Marcus P. Corrosion mechanisms in theory and practice

Подождите немного. Документ загружается.

work suggested). In this model the rate C of current rise at the pit nucleus is the one

parameter which would vary with the electrode potential. Note that, although the

linear time dependence of the current makes this model different from that proposed

by Gabrielli and colleagues [4b], the power spectral density frequency of the anodic

current has the same asymptotic limit as Gabrielli’s [i.e. f

Ψ (f) = constant when

f →∞; see above]; however, in contrast with the Gabrielli model, no low-frequency

plateau is found.

Going further, the authors proposed that pitting initiation is controlled by

the onset and the maintenance of a sufficient gradient of acidity (provided by the

hydrolysis of the dissolved metal cations) and electrode potential on the scale of

the surface roughness of the specimen. Fluctuations in these gradients, leading to

the birth and death of events, could arise because of fluctuations in the boundary layer

thickness in the liquid at the metal surface; a pit would become stable when its depth,

including the surface roughness, significantly exceeds the thickness of the solution

boundary layer. Let us note that such ideas are not far from those proposed by Okada

[26], which additionally involve the definition of a critical wavelength for

the potential fluctuations parallel to the surface, which would be of the order of

magnitude of the average distance between some “efficient” metallurgical flaws.

However, the idea that local acidification is the cause of pit nucleation does not seem

to be supported by experimental evidence, particularly in unbuffered acidic media,

at least for the initial stages of pit initiation. In our opinion, the electric field at

the filmsolution interface (the electrical potential gradient) and the chloride

ion enrichment are probably responsible for pit nucleation (see later) and their

fluctuations should therefore be at the origin of the prepitting events.

Furthermore [15e], the stable pit generation rate Λ was shown to be

proportional to the microscopic pit generation rate λ, at least for some given alloy and

corrosive medium. However, the proportionality factor λ/Λ = exp(–μτ

) is

much smaller than 1, so that unstable pits can appear (even at very low chloride

concentrations, such as 10 ppm) at much lower potentials than stable ones (for the

definition of which a critical current of the order of 10 μA was chosen, to be compared

to some 100 nAfor the detection of unstable pits). Prepitting a specimen does not lead

to increased sensitivity to further pitting, probably due to reinforcement of the passive

film resistance during the prepitting experiment, which the authors attribute to an

increase in passive film thickness. Although the observed effect is not questionable,

one should not consider as implicit evidence that the pitting resistance improvement

is caused by an increase in passive film thickness, since other beneficial modifications

of passive film (Cr enrichment, film dehydration) may occur during aging.

In another paper, Stewart and Williams [15a] investigated the pitting sensitivity

of some AISI 304 steels with various sulfur contents (in fact, the modern 304 low-

sulfur industrial steels probably have a lower, or at least equivalent, sulfur content

as the high-purity steel considered in the study, in which S is some 30 ppm). Some

sensitive techniques were used to detect unstable pits at the level of some nA. It is

clearly demonstrated that sulfur-rich inclusions dominate as pitting nucleation sites

and that the lifetime of unstable pits is related to the sulfide particle size. It is

proposed that the sulfur derived from the inclusion helps to stabilize the pit growth,

which we reformulate by saying that it helps to stabilize the micropit in its mesoscopic

stage by preventing the repassivation, possibly leading to a growing macroscopic pit.

Last, laser melting treatments considerably reduce the nucleation frequency and the

340 Baroux

lifetime of micropits, which is consistent with the inclusion size effect, since this

process dramatically reduces the inclusion size at the specimen surface.

Studies using microelectrodes [15f] reveal some events whose amplitude is of

the order of 10 to 100 pA, the coulombic charge being of the order of 100 nC,

and the frequency of some 10 mHz, which the authors associate possibly with the

microscopic stage of pitting initiation (i.e., the breakdown and healing of passive films).

An important point is that most of the events are characterized by a sharp rise of the

anodic current followed by a slow decay, in contrast to the current transients associated

with pit nucleation on macroelectrodes for the same MnS-containing steels. Anyway,

these events could not be detected on macroelectrodes because they could not be dis-

tinguished above the background current. This is an important point of the paper, to be

compared with the different forms of current transients shown in the preceding section

for steels containing different types of sulfides. Last, the authors discussed the form of

the observed current decay during the repassivation process and showed that it is con-

sistent with the cluster percolation theory that they proposed [15g] for binary alloy dis-

solution and passivation. The important point is that some arguments are put forward

for the observed prepitting events to occur at the microscopic (atomistic) scale and not

at the mesoscopic one, as for the results generally obtained on macroelectrodes.

Remarks

It is worth noting that, in whatever way a pit nucleus was born, its further development

to form a pit embryo (growing metastable pit) depends on the electrolyte

concentration which locally sets up in the electrolyte when the metal dissolves. The

development of the pit embryo implies the local stabilization of an acidic corrosive

medium, differing from the surrounding one. This local acidification induces a local

dissolution of the metal in the active state, which is compensated only by the

diffusion of the corrosion products. This dissolution in turn provokes an acidification

due to the hydrolysis reactions and the process is self-sustained, provided that the

diffusion or the electromigration is slow enough in the electrolyte. Formally [3h,5d],

if J is the anodic current and X represents the concentration in corrosion products, one

has dX/dt = KJ(V, X) – DX, where the metal-electrolyte potential difference V is a

control parameter (J increases with V) and parameters K and D represent respectively

the production in corrosion products by the anodic dissolution and their dilution into

the electrolyte by a diffusion process. The steady state conditions write dX/dt = 0.

The system is locally stable when

Pitting Corrosion of Stainless Steels 341

This condition is fulfilled when V is smaller that a critical potential V

crit

(the

“protection potential”) which depends on the bulk electrolyte concentration and on

the nature of the pitting site. The critical potential acts as a bifurcation parameter

beyond which a metastable pit can grow and below which it repassivates (as far as

the diffusion parameter D does not decrease). When a pit nucleus repassivates

(V < V

crit

), the current transient is expected to show the form presented in Figure 11a.

This situation occurs generally when no efficient diffusion barrier takes place

(D/K large) but this is of course dependent on the presence or not of some

nonmetallic inclusions (e.g., MnS) whose dissolution can provide harmful chemical

species. In contrast, when a pit embryo grows (V> V

), the current transient

is expected to show the form presented in Figure 11b, with a slow current increase

corresponding to the pit growth, followed by a sharp decrease corresponding to

the breakdown of the diffusion barrier (D drastically increases), for instance, the

breakdown of the pit cap formed by a vestige of the outer metallic surface at the pit

mouth (see Fig. 1). The growth rate of the metastable pit and the conditions for the

diffusion barrier to vanish depend of course on the electrolyte composition inside the

pit and then on the presence or not of dissolving nonmetallic inclusions.

Now considering = {X

} as a set of concentrations for chemical species i, the

master equations of the system evolution are written as dX

/dt = K

J(V, X

) – D

which is a set of ordinary differential equations. Since J(V, X

) is nonlinear in X

the system may produce in some cases (close to V

crit

) some complex oscillations,

explaining qualitatively the existence of a prepitting noise.

Rest Potential Fluctuations

In open-circuit conditions, the prepitting events result in some rest potential

fluctuations as well, although this question is less well documented in the literature

than the anodic current transients. In a general discussion of the phenomena occurring

in the early stages of pitting [16b], Isaacs cited some work on Ni-base alloys, from

which it appears that rest potential prepitting transients consist of some sharp

potential falls followed by slower return to the stationary value. On the other hand,

Magaino [27] analyzed the power spectral density of the rest potential fluctuations

on a 304 type steel and evidenced two results of great metallurgical importance:

(a) The potential fluctuations are more pronounced for cold-rolled specimens than

for solution-annealed ones, and (b) in the 0.05–2-Hz frequency range, the PSD

amplitude increases with the steel sulfur content.

Hashimoto and colleagues [28] published a stochastic analysis of potential

fluctuation during passive film breakdown and repair on iron, later criticized by

Isaacs [16c]. The typical form of the prepitting transient is shown in Figure 20,

leading to the consideration of several stages in pit initiation: (a) nucleation (onset

of the rest potential decrease), (b) micropit growth (quick potential decay), (c)

termination of this growth and (d) pit repassivation. The authors state that:

342 Baroux

Figure 20 Schematic representation of rest potential fluctuations due to prepitting

events. (From Ref. 16c.)

1. Pit nucleation can be approximated to a Poissonian process, the nucleation

rate λ decreasing exponentially with time as shown in an earlier section and

Ref. 3g.

2. In the micropit growth, the pit area A

pit

is proportional to the square of the

time and the potential fall ΔV (measured from the average rest potential V

corresponding to the passive state, so that V = V

– ΔV) is given by ΔV =

b ln(1 + A

pit

/Ai

), where A is the specimen area, b a constant, i

the current

density in the passive state, and i

pit

the current density into the dissolving pit

(assumed to be a constant).

3. The maximum potential fall ΔV

(stage 3) follows a Gaussian distribution

law. The average value V

= V

– ΔV

of the lowest reached potential

appears to be independent of the immersion time. At the opposite, the

average rest potential V

increases with time due to the reinforcement of the

passivity during rest potential aging (which is briefly said in this paper to be

the consequence of the film thickening). Thus, ΔV

increases with the

immersion time. V

is assumed to be a critical potential and (V– V

) the

driving force for pitting initiation. The pit initiates when a critical driving

force is reached but, in the following stages, the potential and then the

driving force decrease, leading finally to termination of growth when

ΔV = ΔV

4. The potential decay reflects the repassivation kinetics. For the authors,

this stage is controlled by the passive film growth inside the pit after the

potential has decreased below V

, which progressively reduces the anodic

current i

pit

. Using a direct logarithmic law for the passive film growth and

reformulating slightly the authors’discussion, one finds exp ΔV= b ln(t/τ + exp

Δ V

/b)/(t/τ + 1) where τ is a time characteristic of the transition from the

high dissolution rate situation to the situation where i ∼ 1/t, as expected

when the current is controlled by the passive film growth.

This last point (the repassivation mechanism) is debated by Isaacs [16c],

particularly as the formation and growth of a passive layer should not be sufficient

to provide the anodic current for the potential difference ΔV to nullify, considering

the balance with cathodic reactions which occur inside an active pit. For this

author, the role of the capacitance of the passive surface is crucial to the interpre-

tation of the transient potential changes during pitting. Following this review,

cathodic current is drawn from the interfacial capacitance of the passive surface

during the micropit growth (stage 2) and repassivation (stage 4) is the reverse

process (a slow recharging of the capacitance). Then the slow potential decay in

stage 4 is not controlled by film growth but by the progressive polarization of

the cathodic reaction. It is worth noting that the capacitance model also has

some implications in stage 2, since the potential changes should be related to

the charge of the pit [16d].

Oltra and colleagues [29] analyzed the role of the capacitance during a “birth

and death” pitting event, with reference to some 304 stainless steels. The time

constant for repassivation is found to depend on the impedance of the passive surface

in open-circuit conditions. From these results it can be inferred that the micropit

growth is under the control of the electrode area, which not only has some practical

Pitting Corrosion of Stainless Steels 343

consequences but also sheds light on the applicability of some current labora-

tory techniques (such as noise analysis), since the cell geometry may be a determining

factor and the actual pit current may be distorted before reaching the meas-

urement setup.

CONCLUDING REMARKS

The results presented in this chapter show that pitting initiation on industrial stain-

less steels exhibits a combination of metallurgical influences, such as nonmetallic

inclusion effects, and stochastic behavior, as evidenced by both the scattering of

the measured pitting potentials and the prepitting noise. A third essential factor is

the influence of the passive film properties on the establishment of a stable pit.

However, many problems come in mind when considering the relative or combined

effects of these three factors. First, the proper effect of the passive film is not clear

when pitting initiates on nonmetallic inclusions. The current models for pit

initiation generally refer to the passive film breakdown, while the investigated

steels contain many nonmetallic inclusions, which probably act as pitting sites.

Second, since many stages are involved in the formation of a metastable pit, at the

onset of stable pitting, the question remains open which stage is deterministic and

which stage is probabilistic. Third, the size effects are often disregarded, and

microscopic models are frequently proposed to account for macroscopic effects.

However, the recent findings of several workers make the situation clearer, and we

feel that the route is now open for a better understanding and then better modeling

in this rather complex field.

For simplicity, the elementary pitting process can be divided into three phases

(Fig. 21).

1. Pit nucleation, which leads to the formation of a small area of bare, unfilmed,

metal (pit “nucleus”). Pit nucleation as a whole is likely to be a probabilistic

process but can itself be the result of several elementary steps at the atomic

scale, involving possibly deterministic and probabilistic mechanisms. The

nucleation rate (cm

–2

–1

) is denoted G.

2. The development of a metastable pit, which leads to local dissolution of a

significant part of the underlying metal. The dissolving metal area is referred

to as a pit “embryo.” The result is either the formation of a stable pit (pit

stabilization) or the repassivation of the embryo. The development and the

repassivation of a metastable pit are probably deterministic processes for each

embryo but may be probabilistic for a population of embryos, growing on

different pitting sites. The formation of a stable pit (stages 1 and then 2) will be

referred to as the initiation stage. The rate of formation (cm

–2

–1

) of a stable

pit is the pit generation rate and is de noted g. In some experimental conditions

(for instance, high scanning rate potentiokinetic tests) the repassivation rate of

the metastable pits is expected to be constant and then g is proportional to

the nucleation rate G. In this case, the measurement of g provides some infor-

mation on the nucleation stage. In other experimental conditions (for instance,

potentiostatic tests at suitable potentials), some information is obtained on

the repassivation processes involved in stage 2. Last, it is believed that

344 Baroux

prepitting noise results from nonlinear feedback phenomena [15h] occurring

in phase (2).

3. The growth of a stable pit, which is deterministic. This third step is outside the

present concern but it is itself a complex stage of pitting, involving possible

repassivations at a macroscopic scale.

As far as nucleation is concerned, three situations may occur. First, pit can nucle-

ate directly on the passive film, far from any nonmetallic inclusion. This is the case

generally considered in most theoretical models, but there is no evidence that it is

the prevailing situation for industrial stainless steels and more work is needed to

clarify this point. It is a point for future work to study in more detail both the effect

of the electronic properties of the passive film and the reactions occurring at its

outer interface (adsorption, complex formation, etc.), seeking a better understand-

ing of the distribution of potentials throughout the interface. Second, pitting can

arise from the dissolution of unstable inclusions (such as MnS), leading to a small

area of bare metal when the inclusion is entirely dissolved. The point to be clarified

is then the proper role of the passive film properties in the pit nucleation process.

Last, passive film and inclusions can interact together. Referring to the results pre-

sented earlier, a simple model can be proposed in the case of the MnS inclusions.

It has been pointed out that MnS dissolution is pH and chloride assisted, producing

some Mn

ions and some sulfur-containing species, followed by a pH decrease

and a local increase in chloride concentration. This increase in chloride concen-

tration is provoked by the electromigration of chloride ions from the solution bulk

to the pit embryo (due to a local excess of positive charges close to the

dissolving sulfur), resulting in an increase in the [Mn

][Cl

–

] product, then the

precipitation of a nonprotective salt layer. However, only a small number of MnS

sulfides produce a pit embryo, which is generally explained by differences

between the sulfide properties (or the properties of the metal-inclusion interface).

Another idea is that at the beginning of MnS dissolution, the local excess of

positive charges is not sufficient to increase significantly the local chloride concen-

tration and that this process should be assisted by passive film breakdown

Pitting Corrosion of Stainless Steels 345

Figure 21 Multistep mechanism for the onset of a stable pit.

in the inclusion neighborhood, which produces a large amount of metal cations,

leading to local Cl

–

enrichment due to electromigration. One suggests that beyond

a critical electrode potential a large number of microscopic passive film break-

down events are randomly produced, generally followed by passive film healing,

except close to the MnS sulfides, where some dissolved sulfur species can prevent

the repassivation. This phenomenon is a low-probability event, explaining the

small number of the inclusions acting as pitting sites. From this situation, the stable

sulfide dissolution may arise, as proposed above. The question remains open

whether the sulfur species act only as inhibitors of the passive film healing or

also increase the film breakdown probability at the atomistic scale. Anyway, such

a model shows that the pitting rate is influenced by the passive film properties at

the early stages of pit nucleation and accounts for the fact that pits initiate close

to the inclusions. In a further step the inclusion may dissolve, producing a bare

metal area (pit nucleus).

More generally, it should be pointed out that, whatever the pit nucleation

mechanism, the survival probability of a pit nucleus (or even of a pit embryo)

depends on the state of the surrounding passive film, since different passive films

formed on the same alloy do not exhibit the same pitting resistance. The distribution

of potentials and electric charges around the pit nucleus should likely provide a

better assessment of the intensity of the anodic reactions than the average potential

difference Vand are probably a function of both the pitting site characteristics and the

surrounding passive film properties [3i].

REFERENCES

1. (a) Z. Szlarska-Smialowska, Pitting Corrosion of Metals, Nace, Houston, 1986;

(b) Z. Szlarska-Smialowska, Corrosion (Nace) 28: 389 (1972); (c) Z. Szklarska-

Smialowska and E. Lunarska, Werk. Korr. 32: 478 (1981), cited in Ref. 8.

2. T. Shibata and T. Takeyama, Corrosion (Nace) 33: 243 (1977).

3. (a) B. Baroux, Corros. Sci. 27: 969 (1988); (b) B. Baroux, Mem. Sci. Rev. Met.

December 683 (1988) (in French); (c) B. Baroux, F. Dabosi, and C. Lemaitre, Stainless

Steels (P. Lacombe, B. Baroux, and G. Béranger, eds.), Les Editions de Physique Les

Ulis (France), 1993, p. 314; (d) D. Gorse, J. C. Joud, and B. Baroux, Corros. Sci. 33:

1455 (1992); (e) B. Baroux and D. Gorse, Proceedings of the European Symposium on

Modification of Passive Film (P. Marus, B. Baroux, and M. Keddam, eds.). The

Institute of Metals, 1994, p. 300.; (f) B. Baroux, 11th international Congress of

Corrosion, Florence, Italy April 1990, 5, 469; (g) B. Baroux and D. Gorse, NATO

Workshop on Aqueous Corrosion modeling. Plymouth, 1993, in press.; (h) B. Baroux,

Passivity of Metals and Semiconductors (M. Froment, ed.), Elsevier, Amsterdam, 1983;

(i) D. Gorse and B. Baroux, to be published.

4. (a) G. Gabrielli, F. Huet, M. Keddam, and R. Oltra, Corrosion (Nace) 46: 266 (1990);

(b) M. Keddam, M. Krarti, and C. Pallotta, Corrosion (Nace) 43: 454 (1987).

5. J. L. Crolet, L. Seraphin, and R. Tricot, (a) C.R. Acad. Sci. Paris 280C: 333 (1975)

(in French); (b) Mem. Sci. Rev. Met. 73: 703 (1976) (in French); (c) Mem. Sci. Rev.

Met. 74: 281 (1977) (in French); (d) Mem. Sci. Rev. Met. 74: 647 (1977) (in French).

6. G. S. Eklund, J. Electrochem. Soc. 121: 467 (1974).

7. A. J. Sedriks, Int. Met Rev. 28: 295 (1983).

8. G. Wranglen, Corros. Sci. 14: 331(1974).

346 Baroux

9. A. J. Betts and R. C. Newman, Corros. Sci. 34: 1551 (1993).

10. (a) M. A. Baker and J. E. Castle, Corros. Sci. 33: 1295 (1992); (b) Ruoru Ke and

J. E. Castle, Corros. Sci. 30: 409 (1990); (c) M. A. Baker and J. E. Castle, Corros.

Sci. 34: 667 (1993).

11. P. Poyet, A. Desestret, H. Coriou, and L. Grall, Mem. Sci. Rev. Met. February 133

(1975) (in French).

12. (a) A. Szummer, K. Lublinska, and M. Janik-Czachor, Werk. Korr. 41: 618 (1990).

(b) A. Szummer and M. Janik-Czachor, Br. Corros. J. 9: 216 (1974).

13. B. R. T. Anderson and B. Solly, Scand. J. Metall. 4: 85 (1975).

14. P. E. Manning, D. J. Duquette and W. F. Savage, Corrosion, (Nace), 36: 313 (1980).

15. J. Steward and D. E. Williams, Corros. Sci. 33: 457 (1992); (b) A. R. J. Kucernak,

R. Peat, and D. E. Williams, J Electrochem. Soc. 139: 2337 (1992); (c) D. E. Williams,

C. Westcott, and M. Fleismann, J. Electroanal. Chem. 180: 549 (1884); (d)

D. E. Williams, C. Westcott, and M. Fleismann, J. Electrochem. Soc. 132: 1796 and 1804

(1985); (e) J. Steward and D. E. Williams, Advances in Localized Corrosion, Nace 9,

Orlando, FL, 1987, p. 131; (f) A. M. Riley, D. B. Wells, and D. E. Williams,

Corros. Sci.

32: 1307 (1991); (g) D. E. Williams, R. C. Newman, Q. Song, and R. G. Kelly, Nature

350: 216 (1991); (h) D. E. Williams, J. Steward, and P. H. Balkwill, Critical Factors in

Localized Corrosion (G. S. Frankel and R. C. Newman, eds.), The Electrochemical

Society, Corrosion Division, Proceedings volume 92-2, p. 36.

16. (a) U. Steinsmo and H. S. Isaacs, J. Electrochem. Soc. 140: 643 (1993); (b) H. S.

Isaacs, Corros. Sci. 29: 313 (1989); (c) H. S. Isaacs, Corros. Sci. 34: 525 (1993);

(d) H. S. Isaacs and Y. Ishikawa, J. Electrochem. Soc. 132: 1288 (1985).

17. S. E. Lott and R. C. Alkire, J Electrochem. Soc. 136: 973 (1989).

18. J. M. Defranoux, Corros. Sci. 3: 75 (1963).

19. U. Bertocci and Y. Yang-Xiang, J. Electrochem. Soc. 131: 1011 (1984).

20. J. J. Podesta, R. C. V. Piatti, and A. J. Arvia, Corros. Sci. 22: 193 (1982).

21. C. Cao, Q. Shi, and H. Lin, Bull. Electrochem. 6: 710 (1990).

22. T. Tsuru and M. Saikiri, Corros. Eng. 39: 401 (1990).

23. (a) P. C. Pistorius and G. T. Burstein, Mater. Sci. Forum 111–112: 429 (1992); (b)

G. T. Burstein and S. P. Martin, Philos. Mag. Lett

. 66: 127 (1992); (c) P. C. Pistorius

and G. T. Burstein, Corros. Sci. 33: 1885 (1992).

24. (a) P. Schmucki and H. Bohni, Werkst Korr. 42: 203 (1991); (b) P. Schmucki and H.

Bohni. J. Electrochem. Soc. 139: 1908 (1992); (c) P. Schmucki and H. Bohni, Oxide

Films on Metals and Alloys, The Electrochemical Society, Toronto, 1992, p. 326;

(d) F. Hunkeler, G. S. Frankel, and H. Bohni, Corrosion (Nace) 43: 189 (1987).

25. R. Doelling and K. E. Heusler, Z. Phys. Chem. 139: 39 (1984).

26. T. Okada, (a) J. Electrochem. Soc. 132: 537 (1985); (b) Corros. Sci. 26: 839 (1986);

The Electrochemical Society, Corrosion Division, Proceedings volume 92–2, p. 65.

27. S. Magaino, J. Electroanal. Chem. 258: 227 (1989).

28. M. Hashimoto, S. Miyajima, and T. Murata, Corros. Sci. 33: 885 (1992).

29. R. Oltra, G. M. Indrianjafy, and R. Roberge, J. Electrochem. Soc. 140: 343 (1993).

Pitting Corrosion of Stainless Steels 347

Crevice Corrosion of Metallic Materials

Pierre Combrade

FRAMATOME, Centre Technique, Le Creusot, France

INTRODUCTION

Crevice corrosion is a form of localized corrosion that occurs in zones of restricted

flow where a metallic material surface is in contact with a small volume of

confined, stagnant liquid whereas most of the material surface is exposed to the bulk

environment.

Crevice zones may result from the design of the component (see Fig.1) or

from the formation of deposits during service, shutdown, or even fabrication.

These deposits may come from suspended solids in the environment, corrosion

products, or biological activity. Low-flow areas are prone to the formation of such

deposits.

Crevice corrosion occurs mainly (but not exclusively) on passive materials. The

most important problem is the crevice corrosion of stainless steels, nickel-base alloys,

aluminum alloys, and titanium alloys in aerated chloride environments, particularly

in sea or brackish water, but also in environments found in chemical, food, and oil

industries. Other cases of crevice corrosion are also known such as the so-called

corrosion by differential aeration of carbon steels, which does not require the

presence of chloride in the environment. Also mentioned in the literature is the

crevice corrosion of steels in concentrated nitric acid and inhibited cooling water and

of titanium alloys in hot sulfuric environments.

In this chapter, the basic mechanisms of crevice corrosion are briefly presented

but most of the text is devoted to the crevice corrosion of passive alloys, particularly

Fe-Ni-Cr- … alloys in aerated chloride environments. Phenomenological aspects,

the mechanisms of initiation, the conditions of propagation, the modeling, the

experimental techniques, and the possibility of prevention are successively described.

BASIC MECHANISMS OF CREVICE CORROSION

The specificity of the crevice zones arises (Fig. 2) from (a) the limited mass transport

by diffusion and convection between the inside of the crevice and the bulk

environment, (b) the high (metal surface area)/(solution volume) ratio inside the

349

crevice, (c) the presence of large external surfaces exposed to the bulk environment,

and, in many cases, (d) the significant solution resistance between the inside and the

outside of the crevice.

Crevice corrosion is caused by a change of the local environment inside the

crevice zone into more aggressive conditions. On passive materials, the local

conditions must become severe enough to cause passivity breakdown and the onset

of active dissolution.

Usually the local change of environment occurs in two steps. The first step

is due to the local exhaustion of reactants, usually the dissolved oxygen, and/or the

local accumulation of corrosion products as a consequence of the restricted transport

kinetics between the crevice and the bulk environment. This leads to the buildup

350 Combrade

Figure 1 Examples of crevice zones due to design.

Figure 2 Characteristics of a crevice zone: reduced diffusional exchanges with the bulk

solution, low Φ

diffusion

; small solution volume in contact with large surface area, L/h large;

large external surfaces S

; solution resistance between crevice gap and bulk solution, IR drop.