (1941) fuel medium forat least 7 days. A 1-ml portion of each inoculum wasadded to each test jar. The inoculum used in the deion-ized water test setups was prepared in tubes of deionizedwater-fuel before use. Each inoculum was checked for vi-ability by the streak plate method. The test media usedwere (i) Bushnell-Haas fuel, (ii) deionized water-fuel, (iii)Bushnell-Haas-cystine (50 mg per liter)-fuel, and (iv) sea-water medium-fuel (Sisler, 1961). All of the media wereused at a 2:1 medium-fuel ratio. Test specimens (1 by 4by 0.125 in.) were prepared from aluminum alloy 7178-T651.

All specimens were coded, cleaned, weighed, and thensterilized with the media in 16-oz culture jars at 121 Cfor 15 min. The position of the specimens in the jars was

FIG. 1 Metal-strsp test setup for screening the microbiologicalcorrosion systems.

197

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

HEDRICK ET AL.

such that the system provided a water, fuel, and vapor

phase. The prepared jars were inoculated, and sterile JP-4fuel was added. Duplicate jars were incubated at 30 Cuinder both static and aerated conditions for 90 days. Theaeration was accomplished by passing air through a 0.45-,iMillipore filter fitted to a sterile manifold connected toeach jar with rubber tubing. The metal-strip test setupis shown in Fig. 1.

Tank-type test. Two simulated integral fuel tanks (12 by6 by 1.5 in.), were constructed of aluminum alloy 7178-T651 and faced with Plexiglass so that they could be sealed.The tanks were cleaned and aseptically sterilized with70 % ethyl alcohol, and sterile Bushnell-Haas fuel mediumwas added to one, and deionized water-fuel medium tothe other. Both tanks were inoculated with the naturalcontaminated sample from Ramey Air Force Base andincubated at 30 C in a position to provide the greatestexposed surface area.

The metal-strip and tank tests were observed weeklyfor visual corrosion and growth. The media levels in allsetups were maintained by addition of sterile fuel and de-ionized water. The specimens were removed and examinedafter 45- and 90-day exposure periods. Attack depths were

measured by use of an Ace Optical Micrometer, and thetypes of corrosion were determined by metallographic ex-

amination of sections through the major corroded areas.

RESULTS AND DISCUSSIONObservations made during a 90-day exposure period re-

vealed the following. (i) The amount of visible culturegrowth had no apparent correlation to the amount of cor-

rosion noted on the metal-strip specimens. (ii) Corrosionof the metal-strip specimens was more extensive in theinoculated deionized water-fuel medium, followed in de-

scending order by the seawater medium-fuel and the twobasic salts-fuel media (which were equally poor as littleor no corrosion occurred in either one of them). (iii)Heaviest corrosion per specimen was caused by the arti-ficially mixed culture and the natural (Ramey) culture.(iv) More extensive corrosion was caused by bacteria thanby fungi. (v) Corrosion was more extensive under theaerated than static conditions. In the noninoculated con-

trols, corrosion was evident as a slight surface attack inthe seawater medium-fuel setups.Weight measurements indicated that the per cent weight

loss of the specimens varied from 0.015 to 0.574. The finalpH values of the water phase in the inoculated and con-

trol jars ranged from 5.0 to 7.5 and 6.0 to 6.5, respectively.Observations of the corrosion progress revealed little

corrosion worthy of depth measurements on most of themetal-strip specimens removed from the two basic salts-fuel media. In the deionized water-fuel medium, more ex-

tensive corrosion in general was observed than in the othermedia. Depth measurements on specimens exposed in de-ionized water-fuel with the various inocula are given inTable 1. In most cases, the deepest corrosion occurred inthe water phase or at the water-fuel interface. Corrosionwas also more extensive in aerated bacterial and mixedcultures.

Metallographic examinations of typical corrosion areas

on specimens taken from the inoculated deionized water-fuel and seawater medium-fuel setups after 45 days of ex-

posure showed pits and blisters at the water-fuel interfaceof the specimen (Fig. 2 to 7). Figures 8 to 13 show theprogress of corrosion after the 90-day exposure period.Figures 8 and 9, and 12 and 13, show a comparison betweencontrol and inoculated specimens. The presence of the mi-

TABLE 1. Depth measurements on 7178-T651 alloy exposed 90 days in deionized water-fuel medium

Maximal depth of corrosion pits (in.)

Specimen no. Culture Incubation conditionWater phase Water-fuel Fuel phase Fuel-vapor Vapor phase

38 Natural Ramey sample Static 0.0097 0.0115 None None None56 Natural Ramey sample Aerated 0.0097 0.0110 0.0060 0.0010 None40 Pseudomonas aeruginosa Static 0.0115 0.0066 None None None58 P. aeruginosa Aerated 0.0104 0.0036 None None None42 Desulfovibrio desulfuricans Static 0.0095 0.0100 None None None60 D. desulfuricans Aerated 0.0099 0.0109 0.0100 0.0100 0.005044 Cladosporium resinae Static 0.0049 0.0010 None None None62 C. resinae Aerated 0.0048 None None None None46 Aspergillus niger Static 0.0085 0.0087 None None None64 A. niger Aerated 0.0010 0.0010 None None None48 Mixed bacterial sample Static 0.0092 None None None None66 Mixed bacterial sample Aerated 0.0113 0.0061 0.0063 None None50 Mixed fungal sample Static 0.0084 0.0055 None None None68 Mixed fungal sample Aerated 0.0063 0.0063 None None None52 Mixed bacterial and fungal sample Static 0.0110 0.0092 None None None70 Mixed bacterial and fungal sample Aerated 0.0122 0.0115 0.0050 0.0040 0.003054 None, control Static None None None None None72 None, control Aerated 0.0022 None None None None

198 APPL. MICROBIOL.

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

MICROBIOLOGICAL CORROSION SYSTEM19

2,, 'r.',-, - ,, ^ - . ; ; .,,,.<;,>.,~;4.

.

3 --

_4.'z-.-* .. :..i

, 9;.;-* - u .

6.- *:.

1.2.s_t* 7i _ sIt

70 ~;

FIG. 2 Pittingandintergranular corrosion.on 7178-T651 aluminumalloy by mixed bacterial and fungal inoculum in static deionizedwater-fuel medium. Magnification, 1OOX.

FIG. 3 Blister, exfoliation, and intergranular corrosion in 7178-T661 aluminum alloy by mixed bacterial and fungal inoculum instatic deionized water-fuel medium. Magnification, 1OOX.

FIG. 4 Blister, exfoliation, and intergranular corrosion on 7178-T661 aluminum alloy by mixed bacteriat and fungal inoculum.Magnification, IOOX.

FIG. 5 Blister, pitting, and intergranular corrosion on 7178-T651aluminum alloy by Desulfovibrio desulfuricans in aerated deionizedwater-fuel medium. Magnification, 1OOX.

FIG. 6 Pitting corrosion on 7178-T651 aluminum alloy by Desulfo-vibrio desulfuricans in aerated seawater medium-fuel medium.Magnification, 1OOX.

FIG. 7 Blister, pitting, intergranular, and grain fragmentationcorrosion on 7178-T651 aluminum alloy by Desulfovibrio desul-furicans in aerated seawater medium-fuel medium. Magnification,boox.

FIG. 8 Surface corrosion on 7178-T651 aluminuim in aerateddeionized water-fuel medium in the control. Magnification, 1OOX.

FIG. 9 Surface, exfoliation, and intergranular corrosion on 7178-T651 aluminum alloy in aerated deionized water-fuel medium inocu-lated with a mixture of bacterial and fungal sludge.

FIG. 10 Surface and intergranular corrosion on 7178-T651 alumi-num alloy by Desulfovibrio desulfuricans in aerated deionized water-fuel medium. Magnification, 10OX.

FIG. 11 Surface, exfoliation, and intergranular corrosion on 7178-T661 aluminum alloy in aerated deionized water-fuel medium by anatural contaminated Ramey fuel sample. Magnification, 10OX.

FIG. 12 Surface corrosion on 7178-T661 aluminum alloy in aerateddeionized water-fuel medium in the control. Magnification, 10OX.

FIG. 13 Surface corrosion on 7178-T651 aluminum alloy by arti-ficial mixed culture in aerated deionized wnater-fuel. Magnification,IOox.

199VOL. 12. 1964

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

HEDRICK ET AL.

croorganisms apparently enhanced the corrosion process,resulting in a greater degree of attack on the specimens.

In the tank tests, the combinations of media and cul-tures failed to produce corrosion during the 90-day incu-bation period on any of the metal surface contact areas.Microbial growth was extensive, and the condition of in-cubation was similar to that in the metal-strip tests. Itwas concluded that the tank type test would not be feas-ible for producing large numbers of test specimens by lab-oratory methods.These results indicate that microorganisms are able to

bring about corrosion of an aluminum alloy under labora-tory conditions. This corrosion varies from surface, pitting,and blistering to intergranular attack and exfoliation.These results also indicate that microorganisms cause amore extensive corrosion of a structural aluminum alloyin a deionized water-fuel environment than occurs in thecontrols in which the microorganisms were omitted. Thecorrosion was also more extensive in the water phase orat the water-fuel interface than at other areas in the sys-tem.

ACKNOWLEDGMENT

This study was done as part of USAF contract numberAF 33(657)-8752 on "Microbiological Corrosive Effects on

Structural Materials Used in Aircraft Fuel Tanks" withC. B. Ward as project monitor. Aid and suggestions werealso contributed by the project leader, D. C. Wilson, andothers.

LITERATURE CITED

BAUDON, L. 1958. Le r6le des micro-organismes dans certainsph6nomenes de corrosion. Ind. Chim. Belge 23:983-990.

BAUMGARTNER, A. W. 1962. Microbiological corrosion-whatcauses it and how it can be controlled. J. Petrol. Technol.13:1074-1078.

BUSHNELL, L. D., AND H. F. HAAS. 1941. The utilization of certainhydrocarbons by microorganisms. J. Bacteriol. 41:653-673.

CHURCHILL, A. V. 1963. Microbial fuel tank corrosion: mechanismsand contributing factors. Materials Protection 2:19-23.

HARRIS, J. 0. 1962. The role of soil microorganisms in corrosion.Proc. 7th. Ann. Appalachian Underground Corrosion, ShortCourse. W. Va. Univ. Eng. Expt. Sta. Tech. Bull. 66:273-282.

ROGERS, T. H. 1948. The promotion and acceleration of metalliccorrosion by microorganisms. J. Inst. Metals 75:19-38.

SISLER, F. D. 1961. Electrical energy from biochemical fuel cells.New Scientist 12:110-111.

STARKEY, R. L. 1956. Microorganisms and metal corrosion in metalmetabolism and microbiological deterioration. Prevention ofDeterioration Center. Natl. Acad. Sci.-Natl. Res. CouncilPubl. 514, p. 21-22.

WARD, C. B. 1963. Corrosion resulting from microbial fuel tankcontamination. Mater. Prot. 2:10-16.

200 APPL. MICROBIOL.

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from