Adsorption of Anions of Higher Carboxylic Acids on Magnesium from Weakly Alkaline Aqueous Solutions

Article abstract of DOI:10.1134/S0036024420060187

Abstract: The adsorption of sodium salts of higher carboxylates on oxidized magnesium is studied via in situ reflective ellipsometry. It is shown that the free energies of adsorption of sodium oleyl sarcosinate (OsS) and sodium linoleate (LiS) is >55 kJ/mol, indicating the chemisorption of carboxylates on oxidized magnesium surfaces. Electrochemical impedance spectra, voltammetry, and corrosion testing show that sodium oleate (OlS) has the best protective properties on pure and oxidized magnesium. The strong protective properties of OlS are confirmed by Mg plate testing under conditions of a wet atmosphere with daily condensation. Tentative passivation of chemically oxidized Mg in a 16 mmol/L OlS solution protects against corrosion for 92–96 h.

Full text of DOI:10.1134/S0036024420060187

ISSN 0036-0244, Russian Journal of Physical Chemistry A, 2020, Vol. 94, No. 6, pp. 1104–1110. © Pleiades Publishing, Ltd., 2020.
Russian Text © The Author(s), 2020, published in Zhurnal Fizicheskoi Khimii, 2020, Vol. 94, No. 6, pp. 829–836.
PHYSICAL CHEMISTRY
OF SOLUTIONS
Adsorption of Anions of Higher Carboxylic Acids on Magnesium
from Weakly Alkaline Aqueous Solutions
V. A. Ogorodnikovaa, Yu. I. Kuznetsova,*, N. P. Andreevaa, A. Yu. Luchkina, and A. A. Chirkunova
aFrumkin Institute of Physical Chemistry and Electrochemistry, Russian Academy of Sciences, Moscow, 119071 Russia
*e-mail: kuznetsov@ipc.rssi.ru
Received June 21, 2019; revised July 11, 2019; accepted September 17, 2019
Abstract—The adsorption of sodium salts of higher carboxylates on oxidized magnesium is studied via in situ
reflective ellipsometry. It is shown that the free energies of adsorption of sodium oleyl sarcosinate (OsS) and
sodium linoleate (LiS) is >55 kJ/mol, indicating the chemisorption of carboxylates on oxidized magnesium
surfaces. Electrochemical impedance spectra, voltammetry, and corrosion testing show that sodium oleate
(OlS) has the best protective properties on pure and oxidized magnesium. The strong protective properties of
OlS are confirmed by Mg plate testing under conditions of a wet atmosphere with daily condensation. Ten-
tative passivation of chemically oxidized Mg in a 16 mmol/L OlS solution protects against corrosion for 92–
96 h.
Keywords: adsorption, ellipsometry, corrosion, magnesium, corrosion inhibitors, electrochemical impedance
spectroscopy, carboxylates
DOI: 10.1134/S0036024420060187
INTRODUCTION
formed on Mg surfaces have a number of advantages:
environmental friendliness, inhibition of anodic metal
dissolution, reduced corrosion current, and the ability
to improve the adhesion and protective properties of
polymer coatings.
The use of corrosion inhibitors (CIs) is a promising
way of improving the corrosion resistance of Mg. For
example, it is known [2] that sodium decanoate in a
solution with neutral pH slows the rate of corrosion
and ensures the passivation of Mg.
Surfactants predominate among organic CIs. The
authors of [5] thus investigated the effect 151 organic
compounds had on the corrosion behavior of nine Mg
alloys. In this work, a number of new CIs were found
that provided effective protection comparable to and
even higher than that of chromate.
Controlling the corrosion of Mg alloys is worthy of
special attention. It is achieved by dissolving their elec-
tropositive metal impurities, which settle on a surface to
become cathodes. The ability of CIs to bind such cations
(i.e. iron ions) in complexes is thus of interest [5, 6].
Sodium salts of derivatives of pyridinedicarboxylic and
salicylic acids are the ones most effective and universal.
Magnesium and its alloys are in high demand in
industry, since they have unique combinations of
physical and mechanical properties. Magnesium is the
lightest metal, with a density of ∼1.7 g/cm3. This is
lower than that of Al (∼2.7 g/cm3), Ti (∼4.5 g/cm3),
and Fe (∼7.9 g/cm3) [1]. Magnesium is eighth among
the elements of the Earth’s crust, so its alloys are
widely available. Their useful properties are high
strength, rigidity, good damping capability, machin-
ability, excellent casting characteristics, and low heat
capacity. However, weak corrosion resistance hinders
the broad use of Mg alloys in practice, so anticorrosion
treatment of their surfaces is needed [2].
One of the most effective ways of protecting Mg
alloys against corrosion is to create barrier coatings on
their surfaces to mask the basic material from the envi-
ronment. This is achieved by depositing either a con-
version coating (CC), including those obtained via
anodization and microarc oxidation (plasma electro-
lytic oxidation) (MAO, PEO), or by applying paint or
lacquer. According to [3, 4], now there is a number of
ways of producing chromate-free CCs for Mg alloys
that have their own merits and demerits. An interest-
ing strategy for developing new and more ecological
CCs for Mg alloys is treating alloys with natural
organic acids that form insoluble salts with magne-
sium. The authors of [4] considered vanillic acid (4-
hydroxy-3-methoxybenzoic acid C8H8O4) on alloy
There is interest in passivating Mg alloys with
aqueous solutions of organic CIs to form oxide films in
air [7–9] or CCs obtained via PEO [10, 11]. Sodium
salts of higher carboxylic acids, i.e. oleic (OlA), stearic
(SA), tridecanoic (TDA), are used as CIs. Sodium
oleate (OlS), sodium oleyl sarcosinate (OsS), and
AZ31 as substitutes for toxic chromatizing. Such CCs IFKhAN-25 are of special interest [7–9].
1104

ADSORPTION OF ANIONS OF HIGHER CARBOXYLIC ACIDS
1105
Not only is using alkyl carboxylates as CIs an effec- 2. An electrode was oxidized for 1.5 h in a 5.0 М
tive way of reducing metal corrosion; it is relatively NaOH solution, washed with distilled water, and dried
environmentally friendly. The aim of this work was to in air for 60 min. The dried oxidized electrode was
study the adsorption of higher aliphatic carboxylates then held in an CI solution for 10 min and dried again
on Mg to form passivating layers that are stable in an in air.
aggressive wet atmosphere.
After such preparation, each sample was immersed
in a background solution and anodic polarization was
immediately switched on with a scanning rate of
0.2 mV/s.
EXPERIMENTAL
Our investigations were performed using technical
magnesium Mg90 of the following composition: Mg,
99.9%; Fe, up to 0.04%; Mn, up to 0.03%; Al, up to
0.02%; Ni, up to 0.001%; Cu, up to 0.004%; Si, up to
0.009%; and Сl, up to 0.005%.
Sodium salts of carboxylic acids were used as CIs:
oleate (OlS, (СН3(СН2)7СН=СН(СН2)7СООNa),
linoleate (LiS, CH3(CH2СН=СН)3(CH2)7COONa),
stearate (SS, CH3–(CH2)16COONa), lauryl sarcosinate
(LsS, С15Н28NNaO3), and oleyl sarcosinate (OsS,
СН3(СН2)7СН=СН(СН2)7СОN(CH3)СН2COONa).
Solutions of OsS, LiS, and SS were prepared by neu-
tralizing the corresponding acids with equimolar
amounts of NaOH. Solutions of OlS and LsS were
prepared from commercially available reagents. All
reagents were of pure grade.
Protective CI films were obtained by holding an
electrode in aqueous solutions of CIs (OsS, OlS, LiS,
and LsS) with concentration Сin = 16 mmol/L at room
temperature (t). The SS CI film was deposited from a
water–alcohol solution at t = 55°С.
Electrochemical investigations were performed
using cylindrical electrodes machined from technical
Mg, embedded in epoxy resin. The working surface
area of the electrodes was 0.75 cm2. Polarization
curves for Mg90 were measured in a glass cell with sep-
arated electrode spaces using IPC-Pro MF potentio-
stat (Russia). Potential (Е) of Mg90 electrode was mea-
sured vs. a silver chloride reference electrode, and recal-
culated to the standard hydrogen scale. The auxiliary
electrode was Pt.
The electrode for electrochemical investigations
was polished with abrasive papers with different grain
sizes and degreased with acetone. Two series of exper-
iments were performed. In the first, we obtained the
anodic and cathodic polarization curves of Mg90 in
borate buffer solutions with pH 9.2 that contained
1 mmol/L of NaCl with and without 8 or 16 mmol/L
of added CI. The electrode was immersed in a borate
buffer containing one inhibiting additive and held in it
for 15 min to determine the free corrosion potential
Ecor. Finally, polarization with a scanning rate of
0.2 mV/s was switched on.
In the second, polarization curves were measured
for the specimens with preliminarily formed adsorp-
tion CI films. The films were produced in two ways.
1. A polished and degreased electrode was held for C is an independent variable, С0 is the minimal con-
10 min in an CI solution and dried in air for 60 min.
Potential Еcd of local depassivation was found from
the polarization curves. The protective properties of
the passive films were found using the difference ΔЕ =
in
bg
bg
in
−
, where
and
are values Еcd, mea-
Е
Е
Е
Еcd
cd
cd
cd
sured for the electrodes with and without CI treat-
ment.
The adsorption of OsS and LiS on technical Mg
was investigated in situ via reflective ellipsometry in a
borate solution with рН 11.2 and 0.05 М of Na2B4O7 ·
10H2O + 0.1 М NaOH. Ellipsometric angles ∆ and Ψ
were measured on a portable ellipsometer (Rudolph
Research Co.) in a cell used for simultaneous electro-
chemical and ellipsometric investigations. The source
of radiation was a helium–neon laser with wavelength
λ = 640 nm, the angle of incidence on the sample was
68.5°. To ensure a stable surface (the constancy of
ellipsometric angles ∆ and Ψ), magnesium electrode
was subjected to special preliminary passivation. The
stripped, polished, and degreased electrode was
chemically oxidized in 5.0 М of NaOH solution for
1.5 h. The electrode was transferred to the working
solution (a borate solution with рН 11.2), where it was
held for another 17 h. The electrode’s surface retained
a high gloss after such treatment. After 17 h, a poten-
tiostat was used to maintain the electrode potential
20 mV more negative than Еcor. After angles ∆ and Ψ
became stable, the concentrated inhibitor solution was
added to the cell. For every concentration of CI (Сin),
angle ∆ was measured over time until it became con-
stant. For every change in Сin, the angle was
(1)
δΔ = Δ – Δ0,
where ∆ is the current angle obtained after the inhibi-
tor additive was introduced into the solution, with ∆0
is the initial value. We obtained the experimental
dependence of δ∆ on Сin; i.e., isotherm δ∆−ƒ(Сin),
which was reconstructed into isotherm of adsorption
Θ−ƒ(ln C). This isotherm is described by the Temkin
complete equation proposed in [12]:
1 + n−1Bmax(C − C0)
1
f
(2)
Θ = ln
.
1 + n−1Bmin(C − C0)
Here, f is the factor of energy surface inhomogeneity;
Bmax and Bmin are constants of the adsorption equilib-
rium corresponding to the highest and the lowest
energies of adsorption. In this equation, concentration
centration obtained via extrapolation Θ → 0, and n is
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 94 No. 6 2020

1106
OGORODNIKOVA et al.
(Russia) were used for electrochemical impedance
−δΔ, deg
spectroscopy (EIS). A three-electrode cell with undi-
vided electrode space was used in each experiment.
The reference electrode was a two-key silver chloride
electrode, the auxiliary electrode was Pt, and the elec-
trolyte was a borate buffer solution with рН 9.2 and
1 mМ of sodium chloride. Electrodes for electrochem-
ical impedance spectroscopy were prepared as for the
electrochemical and corrosion investigations. We
placed each treated sample in an electrolytic cell, held it
for 15 min to stabilize Еcor, fixed the potential, and
obtained spectra in the 0.01–100000 Hz range of fre-
quencies at a 0.01 V amplitude of potential oscillation.
1.4
(a)
1
1.2
1.0
0.8
0.6
0.4
0.2
0
2
−22.3
−21.3
−20.3
−19.3
−18.3
lnC [mol/L]
In calculating the EIS parameters, an equivalent
scheme was used to describe the impedance spectra of
magnesium alloys [13]:
θ
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0
(b)
2
1
Rs
Rsl
Qdl
Rst
Qsl
where Rs is the resistance of the bulk electrolyte solu-
tion between the auxiliary and operating electrodes;
Rs does not affect electrode processes, and its value
depends on the conductivity of the test medium and
cell geometry; Rsl is the resistance of the film of the
surface layers (both oxide and inhibitor); and Rst is the
polarization resistance characteristic of the electro-
chemical kinetics of corrosion.
−22.4
−21.4
−20.4
−19.4
−18.4
lnC [mol/L]
Fig. 1. Dependence of the change in (a) ellipsometric
angle (−δΔ) and (b) degree of filling (θ) on lnС for
in
Instead of pure capacities, elements of constant
phase Q were used in the equivalent scheme that were
due to the degree of inhomogeneity of surface films
and the electrochemical double layer [14]. Qsl is the
element of the constant phase characteristic of the
capacity of surface films (oxide and inhibitor). Qdl is a
component of the model that expresses the capacity of
electrical double layer of film defects.
(1) OsS and (2) LiS.
the normalization factor equal to the number of water
moles in 1 L. The procedure for calculating f, Bmax
,
and Bmin from the results of ellipsometric measure-
ments was given in [12]. Using Bi, we calculated the
corresponding values of the standard free energy of
adsorption using the relation
The results were processed and the data calculated
using the Dummy Circuits Solver program (version 2.1).
The correspondence between the experimental and
calculated data was at least 97%. The degree of mag-
nesium electrode protection was calculated according
to the formula
Bi = exp(–ΔGa0,i /(RT )),
(3)
where index i = max or min.
The calculations for passivating films were performed
according to the program at Internet resource http://
www.ccn.yamanashi.ac.jp/~kondoh/ellips_e.html.
Z = ((Rinh – Rbg)/R ) ×100%,
inh
Corrosion testing was done with periodic conden-
sation of moisture on both unoxidized and chemically
oxidized rectangular samples 20 × 30 × 5 mm in size.
The plates for corrosion testing were prepared as in
electrochemical investigations. Prepared samples were
hung in glass cells filled with 40–50 mL of distilled water
at t = 40–50°С. The cells with samples were placed in a
drying oven for 8 h at t = 40 2°С. Heating was then
switched off, ensuring the condensation of moisture on
a sample’s surface (State Standard 9.308-85).
where Rbg and Rinh is the total resistance of the Mg90–
electrolyte interphase interaction with the participa-
tion of Rst and Rsl before and after electrode passiva-
tion, respectively. All of the values given in this work
are the results from averaging 5–10 independent
experiments.
RESULTS AND DISCUSSION
Measuring the adsorption of inhibitor from the
The samples were examined hourly to fix the time background solution ellipsometrically allowed us to
of the first signs of corrosion. An IPC PRO-MF estimate the strength of its bond with the metal sur-
potentiostat and a FRA frequency response analyzer face. Unfortunately, because of the high reactivity of
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 94
No. 6
2020

ADSORPTION OF ANIONS OF HIGHER CARBOXYLIC ACIDS
1107
40
3
4*
1
3*
2*
4
30
20
10
2
−1.6 −1.4 −1.2 −1.0 −0.8 −0.6 −0.4 −0.2 0
0.2 0.4
i, μА/сm2
E, V
40 −10
3
1
4*
3*
4
30
20
−20
−30
−40
E, V
−1.6
10
0
2*
−1.5
−1.4
−1.3
−10
−20
−30
−40
4
1
3
3*
4*
2
Fig. 2. Anodic and cathodic polarization curves for technical magnesium (without chemical oxidation pretreatment) in a borate
buffer with pH 9.2 and 1 mM NaCl (1) without CI and after adding 8 mmol/L of CI: (2) OlS, (3) OsS, (4) LiS, and 16 mM of CI
(2*) OlS, (3*) OsS, (4*) LiS.
Mg, measuring the adsorption of CIs ellipsometrically
was possible only when the electrode’s surface was in a
stable passive state, which was attained via oxidation
according to the procedure described above.
As is seen from Fig. 1a, OsS is absorbed earlier than
LiS, so its free energy of adsorption is greater.
The thicknesses of the CI monolayers were found
from the values of angles Δ and Ψ. Because OsS
monolayer thickness δ = 1.2–1.5 nm and molecule
length lm = 2.5 nm, we may assume that each molecule
is adsorbed obliquely on the surface at an angle 36.9°.
A monolayer with δ = 0.9 nm forms on Mg90 oxidized
LiS with lm = 2.4 nm, so the angle of its slope is prob-
ably 66°.
Figure 1a presents the experimental dependences
(−δΔ) on ln Cin for OsS and LiS in a borate solution
with pH 11.2. The adsorption of carboxylates starts
within the region of low Сin = (2.2–5.5) × 10−10 mol/L
and manifests as a change in angle Δ. After reaching a
certain Сin, Δ becomes constant and a plateau can be
distinguished in the curve that is responsible for the
formation of a conventional inhibitor monolayer with
a surface filling of Θ = 1. However, angle Δ changes as
Сin continues to grow, testifying to further adsorption
of the inhibitor. According to the isotherm of OlS
adsorption obtained in [15], the conventional mono-
layer in this case formed at С = (0.42–1.58) ×
10‒6 mol/L.
It was shown in [9, 15] that OlS is a promising CI
for Mg alloys in neutral and weakly alkaline media.
The curve of polarization in a borate buffer with рН
9.2 and 8 mM of OlS is shown in Fig. 2. Not only does
this CI inhibit the cathodic reaction and raise Еcor by
0.15 V, it also provides a wide range of potentials in
which Mg90 is in the passive state.
When OsS and LiS (8 mmol/L) were introduced
into the borate buffer solution with рН 9.2, Еcor grew
by 0.024 and 0.092 V, respectively, but only for the LiS
solution can we note an increase in electrode polariz-
ability, testifying to the inhibition of both electro-
chemical reactions. The protective effect grew when
Сin was raised from 8 to 16 mM (Fig. 2), as is seen from
the increase in Еcor and the polarizability of Mg90 in
OsS and LiS solutions. In the anodic polarization
curve with an additive of 16 mM of OlS, the passive
We constructed the Θ−ƒ(ln C) isotherms of the
adsorption of these inhibitors from the linear depen-
dence of the changes observed experimentally in angle
Δ on degree Θ of the filling of the electrode’s surface
with CI particles. They are shown in Fig. 1b.
ꢀ
Table
1
presents values
(
)
and
−ΔGa,min
ꢀ
(
), coefficient f, Вmax, and Bmin, obtained
−ΔGa,max
using formulas (2), (3).
Table 1. Dependence of the logarithmic partition coefficients (logР), as well parameters f, (
chemical structures of carboxylates. logP values were obtained using the ACDLABS program
), (
) on
−ΔGa,max
−ΔGa,min
ꢀ
,
,
−ΔGa,max
kJ/mol
−ΔGa,min
i,max, mol−1
B
i,min, mol−1
Inhibitor
logP
f
В
kJ/mol
OlS
OsS
LiS
7.70
7.00
6.50
1.59 × 1010
2.15 × 1011
5.21 × 1010
3.65 × 109
3.57 × 1010
1.36 × 1010
53.65
59.20
56.86
57.22
63.58
60.12
1.47
1.8
1.31
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 94 No. 6 2020

1108
OGORODNIKOVA et al.
potential range is higher than when 8 mM of OlS was
introduced into the solution.
i, μА/сm2
60
(a)
3
2
The passivating action of CI aqueous solutions was
studied at Сin = 16 mM. The duration of the formation
of adsorption layers on Mg90 with an oxide film
formed in air (Fig. 3a) or via oxidation in alkali
(Fig. 3b) was 10 min. The electrode treated in this way
was placed in the borate buffer solution with рН 9.2
and immediately subjected to polarization.
50
40
30
20
5
1
4
6
10
0
The LiS film displayed a larger shift of initial
potential Еn of the electrode (−0.92 V), than that of
OlS. A region with low current density observed in the
anodic polarization curve is shorter than the one with
OlS in both the neutral buffer [9] and the BBS with pH
9.2. At the same time, the presence of LiS in the back-
ground electrolyte did not result in an emergence of
passive potential range, as it is shown above. It is likely
that the cross-linking the LiS molecules proceeds with
the participation of double bonds during preliminary
treatment and drying, but not when LiS is introduced
into the solution.
−1.65
−1.45
−1.25
−1.05
(b)
−0.85
E, V
i, μА/сm2
60
5
3
4
1
50
40
30
20
10
0
2
6
The inhibition of intense anode dissolution was not
observed in the anodic polarization curves when
unoxidized Mg was treated in 16 mM solutions of LsS
and SS, and En was shifted anodically by 50 and
159 mV, respectively.
−1.5 −1.4 −1.3 −1.2 −1.1 −1.0 −0.9 −0.8 −0.7 −0.6
E, V
Fig. 3. Anodic and cathodic polarization curves for techni-
cal magnesium (a) with no and (b) after chemical oxidation
in a borate buffer with рН 9.2 and 1 mM of NaCl (1) without
tentative adsorption, and after holding in 16 mM of CI solu-
tion for 10 min: (2) OlS, (3) OsS, (4) LiS, (5) SS, (6) LsS.
OsS, another anionic surfactant that also contains
one double bond and 17 carbon atoms in its hydrocar-
bon chain, is markedly inferior to OlS in inhibiting the
anodic dissolution of Mg. One difference between
these CIs is the higher hydrophobicity of OlS anions,
relative to those of OsS. The hydrophobicity of the
investigated compounds can be described by the loga-
rithm of the partition coefficient in the octanol–water
system, logР [16]. It helps to calculate anion hydro-
phobicity logD, allowing for pKa of the corresponding
acid [17] and the pH of the solution:
Ex situ ellipsometry showed that when magnesium
electrodes are preliminarily oxidized for 1.5 h in 5.0 М
NaOH, an oxide–hydroxide film ≈80 nm thick forms
on the surface. The picture changes somewhat after
oxidized Mg90 is treated with CI solutions (Fig. 3b).
The Еn value for the oxidized electrode is far more
positive than that of the electrode not subjected to
alkaline oxidation: 1.13 V instead of −1.57 V. This dif-
ference shows the protective action of the oxide–
hydroxide film. Subsequent treatment with the OlS
solution affects the Еn value slightly, but the passive
potential range becomes wider. This changes only
when Еcd ≈ −0.69 V, while OlS adsorption without
oxidation yields Еcd ≈ −0.85 V. The protective action
logD = logР – log[1 + 10pH−pK ].
a
These values can be calculated with the resource
ACD/I-Lab (https://ilab.acdlabs.com/). The logР
values for oleic acid and oleyl sarcosine are 7.70 and
7.0; logD at рН 7.4 is 5.10 and 3.61, respectively; at
рН 9.2, it is 3.3 and 2.6, respectively. However, hydro-
phobicity does not play a key role in Mg protection.
This is confirmed by the adsorption film of SS having of the LiS film is markedly improved, as is seen from
the highest hydrophobicity among the studied carbox-
ylates: logР = 8.23 and logD = 5.62 (рН 7.4) and 4.06
(рН 9.2). In protection effectiveness, it is markedly
inferior to analogous films formed with OlS and dif-
fers little from OsS. The great effectiveness of Mg90
protection with OlS and LiS is likely due to the ability
of their molecules to be cross-linked on a surface
because of the presence of double bonds. There was
virtually no protective effect after treating with the
solution of the least hydrophobic LsS (logР = 4.33).
the drop in current density in the passive potential
range. The protective action of adsorption films of
LsS, SS, and OsS declines, as is seen from the drop in
Еn to −1.42, −1.43, and −1.36 V, respectively. This was
likely due to the partial dissolution of the oxide–
hydroxide layer when the films were supported from
solutions.
Under conditions of the periodic condensation of
moisture onto Mg samples with naturally formed
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 94
No. 6
2020

ADSORPTION OF ANIONS OF HIGHER CARBOXYLIC ACIDS
1109
Table 2. Results from corrosion tests of Mg90 with periodic condensation of moisture
No.
1
Composition of passivating solution
pH
—
t, °С
τ, h
Without treatment
Stripped
—
0.5–1
17–20
18–20
92–96
8
43
2
33.5
8
25
Oxidized
Stripped
Oxidized
Stripped
Oxidized
Stripped
Oxidized
Stripped
Oxidized
Stripped
Oxidized
2
3
4
5
6
16 mM OlS
8
8
20
20
20
20
55
16 mM OsS
16 mM LsS
16 mM LiS
9.4
16 mM SS
From water–alcohol solutions
8
20.5
t is temperature of treatment; τ is the time before the first attack.
oxide film, corrosion begins at τcor = 30–60 min. The tection: τcor = 25, 33, and 43 h, respectively. Treating
least efficient CI on stripped Mg is LsS (τcor = 2 h). the oxidized Mg90 with 16 mM OlS aqueous solution
provided the greatest protection: τcor = 94 2 h.
Films produced with OsS, SS, and LiS (τcor = 8 h) are
four times more effective. Preliminary treatment of
Mg with 16 mM OlS is most effective, providing pro-
tection up to τcor = 18–20 h (Table 2). The thickness of
OlS after such Mg treatment in 16 mM aqueous solu-
tion without subsequent washing was ≈60–62 nm.
EIS is a well known means of surface investigation
used to study magnesium and its alloys. It allows us to
study the mechanism and kinetics of electrode processes,
along with the characteristics of passive films, with a
minimal effect on films. EI spectra obtained for stripped
and oxidized Mg90 samples are presented in Fig. 4, along
with ones for samples passivated with OlS.
Mg90 samples after oxidation and/or passivation in
OlS solutions have larger hodograph radii than ones
that are not treated, testifying to their greater corro-
sion resistance. The calculated nominals are given in
Table 3. Rct grows considerably for oxidized and/or
extrapassivated samples, due to a drop in the rate of
electrode processes. Rsl also grows, which could indi-
cate a thicker protective layer with improved stability
in a solution.
An oxide–hydroxide film of 80 nm thick formed in
5 М NaOH solution has a protective effect commen-
surable with the one for OlS on Mg90 (τcor = 17–20 h).
In the subsequent treatment of oxidized Mg90 with a
16 mM SS water–alcohol solution, a nonuniform
white deposit that was easily removed with filtering
paper formed on the surface. The τcor value for oxi-
dized samples covered with the SS film was 20.5 h.
LiS, LsS, and OsS films provided some enhanced pro-
The oxidation of electrodes results in a slight rise in
Rct and a growth of 1.5 times in Rsl. This indicates a
drop in the rate of corrosion processes. A small reduc-
tion in Qdl was also observed, indicating a drop in the
fraction of the elecrtrochimically active surface of Mg.
The small increase in Qsl is a consequence of a thicker
layer of magnesium oxide in the alkaline oxidation
process. The values of n allow us to assume that the
corrosion rate of the stripped sample was due to diffu-
sion of the reagents and products of corrosion to or
from the surface. However, oxidation reduced the
fraction of diffusive processes on the electrode’s sur-
face.
Treating pure and oxidized Mg90 electrodes with
OlS raises Rct hundreds of times. This demonstrates
the strong inhibition of electrode processes. The Qdl
value falls relative to background by two orders of
magnitude, testifying to the large drop in the area of
the elecrtrochimically active surface. The n values
ZIm, Ω сm2
3500
2*
3000
2500
ZIm, Ω сm2
2000
200
150
100
50
0
2*
2
1500
1000
500
2
1*
1
0
200
400
ZRe, Ω сm2
0
2000 4000 6000 8000 10 000
ZRe, Ω сm2
Fig. 4. Nyquist plots for technical magnesium in a borate
buffer with рН 9.2 and 1 mM of NaCl: (1) without treat-
ment, (1*) after preoxidation in 5 М NaOH, (2) after pas-
sivation of stripped Mg90 in 16 mM of OlS aqueous solu-
tion, (2*) after passivation of preoxidized Mg90 in 16 mM
of OlS aqueous solution.
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 94 No. 6 2020

1110
OGORODNIKOVA et al.
Table 3. Calculated values for Mg90 electrodes in BBS with 1 mM NaCl pH 9.2
Rct, Ω cm2
Rsl, Ω cm2
Rr, Ω cm2
CPEdl, Т
0.003
ndl, Ф
CPEsl, Т
nsl, Ф
200.83 3.9 × 10–5 0.875
300.21 5.2 × 10–5
0.812
Z1
Z2
Samples
Mg90 stripped
Mg90 oxidized
16 mM OlS on
stripped Mg90
16 mM OlS on
oxidized Mg90
16.48
23
4803
0.62516
5.3 × 10–4 0.90000
1.4 × 10–5 0.57935
36.635
29.619
43.382
—
32.77
96.49
—
—
—
1391.00 1.5 × 10–6 0.963
5741
1.1 × 10–5
1
6880.60 1.8 × 10–5 0.607
14.197
98.28
97.44
Z and Z are relative to stripped and oxidized Mg90, respectively.
1
2
allow us to assume that on electrodes passivated with
OlS without preliminary surface oxidation, diffusive
processes make a substantial contribution to the
observed inhibition. This contribution is far less on
preliminarily oxidized electrodes passivated with OlS.
REFERENCES
1. K. U. Kainer, Magnesium: Alloys and Technology (Wi-
ley, Chichester, 2003).
2. A. Mesbah, C. Juers, et al., Solid State Sci. 9, 322
(2007).
When inhibiting films form, the Rsl value falls sub-
stantially (which in this case characteristically
describes surface oxide–inhibitor layers), so their val-
ues on preliminarily oxidized samples are far higher
than on unoxidized ones. The Qsl values also fall
somewhat. The protective effect the inhibitor film has
on preliminarily formed oxide is thus more effective
than protection with an inhibitor of Mg90 surface
without oxidation.
The nst value testifies to diffusion processes pro-
ceeding in surface layers. However, their contribution
is far greater for inhibiting treatment of an oxidized
surface than when magnesium is passivated without
preliminary oxidation in an alkaline solution.
3. X. B. Chen, M. A. Easton, and N. Birbilis, in Corrosion
Prevention of Magnesium Alloys (Woodhead, Philadel-
phia, 2013), p. 282.
https://doi.org/10.1533/9780857098962.2.282
4. P. A. Guilherme, T. Alfredo, P. Nunes, et al., J. Alloys
Compd. 738, 224 (2018).
https://doi.org/10.1016/j.jallcom.2017.12.115
5. S. V. Lamaka, B. Vaghefinazari, Di Mei, et al., Corros.
Sci. 128, 224 (2017).
https://doi.org/10.1016/j.corsci.2017.07.011
6. S. V. Lamaka, D. Höche, R. P. Petrauskas, et al., Elec-
trochem. Commun. 62, 5 (2016).
https://doi.org/10.1016/j.elecom.2015.10.023
7. Yu. I. Kuznetsov, Organic Inhibitors of Corrosion of Met-
als (Plenum, New York, 1996).
8. Yu. I. Kuznetsov, Korroz.: Mater., Zashch., No. 1, 1
(2017).
9. A. M. Semiletov, Yu. I. Kuznetsov, and A. A. Chirkun-
ov, Korroz.: Mater., Zashch., No. 7, 25 (2016).
Our investigations showed that the best protective
action was Mg90 passivation with preliminary oxida-
tion of its surface in an alkaline solution, which cor-
relates with the corrosion and voltammetry data.
10. S. V. Oleinik, V. C. Rudnev, Yu. A. Kuzenkov, et al.,
Korroz.: Mater., Zashch., No. 10, 39 (2015).
11. S. V. Oleinik, V. C. Rudnev, Yu. A. Kuzenkov, et al.,
Korroz.: Mater., Zashch., No. 12, 29 (2016).
CONCLUSIONS
The values of the free energy of OsS and LiS
adsorption are
> 55 kJ/mol, so we may assume
−Ga,max
they are chemisorbed on an oxidized surface of Mg.
Treating oxidized Mg90 with an aqueous 16 mM solu-
tion OlS effectively inhibits its anodic dissolution in a
borate solution containing chloride ions and slows
atmospheric corrosion even under conditions of daily
condensation of moisture on samples. Preliminary oxi-
dation of an Mg surface in an alkaline solution enhances
the passivating action of CI aqueous solutions, particu-
larly the one most effective, sodium oleate.
12. M. O. Agafonkina, N. P. Andreeva, and Yu. I. Kuznetsov,
Russ. J. Phys. Chem. A 91, 195 (2017).
13. L. M. Calado, M. G. Taryba, M. J. Carmezim, et al.,
Corros. Sci. 142, 12 (2018).
https://doi.org/10.1016/j.corsci.2018.06.013
14. L. J. Zhang, J. J. Fan, Z. Zhang, et al., Electrochim.
Acta 52, 5325 (2007).
https://doi.org/10.1016/j.electacta.2006.12.083
15. V. A. Ogorodnikova, Yu. I. Kuznetsov, and N. P. An-
dreeva, Korroz.: Mater., Zashch., No. 5, 27 (2018).
FUNDING
16. C. Hansch and A. Leo, Correlation Analysis in Chemis-
try and Biology (Wiley New York, 1981).
This work was supported by the Russian Science Foun-
dation, project no. 17-13-01413 “Fundamental Aspects of
the Formation of Hyperfine Passivating Films of Organic
Compounds on Metals to Protect against Atmospheric
Corrosion.”
17. R. A. Scherrer and S. M. Howard, Med. Chem. 20, 53
(1977).
Translated by S. Lebedev
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 94
No. 6
2020