AD-A095 b42 DEFENCE RESEARCH ESTABLISHMENT ATLANTIC DARTMOUTH (NO--ETC F/6 13/11

CORROSION OF BUTTERFLY VALVES IN SEA WATER SERVICE.IU)

SEP 80 D R LENARD, L C MACLEODUNCLASSIFIED DREA-TM-I/H NLuEEElhllllEE

I lllflllll.f f

04-1

REEAC AN EEOMETBAC

DIPCORROSION OAIOA BUTERFY VLVECNSEAWTEASRVC

81) :2

DEFENCE RESEARCH ESTAILUSHMEN1T ATLANIMC CENTE DE RECHERCHE$ FOUR UA ODFnME ATLANTIQUE9 GROVE STREET P 0 Box 1019 9 GROVE STREET C P t012

DARTMOUTH. N.S TELEPHONE DARTMOUTH. N E82 y 3Z7 (9021 426-3100 82v 3Z7

RESEARCH AND DEVELOPMENT BRANCHDEPARTMENT OF NATIONAL DEFENCE

CANADA

p.rI.

Ir

DfffflCf RfSfflRC~ fSIfl~1IS~~fflI flT[~flTIC

4

DEFENCE RESEARCH ESTABLISHMENT

ATLANTIC

DARTMOUTH N.S.

D.R.E.A. TECHNICAL MEMORANDUM 80/H

CORROSION OF BUTTERFLY VALVESIN SEA WATER SERVICE

D. R. Lenard L. C. MacLeod

September 1980

Approved by J. R. Brown A/ Director / Technology Division

DISTRIBUTION APPROVED BY

CHIEF D.R.E.A.

RESEARCH AND DEVELOPMENT BRANCHDEPARTMENT OF NATIONAL DEFENCE

CANADA

r

ABSTRACT

- ">Electrochemical techniques were used to compare theccorrosion resistance in sea water of butterfly valve compon-ents made from an alloy which did not meet compositionalrequirements to components made from the specified Monel alloy400. Polarization resistance studies indicated that the alloywhich did not meet the specification had corrosion ratesseveral times higher than Monel in both aerated and de-aeratedsea water. Anodic polarization studies also showed that thisalloy had a much higher probability for propagation oflocalized corrosion than Monel. Thus components made fromthis alloy can be expected to provide poorer service thanthose made of Monel.

Np-K

RESUME

L'utiiisation de techniques 6iectrochimiques a perrrisde comparer, pour ce qui est de la resistance A la corrosiondans i'eau de met, des 616ments de robinets A papilionfabriquC-s en alliage non conforme aux exigences de compositionet des i.ments fait de i'alliage prescrit monel. 400. Desitudes de resistance de polarisation ont montrg que i'ailiagenon conforme aux sp~cifications pr~sentait des taux de corrosionplusieurs fois sup~rieurs au monel Iorsqu'ii 6tait mis enIr~sence d'eau de nier a~r~e et dea~r~e. Des 6tudes depokai.3sation anodique ont 6galementr v~.6 pour cet alaeutitl probabiliLi! de propagation de la corrosion iocalisi3ebeaticouip plus e levee que pour le monel. Par consequent, onpenlt ' attendre quje les 616ments fabriqu~s i partir de ceta1Iiage donneront tin rendement inf~rieur aux 616ments de mnoiuI.

TABLE OF CONTENTS

Page No.

ABSTRACT i.

NOTATION V

1, INTRODUCTION 1

2, BACKGROUND 1

3, EXPERIMENTAL PROCEDURE 4

3..1 Materials 4

3.2 Apparatus 4

3,3 Calibration 53..4 Procedure 5

4. RESULTS 6

4.1 Polarization Resistance Measurements 64.2 Anodic Polarization Curves 7

5. CONCLUSIONS 8

ACKNOWLEDGEMENT 9

TABLES 10

FIGURES 13

REFERENCES 20

DOCUMENT CONTROL DATA 21

iv

7!777'--

NOTATION

Ir corrosion potential

primary passive potent ial

critical pit potential

polarization current

ba anodic Tafel slope

bc cathodic Tafel slope

R polarization resistance

A4 difference between applied potential andcorrosion potential

i instantaneous corrosion rate

SCE saturated calomel electrode

V

i, INTRODUCTION

Butterfly valves which were used in a sea waterapplication as part of a ship service piping system on severalHMC Ships have recently shown instances of failure due tosevere localized corrosion at the interface between the discand shaft areas (Figure 1). A chemical analysis of thesevalve components, which should have met the chemical require-ments of the standard specification I for Monel alloy 400(ASTM B164), indicated a composition which did not conform tothis specification or to that of any other known commercialalloy. Table I compares the chemical compositions of thespecified material and the valve alloy.

Nickel and its alloys, such as Monel, have excellentresistance to corrosion in aerated, flowing sea water. Thisresistance results from the formation of a passive film onthe alloy surface which imparts to the alloy the electrochemicalbehaviour of an appreciably more noble metal (that is, onewith a more positive corrosion potential and a lower corrosionrate). The nature of this passive film is not yet fullyunderstood. It has been described either as a protective,self-repairing film of oxide or hydrated oxide2 or as achemisorbed film of oxygen 3 . In circumstances where there isinsufficient oxygen available to maintain the film, such asstagnant sea water trapped in the crevice between the discassembly and the Buna N rubber seat of the butterfly valves,pitting may develop.

A study was initiated in order to establish therelative corrosion susceptibilities of Monel and the valvealloy in a sea water environment. The response of the alloysto both aerated and de-aerated sea water was studied inorder to simulate the operating conditions of the valve discand shaft areas. The study was part of a larger investigationof localized corrosion damage experienced by Monel buttcrl[yvailves in various aqueous environments on IIMC Ships.

2. BACKGROUND

Electrochemical methods have proven to be valuablein determining the susceptibility of alloys to crevicecorrosion 4 ,5. At the corrosion potential (Er), the rates ofanodic and cathodic processes on a specimen surface are equal.

However, if the potential of the specimen is forced away fromits corrosion potential, the partial processes are no longer

equal and an external current flows. A plot of the resultingcurrent versus the applied potential produces a polarizationcurve.

If the potential of the specimen is forced away fromthe corrosion potential in the cathodic (negative) direction,the anodic areas on its surface are extinguished and thespecimen behaves essentially as a cathode. In sea water theprincipal cathodic reactions are oxygen reduction and hydro-gen evolution, with the former being predominant as long assufficient oxygen is available. The polarization resistancetechnique, which will be discussed in more detail later,provides information concerning the cathodic reaction involvedin a corrosion process.

If the potential of the specimen is forced away fromthe corrosion potential in the anodic (positive) direction,the sample behaves essentially as an anode, the primary anodicprocess being metal dissolution. The potentiodynamic anodicpolarization technique, in which the potential is increasedabove the corrosion potential at a constant rate, provides amethod for evaluating the effects of various environments onthe rate of metal dissolution.

A typical anodic polarization curve for an alloywhich can be repassivated is shown in Figure 2. Above thecorrosion potential, the curve exhibits an increase in currentwith increasing potential until the passive film is formed.'rie potential at which the current ceases to increase iscalled the primary passive potential (EpD). As the passivefilm develops, it protects the alloy surface from furthercorrosion. Thus the current decreases wjth increasingpotential until the film covers the the entire surface. Oncethe film is completely formed there is no substantial changein current until the potential reaches a sufficiently noblevalue that the passive film begins to break down and localizedco,-rosion is initiated. The potential at which localizedcorrosion is first observed is called the critical pitpotential (Ep). The region above this potential is called thetranspassive region and is characterized by an exponentialincrease in current with increasing potential. The proximityof the pit potential to the corrosion potential provides anindication of the susceptibility of the alloy to initiationof localized corrosion.

If the direction of the potential scan is reversedin the transpassive region, hysteresis is observed in the

2

back scan. This hysteresis occurs because the specimensurface has been irreversibly altered by the localizedcorrosion. The value of the current at which the potential isreversed is chosen to be sufficiently large to ensureappreciable hysteresis yet must also be low enough to avoid

- additional polarization effects that can occur with highcurrent densities. Furthermore, it is very important that thesame current value be used for each spec imen in an exper iment.As the reverse scan is continued, a potential will be attainedat which all localized corrosion is extinguished and thecurrent returns to a passive value. This potential is knownas the repassivation potential (Erp). The degree of hysteresisprovides qualitative information concerning the probabilitythat existing localized corrosion will propagate.

Another technique, the polarization resistancetechnique, provides numerical values for important kineticparameters of corrosion reactijns. In this technique, aspecimen is polarized over a narrow potential range in theorder of ±30 mv from the corrosion potential. Specimensurface changes which may result from high current densitiesare consequently eliminated.

This method, recently discussed in detail by Mansfeld ,

is based on the derivation of Equation (1) from mixed-potcntialt he o rv

b a c 1 x2. exp 3A2.3 babc R epI bj -

where:

I polarization current

b a b C anodic and cathodic Tafel slopesa c

R reciprocal of the slope of thepolarization curve at the corrosion

potential (polarization resistance)

A d if f erenl, between the appl iedpotential and the corrosionpotent ial

3

A manual or computer fit of polarization data to equationsderived from Equation (I) yields the polarization resistance(R ) and the anodic and cathodic Tafel slopes. The Tatel

pslopes ba anu b c are dependent on the mechanisms of the anodicand cathodic reactions, respectively. The instantaneouscorrosion rate (i cor) can then be determined from Equation 2:

bbba bc i

corr 2.303 (b +b) R

...(2)

3. EXPERIMENTAL PROCEDURE

3.1 Materials

Cylindrical specimens were sectioned from shafts ofthe appropriate valves to a length of I inch and machined toa diameter of ' inch. One end was drilled and tapped for3-48 N.C. threads. The specimen surface was gound through aseries of silicon carbide abrasive papers from 120 to 400 grit.Prior to immersion in the electrolyte, the surface was givena final 600 grit polish to remove any oxide films, degreasedin isopropanol and rinsed with distilled water.

The electrolyte used was natural sea water at roomtemperature. Aerated or de-aerated conditions were ensuredby passing either compressed air or nitrogen through theelectrolyte continuously, commencing one hour before immersingthe specimen.

The corroded butterfly valve components examined bythe Dockyard Laboratory exhibited extensive porosity in thecastings (Figure 3). The cavities resulting from this porositywould tend to enhance crevice corrosion susceptibility abovethat for a properly cast component of the same -.llov. IHowever,all specimens, whether made of Monel or of the tnusual alloy,wer, machined from valve components obtained from I1MC Ship';.Thus this study provided a direct comparison of the elcoftro-chemical properties of the components found ii" service.

1.2 Apparatus

A schematic diagram of the circuit used for thepolarization experiments is shown in Figure 4. The system

4

utilized the following apparatus:

Wenking potentiostat (Model 70HC3)Wenking scanning potentiometer (Model SMP 69)

Standard polarization cell 7 and Luggin probe

salt bridgeFisher calomel reference electrodeHewlett-Packard - XY recorder (MoJel 7004B)

Programmable calculator (Model 9810A)Plotter (Model 9862A)

Decade resistance box

33 Calibration

The Venking potentiostat had a feature which permittedthe recording of a wide range of current values, for full anodicscans, without the inconvenience of changing the recorder rangeswitch. A pair of diodes in parallel with the meter resistancewould conduct a current after a certain potential drop acrossthe meter was exceeded. This resulted in a logarithmic outputsignal above this potential value as part of the total currentwas shunted across the diodes. Thus it was necessary todetermine the relationship between the current flowing ir thecell and the corresponding output voltage to the X-Y recorderin order to calibrate the pen response.

The calibration was performed with the circuit shownin Figure 5. The potentiostat sensitivity switch was se. tothe 0 - 100 pamp range. A constant current flow through theexternal circuit was regulated by adjustments of the counterelectrode potentiometer control and/or the decade resistancebox during which the corresponding output potential wasmeasured.

The calibration curve had three distinct regions -linear, parabolic (transitional) and logarithmic. Curvefitting routines obtained the appropriate coefficients whichdefined the three regions. A program was written for theHewlett-Packard calculator-plotter combination which wouldoutput the calibration curve of "Output Potential vs Current"from an input of the coefficients defining the three regions.The calibration curve generated by this program fnr theWenking 70HC3 potentiostat is shown in Figure 6.

1.4 Procedure

ilefore the specimen was mounted and immcrsed in the

,lcctrolyte, the exposed spec imen area was determined. A 2 4

........I ......................................

hour period was allowed for the specimen to reach its steady-state corrosion potential in the electrolyte before polariza-tion was performed. The tip of the Luggin probe salt bridgewas placed I - 2 mm from the surface of the specimen. It wasfound convenient to obtain the polarization resistance data(within +30 mv of the corrosion potential) before proceedingwith the destructive full anodic scan. External potentialswere applied by the scanning potentiometer at a constant rateof 5 mv/min in 5 mv increments for both methods. In the fullscan, the direction of the potential steps was reversed whenthe current reached a value of 100 microamps/cm 2 .

Initially a condition of zero current ±low wascreated by matching the counter electrode potential to thespecimen corrosion potential. The specimen was then polarizedby scanning to 30 mv away from the corrosion potential in thecathodic direction. The specimen was then allowed to returnto its original corrosion potential before being polarized to30 mv away from Er in the anodic direction. Several of theseexperiments were performed over a 24 hour period for eachspecimen before full anodic polarization scans were made.

The polarization curves were analyzed by the methodof linear least squares using a computer program initiallydescribed by Mansfeld 8 . The output from this program providedthe polarization resistance, anodic and cathodic Tafel slopes,the instantaneous corrosion current and several statisticalparameters.

4, RESULTS

4.1 Polarization Resistance Measurements

A typical polarization curve obtained within f30 mvof the corrosion potential is shown in Figure 7. The kineticparameters obtained from the polarization resistance experi-ments in each of the metal-electrolyto systems are giv veU infable I I . The reporL(,d values vre average resut] s obta inedover a 24 hour period following tho attainment of steady-s-tatLaconditions. The least squares analysis using a DECsvstem-20computer invol.ved no less than 25 experimental current-potential pairs for each curve and fits to the data wereobtained with standard deviations ranging from 2 percent to7 percent.

If the only electrochemical reactions occurring in asystem are metal dissolution and the corresponding reductionof a single oxidizing species, then the Tafel slopes depend

6

on the rate determining step in the reaction mechanism. Insea water, oxygen reduction and hydrogen evolution are themost common cathodic reactions. The infinite cathodic Tafelslopes observed in aerated sea water for both alloys indicatethat the dominant cathodic reaction is oxygen reductionoccurring under diffusion control 9 . In de-aerated sea water,the cathodic Tafel slope and corrosion potential determinedfor Monel alloy 400 are consistent with hydrogen evolution asthe cathodic reaction1 0 . The corrosion potential of the valvealloy in de-aerated sea water was too noble for hydrogenevolution and no oxygen was present so the cathodic reactioncould not be determined.

Even without knowledge of the reaction mechanism, theTafel slopes can be used to determine the instantaneo,,corrosion rate. In both aerated and de-aerated sea water, thcaverage corrosion rate of the valve alloy was found to beseveral times that of Monel alloy 400 in the same environment

(Table II). Thus the valve alloy was found to be inferiorto Monel by the polarization resistance method.

4.2 Anodic Polarization Curves

The anodic polarization curves used to determinecrevice corrosion susceptibilities for Monel and the valvealloy in de-aerated sea water are shown in Figures 8 and 9.The critical potentials extracted from these curves aresummarized in Table III.

The Monel specimen displayed typical behaviout foran alloy with an active-passive transition (Figure 8). Fromthe active corrosion potential of -333 millivolts versus thesaturated calomel electrode (SCE) the dissolution rateincreased with increasing applied potential. Above the

primary passive potential of -2W millivolts (SCE) thedissolution rate decreased and then remained independent ofpotential for 100 millivolts. Above -60 millivolts (thepitting potential), the current increased dramatically,accompanied by the initial appearance of pits on the specimensurface. Finally, the repassivation potential, below whichno localized corrosion can occur, was found to be -130 milli-volts (SCE).

The valve alloy had an initial corrosion potential(-103 millivolts vs. SCE) that was 230 millivolts more noblethan that of Monel. Furthermore, it did not display theS-shaped anodic scan typical of an active-passive metal

7

(Figure 9). There was, however, an inflection in the anodicscan at -45 millivolts (SCE) which corresponded to theappearance of pits on the specimen surface. The repassivationpotential of -240 millivolts (SCE) was much lower than thatof Monel.

The corrosion petential of Monel in aerated seawater (-75 millivolts vs. SCE) was within the passive regionshown in Figure 8 but was only 15 millivolts below the pittingpotential. Thus Monel should have a marked tendency for theinitiation of localized corrosion under conditions where thepassive film cannot be maintained. This tendency is wellknown and has been demonstrated in this laboratory (Figure 10).However, the repassivation potential is well above the activecorrosion potential (203 millivolts more noble) and only /0millivolts below the passive corrosion potential, indicatingthat Monel can be readily repassivated.

The corrosion potential of the valve alloy in aeratedsea water was only 10 millivolts above that in de-aeratedwater and 49 millivolts below the pitting potential. Thusthis alloy should show a marked tendency for localizedcorrosion which is apparently not dependent on the presenceof oxygen in the water. Furthermore, the repassivationpotential is so far below the rest potential that repassiva-tion is very unlikely. Thus there is a very high probabilityfor propagation of any localized corrosion which occurs.

5 CONCLUSIONS

Valve components made from the alloy which did notmoet specifications were found to have a corrosion rateseveral times that of Monel components. Furthermore, theywere found to have a much higher probability for propagationof crevice corrosion than their Monel counterparts. Thisstudy has demonstrated that the nickel-copper alloy containingexcessive concentrations of chromium and molybdenum is notsuitable for use in a sea water environment. Any componentsmade from this alloy can be expected to exhibit severelocalized corrosion.

8

ACKNOWLEDGEMENT

The authors wish to express their thanks toMs. B. Trenholm and Mr. P. Gergely of the Applied Vathemat isSection at Defence Research Establishment Atlantic fordevelopment of the computer program used in thisinvestigation.

-

FA bI.L L

CIIE.MICAL COMPOSITION 01 BUTTLRFLY VALVES

Composition, percentLI emen t

A.S.T.M. Bl4 Valve

Nickel o3.0 - 70.0 09.4

Copper Remai nde r 19.,

I ron 2.5 (max) 1. 0

Manganese c2.0 (max) -

Ca lbull 0 . 3 (ma x )

Silicoi O.) (Max) 1I.

Sulphur 0.024 (max)

Ch rom i un - 3.S

.olybdenum 4.8

1 0

_A

TABLL 11

CORROSION PARAMTI-RS FOR MONL. AND VAiVI.

ALLOI IN AERATLE AND 1)1 -ALRATI-) SLA WATI-R

S': Watt K r R ha b c icorr

Alloy Medium (mv vs. SCE) (m v) (my) (pa/cm')

Monel Aerated - 74 97.8 81 0.10

De-Aerated -337 35.9 53 179 0.11

Valve Aerated -94 21.3 104 0.54l loy

De-Aerated -104 26.3 161 0 .6)9

I-: rcorrosion potentialr

R polarization resistancep

b anodic Tafel slope

I c. odthodic TafeI slope

i I ) [ ,ill I le oil:; ( ( ) S r €1 I :1 i('1)! r

?ilii

TABLIE I I I

CRITICAL POTEN"IALS FOR MONLL AND VALVL

ALLOY IN )li-ALRATHl) SLA WAILR

Corrosion I~i tting RepassivationPotential Potential Potential

Sea Cater Er (mv vs, p (my vs. Er (mv vs.Allo Medium SeE) SCL) sc)

Nionel De-Aerated -S33 -60 -130

Valve De-Aerated -103 -45 -240Alloy

12

Fisure l Corroded butterfly valve componentsin sea water service.

Ep-

E-p

Ep-

Er

CURRENT DENSITY, LOG SCALE

Figure 2: Typical anodic polarization curve foran active-passive alloy.

13

Figure 3: Porosity (dark areas) in a valveshaft made of the unusual composition.(250X), unetched.

14

MOM&"

PPOTENTIOMETT

POARZAIN EFREC

CELL CELL

Figure 4: Schematic diagram of polarization circuit.

SCANNINGPOTENTIOMETER

XY RECORDER

REF 0 ECD

CEL RESISTANCE

WE O

POTENTIOSTAT

F igurC 5: Schemnatic' d iagram of call brat ioll ci rcu it.

16

E84

0 -.

0

0.01 .10 1.0 10 D00

CURRENT (mA)

Figure 6: Calibration curve for Wenking70HC3 potentiostat,

2-2

F- 0

ILl

LL

-2L -20 -10 0 10 20 30

A*~ (mV)

Figure 7: Experimental polarizationresistance curve for Nonelin de-aerated sea water.

17

> 0E7

0 - - - - - - -

0 -

3 jEr_-

CURRENT DENSITY (MkLA/cm )

Figure 8: Anodic Polarization curve ofMonel in dc-aerated sea water.

0 -P

0 ------------

-2

-0

0.1 ID 10 00

CURRENT DENSITY (MLA/cm 2 )

Figure 9: Anodic polarization curve ofvalve alloy in dc-aerated seawater.

18

Figure 10: Crevice corrosion occuring undernylon retaining screw on Monelalloy 400 immnersed in sea waterfor 40 days.

19

RE F E RENC I;S

I. "Annuai I Book of ASTM St and;i rds'', Part 0, ASTM,Phi I aIde I ph i a, p . 13 Lt 1979 )

2. I'a n5 u . R. , ''The Cor ros i on aind Ox i da ion oi lf Met a I z,'A rno I d, London , Ch . 7 (1960);1 ''First SUppl Iementa ry Vo I um(i'',A rno' I d, London ,

lUh I g, If. 11,., "Corrosion and Corros ion Cont roil'', 2nd cd.W'iley, Toronto, Ch. 5 (1971).

4. Leckie, H.P. and Uhlig, HIJ. ELectrochen. Soc., -11J,1262 (1966),

5. Morris, P.E. and Scarberry, R.C., Corrosion, 30, S3 (197-4).

0, Mansfeld, F,~, "'[he Polarization Resistance Technique forMeasuring Corrosion Currents" in "Advances in CorrosionScience and Technology", Vol 6, Edited by Fontana, M.G.and Stachie, R.W., Plenum, New York, p. 103 (1970).

7. G re ence, N. ., " Expe r iment al Elcc trod e K in ec t ics ",Rensselaer Polytechnic Institute, New~ York, p. 3 (l9tvrB.

S. Mainsfeld, F., Corrosion, 29, 397 (19731).

9 . Iloare, J . P1., "The Electrochemiist ry of Oxygen", Intersc ience,New York (190,8).

10. ctt e r, K .J .,I "LIc t ro ch enmiic a I kinct ic s" A ca de m ic P r ess,Now York (1967).

20

UNCLASSIFIEDSecurity CUIlaaIef

DOCUMENT CONTROL DATA -R & D"'-" I ,s~a~. At title, NiodV of absuaed anj indexisng annotation niust he. entered when os ove~rall docurne to clasilsedl

" IllINAT'NGrACe'VITY 2.. DOCUMPNT SECURITY CLASSIFICATIONDe ence Reerch Establishment Atlantic, UnclassifiedDockyard Laboratory, 21, GRHOUP

FRO Hiif.A*NS.D ZXa ..

UOCUMENT TITIE

Corrsionof Butterfly Valve-; in Sea Water Servie

4 OESCRIPTIvE NL S Tvp, of report and inclusive dares) Tech Memo

5 AUTHORIS) ~ 1l61

Lenard, Derek R."-/'MacLeod, Leslie" C. /041 a- A/u-J _____

F OUM rF*A-- - - 1* 7a. TOTAL NO. OF PAGES 7b. NO. OF REFS 1-Seplowew k980/4 27

PA. PROJFCT OR GRANT NO. --s 9a. ORIGINATOR'S DOCUMENT NUMBERISI

(L~'~,9DREATECHNICALMEMRANDUMI 80/H

Ob CONTRACT NO. 9b. OTHER DOCUMENT NO(S) lAisy other numbers 115.1 mray beaUignW this docunenl

10 DISTRIBUTION STATEMENT.

it SUPPLEMENTARY NOTES 12. SPONSORING ACTIVITY

Defence Research Establishment Atlantij9 Grove Street

______________________Dartmouth. N.S. B2Y 3Z713. ABSTRACT

Electrochemical techniques were used to compare the corrosionresistance in sea water of butterfly valve components made from an alloywhich did not meet compositional requirements to components made from thespecified Monel alloy 400. Polarization resistance studies indicatedthat the alloy which did not meet the specification had corrosion ratesseveral times higher than Monel in both aerated and de-aerated sea wate~r.Anodic polarization studies also showed that this alloy had a much higherprobability for propagation of localized corrosion than Monel. Thuscomponents made from this alloy can hc expected to provide poorer servicethan those made of Monel.

21 111

UNCLASSIFIEDsecurity cIaosiilar,

KEY WORDS

Corrosion

Sea water

Mone I

Electrochemical techniques

INSTRUCTIONS

IORiGINAT ING ACTIVITY Enter the name end address of the 9b. OTHER DOCUMENT NUMBERIS). If the document has beenorgrn.,ation issuing the document, assigned any other document numbers itolner by the originator

or by the sponsor), also enter this numberlsl,DOCUMENT SECURITY CLASSIFICATION. Enter the overallitctriry classification of the document including special warning 10. DISTRIBUTION STATEMENT. Enter any limitations onterms whenever applicable. further dissemrination of the document. other than those trmoo..l

by security classification, using standard statements such as,2b. rROLUP Enter security re~classiication group number. The three

orwrips are dlefineid in Appiendix 'M of the DO Security Regulations. (1) "Qualified requesters may obtasin copies of this

3 DOCUMENT TITLE: Enter the com'plete document title in all dcmn rmterdfnedcmnaincne.cauttr letters. Titles in alt cases should be unclassified. If a (21 "Announcement and dissetminatione of this documentusitciently descriptive title cannort be selected without classifi is not authorized without prior approval fromcation, show title classification with the usual onwcapital-letter originating aectivity.-abbrweviation in parenthieses imrmedlitl following the title.

1t. SUPPLEMENTARY NOTES. Use for additional eisplainaltory4 DESCRIPTIVE NOTES: Enter the category of documnelrt, e.g. noses.

technical report, technical note or technical letter, If asipropri-ate, enter the type of document, e.g. interim, progress. 12. SPONSORING ACTIVITY Enter the name of the depacrentalf.1irmary, annual or final. Give the inclusive dates when a project off ice or laboratory sponsoring the reseairch end

specific reporting period is covered. deveolopmnent. Include address.

rAUTHORISI: Enter the namels) of authoris) as shown on or 13. ABSTRACT: Enter en abstract giving a brief end fectualin the document. Enter lest name, first name, middle Intial. summary of the document, even though it may also appearIf military, show rank. The name of the principal author is en elsewhere in the body of the document itself. It is highlyabsolute minimum requirement. desirable that the abstract of classified documentis be unschaesi

fled. Each paragraph of the abstract shelf end wixth en6. DOCUMENT DATE: Enter the date (month. yeer) of indication of the security classification of the hiformestio

Establishment approval for publication of the document. In the paragraph (unless the document Itself is uuscieefiesd)represented as ITS). [S), (C), 1111. or 11.1).

/a. TOTAL NUMBER OP PAGES: The total page count shouldhoillow normal pagination procedures, ICe. enter the number The length of the abstract should be limited to 20 slsghe-spacedof pagei containing information, standard typewritten lines; 7%s inches long.

71a Ni IMSER OP REFERENCES: Enter the total number of 14. KEY WORDS: Key words oe technically meainingful terms orraeences cited in the document. sort phrases that characterize a documenst end could be helpful

in cataloging the document. Key words should be setectatd soasl~ PROJECT OR GRANT NUMBER, If appropriate. anti-nr the that no security classification Is reqirlied. identifiers. sucs as

.,irplicab, iiose-irch #end development project or grant nuomber equipment model designation, trade noaine military project Codetederi which the dri..iimeiit was written, nafome, geographic location. mayti be used "s key woords but will

iii~il~i uner hichthuiliizisantwas riten.be followed by en indication of technical contest.

'itiir NATON'S rDOCUMfNT NUM111,tfSI Enter the

; 1 wroolen e irn by which tha document *411 theit u(I cow-iiiile~ ty the originaliong activity. This

.,nmbaf must Iba uniiut- it) this idocument.

22