Electrochimica Acta 76 (2012) 94–101

Contents lists available at SciVerse ScienceDirect

Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta

Crevice corrosion kinetics of nickel alloys bearing chromium and molybdenum

N.S. Zadorozne a , C.M. Giordano c , M.A. Rodríguez a,c,∗ , R.M. Carranza c , R.B. Rebak b
a
Consejo Nacional de Investigaciones Científicas y Técnicas,Argentina
b
GE Global Research, 1 Research Circle, Schenectady, NY 12309, USA
c
Gerencia Materiales, Comisión Nacional de Energía Atómica, Instituto Sabato, UNSAM/CNEA, Av. Gral. Paz 1499, San Martín, B1650KNA, Buenos Aires, Argentina

a r t i c l e i n f o a b s t r a c t

Article history: The crevice corrosion kinetics of alloys C-22, C-22HS and HYBRID-BC1 was studied in several chloride
Received 17 February 2012 solutions at 90 ◦ C. The crevice corrosion resistance of the alloys increased with PREN (Pitting Resistance
Received in revised form 16 April 2012 Equivalent Number), which is mainly affected by the Mo content in the alloys. The crevice corrosion
Accepted 17 April 2012
kinetics of the three alloys was analyzed at potentials slightly higher than the repassivation potential.
Available online 22 May 2012
Crevice propagation was controlled by ohmic drop in the more dilute chloride solutions, and by charge
transfer in the more concentrated chloride solutions. Ohmic drop was not a necessary condition for
Keywords:
crevice corrosion to occur.
Nickel alloys
Crevice corrosion © 2012 Elsevier Ltd. All rights reserved.
Repassivation
N06022
Chloride

1. Introduction (UNS N06022) [5] is one of the most versatile members of the
Ni–Cr–Mo family. This alloy has shown excellent resistance to
Crevice corrosion is a form of localized corrosion that may occur pitting corrosion, crevice corrosion and environmentally assisted
within cracks and crevices where an acidic concentrated solu- cracking in hot concentrated chloride solutions [3,6,7]. Alloy C-22
tion is developed [1]. Pitting corrosion and crevice corrosion are is used in harsh industrial environments and it was proposed as a
essentially the same phenomena from an electrochemical point of corrosion-resistant barrier of high-level nuclear waste containers
view, although there are geometrical differences between them [2]. [3,6–8]. The increasing demand of the industry for corrosion resis-
Crevice corrosion is of particular concern for stainless steels and tant alloys with particular properties of corrosion and mechanical
nickel alloys when in contact with chloride containing solutions resistance has led to the development of new alloys. HASTELLOY®
[1,3,4]. C-22HS® (UNS N07022) is a novel high-strength corrosion resis-
Several families of nickel alloys have been developed based tant material. This alloy provides a corrosion resistance comparable
on the ability of nickel to retain large amounts of different with that of other Ni–Cr–Mo alloys, and it can also be age-hardened
alloying elements in solid solution [3,4]. Each alloy family is to effectively double its yield strength [9–11]. High yield strength
intended for specific applications. Commercially pure nickel is corrosion resistant alloys, such as alloy C-22HS, are candidate mate-
widely used in caustic environments. Nickel–Copper (Ni–Cu) rials for oil and gas applications [12]. HASTELLOY® HYBRID-BC1®
alloys are able to withstand hot wet hydrofluoric acid; and (UNS N10362) is a new alloy intended for filling the gap between
Nickel–Molybdenum (Ni–Mo) alloys are the selected materials Ni–Mo and Ni–Cr–Mo alloys. Despite the versatility of Ni–Cr–Mo
for handling hot reducing (non-oxidizing) acids (HCl and H2 SO4 ), alloys, there is a need within the chemical process and pharmaceu-
provided that no oxidants are present. The most versatile family tical industries for materials with even higher resistance to HCl
is the Nickel–Chromium–Molybdenum (Ni–Cr–Mo), which pro- and H2 SO4 . This need is partially satisfied by the Ni–Mo alloys.
vides corrosion resistance in reducing and oxidizing hot acids [3]. However, Ni–Mo alloys cannot withstand even small quantities of
Chromium and molybdenum provide protection against oxidizing dissolved oxygen [13]. Alloy HYBRID-BC1 is able to withstand HCl
and reducing conditions, respectively [3,4]. HASTELLOY® C-22®1 and H2 SO4 , even in the presence of dissolved oxygen and other
oxidizing species, and its resistance to chloride-induced pitting
corrosion, crevice corrosion and stress corrosion cracking is also
remarkable [13].
∗ Corresponding author. Tel.: +54 11 6772 7353; fax: +54 11 6772 7362.
The crevice corrosion kinetics of Ni–Cr–Mo alloys has been stud-
E-mail addresses: martinrz@gmail.com, maalrodr@cnea.gov.ar (M.A. Rodríguez).
1
HASTELLOY® , C–22® , C–22HS® and HYBRID–BC1® are trademarks of Haynes ied in some detail [14–19]. Results from a previous work suggest
International, Inc. that the charge transfer process controls the anodic dissolution

0013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.electacta.2012.04.157
 N.S. Zadorozne et al. / Electrochimica Acta 76 (2012) 94–101 95

Table 1
Nominal composition (wt%) of the tested alloys (* maximum values).

Alloy Ni Cr Mo W Fe Co Si Mn C V Al B

C-22 56 22 13 3 3 2.5* 0.08* 0.5* 0.01* 0.35* – –

C-22HS 61 21 17 1* 2* 1* 0.08* 0.8* 0.01* – 0.5* 0.006*
HYBRID-BC1 62 15 22 – 2* – 0.08 0.25 0.01* – 0.5* –

within the crevice of alloy C-22 [16]. However, ohmic drop may test. A water-cooled condenser combined with a water trap was
be a necessary condition for crevice corrosion to occur [20]. Lillard used to avoid evaporation of the solution and to prevent the ingress
et al. indicate that crevice corrosion of alloy 625 in artificial ocean of air (oxygen). The temperature of the solution was controlled by
water is under ohmic control [14]. immersing the cell in a water bath, which was kept at a constant
The objective of the present work was to study the crevice corro- temperature. All the tests were performed at ambient pressure. The
sion kinetics of nickel alloys bearing chromium and molybdenum reference electrode was a saturated calomel electrode (SCE), which
in chloride solutions, at 90 ◦ C. Alloys C-22, C-22HS and HYBRID- has a potential 0.242 V more positive than the standard hydrogen
BC1 were selected to assess the effect of the chemical composition. electrode (SHE). The reference electrode was connected to the solu-
Testing in hydrochloric acid solution was performed in order to tion through a water-cooled Luggin probe. The reference electrode
simulate the environment within active crevices. The results are was kept at room temperature. The electrode potentials were not
analyzed in terms of the localized acidification model [2]. corrected for the thermal liquid junction potential, since it was
assumed to be on the order of only a few mV. The counter electrode
2. Experimental consisted in a flag of platinum foil (total area 50 cm2 ) spot-welded
to a platinum wire. All the potentials in this paper are reported in
The test specimens were prepared from wrought mill annealed the SCE scale.
plate stock. The nominal chemical compositions of the tested alloys
in weight percent are listed in Table 1. Fig. 1 shows the two different
types of specimens used. Fig. 1a shows the prism crevice assemblies 2.1. Corrosion behavior in 1 M HCl solution
(PCA), fabricated based on ASTM G 48 and G192 [21,22] which con-
tained 24 artificially creviced spots formed by a ceramic crevice The tests were performed in deaerated 1 M HCl at 90 ◦ C, to
former wrapped with a 70 ␮m-thick PTFE tape. The torque applied simulate the acidic conditions in active crevices using prismatic
to the crevice formers was 5 N m. Fig. 1b shows the prismatic specimens. The open circuit or corrosion potential (ECORR ) of the
specimen fabricated based on a variation of the ASTM G 5 speci- tested alloys was monitored and recorded for 24 h in deaerated
men [23]. The tested surface areas were approximately 14 cm2 for conditions. For the three tested alloys, ECORR remained approx-
PCA specimens and 10 cm2 for prismatic specimens. The specimens imately constant over the testing time varying only within a
had a finished grinding of abrasive paper number 600 and were range of 20–30 mV. Electrochemical impedance spectroscopy (EIS)
degreased in acetone and washed in distilled water within the hour measurements were performed after 24 h of immersion. A 10 mV-
prior to testing. Prism crevice assemblies were used for determin- amplitude sinusoidal potential signal was superimposed to ECORR .
ing the repassivation potential of the tested alloys in near neutral The frequency scan was started at 10 kHz and ended at 10 mHz.
chloride solutions. Prismatic specimens were used for studying the Parameters of simple equivalent circuit mathematical models were
general corrosion of the tested alloys in 1 M HCl solution. Although fitted to these data in order to obtain the polarization resistance
this specimen forms a crevice (Fig. 1b), it is not severe enough for (RP ), which led to the calculation of the instantaneous uniform
producing crevice corrosion of the tested alloys [16]. corrosion rate. The equivalent circuits used for fitting to the EIS
All the electrochemical tests were conducted in a one-liter, experimental results were simple combinations of R//CPE in series
three-electrode vessel (ASTM G 5) [23]. Nitrogen (N2 ) was purged with an ohmic resistance, where CPE stands for Constant Phase
through the solution from 1 h prior to testing until the end of each Element. Circuits with two or three time constants were used. RP
was defined as the sum of the resistances of all the time constants
observed. Further analyses of the impedance data was out of the
scope of the present work.
Calculation of the corrosion rate of the tested alloys was made
according to ASTM G 102 [24]. Since the anodic behavior of the
tested alloys in hot HCl solutions does not show a linear Tafel
region, the Tafel slopes cannot be obtained directly from polariza-
tion curves [25,26]. The value of the anodic Tafel slope (bA ) used
in the present work was 0.053 V. This value is reported in the lit-
erature for alloy C-22 when tested in a variety of low pH chloride
solutions at 90 ◦ C [25]. The cathodic behavior of the tested alloys
in 1 M HCl at 90 ◦ C did not show a linear Tafel region either. The
value of the cathodic Tafel slope (bC ) used in the present work was
0.023 V, which was extrapolated from data of alloy C-22 at slightly
higher pH values, where a Tafel behavior is observed [26]. There is
no reported data about Tafel slopes for alloys C-22HS and HYBRID-
BC1 in hot HCl solutions. The above mentioned anodic and cathodic
Tafel slopes were used for the three tested alloys and they rendered
a Stern–Geary constant (B) of 0.007 V.
The general corrosion rate (CR) was calculated using Eq. (1),
where jCORR is the corrosion current density, EW is the equivalent
Fig. 1. Scheme of PCA and prismatic specimens. weight,  is the alloy density and K is the faradaic conversion factor
96 N.S. Zadorozne et al. / Electrochimica Acta 76 (2012) 94–101

Table 2
Constants used in corrosion rate calculations (* assuming congruent dissolution of
the major alloying elements as Ni2+ , Cr3+ , Mo6+ , Fe2+ , and W6+ ) [24].

Alloy EW ␳/g cm−3

*
C-22 23.28 8.68
C-22HS 22.78* 8.60
HYBRID-BC1 22.77* 8.83

(K = 3.27 m g cm−1 A−1 yr−1 ). The values of EW and  used for each
tested alloy are listed in Table 2 (ASTM G 102) [24].
K · jCORR · EW
CR = (1)

Potentiodynamic polarization curves were performed in deaerated
1 M HCl solutions at 90 ◦ C using prismatic specimens. The anodic
polarization started a few mV below the ECORR and ended when an
anodic current density of 1 mA cm−2 was attained. A potential scan
rate of 0.167 mV s−1 was used.

2.2. Crevice corrosion in near neutral chloride solutions

Fig. 2. Complex plane diagram and Bode plots (inset) corresponding to EIS tests on
The Potentiodynamic–Galvanostatic–Potentiodynamic the tested alloys in deaerated 1 M HCl at 90 ◦ C, after 24 h of immersion. Dots: data;
(PD–GS–PD) technique was used to determine susceptibility line: equivalent circuit fittings.

to localized corrosion [27,28]. This technique is a modification of

the Tsujikawa–Hisamatsu electrochemical (THE) method (ASTM by alloys C-22HS (0.3 mm yr−1 ) and HYBRID-BC1 (0.08 mm yr−1 ). A
G 192) [22] for determining the crevice corrosion repassivation slight increase of ECORR was observed along with the decrease of CR
potential It consists of three stages: (1) a potentiodynamic polar- (Fig. 3).
ization (at a scan rate of 0.167 mV s−1 ) in the anodic direction Fig. 4 shows the polarization curves of the tested alloys in 1 M
until reaching a total anodic current of 30 ␮A, (2) the applica- HCl at 90 ◦ C. Alloys C-22 and C-22HS showed an active to passive
tion of a constant anodic current of IGS = 30 ␮A (approximately transition peak, followed by a wide passive range, and a current
jGS = 2 ␮A cm−2 ) for 2 h, and (3) a potentiodynamic polarization increase due to transpassivity. The peak current density of alloy C-
(at 0.167 mV s−1 ) in the cathodic direction, from the previous 22HS was smaller than that of alloy C-22. Alloy HYBRID-BC1 did
potential until reaching alloy repassivation. The selected current not show an active to passive transition peak, and a passive or
density of stage 2 (2 ␮A cm−2 ) was slightly higher than the passive pseudo-passive range was observed at potentials higher than ECORR .
current density of the alloys in the tested conditions. This assures The passive current density of alloy HYBRID-BC1 was one order of
a minimal interference of transpassive dissolution on the crevice magnitude higher than those of alloys C-22 and C-22HS.
corrosion propagation. Furthermore, those methods which pro-
duce the lowest propagation of crevice corrosion lead to the most
3.2. Crevice corrosion in near neutral chloride solutions
conservative repassivation potentials [28]. Stage 2 was meant to
sustain a precise amount of crevice corrosion propagation.
Fig. 5 shows the results from a PD–GS–PD test performed to alloy
The crevice corrosion repassivation potential for this technique
HYBRID-BC1 in 1 M NaCl, at 90 ◦ C. The crevice corrosion repassiva-
is the cross-over potential (ECO ) determined at the intersection of
tion potential (ECO ) was determined as the cross-over between the
the forward (stage 1) and reverse (stage 3) scans. The PD–GS–PD
forward and the backward scans (Fig. 5). Figs. 6–8 show crevice
tests were performed in deaerated chloride solutions at near neu-
corrosion attack in alloys C-22, C-22HS and HYBRID-BC1, respec-
tral pH, using PCA specimens. The testing solutions included 0.1 M
tively, after testing in 1 M NaCl at 90 ◦ C. Corrosion products were
NaCl, 0.5 M NaCl, 1 M NaCl, 4 M NaCl and 5 M CaCl2 . All the speci-
detected around and over the attacked areas. The localized attack
mens were examined after testing with a light optical microscope
and some of them were observed in the scanning electron micro-
scope (SEM).

3. Results and discussion

3.1. Corrosion behavior in 1 M HCl solution

Fig. 2 shows the complex plane diagram and bode plots cor-
responding to EIS tests in deaerated 1 M HCl at 90 ◦ C, after 24 h
of immersion. Experimental data are shown along with the corre-
sponding equivalent circuit fittings. EIS tests showed two or three
time constants indicating complex anodic dissolution processes for
the three tested alloys. Alloy HYBRID-BC1 showed the highest cor-
rosion resistance in the tested conditions (Fig. 2).
Fig. 3 shows average values of CR of the tested alloys as a function
of ECORR in deaerated 1 M HCl, at 90 ◦ C. Error bars represent one
standard deviation of two to three values of CR and ECORR . Alloy C-22 Fig. 3. Corrosion rate as a function of potential for the tested alloys in deaerated 1 M
showed the highest CR of the three alloys (0.8 mm yr−1 ), followed HCl at 90 ◦ C, after 24 h of immersion.
 N.S. Zadorozne et al. / Electrochimica Acta 76 (2012) 94–101 97

Fig. 4. Potentiodynamic polarization curves for the tested alloys in deaerated 1 M

HCl at 90 ◦ C.

Fig. 6. Image of an alloy C-22 specimen after a PD–GS–PD test in pH 6, 1 M NaCl at

90 ◦ C.

showed similar features in 0.1 M, 0.5 M and 4 M chloride solutions,

for the three tested alloys (not shown). Figs. 9–11 show specimens
of alloys C-22, C-22HS and HYBRID-BC1 respectively, after testing
in 5 M CaCl2 , where the crevice corrosion attack spread outside the
area occluded by the crevice former teeth. Figs. 12 and 13 show SEM
images of crevice corrosion on alloy C-22 after testing in 1 M NaCl
and 5 M CaCl2 , respectively. Fig. 12 shows crevice corrosion of alloy
C-22 below a crevice former tooth in 1 M NaCl. Corrosion products
were observed covering the attacked areas. Fig. 13 shows crevice
corrosion of alloy C-22 in 5 M CaCl2 . The attack spread outside the
occluded area. No corrosion products were observed. It is not clear
if this difference is only result of the total chloride concentration or
if there is a cation effect.
Fig. 14 shows ECO obtained from PD–GS–PD tests as a func-
tion of chloride concentration (cCl − ) for the three tested alloys, at
90 ◦ C. The symbols are average values from at least three mea-
surements, and the error bars indicate one standard deviation.
Alloy HYBRID-BC1 showed the highest resistance to localized cor-
rosion, since its ECO values were approximately 0.2 V higher than
those of alloys C-22 and C-22HS. Alloy C-22HS showed ECO values
slightly higher than those of alloy C-22. As expected, ECO for the
three alloys decreased as the cCl − increased. However, ECO slightly Fig. 7. Image of an alloy C-22HS specimen after a PD–GS–PD test in pH 6, 1 M NaCl
at 90 ◦ C.
increased from cCl − = 4 M to cCl − = 10 M for alloys C-22 and C-22HS.
This behavior has been reported before for alloy C-22 [30].

Fig. 5. PD–GS–PD test for an alloy HYBRID-BC1 specimen, in pH 6, 1 M NaCl at 90 ◦ C. Fig. 8. Image of an alloy HYBRID-BC1 specimen after a PD–GS–PD test in pH 6, 1 M
NaCl at 90 ◦ C.
98 N.S. Zadorozne et al. / Electrochimica Acta 76 (2012) 94–101

Fig. 9. Image of an alloy C-22 specimen after a PD–GS–PD test in pH 6, 5 M CaCl2 at Fig. 12. SEM image of an alloy C-22 specimen after a PD–GS–PD test in pH 6, 1 M
90 ◦ C. NaCl at 90 ◦ C.

Fig. 10. Image of an alloy C-22HS specimen after a PD–GS–PD test in pH 6, 5 M CaCl2
at 90 ◦ C.
Fig. 13. SEM image of an alloy C-22 specimen after a PD–GS–PD test in pH 6, 5 M
CaCl2 at 90 ◦ C.

Fig. 14. Crevice corrosion repassivation potentials (ECO ) from PD–GS–PD tests as a
Fig. 11. Image of an alloy HYBRID-BC1 specimen after a PD–GS–PD test in pH 6, 5 M function of the chloride concentration for all the tested alloys at 90 ◦ C.
CaCl2 at 90 ◦ C.
 N.S. Zadorozne et al. / Electrochimica Acta 76 (2012) 94–101 99

Fig. 15. ECO in NaCl solutions and ECORR at 24 h of immersion in deaerated 1 M HCl,
as a function of PREN for all the tested alloys.
∗
Fig. 16. ECO − ECORR =  + ˚ as a function of PREN for all the tested alloys.

3.3. Effect of alloy composition

Fig. 15 shows ECO as a function of PREN in chloride solutions of
Considering the polarization curves in Fig. 4, three ranges of different concentrations at 90 ◦ C. The value of ECORR after 24 h
potential may be identified as follows: of exposure to deaerated 1 M HCl is also plotted. ECO and ECORR
increased with PREN (Fig. 15). It is well known that the elec-
• Active range: from ECORR to −0.1 V vs SCE. trochemical behavior of metals in acidic chloride environments
• Passive range: from −0.1 to 0.8 V vs SCE. determines their susceptibility to localized corrosion in neutral
• Transpassive range: above 0.8 V vs SCE. chloride solutions [1,2]. Therefore, the present results are analyzed
in terms of the localized acidification model by Galvele [2]. This
In the active range, alloy HYBRID-BC1 showed the highest corro- model was originally developed for predicting the pitting corro-
sion resistance, followed by alloys C-22HS and C-22 (Figs. 3 and 4). sion potential. Applications of the model to the crevice corrosion
Alloy HYBRID-BC1 has 22% of molybdenum, while alloys C-22HS behavior of stainless alloys have been also reported [17,29]. Rincón
and C-22 have 17% and 13% of molybdenum, respectively (Table 1). Ortíz et al. [30] demonstrate that the initiation and repassivation
The corrosion rate data shows that the anodic dissolution rate in potentials for crevice corrosion are essentially the same, if they are
the active range of potentials was mainly controlled by the molyb- measured at near steady state conditions. Thus, the localized acid-
denum content in the alloys. On the contrary, alloy HYBRID-BC1 ification model may be used for predicting the crevice corrosion
showed the lowest corrosion resistance in the passive range (Fig. 4). repassivation potential according to Eq. (3).
The performances of alloys C-22HS and C-22 were identical in the ∗
ECO = ECORR +  + ˚ (3)
passive range. The chromium content of alloy HYBRID-BC1 (15% Cr)
is significantly lower than those of alloys C-22HS (21% Cr) and C-22 ECO may be expressed as the sum of three contributions: ECORR ∗ ,
(22% Cr) (Table 1). The anodic dissolution rate in the passive range which is the corrosion potential in the acidified solution within
is controlled by the content of chromium since a chromium-rich the crevice; , which is the polarization necessary to obtain a cur-
protective film is developed. In the transpassive range, the three rent density high enough to maintain a critical acidity within the
alloys showed a sharp increase in the current density. This may be crevice; and ˚ = IR, which is the ohmic potential drop within the
due to the oxidation of Cr3+ to Cr6+ within the passive film and its crevice (I is the circulating current and R is the ohmic resistance).
further release into the acidic solution [25]. ECO was measured in the present work using the PD–GS–PD tech-
Molybdenum is known to enhance the corrosion resistance in ∗
nique (Fig. 14). ECORR was ascribed as the ECORR measured at 24 h of
reducing acidic conditions (low potentials), while chromium is a immersion in deaerated 1 M HCl at 90 ◦ C. This solution was consid-
beneficial element under oxidizing acidic conditions (high poten- ered to be representative of the critical crevice solution (CCS) for
tials) [3,4,6,7]. That is the reason why alloy HYBRID-BC1 shows the the tested alloys. This is an approximation since the aggressiveness
best performance in hot deaerated HCl solutions. If ECORR of the cor- of the CCS is expected to increase as the corrosion resistance of the
roding environment is set to potentials higher than −0.1 V vs SCE alloys increases. However, ECORR for the three alloys in HCl is not a
because of the presence of strong oxidizing species such as oxy- strong function of the acid concentration provided that the alloys
gen, cupric or ferric ions, alloys C-22HS and C-22 will show a better remain in the active state [25].
performance than alloy HYBRID-BC1. The difference between ECO and ECORR ∗ is given by  + ˚ (Eq.
(3)). Fig. 16 shows  + ˚ for the tested alloys in NaCl solutions
3.4. Crevice corrosion kinetics as a function of PREN. The value of  + ˚ increased as cCl − of
the bulk solution decreased (Fig. 16). According to the model, 
3.4.1. The localized acidification model does not depend on cCl − and ˚ increases as cCl − decreases. It is
The PREN (Pitting Resistance Equivalent Number) is frequently observed that  + ˚ did not change significantly from cCl − = 4 M
used as an indicative measure of Cr-passivating alloys resistance to to cCl − = 10 M, indicating that ˚ was negligible (˚ → 0) for the
localized corrosion [1]. For nickel alloys, PREN is defined in Eq. (2) highest tested cCl − . Therefore, the value of  was estimated under
as a function of weight percentages of the alloying elements Cr, Mo these conditions of high cCl − .
and W. The PREN value is mainly affected by the Mo content and, Fig. 17 shows  and jCORR in the CCS as a function of PREN.
in a lesser extent, by the W content.  increased as PREN increased, while jCORR in the CCS decreased
as PREN increased. The localized acidification model states that
PREN = %Cr + 3.3(%Mo + 0.5%W) (2)  is the polarization necessary to reach a critical value of the
100 N.S. Zadorozne et al. / Electrochimica Acta 76 (2012) 94–101

drop. The above mentioned facts support the model prediction of a

negligible ohmic drop in concentrated chloride solutions.
There is further evidence in the recent literature and in the
present work supporting crevice corrosion propagation with neg-
ligible ohmic drop. There is a critical distance from the mouth of
the crevice at which crevice corrosion starts [33]. This critical dis-
tance is proportional to the ohmic drop in the context of the IR
mechanism of localized corrosion [20]. However, Evans et al. and
Carranza et al. observe crevice corrosion of alloy C-22 outside the
area occluded by the crevice former teeth, in 5 M CaCl2 , at 90 ◦ C
[34,35]. Bocher et al. report that crevice corrosion of 316 stainless
steel in 0.6 M NaCl at 50 ◦ C initiates close to the crevice mouth at
low potentials (E = 0 V vs SCE) and spread both inward and outside
the crevice with time [15]. Bocher et al. used a sodium salt suggest-
ing this type of attack it is not a cation effect, but a result of the
high chloride concentration. In the present work, crevice corrosion
Fig. 17.  and jCORR as a function of PREN for all the tested alloys. attack was also observed spreading outside the area occluded by the
crevice former teeth in the three tested alloys (Figs. 9–11 and 13).
product x·j, where x is the diffusion path and j is the current density In conditions of high chloride concentration and high temperature
within the pit or crevice. In the present work, the value of x was the critical distance at which crevice corrosion starts tends to be
constant since the crevicing mechanism was the same for all the near zero, and consequently negligible ohmic drop occurs.
alloys. The value of jCORR in the CCS for alloy C-22 was high enough It may be concluded that crevice corrosion may propagate with-
for the crevice corrosion to occur without a significant polariza- out any ohmic drop. The ohmic drop is not a necessary condition for
tion ( ≈ 0.010 V). For alloy C-22HS, a slightly higher polarization crevice corrosion to occur, but a consequence of the crevice corro-
was needed for the crevice corrosion to occur ( ≈ 0.025 V). On the sion process in certain conditions. The criteria of a critical potential
contrary, the jCORR in the CCS for alloy HYBRID-BC1 was low and a drop for the crevice corrosion initiation [20] cannot be applied to
significant polarization was needed for crevice corrosion to occur the systems studied in the present work.
( ≈ 0.135 V). It may be argued that the critical value of the product
x·j, is the same for the three tested alloys. However, the quantitative 4. Conclusions
condition of x·j > 0.01 A cm−1 used by Bocher et al. [17] is not rea-
sonable, since x > 100 cm would be necessary for j = 100 ␮A cm−2 . The crevice corrosion resistance of the tested nickel alloys bear-
Such a large diffusion path is unrealistic. ing chromium and molybdenum increased with PREN, which is
mainly affected by their Mo content. As PREN increased, the corro-
3.4.2. Ohmic control and charge transfer control of the kinetics sion rate in the critical crevice solution decreased and the resistance
The present results indicate that crevice corrosion propagation to crevice corrosion increased.
was under ohmic control in the more dilute chloride solutions, since Analyses of the data in terms of the localized acidification
large values of ˚ were calculated (Fig. 16). However, in the con- model clarified the crevice corrosion kinetics of the three alloys
centrated chloride solutions ˚ was negligible. Then, Eq. (3) leads at potentials slightly higher than the repassivation potential. The
∗
to ECO = ECORR + , which means that crevice corrosion propaga- polarization necessary to develop a critical chemistry within the
tion was under charge transfer control. This applies to the crevice crevice increased as PREN increased. Crevice corrosion propagation
corrosion of the three tested alloys at potentials slightly higher than was found to be under ohmic control in the more dilute chloride
the repassivation potential in 10 M chloride solutions, at 90 ◦ C. For solutions, but under charge transfer control in the more concen-
alloy C-22, Eq. (3) may be further reduced to ECO = ECORR ∗ , since  trated chloride solutions. Ohmic drop was not a necessary condition
was negligible (Fig. 17). This indicates that the ECO values obtained for crevice corrosion to occur, but a consequence of the crevice
in this work were conservative, since lower potentials than ECO corrosion process in certain conditions.
would lead to proton reduction and pH increase within the crevice.

Acknowledgments
3.4.3. Ohmic drop
˚ is a function of the crevice corrosion current density and
Financial support from the Agencia Nacional de Promoción
of the resistivity of the crevice solution and that of the corrosion
Científica y Tecnológica of the Ministerio de Educación, Ciencia y
products layers, usually precipitated on the surface of the corroding
Tecnología from Argentina and from the Universidad Nacional de
alloy [1]. Corrosion products formed within active crevices of alloy
San Martín is acknowledged. The authors are grateful to N. S. Meck
C-22 are rich in molybdenum and tungsten [16,31,32]. Polymeric
from Haynes International, who kindly supplied the tested alloys.
molybdates were detected within the crevice corroded region of
alloy C-22 in 5 M NaCl, at 120 ◦ C [32]. In the present work, cor-
rosion products were not observed in the tests in 10 M chloride References
solutions, but they were observed in the tests in less concentrated
[1] Z. Szklarska-Smialowska, Pitting and Crevice Corrosion of Metals, NACE Intl.,
solutions (Figs. 6–13). The crevice corrosion current density of alloy
Houston, TX, 2005.
C-22 at potentials slightly above ECO is reported to be less than [2] J.R. Galvele, Journal of the Electrochemical Society 123 (1976) 464.
100 ␮A cm−2 [16]. Other authors report a slightly higher current [3] R.B. Rebak, in: R.W. Cahn, P. Haasen, E.J. Kramer (Eds.), Corrosion and Environ-
density of 250 ␮A cm−2 [18]. These values are in agreement with mental Degradation, vol. II, Wiley-VCH, Weinheim, Germany, 2000, Ch. 2.
[4] D.C. Agarwal, N. Sridhar, in: R. Winston Revie (Ed.), Uhlig’s Corrosion Handbook,
the jCORR in 1 M HCl obtained in the present work (Fig. 17). If the 3rd ed., John Wiley & Sons, New York, NY, 2011, Ch. 59.
potential drop in the solution is neglected, as in the case of concen- [5] ASTMB575, Annual Book of ASTM Standards, vol. 02.04, ASTM Intl., West Con-
trated solutions, the overall potential drop will occur entirely over shohocken, PA, 2002.
[6] A.I. Asphahani, The Arabian Journal of Science and Engineering 14 (1989) 317.
the salt layer. For a salt layer thickness of 10 ␮m [31], a salt resistiv- [7] P.E. Manning, J.D. Smith, J.L. Nickerson, Materials Performance 27 (1988) 67.
ity as large as 100 m will be needed for causing a 1 mV-potential [8] G.M. Gordon, Corrosion 58 (2002) 811.
 N.S. Zadorozne et al. / Electrochimica Acta 76 (2012) 94–101 101

[9] L.M. Pike, D.L. Klarstrom, Paper 04239, Corrosion/2004, NACE Intl. Houston, TX, [22] ASTMG192-08, Annual Book of ASTM Standards, vol. 03.02, ASTM Intl., West
2004. Conshohocken, PA, 2008.
[10] N.S. Meck, L.M. Pike, P. Crook, Paper 08181, Corrosion/2008, NACE Intl. Houston, [23] ASTMG5-94(2004), Annual Book of ASTM Standards, vol. 03.02, ASTM Intl.,
TX, 2008. West Conshohocken, PA, 2005.
[11] L.M. Pike, P.E. Manning, E.L. Hibner, Paper 10319, Corrosion/2010, NACE Intl. [24] ASTMG102-89(2004), Annual Book of ASTM Standards, vol. 03.02, ASTM Intl.,
Houston, TX, 2010. West Conshohocken, PA, 2005.
[12] R.B. Rebak, Environmentally assisted cracking behavior of nickel alloys in oil [25] M.A. Rodríguez, R.M. Carranza, R.B. Rebak, Journal of the Electrochemical Soci-
and gas applications. A review, in: Proceedings of Eurocorr/2010, Moscow, ety 157 (2010) C1.
Russia, 13–17 September, 2010. [26] M.A. Rodríguez, Ph.D. Thesis, Instituto Sabato, UNSAM/CNEA, 2008.
[13] P. Crook, N. S. Meck, N. E. Koon, Paper 08190, Corrosion/2008, NACE Intl. Hous- [27] A.K. Mishra, G.S. Frankel, Corrosion 64 (2008) 836.
ton, TX, 2008. [28] C.M. Giordano, M. Rincón Ortíz, M.A. Rodríguez, R.M. Carranza, R.B. Rebak,
[14] R.S. Lillard, M.P. Jurinski, J.R. Scully, Corrosion 50 (1994) 251. Corrosion Engineering Science and Technology 46 (2011) 129.
[15] F. Bocher, F. Presuel-Moreno, J.R. Scully, Journal of the Electrochemical Society [29] N. Sridhar, G.A. Cragnolino, Corrosion 49 (1993) 885.
155 (2008) C256. [30] M. Rincón Ortíz, M.A. Rodríguez, R.M. Carranza, R.B. Rebak, Corrosion 66 (2010)
[16] M.A. Rodríguez, R.M. Carranza, R.B. Rebak, Corrosion 66 (2010) 015007. 105002.
[17] F. Bocher, R. Huang, J.R. Scully, Corrosion 66 (2010) 055002. [31] X. Shan, J.H. Payer, Journal of the Electrochemical Society 156 (2009) C313.
[18] P. Jakupi, J.J. Noël, D.W. Shoesmith, Corrosion Science 53 (2011) [32] P. Jakupi, F. Wang, J.J. Noël, D.W. Shoesmith, Corrosion Science 53 (2011) 1670.
3122. [33] B.A. Kehler, G.O. Ilevbare, J.C. Scully, Corrosion 57 (2001) 1042.
[19] P. Jakupi, J.J. Noël, D.W. Shoesmith, Corrosion Science 54 (2012) 260. [34] K.J. Evans, L.L. Wong, R.B. Rebak, Transportation, storage and disposal of
[20] H.W. Pickering, Journal of the Electrochemical Society 150 (2003) K1. radioactive materials, in: Proc ASME Pressure Vessel and Piping, 483, San Diego
[21] ASTMG48-03, Annual Book of ASTM Standards, vol. 03.02, ASTM Intl., West Conf. CA, 25–29 July 2004, 2004, p. 137.
Conshohocken, PA, 2005. [35] R.M. Carranza, M.A. Rodríguez, R.B. Rebak, Corrosion 63 (2007) 480.