Marcus P. Corrosion mechanisms in theory and practice

Подождите немного. Документ загружается.

17. P. Marcus and I. Olefjord, Corrosion (NACE) 42:91 (1986).

18. P. Marcus and J.M. Grimal, Corros. Sci. 31:377 (1990).

19. P. Marcus and H. Talah, Corros. Sci. 29:455 (1989).

20. D. Tromans and L. Frederick, Corrosion (NACE) 40:633 (1984).

21. R.C. Newman, H.S. Isaacs, and B. Alman, Corrosion (NACE) 38:261 (1982).

22. A. Garner, Corrosion (NACE) 41:587 (1985).

23. R.C. Newman, Corrosion (NACE) 41:450 (1985).

24. R.C. Newman, K, Sieradski, and H.S. Isaacs, Met. Trans. 13A:2015 (1982).

25. R.C. Newman and K. Sieradski, Corros. Sci. 23:363 (1983).

26. S.E. Lott and R.C. Alkire, J. Electrochem. Soc. 136:973 and 3256 (1989).

27. C. Duret-Thual, D. Costa, W.P. Yang, and P. Marcus, Corros. Sci. 39:913 (1997).

28. H.-S. Kuo, H. Chang, and W.-T. Tsai, Corros. Sci. 41:669 (1999).

29. T. Haruna, T. Shibata, and R. Toyota, Corros. Sci. 39:1935 (1997).

30. Z. Szklarska-Smialowska, Corrosion (NACE) 28:388 (1972).

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

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

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

34. P. Marcus, C.R. Acad. Sci. Paris Ser. II 305:675 (1987).

35. P. Marcus and M. Moscatelli, J. Electrochem. Soc. 136:1634 (1989).

36. P. Marcus and M. Moscatelli, Mem. Et. Sci. Rev. Met. 85:561 (1988).

37. X. Cheng, H. Ma, S. Chen, L. Niu, S. Lei, R. Yu, and Z. Yao, Corros. Sci. 41:112 (1999).

38. J.E. Castle and R. Ke, Corros. Sci. 30:409 (1990).

39. J. Stewart and D.E. Williams, Corros. Sci. 33:457 (1992).

40. M. Janik-Czachor, Modifications of Passive Films (P. Marcus, B. Baroux, and

M. Keddam, eds.), The Institute of Materials (EFC 12), 1994, p. 280.

41. J.L. Crolet, L. Seraphin, and R. Tricot, Met. Corros. Ind.,

no. 616 (1976).

42. D.E. Williams, T.F. Mohiuddin, and Ying Yang Zhu, J. Electrochem. Soc. 145:2664

(1998).

43. P. Combrade, O. Cayla, M. Foucault, D. Vançon, A. Gelpi, and G. Slama, Proceedings

of the 3rd International Symposium on Environmental Degradation of Materials in

Nuclear Power Systems-Water Reactors, Traverse City, (G.J. Theus and J.R. Weeks, eds.),

The Metallurgical Society (1987) p. 525.

44. P. Marcus and E. Protopopoff, J. Electrochem. Soc. 137:2709 (1990).

45. P. Marcus and E. Protopopoff, J. Electrochem. Soc. 140:1571 (1993).

46. P. Marcus and E. Protopopoff, J. Electrochem. Soc. 144:1586 (1997).

47. P. Marcus and E. Protopopoff, Corros. Sci. 39:1741 (1997).

48. E. Protopopoff and P. Marcus, to be published.

49. V. Maurice, H.-H. Strehblow, and P. Marcus, Surf. Sci. 458:185 (2000).

310 Marcus

Further Insights on the Pitting Corrosion of

Stainless Steels

Bernard Baroux

Ugine Research Center, Ugine-Savoie, Ugine, France

INTRODUCTION

Fundamentals of the pitting mechanisms have been discussed in Chapter 7 and the

whole question of pitting corrosion was reviewed by Smialowska [1a] in 1986. It

is intended here to shed light on some points of practical or theoretical importance

which were recently investigated or revisited and to propose a comprehensive

viewpoint on some rather scattered experimental evidence. The examples are

restricted to stainless steels, in weakly acid chloride-containing aqueous media

whose pH is larger than the depassivation pH, so that the passive film remains

stable except in the pit itself.

The role of the chloride ions in pitting initiation can be related to local pas-

sive film destabilization, together with a counteracting effect vis-à-vis the passive

film healing. However, it should be noted that the passive film breakdown and

healing phenomena act on the microscopic scale (some nanometers), whereas the

pits observed in the practical situations are on a macroscopic scale (some 100

micrometers). Between the microscopic and the macroscopic size scales, many

phenomena can occur, such as pit growth, modifications of the local solution com-

position, and dissolution of the nonmetallic inclusions present in the steel, which

act generally at the scale of some micrometers. In the same way, electrochemical

assessment of the pitting corrosion resistance generally involves the measurement

of anodic currents of the order of some microamperes (or some 10 μA). Measuring

1 μA during 1 s corresponds to a pit size of the order of some micrometers. This

size scale will be referred to as the mesoscopic scale. It should be pointed out that

the pit repassivation can occur in these mesoscopic stages as well as in the micro-

scopic one, so that the passive film breakdown theories are often unable to predict

the actual behavior of the alloy. Figure 1 shows a semideveloped pit observed at

the scale of 10 μm on a 17% Cr stainless steel immersed in an NaCl-containing

solution. The pit consists of an undermining hollow, covered by a thin metallic

cap, leading to the formation of an occluded zone. Some secondary pits are visible

all around the main hole. The possible collapse of the metallic cap can suppress

the occlusion and provoke pit repassivation.

311

Furthermore, pit occurrence is often considered a random phenomenon.

[2,3a,4a], which makes the probability of survival of the passive state difficult to

predict and the measurement of pitting potentials uncertain. This random behavior

can result either from the passive film breakdown and healing mechanisms

(microscopic stage) or the phenomena occurring in the mesoscopic stages. Moreover,

the pit occurrence probability varies with time, increasing to severe conditions but

decreasing in milder ones so that the pitting sensitivity depends on the surface aging.

Last, pitting of industrial steels also depends on their metallurgical properties.

It is obvious that the steel composition plays a major role, but the phenomena can

be rather complex, depending on the pitting stage under consideration. The steel

microstructure is a relevant factor as well, although scarcely taken into account by

the current models. The unavoidable presence of some nonmetallic inclusions in

industrial steels and the properties of these inclusions (which may act as pitting

sites) are often more determining for the pitting resistance than the addition of

expensive alloying elements.

In the following, some experimental results pertaining to these concerns are

presented, making the relation between pitting theories and practical behavior more

realistic and motivating future works to shed light on some poorly understood

domains.

PITTING POTENTIAL MEASUREMENTS

The Experimental Procedure

Measuring the pitting potential provides some information on the pitting resistance in

the test solution. However, the result of the measurement depends not only on the

312 Baroux

Figure 1 Scanning electron microscopy, × 1000. Typical view of a semideveloped pit at the

end of the mesoscopic stage. A thin metallic film is still present and covers part of the pit.

Secondary pits are visible all around the main hole. The pit may either go on or repassivate if

the thin “top” film breaks down. (From Ref. 28.)

material itself but also on the passive film characteristics. On the other hand, the

properties of the passive film depend on (a) its initial state before the test, then on the

surface preparation procedure, and (b) the test procedure itself, since the film can be

modified when immersed in the electrolyte. Moreover, a sample is considered to be

pitted when the anodic current increases sharply and irreversibly. Then the question

arises of when pitting becomes irreversible. In the most frequently used techniques,

the anodic current increase associated with the onset of irreversible pitting is larger than

some l0 μA; then pitting potential characterizes the occurrence of a macropit, or at least

of a pit in the mesoscopic stage, but certainly not the nucleation of a micropit.

Two procedures are used for measuring the pitting potential. In the

potentiostatic method, the sample is polarized at a potential V and the induction

time τ(V) for pit occurrence is recorded. The pitting potential is the highest

potential for which τ = ∞. In practice, a large, but finite, value of τ is chosen. Note

that when pit initiation is a purely deterministic process, τ can be considered as the

“incubation time” necessary for the conditions for irreversible passive film break-

down to occur.

In the potentiokinetic method, the electrode potential is raised from the rest

potential up to pitting. The measured pitting potential depends on the scanning rate

s. Using a deterministic model, one finds [5a]

Pitting Corrosion of Stainless Steels 313

and then the pitting potential decreases when the scanning rate is lowered.

However, it is believed that, at least for low scanning rates, the passive film is modi-

fied during the potentiokinetic scan, resulting in a pitting resistance improvement

and in an increase of the incubation time [(

∂τ

–

∂t

)

> 0]. Then, under such conditions,

the measured pitting potential increases when the scanning rate is lowered, leading

to the occurrence of a critical scanning rate, s

for which the pitting potential is

minimum [3a,b].

Another characteristic potential is sometimes drawn from the potentiokinetic

measurements, namely the so-called repassivation potential V

rep

. The potential is first

raised up to pitting; then, once the anodic current has reached a predetermined value

corresponding to anodic dissolution into the pit, the scanning is reversed toward the

cathodic potentials. The anodic current then decreases and the repassivation potential

is obtained when the current measured during the backward scanning equals the one

was found during the forward scanning. For potentials larger than V

rep

the pit is active

and for smaller values it repassivates. However, one should keep in mind (a) that the

repassivation potential does not characterize the resistance to pitting initiation and (b)

that it depends on the degree of pit development when the scanning is reversed, then

on the arbitrary current density i

Whatever the test procedure and the method chosen for analyzing the results,

one must be aware of the fact that, even when the corrosive solution is well

defined, associating an intrinsic pitting potential with a given material as a criterion

of resistance to pitting should be definitely ruled out. The measured values are

strongly dependent not only on the steel surface condition but also on the experi-

mental procedure.

In the following, some examples are given of various stainless steels tested at

constant temperature in carefully deaerated NaCl aqueous solution (distillated water)

whose pH was adjusted to the test value by adding either NaOH or HCl. The samples

(S = 0.785 cm

) were cut from thin cold-rolled sheets (thickness 0.5 to 1 mm), then

mechanical polished (paper SiC grade 1200, under water) and aged 24 h in air (this

aging procedure was shown to ensure better reproducibility of the results). Last, the

specimens were immersed in the corrosive solution and left at rest potential for 15min

before testing.

Probabilistic Behavior

Because pitting exhibits probabilistic behavior, the induction time τ (the time for

the occurrence of the first pit) in potentiostatic conditions is a probabilistic value

and here the model described above does not hold. In the same way, the pitting

potential measured in potentiokinetic conditions is probabilistic.

For a set of N samples tested under strictly the same conditions (potentiostatic

or potentiokinetic), the pitting probability at a time t (or at a potential V) is defined as

n/N, and the survival probability P as (N – n)/N, where n is the number of pitted

specimens. When N is large enough (which should be discussed), P also represents

the survival probability (probability for no pit to occur) of a single specimen.

Now, let us consider the pitting probability ω

−

. δS of an infinitesimal area δS of this

single specimen (which implicitly assumes that ω is much smaller than the

number of pitting sites per unit area). In the following, ω will be referred to as the

elementary pitting probability [3a,b] (EPP). The survival probability of δS is

δP = 1 – ω

−

δS. The survival probability of the whole sample is then P = (1 – ω

−

δS)

S/δS

which tends to exp-ω

−

S when δS tends to zero. Then the EPP is given by ω

−

= (–In P)S,

where P is approximated by P =(N – n)/N. The elementary probability (for the

occurrence of a single pit) is a probability per unit area. A more detailed statistical

analysis shows that the smallest reliable value of ωwhich can be measured is of the

order of 1/(NS). In the following examples, N is equal to 24 or 48.

The time derivative g = dω

−

/dt of the elementary pitting probability ω is

referred to as the pit generation rate. The intrinsic pitting potential (if it really exists)

is given by g = 0 and can be deduced from the transition from P = 1 to P = 0 for

infinite surface areas S. This condition is not easily fulfilled in laboratory experiments,

and it is often more convenient to use small surface areas and to characterize the

pitting resistance by the probabilistic function ω or g.

Let us note that this description of the probabilistic behavior for the appearance

of observable pits does not imply, at this stage, any mechanistic assumption about the

mechanisms responsible for this behavior. It is only a convenient way to describe

the experimental results. Shibata and Takeyama suggested another way to analyze the

probabilistic behavior. They proposed a stochastic model including a pit initiation

λ(V) frequency and a pit repassivation μ(V) frequency, leading to a conventional

pitting potential V

pit

defined by λ(V

Pit

)=μ(V

pit

Potentiostatic measurements on a sufficient number of samples, or

simultaneous measurements using a multichannel device [2,3ab], given the survival

probability P and then the elementary pitting probability ω

−

(V, t) as a function of the

polarization time t (Fig. 2a). The probability density for the random function τ is

–dP/Pdt = Sg(V, t). Figure 2a and b show the pit generation rate time dependence

314 Baroux

for several electrode potentials, which can be denoted g = g

(V)h(V, t), where

h(V, 0) = 1. One sees (Fig. 2b) that for low enough potentials h is a decreasing

function of time (which can be approximated by an exponential decay), showing

that potentiostatic holding in the corrosive solution itself may improve the further

pitting resistance. The initial pit generation rate g

(V) increases with V (Fig. 2c) and

provides some information on the pitting resistance of the tested material before any

polarization. A conventional pitting potential V

pit

can be defined, for which g

pit

)

Pitting Corrosion of Stainless Steels 315

Figure 2 (a) Elementary pitting probability in NaCl (0.5 M) for an AISI 304 type

stainless steel in potentiostatic conditions for different electrode potentials V; short

holding times. (b) Same for long holding times. (c) Pit generation rate for t = 0 determined

either in potentiostatic or potentiokinetic conditions. (From Ref. 3a.)

has a small, but measurable, arbitrary value (for instance 0.1cm

–2

–1

in the

following examples).

Potentiokinetic measurements give the elementary pitting probability ω

−

(V)- for

the potential V and the potentiokinetic pit generation rate g(V)=dω

−

/dt = s dω

−

/dV,

both depending on the scanning rate s. For high enough scanning rates, the

potentiokinetic scanning does not modify the pitting resistance. Therefore

g = g

(V) and does not depend on the scanning rate. The elementary pitting

probability is then ω

−

(V)=∫

rest

(V)dV, which does not depend on V

rest

since

rest

) = 0. For smaller scanning rates, g < g

(V) and then ω

−

(V) < ω

−

(V).

It is convenient to define a conventional potentiokinetic pitting potential V

pit

−

pit

) = ω

−

, where ω

−

is a conventional pitting probability (ω

−

= 0.1 cm

–2

in the

following examples). One sees that for large enough scanning rates, V

pit

increases

with s. For scanning rates smaller than a critical value, the pitting sensitivity of the

surface decreases during the test and then V

pit

increases when s decreases [3a]. For

this reason, it can be better to choose a large enough scanning rate, which does not

provide any change in the pitting sensitivity during the test and whose kinetic effect

(s = dV/dt) on V

pit

is readily taken into account. Following this idea, the effect of the

passive film modifications can be studied either by using the potentiostatic mode

or by prepolarizing the specimen at constant potential before a high scanning rate

potentiokinetic test. Otherwise, the pitting potential obtained by the potentiokinetic

method is a complex function of the scanning rate, including kinetic and surface

modification effects. In the following examples, s = l00mV/min, which was shown

to be larger than the critical scanning rate for all the studied steels, and the

conventional pitting potential was defined as giving ω

−

= 0.1 cm

–2

METALLURGICAL ASPECTS

The Alloying Elements and the Metallurgical Processing

For stainless steels [3c], the chromium content is obviously one of the major

parameters which controls the passive film properties and then the pitting resistance,

at least for some conditions of passive film formation. This is illustrated in standard

conditions by the results shown in Figure 3. Other oxidizable elements, such as

silicon, which is present at the level of some tenths of a weight % in industrial AISI

430 or 304 type steels, can also enter the passive film and improve the pitting

resistance.

The sulfur content is also a determining parameter for the pitting resistance, due

to the formation of sulfides, which may act as pitting sites (see later). The free

machining steels, such as AISI 303 type austenitic steels, containing 0.2% to 0.3%

sulfur, which forms some numerous and large manganese sulfides. These inclusions

are known to improve the steel machinability dramatically, but also to decrease the

pitting resistance. AISI 430 or 304 contains less than 0.03% S, which leads to less

numerous and smaller sulfides. Among them, the grades used for long products (bars

and wires) often contain typically 0.02% S in order to ensure sufficient machinability.

In contrast, the grades used for flat products (sheets and plates) generally contain less

than 50 ppm S, which leads to improved pitting resistance. The precise role of Mn

sulfides in pit initiation has been extensively studied [1b,6,7] and will be detailed

later. It is known that sulfide dissolution provides some harmful sulfur species, but that

316 Baroux

alloying elements [8] such as Cu, Ni, and also Mo

may combine in the aqueous

solution with these sulfur species and then decrease their noxious effect.

The reason for the beneficial effect of some alloying elements which do not

enter the passive film to a large extent (Ni in 304 or 316 steels, Mo in 434 or 316

steels) is not so clear for industrial steels, despite the numerous hypotheses which

have been proposed on the basis of academic work. They might (a) lower the ionic

current throughout the metal–passive film interface, (b) favor pit repassivation by

changing the solution composition in an initiated pit (for instance, by preventing too

large a pH decrease inside the pit), or (c) counteract the harmful effect of sulfur

species as indicated above. The third mechanism accounts fairly for the Mo effects in

434 and 316 type stainless steels: accepting the idea that adsorbed sulfur plays a

major role in pit initiation on such materials, Mo could act [9] by favoring the

desorption of sulfur, possibly in a combined state with Mo, adsorbed uncharged S

being unstable in the presence of Mo and then reducing finally to soluble H

The metallurgical processing of the steel (hot and cold rolling, heat treat-

ments, welding, surface treatments) may dramatically modify the pitting resistance

of a steel grade as well. The effect of heat treatments on the phase equilibria,

recrystallization, and minor element segregation will not be discussed here, but

some known or less known evidence must be recalled. First, the steel generally

contains some nonmetallic phases such as precipitates (Cr carbides in ferritic

steels, for instance) or inclusions (oxides, sulfides, etc.). The difference between

precipitates and inclusions is that the former are produced in the metal solid

state and the latter during (or at the end of) the melting process, in the metal

liquid state. Then the inclusions do not generally produce any segregation at

their interface with the metallic matrix, which is not the case for the precipi-

tates (e.g., the chromium carbides). However, nonmetallic inclusions can be

unstable inthe corrosive medium; that is the case for sulfides and even for

oxides in some cases [10a]. Furthermore, because their ductility is generally not

the same as that of the metallic matrix, hot and cold rolling may produce some

Pitting Corrosion of Stainless Steels 317

Figure 3 Effect of the Cr content on the pit generation rate in NaCl (0.02 M) for some

Ti-bearing (Ti=0.4%) FeCr industrial alloys with different Cr contents. V is the electrode

potential. (From Ref. 3c.)

microdecohesions at the metal-inclusion interface, which behave as micro-

crevices and may be some preferential pitting sites, depending mainly on the

cold rolling ratio. Anyway, it is now well established that, in industrial steels,

inclusions may act as pitting sites [11]. On the other hand, chromium carbides may

provoke some chromium depletion in the metal around the precipitate, after a

welding treatment or even after the metal cooling at the end of a hot transformation

process. Even when these local negative segregations are too weak to produce

intergranular corrosion, they may lead to a decrease in the pitting resistance. For

this reason, the final annealing must be carefully controlled, mainly for ferritic

steels [due to the low carbon solubility in body-centered cubic (bcc) structures].

Last, the final phase for stainless sheet processing is generally annealing, or

solution treatment in the case of austenitic steels, followed or not by pickling. When

the annealing is performed in hydrogen-containing atmosphere (bright annealing), no

pickling is needed but the composition of the annealing atmosphere (for instance, its

residual water content) markedly influences the passive film properties [3d]. When

annealing is performed in oxidant atmospheres, a pickling treatment should follow in

order to remove the thick oxide film which was formed. The composition and the

properties of the resulting passive film, together with a change in the metallic

inclusion distribution at the metal surface, noticeably affect the pitting resistance. In

the same way, so-called nitric acid decontamination treatments are sometimes

performed after at the end of the process and also modify the surface properties. Last,

the surface condition in the delivery state also depends on the surface roughness,

which is itself controlled by the rolling conditions. One sees that the practical pitting

resistance of an industrial material cannot be entirely predicted by the simple models

developed on model systems!

The Sulfide Inclusions

Sulfide inclusions play a major role in pitting initiation on iron alloys, among them

stainless steels. It is worth drawing attention to some of their physicochemical and

metallurgical properties.

For AISI standards 430, 434, 304, 316, and more generally for all Ti-free

stainless steels, sulfur is combined with Mn to form manganese sulfides, MnS, which

act as pitting sites in current corrosive media (see later). Following the sulfur content

of the steel and the metallurgical processing, MnS is found either isolated or stuck to

other inclusions (generally oxides) or even combined in complex inclusions.

Chromium may also substitute for Mn to form some (Mn,Cr) S sulfides, whose Cr

content depends on the Mn content of the steel (currently 0.4% for ferritic steels and

1% or more for austenitic ones) as shown in Figure 4. The higher the chromium

content in the sulfide, the higher the pitting resistance of the steel, resulting in a

better pitting resistance for low-Mn steels [12a], at least when other metallurgical

parameters are not modified by the Mn content lowering. In Ti-bearing steels

(Ti > some 0.1% by weight or more), the MnS formation is prevented, since Ti

sulfides (Ti

S) are formed at higher temperature during the melting process. Ti

sulfides are considered less harmful than MnS for pitting corrosion but are also able

to act as pitting sites in some conditions [3e,12a,13].

Three basic characteristics of manganese sulfides are worthy of notice [5b]

as far as they can influence the pitting initiation process: (a) They are electronic

318 Baroux

conductors, but less than the surrounding metal, leading to pit initiation at the

inclusion-matrix interface. Their conductivity should depend on their exact chemical

composition. (b) They are polarized together with the surrounding metal, but should

be unstable in the potential range where the metal is passive, and then readily

dissolve, providing some local chemical and electrochemical conditions which differ

from ones which prevail on other parts of the surface. (c) Their size morphology,

together with their composition, plays an important role in their ability to promote

pitting [14]. Rapid solidification or laser surface melting [15a] provides the removal

or at least the redistribution of the Mn sulfides, resulting in a strong improvement in

pitting resistance.

It should also be noticed that not all the sulfides present at the steel surface

promote pit initiation, which some authors explained by differences in their

composition and the morphology of the metal-sulfide interface. However, the

question is not so clear, because the number of Mn sulfides acting as pitting sites is

much smaller than the total number of Mn sulfides at the steel surface. In our

opinion, the competition between sulfide dissolution and local modifications in the

passive film properties could result in a low-probability stochastic initiation process,

which might account for the small number of sulfides which are found to initiate

pitting. Some authors underlined the importance of the pitting test procedure for the

relation between the MnS dissolution kinetics and the pitting sensitivity [14], which

shows once more the absence of any “intrinsic” pitting potential.

Last, the recognition that manganese sulfides are probably the least resistant pit

initiation sites in standard industrial stainless steels motivated severe control of the

steel microstructure, either by lowering the Mn content (which unfortunately also has

some undesirable secondary effects) or, better, by adding small amounts of Ti as

alloying elements, preventing the formation of Mn sulfides. However, this means of

improving the pitting resistance is itself limited, because in the absence of MnS, pits

can initiate on other inclusions—namely Ti sulfides, oxides, or silicates—or

possibly on the passive surface itself, as the electrode potential or the chloride

concentration increases.

Pitting Corrosion of Stainless Steels 319

Figure 4 Composition of sulfide inclusions in type 416 (13%Cr 0.3%S) steel as a function

of the steel Mn content. (From Ref. 8.)