A Pitting Mechanism For Passive 304 Stainless Steel In Sulphuric Acid Media Containing Chloride Ions

Corrosion Science, Vol. 34, No. 8, pp. 1343-1354, 1993

0010--938X/93 $6.00 + 0.(X) © 1993 Pergamon Press Ltd

Printed in Great Britain.

A PITTING M E C H A N I S M F O R PASSIVE 304 STAINLESS STEEL IN S U L P H U R I C ACID M E D I A C O N T A I N I N G C H L O R I D E IONS P. Q. ZHANG, J. X. W u , W. Q. ZHANG, X. Y. L u and K. WAN~ University of Science and Technology, Beijing, Beijing 100083, P.R.C. Abstract--The pitting behavior of passive 304 stainless steel in 0.25 M Na2SO4 media of pH 1 containing chloride ions of different concentrations has been investigated using EIS technique. The characteristics of measured impedance spectra have been analysed. It is proposed that an intermediate complex (MOHC1)com will form in both the pit initiation and propagation stages. The reaction mechanism models for pit initiation and propagation, and an equivalent circuit model to simulate the electrode impedance in pit propagation stage have been suggested. The factors which control the pit growth kinetics have been discussed.

INTRODUCTION

MANY MODELS of the pit initiation mechanism of passive metals have been proposed. 1-8 The impedance measurement technique has been applied to the study of pitting corrosion and other localised corrosion, 9-12 and some beneficial results have been obtained. The impedance technique has marked advantages in the study of interfacial reactions and other interfacial phenomena. The impedance information obtained has time-resolved and surface-averaged characteristics. When the impedance technique is applied to investigating a changing localised corrosion system, the influences of the time and area factors must be taken into account. Cao e t al. 12'13 investigated the impedance spectra characteristics of passive electrode with both pit initiation and propagation stages. A capacitance loop occurring in a high frequency range was found before passive film breakdown. An inductance loop or inductive shrinkage of the real part occurs in the low frequency range in the Sluyters plot; after passive film breakdown, two capacitance loops occur in the Sluyters plot. Similar behavior was found in another study.14 In the present work, the pitting behavior of passive 304 stainless steel in chloride containing sulphuric acidic media was investigated using the EIS technique. EXPERIMENTAL

METHOD

The specimens were fabricated from a Japanese commercial AISI 304 stainless steel SS plate with a thickness of 2 mm, and sealed with epoxy resin. Each specimen was abraded with decreasing grit size paper from No. 400 to No. 1000 and degreased before experiment. The chemical composition of 304 SS is listed in Table 1. The solution used was a 0.25 M Na2SO 4 aqueous solution of pH 1 containing NaCI of different concentrations. All solutions were de-aerated with high purity nitrogen for >12 h and kept under a nitrogen atmosphere during experiments. A specially designed glass installation was used for the cell Manuscript received 4 August 1990; in revised form 27 July 1992. 1343

P, Q. ZHANGet al.

1344

CHEMICAL COMPOSmONOr 304 STAINLESSSTEEL

TABLE 1.

Element wt%

Cr

Ni

18.25

8.5

Si

Mn

Mo

Ti

A1

V

S

C

Cu

0 . 5 8 1.17 0.06 0.006 0.02 0.06 0.005 0.17 0.11

system to allow all experiments to be carried out in the absence of air. A saturated calomel electrode (SCE) was used for the reference electrode. The pitting experiments were carried out after the specimen had been passivated in 0.25 M Na2SO4 solution ofpH I at Ep = 0.4 V(SCE) for 30 min to allow a complete and steady passive filmto be formed on the specimen surface. Before passivation the specimen was kept at a constant cathodic current density of 10 mA cm-2 for 15 min to remove a thin layer of oxide film formed in air. Just after the solution had been changed to that containing chloride ions, a potential was applied to the specimen immediately and impedance measurement was taken continually. An impedance measurement system composed of Solartron 1250 FRA, Solartron 1286 ECI and HP 7475A Plotter was used for impedance measurement. E X P E R I M E N T A L RESULTS T h e b r e a k d o w n p o t e n t i a l (Eb) was d e t e r m i n e d by p o t e n t i o d y n a m i c p o l a r i z a t i o n curve m e a s u r e m e n t . It is f o u n d that E b of 304 SS in 0.25 M NaESO 4 s o l u t i o n of p H 1 c o n t a i n i n g 10,100, 1000 p p m a n d 0.1 N CI is almost the same, E b (i = 1 0 p A cm -2) = 0.89 V ( S C E ) . T h e following pitting e x p e r i m e n t s were carried out u n d e r E b a n d at p o t e n t i a l s a b o v e E b.

Measurement results in 0.25 M NaESO 4 q- 10 p p m C l - medium o f p H 1 T o o b t a i n i n f o r m a t i o n o n the early stages of pitting, c o n t i n u a l i m p e d a n c e m e a s u r e m e n t in the high f r e q u e n c y r a n g e was m a d e i m m e d i a t e l y after a p o t e n t i a l h a d b e e n a p p l i e d to the s p e c i m e n . F i g u r e I shows the results m e a s u r e d at E = 0 . 9 4 V. T h e total m e a s u r e m e n t time for the first four m e a s u r e m e n t s was 5 min. Figure l ( b ) shows the later i m p e d a n c e m e a s u r e m e n t results in the low f r e q u e n c y range. It is seen that a c a p a c i t a n c e loop occurs in the high f r e q u e n c y region while a n i n d u c t a n c e loop occurs in the low f r e q u e n c y region.

Measurement results in 0.25 M Na2SO 4 + 100 p p m C1- medium o f p H 1 T o c o n f i r m the i m p e d a n c e characteristics o c c u r r i n g in the low f r e q u e n c y r e g i o n , two groups of i m p e d a n c e m e a s u r e m e n t were carried o u t u n d e r the same c o n d i t i o n s : 1.5

" • 1.07

-

E

4th A1.07 ~3rd ~ •2nd •1"07 200(~ l~t o1,07 L,,.

1.0f

g

N ~-

* 1.16

• o o o

o0.23 0 °

i[ %0

2CI0Re(Z), ~ ohm6~

800

-1.

2nd

1st

1:0

3.4

0.065 . • .

2:0 3:0 Re(Z), kohm

FIO. 1. (a) Continual measurement results in the high frequency range for passive 304 SS in 0.25 M Na2SO 4 medium of p H i containing 10 ppm C1- at E = 0.94 V (total measurement time is 5 rain, numbers in plots refer to frequency in Hz); (b) the later impedance measurement results in the low frequency range under the same conditions as in Fig. 1 (a).

4:0

Pitting of 304 steel in H2SO 4

1345

one a sweep-down type (measuring from high frequency to low frequency), another a sweep-up type (measuring from low frequency to high frequency). The results are shown in Fig. 2 (a) and (b) respectively. From Fig. 2 it is seen that inductance in the low frequency range occurs in both groups of experiments. As the measured system is changing during experiment, some different features are reflected in the Sluyters plots depending upon the frequency sweep direction. The first measurement result in Fig. 2(a) (which needs about 11 min) shows a "negative capacitance" phenomenon in the low frequency part. This phenomenon becomes more obvious with the increase of potential or/and C1- concentration. The first measurement in Fig. 2(b) shows a low frequency inductance phenomenon. The "negative capacitance" is intrinsically inductance. Its occurrence is attributed to the violent change of the electrode surface in the early stage of pitting. When pitting enters its stable propagation stage, an inductance loop will occur in the low frequency region independently of the sweep direction. Under the same conditions, a potentiostatic experiment was carried out at E = 0.925 V. The anodic current density changed from 1.9 mA cm -2 to 27.5/~A cm -2 in 15 min, when it was found that there were some micropits on the surface. This showed that the passive film had suffered local damage. The long period potentiostatic experiment showed that no accelerated pit propagation took place. The impedance measurement results reflect the characteristics in pit initiation (during the first measurement) and propagation stages. Measurement results in 0.25 M Na2SO 4 + 0.1 N C1- medium o f p H 1 Figure 3 shows the measurement results at E=0.925 V. The impedance spectra characteristics and the changing tendency are the same as above. The results obtained in 0.25 M Na2SO4 medium containing 1000 ppm CI- and at other potentials above E b are similar to those above. Measurement results at potentials below E b Figure 4 shows a group of results for passive 304 SS in 0.25 M Na2SO4 + 100 ppm C1- medium of p H i at E = 0.60 V. It is seen that in the initial 11 min, an inductance phenomenon in the low frequency region also occurs in Sluyters plots as in the case of E -> Eb (see Figs 1-3); after 30 min the inductance loop in the low frequency region

1.5[ • ,,[

0.23• &~. 0.~3

. .

.

_

/ ~" / ~I B 7-4

,, -

• ~

a)

¢

~ F '''

.

.

0.23

• 2nd

•

/

/ -1 0'_ • 0

E "1.0 o

,, 3rd 4th •

":

• w°

'

1.0

6.5mo. °

"

--" 5

2.0

Re(Z), k o h m

~ •

~ •

3.0

2 ~ o o i u.oo 15T

N IFok _E C ' 12om o

'~ 12m__

b) 0.23

kf~aet)lst

-E

2.0

,i.o

-1 .o .

110

• ~3rd • 4th 2nd •

o

1st o o o6.5m

2.0

"~•

"

3.0

Re(Z), k o h m

FIG. 2. (a) Continual measurement results for passive 304 SS in 0.25 M Na2SO 4 + 100 ppm Cl- medium of pH at E = 0.925 V (sweep up); (b) continual measurement results under the same conditions as in (a) (sweep down).

,=i0

P.Q. ZHANGet al.

1346

2.0

0.23 L$ A •

•

~• ~Oo 2.o3 o

E 1.0

..c o

N E

L$

0.23

~>

• Olst

P65 0 k

";"

2nd

o o

12m O

9

0

o

• 6.5m

0

110

/',

•

•

-1.0

3rd

6.5m A

20

A

30

4'0

50

Re(Z), k o h m

F]6.3

Continual measurement results for passive 304 SS in 0.25 M Na2SO 4 + 0.1 N CI-

medium of pH 1 at E = 0.925 V (sweep up).

disappears and the time constant of the capacitance loop in high frequency region increases markedly; 1 h later, the m e a s u r e m e n t results are almost the same as that for the specimen passivated directly at E = 0.60 V for 30 min in 0.25 M Na2SO 4 medium of p H 1 in the absence of C I - . Figure 5 shows the measurements for passive 304 SS in 0.25 M Na2SO4 + 100 p p m C I - medium of p H 1 first kept at E = 0.925 V for 30 min (with pits occurring on the specimen surface), then kept at E = 0.80 V for 50 min. It is seen that only one capacitance loop occurs in the Sluyters plots, which is similar to the result for the specimen passivated directly in 0.25 M Na2SO4 medium of p H 1 in the absence of C1- at Ep = 0.80 V for 30 min.

SEM Observation of the morphology of pits Figure 6 shows the SEM photographs of the morphology of pits for 304 stainless steel pitting corroded in 0.25 M Na2SO 4 ÷ 100 p p m C I - medium of p H 1 for 3 h and in 0.25 M Na2SO4 + 0.1 N C I - medium o f p H 1 for 2.5 h respectively. A striped pattern appears at the b o t t o m of pits and some pits initiate and propagate along grain boundaries.

20~ •

•

33m

1-~'--

~

6m

N "~

2nd

"-~. "

E 10C / o =V;

\ \ .X

~ ~

3"7mX I st

-'

l~rn -10(; 0

i 100

i 200 Re(Z), k o h m

i 300

4OO

FI6.4. Impedance measurement results for passive 304 SS in 0.25 M N a 2 S O 4 + 100 ppm CI- solution of pH 1 at E = 0.6 V (C)---in the initial 11 min, Q---after 1 h, dotted curve--directly passivated in 0.25 M Na2SO4solution of pH 1 at E = 0.4 V for 30 min).

Pitting of 304 steel in H2SO 4

1347

E 10(1 tO

.~ N

E 50 '

o ~ • ~

33rn

o/.1 0 ./

o

6m o

----.

\"

/

3.7m

\ '

200 Re(Z),

kohm

Fx6.5. Impedance measurement results for passive 304 SS in 0.25 M Na2SO 4 + 100 ppm CI solution first kept at E = 0.925 V for 30 min and then kept at E = 0.8 V for 50 min (dotted curve---specimendirectly passivated in 0.25 M Na2SO 4 solution of pH 1 at E = 0.8 V for 30 min). DISCUSSION

The impedance spectra characteristics and the reaction mechanism for pit initiation and propagation Under the strong acidic media containing CI-, the measured impedance spectra show different characteristics compared with that obtained in weakly basic H 3 B O 3 - N a 2 B 4 0 7 media containing C1-. 14 In the very initial stages of pit initiation, after a sharp decrease the electrode impedance increases immediately (see Fig. la). With the occurrence of pits, inductance occurs in the low frequency region. The sharp decrease of electrode impedance in the initial time of pitting is attributed to the rapid changes of electrical properties in some local points of the passive film (e.g. electrical field increases, film resistance decreases etc.). The increase of electrode impedance in the later period occurs because the adsorbed CI- in some points are replaced by SO 2 - or O H - , and that restores the passive state. At other points, the adsorbed C1- can form an intermediate complex with the surface oxide. The adsorption of more CI- in the intermediate complex causes rapid dissolution of the passive film to take place at these points, which leads to pit nucleation. When the pit nucleus attain a critical size, the pit will grow stably in these positions. In most areas in which no CI- is adsorbed, passivation reactions still take place. Considering that the surface structure of passive film exists mainly in the form of H 2 0 - M - O H 2 and HO-M-OH,15 the following reactions will take place on the film surface: (MOH)ad~ + Cl- "-> (MOH" C1-)ads

(1)

(MOH- Cl )ads + SO2 ~ (MOH. SO2-)ads + CI-

(2)

(MOH. SO]-)~ds---~ ( M O ) p a s + H + + SO ]- +e

(3)

Cl-)ads + O H - ~ (MOH. OH-)ads + CI-

(4)

(MOH-

(MOH. OH-)ads-+ [M(OH)2]ads + e (MOH)ads ~ (MO)pas + H + + e (MOH)ads + H20 ~ [(M(OH)2]ad~ + H ÷ + e

(5) (6) (7)

1348

P.Q. ZHANG et al.

[M(OH)E]ads + H20--+ [M(On)3]ads + n + + e

(8)

[M(On)2]ads--> ( M O O H ) p a s + n + + e rds ( M O H " Cl-)ads--> (MOHC1)com + e

(9)

(MOHCl)com + nCl---> (MOUC1-Cln)ad s

(10)

(11)

(MOnCI-Cln)ads + n + --> Ms2~ + n 2 0 + (n + 1)C1-

(12)

(MOn)ads---> (MOH)+ol + e

(13)

(MOn)+ol + U + ~ M~2-~+ n 2 0

(14)

where ads, pas, com and rds represent adsorbed, passive, complex and rate determination step, respectively. Reactions (2)-(9) are passivation reactions, and reactions (1), (10)-(14) are depassivation reactions. Only when a certain critical potential (Ecp) is reached can reaction (10) take place and reactions (11)-(12) follow it, leading to high dissolution rates at some local points, which results in pit nucleation. The results shown by Figs 4 and 5 support this point. They show that below Eb, CI- can not lead to pitting, and that if pitting has occurred at E > Eb, if the potential is lowered to below Eb, the specimen will go back to passive state. Ecp is the lowest potential at which reaction (10) can take place, i.e. the pit nucleation potential (Enp) or film breakdown potential (Eb). The induction time of pitting is the time required for complete dissolution of local passive film through reactions (10)-(12) under certain potential, temperature and medium conditions. Under certain conditions, the ohmic potential drop inside a pit, or diffusion due to the concentration differences between the internal and the external bulk liquid, or salt film at the bottom of a pit can all become controlling factors of pit growth kinetics. If pit propagation is controlled by one of the above three factors, interpretation of the character of low frequency inductance is not possible. According to Frankentha116 and Vetter 17 the striped pattern appearing at the pit bottom shown by Fig. 6(a) and (b) indicates that pit propagation is in the active dissolution state, since if pit propagation is controlled by one of the above three factors, the isotropy of pit growth will lead to the occurrence of polished smooth surface of pit morphology. 16,18 The following reactions will take place first on bare metals inside pits: M + C1- ~ (MCl)ads ÷ e rds (MCl)ads + H 2 0 "-* (MOHCl)com + H + + e

(15) (16)

Pit propagation will follow reactions (11) and (12). Assuming that: (1) the total area fraction of active pits (00) can be considered as constant within a certain period of time during pit propagation; (2) a mass transfer process is not involved; (3) the area fraction of (MCl)ads in the total active area is 01. Then the current densities of reactions (15) and (16) should be: i15 = k15(1

-

01)" 00 exp (q~15E)

i16 = kl60100C exp (q)16E)

(17) (18)

a), x2000

b), x6000 FIG. 6. S E M p h o t o g r a p h o f m o r p h o l o g y o f p i t s f o r 3 0 4 S S i n ( a ) 0 . 2 5 M N a e S O 4 + 100ppm CI- medium o f p H 1 at E = 0.925 V for 3 h and (b) 0.25 M Na2SO 4 + 0.1 N C1- medium of pH 1 at E = 0.925 V for 2.5 h.

1349

Pitting of 304 steel in H2SO4

1351

Ci

II

F~G. 7. Equivalentcircuit model to simulate the electrode impedance in the pit propagation stage.

where k15, k16 are the reaction rate constant of reactions (15) and (16) respectively. A = [Cl-], C = [H20], 015 = a15 F / R T , el6 = a 1 6 F / R T . Considering that reaction (16) is the rate determining step, the Faradaic admittance (Z~-1) can be deduced. Zr71 has following form: (19)

Z F 1 = R t I + ( R o + j o L o ) -1

where Rt is the charge transfer resistance. R t = (Rtl 5 × Rtl6)/(Rtl 5 + Rtl6)

(20)

Ro =

(21)

(1 -- 01)i16 + 01i15 > 0 [(1 -- 01)i16 -- 01i15](q915i15 -- ~b16i16) 01(1 - 01)

~> 0

(22)

L° = Ko[(1 - 01)i16 - 01i15](q)15i15 - ~16i16) where Ko (>0) is a constant with dimension of C -1 cm 2. The Faradaic impedance of above reactions has an inductance component. The measured impedance is the parallel coupling of the impedance of the passive area and the impedance of the active area. As the impedance of the passive area is much higher than that of the active area, the measured impedance spectra reflect the interfacial impedance of the active area. An equivalent circuit model can be used to simulate the electrode impedance in pit propagation stage as Fig. 7 shows. Rso~is the solution resistance between working electrode and reference electrode, Ci the 12 A 10 8 6

o;

• A • o

10 ppm CI-, 0.925 V 100 ppm CI-, 0.925 V 100 ppm CI', 0.94 V 0.1 N CI-, 0.925 V

2'0

io

6b

8b

t(min)

FIG. 8. Changeof R t I with time (t).

P.Q. ZHAt~6et al.

1352 140

0.90

120 ~

100

,/:

~

0.85

0.80

u~ 80:

v

0.75

60

0.70

40 v 20

~7 v100ppmO.925 0.65 o • 100 ppm 0.94 0.60 z~ • 0.1 N 0.925 ' ' ' 328 55

°c

t (min)

FIG.9. Changeof Cdlandfl with time (t). interfacial capacitance of surface passive film, R a the solution resistance inside pit, Cdl the double layer capacitance inside pits, and Rtls//Rt16//(Ro-Lo) ("//" means in parallel and "-" means in series) the Faradaic impedance of the electrochemical reactions taking place inside pits. The characteristics of impedance spectra measured in pit propagation stage can be well interpreted using above model.

The change of the parameters of equivalent circuit with time and the factors controlling pit growth kinetics The conductivity (k) of the experimental medium is about k = 0.25 12-1 cm -1. When pit depth (h) attains 100/~m, the solution resistance inside a single pit is: R~ = h/k = 0.04 f~ cm2. Surface examination after a pitting experiment shows that the total number (n) of pits on specimen surface is generally over 50. Assuming that n = 50, the shape of pits is hemispherical and eliminating the area factor of pits, the total solution resistance inside pits should be: R~ = RS~/(n2:rr2) = 5.09. It is seen that R~ is very small compared with R t. So the measured impedance essentially reflects Ci//Cd/ZF, i.e. Ct//ZF (Ct = Ci+ Cd]). Considering the dispersion effect caused by the occurrence of pits on the electrode surface, the parameters of the capacitance loop in the high frequency region, the charge transfer resistance Rt, the overall capacitance Ct and dispersion effect index fl were fitted using a computer. 19 The parameters of inductance loop in the low frequency range (resistive component Ro and inductive component Lo) were also calculated.

6 5

6 o Ro • Lo

i

|

60

5

i

[

IO0 t (rain)

h

[

140

180

F;G. 10. Changeof Ro and Lo withtime (t).

Pitting of 304 steel in HzSO 4

1353

Figure 8 shows the changing law o f Rt -1 with time. It is seen t h a t R t 1 decreases with time. Figure 9 shows the changing law of Ct and fl with time. Figure 10 shows the change of the parameters of inductance loop with time. From Fig. 9, it is seen that Ct increases with pit propagation, while fl stays at around 0.85, and does not decrease continually. This shows that the roughness of electrode surface does not increase with pit propagation. This is inconsistent with the observation after the pitting experiment, which shows that the pit number does not increase with pit propagation. Pits will grow mainly in the existing pits formed in the early stage of pitting. The increase of Ct is mainly attributed to the increase of Cal due to the production of corrosion products inside pits. From Fig. 10, it is seen that Ro and Lo increase as the pit propagates, and the frequency range in which inductance occurs lower, which makes the impedance measurement get more difficult. The change of Ro and Lo also reflects some changes of the electrochemical processes taking place inside the pits. For an electrochemically controlled process of metal dissolution, under strong anodic polarization condition, the cathodic process can be neglected. Then: i = i° exp(2.3 AE/ba)

(23)

where i ° is the exchange current density, ba is the Tafel slope of the anodic process, AE is the overpotential. Differentiating equation (23): i = b J 2 . 3 R t.

(24)

Eliminating the area factor, taking R t as the overall charge transfer resistance, from equation (24), the overall current for pit propagation can be obtained. Taking a = 0.5 and n = 2, then b a = 2 . 3 R T / a n F = 0.059 V. The current directly measured during pit propagation is about one order larger than the current calculated from equation (24) (neglecting the current flowing through the passive area). This shows that the apparent Tafel slope should be one order bigger than the above real Tafel slope. This may be due to the decrease of active area inside pits caused by the adsorption of SO 2- on the surface of the pit bottom. Zhu 2° calculated the effect of anion adsorption on the apparent Tafel slope. The results showed that when the real Tafel slope is between 2.3RT/2ctF and 2.3RT/ctF, the apparent Tafel slope can change from several decades of millivolts to infinity, while the law of active dissolution inside pits does not change. Therefore, the changing law of overall current of pitting can still be represented by the changing law of R t 1. From Fig. 8, it can be concluded that the overall current of pitting decreases with time, i.e. pit growth rate decreases with time. In all the experimental systems related to this work, the concentration of SO 2 is higher or much higher than the concentration of CI-. The adsorption of SO 2- on pit surfaces will decrease the area fraction of adsorbed C1- (01) greatly. From equation (18), it is known that the reaction rate of the rate determining step (equation 14) will decrease greatly, so the pit growth rate is inhibited. It can be concluded that the competitive adsorption of SO42- and CI- on the pit surface is the controlling factor of pit growth kinetics. CONCLUSIONS

(1) The reaction mechanisms for pit initiation and propagation, and an equivalent circuit model to simulate the electrode impedance in pit propagation stage have been

1354

P.Q. ZHANGet al.

p r o p o s e d . T h e e x p e r i m e n t a l results have b e e n satisfactorily i n t e r p r e t e d using the proposed model. (2) T h e c o m p e t i t i v e a d s o r p t i o n of SO42- a n d C1- o n the pit surface is the c o n t r o l l i n g factor of pit growth kinetics. (3) T h e c h a n g i n g law of overall c u r r e n t of pitting can be reflected by the c h a n g i n g law of the reciprocal of overall charge transfer resistance of pitting.

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

REFERENCES Y. M. KOLOTYRKIN,J. electrochem. Soc. 108,209 (1961). H. P. LECKIEand H. H. UHLIG,J. electrochem. Soc. 113, 1262 (1966). T. P. HOARand W. R. JACOB,Nature, 126, 1299 (1967). T. P. HOAR,D. C. MEARSand G. P. ROTHWELL,Corros. Sci. 5, 279 (1965). C. Y. CHAO,L. F. LIN and D. D. MACDONALD,J. electrochem. Soc. 128, 1187 (1981). J. R. GALVELE,J. electrochem. Soc. 123,464 (1976). J. R. GALVELE,J. electrochem. Soc. 125, 1204 (1978). K. J. VETrER,Bet. Bunsen. Phys. Chem. 69, 589 (1965). M. OraDOand T. OK1,Light Metals 36,416 (1986). J. L. DAWSONand M. G. S. FERREIRA,Corros. Sci. 26, 1009 (1986). F. MANSFELDand H. Slain, J. electrochem. Soc. 135, 1171 (1988). J. WANG,C. N. CAOand H. C. LIN, J. Chinese Soc. Corros. Prof. 9,271 (1989). C. N. CAD,J. WANGand H. C. LIN, J. Chinese Soc. Corros. Prot. 9,261 (1989). P. Q. ZHANG,J. X. WU, W. Q. ZHANG,X. Y. LU and K. WANG,J. Univ. Sci. Tech. Beijing 12,412 (1990). G. Or.AMOTOand T. SHIBATA,Corros. Sci. 10,371 (1970). R. P. FRANKENTHALand H. W. PICKERING,J. electrochem. Soc. 119, 1304 (1972). K. J. VETFERand H.-H. STREHBLOW,in Localized Corrosion, p. 240. NACE-3 (1974). N. SATO,J. electrochem. Soc. 129,260 (1982). J. WANG,C.Y. Sin, S. Z. SONGand C. N. CAO,J. Chinese Soc. Corros. Prot. 9, 11(1989). Y. Y. Znu, Ph.D Thesis, Univ. of Sci. and Tech. Beijing, (1986).