Downloaded from [Link] on: iul.

02, 2025

Delamination of thermally sprayed ZnAl15/organic duplex coating systems on

structural steel for offshore wind applications

Li, Ziyou; Bi, Huichao; Gunnarsson, Sverrir; Weinell, Claus Erik; Dam-Johansen, Kim

Published in:
Surface and Coatings Technology

Link to article, DOI:

10.1016/[Link].2024.130796

Publication date:
2024

Document Version
Publisher's PDF, also known as Version of record

Link back to DTU Orbit

Citation (APA):
Li, Z., Bi, H., Gunnarsson, S., Weinell, C. E., & Dam-Johansen, K. (2024). Delamination of thermally sprayed
ZnAl15/organic duplex coating systems on structural steel for offshore wind applications. Surface and Coatings
Technology, 485, Article 130796. [Link]

General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright
owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

 Users may download and print one copy of any publication from the public portal for the purpose of private study or research.
 You may not further distribute the material or use it for any profit-making activity or commercial gain
 You may freely distribute the URL identifying the publication in the public portal

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
 Surface & Coatings Technology 485 (2024) 130796

Contents lists available at ScienceDirect

Surface & Coatings Technology

journal homepage: [Link]/locate/surfcoat

Delamination of thermally sprayed ZnAl15/organic duplex coating systems

on structural steel for offshore wind applications
Ziyou Li a, Huichao Bi a, *, Sverrir Gunnarsson b, Claus Erik Weinell a, Kim Dam-Johansen a
a
CoaST, Department of Chemical and Biochemical Engineering, Technical University of Denmark (DTU), Building 229, 2800 Kgs. Lyngby, Denmark
b
Hempel A/S, Lundtoftegårdsvej 91, 2800 Kgs. Lyngby, Denmark

A R T I C L E I N F O A B S T R A C T

Keywords: Duplex coating systems incorporating both thermally sprayed metal (TSM) and organic coatings have demon­
Thermally sprayed metal strated excellent corrosion protection performance in highly corrosive offshore applications. However, the
ZnAl15 present work showed extensive delamination of organic coatings from the TSM (ZnAl15) layer during ISO 12944-
Duplex coating
9 cyclic ageing test. A positive correlation was uncovered between cathode-to-anode ratio and the extent of the
Cyclic ageing test
Corrosion
delamination, which was further verified by delamination measurement on samples with varying artificial
Offshore structures scribes. A depletion of aluminum at the tip of propagating cracks in the TSM layer was detected by scanning
electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDS), suggesting the occurrence of local
acidification driven by hydrolysis of Al3+. X-ray diffraction (XRD) analysis confirmed ZnO, Zn5(OH)8Cl2⋅H2O,
and AlO(OH) as the main corrosion products in the artificial scribe. Careful considerations and supplementary
real-field assessment were recommended when employing ISO 12944-9 cyclic ageing test for a robust anticor­
rosive performance evaluation of this type of duplex coating system.

1. Introduction coatings, resulting in an extension of service life by 1.5 to 2.3 times

[4,5]. Considerable longer time interval to the first maintenance for TSM
The escalating demand for renewable energy is consistently stimu­ duplex coating systems in an environment with very high corrosivity
lating the deployment of offshore wind turbines. The European Union (C5) was also highlighted by ISO 12944-5 [6], compared to pure paint
has published a dedicated plan that aims to have at least 60 GW of systems of similar thickness. Numerous real-field application examples
offshore wind installed by 2030, representing a 4-fold increase have shown the effective protection provided by TSM duplex coatings.
compared to the 14.6 GW capacity recorded in 2021 [1]. Offshore wind For instance, the Vilsund Bridge, which spans a strait in Denmark's
energy devices (OWEAs) face aggressive environments including high Limfjord, was coated with TSM duplex coatings in 1979, and the TSM
humidity, high salinity, and intensive UV radiation. Moreover, the layer remained in sound condition during a patch paint repair in 2013
remote and difficult-to-access nature of offshore installations imposes after 34 years of service in the splash zone [7]. Similarly, the Vindeby
additional durability requirements. OWEAs are typically designed for a wind farm, which commenced operations in 1991, was coated with a
lifespan of 20–25 years, whereas the current trend in life extension system consisting of 120 μm TSM and 475 μm of epoxy/polyurethane.
strategies aims to further prolong it to 35 years [2], which necessitates Following its decommissioning in 2017, the inspection revealed that the
the design of more sustained and reliable corrosion protection systems coatings were still in excellent condition [8].
for structural steel. Despite the favorable outcomes observed in the real applications, the
Thermally sprayed metal (TSM) duplex coatings, comprised of a base understanding of the anticorrosive properties and degradation mecha­
layer of pure zinc, pure aluminum, or ZnAl alloy along with protective nisms of the TSM duplex coating system remains limited. The majority of
organic coatings on top, have gained widespread usage for safeguarding disclosed research has concentrated on the pure TSM system. Shaw et al.
steel structures from corrosion. Among various alloy options, ZnAl15 is [9] performed laboratory electrochemical tests and 1-year field expo­
particularly favored due to its demonstrated reduced corrosion rate [3]. sure on TSM ZnAl15 coatings. The anodic polarization curves indicated
Studies have reported a synergistic effect between the TSM and organic the absence of passivation, with galvanic action being the prevailing

* Corresponding author.
E-mail address: hubi@[Link] (H. Bi).

[Link]
Received 6 February 2024; Received in revised form 26 March 2024; Accepted 12 April 2024
Available online 14 April 2024
0257-8972/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license ([Link]
Z. Li et al. Surface & Coatings Technology 485 (2024) 130796

mechanism. Upon 1-year field exposure, depletion of zinc and the for­ 2. Materials and methods
mation of voids in the zinc-rich areas were observed under both atmo­
spheric splash and immersion conditions. These observations were 2.1. Materials
attributed to the presence of porous oxide layers and a weak aluminum-
rich matrix, which facilitated the creation of continuous pathways for Carbon steel S235, size 150 mm × 75 mm × 5 mm, grit blasted to a
seawater ingress, leading to the progressive degradation of the coating. surface cleanliness of Sa 21/2 [17] was used as substrate. ZnAl15 alloy
Hideki and Seiji [10] conducted a comparative analysis of the atmo­ (85 wt% Zn and 15 wt% Al) was arc-sprayed (applied voltage 30 V,
spheric corrosion properties of thermally sprayed Zn, Al, and Zn–Al current 85 A, nitrogen as carrier gas) on the substrate in-house with a
coatings after 33 years of field exposure. The Zn–Al coating demon­ nominal thickness of 60, 80, 100, and 150 μm. The porosity of the ‘as-
strated fine corrosion performance. The researchers observed a thicker, sprayed’ TSM layer was measured and presented in Fig. S1 of Supple­
nonuniform corrosion layer with cracks in the Zn–Al coating, which mental Materials. Considering the cavities and pores on the TSM
was attributed to the selective dissolution of Zn in the alloy. The main coating, a thin layer of mist coat (or sealer) was applied prior to the
corrosion products were identified as Zn6A12(OH)16CO3⋅4H2O. application of organic coatings to prevent pinholes and provide addi­
The abovementioned investigations may indirectly support selecting tional adhesion. Specific details of the duplex coating systems can be
TSM duplex coating systems, whereas organic coating manufacturers are found in Table 1. The paint application was carried out by airless spray
still required to conduct specific laboratory performance tests to assess at 23 ◦ C, 50 % relative humidity, and the overcoat interval was followed
the characteristics of the organic coatings, such as compatibility, dura­ according to the instructions provided by individual suppliers. All
bility, and other parameters relevant to corrosion protection. One of the coated samples were kept at ambient temperature for 7 days to be
widely recognized standards for such evaluations is ISO 12944-9:2018 properly cured. To ensure data reproducibility, a minimum of three
[11], which has a reputation for better correlation to field perfor­ replicates were prepared for each coating system.
mance [12]. This standard outlines a 25-week (4200 h) cyclic ageing test In accordance with ISO 12944-92,018 A.1 guidelines, coated samples
program designed to evaluate coating samples with artificial scribes. underwent the creation of a standard artificial scribe (width = 2 mm,
The test follows a weekly routine consisting of 3-day UV/condensation length = 50 mm, down to the steel substrate) using a computer numeric
exposure, 3-day exposure to neutral salt spray, and 1-day freezing controlled (CNC) miller with a 1 mm diameter drill. In addition,
conditions. However, both observations in this study and evidence from specialized artificial scribes were generated on samples from the duplex
others [13,14] have indicated that subjecting TSM duplex coatings coating system S1 with 100 μm TSM layer. There are 8 distinct scribe
(containing ZnAl15 alloy) to the cyclic ageing test mentioned earlier patterns characterized by variations in both depth and width. 5 different
results in significant delamination of organic coatings, which starkly modifications to the cathode-to-anode ratio were achieved by adjusting
contrasts with their behavior under real-field conditions. The validity the exposed area of the TSM layer (anode) and the steel substrate
and relevance of this testing method warrant further investigation. A (cathode), as illustrated in Fig. 1. These artificial scribes were generated
similar degradation pattern on thermally sprayed aluminum (TSA) to replicate various scratch conditions encountered in real-field sce­
duplex coatings was documented by Knudsen et al. [15], where exten­ narios, to study the effects of scribe patterns (cathode-to-anode ratio) on
sive blistering and a large amount of aluminum corrosion products organic coating delamination. The scribes were cut to a depth of
appeared after 24 months of saltwater immersion. The formation of approximately 50 μm within the TSM layer (D1) and down to the steel
hydrochloric acid under the organic coating was believed to explain this substrate (D2), respectively.
phenomenon. Moreover, metallic coating systems may exhibit more
sensitivity to this cyclic ageing test than others. A recent publication 2.2. Rust creep and delamination measurement
demonstrated a 7-fold increase in accelerated factor for zinc-rich
primer/polyureathane systems compared to epoxy/polyurethane sys­ The prepared samples underwent a cyclic ageing test following ISO
tems, although it was assessed based on rust creep, and no delamination 12944-9 Annex B [13]. After 25-week ISO 12944-9 cyclic ageing test, a
was noticed during the test [16]. In this case, there is a clear lack of study scraper was employed to gently remove the paint that had delaminated
focusing specifically on the premature degradation of TSM (ZnAl15) until reaching the boundary where pristine bare metal was observed,
duplex coatings during the ISO 12944-9 cyclic ageing test. and optical images of the samples were captured. The rust creep was
The objective of this study is to explore the failure mechanisms evaluated according to ISO 4628-8 [18], and the average corrosion
behind the delamination of organic coatings observed in TSM (ZnAl15)
duplex coating systems during the ISO 12944-9 cyclic ageing test. To the Table 1
best of our knowledge, the comprehensive documentation and elucida­ Specifications of the duplex coating systems.
tion of this phenomenon remain absent, which underscores the novelty
Coatings TSM Organic coatings
and importance of this work. Additionally, it is aimed at providing some code Layer
insights into the design of more reliable accelerated corrosion testing Mist coat /μm 2nd coat /μm 3rd coat /μm
/μm
methods. To ensure the reproducibility of the findings, this study
Two-
selected three types of commercial organic coating systems designed ZnAl15,
component,
Two- Two-
System 60, 80, component, component
specifically for TSM surfaces and offshore applications, and the TSM S1 high-solids
1 100, high solids polyurethane,
layer thickness ranged from 60 μm to 150 μm. After a 25-week cyclic 150
epoxy,
epoxy, 160 60
30
ageing test, the samples were subjected to rust creep measurements
Two- Two- Two-
following the guidelines from ISO 12944-9. Additionally, various arti­ ZnAl15, component, component, component
ficial scribes were introduced to investigate their impact on the extent of System
S2 60, 80, high-solids, high solids, acrylic
2
the delamination and scanning acoustic microscopy (SAM) was utilized 100 zinc phosphate zinc phosphate polyurethane,
as a non-destructive technique for delamination inspection. Further­ epoxy, 30 epoxy, 160 60
Two-
more, cross-sections through the coatings and corrosion products were Two- Two-
ZnAl15, component
analyzed using SEM/EDS and XRD respectively. System
S3 60, 80,
component, component
acrylic
3 high-solids high solid
100 polyurethane,
epoxy, 30 epoxy, 160
60

*The duplex coating systems are labeled as ‘System-thickness of TSM’ in this

paper. For example, ‘S1–60’ represents duplex coating system 1 with 60 μm of
TSM.

2
Z. Li et al. Surface & Coatings Technology 485 (2024) 130796

Fig. 1. The illustration of the artificial scribe patterns: depths ‘D1’ or ‘D2’ of the scribe are within the TSM layer or the steel substrate respectively, and ‘a’ indicates
the width of the scribe, which varies from 1 mm to 4 mm, labeled as ‘W1’ – ‘W4’. The cathode-to-anode ratio is calculated based on the exposed area of steel and TSM
layer created by the artificial scribe.

creep M is calculated based on: reported [16,19,20]. This device can generate ultrasound focusing on
specific spots or planes to collect in-depth and time-resolved informa­
M = (C − W)/2
tion. The capability of SAM to differentiate materials is ascribed to their
where C is the average of 9 width measurements and W is the measured distinctive acoustic impedance Z, which is defined as follows:
width of the artificial scribe (W = 2.0 mm). Further beyond the rust Z=c•ρ
creep area, organic coating delamination from the TSM layer was
observed. The measurement of delamination D was conducted the same where ρ is the material density (kg/m3), c is the sound transmission
way where only C was replaced by the average width of the delamina­ velocity within the material (m/s). At the interface between two mate­
tion zone. An illustrative example showing the measured region for rust rials, the acoustic waves will (a) pass through and (b) reflect, the ratio of
creep and delamination is shown in Fig. 2. reflected amplitude over the incident amplitude is called the reflection
Moreover, to perform the real-time and quantitative delamination coefficient, as given by:
measurement during the cyclic ageing test, scanning acoustic micro­
Z2 − Z1
scopy (SAM) was employed, and its application as a nondestructive R=
Z1 + Z2
technique in detecting the delamination of organic coatings has been
where Z1 and Z2 are acoustic impedance of two materials, apparently
larger acoustic impedance difference leads to a larger R, thereby a
distinctive greyscale will appear on the image. By following the contrast,
the delaminated area could be marked. In this study, the PVA TePla 300
HD2 with a 125 MHz transducer was used, and deionized water was
used as the coupling media. The samples with certain types of artificial
scribe were withdrawn after 3 weeks of ISO 12944-9 cyclic ageing test
for SAM inspection, ensuring that the delamination has not been
affected by the side edges of the sample before the measurement, the
running corrosion products deposited on the coating surface were gently
wiped to avoid interference. C-scan on a 6.2 mm × 3.5 mm rectangular
region around the artificial scribe was conducted. The data acquisition
window was set at the organic coating/TSM interface, aiming at pre­
cisely capturing the area of delamination.

2.3. SEM-EDS analysis of coating cross-section

For the high-resolution scanning electron microscopy, the coated

samples were cut using a Secotom-6 automatic cutting machine (Struers,
Germany). Cross-sections were prepared by broad ion beam (BIB) pol­
ishing using a Gatan Ilion Advantage device. To fit the sample in the
designated holder, the metallic panels were thinned (on the uncoated
side) down to <1 mm by careful, manual grinding on SiC paper and then
glued to Ti masks with Ag-base paste on the uncoated side. Subse­
quently, samples were sputtered with Pt to improve electrical conduc­
tivity, as required to minimize charging effects during SEM imaging.
High-resolution scanning electron microscopy was carried out with a
Gemini instrument (Zeiss, Germany) and EDS analysis was performed
with an XFlash FlatQUAD detector (Brucker, Germany), with accelera­
tion voltage in the interval of 3 to 20 kV.

Fig. 2. An optical image of the rust-creep measurement from a sample with

duplex coating system S1–60 after 25 weeks of ISO 12944-9 cyclic ageing test.

3
Z. Li et al. Surface & Coatings Technology 485 (2024) 130796

2.4. XRD analysis of corrosion products 3.2. SEM and EDS analysis on delaminated TSM duplex coatings

Accumulated corrosion products were carefully collected around the To investigate the morphological and elemental aspects of the
scribe without disrupting the adjacent coatings. Powder X-ray diffrac­ delaminated region, SEM/EDS characterization was conducted on
tion tests were conducted on an Empyrean diffractometer (Malvern samples after 3 weeks of ISO 12944-9 cyclic ageing test, A microscopic
Panalytical, UK), equipped with a 1Der detector, using Cu K-α (1.5406 illustration of the delamination is shown in Fig. 4. It is noteworthy that
Å) radiation in reflection mode. Samples were measured between 3.5 the substrate exhibited little evidence of steel corrosion at this time,
and 90 2θ with 1200 s of acquisition time. while a significant accumulation of white corrosion products originating
from the ZnAl15 alloy was observed, flowing and covering the artificial
3. Results and discussion scribe.
In Fig. 5, the front part of the propagating delamination region shows
3.1. Rust creep and delamination measurement after 25-week ISO a clear contrast between the corroded and pristine areas of the TSM
12944-9 cyclic ageing test layer. A noticeable reduction in Zn and Al intensity can be observed
within the corroded region, with delamination closely following the
Fig. 3 (a) presents the measurement of rust creep on samples with a propagation of cracks and an absence of Al at the crack forefront.
standard artificial scribe following a complete 25-week ISO 12944-9 To analyze more details of elemental distribution at the crack front, a
cyclic ageing test, and the optical images of the samples are shown in higher magnification (×3000 and ×20,000) SEM/EDS was employed for
Fig. S2 of Supplemental Materials. Notably, distinct measurement re­ image acquisition as depicted in Fig. 6. The pristine part of the TSM layer
sults on duplex coating systems with 60 μm TSM were noticed among shows a wavy layered structure with bright regions interspersed with
different suppliers, which can be ascribed to the variations in formula­ thinner darker regions, and the brighter regions contain sub-
tion compositions. As per the standard requirements, M ≤ 8 mm is micrometer-sized darker islands. Some thin, short cracks appear to go
mandated for high-impact areas, while M ≤ 3 mm is required for CX through the middle of the darker elongated regions. This indicated that
applications. Based on the observations, the most severe rust creep took these cavities may have been formed by imperfect flattening of the
place on duplex coating systems with a 60 μm TSM layer, and two (S1 droplets and entrapped gas or air pockets during the spraying process.
and S2) out of three risk not meeting the CX criteria. Moreover, an The brighter matrix region is rich in Zn, while the darker regions are rich
increased thickness of the TSM layer seems to result in a significant in Al. This observation verifies the porous nature of the TSM layer, and
decrease in rust creep. Particularly, the rust creep in S2–100 and S1–150 the crack may propagate preferentially along the area between molten
did not develop outside of the artificial scribe (marked as M = 0). droplets with weaker adhesion. Notably, a distinct chlorine (Cl) signal
The measurement of the delamination of organic coatings is shown in was observed along the crack. By focusing on this specific region, it was
Fig. 3 (b). Similarly, variations among coating suppliers were more further observed that Al appeared to be depleted at the crack tip
prominent on duplex coating systems with 60 μm TSM, whereas at compared to its relatively uniform distribution in other regions, whereas
higher TSM thickness, the difference appeared less significant. Addi­ Zn can still be detected along the crack. The abundance of oxygen across
tionally, the extent of delamination greatly surpassed rust creep, which the TSM layer could be ascribed to the thermal spray process, where
indicated that delamination took place ahead of the rust formulation. outer surface of the molten droplets was oxidized.
Moreover, an inverse relationship between TSM thickness and delami­
nation was evident in systems 1 and 2, where further investigation and
discussions can be found in Section 3.4. It is essential to note that the 3.3. Corrosion products analysis on delaminated TSM duplex coatings
rust creep assessment of the duplex coatings may be compromised when
massive delamination occurs since the delamination of organic coatings The XRD spectra presented minimum variations across different
leads to direct exposure of the TSM layer to the surrounding environ­ duplex coating systems. As an example, Fig. 7 displays the peak fittings
ment, and the steel substrate becomes predominantly reliant on the of corrosion products gathered from the artificial scribe of S1–100
galvanic protection offered by the TSM layer instead of the whole duplex coated samples. Major Bragg peaks can be attributed to ZnO (JCPDS
coating system. data card No. 36-1451) and the presence of Zn5(OH)8Cl2⋅H2O (JCPDS
data card No.07-0155) was also verified. Previous studies suggested that
higher pH at cathodic sites tends to promote the formation of ZnO, while
Zn5(OH)8Cl2⋅H2O preferentially deposits at the anodic sites where
chloride ions migrate [21]. Relatively minor peaks were identified for

Fig. 3. The (a) rust-creep measurement (M value) and (b) delamination measurement (D value) of TSM duplex coatings after the 25 weeks of cyclic ageing test,
where for each system the error bar is based on results of 3 to 6 replicates.

4
Z. Li et al. Surface & Coatings Technology 485 (2024) 130796

Fig. 4. An illustrative SEM image of the delamination (next to the artificial scribe) on a sample from duplex coating system S1–100 after 4 weeks of ISO 12944-9
cyclic ageing test.

Fig. 5. A sample from duplex coating system S1–100 after 4 weeks of ISO 12944-9 cyclic ageing test: The illustration of the cross-section cut and the SEM image with
a magnification of ×200 at the delamination front. EDS mapping of Zn, Al, Cl, and O are presented respectively.

AlO(OH) [22] and α-Fe2O3 (JCPDS data card No.29–0713). Addition­ TSM within the artificial scribe.
ally, The SEM and EDS investigation was also conducted directly on the Worth mentioning that, under most natural exposures (atmospheric,
scribe, showing a good correlation with the XRD results, as shown in freshwater, or seawater), zinc carbonates compounds (ZnCO3 and
Fig. S3 of Supplemental Materials. The iron oxide was absent from the Zn5(CO3)2(OH)6) are commonly found on the surface of ZnAl15 alloy in
EDS results probably because it was covered by corrosion products from abundance [23] and they were widely reported with enhanced corrosion

5
Z. Li et al. Surface & Coatings Technology 485 (2024) 130796

Fig. 6. SEM image of the crack tip on a sample with duplex coating system S1–100 after 4 weeks of ISO 12944-9 cyclic ageing test: at a magnification of (a) × 3000
and (b) × 20,000. (c), (d) and (e) EDS mapping of this region. The white dashed line indicates the path of cracks and the white arrow shows the region with
Al depletion.

Fig. 8. SAM C-scan image of a sample from duplex coating S1–100 with arti­
ficial scribe D2W2, after 3 weeks of ISO 12944-9 cyclic ageing test.

system S1–100. The delamination boundary appeared well identifiable

in this C-scan image and it was further processed by the software
ImageJ, where the delamination extent was quantified as previously
described.
Fig. 9 presents the extent of organic coatings delamination around
the scribe after 3 weeks of ISO 12944-9 cyclic ageing test. Generally, for
samples with scribes not extending down to the steel substrate (referred
Fig. 7. X-ray diffraction spectra for corrosion products collected at the artificial to as D1), the delamination was less compared to samples with scribes
scribe of samples with duplex coating system S1–100 after 25 weeks of ISO extending down to the steel substrate (referred to as D2). Specifically,
12944-9 cyclic ageing test.
more extensive delamination can be found and it shows a correlation
with the elevated cathode-to-anode ratio. This phenomenon can be
resistance in chloride containing environment [24,25]. However, zinc explained as the larger ratio from the cathode area to the anode area,
oxide exhibited more dominance during ISO 12944-9 cyclic ageing test, which leads to greater galvanic corrosion current density. Particularly,
whereas no zinc carbonate compound was detected. This observation the accelerated dissolution of the anode occurs when the cathode area
indicates that the testing method may not truly reflect the coating significantly surpasses the anode area. For D1 samples, uniform corro­
behavior under real-field conditions. sion of the TSM occurs immediately upon exposure. The galvanic couple
could still be created between TSM and steel substrate when electrolytes
3.4. Delamination of organic coatings measured by scanning acoustic penetrate through the porous TSM layer, although the rate may be
microscopy – effects of cathode-to-anode ratio reduced considering the restricted exposure of the cathodic region and
the lower oxygen concentrations present there. The difference among D1
During the rust creep removal, a strong correlation was observed samples can be attributed to the fluctuations in TSM thickness limited by
between the extent of delamination and the thickness of the TSM layer. the scribe milling precision.
To investigate further the effect of TSM layer on organic coating
delamination, a series of cathode-to-anode ratios, with various exposed 3.5. Delamination mechanisms of organic coatings from TSM (ZnAl15)
areas of TSM layer on steel substrate were established using different layer
artificial scribe patterns. Subsequently, delamination was captured non-
destructively by SAM during the exposure. Fig. 8 illustrates an example In a corrosive environment, the introduction of an artificial scribe
of the SAM measurement conducted on a sample with duplex coating that penetrates through TSM duplex coatings, reaching the underlying

6
Z. Li et al. Surface & Coatings Technology 485 (2024) 130796

Al(OH)2+ + Cl− →Al(OH)Cl+ (2)

Al(OH)Cl+ + H2 O ↔ Al(OH)2 Cl + H+ (3)

Zn2+ + 2Cl− →ZnCl2 (4)

O2 + 4e− + 2H2 O→4OH− (5)

2Al(OH)2 Cl + 2OH− →2Al(OH)3 + 2Cl− (6)

Zn2+ + 2OH− →Zn(OH)2 →ZnO + H2 O (7)

Al(OH)3 →AlO(OH) + H2 O (8)

5ZnO + 2Cl− + 6H2 O ↔ Zn5 (OH)8 Cl2 • H2 O + 2OH− (9)

However, the abovementioned scenario may not represent the real-

field exposure conditions, where for example, a constant splash of
seawater makes it less likely to form such a confined condition and there
are various ions beyond NaCl presented in the salt spray test. These
additional ions may also affect the corrosion process. For example,
sulfates were reported with inhibition effects on the corrosion of
aluminum in NaCl solution [31]. Carbonates and bicarbonates are also
Fig. 9. The extent of organic coating delamination measured by SAM on duplex
believed to have an important role in the protection since calcareous
coating system S1–100 with different scribe patterns after 3 weeks of ISO
minerals can impede the diffusion of oxygen to the steel surface [32].
12944-9 cyclic ageing test. For each system, the error bar was based on the
measurement of 3 replicates.
The simplified composition of the testing media used for salt spray test
can be one of the causes contributing to the propagation of cracks and
subsequent delamination of organic coatings from TSM. Additionally,
TSM layer or beyond, leads to the formation of a galvanic couple be­
the rapid transition from low-temperature exposure (− 20 ◦ C) to UV test
tween the TSM layer and the steel substrate. The cathodic reaction and
(60 ◦ C) during the cyclic ageing test also induces a thermal expansion
anodic dissolution take place at the steel substrate and TSM layer
that is unlikely to occur in the real world, potentially hastening the
respectively. This is assumed to be uniform corrosion and the corrosion
delamination process. This phenomenon also raises awareness that the
rate is affected mainly by the area ratio of cathode-to-anode, where a
applicability of ISO 12944-9 for evaluating as-tested TSM/organic
thinner layer of TSM can lead to an accelerated anodic dissolution.
duplex coating systems should be revisited, and real-field tests are rec­
Considering the porous nature of TSM layer, chloride ions from the salt
ommended for a more comprehensive assessment.
spray test can constantly migrate through the porous TSM layer to the
active sites to neutralize the excessive cations generated during anodic
dissolution. The interstices formed during the arc-spray process create a 4. Conclusions
confined condition for the pitting corrosion of Al, where local acidifi­
cation and hydrogen evolution can take place, as shown in Eqs. (1) to (3) In this study, thermally sprayed ZnAl15 (TSM) duplex coatings un­
[26,27]. Under this condition, the regenerated acidic condition will lead derwent evaluation following the ISO 12944-9 cyclic ageing test. Pro­
to the breakdown of the oxide film on the Al-rich part and the propa­ nounced organic coatings delamination was observed on samples with
gation of the cracks through the TSM layer. Similar acidification lower TSM thicknesses (60 μm and 80 μm), where the discrepancy
mechanisms were proposed for the failure of thermally sprayed among organic coating suppliers was more significant than at higher
aluminum (TSA) duplex coating systems [15]. Meanwhile, the acidic TSM thicknesses. The positive correlation between the cathode-to-anode
environment can also facilitate the dissolution of Zn [28], and the ratio and the severity of delamination was verified by scanning acoustic
continuous generation of cracks gives more access to oxygen, and alle­ microscopy. High-resolution SEM/EDS and analysis revealed a prefer­
viation of pH is suspected due to the oxygen reduction reaction at ential aluminum depletion along the cracks in the TSM layer, attributed
cathodic sites. Further, during the UV/condensation part of ISO 12944-9 to accelerated aluminum dissolution prompted by localized acidifica­
cyclic ageing test, Al(OH)3 and Zn(OH)2 can undergo dehydration to tion. As the cracks propagated, more pathways for corrosion agents were
form AlO(OH) and ZnO respectively, as a relatively more stable phase, created and the integrity of the TSM layer was disrupted. The cracking
reactions are shown in Eqs. (7) and (8) [29]. The formation of and dissolution of ZnAl15 alloy ultimately led to delamination of
Zn5(OH)8Cl2⋅H2O (Eq. (9)) can also be expected due to the high chloride organic coatings. XRD analysis identified ZnO, Zn5(OH)8Cl2⋅H2O, and
content and constant wet/dry cycles during the cyclic ageing test. AlO(OH) as the main corrosion products, which was consistent with the
Consequently, cracks may follow a trajectory and are subjected to an mechanisms proposed. Supplementary real-field exposure was recom­
opening mode of deformation at the tip as described by Thouless [30]. mended for evaluating TSM duplex coatings besides the ISO 12944-9
The tensile stress from thermal expansion and contraction of organic cyclic ageing test to comprehensively reveal the corrosion protection
coatings during the exposure will also accelerate the delamination performance of the duplex coating systems.
process. Depending on the distribution of the cracks inside the TSM
layer, the delamination may start either from the organic coating/TSM CRediT authorship contribution statement
interface or within the TSM layer according to observations of this study,
as shown in Fig. S4 in Supplemental Materials. Further clarifications on Ziyou Li: Writing – original draft, Methodology, Investigation,
mechanical aspects are however beyond the scope of this paper. Formal analysis, Data curation. Huichao Bi: Project administration,
Methodology, Conceptualization, Resources, Supervision, Visualization,
Al3+ + H2 O ↔ H+ + Al(OH)2+ (1) Writing – review & editing. Sverrir Gunnarsson: Conceptualization,
Resources, Supervision, Visualization, Writing – review & editing. Claus
Erik Weinell: Conceptualization, Methodology, Supervision, Writing –

7
Z. Li et al. Surface & Coatings Technology 485 (2024) 130796

review & editing. Kim Dam-Johansen: Conceptualization, Resources, performance, Mater Today Commun 31 (2022), [Link]
mtcomm.2022.103729.
Supervision, Visualization, Writing – review & editing.
[13] A. Bjørgum, New Coatings for Corrosion Protection of Offshore Wind Structures,
Corrosion Protection for Offshore Wind, Bremen, 28–30 April, 2015: p. 20.
Declaration of competing interest [14] L.T. Døssland, Protection of Carbon Steel in Marine Environment by Use of TSZ and
Organic Coatings, Specialization project report, Department of Materials Science
and Engineering, Norwegian University of Science and Technology, 2011.
The authors declare that they have no known competing financial [15] O.O. Knudsen, T. Rogne, T. Rossland, Rapid Degradation of Painted TSA, Paper
interests or personal relationships that could have appeared to influence Presented at the CORROSION 2004, Louisiana, New Orleans, 2004.
the work reported in this paper. [16] S. Li, H. Bi, C.E. Weinell, K. Dam-Johansen, A quantitative real-time evaluation of
rust creep propagation in coating systems exposed to field testing and cyclic ageing
test, Prog Org Coat 184 (2023) 107866, [Link]
Data availability porgcoat.2023.107866.
[17] ISO 8501–1, Preparation of Steel Substrates before Application of Paints and
Related Products — Visual Assessment of Surface Cleanliness — Part 1: Rust
Data will be made available on request. Grades and Preparation Grades of Uncoated Steel Substrates and of Steel Substrates
after Overall Removal of Previous Coatings, International Organization for
Acknowledgement Standardization, Geneva, Switzerland, 2007.
[18] ISO 4628-8:2012 Paints and varnishes — Evaluation of degradation of coatings —
Designation of quantity and size of defects, and of intensity of uniform changes in
The financial support from the Hempel Foundation to CoaST (The appearance — Part 8: Assessment of degree of delamination and corrosion around
Hempel Foundation Coatings Science and Technology Centre) is greatly a scribe or other artificial defect, International Organization for Standardization,
2012.
acknowledged.
[19] I. Alig, S. Tadjbach, P. Krüger, H. Oehler, D. Lellinger, Characterization of coating
systems by scanning acoustic microscopy: Debonding, blistering and surface
Appendix A. Supplementary data topology, Prog. Org. Coat. 64 (2009) 112–119, [Link]
porgcoat.2008.09.013.
[20] H. Bi, C.E. Weinell, R. Agudo de Pablo, B. Santos Varela, S. González Carro,
Supplementary data to this article can be found online at [Link] Á. Rodríguez Ruiz, K. Dam-Johansen, Rust creep assessment—a comparison
org/10.1016/[Link].2024.130796. between a destructive method according to ISO 12944 and selected non-
destructive methods, Prog Org Coat 157 (2021), [Link]
porgcoat.2021.106293.
References [21] F. Zhu, D. Persson, D. Thierry, C. Taxen, Formation of corrosion products on open
and confined zinc surfaces exposed to periodic wet/dry conditions, Corrosion 56
[1] European Commission, Offshore renewable energy, [Link] (2000) 1256–1265, [Link]
eu/topics/renewable-energy/offshore-renewable-energy_en (accessed July 8, [22] F. Lai, X. Zhang, H. Wang, S. Hu, X. Wu, Q. Wu, Y. Huang, Z. He, Q. Li, Three-
2023). dimension hierarchical Al2O3 Nanosheets wrapped LiMn2O4 with enhanced
[2] C. Bouty, S. Schafhirt, L. Ziegler, M. Muskulus, Lifetime extension for large offshore cycling stability as cathode material for Lithium ion batteries, ACS Appl. Mater.
wind farms: is it enough to reassess fatigue for selected design positions? Energy Interfaces 8 (2016) 21656–21665, [Link]
Procedia 137 (2017) 523–530, [Link] [23] T.E. Graedel, Corrosion mechanisms for zinc exposed to the atmosphere,
[3] M. Büteführ, Zinc-aluminium-coatings as corrosion protection for steel, Mater. J. Electrochem. Soc. 136 (1989) 193C–203C, [Link]
Corros. 58 (2007) 721–726, [Link] [24] P. Volovitch, C. Allely, K. Ogle, Understanding corrosion via corrosion product
[4] J.F.H. van Eijnsbergen, Duplex systems: hot-dip galvanizing plus painting, Elsevier, characterization: I, Case study of the role of Mg alloying in Zn-Mg coating on steel,
Amsterdam (1994), [Link] Corros Sci 51 (2009) 1251–1262, [Link]
[5] O.Ø. Knudsen, H. Matre, C. Dørum, M. Gagné, Experiences with thermal spray zinc [25] D. de la Fuente, J.G. Castaño, M. Morcillo, Long-term atmospheric corrosion of
duplex coatings on road bridges, Coatings 9 (2019), [Link] zinc, Corros. Sci. 49 (2007) 1420–1436, [Link]
COATINGS9060371. corsci.2006.08.003.
[6] ISO 12944-5: 2018 Paints and Varnishes - Corrosion Protection of Steel Structures [26] S. Khireche, D. Boughrara, A. Kadri, L. Hamadou, N. Benbrahim, Corrosion
by Protective Paint Systems - Part 5: Protective Paint Systems, International mechanism of Al, Al-Zn and Al-Zn-Sn alloys in 3wt.% NaCl solution, Corros. Sci. 87
Organization for Standardization, Geneva, Switzerland, 2018. (2014) 504–516, [Link]
[7] M. Gagné, O.Ø. Knudsen, Duplex zinc coatings for steel bridges, Mater. Perform. [27] Z. Szklarska-Smialowska, Pitting corrosion of aluminum, Corros. Sci. 41 (1999)
(2021) 28–30. [Link] 1743–1767, [Link]
s/2021/11/duplex-zinc-coatings-for-steel-bridges. [28] T.N. Vu, P. Volovitch, K. Ogle, The effect of pH on the selective dissolution of Zn
[8] M. Van Leeuwen, F.E. Goodwin, When good is not good enough - synergistic effects and Al from Zn – Al coatings on steel, Corros. Sci. 67 (2013) 42–49, [Link]
of duplex coatings result in extreme durability, Asia Pacific General Galvanizing org/10.1016/[Link].2012.09.042.
Conference 2019 (2019). [29] J. Zhao, W. Wu, J. Cai, X. Cheng, Evaluating the effect of aluminum on the
[9] B.A. Shaw, L. AM, P.J. Moran, Corrosion Performance of Aluminum and corrosion resistance of the structural steels used for marine engineering, J. Mater.
Zinc–Aluminum Thermal Spray Coatings in Marine Environments, ASM Res. Technol. 18 (2022) 4181–4193, [Link]
International (1987) 246–263. [30] M.D. Thouless, Cracking and delamination of coatings, J. Vac. Sci. Technol. A 9
[10] H. Katayama, S. Kuroda, Long-term atmospheric corrosion properties of thermally (1991) 2510–2515, [Link]
sprayed Zn, Al and Zn-Al coatings exposed in a coastal area, Corros. Sci. 76 (2013) [31] K.H. Na, S. Il Pyun, Effects of SO2− 2−
4 , S2O3 and HSO4 ion additives on the pitting
−

35–41, [Link] corrosion of pure aluminium in 1 M NaCl solution at 40-70 ◦ C, J. Solid State
[11] ISO 12944-9: 2018 Paints and Varnishes - Corrosion Protection of Steel Structures Electrochem. 9 (2005) 639–645, [Link]
by Protective Paint Systems - Part 9: Protective Paint Systems and Laboratory [32] B. Syrek-Gerstenkorn, S. Paul, A.J. Davenport, Sacrificial thermally sprayed
Performance Test Methods for Offshore and Related Structures, International aluminium coatings for marine environments: a review, Coatings 10 (2020) 1–19,
Organization for Standardization, Geneva, Switzerland, 2018. [Link]
[12] O. Knudsen, A.W.B. Skilbred, A. Løken, B. Daneshian, D. Höche, Correlations
between standard accelerated tests for protective organic coatings and field