Accepted Manuscript - GECEAgecea.ist.utl.pt/Publications/SL/26_2011_Lamaka... · Accepted...
Transcript of Accepted Manuscript - GECEAgecea.ist.utl.pt/Publications/SL/26_2011_Lamaka... · Accepted...
Accepted Manuscript
Title: The combined use of scanning vibrating electrodetechnique and micro-potentiometry to assess the self-repairprocesses in defects on “smart” coatings applied to galvanizedsteel
Authors: M. Taryba, S.V. Lamaka, D. Snihirova, M.G.S.Ferreira, M.F. Montemor, W.K. Wijting, S. Toews, G.Grundmeier
PII: S0013-4686(11)00263-5DOI: doi:10.1016/j.electacta.2011.02.048Reference: EA 16789
To appear in: Electrochimica Acta
Received date: 25-8-2010Revised date: 9-2-2011Accepted date: 10-2-2011
Please cite this article as: M. Taryba, S.V. Lamaka, D. Snihirova, M.G.S. Ferreira,M.F. Montemor, W.K. Wijting, S. Toews, G. Grundmeier, The combined use ofscanning vibrating electrode technique and micro-potentiometry to assess the self-repairprocesses in defects on “smart” coatings applied to galvanized steel, ElectrochimicaActa (2010), doi:10.1016/j.electacta.2011.02.048
This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.
Page 1 of 36
Accep
ted
Man
uscr
ipt
Weldable primers were modified with submicron containers loaded with corrosion inhibitors
SVET and micro-potentiometry were used to study the corrosion inhibition ability
Submicron containers do not damage the barrier properties of model primers
Artificial defects of 50x50 μm in a coating can be easily analyzed by SVET and SIET
Inhibiting dissolution of sacrificial Zn may result in detrimental dissolution of Fe
*Research Highlights
Page 2 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
1
The combined use of scanning vibrating electrode technique and
micro-potentiometry to assess the self-repair processes in defects on “smart”
coatings applied to galvanized steel
M.Taryba1, S. V. Lamaka1*, D. Snihirova1, M.G.S. Ferreira1,2, M.F.Montemor1, W.K.
Wijting3, S. Toews3, G. Grundmeier3
1ICEMS, Instituto Superior Tecnico, UTL, Av. Rovisco Pais, 1049-001 Lisbon, Portugal
2CICECO, Dep.Ceramics and Glass Eng., University of Aveiro, 3810-193, Aveiro, Portugal
3Institute for Polymer Materials and Processes, University of Paderborn, 33098 Paderborn,
Germany
Abstract
Model weldable primer coatings for galvanized steel were modified with submicron
containers loaded with corrosion inhibitors. This procedure aims at introducing a new
functionality in the thin coatings: self-repair ability. The assessment of this property demands
new protocols and new approaches, combining conventional electrochemical methods with
electrochemical and analytical techniques of micrometer spatial resolution. Thus, in this work
model defects were created in the coatings by using a focused ion beam (FIB). The coated
samples, containing the model defects, were immersed in a NaCl 0.05 M solution and the
corrosion inhibition ability was studied using the scanning vibrating electrode technique
(SVET) and the scanning ion-selective electrode technique (SIET). SVET-SIET measurements
were performed quasi-simultaneously. Qualitative chemical analysis was performed by SEM
combined with EDS. Complementary studies were carried out by electrochemical impedance
spectroscopy (EIS) to assess the effect of the containers filled with corrosion inhibitors on the
barrier properties of the coatings. The electrochemical results highlight the importance of the
combined use of integral and localized electrochemical techniques to extract information for a
better understanding of the corrosion processes and corresponding repair of active microscopic
defects formed on thin coatings containing inhibitor filled containers.
Keywords: SVET; micro-potentiometry; SIET; galvanized steel; self-repair.
*Corresponding author: [email protected] , +351 218 417 996.
1. Introduction
Organic and inorganic coatings have been widely applied for the protection of metal
alloys against corrosion with significant progress in recent years. However, despite essential
improvements in anti-corrosion coating technologies, problems persist in the long-term
Page 3 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
2
protection of metals in aggressive environments at very high cost. An increased interest has
been observed in approaches that combine different classes of materials, achieving more
complex and feed-back active structures in order to obtain new synergistic effects with respect
to the functional performance. A key strategic topic is the combination of nanostructured
materials and eco-friendly hybrid and/or organic polymers to develop new coatings that
enhance the product performance and introduce new functionalities as for example self-healing
ability.
Nowadays, the study of different coating systems (pre-treatments, primers, inorganic and
hybrid sol-gel films, conductive polymers and water based coatings) with self-healing ability is
an important topic [1], with a significant increase in the number of publications in the last five
years.
Based on literature it is possible to identify two main types of electrochemical mechanisms
for the self-healing of corrosion processes:
a) Healing due to the formation of protective layers of corrosion products, which block the
access of aggressive species and/or oxygen towards the active sites [2-4].
b) Healing due to the formation of protective species, resulting from the presence of organic or
inorganic corrosion inhibitors, often incorporated into “smart” particles [5-13].
These mechanisms of self-healing have been studied in different aggressive environments,
mainly on aluminum and zinc alloys, including cut edges. It has been shown that the corrosion
mechanisms and, consequently, the mechanisms of self-healing are dependent upon the
substrate composition and nature of the corrosion products formed [2-3, 14]. According to [2],
corrosion of zinc coupons immersed in NaCl electrolytes reveals slight acidification of the
anodic zones and major alkalinization of the cathodic sites. Corrosion products precipitate
between anodic and cathodic areas, where pH is favorable for formation of insoluble species.
In coated metallic substrates, both corrosion and self-healing mechanisms can be localized
over very small areas and generally involve changes in the electrochemical potential, current
density and acid-base equilibrium reactions. Furthermore, the processes may become even
more complex due to cathodic delamination or anodic undermining phenomena under buried
interfaces. The effective characterization of the self-repair ability is an acknowledged need due
to the novelty and dynamic behavior of these systems. In this perspective, the combined use of
integral and spatially-resolved electrochemical techniques becomes fundamental. Some of the
most successful spatially resolved electrochemical tools that have been used to investigate
localized corrosion phenomena in uncoated and coated substrates are shortly addressed below.
The Scanning Vibrating Electrode Technique (SVET), that detects local changes of the
current density due to the potential differences originated in the surface as consequence of the
ionic fluxes generated between the anodic and cathodic sites, has become one of the most
effective and widely used techniques to detect localized corrosion phenomena in bare and
coated substrates [2-4, 13, 15-24].
Page 4 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
3
The micro-potentiometric measurements with ion-selective microelectrodes can be
performed by using Scanning Ion-selective Electrode Technique (SIET) and SECM in
potentiometric mode. This method provides information about local activity (concentration) of
specific ions in solution (e.g. H+) making possible to trace changes in the acid-base equilibriums
associated with the electrochemical processes [2, 3, 8, 16-20, 25-28]. Mg2+, Zn2+, Na+ and Cl-
selective microelectrodes have also been reported to be used for studying localized corrosion
[16-18, 29, 30].
The localized electrochemical techniques described above present advantages and
limitations, and the need of complementary application has been stressed in literature [2-4, 16-
20, 24, 28, 29]. In this work a complementary combination of different electrochemical
techniques was used to study the corrosion and self-repair processes in model defects. For this
purpose, model weldable primers were applied to galvanized steel substrates. Additionally the
primers were modified with submicron containers loaded with different corrosion inhibitors.
The barrier properties of five different coatings were first assessed by electrochemical
impedance spectroscopy. Based on EIS ranking, the corrosion mechanisms and the self-repair
ability of two best coatings and the reference sample were studied using quasi-simultaneous
SVET and SIET measurements, complemented with SEM/EDX analysis. The results highlight
the need of new experimental protocols and the synergistic combination of SVET and micro-
potentiometry with conventional electrochemistry and scanning electron microscopy, providing
new information both on the corrosion processes and the self-repair effects occurring on defects
formed in coated galvanized steel.
2. Experimental
2.1. Materials
The substrate used for coating application consisted of hot dip galvanized steel, thickness
0.8 mm, coated with 22 m zinc layers on both sides. This substrate was supplied by the
industrial company coated with model weldable primers without nanocontainers. (Reference
sample M0) or modified with different containers loaded with corrosion inhibitors as described
in Table 1. Layered double hydroxide (LDH) nanocontainers were prepared according to the
procedures described in [31]. Preparation of the titania containers and loading with organic
inhibitors is detailed in [32]. Application of natural halloysites for corrosion protection was first
reported in [6]. The halloysite nanocontainers used in this work were prepared and
characterized as described in [33]. Apart from nanocontainers, the model weldable primer
contains spherical zinc particles (50% w/w) with diameter ranging from 2 to 8 m embedded in
an organic epoxy matrix. Along with Zn particles, the primer also contains silica and organic
additives meant to inhibit the dissolution of the zinc particles and prolong primer’s service life.
This primer is a model system based on a formulation used in the automotive industry. The
Page 5 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
4
cross-section SEM image of the reference sample supplemented with Fe, Zn and Si EDS
elemental mapping is presented in Fig. 1.
Two sets of samples were prepared: one set was used for evaluating the barrier properties
of the modified coatings by EIS. Based on these results the best samples were chosen for the
formation of Focused Ion Beam (FIB) defects and further studied by SVET and SIET.
2.2. Preparation of FIB defects and their SEM/EDS characterization
Defect preparation in organic coatings by means of the FIB technique consists of cutting
the coating through Ga-ion milling while the prepared cut is controlled by means of Field-
Emission Electron Microscopy (FE-SEM) [34]. Based on the relative movement of coated sample
and ion beam, three dimensional defects can be prepared. To the best knowledge of the authors
FIB was not yet applied to prepare 3D defects in organic coatings for a subsequent
electrochemical analysis. The coating around the defects remains effectively unchanged. In
each coated sample a defect with a size of 50 x 50 μm was made by means of the FIB technique
(Zeiss Neon 40). The FIB beam was placed perpendicularly to the surface, milling away the
coating until a defect was created. FIB was used to produce microsized defects that intend to
replicate small defects formed in the coating. The technique is suitable to produce identical
defects because not only the size of the defect can be well defined, but also the depth is
controlled. Regularly, SEM images were taken and EDS analysis was performed to control
composition and depth.
A Scanning Electron Microscope, coupled with Energy Dispersive X-ray Spectroscopy
(SEM/EDS), was used for examining the microstructure and the chemical composition of the
samples after SVET-SIET measurements. A semi-in-lens Hitachi SU-70 UHR Schottky
(Analytical) FE-SEM microscope coupled with a Bruker EDS detector was used. An electron
beam energy of 15 keV or 25 keV was applied for SEM analysis and EDS mapping.
2.3. Electrochemical techniques
All the coated specimens (without defects) were studied by EIS in order to evaluate the
effect of the containers on the barrier properties of the coating. Measurements were taken in 0.5
M NaCl. The EIS were performed using a Gamry FAS2 Femtostat with a PCI4 Controller in a
frequency range from 100 kHz down to 0.01 or 0.003 Hz. All the spectra were recorded at open
circuit potential, applying a 10 mV sinusoidal perturbations (rms signal). A conventional three-
electrode cell was used, consisting of a saturated calomel reference electrode, a platinum wire as
counter electrode and the coated metallic coupon as working electrode. The area of the working
electrode was approximately 3.4 cm2. The cell was placed in a Faraday cage.
The evolution of the corrosion activity and the self-healing processes on the samples
showing the best barrier properties were studied by the SVET and SIET techniques. A
commercial device from Applicable Electronics, controlled by the ASET program
(Sciencewares), was used to perform the SVET and the SIET measurements.
Page 6 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
5
The FIB defect was in the centre of the scanned area. The area exposed to the electrolyte
was approximately 0.9 x 0.7 mm2 for the sample M0, 0.5 x 0.4 mm2 for the M1 sample and 0.5 x
0.5 mm2 for the M4 sample. The rest of the sample’s surface was isolated using bee wax.
Samples M2 and M3 were not tested, since they demonstrated the weakest barrier properties
among the samples tested by EIS. The local ionic current density and pH were mapped at 35 x
35 grid generating 1225 data points. Measurements were taken in 0.05 M NaCl solution.
An insulated Pt-Ir probe (Microprobes Inc.) was used as vibrating electrode for SVET
measurements. A spherical layer (d=12 ± 2µm) of platinum black was deposited on the exposed
tip of the probe. The probe was placed 100±3 µm above the surface, vibrating in the planes
perpendicular (Z) and parallel (X) to the sample’s surface. The vibration frequencies of the
probe were 124 Hz (Z) and 325 Hz (X).
The localized pH measurements were carried out using pH-selective glass-capillary
microelectrodes. Silanized glass micropipettes were back-filled with the inner filling solution
and tip-filled with the H+-selective ionophore-based membrane. The ion-selective membrane
was prepared in-house and was composed of 6 wt% 4-nonadecylpyridine, 12 mol% potassium
tetrakis(4-chlorophenyl)borate and membrane solvent 2-nitrophenyloctyl ether. All reagents for
the membrane of pH-SME were Selectophore grade products from Fluka. The diameter of the
tip of the glass-capillary microelectrodes was 2 ± 0.3 μm. The pH-selective microelectrodes were
calibrated using commercially available pH buffers in a range from pH = 2 to 10 and
demonstrated linear Nernstian response -54.4 ± 0.9 mV/pH. The local activity of H+ was
detected 30±3 µm above the surface.
SVET and SIET scans were taken sequentially to trace the evolution of corrosion process in
the case of the M0 sample. Quasi-simultaneous SVET-SIET measurements were made on the M1
and M4 samples. To this end the vibrating probe of SVET and the pH-selective microelectrode
of SIET were brought up together to map the ionic current density and pH distribution
simultaneously. The 1.5 second time lag between reading SVET and SIET signals required due
to the response time of the pH-selective probe. The details of methodology of quasi-
simultaneous measurements are presented elsewhere [35].
3. Results and discussion
3.1. Barrier properties of the model primers
The addition of submicron particles loaded with corrosion inhibitors in the model primer
aims at introducing self-healing ability in the coating; however it is of outmost importance that
these additives do not damage the protective barrier properties. Otherwise, the protective
performance of the coating will be reduced and the addition of containers is not recommended,
despite the fact that they could impart corrosion inhibiting ability.
Page 7 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
6
EIS measurements were performed to evaluate the barrier properties and, therefore, the
anti-corrosion performance of the coated model systems. Fig. 2 depicts the EIS results obtained
for the samples M0, M1, M2, M3 and M4 (as listed in Table 1) after 2 hours and after 48 hours of
immersion in NaCl 0.5 M. At the early stages of immersion all the spectra revealed a capacitive
response in the high frequency domain, which can be related to the barrier properties of the
coating. In the low frequency range there is a resistive response, which is a consequence of the
presence of zinc microparticles or electrolyte uptake and formation of conductive pathways
through the coating. The M4 and M1 samples showed the highest coating resistance at the early
immersion stages, which was approximately 1 MΩ. At this early stage, all the samples showed
total impedance values identical or above those of the blank sample, showing that the additives
filled with corrosion inhibitors do not damage the barrier properties of the coating.
As the immersion time elapsed, the total impedance decreased for all the samples. After 2
days of immersion all the spectra presented similar behavior, showing impedance values in the
range 30 to 130 kΩ cm2. At this stage, the sample M1 revealed the highest overall impedance
values.
Since the additives must not damage the barrier properties of the coating, its resistance
was determined by fitting procedures. The equivalent circuit described elsewhere [7, 15]
consisted of an association of two time constants connected in parallel. Each time constant
included a resistive element and a constant phase element. This association, also including a
series resistive element to account for the electrolyte resistance, was used to fit the experimental
impedance spectra and to calculate the evolution of the coating resistance during immersion in
0.5M NaCl, Fig. 3. The fast decrease of the coating resistance at the beginning of immersion is
related with penetration of the electrolyte in the pores of the coating. Sample M2 showed
barrier properties that were very close to the blank coating, whereas the M4 and M1 systems
provided the highest coating resistances. This resistance showed a sudden drop during the first
day of immersion and, after 2 days, the coating resistance became stable and similar to that of
the blank sample. Concerning long term barrier properties, the M1 sample revealed the highest
coating resistances.
The impedance results clearly show that the submicron containers added to the model
weldable primer do not damage the barrier properties of the coating. Moreover, in some cases
the additive could improve the protective performance at least at the early stages of immersion.
Since the M1 and M4 samples revealed the best performance regarding the barrier properties,
these samples were chosen to create the FIB defects and to study the self-healing mechanisms as
presented in the next sections.
3.2. FIB defects analysis
To assess the self-healing ability, a model defect was prepared by means of FIB, at 30 kV
and 200 pA. The samples were milled for 3 µm. SEM images of the defect in the M0 coating are
Page 8 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
7
shown in Fig. 4. Since the coating is a weldable primer it contains zinc particles, which are
evident in the SEM/EDS images.
EDS maps of the defect are shown in Fig. 5. In the defect the content of C, O, and Si were
significantly lowered after the defect formation as consequence of the coating removal. The
substrate consists of a galvanized layer, so the signal of zinc in the defect increased,
comparatively to that of the coating.
Fig. 6 depicts the cross-section SEM image and the EDS spectrum of the defect formed in
the M1 coating. The largest and round shaped particles are the zinc powder pigments and the
longest ones, as highlighted in the square, are the added halloysites. The analysis reveals the
presence of silicon from the coating and zinc from the substrate. Rather high content of
aluminum as well as traces of magnesium and phosphorus accounts for the halloysites, which
are aluminosilicate clay mineral [6, 33].
In Fig. 7a and b a cross-section and a top view of the defect formed in the M4 coating are
presented. The EDS spectra obtained over the highlighted areas (continuous and dashed) are
presented in Fig. 7c and d, respectively. Fig. 7c reveals the presence of titanium, which comes
from the titanium dioxide particles used as inhibitor reservoir.
3.3. Combined SVET and SIET study
3.3.1. Reference sample (M0) - blank primer
Figs. 8a,b show the optical micrographs of the area tested by the SVET and SIET. The dark
square dot in the middle is the FIB defect. Fig. 8a was taken after one day of immersion in
0.05M NaCl. A whitish border of corrosion products of roughly circular shape is visible in the
central and lower left part of the sample. Fig. 8b indicates the position of the SVET and SIET
scans. Figs. 8c,d show the SIET and SVET maps, respectively.
The SVET map depicted in Fig. 8d show the anodic activity (positive ionic current fluxes)
over the defect and corresponding cathodic activity (negative current fluxes) in the right part of
the scanned area. Corrosion activity was detected at the very beginning of the immersion and
there was a gradual increase of the anodic and cathodic current density with time. After one
day of immersion the anodic current density attained values above 80 μA∙cm-2 and the cathodic
current density remained around -40 μA∙cm-2.
The pH map (Fig. 8c) reveals significant acidification over the defect and alkalinization in
the upper and right parts of the scanned area. The pH recorded in the alkaline zone attained
values 9.1 while the lowest measured pH in acidic area reached 3.5. Such profound acidification
over the defect is worthy of special attention. Assuming that the anodic dissolution would
generate only Zn2+ due to dissolution of the galvanized layer and zinc particles from the primer,
the pH in the anodic zone would not be lower than 5.0. The change of pH is predetermined by
the hydrolysis of Zn2+ cations:
Page 9 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
8
Zn Zn2+ +2e- (1)
nH[Zn(OH)n]OnHZn n)-(2
2
2
pK 1
hyd = 7.96 K 1
st = 1.10·106 (2)
According to the equation (2), the expected pH values are: pH = 4.0 if [Zn2+] = 1M, pH= 5.0
if [Zn2+] = 0.01M and pH= 6.0 if [Zn2+] = 0.0001M. The Kst1 value was taken from [36]. pK1
hyd was
calculated from K 1
st . For the studied sample, with a defect size of 50x50 µm, the local
concentration of Zn2+ is unlikely to be higher than 0.01M. For comparison, the maximum Zn2+
concentration measured using Zn2+-selective microelectrode over an anodically polarized 500
µm diameter Zn wire was found to be 0.01M [18]. Accessed by a disc electrode, maximum
concentration of Zn2+ during galvanic corrosion of a model Zn/steel couple was found to be
6.5∙10-3M [37].
However, the pH values detected by SIET in the middle of anodic zone are lower than 4. A
plausible explanation for such acidic pH can be the presence of Fe3+ ions, which cause a
considerable decrease of pH, due to hydrolysis of Fe3+ ions:
Fe Fe2+ +2e- (3)
Fe2+ Fe3+ +e- (4)
mH][Fe(OH)OmHFe m)(3
m2
3 pK 1
hyd = 2.20 K 1
st = 6.31·1011 (5)
The equilibrium reactions for the hydrolysis of Fe3+ (eq. 5) show that the expected pH values
are: pH = 2.3 if [Fe3+] = 0.01M and pH= 4.0 if [Fe3+] = 0.0001M. Thus, dissolution and hydrolysis
of Fe3+ are likely to contribute to the highly acidic pH values detected by micro-potentiometry,
Fig. 8c. The presence of the red rust zone around the defect in the optical images (Fig. 8a)
corroborates this explanation.
The reduction of dissolved oxygen is the main cathodic reaction, producing hydroxyl ions,
which are responsible for the increased pH.
O2 + 2H2O + 4e- 4OH- (6)
Comparing the pH map, Fig.8c, with the optical micrograph, Fig. 8a, it is evident that the
distribution of corrosion products follows the pH map. The boarder of the acidified zone, the
distorted circle around the anodic spot characterized by pH values in the range 6-7, Fig. 8c,
match the whitish deposits of corrosion products that are visible in Fig. 8a. The light blue curve
Page 10 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
9
in the lower middle part of the SVET map, Fig. 8d, also indicates that the current densities are
decreased over the area where the precipitation of corrosion products occurred. According to
Fig. 9a, this roughly circular area of corrosion products is likely to be composed of zinc oxide
ZnO or hydroxide Zn(OH)2 or hydroxylchloride Zn5(OH)8Cl2, which precipitate from solution
as pH reaches 6.3 and higher. Investigations performed by Ogle et al. on cut edges also revealed
the presence of zinc-based corrosion products precipitated in zones of intermediate pH, which
were in that case identified as ZnO and 3Zn(OH)2·2ZnCO3 [4] or Zn5(OH)8Cl2 [2] working as
cathodic inhibitors.
The pH values observed on the anodic sites of the M0 sample are more acidic, while the
alkalinization in cathodic areas is milder than reported in previous publications on cut edge
galvanized steel. The two most likely explanations for this fact are: i) A coating with a micro-
size defect (50x50 m) obviously generates less activity than the cut edge with an area of 10000
x 810 m, as, for example, tested by Ogle et al [4]. ii) In their localized pH measurements Ogle’s
group, as well as Ding and Hihara [28], use Hydrogen I cocktail B (Fluka, Ref.95293) for
producing glass-capillary microelectrodes. Remarkably, its functional pH range is 5.5 to 12 [39].
Thus, even if present, lower pH values simply could not be quantified with the tool used. In this
work we used another ionophore to produce pH-selective microelectrodes: 4-
nonadecylpyridine-based cocktail demonstrated a linear Nernstian response in pH ranging
from 2 to 10 making the detection of acidified sites possible. For dynamic and concentration
calibration curves of the microelectrodes based on tridodecylamine (Hydrogen I, Fluka) and 4-
nonadecylpyridine, interested readers can refer to our review [29].
Fig. 10 shows the results of the SEM/EDS measurements performed on the same area as
depicted in Fig. 8a. Depletion of zinc (relatively to initial even distribution of Zn-rich primer)
and accumulation of chlorine around the defect is shown in Fig.10 c and d. Silicon, which is part
of the formulation of the weldable primer, was also detected, being depleted in the same areas
as zinc, Fig.10b. This result reveals decomposition of the primer in the acidified area. SEM
observation of the defect after immersion indicates that the defect is partly blocked with the
corrosion products, Fig. 10e. The EDS spectrum over the defect, Fig. 10f, reveals the presence of
O, Cl, Zn and Fe. Modeling of ionic equilibrium in the solution containing Fe3+, Zn2+, Cl- and
CO32- suggests that the species blocking the defect are likely to be nonstoichiometric iron
hydroxylchloride Fe(OH)2.7Cl0.3, which starts precipitating at pH as low as 1, Fig. 9b. This
supports the statement that co-dissolution of zinc and iron occurred at the corroding defect,
taking active part in the ionic flux detected in the anodic site of the SVET maps.
Indeed, the model weldable primer contains additives which aim to hinder the dissolution
of the zinc particles and prolong primer’s service life. However, if sacrificial zinc is being
dissolved too slowly, the exposed steel substrate can undergo anodic dissolution resulting in
the major acidification of the defect area (eqs. 3-5). SVET alone could not reveal the chemistry
of the anodic reaction, while SIET suggested that both Zn and Fe take part in anodic dissolution.
This essential conclusion about the detrimental effect of the inhibiting additives in a particular
Page 11 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
10
primer formulation could not be attained by only using localized electrochemical technique that
measures net positive and negative currents such as SVET. This example shows the importance
of complementary use of SVET and SIET for elucidating the chemical mechanisms of corrosion
reactions.
3.3.2. Primers modified with nanocontainers loaded with inhibitors (M1 and M4)
Quasi-simultaneous SVET-SIET measurements were performed on the M1 and M4
samples. Figs. 11-a show the optical micrographs of the M1 sample and corresponding SIET
(Figs. 11-b) and SVET (Figs. 11-c) maps after different immersion times.
According to the SVET scan Fig. 11-1c, high corrosion activity could be detected after one
hour of immersion, attaining values around 80 μA∙cm-2, which are very close to those observed
for the blank sample, M0. The SIET map also shows acidification in the defect, characterized by
a decrease of the pH to values around 5, Fig.11-1b.
As the immersion time elapsed, both pH and current density show important changes.
After 8 hours of immersion (Fig. 11-2 a, b, c) the pH measured in the anodic zone ranges
between 4.6 and 5.0, while pH in cathodic zone is around 6. As compared to the reference
sample, the pH changes are more attenuated, both in the anodic and cathodic zones. Again, as it
was mentioned in part 3.3.1, if the anodic dissolution would generate only zinc cations Zn2+,
then the pH in the anodic zones could not decrease to values lower than 5.0. However, the pH
values detected by SIET in the anodic zone reached 4.6, suggesting that Fe3+ ions were also
generated simultaneously with Zn2+ during the corrosion reactions. After approximately one
day of immersion the current is lower than that observed for the reference sample M0 and the
pH values in the anodic zone range between 4.5 and 4.1. Anodic and cathodic current densities
are in the range 40 to -30 μA∙cm-2. Comparatively to the reference sample, there is a slight
decrease of the anodic current density and a narrower pH window. The cathodic areas showed
pH values around 7, much lower than these observed for the reference sample (around 9).
SEM/EDS study was performed right after the SVET/SIET experiments, in the upper part
of the scanned area as limited by the rectangle in Fig. 12c. The results clearly demonstrate the
presence of traces of iron precipitated around the defect. This precipitates may account for
mixed zinc-iron oxides as shown in Fig. 9b. Furthermore, a clear absence of Si was detected over
the defect, Fig. 13 c, but its distribution is nearly uniform around the defect, suggesting that the
primer organic matrix was not delaminated.
Fig. 13 shows the results of SEM/EDS observations of a bigger area around the defect. The
EDS maps were taken for several elements in order to assess the composition of the precipitates
covering the defect.
It can be seen that the defect is mainly covered by Zn-containing corrosion products. Also
significant amounts of P and O are detected, while Cl does not seem to contribute to formation
Page 12 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
11
of corrosion products in the defect. The phosphorous comes from the halloysites, as phosphate
salts are used to separate the halloysite nanotubes from the bulk clay. Accumulation of O, P and
Zn over the defect suggests formation of insoluble corrosion products, like Zn3(PO4)2 which can
be formed at pH as low as 4 [36]. The Si distribution around the defect is nearly uniform. In
contrast to the results obtained for the blank sample, it is evident that the coating is still intact
around the defect and that the corrosion process did not spread laterally as observed for the
blank sample. Although the acidification in the anodic zone does not attain the values observed
for the blank sample, it is still present for the M1 sample. In this case acidification front,
extending from the centre of the defect, is not as strong as for example in M0, delaying
dissolution of the zinc and therefore coating degradation. On the other hand, the halloysites
introduced in M1 are loaded with 2-mercaptobenzothiazole which may contribute for the
narrower pH variations, suggesting a decrease of the corrosion activity. MBTA has been
reported as an effective corrosion inhibitor for copper, aluminum, zinc and iron [40, 41]. The
inhibition efficiency is likely to be the consequence of inhibitor’s adsorption on the metallic
surface and formation of complexes with the metal cations.
The M4 sample was also studied using the same approach: quasi-simultaneous SVET-SIET
measurements followed by SEM/EDS examination of the sample after immersion test. The
visual appearance of the defect and area around it at the beginning of immersion tests is shown
in Fig. 14 1a. Optical micrograph of the same sample after 46 hours of immersion is shown in
Fig.14 1b.
Figs. 14 2a-5a show the SIET maps and Figs.14 2b-5b depict the SVET maps. The pH maps
are in a good agreement with the distribution of ionic currents. In general, the corrosion activity
in the M4 sample remained relatively weak throughout 46 hours of immersion. No corrosion
activity was registered by SVET in the defect of the M4 sample during the first 9 hours of
immersion, Fig. 14-2b and 3b. However, SIET detected the first signs of corrosion attack at the
very early stage of immersion, after 1 hour: slight alkalinization around the defect with the sign
of anodic dissolution in the defect, Fig. 14 2a and 3a. SVET registered corrosion onset after 1 day
of immersion, when a slight increase of the anodic activity was noticed, Fig. 14 4b. Further
progress of corrosion activity results in the appearance of anodic and cathodic activity as
detected by SVET and SIET (Fig. 14 5a and 5b).
The pH measured in the anodic areas ranges between 5.1 and 6.1 during the first day of
immersion, while the pH in the cathodic area remains within 6.2-7.1 during the same period of
time. These pH values can be caused by anodic dissolution and hydrolysis of zinc cations solely,
without involving iron as it was the case for M0 and M1 samples. The pH over the anodic and
cathodic areas change as the corrosion process proceeds and the products of anodic and
cathodic processes accumulate. The pH measured in the anodic zone decreases to 5.1-5.7 and
the pH of the cathodic area slightly increases to 6.5-7.1. After 2 days of immersion, the pH and
the current density remain similar to the values obtained after one day, and one order of
Page 13 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
12
magnitude below the values measured for the blank sample, revealing an important inhibiting
effect in this system.
Corrosion activity measured by SVET remained relatively low (within ±8 μA∙cm-2) for the
M4 sample during the entire immersion period.
The SEM/EDS maps obtained after the immersion test do not reveal significant changes
on the coating and over the defect, which is outlined by the dotted square in the SEM image,
Fig.15. The SEM/EDS study reveals uniform distribution of Zn, O and Si. It also reveals the
presence of titanium from the added nanocontainers modified with 8-hydroxiquinoline. Thus,
the inhibiting effect detected for this sample is a consequence of the presence of this inhibitor.
Inhibition by 8-hydroxyquinoline on zinc substrates was found to be predominantly anodic
resulting in a change of the mechanism of zinc dissolution [42]. Identical behavior has been
reported for mild steel [43]. Literature reports that the main process of inhibition in the presence
of 8-HQ seems to be the adsorption of the 8-HQ on the anodic sites, which slows down the
corrosion rate and prevents adsorption of chloride ions. Formation of complexes between 8-
hydroxyquinoline and dissolved cations is also likely to occur, which account for some mixed
inhibition effect [44, 45].
The results obtained in this work show that the containers added to the weldable model
primer, do not damage the barrier properties. The M1 and M4 systems presented improved
barrier properties comparatively to the reference system. The self-healing ability of these
nanocontainers was demonstrated by the quasi-simultaneous SVET and SIET measurements.
Correlation of local current density and pH is performed by matching SVET and SIET maps,
Figs 8, 11 and 14. A very good interdependence is observed: acidified zones correspond to
anodic currents in SVET maps and alkaline sites match cathodic currents.
Part of the corrosion activity might be hidden below the planes of SVET and SIET scans
due to formation of galvanic couples between the steel substrate and zinc coating in the defect.
Even though, it should be proportional for defects in all samples. One of the advantages of the
FIB technique is that it allows for the formation of reproducible defects. As the same primer was
used and similar defects were created, the comparative analysis of several different samples
tested under the same conditions produces reliable data.
Complementary SVET-SIET experiments allow assessing relevant information regarding
corrosion activity and self-healing potential. Thus, corrosion activity is relatively high for M0
sample (reference sample) coated with the model primer. The values of current density and pH
remain within the range -40 to +80 μA∙cm-2 and 3.5 to 9.1 respectively for 24 hours of immersion
in 0.05M NaCl solution. Low pH values are related with important iron dissolution. Presence of
corrosion products containing Fe was confirmed by the SEM/EDS analysis. Unexpected
dissolution of iron in the microdefect is an important warning that inhibiting of the zinc anode
may impair its protective properties. Thus, addition of corrosion inhibitor into the weldable
primer needs to be fully tested to optimize inhibition and dissolution of the sacrificial zinc.
Page 14 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
13
The sample M1 coated with the model primer modified by MBTA loaded halloysites
demonstrated corrosion activity with peak current densities reaching -40 to +80 μA∙cm-2.
Decrease of corrosion activity to -7 to + 32 μA∙cm-2 after 8 hours suggests a weak self-healing
effect which did not last long as corrosion intensified again to -20 to +60 μA∙cm-2 after 26 hours
of immersion.
The sample M4 coated with model primer modified by 8-hydroxyquinoline loaded in TiO2
nanoparticles exhibits the lowest corrosion activity and resists the corrosion attack for several
hours of immersion. Ionic current density did not exceed ±8 μA∙cm-2 and pH values remain in
the range 5.1 – 7.1 suggesting very weak anodic activity, which did not intensify throughout
two days of immersion. This reveals that these containers have an inhibiting effect on the
corrosion activity, and therefore induce self-healing ability in the FIB defects.
4. Conclusions
Electrochemical impedance spectroscopy revealed that the addition of containers filled
with organic inhibitors do not damage the barrier properties of weldable model primers applied
to galvanized steel substrates.
It was found that an artificial defect of 50x50 μm in a coating on HDG-steel size can be
easily studied by localized electrochemical techniques as SVET and SIET. The SVET-SIET
measurements clearly identify zones of cathodic and anodic activity, both in the maps of pH
and ionic current distributions for all three samples. The advantages of co-operative application
of SVET and micro-potentiometry measurements are emphasized. The combination of these
measurements with localized chemical analysis highlights the changes in the composition and
supports the statements made from the pH changes. On the reference sample the intense
acidification over the defect shows that Fe3+ species are formed together with Zn2+ species,
which was confirmed by the chemical EDS analysis. Thus, iron was clearly identified in the
anodic zones, when strong acidification was measured. The addition of inhibiting species
extending the service life of weldable primer led to the passivation of sacrificial Zn, with the
detrimental dissolution of Fe as a result.
The study reveals that the coatings modified with the containers filled with corrosion
inhibitors are able to induce an important corrosion inhibitive effect and that, in the case of the
M4 sample the corrosion activity was very low, indicating the self-healing ability of the
containers loaded with the corrosion inhibitor.
Summarizing, this work demonstrates the complementary use of the localized
electrochemical techniques to detect local changes both on the ionic activity and pH. This
information provides important insights into mechanism of corrosion processes and
demonstrates how to assess the self-healing ability of modified coatings over localized defects.
Page 15 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
14
Acknowledgments: The authors acknowledge financial support from projects
REDE/1509/RME/2005, PTDC/CTM/108446/2008 (FCT, Portugal), and European FP7
“MUST” NMP3-LA-2008-214261 and Marie Curie IRSES project “SISET” GA-2010-269282.
References
1. W. Feng, S.H. Patel, M-Y. Young, J.L. Zunino, M. Xanthos, Adv. Polym. Technol. 26 (2007) 1.
2. K. Ogle, S. Morel, D. Jacquet, J. Electrochem. Soc. 153 (2006) B1.
3. F. Thébault, B. Vuillemin, R. Oltra, K. Ogle, C. Allely, Electrochim. Acta 53 (2008) 5226.
4. K. Ogle, V. Baudu, L. Garrigues, , X. Philippe, J. Electrochem. Soc. 147 (2000) 3654.
5. S.V. Lamaka, M.L. Zheludkevich, K.A. Yasakau, R. Serra, S.K. Poznyak, M.G.S. Ferreira, Prog. Org. Coat. 58 (2007) 127.
6. D.G. Shchukin, S.V. Lamaka, K.A. Yasakau, M.L. Zheludkevich, H. Möhwald, M.G.S. Ferreira, J. Phys. Chem. C, 112 (2008) 958.
7. M.F. Montemor, R. Pinto, M.G.S. Ferreira, Electrochim. Acta 54 (2009) 5179.
8. D. Snihirova, S.V. Lamaka, M. Taryba, A.M. Salak, S. Kallip, M.L. Zheludkevich, M.G.S. Ferreira, M.F. Montemor, ACS Appl. Mater. Interfaces 2 (2010) 3011.
9. G.Williams, H.N. McMurray, M.J. Loveridge, Electrochim. Acta 55 (2010) 1740.
10. P.J. Kinlen, Y. Ding, D.C. Silverman, Corrosion 58 (2002) 490.
11. G. Paliwoda-Porebska, M. Stratmann, M. Rohwerder, K. Potje-Kamloth, Y. Lu, A.Z. Pich, H.J.Adler, Corros. Sci. 47 (2005) 3216.
12. M. Kendig, M. Hon, L. Warren, Prog. Org. Coat. 47 (2003) 183.
13. S.V. Lamaka, D. G. Shchukin, D.V. Andreeva, M. L. Zheludkevich, H. Möhwald, M. G.S. Ferreira, Adv. Funct. Mat. 18 (2008) 3137.
14. M.W Kendig, R.G. Buchheit, Corrosion (Houston, TX, U. S.) 59 (2003) 379.
15. W. Trabelsi, P. Cecilio, M.G.S. Ferreira, M.F. Montemor, Prog. Org. Coat. 54 (2005) 276.
Page 16 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
15
16. S.V. Lamaka, O.V. Karavai, A.C. Bastos, M.L. Zheludkevich, M.G.S. Ferreira, Electrochem. Commun. 10 (2008) 259.
17. O.V. Karavai, A.C. Bastos, M.L. Zheludkevich, M.G. Taryba, S.V. Lamaka, M.G.S.Ferreira, Electrochim. Acta 55 (2010) 5401.
18. A.C. Bastos, M.G. Taryba, O.V. Karavai, M.L. Zheludkevich, S.V. Lamaka, M.G.S.Ferreira, Electrochem. Commun. 12 (2010) 394.
19. F. Thébault, B. Vuillemin, R. Oltra, C. Allely, K. Ogle, E. Tada, ECS Transactions 11 (2008) 91.
20. S.V. Lamaka, G. Knörnschild, D.V. Snihirova, M.G. Taryba, M.L. Zheludkevich, M.G.S. Ferreira, Electrochim. Acta 55 (2009) 131.
21. C. Scheffey, Rev. Sci. Instrum. 59 (1988) 787.
22. L.F. Jaffe, R. Nuccitelli, J. Cell Biol. 63 (1974) 614.
23. H.S. Isaacs, A.J. Aldykiewicz Jr., D. Thierry, T.C. Simpson, Corrosion (Houston, TX, U.S.) 52 (1996) 163.
24. A.M. Simões, A.C. Bastos, M.G.S. Ferreira, Y. González-García, S. González, R.M. Souto, Corros. Sci. 49 (2007) 726.
25. C. Wei, A.J. Bard, G. Nagy, K. Toth, Anal. Chem. 67 (1995) 1346.
26. B.R. Horrocks, M.V. Mirkin, D.T. Pierce, A.J. Bard, G. Nagy, K. Toth, Anal. Chem. 65 (1993) 1213.
27. M. Marenzana, A.M. Shipley, P. Squitiero, J.G. Kunkel, A. Rubinacci, Bone 37 (2005) 545.
28. H.B. Ding, L.H. Hihara, J. Electrochem. Soc. 155 (2008) 226.
29. S.V. Lamaka, R. M. Souto, M.G.S. Ferreira, in: A.M. Vilas (Ed.), Microscopy: Science,
Technology, Applications and Education, Edition 2010, Formatex Research Center, Badajoz,
Spain, 2010, vol. 3, pp. 2162-2173.
30. C.J. Lin, R.G. Du, T. Nguyen, Corrosion 56 (2000) 41.
31. M.L. Zheludkevich, S.K. Poznyak, L.M. Rodrigues, D. Raps, T. Hack, L.F. Dick, T. Nunes, M.G.S. Ferreira, Corros. Sci. 52 (2010) 602.
32. I.A. Kartsonakis, I. L. Danilidis, G. S. Pappas, G. C. Kordas, J.Nanosci. Nanotech. 10 (2010) 5912.
Page 17 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
16
33. D. Fix, D.V. Andreeva, Y.M. Lvov, D.G Shchukin, H. Möhwald, Adv. Func. Mater. 19 (2009) 1720.
34. R .Wirth, Eur. J. Mineral 16 (2004) 863.
35. S.V. Lamaka, M.G. Taryba, M.F. Montemor, H.S. Isaacs, M.G.S. Ferreira, Electrochem.
Commun. 13 (2011) 20.
36. Y.Y. Lur’e, Handbook of Analytical Chemistry, 6th ed., Himiya, M., 1989.
37. E. Tada, S. Satoh, H. Kaneko, Electrochim. Acta 49 (2004) 2279.
38. I. Puigdomenech, Program MEDUSA (Make equilibrium diagrams using sophisticated algorithms), Royal Institute of Technology, Stockholm, 1999.
39. Fluka ionophore catalogue. Available at
http://www.sigmaaldrich.com/etc/medialib/docs/Sigma/Instructions/h_hydrogen.Par.0001.
File.tmp/h_hydrogen.pdf. Accessed: February 04, 2011.
40. M. Ohsawa, W. Suetaka, Corros. Sci. 19 (1979) 709.
41. M.L. Zheludkevich, K.A. Yasakau, S.K. Poznyak, M.G.S. Ferreira, Corros. Sci. 47 (2005) 3368.
42. M.S. Abdel-Aal, Z.A. Ahmed, M.S Hassan, J. Appl. Electrochem. 22 (1992) 1104.
43. M.S. Abdel-Aal, M.S. Morad, Br. Corros. J. 36 (2001) 253.
44. S. Szunerits, D.R. Walt, Anal. Chem. 74 (2002) 886.
45. S.V Lamaka, M.L. Zheludkevich, K.A. Yasakau, M.F. Montemor, M.G.S. Ferreira, Electrochim. Acta 52 (2007) 7231.
Figure Captions
Fig. 1. SEM micrograph of cross-section of reference coating M0 (a), and corresponding EDS
elemental mapping of Fe (b), Zn (c) and Si (d).
Fig. 2. EIS Bode plots for the galvanized steel coated with the weldable primer modified with
different additives. Data is presented for 2 and 48 h of immersion in 0.5M NaCl.
Fig. 3. Evolution of the weldable coating resistance (determined by fitting) during immersion in
0.5M NaCl.
Page 18 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
17
Fig. 4. 50 x 50 µm FIB defect formed in the M1 primer coating applied to galvanized steel. Top
image: top view. Bottom image: Taken under an angle of 54 °.
Fig. 5. Sample M0. EDS mapping of the defect, showing the Carbon and Oxygen (left column
images) and Zinc and Silicon (right column images).
Fig. 6. FIB defect formed in the M1 primer coating applied to galvanized steel. a) Cross-section
view; b) EDS spectrum according to the white rectangle.
Fig. 7. 50 x 50 µm FIB defect formed in the M4 primer coating applied to galvanized steel. a)
Cross-section view; b) Top view; c) EDS spectrum taken at the white circle; d) EDS spectrum
taken at the dashed circle.
Fig. 8. Sample M0. Optical micrograph of the sample after one day of immersion in 0.05 M
NaCl, showing the whitish contour of corrosion products around the defect (a); optical
micrograph indicating the data points of the SVET and SIET scans (b); pH distribution (c), map
of ionic current density (d).
Fig. 9. Modelling of ionic equilibria as a function of pH in solution containing the following
ions: a) [Zn2+]= 0.01M, [Cl- ]= 0.05M, [CO32+]= 7·10-11M; b) [Zn2+]= 0.01M, [Fe3+]= 0.0005M, [Cl-
]= 0.05M, [CO32+]= 7·10-11M. The diagrams were made using Hydra/Medusa software [38].
Fig. 10. Sample M0. a) SEM images and corresponding EDS elemental distribution of b) Si; c)
Zn; d) Cl; e) SEM image of the defect; f) EDS spectrum of the area shown in e).
Fig. 11. Sample M1. Optical micrographs of the scanned sample (1a, 2a, 3a) with the white
rectangle limiting the area of the SVET and SIET scans; pH distributions (1b, 2b, 3b); and
current density distribution (1c, 2c, 3c). The measurements were taken after 1, 8 and 26 hours of
immersion in 0.05M NaCl.
Fig. 12. Sample M1. EDS elemental distribution of a) Fe and b) Si; c) pH map with white
rectangle showing the area of the EDS mapping.
Fig. 13. Sample M1. SEM image and EDS elemental mapping of the defect and area around it.
Fig. 14. Sample M4. Optical micrograph of the sample with the FIB defect after 1 hour (1a) and
46 hours (1b) of immersion in 0.05M NaCl; pH distribution (2a, 3a, 4a, 5a) and corresponding
Page 19 of 36
Accep
ted
Man
uscr
ipt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
18
current density distribution (2b, 3b, 4b, 5b). The measurements were made after 1, 9, 21 and 46
hours of immersion. The white rectangle limits the area of the SVET and SIET scans.
Fig. 15. Sample M4. SEM image and EDS elemental mapping taken after 46 hours of immersion.
Page 20 of 36
Accep
ted
Man
uscr
ipt
Table 1. Composition of the model weldable primers containing different nanocontainers.
Nanocontainer / Inhibitor
M0 Primer (reference, without nanocontainers)
M1 Primer modified with halloysites/2-mercaptobenzothiazole (MBTA)
M2 Primer modified with LDH/vanadium oxide (VOx)
M3 Primer modified with TiO2/2-mercaptobenzothiazole (MBTA)
M4 Primer modified with TiO2/8-hydroxyquinoline (8-HQ)
Table 1
Page 21 of 36
Accep
ted
Man
uscr
ipt
Figure 1
Page 22 of 36
Accep
ted
Man
uscr
ipt
Figure 2
Page 23 of 36
Accep
ted
Man
uscr
ipt
Figure 3
Page 24 of 36
Accep
ted
Man
uscr
ipt
Figure 4
Page 25 of 36
Accep
ted
Man
uscr
ipt
Figure 5
Page 26 of 36
Accep
ted
Man
uscr
ipt
Figure 6
Page 27 of 36
Accep
ted
Man
uscr
ipt
Figure 7
Page 28 of 36
Accep
ted
Man
uscr
ipt
Figure 8
Page 29 of 36
Accep
ted
Man
uscr
ipt
Figure 9a
Page 30 of 36
Accep
ted
Man
uscr
ipt
Figure 9b
Page 31 of 36
Accep
ted
Man
uscr
ipt
Figure 10
Page 32 of 36
Accep
ted
Man
uscr
ipt
Figure 11
Page 33 of 36
Accep
ted
Man
uscr
ipt
Figure 12
Page 34 of 36
Accep
ted
Man
uscr
ipt
Figure 13
Page 35 of 36
Accep
ted
Man
uscr
ipt
Figure 14
Page 36 of 36
Accep
ted
Man
uscr
ipt
Figure 15