CHAPTER I

I n t r o d u c t i o n t o S u r f a c e

E n g i n e e r i n g f o r

C o r r o s i o n a n d W e a r

R e s i s t a n c e

SURFACE ENGINEERING is a multidisciplinary activity intended totailor the properties of the surfaces of engineering components so thattheir function and serviceability can be improved. The ASM Handbook de-fines surface engineering as "treatment of the surface and near-surfaceregions of a material to allow the surface to perform functions that aredistinct from those functions demanded from the bulk of the material"(Ref 1). The desired properties or characteristics of surface-engineeredcomponents include:

Improved corrosion resistance through barrier or sacrificial protectionImproved oxidation and/or sulfidation resistanceImproved wear resistanceReduced frictional energy lossesImproved mechanical properties, for example, enhanced fatigue ortoughnessImproved electronic or electrical propertiesImproved thermal insulationImproved aesthetic appearance

As indicated in Table 1, these properties can be enhanced metallurgically,mechanically, chemically, or by adding a coating.

The bulk of the material or substrate cannot be considered totally inde-pendent of the surface treatment. Most surface processes are not limitedto the immediate region of the surface, but can involve the substrate by

Table 1 Surface engineering options and property benefitsSurface treatment/coating type

Changing the surface metallurgyLocalized surface hardening (flame, induction, laser,

and electron-beam hardening)Laser melting

Shot peening

Changing the surface chemistryPhosphate chemical conversion coatings

Chromate chemical conversion coatings

Black oxide chemical conversion coatingsAnodizing (electrochemical conversion coating)

Steam treating

Carburizing

Nitriding

CarbonitridingFerritic nitrocarburizingDiffusion (pack cementation) chromizingDiffusion (pack cementation) aluminizingDiffusion (pack cementation) siliconizingBoronizing (bonding)Ion implantationLaser alloying

Adding a surface layer or coatingOrganic coatings (paints and polymeric or elastomeric

coatings and linings)Ceramic coatings (glass linings, cement linings,

and porcelain enamels)Slip/sinter ceramic coatingsHot-dip galvanizing (zinc coatings)Hot-dip aluminizingHot-dip lead-tin alloy-coatings (terne coatings)Tin plate (continuous electrodeposition)Zinc-nickel alloy plate (continuous electrodeposition)Electroplating

Electroless plating

Mechanical platingWeld overlays

Thermal spraying

Cladding (roll bonding, explosive bonding, hotisostatic pressing, etc.)

Laser claddingCarbide (salt bath) diffusionChemical vapor deposition (CVD)

Physical vapor deposition (PVD)

Primary property benefits

Improved wear resistance through the development of a hard martensitic surface

Improved wear resistance through grain refinement and the formation of finedispersions of precipitates

Improved fatigue strength due to compressive stresses induced on the exposedsurface, also relieves tensile stresses that contribute to stress-corrosion cracking

Used primarily on steels for enhanced corrosion resistance, increased plating or paintadhesion, and for lubricity (e.g., to increase the formability of sheet metals)

Enhanced bare or painted corrosion resistance, improved adhesion of paint orother organic finishes, and provides the metallic surface with a decorative finish

Used for decorative applications, e.g., the "bluing" on steel gun barrelsUsed primarily for aluminum for increased corrosion resistance, improved decorative

appearance, increased abrasion resistance (hard anodizing), improved paint adhe-sion, and improved adhesive bonding (higher bond strength and durability)

Used on ferrous powder metallurgy parts to increase wear resistance and transverserupture strength

Used primarily for steels for increased resistance to wear, bending fatigue, androlling-contact fatigue

Used primarily for steels for improved wear resistance, increased fatigue resistance,and improved corrosion resistance (except stainless steels)

Used primarily for steels for improved wear resistanceImproved antiscuffing characteristics of ferrous alloysImproved molten-salt hot corrosionImproved oxidation resistance, sulfidation resistance, and carburization resistanceImproved oxidation resistanceImproved wear resistance, oxidative wear, and surface fatigueImproved friction and wear resistance for a variety of substratesImproved wear resistance

Improved corrosion resistance, wear resistance, and aesthetic appearance

Improved corrosion resistance

Improved wear resistance and heat resistanceImproved corrosion resistance via sacrificial protection of steel substrateImproved corrosion and oxidation resistance of steel substrateImproved corrosion resistance of steel substrateImproved corrosion resistance of steel substrateImproved corrosion resistance of steel substrateDepending on the metal or metals being electrodeposited, improved corrosion

resistance (e.g., nickel-chromium multilayer coatings, and cadmium and zinc sacri-ficial coatings), wear resistance (e.g., hard chromium coatings), electrical proper-ties (e.g., copper and silver), and aesthetic appearance (e.g., bright nickel or deco-rative chromium plating)

Improved corrosion resistance (nickel-phosphorus) and wear resistance(nickel-phosphorus and nickel-boron)

Improved corrosion resistanceImproved wear resistance (hardfacing alloys) and corrosion resistance (stainless steel

or nickel-base overlays) and dimensional restoration (buildup alloys)Primarily used for improved wear resistance (many coating systems including ceramics

and cermets), but also used for improved corrosion resistance (aluminum, zinc, andtheir alloys) and oxidation resistance (e.g., MCrAlY), thermal barrier protection(partially stabilized zirconia), electrically conductive coatings (e.g., copper and sil-ver), and dimensional restoration

Improved corrosion resistance

Improved wear resistanceUsed primarily for steels for improved wear resistance in tooling applicationsImproved wear (e.g., tools and dies), erosion, and corrosion resistance; also used for

epitaxial growth of semiconductorsImproved wear (e.g., tools and dies) and corrosion resistance, improved optical and

electronic properties, and decorative applications

exposure to either a thermal cycle or a mechanical stress. For example,diffusion heat treatment coatings (e.g., carburizing/nitriding) often havehigh-temperature thermal cycles that may subject the substrate to temper-atures that cause phase transformations and thus property changes, orshot-peening treatments that deliberately strain the substrate surface to in-duce improved fatigue properties. It is the purpose of this book, and inparticular Chapters 4 to 6, to review information on surface treatmentsthat improve service performance so that metallurgists, chemists, me-chanical engineers, and design engineers may consider surface-engi-neered components as an alternative to more costly materials.

Surface Engineering to Combat Corrosion and Wear

The Economic Effects of Corrosion and Wear. The progressive deteri-oration, due to corrosion and wear, of metallic surfaces in use in major in-dustrial plants ultimately leads to loss of plant efficiency and at worst a shut-down. Corrosion and wear damage to materials, both directly and indirectly,costs the United States hundreds of billions of dollars annually. For exam-ple, corrosion of metals costs the U.S. economy almost $300 billion per yearat current prices. This amounts to about 4.2% of the gross national product.

However, about 40% of the total cost could be avoided by proper corro-sion prevention methods. Table 2 provides a breakdown of the cost ofmetallic corrosion in the United States. Similar studies on wear failureshave shown that the wear of materials costs the U.S. economy about $20billion per year (in 1978 dollars) compared to about $80 billion annually(see Table 2) for corrosion during the same period. Table 3 illustrates theextent of wear failures by various operations within specific industrialsegments. Highway vehicles alone use annually 14,600 X 1012 Btu/ton ofenergy represented in lost weight of steel and 18.6% of this energy couldbe saved through effective wear-control measures.

Table 2 Cost of metallic corrosion in the United StatesBillions of U.S. dollars

Industry 1975 1995

All industriesTotal 82.0 296.0Avoidable 33.0 104.0

Motor vehiclesTotal 31.4 94.0Avoidable 23.1 65.0

AircraftTotal 3.0 13.0Avoidable 0.6 3.0

Other industriesTotal 47.6 189.0Avoidable 9.3 36.0

Source: Ref 2

Table 3 Industrial operations with significant annualwear economic consequencesIndustry Operation Loss mass (a), 1012 BtuUtilities (28% total U.S. Seate 185

consumption) Accessories 120Bearings 55Reliability 145Total 505

Transportation (26% total Brakes (b)U.S. consumption) Valve trains (b)

Piston ring assemblies (b)Transmission (b)Bearings (b)Gears (b)Total (b)

Mining Ore processing 22.80Surface mining 13.26Shaft mining 10.70Drilling 5.58Total 52.34

Agriculture Tillage 16.85Planting 2.47Total 19.32

Primary metals Hot rolling 14.30Cold rolling 0.14Total 14.44

(a) Assumes 19.2 X 106 Btu per ton of energy represented in lost weight of steel.(b) Lost mass not estimated. Source: Ref 3

Corrosive Wear. Complicating matters is the fact that the combined ef-fects of wear and corrosion can result in total material losses that are muchgreater than the additive effects of each process taken alone, which indi-cates a synergism between the two processes. Although corrosion canoften occur in the absence of mechanical wear, the opposite is rarely true.Corrosion accompanies the wear process to some extent in all environ-ments, except in vacuum and inert atmospheres. Corrosion and wear oftencombine to cause aggressive damage in a number of industries, such asmining, mineral processing, chemical processing, pulp and paper produc-tion, and energy production. Corrosion and wear processes involve manymechanisms, the combined actions of which lead to the mutual reinforce-ment of their effectiveness. As listed in Table 4, 17 synergistic relation-ships among abrasion, impact, and corrosion that could significantly in-crease material degradation in wet and aqueous environments have beenidentified.

The combined effects of corrosion and wear can also lead to galvaniccorrosion in some applications, such as crushing and grinding (comminu-tion) of mineral ores. Wear debris and corrosion products that are formedduring comminution affect product quality and can adversely affect sub-sequent benefication by altering the chemical and electrochemical proper-ties of the mineral system (Ref 5-8). Electrochemical interactions be-tween minerals and grinding media can occur, causing galvanic couplingthat leads to increased corrosion wear. More detailed information on gal-vanic corrosion can be found in Chapter 2.

Methods to Control Corrosion. Owing to its many favorable charac-teristics, steel is well suited and widely used for a broad range of engi-neering applications and is referenced here to demonstrate the various cor-rosion-control steps that can be considered. Steel has a variety of excellentmechanical properties, such as strength, toughness, ductility, and dent re-sistance. Steel also offers good manufacturability, including formability,weldability, and paintability. Other positive factors include its availability,ferromagnetic properties, recyclability, and cost. Because steel is suscep-tible to corrosion in the presence of moisture, and to oxidation at elevatedtemperatures, successful use of these favorable characteristics generallyrequires some form of protection.

Methods of corrosion protection employed to protect steel include:

Altering the metal by alloying, that is, using a more highly alloyed andexpensive stainless steel rather than a plain carbon or low-alloy steelChanging the environment by desiccation or the use of inhibitorsControlling the electrochemical potential by the application of ca-thodic or anodic currents, that is, cathodic and anodic protectionApplying organic, metallic, or inorganic (glasses and ceramics) coat-ings

Application of corrosion-resistant coatings is one of the most widely usedmeans of protecting steel. As shown in Table 1, there are a wide varietyof coatings to choose from, and proper selection is based on the compo-nent size and accessibility, the corrosive environment, the anticipated

Table 4 Synergistic relationships between wear and corrosion mechanismsAbrasionRemoves protective oxidized metal and polarized coatings to expose unoxidized metal, in addition to removing metal particles.Forms microscopic grooves and dents for concentration cell corrosion.Increases microscopic surface area exposed to corrosion.Removes strain-hardened surface layers.Cracks brittle metal constituents forming sites for impact hydraulic splitting.Plastic deformation by high-stress metal-mineral contact causes strain hardening and susceptibility to chemical attack.

CorrosionProduces pits that induce microcracking.Microcracks at pits invite hydraulic splitting during impact.Roughens surface, reducing energy needed to abrade away metal.May produce hydrogen with subsequent absorption and cracking in steel.Selectively attacks grain boundaries and less noble phases of multiphase microstructures, weakening adjacent metal.ImpactPlastic deformation makes some constituents more susceptible to corrosion.Cracks brittle constituents, tears apart ductile constituents to form sites for crevice corrosion, hydraulic splitting.Supplies kinetic energy to drive abrasion mechanism.Pressurizes mill water to cause splitting, cavitation, and jet erosion of metal and protective oxidized material.Pressurizes mill water and gases to produce unknown temperatures, phase changes, and decomposition or reaction products from ore and water

constituents.Heats ball metal, ore, fluids to increase corrosive effects.

Source: Ref 4

Thickness, mmFig. 1 Approximate thickness of various surface engineering treatments

temperatures, component distortion, the coating thickness attainable(Fig. 1), and costs. Many of these selection criteria are addressed in Chap-ters 6 to 8 in this book.

Painting is probably the most widely used engineering coating used toprotect steel from corrosion. There are a wide variety of coating formula-tions that have been developed for outdoor exposure, marine atmospheres,water immersion, chemical fumes, extreme sunlight, high humidity, andmoderately high temperatures (less than about 200 0C, or 400 0F).

The most widely used corrosion-resistant metallic coatings are hot-dipped zinc, zinc-aluminum, and aluminum coatings. These coatings ex-hibit excellent resistance to atmospheric corrosion and are widely used inthe construction, automobile, utility, and appliance industries.

Other important coating processes for steels include electroplating,electroless plating, thermal spraying, pack cementation aluminizing (forhigh-temperature oxidation resistance), and cladding (including weldcladding and roll-bonded claddings). Applications and corrosion perform-ance of these coatings are described in Chapter 6 in this book.

Methods to Control Wear. As is described in Chapter 3 in this book,there are many types of wear, but there are only four main types of wearsystems (tribosystems) that produce wear and six basic wear control steps(Ref 9). The four basic tribosystems are:

Weld overlayFriction surfacing

Thermal sprayingCarburizing

CarbonitridingNitrocarburizing

NitridingMechanical working

Electrochemical plate + diffusionTransformation hardening

Surface alloyinglasersHot dipping (galvanizing and aluminum)

Mechanical platingElectroless plating

Electrolytic platingChemical vapor deposition

Physical vapor depositionResin or laquer"bonding

Ion implantation

Relatively smooth solids sliding on other smooth solidsHard, sharp substances sliding on softer surfacesFatigue of surfaces by repeated stressing (usually compressive)Fluids with or without suspended solids in motion with respect to asolid surface

As shown in Fig. 2, the wear that occurs in these tribosystems can be ad-dressed by coatings or by modifications to the substrate metallurgy orchemistry.

The six traditional techniques applied to materials to deal with wearproduced in the preceding tribosystems include:

Separate conforming surfaces with a lubricating film (see Chapter 3 inthis book).Make the wearing surface hard through the use of hardfacing, diffu-sion heat treatments, hard chromium plating, or more recently devel-oped vapor deposition techniques or high-energy processes (e.g., ionimplantation).Make the wearing surface resistant to fracture. Many wear processesinvolve fracture of material from a surface; thus toughness and frac-ture resistance play a significant role in wear-resistant surfaces. Theuse of very hard materials such as ceramics, cemented carbides, andhard chromium can lead to fracture problems that nullify the benefitsof the hard surface.Make the eroding surface resistant to corrosion. Examples include theuse of cobalt-base hardfacing alloys to resist liquid erosion, cavitation,and slurry erosion; aluminum bronze hardfacing alloys to prevent cav-itation damage on marine propellers or to repair props that have

Substrate treatments toreduce wearThrough hardeningSurface hardening (flame,induction, EB, laser)Diffusion of a hardeningspecies (carburizing,nitriding, etc.)Laser/EB alloyingIon implantationWork hardening

Coatings to reduce wearPolymers/elastomersElectrochemical(plating, etc.)Chemical (CVD,electroless plating)Thermal sprayingFusion weldingThin films (PVD,sputtering, ion plating)Wear tilesCladding (cast, explosion,hot rolling)Lubricants

Wear-causingeffects

Tribosystem SurfacewearFlC. 2 Surface engineering processes used to prevent wear. CVD, chemical vapor deposition; PVD,

" physical vapor deposition; EB, electron beam

suffered cavitation damage; nickel-base hardfacing alloys to resistchemical attack; and epoxy-filled rebuilding cements used to resistslurry erosion in pumps.Choose material couples that are resistant to interaction in sliding(metal-to-metal wear resistance). Hardfacing alloys such as cobalt-base and nickel-chromium-boron alloys have been used for manyyears for applications involving metal-to-metal wear. Other surface-engineering options include through-hardened tool steels, diffusion(case)-hardened surfaces, selective surface-hardened alloy steels, andsome platings.Make the wearing surface fatigue resistant. Rolling-element bear-ings, gears, cams, and similar power-transmission devices oftenwear by a mechanism of surface fatigue. Repeated point or line con-tact stresses can lead to subsurface cracks that eventually grow toproduce surface pits and eventual failure of the device. Prevention ispossible through the use of through-hardened steels, heavy case-hardened steels, and flame-, induction-, electron beam-, or laser-hardened steels.

More details on these surface-engineering techniques can be found inChapters 5 through 8 in this book.

Material/Process Selection (Ref 10). Faced with the wide range ofpossibilities indicated in Table 1 and the discussions on "Methods to Con-trol Corrosion" and "Methods to Control Wear," selection of surface engi-

Predict workingenvironment fromconsideration of design

Proceed with one-piececonstruction (see notebelow)Note: One-piece construction is often leastexpensive for small parts as some surfacing alloysare available as castings machined to finished sizeor as powder metallurgical parts.

Identify materialrequirements forstructure and surface

Consider one-piececonstructionYes No

Select surfacing materialto suit requirements

Analyze service failuresto assist selection ofbetter materials

Select substrate materialto suit strength, heat, andcorrosion needs

Select from surfacing processessuitable for chosen material and job,(must satisfy needs for coating density,thickness, dilution, etc.)

Reconsidermaterials

Decide if chosen processsuits substrate materialand design (adhesion,HAZ, access, distortion,etc.)

Yes

Identify quality assuranceand control needs Finalize choice ofmaterials and process

Decide manufacturingdetails, procedures,health and safetyrequirements, etc.

Fig, 3 Checklist for surface engineering material/process selection. HAZ, heat-affected zone

Reconsider processand/or material

None

Yes No

neering material and process may seem difficult, but it is normallystraightforward. Often there are constraints placed on the choice becauseof availability (e.g., laser melting and/or alloying are not widely used, andthese processes can only be obtained by a special arrangement with laserjob shops). In many cases there is a precedent, but when considering anew problem it helps to follow a checklist of the type shown in Fig. 3.

The sequence of decisions to be made covers several fundamentalpoints. The first is the need to be clear about service conditions, based onexperience or plant data. This is the key to material selection. The seconddecision is the choice of application process for the material. This involvesthe question of compatibility with the coating material; that is, not all ma-terials can be applied by all processes. A further question of compatibilityarises between both material and process with the substrate, for example,whether distortion from high-temperature processes be tolerated. All theseissues are covered in subsequent chapters in this book (see, in particular,Chapters 7 and 8).

References

1. CM. Cotell and J.A. Sprague, Preface, Surface Engineering, VoI 5,ASM Handbook, ASM International, 1994, p v

2. Economic Effects of Metallic Corrosion in the United States, BattelleColumbus Laboratories and the National Institute of Standards andTechnology, 1978 and Battelle updates in 1995

3. "Tribological Sinks in Six Major Industries," Report Number PNL-5535, Sept 1985, Pacific Northwest Laboratory, Richland, WA, oper-ated for the U.S. Department of Energy by Battelle Memorial Institute(NTIS No. DE86000841)

4. DJ. Dunn. Metal Removal Mechanisms Comprising Wear in MineralProcessing, Wear of Materials, K.C. Ludema, Ed., American Societyof Mechanical Engineers, 1985, p 501-508

5. R.L. Pozzo and I. Iwasaki, Pyrite-Pyrrhotite Grinding Media Interac-tions and Their Effects on Media Wear and Flotation, /. Electrochem.Soc, VoI 136 (No. 6), 1989, p 1734-1740

6. R.L. Pozzo and I. Iwasaki, Effect of Pyrite and Pyrrhotite on the Cor-rosive Wear of Grinding Media, Miner. Metall. Process., Aug 1987, p166-171

7. K.A Natarajan, S.C. Riemer, and I. Iwasaki, Influence of Pyrrhotite onthe Corrosive Wear of Grinding Balls in Magnetite Ore Grinding, Int.J. Miner. Process., VoI 13 1984, p 73-81

8. R.L. Pozzo and I. Iwasaki, An Electro-chemical Study of Pyrrhotite-Grinding Media Interaction Under Abrasive Conditions, Corrosion,VoI 43 (No. 3), 1987, p 159-169

9. K.G. Budinski, Surface Engineering for Wear Resistance, Prentice-Hall, Inc., 1988, p 6-10

10. Engineering CoatingsDesign and Application, 2nd ed., S. Graingerand J. Blunt, Ed., Woodhead Publishing Ltd., 1999, p 7

CHAPTER Mm

P r i n c i p l e s o f C o r r o s i o n

CORROSION of metal is a chemical or electrochemical process inwhich surface atoms of a solid metal react with a substance in contact withthe exposed surface. Usually the corroding medium is a liquid substance,but gases and even solids can also act as corroding media. In some in-stances, the corrodent is a bulk fluid; in others, it is a film, droplets, or asubstance adsorbed on or absorbed in another substance.

All structural metals corrode to some extent in natural environments(e.g., the atmosphere, soil, or waters). Bronze, brass, most stainless steels,zinc, and pure aluminum corrode so slowly in service conditions that longservice life is expected without protective coatings. Corrosion of struc-tural grades of cast iron and steel, the 400 series stainless steels, and somealuminum alloys, however, proceeds rapidly unless the metal is protectedagainst corrosion. As described in Chapter 1, corrosion of metals is of par-ticular concern because annual losses in the United States attributed tocorrosion amount to hundreds of billions of dollars.

Although emphasis in this Chapter has been placed on irons and steels,the electrochemical corrosion basics and the forms of corrosion describedare applicable to all metallic materials. For more detailed information onthe corrosion resistance of various metals and their alloys, the readershould consult the selected references listed at the conclusion of thisChapter, as well as Corrosion, VoI 13, of the ASM Handbook or Corro-sion: Understanding the Basics, published by ASM International in 2000.

Electrochemical Corrosion Basics

Electrochemical corrosion in metals in a natural environment, whetheratmosphere, in water, or underground, is caused by a flow of electricityfrom one metal to another, or from one part of a metal surface to anotherpart of the same surface where conditions permit the flow of electricity.

Fig. 1 Simple electrochemical cell showing the components necessary for corrosion

For the flow of energy to take place, either a moist conductor or anelectrolyte must be present. An electrolyte is an electricity-conducting so-lution containing ions, which are atomic particles or radicals bearing anelectrical charge. Charged ions are present in solutions of acids, alkalis,and salts. The presence of an electrolyte is necessary for corrosion tooccur. Water, especially salt water, is an excellent electrolyte.

Electricity passes from a negative area to a positive area through theelectrolyte. For corrosion to occur in metals, one must have (a) an elec-trolyte, (b) an area or region on a metallic surface with a negative charge,(c) a second area with a positive charge, and (d) an electrically conductivepath between (b) and (c). These components are arranged to form a closedelectrical circuit. In the simplest case, the anode would be one metal, suchas iron, the cathode another, perhaps copper, and the electrolyte might ormight not have the same composition at both anode and cathode. Theanode and cathode could be of the same metal under conditions describedlater in this article.

The cell shown in Fig. 1 illustrates the corrosion process in its simplestform. This cell includes the following essential components: (a) a metalanode, (b) a metal cathode, (c) a metallic conductor between the anode andthe cathode, and (d) an electrolyte in contact with the anode and the cath-ode. If the cell were constructed and allowed to function, an electrical cur-rent would flow through the metallic conductor and the electrolyte, and ifthe conductor were replaced by a voltmeter, a potential difference betweenthe anode and the cathode could be measured. The anode would corrode.Chemically, this is an oxidation reaction. The formation of hydrated rediron rust by electrochemical reactions may be expressed as follows:

Metallic conductor betweenthe anode and the cathodeCurrent flow in

conductor

Metal anode

Oxidation reactionoccurs at anode

Metal cathodeOxygen or otherdepolarizer in

electrolyte

Electrolyte, water containingconductive salts

Reduction reactionoccurs at cathode

Current flow throughthe electrolyte

(EqI)

(Eq 2)

During metallic corrosion, the rate of oxidation equals the rate of re-duction. Thus, a nondestructive chemical reaction, reduction, would pro-ceed simultaneously at the cathode. In most cases, hydrogen gas is pro-duced on the cathode. When the gas layer insulates the cathode from theelectrolyte, current flow stops, and the cell is polarized. However, oxygenor some other depolarizing agent is usually present to react with the hy-drogen, which reduces this effect and allows the cell to continue to func-tion.

Contact between dissimilar metallic conductors or differences in theconcentration of the solution cause the difference in potential that resultsin electrical current. Any lack of homogeneity on the metal surface or itsenvironment may initiate attack by causing a difference in potential, andthis results in localized corrosion. The metal undergoing electrochemicalcorrosion need not be immersed in a liquid but may be in contact withmoist soil or may have moist areas on the metal surface.

Corrosive Conditions

If oxygen and water are both present, corrosion will normally occur oniron and steel. Rapid corrosion may take place in water, the rate of corro-sion being accelerated by several factors such as: (a) the velocity or theacidity of the water, (b) the motion of the metal, (c) an increase in tem-perature or aeration, and (d) the presence of certain bacteria. Corrosioncan be retarded by protective layers or films consisting of corrosion prod-ucts or adsorbed oxygen. High alkalinity of the water also retards the rateof corrosion on steel surfaces. Water and oxygen remain the essential fac-tors, however, and the amount of corrosion is generally controlled by oneor the other. For example, corrosion of steel does not occur in dry air andis negligible when the relative humidity of the air is below 30% at normalor lower temperatures. This is the basis for prevention of corrosion by de-humidification.

Water can readily dissolve a small amount of oxygen from the atmos-phere, thus becoming highly corrosive. When the free oxygen dissolved inwater is removed, the water becomes practically noncorrosive unless itbecomes acidic or anaerobic bacteria incite corrosion. If oxygen-freewater is maintained at a neutral pH or at slight alkalinity, it is practically

noncorrosive to structural steel. Steam boilers and water supply systemsare effectively protected by deaerating the water. Additional informationon corrosion in water can be found in Ref 1.

Soils. Dispersed metallic particles or bacteria pockets can provide a nat-ural electrical pathway for buried metal. If an electrolyte is present and thesoil has a negative charge in relation to the metal, an electrical path fromthe metal to the soil will occur, resulting in corrosion. Differences in soilconditions, such as moisture content and resistivity, are commonly re-sponsible for creating anodic and cathodic areas (Fig. 2). Where a differ-ence exists in the concentration of oxygen in the water or in moist soils incontact with metal at different areas, cathodes develop at points of rela-tively high-oxygen concentrations and anodes at points of low concentra-tion. Further information on corrosion in soils is available in Ref 2.

Chemicals. In an acid environment, even without the presence of oxy-gen, the metal at the anode is attacked at a rapid rate. At the cathode,atomic hydrogen is released continuously, to become hydrogen gas. Cor-rosion by an acid can result in the formation of a salt, which slows the re-action because the salt formation on the surface is then attacked.

Corrosion by direct chemical attack is the single most destructive forceagainst steel surfaces. Substances having chlorine or other halogens in theircomposition are particularly aggressive. Galvanized roofing has beenknown to corrode completely within six months of construction, the build-ing being downwind of an aluminum ingot plant where fluorides were al-ways present in the atmosphere. Consequently, galvanized steel should nothave been specified. Selection of materials and evaluation of service con-ditions are extremely important in combating corrosion. The response ofvarious materials to chemical environments is addressed in Ref 3 and 4.

Atmospheric corrosion differs from the corrosion action that occurs inwater or underground, because sufficient oxygen is always present. In at-

pjo 2 A metal pipe buried in moist soil forming a corrosion cell. A difference^* in oxygen content at different levels in the electrolyte will produce a

difference of potential. Anodic and cathodic areas will develop, and a corrosioncell, called a concentration cell, will form.

Oxygen diffusing into earthfrom ground surfaceElectrolyte 1 (soil withground water high in

oxygen content)Current flow

Electrolyte 2 (soil withground water deficient

in oxygen content)Fe2+ (rust)

Anodic area (steelat bottom of pipe)

Buried pipe

Cathodic area (steelat top of pipe)

mospheric corrosion, the formation of insoluble films and the presence ofmoisture and deposits from the atmosphere control the rate of corrosion.Contaminants such as sulfur compounds and salt particles can acceleratethe corrosion rate. Nevertheless, atmospheric corrosion occurs primarilythrough electrochemical means and is not directly caused by chemical at-tack. The anodic and cathodic areas are usually quite small and close to-gether so that corrosion appears uniform, rather than in the form of severepitting, which can occur in water or soil. A more detailed discussion on at-mospheric corrosion can be found in Ref 5.

Forms of Corrosion

The differing forms of corrosion can be divided into the following eightcategories based on the appearance of the corrosion damage or the mech-anism of attack:

Uniform or general corrosionGalvanic corrosionPitting corrosionCrevice corrosion, including corrosion under tubercles or deposits, fil-iform corrosion, and poultice corrosionErosion-corrosion, including cavitation erosion and fretting corrosionIntergranular corrosion, including sensitization and exfoliationDealloyingEnvironmentally assisted cracking, including stress-corrosion crack-ing (SCC), corrosion fatigue, and hydrogen damage (including hydro-gen embrittlement, hydrogen-induced blistering, high-temperature hy-drogen attack, and hydride formation)

Figure 3 illustrates schematically some of the most common forms of cor-rosion. More detailed information pertaining to recognition and preven-tion of these forms of corrosion can be found in Ref 6 and 7.

Uniform CorrosionGeneral Description. Uniform or general corrosion, as the name im-

plies, results in a fairly uniform penetration (or thinning) over the entireexposed metal surface. The general attack results from local corrosion-cellaction; that is, multiple anodes and cathodes are operating on the metalsurface at any given time. The location of the anodic and cathodic areascontinues to move about on the surface, resulting in uniform corrosion.Uniform corrosion often results from atmospheric exposure (especiallypolluted industrial environments); exposure in fresh, brackish, and saltwaters; or exposure in soils and chemicals.

Flg. 3 Schematics of the common forms of corrosion

Metals Affected. All metals are affected by uniform corrosion, al-though materials that form passive films, such as stainless steels or nickel-chromium alloys, are normally subjected to localized forms of attack. Therusting of steel, the green patina formation on copper, and the tarnishingof silver are typical examples of uniform corrosion. In some metals, suchas steel, uniform corrosion produces a somewhat rough surface by re-moving a substantial amount of metal, which either dissolves in the envi-ronment or reacts with it to produce a loosely adherent, porous coating ofcorrosion products. In such reactions as in the tarnishing of silver in air,the oxidation of aluminum in air, or attack on lead in sulfate-containingenvironments, thin, tightly adherent protective films are produced, and themetal surface remains smooth.

Prevention. Uniform corrosion can be prevented or reduced by propermaterials selection, the use of coatings or inhibitors, or cathodic protec-tion. These corrosion prevention methods can be used individually or incombination.

Galvanic CorrosionGeneral Description. The potential available to promote the electro-

chemical corrosion reaction between dissimilar metals is suggested by thegalvanic series, which lists a number of common metals and alloysarranged according to their tendency to corrode when in galvanic contact(Table 1). Metals close to one another on the table generally do not havea strong effect on each other, but the farther apart any two metals are sep-arated, the stronger the corroding effect on the one higher in the list. It ispossible for certain metals to reverse their positions in some environ-ments, but the order given in Table 1 is maintained in natural waters andthe atmosphere. The galvanic series should not be confused with the sim-

Pitting Exfoliation Dealloying lntergranular Stress-corrosioncracking

Corrosionfatigue

Tensile stress Cyclic stress

CreviceFrettingErosionGalvanicUniformNo corrosion

More noblemetal

Flowingcorrodent

Cyclicmovement Metal ornonmetal

Table 1 Galvanic series in seawater at 25 0C (77 0F)Corroded end (anodic, or least noble)MagnesiumMagnesium alloysZincGalvanized steel or galvanized wrought ironAluminum alloys 5052, 3004, 3003, 1100, 6053, in this orderCadmiumAluminum alloys 2117, 2017, 2024, in this orderLow-carbon steelWrought ironCast ironNi-Resist (high-nickel cast iron)Type 410 stainless steel (active)50-50 lead-tin solderType 304 stainless steel (active)Type 316 stainless steel (active)LeadTinCopper alloy C28000 (Muntz metal, 60% Cu)Copper alloy C67500 (manganese bronze A)Copper alloys C46400, C46500, C46600, C46700 (naval brass)Nickel 200 (active)Inconel alloy 600 (active)Hastelloy alloy BChlorimet 2Copper alloy C27000 (yellow brass, 65% Cu)Copper alloys C44300, C44400, C44500 (admiralty brass)Copper albys C60800, C61400 (aluminum bronze)Copper alloy C23000 (red brass, 85% Cu)Copper C! 1000 (ETP copper)Copper alloys C65100, C65500 (silicon bronze)Copper alloy C71500 (copper nickel, 30% Ni)Copper alloy C92300, cast (leaded tin bronze G)Copper alloy C92200, cast (leaded tin bronze M)Nickel 200 (passive)Inconel alloy 600 (passive)Monel alloy 400Type 410 stainless steel (passive)Type 304 stainless steel (passive)Type 316 stainless steel (passive)Incoloy alloy 825Inconel alloy 625Hastelloy alloy CChlorimet 3SilverTitaniumGraphiteGoldPlatinum

Protected end (cathodic, or most noble)

ilar electromotive force series, which shows exact potentials based onhighly standardized conditions that rarely exist in nature.

The three-layer iron oxide scale formed on steel during rolling varieswith the operation performed and the rolling temperature. The dissimilar-ity of the metal and the scale can cause corrosion to occur, with the steelacting as the anode in this instance. Unfortunately, mill scale is cathodicto steel, and an electric current can easily be produced between the steeland the mill scale. This electrochemical action will corrode the steel with-out affecting the mill scale (Fig. 4).

A galvanic couple may be the cause of premature failure in metal com-ponents of water-related structures or may be advantageously exploited.

Fig, 4 Mill scale forming a corrosion cell on steel

Galvanizing iron sheet is an example of useful application of galvanic ac-tion or cathodic protection. Iron is the cathode and is protected against cor-rosion at the expense of the sacrificial zinc anode. Alternatively, a zinc ormagnesium anode may be located in the electrolyte close to the structureand may be connected electrically to the iron or steel. This method is re-ferred to as cathodic protection of the structure. Iron or steel can become theanode when in contact with copper, brass, or bronze; however, iron or steelcorrode rapidly while protecting the latter metals. Also, weld metal may beanodic to the basis metal, creating a corrosion cell when immersed (Fig. 5).

While the galvanic series (Table 1) represents the potential available topromote a corrosive reaction, the actual corrosion is difficult to predict.Electrolytes may be poor conductors, or long distances may introducelarge resistance into the corrosion cell circuit. More frequently, scale for-mation forms a partially insulating layer over the anode. A cathode hav-ing a layer of adsorbed gas bubbles, as a consequence of the corrosion cellreaction, is polarized. The effect of such conditions is to reduce the theo-retical consumption of metal by corrosion. The area relationship betweenthe anode and cathode may also strongly affect the corrosion rate; a highratio of cathode area to anode area produces more rapid corrosion. In thereverse case, the cathode polarizes, and the corrosion rate soon drops to anegligible level.

The passivity of stainless steels is attributed to either the presence of acorrosion-resistant oxide film or an oxygen-caused polarizing effect,

FlC, 5 Weld metal forming a corrosion cell on steel. Weld metal may be an-^* odic to steel, creating a corrosion cell when immersed.

Electrolyte (water) (rust)Current flow

Cathode (steel)Anode(weld metal) \

Electrolyte (water)(rust) Cathode(broken mill

scale)Current flow

Anode(steel)

durable only as long as there is sufficient oxygen to maintain the effect,over the surfaces. In most natural environments, stainless steels will re-main in a passive state and thus tend to be cathodic to ordinary iron andsteel. Change to an active state usually occurs only where chloride con-centrations are high, as in seawater or reducing solutions. Oxygen starva-tion also produces a change to an active state. This occurs where the oxy-gen supply is limited, as in crevices and beneath contamination onpartially fouled surfaces.

Prevention. Galvanic corrosion can be prevented or reduced by propermaterials selection (i.e., select combinations of metals as close together aspossible in the galvanic series), insulating dissimilar metals, applying abarrier coating to both the anodic (less noble) and cathodic (noble) metal,applying a sacrificial coating (aluminum, zinc, or cadmium) to the ca-thodic part, applying nonmetallic films (e.g., anodizing aluminum alloys),and by providing cathodic protection.

PittingGeneral Description. Pitting is a type of localized cell corrosion. It is

predominantly responsible for the functional failure of iron and steelwater-related installations. Pitting may result in the perforation of waterpipe, rendering it unserviceable, even though less than 5% of the totalmetal has been lost through rusting. Where confinement of water is not afactor, pitting causes structural failure from localized weakening whileconsiderable sound metal still remains.

Pitting develops when the anodic or corroding area is small in relationto the cathodic or protected area. For example, pitting can occur wherelarge areas of the surface are covered by mill scale, applied coatings, ordeposits of various kinds and where breaks exist in the continuity of theprotective coating. Pitting may also develop on bare, clean metal surfacesbecause of irregularities in the physical or chemical structure of the metal.Localized, dissimilar soil conditions at the surface of steel can also createconditions that promote pitting.

Electrical contact between dissimilar materials or concentration cells(areas of the same metal where oxygen or conductive salt concentrationsin water differ) accelerates the rate of pitting. In closed-vessel structures,these couples cause a difference of potential that results in an electric cur-rent flowing through the water or across the moist steel from the metal-lic anode to a nearby cathode. The cathode may be copper, brass, millscale, or any portion of a metal surface that is cathodic to the more activemetal areas. In practice, mill scale is cathodic to steel and is found to bea common cause of pitting. The difference of potential generated betweensteel and mill scale often amounts to 0.2 to 0.3 V. This couple is nearlyas powerful a generator of corrosion currents as is the copper-steel cou-ple. However, when the anodic area is relatively large compared with the

cathodic area, the damage is spread out and usually negligible, but whenthe anode is relatively small, the metal loss is concentrated and may bevery serious.

On surfaces having some mill scale, the total metal loss is nearly con-stant as the anode is decreased, but the degree of penetration increases.Figure 4 shows how a pit forms where a break occurs in mill scale. Whencontact between dissimilar materials is unavoidable and the surface ispainted, it is preferred to paint both materials. If only one surface ispainted, it should be the cathode. If only the anode is coated, any weakpoints such as pinholes or holidays in the coating will probably result inintense pitting.

As a pit, perhaps at a break in mill scale, becomes deeper, an oxygenconcentration cell is started by depletion of oxygen in the pit. The rate ofpenetration by such pits is accelerated proportionately as the bottom of thepit becomes more anodic. Fabrication operations may crack mill scale andresult in accelerated corrosion.

Metals Affected. Pitting occurs in most commonly used metals and al-loys. Iron buried in the soil corrodes with the formation of shallow pits,but carbon steels in contact with hydrochloric acid or stainless steels im-mersed in seawater characteristically corrode with the formation of deeppits. Aluminum tends to pit in waters containing chloride ions (for exam-ple, at stagnant areas), and aluminum brasses are subject to pitting in pol-luted waters.

Despite their good resistance to general corrosion, stainless steels aremore susceptible to pitting than many other metals. High-alloy stainlesssteels containing chromium, nickel, and molybdenum are also more re-sistant to pitting but are not immune under all service conditions.

Pitting failures of corrosion-resistant alloys, such as Hastelloy C,Hastelloy G, and Incoloy 825, are relatively uncommon in solutions thatdo not contain halides, although any mechanism that permits the estab-lishment of an electrolytic cell in which a small anode is in contact with alarge cathodic area offers the opportunity for pitting attack.

Prevention. Typical approaches to alleviating or minimizing pittingcorrosion include the following:

Use defect-free barrier coatingsReduce the aggressiveness of the environment, for example, chlorideion concentrations, temperature, acidity, and oxidizing agentsUpgrade the materials of construction, for example, use molybdenum-containing (4 to 6% Mo) stainless steels, molybdenum + tungstennickel-base alloys, overalloy welds, and use corrosion-resistant alloyliningsModify the design of the system, for example, avoid crevices and theformation of deposits, circulate/stir to eliminate stagnant solutions,and ensure proper drainage

Crevice CorrosionGeneral Description. Crevice corrosion is a form of localized attack

that occurs at narrow openings or spaces (gaps) between metal-to-metal ornonmetal-to-metal components. This type of attack results from a con-centration cell formed between the electrolyte within the crevice, which isoxygen starved, and the electrolyte outside the crevice, where oxygen ismore plentiful. The material within the crevice acts as the anode, and theexterior material becomes the cathode.

Crevices may be produced by design or accident. Crevices caused bydesign occur at gaskets, flanges, rubber O-rings, washers, bolt holes,rolled tube ends, threaded joints, riveted seams, overlapping screen wires,lap joints, beneath coatings (filiform corrosion) or insulation (poulticecorrosion), and anywhere close-fitting surfaces are present. Figure 6shows crevice corrosion in a riveted assembly caused by concentrationcells. Occluded regions are also formed under tubercles (tuberculation),deposits (deposit corrosion), and below accumulations or biological ma-terials (biologically influenced corrosion). Similarly, unintentionalcrevices such as cracks, seams, and other metallurgical defects could serveas sites for corrosion.

Metals Affected. Resistance to crevice corrosion can vary from onealloy-environment system to another. Although crevice corrosion affectsboth active and passive metals, the attack is often more severe for passivealloys, particularly those in the stainless steel group. Breakdown of thepassive film within a restricted geometry leads to rapid metal loss andpenetration of the metal in that area.

Low metal ion concentration Metal ion concentration cell

High metal ion concentration

Oxygen concentration cellHigh oxygen concentration

Low oxygen concentrationFig. 6 Corrosion caused at crevices by concentration cells. Both types of con-

centration cells shown sometimes occur simultaneously as in a reentryangle in a riveted seam.

Prevention. Crevice corrosion can be prevented or reduced through im-proved design to avoid crevices, regular cleaning to remove deposits, byselecting a more corrosion-resistant material, and by coating carbon steelor cast iron components with epoxy or other field-applied or factory-applied organic coatings.

Erosion-CorrosionGeneral Description. Erosion-corrosion is the acceleration or increase

in the rate of deterioration or attack on a metal because of mechanicalwear or abrasive contributions in combination with corrosion. The combi-nation of wear or abrasion and corrosion results in more severe attack thanwould be realized with either mechanical or chemical corrosive actionalone. Metal is removed from the surface as dissolved ions, as particles ofsolid corrosion products, or as elemental metal. The spectrum of erosion-corrosion ranges from primarily erosive attack, such as sandblasting, fil-ing, or grinding of a metal surface, to primarily corrosion failures, wherethe contribution of mechanical action is quite small.

All types of corrosive media generally can cause erosion-corrosion, in-cluding gases, aqueous solutions, organic systems, and liquid metals. Forexample, hot gases may oxidize a metal then at high velocity blow off anotherwise protective scale. Solids in suspension in liquids (slurries) areparticularly destructive from the standpoint of erosion-corrosion.

Erosion-corrosion is characterized in appearance by grooves, waves,rounded holes, and/or horseshoe-shaped grooves. Analysis of these markscan help determine the direction of flow. Affected areas are usually free ofdeposits and corrosion products, although corrosion products can some-times be found if erosion-corrosion occurs intermittently and/or the liquidflow rate is relatively low.

Metals Affected. Most metals are susceptible to erosion-corrosionunder specific conditions. Metals that depend on a relatively thick protec-tive coating of corrosion product for corrosion resistance are frequentlysubject to erosion-corrosion. This is due to the poor adhesion of thesecoatings relative to the thin films formed by the classical passive metals,such as stainless steels and titanium. Both stainless steels and titanium arerelatively immune to erosion-corrosion in many environments. Metals that

Water flowImpingementcorrosion pits Original metalsurfaceCorrosion film

Metal tube wall

Fig, 7 Schematic of erosion-corrosion of a condenser tube

are soft and readily damaged or worn mechanically, such as copper andlead, are quite susceptible to erosion-corrosion. Even the noble or pre-cious metals, such silver, gold, and platinum, are subject to erosion-cor-rosion. Figure 7 shows a schematic of erosion-corrosion of a condensertube wall. The direction of flow and the resulting attack where the protec-tive film on the tube has broken down are indicated.

Prevention. Erosion-corrosion can be prevented or reduced through im-proved design (e.g., increase pipe diameter and/or streamline bends to re-duce impingement effects), by altering the environment (e.g., deaeration andthe addition of inhibitors), and by applying hard, tough protective coatings.

CavitationGeneral Description. Cavitation is a form of erosion-corrosion that is

caused by the formation and collapse of vapor bubbles in a liquid againsta metal surface. Cavitation occurs in hydraulic turbines, on pump im-pellers, on ship propellers, and on many surfaces in contact with high-ve-locity liquids subject to changes in pressure. The appearance of cavitationis similar to pitting except that surfaces in the pits are usually muchrougher. The affected region is free of deposits and accumulated corrosionproducts if cavitation has been recent.

Figure 8 is a simplified representation of the cavitation process. Figure8(a) shows a vessel containing a liquid. The vessel is closed by an airtightplunger. When the plunger is withdrawn (Fig. 8b), a partial vacuum iscreated above the liquid, causing vapor bubbles to form and grow within

Partialvacuum

Pressurized

(a) RestQuiescent liquid

at standardtemperatureand pressure

(b) ExpansionLiquid boiling

at roomtemperature

(c) CompressionCollapse of

vapor bubbles

Metal

(d)

Metaloxide

Approachingmicrojettorpedo

Destruction ofmetal oxideon impact

Repair ofmetal oxide at

expense of metal

P J o- 8 Schematic representation of cavitation showing a cross section through a vessel and plunger enclosing a f luid.

" (a) Plunger stationary, l iquid at standard temperature and pressure, (b) Plunger withdrawn, l iquid boils at roomtemperature, (c) Plunger advanced, bubbles collapse, (d) Disintegration of protective corrosion product by impacting mi-crojet "torpedo." Source: Ref 8

SurfaceOxide

Fig, 9 Schematic of the fretting process

BareMetalMetal andOxide Debris

the liquid. In essence, the liquid boils without a temperature increase. Ifthe plunger is then driven toward the surface of the liquid (Fig. 8c), thepressure in the liquid increases, and the bubbles condense and collapse(implode). In a cavitating liquid, these three steps occur in a matter of mil-liseconds. As shown in Fig. 8(d), implosion of a vapor bubble creates amicroscopic "torpedo" of water that is ejected from the collapsing bubbleat velocities that may range from 100 to 500 m/s (330 to 1650 ft/s). Whenthe torpedo impacts the metal surface, it dislodges protective surface filmsand/or locally deforms the metal itself. Thus, fresh surfaces are exposedto corrosion and the reformation of protective films, which is followed bymore cavitation, and so on. Damage occurs when the cycle is allowed torepeat over and over again.

Prevention. Cavitation can be controlled or minimized by improving de-sign to minimize hydrodynamic pressure differences, employing stronger(harder) and more corrosion-resistant materials, specifying a smooth finishon all critical metal surfaces, and coating with resilient materials such asrubber and some plastics.

Fretting CorrosionGeneral Description. Fretting corrosion is a combined wear and cor-

rosion process in which material is removed from contacting surfaceswhen motion between the surfaces is restricted to very small amplitudeoscillations (often, the relative movement is barely discernible). Usually,the condition exists in machine components that are considered fixed andnot expected to wear. Pressed-on wheels can often fret at the shaft/wheelhole interface.

Oxidation is the most common element in the fretting process. In oxi-dizing systems, fine metal particles removed by adhesive wear are oxi-dized and trapped between the fretting surfaces (Fig. 9). The oxides actlike an abrasive (such as lapping rouge) and increase the rate of materialremoval. This type of fretting in ferrous alloys is easily recognized by thered material oozing from between the contacting surfaces.

Fretting corrosion takes the form of local surface dislocations and deeppits. These occur in regions where slight relative movements have oc-curred between mating, highly loaded surfaces.

Prevention. Fretting corrosion can be controlled by lubricating (e.g.,low-viscosity oils) the faying surfaces, restricting the degree of movement,shot peening (rough surfaces are less prone to fretting damage), surfacehardening (e.g., carburizing and nitriding), anodizing of aluminum alloys,phosphate conversion coating of steels, and by applying protective coat-ings by electrodeposition (e.g., gold or silver plating), plasma spraying, orvapor deposition (Ref 9).

lntergranular CorrosionGeneral Description. lntergranular corrosion is defined as the selec-

tive dissolution of grain boundaries, or closely adjacent regions, withoutappreciable attack of the grains themselves. This dissolution is caused bypotential differences between the grain-boundary region and any precipi-tates, intermetallic phases, or impurities that form at the grain boundaries.The actual mechanism differs with each alloy system. Although a wide va-riety of alloy systems are susceptible to intergranular corrosion under veryspecific conditions, the majority of case histories reported in the literaturehave involved austenitic stainless steels and aluminum alloys and, to alesser degree, some ferritic stainless steels and nickel-base alloys.

Precipitates that form as a result of the exposure of metals at elevatedtemperatures (for example, during production, fabrication, and welding)often nucleate and grow preferentially at grain boundaries. If these pre-cipitates are rich in alloying elements that are essential for corrosion re-sistance, the regions adjacent to the grain boundary are depleted of theseelements. The metal is thus sensitized and is susceptible to intergranularattack in a corrosive environment. For example, in austenitic stainlesssteels such as AISI type 304, the cause of intergranular attack is the pre-cipitation of chromium-rich carbides ((Cr5Fe)23C6) at grain boundaries.These chromium-rich precipitates are surrounded by metal that is depletedin chromium; therefore, they are more rapidly attacked at these zones thanon undepleted metal surfaces.

Impurities that segregate at grain boundaries may promote galvanic ac-tion in a corrosive environment by serving as anodic or cathodic sites.Therefore, this would affect the rate of the dissolution of the alloy matrixin the vicinity of the grain boundary. An example of this is found in alu-minum alloys that contain intermetallic compounds, such as Mg5Al8 andCuAl2, at the grain boundaries. During exposures to chloride solutions,the galvanic couples formed between these precipitates and the alloy ma-trix can lead to severe intergranular attack. Susceptibility to intergranularattack depends on the corrosive solution and on the extent of intergranu-lar precipitation, which is a function of alloy composition, fabrication, andheat treatment parameters.

Prevention. Susceptibility to intergranular corrosion in austenitic stain-less steels can be avoided by controlling their carbon contents or by

adding elements (titanium and niobium) whose carbides are more stablethan those of chromium. For most austenitic stainless steels, restrictingtheir carbon contents to 0.03% or less will prevent sensitization duringwelding and most heat treatment.

Intergranular corrosion in aluminum alloys is controlled by material se-lection (e.g., the high-strength Ixxx and Ixxx alloys are the most suscep-tible) and by proper selection of thermal (tempering) treatments that caneffect the amount, size, and distribution of second-phase intermetallic pre-cipitates. Resistance to intergranular corrosion is obtained by the use ofheat treatments that cause precipitation to be more general throughout thegrain structure (Ref 10).Exfoliation

General Description. Exfoliation is a form of macroscopic intergran-ular corrosion that primarily affects aluminum alloys in industrial or ma-rine environments. Corrosion proceeds laterally from initiation sites onthe surface and generally proceeds intergranularly along planes parallel tothe surface. The corrosion products that form in the grain boundaries forcemetal away from the underlying base material, resulting in a layered orflakelike appearance (see, for example, the schematic shown in Fig. 3).

Prevention. Resistance to exfoliation corrosion is attained throughproper alloy and temper selection. The most susceptible alloys are thehigh-strength heat-treatable Ixxx and Ixxx alloys. Exfoliation corrosion inthese alloys is usually confined to relatively thin sections of highlyworked products. Guidelines for selecting proper heat treatment for thesealloys can be found in Ref 10.

Dealloying CorrosionGeneral Description. Dealloying, also referred to as selective leaching

or parting corrosion, is a corrosion process in which the more active metalis selectively removed from an alloy, leaving behind a porous weak de-posit of the more noble metal. Specific categories of dealloying oftencarry the name of the dissolved element. For example, the preferentialleaching of zinc from brass is called dezincification. If aluminum is re-moved, the process is called dealuminification, and so forth. In the case ofgray iron, dealloying is called graphitic corrosion.

In the dealloying process, typically one of two mechanisms occurs:alloy dissolution and replating of the cathodic element or selective disso-lution of an anodic alloy constituent. In either case, the metal is leftspongy and porous and loses much of its strength, hardness, and ductility.Table 2 lists some of the alloy-environment combinations for which deal-loying has been reported. By far the two most common forms of dealloy-ing are dezincification and graphitic corrosion.

Copper-zinc alloys containing more than 15% zinc are susceptible todezincification. In the dezincification of brass, selective removal of zinc

leaves a relatively porous and weak layer of copper and copper oxide.Corrosion of a similar nature continues beneath the primary corrosionlayer, resulting in gradual replacement of sound brass by weak, porouscopper.

Graphitic corrosion is observed in gray cast irons in relatively mild en-vironments in which selective leaching of iron leaves a graphite network.Selective leaching of the iron takes place because the graphite is cathodicto iron, and the gray iron structure establishes an excellent galvanic cell.

Prevention. Dezincification can be prevented by alloy substitution.Brasses with copper contents of 85% or more resist dezincification. Somealloying elements also inhibit dezincification (e.g., brasses containing 1%tin). Where dezincification is a problem, red brass, commercial bronze, in-hibited admiralty metal, and inhibited brass can be successfully used.

Attack by graphitic corrosion is reduced by alloy substitution (e.g., useof a ductile or alloyed iron rather than gray iron), altering the environment(raise the water pH to neutral or slightly alkaline levels), the use of in-hibitors, and avoiding stagnant water conditions.

Stress-Corrosion CrackingGeneral Description. Stress-corrosion cracking (SCC) is a cracking

phenomenon that occurs in susceptible alloys and is caused by the con-joint action of a surface tensile stress and the presence of a specific cor-rosive environment. For SCC to occur on an engineering structure, threeconditions must be met simultaneously, namely, a specific crack-promot-ing environment must be present, the metallurgy of the material must besusceptible to SCC, and the tensile stresses must be above some thresholdvalue. Stresses required to cause SCC are small, usually below the macro-scopic yield stress. The stresses can be externally applied, but residualstresses often cause SCC failures. This cracking phenomenon is of partic-ular importance to users of potentially susceptible structural alloys be-cause SCC occurs under service conditions that can result, often with nowarning, in catastrophic failure. Failed specimens exhibit highly branched

Table 2 Combinations of alloys and environments subject to dealloying and elements preferentiallyremovedAlloy

BrassesGray ironAluminum bronzesSilicon bronzesTin bronzesCopper-gold single crystalsMonelsGold alloys with copper or silverTungsten carbide-cobaltHigh-nickel alloysMedium- and high-carbon steelsIron-chromium alloysNickel-molybdenum alloys

Environment

Many waters, especially under stagnant conditionsSoils, many watersHydrofluoric acid, acids containing chloride ionsHigh-temperature steam and acidic speciesHot brine or steamFerric chlorideHydrofluoric and other acidsSulfide solutions, human salivaDeionized waterMolten saltsOxidizing atmospheres, hydrogen at high temperaturesHigh-temperature oxidizing atmospheresOxygen at high temperature

Element removedZinc (dezincification)Iron (graphitic corrosion)Aluminum (dealuminification)Silicon (desiliconification)Tin (destannification)CopperCopper in some acids, and nickel in othersCopper, silverCobaltChromium, iron, molybdenum, and tungstenCarbon (decarburization)Chromium, which forms a protective filmMolybdenum

Table 3 Some environment-alloy combinations known to result in stress-corrosion cracking (SCC)

Environment

Alloy system

Aluminumalloys

Carbonsteels

Copperalloys

Nickelalloys

Stainless SteelsAustenitic Duplex Martensitic

Titaniumalloys

Zirconiumalloys

Amines, aqueousAmmonia, anhydrousAmmonia, aqueousBromineCarbonates, aqueousCarbon monoxide, carbon

dioxide, water mixtureChlorides, aqueousChlorides, concentrated,

boilingChlorides, dry, hotChlorinated solventsCyanides, aqueous,

acidifiedFluorides, aqueousHydrochloric acidHydrofluoric acidHydroxides, aqueousHydroxides, concentrated,

hotMethanol plus halidesNitrates, aqueousNitric acid, concentratedNitric acid, fumingNitrites, aqueousNitrogen tetroxidePolythionic acidsSteamSulfides plus chlorides,

aqueousSulfurous acidWater, high-purity, hot

X, known to result in SCC

Stress-corrosion cracking control

Mechanical Metallurgical Environmental

Avoid stressconcentrators

Change alloycomposition

Modifyenvironment

Relieve fabricationstresses

Change alloystructure

Apply anodic orcathodic protection

Introduce surfacecompressfve

stresses

Use metallicor conversion

coatingAdd inhibrtor

Reduce operatingstresses

Use organiccoating

Nondestructivetesting implications

for design

Modifytemperature F i g . 1 0 Me thods used to control SCC. Source:

Ket I I

cracks (see Fig. 3) that propagate intergranularly and/or transgranularly,depending on the metal-environment combination.

Table 3 lists some of the alloy-environment combinations that result inSCC. This table, as well as others published in the literature, should beused only as a guide for screening candidate materials prior to further in-depth investigation, testing, and evaluation.

Prevention. Figure 10 summarizes the various approaches to control-ling SCC. Surface engineering treatments like shot peening, metallic coat-ings, and organic coatings play a key role in controlling SCC.

Corrosion FatigueGeneral Description. Corrosion fatigue is a term that is used to de-

scribe the phenomenon of cracking, including both initiation and propa-gation, in materials under the combined actions of a fluctuating or cyclicstress and a corrosive environment. Corrosion fatigue depends strongly onthe interactions among the mechanical (loading), metallurgical, and envi-ronmental variables listed in Table 4.

Corrosion fatigue produces fine-to-broad cracks with little or no branch-ing (see Fig. 3); thus, they differ from SCC, which often exhibits consid-erable branching. They are typically filled with dense corrosion product.The cracks may occur singly but commonly appear as families or parallelcracks. They are frequently associated with pits, grooves, or some otherform of stress concentrator. Transgranular fracture paths are more com-mon than intergranular fractures.

Table 4 Mechanical, metallurgical, and environmental variables thatinfluence corrosion fatigue behaviorVariable TypeMechanical Maximum stress or stress-intensity factor, amax or Kmax

Cyclic stress or stress-intensity range, ACT or AKStress ratio, RCyclic loading frequencyCyclic load waveform (constant-amplitude loading)Load interactions in variable-amplitude loadingState of stressResidual stressCrack size and shape, and their relation to component size and geometry

Metallurgical Alloy compositionDistribution of alloying elements and impuritiesMicrostructure and crystal structureHeat treatmentMechanical workingPreferred orientation of grains and grain boundaries (texture)Mechanical properties (strength, fracture toughness, etc.)

Environmental TemperatureTypes of environments: gaseous, liquid, liquid metal, etc.Partial pressure of damaging species in gaseous environmentsConcentration of damaging species in aqueous or other liquid environmentsElectrical potentialpHViscosity of the environmentCoatings, inhibitors, etc.

Prevention. All metals and alloys are susceptible to corrosion fatigue.Even some alloys that are immune to SCC, for example, ferritic stain-less steels, are subject to failure by corrosion fatigue. Both temporaryand permanent solutions for corrosion involve reducing or eliminatingcyclic stresses, selecting a material or heat treatment with higher corro-sion fatigue strengths, reducing or eliminating corrosion, or a combina-tion of these procedures. These objectives are accomplished by changesin material, design, or environment and by the application of surfacetreatments. Shot peening, nitriding of steels, and organic coatings cansuccessfully impede corrosion fatigue. Noble metal coatings (e.g.,nickel) can be effective, but only if they remain unbroken and are of suf-ficient density and thickness. The relatively low corrosion-fatiguestrength of carbon steel is reduced still further when local breaks in acoating occur.

Hydrogen DamageGeneral Description. The term hydrogen damage has been used to

designate a number of processes in metals by which the load-carrying ca-pacity of the metal is reduced due to the presence of hydrogen, often incombination with residual or applied tensile stresses. Although it occursmost frequently in carbon and low-alloy steels, many metals and alloysare susceptible to hydrogen damage. Hydrogen damage in one form or an-other can severely restrict the use of certain materials.

Because hydrogen is one of the most abundant elements and is readilyavailable during the production, processing, and service of metals, hydro-gen damage can develop in a wide variety of environments and circum-stances. The interaction between hydrogen and metals can result in theformation of solid solutions of hydrogen in metals, molecular hydrogen,gaseous products that are formed by reactions between hydrogen and ele-ments constituting the alloy, and hydrides. Depending on the type of hy-drogen/metal interaction, hydrogen damage of metal manifests itself inone of several ways.

Specific types of hydrogen damage, some of which occur only in spe-cific alloys under specific conditions include:

Hydrogen embrittlement: Occurs most often in high-strength steels,primarily quenched-and-tempered and precipitation-hardened steels,with tensile strengths greater than about 1034 MPa (150 ksi). Hydro-gen sulfide is the chief embrittling environment.Hydrogen-induced blistering: Also commonly referred to as hydro-gen-induced cracking (HIC), it occurs in lower-strength (unhardened)steels, typically with tensile strengths less than about 550 MPa (80ksi). Line pipe steels used in sour gas environments are susceptible toHIC.

As described in the previous section, surface treatments, and in particu-lar protective coatings, are widely used to control corrosion in its varyingforms. The problems of corrosion should be approached in the designstage, and the selection of a protective coating is important. Paint systemsand lining materials exist that slow the corrosion rate of carbon steel sur-faces. High-performance organic coatings such as epoxy, polyesters,polyurethanes, vinyl, or chlorinated rubber help to satisfy the need for cor-rosion prevention. Special primers are used to provide passivation, gal-vanic protection, corrosion inhibition, or mechanical or electrical barriersto corrosive action.

Corrosion Inhibitors. A water-soluble corrosion inhibitor reduces gal-vanic action by making the metal passive or by providing an insulatingfilm on the anode, the cathode, or both. A very small amount of chromate,polyphosphate, or silicate added to water creates a water-soluble inhibitor.A slightly soluble inhibitor incorporated into the prime coat of paint mayalso have a considerable protective influence. Inhibitive pigments in paintprimers are successful inhibitors except when they dissolve sufficiently toleave holes in the paint film. Most paint primers contain a partially solu-ble inhibitive pigment such as zinc chromate, which reacts with the steel

Cracking from precipitation of internal hydrogen: Examples includeshatter cracks, flakes, and fish eyes found in steel forgings, weld-ments, and castings. During cooling from the melt, hydrogen diffusesand precipitates in voids and discontinuities.Hydrogen attack: A high-pressure, high-temperature form of hydro-gen damage. Commonly experienced in steels used in petrochemicalplant equipment that often handles hydrogen and hydrogen-hydrocar-bon streams at pressures as high as 21 MPa (3 ksi) and temperaturesup to 540 0C (1000 0F)Hydride formation: Occurs when excess hydrogen is picked up duringmelting or welding of titanium, tantalum, zirconium, uranium, andthorium. Hydride particles cause significant loss in strength and largelosses in ductility and toughness.

Prevention. The primary factors controlling hydrogen damage are ma-terial, stress, and environment. Hydrogen damage can often be preventedby using more resistant material, changing the manufacturing processes,modifying the design to lower stresses, or changing the environment. In-hibitors and post-processing bake-out treatments can also be used. Bakingof electroplated high-strength steel parts reduces the possibility of hydro-gen embrittlement (see Chapter 8 for additional information).

Coatings and Corrosion Prevention

substrate to form the iron salt. The presence of these salts slows corrosionof steel. Chromates, phosphates, molybdates, borates, silicates, andplumbates are commonly used for this purpose. Some pigments add alka-linity, slowing chemical attack on steel. Alkaline pigments, such as metab-orates, cement, lime, or red lead, are effective, provided that the environ-ment is not too aggressive. In addition, many new pigments have beenintroduced to the paint industry such as zinc phosphosilicate and zincflake.

Barrier coatings are used to prevent the electrolyte from reaching thecomponent surface. Examples of barrier coatings include painted steelstructures, steels lined with thick acid-proof brick, steels lined with rub-berlike materials, or steels electroplated with a noble (see Table 1) metal(e.g., chromium, copper, or nickel). Protection is effective until the coat-ing is penetrated, either by a pit, pore, crack, or by damage or wear. Thesubstrate will then corrode preferentially to the coating (since it is anodicto the coating material), and corrosion products will lift off the coatingand allow further attack (Fig. 11).

Generally, electroplated coatings that are completely free of pores andother discontinuities are not commercially feasible. Pits eventually format coating flaws, and the coating is penetrated. The resulting corrosion cellis shown in Fig. 12. The substrate exposed at the bottom of the resultingpit corrodes rapidly. A crater forms in the substrate, and because of the

p jo "I \ Illustration of the mechanism of corrosion for painted steel, (a) A void" in the paint results in rusting of the steel, which undercuts the paint

coating and results in further coating degradation, (b) Photograph showing blis-tering and/or peeling (undercutting) of paint where exposed steel is rusting.

(b)

(a)

PaintSteel

Rust

f\a 1 2 Crater formation in a steel substrate beneath a void in a noble metal" coating, for example, passive chromium or copper. Corrosion pro-

ceeds under the noble metal, the edges of which collapse into the corrosion pit.

Noble metal coating(cathode)Moist air

Steel substrate(anode)

Coating (M1)Water drop

Coating (M2)Substrate (M3)FlC, 1 3 Corrosion pit formation in a substrate beneath a void in a duplex

^* noble metal coating. The top coating layer (M1) is cathodic to thecoating underlayer (M2), which is in turn cathodic to the substrate (M3). As inFig. 12, the coating tends to collapse into the pit.

large area ratio between the more noble coating and the anodic crater, thecrater becomes anodic, and high corrosion current density results. Elec-trons flow from the substrate to the coating as the steel dissolves. Hydro-gen ions (H+) in the moisture accept the electron and, with dissolved oxy-gen, form water at the noble metal surface near the void. Use of anintermediate coating that is less noble than a surface coating but morenoble than the base metal can result in the mode of corrosion shown inFig. 13. This would be typical of a costume jewelry item with a brass sub-strate, an intermediate nickel coating, and a tarnish-resistant gold top coat.It is also exemplified by nickel-chromium coating systems.

Sacrificial coatings, which corrode preferentially to the substrate, in-clude zinc, aluminum, cadmium, and zinc-rich paints. Initially these sac-rificial coatings will corrode, but their corrosion products are protectiveand the coating acts as a barrier layer. If the coating is damaged or defec-tive, it remains protective as it is the coating that suffers attack and not thesubstrate. Figure 14 shows the sacrificial (galvanic) protection offered bya zinc coating to a steel substrate.

Cathodic protection involves the reversal of electric current flowwithin the corrosion cell. Cathodic protection can reduce or eliminatecorrosion by connecting a more active metal to a metal that must be

FlC. 14 Principles and mechanism of galvanic protection of a substrate by a^* coating. Galvanic protection of a steel substrate at a void in a zinc

coating. Corrosion of the substrate is light and occurs at some distance from thezinc.

protected. The use of cathodic protection to reduce or eliminate corrosionis a successful technique of long-standing use in marine structures,pipelines, bridge decks, sheet piling, and equipment and tankage of alltypes, particularly below water or underground. Typically, zinc or magne-sium anodes are used to protect steel in marine environments, and the an-odes are replaced after they are consumed.

Cathodic protection uses an impressed direct current (dc) supplied byany low output voltage source and a relatively inert anode. As is the casein all forms of cathodic activity, an electrolyte is needed for current flow.Cathodic protection and the use of protective coatings are most often em-ployed jointly, especially in marine applications and on board ships whereimpressed current inputs do not usually exceed 1 V. Beyond 1 V, manycoating systems tend to disbond. Current source for cathodic protection insoils is usually 1.5 to 2 V.

Choice of anodes for buried steel pipe depends on soil conditions. Mag-nesium is most commonly used for galvanic anodes; however, zinc canalso be used. Galvanic anodes are seldom used when the resistivity of thesoil is over 30 fl m (3000 ft cm); impressed current is normally usedfor these conditions. Graphite, high-silicon cast iron, scrap iron, alu-minum, and platinum are used as anodes with impressed current. Theavailability of low-cost power is often the deciding factor in choosing be-tween galvanic or impressed current cathodic protection. Figure 15 illus-trates both types of galvanic protection systems.

Protective coatings are normally used in conjunction with cathodic pro-tection and should not be disregarded where cathodic protection is con-templated in new construction. Because the cathodic protection currentmust protect only the bare or poorly insulated areas of the surface, coat-ings that are highly insulating, very durable, and free of discontinuitieslower the current requirements and system costs. A good coating also en-ables a single-impressed current installation to protect many miles of pip-ing. Coal-tar enamel, epoxy powder coatings, and vinyl resin are exam-

steel substrate(cathode)

Water drop Zinc coating(anode)

Fig, 1 5 Cathodic protection for underground pipe, (a) Sacrificial or galvanic anode, (b) Impressed-cur-^* rent anode, ac, alternating current

pies of coatings that are most suitable for use with cathodic protection.Certain other coatings may be incompatible, such as phenolic coatings,which may deteriorate rapidly in the alkaline environment created by thecathodic protection currents. Although cement mortar initially conductsthe electrical current freely, polarization, the formation of an insulatingfilm on the surface as a result of the protective current, is believed to re-duce the current requirement moderately.

Cathodic protection is used increasingly to protect buried or submergedmetal structures in the oil, gas, and waterworks industries and can be usedin specialized applications, such as for the interiors of water storage tanks.Pipelines are routinely designed to ensure the electrical continuity neces-sary for effective functioning of the cathodic protection system. Thus,electrical connections or bonds are required between pipe sections in linesusing mechanically coupled joints, and insulating couplings may be em-ployed at intervals to isolate some parts of the line electrically from otherparts. Leads may be attached during construction to facilitate the cathodicprotection installation when needed.

Corrosion Testing

Many tests exist for establishing the reliability of protective coatings onmetal substrates. Existing tests and standards are under continuous devel-opment, and new tests are being designed. Organizations active in the de-velopment and standardization of corrosion tests for coatings includeASTM, NACE International, the Society of Automotive Engineers (SAE),the National Coil Coaters Association (NCCA), the InternationalStandards Organization (ISO), international systems (e.g., DIN), andcommercial (e.g., automotive, architectural, electronics), proprietary, and

Insulatedcopper wire

Anode

BackfillCurrent

ac line

Rectifier.

(b)

Pipeline

Soil

Insulated copper wire

Soil

Pipeline Activemetalanode

CurrentBackfill

(a)

military organizations. This section provides a brief review of the mostwidely used test methods including:

Field testsSimulated service testsLaboratory (accelerated) tests (e.g., salt spray tests, humidity tests,and electrochemical tests)

Table 5 lists selected tests used for determining the effectiveness of pro-tective coatings in corrosive environments.

More detailed information on testing of coated specimens can befound in several excellent sources. Gaynes (Ref 13) and Munger (Ref14) give descriptions and the framework for effective use of tests andstandards. Gaynes provides detailed descriptions including photo-graphs, cross-listing ASTM to federal tests and a broader perspectiveencompassing the federal standard, miscellaneous tests, and somecaveats of traditional testing. Munger offers practical material directedtoward large structures and provides a listing based on ASTM stan-dards. Altmayer (Ref 15) compiled a table of 13 applicable corrosiontests for 30 metallic, inorganic, and organic coating/substrate combina-tions. Other useful sources of information can be found in review arti-cles by Simpson and Townsend (Ref 16) and Granata (Ref 12), whichdescribe tests for metallic coatings and nonmetallic coatings, respec-tively.

Field TestsThe most reliable performance data are obtained by field tests/surveys.

One example would be to monitor and test the corrosion of autobody pan-els that sit in junkyards. Another example of in-service testing would beto monitor the behavior of the materials in a fleet of captive vehicles. Thisenables better control and recording of the exposure and driving condi-tions. The use of fleet vehicles also makes it possible to test coupons rep-resenting a larger database of materials.

Simulated Service TestsThe most widely used simulated service test for static atmospheric test-

ing is described in ASTM G 50, "Practice for Conducting AtmosphericCorrosion Tests on Metals." It is used to test coated sheet steels for a va-riety of outdoor applications. Test materials, which are in the form of flattest panels mounted in a test rack (Fig. 16), are subjected to the cycliceffects of the weather, geographical influences, and bacteriological factorsthat cannot be realistically duplicated in the laboratory. Test durations canlast from several months up to many years. Some zinc-coated steel speci-mens have undergone testing for more than 30 years.

Table 5 Widely used tests for determining the corrosion resistance of protective coatingsTest

Salt spray (ASTM B 117)

100% relative humidity (ASTM D 2247)

Acetic acid-salt spray ASTM G 85, Al (formerlyASTM B 287)

Sulfur dioxide-salt spray (ASTM G 85, A 4)

Copper-accelerated salt spray, or CASS(ASTM B 368)

FACT (formerly ASTM B 538)

Accelerated weathering

Lactic acid

Acidified synthetic seawater testing or SWAAT(ASTM G 85, A3; formerly ASTM G 43)

Electrographic and chemical porosity tests

Adhesion (ASTM D 3359-90)

T-bend adhesion (ASTM D 4145)

Description and remarks

Most widely specified test. Atomized 5% sodium chloride (NaCl), neutral pH, 35C (95 0F)(a), follow details of ASTM B 117, Appendix Xl. Emphasizes wet surfaces (nondrying),high oxygen availability, neutral pH, and warm conditions. Control of comparative speci-mens should be run simultaneously. Corrosivity consistency should be checked as de-scribed in ASTM B 117, Appendix X3. Notes: May be the most widely misused test. Re-quires correlation to service tests for useful results. Do not assume correlation exists.

Widely used test. Condensing humidity, 100% RH, 38 0C (100 0F). Emphasizes sensitivity towater exposure

Widely used test. Atomized 5% NaCl, pH 3.2 using acetic acid, 35 0C (95 0F). More severethan ASTM B 117. The lower pH and the presence of acetate affect the solubility of corro-sion products on and under the protective coatings.

Atomized 5% NaCl, collected solution pH = 2.5-3.2, 35 0C (95 0F), SO2 metered (60 min 35 cm3/min per m3 cabinet volume) 4 times per day

Atomized 5% NaCl, pH 3.2 with acetic acid, 0.025% cupric chloride-dihydrate, 35 0C (95F).Galvanic coupling due to copper salt reduction to copper metal. More severe than ASTM B117

Testing anodized aluminum specimens. Electrolyte as in salt spray or CASS test. Specimen ismade the cathode to generate high pH at defects.

Exposure of coated specimens to effects of ultraviolet radiation experienced in outdoor sunlightconditions, which may be combined with other exposures such as moisture and erosion.Exposure cabinets use carbon arc (ASTM D 822), xenon lamp (ASTM G 26), or fluores-cent lamp (ASTM G 53).

On substrates of brass and copper alloys, determines coatings porosity and resistance to handling(perspiration). Consists of immersion in 85% lactic acid solution, drying, and incubatingabove acetic acid vapors for 20 h to reveal discoloration spots at failure points or delami-nations

Atomized synthetic seawater (ASTM D 1141) with 10 mL glacial acetic acid per L of solution,pH 2.8 to 3.0, 35 0C (95 0F). More severe than ASTM B 117. The lower pH and the pres-ence of acetate affect the solubility of corrosion products on and under the protective coat-ings.

Pores and active defects in nonmetallic coatings can be revealed by color indication or depositformation. On nickel substrates, dimethylglyoxime, or steel, potassium ferricyanide (fer-roxyl test) indicator can be applied to surface on filter paper while substrate is made theanode. Alternatively, a substrate immersed in acidic copper sulfate can be made the cath-ode to form copper nodules at conductive coatings d