Hvt Ise F.Jnkn

Noeme 197

Ri..nd Washington '"352

fLd-- CIVIL [NC4IEtI•ISiNC E •UU rCL•'YCs Naval Construction Battaliorn Center, Port Hueneme, California 93043 ,

CORROSION AND BIOFOULING OF OTEC t

SYSTEM SURFACES -, DESIGN FACTORS !

By ]tmes P. Jenkins

November 197/8

; Sponsortd by

) •ENER(;Y RESFAHCH AND DEVELOPMENT ADM(NISTRATION

il Richlknd. Washi~ngto)n )9352

AiilO• d fo pubic"ren-e;dluerbutkn unlmite .( ) 20) 0!

-LI______

UnclassifiedREAD) INSTRUCTIONSREPORT DOCUMENTATIOt4 PAGE flEFORV. COMPtLETING FORM

f-4 SFR12. GOVT ACCCSSIO NO00 RECtA,

TR-868 DN787083 )___OROSION-ATOS D VFE

Naval IO C AsrciND Batta Ulon Center.C 52046; W6-~rW?

IS PERF IORM, N or GAN.EIJC N AME 404 ADDRESSfI' EUI V 0 ~ P ~AM LEEN ' ,PROET TASK

t AI 1. 0 ELL U~~r N'IT OONONAOIN

IS I CON"RISUTIOFFC N AtSMEAND t ,~ RAo,,,II

Apprgy Reseforc aubnd Deelepse;nit~bti, Adm !inistraiod.vm

Rchlraind, Wash~ingto 9935C paJet53cagr, rrcic ornM

tAphoed fo publect release; doistic to urkim..

C ifuin n orrosion, bif-fig OcC ~n ean -Tuhir ngergy pronetve coaiongsT, plnsr

'2ri-ical factors in thcecooi feasibility of the OTEC corcept. As the missi,)n andoper~ting requirements of an OTFC plant arc significantly different rLan those foir anyexisting facility 0any unique materials and design Froblems must be addressed. Thisreport identifies factors, that inf;luence biofouling and corrosion, recommends specificdesign features where appropriate, and identifies items requiring further investigation,

DO I* IA1 3 ~ iUnrlassified 4SfC,'N-T 0 CL A,;',,* rAtION OF tHIO PAGE fla.ot'. InI'IA,.I

mun, Unclassifiedi• ~SICURITY CI-ASS1 FICA 'ION Of" VNIS PAGE(IWhai DV&t 1919otd)

I

I.

S,

Library Card

Civil Engineering LaboratoryCORROSION AND BIOFOULING OF OTEC SYSTEMSURFACES - DESIGN FACTORS (Final), byIJmr F. Jenkins

k TR-868 53 pp illus November 1978 Unclassified

.I OTF(: systems 2. Protective coatings I. 52-016

I lIdofouling and corrosion of Ocean Thermal Energy Conversion (OTEC) plants areI critical favrors in the cconcmic feasibility of the OTEC concept. As the mision and operating

require. . tts of .n OTEC plant are significantly different man .hose for any er:isting facility.I many unique materials and design problems must be addressed. T1i,. report identifies fa,.tors

that infucnce •iofouling 3nd 'orrosior,, recommends specific design ferzures wher, appropriate,r and identifies items requiring further investigation. -7

!, I

Uncdassifiedi'ECURITY CLASWISICATION OF Tw1S PAO5'Whqý,.me. Ft-,,•tt )

CONTENTS

Page

INTRODUCTION. .. .................................. .. .. .. ......

IMPORTANCE OF COR~ROSION CONTROL .. ................................2

Heat Exchangers. .. .............................. ........... 2Auxiliary Equipment. .. .............................. ....... 3Support Platform. .......... ................................3

IMPORTANCE OF BIOFOULING CONTROL. .. ..............................3

Heat Exchangers. .. ................................ ......... 4Auxiliary Equipment......... ..... . . ..... .. .. .. .. ... 4Support Platform. .......... ................................5

DESIGN FACTORS INFLUENCING CORROSION. .. .......................... 5

L M-'chanical Design Factors. .. ................................6Geometry and Orientation. .. ............................6Layout .. ............ ..................................15ISurface Finish .. .......... ............................16Stress .. ............ ..................................16Cathodic Protection. ............ ......................18Protective Coatings. ........... .. ....................19Cleaning Systems .. .......... ..........................19

Environmental Design Factors. .............. ................20Chemical .. ............ ................................20

Velocity .. .......... ..................................25Flow Regime . . . . . . . . . . . . . . . . . . . . . %Pressure .. ............ ................................30Temperature. .......... ................................30

DESIGN FAC1CRS INFLUENCING BIO]FOULING. ............ ..............30

Mechanical Design Factors .. ............ ....................31Geomietry and Orientation .. ........ ....................31Layout. .......................... 32Surface Finish .. .......... ............................32Ca!'hodic Protection. ............ ................ . .33Antifouling Coatings .. ............ ....................33Cleaning Systems....... ... . . .... .. .. .... 33

Environmental Design Factors .cý....3

Chemical. .. ............... ..- ' 34Y.clocity . . . . . . . . .. . . . 34Temperpture. ............................... 35Preesure .. ................. . . . ... 35

______

Page

SUIKIARY ....... ........... ............................. .... 35

tdECOMM1ENDATIONS ....... ......... ......................... ... 36

REFERENCES ...... ........... ............................ ... 36

APPENDIX - Forms of Corrosion Attack ........ ............. ... 42

v

L. .

r

vi

"L1

INTRODUCTION

The biofouling and corrosion of Ocean Thermal Energy Conversion(OTEC) plants are critical factors in the economic feasibility of theseplants. Biofouling can decrease heat transfer capacity of the criticalheat exchangers, increase pumping requirements due to clogging of pipesand valves, and increase the drag and weight of the platform hull.Corrosion can result in high maintenance and replacement costs and largeamounts of downtime. Since OTEC plants Pe significantly different fromany system built to date in both size and operating characteristics,there is little direct experience from which specific anticorrosion orantibiofouling measures can be adopted. Past experience has shown,however, that there are certain basic principles of corrosion and foul-ing control which, if properly interpreted, can be applied to manyspecific applications.

Corrosion and biofouliag con.-ol should be carefully differentiatedfrom corrosion ind biofouling prevention. Almost all real industrialplants corrode. Almost all industrial plants that use seawater aresubject to biofoulirg. It has been found that it is not economicallyfeasible to completely prevent corrosion or biofouling. Corrosion andbiofouling control is the art/science of maintaining corrosion and bio-fouling at economically tolerable levels. The basic tactic of corrosionand biofouling control is to identify what levels and types of corrosionand biofouling can be tolerated in specific areas and to control them tothose levels. An alternative tactic is to predict the level and type ofcorrosion and biofouling that will occur and then devise equipment thatwill tolerate that amount of corrosion and biofouling.

The incorporation of corrosion and biofouling control into thedesign of any system is essentially an iterative process. A specificdesign is proposed using available corrosion and biofouling controlmethods. The results are analyzed to predict corrosion and biofoulingbehavior. Then the design is modified as necessary until a balancebetween cost and performance is achieved. The corrosion engineer has athis disposal only two basic tools. He can selecL a material that hastolerable corrosion resistance in a specific environment or he canmodify the environment such that a specific material will have tolerablecorrosion resistance. The corrosive environment is not limited to thechemical composition of the corroding substance; it is the "universalenvironment," which includes all corrosion-related parameters. Thebiofouling specialist likewise has only two basic tools at his disposal.Ho can modify the environment so that only tolerable levels of biofoul-ine will occur or he can devise methods for removing biofouling from thesur.faces at su:h a rate that their accumulation is tolerable. Again,the c2vironment in this case is the "universal environment," whichincludes all fouling-related parameters.

1.

This document will liiwit its scope to the effect of the "universalenvironment" on corrosion and biofrliling. Since this "universal envi-ronment" is a function of system design, the system design s'iould incor-porate, where possible, those design features that will result in a"uni-rersal environment" which is compatible with the materials andoperational requireme7,Ls of the OTEC plant.

The design factors that influence the "t-niversal environment" are,for the purposes ot this report, divided into two broad categories:mechanicil and envi'.onmental. Mechanical design factors are the combi-nation of size, shiape, location, stress, etc., which influence corrosionand biofouling. Environmental design factors are the combination ofchemical composition, valocity, flow regime, temperature, pressure,etc., which influence corrosion and biofouling.

The design engineer will find that this document does not give manyspecific design recommendations. However, it is hoped that the basicdesign principles described here can be applied to many specific consid-erations in the design of OTEC systems so that optimum use is made ofthe existing applicable knowledge of the influence of design on thecontrol of biofouling and corrosion. This report can serve as a basisfor the formulation and appraisal of proposed OTEC designs with respectto avoiding improper features and taking advantage of the properapproaches identified.

IMPORTANCE OF CORROSION CONTROL

Good corrosion control can be defined as the achievement of anoptimum balance between initial material ond equipment costs, mainte-nance costs, operational requirements, and system life. Achieving goodcorrosion control will result in optimum long-term overall system effi-ciency. The critical requirements for good corrosion control, or per-haps more appropriately the adverse effect of poor corrosion control,varies considerably among the various OTEC subsystems.

Heat Exchangers

The heat exchangers for the evaporation and cordensation of theworking fluid are probably the most critical subsystem of an OTEC plant.The selection of materials for these heat exchanger3 is limited by thech,.racteristics of the working fluid, the cost of the materials, andtheir availability ini the massi",e quantities required. Thus, it may benecessary to specifically design the corrosion control for much of theOTEC system around the material selected for the heat exchangers.Corrosion of heat exchangers is a critical problem in all industrialplants, but is a supercritical problem for OTEC. The corrosivity ofseawater, the thin-walled tubes required for good heat transfer, thelarge surface area of material exposed to the seawater, and the adverseeffects of mixing seawater with the working fluid all contribute to thesupercritical nature of the heat exchanger problem. The availability ofheat exchanger maintenance may be limited, unless a large number of

2

small units are used. However, a large number of small units wouldcause problems in other areas, such as flow distribution, system com-plexity, and system efficiency. Thus, heat exchanger maintenance willbe, at best, a compromise.

Auxiliary Equipment

The corrosion of auxiliary equipment, such as pumps, piping,valves, turbines, screens, etc., is somewhat less critical than *.hecorrosion of heat exchangers, but is no less important. Corrosion ofauxiliary equipment is less critical primarily because there are fewerrestrictions on material selection and more options for corrosion con-k trol methods. Also, the scale-up factor between existing auxiliaryequipment and thE size of equipment required for OTEC is not nearly aslarge as the scale-up factor for heat exchangers. Design for corrosioncontrol of these systems, although perhaps less critical, is more com-plex as there are such a large number of variables that can be manipu-lated in the design. The possible effects of system contamination duetc corrosion of auxiliary equipment also make the design of thesesystems more complex.

Support Platformi Corrosion of the support platform is important, but is much morestraightforward than for the other systems. This is because of thelarge amount of experience with similar systems in similar environments.If good standard practices used by marine "rchitects in the design oflarge ocean-going ships and offshore structures are applied to thedesign of the OTEC platt, a reliable and efficient support platform canbe built. This must not be interpreted to indicate the design of corro-sion control for the OTEC support platform will not present complex and

unique problems; only that these problems can most likely be solved bythe application of well-known corrosion control techniques.

IMPORTANCE OF BIOFOULING CONTROL

As for corrosion control the successful control of biofouling canbe defined as the achievem:ent of an optimum balance between initiali material and equipment costs, maintenance costs, operational require-ments, and system life. Achieving good biofouling control will resultin optimum long-term overall system efficiency. It should be re-emphasized that biofouling control is not biofouling prevention orelimination. Biofouling control is simply the capability of limitingbiofouling accumulations to levels that can be operationally tolerated.The amount of biofouling accumulations that can be tolerated in thevarious OTEC sdbsystems varies con;uiderably. Thus, the requirements andtechniques for biofouling control will vary considerably among theoesubsystems.

3

L06'a

Heat Exchangers

As the heat exchangers for evaporation and condensation of theworking fluid In an OTEC plant must transfer a large amount of heat atvery low temperature gradients, the presence of even thin films on theheat exchanger surfaces will seriously degrade heat exchanger perfor-mance. The presence of macrofouling organisms in the heat exchangers isimportant because of clogging, crevice, and turbulence effects. Macro-fouling is, however, not the critical problem. Even the thin, soft

films of microfouling are sufficient to cause enough degradation in heatexchanger efficiency to seriously affecL the OTEC power cycle. (See p

t 32 for detailed explanation.) Because the control of microfoulingusually results in the control of macrofouling, microfouling controlwill be the key factor in biofouling control of OTEC heat exchangers.

The evaporators in an OTEC plant will utilize warm, oxygen-rich andbiologically active surface waters, while the condensers will use cold,oxygen-poor, nutrient-rich, biologically less active waters from depthsof 600 meters (2,000 ft) or more. Because the evaporator and condenserwill operate with seawater of different chemical, physical, and biofoul-ing characteristics, it is probable that the most effective biofoulingcontrol techniques will be different for these subsystems. It has beenassumed, but not necessarily proven, that biofouling of the condenserswill be less of a problem than biofouling of the evaporators. Althoughthis may not be the case, it is certain that the biofouling control foreach of the systems should be approached with these potential differ-ences in mind.

Auxiliary Equipment

The biofouling of auxiliary equipment, such as pumps, piping,valves, screens, etc., has a less immediate impact on the operation ofan OTEC plant than biofouling of the heat exchangers, but it must beconsidered in order to obtain efficient long-term operation. Macrofoul-ing will probably be the major factor in biofouling of auxiliary OTECequipment, because this type of equipment usually has a high tolerancefor microfouling. Macrofouling, however, can cause restriction of fluidflow, turbulence, and other problems in piping systems. It can alsocause restriction, clogging, or jamming of valves, screens, and pumps;particularly when they are operated intermittently. The mass of macro-fouling accumulations can even cause problems in rotating machinery dueto an increase in rotating mass or an imbalance condition.

The control of biofouling in the auxiliary equipment of an OTECplant may be in some ways more difficult than biofouling control for theheat exchangers. Due to the wide variety of auxiliary equipment andbiofouling control requirements and techniques applicable to each compo-uent, the implementation of an overall biofouling control program forthe auxiliary equipment will be complex.

4

Support Platform

Biofouling of the support platform will have little immediateeffect on the operation of an OTEC plant; however, long-term effects maybe significant. The increased drag o& the platform due to biofoulingwill result in increased power requirements for relocation or dynamicstation keeping. If the plant is moored, the forces on the mooringsystem could significantly increase due to increased drag. The mass offouling accumulations can also change draft and trim of the supportplatform. Biofouling of the support platform can also cause clogging ofseawater intakes and other through-hull connections.

While the biofouling of an OTEC support platform will be similar tothat encountered on large ships and fixed ocean platforms, there aresignificant differences between the operation of OTEC and existing shipsand platforms that may lead to significant differences in biofouling and

i!biofouling control requirements. Ships normally operate at significant

velocities in the open ocean and are normally at rest in harbors thatare, to some extent, polluted and are not typical of the open ocean.Because of high fuel costs, the drag induced by biofouling is a signifi-cant factor in the economy of ship operation; therefore, biofouling iscontrolled at low levels. The significance of biofouling on the opera-tion of fixed ocean platforms is much less than that for ships, so it isnormally not controlled. OTEC platforms will operate at very low veloc-ities in the open ocean but, as discussed above, the effect of biofoul-ing on the OTEC platform may be significant. Thus, biofouling controlfor the OTEC support platform must be based upon existing experiencewith biofouling of similar structures. However, this experience is notdirectly applicable and must be adapted to unique OTEC plant operationsand requirements.

[, DESIGN FACTORS INFLUENCING CORROSION

The "universal environment" experienced by a metal will ietermineits corrosion behavior. By avoiding those design features tVat lead toalt aggressive "universal environment," the corrosion of an OTEC plantcan be significantly reduced. Proper design, combined with other corro-sion control measures, will result in an OTEC plant with optimum resis-tance to deterioration.

The design factors that influence the "universal environment" canbe divided into two broad categories: mechanical and environmental.Factors discussed under these broad categories often have significanteffects on the various forms of corrosive attack discussed in theAppendix. If the acceptable forms and rates of attack and the designfactors influencing them are considered, most adverse conditions can beavoided. I

54

Mechanical Design Factors

Mechanical design factors are those conditions of the metal sub-strate that influence the "universal environment" and, thus, the corro-Ision behavior of the metal. This influence may be direct, as in thecase of the effect of induced stress on stress corrosion cracking, or itmay be indirect, as in the case of the effect of equipment layout onmaintainability and, thus, tolerance of corrosion.

Geometry and Orientation. The broadest L' tegory of mechanicaldesign factors that influence corrosion is geometry. By proper controlof geometry a given material can be allowed to exhibit its best corro-sion resistance. All alloys can be made to corrode if the proper, orrather, improper, conditions of exposure are present. If the conditionsunder which a given alloy is susceptible to corrosion are known, theseconditions can be avoided or controlled.

Certain basic principles of corrosion can be applied to the selec-tion of proper geometry for a given function and materiail. The first ofthese principles is that any difference in condition between differentareas on a given member can lead to accelerated attack by a differentialenvirornment cell. These differences can be direct, such as bimetallicor multimetallic connections, or they can be subtle, such as in attackdue to crevices.

Whenever two or more dhissimilar metals are placed in electricalcontact and exposed to a conductive environme-..t, intolerable galvanicattack may occur. Therefore, its effect must be carefully considered.Several basic factors control the direction and magnitude of galvanicattack which will occur in such a coujde. The first is the position ofthe material on the electrochemical potential scale for the exposureconditions. Such a list is given in Table 1. This so-called "GalvanicSeries" has been develuped from many experiments and experiences withbimetallic couples in seawater. Materials listed near the top of theseries are active, or anodic, with respect to the more noble, or cath-odic, materials below them on the list. In general, the greater theseparation be -.ween the alloys on the list, the more severe the galvanicattack on the anodic material will be. This galvanic series is forsurface seawater and is directly applicable only in that environment.Considerable deviation from this series is possible in environmentsoignificantly different than surface seawater. Galvanic series forthese other environments are generally not well developed; they remain

to be developed from related experience or experiment.The actual rate of galvanic attack is dependent on many undeter-

However, several geometric factors can be manipulated to minimize theadverse effects of galvanic coupling. The first of these is area ratio.The most severe galvanic attack will occur when the exposed area of thecathodic material is large with respect to the exposed area of anodicmaterial. Even when the materials involved in the couple are fairlyclose in the galvanic series, the rate of galvanic corrosion can be highif the area ratio is adverse. Thus, it is good design practice to avoid

6

galvanic couples where possible and, when they cannot be avoided, tomake the exposed anodic area large with respect to the exposed -thodicarea. A typical case where adverse anode/cathode area ratios inad-vertently designed into a system is when an innocuous galvani e isimproperly coated. Uninformed logic would indicate that the cor-rodable material could be left bare while the more corredabl, ýrialshould be coated. When coating defects occur on the coated a , thena small anode/large cathode situation occurs. Thus, when 'iuiag agalvanic couple, coat both materials or coat the cathode only. To coatthe anode only is to invite rapid attack.

Galvanic effects can be used to advantage in many applications.When a critical system component is relatively small it can be fabri-cated from a material that is cathodic to the material from which thebulk of the system is fabricated. Thus, the critical component willreceive cathodic protection from the bulk of the system. The accelera-tion of corrosion of the relatively large exposed area of the bulksystem will be minimal. This tecl~nique is commonly applied. Examplesare cathodic "itrim"t seats, discs and stems of valves having anodicbodies, and cathodic impellers, shafts and w%--ar rings of pumps withanodic casings.

An additional factor influencing the rate of galvanic attack isexternal circuit electrical resistance. As shown in Figure 1, -. gal-vanic cell is essentially a battery/load system. When the intermetallicresistance, R1 , or the environmental resistance,' R 2 , is raised, thecurrent flow, I0, is lowered, thus reducing the amount of galvaniccorrosion. Thus, the distance between the active members of a coupleshould be maximized when possible. Also, insulating or high resistance

F bushings, etc., can reduce or el4!.minate galvanic corrosion.Concentration cell corrosion is similar to galvanic corrosion in

that it is a result of an electroche~mical ;ifference between electri-cally connected portions of a member. However, in concentration cellsthis electrochemical difference is due to a difference in environmentrather than a difference in substrate composition. In seawater, themost common form of concentration cell attack is crevice corrosion. Inthis form of attack, seawater is trapped within small gaps and stag-nates. Either the consumption of a constituent of the solution, such asoxygen, or the buildup of a substance, such as a metal ion, leads to asignificant difference in environment between the inside of the creviceand the outside. According to LaQue (Ref 1, p 39), metals near the topof the galvanic series, such as iron, tend to crevice corrode by theii oxygen starvation type of attack, while materials, such as copper (lowon the galvanic series), tend to crevice corrode by the metal ion con-centration type of attack. Alloys near the middle of the galvanicseries, such as the copper-nickel alloys, are less susceptible to crev-ice attack. Alloys, such as stainless steels, which exhibit a morenoble potential than their constituents, are often particularly sensi-tive to oxygen starvation attack.

7

Table 1. Galvanic Series in Surface Seawater

MagnesiumIHigh Purity AluminumZincGalvanized Steel (New)Aluminum Alloy 5052Alclad Aluminum (X7002)Aluminum Alloy 3003Aluminum Alloy 6061-T6Aluminum Alloy 7079-T6

Aluminum Alloy 2024Ca.•miumCadmium Coated Steel (New)Mild SteelWrought IronAlloy SteelCast IronNi-Resist Cast IronMonel Alloy 400 (ative)Stainless Steel 410 (active)

Stainless Steel 430 (active)Solder (60% P6 - 40% Sn)Stainless Steel 304 (active)

Stainless Steel 316 (active)Stainless Steel Alloy 20-Cb (active)Lead

More Active (anodic material) TinMuntz MetalManganese BronzeNaval BrassNickel (active)

More Passive (noble or cathodic material) Inconel &60 (active)Yellow BrassAluminum BronzeRed BrassCopperSilicon BronzeNickel SilverCupro-Nickel 95-5Cupro-Nickel 90-10Cupro-Nickel 80-20Cupro-Nickel 70-30 low FeCupro-Nickel 70-30 high FeG-BronzeM-BronzeNickel (passive)Inconel 600 (passive)Silver SolderMonel (passive)Stainless Steel 410 (passive)Stainless Steel 430 (passive)Stainless Steel 304 (passive)Stainless Steel 316 (passive)

Stainless Steel Alloy 20-Cb (passive)Stainless Steel A.I6XSilverInconel625

NOTE: In general, the greater the separation Ilastelloy Cbetween alloys, the more severe the galvanic Titaniumattack will be on the anodic material. Graphite

GoldPlatinum

J,

II

nt ntact resistancee

R, ontact resistance

n A

R2 i ontact resistance ,

Figure 1. Galvanic corrosion equivalent circuit..A•

9•

rhe most reliable technique for avoiding crevice corrosion is toavoid crevices. Crevices from bolted or riveted joints (Ref 2, p 17),gasketed joints (Ref 1, p 164), discontinuous welds or welds withoutfull penetration (Ref 2, p 17) as well as other situations where solu-tions can be trapped or held should be avoided where materials suscepti-ble to crevice corrosion are used. Good design practice for OTECsystems is to use continuous, full penetration welds wherever possible.When bolted or other lap type joints are requitced, a nonwater ab'.rbinggasket materi'. (Ref 2, p 20) should be placed between the mating sur-faces. When such gaskets are used, they should be carefully sized andplaced in order to avoid aggravating crevice attack (Ref 3, p 10.16).The effect of a poor gasket fit is shown in Figure 2. Nonwater absorb-ing caulking materials, such as silicone rubber, can also be used toseal crevices when the crevices cannot be avoided. Such gasket designfeatures cannot prevent crevice attack on vulnerable materials, such asmost stainless steels and some copper alloys. Special attention shouldbe given to all joints of this type to avoid crevice attack by usingnonvulnerable materials or proper sealants, or by substituting an alter-native connection method.

Another concentration cell related type of attack is c.,used bysurface contamination. The accumulation of sediments, corrosion prod-ucts, or other contaminants on metal surfaces often leads to concentra-tion cell type attack (Ref 3, p 10.16). A system that is clean, willremain clean, or can be easily cleaned will be more cocrosion free thaua system that will accumulate and hold contaminants (Ref 2, p 15).There are a multitude of structure geometries that can lead to thetrappir'g and holding of contaminants. Figures 3 (Ref 3, p 10.15) and 4(Ref 2, p 25) show typical situations where contaminant trapping islikely and recommended alternative practice.

Interfaces between gases and liquids, such as is present in par-tially filled tanks, are frequently the site of corrosion (Ref 2, p 27).Piping and tanks should be designed so that they are completely full ofeither liquid or gas. Figure 5 shows typical situations of this typeand recommended alternative practice.

Erosive damage can also occur from vapor impingement, especially ifdroplets of condensed liquid ar'e present. Proper baffling of the vaporstream can be used to prevent this type of attack.

In general, smooth flow conditions in piping systems are desirableto reduce corrosion due to impingemcnt and other forms of turbulence-related attack. Desirable piping designs, suzh as those shown inFigure 6 (Ref 2, p 29), should be used whenever practical.

Dead ends, where stagnation can occur, should be avoided wheneverpossible. When they cannot be avoided, the dead end should enter fromthe top of a line or tank in order to minimize contaminant buildup(Ref 2, p 27). The placement of valves in horizontal runs of pipingrather than in vertical runs also minimizes corrosion from this effect(Ref 2, p 27).

Structural geometry also affects the performance of protectivecoatings. Sharp corners are to be particularly avoided on structuresthat are to be painted. In Figure 7 the effect of sharp corners oncoating thickness is shown (Ref 3, p 10.22).

10

• . . -•

I ti.

Efiect offus (aietft

Poor Practice Good Practice

Figure 2. Effect of gasket fit,

1I~i L7Z~orLdrain holes *,i

or

drain holes 1

weld

Poov Practice Good Practice

Figure 3. Structural member effects. 19

;i/

f ' 11

121

.94U

0.1

4)

cc

4.4

.P4

P4.9.4

41'

a'wr

. . . . . . . . . . . .........

iiti

L Figure7.Efiguec 6. Piapin edgesign efcotsn. hckes

14I

e -raeoinainwt esett h oiotlas has aneffet o coroson. Thi is rimril du totheaccumulation ofcorrsio prouct or edient on he ppersid ofhor-Azontal sur-

face. Pope sytem eomtrytha miimizs cntainaiondue toaccuulaionof o.- osin poduts ad sdimntsandmaximizes themainainbilty ad ceanbiliy o thse srfaes illminimize the

Ohrgeometry- related design factors should be considered in thedesign of OTEC systems. Such an example is the possibility of corrosionfatigue induced by structural vibration (Ref 4, p 163). In general, asimple design is preferable to a complex one. This includes the reduc-tion of the number of structural members whenever possible, particularlyin areas where corrosion is a problem. Redundant members in a corrosivezone usually only result in redundant failures rather than in addedsafety. A simple design, easy to maintain and repair, which avoidsdissimilar metal connections, crevices, etc., will maximize the servicelife of properly selected materials.

Layout. The layout of equipment in an OTEC plant will have littledirect effect on the corrosion of the equipment. However, the indirecteffects on service life can be significant. The most important factorin the relationship between layout and service life is cleanability andmaintainability. A system that is easy to clean and maintain willnormeally receive more frequent and thorough care than a system that isdifficult to maintain. Convenient access to~ maintenance points isprimarily a function of layuat and is an important design factor.Convenient access to and ease of replacement of failure-pione componentswill also reduce system downtime and unexpected failures. For equipmentthat must be drained for cleaning and maintenance, proper geometry suchas that indicated in the previous section will allow for complete drainageand easy flushing of cleaning residues.

Not only is easy access to equipment important for cleaning andmaintenance, but it is also important in order to be able to properly

F inspect and assess the equipment or structure for cleaning, maintenancerepair, or replacement. Proper cleaning and maintenance schedules canoften only be derived from the results of thorough, carefully plannedinspections. Proper inspections also have the beneficial effect ofreducing downtime due to unexpected failures.

One possible direct effect of equipment layout on corrosion is that

of metal ion contamination. Metal ion contamination of the environmentI can significantly accelerate the corrosion of some materials, especiallyaluminum alloys. Even very small amcunts of copper from the corrosionof copper- containing alloys will deposit on the surfaces of aluminuma11'ys and result in rapid at~tack of the aluminum (Ref 1, p 19). Thiseffect can be prevented by eliminating the source of metal ions or bylocating the equipmient such that the copper ions are produced downstreamof the aluminum alloy components. Thus, if flow is maintained, thecopper salts will not contact the aluminum alloys. Another potentialsource of copper salts are the cop ~er-containing antifouling coatings;

15

rM .

r[ thesecoatings should be avoided on aluminum stfuctures (Ref 1, p 19).Othermeaiocotmntoprbeseitwtmecranalmu,and nercury and copper.

In some cases, the effect of dissolved metal salts can be benefi-cial. Iron salts can reduce the corrosion of certain copper alloys(Ref 5). Also, corrosioni products from aluminum and zinc have beenreported to have certain beneficial effects on the corrosion of steel(Ref 6).

Surface Finish. No studies of the effect of surface finish on thebehavior of metals in seawater were located in the literature surveyperformed for this report. However, surface finish can be expected tohave some effect on the corrosion behavior of some materials. Materialssusceptible to crevice corrosion will probably suffer more extensiveattack when exposed in a mechanically roughened condition. This is dueto the large number of crevices formed during the roughening process.This is simailar to thz adverse effects of sheared edges on the corrosionbehavior of similar materials (Ref 7, p 89).

Another effect not directly related to surface finish, but with asimilar result to the effect of sheared edges is the orientation of theexposed grain on corrosion behavior. Most ma,-rials are anisotropic inboth their mechanical and chemical behavior. Some alhminum alloys areparticularly anisotropic in their corrosion behavior (Ref 8). In thesealloys, corrosion attack is many times greater in the longitudinaldirection (parallel to grain orientation) than in the transverse direc-tion (perpendicular to grain orientation). An extreme example of thisform of attack is the exfoliation attack noted on some very highstrength alumninum alloys (Ref 7, p 190); the corrosion products exertsufficient preqsure to delaminate the material, resulting in catastroph-ic attack.

Stress. Tensile stresses below the level required for plasticdeformation of the metal, or surface films (such as mill scale or pro-tective corrosion products) do not change the rate of general corrosionof metals exposed to seawater (Ref 9, p 236). The presence of a corro-sive agent, such as seawater, can, however, significantly reduce theload-carrying capability of metals, even when no significant generalattack occurs. Also, a given amount of corrosion damage on a highlystressed member will usually have more serious consequences than thesame amount of damage on a member with lower applied stress.

The reduction of the ability of a member to sustain a static stresswhen exposed to a corrosive environment is defined ats stress corrosioncracking. While the mechanism of this phenomenon is, at best, poorlyunderstood, the effects of a given corrosive environment on the static jload-carrying capacity of a given material can be well defined by theuse of fracture mechanics. As all engineering structures contain flaws,test results of stress corrosion cracking coupled with fracture mechan-ics can be used to determine the maximum flaw size that can be toleratedin a given material and to define the level of inspection required to

locate flaws of this critical size.

ILI

The following materials are recognized to be susceptible to stresscorrosion cracking in marine environments (Ref 10, p 3).

1. High strength steels

2. Austenitic stainless steels

3. Precipitation-hardened stainless steels

4. Quenched and tempered Martensitic stainless steels

5. High strength titanium alloys

6. High strength aluminum alloys

7. Copper zinc alloys (brasses)

8. Aluminum bronzes

9. Manganese brinzes

The suscejtibility of high strength steels to stress corrosion crackingis related co the strength of the material. For a given alloy composi-tion, the higher the strength level, the greater the susceptibility tostress corrusion cracking. Stress corrosion cracking is the principallimiting factor in the strength level of steels that can be used safelyin seawater (Ref 1, p 216). Austenitic stainleis steels are susceptibleto stress corrosion cracking in seawater when they have become sensi-tized to intergranular corrosion by improper heat treatment or improper

welding techniques (Ref 1, p 221). The avoidance of sensitization byheaL treatment or alloy additions i. well documented (Ref 11) and canessentiafly elirvinate this form of attack. Austenitic stainless steelscan suffer stress corrosion crackiig in seawater even when not sensi-tized if the t=mperacure of the seawater is above about 140OF and thereis an opportunity for concentrations of chl.rides in crevices, in splash4reas, or under insulation. These czitical conditions should not beencountered in OTEC components operating well below 140*F. The resis-tance of austenitic stainless steels to stress corrosion cracking isinfluenced also by alloy composition. A high nickel content is espe-cially desirablL. TIhe 25% nickel content of A16X alloy io high enoughto prcvent streis corrosion cracking under OTEC conditions.

The prtipitation hardening stainless steels are also susceptibleto stress corrosion cracking. The susceptibility of these alloys is, asfor the alloy steels, a function of the strength level. Alloys heat-treated to high strength levels, particularly when age-hardened attemperatures below about 1,000*F, are most susceptible (Ref 7, p 134).

The quenched and tempered martensitic stainless steels when hard-ened above about 32 Rockwell C are susceptible to stress corrosioncracking in seawater (Ref 12, p 33).

The stress corrosion behavior of titanium alloys in seawater iswell documented (Ref 7,10,13-17). The incidence of stress corrosioncracking of titanium alloys in marine service has been low because ofthe uL- of resistant alloys, the low stress levels encountered, theknowlr.,, from earlier failures, and the application of the results of

17

[ laboratory studies. Future failures can be avoided by the use of alloysknown to be resistant to this formu of attack under the conditions ofexposure (Ref 15, pp 1-2).

Aluminum alloys, particularly the high strength alloys, are alsosusceptible to stress corrosion cracking in seawater (Ref 1,7,18,19).As is the case for titanium alloys, only those aluminum alloys that havebeen found to be resistant to stress corrosion cracking in seawatershould be used in applications where stress corrosion cracking wouldresult in component failure.

Copper zinc alloys are susceptible to stress corrosion cracking inseawater (Ref 1, p 223). The presence of ammonia, even in very smallconcentrations, greatly increases the severity of this attack (Ref 20, p147).

The 90:10 copper-nickel candidate alloy CA706 is not susceptible tostress corrosion cracking in seawater or in ammonia -contaminated sea-water.

Aluminum bronzes are also susceptible to stress corrosion crackingin seawater. Heat treatment or nickel additions can effectively elimi-nate this effect (Ref 1, p 223). AMmonia also can be expected to accel-erate this attack.

Careful consideration mu0 be given to the use of alloys in an OTECplant that are susceptible to stress corrosion cracking. Unless furthertesting shows the material to be suitable, it should not be used in anOTEC plant. Usually the stresses induced during fabrication of a struc-ture are as serious in leading to stress corrosion cracking as areapplied loads (Ref 19, p 18).

The reduction of the ability of a member to sustain a dynamicstress when exposed to a corrosive environment is defined as corrosionfatigue (Ref 21, p 148). The combined effect of corrosion and fatigueproduces an adverse cojoint action that leads to premature failure(Ref 22, p 1). Any material that is subject to corrosion in seawaterwill be affected by cor'rosion fatigue to a greater or lesser '_ýxtent.Corrosion fatigue test data, preferably coupled with actual serviceexperience in seawater, must be considered to predict the service lifeof materials under the conditions of cyclic loading expected in an OTEC

F application.

Cathodic Protection. Cathodic protection can reduce, or in somecases eliminate, corrosion of materials in seawater. While the designof cathodic protection systems is outside the scope of this report,several factors in overall system design can affect the applicability ofcathodic protection. A system that is designed according to the princi-ples outlined earlier will not only result in a system that is inherent-ly more corrosion-resistant, but it will also result in a system that iseasier to cathodically protect.

In general, cathodic protection can be applied easily to externalsurfaces of components continuously immersed in an electrolyte, such asseawater. The more complex the surface geometry, the more difficult thedesign of the cathodic protection system will be. The internal portions

18

of large tanks, condenser waterboxes, pipes, etc. , can be cathodicallyprotected. The internal surfaces of small diameter tubes can be cathod-ically protected if there is no fluid flow in the tubes (Ref 23).

Specific criteria for cathodic protection of large steel structuresare well defined (Ref 24-28). Specific criteria for cathodic protectionof some stainless steels, copper alloys, aluminum alloys, and nickelalloys have been developed (Ref 28-32), but they have not been validatedto the extent of tLhe criteria for protection of steel structures.

Protective Coatings. Protective coatings are widely used in themarine industry to mitigate corrosion of structures and equipment. Forequipment exposed to the marine atmosphere, protective coatings canprovide long-lasting protection, if they are properly selected andproperly applied to well-designed structures. Good design practices, asoutlined previously, allow for easy access for metal preparation andcoating application, and for periodic inspection and maintenance.Simple designs with rounded corners allow for uniform application of theselected coatings. It must be remembered that all protective coatingswill contain defects, and that the coatings will deteriorate with time.Thus, unless additional protection is provided, corrosion will occur atcoating defects. This is of particular concern on alloys that corrodenonuniformly.

Protective coatings on submerged surfaces often serve a dual ortreble purpose. The primary objective of applying the coating is toisolate the environment from the metal and, thus, eliminate corrosiveattack. However, with simple and always imperfect barrier coatings,corrosion will continue at the coating defects at least at its normalrate. When a barrier coating is combined with a cathodic protectionsystem, a dual benefit is achieved. The cathodic protection systemneeds only to protect the areas where coating defects or deteriorationhave exposed the unateclying material. While there is considerablecontroversy over the economic merits of the combination of these twoprotective systems, it is generally recommended that cathodic protectionof large, submerged surfaces should be coupled with a protective coatingsystem (Ref 28, p 135). Cathodic protection also generally increasesthe life of protective coatings by reducing undercutting and lifting due

k' to substrate corrosion. Protective coatings also can be formulated withpoisonous compounds to reduce or eliminate biofouling of the submergedsurfaces.

Inorganic coatings (such as the zinc-rich paints), hot-dip galva-nizing (zinc coating), and aluminizing are also useful methods forprotecting steel and other metals when submerged in seawater (Ref 1, p296).__________________

Cleaning Systems. It is generally accepted that a clean systemwill be less subject to corrosive attack than a system in which sediment

Fand debris are allowed to accumulate. A well-designed system will tendto remain cleaner for a longer period than a poorly designed system.Eventual accumulation of debris or biofouling is inevitable in a sea-water environment, and some critical systems 'will require periodic oreven continuous cleaning in order to maintain the desired operational

19

characteristics. This is particularly true for heat exchangers andsmall diameter piping. The effect of this periodic cleaning on thecorrosion of the base material or protective coatings must be evaluated.Experience with cleaning of materials that form thin, tightly adherent,corrosion-product films, such as the copper alloys and stainless steels,has shown that these surfaces can be cleaned with little or no adverseeffect. Improper mechanical or chemical cleaning can, of course, harmany material. The effect of cleaning on materials that form looselyadherent ...orrosion-product films, such as steels, can be cexpected tohave conasiderable adverse effect, as these alloys depend on a buildup ofthe corrosion products to achieve relatively low corrosion rates (Ref 1,p 44). Removal of theoe products can result in rapid corrosion.

Protective coatings can also be easily damaged by improper clean-ing. This is particularly true at projections and sharp edges. Theseadverse design features should be avoided on coated systems that requirecleaning.

Environmental Design Factors

Intelligent design must consider the environment to which thestructure or equipment will be exposed. In some cases, it could bepractical to modify the environment to improve the behavior of certain

materials.I Seawater is an extremely complex medium that contains nearly dli ofthe known elements. In addition, seawater contains large amounts ofcomplex organic substances, both living and nonliving (Ref 33). Due tothe complex nature of seawater and to the local generation, introduc-tion, and contsamption of important constituents, the composition ofseawater varies considerably with both location and depth in the ocean.

In addition to the chemical factors in the environment, otherphysical factors must be considered in order to predict and controlcorrosion of an OTEC plant. Velocity of the seawater is a very impor-tant factor and may indeed prove to be a critical factor for such compo-nents as heat exchangers. Flow regime, turbulent or laminar, also canaffect corrosiou and must be considered along with velocity. Otherfactors, such as temperature, pressure, and heat flow, have been shownto have some effect on corrosion and must be considered in an intelli-gent system design.

Chemical. Seawater is a complex and nonhomogeneous environment(Ref 34, p 187). Although many chemical constituents of the environment

vary considerably with location and depth, experience with ocean struc-tures and corrosion exposure testing has shown that material performancecan be correlated with the value of a few key variables. In naturalsurface seawater, the pý 'mary chemical variables of interest are salin-ity, organic material, and pollution.

OTEC plants will probably be located a reasonable distance fromsources of diluent freshwater that would affect salinity to the extentnecessary to significantly affect corrosion behavior. Therefore, the

20

effect of salinity variations in natural surfa:e seawater will be small(Ref 35, p 5). Likewise, the effects of industrial pollution on OTECsystems will be small, unless they are accompanied by a large floatingindustrial plant. If an OTEC plant will be exposed to either low salin-ity or polluted waters during construction, maintenance or repair, thencareful consideration must be made as to the effect of these variableson material performance. For example, if some copper alloys are ini-tially exposed to polluted seawater, their cnrrnsi-n resistanc,' uponsubsequent exposure to unpolluted seawater will be significintlyreduced. Initial exposures to fresh seawater allow these alloyr todevelop protective corrosion-product films that will improve theircorrosion resistance even when subsequently exposed to pollut-d seawater(Ref 36).

The concentration of organic material in seawater is important inmany corrosion processes (Ref 33). The effect of this concentration onthe corrosion of metals in surface seawater has not been well studied.Open ocean testing in environments similar to those expected at OTECoperating sites may be necessary to precisely determine the cocrosionbehavior of materials for critical systems in the OTEC plant.

There is a greater variation in chemicals and corrosion behaviorbetween surface waters and waters at depth than there is between surfacewaters at various locations. Oxygen content, pH, and organic materialconcentration and type are the major variables affecting cor-osion atdepth (Ref 35, p 3-6). The variation of oxygen content and pH withdepth in the Pacific Ocean is shown in Figure 8 (Ref 7, p 4). Thevariations in oxygen content and pH with depth in the Atlantic Ocean aresignificantly less than those in the Pacific Ocean (Ref 34, p 187). Asshown in Figures 9 and 10, the corrosion behavior of some materialsvaries considerably with differences in oxygen content (Ref 7, pp 15,213,214). Recent research has indicated that for some materials, suchas aluminum alloy 5052, the variation in pH with depth may have a moresignificant effect than the variation in oxygen content (Ref 37).Actual environmental exposure tests in the deep ocean have been limited,primarily due to high cost. However, when deep-ocean waters are to beencountered by an OTEC plant, References 1, 7, 10, 16, 17, 34, 35 and 38through 44 should be consulted in order to obtain valid data for mate-rial performance under low velocity conditions.

In addition to the direct effects of variations in environment withdepth, these variations can possibly lead to concentration cell attackon electrically conductive members that pass through waters of varyingcomposition (Ref 34, p 188).

Although the accumulation of fouling organisms on exposed surfacesat depth is much lesq than near the surface, there are organisms presentat all depths. The concentration of uncombined organic material may besignificantly different at depth. The effect of such differences inorganic constituents has not been determined.

21

WI

Surface din

1,000 0

ZoxyBgf

2,000 I0

pitWIk-k3,000 0

4.b000C

5,0000

6,000

6,500 1UI I II

0 1 2 3 4 5 6

Oxygen (mil/)

6.4 6.6 6.8 7.0 7.2 7.4 7.6 7.3 8.0 8.2

pi1

Figure 8. Oxygen content and pH versus depth - Pacific Ocean.

22

* •r-"r ".Wwqu''.; .

M77

Ii

6

C-

o

83

'II I I-.! .I*,,- ,aI 2 34 3 6.

Oxygcn (mI/I)

Figure 9. Effect of oxygen content in seawater on thecorrosion of steel after 1 year of exposure.

23

SI 231

81,1

711

L51)

40

ill

Io

O)xygen (m/l D

Figure 10. Effect of oxygen content in seawater on

the pit depth of aluminum alloy 6061-T6after 1 year of exposure. A

24

LAI

The solubility of several constituents of seawater is also related,either directly or indirectly, to depth (Ref 45). Thus, it is possiblethat precipitation of materials such as calcium carbonate may occur whendeep waters are brought to the surface. Such possible precipitationswould have an adverse effect on heat exchanger performance and should beinvestigated further.

Velocity. Flow velocity may be one of the most significant, factorsaffecting the selection of materials for key components of an OTECplant. Most materials exhibit satisfactory performance over only arelatively small range of velocity and flow conditions. Aluminum alloysand stainless steels perform well at moderate velocities, but are se-verely attacked under stagnant conditions. Copper alloys generallyexhibit a threshold velocity above which they corrode very rapidly butbelow which they have satisfactory resistance' (Ref 46, p 1745). Mate-rials such as titanium often show no effect of velocity within thelimits expected for OTEC plant operation (Ref 1, p 249).

Specific data from corrosion tests, coastal power plants, and shipservice may iiot be directly applicable to the design of an OTEC plantdue to the differences in chemical characteristics of the environmentand sizes of equipment tested. The reasons for the effect of variationsin chemical characteristics are the same as those for low flow condi-tions as discussed previously. The effect of equipment size on velocitytolerance is more subtle. It has been shown that velocity per se is notthe critical factor in the behavior of some alloys. For these, thefluid shear stress at the metal surface is the critical factor (Ref 47).For most equipment, the fluid shear stress is a function of velocity,size, geometvy, and fluid viscosity (Ref 48). Thus, the critical fluidshear stress or tolerable shear stress range should be determined formaterials that are candidates for critical OTEC systems and should beapplied to specific equipment designs. Interim velocity limits can beapplied to noncritical candidate materials for validation testing. Suchdata are presented in References 1, 21, 35, 46, and 49 through 51 aswell as in many other documents that outline specific equipment ormaterial performance. From these data the following liwiting flowvelocities, based on resistance to impingement attack, are suggested foruse until more OTEC specific data can be obtained:

25

Limiting Maximum VelocityComponent Material

m/sec ft/sec

Heat Aluminum 1.5 to 1.6 5 to 6exchanger Titanium no limit no limittubes Stainless steel (AM6X) no limit no limit

90:10 copper nickel (CA706) 3 In

Salt Steel 1.5 to 1.6* 5 =, 6*water Alumitium 1.5 to 1.6 5 to 6piping Titanium no limit no limit

Stainless steel (A16X) no limit no limit90:10 copper nickel (CA706) 2.5 to 3.5** 8 to 12"wr

*Based upon corrosion rate of 0.5 mm (0.02 in.) per ye0r. If macro-fouling is present and covers surface, then i s 0.125 ye n (0.005 nco)per year.

,,Higher limit for larger diameter pipes.

Flow Regime. As discussed in the previous section, the fluid shearstress at the metal surface is the critical variable in the performanceof metals under velocity conditions. Under turbulent-versus-laminar

flow conditions, the fluid shear stress can vary considerably at a givenvalue of bulk flow due to the variation in Reynolds number or frictionfactor (Ref 47). Flow regime also affects mass transfer processes whichoften control corrosion behavior (Ref 52). OTEC heat exchangers willprobably operate in the turbulent regime. Both velocity and flow regimemust be considered in the determination of fluid shear stresses on metalsurfaces. Localized velocity gradients are likely to cause more attackthan higher uniform velocities. Damaging velocity gradients are oftencaused by small radius bends in piping, improperly designed pump dis-charges, and partially closed (throttled) valves. Debris, such as

fouling organisms, that partially obstruct small diameter tubes can alsocause damaging velocity gradients and should be avoided by proper up-stream screening.

Cavitation-erosion is a drastic form of attack associated with highvelocities and high velocity gradients such as are usually encounteredin pumps and valves. Critical components, such as impellers, diffusers,and valve trim, should be fabricated from metals such as titanium andhigh nickel-chromium-molybdenum stainless steels that. are resistant toboth cavitation-erosion during operation and crevice attack duringstand-by periods.

Heat exchanger water boxes are often subject to turbulence- relatedattack, and the induced turbulence can also affect the performance ofthe heat exc:hanger tubes. Proper water box design features are shown in

il Figure 11 (tef 21).

26

~ ~ ~~~~~~~~~~~~~~4 .. .. ..... .. ... . .......... ..... " ....... . ..

AAIR SEVELACCUMULATION HEREHEMI DO- OUNDED

HERE

FAULTY OE&GN, L IMPROVED DESIGN

-q--, g~w 4g~qMOkZE STRAINER

-w- BW -WTOP VIEW

Faulty and Improved Design of Injection Strainers for Use on S4hips.

WILL RI CEN HENSM O

_Aiwj:CTiON FREE OF AIR O H FALGMN

Heirtio of ilg Kel Cu Awy toPreentMisalignment of Pump Tm-

Air Etrapent.pelierli InCrer An Turbulence.

Figure 11. Water box design.

27

a. Objecttonable. Rlestricted entrance dv,4 to b. Iriepro wd. Air pocket eliminated by stream.-ahallow water box. lined entrance and deepened water box.

Decignie )f Injection Noasle and Water Box.

a. ObjecH010r1010. Sharp angtles at juncotion of nossle b.fInsroved. Gradually increased nozzle areaatand water box. approach to water box.

Itziunz of Entrance to Water Box.

"IERI

0. Objectionc.Nv. Cylinilriril nozzle; abrupt b. Iempvred. Nosale atradually increasing in .increase its areat. area at apprsael t,, water ijx.

Desigsss of Entrance to Waiter Box.

Figure 11. Continued.

28

HEA 0/

PASS

(a) Reverse-Flow Water Box with Reinforcing Plate in Center Causes Turbulenceand Air Separation in Water Entering Second Pass Tubes. (b) Well-Designed Reverse-Flow

Water Box with Test Holes Drilled as Shown.

Injection Nozzle Too Far from Tube Sheet. Air Vents in Division Plate ReduceIthe Harmful Effect.

TOP VIEW

1'URSULENCE HERE

Objectionable Internal Reinforcing Ribs in Inlet Water Box.

Figure 11. Continued.

29

Pressure. Pressure has not been found to have a direct effect onthe corrosion behavior of metals (Ref 7,35). Pressure can, aside frommechanical effects such as stress, indirectly influence corrosion behav-ior. Under high pressure, thermodynamic considerations indicate that alowering of pH will occur, thus affecting both corrosion and the forma-tion of protective carbonate scales (Ref 35, p 6). Pressure shouldreduce tho tendency of bubbles to form that are responsible for impinge-ment attack (Ref 1, p 124).

Temperature. In surface seawater the effect of temperature oncorrosion is generally balanced by two opposing factors. At low temper-atures, the solubility of oxygen in the seawater is increased, and theamount of fouling organisms that accumulate on exposed surfaces is low.In high temperature waters, the oxygen solubility is lower, and thefouling accumulations are normally heavier. Thus, the expected increaseof corrosion rate with temperature is offset by other factors within thenormal range of surface ocean temperatures (Ref 1, p 133).

DESIGN FACTORS INFLUENCING BIOFOULING

There is a natural tendency for any surface exposed to seawater tobecome covered with biological material. The films formed can rangefrom the thin slines and sparse macroorganisms typical of the deep oceanto the heavy accumulation of both hard-shelled and soft-bodied plantsand animals of the surface. Specific equipment in an OTEC plant willvary greatly in its tolerance to fouling organisms. Considerableamounts of heavy fouling can be accommodated on the platform hull,whereas very thin slime films can adversely affect the performance ofheat exchangers. Thus, a wide range of fouling control requirementsmust be applied to the various OTEC system components.

Fouling organisms also have a direct effect on the corrosion ofmany materials. Heavy accumulations of growth on steel can limit theaccess of oxygen to the metal surface and, thereby, reduce corrosion tolow values. These low, consistent values are the result of the sulfate-reducing bacteria in sustaining the corrosion of the steel rather thanthe classical oxygen reduction process (Ref 53). Oxygen and metal ionconcentration cells can also be formed by the attachment of certainmarine organisms, notably barnacles. These cells can adversely affectthe performance of alloys that are susceptible to localized attack,particularly crevice corrosion (Ref 47).

In special cases, mobile organisms, such as sea urchins, can con-tinuously remove protective corrosion product films, thereby producinghigh corrosion rates (Ref 1, p 14). Sulfides produced by the decay oforganisms are a form of natural pollution and should be considered,particularly when alternate filling, stagnation or intermittent anti-fouling treatments are utilized. Organisms, both micro and macro, canalso cause accelerated degradation of protective coatings (Ref 1, p I5).

30

Direct mechanical effects of fouling accumulations must also beconsidered. The weight and bulk of organisms can increase stresses inequipment, alter flow rates, and cause turbulence, all of which canaffect both the short-term mechanical performance and the long-termcorrosion behavior of not only the equipment in which the fouling hasoccurred but also other system components (Ref 1, p 14).

By proper design and fouling control, the accumulation and effectsof fouling organisms can be kept within reasonable limits at minimumexcess cost. Many factors must be ccrnsidered in order to obtain optimum(lowest cost for required resistance) fouling control. As in design forcorrosion control, these factors can be separated into two broad groups:mechanical and environmental.

Mechanical Design Factors

Mechanical design factors that can influence the attachment andgrowth of fouling organisms include geometry, orientation, layout,surface finish, cathodic protection, antifouling coatings, and cleaningsystems. These factors must not only be considered individually, butmust be considered as a synergistic system. Any design is necessarily acompromise. Often factors that decrease fouling will increase corrosionand vice versa. Only by carefully defining required corrosion andfouling resistance and adjusting material selection and design toachieve these requirement~s can an optimum design be formulated.

jGeometry and Orientation. Biofouling will occur on all nontoxicsurfaces exposed to ambient seawater. The type and extent of foulingvaries considerably with both location and season, and it must either beprevented or tolerated. The major effects of geometry and orientationon fouling are in the area of tolerance rather than prevention. Thebasic factors of simplicity, accessibility, and m1aintainability outlinedin the corrosion section also apply here.

The adverse effect of biofouling on system performance should beestimated to determine if fouling control is necessary. Fouling masscan be estimated for both near-shore and open ocean conditions (Ref 54).An additional weight. of 17 kg/sq m (1.7 lb/sq ft) can be expected on acompletely fouled i~nprotectLed surface at warm water sites typical ofthose proposed for OTEC. This is the weight of hard shelled organisms;soft fouling has essentially the same density as seawater. After thesurface becomes completely fouled, a stable community is established,and no significant additional weight accumulates. From the surface to adepth of 5 meters (16 ft), surfaces can be expected to be completelyfouled within 3 years. At a depth of 18 meters (60 ft), surfaces can beexpected to be covered within 12 years. At a depth of 30 meters (100ft), surfaces will be approximately 50% fouled within 12 years. At adepth of 100 meters (330 ft), surfaces can be expected to be only 5%covered within 12 years. The shape of fouling rate curves indicatesthat a maximum fouling population will be achieved after I to 2' veersand will remain fairly stable after that time (Ref 54). The foul~ing

31

Li J.iW4'

- -- --- - -

rate estimations can be used to determine the cleaning interval re-quired. These forecasts are for external surfaces and may not be appli-

L cable to the large internal volumes to be encountered in an OTEC plant.One indirect effect of geometry is that of light on fouling.

Fouling organisms are sensitive to light. Light is necessary for photo-synthesis, and plants will not grow in its absence. Animal species,however, seem to flourish in dark areas. Thus, the maximum amount offouling seems to occur on strongly illuminated or completely dark sur-faces (Ref 55). This effect should be considered as a predictive tool,not a preventive measure (Ref 56).

The increase in resistance to fluid flow due to fouling is also anadverse effect. Data from fouling of ships' hulls and increase fluidflow resistance in small diameter piping can be used only to estimatethese effects on the interior portions of an OTEC plant.

The effect of fouling on heat transfer surfaces is a criticalfactor in the viability of the OTEC concept (Ref 57). The geometry ofthe heat transfer surfaces will have little direct effect on theirpropensity to foul, but it will have a significant effect on the effi-cacy of various antifouling treatments. Geometry can, however, indi-rectly affect the fouling of heat transfer surfaces by influencing otherfactors, such as velocity and turbulence, which will be discussed laterin this paper.

Orientation of a surface has a direct effect on the fouling of thatsurface. In general, the lower side of a horizontal surface fouls moreheavily than the upper side of a horizontal surface, with there being agradation of intensity between these extremes (Ref 55). The variationis small, however, and, taking the vertical orientation as an average,will vary approximately ±10% (Ref 54).

Layout. Like geometry and orientation, layout does not have asignificant direct effect on fouling. The indirect effects of layout onaccessibility, maintainability, etc., which were outlined in the pre-ceding section on corrosion, apply also to fouling.

Surface Finish. There have been some studies made that indicate apolished surface is initially more resistant to fouling than a roughenedone (Ref 58). This effect seems to be short-lived, and the fouling onpolished surfaces approaches that on roughened surfaces within a fewmonths (Ref 58). Considerable information on other surface charac-teristics was located, but it was primarily related to surface chemistryand material properties.

Metals can be grouped into three broad categories of fouling prop-erties. Highly corrodable metals, such as steel, foul readily. Thefouling can, however, be readily removed with the typical loose corro-sion products formed. Passive metals, such as the stainless steels,will also readily foul, but the fouling is tightly adherent. Somemetals, such as the copper alloys, form corrosion-product films (ini thiscase, cuprous oxide) that are toxic to organisms (Ref 59).

32

The fouling resistance of inorganic materials is related primarilyto the initial adhesion of a proteinaceous "conditioning film," whichserves as a substrate for the attachment of subsequent fouling communi-ties (Ref 60). Surface tension appears to have primary influence on theadhesion of these films. Surface tension in the range of 20 to 30dynes/cm appears to result in a minimum adhesion, although a totallyfouling free surface has not been achieved through application of thisphenomenon (Ref 60).

Cathodic Protection. Cathodic protection, or rather the impositionof electrical currents, has been widely studied as an antifouling mea-sure (Ref 61-63). Except for the in-situ generation of chlorine gas(Ref 62) these methods are considered either ineffective or impracticalfor large surfaces, primarily due to the high current densities requiredfor protection (Ref 64,55). The dissolution of heavy metals from coat-ings or alloy matrices by anodic currents and the periodic formation andremoval of calcareous coatings by periodically reverse1 currents may,however, be applicable to certain critical systems where no other meansof fouling control is practical. Some studies indicoite that steelprotected with low current densities (1 to 30 mA/sq m) may be subject toheavier fouling than unprotected surfaces. With currents above 30 mA/sqm, fouling was reduced (Ref 55).

Antifouling Coatings. Antifouling coatings that contain toxicmaterials can be used to prevent fouling. However, these coatings havelimited life, because the toxic materials must leach from the coatingsin order to provide fouling resistance. Unless the system can be de-watered and recoated, antifouling coatings should be considered as atemporary measure. If the equipment was properly designed to allow forsuch dewatering and recoating, then such coatings could be practical insome instances. The selection of these coatings for specific applica-tions is outside the scope of this report. Certain precautions must betaken, however, in the selection of antifouling coatings in order toprevent adverse corrosion effects. The toxic products from antifoulingcoatings containing heavy metals, such as copper and mercury, can causerapid corrosion of some metals, particularly aluminum alloys. Thesecoatings should not be applied directly to aluminum surfaces, nor shouldseawater flowing past them be allowed to contact aluminum surfaces(Ref 1, p 18).

Radioactive coatings can prevent fouling, but, for the coatingsdeveloped to date, the levels of activity required present serioussafety and environmental problems (Ref 63).

Cleanin Shystems. Fouling can be removed from any surface bymechanical cleaning. Initial proteinaceous and bacterial films can beremoved by simply wiping or brushing, whereas hard-shelled organismsrequire vigorous scraping to dislodge them from many surfaces. Thedesign of a system will be a major factor in the efficacy of any clean-ing system. Smooth and simple surface contours lend themselves to easycleaning, whereas complex contours with re-entrant angles, etc., are

33

j .• , r

more difficult to clean. For example, the interior of heat exchangertubes are relatively simple to clean whereas their exterior surfaces,particularly when they are closely spaced, present a much more difficultsituation.

Environmental Design Factors

By properly controlling or modifying the chemical or physicalcharacter of seawater, the ability of fouling organisms to live and growcan be limited or totally inhibited. Any modification of the largevolumes of seawater required for the operation of an OTEC plant willrequire large inputs of chemicals or energy. Therefore, the use ofenvironmental modification as an antifouling method should be carefullyanalyzed. Certain environmental factors will be modified by the normaloperation of the plant and, where possible, the effect of these modifi-cations on fouling should be used to the maximum overall benefit.

Chemical. Introduction of toxic materials into the environment cankill marine organisms. Chlorine is the most widely used chemical agentfor this purpose. Chlorine may be added as a gas or as a hypochlorite(Ref 65). It can also be generated in situ by electrolysis of theseawater itself (Ref 66). Chlorine residualis on the order of 0.25 ppmapplied continuously will provide complete control in most instanices.Chlorine additions must, of course, be in excess of these residualrequirements in order to allow for reaction of the chlorine with theseawater (Ref 65). The amount of chlorine required will vary consider-ably with local conditions (Ref 65). This may be particularly true fordeep-ocean waters where the fouling population is limited. The inter-mittent addition of chlorine can also eliminate fouling (Ref 67, p 359).The actual amount and optimum addition cycle of chlorine for inhibitingfouling in the various portions of an OTEC system can be expected to besignificantly different from the dosages normally used in coastalplants. Specific recommendations as to the feasibility of chlorinationas the method for inhibiting fouling of an OTEC plant cannot be madeuntil these data are obtained.

Velocity. Relative velocity between seawater and a surface caninhibit or prevent the attachment of fouling organisms (Ref 1, p 132).The velocity of this flow is the major antifouling factor that can bedirectly controlled by system design. While the critical velocityrequired to inhibit or prevent fouling has not been well defined, itprobably lies in the 2-to-4-fps range (Ref 63). The use of velocity tocontrol fouling depends on continuous flow. If the flow is interrupted,even for short periods, fouling organisms will attach and continue togrow under subsequent high velocity conditions (Ref 65, p 39). Thus, inorder to be practically applicable to OTEC systems, the use of velocityto control fouling probably will need to be combined with some otherantifouling method in order to accommodate periods of low flow.

Design also affects the uniformity of velocity and, thus, theefficacy of velocity in inhibiting fouling. Where the flow is notsmooth and evenly distributed, areas of low velocity will occur and, if

34

.......................................................,....

below the critical velocity, fouling will occur. Fouling downstream ofobstructions, such as ells, screens, etc., is coimon even in systemsoperating at high velocity. It is suggested that an approach analogousto the surface fluid shear concept for velocity effects on corrosion(Ref 47) may be applicable to the study of the effects of velocity onfouling.

Temperature. Temperature has a significant effect on foulingorganisms and, thus, on fouling rates. In warm, tropical waters thefouling rates are typically high, whereas they are typically lower inLHtemperate waters. A rapid temperature change on the order of 20 to 401Ccan inhibit or prevent fouling (Ref 63,68,69). However, providing forsuch temperature changes in the volumes of water involved in the 3pera-tion of an OTEC plant presents many engineering and economic difficul-ties. It is possible that the effect of heating may be significantlygreater on deep ocean organisms than on common surface organisms(Ref 70, p 28).

Pressure. The effect of pressure, or rather changes of pressure,on fouling organisms is not well documented. As OTEC will bring largequantities of deep water to the surface and utilize this water in equip-inent, such as heat exchangers, that have critical fouling requirements,this effect must be studied further. It is generally assumed thatfouling will be minimal in the deep water, even when brought to thesurface. The heavy accumulations of typical hard fouling found in heavyI surface exposures do not generally occur in deep waters, but the marineorganisms capable of forming slime films and heavier accumulations ofanimal fouling are known to occur at all depths proposed for OTEC plantoperations (Ref 71). The effect of these organisms on OTEC plant con-densers should not be neglected, unless valid test data can be obtainedthat show the fouling rates are within acceptable limits.

S~SUMMARY

SU:lMany factors that influence the biofouling and corrosion of OTECsystem surfaces can be manipulated by system design. In general, simpledesigns are preferable to complex designs; maintainability and accessi-bility are key factors; and, as for any engineering endeavor, the finaldesign will be a compromise between many factors. In some instances,corrosion and biofouling will be of little concern, while the influenceof other factors will predominate. In other instances, corrosion andbiofouling will be critical factors and will predominate over otherconcerns. Careful attention to details of design and application of thebasic design principles outlined in this report is necessary to controlbiofouling and corrosion of an OTEC plant. Only by a careftl balance ofthese many factors through careful engineering judgment can an optimumOTEC plant be placed into successful operation.

35 .I

RECOMMENDATIONS

Because of the unique nature and scale of the OTEC concept, muchextrapolation of presently available experience and experimental datamust be made to obtain a first approximation of the biofouling and

corrosion behavior that may occur in a full-scale OTEC plant. Suchtopics as the rate of biofouling of heat exchanger tubes in the openocean using both near-surface and deep-ocean water, the precipitation ofmaterials from deep water when brought to the surface, the effect ofenvironmental gradients on structural attack, the effect of velocity onthe corrosion of very large components, as well as many other OTECspecific problems must be studied under actual OTEC conditions in orderto refine the atility to predict and control biofouling and corrosion inan OTEC plant.

REFERENCES

1. La Que, F L. Marine Corrosion, Causes and Prevention. John Wileyand Sons, 1975.

2. Perrigo, L. D. "Fundamrntals of corrosion control design," Proceed-ings of the Ninth Western States Corrosion Seminar, National Associationof Corrosion Engineers, Houston, Tex, May 1975.

3. Shrier, L. L. Corrosion. John Wiley and Sons, 1963.

4. Sebald, J. F. "Control of tube vibration in steam surface conden-sers," P'roceedings of Seminar on Performance of Surface Condensers inNuclear and Fossil Power Plants, Ohio State University, Columbus, Ohio,2-4 June 1975.

5. Bostwick, T. W. "Reducing corrosion of power plant condenser tubingwith ferrous sulphate," Corrosion, vol 17, no. 8, Aug 1961.

6. Carson, J. A. H. "An evaluation of an electrolytic process for thereduction of corrosion in a steel sea water piping system," Proceedingsof the 1st International Congress on Metallic Corrosion, London, 10-15April 1961.

7. Civil Engineering Laboratory. Technical Report R834: Corrosion ofmetals and alloys in the deep ocean, by F. H. Reinhart. Port Hueneme,C3lif., Feb 1976.

8. Brown, B. F. "Corrosion itt the marine environment," Marine Technol-ogy 1970; preprints of 6th annual conference and exposition, June 29-July 1, 1970, Washington, D.C., Marine Technology Society, vol 2.

9. Uhlig, H. H. "How to make use of corrosion data," Proceedings ofthe Third International Congress on Marine Corrosion and Fouling,October 2-6, 1972, National Bureau of Standards, Gaithersburg, Md.

36 J•

10. U.S. Naval Research Laboratory. NRL Memorandum Report 1634:Marine corrosion studies; stress-corrosion cracking, deep ocean technol-ogy, cathodic protection, corrosion fatigue; third interim report ofprogress, by B. F. Brown et al. Washington, D.C., July 1975.

11. Zapffe, C. A. Stainless Steels, American Society for Metals.Cleveland, Ohio, 1949.

12. Peters, B. F. "Stress corrosion cracking," Materials Protection,vol 4, no. 5, May 1965.

13. Naval Research Laboratory. NRL Report 6564: Stress-corrosioncracking characteristics of alloys of titanium in splt water, by R. J.Goode et al. Washington, D.C., Jul 1967.

14. Judy, R. W., Jr., and R. J. Goode. "Crack growth in heavy sectiontitanium," Stress Corrosion - New Approaches, ASTM STP 610, AmericanSociety for Testing and Materials, 1976.

15. The American University. "Stress corrosion cracking control mea-sures, Pt. 5. Titanium alloys," by B. F. Brown Washington, D.C., Jun1•75.

16. U.S. Naval Research Laboratory. NRL Memorandum Report 1574:Marine corrosion studies; deep ocean technology, stress-coirosion crack-ing, cathodic protection; second interim report of progress, by B. F.Brown et al. Washington, D C., Nov 1964.

17. NRL Memorandum Report 1711: Marine corrosion studies;stress corrosion cracking, deep ocean technology, cathodic protection,corrosion fatigue; fourth interim report of progress, by T. J. Lennox etal, Washington, D.C., May 1966.

18. Goddard, H. P., et al. The Corrosiin of Light Metals. John Wileyand Sons, Inc., New York, 1967.

19. The American University, "Stress corrosion cracking control mea-sures, Pt. 4. Aluminum alloys," by B. F. Brown. Washington, D.C., Jun1975.

20. Leidheiser, H., Jr. The Corrosion of Copper, Tin, and Their

Alloys. John Wiley and Sons, Inc., New York, 1971.

21. Uhlig, H. H. , ed. Corrosion Handbook. John Wiley and Sons, Inc.,New York, 1948.

22. Rensselaer Polytechnic Institute. "Corrosion fatigue of metals andalloys," by D. J. Duquette. Troy, N.Y., May 1972.

37

IY

i •.

23. Groover, R. D., and M. H. Peterson. "~Cathodic protection of inter-nal surfaces of pipes containing seawater," Materials Performance, vol13, no. 11, Nov 1974.

24. Compton, K. G. "Cathodic protection of structures in sea water,"Paper no. 13, preprints of Corrosion '75, April 14-18, 1975, Toronto,Canada. National Association of Corrosion Engineers, Houston, Tex.

25. Compton, K. G., et al. "Considerations of importance in the cath-odic pr'rcection of marine structures," Paper no. 85, pr-eprints of Corro-sion '74, March 4-8, 1974, Chicago, Illinois. National Association ofCorrosion Engineers, Houston, Tex.

26. Morgan, 3. H. , and J. A. Lehmiann. "Cathodic protection of jettiesand offshore structures: American and European practices," Paper no.112, preprints of Corrosion '74, March 4-8, 1974, Chicago, Illinois.

F National Association of Corrosion Engir~eers, Houston, Tex.

27. Davis, J. G., et al. "New aspects of the application of cathodicprotection to offshore facilities," Paper no. 15, preprints of Corrosion'75, April 14-18, 1975, Toronto, Canada. National Association of Corro-sion Engineers, Houston, Tex.

28. Brown, B. F. "Corrosion standards and control in the marine indus-try," Manual of Industrial Corrosion Standards and Contriol, ASTM SpecialTechnical Publication 534, F. H. Cocks, ed. American Society for Test-ing and Materials, Philadelphia, Penn., 1973.

29. Naval Research Laboratory. NRL Memorandum Report 1948: Marinecorrosion studies: The corrosion characteristics and response to cath-odic protection of several stainless steel alloys in seawater; with apartially annotated bibliography, sixth interim report of progress, by

T. J. Lennox et al. Washington, D.C., Nov 1968.

30. .NRL Memorandum Report 2374: Marine corrosion studies:Cathodic protection of 304 stainless steel and copper crevices in saltsolutions, thirteenth interim report of progress, by T. 3. Lennox and M.

H. Peterson. Washington, D.C., Oct 1971.A 31. .NRL Memorandum Report 2348: Marine corrosion studies:Corrosion characteristics of a Ni-Cr-Mo alloy in quiescent seawater andits response to cathodic protection, twelfth interim report of progress,by T. J. Lennox et al. Washington, D.C., Aug 1971.

32. Morgan, John H. Cathodic Protection, Leonard Hill [Books] Ltd.,London, 1959.

33. Compton, K(. G. "The unique environment of sea water and its effecton the corrosion of metals," Corrosion, vol 26, no. 10, Oct 1970.

38

34. Di Gregoria, J. S., and J. P. Fraser. "Corrosion tests in the Gulf"floor," Corrosion in Natural Environments, ASTM STP 558, AmericanSociety for Testing and Materials, 1974.

35. Fink, F. W., and W. K. Boyd. The Corrosion of Metals in MarineEnvironments, Bayer and Company, Inc., 1970. (DMIC Report 245)

36. Gudas, J. P., et al. "Accelerated corrosion of copper-nickelalloys in polluted waters," Paper No. 76, preprints of Corrosion '76,March 22-26, 1976, Houston, Texas. National Association of CorrosionEngineers, Houston, Tex.

37. University of Delaware. "Effects of dissolved oxygen and heavymetal contamination of seawater on pitting corrosion of aluminum alloysfor seawater heat exchanger and condenser tubing, Progress report forperiod 15 June 1976 - 15 March 1977," by S. C. Dexter. Lewes, Del., Mar1977.

38. U.S. Naval Research di-oratory. NRL Memorandum Report 1282:Abyssal corrosion and its ,iitigation, Part 1. Details of pilot testexposure, by L. J. Waldron et al., Washington, D.C., Mar 1962.

39. U.S. Navy Marine Engineering Laboratory. MEL Report 429/66: Metalcorrosion in deep ocean environments, by W. L. Wheatfall. Annapolis,Md, Jan 1967.

40. U.S. Naval Applied Science Laboratory. Lab Project 930-6, ProgressReport No. 1: Installation of NASL deep sea materials exposure arraysin the Tongue of the Ocean, by A. Macander and C. D. Francy. Brooklyn,N.Y., Jul 1967.

41. . Lab Project 930-6, Progress Report No. 2: Performanceof a NASL deep sea array after one year exposure i, the Tongue of theOcean, by A. Macander. Brooklyn, N.Y., Dec 1968.

42. U.S. Naval Underwater Ordnance Station. Technical Memorandum TMNo. 359: Sample tests exposures to examine corrosion and foulingeffects on materials in the deep ocean," by S. Milligan. Newport, R.I.,Jan 1966.

43. WJoods Hole Oceanographic Institution. Technical Memorandum WHO1-4-72: Handbook of oceanographic engineering materials, Volume I. Metalsand alloys, by S. C. Dexter. Woods Hole, Mass., Dec 1972.

44. U.S. Naval Research Laboratory. NRL KRport 7447: Behavior ofmaterials in a subsurface ocean environment, by L. A. Beaubien et al.Washington, D.C., Jul 1972.

45. University of Miami. "The saturation state of seawater with re-spect to calcium carbonate and its possible significance for scaleformation on OTEC heat exchangers," by J. W. Morse et al. Iliami, Fla.

39

Sc

46. Copson, H. R. "Effects of velocity on corrosion by water," Indus-trial and Engineering Chemistry, vol 44, Aug 1952.

47. Efird, K. D. "Effect of fluid dynamics on the corrosion of copper-base alloys in sea water," Corrosion, vol 33, no. 1, Jan 1977.

48. Perry, R. H., and C. H. Chilton. Chemical Engineers Handbook, 5thed, McGraw-Hill Book Co., New York, 1973.

49. Niederberger, R. B., et al. "Corrosion of nickel alloys in quietand low velocity sea water," Materials Protection and Performance, vol9, no. 8, Aug 1970.

50. Fleetwood, M. J. "Non-ferrous alloys for marine applications,"Corrosion Prevention and Control, vol 22, no. 2, Apr 1975.

51. Huston, K. M., and R. B. Teel. "Some observations of the poten-tials of stainless steels in flowing sea water," Corrosion, vol 8, no. 7Jul 1952.

52. Beck, T. R. "Effects of hydrodynamics on pitting," Paper No. 77,preprints of Corrosion '76, March 22-26, 1976, Houston, Texas. NationalAssociation of Corrosion Engineers, Houston, Tex.

53. U.S. Naval Research Laboratory. NRL Report 7672: Influence of

marine organisms on the life of structural steels in seawater by C. R.Southwell and J. D. Bultman. Washington, D.C., Mar 1974.

54. De Palma, J. R. "Fearless fouling forecasts," Proceedings of theThird International Congress on Marine Corrosion and Fouling, October2-6, 1972. National Bureau of Standards, Gaithersburg, Md.

55. Zachev, B. "The fouling of ships hulls and marine installations,"translated by Defence Research Information Centre, Orpington (England),Mar 1974 (DRIC-TRANS-3524).

56. Saroyan, J. R. "Antifouling paints - The fouling problem," Naval

Engineers Journal, vol 80, no. 4, 1968.

57. Allied Chemical Corp. "Prevention of biofouling on heat transfersurfaces of ocean thermal energy converters, Progress report," by R. L.Ostrozynski and P. E. Jones. Buffalo, N.Y., Dec 1975.

58. David Taylor Model Basin. Report No. 396: Fouling of ships bot-toms; Effect of physical character of surface, by A. S. Pitre et al.Washington, D.C., Apr 1935.

59. Efird, K. D. "The inter-relationship of corrosion and fouling formetals in seawater," Materials Performance, vol 15., no. 4, Apr 1976.

40

-• -z.-.

L

60. Baier, R. E. "Influence of the initial surface condition of mate-rials on bioadhesion," Proceedings of the Third International Congresson Marine Corrosion and Fouling, October 2-6, 1972. National Bureau ofStandards, Gaithersburg, Md.

61. Heathfield, P. E. H. "An electrolytic system for controllingcorrosion and marine growth," Underwater Science and Technology Journal,

V vol 2, no. 3, Sep 1970.

62. U.S. Patent No. 3,984,302 Oct 5, 1976. "Apparatus for controllingmarine fouling of salt water coolant heat exchangers, piping systems andthe like."

63. Benson, P. H., et al. "Marine fouling and its prevention," MarineTechnology, Jan 1973.

64. Littauer, E. L., and E. Jennings. "Prevention of marine fouling byelectrical currents," Proceedings of the Second International Congresson Marine Corrosion and Fouling, Athens, Greece. Sep 1968.

65. Office of Saline Water. INT-OSW RDPR-73-858: Methods for control-"ling marine fouling in intake systems, by D. C. Mangum et al.Washington, D.C., Jun 1973.

66. Lovegrove, T., and T. W. Robinson. "The prevention of fouling bylocalized chlorine generation," Biodeterioration of Materials, A. H.Walters and J. J. Elphich, ed. Elsevier Publ. Co. Ltd., London, 1968.

67. Mangum, D. C., et al. "Methods of controlling marine fouling indesalination plants," Proceedings of the Third International Congress on

Marine Corrosion and Fouling, October 2-6, 1972. National Bureau ofStandards, Gaithersburg, Md.

68. Graham, J. W., et al. "Heat treatment for the control of marinefouling at coastal electric generating stations," Ocean 75; ProceedingsS~of the combined meeting of 1975 IEEE conference on Engineering in the

Ocean Environment and 11th Annual Meeting of the Marine TechnologySociety, San Diego, California, Sep 22-25, 1975. IEEE, New York, 1975.

69. U.S. Naval Engineering Experiment Station, E.E.S. Report4F.(FI)101717: Control of marine fouling in piping system by treatmentwith I-lot water, Summary of progress reports, by William F. ClappLaboratories (Duxbury, Mass.). Annapolis, Md., Jan 1952.

70. Department of the Navy, Bureau of Ships. Technical Report No.1240762: Marine corrosion and fouling, by Arthur D. Little, Inc.

k (Cambridge, Mass.), Jul 1962.

71. U.S. Naval Institute. "Marine fouling 3nd its prevention," byWoods Hole Oceanographic Institute. Annapolis, Md., 1952.

41

Appendix

FORMS OF CORROSION ATTACK

For all the combinations of metals, ocean environments, and struc-tural applications, only a few forms of corrosive attack occur. Severalcorrosion mechanisms may result in the same form of attack in varioussituations. The description of these forms of attack given below areintended to provide the design engineer with a mental picture so thatthe effect of the attack on structural integrity or equipment operationcan be better evaluated.

1. No Corrosion. While, strictly speaking, this is not a form ofattack, it- is a form of material behavior exhibited by a few alloys. Inthis form of attack there is essentially no interaction with the marineenvironment and no change in weight, strength, ductility or other prop-erty that can be attributed to the marine exposure. Typical materialsthat undergo no corrosion when exposed to seawater are: platinum, gold,titanium (selected alloys), nickel alloys (N06625* and N10002*), cou~altalloy (R30035-), and the complex stainless steel Al6X.

2. Uniform Corrosion. When a metal corrodes at substantially thesame rate over its entire exposed surface, uniform corrosion is said tohave occurred. As for all natural phenomena, the attack is not strictlyuniform. Surface roughening always occurs. The amount and distributionof this roughening vary considerably and can grade into pitting corro-sion. Weight loss is a valid measurement of uniform corrosion.. Typicalmaterials that are subject to uniform corrosion in seawater are carbonsteels and most alloy steels (when mill scale has been removed), copper,beryllium copper (C17200), and 90-10 copper-nickel (C70600).

3. Pitting Corrosion. Pitting is a form of localized attack wherethe corrosion rate is substantially greater on some areas of a metalsurface than on others. The result of pitting attack can range frombroad, shallow cratering to narrow, deep pits resembling drilled holes.The surfaces surrounding the pits can remain virtually unattacked.Weight loss cannot be used to evaluate pitting corrosion. Pit depth andfrequency measurements, tensile tests, and visual observations arenormally used to evaluate pitting attack. The gradation of uniformcorrosion into the broad, shallow cratering form of pitting attack canbe subtle. The two forms of attack can be differentiated by use of thepitting factor (P.F.) where

P.F. - Depth of Deepest PitAverage Depth of Attack

Numbers refer to assignments in "Unified Numbering System for Metalsand Alloys," Society of Automotive Engineers, SAEJ 1086, latest issue.

42

When the pitting factor is less than two, uniform corrosion is said tohave occurred. When the pitting factor is great-,r than two then pittingcorrosion is said to have occurred. Typical materials that are subjectto pitting attack are the aluminum alloys and stainless steels, exceptthose containing appropriate amounts of ,nickel and molybdenum such asthe Al6X stainless steel proposed for OTEC applications.

4. Galvanic Corrosion. Galvanic corrosion is attack acceleratedby electrical coupling of two or more dissimilar metals. Galvanicattack can occur rapidly and can result in either uniform or localizedattack. Not only the composition of the metals involved, but also theirrelative exposed surface areas, distance apart, geometric distribution,and the resistance of the electrical contact, determine the rate andtype of attack that will occur. Galvanic corrosion tendencies in sea-water can be predicted. In most cases, however, quantitative galvaniccorrosion rates cannot be predicted. Galvanic corrosion should, there-fore, be avoided wherever possible by electrically isolating dissimilarmetals. Although nearly any metal can be subject to galvanic corrosion,the aluminum alloys, carbon, and alloy steels are most prone to thistype of attack due to their relatively high activity in seawater.

5. Concentration Cell (Crevice) Corrosion. Concentration cellcorrosion is attack caused by a difference in environment between dif-ferent points on a single metal. While differences in temperature,light, etc. , can cause concentration cell corrosion in special circum-stances, the most common cause of concentration cell corrosion in marineenvironme.nts is the differential environment caused by the trapping ofseawater in small gaps or crevices. When the seawater is trapped andId in a crevice, its composition will change considerably from that of

_aie bulk seawater outside the crevice. For some materials, such as thealuminum alloys and stainless steels, the composition change of signifi-cance is the consumption of dissolved oxygen within the crevice. This,combined with changes in acidity within the crevice, results in acceler-ated attack within the crevice. This form of attack is insidious,because it can occur rapidly and is often revealed only by equipmentfailure.

A second form of crevice attack is commonly found to occur on thecopper alloys. In this form of attack, the buildup of metal ions withinthe cre,ý`-e causes the area inside the crevice to become less active

unth-, . ea just outside the crevice. Accelerated attack occurs justoutsi~jc the crevice. Weight loss cannot be used to evaluate this formof localized attack. Depth and frequency of attack are often used toevaluate crevice corrosion.

6. P Il~oying. Dealloying is the selective attack of one constit-uent of a ilby. As alloys, by definition, consist of solutions or1'mixtures more than one element, it is possible that, in some in-stances, one or more constituent of the alloy will be sufficiently moreactive than the bulk alloy to be subject to selective attack. In mostcases the residual material is porous and spongy and has little mechani-cal strength. Typical materials subject to this form of attack in

43

.......................................... * it,

seawat~er are the cast irons, yellow brass (C27000O), and aluminum bronze(C95400). Weight loss cannot be used to properly evaluate this form ofattack. Penetration of the attack into the metal as measured on a crosssection is often the method of evaluation.

7. Intergranular Corrosion. Intergranular corrosion is the inter-nal attack of a metal at or adjacent to the grain boundaries. Allengineering materials are agglomerations of individual crystals orgrains. The areas where these grains meet are called grain boundaries.In some metals, such as some aluminum alloys, the grain boundariesthemselves are more active than the bulk material. In other alloys,such as the stainless steels, improper heat treatment (sensitization)can lead to rapid attack of the material adjacent to the grain bounda-ries. In both cases, the structural integrity of the alloy is essen-tially destroyed in the affected area with a very small amount of actualmaterial loss. Weight loss cannot be used to evaluate this form ofattack. Microscopic examination and tensile testing are the most com-monly used methods of evaluation.

8. Stress Corrosion Cracking. Stress corrosion cracking resultsin the failure of a component in a shorter time at a given load or alower '.oad in a given time when exposed to static stress in a particularenvironment as compared to failure loads or times in air. Stress corro-sion cracking can occur rapidly and often occurs due to residual fabri-cation stresses rather than service loading. High strength materialsare most susceptible to this form of attack. Typical materials subjectto this form of attack in seawater are some aluminum alloys, some tita-nium alloys, and some precipitation-hardening stainless steels. Fail-ures of structures by stress corrosion cracking can be predicted byfracture mechanics techniques when, by experiment, critical stressintensity factors for the specific material,; have been determined.

9. Corrosion Fatigue. Corrosion fatigue is the dynamic analogy ofstress corrosion cracking. In this form of attack failure occurs at alower stress level or in a fewer number of cycles when exposed to cyclicloading in an aggressivp environment than when exposed to cyclic loadingin air. Although nearly every material will show some effects of corro-

L sion fatigue, the most significant reductions are noted on materials

such as mild steel, ultra high strength steels, and some copper alloys.Reductions to as little as 5 to 10% of the in-air fatigue strength arenot uncommon.

10. Hydrogen Embrittlement. Hydrogen embrittlemerit is similar tostress corrosion cracking in its effect on the load-bearing capacity ofa material. In some cases, these two forms of attack are identical.The source of hydrogen can be from initial manufacture, electroplating,cathodic protection, the presence of sul fate- reducing bacteria, or self

F corrosion of the material itself. Ultra high strength steels are mos~tsusceptible to this form of attack as are many other materials, such assome stainless steels.

44

.ism

11. Erosion Corrosion. Erosion corrosion is the attack of a metalcaused by the rapid flow of a fluid past or the impingement of a fluidon the metal surface. Steels, copper alloys, and aluminum alloys aremost subject to this form of attack in seawater.

12. Cavitation Corrosion. Cavitation corrosion is the result ofthe impingement and collapse of vacuum bubbles on a metal surface. Thebubbles are normally formed in areas of low pressure in a high velocitystream, such as are commonly found downstream of obstructions to thefluid flow. Cavitation corrosion is particularly severe on copperalloys, steels, and aluminum alloys and should be avoided by properhydraulic design.

13. Fretting Corrosion. Fretting corrosion is attack acceleratedby relative motion between two surfaces in close contact. The attackmay be due to the repeated rupture of inherently protective corrosionproduct films or may be due to direct abrasion effects with the corro-sion products serving as an abrasive. Fretting corrosion can result inrapid attack and should be avoided by design such that relative motionbetween faying surfaces is avoided. Fretting corrosion of aluminum heatexchanger tubes at tube supports is an example of this type of attack.Welding of joints is an effective method of preventing fretting corro-sion.

45

DISTRIBUTION LS

Al ENVIRON. HE 'ALTH LAB NMcClellan AEB CAAFB (AFI'l'IDl). Wright-Patterson OH: ABG/I)EE (F. Nethers). (ioodfellos- AFBT'X: AFCEiC/XR,l'yndall Ii.:

CEISCH. Wright-Patterson: HQ T'actical Air Cmd (R. E. Fisher), lhangky AFB VA; MAC/lET (Col. P.T'hompson)Scott. Il.: SAMSO/MNNF. Norton AFB CA: Stinfo L~ibrary, Offuti NE

ARCTICSUBI.AB ('ode 5s1'. S,-n D~iego. CAARMY BNIISC-RE (H. McClellan) Huntsville At.: DAEN-CWE-M 0tilCDI Binning). Washington IDC: DAEN-FEU1.

Washingtoni DC: I)AEN-FEU-Ei 0. Ronan). Washington DC: l)AEN-MCE-l) Washington DC: ERAI)C)M TechStipp Dir. (I)E.I.Sl)-la Ft. Monmouth. NJ: HQ-l)AEN-FEB-P (Mr. Price): Natick LabIvratories tK~wh Hu) NatickMA: 'Tech. Red. Div., Fort Huachuca. AZ

ARMY('R.lbayCh pigI.ARMY CO)ASTAL. E NCR RSCH CEN Fort Belvoir VA: R. Jachowski, Fort Belvoir VAARMY CORIPS 01: ENGINEERS MRD-Eng. D~iv., Omnaha NE: Seattle D~ist. L~ibrary,. Seattle WAARMY CRREI. A. Kovacs. Hanoiver NH: Constr. E~ngr Re% Branch. tAamnott: G. Phetteplace Hanover, NitARMY CRREI. R.A. EatonARKN'Y lARC()NI ANICPMi-(S tJ. Carrt. Alevandria VAARiY EN(i DIV HNI)EI-CS. Huntsville Al.: Hinded-Sr. Huntsville. Al.ARMY E-'NG WAIERW\AYS EXP STA Library. Vicksburg NISARKN'Y ENGR DI1ST. librar%. Portland 0RARMYIN' FVIRON. HYGIENE: AGCY Water Qual D~i\ t loner). Aberdeen Proy Ground. MD)ARKN'Y NIArIRiAI.S & MECHANICS RFSE.ARCH (T'ENTER D~r. l enoe. Watertosun NIAARNMY NIISSILI.ER&I) (Nil Redstone Arsenal Al. Sci. Info. ('en kl~ocurnents)ARNMY NIOBII. EQUIPI~R&I)COM M~r. Cevasco. Fort Belvoir Nil)ARNIY-PI.ASII:C Picaiinn\ Arenia: (A Ni Anzalone. SNiLTA-1R9-M-Di lDover NJASO) 11W) WNS J.A. Jenkinis). Philadelphia. PAASSI SECRH ARY FTH.'I i: NAVY Spec. A,%ist EiiergY (1P. \\oatemian). Wkashington DC: Spec. Assist Submarines.

'A ashmgtoti IXCIlLREAU OF '(iM NERCIAI FISH FRI ES \ktltods Hole MiA t Biological l-ab. L ib.)IlLREAL' A) RE'L AMiAT10 N (Code 1512 t(C. Selatider) D~enver ('0('IN( 'ANIl Ci'iti Engr. Stipp. Plans. Ofr Norfolk. VA('IN(PA( i~c Fngrng Di\ 4.1440 Ntakilpa. HI('NAVRI.S ('tite 13 IDir. IFacilitic%) Noss (;reans. LA'Ni ('ode NIX 1-OKI 3. Washington. DC:, NMiAT (11(246 tIDieterlet WAash. IDC

('NO ('ode N011-9W. \kashingtovn DC: ('ode OP 323. Washington D)C: ('ode Oil 987 Washington D)C: ('ode 0P-413'Aash. iX ('ode O0PNAV 0t9H24 (H): ('oe OPNAV 22. \\;ish IX': ('ode OPNAV 23. Wash D)('; OPJ87J1 (J.Hoosnma ,t Pentagon

('1)NiBP(' ipraion Of. Nlakalapa HII('1INIII EA I ( KlNW ('mmader Kden I kitas.l:PW().Kadena. (kinasa

(ONNAYViARIANAS ('ode N4. Guam(ONI (WEANS) SPAC(' C.. Plearl Harlitr HI4('ItMtSVI' *V'iRt' N F Opecrations If~r. San Diego. ('ADFENSE IX KINIEN Al I 10N CTR Aleitandria. VADEFE1-IN SE INTEL -'.1.I( iENCEI AC ;FN(Y D~ir.. Woshington DCDE P1 . (W' EN R( iY I .IDivonec. Wash IX'

IXW DElr. C'ohen: D~r. Vanderrvn. Washington. DC: F. F. Parr\ . Wash ington DU(: L:('N t' F UFIt Washington IX': P.

IYI'NSRIX'('ode 1706, Bethesda Nil: ('ode 172 t(i Kren.,ke). Bethesda Nil)11)TN SR IX'C('ode 284 (A. Rufolot. Annapolis Nil)l)TNSRIX'Codte 4111 iR. (iierichi. Bethesda Nil)I1WNSRIX'0('ode 12I (R. Rivers), Annapoili. Nil)lDlNSRIX'G('oe 42. Bethesdta Nil)Il)'N SR IX'('ode 522 (Library). Annapolis Nil)ENFERGY R&l ADMNIN. INEl. I ech. ILib. tReports ectiont. Idaho Falls ID): I it lick. Richmond. 'AA'ENVIRONNMENTAL. PROTFCTI0N AGEIN('Y' Reg. VIII. HNI-ASI.. Ioeniter uo(

FI.T'O~BAT''RAFNL.NT WO.Virginia Hch VA

46

FMFI.ANT 'CPU Offti. Norfolk VA0SA F.ed. Slip. Serv. (FMHP), Washington DC: Office of const. Mgnit (M. Whitley). Washington DCr ~HPISUPPAC'I'PWO. Taipei. TaiwanKWAJALEIN MISRAN BMI)SC-RKI.-CMARINE CORPS HASP Camp Pendleton CA 92055: Code 43-260. Camp Lejeune NC: M & R IDivision, Camp Lejeiene

L NC: PWO. Camp S. 1). Butler, Kassasaki JapanMARINE CORPS DISTU9. Code 043, Overland Park KS

[ ~MARINE CORPS HQS Code L.FF-2. Washington DCMCAS Facil. Engr. D~iv. Cherry Point NC; CO, Kaneohe Bay HI: Code PWE. Kaneohe Bay HI: Code S4, Quantico

* ~VA: J. Taylor. Iwakuni Japan: PWD. Dir. Maint. Control Div., Iwakuni Japan, PW() Kaneohe Bay HI: [IWOUtilities (Paro), Iwakuni, Japan: PWO. Yuma AZ: SCE, Futema Japan: UTC lDupalo, Issakoni, Japan

MCIPC NSAP REP, Quantico VA: P&S lDiv Quantico VAMCI SBPAC B520, Barstow CA:. PWO. Barstow CAMCRID PWO. San Diego CaNAI) Engr. Dir, Hdwthorne. NVNAF PW() Sigonella Sicily: PWO. Atsugi JapainNAS Asst C/S ('P Corpus Chrksti,'I"X: CO, Gjuantanamo Blay Cuba: Code 114, Alameda CA: Code 183 jFac. Plan HR

MGR): Code 187MX. Hrunsssick ME: Code 18U (ENS P.J. Hickey), Corpus Christi TX:, Code 6234 ~(i. rask).Point Mugu CA: (Code 70). Atlanta, Marietta GA: Code 8E. Patuxent Riv.. MD. D~ir. Maint. Control IDiv., Key WestFL.: Dir. Util. IDiv.. Bermuda: ENS Buchholz, Pensah.ola, FL.: Lakehurst, NJ: Lead. Chief. Petty Offr. PW/SelfHelp Div. Heeville' TX: OW. CUH 417. Oak Harbor WA: PW (J. Maguire), Corpus Christi TX: PWI ) Mlaint. Cont.D~ir., Fallon NV: PWD Ntaint. Div., New Orleans. Belle Chasse LA: PNVD. Maintenanc Control D~ir., Bermuda:PNNI). Willosu Grove PA: PW() (NI. Elliott). L~os Alamitos ('A: PW() Belle Chasse. L.A: PW() Chase Field Hcevilic,'TX PWO Key West Ft.; PW() Whiting Fld. Milton FL: PWO. Dallas TX: PWO. Glenviess IL.: PWO, Kingsville

F *rTX: PWO. Millington TN: PWO, Miramar. San Diego CA: PWO.. Mof fett Field C'A: ROI('C Key West Fl.: SCELant Fleet Norfolk, VA: SC'E Norfolk. VA; SCE. Bairbers Point HI: Security Offr. Alameda C'A

NAIl. BUREAU OF STAND[ARDS B-348 BR tlDr. Campbell), Washington DC'NATI. RESEARCH COUNCIL Naval Studies Hoard, Washington DCNAi'NAVMPI)CPN PW() Bethesda. MI)NATPARACHUTETESTRAN PW Engr. El Centro CANAVACT PWO, London UK aNAVACTI)ET PWO. Holy Lock UKNAVA EROSPREGM E IX'N SCE, Pensacola Fl.NAVAL. FACILITY PWO, Barbxados: PWO. Cape Hatteras. Buxton NC': PWO. Centerville Bch. Ferndale CA: PWNO.

GjuamNAVAV ION ICFAC PWD D~eputy D~ir. D/701,I Indianapolis, INNAVCOASTSYSI.AB CO. Panama City Fl.: Code 423(1) Gowodt. Panama City Ft.: Code 715 (J. Mittleman) Panama

City, FL:, Code 715 tJ. Quirk) Panamai City. Ft.: Library Panama City. Ft.NAVCOMMAREAMSTRSTA Code W-(A02. Honolulu. Wahiawa HI: PWO. Norfolk VA: PWO, Wahiassa, HI: SCE

Unit I Naples ItalyNAVCOMMSTA CO (61 E) Puerto Rico: ('ode 401 Nea Makri, Greece: PWO. Adaik AK: PWO. E xmouth. Australia:

PWO. Fiort Amador ('anal ZoneNAV('OMMUNIT (utler/E. Machias ME PW (jen. For.)NAVCONSI*RACEN Code 74(MN) (Bodwell) Port Hueneme, CANAVE I)TRAPRODEVCEN Tech. Library

NAVEI)UTRACEN Engr D~ept (Codc 42) Newport. RINAV PLEXSYSCOM ('ode PM E- 124-61. Washington D)CNAVENVIRHI.THCEN CO. Cincinnati. OHNAVEODFAC Code 605, Indian Head MI1)

L NAVFAC PWO. L~ewes DENAVFACEiNGCOM Cosk 043 Alexandria. VA: (Code (44 Alexandria. VA: ('ode 0451 Alexandria. VA: Code 0453 (1).

Potter) Alexandria. VA: Code 0454B Alexandria. Va: Code 046: Code (4611) (V M Spaulding) Alexandria. VA:

C~ ~ ~ ~~V:(ode 0414Bleadra VA: Code 0413 Alexandria. VA (' ode (*IBAeanratA ill 0 Alexandria. VA:Code 01330, eimanis) Alexandria. VA: (Code 1023 (M. Carr) Alexandria. VA: Code 1023 Ti. 1). Stevens).

AlexadriaVA. Cde 123 (T. Steven%) Alexandria. VA: Code 104 Alexandria. VA: ('odec 2014 (Mr. i'aamt. PearlHarui H MorionYap. CaoieI. C2 E pne)Aeadi.V.P1.-? oc t Aeadi.V

NAVFCF'NGCO -'HES D)IV. Code 101 Wash. DX': Code 102, (Wildman), Wash. IX': ('ode 40)3 (11. lkeVoetWas. D: Cde 05Was~h. DC: ('ontracts. ROI('(. Annapolis MI): Schee~ssle. ('ode 4(12, Wash. IX'

47

F NAVFACENGeC(M- LANT I'DIV.. Code IOA, Norfolk VA; Code Ill1, Norfolk. VA; Ijur. BR Deputy IDir, NaplesItl;NSNorfolk, VA, RDT&EL.0 09P2, Norfolk VA

NAVFCENGOM -NORTH DIV. (Boretsky) Philadelphia, PA: AROICC, Brooklyn NY: CO: Code 09P (L.CDR AJi.Steart); ode1028, RDT&ELO.0 Philadelphia PA: Code I II (Castranovo) Philadelphia, PA: Code 114 (A.Rhods) DeignDiv. tR. Mi'sino), Philadelphia PA: ROICC, Contracts, Crane IN

NAVFAENGCNI -PAC DIV. Code 09DG (Donovan). Pearl Harbor. HI: Code 402, RDT&E, Pearl Harbor HI:t Comandr. ParlHarbor, HI

NAVFCENGOM -SOUTH DIV. Code W,. RDT&ELO. Charleston SC: D~ir., New Orleans LA: ROICC ([.CIR R.MelrContracts. Corpus ChristiT'X

NAVFACFNGCOM - WEST D)IV. 102: 112: AROICC. Contracts. Twentynine Palms CA: Code ()4B: 09P/20:RI)T&EI.( Code 2011 San Bruno, CA

NAVFACENGCOM CONTRACT AROICC. Point Mugu CA: AROICC, Quantico, VA: code 05, 'rRDENT,Bremerton WA: Code 09E. TRIDENT, Bremerton WA: Dir, Eng. Div.. Exmouth, Australia; Eng IDiv dir,Southwest Pac, Manila, P1, OICC (Knowlton), Kaneohe. HI: 01CC, Souihawest Pac. Manila, PI: OlCCI'ROICC.Balboa Canal Zone: ROICC (Erv in) Paget Sound Naval Shipyard. Bremerton, WA: ROICC dl.CDR J.G. Leech).Subic Baty, R.P.: ROICC AF Guam: I OICC 1.ANT DIV.. Norfolk VA: ROICC Off Point Mugu. CA: ROICC,D~iego Garcia Island: ROICC. Keflavik, Iceland: ROICC. Pacific. San Bruno CA

NAVHOSP LT R, lilshernd. Puerto RicoNAVMAG SCE. GuamNAV MI I RO OIC. Philadelphia PANAVNUPWRU MUSE DET Coide NPU-30 Port Hueneme. CANAV()CEAN() Code 16N8 Bay St. L~ouis, MS: Code 3408 (J. Kravitz) Bay St. Louis: Code 34321 0. 1)ePaltaa). Bay St.

Llouis NISNAVOCEANSYSCliN Code 2010 San Diego. CA: Code 34(X) San Diego CA: Code 409~ (D. G. Moore). San Diego CA:

Code 4473 Batyside Library. San D~iego. CA: Code 52 (H. Talkington) S'-n Diego CA: Code 5204 (J. Stachiw). SanDiego. ('A: Code 5214 (H. Wheeler), San Diego CA: Coide 5224 (RiJones) San Diego CA: Code 6565 (Tech. L.ib.),San Diego CA: ('ode 67(X). San Diego. CA: Code 7511 (PWO) San Diego. ('A: Code 811 San D~iego, CA: ResearchI ih. Sun Diego CA: SC'E (Code 66(X)). San Diego CA

NAVORDSTA PWO. L~ouisville KYNAVPErOFF Code 30. Alcujndria VANAVPETRES Director. Washington DCNAVPGSCOI. 1). Leipper. Monterey CA: Ei. Thornton. Monterey CA: J. Garrison Monterey CA: LCI)R KUC. Kelley

Monterey CANAVPHIBASE ('0. ACB 2 Norfolk. VA: Code S3T. Noriolk VA: Harbor Clearance Unit Two. L.ittle Creek. VA:

OIC, LJ(1 ONE Norfolk. VaNAVRAI)RECFAC PWO. Kami Seya1 JapaknNAVREG.%IEDCEN Code 3041. Memphis. Millington l'N; PWO Newport RI: PWO Portsmouth. VA: SCE (D. Kaye);

SC'E (Il.CI)R B. E. Thurston), San Diego CA: SCE. Camp Pendleton CA: SCE, GuamNAVSUOI.CECOFF (C35 Port Hueneme. CA: CO, Code C44A Port Hueneme, CANAVSEASYSCOM Code 0325. Program Mgr, Washington. DC; Code 00(7 (LT R. MacDougal). Washington D)C,

Code SEA O()C Washington. DCNAVSEC ('ode 6034 (L~ibrary), Washington DCNAVSECGRUACT Facil. Off.. Uialeta Is. Canal Zone: PWO. Edzell Scotland: PWO. Puerto Rico: PWO. lorri Sta.

Okina%%aNAVSHIPREPFA('library. Guam: SCE Suhic BayNAVSHIPYI): ('0 Marine Barracks. Norfolk, Portsmouth VA: Code 202.4. Long Beach CA: Code 202.5 ti.ihrary)

Puget Soutnd. Bremerton WA: Code 380. (Woodroff) Norfolk, Portsmouth. VA: Code 4(N), Puget Sotund: CodeL ~41X).03 Long Beach, CA; Code 404 (I.T J. Ricciot. Norfolk. Portsmouth VA: Coide 410. Mare Is.. Vallejio CA: Code

440) Portsmouth NH: Code 440). Norfolk: Code 440. Puget Sound, Biemrtiron WA: Code 440.4. Charleston S(:C~ode 450. Chuarleston S(': Code 453 (Util. Supr). Valle~jo CA: 1..1), Vivian: L~ibrary, Portsmouth N H: PWI) (('oide

4W8), Philadelphia PA: PWO. Mare Is., PWO. Puget Sound: SCE. Pearl Harbor HI: Tech L~ibrary. Vallejio. CANAVSI'A CO Naival Station. Mayport Fl.: CO Roosevelt Roads P.R. Ptuerto Rico: Engr. Dir.. Rota Spatin: Makint.

Cont. Div., (Guantanamo Bay Cuba: Maint. Div, Dir/Code 531. Rodman Canal Zone: PWI) (1.T W.H. Rigby).('uantam~mo Bay Cuba:. PWI) (I.'UJG.P.M. Motolenich). Puerto Rico: PW() Mid~tay Island: PWO. GuantanamoBav ('ubu; PWO. Keflavik Iceland: PWO, Makyport Fl.; ROICC'Rota Spakin: ROICC. Rota Spain: S('E. Guam:S('E. San Diego CA: SCEF..Subic Ray. R.P.: Utilities Engr Off. (L.'IJG A.S. Ritchie). Roia Spain

NAVS'IA BISHOPS POINT Harbor C~lear. Unit one. Pearl Harbor. HINAVSLJBASE FNS S, Dove. Groton. CT. I.TiG D.W. Peck. Croton, CT; SCE,. Pearl Harbor HI

48

k

NAVSU BSCOL .T J.A. Nelson Groton. CTK_: iNAVSUPPACF CO, Brooklyn NY- CO, Seattle WAk Code 4, 12 Marine Corps Dist, Treasure Is,, San Francisco CA,

Code 413, Seattle WA: LTJG McGarrah, Vallejo CA, Plan/Engr Div., Naples ItalyNAVSURFWPNCEN PWO, White Oak, Silver Spring, MDNAVTECHTRACEN SCE, Pensacola Ft.NAVUSEAWARENGSTA Keyport, WANAVWPNCEN Code 2636 (W. Bonner), China Lake CA: PWO (Code 26), China Lake CA; ROICC (Code 702), China

Lake CANAVWPNEVALFAC Technical Library, Albuquerque NM

L NAVWPNSTA EARLE (Clebak) Colts Neck, NJ; Code 092, Colts Neck NJ; Code 092A (C. Fredericks) Seal BeachCA: ENS G.A. Lowry, Fallbrooik CA; Maint. Control Dir., Yorktown VA: PW Office (Code 09CO) Yorktown, VA:PWO, Seal Beach CA

NAVWPNSUPPCEN Code 09 Crane INNAVXDIVINGU LT A.M. Parisi, Panama City FLNCBU 405 OIC, San Diego. CANCBC CEL AOIC Port Hueneme CAk Code 10 Davisville, RIE Code 155, Port Hueneme CA; Code 156, Port Hueneme.

CA; Code 25111 Port Hueneme, CA. Code 400, Gulfport MS: NESO Code 251 P.R. Winter Port Hueneme, CA,PW Engrg, Gulfport MS; PWO (Code 80) Port Hueneme. CA; PWO. Davisville RI

NCBU 411 OIC. Norfolk VANCR 20, CommanderNCSO BAHRAIN Security Offr, BahrainNMCB 133 tENS T.W. Nielsen); 5. Operations Dept.: 74. CO- Forty, CO,: THREE, Operations Off.NOAA Librarym Rockville, MDNORDA Code 410 Bay St. Louis, MS, Code 440 (Ocean Rsch Off) Bay St. Louis MSNRL Code 84(W) (J. Walsh), Washington DC; Code 8441 (R.A, Skop), Washington DC: Rosenthal, Code 8440, Wash,

DCNSC Code 54.1 (Wynne). Norfolk VANSD SCE, Subic Bay, R.P.: Security Offr. Yokosuka, JapanNTC Code 54 (ENS P. G. Jackel), Orlando FL: Commander Orlando. FL: OICC, CBU-401. Great Lakes ILNAVOCEANSYSCEN Hawaii Lab (D. Moore), HawaiiNUCLEAR REGULATORY COMMISSION T.C. Johnson. Washington, DCNUSC Code 131 New London. CT: Code EA123 (R.S. Munn). New London CT; Code S332. B-80 Ji. Wilcox): Code

SB 331 (Brown). Newport RIE Code TAI31 (G. De la Cruz). New London CT(OCEANAV Mangmt Info Div., Arlington VAS(CEANSYSLANT LT A.R. Giancola, Norfolk VAOFFICE SECRETARY OF DEFENSE OASD IMRA&L) Pentagon (T. Casberg), Washington, DCONR CDR Harlett, Boston MA: BROFF, CO Boston MA: Code 221, Arlington VA. Code 481, Arlington VA, Code

481. Bay St. L.ouis, MS: Code 700F Arlington VA: Dr. A. Laufer, Pasadena CAPACMISRANFAC CO. Kekaha HI

PHIBCB I P&E, Coronado, CAPMTC Covde 3331 (S. Opatowsky) Point Mugu. CA, Code 4253-3. Point Mugu. CA- EOD Mobile Unit, Point Mugu, CA;

Pat. Counsel, Point Mugu CAPWC ACE Office (I.TJG St. Germain) Norfolk VA: CO Norfolk, VA: CO. Great l~akes It.- Code 120, Oakland CA: 3,

Code 120C (Library) San Diego, CA: Code 128. Guam; Code 200, Great Lakes IiL: Code 200, Guam: Code 2(W).Oakland CA: Code 220 Oakland, CA: Code 220.1, Norfolk VA: Code 40 (C, Kolton) Pensacola, FL: Code 400.Pearl Harbor, HI: Code 42B (R. Pascua). Pearl Harbor HI; Code 505A (H. Wheeler): Code 680, San Diego CA:library. Subic Bay, R.P.; OIC CBU-405. San Diego CA: Utilities Officer. Guam: XO Oakland. CA

SPCC Code 122B, Mechanicsburg, PA; PWO (Code 120) Mechanicsburg PATVA Smelser, Knoxville, Tenn.UCT TWO OIC, Port Hueneme CAU.S. MERCHANT MARINE ACADEMY Kings Point, NY (Reprint Custodian)US DEPT'OF AGRIC Forest Products Lab. Madison WI. Forest Products Lab. (R. DeGroot), Madison WIUS DEPT OF INTERIOR Bureau of Land MNGMNT - Code 733 (T.E. Sullivan) Wash, DCUS GEOLOGICAL. SURVEY Off. Marine Geology, Piteleki, Reston VAUS NATIONAL MARINE FISIIERIES SERVICE Highlands NY (Sandy Hook Lab-Library)USAF Maj. Riffel, Rumstein, GermanyUSAF REGIONAL HOSPITAl. Fairchild AFB, WAUSAF SCH(X)O. OF AEROSPACE MEDICINE Hyperbaric Medicine Div, Brooks AFB, TX

49

i'SU((, 4(-LVj V.':hingi tn Dc: ((i-FCV161 I4 iHwkharir %kashingwil. DC): tG-.-II'-3/L'SR!X2) Wat~hingion DC:G-FF )t44, 1 1I. Dim) tI . " ashingtonIW

USCG A(AIDFIY 1.I N. Stramrandi. Nel I owiton ("IUSCG R&D CE NTER( CO: (iobon. CT: I). %IothtcrLIa. Grown C'I ; IIJ( R. Dair. (irotoin (AI I cch. D~iv. Cioon I

1.SNA C h. NkIch. Ingi. Decpt Anna~polis NIý I: ':wg -Lnp it kil Studkl Gip, Animix~iso. Nil: Liigi. D~iv. (C. WAi)Annaplolk Nil): Emivron. Ili-tt. R&D P 0.(.~i~a~. Annapolis MID: Oct.- S~s. Ing Dept (Dr). NionncN IAnnmlA jh~s. NIlI). PWD I)Ingr. D~iv. 4(C. Britiford 4 Annapoklis NIP: PW( A nnapolv.ý NilI)

AMEICHIAN C( \KIA 1l% INISI FIU I[ Ikirokit Ni1 tLibiarl. IAMF RI( AN i \IVERSHlY W\ashington X* 00i. Norton)ARIZO N A StalLc I licig Il'ogi;ains O ff.. Ilhocniis AZ.tBO NNE[VILLE I P( )W[R ADlMIIN PortlandJ C R [ncigp Uon .r%. Oft.. 1). IDaiic%('A IW. D) I ()I O FISH & GA Ni I ling Hen:wh (A MmiNarne tech Into CItrt(At II-. PltIYC (4- NAV I(GATIi #N & OCT A N 1)1:V. Sacrarntnti.. (A (G. Armstrong)CAl I F. NIARITIMiI A(AIWNIY Vidkjo. (A (1.ihriryICALIFOR ~NIA INSt I1ll.1 I OF 'I [:(*Vl\4 JI, OG; PIwadena CA C K~cc Ref Rill

CAl.11IC (NIA S*I All. F~ VN IVRSIh F ONG HI-ACH. CA 4(H IA APAIIk LONG fIt-.C11 I. (A (YE N)1: LO)SANGII .1-S. (A (KI10 I

(AttI H1 IC I NIN'. \ii.01 Iiigr D~ept. P'rof.Nct'ck. WaNsh.. DC()RN 1-1.1. I 'NI I .RSHt Y ltha.ic, N fiScrials Dept. tFngr I h. 4

I LANIIS&NAI LN )CK [INt)R IRiASNI I.NIW~A RA ((N.1:.(NIA~LISITIRi: Boma Raton H:. (Occan Fipgr Detkp C.

V~~ ~~ It (JR I),Al XIANAWt( NI V F SY Bocau Raton H I W. lessin)FLO RIDIA I FCH~IINC iI C- GI l. NIVI;RSI IY ORL ANDOX. 1:1 iHA\Rt NIAN)ICR [Si I NSl. F( K 4 (T.AN & MOC)LNI 1NIN (arscon ('it Nsi' (Stuiv~us - hI ahurv 4

INIC IN I' .11111) k1

(IFO.1 iN.Sl It( N 1- At.' I- AR..H; RICnt HA ARcoo (if4 Civilcn PA'n -m

61-010A NS 111 f . 1- 114 No 0Y inmi nia GA (1 . Naio.3t Fit)itr

INDLIIANA I)IVNTRiYI+RL.C [OLCF &nml [NERON In~np lsp. I hN s~~uii.hatnKnc

INISS )I'RLI' ON[F MARINI CYIN(F Jdkrso, e (i NI i\Niiet10% A S'IA'N 1 (R , A I'I. A. it'll'TI bepASFF SC (IAC ýG

NIAINE- SIAh4IIM ACO] 1:61 'enc %! ) (Cuninghia:m) rN. F4 IRAY

HBiAI N I fl~CO N61RESS WAHINGO )LKCES FNC-' A gisa.IM)LOISICI(AN.A( DIVNI URaAL. (NIVIR(Sl & I-huttoDep. Nit Conaserain)ao og .

M1IT (,mnridge NIA: (.mbrnhige MA (Rm 144-51H). lch. Repor~s. Fngr. .iri~.4: (ambhridge MiA (Whitman): Cambridge.

MOc NTAN AINEG luHc.:NAIl ACADEMY 01 I-NC. At FX.ANIRIA. VA tS11"ARI+. JR.)NE'A It ,MNtSiIkIRI Concix 4 Nil. (Governor', Coun,:.; on I- nerpylNEW MiEXICO SOLIAR FNI WOY INSIl. D~r. Z..ihch1 Las Cytices NMNC RTI-W!-SI'P`RN N IV Z.11. Hia/ans F toston Ii.N'* CITN (C04IM% NI'i (CO1.11 tAI i) OK.YN. NY tII!~ARY)NVS 1;"N[ICC *V C L(Fibrar% . Almianki Is'1I IIV. NC iRF; D)AME Katonia, Notre P.I4L. IN

OR11 ' N SI AllTI N I V 1: R S I Y i F De pi C iracv) Cor s'IIlis. 0 RK: (OR V AI ,1. 1 S,)R (CF. 1)1:VI'I. .3 El: 1.1,:)RVAI H.S. ()I- iI([ i1101T..... Corvalis (4K (Svchool oft Ov)eanograph%

PINNSYI VAN[IAS FA II- I'N IV[;RSIIY FI-AT 1 COI TEG. PA (SN YIDF)l;, tt tee A(pýdKc

I V~ N IVERNSt Y PARIK, PA iC 41(4 OLS(I,94'PURLK)UINIVI. `;HIlY I alasette IN 0.csn~irmasi: I afavitte. IN 4Altwshacftt4: Lafayette. IN (CT- ingr. ~ill

50

0" )N N Ut11(1. AH artford It I ii )tl of lP~n. & [irjg Pohc%SA\, DI)I)SIAn IV IN IV. 1. N~ooaný, Sin lDicg, (A: Dr~. Krish~xramiurihv. San D~iego ("ASCRIPPS INSTFITLITE ( U O(ANO RAMIN I A JOLL.1A, CA ODAMS)AN: San D~iego. CA IMarina Phý. I a!'. SpwieoSAI: VIT U 1' rof Scht ieglr Scauk V. AA4 114 %% > VS Ur RS(!Ii I N S' I King. Sani Antoni~o. I X; R. Del-Hart. Sain Antonio'TXSI.ANII()RD) UNIVFRKS!! N 1Ingr I ib. Sutoford CA: S1 ANFN)REJ. CA (II)L'6I.ASiSt ATE lit 115 ALCS I A. MU i lAINI. SA AIT. F UEL A.*NIAC & (ONNSI-V. 01+.)SI ATE UNIV. 0F NUM. N' iRK Buiffalo,) NNY- Fort Schu~icr. NY (Lon o ardilI I-XAS -\&\ ~I\ K!I W.H. I cd!'c~tcvC0dkgc Staiuon. IV: College Station IX (:I- Dept. 1krhichi1:NIV I-HSUN, 0I. (Al 1k )RNIA HI:RKFI+.N.\ CA (CI- IIPI. (jU:RWIIK : BFWRKEIY CA (CI DEV !I.~I Ni! 1.. ( iii HeI: e (~Ck(A it. Bresekr): Bekrleký CA .cpt of Na%,al Arch.i: Berkceýe C.A if. llcarm.ni: IDAVIS,

CA (CF lW! 1!. NAY! 04ý. I15 CRI)R.A iI.A\%RIHNCE IJVIVFRMOH!I I Al).*M )KAR/i: La lolLI (AiAk.q.Diept. Lib'. C4r7,5,A: MI Dmncan. (ekl A: Oflf-Carpu- FacN & Fnergý Conserl... D!)i , CA (A: ff-(amru1%

F;c & FII g ( onwci\_ Dais. CA: SAN ODI6t CA1. A1...CASFKll VcPrsdnhke'.

I. N IV [IS!!Y~ IiA\,%A1 A' A! H)NOC (!.I'. 1-!! iS( IN(T ANOI I I.( it. I)IV .) l:onolulu fit 0)i. S/ilordiUN('!R IY - LLNOIASSA %let/ RellS m ( eroneitim All ia URAA. (U lAISN, RA'A LitpR

t RHNAU. NII(HI1~ Urbauna1.W Dept.r \kl Hicari iLNiVURSII-) 0F KNE ASMKanw lIci II N lincoln. IE aRo%%ce KShl ij

I ' NI V 1R9S11N ( )F NJ- HIAMPSHIRE [)ILRIIAMl. Ni it lAVOW)iLN'IV:RSrI*Y 0F PENhNSYLVNAN IA PHIiIIADVITHIA. PA (SCIIB) 10F 01 N(iR & A1P111.111) S'II:N( I-. ROLL)!.

I. NIVI RSII'N O1: RHO!)!. ISL AN!) KIN1(iSF( IN. RI (P`A/IS): Narr..ganwi2t RI W01l Marine Sci. I ib.iI. NVi RS~l N0! So). ('Al IF0 RNIA Unix; So. CalifU N IV ERSITY( I) FX AS List . Mkiarw S6ciI .ihrzkrN . Port Arkauma IXI NIVURSIIY0OF lITXAS AlI AUSTIN AUStIT\. IX i Il-Ht)PSON):. Ausi~n. IX igrenii

IN NIV F RSII 1) 01 WASH-!IN(. C N Seattle %A A (MI. Sherif): IDept of Civil Engr ( Dr. S1ittoo).. Sem:ttv \k A:SE.F 'AI IT*. WA (AI'PIIIA l PH!!YSIC;S LAB). SEATTI.F.. %A ((X+lAN FINt R-S( If LAB4. (,RAY:- SEATTI I.%\A (!'A(!E!C NIARIN. I-.N% IRON. LAW.. H-Al .P~kW) S,;.i k .A 4F. l~ingeri. Scttle. AA Traostuortation.Constraci ion & (icorn. Di%

(XI!V !RSI I'N 01:-'AISCO(N'SIN- Millur.kt~Ce A 1 Cu1r ofl Greatl LikeS StUdies)('RS RISi:AR( " (C0. LIBRARY SAN NIATY%0. CAVFNlI;RA COUNTY ENVIRON RFSORCR(F AGIFNCY Ifentura. (A'Alee!' LibraivVIRGINIA INST. OF MARINE. SC!. (ioucc%tvr l'oiit VA it ibraurýAGIIABIAN ASS( iC. C. Bagge. IA Segundo CAAlFRI 1: ) A V;F. & ASS0C. Honolulr HII

AN1IFI 'K t11fkhore Res. & IFngr Div

ANISC( Dr. R. 'McCoy. 1`tic. PA

ARVID (iRAN! OLYMPIA. WAATLANTIt.'R('ICI-IFILD), CO )AI IlAS. IX (SMITH)AUSTRALIA !)pt. l'W iA. Hick%(. MelN'wneBIECtl , .11:,1 C(ORP. SAN FRANCISCO, ('A (PHEL.PS)411*1i0Uý'I HAEC0N. N.V., GentBFHI IHE M STEI. CO. IDimukik. Sethelchern. PABOUW KAMP INC Ho'keleyBRANDI) NDU~S SEHV INC. J. Bueihler. Haciendat Heights ('ABRVI 15I11 F MI6ASSY Sci. & Tcch. Dept. (i. McAukleI. Waiv.,ington DilRRO\% N & ('AlIMMT 1.1. U; M Sauunders kAdnut (rec'k. (ABROWN & R(X~lI Hou%%oh IX it). Ward)CANAD)A Adriati. Andvr%on & Assoc-. Winnipeg: (Lribv ci ~~(nl~) ot ~rov ra ~l~r\ . (algar-

Alberta: I ockheed Petro. Serv. Lid. Nc% Wemmtnuinter HI'.; Vivk.cIhed Petroll Srv. I td.. Nei% ' %inhjnmer OIC:

Nenninger & heneveri Inc.. Monirecul

;z

('AYWOOD, NOPP. WARD). AlA Sacramento. CA(CF BRAUN CO Dui Blochet. Murray Hill, NJCHEMEI)CORP L~ake Zurich IL. (IDearlwirn (Chem, IDiv.ILih.)CHEVRON Oil. FIE~LD RESEARCH CO. LA HABRA. CA (BROOKS)I ~ COLUMH!A GULF TRANSMISSION ('0. HOUSTON. TX (ENG. LB.)~C'ONCRETEF TECH NOI.OGY (ORP. TACOMA, WA (AN DERSONiD)ESIGN SERVICES Beck. Ventura, ('ADILL.INGHAM PRECAST F. McHale. Honolulu HIDIXIE D)IVING CENTrER Decatur. GAI)RAVO C'ORP Pittsburgh PA (Giannino): Pittsburgh PA (Wright)I)URI.ACH. O'NFAI.. JENKINS & ASSOC'. Columbia SC'NORWAY l)EINOkSKE VERITAS (Lihrar)). OsloEVAL.UATION ASSOC. INC KING OF PRUSSIA. PA 4FEI)EIE)EXXON PROD)UCTION RESEARC'H Co Housbon TX (A. Butler Jr): Houston- TX (Chaoj

FORD). BACON & DAVIS. INC. Nce% York- Iibrary)

F:RANCE Dr~. lNterur. Boulogne. IL. Plij4i. zteris; P. Jensen. Boulogne; Rouder L.aCroix. FKtriýUENERAl. DYNAMIICS Flec. B~oat D~iv., Environ. EngrifH. Wallnia"p. Growon CTGEOTECHNICA!. ENGINEERS IN('. Winchc%Ier. MA (Pa~ulding)61l IDI)EN ('0. STRONGSVILI.I. OH I RSCH L.IB)CjOULD! INC. Shady Side MI1) ('hes. Ins[. Div'.. %%. Paul)GRUMMAN A2-.RISPA('E CORP. Bethpage NY (Tlech. Info. Cir)HALEY & AILDRICHI. INC. Caimbridge MA (Aldrich. Jr.)HUGHLS AIR('RArI- Culver 0qt ('A (Tech. Doec. ('tip

ITALY MI. ('aironi. Milan: Sergio lationi Mijlano; '1,rino ( F. Lecvi)NIAKAI O(TAN EN61RNG INC. Kailua. HI'AMESC('. R. Girdley. Orlando Fl.KENN FTH TATOR ASS(K'CORAOPOLIlS. ?A (lIBRARYV)4KOREA Korea Rsch In%t. Ship & Ocean 04. (Choi). SeoulI.ANI ONT- xHI-HETY (iJ(OLOGlLA 1, OBSF RV. Pali~adc% NY I McCo y): P~ali%adc- NY tScl%% ynpLIN OFFSHORE ENG(;R P. Cho%%. San Franci~.co ('AL,(K'KHEI-'.) MISSII.:S & SPA('E (0. lN('.I.. T'rinmble. Suinn)ývale(',A: Nlgr Naval Arch & Mar Enig Sunn~vaIl'.

('A; Sunnyvaile C"A tkyneuicz): Sunnyvale. CA (Phillip%)I.(X'KHEUD-.I(X'EAN lABORATO'(RY San Dicuto C.A F. Sinipsten: San liego. (A4Springco-MARATHON Ofl. (0) H-ou%ion TX (U. Scay)MARINE CONC'RETE STIRUCFURI:S INC. NIEFAIRIF. L.A (INGRAHAMsMC ('I.EI.I.ANI) ENGINEERS INC Hou%ton" X III. McClelland)

CIX)ONNEI. AIRCRAFFC('. Dept 501 (R.N. Fa~man). St L~ouis MO)MEiDALLI A ASS(K. INC. i.T. 6AFFFY If SAN'IA ANA. ('ANIEIWFRMO`I1A CO ('DIiving Division. flarve%. LA

MOB1L. PIPE lINE (0. D)ALLIAS. TX N1.161 OF ENGR (NOACKiMIJESEII. RUTI.FiXiI;. WENTWORTH ANI)JOHNSTON NEW YORK (RICHARI)SiNFV% ZEALAND Net% Zealand Concrete Research Assoc. (Libiarmut). PoriruaN.W'P(OR (NEWS SHIPIIHIIXi & DRYX(K'K (0( Newport NeASVA tTech. I.ihj)NORWAY A. Forum. Trondheim: D)LT NO RSK I: VRI'rAS (Roren) Oslo: 1. F'o%%. Osl: I. Creed, Ski; I.D. HoI%t.

Oslo; Jakobsen. O%I4': Nor~vgian'Tech Univ (lBrandt/acz). TrondheimO('VAN FNGiINEERS SAUSALITO. ('A (RYNFKCKI)(XKlAN RESOURCI FNG INC. HOUSTON.'IX tANDERSONp I()h&le

PUEIRl 0 RICO Puerto Rico (R'ch L~ih,. klayaque/ P RIR.l J ROWN ASSO(X(IMcKeehan), Houston. TXRA ND CORP. Santa Monica ('A (A. Laupa)RAYMNION)INTF.RNAl14)NAI INC. 1: ('oll Soijl Tech Dept. .Penn',auken. NJRIVERSIDE ('ENI-N I C'0 Riverside "'A tW. Srritt)

SA~lIN I RI~-S. INC A I'Won. Nrovidence kI

52

SAN DIA 1.ABORATORI ES Albuquerque. NMN (Vortman): I ihrary Div.. I .ivermoi e CASCHUPACK ASSOC' So. NORWALK. CT iSCHUPACK)SEAFOOD) LABORATORY MOREiHEADI CITY. NC ILIBRARY)SEATECH CORP. MIAMI. Ft. (PERONI)SHELL! DEVELOPMENT CO. Houston TX (C. Sellars Jr.)SHELL. O11. Co. HOUSTON. TX (MARSHALL). Houston TX (R. de Castongrene)SAWUTH AMERICA N. Nouci. Valencia, VenazuelaSWED)EN Cement & Concrete Research Inst.. Stockholm: fleoTech Inst: VI38 It.ibrary). StockholmITCHNICA!. COATINGS CO Oakmont PA (Library)TEXTRON INC BUFFAL.O. NY (RESEARCH CFNTER I.I13.)TlIDE% ATER CONSTR. Co( Norfolk VA (Fowlerl

-RW SYTM -.'E.N.O EG Ii RFIX)NIX) BEACH. ('A (I)AI IUNION C'ARBIDE CORP. RiJ. Martell Boton. MAUNITED) KIN(IX)NI Briti%h EnibasVy Onfo. Of fr). Washington DC: Cement & Concrete Assoc Wexham Springs.

Slough Btack4. 'cm.,ýnt & C'oncrete Asstw. (Library). Wexhamn Springs, Slough: Cement & Concrete Assoc. (Lit.Ex). Bucks: D. L~ee. L~ondon; D. New. G. M4ItinseII & Partners. L~ondon; J. Derringion. London: L~ibrary, Bristol.R. Rudhaim Oxfordshirc. Shaw & Hatton (F. Hansen). London; Taylor, Wotsro% Constr (014P). Southoll.Middlesc%: Taylor, Wodtxlow Constr (Stubbs). Southall. Middlesex: Univ. of Briisol (R. Mlorgain). Bristol

LUNI'-I:DTEFCHNOL.OGIES Windsor Locks ("I (H-amilton SOd Div.. LibrarylWATI BRIAN ASSOX' INC. Houston, TXWESTINGHOUSE ELET.FRIC CORP. Annufpolis MD (Oceanic IDiv Lib. Bryan): L~ibrary. Pittsburgh PAWEYERHAEUSER CO. I.ON6VllEW. WA (TECH CTR t.1ij

~ISS. JANNEY. EI.STNER. & ASSOC Northbrook. It. (D.W. Pfeifer)WMI ('[APP LABS - BATTELLE I)UXBURY. MA itLIBRARY): Duxbur). MA tkiciag-dsWOOI)WARD-CINDIE ('ONSULTAN' S PLYMOUTH MEET'ING PA (CROSS. fit)ADAMS, CAPI' R[ET) Irvine. CAANTION TEIESK() fronxville NYBRAHTZ7 L, Jolla. CABRYANT ROSE Jol'nson Div. UOP. Glendora (CABUI1.ILOCK( La CanadaF. H1:U ZE Boulder COKRUZIC. i.P. Silver Spring. MD('API MURPHY Sunnyvitle. CAG;REG PAGF. EiUGENE. ORR.I. HI-' NI I- Old 'Sayhrook CT'R.Q. PAI.M[E*R K;,tua. H?FNERCY R&D) ADMIN. H. Skvvtbo. Washington. DCSMITrH iUllfpirt. MS'rW. NIERMEi. Washington DC'WM TrA L.BO'Orange ('ACEC D~onofrio. John L., l.T; Morris. I)oinald 'i.. LT

53