Corrosion Science 52 (2010) 32433257

Contents lists available at ScienceDirect

Corrosion Science
journal homepage: www.elsevier.com/locate/corsci

A newly synthesized glycine derivative to control uniform and pitting

corrosion processes of Al induced by SCN anions Chemical, electrochemical
and morphological studies
Mohammed A. Amin *
Department of Chemistry, Faculty of Science, Ain Shams University, 11566 Abbassia, Cairo, Egypt
Materials and Corrosion Lab (MCL), Chemistry Department, Faculty of Science, Taif University, 888 Hawiya, Kingdom of Saudi Arabia

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

Article history: A newly synthesized glycine derivative (termed GlyD), 2-(4-(dimethylamino)benzylamino)acetic acid
Received 6 April 2010 hydrochloride, was used to inhibit uniform and pitting corrosion processes of Al in 0.50 M KSCN solutions
Accepted 29 May 2010 (pH 6.8) at 25 C. For uniform corrosion inhibition study, Tafel extrapolation, linear polarization resis-
Available online 4 June 2010
tance and impedance methods were used, complemented with SEM examinations. An independent
method of chemical analysis, namely ICP-AES (inductively coupled plasma atomic emission spectrome-
Keywords: try) was also used to test validity of corrosion rate measured by Tafel extrapolation method. GlyD inhib-
A. Aluminium
ited uniform corrosion, even at low concentrations, reaching a value of inhibition efciency up to 97% at a
C. Pitting corrosion
concentration of 5  103 M. Results obtained from the different corrosion evaluation techniques were in
good agreement. This new synthesized glycine derivative was also used to control pit nucleation and
growth on the pitted Al surface based on cyclic polarization, potentiostatic and galvanostatic measure-
ments. The pitting potential (Epit) and the repassivation potential (Erp) increased by the addition of GlyD.
Thus GlyD suppressed pit nucleation and propagation. Nucleation of pit was found to take place after an
incubation time (ti). The rate of pit nucleation and growth decreased with increase in inhibitor concen-
tration. Morphology of pitting was also studied as a function of the applied anodic potential and solution
temperature. Cross-sectional view of pitted surface revealed the formation of large distorted hemispher-
ical and narrow deep pits. GlyD was much better than Gly in controlling uniform and pitting corrosion
processes of Al in these solutions.
2010 Elsevier Ltd. All rights reserved.

1. Introduction Pure Al and Al alloys nd a wide spread spectrum of technolog-

Pitting is a form of localized corrosion in which metal is re-
moved preferentially from vulnerable areas on the surface [1].
The process of localized corrosion can be divided into a sequence
of steps [1]: initiation by the breakdown of the passive lm; meta-
stable growth of small pits on the verge of stability; stable growth
of localized corrosion sites that can grow quite large; and nally, if
conditions permit, repassivation or cessation of attack. Localized
corrosion is known to initiate above a critical potential and repass-
ivate below another, lower potential [1].
Determination of this critical potential (previously called break-
down potential of passivity, and more recently, critical potential
for pit nucleation, or pitting potential) is one of the primary goals
of pitting corrosion research because it permits quantitative evalua-
tion of the resistance versus susceptibility to pitting of a given mate-
rial under dened environmental conditions.
* Address: Department of Chemistry, Faculty of Science, Ain Shams University,
11566 Abbassia, Cairo, Egypt. Tel.: +20 222509331; fax: +20 24836831.
E-mail address: maaismail@yahoo.com
ical applications because of their particular properties such as low
density, good appearance and corrosion resistance. The breakdown
of the passive oxide lms on Al and its alloys, by aggressive anions
such as halides at sufciently positive anodic potentials is fre-
quently responsible for the failure of Al and its alloys in aqueous
halide solutions, because it usually leads to severe pitting of the
underlying metal. While there is plenty of data on pitting corrosion
of Al in different environments, especially those containing
Cl ions [212], no information, to the author knowledge, is avail-
able concerning the aggressive attack of anions other than Cl like,
for example, SCN anion. Recently, our research group presented
detailed study regarding corrosion, passivation and breakdown of
passivity of Al and AlZn alloys in KSCN solutions [13].
Indeed, KSCN solutions are not very commonly used in corro-
sion research of Al alloys. However, there are strong evidences
from our previous study proving the occurrence of the pitting cor-
rosion on the surface of Al and AlZn alloys as a result of passivity
breakdown by the aggressive attack of SCN anions [13]. The
breakdown of a protective passive lm leading to accelerated dis-
solution at localized sites is an important practical issue and a

0010-938X/$ - see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.corsci.2010.05.041
3244 M.A. Amin / Corrosion Science 52 (2010) 32433257
vexing scientic problem [1]. In addition, the study of interactions
between anions and aluminum is always interesting to understand
the behaviour of the metal in different environments. This makes
studying and controlling passivity breakdown and initiation of pit-
ting corrosion of Al in these solutions important, at least from the
academic point of view.
Recently, glycine (Gly) was successfully reported as a corrosion
safe inhibitor for some metals and alloys [1418]. In view of the
inhibition characteristics of Gly, a newly Gly derivative (GlyD),
namely 2-(4-(dimethylamino)benzylamino)acetic acid hydrochlo-
ride was synthesized in our lab to effectively inhibit uniform and
pitting corrosion processes of Al in KSCN solutions. Measurements
were carried out with the use of various electrochemical tech-
niques including Tafel polarization, linear polarization resistance
(LPR), impedance, cyclic polarization, potentiostatic, and galvano-
static techniques. An independent method of chemical analysis
(ICP-AES) was also used to determine the concentration of Al3+ in
solution due to uniform corrosion of Al in KSCN solutions without
and with inhibitor. The aim of the ICP method is to conrm electro-
chemical measurements and to test the validity of corrosion rate
measured by Tafel extrapolation method. The obtained results
were compared with results recorded for Gly. Morphologies of uni-
form and pitting corrosion processes were investigated as a func-
tion of the concentration of the tested inhibitor. The effect of
applied anodic potential and solution temperature on the mor-
phology of the pitted Al surface was also studied.
2. Experimental
The working electrode employed in this work was made of Al wire
(SigmaAldrich, 1.0 mm diam., 99.999). The pure Al wire was
mounted into Teon leaving a constant surface area of 0.00785 cm2
to contact the solution. Prior to each experiment, the electrodes were
rst briey ground with no. 600 emery paper, subsequently abraded
with no. 2000 emery paper, washed with deionized water, followed
by immediate rinsing with absolute ethanol.
The glycine derivative (GlyD) was synthesized in the laboratory
according to the procedure described below.
Glycine (0.75 g, 10 mmol) and 4-(dimethylamino)benzaldehyde
(1.49 g, 10 mmol) were dissolved in ethanol (20 ml) in the pres-
ence of KOH (0.56 g, 10 mmol) and vigorously stirred at room tem-
perature and reuxed for 2 h. A white precipitate was formed on
cooling, ltered off, washed with cold ethanol, and dried in vacuo.
The produced imine derivative was then added to a solution of
0.5 g (14 mmol) of NaBH4 in ethanol and the reaction was let to stir
for 4 h at room temperature. The result at solution was acidied by
concentrated HCl. A white precipitate of L.HCl was formed, ltered,
and dried up. Yield 1.98 (81%), yield 123 mg (89%). FT-IR (KBr): m
(cm1): 3428, 2947, 2360, 1751, 1635, 1564, 1539, 1472, 1420,
1209, 1051, 908, 665, 607, and 516. 1H NMR (500 MHz, DMSO-
d6) d (ppm): 13.56 (s, 1H, OH), 6.79 (s, 1H, NH), 7.62 (d, J = 7.8,
2H, CHbenzene), 7.42 (d, J = 7.8, 2H, CHbenzene), 4.07 (s, 2H,
CH2), and 2.95 (s, 6H, CH3).
All experiments (either chemical or electrochemical) were carried
out in 0.50 M KSCN solutions without and with various concentra-
tions (1055  103 M) of Gly or GlyD. All solutions were freshly
prepared from analytical grade chemical reagents using doubly dis-
tilled water and were used without further purication. For each
run, a freshly prepared solution as well as a cleaned set of electrodes
was used. Each run was carried out in aerated stagnant solutions at
the required temperature (1 C), using water thermostat.
Full immersion tests were performed in accordance with the
ASTM-G31 standard [19]. Exposure times between 1.0 and 30 days
were used. Corrosion rates were monitored in 0.50 M KSCN solu-
tions without and with various concentrations of Gly or GlyD as
a function of the immersion time (130 days) at 25 1 C using
inductively coupled plasma atomic emission spectroscopy (ICP-
AES) method of chemical analysis. After these immersion essays,
microstructural features of some of the exposed samples were ana-
lyzed by SEM examinations using Analytical Scanning Electron
Microscope JEOL JSM 6390 LA.
In the ICP method, the Al3+ ions concentration was determined
using PerkinElmer Optima 2100 Dual View inductively coupled
plasma atomic emission spectrometry (ICP-AES) instrument con-
nected with AS 93 Plus autosampler. The 40-MHz free-running
generator was operated at a forward power of 1.3 kW; the outer,
intermediate and Ar carrier gas ow rates were 15.0, 0.2 and
0.8 L/min, respectively. The carrier gas ow rate was optimized
to obtain maximum signal-to-background ratios. In all experi-
ments, the measured samples were nebulized downstream to the
plasma by the autosampler and the concentrations were automat-
ically determined using the standard calibration graph. The system
adjusted to measure the samples in triplicates and the relative
standard deviation was calculated. The RSD was <2% and the corre-
lation coefcient was >0.99998.
Electrochemical experiments were performed in a 100 ml vol-
ume Pyrex glass cell using Pt wire and a saturated calomel electrode
(SCE) as auxiliary and reference electrodes, respectively. The SCE
was connected via a Luggin capillary, the tip of which was very close
to the surface of the working electrode to minimize the IR drop. All
potentials given in this paper are referred to this reference electrode.
Electrochemical measurements were performed using Autolab fre-
quency response analyzer (FRA) coupled to an Autolab Potentio-
stat/Galvanostat (PGSTAT30) with FRA2 module connected to a
personal computer.
 M.A. Amin / Corrosion Science 52 (2010) 32433257 3245

Tafel extrapolation method and linear polarization resistance 300

1
(LPR) technique have been used alongside the ICP method to eval- (a)
uate the degree of protection against uniform corrosion supplied 2
250
by Gly and GlyD, as well as to obtain information about the mech-
3

-2
-2 -1
anism of the inhibition process. The linear Tafel segments of the Slope = 9.67 mg cm day

Dissolved Al / mg cm
cathodic curves were extrapolated to the corrosion potential (Ecorr) 200 -2 4
= 1.2 mA cm
to obtain the corrosion current densities (jcorr). Impedance mea-
surements were carried out using AC signals of amplitude 5 mV 150 5
peak to peak at the open circuit potential in the frequency range
100 kHz to 1.0 mHz.
100 6
Pitting corrosion inhibition degree has been evaluated in terms 7
of both nucleation of pits and growth of pre-existing pits, using
cyclic polarization, potentiostatic and galvanostatic measure- 50
ments. The potentiostatic j/t measurements were carried out after
holding the working electrode at constant anodic potential (Ea) 0
where the anodic current was recorded with time. The galvanostat- 0 5 10 15 20 25 30 35
ic E/t curves were recorded after holding the electrode at constant Time / day
anodic current (ja) where the anodic potential was recorded as a
function of time. The stabilization period prior to collecting data 300
was 12 h. The open circuit potential of the working electrode was 1
(b)
measured as a function of time during this stabilization time. This
250
time was quite sufcient to reach a quasi-stationary value for the
2

-2
open circuit potential.

Dissolved Al / mg cm
For surface morphology studies regarding pitting corrosion and 200
3
its control, some Al samples were exposed to pitting attack in
0.50 M KSCN solutions as a function of applied anodic potential 150
and solution temperature. Other samples were exposed to pitting
attack at a xed anodic potential in 0.50 M KSCN solutions in pres- 4
ence of various concentrations of Gly or GlyD at 25 C. Each sample 100

was held at the given potential for 2.0 min, and nally washed
5
thoroughly and submitted to 20 min of ultrasonic cleaning in order 50
to remove loosely adsorbed ions. Morphology of uniform corrosion 6
was also studied in the absence and presence of various concentra- 7
0
tions of Gly or its derivative. 0 5 10 15 20 25 30 35
Time / day
Fig. 1. Corrosiontime plots recorded for Al in aerated 0.50 M KSCN solutions
3. Results and discussion without and with various concentrations (1055  103 M) of (a) Gly or (b) GlyD at
25 C. (1) Blank; (2) 105 M inhib.; (3) 5  105 M inhib.; (4) 104 M inhib.; (5)
3.1. Inhibition of uniform corrosion 5  104 M inhib.; (6) 103 M inhib.; (7) 5  103 M inhib.

3.1.1. ICP-AES method of chemical analysis

these conditions was further conrmed here, as will be seen, using
The advantages offered by the ICP technique include a multi-
SEM.
element measurement capability, high stability, optimum selectiv-
Regarding to Fig. 1, the slope of each line (mass, in mg, of Al dis-
ity and superior dynamic range for concentration as well as
solved as Al3+ per unit area per unit time; mg cm2 h1) represents
calculation, data reporting and data reprocessing. These make the
the corrosion rate of Al at the specied conditions. The numerical
ICP-AES instrument is superior in the trace and environmental
values of these slopes were converted into the corresponding
analysis in aqueous and organic media in complex samples. In
the corrosion studies, such sophisticated instrument provides an
excellent continuous chemical monitoring and robust analytical Table 1
method for corrosion rate. (ICP-AES) provides therefore the most Various parameters derived from the ICP-AES measurements of Al in 0.50 M KSCN
reliable results concerning the efciency of a given inhibitor com- solutions without and with different concentrations of Gly or GlyD at 25 C.
pound, so that the corresponding corrosion data obtained from Inhibitor [Inhib.] Slope j IICP- tICP-AES
them approach service conditions more accurately than the data (M) (mg cm2 day1) (mA cm2) AES (%) (lm year1)
obtained with any other test, even weight loss. Blank 9.67 1.2 13.12
In the ICP experiments, known aliquots of the solution contain- Gly 105 8.70 1.08 10 11.81
ing dissolved Al3+ ions were withdrawn as a function of time and 5  105 7.57 0.94 21.76 10.28
Gly or GlyD concentration and were analyzed to produce the corro- 104 6.61 0.82 31.76 8.97
5  104 5.08 0.63 47.65 6.89
siontime curves depicted in Fig. 1a and b, respectively. In the 103 3.38 0.42 64.71 4.59
course of ICP experiments, localized corrosion is not supposed to 5  103 2.82 0.35 70.59 3.83
be formed. Otherwise, the results cannot be compared with those GlyD 105 7.82 0.97 18.82 10.61
obtained from the Tafel extrapolation method (see later), as this 5  105 6.29 0.78 35.29 8.53
DC electrochemical technique measures only the uniform corro- 104 3.95 0.49 59.41 5.36
sion rate. Indeed, after 30 days of immersion no pits were observed 5  104 2.01 0.25 78.82 2.73
103 1.05 0.13 88.82 1.42
under the optical microscope (50 magnication). Thus, extensive
5  103 0.32 0.04 97.06 0.44
pitting attack does not exist. The absence of pitting attack under
3246 M.A. Amin / Corrosion Science 52 (2010) 32433257

Fig. 2. SEM images obtained for Al, after a full immersion test of 30 days at 25 C, in aerated 0.50 M KSCN solutions without (image a) and with various concentrations of Gly
(image be), or GlyD (image fi).
 M.A. Amin / Corrosion Science 52 (2010) 32433257 3247

corrosion current density (jcorr) values using Faradays law, and
subsequently converted into corrosion rates, tICP-AES, in lm year1
(micrometer per year; the penetration rate of corrosion through a
metal) using the expression [20]:
mICP-AES 3280jcorr M=nd
where M is the atomic weight of Al (26.98154 g), n the number of
electrons transferred in the corrosion reaction (n = 3) and d the den-
sity of Al (2.70 g cm3). The corrosion currents obtained from the
ICP method were also used to evaluate inhibition efciency values,
IICP-AES (%), as a function of inhibitor concentration, using Eq. (2):
0 0
IICP-AES % 100jcorr  jcorr =jcorr 
0
where jcorr and jcorr are the corrosion current densities for uninhib-
ited and inhibited solutions. Table 1 collects all parameters derived
from the ICP method for Al in 0.50 M KSCN solutions without and
with various concentrations of Gly or GlyD at 25 C.
As the corrosion rate depends directly on the mass of Al dis-
solved, the addition of Gly or GlyD results in a decrease in this
parameter. This is clearly seen from the decrease of the mass of dis-
solved Al with inhibitor concentration. This decrease of the mass of
dissolved Al is signicant in presence of GlyD (Fig. 1b). These nd-
ings indicate that Gly and GlyD can be considered as inhibitors of
Al uniform corrosion in KSCN solutions, with GlyD being more
effective than Gly. It follows from the data of Table 1 that the cor-
rosion rate is suppressed, and therefore the corrosion inhibition
1 strengthened, with increase in Gly or GlyD concentration. GlyD
presented the lowest rates of corrosion, reecting its high inhibi-
tion performance compared to Gly, see later.
A surface lm of inhibitor is assumed to be formed, protecting
Al against uniform corrosion in SCN solutions. This was conrmed
from SEM examinations performed for the corroded and inhibited
Al samples, as shown in Fig. 2. In absence of inhibitor, SEM exam-
inations presented a corroded Al surface. The appearance of such
2
corroded areas may be attributed to the dissolution of alumina
layer due to corrosive attack of SCN anions and the increased
alkalinity of the solution in the vicinity of the electrode surface,
see Section 3.1.3.
On the other hand, in presence of inhibitor, there is an improve-
ment in the surface morphology due to the decrease in the cor-
roded areas caused by the inhibitor layer covering the electrode
surface. This improvement in the surface morphology is also signif-
icant in case of GlyD (images fi) more than Gly (images be).

-1190 -1190
1 1
(a) -1195 (a)
-1195 2 -1200
3 2
-1205
4 3
-1200
-1210
E / mV(SCE)

E / mV(SCE)

-1215
-1205 4
5 -1220

-1225
-1210
-1230
5
-1215 -1235

-1240

-1220 -1245
0 2 4 6 8 10 12 0 2 4 6 8 10 12

Time / h Time / h

-1200 -1215
tm 6
(b) tm (b)
-1220
-1205 7
tm -1225
6
-1210 8
7 -1230
E / mV(SCE)

-1235
E / mV(SCE)

-1215 12
8 -1240
9
-1220 12 -1245
9 11

11 -1250
-1225
-1255 10
-1230 -1260
10
-1265
-1235 0 2 4 6 8 10 12
0 2 4 6 8 10 12
Immersion time / h Immersion time / h
Fig. 3. Open circuit potential vs. time plots recorded for Al in aerated 0.50 M KSCN Fig. 4. Open circuit potential vs. time plots recorded for Al in aerated 0.50 M KSCN
solutions without and with various concentrations of Gly at 25 C. (1) Blank; (2) solutions without and with various concentrations of GlyD at 25 C. (1) Blank; (2)
105 M Gly; (3) 2.5  105 M Gly; (4) 5  105 M Gly; (5) 104 M Gly; (6) 105 M GlyD; (3) 2.5  105 M GlyD; (4) 5  105 M GlyD; (5) 104 M GlyD; (6)
2.5  104 M Gly; (7) 5  104 M Gly; (8) 103 M Gly; (9) 2.5  103 M Gly; (10) 2.5  104 M GlyD; (7) 5  104 M GlyD; (8) 103 M GlyD; (9) 2.5  103 M GlyD;
5  103 M Gly; (11) 7.5  103 M Gly; (12) 102 M Gly. (10) 5  103 M GlyD; (11) 7.5  103 M GlyD; (12) 102 M GlyD.
3248 M.A. Amin / Corrosion Science 52 (2010) 32433257

Therefore, it can be concluded that on the Al surface, after a full 2

1
immersion test of 30 days in aerated 0.50 M KSCN solutions con- (a)
taining various concentrations of Gly or GlyD, a protective lm of
the inhibitor is developed, and the protectiveness of such lm de- 1

log(j / mA cm-2 )
pends on the type and concentration of the introduced inhibitor. It 1
seems that the corrosion inhibition effect of the two tested inhib-
0 7
itors is due to their adsorption onto the aluminum surface blocking
the corrosion process, see more details in Section 3.3.
7
-1
3.1.2. OCP vs. time measurements
The open circuit potential curves were recorded as a function of
time for up to 12 h of immersion time to achieve a stable value. 1
-2
Figs. 3 and 4 show the variation of the open circuit potential
(OCP) with immersion time of Al in aerated solutions of 0.50 M
KSCN without and with various concentrations of Gly or GlyD at 7
-3
25 C, respectively. In the absence of inhibitor (curve 1), it can be -1600 -1400 -1200 -1000 -800 -600
seen, in the rst moments of immersion, how the open circuit po-
tential moves rapidly towards less negative values due to the ini- E / mV(SCE)
tial formation and growth of the passive oxide lm. Then the
electrode potential shifts rapidly towards the more negative values 2
1
during the rst 20 min of immersion. This trend may be attributed (b)
to dissolution of the protective alumina layer as the result of SCN
1
adsorption and subsequent formation of AlSCN soluble complex
log(j / mA cm-2 )
1
species (see later). Increased alkalinity of the solution layer close
to the electrode surface, due to cathodic processes, may also coop-
0
erate with SCN anions to damage the protective alumina layer;
more details are presented in Section 3.1.3. 7
For increasing immersion times, the open circuit potential goes -1
towards a more positive value and keeps a constant value (steady 7
potential) after about 90 min of immersion due to the formation of
1
a porous layer of alumina [21]. This steady potential corresponds to -2
the free corrosion potential, Ecorr, conrmed from polarization
studies, see later, of the metal. Similar results were previously ob-
7
tained [22]. -3
On the other hand upon introducing the inhibitor, new curve -1600 -1400 -1200 -1000 -800 -600
features were recorded, depending on the type and concentration
E / mV(SCE)
of the tested inhibitor. Addition of dilute concentrations of Gly
(curves 25 in Fig. 3a) or GlyD (curves 2 and 3 in Fig. 4a), for the
2
rst moment of immersion, shifts signicantly the electrode poten- Cathodic Tafel
region
tial to more negative values in comparison with the solution with-
out inhibitor (curve 1). The magnitude of this negative shift of (c)
potential depends on the type and concentration of the tested 1
log (j / mA cm-2 )

inhibitor. At the same time, the growth of the oxide lm is slowed

down due to the decrease in the slope of the initial linear part of E
vs. t plots. In addition, the steady state Ecorr drifts to more negative
0 jcorr
values without changing the general shape of the OCP vs. time Extrapolated
cathodic Tafel line
plots. This negative shift in Ecorr is a function of the concentration
Ecorr = -1192 mV(SCE) Ecorr
and type of the inhibitor. Values of Ecorr are always more negative
-2
in presence of GlyD. These results may refer to the signicant -1 jcorr = 1.12 mA cm
cathodic inhibiting effect of the GlyD compared with Gly, see also
Section 3.1.3. = - 230 mV dec
-1
c
For higher concentrations of the additive (>104 M Gly; Fig. 3b,
curves 612) and (>2.5  105 M GlyD; Fig. 4b, curves 4 and 5), the -2

OCP shifted rst to less negative values reaching a plateau. After a
certain time, the potential declined to a reasonably steady value.
This trend indicated the occurrence of two counter-acting pro-
cesses. The rst process being the formation of a protective layer
of the inhibitor on the electrode surface, and consequently de-
layed-action corrosion occurred shifting the OCP to nobler values.
The second process is corrosion, may be due to the corrosive attack
of SCN and OH anions (Section 3.1.3), which dragged the poten-
tial back towards active values.
This competition may explain the appearance of such plateau in
the corresponding OCP vs. time curves. The life-time of this pla-
teau, designated here as tm, is dependent upon the concentration
and type of the introduced inhibitor. Addition of increasing con-
-1600 -1400 -1200 -1000 -800 -600
E / mV(SCE)
Fig. 5. Potentiodynamic cathodic and anodic polarization curves recorded for Al in
aerated 0.50 M KSCN solutions without and with various concentrations
(1055  103 M) of (a) inhib. or (b) GlyD at 25 C. (1) Blank; (2) 105 M inhib.;
(3) 5  105 M inhib.; (4) 104 M inhib.; (5) 5  104 M inhib.; (6) 103 M inhib.; (7)
5  103 M inhib. (c) A cathodic and anodic polarization plot obtained for Al in
aerated 0.50 M KSCN solution at 25 C, representing calculations of the various
electrochemical parameters based on Tafel extrapolation method.
centrations of inhibitor enhances values of tm. These ndings indi-
cate that Gly and GlyD inhibit uniform corrosion of Al in SCN
solutions, conrming results of ICP, to an extent depending on their
 M.A. Amin / Corrosion Science 52 (2010) 32433257 3249

concentration. Thus, surface coverage, and therefore inhibition
performance of the additive increases with increase in its concen-
tration. It is obvious that the values of tm are always longer in pres-
ence of GlyD, reecting again the high inhibition performance of
GlyD compared with Gly.
However, the appearance of this arrest indicates that Gly and
GlyD, under these conditions, are not able to ensure effective cor-
rosion inhibition. The progressive negative shift in Ecorr with
increasing additive concentration may be explained on the basis
that the additive adsorbs preferentially on the electrode surface
impeding the cathodic sites (see more details in Section 3.1.3).
It is interesting to note that an increase of GlyD concentration
above 104 M almost prevented the decay of potentials to negative
values (see curves 612 in Fig. 4b); this trend not achieved for Gly
(inspect again Fig. 3). These results mean that GlyD can ensure ef-
cient corrosion inhibition much more than Gly, at least during the
tested immersion time (12 h). At Cinhib > 103 M, see curves 11 and
12 in Figs. 3 and 4, the OCP shifted to less negative values reaching
a reasonably steady value before declining to Ecorr, as in the case of
Gly. These ndings may also reect the ability of these two addi-
tives to inhibit the anodic process too. Similar results were previ-
ously obtained by K.M. Ismail [23] during inhibition of copper
corrosion in NaCl and HCl solutions using cysteine.
3.1.3. Tafel polarization
The nature of the inhibition process was also established on the
basis of Tafel polarization measurements. Thus, changes observed
in the polarization curves after inhibitor addition are usually used
as criteria to classify the inhibitor as cathodic, anodic or mixed
[24]. In addition, Tafel polarization measurements were performed
here in order to verify the correlation with the ICP measurements.
In our case (Fig. 5a and b), the modications caused by Gly or GlyD
addition are a negative shift in the corrosion potential (Ecorr), in the
direction of what was seen by means of OCP vs. time plots (Figs. 3
and 4), and a marked leftward displacement in the cathodic branch
and a rightward displacement in the anodic branch.
This displacement in the cathodic and anodic branches depends
on the type and concentration of the introduced inhibitor. These
are the typical features of mixed-type inhibitors with cathodic pre-
dominance. The presence of Gly or GlyD in KSCN solutions there-
fore inhibits both the anodic and cathodic processes to an extent
depending on their concentration.
Following Bethencourt et al. [25], the cathodic processes may be
reduction of dissolved O2 and/or H2O, as indicated in reactions (3)
and (4):
2H2 O 2e ! H2 " 2OH
O2 2H2 O 2e ! 4OH
As a result of these reactions, the local pH becomes more alka-
line. This increases the abundance of OH ions in the layer of solu-
tion adjacent to the electrode surface. These OH ions may adsorb
promoting Al dissolution with the generation of aluminate species
AlOH4  [26]:

Al OH AlOHads ! AlOHads e 5
AlOHads OH ! AlOH2ads e 6
AlOH2ads OH ! AlOH3ads e 7
AlOH3ads OH ! AlOH4 aq 8
The overall reaction of the Al dissolution process is described by
Eq. (9) [26]:
Al 4OH ! AlOH4 aq 3e 9
In parallel, Al oxidation by uniform corrosion may occur, and a
layer of aluminum oxide is formed, with the physical detachment
of gaseous hydrogen, according to the reaction [25]:
2Al 3H2 O ! Al2 O3 H2 " 10
 
SCN anions may also cooperate with OH to damage the pro-
tective alumina layer. This could be explained on the basis of the
coordination chemistry of SCN. It is well-known that SCN is an
ambidentate ligand, i.e., can be bonded with metal ions either via
nitrogen, sulphur or both. Coordination of a SCN ligand with me-
tal ions has been considered in several books and journals [2729].
Sasic et al. [30] studied the interaction between Al(III) and SCN
anion in aqueous solutions using Raman Spectroscopy Authors in
this study showed that SCN ligand can displace water molecules
in AlH2 O3
6 binding through nitrogen in aqueous solutions. They
revealed the presence of two soluble complex species, namely
AlNCSH2 O2
5 and AlNCS2 H2 O4 .
Based on the above arguments, the anodic dissolution of Al in
KSCN solutions could be explained as a series of adsorption reac-
tions involving catalysis by SCN anion at active surface sites fol-
lowing the equations:
Al NCS ! AlNCSads e 11
AlNCSads nH2 O ! AlH2 O3
n NCS 2e

12
where Eq. (12) is the dissolution step and AlH2 O3
n represents the
solvated Al3+ ion. If the rate of oxide lm formation (Eq. (10)) is fas-
3
ter than the rates of Eqs. (9), (11), and (12), lm healing (passiv-
ation) predominates. The converse causes uniform corrosion to
4
dominate. In all probabilities, if SCN anions adsorb strongly on Al
surface to promote dissolution (see Eqs. (11) and (12)), then these

Table 2
Various electrochemical parameters obtained from Tafel and LPR measurements of Al in 0.50 M KSCN solutions without and with different concentrations of Gly or GlyD at 25 C.

Inhibitor [Inhib.] (M) Ecorr (mV (SCE)) bc (mV dec1) jcorr (mA cm2) ITafel (%) tTafel (lm year1) Rp (kX cm2) ILPR (%)
Blank 1192 230 1.12 12.24 4.25
Gly 105 1195 232 1.02 9.25 11.15 4.66 8.87
5  105 1196 231 0.87 22.05 9.51 5.41 21.50
104 1197 228 0.78 29.95 8.53 6.30 32.55
5  104 1198 229 0.58 48.00 6.34 7.85 45.84
103 1226 230 0.40 63.88 4.37 11.48 62.98
5  103 1238 230 0.32 71.02 3.50 15.44 72.48
GlyD 105 1201 228 0.90 19.22 9.84 5.18 17.97
5  105 1210 227 0.71 37.00 7.76 6.65 36.05
104 1216 232 0.44 61.10 4.81 11.19 62.02
5  104 1226 231 0.25 77.78 2.73 18.42 76.93
103 1237 230 0.12 89.21 1.31 42.8 90.07
5  103 1252 230 0.03 96.89 0.33 166.67 97.45
3250 M.A. Amin / Corrosion Science 52 (2010) 32433257

species may be expected to remain bound to the solvated ion to 3.0

form soluble AlNCS complex species [30]. (a)
Q
Extrapolation of Tafel lines is one of the most popular DC tech-
2.5
niques for estimation of corrosion rate. However, it is obvious from Rs
the polarization plots of the present work that accurate evaluation
2.0 Rct

-Z im / kohm cm2
of corrosion rate (i.e., corrosion current density, jcorr) from anodic
branches, and therefore the anodic Tafel slope (ba), is impossible,
simply because the recorded experimental anodic polarization 1.5
curves do not exhibit linear Tafel regions. This deviation from Tafel Simulation
behaviour may be attributed to roughening of the surface as a re-
1.0
sult of the aggressive attack of SCN anions [13]. The absence of
linearity in anodic branches prevents linear extrapolation to the Measurements
corrosion potential, Ecorr. This was the reason why values of ba, cal- 0.5
culated from the software, were not included here. However, the
cathodic branch is under activation control and exhibits linearity 0.0
in accord with Tafel relationship. 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
The corrosion rate, and therefore the cathodic Tafel slope (bc) Z re / kohm cm2
may now be estimated accurately by extrapolating the cathodic
linear region back to Ecorr, as shown in Fig. 5c. Since the corrosion 8
rate (t) is directly proportional to the value of the corrosion current
density, the inhibition efciency, ITafel (%), was evaluated from the
(b)
measured jcorr values using Eq. (2). The recorded jcorr values were
also introduced in Eq. (1) to obtain the corrosion rate, tTafel, values 6
as a function of inhibitor concentration, see Table 2.
A decrease in the corrosion rate, corresponding to an increase in -Z im / kohm cm2
the inhibition efciency, can be clearly noticed. These results par-
4
allel those recorded in Table 1 coming from the ICP method of
chemical analysis. This agreement between ICP and Tafel extrapo-
lation methods conrms the validity of corrosion rates measured 7
by the later. 2 6
It is obvious that the shapes of the polarization plots for inhib- 5
ited electrodes are not substantially different from those of unin- 3 4
1 2
hibited electrodes. The presence of inhibitor decreases the
0
corrosion rate but does not change other aspects of the behaviour. 0 2 4 6 8 10 12 14 16
This means that the inhibitor does not alter the electrochemical
reactions responsible for corrosion. In addition, the absence of sig- Z re / kohm cm2
nicant changes in the cathodic Tafel slope in the presence of
4
inhibitor indicates that the hydrogen evolution is slowed down 50
by the surface blocking effect of the inhibitor. This indicates that
-Zim / kohm cm 2

the inhibitive action of Gly and GlyD, may be related to their

2
adsorption and formation of a barrier lm on the electrode surface. 40
3
2
-Z im / kohm cm2

30 0
0 2 4 6 8
Zre / kohm cm2
3.1.4. Linear polarization resistance measurements
For linear polarization resistance (LPR) measurements, a sweep
(c)
from 20 to +20 mV vs. open circuit potential at a sweep rate of 20
0.5 mV s1 was used and the polarization resistance (Rp) was mea- 7
sured from the slope of g vs. j curve in the vicinity of corrosion po-
tential. Since the electrochemical theory assumes that 1/Rp is 10
6
directly proportional to the corrosion rate, the inhibition efcien-
cies, ILP (%) of the tested inhibitors were calculated from (Rp) values 5
4
obtained from LP data at different inhibitor concentrations and 0
0 20 40 60 80 100
temperatures, using Eq. (13):
2
Z re / kohm cm

ILP % fRp  R0p =Rp g100 13
where R0p and Rp are the polarization resistance values without and
with the addition of inhibitor, respectively. The polarization resis-
tance (Rp) values of Al in 0.50 M KSCN solutions increase with in-
crease in the concentration of the two inhibitors (Table 2). The
increase in the Rp value suggests that the inhibition efciency in-
creases with the increase in the inhibitor concentration, with GlyD
being more effective than Gly (inspect ILP (%) values presented in
Table 2.
Fig. 6. Complex-plane impedance plots recorded for Al in aerated 0.50 M KSCN
solutions (a) without and with various concentrations (1055  103 M) of (b)
inhib. or (c) GlyD at the respective corrosion potentials at 25 C. (1) Blank; (2)
105 M inhib.; (3) 5  105 M inhib.; (4) 104 M inhib.; (5) 5  104 M inhib.; (6)
103 M inhib.; (7) 5  103 M inhib.
3.1.5. Impedance measurements
Fig. 6a represents a complex-plane impedance plot recorded for
Al in 0.50 M KSCN solution at Ecorr at 25 C. It follows that a de-
pressed charge-transfer semi-circle was observed. Brett [3,4,31]
 M.A. Amin / Corrosion Science 52 (2010) 32433257 3251

assigned this semi-circle to the reactions involved at the oxide
layer formation. Although all these reactions involved at the oxide
layer formation are present, just one semi-circle (i.e., time con-
stant) in the Nyquist diagrams is observed. This could either mean
that the time constants of the individual processes, suggested by
Brett [3,4,31], strongly overlap, or that one process dominates
and, therefore, excludes the other processes in this frequency
range.
To determine the impedance parameters of Al in these solu-
tions, the measured impedance data were analyzed by using the
computer program EQUIVCRT [32], based upon the electric equiv-
alent circuit presented in the insert of Fig. 6a. This equivalent cir-
cuit was also successfully used to interpret the impedance data
presented in our previous study [13]. A good t with this model
was obtained with our experimental data. It is observed that the
tted data match the experimental, with an average error of about
46%. The symbols in all impedance plots represent the experi-
mental data, while the solid lines represent the best ts.
Fig. 6b and c presents the complex-plane impedance plots re-
corded for Al in 0.50 M KSCN solutions without and with various
concentrations of Gly or GlyD, respectively. Measurements were
carried out at the respective corrosion potentials at 25 C. In all
cases, the Nyquist plots are also characterized by a depressed
capacitive time constant, the diameter of which is a function of
the type and concentration of the tested inhibitor.
In all cases, it is clear that the shapes of the impedance plots for
inhibited electrodes are almost similar to those of uninhibited elec-
trodes. The presence of either Gly or its derivative increases the
impedance but does not change other aspects of the impedance
behaviour. These results support the results of polarization mea-
surements that the inhibitor does not alter the electrochemical
reactions responsible for corrosion. It inhibits corrosion primarily
through its adsorption on the metal surface.
Table 3 collects the tting results recorded for Al in 0.50 M
KSCN solutions as a function of the concentration of the two tested
inhibitors. Data presented in Table 3 reveal that the charge-trans-
fer resistance (Rct) increases with increase in inhibitor concentra-
tion. At the same time the double layer capacitance (Cdl) has
opposite trend at the whole concentration range.
The decrease in Cdl is due to the gradual replacement of water
molecules by the adsorption of the organic molecules at metal/
solution interface, resulting in a protective lm formation on the
electrode surface, then decreasing the extent of dissolution reac-
tion [33]. The values of Rct are always greater, while those of Cdl
are smaller in presence of GlyD. These ndings conrm the highest
inhibition performance of ClyD. The systematic increase of n values
with concentration of the two inhibitors may be explained by
decreasing in surface heterogeneity, due to the adsorption of the
inhibitor on the most active adsorption sites. The value of n is al-
ways higher, at any given concentration, in presence of GlyD,
reecting the higher adsorption afnity of GlyD, and hence its high
protection performance, compared to Gly itself.
The inhibition efciency, IZ (%) was estimated from the mea-
sured Rct values using the relation:
IZ % fRct  R0ct =Rct g100 14
where R0ct
and Rct are the charge-transfer resistance values in the ab-
sence and presence of the inhibitor, respectively. It is obvious that
the inhibition efciency values obtained from impedance technique
(Table 3) are in good agreement with those recorded using ICP-AES
(Table 1) and Tafel extrapolation and LPR (Table 2) methods.
3.2. Morphology of pitting and its inhibition
3.2.1. Cyclic polarization measurements
Cyclic polarization measurements were recorded for Al in
0.50 M KSCN solutions devoid of and containing various concentra-
tions of Gly or GlyD at a scan rate of 0.50 mV s1 at 25 C. In all
cases, the potential scan was reversed at the same current density
to evaluate the inuence of Gly and GlyD on the pit growth.
Fig. 7ac depict, as an example, results obtained for Al in 0.50 M
KSCN solutions without and with 103 M Gly or GlyD. Similar re-
sults were obtained for the other tested concentrations of the
two inhibitors (not included here). The obtained plots had the
familiar form for Al showing a well-dened corrosion potential,
Ecorr, followed by a passive region.
The current passing through this passive region is termed the
passive current (jpass). The passive region extends up to the pitting
potential (Epit). When Epit is reached, a drastic increase in jpass is ob-
served, denoting breakdown of the protective alumina layer and
initiation of pitting attack. Once initiated, the attack is stabilized
by the delocalized development of aggressive conditions as the re-
sult of metal cation hydrolysis and migration of the aggressive an-
ion [13].
Fig. 8 shows morphologies of passivated (image a) and pitted Al
surfaces in 0.50 M KSCN solutions as a function of the applied ano-
dic potential and solution temperature (images bh). Morphologies
of pitted surfaces in presence of Gly or GlyD (images il) were also
studied, see later. It was observed that no pitting phenomena oc-
curred on the Al specimen exposed to an anodic potential within
the passive region at room temperature, see image (a). On the con-
trary, if the sample is polarized at xed anodic potentials Ea beyond
Epit, passivity breakdown and initiation and propagation of pitting
occur (see images bd). In all cases, the formed pits are surrounded
on all sides by the regions covered with protective oxide layer. It is
obvious that, with increase in potential, the pitted areas increase at
the expense of the areas of the protective oxide layer. This means
that the ratio of pitted area to total surface area (i.e., pit area den-
sity) increased with increase in potential, corresponding to in-
creased severity of pitting.

Table 3
Various electrochemical parameters obtained from impedance measurements of Al in 0.50 M KSCN solutions without and with different concentrations of Gly or GlyD at 25 C.

Inhibitor [Inhib.] (M) Rs (kX cm2) Q (sn (x1 cm2)) Rct (kX cm2) n Cdl (lF cm2) IZ (%)
Blank 0.005 10.01 4.32 0.82 22.88
Gly 105 0.005 9.51 4.84 0.82 22.05 10.75
5  105 0.007 9.38 5.46 0.83 21.00 20.88
104 0.008 9.14 6.44 0.85 18.77 32.92
5  104 0.009 8.44 8.04 0.87 15.84 46.25
103 0.009 7.01 12.24 0.88 12.87 64.71
5  103 0.011 6.16 16 0.90 10.25 73.00
GlyD 105 0.006 10.72 5.39 0.86 20.75 19.81
5  105 0.009 11.24 6.75 0.91 17.25 35.96
104 0.009 8.89 10.81 0.92 13.22 60.05
5  104 0.012 7.54 20.69 0.93 11.02 79.12
103 0.014 5.60 48.11 0.95 7.52 91.02
5  103 0.017 4.61 105.62 0.96 5.97 95.91
3252 M.A. Amin / Corrosion Science 52 (2010) 32433257

ently composed of several smaller pits and some pits were formed
2 (a) at the bottom of previous formed pits. As pits begin to form, the lo-
cal environments change. Solution concentration and potential
Erp gradients develop, which can affect pitting susceptibility on adja-
1
cent surface areas [34,35]. The increase in pitting susceptibility
log (j / mA cm-2)

with increase in potential could be explained on the basis that

increasing the applied potential increases the active sites available
0 for pit nucleation [36]. Moreover, an increase in the applied poten-
Epit tial may increase the electric eld across the passive lm and
therefore enhances the adsorption of the aggressive SCN anions
-1 on the passive electrode surface [13]. Previous studies, based on
XPS examinations, also revealed increased adsorption of Cl [10]
 
and ClO3 and ClO4 anions [37] on the passivated Al surface.
-2 Ecorr o
Surface morphologies also revealed the occurrence of intense
R = 1280 pitting corrosion on the specimen exposed to high temperatures
(compare between images d, e and f), where deep pits with wide
-1600 -1200 -800 -400 0 400 800 1200 1600 2000 mouths result. Also, it can be said here that the pit area density in-
E / mV(SCE) creased with solution temperature. The radial growth of large pits
seems to begin by the coalescence of small pits. It seems that the
2
rate of radial growth of large pits as well as their penetration rate
(b) increase with temperature, yielding a rough etched surface with
deep pits, and therefore intense pitting takes place.
1 Increasing temperature enhances the solubility of Al2O3 [38].
Erp This makes the thickness of the passive lm decrease and its poros-
ity increase and hence a less protective lm results. An increase in
log (j / mA cm-2)

0 temperature also enhances the rate of diffusion and migration of

the reactant and product species to and from the electrode surface.
-1
In our previous study [13], EDX spectra revealed increased adsorp-
Epit tion of aggressive SCN anions on Al at high temperatures as a re-
sult of increased porosity of the passive lm. Augustynski et al.
-2 [39], using XPS method, came to the same conclusion. They also
observed an increased adsorption of Cl at high temperatures.
Ecorr The micrographs presented in Fig. 8g and h represent section of
-3
R = 1740
two pitted areas showing the mouth of the pit at the top of the
micrograph, and the cavity developing underneath the surface. A
-1600 -1200 -800 -400 0 400 800 1200 1600 2000 distorted hemispherical (image g) and a narrow deep (image h)
E / mV(SCE) pits with a wide mouth were observed. It seems that the pits are
growing underneath the surface.
Referring again to Fig. 7, the reverse scan shows a wide hyster-
2 esis cycle, this being characteristic of passivity breakdown on the
(c)
upward sweep and repassivation. Finally, a repassivation potential,
1 Erp, was attained. The existence of a hysteresis loop in a cyclic
potentiodynamic polarization curve indicates that repassivation
Erb
of an existing pit is more difcult when the potential is scanned to-
0
log (j / mA cm-2)

ward the negative direction. The larger the hysteresis loop, the
more difcult the repassivation is. Epit represents the potential lim-
-1 it above which the formation of pitting begins. Erp refers to the lim-
Epit it below which the metal remains passive, and it is dened as the
potential where the forward and reverse scan crosses; it marks the
-2
division between stable and unstable passivity. Intermediate val-
ues between Epit and Erp did not permit the formation of new pits,
-3
Ecorr
but allowed the development of those which already existed. The
R = 2618
narrower the hysteresis loop the easier it becomes to repassivate
-4 the pit.
-1600 -1200 -800 -400 0 400 800 1200 1600 2000 It is obvious that the pitting potential and the repassivation po-
tential were shifted towards positive potentials by the addition of
E / mV(SCE)
either Gly or GlyD to an extent depending on type of the tested
Fig. 7. Cyclic polarization curves recorded for Al in aerated 0.50 M KSCN solutions inhibitor and its concentration. These ndings indicate suppression
(a) without and with (b) 103 M Gly or (c) 103 M GlyD at a scan rate of 0.5 mV s1 of both pit nucleation and propagation. This positive shift of Epit
at 25 C. All polarization curves were reversed at the same current density with the and Erp was found to be signicant in presence of GlyD, suggesting
same scan rate. again that GlyD is more effective than Gly in inhibiting both pit
nucleation and growth on the pitted Al surface.
Based on the above argument, cyclic polarization technique is
For the electrode polarized just after Epit (image b), small pits useful in determining the degree of protection against pitting
begin to form on the surface. As the potential is made more posi- corrosion provided by Gly and its derivative as well as their effect
tive, larger pits form (images c and d). These large pits were appar- on the characteristics steps of this form of localized attack, i.e.,
 M.A. Amin / Corrosion Science 52 (2010) 32433257 3253

Fig. 8. SEM images obtained for Al subjected to passivation conditions (image a) and pitting attack in aerated 0.50 M KSCN solutions as a function of applied anodic potential
(images bd) and solution temperature (images df). Other SEM images were recorded for Al in 0.50 M KSCN solutions in presence of various concentrations of Gly (images i
k) or GlyD (images ln). Conditions of pitting attack; each sample is potentiostatically held at the given potential for 2.0 min at the specied conditions.

nucleation of new pits and growth of pre-existing pits. According
to [40], two factors should contribute to the overall pitting corro-
sion resistance. First, the resistance against the nucleation of new
pits and, secondly, a growth resistance of pre-existing pits. The for-
mer has been evaluated as the difference, R, between the pitting
potential, Epit, and the corrosion potential, Ecorr. Though this R
parameter cannot be used in absolute terms, it is very useful when
used as a comparative parameter. Thus, a higher R value refers to a
higher resistance against pit nucleation.
The protection degree against pitting nucleation, designated
here as Ppit, was therefore estimated in terms of the percentage in-
crease of parameter R. The reference value of this parameter, i.e.,
R0, was taken in the cyclic polarization experiment carried out in
bare KSCN solution, as shown in Fig. 7a. Fig. 9 shows the degree
of protection against pitting nucleation, Ppit, as a function of Gly
and GlyD concentration. Note the increase of Ppit with concentra-
tion of additives. Values of Ppit are always greater, as expected, in
presence of GlyD.
The resistance against growth of pre-existing pits was evaluated
by comparing the areas of the anodic hysteresis loops observed in
the cyclic polarization diagrams plotted in the j/E style. Using a
specic routine of the Software, the areas of the anodic hysteresis
loops were calculated. This area is related to the charge consumed
during the growth of the pre-existing pits. The percentage decrease
of this area, from the value observed for the blank solution, was ta-
ken as a measure for the protection degree, designated as Pg,
against pitting growth, see Fig. 10. From Figs. 9 and 10, addition
of Gly or GlyD to the aggressive SCN solutions resulted in an im-
proved behaviour against pitting corrosion events (i.e., pit initia-
tion and propagation), with GlyD being the most effective one.
Potentiostatic and galvanostatic measurements were also per-
formed to gain more information concerning the ability of Gly
and GlyD to control pit initiation and growth of Al in KSCN solu-
tions. Fig. 11a and b represent, as an example, results obtained
for Al in 0.50 M KSCN solutions without and with 103 M Gly of
GlyD. Similar trends were obtained for the other concentrations
of the two tested inhibitors. It follows from Fig. 11a that, in absence
of the inhibitor, the anodic current density initially decreases, due
to oxide lm growth, to a minimum value determining the charac-
teristic pitting parameter, namely the incubation time (ti). These
data reveal that an incubation time is necessary before pit initia-
tion. The magnitude of ti reects the susceptibility of sample to pit-
ting corrosion.
Beyond ti, the current density increases rapidly and nally
reaches a steady state value. The rise in the current density (i.e.,
the pit growth current density (jpit)) indicates pit growth with
3254 M.A. Amin / Corrosion Science 52 (2010) 32433257

Fig. 8 (continued)

time, while the steady state is due perhaps to pits blocked by cor-
rosion products that accumulated on the anode surface. Addition of
either Gly or GlyD decreases Al susceptibility towards pitting. This
is clearly seen from the decrease in jpit and the subsequent increase
in ti. This decrease in susceptibility towards pitting depends on the
type of the introduced inhibitor, with GlyD being more effective
than Gly.
Regarding to Fig. 11b, the linear increase in potential at the
commencement of anodization accompanies the initial formation
and growth of the oxide lm. A potential maximum appears at a
certain time, i.e., the incubation time (ti). This maximum is thought
to correspond, as will be shown, to the pitting potential (Epit). The
appearance of which is the correspondence of competition be-
tween two processes, namely, further oxide lm growth and its
breakdown.
The decline in potential following the maximum corresponds to
breakdown of passive layer followed by the nucleation and forma-
tion of pits as a result of the aggressive attack of SCN anions [13].
After prevalence of the latter reaction at Epit, or in other words,
after ti, the potential drops to a steady value, corresponding to
the repassivation potential (Erp). It was found that the values of Epit
and Erp obtained from cyclic polarization and galvanostatic
 M.A. Amin / Corrosion Science 52 (2010) 32433257 3255

160 60
(a) blank
140 GlyD
50
120
Gly
40
100

j / mA cm-2
80
Ppit (%)

30

60
Gly
20
40
ti
10 ti ti
GlyD
20

0 0
0.000 0.001 0.002 0.003 0.004 0.005 0.006 0 10 20 30 40 50 60 70 80 90 100

C inhib / M Time / s

Fig. 9. Estimated values of the degree of protection against pitting nucleation, Ppit, ti
as a function of Gly and GlyD concentration at 25 C.
1600 (b) Epit

1200
100 ti Erp
GlyD
800
90
E / mV(SCE)
GlyD Epit
ti
80 400
70 Epit
0
60
Gly
Erp
Pg (%)

50 Gly -400 Erp

blank
40
-800
30

20 0 10 20 30 40 50 60 70 80 90 100
Time / s
10

0 Fig. 11. (a) Potentiostatic currenttime plots recorded for Al in 0.50 M KSCN
solution without and with 103 M Gly or GlyD at an applied anodic potential of
0.000 0.001 0.002 0.003 0.004 0.005 0.006
800 mV (>Epit) at 25 C. (b) Galvanostatic potentialtime plots recorded for Al in
C Gly / M 0.50 M KSCN solution without and with 103 M Gly or GlyD at an applied anodic
current of 50 lA at 25 C.
Fig. 10. Degree of protection values, Pg (%), against pit growth as a function of Gly
and GlyD concentration at 25 C.
36
a
32

measurements are in good agreements (compare between Figs. 7 28

Ecorr
and 11b). These ndings conrm that the potential maximum PZC
Cdl / F cm-2

observed in the E/t plots corresponds to Epit. 24

It follows from the data of Fig. 11b that the rate of potential in-
crease (i.e., the rate of oxide growth) and the value and location of 20
potential maximum depend on the type of introduced inhibitor. Ecorr
b
The value of the potential maximum (i.e., Epit) increases and its 16 PZC
location shifts towards more positive value (i.e., longer ti values) c
in presence of these inhibitors. These ndings conrm results of 12 Ecorr
PZC
cyclic polarization measurements that these organic compounds
inhibited pitting corrosion of Al in KSCN solutions. 8

Referring again to the SEM images presented in Fig. 8, it is obvi- -1500 -1400 -1300 -1200 -1100 -1000 -900 -800 -700
ous that the surface morphology varies depending on the concen-
E / mV(SCE)
tration and type of tested inhibitor, compare image (d) (in absence
of the inhibitor) with images (in) (in presence of either Gly or Fig. 12. The Cdl vs. Ea plots recorded for Al in 0.50 M KSCN solutions (a) without and
GlyD). The pitted areas obviously diminished in presence of Gly (b) with 03 M Gly or (c) 103 M GlyD.
(images ik) or GlyD (images ln). Signicant decrease in pitted
areas was again observed in presence of GlyD more than in pres- pitting attack is prevented and the surface is efciently separated
ence of Gly. Surface morphologies in presence of GlyD showed that from the corrosive medium.
3256 M.A. Amin / Corrosion Science 52 (2010) 32433257
3.3. Mechanism of inhibition
It is well-known that, the phenomenon of adsorption is inu-
enced by the nature and surface charge of the metal and by the
chemical structure of inhibitors. Terminal oxygen atoms at metal
oxide surfaces react with water, forming hydroxylated sites that
impart a pH-dependent surface charge. Surface charge depends
on activities of potential-determining ions (H+ and OH) and elec-
trolyte concentrations (ionic strength, I) [41]. The pH where the net
total particle charge is zero is called the point of zero charge (PZC).
The point of zero charge (PZC) is an important parameter charac-
terizing the adsorption properties of metal oxides and related
materials [42,43].
The pH of the potential of zero charge (PZC) for aluminum oxi-
des/hydroxides is 9 [44,45]. Depending on solution pH, the elec-
trode surface can bear net negative, or positive, or no charge. In
0.04 M KSCN solution (pH 6.8), therefore oxide surface is positively
charged. This positive charge of the electrode surface at Ecorr was
further conrmed via dening the position of Ecorr with respect
to the respective potential of zero charge (PZC) Eq = 0 [46]. When
the difference, u (u = Ecorr  Eq = 0), is negative, the electrode sur-
face acquires a negative net charge and the adsorption of cations
is favoured. On the contrary, the adsorption of anions is favoured
when u becomes positive.
It has been shown in the literature that ac impedance studies
can be used to evaluate the potential of zero charge (PZC) from
the capacitance (Cdl) versus applied potential (E) plot [4749].
Fig. 12 represents the variation of Cdl with E recorded for Al in
0.50 M KSCN solutions in the absence (curve a) and presence of
103 M Gly (curve b) or GlyD (curve c). The minima on the Cdl ver-
sus E curves are considered as the value of PZC of the electrode. It is
obvious from Fig. 12 that the surface charge of Al in KSCN solutions
at the free corrosion potential is positive (u = [(1192) 
(1254) = +62 mV (SCE)]). In absence of the inhibitor, this positive
charge of the oxide lm may favour the electrostatic adsorption of
the aggressive SCN anions.
Addition of GlyD provides KSCN solution with Cl anions. These
Cl anions adsorb, in competition with SCN anions, at the oxide/
 the repassivation kinetics of pure aluminium, Electrochim. Acta 40 (1995)
solution interface through electrostatic attraction force because
of the excess positive charge at this interface at the corrosion po-
tential. This process changes the solution side of the interface from
positive to negative. Thus, the quaternary ammonium cations in
the GlyD molecule, HN+(CH3)2, electrostatically adsorb on the
oxide surface covered with primarily adsorbed Cl and/or SCN an-
ions. On the other hand, the inhibitive inuence of Gly could be ex-
plained on the basis that negative charge is present at the Al
surface, due to the adsorption of SCN anions, and this favours
the electrostatic adsorption of cations of Gly.
This type of electrostatic adsorption decreases the positive ex-
cess charge on the Al surface. This may explain the positive shift
of PZC upon adding Gly or GlyD (see again curves b and c in
Fig. 12) compared with PZC for bare KSCN solutions (curve a). It
is probable that, in case of GlyD, the adsorbed nitrogen atoms are
oriented with the benzene ring structure parallel to the metal sur-
face (due to aromaticity of the GlyD molecule). This mode of
adsorption suggests that the adsorbed GlyD molecule covers a
large area, thereby inhibiting Al corrosion effectively more than
Gly. Finally, I would say, this study could be initial ndings to
introduce Gly and its derivatives to control corrosion on Al and
Al alloys.
4. Conclusion
The presence of a new synthesized glycine derivative in KSCN
solutions inhibited uniform and pitting corrosion process of Al to
an extent depending on its concentration. Corrosion and passive
currents were reduced, and the pitting potential was increased
with increase in additive concentration. Galvanstatic and potentio-
static measurements showed that the rate of pit initiation and
growth decreased with increase in additive concentration. Polari-
zation measurements showed that the glycine derivative acted as
a mixed-type inhibitor with cathodic predominance. SEM examin-
ations of the electrode surface showed that the aggressive pitting
attack of SCN anions decreased with increase in additive concen-
tration. The inhibition efciency of the newly synthesized glycine
derivative was found to be higher than that of glycine itself.
Acknowledgment
Dr. Mohamed M. Ibrahim (Permanent address: Chemistry
Department, Faculty of Education, Kafrelsheikh University, Kafrel-
sheikh 33516, Egypt; Present address: Department of Chemistry,
Faculty of Science, Taif University, 888 Hawaiya, KSA) is acknowl-
edged for providing me with the newly synthesized glycine
derivative.
References
[1] K. Videm, The Electrochemistry of Uniform Corrosion and Pitting of
Aluminium, Kjeller Report 62, Institute for Atommenergi, Kjeller, Norway,
1974.
[2] H.J.W. Lenderink, M.V.D. Linden, J.H.W.D.E. Wit, Corrosion of aluminium in
acidic and neutral solutions, Electrochim. Acta 38 (1993) 19891992.
[3] C.M.A. Brett, Studies on aluminium corrosion in hydrochloric acid solution,
Portug. Electrochim. Acta 7 (1989) 123126.
[4] C.M.A. Brett, The application of electrochemical impedance techniques to
aluminium corrosion in acidic chloride solution, J. Appl. Electrochem. 20
(1990) 10001003.
[5] P.L. Cabot, F.A. Centellas, E. Perez, R. Loukili, Pitting and repassivation
processes of AlZnMg alloys in chloride solutions containing sulphate,
Electrochim. Acta 38 (1993) 27412748.
[6] F.M. Al-Khara, W.A. Badawy, Inhibition of corrosion of Al 6061, aluminum,
and an aluminumcopper alloy in chloride-free aqueous media: Part 2
behavior in basic solutions, Corrosion 54 (1998) 3773892.
[7] J.-D. Kim, S.-I. Pyun, Effects of electrolyte composition and applied potential on
18631869.
[8] W.C. Moshier, G.D. Davis, J.S. Ahearn, The corrosion and passivity of aluminum
exposed to dilute sodium sulfate solutions, Corros. Sci. 27 (1987) 785801.
[9] E. McCafferty, A competitive adsorption model for the inhibition of crevice
corrosion and pitting, J. Electrochem. Soc. 137 (1990) 37313737.
[10] A. Kolics, J.C. Polkinghorne, A. Wieckowski, Adsorption of sulfate and chloride
ions on aluminum, Electrochim. Acta 43 (1998) 26052618.
[11] H. Bohni, H.H. Uhlig, Environmental factors affecting the critical pitting
potential of aluminum, J. Electrochem. Soc. 116 (1969) 906910.
[12] S.S. Abdel Rehim, H.H. Hassan, M.A. Amin, Chronoamperometric studies of
pitting corrosion of Al and (AlSi) alloys by halide ions in neutral sulphate
solutions, Corros. Sci. 46 (2004) 19211938.
[13] M.A. Amin, S.S. Abd El-Rehim, E.E.F. El-Sherbini, S.R. Mahmoud, M.N. Abbas,
Pitting corrosion studies on Al and AlZn alloys in SCN solutions, Electrochim.
Acta 54 (2009) 42884296.
[14] M. Zerfaoui, H. Oudda, B. Hammouti, S. Kertit, M. Benkaddour, Inhibition of
corrosion of iron in citric acid media by aminoacids, Prog. Org. Coat. 51 (2004)
134138.
[15] M.A. Amin, K.F. Khaled, S.A. Fadl-Allah, Testing validity of the Tafel
extrapolation method for monitoring corrosion of cold rolled steel in HCl
solutions experimental and theoretical studies, Corros. Sci. 52 (2010) 140
151.
[16] M.A. Amin, S.S. Abd El Rehim, H.T.M. Abdel-Fatah, Electrochemical frequency
modulation and inductively coupled plasma atomic emission spectroscopy
methods for monitoring corrosion rates and inhibition of low alloy steel
corrosion in HCl solutions and a test for validity of the Tafel extrapolation
method, Corros. Sci. 51 (2009) 882894.
[17] W.A. Badawy, K.M. Ismail, A.M. Fathi, Corrosion control of CuNi alloys in
neutral chloride solutions by amino acids, Electrochim. Acta 51 (2006) 4182
4189.
[18] H. Ashassi-Sorkhabi, M.R. Majidi, K. Seyyedi, Investigation of inhibition effect
of some amino acids against steel corrosion in HCl solution, Appl. Surf. Sci. 225
(2004) 176185.
[19] ASTM G 31, Standard Practice for Laboratory Immersion Corrosion Testing of
Metals, Licensed for BASF, p. 5.
[20] Electrochemistry and Corrosion-Overview and Techniques, Application Note
CORR-4, EG and G, Princeton Applied Research, USA.
 M.A. Amin / Corrosion Science 52 (2010) 32433257
[21] J.O.M. Bockris, L.J.V. Minevski, On the mechanism of the passivity of aluminum
and aluminum alloys, J. Electroanal. Chem. 349 (1993) 375414.
[22] L. Garrigues, N. Pebere, F. Dabosi, An investigation of the corrosion inhibition
of pure aluminum in neutral and acidic chloride solutions, Electrochim. Acta
41 (1996) 12091215.
[23] K.M. Ismail, Evaluation of cysteine as environmentally friendly corrosion
inhibitor for copper in neutral and acidic chloride solutions, Electrochim. Acta
52 (2007) 78117819.
[24] B.G. Cubley (Ed.), Chemical Inhibitors for Corrosion Control, The Royal Society
of Chemistry, 1990.
[25] M. Bethencourt, F.J. Botana, M.J. Cano, M. Marcos, J.M. Sanchez-Amaya, L.
Gonzalez-Rovira, Behaviour of the alloy AA2017 in aqueous solutions of NaCl.
Part I: corrosion mechanisms, Corros. Sci. 51 (2009) 518524.
[26] S.-M. Moon, S.-I. Pyun, The formation and dissolution of anodic oxide lms on
pure aluminium in alkaline solution, Electrochim. Acta 44 (1999) 24452454.
[27] J.L. Burmeister, Chemistry and Biochemistry of HCN and its Derivatives,
Academic Press, London, 1975.
[28] A.H. Norbury, Coordination chemistry of the cyanate, thiocyanate, and
selenocynate ions, Adv. Inorg. Radiochem. 17 (1975) 231386.
[29] A. Bahta, G.A. Parker, D.G. Tuck, Critical survey of stability constants of
complexes of thiocyanate ion, Pure Appl. Chem. 69 (1997) 14891548.
[30] S. Sasic, D. Zoric, M. Jeremic, A. Antic-Jovanovic, Study of complex formation in
Al(III)thiocyanatewater system via Raman spectra and factor analysis,
Polyhedron 20 (2001) 839847.
[31] C.M.A. Brett, On the electrochemical behaviour of aluminium in acidic chloride
solution, Corros. Sci. 33 (1992) 203210.
[32] B.A. Boukamp, Equivalent Circuit, Princeton Applied Research Corporation,
Princeton, NJ, 1990.
[33] C. Bataillon, S. Brunet, Electrochemical impedance spectroscopy on oxide lms
formed on zircaloy 4 in high temperature water, Electrochim. Acta 39 (1994)
455465.
[34] J.R. Galvele, Anodic behavior of aluminum straining and a mechanism for
pitting, J. Electrochem. Soc. 121 (1976) 12711276.
[35] N.D. Budiansky, J.L. Hudson, J.R. Scully, Origins of persistent interaction among
localized corrosion sites on stainless steel, J. Electrochem. Soc. 151 (2004)
B233B243.
3257
[36] Metikos-Hukovic, R. Babic, Z. Grubac, S. Brinc, Inhibition of the hydrogen
evolution reaction on aluminium covered by spontaneous oxide, J. Appl.
Electrochem. 22 (1992) 325331.
[37] M.A. Amin, Metastable and stable pitting events on Al induced by chlorate and
perchlorate anionspolarization, XPS and SEM studies, Electrochim. Acta 54
(2009) 18571863.
[38] T.H. Nguyen, R.T. Foley, The chemical nature of aluminum corrosion, J.
Electrochem. Soc. 127 (1979) 25632566.
[39] J. Augustynski, R.P. Frankenthal, J. Kruger (Eds.), Proceedings of the Fourth
International Symposium on Passivity, The Electrochemical Society,
Pennington, NJ, 1978. p. 997.
[40] S. Szklarska-Smialowska, Pitting Corrosion of Metals, D. 296, NACE, Houston,
TX, 1986.
[41] H. van Olphen, An Introduction to Clay Colloid Chemistry, second ed.,
Interscience, Publications, New York, 1977.
[42] F.I. Morais, A.L. Page, L.J. Lund, The effect of pH, salt concentration, and nature
of electrolytes on the charge characteristics of Brazilian tropical soils, Soil Sci.
Soc. Am. J. 40 (1976) 521527.
[43] G.A. Parks, P.L. de Bruyn, The zero point of charge of oxides, J. Phys. Chem. 66
(1961) 967973.
[44] H. Hohl, M. Stumm, Interaction of Pb2+ with hydrous c-Al2O3, J. Colloid Int. Sci.
55 (1976) 281288.
[45] R. Wood, D. Fornasiero, J. Ralston, Electrochemistry of the boehmitewater
interface, Colloids Surf. 51 (1990) 389403.
[46] L.I. Antropov, E.M. Makushin, V.F. Panasenko, Metal Corrosion Inhibitors, Kiev,
Technika, 1981.
[47] M.A. Amin, Weight loss, polarization, electrochemical impedance
spectroscopy, SEM and EDX studies of the corrosion inhibition of copper in
aerated NaCl solutions, J. Appl. Electrochem. 36 (2006) 215226.
[48] H.H. Hassan, E. Abdelghani, M.A. Amin, Inhibition of mild steel corrosion in
hydrochloric acid solution by triazole derivatives: Part I. Polarization and EIS
studies, Electrochim. Acta 52 (2007) 63596366.
[49] M.A. Amin, S.S. Abd El-Rehim, E.F.E. El-Sherbini, R.S. Bayoumi, The inhibition of
low carbon steel corrosion in hydrochloric acid solutions by succinic acid: Part
I. Weight loss, polarization, EIS, PZC, EDX and SEM studies, Electrochim. Acta
52 (2007) 35883600.