Corrosion Science 53 (2011) 3780–3787

Contents lists available at ScienceDirect

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

Co2 corrosion inhibition of X-120 pipeline steel by a modified imidazoline

under flow conditions
D.M. Ortega-Toledo a, J.G. Gonzalez-Rodriguez b,⇑, M. Casales c, L. Martinez c, A. Martinez-Villafañe a
a
Centro de Inv. en Materiales Avanzados, Miguel de Cervantes 120, Chih., Mexico
b
U.A.E.M.–C.I.I.C.Ap, Av. Universidad 1001, 6225 Cuernavaca, Mor., Mexico
c
Universidad Nacional Autonoma de Mexico, Instituto de Ciencias Fisicas, Av. Universidad S/N, 6225 Cuernavaca, Mor., Mexico

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

Article history: A study of the effect of flow conditions on the performance of a carboxyethyl-imidazoline as CO2-
Received 17 March 2011 corrosion inhibitor for API X-120 pipeline steel has been evaluated. Testing techniques include polariza-
Accepted 16 July 2011 tion curves, linear polarization resistance, electrochemical impedance spectroscopy and electrochemical
Available online 22 July 2011
noise measurements under stirred conditions (0, 250, 500, 1000 and 2500 rpm). All these techniques
show that for the uninhibited solution, corrosion rate increases with an increase in the rotation speed,
Keywords: but for the inhibited solution, the lowest corrosion rate is obtained at 500 rpm, and it increases at lower
A. Carbon steel
or higher rotation speeds.
B. Electrochemical noise
C. CO2 corrosion
Ó 2011 Elsevier Ltd. All rights reserved.
C. Flowing conditions

1. Introduction loss tests and electrochemical impedance spectroscopy (EIS)

In oil and gas industries, carbon dioxide (CO2) corrosion often
causes severe problems for carbon steel pipelines [1,2]. It is very
well known that CO2 corrosion would cause the failure of pipelines
and structural components in petroleum production, resulting in
significant economic losses and catastrophic accidents as well as
water resource and environment pollution [3–5]. CO2 corrosion is
complex, with a number of factors affecting the steel corrosion
process [5–10]. Particularly, the fluid hydrodynamics plays a sig-
nificant role through a complex interrelationship of near-wall
hydrodynamic momentum and mass transfer as well as fluid
parameters defining general flow properties with the corrosion
reactions of steel [11–15]. So far, corrosion inhibitor injection,
especially nitrogen-based organic surfactant, such as amine,
imidazoline and their salts is still the most cost-effective method
to solve this problem. However, most inhibitor evaluations are
generally only based on test results under stagnant conditions.
There exist a limited number of studies on the effect of flow veloc-
ity on inhibitors performance. For instance, Jiang et al. [16] carried
out a study of the inhibition of CO2 corrosion of N80 steel by
quaternary alkynoxymethyl-amine and imidazoline in a 3% NaCl
solution under static and flowing conditions by using a modified
rotating disk apparatus. The effect of flow velocity and inhibitor
concentration was studied by using polarization curves, weight
⇑ Corresponding author. Tel.: +52 045 777 3297084.
E-mail address: ggonzalez@uaem.mx (J.G. Gonzalez-Rodriguez).
measurements. They have found an optimum inhibitor concentra-
tion under static conditions, finding a critical effect of the flow
velocity depending upon the inhibitor concentration. Usually,
inhibitor efficiency increases with an increase in the inhibitor
concentration until a steady state is reached. However, some
inhibitors may exhibit the peak-value-phenomenon of inhibitor
concentration [16,17]. The concentration, where the lowest corro-
sion rate is exhibited, is called optimum concentration. Currently,
research works on the optimum concentration of inhibitor exist
in literature [16,17] mainly for static conditions but very few
works on this matter exist for flowing conditions. Bommersbach
et al. [17] studied the inhibitor efficiency of a carboxylic acid and
a tertiary amine on the corrosion of SAE 1038 carbon steel in a mix-
ture of Na2SO4 + Na2CO3 + NaCl at different rotation speeds (0,
1600 and 3200 rpm) by using EIS measurements. They found that
inhibitor film formation depends on the electrode rotation rate:
film forms faster under high rotation speeds. Ochoa et al. [18,19]
studied the inhibition mechanism of a carbon steel by a non-toxic
multi-component inhibitor by using EIS measurements, and found
that the properties of the protective formed layers on the metal
surface depend on the electrode rotation rate.
Among the different electrochemical techniques that can be
used to evaluate inhibitors, electrochemical impedance spectros-
copy (EIS) is a powerful tool in addition to traditional techniques
such as polarization curves or linear polarization resistance (LPR)
measurements. However, electrochemical noise (EN) measure-
ments have also been successfully applied to the study of corrosion
inhibitors performance [20–23]. These measurements are made

0010-938X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.corsci.2011.07.028
 D.M. Ortega-Toledo et al. / Corrosion Science 53 (2011) 3780–3787
without any external perturbation of the systems, and provide
information about the actual system being studied with little pos-
sible artifacts. EN technique involves the estimation of the electro-
chemical noise resistance, Rn, which is calculated as the standard
deviation of potential, rv, divided by the standard deviation of cur-
rent, ri,
Rn ¼ rv =ri
where Rn can be taken as the linear polarization resistance, Rp in the
Stern–Geary equation:
ba bc 1
icorr ¼
2:3ðba þ bc Þ RP
thus, inversely proportional to the corrosion rate, icorr, but with the
necessary condition that a trend removal applied over an average
baseline, as previously established [24]. It has been demonstrated
that the noise signal provides information about the dynamics that
occur on the surface of the electrode and about the type of corrosion
that is occurring, either uniform or localized. In past years there has
been a trend to increase the strength levels of used pipeline steels
such as API Grades X-60, X-80, X-100, etc., but it brings a higher
susceptibility of failure. Thus, the aim of this work is to study the
performance of a simple organic compound, i.e. carboxyethyl-
imidazoline, as a CO2 corrosion inhibitor of API X-120 pipeline steel
by using different electrochemical techniques such as polarization
curves, LPR, EIS and EN measurements under hydrodynamic
conditions.
2. Experimental procedure
2.1. Test material
Material tested was an API X-120 pipeline steel with a chemical
composition as given in Table 1. Cylindrical specimens 10 mm long
with a diameter of 6 mm were machined and embedded in PTFE
exposing an effective surface area of 0.28 cm2 to the electrolyte.
This system was machined to form the rotating disk electrode,
which was ground up to 2400 grade emery paper, rinsed with
distilled water, acetone, and dried under an air flow.
2.2. Test solution
Inhibitor used in this work was a commercial carboxyethyl-
imidazoline, from Lakeland Labs. (UK), with a molecular structure
as shown in Fig. 1, which is composed of a five member ring
containing nitrogen elements, a C-14 saturated hydrophobic head
group and a pendant hydrophilic carboxy group attached to one
of the nitrogen atoms. This inhibitor was used in this work,
without the addition of another component, and it was previously
dissolved in pure 2-propanol. The concentration of the inhibitor
was the recommended by the manufacturer, 8.1  105 mol/l at a
testing temperature of 50 °C, which was kept constant with a hot
plate. Test solution consists of 3% NaCl solution, heated, de-aerated
by purging with CO2 gas during 2 h prior the experiment and kept
bubbling throughout the experiment. Inhibitor was added 2 h after
pre-corroding the specimens in the CO2-containing solution,
Table 1
Chemical composition of used X-120 steel (wt%).
C Si Mn P S
0.027 0.24 1.00 0.003
Cu Nb Ti Fe
0.010 0.024 0.015 96.6
3781
ð1Þ
ð2Þ
Fig. 1. Molecular structure of carboxyethyl-imidazoline.
starting the measurements 1 h later. Rotation speeds were 0, 250,
500, 1000 and 2500 rpm.
2.3. Electrochemical measurements
Employed electrochemical techniques includes potentiodynam-
ic polarization curves, linear polarization resistance, LPR, electro-
chemical impedance spectroscopy, EIS, and electrochemical noise,
EN, measurements, both in current and voltage. Experiments were
repeated three times. Polarization curves were recorded at a con-
stant sweep rate of 1 mV/s from 300 to +300 mV interval with
respect to the open circuit potential (Ecorr), i.e. it started at 900
and ended at 300 mVSCE. Measurements were obtained by using
a conventional three electrode glass cell with two graphite rods
as auxiliary electrodes, and a saturated calomel electrode (SCE)
reference electrode with a Luggin capillary bridge. Since the corro-
sion process was under cathodic diffusion control, corrosion cur-
rent density values, icorr, were calculated by extrapolating the
limit cathodic diffusion current density towards the Ecorr value.
Inhibitor efficiency, g, is calculated as follows:
 0 
icorr  icorr
gð%Þ ¼  100 ð3Þ
icorr
0
where icorr and icorr are the corrosion current values with and with-
out inhibitor, respectively. LPR measurements were carried out by
polarizing the specimen ±10 mVSCE with respect to Ecorr at a scan-
ning rate of 1 mV/s every 20 min, during 24 h. EIS tests were carried
out at the Ecorr value by using a signal with amplitude of 10 mVSCE
and a frequency interval of 0.1–30 kHz. An ACM potentiostat
controlled by a desk top computer was used for the LPR tests and
polarization curves, whereas for the EIS measurements a PC4 300
Gamry potentiostat was used. EN measurements in both current
and potential were recorded using a working, cylindrical electrode
with an exposed area of 0.28 cm2 as described above, and a
platinum rod as second working electrodes, 0.25 mm diameter; this
platinum rod was embedded in epoxy resin, with an effective
exposed area of 0.049 mm2, and a reference SCE electrode, in such
a way that the current flows towards the platinum rod. EN
measurements were made recording simultaneously the potential
and current fluctuations at a sampling rate of 1 point per second
in blocks of 1024 readings. The readings were taken every hour
Cr Mo Ni Al Co
0.004 0.42 0.18 1.35 0.045 0.004
3782 D.M. Ortega-Toledo et al. / Corrosion Science 53 (2011) 3780–3787

during 24 h. The 24 blocks of readings were within the same order

-5
of magnitude, thus all the data had 100% of reproducibility, and in 8.1 X 10 mol/l
this work only the most representative are shown. Current and -400
potential standard deviation values, ri and rv, increased with
rotation speed; for instance, at 0 and 2500 rpm, ri was 1  106
and 5  104 mA/cm2 respectively, whereas rv had values of 0.5
and 2 mV. A fully automated zero resistance ammeter (ZRA) from
-600
ACM instruments was used in this case. Finally, the noise resistance,

E (mV SCE )
1000 rpm
Rn, was then calculated as the ratio of the potential noise standard
deviation, rv, over the current noise standard deviation, ri, 2500 rpm
according to Eq. (1).
-800
500 rpm
3. Results and discussion
250 rpm
0 rpm
Polarization curves for X-120 steel in uninhibited CO2-saturated
3% NaCl solution in stagnant and stirred solutions are shown in
-1000
Fig. 2. This figure shows an active behavior in the anodic branch 10
-4
10
-3
10
-2
10
-1
10
0
10
1

of the curve, with the current increasing with the applied potential
without the presence of a passive layer. The anodic current density
value remains more or less constant regardless of the rotation
speed. On the other hand, the cathodic current density exhibits a
limit value, which increases with the rotation speed, which is
due to the hydratation of CO2 to give carbonic acid as follows [9]:
Current density (mA/cm 2 )
Fig. 3. Effect of rotation speed in the polarization curve for X-120 steel in
CO2-saturated 3% NaCl solution containing 8.1  105 mol/l of carboxyethyl-
imidazoline.

CO2 þ H2 O ! H2 CO3
Since the solution is de-aerated, the dominant cathodic reaction
is the reduction of H+ ions which are supplied by dissociation of
carbonic acid [25]:
H2 CO3 ! Hþ þ HCO3
HCO3 ! Hþ þ CO2
3
Hþ þ e ! 1=2H2
Thus, this cathodic limit current value is due to the diffusion of
H+ ions. The free corrosion potential value, Ecorr, becomes nobler
and the corrosion current density value, icorr, increases as the rota-
tion speed increases. On the other hand, for the inhibited solution,
Fig. 3, the Ecorr value becomes nobler and the icorr value increases
with an increase in the rotation speed, reaching a more or less
constant value after 500 rpm. An abrupt increase in the current
density can be observed on the anodic branch in presence of the
ð4Þ inhibitor when the polarization potential attains a relatively posi-
tive value, except at 1000 and 2500 rpm, which indicates an anodic
desorption process of the inhibitor [7]. Fig. 4 shows a plot of log
(icorr) versus log (x) where x is the angular velocity, where it
can be seen that the data for uninhibited solution follow the Levich
ð5Þ equation, indicating that the corrosion process is under diffusion
control; on the other hand, data for the inhibited solution do not
ð6Þ follow this equation, which indicates that in this case the process
is only partially controlled by diffusion of species.
The surface coverage of adsorbed inhibitive species, H usually
ð7Þ
decreases with the increase of potential in the anodic branch.
The increase of polarization potential accelerates metal dissolution
not only on the surface without adsorbed species but also with
adsorbed inhibitive species because they become unstable at a
high polarization potential. This desorption process seems to end
at a potential value close to 530 mVSCE at 0, 250 and 500 rpm,
but it does not exist at 1000 or 2500 rpm, which indicates that at
these rotation speeds any inhibitor film is very unstable. In all
cases, however, an anodic limit current density is observed, which

-300
0 mol/l 0.0

-0.2

-0.4
-600
log (i corr )
E (mV SCE )

-0.6

0 mol/l
2500 rpm -0.8
500 rpm
1000 rpm
-900 250 rpm
-1.0 -5
0 rpm 8.1 X 10 mol/l

-1.2

-3 -2 -1 0 1
0.0 0.3 0.6 0.9 1.2 1.5
10 10 10 10 10 log (ω)
2
Current density (mA/cm )
Fig. 4. Variation of log (icorr) as a function of log (x) for X-120 steel in
Fig. 2. Effect of rotation speed in the polarization curve for X-120 steel in CO2-saturated 3% NaCl solution containing 0 or 8.1  105 mol/l of carboxyethyl-
uninhibited CO2-saturated 3% NaCl solution. imidazoline.
 D.M. Ortega-Toledo et al. / Corrosion Science 53 (2011) 3780–3787 3783

Table 2
Electrochemical parameters obtained from the polarization curves (the standard deviation values for Ecorr and icorr are 11 mV and 5  103 mA/cm2 respectively).

Carboxyethyl-imidazoline concentration (mol/l) 0 rpm 250 rpm 500 rpm 1000 rpm 2500 rpm
0 Ecorr (mVSCE) 700 668 686 670 647
icorr (mA/cm2) 0.15 0.30 0.35 0.40 0.60
5
8.1  10 Ecorr (mVSCE) 640 627 -610 590 610
icorr (mA/cm2) 0.07 0.08 0.1 0.12 0.11
g (%) 53.3 73.3 71.4 70 81.6
H 0.53 0.73 0.71 0.7 0.81

could be due to the diffusion of the electrolyte through a film
formed inhibitor. On the cathodic branch, a cathodic limit current
density can be seen, which is due to the hydratation of CO2 as
explained above. Table 2 summarizes these results. This table
shows more or less an increase in both the inhibitor efficiency
and its degree of surface coverage as the rotation speeds increases.
To see the effect of inhibitor on the surface metal as time
elapses at the different rotation speeds, Figs. 5 and 6 show the
change in the polarization resistance value with time, Rp, without
80
0 mol/l
70 0 rpm
and with inhibitor respectively. For the uninhibited solution,
Fig. 5 shows that the highest Rp value, thus the lowest corrosion
rate is obtained under stagnant conditions and this value decreases
as the rotation speed increases, which is due to an increase in the
mass transport. During most of the test, the lowest Rp value is
obtained at 2500 rpm. The Rp values decrease as time elapses,
reaching a more or less steady state value towards the end of the
test. On the other hand, for the inhibited solution, Fig. 6, the Rp
value decreases at 250 rpm, but it reaches its highest value at
500 rpm for more than one order of magnitude; with a further
increase in the rotation speed, a decrease in the Rp value is
observed. Thus, for the inhibited solution, the lowest corrosion rate
is obtained at 500 rpm; a further increase in the rotation speed
brings an increase in the corrosion rate.

(a) -110

-100
R p (Ohm cm )

60 250 rpm
2

1000 rpm -90

-80
50 -70 0 mol/l
Z im (Ohm cm 2 )

-60
2500 rpm
500 rpm -50
40
-40
250 rpm
-30 0 rpm
30
0 5 10 15 20 25 -20
Time (h) -10 1000 rpm 500 rpm

Fig. 5. Effect of rotation speed in the change on the Rp value with time for X-120 0 2500 rpm
steel in uninhibited CO2-saturated 3% NaCl solution (the standard deviation value
0 10 20 30 40 50 60 70 80 90 100 110
for Rp is 10 X cm2).
Z re (Ohm cm 2 )

5000
-5
8.1 X 10 mol/l (b) -60
-50
0 rpm
4000 0 mol/l
500 rpm 250 rpm
-40
Rp (Ohm cm )
2

3000 500 rpm

Θ (Degrees)

-30

1000 rpm
2000 -20
2500 rpm
2500 rpm 0 rpm
1000 -10

1000 rpm 250 rpm

0
0
0 5 10 15 20 25
0.1 1 10 100 1000 10000 100000
Time (h)
Frequency (Hz)
Fig. 6. Effect of rotation speed in the change on the Rp value with time for X-120
steel in CO2-saturated 3% NaCl solution containing 8.1  105 mol/l of carboxy- Fig. 7. Effect of rotation speed in the (a) Nyquist and (b) Bode diagrams for X-120
ethyl-imidazoline (the standard deviation value for Rp is 10 X cm2). steel in uninhibited CO2-saturated 3% NaCl solution.
3784 D.M. Ortega-Toledo et al. / Corrosion Science 53 (2011) 3780–3787

(a) -300
-4000 0 rpm
-250

-200

Zim (Ohm cm 2)
-5 -150
8.1 X 10 mol/l
-3000 -100
Z im (Ohm cm 2 )

-50

0
0 50 100 150 200 250 300
-2000 Zre (Ohm cm2)

500 rpm

-1000
Fig. 9. Equivalent circuits used to model EIS data (a) uninhibited solutions, and (b)
1000 rpm 2500 rpm inhibited solutions.
250 rpm
0
0 1000 2000 3000 4000
The resulting complex, depending on its relative solubility, can
2
Z re (Ohm cm ) either inhibit or catalyze further metal dissolution. At low concen-
trations the amount of carboxyethyl-imidazoline is not enough to
(b) -80 form a compact complex with the metal ions, so that the resulting
-5 adsorbed intermediate will be readily soluble in the acidic environ-
-70 8.1 X 10 mol/l
ment. But at relatively higher inhibitor concentrations, more car-
-60 boxyethyl-imidazoline molecules become available for the
complex formation, which subsequently diminishes the solubility
-50 of the surface layer, leading to improve the inhibition of metal cor-
500 rpm rosion, because under flow conditions there are different effects on
Θ (Degrees)

-40 the inhibition performance:

0 rpm 1000 rpm
-30 250 rpm (i) Flow can increase mass transport of inhibitor molecules that
2500 rpm
causes more inhibitor presence at metal surface. This effect
-20
can improve the inhibitor performance [27].
(ii) Hydrodynamic conditions can increase mass transport of
-10
metal ions (Fe2+) produced during metal dissolution from
0 the electrode surface to the bulk solution and hence lead
to less [Fe–Inh]2+ complex presence on the electrode; this
0.1 1 10 100 1000 10000 100000 is a harmful effect for inhibition performance.
Frequency (Hz) (iii) The high shear stress resulted from high flow velocity can
also separate the layer of adsorbed [Fe–Inh]2+ complex and
Fig. 8. Effect of rotation speed in the (a) Nyquist and (b) Bode diagrams for X-120 cause more desorption from the metal surface, which acts
steel in CO2-saturated 3% NaCl solution containing 8.1  105 mol/l of carboxy-
as a negative factor on the inhibitor efficiency [27]. The bal-
ethyl-imidazoline.
ance of the above mentioned effects lead to changes on
inhibitor efficiency with rotation rate.
It is generally accepted that the first step during the adsorption
of an organic inhibitor on a metal surface usually involves replace- Nyquist and Bode plots for the uninhibited solution at the dif-
ment of water molecules absorbed on the metal surface: ferent rotation speeds are shown on Fig. 7. Nyquist plot shows a
capacitive semicircle at high and intermediate frequency values,
Inhsol þ xH2 Oads ! Inhads þ xH2 Osol ð8Þ Fig. 7a, which arise from the charge transfer from the metal to elec-
2+ trolyte through the double electrochemical layer. The diameter of
The inhibitor may then combine with freshly generated Fe ions on
steel surface, forming metal inhibitor complexes [16,26]: this semicircle, the charge transfer resistance, Rct, is equivalent to
the polarization resistance, Rp, which decreases as the rotation
Fe ! Fe2þ þ 2e ð9Þ speed increases, increasing, thus, the corrosion rate, as shown by
the polarization curves; the semicircle diameters were more or less
Fe2þ þ Inhads ! ½Fe  Inh2þ
ads ð10Þ the same at 500, 1000 and 2500 rpm. At lower frequency values, a

Table 3
Calculated parameters used to simulate the EIS data for X-120 steel in uninhibited CO2-saturated 3% NaCl solution.

Element Rotation speed (rpm)

0 250 500 1000 2500
Rs (X cm2) 5.6 5.7 5.2 6.5 4.8
Yct (X1 cm2 sn) 8  104 2  103 1  103 1  103 1  103
ndl 0.80 0.85 0.79 0.78 0.79
Rct (X cm2) 88 63 56 50 43
RL (X cm2) 5.04  107 5.4  107 5.1  107 5.8  107 4.3  107
L (H cm2) 6.8 6.1 5.5 5.2 5.1
 D.M. Ortega-Toledo et al. / Corrosion Science 53 (2011) 3780–3787 3785

Table 4
small inductive loop can be observed at all rotation speeds, which
Calculated parameters used to simulate the EIS data for X-120 steel in inhibited CO2-
is due to relaxation process obtained by the adsorption of species saturated 3% NaCl solution.
such as [H+]ads on the electrode surface. This loop is present at all
Element Rotation speed (rpm)
rotation speeds, indicating that the adsorbed species remain on
the steel surface. On the other hand, Bode plots show only one peak 0 250 500 1000 2500
around 30 Hz, Fig. 7b, and remains constant with an increase in the Rs (X cm2) 7.1 6.7 6.6 6.1 6.7
rotation speed. No evidence of any protective layer was found due Yct (X1cm2 sn) 8  104 3  103 6  105 1  103 8  105
to the presence of only one peak. ndl 0.7 0.68 0.78 0.76 0.73
Rct (X cm2) 284 429 4739 204 1837
For the inhibited solution, Nyquist diagrams show a capacitive
nf 0.9 0.87 0.85 0.78 0.8
semicircle at all rotation speed values, Fig. 8a, which is due to Yf (X1 cm2 sn) 5  105 2  102 5  102 2  102 5  102
the charge transfer from metal to electrolyte through the double Rf (X cm2) 5 48 67 57 44
electrochemical layer. The semicircle diameter increases with an

0 .0 0 0 9 5 -6 9 9 .0
0 m o l/l
0 .0 0 0 9 0 0 rp m
Current density (mA/cm 2 )

-6 9 9 .5
0 .0 0 0 8 5 -7 0 0 .0

Potential (mV)
P o te n tia l
0 .0 0 0 8 0 -7 0 0 .5
0 .0 0 0 7 5 -7 0 1 .0
0 .0 0 0 7 0 -7 0 1 .5
0 .0 0 0 6 5 -7 0 2 .0
0 .0 0 0 6 0 -7 0 2 .5
0 .0 0 0 5 5 C u r r e n t d e n s ity -7 0 3 .0
0 200 400 600 800 1000
T im e ( s )

0 .0 0 1 0 5
0 m o l/l -6 8 3
Current density (mA/cm 2 )

0 .0 0 1 0 0 2 5 0 0 rp m
C u r r e n t d e n s ity
0 .0 0 0 9 5 -6 8 4

Potential (mV)
P o te n tia l
0 .0 0 0 9 0
-6 8 5
0 .0 0 0 8 5
0 .0 0 0 8 0
-6 8 6
0 .0 0 0 7 5
0 .0 0 0 7 0 -6 8 7
0 .0 0 0 6 5
-6 8 8
0 200 400 600 800 1000
T im e ( s )

0 .0 0 0 2 4 -6 6 1
-5
8 .1 X 1 0 m o l/l
0 rp m -6 6 2
0 .0 0 0 2 2
Current density (mA/cm 2 )

-6 6 3
0 .0 0 0 2 0 P o te n tia l
Potential (mV)

-6 6 4
0 .0 0 0 1 8 -6 6 5
-6 6 6
0 .0 0 0 1 6
-6 6 7
0 .0 0 0 1 4 C u r r e n t d e n s ity
-6 6 8
0 .0 0 0 1 2 -6 6 9

0 .0 0 0 1 0 -6 7 0
0 200 400 600 800 1000
T im e ( s )

0 .0 0 0 7 0 -5
-5 9 4
8 .1 X 1 0 m o l/l
0 .0 0 0 6 5 2 5 0 0 rp m
-5 9 6
Current density (mA/cm 2 )

0 .0 0 0 6 0 P o te n tia l
-5 9 8
Potential (mV)

0 .0 0 0 5 5
0 .0 0 0 5 0 -6 0 0
0 .0 0 0 4 5
0 .0 0 0 4 0 -6 0 2

0 .0 0 0 3 5 -6 0 4
0 .0 0 0 3 0 C u r r e n t d e n s ity

0 .0 0 0 2 5 -6 0 6
0 200 400 600 800 1000
T im e ( s )

Fig. 10. Noise in current and potential for X-120 steel in CO2-saturated 3% NaCl solution containing (a) and (b) 0 and (c) and (d) 8.1  105 mol/l of carboxyethyl-imidazoline
at (a) and (c) 0, and (b) and (d) 2500 rpm.
3786 D.M. Ortega-Toledo et al. / Corrosion Science 53 (2011) 3780–3787

increase in the rotation speed, obtaining the highest semicircle

240 0 rpm 0 ppm
diameter value at a rotation speed of 500 rpm, similar to the results
obtained by the LPR measurements; a further increase in the rota-
tion speed brings a decrease in the semicircle diameter, and thus, 200
an increase in the corrosion rate. Bode diagrams, Fig. 8b, shows 1000 rpm 500 rpm
one peak at 0 and 2500 rpm but two peaks at 250, 500 and 160

Rn (Ohm cm )
2
1000 rpm, indicating the presence of a film formed by the inhibitor
at these rotation speeds. 2500 rpm
120
Impedance spectra obtained can be modeled as electric equiva-
lent circuits given in Fig. 9. These are simple circuits for the
corrosion system under charge transfer control, where no mass 80
250 rpm
transport was detected. In these figures, Rs is the electrolyte resis-
tance, Rct the charge transfer resistance, Cdl the double layer capac- 40
itance, L the inductive element, RL its resistance, Rf the resistance of
any formed film on the surface and Cf its capacitance [14]. When a 0
non-ideal frequency response is present, it is commonly accepted 0 5 10 15 20 25
to employ distributed circuit elements in an equivalent circuit. Time (h)
The most widely used is constant phase element (CPE), which
has a non-integer power dependence on the frequency. The imped- Fig. 11. Effect of rotation speed in the change on Rn with time for X-120 steel in
ance of a CPE is described by the expression: uninhibited CO2-saturated 3% NaCl solution.

Z CPE ¼ Y 1 ðiwÞn ð11Þ

p
where Y is a proportional factor, i is 1, x is 2pf, f the frequency
500 rpm
and n has the meaning of a phase shift [14]. Often a CPE is used in a
model in place of a capacitor to compensate for non-homogeneity in
3
the system to take into account irregularities of the surface such as 10 250 rpm
roughness or because properties such as double layer capacitance,
R n (Ohm cm )

charge transfer rate, etc., are non uniformly distributed. For an ideal 1000 rpm
2

capacitor Y is 1/C and n = 1, whereas for a non ideal capacitor n < 1.

The fitting parameters for uninhibited and inhibited solutions are 0 rpm
given on Tables 3 and 4 respectively. It can be noticed that the Rct 10
2
2500 rpm
value for the uninhibited solution is much smaller than that for
inhibited solution and the highest value is attained for unstirred
solution; the Rct value decreases as the rotation speed increases, just
like the Rp values did, Fig. 5. On the other hand, for inhibited solu-
tion, the highest Rct value is obtained at a rotation speed of 1
10
500 rpm; lower or higher rotation speeds its value decreases. A sim-
ilar behavior is observed for the Rp values, Fig. 6.
0 5 10 15 20 25
To study the susceptibility of this steel towards a localized type
of corrosion, some electrochemical noise tests in both potential Time (h)
and current were performed, and Fig. 10 shows typical time series
Fig. 12. Effect of rotation speed in the change on Rn with time for X-120 steel in
at different rotation speeds for uninhibited and inhibited solutions. CO2-saturated 3% NaCl solution containing 8.1  105 mol/l of carboxyethyl-
Time series at 0 rpm consists of transients with low intensity and imidazoline.
high frequency, combined with some transients of higher intensity
but lower frequency, indicating a mixture of uniform and localized
type of corrosion. As the rotation speed increases, both the inten-
Figs. 11 and 12 for uninhibited and inhibited solutions respec-
sity and frequency of these transients increase, indicating that
tively. The behavior for uninhibited solution is very similar to that
the steel surface is now susceptible towards a uniform type of cor-
observed by Rp in Fig. 5, with the highest Rn value obtained under
rosion. When the inhibitor is added, Fig. 10c, the time series under
stagnant conditions and decreases as the rotation speed increases.
stagnant conditions consists of transients of both much lower
Thus, for uninhibited solution, noise data indicate that the lowest
intensity and frequency than that found for uninhibited solution,
corrosion rate is obtained in static conditions. For inhibited solu-
but both the intensity and frequency increases with an increase
tion, Fig. 12 shows some fluctuations for the Rn values; the highest
in the rotation speed. These transients are due to the breakdown
Rn value is obtained at a rotation speed of 500 rpm, with a decrease
of a film formed on the steel surface and they are typical of a local-
in its value with a further increase in the rotation speed up to
ized type of corrosion such as pitting [28]. This is because although
2500 rpm, similar to the behavior exhibited by Rp in Fig. 6.
the inhibitor in transported towards the electrode surface as the
rotating speed increases, some places remain uncovered by the
inhibitor, and these places remain as active sites where a localized 4. Conclusions
type of corrosion can take place.
The combined effect of potential standard deviation, rv, and A study of the effect of rotation speed on the CO2 corrosion
current standard deviation, ri, results in the noise resistance value, inhibition of X-120 pipeline steel by carboxyethyl-imidazoline
Rn, (eq. (1)) which is inversely proportional to the corrosion rate, has been carried out by using electrochemical techniques. Different
icorr, as previously done with the Rp values, Fig. 5. An estimation techniques have shown that for the uninhibited solution, an in-
of the corrosion rate can be completed through the variation of crease in the rotation speed increases the corrosion rate, but for
Rn with time for the different rotation speeds, as can be seen in the inhibited solution, the lowest corrosion rate is obtained at a
 D.M. Ortega-Toledo et al. / Corrosion Science 53 (2011) 3780–3787 3787

rotation speed of 500 rpm, but it increases again with a further [11] G.T. Burstein, K. Sasaki, Corros. Sci. 42 (2000) 841–860.
[12] K.D. Efird, E.J. Wright, J.A. Boros, T.G. Hailey, Corrosion 49 (1993) 992–1004.
increase in the rotating speed. Additionally, EN measurements
[13] S. Zhou, M.M. Stack, R.C. Newman, Corros. Sci. 38 (1996) 1071–1084.
have shown that for uninhibited solution, the steel was susceptible [14] E.A.M. Hussain, M.J. Robinson, Corros. Sci. 49 (2007) 1737–1754.
to uniform type of corrosion, whereas for the inhibited solution, [15] K. Sasaki, G.T. Burstein, Corros. Sci. 49 (2007) 92–102.
the film formed by the inhibitor leaves some active sites, making [16] X. Jiang, Y.G. Zheng, W. Ke, Corros. Sci. 47 (2005) 2636–2658.
[17] P. Bommersbach, C. Alemany-Dumont, Jean-Pierre Millet, Bernard Normand,
the steel susceptible to localized corrosion. Electrochem. Acta 51 (2005) 1076–1084.
[18] N. Ochoa, F. Moran, N. Pébère, B. Tribollet, Corros. Sci. 47 (2005) 593–604.
References [19] N. Srisuwan, N. Ochoa, N. Pébère, B. Tribollet, Corros. Sci. 50 (2008)
1245–1250.
[20] A. Popova, S. Raicheva, E. Sokolova, M. Christov, Langmuir 12 (1996)
[1] D.A. Lopez, S.N. Simison, S.R. de Sanchez, Electrochim. Acta 48 (2003) 845–854. 2083–2089.
[2] M.B. Kermani, A. Morshed, Corrosion 59 (2003) 659–682.
[21] L. Nyikos, T. Pajkossy, Electrochim. Acta 30 (1985) 1533–1540.
[3] B.R. Linter, G.T. Burstein, Corros. Sci. 42 (1999) 117–139. [22] E. Garcia-Ochoa, J. Genesca, J. Surf. Coat. Technol. 184 (2004) 322–330.
[4] Z. Xia, K.C. Chou, Z.S. Smialowska, Corrosion 45 (1989) 636–643. [23] E. Samiento-Bustos, J.G. González Rodriguez, J. Uruchurtu, G. Dominguez-
[5] K.S. George, S. Nesic, Corrosion 63 (2007) 178–187. Patiño, V.M. Salinas-Bravo, Corros. Sci. 50 (2008) 2296–2303.
[6] K. Videm, J. Kvarekval, Corrosion 51 (1991) 260–266. [24] P.H. Tewart, A.B. Campbell, Can. J. Chem. 57 (1979) 188–197.
[7] T. Hong, Y.H. Sun, W.P. Jepson, Corros. Sci. 44 (2002) 101–112.
[25] C. de Waard, D.E. Williams, Corrosion 31 (1975) 131–140.
[8] G.I. Ogundele, W.E. White, Corrosion 42 (1986) 71–82. [26] E.E. Oguzie, Y. Li, F.H. Wang, J. Colloid Interface Sci. 310 (2007) 90–98.
[9] S. Nesic, J. Postlethwaite, S. Olsen, Corrosion 52 (1996) 280–295. [27] V. Branzoi, F. Branzoi, M.M. Baibarac, Mater. Chem. Phys. 65 (2000) 288–297.
[10] D.A. Lopez, W.H. Scheiner, S.R. de Sanchez, S.N. Simison, Appl. Surf. Sci. 207 [28] K. Hladky, J.L. Dawson, Corros. Sci. 22 (1982) 231–237.
(2003) 69–85.