Construction and Building Materials 411 (2024) 134284

Contents lists available at ScienceDirect

Construction and Building Materials

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

Corrosion behavior and mechanical performance of weathering steel in

industrial and rural atmospheric environments
Chengshuo Han a, Zhigang Li b, Xia Yang a, *, Jiuyuan Wang a
a
School of Urban Construction, Yangtze University, Jingzhou 434023, China
b
CCCC Second Highway Consultants Co., Ltd., Wuhan 430000, China

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

Keywords: In this study, a series of alternate immersion tests were conducted to simulate the corrosion processes of
Weathering steel weathering steel and control material (Q235 steel) in industrial and rural atmospheric environments. The
Atmospheric environment corrosion rates of the two steels were regularly monitored and compared using weight loss method. Scanning
Corrosion resistance
Electron Microscope (SEM) and X-ray Diffraction (XRD) were employed to analyze the microstructure and phase
Alternate immersion test
Constitutive model
compositions of the rust layer formed on steel. Finally, the tensile tests were performed to examine the me­
chanical performance degradation of weathering steel. The results indicated that the corrosion rates of weath­
ering steel decreased dramatically at first and then tended to be steady. In industrial environment, the
weathering steel maintains a 30 % lower corrosion rate than Q235 steel, whereas in rural environment, the
difference between the two corrosion rates decreased dramatically over time and disappeared after 360 h. The
cause for the difference between the corrosion rates is that weathering steel has higher content of α-FeOOH, CuO
and Ni compounds in the corrosion product which make its rust layer more compact and denser than that of
Q235 steel, leading to better corrosion resistance. Existing stress-strain curve models for ordinary steel gave poor
prediction effect in the hardening stage of weathering steel. Therefore, a new constitutive model for the
weathering steel was proposed in this study. It is demonstrated that the proposed model can well reflect the
mechanical behaviors of weathering steel before and after corrosion.

1. Introduction [2,3]. It is estimated that the corrosion rate of weathering steel is about
2–8 times that of ordinary steel in atmospheric environments [4,5]. This
Mild steel has been massively applied in engineering structural can be attributed to the alloying elements contained in weathering steel,
components owing to its advantages of excellent physical properties and such as Cr, Ni, P, Cu, Mo, etc. [6–8] which helps to form dense and
cost efficiency [1]. Nevertheless, as is well known that steel is prone to tightly adherent protective rust layer on steel surface when exposed to
corrosion in atmospheric environments especially in aggressive envi­ air, and further effectively obstruct the transport of corrosive agents
ronments. Corrosion not only weakens the section of steel components, such as H2O and O2 to steel substrate [9].
but also causes the occurrence of corrosion pits, which can further cause So far, there has been many studies about the corrosion perfor­
stress concentration, and result in deterioration of toughness, ductility mances of ordinary steel in atmospheric environment, and the work
and capacity. Therefore, it is crucial to explore the corrosion perfor­ focused on the corrosion resistance of weathering steel is much less. It is
mances of steel in atmospheric environments. Coating is the most reported weathering steel and ordinary steel had essentially the same
common way to improve the resistance of steel to atmospheric corro­ corrosion rate in the initial stage of marine corrosion, but then the
sion, but there are also some shortcomings, such as heavy workload and corrosion rate of weathering steel became slower, causing the corrosion
requiring periodic re-painting, which highlights the requirement for rate of ordinary steel higher than that of weathering steel [10]. The rust
more anti-corrosion materials such as weathering steel. Although a little layer formed on weathering steel in atmospheric environment mainly
expensive, weathering steel has been frequently applied as building comprises α-FeOOH, γ-FeOOH, γ-Fe3O4 and γ-Fe2O3, while that of or­
material in transportation, aviation, infrastructure, and other fields due dinary steel has similar compositions but much more voids and cracks
to its superior corrosion resistance and excellent mechanical property [11,12]. The corrosion product γ-FeOOH is unstable, which can further

* Corresponding author.
E-mail address: [email protected] (X. Yang).

https://doi.org/10.1016/j.conbuildmat.2023.134284
Received 29 September 2023; Received in revised form 14 November 2023; Accepted 20 November 2023
Available online 6 December 2023
0950-0618/© 2023 Elsevier Ltd. All rights reserved.
C. Han et al. Construction and Building Materials 411 (2024) 134284

transform into metastable amorphous FeOx (OH)3–2x and ultimately 2. Experimental method procedures
evolves into protective nanoscale α-FeOOH, forming a protective rust
layer on the steel surface [13,14]. It is validated that the rust layer of 2.1. Experimental materials and procedures
weathering steel has a higher α-FeOOH content and lower γ-FeOOH
content compared to that of ordinary steel, and thus the rust layer of The chemical compositions of the weathering and Q235 steels used
weathering steel is denser and provides a stronger protective effect in this study are shown in Table 1. Three sizes of specimens were made
against the atmospheric corrosion [15,16]. The increase of Cr, Cu and Ni for each type of steel according to different experimental purposes, as
contents is benefit to the corrosion resistance of weathering steel, but an shown in Fig. 1.
excessive Cr content can cause an aggravation of localized corrosion and All the specimens were cleaned firstly, and then the alternate im­
reduce the steel corrosion resistance [17]. So far, many achievements mersion test was conducted. After exposure to corrosive solution for a
have been made on the corrosion resistance of weathering steel, but specified period of time, a serial of tests was carried out on each spec­
these studies are mainly focused on the weathering steel in marine or imen, including the corrosion weight loss measurement, the rust layer
industrial environment. Research on the weathering steel in rural micro-morphology observation, the XRD analysis, the electrochemical
environment is rare. Moreover, the existing studies generally involve measurement and the static tensile test. Fig. 2 and Table 2 give the
only one type of steel, and differences between weathering steel and number of specimens and their uses.
ordinary steel are seldom systematically compared and discussed with
respect to the corrosion resistance and corrosion mechanism. Therefore, 2.2. Alternate immersion test
there is still a lack of quantitative comparison of the corrosion perfor­
mances between weathering steel and mild steel. To quickly evaluate the corrosion resistance of weathering steel and
The constitutive model intuitively reflects the mechanical properties Q235 steel in rural and industrial environments, the alternate immer­
of steel [18,19]. Thus, it is of great significance to obtain the stress-strain sion test was conducted under the guidance of Chinese standard GB/
relation of corroded steel in the safety evaluation of existing steel T19746-2018 [21]. Fig. 3 lists the materials and the specimens
structures. To date, extensive work has been carried out on the consti­ required for tests. Fig. 4 displays some tensile specimens before and after
tutive model of ordinary steel. In contrast, little work has been done on corrosion. A dry and wet cycle testing machine was used for this test,
the stress-strain relationship of weathering steel, let alone that after with the experimental temperature of (25 ± 2) ◦ C. Distilled water and
corrosion. As is known that weathering steel and Q235 steel differ in 3.5 % NaCl + NaHSO3 (0.02 mol/L) solution were respectively prepared
chemical composition, indicating that the existing constitutive models to simulate the rural and industrial atmospheric environments. Each
for ordinary steel may be not adequate in reflecting the mechanical immersion cycle takes 60 min, including 12 min of immersion and
behavior of weathering steel. Therefore, further research is still needed 48 min of drying [22]. Four levels of cycle number were considered as
to obtain the stress-strain relationship of weathering steel before and follows: 24 cycles (24 h), 168 cycles (168 h), 360 cycles (360 h), and
after corrosion. 720 cycles (720 h).
In this paper, alternate immersion tests were carried out on weath­
ering steel and Q235 steel to quantitatively compare their corrosion
behaviors [20]. Two corrosion environments, namely, industrial atmo­ 2.3. Corrosion kinetics
spheric environment and rural atmospheric environment, were consid­
ered and simulated. By monitoring weights of the two steels, their The corrosion weight loss of weathering steel and Q235 steel in rural
corrosion rates in the two corrosion environments were determined and and industrial environments was monitored regularly by the following
discussed, and meanwhile, the change of their corrosion potentials with steps. Firstly, each steel specimen (with dimension of 50 mm × 50 mm)
time were plotted by electrochemical test. Then the microstructural was cleaned and weighed using a balance to record its initial weight
observations and phase composition analyses were performed by scan­ (W0), and after that the alternate immersion test was started. Specimens
ning electron microscopy (SEM) and X-ray diffraction (XRD). On this were periodically sampled after 24, 168, 360, and 720 h, and then
basis, a comparative analysis of the corrosion mechanisms of the two treated by HCl solution to remove the corrosion products that might
steels was conducted. Finally, the mechanical performance degradation have formed on the surface. Afterwards, the specimens were washed
of weathering steel and Q235 steel in the two corrosion environments with deionized water, wiped with ethanol solution, dried and stored in a
were examined by tensile tests, based on which a constitutive model for drying oven. Finally, the weight was measured again (W1) after drying at
weathering steel before and after corrosion was proposed and validated. 70 ◦ C for 24 h. Fig. 5 shows the procedure of weightlessness measure­
Unlike previous studies that only focused on one type of steel or just use ment. The weight loss of each specimen was calculated using the for­
the constitutive model of ordinary steel to predict the mechanical mula below:
behavior of weathering steel, the present work makes a systematic W0 − W1 ( )
ΔW = × 104 g⋅m− 2 (1)
comparative study on the corrosion resistance of weathering steel and 2⋅(a⋅b + b⋅c + a⋅c)
Q235 steel, and explored the difference between the tress-strain re­
lationships of weathering steel and Q235 steel. The research results are where W0 and W1 are the masses of the specimen before and after
expected to provide basic and data support for the research on the alternate immersion test (g); a, b and c are the original length, width and
corrosion resistance of weathering steel. thickness (cm), respectively.
The corrosion rates of weathering steel and Q235 steel in rural and
industrial environments were further calculated as follows:
KΔw
R= (2)
s×t×ρ

Table 1
Chemical composition of weathering steel and Q235 steel.
Steel C Si Mn S P Cu Ni Cr Fe

Q235 steel 0.16 % 0.20 % 0.61 % 0.023 % 0.019 % — — — 98.988 %

Weathering steel 0.01 % 0.36 % 0.43 % 0.007 % 0.09 % 0.3 % 0.15 % 0.5 % 98.153 %

2
C. Han et al. Construction and Building Materials 411 (2024) 134284

Fig. 1. Experimental specimens (unit: mm).

Fig. 2. Schematic diagram of experimental apparatus.

electron microscopy (SEM) and X-ray diffraction (XRD). Specifically,

Table 2
SEM was used to observe the corrosion layers, while XRD was employed
Experimental specimens and experimental instructions.
to measure and analyze the phase compositions of corrosion products.
Steel Experimental specimens (numbers) By performing XRD and SEM tests, the important information
(numbers)
Specimen 1 Specimen 2 Tensile specimen including the composition, the morphology, the structure, and the
characteristics of the rust layers were discussed, based on which the
corrosion resistance mechanism of weathering steel was investigated.

2.5. Electrochemical test

Q235 Steel Electrochemical Corrosion weight Static tensile test

Polarization curves are of great guiding significance in studying the
(64 pieces) test loss measurement (48 pieces)
Weathering (16 pieces); (32 pieces) corrosion behavior of steel [23]. To obtain the Tafel polarization curves
steel SEM test of weathering steel and Q235 steel in rural and industrial environments,
(64 pieces) (16 pieces); the two steels (with dimension of 10 mm × 10 mm) were periodically
XRD test
sampled after 24, 168, 360 and 720 h of corrosion and then their elec­
(16 pieces)
trochemical properties were studied by an electrochemical test appa­
ratus (see Fig. 6). Fig. 7 sketches the three-electrode system used for the
where Δw is the weight loss (g); t is the period of alternate immersion electrochemical test, in which the platinum electrode is the counter
test (h); s and ρ (7.86 for carbon steel and 7.85 for weathering steel) are electrode, the saturated glycerol electrode (SCE) is the reference elec­
the exposed area (m2) and the density (g⋅m− 3) of the specimen, trode, and the rusty sample is the experimental electrode. The potential
respectively. scan range and scan rate were set as − 250 to + 500 mV and a 0.5 mV/s,
respectively. The NaCl solution with a mass fraction of 3.5 % was used
2.4. Morphology and composition characteristics of rust layer for this test.

After being corroded for specified periods (24, 168, 360, and 720 h), 2.6. Mechanical property test
the morphology and phase composition of the rust layers formed on the
surface of weathering steel and Q235 steel were examined by scanning In order to explore the mechanical property degradation of

3
C. Han et al. Construction and Building Materials 411 (2024) 134284

Fig. 3. Explanation diagram of alternate immersion test.

Fig. 4. Tensile specimens before and after corrosion.

Fig. 5. Procedure of weightlessness measurement.

weathering steel and Q235 steel in industrial and rural environments, a

total of 48 pieces of tensile specimens are prepared (see Fig. 1). Simi­
larly, the standard-sized weathering steel and Q235 steel were periodi­
cally sampled after being subjected to corrosion for 24, 168, 360, and
720 h, respectively. Physical method was adopted to remove the rust
layer on the surface. The static tensile test was performed using an EHC-
3100 universal testing machine with a maximum capacity of 1000 kN
(see Fig. 8). To eliminate the residual stress of the steel samples, the
specimens were pre-loaded with a load value of about 0.4 times the yield
strength. During the formal loading, the tensile test was conducted at a
speed of 1 mm/min until the specimen fracture. For reducing the error,
three parallel tests were carried out for each load case.

Fig. 6. Electrochemical test.

4
C. Han et al. Construction and Building Materials 411 (2024) 134284

3. Results and discussion

3.1. Corrosion kinetics

Fig. 9 illustrates the relationships between corrosion weight loss and

exposure time in rural and industrial environments. It is shown that the
corrosion weight loss of Q235 steel and weathering steel are roughly the
same at the beginning, but as the exposure time increase, the difference
between them increases dramatically, especially under the industrial
environment. For example, the weight losses of weathering steel and
Q235 steel in industrial environment after 24 h are 85 g⋅m− 2 and
95 g⋅m− 2, respectively, while after 720 h, the losses increase to
911 g⋅m− 2 and 1379 g⋅m− 2, respectively.
The corrosion solution plays an important role in the weight loss. The
industrial environment causes much more severe corrosion in steels as
Fig. 7. The three-electrode system. compared to the rural environment, leading to higher weight loss. The
weight losses of Q235 steel in rural and industrial environments after
24 h are 55 g⋅m− 2 and 95 g⋅m− 2, respectively, with a difference of 72.3
%, while after 720 h the losses increase to 660 g⋅m− 2 and 1379 g⋅m− 2,
respectively, with a difference of 108.9 %. The weight losses of weath­
ering steel in rural and industrial environments after 24 h are 44 g⋅m− 2
and 85 g⋅m− 2, with a difference of 93.2 %, respectively, while after
720 h the losses increase to 560 g⋅m− 2 and 911 g⋅m− 2, respectively,
with a difference of 62.7 %.
It is reported that there is a power-law relationship between weight
loss of carbon steel and corrosion time as below [24,25]:
W = Atn (3)

where t represents the corrosion time (h), A is a constant related to

environment and material, and n is the environment-related corrosion
coefficient [26]. When n > 1, it indicates that rust layer makes no
contribution in protecting steel substrate; conversely, it means that rust
layer has a protective effect on steel substrate.
Fig. 9 gives the curve fitting of the weight losses of weathering steel
and Q235 steel over time. It is indicated that as expected, the weight
losses have significant power function relation with time W– –Atn.
Table 3 lists the parameters of the power exponent curves. It is shown
that both weathering steel and Q235 steel have the n values less than 1,
suggesting that the rust layers on both the two steels can provide pro­
tection to their steel substrate. The parameters A and n of weathering
Fig. 8. Tensile test. steel are smaller than those of Q235 steel, indicating that weathering
steel has stronger corrosion resistance as compared to Q235 steel.
Fig. 10 depicts the corrosion rates of weathering steel and Q235 steel
over time. It can be observed that under both the rural and industrial

Fig. 9. Weight loss of steel samples with corrosion time.

5
C. Han et al. Construction and Building Materials 411 (2024) 134284

Table 3 of corrosion are 4.43 and 2.24, respectively, with a difference of 49.4 %,
Fitting results of weight loss. while after 720 h, the corrosion rates are 1.35 and 0.95, respectively,
Environmental category Steel grade Fitting results with a difference of 29.6 %.
A n R2

Industrial environment Weathering steel 12.3 0.5 0.986 3.2. Corrosion product analysis
Q235 Steel 18.8 0.7 0.997
Rural environment Weathering steel 3.8 0.6 0.992
Figs. 11 and 12 give the phase composition of the rust layer formed
Q235 Steel 6.8 0.8 0.986
on weathering steel and Q235 steel in rural and industrial environments.
Referring to previous studies [27–29], corrosion products of the two
environments, the corrosion rates of the two steels decrease rapidly over steels are approximately the same, mainly including two kinds of iron
exposure time in the early stage and then tend to be stable, with a hydroxyl groups—goethite(α-FeOOH), and lepidocrocite (γ-FeOOH)—
turning point at 360 h. In term of this, the curves are divided into two as well as the ironoxides (Fe3O4, γ-Fe2O3), which have the highest
parts as shown in Fig. 10, i.e., the initial stage and the stable stage. diffraction peak. Among them α-FeOOH is the most stable product and
In rural environment, the corrosion rates of Q235 steel and weath­ has a positive impact on the corrosion resistance [30]. The α-FeOOH
ering steel differ significantly at the beginning, with a difference of mainly exists in the outer rust layer and protects the steel matrix to some
about 45 %, but then the difference diminishes gradually. During the extent. On the contrary, the product of γ-FeOOH is unstable and prone to
stable stage, the corrosion rates of the two steels are basically the same, decomposition, and thus has little effect on preventing steel from
indicating that weathering steel has better corrosion resistance than corrosion. The stability of Fe3O4/ γ- Fe2O3 is between α-FeOOH and
Q235 steel initially, but with the exposure time increasing, the two steels γ-FeOOH, and its content is the highest in the rust layer. This product
show no difference. On the other hand, under the industrial environ­ mainly exists in the inner rust layer and is exposed to the corrosive
ment, weathering steel maintains a roughly 0.5 mm/a higher corrosion environment when the outer layer peels off.
rate than Q235 steel, and their corrosion rates show similar trends It can be seen from Fig. 11 that the diffraction peak intensity of
throughout the test. The weathering steel consistently exhibits better Fe3O4/ γ- Fe2O3 of weathering steel shows an increasing tendency with
corrosion resistance than Q235 steel, and this advantage becomes more increasing exposure time, and the diffraction peak intensity of Fe3O4/ γ-
prominent with the increase of exposure time. For example, the corro­ Fe2O3 of Q235 steel first increases, then decreases and then increases
sion rates of weathering steel and Q235 steel after 24 h in industrial again, indicating that the crystallinity of Fe3O4/ γ- Fe2O3 formed in
environment are 4.43 mm/a and 5.04 mm/a, respectively, with a dif­ weathering steel increases over exposure time in industrial environ­
ference of 13.8 %, while after 720 h, the rates of the two steels are ment. While the diffraction peak intensity of α- FeOOH decreases at first
1.35 mm/a and 1.92 mm/a, respectively, with a difference of 42.2 %. and then increases slightly. The reason for this variation is that in the
Therefore, it is comparatively more recommended to use the weathering initial stage, the iron dissolution reaction increases dramatically due to
steel in industrial environment. the presence of corrosive ions, such as Cl- and SO2-4 ions and without the
The environment significantly influences the corrosion rates of the protection of rust layer, resulting in a rapid increase of α-FeOOH and
two steels. The steels in industrial environment exhibit faster corrosion magnetite. Afterwards, with the rapid increase of Fe3O4/ γ- Fe2O3 be­
rate as compared to that in rural environment, especially for the Q235 tween the α-FeOOH and steel substrate, the relative content of α-FeOOH
steel. For example, the corrosion rates of Q235 in rural and industrial decreases accordingly. It should be noted that the diffraction peak of
environments after 24 h of corrosion are 5.04 and 3.13, respectively, NaCl was also detected on the samples in industrial environment after
with a difference of 37.9 %, while after 720 h, the corrosion rates are 720 h. This should be because the samples were not processed before
1.82 and 0.98, respectively, with a difference of 46.2 %. This can be conducting XRD test. It follows from Fig. 12 that with the increase of
attributed to the fact that industrial environment is rich in corrosive exposure time, the diffraction peak intensity of Fe3O4/ γ- Fe2O3 keeps
ions, such as Cl- and SO2- -
4 , Cl not only has electrical conductivity, but the highest throughout the process. In general, the diffraction peak in­
also can accelerate the steel corrosion. The corrosion of weathering steel tensity of α-FeOOH formed on weathering steel changes little over time
is also remarkably affected by the environment, but the impact is in rural environment, while that formed on Q235 steel decreases firstly
constantly decreasing with corrosion time. For example, the corrosion and then increases slightly. By combining Figs. 11 and 12, it is shown
rates of weathering steel in rural and industrial environments after 24 h that the difference between phase contents of the rust layers formed on

Fig. 10. Corrosion rate of weathering steel and Q235 steel.

6
C. Han et al. Construction and Building Materials 411 (2024) 134284

Fig. 11. XRD pattern of rust layer in industrial environment.

Fig. 12. XRD pattern of rust layer in rural environment.

weathering steel and Q235 steel has no obvious regularity. layer develops into a cluster structure which is extremely compact, and
By comparing Figs. 11 and 12, it can be found that the phase com­ the lamellar products are basically disappeared. By comparing Fig. 13
positions of rust layer in industrial environment vary at a faster rate and Fig. 14, it can be seen that the products formed on Q235 steel after
compared with that in rural environment. This is because that the steels 24 h of corrosion are far more than those on weathering steel. After­
in industrial environment experience higher corrosion rate than that in wards, the rust layer of Q235 develops into a flower-shaped structure
rural environment. with lots of pores. The pore volume decreases with the prolonged
exposure time, and meanwhile cracks in the rust layer develop. By
720 h, the pore space has become relatively little and small, and the
3.3. Corrosion micromorphology analysis flower-shaped structure becomes dense. The variant process of rust layer
helps explain why the corrosion rates of the two steels decline over
Fig. 13 to Fig. 16 show the microscopic morphology of weathering exposure time. By contrasting Fig. 13(d) and Fig. 14(d), it is shown that
steel and Q235 steel after exposure to rural and industrial environments both the two steels have compact and dense rust layer after 720 h of
for 24, 168, 360, and 720 h, respectively. It can be observed that the corrosion, and thus the two steels show similar corrosion rates at the
corrosion products increase dramatically over time. And meanwhile, the later stage.
products play a role in keeping the steel substrate off external envi­ Figs. 14 and 15 show that in industrial environment, the weathering
ronment to some extent, leading to the decline of corrosion rate. and Q235 steel substrates are completely covered by the corrosion
It follows from Fig. 13 that in rural environment, only a few scattered products after 24 h. With prolonged exposure time, the products in­
lamellar products are formed on the weathering steel after 24 h. But it crease significantly, and meanwhile pores, pitting and cracks appear in
increases significantly over exposure time. By 168 h, the weathering the rust layer in succession. By 720 h, cluster structure appears in the
steel substrate has been completely covered by block-like and lamellar products of weathering steel, but it not so uniform and compact as that
products which shows a loose structure. Afterwards, the intermingled in rural environment. As for Q235 steel, the cracks appear in rust layer at
lamellar products gradually diminish, and the rust layer exhibits a initial stage and develop significantly over time. By 720 h, pores, long
flocculent structure. When the exposure time reaches 720 h, the rust

7
C. Han et al. Construction and Building Materials 411 (2024) 134284

Fig. 13. Surface morphology of rust layer on weathering steel in rural environment.

and wide cracks are observed in the rust layer. The pores and cracks reduction reaction of rust layer increases. On the contrary, in the anodic
provide channels for corrosive media (H2O, O2, and Cl− ) to pass process, the current density presents a decreasing trend, indicating that
through the rust layer and attack the steel substrate, and further accel­ the iron dissolution reaction is decreasing. For example, at the same
erate the steel corrosion. By contrasting Fig. 15(d) and Fig. 16(d), it can potential of − 0.2 V, the current density of weathering steel after 24 h
be found that under the industrial environment, the rust layer is much and 720 h of corrosion in industrial environment are 9.51 × 10− 3A.
more compact and denser than that on Q235 steel. This helps explain cm− 2 and 3.34 × 10− 3 A.cm− 2, respectively, indicating that with the
why the weathering steel exhibits lower corrosion rate than Q235 steel increase of exposure time, the iron dissolution reaction decreases. This
in the later stage. By contrasting Figs. 13–14 and Figs. 15–16, it is helps explain why the corrosion rates of the two steels decrease over
illustrated that the rust layers of the two steels in rural environment are exposure time.
more compact and denser than those in industrial environment. This Moreover, in both rural and industrial environments, the corrosion
helps explain why the steel in industrial environment shows higher potential of the weathering steel and Q235 steel increases over time,
corrosion rate as compared to that in rural environment. which illustrates the protective effect of rust layer is increasing. On the
other hand, it can be found that in both rural and industrial environ­
3.4. Electrochemical analysis ments, the weathering steel exhibits higher corrosion potential than
Q235 steel, indicating its superior corrosion resistance.
Figs. 17 and 18 illustrate the polarization curves of weathering steel
and Q235 steel in rural and industrial environments, in which “E” rep­ 3.5. Corrosion mechanism analysis
resents the electric potential, and “I” denotes the current density. It can
be seen that the electrochemical property of each steel sample is In atmospheric environments, gases such as oxygen, carbon dioxide,
depicted by two curves, namely, anodic (upward) polarization curve and water vapor, and liquid corrosion agents such as rain and dew are the
cathodic (downward) polarization curve, respectively. It follows from main factors leading to steel corrosion. Fig. 19 provides a detailed
Figs. 17 and 18 that in the equipotential case (E), the current density description for the formation and evolution of the corrosion products on
shows an increasing trend with the increase of exposure time in the steel surface. When the steel surface is covered by a liquid film, carbon
cathodic process, indicating that the rust layer involved in reduction or manganese at one location acts as cathode, while iron at another
reaction is increasing. For example, at the same potential of − 1.2 V, the location acts as anode, forming many microcells in steel. The redox re­
current density of weathering steel after 24 h and 720 h of corrosion in actions of weathering steel and Q235 steel in initial stage occur as fol­
rural environments are 1.21 × 10− 2 A.cm− 2 and 1.92 × 10− 2 A.cm− 2, lows [31,32]:
respectively, indicating that with the exposure time increasing, the

8
C. Han et al. Construction and Building Materials 411 (2024) 134284

Fig. 14. Surface morphology of rust layer on Q235 steel in rural environment.

Anode : Fe − 2e− →Fe2+ (4)

HSO−3 + O2 + H2 O + Fe2+ →FeOOH + SO2−4 (11)
Cathode : O2 +2H2 O + 4e− →4OH− (5) Figs. 20 and 21 show the corrosion evolution of weathering steel and
Q235 steel in industrial environment. It can be seen that the corrosion
Fe2+ +2OH− →Fe(OH)2 (6) products, as well as pitting and cracks, gradually appear on the surface
Next, the Fe 2+
undergoes hydrolysis and oxidation in the liquid film: of the two steels. The products can generally be divided into two layers
as shown Figs. 20–21. The outer rust layer is composed mainly of
1
Fe2+ + O2 + H2 O→Fe3+ + 2OH − (7) α-FeOOH, which is rather compact and offers an effective protection for
2 the steel substrate. The inner rust layer is made up of γ-FeOOH and
The Fe3+ reacts with H2O to form [Fe (H2O)6 3+] which then un­ Fe3O4/ γ- Fe2O3. Apparently, it is not so compact as the outer layer since
dergoes proton-transfer and hydrolysis to form [Fe (OH)n (H2O)6-n 3+] there are lot of pores.
(n = 1–6) It is reported that the protection provided by the corrosion film is
[ ] [ ] closely related to its porosity and thickness [34,35]. As shown in Figs. 20
Fe(H2 O)3+
6 + nH2 O→ Fe(OH)n (H2 O)3+
6− n + nH3 O
+
(8) and 21, the α-FeOOH mainly exists in the outer rust layer and protect the
After that, the product [Fe (OH)n (H2O)6-n 3+] undergoes deproto­ steel matrix from corrosion. But with the rust layer thickens, lots of
nation, oxygen conjugation and hydroxyl reaction, followed by nucle­ cracks and even exfoliation occurs on the outer rust layer, exposing the
ation and crystal growth, forming corrosion products γ- FeOOH which is inner rust layer to the corrosive environment. The inner rust layer be­
unstable. A proportion of this product can further convert into α-FeOOH comes the outer rust layer accordingly. As shown in Fig. 20, the inner
and Fe3O4, whereas others is dissolved, forming γ-Fe2O3 rust layer of Q235 has lots of pores, thus the corrosive medium such as
H2O and O2 can react with the steel matrix through the pores, and
2FeOOH→Fe2 O3 + H2 (9) further cause corrosion damage and result in the rust layer being looser.
However, different from the ordinary steel, the weathering steel con­
6FeOOH + 2e− →2Fe3 O4 + 2H2 O + 2OH− (10) tains the alloying elements Cu, Ni, which can form CuO, Ni compounds
On the other hand, the HSO-3 ion existing in industrial environment is during the corrosion process. As shown in Fig. 21, these compounds can
2+
prone to be oxidized and combine with Fe to form hydroxy iron oxide, fill the pores in inner rust layer of weathering steel, and thus helps to
as shown in Eq. (11). Therefore, the presence of NaHSO3 accelerates the enhance the density of the inner rust layer. This aids in the formation of
initial corrosion rate of steel [33]. a dense oxide layer that acts as a protective barrier, preventing the
diffusion of corrosion ions and dissolved oxygen to the substrate,

9
C. Han et al. Construction and Building Materials 411 (2024) 134284

Fig. 15. Surface morphology of rust layer on weathering steel in industrial environment.

thereby improving the corrosion resistance of weathering steel. weathering steel decrease by 7.4 % and 27.8 %, respectively. Moreover,
owing to the less corrosion pits and cracks, the weathering steel shows
less change in the yield strength and ultimate strength after corrosion as
3.6. Mechanical property analysis
compared to Q235 steel. For example, the ultimate strength of Q235
steel in rural and industrial environments after 720 h of corrosion de­
Figs. 22 and 23 show the stress-strain curves of weathering steel and
creases by 6 % and 18.5 %, respectively, while those of weathering steel
Q235 steel in rural and industrial environments. It can be observed that
decreases by 2.6 % and 13 %, respectively.
as the corrosion time is prolonged, both the weathering steel and Q235
Since the exposure time to industrial environment plays a crucial role
steel experience a certain decrease in yield strength and tensile strength,
in the material property of weathering steel, the relation between the
accompanied by a decrease in fracture strain and a shortened yield
exposure time and the mechanical property of weathering steel are
stage, especially under the industrial environment. For example, both
further explored herein. The yield platform, ultimate strain and yield
the two steels exhibit obvious yield platforms initially which then
ratio of weathering steel after corrosion is studied by the model as fol­
disappear after exposure to the industrial environment for more than
lows
168 h. The main cause for these variations is due to the occurrence of
corrosion pits on the steel samples, which further causes stress con­ εst
k1 = (12)
centration, leading to the degradation of mechanical properties. εy
In industrial environment, the impact of corrosion on the material
properties of weathering steel is primarily manifested in degradation of εu
k2 = (13)
fracture strain, while that on Q235 steel is primarily manifested in εy
degradation of strength. It is illustrated that the fracture strains of σu
weathering steel decrease by 1.3 %, 4.6 %, 14.6 %, and 27.8% after k3 = (14)
σy
exposure to the industrial environment for 24, 168, 360, and 720 h,
respectively. Likewise, the yield strength of Q235 steel decreases by 5.3 where k1, k2, and k3 are the shape control parameters of the stress-strain
%, 10 %, 13.2%, and 18.5 %, respectively, and the tensile strength de­ curve as shown in Fig. 24; εst is the strengthening strain; εy is the strain
creases by 2.2 %, 3.6 %, 5.1 %, and 10.5 %, respectively. Comparatively, corresponding to yield strength; εu denotes the ultimate strain; σy is the
the steels in rural environment undergo little change in mechanical yield strength; σu is the ultimate strength;.
properties, especially for the weathering steel. For example, the fracture Fig. 25 illustrates the change trends of k1, k2 and k3 with corrosion
strains of Q235 steel in rural and industrial environments after 720 h of time. It is revealed that the parameters k1, k2 and k3 have linearity
corrosion decrease by 6.4 % and 10.6 %, respectively, while those of

10
C. Han et al. Construction and Building Materials 411 (2024) 134284

Fig. 16. Surface morphology of rust layer on Q235 steel in industrial environment.

Fig. 17. Polarization curves of the two steels in rural environments.

relation with t. The parameter k1 is negatively correlated with t, while k2 = 95.24 + 0.00576t (16)
the parameters k2 and k3 increases slightly over t. The functional re­
lationships between these parameters (k1, k2 and k3) and corrosion time k3 = 1.23 + 0.00025t (17)
are obtained through linear regression as follows:
It follows from Figs. 22 and 23 that the stress-strain curves of
k1 = 9.3 − 0.0061t (15) weathering steel are similar to that of ordinary steel, but its hardening

11
C. Han et al. Construction and Building Materials 411 (2024) 134284

Fig. 18. Polarization curves of the two steels in industrial environment.

Fig. 19. Flow diagram of corrosion products.

Fig. 20. Corrosion process diagram for Q235 steel.

Fig. 21. Corrosion process diagram for weathering steel.

12
C. Han et al. Construction and Building Materials 411 (2024) 134284

Fig. 22. Stress-strain curves of weathering steel.

Fig. 23. Stress-strain curves of Q235 steel.

stage is more pronounced. Generally, the stress-strain curve of weath­ elastic-plastic model are only suitable for ordinary steel, and thus could
ering steel can be roughly divided into five stages: elastic stage, elastic- not give a proper description to the hardening stage of weathering steel.
plastic stage, plastic stage, hardening stage, and failure stage. By refer­ To further improve the reliability and accuracy of the model for
ring to the secondary plastic flow model proposed by Esmaeily [36] and weathering steel, a modification coefficient μ is proposed to modify the
the elastic-plastic model proposed by Han [37], the stress-strain curve of strengthening stage as shown in Eq. (19).
weathering steel after corrosion can be expressed as follows:
Es (1 − k3 ) ( )2 ( )
⎧ k3 σ y + μ 2
ε − k2 εy k1 εy ≤ ε ≤ k2 εy (19)
⎪
⎪ Eε (ε ≤ εe ) εy (k2 − k1 )
⎪ s
⎪
⎪
⎪ 2
⎪
⎪ − A ε + B ε + C ( εe ≤ ε ≤ ε1 ) ( )
⎪
⎪ ε − k1 εy ( )
⎪
⎨f (
ε1 ≤ ε ≤ k1 εy
) μ = 1− k1 εy ≤ ε ≤ k2 εy (20)
σ=
y
(18) k2 εy − k1 εy
⎪
⎪ Es (1 − k3 ) ( )2 ( )
⎪
⎪
⎪ k3 σy + ε − k2 εy k1 εy ≤ ε ≤ k2 εy Fig. 27 illustrates the comparison between the stress-strain curves
⎪
⎪ εy (k2 − k1 )2
⎪
⎪
⎪ obtained by the proposed model and experimental data. Clearly, the
⎪
⎩f
u (ε ≥ k ε )
2 2 proposed model can accurately capture the hardening stage of weath­
ering steel before and after corrosion.
where σ is the stress; ε is the strain; εy is the strain corresponding to yield Fig. 28 gives the stress-strain curves of weathering steel before and
strength; εe = 0.8fy/Es; ε1 = 1.5εe; fy is the yield strength; and fu is the after corrosion predicted by proposed model, as well as the curves
tensile strength. predicted by Han’s and Hollomon’s models. It is shown that the stress-
Fig. 26 gives the comparison of the stress-strain curves obtained by strain curves predicted by the proposed model exhibit better agree­
Eq. (18) and the test data. It is shown that the stress predicted by Eq. (18) ment with the test data in both rural and industrial environments.
is higher than the test results during the strengthening stage. This can be Therefore, it can be concluded that the proposed model can properly
attributed to the fact that both the secondary plastic flow model and the predict the stress-strain behavior of weathering steel before and after

13
C. Han et al. Construction and Building Materials 411 (2024) 134284

Fig. 24. Unidirectional loading curve of steel.

Fig. 25. The relation curves of parameter (k1, k2 and k3) with corrosion time.

Fig. 26. Comparison of stress-strain relation obtained by secondary plastic flow Fig. 27. Comparison of stress-strain relation between obtained by proposed
model and experimental data in industrial environment (weathering steel). model and experimental data in industrial environment (weathering steel).

14
C. Han et al. Construction and Building Materials 411 (2024) 134284

Fig. 28. Stress-strain curves of weathering steel obtained by constative models and test data.

corrosion. equation WAtn. The corrosion rates of the two steels decrease initially and then tend
to be steady. In industrial environment, the weathering steel consistently exhibits lower
corrosion rate than Q235 steel, and this advantage becomes more prominent with the
4. Conclusions
increase of exposure time. Whereas in the rural environment, the difference between
corrosion rates of two steels decreases dramatically over time and disappears after 360 h of
In this paper, the corrosion behavior and mechanical property
corrosion.
degradation of weathering steel in industrial and rural environments are
analyzed in comparison to that of Q235 steel. Based on this, the (2) The rust layers of weathering steel and Q235 steel have similar
following conclusions are drawn. compositions. The rust layer of steel in industrial environment
has more pores and cracks as compared to that in rural environ­
(1) –
–In both industrial and rural environments, the corrosion mass ment. In industrial environment, both the two steels have
losses of weathering steel and Q235 steel follow the empirical compact and dense rust layer after 720 h of corrosion. Whereas in

15
C. Han et al. Construction and Building Materials 411 (2024) 134284

the rural environment, the rust layer formed on weathering steel [9] M. Yamashita, H. Miyuki, Y. Matsuda, H. Nagano, T. Misawa, The long-term
growth of the protective rust layer formed on weathering steel by atmospheric
is more compact and denser than that on Q235 steel. This is
corrosion during a quarter of a century, Corros. Sci. 36 (2) (1994) 283–299,
because the Ni and Cu compound can fill the pores in the rust https://doi.org/10.1016/0010-938X(94)90158-9.
layer of weathering steel, causing higher corrosion resistance. [10] Q.C. Zhang, J.S. Wu, J.J. Wang, W.L. Zheng, J.G. Chen, A.B. Li, Corrosion behavior
(3) In both the industrial and rural environments, the impact of of weathering steel in marine atmosphere, Mater. Chem. Phys. 77 (2) (2003)
603–608, https://doi.org/10.1016/S0254-0584(02)00110-4.
corrosion on the mechanical property of weathering steel is pri­ [11] M. Morcillo, I. Díaz, H. Cano, B. Chico, D. de la Fuente, Atmospheric corrosion of
marily reflected in the degradation of fracture strain, while that weathering steels. Overview for engineers. Part II: testing, inspection,
on Q235 steel mainly manifests as a reduction in mechanical maintenance, Constr. Build. Mater. 222 (2019) 750–765, https://doi.org/10.1016/
j.conbuildmat.2019.06.155.
strength. [12] J.D. Fu, S. Wan, Y. Yang, Q. Su, W.W. Han, Y.B. Zhu, Accelerated corrosion
(4) A constitutive model is proposed for the weathering steel before behavior of weathering steel Q345qDNH for bridge in industrial atmosphere,
and after corrosion, and compared with the commonly used Constr. Build. Mater. 306 (2021), 124864, https://doi.org/10.1016/j.
conbuildmat.2021.124864.
constitutive model for ordinary steel. The results show that the [13] T. Misawa, K. Hashimoto, S. Shimodaira, The mechanism of formation of iron
proposed model can better predict the hardening stage of oxide and oxyhydroxides in aqueous solutions at room temperature, Corros. Sci. 14
weathering steel, making it more suitable for numerical (2) (1974) 131–149, https://doi.org/10.1016/S0010-938X(74)80051-X.
[14] T. Misawa, K. Asami, K. Hashimoto, S. Shimodaira, The mechanism of atmospheric
modeling. rusting and the protective amorphous rust on low alloy steel, Corros. Sci. 14 (4)
(1974) 279–289, https://doi.org/10.1016/S0010-938X(74)80037-5.
CRediT authorship contribution statement [15] M. Natesan, S. Selvaraj, T. Manickam, G. Venkatachari, Corrosion behavior of
metals and alloys in marine-industrial environment, Sci. Technol. Adv. Mater. 9 (4)
(2008), 045002, https://doi.org/10.1088/1468-6996/9/4/045002.
Wang Jiuyuan: Validation, Software, Investigation, Data curation. [16] M.C.L. de Oliveira, V.S.M. Pereira, O.V. Correa, R.A. Antunes, Corrosion
Han Chengshuo: Writing – original draft, Methodology, Investigation, performance of anodized AZ91D magnesium alloy: effect of the anodizing potential
on the film structure and corrosion behavior, J. Mater. Eng. Perform. 23 (2) (2014)
Formal analysis. Yang Xia: Writing – review & editing, Supervision, 593–603, https://doi.org/10.1007/s11665-013-0755-0.
Resources, Project administration, Funding acquisition, Formal analysis. [17] X.J. Yang, Y. Yang, M.H. Sun, J.H. Jia, X.Q. Cheng, Z.B. Pei, D. Xu, K. Xiao, X.G. Li,
Li Zhigang: Writing – review & editing, Software, Formal analysis, Data A new understanding of the effect of Cr on the corrosion resistance evolution of
weathering steel based on big data technology, J. Mater. Sci. Technol. 104 (2022)
curation.
67–80, https://doi.org/10.1016/j.jmst.2021.05.086.
[18] H.F. Isleem, D.S. Augustino, A.S. Mohammed, A.M. Najemalden, P. Jagadesh,
S. Qaidi, M.M.S. Sabri, Finite element, analytical, artificial neural network models
Declaration of Competing Interest for carbon fibre reinforced polymer confined concrete filled steel columns with
elliptical cross sections, Front. Mater. 9 (2023), 1115394, https://doi.org/
The authors declare that they have no known competing financial 10.3389/fmats.2022.1115394.
[19] B.A. Tayeh, M.H. Akeed, S. Qaidi, B.H.A. Bakar, Ultra-high-performance concrete:
interests or personal relationships that could have appeared to influence impacts of steel fibre shape and content on flowability, compressive strength and
the work reported in this paper. modulus of rupture, Case Stud. Constr. Mater. 17 (2022), e01615, https://doi.org/
10.1016/j.cscm.2022.e01615.
[20] Q.F. Liu, G.D. Pei, H.T. Hou, Y.Q. Chen, Probabilistic similarity of non-uniform
Data availability corrosion pattern between natural corrosion and accelerated experiment, Constr.
Build. Mater. 392 (2023), 132000, https://doi.org/10.1016/j.
Data will be made available on request. conbuildmat.2023.132000.
[21] GB/T 19746-2018. Corrosion of Metals and Alloys-alternate Immersion Test in Salt
Solution (in Chinese).
Acknowledgements [22] A.H. Chen, J.Q. XU, R. Li, H.L. Li, Corrosion resistance of high-performance
weathering steel for bridge building applications, J. Iron Steel Res. 19 (06) (2012)
59–63, https://doi.org/10.1016/S1006-706X(12)60128-9.
The author would like to acknowledge the financial support provided [23] I. Martínez, C. Andrade, Polarization resistance measurements of bars embedded in
by the Youth Project of Hubei Provincial Natural Science Foundation concrete with different chloride concentrations: EIS and DC comparison, Mater.
(2023AFB391) and the Young Talent Project of Hubei Provincial Corros. 62 (10) (2011) 932–942, https://doi.org/10.1002/maco.200905596.
[24] F. Corvo, T. Perez, L.R. Dzib, Y. Martin, A. Castañeda, E. Gonzalez, J. Perez,
Department of Education (Q20211305).
Outdoor–indoor corrosion of metals in tropical coastal atmospheres, Corros. Sci. 50
(2008) 220–230, https://doi.org/10.1016/j.corsci.2007.06.011.
References [25] W. Han, C. Pan, Z.Y. Wang, G.C. Yu, A study on the initial corrosion behavior of
carbon steel exposed to outdoor wet-dry cyclic condition, Corros. Sci. 88 (2014)
89–100, https://doi.org/10.1016/j.corsci.2014.07.031.
[1] S.M. Abd, I.S. Mhaimeed, B.A. Tayeh, H.M. Najm, S. Qaidi, Investigation of the use
[26] I. Diaz, H. Cano, D. de la Fuente, B. Chico, J.M. Vega, M. Morcillo, Atmospheric
of textile carbon yarns as sustainable shear reinforcement in concrete beams, Case
corrosion of Ni-advanced weathering steels in marine atmospheres of moderate
Stud. Constr. Mater. 18 (2023), e01765, https://doi.org/10.1016/j.cscm.2022.
salinity, Corros. Sci. 76 (2) (2013) 348–360, https://doi.org/10.1016/j.
e01765.
corsci.2013.06.053.
[2] Y.W. Liu, H.G. Zhao, Z.Y. Wang, Y.H. Wei, C. Pan, C.X. Lv, Corrosion behavior of
[27] D. de la Fuente, I. Díaz, J. Simancas, B. Chico, M. Morcillo, Long-term atmospheric
low-carbon steel and weathering steel in a coastal zone of the spratly islands: a
corrosion of mild steel, Corros. Sci. 53 (2) (2011) 604–617, https://doi.org/
tropical marine atmosphere, Int. J. Electrochem. Sci. 15 (7) (2020) 6464–6477,
10.1016/j.corsci.2010.10.007.
https://doi.org/10.20964/2020.07.48.
[28] S. Nasrazadani, A. Raman, Formation and transformation of magnetite (Fe3O4) on
[3] T.Y. Zhang, X.X. Xu, Yong Li, X.W. Lv, The function of Cr on the rust formed on
steel surfaces under continuous and cyclic water fog testing, Corrosion 49 (4)
weathering steel performed in a simulated tropical marine atmosphere
(1993) 294–300, https://doi.org/10.5006/1.3316052.
environment, Constr. Build. Mater. 277 (2021), 122298, https://doi.org/10.1016/
[29] T. Misawa, T. Kyuno, W. Suetaka, S. Shimodaira, The mechanism of atmospheric
j.conbuildmat.2021.122298.
rusting and the effect of Cu and P on the rust formation of low alloy steels, Corros.
[4] J.H. Dong, E.H. Han, W. Ke, Introduction to atmospheric corrosion research in
Sci. 11 (1) (1971) 35–48, https://doi.org/10.1016/S0010-938X(71)80072-0.
China, Sci. Technol. Adv. Mater. 8 (7–8) (2007) 559–565, https://doi.org/
[30] M. Natesan, S. Selvaraj, T. Manickam, G. Venkatachari, Corrosion behavior of
10.1016/j.stam.2007.08.010.
metals and alloys in marine-industrial environment, Sci. Technol. Adv. Mater. 9 (4)
[5] M. Morcillo, B. Chico, I. Díaz, H. Cano, D. de la Fuente, Atmospheric corrosion data
(2008), 045002, https://doi.org/10.1088/1468-6996/9/4/045002.
of weathering steels. A review, Corros. Sci. 77 (2013) 6–24, https://doi.org/
[31] U.R. Evans, The mechanism of rusting, Corros. Sci. 9 (11) (1969) 813–821, https://
10.1016/j.corsci.2013.08.021.
doi.org/10.1016/S0010-938X(69)80074-0.
[6] D. Wang, Q.D. Zhong, J. Yang, S.J. Zhang, Effects of Cr and Ni on the
[32] U.R. Evans, C.A.J. Taylor, Mechanism of atmospheric rusting, Corros. Sci. 12 (3)
microstructure and corrosion resistance of high-strength low alloy steel, J. Mater.
(1972) 227–246, https://doi.org/10.1016/S0010-938X(72)90671-3.
Res. Technol. 23 (2023) 36–52, https://doi.org/10.1016/j.jmrt.2022.12.191.
[33] D.D.N. Singh, S. Yadav, J.K. Saha, Role of climatic conditions on corrosion
[7] Z. Malaibari, R. Kahraman, H. Saricimen, A. Quddus, Investigation of atmospheric
characteristics of structural steels, Corros. Sci. 50 (1) (2008) 93–110, https://doi.
corrosion of mild steel after treatment by several inhibitor solutions, Corros. Eng.
org/10.1016/j.corsci.2007.06.026.
Sci. Technol. 42 (2007) 112–118, https://doi.org/10.1179/174327807×196825.
[34] Q.H. Zhao, W. Liu, Y.C. Zhu, B.L. Zhang, S.Z. Li, M.X. Lu, Effect of small content of
[8] Z.C. Feng, X.Q. Cheng, C.F. Dong, L. Xu, X.G. Li, Passivity of 316L stainless steel in
chromium on wet-dry acid corrosion behavior of low alloy steel, Acta Metall. Sin.
borate buffer solution studied by Mott–Schottky analysis, atomic absorption
(Engl. Lett. ) 30 (2) (2017) 164–175, https://doi.org/10.1007/s40195-016-0513-z.
spectrometry and X-ray photoelectron spectroscopy, Corros. Sci. 52 (11) (2010)
3646–3653, https://doi.org/10.1016/j.corsci.2010.07.013.

16
C. Han et al. Construction and Building Materials 411 (2024) 134284

[35] S. John, A. Salam, A.M. Baby, A. Joseph, Corrosion inhibition of mild steel using [37] L.H. Han, Y.F. An, Performance of concrete-encased CFST stub columns under axial
chitosan/TiO2 nanocomposite coatings, Prog. Org. Coat. 129 (2019) 254–259, compression, J. Constr. Steel Res. 93 (2014) 62–76, https://doi.org/10.1016/j.
https://doi.org/10.1016/j.porgcoat.2019.01.025. jcsr.2013.10.019.
[36] A. Esmaeily, Y. Xiao, Behavior of reinforced concrete columns under variable axial
loads: analysis, Acids Struct. J. 102 (5) (2005) 736–744, https://doi.org/10.1016/
j.engstruct.2012.12.040.

17