Journal of Materials Research and Technology 33 (2024) 4290–4302

Contents lists available at ScienceDirect

Journal of Materials Research and Technology

journal homepage: www.elsevier.com/locate/jmrt

Corrosion mechanism of a high corrosion-resistance Zn–Al–Mg coating in

typical extremely harsh marine and cold environments
Yuwei Liu a,c,*, Tianzhen Gu a, Miaoran Liu b, Zhenyao Wang a,c , Gongwang Cao a,c ,
Quanzhong Guo a,c , Chuan Wang a,c,**
a
Corrosion Centre, Institute of Metal Research, Chinese Academy of Sciences, Shenyang, 110016, China
b
National Key Laboratory of Environmental Adaptability for Industrial Products, China National Electric Apparatus Research Institute Co., Ltd, Guangzhou, 510663,
China
c
Liaoning Shenyang Soil and Atmosphere Corrosion of Material National Observation and Research Station, Shenyang, 110016, China

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

Keywords: The corrosion mechanism of a high corrosion-resistant Zn–Al–Mg coating exposed to typical extremely harsh
Zn-Al-Mg coating environments in Wanning and Mohe was investigated. The findings revealed that the corrosion behavior of
Low temperature atmosphere Zn–Al–Mg coating is significantly influenced by environmental conditions, exhibiting varying rates and processes
Marine atmosphere
in high-temperature marine and extremely cold urban environments, which lead to distinct corrosion patterns
Atmospheric corrosion
and pit morphologies over time. In the marine atmosphere of Wanning, characterized by high temperature and
humidity, corrosion preferentially occurs in the eutectic phase and extends towards the interior of the coating
along the eutectic phase near the zinc-rich phase. The corrosion rate on the skyward side is smaller than that on
the field-facing side, and this difference between the two sides initially decreases and later increases. However, in
the urban atmosphere of Mohe, with extremely low temperatures, the phenomenon of localized corrosion is more
pronounced. Corrosion pits tend to form in the eutectic phase near the zinc-rich phase, with their shapes
gradually transitioning from sharp to shallow. After 6 and 12 months of exposure, the corrosion rate on the
skyward side is smaller than that on the field-facing side, and after 24 months of exposure, both sides exhibit
similar corrosion rates.

1. Introduction galvanized layer, but also change its electrochemical stability. Conse­
quently, the corrosion process of ZAMc becomes more complicated
To significantly enhance the longevity of components, surface pro­ compared to galvanized steel.
tection techniques are becoming increasingly critical. Over the past two During the service process, zinc will react with pollutants (CO2, SO2,
years, researchers have continued to develop new organic coatings, Cl− ) in the surrounding atmosphere to produce a series of corrosion
metal coatings, and high-efficiency composite coatings [1–5]. Among products. These products play different roles, among which
these advancements, Zn–Al–Mg coating (ZAMc) has gained remarkable Zn5(OH)6(CO3)2⋅2H2O, NaZn4Cl(OH)6SO4⋅6H2O, Zn5(OH)8Cl2⋅H2O and
international attention due to its exceptional corrosion resistance Zn4(OH)6SO4⋅nH2O can effectively hinder the transmission of aggressive
through adding Al and Mg to galvanized coatings [6–8]. It is gradually media to the substrate surface, while ZnO, Zn(OH)2 and amorphous
being commercialized and has a wide range of application prospects in phases tend to promote corrosion [15–20]. The influence of magnesium
the automotive, photovoltaic power generation and construction in­ on corrosion process in ZAMc is mainly manifested by the higher elec­
dustries [9–11]. The addition of magnesium and aluminum elements trochemical activity of MgZn2 compared to the zinc-rich phase, making
promotes the formation of new alloy phases within the coating, such as it more susceptible to preferential corrosion. The generated Mg2+ not
zinc-rich phase, Al-rich phase, MgZn2 phase, MgZn2/Zn binary eutectic only stabilizes protective corrosion products such as Zn5(OH)6(­
phase, and MgZn2/Zn/Al ternary eutectic phase [12–14]. These newly CO3)2⋅2H2O, Zn5(OH)8Cl2⋅H2O and Zn4(OH)6SO4⋅nH2O, but also par­
formed phases not only alter the microstructure of the original ticipates in the formation of MgAlLDH (Mg6Al2(OH)16CO3⋅4H2O) or

* Corresponding author. Corrosion Centre, Institute of Metal Research, Chinese Academy of Sciences, Shenyang, 110016, China.
** Corresponding author. Corrosion Centre, Institute of Metal Research, Chinese Academy of Sciences, Shenyang, 110016, China.
E-mail addresses: ywliu12s@imr.ac.cn (Y. Liu), cwang@imr.ac.cn (C. Wang).

https://doi.org/10.1016/j.jmrt.2024.10.080
Received 22 August 2024; Received in revised form 25 September 2024; Accepted 9 October 2024
Available online 15 October 2024
2238-7854/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

ZnAl-LDH (Zn2Al(OH)6(CO3)0.5⋅xH2O), which significantly contributes both skyward and field-ward sides after exposed to both high tempera­
to corrosion resistance [21]. These protective corrosion products effec­ ture and high humidity marine atmosphere and extremely cold atmo­
tively fill pits or cracks in the coating structure after corrosion, impeding sphere. This study provides crucial data support for the development of
diffusion of corrosive media within these cracks and inhibiting oxygen marine and polar resources while also serving as an important basis for
reduction at the cathode, ultimately reducing the overall corrosion rate. enhancing performance and promoting new ZAMc products.
Aluminum also plays a significant role in forming these corrosion
products. Initially, aluminum appears as insoluble corrosion products 2. Experimental material and methods
during atmospheric exposure and later transforms into LDH form. Al3+
produced from the ternary eutectic phase can stabilize compounds like 2.1. Material
Zn5(OH)8Cl2⋅H2O and Zn4SO4(OH)6⋅nH2O, preventing their trans­
formation into ZnO by inhibiting pH rise. Additionally, aluminum par­ The material used in this study is the low aluminum ZAMc plates
ticipates in forming MgAl-LDH (Mg6Al2(OH)16CO3⋅4H2O) or ZnAl-LDH provided by HBIS, and the substrate material is cold-rolled carbon steel
(Zn2Al(OH)6(CO3)1/2⋅xH2O), thereby contributing to effective corro­ sheet (SPCC). The mass of the coating is 275 g m− 2, with a thickness of
sion protection [22,23]. approximately 20 μm. The main chemical composition shown in Table 1.
The marine environment is inherently complex and dynamic, char­ The coating is composed of zinc-rich phase, Zn–MgZn2 binary eutectic
acterized by high salinity, humidity, intense corrosion, and relentless phase and Zn–Al–MgZn2 ternary eutectic phase [33]. All the sample
ultraviolet radiation. These factors pose formidable challenges to the surfaces were cleaned with alcohol and acetone, and then trimmed on all
corrosion resistance of metallic materials, necessitating robust protec­ sides before outdoor exposure. Four parallel samples, each with a size of
tive measures. Traditional anti-corrosion materials, like galvanized 150 mm × 70 mm × 0.85 mm, were prepared for each exposure period.
sheets and aluminized zinc plates, frequently struggle to sustain their Three of these samples were used for corrosion loss analysis, and one
protective properties over extended periods in such a hostile marine was used for corrosion product characterization.
setting. In the realm of ocean engineering, critical structures such as pile The method for removing corrosion products after exposure is as
foundations and beam plates are continually threatened by the corro­ follows: according to ISO 8407, remove corrosion products with a soft
sion. The need for materials that can withstand this corrosive environ­ brush firstly. Then place the sample in a saturated glycine solution at
ment is paramount. Similarly, the polar region’s environment, with its room temperature for 1–10 min to remove the remaining corrosion
extreme low temperatures, high humidity, fierce winds, and pervasive products. Finally, rinse the sample with deionized water, dehydrate it,
salt spray, imposes even stricter demands on the corrosion resistance of and dry it with alcohol. After 24 h of storage in the drying oven, the
metallic materials. Structural components and peripheral installations of sample mass was determined by an analytical balance.
polar research stations are confronted with severe corrosion issues that
threaten their integrity and functionality. Therefore, there is a pressing 2.2. Exposure test
need for innovative anti-corrosion solutions that can effectively shield
metallic structures in both marine and cold environments, ensuring their The exposure sites are Wanning (110◦ 30′E 18◦ 58′N), with a typical
longevity and reliability. ZAMc is renowned for its exceptional corrosion high temperature and high humidity marine atmosphere, and Mohe
resistance and comprehensive performance, making it a potential pro­ (122◦ 23′E, 53◦ 0′N), with an extremely low temperature atmosphere,
tective measure to mitigate significant issues caused by corrosion- respectively. The main environmental parameters of the two exposure
induced failures of components in these extremely atmospheric envi­ sites are shown in Table 2. The exposure period was from September
ronments. It can effectively enhance the service life and reliability of 2020 to September 2022, and the corroded samples were retrieved after
these components. 6, 12 and 24 months. All samples are fixed on the exposure rack at a 45◦
As known, the service life of ZAMc depends on its corrosion resis­ angle from the horizontal direction, facing south.
tance and is closely related to its evolution process under specific at­
mospheric conditions [24–27]. Infrastructure construction and high-end 2.3. Characterization of the corrosion products
equipment manufacturing are fundamental for scientific understanding,
rational exploitation, and utilization of marine and polar resources, An ESEM XL30 FEG scanning electron microscope (SEM) and energy
which often face harsh climates necessitating materials with high levels dispersive spectrometer (EDS) were used to determine the surface and
of corrosion resistance during resource development. Moreover, current cross-sectional morphologies, as well as the element distribution, of the
research by some scholars has investigated the corrosion characteristics corrosion products. The corroded samples for cross-section investigation
of different components of ZAMc in typical atmospheric environments. were embedded in epoxy resin, wet-ground to 2000 grit and polished
Dier et al. [28] conducted research on characterizing the corrosion with 1.5 μm diamond paste. X-ray diffractometer (XRD, XPERT-PRO)
products of ZAMc exposed to a marine atmosphere with a corrosion with a Cu target, and X-ray photoelectron spectrometer (XPS, ESCA­
category C3 for 6 months. Tomandl et al. [29] studied the evolution of LAB 250) with Al target were used to analyze the composition of the
corrosion morphology of ZAMc exposed to highly corrosive marine at­ corrosion products. For the XRD test, the test voltage and current were
mospheres with a corrosion category C5~CX. However, these researches 40 kV and 40 mA, respectively. The test range was 4◦ –90◦ at a scanning
solely focus on the corrosion behavior from a single aspect, neglecting rate of 4◦ /min. The phases were identified using the JCPDS database.
the combination of morphology and composition evolution of corrosion For the XPS test, all peaks were calibrated using standard carbon
products. Although Thierry et al. [30] examined the long-term exposure contamination (284.6 eV). The data analysis was performed using XPS
effects on corrosion rate, morphology and composition evolution of PEAK 4.1 software.
ZAMc in various environments, yet the electrochemical evolution pro­
cess remains unclear, necessitating further investigation. Given the 2.4. Electrochemical analysis
above, and considering recent findings that atmospheric corrosion in
extremely cold areas can be severe [31,32], further investigation into Electrochemical impedance spectroscopy (EIS) and potentiodynamic
the atmospheric corrosion behavior of ZAMc in such conditions is
essential to promote its application potentiality in these areas. Table 1
In order to understand the evolution mechanism of corrosion prod­ Chemical composition of ZAMc.
ucts formed on ZAMc in harsh atmospheric environments, we conduct a Elements Mg Al Si P Zn
series of studies examining corrosion rates, morphology, structure,
Mass fraction/% 4.4 6.8 4.8 0.5 Bal.
composition and electrochemical properties of corrosion products on

4291
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

Table 2
The environmental parameters of Wanning and Mohe.
Exposure Classification Continuous Continuous Relative humidity Temperature
site Sulfate precipitation rate Sea salt particle precipitation rate (%) (◦ C)
(mg⋅m− 2⋅d− 1) (mg⋅m− 2⋅d− 1)

Wanning Marine 44.53 652 86 24.6

Mohe Extremely low 0.39 – 66 1.57
temperature

polarization (PDP) and were measured using the Gamry Reference 600 which was approximately 2.14 times higher than that observed in first 6
electrochemical workstation. The corrosion sample is treated with Cu months (7.96 g m− 2⋅a− 1). Moreover, there was no significant difference
wire, raw material strip and rosin paraffin to ensure a test area of 1 cm2. between first-year (12.50 g m− 2⋅a− 1) and second-year (11.63 g m− 2⋅a− 1)
Three electrode system is selected (a platinum plate worked as counter average values. In contrast, when subjected to Mohe atmospheric
electrode, a saturated calomel electrode as reference electrode, and environment, there exists an initial decrease followed by an increase in
corroded sample as working electrode). In order to ensure the stability of ZAMc’s corrosion rate over time.
the test system and keep the corrosion system similar to the outdoor
atmospheric environment, 0.1 mol/L NaCl solution and 0.1 mol/L
Na2SO4 solution are chosen to test the corroded samples after exposure 3.2. Morphology of corrosion products
in Wanning and Mohe, respectively. Before the EIS and PDP tests, the
open circuit potential (OCP) was measured for 10 min to obtain a stable In order to investigate the factors contributing to the variation in
system. The frequency of EIS was 105 - 10− 2 Hz with a 10 mV amplitude corrosion rate of ZAMc in these two atmospheric environments, this
signal. The scanning rate of PDP measurement is 0.33 mV/s, and the study conducted a comprehensive analysis of the surface and cross-
scanning range is − 0.25V (vs. OCP) ~ 0.3V (vs. OCP). To ensure the sectional morphologies of ZAMc on both skyward and field-ward sides.
reliability of test data, three parallel samples were prepared for each
group. 3.2.1. Microstructure of ZAMc after exposure to Wanning atmosphere
Fig. 2 illustrates the surface morphologies of ZAMc after different
3. Results exposure durations in Wanning. It is evident that the corrosion products
on the skyward side are significantly more abundant compared to those
3.1. Corrosion rate on the field-ward side, indicating a higher degree of corrosion on the
skyward side. Specifically, after 6 months’ exposure (Fig. 2a), localized
After conducting an outdoor exposure test, corrosion weight loss is clusters and accumulation of surface corrosion products were observed
the most intuitive and accurate analysis method for evaluating mate­ on the skyward side. Upon local magnification, it was found that these
rials’ corrosion resistance and environmental corrosion category. Fig. 1 corrosion products exhibited loose structures with visible cracks. After
illustrates the corrosion weight loss and corrosion rate of ZAMc exposed 12 months’ exposure (Fig. 2b), there was a more uniform distribution of
to atmospheric environment of Wanning and Mohe at different time. As corrosion products across the surface, accompanied by an increased
depicted in Fig. 1a, the weight loss of ZAMc gradually increases with number of cracks and denser formation of corrosion products, as
prolonged exposure time in both areas. The corresponding corrosion revealed by local magnification analysis. Furthermore, after 24 months’
rate (Vn) is calculated using Eq. (1), exposure (Fig. 2c), there was a continuous increase in density and the
quantity of corrosion products compared to those observed at 6 months.
Wn - Wn-1
Vn = × 12 (1) The local magnification also indicated longer cracks within these
tn - tn-1 accumulated corrosion products. On the other hand, for the field-ward
side, only localized areas exhibited light corrosion after 6 months’
Where Vn represents corrosion rate measured in g⋅m− 2⋅a− 1, t represents
exposure (Fig. 2d). Subsequently, after 12 months’ exposure (Fig. 2e),
the exposure time measured in months (m), W represents weight loss per
spot-like accumulations of corrosion products were observed along with
unit area measured in g⋅m− 2, and n represents exposure period. The
some generated cracks within them. Finally, after 24 months’ exposure
variation of ZAMc’s corrosion rates over time after exposure to atmo­
(Fig. 2f), similar surface morphology characteristics were noted for
spheric conditions is presented in Fig. 1b. The results demonstrate that
corrosion products as seen previously on the skyward side at 12 months.
initially there is an increase followed by a decrease in ZAMc’s corrosion
According to the cross-sectional morphology depicted in Fig. 3, it is
rate after exposed in Wanning atmosphere. Specifically, from June to
evident that there is a significant accumulation of corrosion products on
December, the maximum value of corrosion rate was 17.03 g m− 2⋅a− 1
the skyward side. Initially, after 6 months’ exposure, the corrosion

Fig. 1. Mass loss (a) and corrosion rate (b) of Zn–Al–Mg in Wanning and Mohe at different exposure time.

4292
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

Fig. 2. Surface morphologies of ZAMc in the Wanning at different time (ãc) of skyward sides, (d ~ f) field-ward sides.

Fig. 3. Cross–sectional morphologies of Zn–Al–Mg coating and corrosion products in the Wanning at different time (ãc) skyward sides, (d ~ f) field-ward sides.

products exhibit a loose structure with cracks distributed throughout the months’ exposure, a thin layer of corrosion products covers the surface
corrosion layer (Fig. 3a). After 12 months’ exposure, the surface of the coating, while a significant amount accumulates internally in the
corrosion products became denser, but notable cracks emerge at the eutectic near the zinc-rich phase, and exhibiting a denser structure
interface between the corrosion products and the coating, suggestive of (Fig. 3c). Concerning the field-ward side, a small amount of corrosion
a weakening binding force. Additionally, corrosion has penetrated products is observed near the eutectic of the zinc-rich phase, which
deeper into the coating, with the eutectic phase near the zinc-rich phase exhibits cracks after 6 months’ exposure (Fig. 3d). With increasing
experiencing more severe corrosion as shown in Fig. 3b. After 24 exposure, selective corrosion becomes apparent within the eutectic

4293
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

interaction, resulting in a thin layer of corrosion products on the coating corrosion products (Fig. 5d). After 24 months of exposure, the corrosion
surface (Fig. 3e). The corrosion products on the coating surface gradu­ became more pronounced, with an increase in the number and width of
ally thicken, and cracks appear at the junction between the corrosion cracks on both sides. Notably, the length of the crack on the field-ward
product layer and the coating as shown in Fig. 3f, resembling the side did not change significantly, and they remained shorter than those
morphology observed on the skyward side after 12 months’ exposure. on the skyward side. Additionally, the cracks that formed on the field-
This indicates a certain degree of orientation is occurring within zinc- ward side were also located within the eutectic phase (Fig. 5f).
rich interaction. Consistent with previous literature [34,35], it is From the cross-sectional morphology depicted in Fig. 6, the evolu­
well-established that zinc preferentially corrodes at grain boundaries. tion of pits and cracks can be further analyzed. After 6 months of
Therefore, it is reasonable to infer that a certain degree of intergranular exposure, the surface of the coating exhibited signs of unevenness, with
corrosion is occurring in the zinc-rich interaction. corrosion preferentially occurring within the eutectic phase. On the
The atmospheric corrosive medium in Wanning primarily consists of skyward side, the corrosion products were dense and measured
sea salt particles, along with a minor quantity of sulfate, as detailed in approximately 0.25 μm (Fig. 6a). Meanwhile, on the field-ward side,
Table 2. To gain a deeper understanding of the corrosion process, EDS holes began to appear in the eutectic phase, particularly in areas sur­
analysis was conducted on corrosion products after 24 months’ expo­ rounding the zinc-rich phase (Fig. 6d). After 12 months and 24 months
sure, as depicted in Fig. 4. On the skyward side, the corrosion products of exposure, a distinct pattern of corrosion pits became evident on both
exhibit a distinct layered structure, with a dense inner layer and a looser sides. These pits exhibited a conical shape, with a wider opening at the
outer layer. At point A there is a notable enrichment of sulfur(S), sug­ top and narrowing towards the bottom. This morphology suggests a
gesting its presence primarily in the outer corrosion layer. Conversely, at progressive deepening of the corrosion process over time. The presence
points B1 and B2, there is a significant accumulation of chlorine (Cl), of these conical pits indicates that the corrosion is not only occurring on
indicating its tendency to migrate longitudinally within the coating and the surface but is also penetrating deeper into the material. The pref­
infiltrate its interior, thereby initiating corrosion. These observations erential corrosion in the eutectic phase can be attributed to its chemical
imply that Cl is more prone to penetrating the coating and causing composition and microstructure, which may make it more susceptible to
corrosion, while S is more likely to remain in the outer corrosion attack by corrosive agents in the atmosphere. The holes and pits
products. On the field-ward side, the distribution of Cl and S elements observed in the eutectic phase likely result from the dissolution of
follows a similar pattern to the skyward side. Although the cross- weaker components or the formation of galvanic cells between different
sectional morphology does not reveal obvious corrosion at point C phases. The dense corrosion products on the skyward side may be due to
within the eutectic phase, the EDS spectrum reveals a uniform distri­ the accumulation of corrosion by-products or the formation of protective
bution of oxygen (O), indicating the occurrence of corrosion. Further­ oxide layers, which can affect the rate and nature of further corrosion.
more, the S element appears to cause corrosion primarily on the surface To gain a deeper understanding of the corrosion pit evolution, the
of the coating and does not easily penetrate its interior. In contrast, Cl analysis of corrosion pits’ width and depth, as presented in Fig. 7, re­
accumulates predominantly at the interface between the eutectic phase veals several key insights. Firstly, it is evident that both the depth and
and the zinc-rich phase, resulting in more severe corrosion. This suggests width of the corrosion pits increase with corrosion time. Notably, the
that Cl has a stronger propensity to migrate longitudinally within the depth of the corrosion pits on skyward side is comparably shallower
coating, exhibiting characteristics that can induce crevice corrosion. than that on field-ward side, whereas the width on skyward side is
broader. Secondly, a comparison of the depth-to-width ratio β (D/W) of
3.2.2. Microstructure of ZAMc after exposure to Mohe atmosphere the pits on skyward side after 12 months and 24 months of exposure
In contrast to the corrosion observed in the atmospheric conditions indicates a decrease over time. This ratio is also consistently lower on
of Wanning, the corrosion of ZAMc in the atmosphere of Mohe is rela­ the skyward side compared to the field-ward side. According to existing
tively mild after various exposure durations, as depicted in Fig. 5. After 6 studies, a β(D/W) value below 0.7 signifies a shallow pit, whereas a
months of exposure, there were no visible corrosion products detected value exceeding 0.7 characterizes a sharp pit [36]. Therefore, it is
on either the skyward or field-ward sides. However, partial detachment apparent that the prolongation of the corrosion time from 12 to 24
of the oxide film was observed on the skyward side (Fig. 5a), while months, the corrosion pits transition from a sharp to a shallow form. The
cracks began to appear in the eutectic phase on the field-ward side change rate of depth or width, denoted as α, is calculated using Eq. (2).
(Fig. 5d). After 12 months of exposure, multiple cracks were found near This equation takes into account the depth or width measurements after
the corrosion products on both sides. On the skyward side, cracks were 12 months (L1) and 24 months (L2) of exposure, respectively. The re­
longer but narrower in width (Fig. 5c). Conversely, on the field-ward sults of these calculations are summarized in Table 3, providing a
side, the cracks were shorter but wider, often spreading along the quantitative assessment of the corrosion pit’s evolution over time.

Fig. 4. Elements distribution (EDS map) of Zn–Al–Mg in the Wanning at 24 months: (a) skyward side, (b) field-ward side.

4294
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

Fig. 5. Surface morphologies of Zn–Al–Mg in the Mohe at different time (ãc) skyward sides, (d ~ f) field-ward sides.

Fig. 6. Cross–sectional morphologies of Zn–Al–Mg coating and corrosion products in Mohe extremely cold atmosphere at different time (ãc) skyward sides, (d ~ f)
field-ward sides.

L2 -L1 corrosion due to potential differences between various phases. Obser­

α= (2)
L1 vations reveal that sharp, conical corrosion pits predominantly form
within the eutectic phase, proximal to the zinc-rich phase. The charac­
The corrosion pits exhibited differential growth patterns based on
teristic shape of these pits, wider at the top and narrower at the bottom,
exposure time and location. Specifically, on both exposure sides, the pits
is attributed to the preferential corrosion along the zinc-rich phase. The
expanded less significantly in the depth direction as the exposure time
corrosion process is hindered along the eutectic phase, away from the
increased. Instead, they demonstrated a greater tendency towards
zinc-rich phase, resulting in this asymmetric pit growth. With the
lateral expansion. Notably, the width of the corrosion pit after 24
extension of exposure time, the corrosion pit tends to undergo lateral
months of exposure increased by over 200% compared to its width after
development, evolving from a sharp to a shallow form. This trend can be
12 months of exposure, indicating a substantial lateral growth over time.
attributed to the unique environmental conditions in Mohe, character­
The corrosion behavior of ZAMc exhibits distinct characteristics,
ized by an extremely cold atmosphere that is relatively free of Cl− ,
particularly in relation to the eutectic phase. This phase, especially when
which, with its small ionic radius, are known to facilitate pitting
located in proximity to the zinc-rich phase, is highly susceptible to local

4295
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

elements (S, Cl), the corrosion products can be categorized into three
groups: those containing only Cl (Zn5(OH)8Cl2⋅H2O), only S (including
Zn4(SO4)(OH)6⋅xH2O, Zn4(SO4)(OH)6⋅4H2O and Zn4(SO4)(OH)6⋅H2O),
Na6Zn4(SO4)4⋅2H2O and MgSO4⋅5H2O), and those containing both S and
Cl (NaZn4(SO4)(OH)6Cl⋅6H2O and Zn12(SO4)3Cl3(OH)15⋅xH2O). The
evolution of corrosion products within the first 12 months is particularly
rapid. On the skyward side, corrosion products primarily consist of those
containing S or both S and Cl. Conversely, on the field-ward side, the
corrosion products are mainly composed of those containing Cl or both S
and Cl. However, after 24 months of exposure, the primary corrosion
products on both sides transform into Zn5(OH)8Cl2⋅H2O and NaZn4(SO4)
(OH)6Cl⋅6H2O. Previous studies have demonstrated that the composi­
tion of zinc-containing corrosion products is influenced by the molar
Fig. 7. Depth and width of corrosion pits formed on Zn–Al–Mg in Mohe
deposition rate ratio of Cl and S [35]. Therefore, the observed differ­
extremely cold atmosphere.
ences in corrosion product composition between the two exposure sides
may be attributed to variations in the deposition rates of Cl and S on the
Table 3 surface.
The change rate of depth or width α of the corrosion pits. The XRD analysis of the corrosion products formed on ZAMc
following exposure to the atmosphere of Mohe for different durations is
Depth Width
presented in Fig. 9. The corrosion degree of ZAMc in this environment is
Skyward sides 39% 212%
relatively mild, resulting in fewer corrosion products. It is possible that
Field-ward sides 14% 218%
the corrosion products are easily dissolved or fail to adequately cover
the surface of the coating during the exposure process. After 6 months of
corrosion. In the absence of these ions, the corrosion process is less exposure, almost no corrosion products were detectable. However, after
aggressive. Additionally, the pit’s geometry, wider at the top, hinders 12 months, a trace amount of corrosion products was identified. Spe­
the penetration of larger molecules like SO2 and O2 to the pit’s bottom. cifically, the skyward side was primarily consisted of Zn4(SO4)
This prevents deepening of the pit and instead promotes lateral (OH)6⋅4H2O, while the field-ward side was primarily composed of
expansion. ZnSO4. After 24 months of exposure, the skyward side exhibited the
formation of Al2O3 and AlOOH. The detailed information regarding the
corrosion products of ZAMc after 6 months of exposure to the Mohe
3.3. Analysis of corrosion product components atmosphere, analyzed by XPS, has been reported in our previous
research. According to those findings [37], Zn5(CO3)2(OH)6 and sulfate
The atmospheric conditions in Wanning significantly impact the were identified as the primary corrosion products.
corrosion behavior of ZAMc, resulting in more severe corrosion and the Overall, the corrosion behavior of ZAMc in the Mohe atmosphere
formation of numerous corrosion products. Analysis of the corrosion appears to be slower compared to Wanning. This could be due to dif­
products’ morphology and element distribution reveals distinct pat­ ferences in temperature, humidity, and contaminant levels between the
terns. The composition of these corrosion products, as shown in Fig. 8 by two locations. The specific atmospheric conditions in Mohe may favor
XRD patterns, is diverse and exhibits notable differences between the the formation of fewer or less stable corrosion products, or the corrosion
skyward and field-ward exposure sides, even after the same exposure products may be more soluble and thus less likely to accumulate on the
time. Notably, after 12 months of exposure, the field-ward side exclu­ surface of the coating.
sively exhibited the presence of Mg-containing corrosion product,
MgSO4⋅5H2O. In contrast, the majority of the corrosion products on both
sides contained zinc. Based on the presence of different corrosive

Fig. 8. XRD patterns of Zn–Al–Mg in the Wanning marine atmosphere after (a) 6 months, (b) 12 months, (c) 24 months.

4296
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

Fig. 9. XRD patterns of Zn–Al–Mg in Mohe after (a) 6 months, (b) 12 months, (c) 24 months.

3.4. Electrochemical analysis

3.4.1. Potentiodynamic polarization

The atmospheric corrosion of ZAMc, as an essentially electro­
chemical process occurring under a thin liquid film, exhibits distinct
behaviors when exposed to different environments. Figs. 10 and 11
present the polarization curves of ZAMc exposed to Wanning and Mohe,
respectively, after varying durations. In Wanning, after 6 and 12 months
of exposure, it is evident that both the anode and cathode polarization
currents of the field-ward sides are significantly larger than those of the
skyward sides. This suggests that the corrosion rate on the field-ward
side is higher. After 24 months, while the anode polarization current
on the field-ward side remains slightly larger, there is a noticeable
reduction in the difference between the two sides, indicating a gradual
convergence in corrosion rates.
On the other hand, in Mohe as shown in Fig. 11, the anode polari­
zation curves of both sides show a high degree of overlap, indicating
similar corrosion rates on both the skyward and field-ward sides. Fig. 11. Potentiodynamic curves of Zn–Al–Mg after being exposed to Mohe for
However, after 6 and 12 months of exposure, the cathode polarization (a) 6 months, (b) 12 months, (c) 24 months.
currents of the skyward side are observed to be smaller than those of the
field-ward side. This difference suggests that the rate of oxygen reduc­ indicating that the corrosion behavior on both sides has converged.
tion, the primary cathodic reaction in the electrochemical corrosion The Tafel extrapolation method was used to analyze the polarization
process of ZAMc, is slower on the skyward side. After 24 months, the curves obtained from the Wanning and Mohe exposures. The results,
anode and cathode polarization curves of both sides nearly overlap, summarized in Table 4, provide valuable insights into the corrosion
behavior of the samples under different environmental conditions. For
the samples exposed to Wanning, it is evident that the corrosion current
densities of the field-ward sides are higher than those of the skyward
sides. This suggests that the field-ward sides experience a higher rate of
corrosion. Furthermore, the corrosion current densities initially increase
and then decrease with exposure time, indicating a fluctuating corrosion

Table 4
The fitting parameters of polarization curves.
Wanning Mohe
2 2
Ecorr/V icorr/μA⋅cm− Ecorr/V icorr/μA⋅cm−

6m-s − 0.97 0.50 − 1.10 0.251

6m-f − 1.01 1.40 − 1.04 17
6m-av − 0.99 0.95 − 1.07 8.626
12m-s − 0.98 1.42 − 1.09 0.1
12m-f − 1.03 7.18 − 1.02 12.6
12m-av − 1.005 4.30 − 1.055 6.35
24m-s − 1.01 0.15 1.08 77.8
24m-f 1.01 0.23 1.06 78.1
Fig. 10. Potentiodynamic curves of Zn–Al–Mg after being exposed to Wanning
−
24m-av − 1.01 0.19 1.07 77.95
for (a) 6 months, (b) 12 months, (c) 24 months.

4297
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

rate on both sides. Combined with the cross-sectional morphologies in low-frequency region reflects the Faraday reaction process, which
Fig. 6, it can be seen that with the progress of corrosion, MgZn2 in the further confirms that the corrosion resistance of the corrosion products
eutectic phase structure of the coating is easy to dissolve preferentially on field-ward side after 24 months of exposure is relatively poor.
as the anode [23], and the proportion of exposed surface zinc-rich phase The equivalent circuit shown Fig. 13 was used to fit the EIS results.
increases, which may lead to a positive potential shift. Secondly, the As shown, Rs represents the solution resistance, while Rr and Qr repre­
increase in local corrosion pits on the surface of the coating may also sent the resistance and capacitance of the corrosion product film,
increase the tendency towards non-uniform corrosion. respectively. Additionally, Rct and Qdl signify the charge transfer resis­
The average corrosion current densities (-av) also follow a similar tance and double layer capacitance, respectively. The polarization
trend, initially increasing and then decreasing. This trend indicates that resistance, Rp was defined as the sum of Rct and Rr, reflecting the
the overall corrosion rate of the sample, considering both exposure corrosion resistance imparted by the corrosion products. The changes in
sides, follows a pattern of initial acceleration followed by deceleration. Rp over time, as obtained from the fitting results, are depicted in Fig. 14.
This finding is in agreement with the results obtained from weight loss It can be seen that the Rp of both sides initially decreased and then
calculations, providing further confirmation of the observed corrosion increased with the extension of exposure time. This indicates that the
behavior. On the other hand, the samples exposed to Mohe exhibit a corrosion resistance of both sides initially decreased and subsequently
different corrosion behavior. After 6 and 12 months of exposure, the increased, resulting in a corrosion rate that first accelerated and then
corrosion current densities of the field-ward sides are significantly decelerated. This trend was consistent with the corrosion rate trends
higher than those of the skyward sides, differing by two orders of reflected by PDP and corrosion weight loss measurements.
magnitude. This indicates a much more severe corrosion process on the The changes of the electrochemical impedance of ZAMc after expo­
field-ward sides under Mohe conditions. However, after 24 months of sure in Mohe are presented in Fig. 15. It is evident that the radius of the
exposure, the corrosion current densities of both sides converge, sug­ capacitive arc on the skyward side is significantly larger than that on the
gesting that the difference in corrosion rates between the two sides has field-ward side, suggesting that the corrosion resistance of the skyward
diminished. The trend in corrosion current densities with exposure time side is superior to that of the field-ward side. To gain a more detailed
in Mohe is also noteworthy. Initially, the corrosion current densities mechanism of the corrosion process, the circuit diagram shown in
decrease, indicating a reduction in the corrosion rate. However, after a Fig. 13 was utilized for fitted. The changes in Rp over time, as observed
certain point, the corrosion current densities begin to increase, sug­ from the fitting results, are depicted in Fig. 16. It can be observed that
gesting that the corrosion rate is once again accelerating. This trend is the Rp of both sides initially increased and then decreased with the
reflected in the average corrosion current densities as well, which extension of exposure time, indicating that the corrosion rate initially
decrease initially and then increase. Consequently, the overall corrosion decreased and subsequently increased. This trend was consistent with
rate of the sample in Mohe follows a pattern of initial decrease, followed the corrosion rate trends reflected by PDP and corrosion weight loss
by an increase. The consistency between the trends observed in the Tafel measurements. Additionally, it is noteworthy that after 6 and 12 months
extrapolation analysis and the weight loss calculations provides strong of exposure, the corrosion resistance of the skyward side was signifi­
evidence for the reliability of the experimental results. cantly higher than that of the field-ward side. However, after 24 months
of exposure, there is little difference in corrosion resistance between two
3.4.2. Electrochemical impedance spectroscopy (EIS) exposure sides, which is consistent with the trend of potentiodynamic
In order to better understand the electrochemical corrosion proper­
ties of samples exposed to different time. Electrochemical impedance
spectra of two different exposure sides varing with time are shown in
Fig. 12. As shown in Fig. 12a and b, after 6 and 12 months of exposure,
the radius of the capacitive arc of the skyward side is significantly larger
than that of the field-ward side, indicating that the corrosion resistance
of the skyward side is superior to that of the field-ward side in these two
exposure stages [38,39]. After 24 months of exposure, the capacitive
reactance arc radius of skyward side is similar to that of field-ward side
in the high-frequency region, whereas the capacitive reactance arc
radius of skyward side is larger than that of field-ward side in the
low-frequency region. Generally, the high-frequency region of the
electrochemical impedance spectroscopy reflects the corrosion resis­
tance characteristics of the corrosion product layer, while the Fig. 13. Equivalent circuit of EIS.

Fig. 12. Nyquist diagrams of Zn–Al–Mg at Wanning after (a) 6 months, (b) 12 months, (c) 24 months.

4298
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

(OH)6Cl⋅6H2O.

Zn(OH)2 + 4Zn2+ + 6OH- + 2Cl- + H2 O→Zn5 (OH)8 Cl2 ⋅H2 O (7)

Zn(OH)2 + 4Zn2+ + 4OH- + 2CO23- →Zn5 (OH)6 (CO3 )2 (8)

3ZnO + Zn2+ + 7H2 O + SO24- →Zn4 (SO4 )(OH)6 ⋅4H2 O (9)

As depicted in Fig. 8, the corrosion products formed on the skyward

and field-ward sides of ZAMc exposed to Wanning for varying durations
exhibit notable differences, which have a direct impact on the coating’s
corrosion resistance. Specifically, Zn5(OH)8Cl2⋅H2O and NaZn4(SO4)
(OH)6Cl⋅6H2O demonstrate superior corrosion resistance, whereas ZnO
exhibits poorer resistance [20,40]. A comprehensive analysis incorpo­
rating XRD data suggests that the disparity in corrosion resistance be­
tween the two exposure sides arises from variations in the composition
Fig. 14. Variations of Rp at Wanning after different corrosion periods. of the corrosion products. On the skyward side, a greater quantity of
corrosion products with high corrosion resistance is formed, thus
polarization. This further validates the accuracy of the electrochemical explaining the observation in electrochemical analysis that the corrosion
testing methodology employed. resistance on the skyward side is superior to that on the field-ward side.
Furthermore, when combined with the cross-sectional morphologies, it
4. Discussion is evident that after 6 months of exposure, the eutectic phase and
zinc-rich phase on the surface of the coating were covered with a layer of
Given that the atmospheric environment of Wanning is characterized corrosion products that possess good electrochemical stability. Howev­
as a marine climate, featuring consistently high humidity and temper­ er, as the exposure time reached 12 months, the corrosion products
ature, with an average atmospheric relative humidity exceeding 80% could no longer effectively prevent the anodic dissolution of the eutectic
throughout the exposure period, a thin liquid film promptly adheres to phase, leading to the formation of an anodic region at the covered area
the surface of the sample upon exposure, as shown in Fig. 17. Conse­ of the product, while the area covered by denser corrosion products
quently, during the electrochemical corrosion process, the anodic re­ became the cathode region [41]. Consequently, this induced a local
action involves the dissolution of zinc within the eutectic phase (Eq. corrosion rate difference on the coating surface, causing the corrosion
(3)), whereas the cathodic reaction entails the reduction of oxygen (Eq. rate to increase. With the progression of corrosion to 24 months,
(4)).

Zn→Zn2+ + 2e- (3)

O2 + 2H2 O + 4e- →4OH- (4)

As the corrosion progresses, OH migrates towards the anode,
−

resulting in the formation of Zn(OH)2 (Eq. (5)). Subsequently, Zn(OH)2

undergoes dehydration to form ZnO (Eq. (6)). This is because the Gibbs
free energies for Eq. (5) and Eq. (6) are − 77.2 kJ/mol and − 11.9 kJ/mol,
respectively, indicating that both reactions are thermodynamically
spontaneous.

Zn2+ + 2OH- →Zn(OH)2 (5)

Zn(OH)2 →ZnO + H2 O (6)

Zn(OH)2 serves as the fundamental nucleation point for the forma­

tion of various corrosion products, including Zn5(OH)8Cl2⋅H2O (Eq. (7),
Zn5(CO3)2(OH)6 (Eq. (8), Zn4(SO4)(OH)6⋅H2O (Eq. (9), and NaZn4(SO4) Fig. 16. Variations of Rp at Mohe after different corrosion periods.

Fig. 15. Nyquist diagrams of Zn–Al–Mg at Mohe after (a) 6 months, (b) 12 months, (c) 24 months.

4299
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

Fig. 17. Schematic diagram of corrosion mechanism under marine atmosphere.

corrosion products appear inside the coating. These corrosion products reactions, are exothermic processes, and the equilibrium concentration
are very dense and can effectively prevent the intrusion of corrosive of these adsorbed and dissolved gases increases as the temperature de­
media, thereby reducing the corrosion rate. creases [44]. Winter in Mohe brings significant day-night temperature
However, the atmospheric conditions in Mohe are typical of a variations, sometimes exceeding 30 ◦ C (− 0.7 ◦ C ~ − 32.5 ◦ C), facili­
northern urban environment, characterized by prolonged periods of low tating frequent melting and condensation of snow and ice. This pro­
temperature throughout the year. Since the exposure test commenced in longed presence of a water film on the surface might have accelerated
September, when the atmospheric temperature surpasses 0 ◦ C, a slight corrosion [32], possibly explaining the higher corrosion rate observed in
liquid film tends to form on the sample surface after exposure, leading to the first six months.
the generation of Zn(OH)2 over time (refer to Eq. (5)). Influenced by the After 12 months of exposure, Zn4(SO4)(OH)6⋅4H2O and ZnSO4 were
winter climate, the average temperature remains below 0 ◦ C for an produced, with Zn4(SO4)(OH)6⋅4H2O forming through the combination
extended duration. Fig. 18 illustrates the fluctuations in average atmo­ of ZnSO4 and Zn(OH)2 according to Eq. (10). As previously mentioned,
spheric temperature and humidity in Mohe during the initial year of Zn4(SO4)(OH)6⋅4H2O is a slightly soluble and corrosion-resistant sub­
exposure. Notably, the temperatures in the first six months are pre­ stance, potentially contributing to a reduction in the corrosion rate.
dominantly lower compared to the latter six months, with the average Additionally, Svensson and colleagues [45] discovered that the nucle­
temperature mostly dipping below 0 ◦ C in the first half and rising above ation of Zn4(SO4)(OH)6⋅4H2O exhibits a stronger temperature depen­
0 ◦ C in the second half. Interestingly, the corrosion rate is higher in the dence. Therefore, when exposed to direct sunlight, the formation of
first six months compared to the latter six months. This can be attributed Zn4(SO4)(OH)6⋅4H2O occurs more rapidly on the skyward side,
to the complex relationship between temperature and corrosion rate. explaining the variation in the corrosion resistance between the two
Electrochemical reactions and electrolytic conductivity are thermally exposed surfaces.
activated processes [42,43], implying that the electrochemical reaction
ZnSO4 + 3Zn(OH)2 +4H2 O→Zn4 (SO4 )(OH)6 ⋅4H2 O (10)
rate typically increases with temperature. Conversely, the adsorption
and dissolution of gases like O2, SO2 or CO2, involved in electrochemical The evolution of the corrosion rate difference between two exposed
surfaces over time is indeed influenced by the structure of corrosion
products. During the initial six months of exposure, the corrosion pro­
cess is susceptible to external influence due to the limited formation of
corrosion products. In Mohe, where snowfall can exceed 50 cm in
winter, the skyward side of the material is often covered by a thick layer
of snow. This snow acts as a natural barrier, reducing the contact be­
tween corrosive particles and the coating surface. Furthermore, the
significant temperature fluctuations in the region, with day-night dif­
ferences exceeding 30 ◦ C, exacerbate the corrosion rate. Condensation
forms easily on the material surface due to these temperature swings,
and the repeated freezing and thawing of water molecules can cause
thermal expansion and contraction. This, in turn, can lead to partial
separation between different phases, resulting in cracks and holes in the
eutectic phase. The combination of temperature extremes and pollutants
often leads to more severe corrosion in the eutectic phase, reducing the
Fig. 18. Variations of atmospheric average temperature (a) and humidity (b) in corrosion resistance, especially on the field-ward side. After 12 months,
Mohe during the first year of exposure.

4300
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

the zinc-rich phase on the field-ward side becomes increasingly exposed, Centre for Precision Manufacturing and Surface Functionalization of
while the skyward side remains partially protected by the eutectic Light Alloys, Liaoning Key Laboratory of Environmental Corrosion and
phase. Corrosion primarily occurs at the junction of these two phases. Evaluation of Materials, and Shenyang Key Laboratory of Environmental
Without the protective barrier of snow on the field-ward side, the zinc- Corrosion Evaluation and Standardization of Materials.
rich phase and local corrosion products are more exposed, accelerating
the corrosion process. After 24 months of exposure, the thickness of References
corrosion products on both sides is relatively uniform, and the eutectic
phase protecting the zinc-rich phase on the skyward side has dis­ [1] Feng K, Mei H, Chen Y, Yan H, Song H, Luo C, Guan B, Guan H, Monacelli E, Hu Q,
Hu Z. A novel high corrosion-resistant and superhydrophobic MAO-30/PA/VTES
appeared. This exposure of the zinc-rich phase leads to accelerated composite coating on Cu 15Fe alloy for long-term corrosion protection. Surf Coat
corrosion of the surrounding eutectic phase, resulting in a noticeable Technol 2024;493:13127.
decrease in corrosion resistance. Overall, the interaction between [2] Huang H, Hou L, Du H, Wei H, Liu X, Wang Q, Wei Y. Efficient dual defense: PDA-
Cu coating for simultaneous corrosion resistance and antibacterial protection of Mg
environmental factors, such as snow cover and temperature fluctuations, alloys. Corros. Sci. 2024;233:112103.
and the formation of corrosion products significantly influences the [3] Huang H, Hou L, Du H, Wei H, Liu X, Wang Q, Wei Y. In situ electrochemical
corrosion rate on different exposed surfaces over time. deposition enables dense polypyrrole grown on AZ31 surface for improving
corrosion resistance. Mater Lett 2024;362:136214.
[4] Yu T, Wu J, Shen Y, Penkova A, Qi W, Su R. Transparent coating based on
5. Conclusion multienzyme-mimicking Janus nanozyme for synergetic biofouling control in
seawater. Chem. Eng. J. 2024;498:155144.
[5] Yu Z, Sun W, Wang L, Yang Z, Ren Y, Li Y, Liu G, Zhang Q. Bioinspired sulfobetaine
This study comprehensively examines the corrosion rates,
borneol fluorinated amphiphilic polymers for marine antifouling and fouling
morphology, structure, composition, and electrochemical properties of release applications. ACS Appl Mater Interfaces 2024;16:46690–702.
corrosion products on both the skyward and field-ward sides of ZAMc [6] Volovitch P, Vu TN, Allély C, Abdel Aal A, Ogle K. Understanding corrosion via
exposed to high-temperature, high-humidity marine and extremely cold corrosion product characterization: II. Role of alloying elements in improving the
corrosion resistance of Zn–Al–Mg coatings on steel. Corros. Sci. 2011;53:2437–45.
atmospheres. The key findings from the detailed analysis are outlined [7] Prosek T, Nazarov A, Le Gac A, Thierry D. Coil-coated Zn–Mg and Zn–Al–Mg: effect
below. of climatic parameters on the corrosion at cut edges. Prog. Org. Coat. 2015;83:
26–35.
[8] Thierry D, LeBozec N, Le Gac A, Persson D. Long-term atmospheric corrosion rates
(1) The corrosion rates of ZAMc vary depending on the exposure of hot dip galvanised steel and zinc-aluminium-magnesium coated steel. Mater
location and duration. Specifically, in the high-temperature, Corros 2019;70:2220–7.
high-humidity marine atmosphere of Wanning, the corrosion [9] Schuerz S, Fleischanderl M, Luckeneder GH, Preis K, Haunschmied T, Mori G,
Kneissl AC. Corrosion behaviour of Zn–Al–Mg coated steel sheet in sodium
rate increases initially and then decreases over time. Conversely, chloride-containing environment. Corros. Sci. 2009;51:2355–63.
in the extremely cold urban atmosphere of Mohe, the corrosion [10] Prosek T, Larché N, Vlot M, Goodwin F, Thierry D. Corrosion performance of Zn-Al-
rate decreases at first and then increases. Mg coatings in open and confined zones in conditions simulating automotive
applications. Mater Corros 2010;61:412–20.
(2) Significant differences in the corrosion processes are observed [11] Keppert TA, Luckeneder G, Stellnberger KH, Mori G, Antrekowitsch H. The effect of
between the skyward and field-ward sides in these two harsh sulphate, phosphate, nitrate and acetate on the corrosion behaviour of Zn-Al-Mg
environments. In Wanning, the corrosion rates on the skyward hot-dip galvanised steel. Mater Corros 2014;65:560–8.
[12] Oh M-S, Kim S-H, Kim J-S, Lee J-W, Shon J-H, Jin Y-S. Surface and cut-edge
sides are consistently lower than those on the field-ward sides,
corrosion behavior of Zn-Mg-Al alloy-coated steel sheets as a function of the alloy
with the disparity decreasing initially and then increasing over coating microstructure. Met Mater Int 2015;22:26–33.
time. In Mohe, the corrosion rates on the skyward sides are lower [13] Sullivan J, Mehraban S, Elvins J. In situ monitoring of the microstructural
than those on the field-ward sides after 6 and 12 months of corrosion mechanisms of zinc–magnesium–aluminium alloys using time lapse
microscopy. Corros. Sci. 2011;53:2208–15.
exposure, but they become comparable after 24 months. [14] Ahmadi M, Salgın B, Kooi BJ, Pei Y. Genesis and mechanism of microstructural
(3) In the Wanning atmosphere, corrosion predominantly occurs in scale deformation and cracking in ZnAlMg coatings. Mater Des 2020;186:108364.
the eutectic phase and spreads inward along the eutectic phase [15] Azmat NS, Ralston KD, Muddle BC, Cole IS. Corrosion of Zn under fine size aerosols
and droplets using inkjet printer deposition and optical profilometry
adjacent to the zinc-rich phase. Oxygen is evenly distributed, quantification. Corros. Sci. 2011;53:3534–41.
sulfur is concentrated in the upper layer, and chlorine exhibits [16] Azmat NS, Ralston KD, Muddle BC, Cole IS. Corrosion of Zn under acidified marine
strong vertical migration, accumulating in the corrosion products droplets. Corros. Sci. 2011;53:1604–15.
[17] Yadav AP, Nishikata A, Tsuru T. Electrochemical impedance study on galvanized
of the coating. In contrast, in the Mohe atmosphere, localized steel corrosion under cyclic wet–dry conditions––influence of time of wetness.
corrosion is more pronounced, and corrosion pits tend to form in Corros. Sci. 2004;46:169–81.
the eutectic phase near the zinc-rich phase. These pits are wider [18] Falk T, Svensson JE, Johansson LG. The role of carbon dioxide in the atmospheric
corrosion of zinc - a laboratory study. J Electrochem Soc 1998;145:39–44.
at the top and narrower at the bottom. As corrosion progresses, [19] Biiestek T, Drys M, Sokolov N, Knotkova D, Ramishvili R, Kozhukharov V,
these pits gradually transition from being deep and narrow to Zeidel M. Atmospheric corrosion of metallic systems. V.Identification of the
being shallow and wide. chemical-compounds in the corrosion products of zinc. Prot Met 1983;19:612–5.
[20] Bernard MC, Hugot-le Goff A, Phillips N. In situ Raman study of the corrosion of
zinc-coated steel in the presence of chloride I. Characterization and stability of zinc
Data availability corrosion products. J Electrochem Soc 1995;142:2162–7.
[21] Salgueiro Azevedo M, Allély C, Ogle K, Volovitch P. Corrosion mechanisms of Zn
(Mg, Al) coated steel in accelerated tests and natural exposure: 1. The role of
The authors do not have permission to share data. Data will be made
electrolyte composition in the nature of corrosion products and relative corrosion
available on reasonable request. rate. Corros. Sci. 2015;90:472–81.
[22] Salgueiro Azevedo M, Allély C, Ogle K, Volovitch P. Corrosion mechanisms of Zn
Declaration of competing interest (Mg,Al) coated steel: 2. The effect of Mg and Al alloying on the formation and
properties of corrosion products in different electrolytes. Corros. Sci. 2015;90:
482–90.
The authors declare that they have no known competing financial [23] Zou J, Lu L. Research progress on relationship between alloy phase and corrosion
interests or personal relationships that could have appeared to influence resistance in Zn-Al-Mg coating. Chin. J. Nonferrous Met. 2022;32:1934–44.
[24] Odnevall Wallinder I, Leygraf C. A critical review on corrosion and runoff from zinc
the work reported in this paper. and zinc-based alloys in atmospheric environments. Corros 2017;73:1060–77.
[25] Veleva L, Acosta M, Meraz E. Atmospheric corrosion of zinc induced by runoff.
Acknowledgement Corros. Sci. 2009;51:2055–62.
[26] Qu Q, Yan CW, Wan Y, Cao CN. Effects of NaCl and SO2 on the initial atmospheric
corrosion of zinc. Corros. Sci. 2002;44:2789–803.
This work was supported by the Natural Science Foundation Project [27] Qiao C, Shen L, Hao L, Mu X, Dong J, Ke W, Liu J, Liu B. Corrosion kinetics and
of Shenyang (Independent Exploration) [Grant number 23503605]. patina evolution of galvanized steel in a simulated coastal-industrial atmosphere.
J Mater Sci Technol 2019;35:2345–56.
The authors would like to thank Liaoning Engineering Research

4301
Y. Liu et al. Journal of Materials Research and Technology 33 (2024) 4290–4302

[28] Diler E, Rouvellou B, Rioual S, Lescop B, Nguyen Vien G, Thierry D. [37] Gu T, Liu Y, Peng C, Zhang P, Wang Z. Initial atmospheric corrosion of zinc-
Characterization of corrosion products of Zn and Zn–Mg–Al coated steel in a aluminum-magnesium coated steel and galvanized steel in regions of extremely
marine atmosphere. Corros. Sci. 2014;87:111–7. cold and industrial climate. Mater Chem Phys 2022;291:126686.
[29] Tomandl A, Labrenz E. The corrosion behavior of ZnAlMg alloys in maritime [38] Zhang X, Zou DN, Zhou YQ, Na X, Zhang W, Han Y. Effect of Cl- concentration on
environments. Mater Corros 2016;67:1286–93. pitting corrosion property of maraging hardened stainless steel based on pourbaix
[30] Thierry D, Persson D, Luckeneder G, Stellnberger K-H. Atmospheric corrosion of diagram. Mater Sci Forum 2018;940:59–64.
ZnAlMg coated steel during long term atmospheric weathering at different [39] Aoki IV, Bernard MC, Torresi S, Deslouis C, Tribollet B. Ac-impedance and Raman
worldwide exposure sites. Corros. Sci. 2019;148:338–54. spectroscopy study of the electrochemical behaviour of pure aluminium in citric
[31] Mikhailov AA, Strekalov PV, Panchenko YM. Atmospheric corrosion of metals in acid media. Electrochim Acta 2002;46:1871–8.
regions of cold and extremely cold climate (a review). Prot Met 2008;44:644–59. [40] Svensson JE, Johansson LG. A laboratory study os the initial stages of the
[32] Cui Z, Ge F, Wang X. Corrosion mechanism of materials in three typical harsh atmospheric corrosion of zinc in the presence of NaCl; influence of SO2 and NO2.
marine atmospheric environments. J. Chin. Soc. Corros. Prot. 2022;42:403–9. Corros. Sci. 1993;34:721–40.
[33] Gu T, Zhang P, Guo M, Peng C, Ma C, Liu Y, Wang Z. Corrosion behavior of zinc- [41] C.C. N.. Principles of electrochemistry of corrosion. 3 rd ed. Beijing: Chemical
aluminum-magnesium coated steel in simulated marine atmosphere. Int J Industry Press; 2008.
Electrochem Sci 2022;17:22054. [42] Salgueiro Azevedo M, Allély C, Ogle K, Volovitch P. Corrosion mechanisms of Zn
[34] Cole IS, Muster TH, Furman SA, Wright N, Bradbury A. Products formed during the (Mg,Al) coated steel: the effect of HCO−3 and NH+ 4 ions on the intrinsic reactivity of
interaction of seawater droplets with zinc surfaces: I. Results from 1-and 2.5-day the coating. Electrochim Acta 2015;153:159–69.
exposures. J Electrochem Soc 2008;155:C244–55. [43] Volovitch P, Allely C, Ogle K. Understanding corrosion via corrosion product
[35] Yin Q, Wang Z-Y, Liu M-R, Pan C. Synergistic effect of NaCl and SO2 on the initial characterization: I. Case study of the role of Mg alloying in Zn–Mg coating on steel.
atmospheric corrosion of zinc under wet–dry cyclic conditions. Acta Metall. Sin. Corros. Sci. 2009;51:1251–62.
Engl. Lett. 2018;32:780–96. [44] Esmaily M, Mortazavi N, Svensson JE, Johansson LG. Evidence for an unusual
[36] Cole IS, Ganther WD, Furman SA, Muster TH, Neufeld AK. Pitting of zinc: temperature dependence of the atmospheric corrosion of zinc. J Electrochem Soc
observations on atmospheric corrosion in tropical countries. Corros. Sci. 2010;52: 2016;163:C864–72.
848–58. [45] Svensson JE, Johansson LG. The temperature-dependence of the SO2-induced
atmospheric corrosion of zinc; A laboratory study. Corros. Sci. 1996;38:2225–33.

4302