@. NOVAK, T. KOSEC: CORROSION AND PROTECTION OF NON-PATINATED, SULPHIDE- ...
697–704
CORROSION AND PROTECTION OF NON-PATINATED,
SULPHIDE- AND CHLORIDE-PATINATED BRONZE
KOROZIJA IN ZA[^ITA NEPATINIRANEGA, SULFIDNO- IN
KLORIDNO-PATINIRANEGA BRONA
@iva Novak, Tadeja Kosec
*
Slovenian National Building and Civil Engineering Institute, Dimi~eva ulica 12, 1000 Ljubljana
Prejem rokopisa – received: 2022-10-05; sprejem za objavo – accepted for publication: 2022-11-07
doi:10.17222/mit.2022.641
The surface of bronze undergoes changes when it is exposed to a polluted atmosphere, and bronze should therefore be protected
from this natural deterioration. The most common protective coating currently in use is Incralac, which includes toxic compo-
nents and is reported to dissolve a few months after application. This work therefore investigates a fluoropolymer-based coating
(FA-MS), and compares it to the protection offered by Incralac. Bronze samples (non-patinated, sulphide-patinated or chlo-
ride-patinated) were exposed to simulated urban rain for four months. The corrosion products formed were characterised using
SEM/EDS and Raman analyses. To study the protection efficiency of the newly developed fluoropolymer coating (FA-MS) and
Incralac protection, various electrochemical methods were used: measurements of open circuit potential linear polarisation and
potentiodynamic measurements. Findings show that the FA-MS coating provides a protection efficiency of 71 % for chlo-
ride-patinated bronze and 99.5 % for sulphide-patinated bronze. Contact angles of the FA-MS samples were higher than those of
the unprotected samples or the samples protected by Incralac, indicating better hydrophobic properties of the FA-MS coating.
Keywords: bronze, corrosion, SEM/EDS analyses, Raman spectroscopy, electrochemistry
Povr{ina brona se z izpostavljenostjo onesna`enemu okolju spreminja, zato mora biti pred naravnim propadanjem za{~itena.
Trenutno je najbolj pogosto uporabljen za{~itni premaz Incralac, ki vsebuje okolju {kodljive snovi, hkrati pa poro~ajo, da po
nanosu razpade `e po nekaj mesecih. Na podlagi tega je bil v tej {tudiji raziskan za{~itni premaz na osnovi fluoropolimera
(FA-MS), ter primerjava z Incralac premazom. Vzorci brona (ne-patiniran, rjavo-patiniran in kloridno-patiniran) so bili za {tiri
mesece izpostavljeni simulaciji de`evnice v urbanem okolju. Tvorjeni korozijski produkti so bili nato ozna~eni s pomo~jo
SEM/EDS in ramanske analize. Za {tudij u~inkovitosti za{~ite novo razvitega fluoropolimernega premaza (FA-MS) in Incralac
za{~ite smo uporabili razli~ne elektrokemijske metode: meritve potenciala odprtega kroga, linearno polarizacijo in
potenciodinamske meritve. Prav tako smo dolo~ili hidrofobnost z meritvami kontaktnih kotov in izmerili spremembe v barvi
pred in po nanosu za{~ite. Ugotovitve preiskav so, da FA-MS premaz nudi u~inkovitost za{~ite v vrednosti 71 % za kloridno
patiniran bron in 99,5 % za sulfidno patiniran bron. Izmerjene vrednosti kontaktnih kotov vzorcev, za{~itenih s FA-MS prevleko,
so bili vi{je kot pri neza{~itenih vzorcih in vzorcih, za{~itenih z Incralacom, kar ka`e na izbolj{ane hidrofobne lastnosti FA-MS
za{~ite.
Klju~ne besede: bron, korozija, SEM/EDS analiza, Raman spektroskopija, elektrokemija
1 INTRODUCTION
Patinas form on bronze surfaces due to exposure to
the atmosphere. They consist of different copper corro-
sion products in various colours, usually reddish, turning
to black, green and blue.
1
Some natural patinas are referred to as šnoble rust’,
due to their artistic look and stability, protecting the
bronze from further corrosion, while others are known as
šbronze disease’, causing a constant loss in the mate-
rial.
1,2
Cuprite (Cu
2
O), for example, successfully protects
bronze from further corrosion, while in a chloride-rich
atmosphere, cuprous chloride (CuCl) is formed, which
continuously releases chlorides that react with copper,
causing cyclic pitting corrosion and the formation of
atacamite.
1–3
Similarly, in the presence of SO
4
2–
,
brochantite usually forms, which, after longer periods of
exposure to acidic water, transforms into antlerite.
4
In
contrast to the formation of natural patinas, artificial
patination is an important final stage in the process of
producing bronze artwork.
5
It both represents the age of
an artefact and adds to its value
6
, so different methods to
create artificial patination have been established.
The stability of products formed on bronze can be de-
termined from the thermodynamic properties of each
mineral, more specifically through the measurement of
the Gibbs free energy.
7
Artificial and natural patinas are
composed of many different copper minerals; the stabil-
ity of a patina can be spectroscopically investigated after
long periods of immersion in a solution, while the stabil-
ity of the corrosion products formed on bronze can be
measured through various electrochemical techniques.
Corrosion resistance can therefore be determined with
linear polarisation, electrochemical impedance spectros-
copy, or potentiodynamic curves. Raman, FTIR,
SEM/EDS or XRD analysis can be used to identify the
minerals formed. The cross-sectional microstructure of a
sample can be observed using the FIB (focused ion
Materiali in tehnologije / Materials and technology 56 (2022) 6, 697–704 697
UDK 620.193.2 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 56(6)697(2022)
*Corresponding author's e-mail:
tadeja.kosec@zag.si (Tadeja Kosec)

beam)-SEM technology, or with an optical microscope
after polishing and etching, and the thickness of the pa-
tina layer or protective coating can be determined. The
composition and binding of the protective coating can
also be determined with XPS spectroscopy.
1
Since, in essence, patination is a form of corrosion,
many different coatings have been studied over the years,
with the desire to retain the colours of an artefact while
simultaneously protecting it.
8
The coating most com-
monly used within the field of conservation is Incralac,
which is composed of Paraloid B-44 with a BTA inhibi-
tor,
9,10
and has been reported to deteriorate over
time.
9,11,12
The use of different fluoropolymers and waxes
has also been explored,
9,13–15
and different inhibitors can
also be added.
15,16
It is, however, important to be aware
of a potential change in the colour following the applica-
tion of a coating. Potential protective coatings must be
suitably characterised and electrochemically tested. In
addition to electrochemical evaluation, the characteris-
tics investigated usually include changes in the contact
angle, colour changes, hardness, information regarding
the thickness of the coating, and the transmittance of UV
light.
In this paper, bare CuSnZnPb and bronze with two
different artificial patinas – brown sulphide and green
chloride – were exposed to a solution of artificial urban
rain for approximately four months in order to study the
development of corrosion over time. The corrosion prod-
ucts formed were identified with SEM/EDS and Raman
spectroscopy.
The protection efficiency of the FA-MS coating on
bare bronze, brown-patinated bronze and green-patinated
bronze was evaluated electrochemically. The hydro-
phobicity and differences in colour between pre- and
post-application were also studied. The results obtained
were compared to those of the commercial coating,
Incralac.
2 EXPERIMENTAL PART
2.1 Bronze
A leaded bronze plate representative of quaternary
bronze
17
(composition in w/%: 5.28 Sn, 3.84 Zn, and
2.71 Pb, with Cu to balance) was cut into discs with a
15-mm diameter. The composition of the quaternary
CuSnZnPb bronze studied was obtained with an optical
emission spectrometer (OES – Oxford Instruments,
2011).
2.2 Surface preparation – patination
The surface of the bronze was prepared for tests in
three different ways (bare bronze, sulphide-patinated
bronze and chloride-patinated bronze), as shown in Fig-
ure 1. All discs were first abraded with 1200 grit SiC pa-
per and ultrasonically cleaned for 3 min in ethanol.
3 w/% K
2
S solution was used for the brown sulphide
patination.
15
The bronze was first heated, then the solu-
tion was applied over the hot surface with a brush a total
of three times. The sulphide-patinated samples were then
washed with deionised water and aged at room tempera-
ture for 24 h. The green chloride patination was applied
over the sulphide-patinated samples at 45 °C. The solu-
tion used for the chloride patina was composed of 2 g
NH
4
Cl,2gNH
4
HCO
3
and 6 mL water. Immediately be-
fore the application, the solution was diluted with addi-
tional 50 % volume of deionised water. These samples
were then placed into a sealed plastic bag containing a
cotton pad soaked in water in order to create a very hu-
mid environment, and left for 24 h.
2.2 Composition and application of the protective coat-
ings
Two different coatings were applied over the bare and
patinated bronze samples. Specifically, Incralac, a com-
mercial protective coating commonly used in restoration
work, was diluted with n-butyl acetate and compared to
the newly developed fluoropolymer coating combined
with an adhesion promotor. 8.4 mg of solution was ap-
plied to each sample, then samples were dried at room
temperature for at least 1 h.
The fluoropolymer (FA-MS) protective coating con-
sisted of5%F Aand10%MSinanequal solvent mix-
ture of n-butyl acetate and diethyl succinate, as previ-
ously reported.
14
Approximately 2–3 mg of FA-MS was
added to the 15-mm-diameter bronze discs, adding ap-
proximately 1.1–1.70 g/m
2
of dry mass. The samples
were then dried at room temperature for at least 1 h fol-
lowed by further 24 h in a drying chamber at 40 °C.
2.3 SEM/EDS and Raman analyses
The bare, brown-patinated and green-patinated
bronze samples were then immersed in an artificial rain
solution composed of 68.6 mg/L SO
4
2–
, 28.7 mg/L Cl
–
and 94.3 mg/L NO
3
–
, simulating 100 times the concen-
tration of urban Ljubljana rain. After 128 d of immer-
sion, a SEM/EDS analysis was performed on the samples
using a Jeol JSM IT500 LV (Japan, 2019), observed at
different magnifications using a 20 keV excitation signal.
Raman spectroscopy was also conducted at this time
in order to study any corrosion products on the unpro-
tected samples of bronze, either non-patinated or covered
@. NOVAK, T. KOSEC: CORROSION AND PROTECTION OF NON-PATINATED, SULPHIDE- ...
698 Materiali in tehnologije / Materials and technology 56 (2022) 6, 697–704
Figure 1: Three different 15-mm-diameter samples of the bronze
studied: a) bare bronze, b) sulphide-patinated bronze, c) green chlo-
ride-patinated bronze

with a sulphide or chloride patina. A LabRAM Horiba
Yvon 800HR was used for measurements, using a laser
wave length of 532 nm at 10 % maximal power
(100 mW). Spectra were collected in a range of
50–4000 cm
–1
. The spectra presented are without back-
ground subtraction.
2.4 Electrochemical tests
An electrochemical investigation was performed us-
ing a Gamry 600+ potentiostat with a three-electrode
system cell. The electrochemical solution was artificial
urban rain with a composition of 685.7 mg/L SO
4
2–
,
287.0 mg/L Cl
–
and 943.2 mg/L NO
3
–
, from sodium
salts, simulating 1000 times the concentration of a
Ljubljana district.
18
The samples were placed in a teflon
holder with a surface area of 0.785 cm
2
exposed to the
solution. An Ag/AgCl reference electrode and a graphite
working electrode were used for the tests. The samples
coated with Incralac were measured at least twice, while
a minimum of three measurements were carried out for
the remaining samples.
Firstly, open circuit potential (OCP) was measured
for 3600 s, followed by linear polarisation resistance
measurements (LP) within ± 20 mV of the OCP, carried
out at a scan rate of 0.1 mV/s. Finally, potentiodynamic
measurements were carried out from –0.25 V below E
corr
to approximately 0.4 V above the reference electrode po-
tential, at a scan rate of 0.167 mV/s.
The corrosion resistance values, R
p
, obtained were
further used to calculate the protection efficiency (PE)
from Equation (1) where R
p
refers to the unprotected
sample and R'
p
to the sample with protection:
PE
R
R
/% % =−
′
⎛ ⎝ ⎜ ⎜ ⎞ ⎠ ⎟ ⎟ ⋅ 11 0
P
P
0 (1)
2.6 Contact-angle measurement
Contact angles were measured using the FTA 1000
DropShape Instrument B FrameSystem (First Ten Ang-
stroms, Newark, USA). At least three measurements
were made on each sample before and after the applica-
tion of the protective coatings, using a 2-μL droplet of
deionised water. Due to a high variance in the results, the
standard deviation of measurements was also provided.
2.7 Colour variations
Colour variations were specified with the CIEL*a*b*
system through the use of an i1 colourimeter (X-Rite,
USA) instrument prior to and following the application
of the protective coatings. Each sample was measured
three times, then the average and total colour differences,
 E*, were calculated using Equation (2):
ΔΔΔΔ ELab
a
****
() =++
222
(2)
When the value of E
ab
* is higher than three, the dif-
ference in colour can be distinguished with the naked
eye, which should be taken into consideration when pro-
tecting bronze statues.
3 RESULTS AND DISCUSSION
3.1 Characterisation of corrosion products
Following the 128-d exposure to simulated urban
rain, corrosion products were identified using SEM-EDS
and Raman spectroscopy, the results of which are shown
in Figures 2 and 3, respectively.
@. NOVAK, T. KOSEC: CORROSION AND PROTECTION OF NON-PATINATED, SULPHIDE- ...
Materiali in tehnologije / Materials and technology 56 (2022) 6, 697–704 699
Figure 2: SEM images and areas of the EDS analysis following 128 d
of immersion in the artificial urban rain: a) bare bronze, b) sul-
phide-patinated bronze, c) chloride-patinated bronze

Two typical morphologies were observed on the bare
bronze after 128 d of immersion in the artificial rain, as
shown in Figure 2a. The area marked as 1 represents a
part of the surface where no visible corrosion products
were formed, but visible grinding marks were present.
The EDS analysis (Table 1, area 1) confirmed the pres-
ence of cuprous oxide, with sulphur also detected. Area
2 is a typical conglomerate of fine needle-like crystals
with smaller and denser clusters. The EDS analysis (Ta-
ble 1, area 2) showed the presence of Cu, O and S.
In general, the flattest part of sulphide bronze con-
sists of Cu, Sn, C, O and S (Table 1, area 3), as can be
seen in Figure 2b. Isolated plate-like crystals formed on
the surface, which also contained Cu, O and S (Table 1,
area 4).
Three different morphologies of corrosion products
were detected on the chloride-patinated bronze following
128 d of exposure to urban rain, as shown in Figure 2c.
The flat, cracked area at point 5 represents the base layer,
which contains a higher amount of Cl (Table 1, area 5).
Islands of flat patches are distributed across the surface.
The EDS analysis confirmed the presence of Cu, O, S
and Cl in area 6 (Table 1, area 6). Tiny clusters can also
form on the surface, as visible at point 7. The elemental
composition of these clusters is similar to that for area 5.
Further characterisation was conducted with Raman
spectroscopy, the spectra of which are shown in Figure
3, with the peaks identified.
The Raman spectrum for area 1 of the bare bronze
(defined in Figure 2a) revealed the presence of cuprite,
Cu
2
O( Figure 3a). The Raman spectrum consisted of
narrow bands at 91 cm
–1
and 213 cm
–1
and broad bands
positioned at 517 cm
–1
and 626 cm
–1
, as reported previ-
ously.
20,21
The Raman spectrum for area 2 (as presented in Fig-
ure 2a) consisted of characteristic bands at 211 cm
–1
and
354 cm
–1
, a triplet at (454, 514 and 604) cm
–1
,av e r y
strong band at 963 cm
–1
, and weak bands at 1069 cm
–1
and 1185 cm
–1
. Broad bands were further observed at
3453 cm
–1
and 3536 cm
–1
, indicating the presence of
posnjakite, while the narrow band at 3572 cm
–1
shows
that brochantite was also present (see Table 2).
The Raman spectrum for area 3 (as presented in Fig-
ure 2b) consisted of narrow bands at (211, 434, 465, 521
and 610) cm
–1
, showing the presence of cuprite, in addi-
tion to the bands at 281 cm
–1
and 395 cm
–1
characteristic
of Cu
2
S, representing the brown patina resulting from the
artificial patination.
22
The Raman spectrum of the corrosion product in area
4 (identified in Figure 2b) consisted of the bands posi-
tioned at (441, 503 and 611) cm
–1
, a very strong band at
972 cm
–1
, weak bands at (1065, 1112 and 1147) cm
–1
,
and broad bands at (3266, 2423 and 3550) cm
–1
, indicat-
ing the presence of brochantite (Table 2).
@. NOVAK, T. KOSEC: CORROSION AND PROTECTION OF NON-PATINATED, SULPHIDE- ...
700 Materiali in tehnologije / Materials and technology 56 (2022) 6, 697–704
Table 1: EDS analysis of the areas following 128 d of immersion in the artificial urban rain
Sample Area Cu Sn Zn C O S Cl impurities
Bare bronze
1 63.9 7.18 2.45 11.2 8.3 6.5 — 0.44
2 48.0 — 0.82 8.02 32.8 10.3 — balance
Sulphide-patinated bronze
3 62.4 2.3 — 13.3 13.6 7.05 0.95 0.37
4 45.8 — 0.49 9.92 36.7 7.06 — balance
Chloride-patinated bronze
5 54.0 — 1.11 12.1 20.7 3.50 8.23 0.39
6 55.0 — — 8.90 30.9 4.65 0.52 0.01
7 55.9 — — 11.2 21.1 2.1 9.72 balance
Figure 3: Raman spectra after 128 d of immersion in simulated urban
rain: a) and b) bare bronze, c) and d) sulphide-patinated bronze, e) and
f) chloride-patinated bronze

Following 128 d of exposure, the Raman spectrum of
the chloride-patinated bronze showed the following char-
acteristic bands: (352, 444 and 502) cm
–1
, a triplet at
(816, 908 and 973) cm
–1
and two bands at 3346 cm
–1
and
3434 cm
–1
. This spectrum was identified as atacamite,
Cu
2
(OH)
3
Cl.
The spectrum obtained for the chloride patina after
128 d of exposure to acid rain (Figure 3f) was very simi-
lar to that for the brown patina, further showing that
brochantite, Cu
4
(SO
4
)(OH)
6
, was also formed (reference
in Table 2).
It was shown that the bronze and artificial patinas un-
derwent changes under stagnant conditions, i.e., a trans-
formation to sulphate minerals. Products such as
posnjakite or langite are predominantly precursors for
brochantite as the final product. This had previously
been reported and demonstrated in the literature.
23
Table 2: RRUFF references for the Raman peaks shown in Figure 3
Antlerite
(cm
–1
)
(R050212)
Brochantite
(cm
–1
)
(R060117)
Posnjakite
(cm
–1
)
(R110172)
Langite
(cm
–1
)
(R060090)
Atacamite
(cm
–1
)
(R050098)
260
343
422
480
607
866
989
1080
1139
1174
3490
3581
153
190
242
319
387
423
479
599
728
773
908
971
1097
1124
3258
3397
3573
183
206
247
335
441
499
607
976
1053
1112
1147
3274
3415
3548
238
434
484
608
764
970
1068
3394
3493
3570
141
354
444
508
585
814
905
970
3205
3343
3435
3.2 Electrochemical measurements
Electrochemical measurements of the open circuit
potential (OCP) of the bare, sulphide-patinated and chlo-
ride-patinated bronze are shown in Figure 4a. Similarly,
linear polarization (LP) measurements are shown in Fig-
ure 4b. Potentiodynamic (PD) curves are shown in Fig-
ure 5. The values of corrosion potential, E
corr
, obtained
from the OCP measurements, corrosion resistance, R
p
,
with the protection efficiency calculated from the LP
measurements, and corrosion current density, j
corr
,o b -
tained from the PD measurements, are shown in Table 3.
The E
corr
value of the bronze in the urban rain solution
was 21.5 mV. The E
corr
values for the protected bronze
samples were negative, being –107 mV and –14.6 mV
for Incralac and FA-MS, respectively. The measurements
(Figure 4b) show that the bronze protected with FA-MS
had a polarisation resistance, R
p
value of 571 k ·cm
2
,
which is about 100 times greater than that of the unpro-
tected sample (5.46 k ·cm
2
). The sample protected with
Incralac had the corrosion resistance 3 decades higher,
365 M ·cm
2
.
The corrosion potential of sulphide-patinated bronze
was more negative than the bare bronze and the R
p
value
(4.88 k ·cm
2
) was smaller (bare bronze: 5.46 k ·cm
2
).
The E
corr
values for sulphide-patinated bronze were simi-
lar, with the lowest value, –56.6 mV, observed for the
sample protected with Incralac. The corrosion resistance
of both coated samples was higher than that of the un-
protected one, about 200 times greater for FA-MS and
7500 times greater for Incralac. This resulted in high
protection efficiencies of 99.5 % and 100 % for FA-MS
and Incralac, respectively.
Compared to the sulphide patina and bare bronze, the
polarisation resistance of the green chloride patina was
lower, 3.36 k ·cm
2
,( Table 3), indicating very unstable
corrosion properties. Application of the FA-MS protec-
tive coating increased the E
corr
values, with the corrosion
resistance of the protected sample representing a protec-
tion efficiency of 71.0 %. Conversely, the E
corr
values de-
creased for the sample coated with Incralac, and the pro-
tection efficiency remained high, 99.9 %.
PD curves were also obtained, as shown in Figure 5.
It can be seen that the lowest corrosion current densities
were observed on the samples protected with Incralac.
On bare bronze, the FA-MS protection primarily af-
fected the cathodic process, while Incralac showed good
barrier properties, as can be observed from the cathodic
and anodic PD curves. Similarly, j
corr
was the lowest in
@. NOVAK, T. KOSEC: CORROSION AND PROTECTION OF NON-PATINATED, SULPHIDE- ...
Materiali in tehnologije / Materials and technology 56 (2022) 6, 697–704 701
Figure 4: Measurements for the samples with and without a protective
coating: a) OCP – bare bronze, b) LP – bare bronze, c) OCP – sul-
phide-patinated bronze, d) LP – sulphide-patinated bronze, e) OCP –
chloride-patinated bronze, f) LP – chloride-patinated bronze

the case of the sample protected with Incralac, with a
value of 0.158 nA/cm
2
.
Application of FA-MS and Incralac on the sulphide
bronze changed the cathodic and anodic behaviour, espe-
cially in the case of Incralac. The j
corr
values were low in
both cases, 39.0 nA/cm
2
and 1.02 nA/cm
2
for the FA-MS
and Incralac coatings, respectively.
The application of the FA-MS coating only slightly
affected the anodic behaviour of the chloride patina,
which was further confirmed by its poor protection effi-
ciency, as estimated from the R
p
values, and high corro-
sion current density, shown by its j
corr
value of
4.6 μA/cm
2
(Table 3).
Although the Incralac protection showed excellent ef-
ficiency and barrier properties, it is known from both the
literature
9
and practice
22
that it degrades very quickly
upon exposure to an aggressive environment. The newly
developed FA-MS coating provided a protection effi-
ciency of 99 % on bare bronze, 99.5 % on brown
patinated bronze and 71 % on chloride-patinated bronze.
This is a promising result for the protection of bare and
brown patinated bronze, while some future adjustments
should be made for the green patina. The lower effi-
ciency is most probably a result of the very thick and
inhomogeneous structure of the chloride green patina.
Such an effect was already observed in previous research
studies on various patinated surfaces.
15
3.3 Contact angle and colour variations
Contact angles of both the unprotected and protected
samples are shown in Table 4. The samples protected
with Incralac have contact angles of around 72°. In the
case of the sulphide- and chloride-patinated bronzes, this
value is lower than that of the unprotected sample. Of all
the samples evaluated, the highest contact angles were
observed on those protected with FA-MS, with values of
117°, 115° and 138° for the bare, sulphide-patinated and
chloride-patinated bronze, respectively.
The colour differences measured are also shown in
Table 4. On the non-patinated bronze, the changes in
colour were similar after the application of each type of
protection: both samples became yellowish, with high
 b* values of 6.70 and 8.57 for Incralac and FA-MS, re-
spectively. A greater change in colour was detected on
the sulphide-patinated bronze, with a E
ab
* value of 13.9
for the Incralac coating and a slightly lower value of 11.1
for the FA-MS coating. The changes were noticeable, as
the samples had visibly darkened, confirmed by large
@. NOVAK, T. KOSEC: CORROSION AND PROTECTION OF NON-PATINATED, SULPHIDE- ...
702 Materiali in tehnologije / Materials and technology 56 (2022) 6, 697–704
Table 3: Corrosion potential, E
corr
, corrosion resistance, R
p
, and corrosion current density, j
corr
, with the protection efficiency, PE, calculated for
each sample measured
Sample E
corr
/mV R
p /(k ·cm
2
) jcorr
/(nA/cm
2
) PE/%
Bronze 21.5 5.46 3 401 —
Bronze – FA-MS –14.6 571 59.5 99.0
Bronze – Incralac –107 365 000 0.158 100
Sulphide patina –30.4 4.88 5 440 —
Sulphide patina – FA-MS –34.9 942 39.0 99.5
Sulphide patina – Incralac –56.6 37 500 1.02 100
Chloride patina –0.32 3.36 6 520 —
Chloride patina – FA-MS 24.0 11.6 4 640 71.0
Chloride patina – Incralac –12.3 2 600 106 99.9
Figure 5: Potentiodynamic curves for the samples with and without
protection: a) bare bronze, b) sulphide-patinated bronze, c) chlo-
ride-patinated bronze

negative values of  L*. The smallest values were mea-
sured on the chloride-patinated bronze. The Incralac
coating on chloride bronze had a colour difference value
of 4.55, representing a visible change in colour, while
with the FA-MS coating it was not possible to perceive
the change in colour with the naked eye, since the value
of E
ab
* (1.10) was lower than three.
4 CONCLUSIONS
The following conclusions can be drawn from this
study:
1) Different copper sulphates formed on non-patinated
and patinated bronze after 128 d of immersion in
simulated urban Ljubljana rain. Posnjakite formed on
the non-patinated sample, while brochantite was
identified on the sulphide- and chloride-patinated
bronze samples. The chloride-patinated sample pri-
marily contained atacamite.
2) Large differences in colour, shown as the yellowing
and darkening of the samples, were present on both
the bare and sulphide-patinated bronzes regardless of
the type of protection, while the application of the
FA-MS coating on the chloride-patinated bronze did
not change the colour of the surface to the extent that
the change would be visible with the naked eye.
3) Contact angles of the samples protected with FA-MS
were high, which may have resulted in the protection
lasting longer.
4) The FA-MS coating exhibited a protection efficiency
of more than 99 % on both bare bronze and brown
sulphide-patinated bronze, but was less efficient on
the chloride-patinated bronze.
Acknowledgments
This work was financed by the SRA under Grant No.
J7 – 9404 and national research funding No. P2-0273.
The authors thank Nina Gartner for performing the
SEM/EDS analysis and Dr. Melita Tram{ek for enabling
the Raman analysis to be performed at the Jo`ef Stefan
Institute.
5 REFERENCES
1
J. M. Daehner, K. Lapatin, A. Spinelli, Artistry in Bronze: The
Greeks and Their Legacy, J. Paul Getty Museum and the Getty Con-
servation Institute, Los Angeles 2017
2
D. MacLeod, Bronze Disease: An Electrochemical Explanation,
ICCM Bulletin, 7 (1981) 1, 16–26, doi:10.1179/ICCM.1981.7.1.002
3
D. A. Scott, A Review of Copper Chlorides and Related Salts in
Bronze Corrosion and as Painting Pigments, Stud. in Conserv., 45
(2000) 1, 39–53, doi:10.1179/sic.2000.45.1.39
4
L. Robbiola, C. Fiaud, S. Pennec, New model of outdoor bronze cor-
rosion and its implications for conservation, ICOM-CC, 1993,
796–802
5
A. Casanova Municchia, F. Bellatreccia, G. D’Ercoli, S. Lo Mastro,
I. Reho, M. A. Ricci, A. Sodo, Characterisation of artificial patinas
on bronze sculptures of the Carlo Bilotti Museum (Rome), Appl.
Phys. A Mater. Sci. Process., 122 (2016), 1–8, doi:10.1007/S00339-
016-0551-4
6
V. Hayez, V. Costa, J. Guillaume, H. Terryn, A. Hubin, Micro Raman
spectroscopy used for the study of corrosion products on copper al-
loys: Study of the chemical composition of artificial patinas used for
restoration purposes, Analyst, 130 (2005), 550–556, doi:10.1039/
B419080G
7
A. H. Zittlau, Q. Shi, J. Boerio-Goates, B. F. Woodfield, J. Majzlan,
Thermodynamics of the basic copper sulfates antlerite, posnjakite,
and brochantite, Chemie der Erde - Geochemistry, 73 (2013), 39–50,
doi:10.1016/J.CHEMER.2012.12.002
8
P. Letardi, Testing new coatings for outdoor bronze monuments: A
methodological overview, Coatings, 11 (2021), 1–16, doi:10.3390/
COATINGS11020131
9
G. Bierwagen, T. J. Shedlosky, K. Stanek, Developing and testing a
new generation of protective coatings for outdoor bronze sculpture,
Prog. Org. Coatings, 48 (2003), 289–296, doi:10.1016/
J.PORGCOAT.2003.07.004
10
J. Wolfe, R. Grayburn, H. Khanjian, A. Heginbotham, A. Phenix, De-
constructing Incralac: A formulation study of acrylic coatings for the
protection of outdoor bronze sculpture, ICOM-CC 18th Triennial
Conference, Copenhagen 2017, 1–9
11
C. J. McNamara, M. Breuker, M. Helms, T. D. Perry, R. Mitchell,
Biodeterioration of Incralac used for the protection of bronze monu-
ments, J. Cult. Herit., 5 (2004), 361–364, doi:10.1016/J.CULHER.
2004.06.002
12
D. A. Scott, Copper and Bronze in Art: Corrosion, Colorants, Con-
servation, Getty Publications, LA, California 2002
13
M. Mihel~i~, M. Gaber{~ek, M. Salzano de Luna, M. Lavorgna, C.
Giuliani, G. Di Carlo, A. K. Surca, Effect of silsesquioxane addition
on the protective performance of fluoropolymer coatings for bronze
surfaces, Mater. Des., 178 (2019), 1–13, doi:10.1016/J.MATDES.
2019.107860
@. NOVAK, T. KOSEC: CORROSION AND PROTECTION OF NON-PATINATED, SULPHIDE- ...
Materiali in tehnologije / Materials and technology 56 (2022) 6, 697–704 703
Table 4: Contact angles of all the samples alongside standard deviations and colour differences, L*, a*, b* E
ab
*, between the unprotected
and protected samples
Sample
Contact angle (°) with
standard deviation
 L*  a*  b*  E
ab
*
Bronze 67 ± 1 ————
Bronze – Incralac 73 ± 0 2.27 2.63 6.70 7.55
Bronze – FA-MS 117 ± 1 0.60 2.67 8.57 8.99
Sulphide patina 95 ± 8 ————
Sulphide patina – Incralac 73 ± 0 −13.82 0.99 1.45 13.93
Sulphide patina – FA-MS 115 ± 1 −10.80 2.54 0.19 11.10
Chloride patina 98 ± 8 ————
Chloride patina – Incralac 71 ± 3 −4.45 0.89 0.24 4.55
Chloride patina – FA-MS 138 ± 6 −0.92 0.48 0.36 1.10

14
T. Kosec, L. [krlep, E. [vara Fabjan, A. Sever [kapin, G. Masi, E.
Bernardi, C. Chiavari, C. Josse, J. Esvan, L. Robbiola, Development
of multi-component fluoropolymer based coating on simulated out-
door patina on quaternary bronze, Prog. Org. Coatings, 131 (2019),
27–35, doi:10.1016/J.PORGCOAT.2019.01.040
15
T. Kosec, @. Novak, E. [. Fabjan, L. [krlep, A. Sever [kapin, P.
Ropret, Corrosion protection of brown and green patinated bronze,
Prog. Org. Coatings, 161 (2021), 1–9, doi:10.1016/J.PORGCOAT.
2021.106510
16
J. Wang, Y. Wu, S. Zhang, A new coating system modified with
nano-sized particles for archaeological bronze protection, Stud.
Conserv., 59 (2014), 268–275, doi:10.1179/2047058414Y.
0000000135
17
T. Kosec, @. Novak, E. [vara Fabjan, L. [krlep, M. Fin{gar, Ex-
ploring the protection mechanism of a combined fluoropolymer coat-
ing on sulphide patinated bronze, Prog. Org. Coatings, 172 (2022),
1–13, doi:10.1016/j.porgcoat.2022.107071
18
P. A. Schweitzer, Fundamentals of Metallic Corrosion: Atmospheric
and Media Corrosion of Metals, 2
nd
ed., CRC Press/Taylor & Francis
Group, Boca Raton 2007
19
M. Del Mar Pérez, A. Saleh, A. Yebra, R. Pulgar, Study of the varia-
tion between CIELAB  E* and CIEDE2000 color-differences of
resin composites, Dent. Mater. J., 26 (2007), 21–28, doi:10.4012/
DMJ.26.21
20
K. Maru{i}, H. Otma~i}-]urkovi}, [. Horvat-Kurbegovi}, H. Take-
nouti, E. Stupni{ek-Lisac, Comparative studies of chemical and elec-
trochemical preparation of artificial bronze patinas and their protec-
tion by corrosion inhibitor, Electrochim. Acta, 54 (2009),
7106–7113, doi:10.1016/J.ELECTACTA.2009.07.014
21
G. Niaura, Surface-enhanced Raman spectroscopic observation of
two kinds of adsorbed OH- ions at copper electrode, Electrochim.
Acta, 45 (2000), 3507–3519, doi:10.1016/S0013-4686(00)00434-5
22
P. Ropret, T. Kosec, Raman investigation of artificial patinas on re-
cent bronze – Part I: climatic chamber exposure, J. Raman
Spectrosc., 43 (2012), 1578–1586, doi:10.1002/JRS.4068
23
V. Hayez, J. Guillaume, A. Hubin, H. Terryn, Micro-Raman spec-
troscopy for the study of corrosion products on copper alloys: setting
up of a reference database and studying works of art, J. Raman
Spectrosc., 35 (2004), 732–738, doi:10.1002/JRS.1194
@. NOVAK, T. KOSEC: CORROSION AND PROTECTION OF NON-PATINATED, SULPHIDE- ...
704 Materiali in tehnologije / Materials and technology 56 (2022) 6, 697–704