materials

Article
Corrosion Behavior of 3104 Aluminum Cans When Used
as Packaging for Chinese Liquor
Mingjie Fan 1, * , Jinyang Chen 1, * , Jie Gu 2 and Zheying Wu 2

1 Chemical Engineering and Technology, Shanghai University, Shanghai 200444, China

2 Shanghai Bao Steel Packaging Co., Ltd., Shanghai 200949, China
* Correspondence: [email protected] (M.F.); [email protected] (J.C.); Tel.: +86-21-6613-7729 (J.C.)

Abstract: Aluminum cans are commonly used for packaging soft drinks and low-alcohol beverages
due to their good recyclability. To enhance the economic cycle and expand the packaging of liquors,
the feasibility of commercial 3104 aluminum cans for packaging Chinese liquor was studied. The
aluminum’s migration into alcoholic solutions was studied using inductively coupled plasma emis-
sion spectroscopy (ICP-OES). Electrochemical impedance spectroscopy (EIS) was used to study the
corrosion process of epoxy coatings on the aluminum cans. Scanning electron microscopy (SEM),
energy dispersive X-ray spectroscopy (EDS), infrared attenuated total reflection (IR-ATR), and X-ray
diffraction (XRD) were used to determine the inner coatings and adhering surfaces of the cans and
the corrosion process. The results showed that the maximum aluminum migration in Chinese liquor
was 4.3572 mg/kg at 60 ◦ C for 30 days. The epoxy coating was corroded enough to decrease the
coating impedance and expose the metal substrate after 25 days. Permeation and aging degradation
of coatings are the main factors to consider when packaging liquor.

Keywords: aluminum cans; Chinese liquor; corrosion; migration; EIS

1. Introduction
Aluminum is widely used for food packaging because of its low density, high me-
chanical strength, high recycling rate, and corrosion resistance. To date, metal cans have
become a popular form of packaging around the world and more than 250 billion cans are
used every year [1], and aluminum cans have gradually become the main type of metal
Citation: Fan, M.; Chen, J.; Gu, J.; Wu,
beverage cans. At present, aluminum cans are mainly used to package water-based soft
Z. Corrosion Behavior of 3104
drinks and low-alcohol beverages such as carbonated drinks and beer. It is known that
Aluminum Cans When Used as
aluminum alloys exhibit corrosion in anhydrous ethanol [2,3] and its water solutions [4–7].
Packaging for Chinese Liquor.
Materials 2024, 17, 3884. https://
Similarly, the original corrosion resistance of aluminum cans may be altered when they
doi.org/10.3390/ma17163884
are in contact with liquor for long time and exposed to the combining effects of multiple
complex components [8]. However, there are few studies about the use of aluminum cans
Received: 27 June 2024 for the packaging of liquors.
Revised: 15 July 2024
In order to reduce the corrosion of metal in contact with beverages, commercial
Accepted: 24 July 2024
aluminum cans are coated with a polymer coating as a food contact material (FCM) [9].
Published: 6 August 2024
These anti-corrosive coatings are complex mixtures that consist of vinyl or phenolic lacquers
and epoxy resins [10]. Epoxy resins are very important because of their heat resistance,
mechanical strength, and corrosion resistance [11]. As a thermosetting polymer, epoxy
Copyright: © 2024 by the authors.
resins can form strong chemical bonds on the surface of metal cans. Although cans have
Licensee MDPI, Basel, Switzerland. this coating as a surface barrier, corrosion sometimes still occurs as a result of beverages.
This article is an open access article The liquid diffuses into the polymer layer and affects the packaging barrier [12]. After this
distributed under the terms and liquid permeation occurs and the coating swells, corrosion and leakage of the aluminum
conditions of the Creative Commons can occur, and excessive aluminum migrating into beverages can cause health hazards such
Attribution (CC BY) license (https:// as Alzheimer’s disease, Parkinson’s syndrome, and multiple sclerosis [13].
creativecommons.org/licenses/by/ The corrosion processes of metal-coated cans are usually estimated using electrochem-
4.0/). ical techniques [14], and electrochemical impedance spectroscopy (EIS) is a relatively fast

Materials 2024, 17, 3884. https://doi.org/10.3390/ma17163884 https://www.mdpi.com/journal/materials

Materials 2024, 17, 3884 2 of 10

and reliable technique for the evaluation of protective coatings. The corrosion resistance of
the inner surface of aluminum cans for guarana, cola-flavored, and tonic water soft drinks
were evaluated with EIS by Esteves et al. [15]. The degradation of aluminum can coatings
for Green Cola and Red Bull cans was examined by Almoiqli et al. [16].
The main purpose of this study was to determine the feasibility of 3104 aluminum alloy
cans for packaging Chinese liquor and to study the related corrosion processes. Inductively
coupled plasma emission spectroscopy (ICP-OES) and EIS were used to determine the leak-
age of aluminum and the corrosion of the cans. In addition, scanning electron microscopy
and X-ray microanalysis (SEM-EDS), infrared attenuated total reflection (IR-ATR), and
X-ray diffraction spectroscopy (XRD) were employed to measure the internal coatings and
the adhering surfaces of the cans. The corrosion mechanism of the liquor was determined.

2. Materials and Methods

2.1. Samples and Reagents
3104 aluminum alloy-coated cans were supplied by Shanghai Baosteel Packaging
Co., Ltd. (Shanghai, China), and 52% vol. strong-flavored Chinese liquor was purchased
from Harbin Junmin Liquor Co., Ltd. (Harbin, Heilongjiang Province, China). Nitric
acid (GR, 65–68%) was obtained from Shanghai Sinopharm Chemical Reagent Co., Ltd.
(Shanghai, China). Anhydrous ethanol (AR, 99.7%) was obtained from Shanghai Boer
Chemical Reagent Co., Ltd. (Shanghai, China). Al standard solution (1000 mg/L) was
acquired from Tanmo Quality Inspection Technology Co., Ltd. (Beijing, China).

2.2. Migration Test of Aluminum Cans

According to EU No. 10/2011 [17] and GB 31604.1-2015 [18] criteria, 20% ethanol,
50% ethanol, 95% ethanol, and 52% Chinese liquor were selected as the food-packaging
simulants. They were poured into the aluminum-coated cans, and the cans were sealed
and then put into a thermostatic oven at 60 ◦ C. Test times were set at 5, 10, 15, 20, 25,
and 30 days. The condition of 10 days at 60 ◦ C corresponds to long-term storage above
6 months at room temperature and below [17,18]. Based on the appropriate extension of
this specific migration condition, the condition of 30 days at 60 ◦ C is about equivalent to a
simulation of 18 months of shelf life at room temperature.
After migration, the solutions were evaporated to dryness on a thermostatic elec-
trothermal plate, subsequently dissolved and fixed with 2% aqueous nitric acid solution,
and then filtered for ICP-OES analysis. Aluminum cans soaked for different lengths of time
were used for the subsequent characterization and electrochemical impedance spectroscopy.
Aluminum migration was calculated as M = (Ct − C0 ) × V/m, where M is the amount
of aluminum migration in the medium (µg/kg); Ct and C0 are the concentrations of the
migration and blank samples (µg/L); V is the volume of the migration sample (mL); and m
is the mass of the sample of the migration solution before digestion (g).

2.3. Characterization and Analysis

The Al concentration was analyzed with ICP-OES (Optima 8000, PerkinElmer, Waltham,
MA, USA). Electrochemical impedance spectroscopy (EIS) of the soaked aluminum can
was measured using an electrochemical analysis workstation (CHI604E, CH Instruments,
Shanghai, China). The morphology of the aluminum cans during corrosion was examined
with scanning electron microscopy (SU1510, Hitachi, Tokyo, Japan). IR-ATR (Nicolet 380,
ThermoFisher, Waltham, MA, USA) was used to study the changes in the properties of
the epoxy coating. X-ray diffraction (D2 Phaser, Bruker, Billerica, MA, USA) was used to
determine the changes in the corrosion composition of the cans.
The operation parameters of the ICP-OES were as follows: RF power: 1250 W; plasma
gas flow rate: 15.0 L/min; auxiliary gas flow rate: 0.9 L/min; nebulizer gas flow rate:
0.85 L/min; sample flow rate: 1.5 mL/min; and Al wavelength: 396.153 nm.
EIS measurements were performed to obtain a stable open-circuit potential by applying
a sinusoidal voltage of ±20 mV in a frequency range of 10−2 to 105 . A three-electrode cell
 The operation parameters of the ICP-OES were as follows: RF power: 1250 W; plasma
gas flow rate: 15.0 L/min; auxiliary gas flow rate: 0.9 L/min; nebulizer gas flow rate: 0.85
Materials 2024, 17, 3884
L/min; sample flow rate: 1.5 mL/min; and Al wavelength: 396.153 nm.
3 of 10
EIS measurements were performed to obtain a stable open-circuit potential by apply-
ing a sinusoidal voltage of ±20 mV in a frequency range of 10−2 to 105. A three-electrode
cell system was constructed at room temperature, containing a saturated calomel elec-
system was constructed at room temperature, containing a saturated calomel electrode as a
trode as a reference electrode, a platinum electrode as a counter electrode, and the alumi-
reference electrode, a platinum electrode as a counter electrode, and the aluminum can as
num can as the working electrode.
the working electrode.
3.
3. Results
Results and
and Discussion
Discussion
3.1.
3.1. Aluminum
Aluminum Migration
Migration
Inductively
Inductivelycoupled
coupledplasma
plasmaoptical
opticalemission
emissionspectroscopy
spectroscopy (ICP-OES)
(ICP-OES) cancanbe be
used to
used
to determine
determine thethe concentration
concentration of elemental
of elemental components
components in samples,
in samples, and and
it wasit was
usedused in
in this
this study
study to determine
to determine the migration
the migration of aluminum.
of aluminum. The results
The results are given
are given in Tablein 1Table 1 and
and show
showthe
that that the migration
migration of aluminum
of aluminum increased
increased with longer
with longer soaking
soaking time.time.
As forAstheforthree
the three
dif-
different aqueous ethanol solutions, the maximum migration of aluminum
ferent aqueous ethanol solutions, the maximum migration of aluminum was observed in was observed
in the
the 50%50% ethanol,
ethanol, followed
followed byby the
the 20%ethanol,
20% ethanol,and
andthe
theleast
leastmigration
migrationof ofaluminum
aluminum was was
observed in
observed in the
the 95%
95% ethanol.
ethanol. However,
However, the the migration
migration ofof aluminum
aluminum in in the
the Chinese
Chinese liquor
liquor
was much
was much higher
higher than
than that
that observed
observed in in the
the other
other simulants.
simulants. This
This result
result may
may be be attributed
attributed
to the
to the complex
complex composition
composition of of various
various trace
trace compounds
compounds contained
contained in in Chinese
Chinese liquor
liquor [19].
[19].

Aluminum migration
Table 1. Aluminum
Table migration (µg/kg)
(µg/kg)under
underdifferent
differentconditions.
conditions.

Aluminum Migration
Aluminum(μg/kg)
Migration (µg/kg)
Time (d) Ethanol
Time (d) Ethanol
Ethanol Ethanol
Ethanol Ethanol Chinese Liquor
Chinese Liquor
20% (v/v) 20%
50%(v/v)
(v/v) 50% (v/v)
95% (v/v) 95% (v/v) 52% (v/v)
52% (v/v)
10 12.16 10 17.76
12.16 17.76 2.56 2.56 387.39387.39
30 13.89 30 13.89
24.02 24.02 7.07 7.07 4357.24357.2

To further understand
To further understandthe theinteraction
interactionbetween
betweencans
cansand
andChinese
Chinese liquor,
liquor, thethe migra-
migration
tion of aluminum
of aluminum in the
in the actual
actual liquor
liquor waswas studied
studied acrossmore
across moretime
timepoints
points(Figure
(Figure1).
1). The
The
results show that the migrated amount of aluminum increased with
results show that the migrated amount of aluminum increased with time. The maximum time. The maximum
migrated
migrated amount
amount ofof aluminum
aluminum in the Chinese liquor was about 4.3572 mg/kg, mg/kg, which
which is
is
equivalent
equivalent to an aluminum
aluminum concentration
concentrationofofabout
about4 4mg/L
mg/Lininthe
themigration
migrationsample.
sample.This
This
is
is much
much higherthan
higher thanthe
theaccepted
acceptedminimum
minimumcontamination
contaminationlevel
levelof
of0.2
0.2mg/L
mg/L concentration
concentration
outlined
outlined by the theAmerican
AmericanAgency
Agencyfor forToxic
Toxic Substances
Substances andand Disease
Disease Registry
Registry (ATSDR)
(ATSDR) [20].
[20].
This This indicates
indicates that Chinese
that Chinese liquor liquor
storedstored in aluminum
in aluminum cans cans has serious
has serious healthhealth
risks.risks.

Figure 1. The aluminum

1. The aluminum migration
migration in
in 52%
52% Chinese
Chinese liquor
liquor at
at 60 ◦ C at different times.
60 °C
Figure at different times.

3.2. Corrosion Morphology of Aluminum Cans

Figure 2 shows the surface morphology of the coated sample cans after being soaked in
Chinese liquor for different lengths of time, as observed by SEM. The surface coating of the
unsoaked cans was relatively flat, whereas after 5 days, circular bubble precursors emerged
 3.2. Corrosion Morphology of Aluminum Cans
Figure 2 shows the surface morphology of the coated sample cans after being soaked
Materials 2024, 17, 3884 4 of 10
in Chinese liquor for different lengths of time, as observed by SEM. The surface coating of
the unsoaked cans was relatively flat, whereas after 5 days, circular bubble precursors
emerged between the coating and the surface of the aluminum alloy without apparent
between the
cracking coating
of the andAfter
coating. the surface
soakingof for
themore
aluminum alloy
than 10 without
days, apparent
the surface cracking
of the coatedofcans
the
coating. After soaking for more than 10 days, the surface of the coated cans became
became rougher with localized bulging and cracking. After 30 days of soaking, a corroded rougher
with was
area localized
clearlybulging
visibleand cracking.
on the surfaceAfter 30sample.
of the days of The
soaking, a corroded
coating area was
lost its basic clearly
protection,
visible on the surface of the sample. The coating lost its basic protection, and the aluminum
and the aluminum alloy substrate developed widespread exposure and numerous small
alloy substrate developed widespread exposure and numerous small pits.
pits.

Figure 2. SEM images of the internal surfaces of aluminum cans after different durations of soaking:
Figure 2. SEM images of the internal surfaces of aluminum cans after different durations of soaking:
(a) 0, (b) 5, (c) 10, and (d) 30 days in Chinese liquor.
(a) 0, (b) 5, (c) 10, and (d) 30 days in Chinese liquor.
Table 2 shows the EDS analysis of the sample cans after soaking for 10 and 30 days
Table 2 shows the EDS analysis of the sample cans after soaking for 10 and 30 days in
in Chinese liquor. The content of the can soaked for 10 days was Al = 9.0, C = 81.5, O = 9.6,
Chinese liquor. The content of the can soaked for 10 days was Al = 9.0, C = 81.5, O = 9.6,
and
and the
the presence
presence of
of the
the epoxy
epoxy coating
coating resulted
resulted in
in aa higher
higher percentage
percentage of
of carbon
carbon content.
content.
After
After 30
30 days
days of
of corrosion,
corrosion, the
the aluminum
aluminum content
content drastically
drastically increased
increased due
due to
to the
the damage
damage
of
of the
the coating
coating (Al
(Al == 38.1,
38.1, C
C ==19.5,
19.5,OO==42.4),
42.4),and
andthe
theincrease
increasein
inthe
theoxygen
oxygen content
content indicates
indicates
the formation of corrosive substances.
the formation of corrosive substances.
Table 2. EDS
Table 2. EDS analysis
analysis of
of aluminum-coated cans after
aluminum-coated cans after soaking
soaking for
for 10
10 and
and 30
30 days
days in
in Chinese
Chinese liquor.
liquor.

Element (At%) Element (At%)

Time (d) Time (d)
Carbon Oxygen
Carbon
Aluminum Aluminum
Oxygen
Total Total
10 81.5 9.6 9.0 100.0
10 81.5 9.6 9.0 100.0
30 19.5 30 42.4
19.5 42.4 38.1 38.1 100.0 100.0

Furthermore, the cross-sectional structure of the aluminum cans during the corrosion
Furthermore,
process the cross-sectional
with Chinese liquor is givenstructure of the
in the SEM aluminum
images belowcans during
(Figure 3). the corrosion
It can be ob-
process with Chinese liquor is given in the SEM images below (Figure 3). It can
served that the thickness of the can was about 100 µm and the layer depth of the epoxy be ob-
served that the thickness of the can was about 100 µm and the layer depth of the epoxy
coating (FCM) was about 20 µm. The arrows show the states of the epoxy coating. At
first, the inter-layer connection between the coating and the metal substrate was tight.
After 10 days of soaking, a gap existed between the coating and the alloy substrate, and
there were granular corrosion products encapsulated inside the bulging position. In this
 coating (FCM) was about 20 µm. The arrows show the states of the epoxy coating. At first,
Materials 2024, 17, 3884 the inter-layer connection between the coating and the metal substrate was tight. After 5 of 10 10
days of soaking, a gap existed between the coating and the alloy substrate, and there were
granular corrosion products encapsulated inside the bulging position. In this case, corro-
case,
sion corrosionthrough
happened happened through infiltration,
infiltration, and the volumetric
and the volumetric expansion expansion
stress ofstress of the
the corrosion
corrosion products accelerated the damage of the coating. After 30 days, the cross-section
products accelerated the damage of the coating. After 30 days, the cross-section became
became
coarser coarser
and and the inter-layer
the inter-layer gap wasThe
gap was obvious. obvious.
coatingThe
fellcoating fellform
off in the off inofthe form
small of
debris.
small debris. Thus, the coating was no longer integral in covering
Thus, the coating was no longer integral in covering the substrate. the substrate.

Figure 3. Cross-sectional morphologies of aluminum-coated cans soaked for (a) 0, (b) 10, and (c) 30
Figure 3. Cross-sectional morphologies of aluminum-coated cans soaked for (a) 0, (b) 10, and (c) 30
days in Chinese liquor.
days in Chinese liquor.

3.3.
3.3.Change
ChangeininEpoxy
EpoxyCoating
Coating
Figure44gives
Figure givesthe
the IR
IR spectra of thethe epoxy
epoxyresin resinbefore
beforeand
andafter soaking
after soakingforfor
30 30
days
days
in Chinese liquor. The band at 2925 cm −−11 is attributed to the stretching vibration of the
in Chinese liquor. The band at 2925 cm is attributed to the stretching vibration of the
aliphaticmethylene
aliphatic methylene (-CH
(-CH22-),
-), and
and the
the low-intensity
low-intensitypeaks peakslocated
located at at
2960 andand
2960 2870 cm−cm
2870
1 −1

indicatethe
thestretching
stretchingmodes
modes of of aliphatic
aliphatic methyl −1
indicate methyl groups
groups(-CH
(-CH33).).The
Thepeak
peakatat1720
1720cm cm−1 is
is stretching
the the stretching vibration
vibration peak peak of carbonyl
of the the carbonylgroupsgroups (C=O).
(C=O). The The stronger
stronger absorption
absorption bands
bands at 1607, 1507, and 1456 cm−1 are the vibrations of the aromatic benzene ring, and
at 1607, 1507, and 1456 cm are the vibrations of the aromatic
−1 benzene ring, and the bend-
the bending of the hydroxyl groups is seen at 1294 cm−1 . The stretching vibration peaks of
ing of the hydroxyl groups is seen at 1294 cm −1. The stretching vibration peaks of the aro-
the aromatic ether are at 1231 and 1180 cm−1 , while the stretching mode of alkyl ether is
matic ether −are at 1231 and 1180 cm−1, while the stretching mode of alkyl ether is at 1037
at 1037 cm . The stretching of the epoxy groups appears at 945 and 770 cm−1 [21]. C-H
1
cm −1. The stretching of the epoxy groups appears at 945 and 770 cm−1 [21]. C-H aromatic
aromatic bending patterns can be noticed at 826, 736, and 556 cm . − 1
bending patterns the
Comparing caninfrared
be noticed at 826,
spectra 736,and
before andafter
556 soaking,
cm−1. it was found that the inten-
sity of the vibrational absorption peaks of the aromatic benzene ring at 1507 cm−1 was
significantly weakened; the intensity of the aromatic ether and the alkyl ether stretching
peaks at 1231 and 1180 cm−1 , as well as that of the alkyl ether stretching absorption peak
at 1037 cm−1 , was reduced; and there was an attenuation of the C-H aromatic bending
absorption at 826 cm−1 . This indicates that soaking with liquor has a disruptive impact on
the polymeric structure of epoxy resin, and leads to degradation.
Materials 2024, 17, x FOR PEER REVIEW 6 of 11

Materials 2024, 17, 3884 6 of 10

Figure 4.Figure 4. The

The IR-ATR IR-ATR
spectra spectra
of epoxy of epoxy
coating coating
before before
and after andfor
soaking after soaking for 30 days.
30 days.

3.4. Change
Comparing the in Composition
infrared spectra of Aluminum
before Cans
and after soaking, it was found that the inten-
sity of the vibrational absorption peaks of the aromatic
Figure 5 shows the XRD patterns of the aluminum benzene ringcansat 1507 cm−1and
before wasafter soaking for
significantly weakened; the intensity of the aromatic ether and the alkyl ether stretching
30 days in Chinese liquor. There was no significant change between the locations of all of
peaks atthe
1231 and 1180
strong cm−1,peaks
feature as wellprior
as that
to of the alkyl and
corrosion ethertheir
stretching absorption
locations peak
after corrosion. The peak
at 1037 cm −1, was reduced; and there was an attenuation of the C-H aromatic ◦
signals with 2-theta angles of 38.5, 45.5, 65.2, and 78.2 were attributed bending ab-to metallic pure
sorption at 826 cm−1. This indicates that soaking with liquor has a disruptive impact on
aluminum (JCPDS No. 89-4037). For the above-indexed signal patterns, there were three
the polymeric structure of epoxy resin, and leads to degradation.
different states of aluminum phases: diffraction peaks with 2-theta angles of 38.5 and
65.2in◦ Composition
3.4. Change were attributed to boehmite
of Aluminum Cans (AlOOH); those with 2-theta angles of 65.2 and 78.2
◦

were identified as face-centered cubic aluminum; and that with a two-theta angle of 45.5◦
Figure 5 shows the XRD patterns of the aluminum cans before and after soaking for
belonged to cubic γ-Al2 O3no[22].
30 days in Chinese liquor. There was
Moreover, there were some slight diffraction signals
significant change between the locations of all of
between the positions of the (111) and
the strong feature peaks prior to corrosion and their (200)locations
crystal planes, which were
after corrosion. related to the effects
The peak
signals with 2-theta angles of 38.5, 45.5, 65.2, and 78.2° were attributed to metallic pure3104 aluminum
of impurities such as manganese, magnesium, iron, and copper in the
aluminum
Materials 2024, 17, x FOR PEER REVIEW
alloy [23]. No.
(JCPDS The89-4037).
intensityForofthe
theabove-indexed
(220) and (311) crystal
signal plane
patterns, peaks
there of7three
were the soaked samples
of 11
differentexperienced a dramatic
states of aluminum phases:decline compared
diffraction to those
peaks with ofangles
2-theta the unsoaked
of 38.5 andaluminum
65.2° cans. This
was duetotoboehmite
were attributed corrosion during the
(AlOOH); thosesoaking process.
with 2-theta angles of 65.2 and 78.2° were
identified as face-centered cubic aluminum; and that with a two-theta angle of 45.5° be-
longed to cubic γ-Al2O3 [22]. Moreover, there were some slight diffraction signals between
the positions of the (111) and (200) crystal planes, which were related to the effects of im-
purities such as manganese, magnesium, iron, and copper in the 3104 aluminum alloy
[23]. The intensity of the (220) and (311) crystal plane peaks of the soaked samples experi-
enced a dramatic decline compared to those of the unsoaked aluminum cans. This was
due to corrosion during the soaking process.

Figure
Figure 5. The
5. The XRD XRD patterns
patterns of the aluminum
of the aluminum cans beforecans before
and after and after
corrosion corrosion
for 30 days. for 30 days.

3.5. Electrochemical Impedance Spectroscopy

EIS was used to determine the anti-corrosion process of epoxy resin coatings treated
with Chinese liquor and soaked at 60 °C for 30 days. Figure 6 shows the Nyquist and Bode
impedance spectra of 3104 aluminum alloy cans soaked in Chinese liquor for 30 days.
Materials 2024, 17, 3884 7 of 10

3.5. Electrochemical Impedance Spectroscopy

EIS was used to determine the anti-corrosion process of epoxy resin coatings treated
with Chinese liquor and soaked at 60 ◦ C for 30 days. Figure 6 shows the Nyquist and
Bode impedance spectra of 3104 aluminum alloy cans soaked in Chinese liquor for 30 days.
Within the first 5 days of immersion, the Nyquist plot shows an incomplete depressed
semicircular arc with a large diameter and the Bode plot exhibits a curve with a slope close
to −1, and the low-frequency modulus was recorded to be 5.55 × 108 Ω·cm2 . At this time,
Materials 2024, 17, x FOR PEER REVIEW 8 of 11
the organic coating was relatively intact and could be considered as an isolation layer with
a large resistance value and a small capacitance value.

Figure
Figure 6. EIS
6. EIS plots
plots of aluminum-coatedcans
of aluminum-coated canssoaked
soakedininChinese
Chineseliquor:
liquor: (a)
(a) Nyquist
Nyquist plot;
plot;(b)
(b)Bode
Bode plot.
plot.
As the soaking time increases, the overall diameter of the capacitive arc of the Nyquist
The bubbling of the coating during the initial 20 days suggests two time-constant
plot decreases continuously, and the low-frequency impedance modulus value of the
characteristics. Nevertheless, the Nyquist plots kept a larger depressed semicircular arc,
corresponding Bode plot also continues to decrease. The curves of each Bode plot nearly
without a clear two-segment capacitance arc for the two time constants. This can be as-
overlap each other in the high-frequency range, transiting to a relatively smooth plateau in
cribed to the overlapping of the small and large capacitive arcs, making it difficult to dis-
the tinguish
low-frequency
the two range.
time constants [26]. One of the capacitive impedance arcs in the high-
It is significant
frequency that the
range reflects a straight
propertyline superimposed
of the epoxy coating, at 45◦the
while to capacitive
both axesimpedance
following the
semicircle
arc in the oflow-frequency
the capacitance arcmay
range appeared
be relatedin to
thethe
Nyquist
corrosionplot after on
reaction soaking for 25
the surface of days,
which
the corresponded
aluminum alloy. to the process of substance diffusion [24]. The appearance of a diffusion
tail at low
Afterfrequencies means the
25 days of soaking, thatimpedance
the coating plotswas
beganbroken, anda combination
to exhibit thus the liquor of theeasily
capacitive
contacted thearc and diffusion
surface tail. The capacitive
of the aluminum arc at
alloy, and thehigher frequencies
electrode processand the dominated
was diffusion by
the impedance
metal substrate tail at [25].
lowerWhen
frequencies
the cancanwasbe attributed
soaked intoliquor
the charge
for 30transfer
days, theprocess of the
low-frequency
metal substrate
modulus in the Bode andplotthe diffusion
dropped process of the corrosion
to a minimum of 1.60 ×substances,
105 Ω·cmrespectively
2 , indicating [27].
that the
The electrochemical impedance curves for two time phases can be illustrated
coating lost its anti-corrosion effect, which corresponded to the macroscopic appearance of by the equiv-
alent circuits in Figure 7 [28].
partial pitting corrosion.
The bubbling of the coating during the initial 20 days suggests two time-constant
characteristics. Nevertheless, the Nyquist plots kept a larger depressed semicircular arc,
without a clear two-segment capacitance arc for the two time constants. This can be ascribed
to the overlapping of the small and large capacitive arcs, making it difficult to distinguish
the two time constants [26]. One of the capacitive impedance arcs in the high-frequency
range reflects the property of the epoxy coating, while the capacitive impedance arc in
the low-frequency range may be related to the corrosion reaction on the surface of the
aluminum alloy.
After 25 days of soaking, the impedance plots began to exhibit a combination of the
capacitive arc and diffusion tail. The capacitive arc at higher frequencies and the diffusion
Figure 7. Equivalent circuits proposed for aluminum-coated cans during different phases of soak-
impedance tail at lower frequencies can be attributed to the charge transfer process of the
ing: (a) phase one; (b) phase two.
metal substrate and the diffusion process of the corrosion substances, respectively [27]. The
electrochemical impedance
For a quantitative curves for
evaluation of two time phases
the coating can be illustrated
performance, the determinedby the equivalent
electro-
circuits in Figure
chemical 7 [28].spectra were further analyzed by using the established equivalent
impedance
physical models. The values of the electrochemical parameters of the fitted equivalent
components are shown in Table 3. These circuit elements are interpreted as the solution
resistance (Rs), coating interface capacitance (Qc), electric double-layer capacitance (Qdl),
resistance of the conductive path through the coating (Rc), charge transfer resistance (Rct),
and diffusion (Warburg) impedance (Zw), respectively. Figure 8 shows the change in each
 the aluminum alloy.
After 25 days of soaking, the impedance plots began to exhibit a combination of the
capacitive arc and diffusion tail. The capacitive arc at higher frequencies and the diffusion
impedance tail at lower frequencies can be attributed to the charge transfer process of the
Materials 2024, 17, 3884 metal substrate and the diffusion process of the corrosion substances, respectively
8 of 10
[27].
The electrochemical impedance curves for two time phases can be illustrated by the equiv-
alent circuits in Figure 7 [28].

Figure 7. Equivalent circuits proposed for aluminum-coated cans during different phases of soak-
7. Equivalent circuits proposed for aluminum-coated cans during different
Figure
Materials 2024, 17, x FOR PEER REVIEW 9 phases
of 11 of soaking:
ing: (a) phase one; (b) phase two.
(a) phase one; (b) phase two.
For a quantitative evaluation of the coating performance, the determined electro-
For a quantitative evaluation of the were coating performance, the determined electro-
essential parameterchemical
related toimpedance spectra
the coating/substrate with further analyzed
the soaking time.by Forusing
30 daysthe in
established equivalent
chemical impedance physicalspectra
models. wereThe further
values analyzed
of the by using
electrochemical
liquor, the electrical resistance of the epoxy coating continued to decrease, while its inter- the established
parameters of equivalent
the fitted equivalent
physical
facial models.
capacitance The values
components
continued areof
to increase. theIt electrochemical
shown isinindicated
Table 3. that
These theparameters
circuit
liquorelements of the
continuously fitted
infil- equivalent
are interpreted as the solution
components
trated areresistance
into the coating shown
throughin(Rthe
Table 3. These
s), coating
micropores interfacecircuit
caused byelements
capacitance (Qare interpreted
c), electric
the manufacturing as the capacitance
double-layer
process, and solution (Qdl),
resistance
the enlarged(R resistance
s ), coating
porosity of theweakened
interface
gradually conductive
capacitance thepath (Qthrough theeffect.
c ), electric
anti-corrosion coating (Rcdecrease
double-layer
The ), charge transfer
capacitance
in resistance
(Qdl ), (Rct),
charge transfer
resistance of the and
resistancediffusion (Warburg)
and thepath
conductive rise through impedance
in double-layer (Zw), respectively.
capacitance
the coating (Rc ), may
charge Figure
be related
transfer8toshows
the the change
resistance (Rct ), in each
regional exposure
and diffusion of the aluminum
(Warburg) impedance alloy substrate and the further
(Zw ), respectively. corrosion
Figure 8 showsof thethe
oxide
change in each
layer on the aluminum surface.
essential parameter related to the coating/substrate with the soaking time. For 30 days
in liquor,
Table the electrical
3. Parameter resistance of
values of electrochemical the epoxy
components coating
at different continued
soaking times. to decrease, while its
interfacial capacitance continued to increase. It is indicated that the liquor continuously
Time (d) 5 infiltrated into 10 the coating 15through the micropores20 25
caused 30
by the manufacturing process,
Rc (Ω·cm2) 5.773 × 104 5.568 × 104 5.103 × 104 2.217 × 104 - -
and the enlarged porosity gradually weakened the anti-corrosion effect. The decrease in
Qc (F·cm−2) 6.837 × 10−9 7.119 × 10−9 7.629 × 10−9 9.830 × 10−9 - -
charge transfer 8resistance and8 the rise in double-layer capacitance may be related to the
Rct (Ω·cm2) 8.906 × 10 8 4.940 × 10 2.143 × 10 3.793 × 107 6.554 × 106 6.450 × 103
regional exposure of the aluminum alloy substrate and the further corrosion of the oxide
Qdl (F·cm−2) 6.828 × 10−9 1.182 × 10−8 1.470 × 10−8 4.084 × 10−8 5.000 × 10−8 6.433 × 10−8
layer on the aluminum surface.

Figure8. 8.
Figure Changes
Changes in theinfitted
the parameters
fitted parameters withtime:
with soaking soaking time:resistance
(a) coating (a) coating resistance
Rc curve; (b) Rc curve;
(b) coating capacitance Qc curve; (c) charge transfer resistance Rct curve; (d) electric double-layer
coating capacitance Qc curve; (c) charge transfer resistance Rct curve; (d) electric double-layer capac-
itance Qdl curve.
capacitance Q curve.
dl

According to the EIS analysis and the characterization of the morphology, it is known
that the coating became damaged as a result of both the degradation of the coating itself
and the destruction of the corrosion products due to stress expansion under the coating.
Materials 2024, 17, 3884 9 of 10

Table 3. Parameter values of electrochemical components at different soaking times.

Time (d) 5 10 15 20 25 30
Rc(Ω·cm2 ) 5.773 × 104 5.568 × 104 5.103 × 104 2.217 × 104 - -
Qc (F·cm−2 ) 6.837 × 10−9 7.119 × 10−9 7.629 × 10−9 9.830 × 10−9 - -
Rct (Ω·cm2 ) 8.906 × 108 4.940 × 108 2.143 × 108 3.793 × 107 6.554 × 106 6.450 × 103
Qdl (F·cm−2 ) 6.828 × 10−9 1.182 × 10−8 1.470 × 10−8 4.084 × 10−8 5.000 × 10−8 6.433 × 10−8

According to the EIS analysis and the characterization of the morphology, it is known
that the coating became damaged as a result of both the degradation of the coating itself
and the destruction of the corrosion products due to stress expansion under the coating.

4. Conclusions
Here, the feasibility of packaging Chinese liquor with 3104 aluminum-coated cans
was studied. The results show that food simulants cannot completely replace liquor in a
packaging test. The level of alcohol alone is not an important factor in the packaging of
liquor with aluminum cans. The epoxy coating on the aluminum cans could not resist the
corrosion caused by the Chinese liquor. The degradation of the coating and the corrosion of
the aluminum, as determined using IR and XRD, and the surface and cross-sectional mor-
phologies showed damage to the coating caused by the liquor. Electrochemical impedance
spectroscopy can be used to analyze the corrosion of aluminum-coated cans. As for the
liquor, the degradation of the coating resulted in an excessive migration of Al into the
packaged liquor.

Author Contributions: Conceptualization, M.F. and J.C.; methodology, M.F.; validation, M.F., J.G.
and Z.W.; formal analysis, M.F.; investigation, M.F.; data curation, M.F.; writing—original draft
preparation, M.F.; writing—review and editing, J.C.; supervision, J.C.; funding acquisition, J.C., J.G.
and Z.W.; project administration, J.G. and Z.W. All authors have read and agreed to the published
version of the manuscript.
Funding: This research was supported by the Program for Innovative Research Team in University
(No. IRT13078).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data are unavailable due to privacy and ethical restrictions.
Conflicts of Interest: Author Jie Gu and Zheying Wu were employed by the company Shanghai Bao
Steel Packaging Co., Ltd. The remaining authors declare that the research was conducted in the absence
of any commercial or financial relationships that could be construed as a potential conflict of interest.

References
1. Kacmary, P.; Rosova, A.; Sofranko, M.; Bindzar, P.; Saderova, J.; Kovac, J. Creation of Annual Order Forecast for the Production of
Beverage Cans—The Case Study. Sustainability 2021, 13, 8524. [CrossRef]
2. Calabrese, L.; Bruzzaniti, P.; Proverbio, E. Pitting corrosion of aluminum alloys in anhydrous ethanol. Mater. Corros. 2018, 69,
1815–1826. [CrossRef]
3. Gazenbiller, E.; Arya, V.; Reitz, R.; Engler, T.; Oechsner, M.; Höche, D. Statistical analysis of AA-1050 localized corrosion in
anhydrous ethanol. Corros. Sci. 2021, 179, 109137. [CrossRef]
4. Brito-Franco, A.; Uruchurtu, J.; Rosales-Cadena, I.; Lopez-Sesenes, R.; Serna-Barquera, S.A.; Hernandez-Perez, J.A.; Rocabruno-
Valdes, C.; Gonzalez-Rodriguez, J.G. Corrosion Behavior of Al in Ethanol–Gasoline Blends. Energies 2020, 13, 5544. [CrossRef]
5. Park, I.J.; Yoo, Y.H.; Kim, J.G.; Kwak, D.H.; Ji, W.S. Corrosion characteristics of aluminum alloy in bio-ethanol blended gasoline
fuel: Part 2. The effects of dissolved oxygen in the fuel. Fuel 2011, 90, 633–639. [CrossRef]
6. Park, I.-J.; Nam, T.-H.; Kim, J.-H.; Kim, J.-G. Evaluation of corrosion characteristics of aluminum alloys in the bio-ethanol gasoline
blended fuel by 2-electrode electrochemical impedance spectroscopy. Fuel 2014, 126, 26–31. [CrossRef]
7. Song, G.-L.; Liu, M. Corrosion and electrochemical evaluation of an Al–Si–Cu aluminum alloy in ethanol solutions. Corros. Sci.
2013, 72, 73–81. [CrossRef]
Materials 2024, 17, 3884 10 of 10

8. Jellesen, M.S.; Rasmussen, A.A.; Hilbert, L.R. A review of metal release in the food industry. Mater. Corros. 2006, 57, 387–393.
[CrossRef]
9. Kato, L.S.; Conte-Junior, C.A. Safety of Plastic Food Packaging: The Challenges about Non-Intentionally Added Substances
(NIAS) Discovery, Identification and Risk Assessment. Polymers 2021, 13, 2077. [CrossRef]
10. Lestido-Cardama, A.; Vazquez Loureiro, P.; Sendon, R.; Paseiro Losada, P.; Rodriguez Bernaldo de Quiros, A. Application of
chromatographic analysis for detecting components from polymeric can coatings and further determination in beverage samples.
J. Chromatogr. A 2021, 1638, 461886. [CrossRef]
11. Blanco, I.; Cicala, G.; Costa, M.; Recca, A. Development of an epoxy system characterized by low water absorption and high
thermomechanical performances. J. Appl. Polym. Sci. 2006, 100, 4880–4887. [CrossRef]
12. Soares, D.S.; Bolgar, G.; Dantas, S.T.; Augusto, P.E.D.; Soares, B.M.C. Interaction between aluminium cans and beverages:
Influence of catalytic ions, alloy and coating in the corrosion process. Food Packag. Shelf Life 2019, 19, 56–65. [CrossRef]
13. Brylinski, L.; Kostelecka, K.; Wolinski, F.; Duda, P.; Gora, J.; Granat, M.; Flieger, J.; Teresinski, G.; Buszewicz, G.; Sitarz, R.; et al.
Aluminium in the Human Brain: Routes of Penetration, Toxicity, and Resulting Complications. Int. J. Mol. Sci. 2023, 24, 7228.
[CrossRef] [PubMed]
14. Wang, Y.-C.; Su, M.; Xia, D.-H.; Wu, Z.; Qin, Z.; Xu, L.; Fan, H.-Q.; Hu, W. Development of an electrochemical sensor and
measuring the shelf life of tinplate cans. Measurement 2019, 134, 500–508. [CrossRef]
15. Esteves, L.; Garcia, E.M.; Castro, M.d.M.R.; Lins, V.F.C. Electrochemical study of corrosion in aluminium cans in contact with soft
drinks. Corros. Eng. Sci. Technol. 2014, 49, 665–668. [CrossRef]
16. Almoiqli, M.; Alharbi, K.N.; Alnuwaiser, M.A.; Yajizi, G.; Alshoshan, S.; Baduways, W.; Albeladi, M.I.; Alsanea, R.S.; Aljohani,
T.A. Corrosion Behavior of Aluminium-Coated Cans. Materials 2023, 16, 1041. [CrossRef]
17. European Commission. Commission Regulation (EU) No 10/2011 of 14 January 2011 on Plastic Materials and Articles Intended
to Come into Contact with Food Text with EEA Relevance. 2011. pp. 1–89. Available online: https://eur-lex.europa.eu/legal-
content/EN/TXT/?uri=CELEX:32011R0010&qid=1722908422783 (accessed on 30 April 2024).
18. GB 31604.1-2015; National Standards for Food Safety General Rules for Migration Testing of Food Contact Materials and Products.
National Standards of China: Beijing, China, 2015; pp. 1–13.
19. Du, P.; Jiao, G.; Zhang, Z.; Wang, J.; Li, P.; Dong, J.; Wang, R. Relationship between Representative Trace Components and Health
Functions of Chinese Baijiu: A Review. Fermentation 2023, 9, 658. [CrossRef]
20. ATSDR. Pubilc Health Statement Aluminum CAS # 7429-90-5 Division of Toxicology and Environmental Medicine. 2008.
Available online: www.atsdr.cdc.gov (accessed on 4 April 2024).
21. Liu, X.; Shao, Y.; Zhang, Y.; Meng, G.; Zhang, T.; Wang, F. Using high-temperature mechanochemistry treatment to modify iron
oxide and improve the corrosion performance of epoxy coating—I. High-temperature ball milling treatment. Corros. Sci. 2015, 90,
451–462. [CrossRef]
22. Nnaji, N.; Nwaji, N.; Mack, J.; Nyokong, T. Corrosion Resistance of Aluminum against Acid Activation: Impact of Benzothiazole-
Substituted Gallium Phthalocyanine. Molecules 2019, 24, 207. [CrossRef] [PubMed]
23. López-Juárez, R.; Razo-Perez, N.; Pérez-Juache, T.; Hernandez-Cristobal, O.; Reyes-López, S.Y. Synthesis of α-Al2O3 from
aluminum cans by wet-chemical methods. Results Phys. 2018, 11, 1075–1079. [CrossRef]
24. Huang, X.; Li, N. Structural characterization and properties of lanthanum film as chromate replacement for tinplate. Appl. Surf.
Sci. 2007, 254, 1463–1470. [CrossRef]
25. Zhou, C.; Wang, J.; Song, S.; Xia, D.; Wang, K.; Shen, C.; Luo, B.; Shi, J. Degradation mechanism of lacquered tinplate in energy
drink by in-situ EIS and EN. J. Wuhan Univ. Technol.-Mater. Sci. Ed. 2013, 28, 367–372. [CrossRef]
26. Chu, Z.; Deng, W.; Zheng, X.; Zhou, Y.; Zhang, C.; Xu, J.; Gao, L. Corrosion Mechanism of Plasma-Sprayed Fe-Based Amorphous
Coatings with High Corrosion Resistance. J. Therm. Spray Technol. 2020, 29, 1111–1118. [CrossRef]
27. Macák, J.; Sajdl, P.; Kučera, P.; Novotný, R.; Vošta, J. In situ electrochemical impedance and noise measurements of corroding
stainless steel in high temperature water. Electrochim. Acta 2006, 51, 3566–3577. [CrossRef]
28. Bonora, P.L.; Deflorian, F.; Fedrizzi, L. Electrochemical impedance spectroscopy as a tool for investigating underpaint corrosion.
Electrochim. Acta 1996, 41, 1073–1082. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.