Received: 2 September 2022 | Accepted: 6 October 2022

DOI: 10.1002/maco.202213533

ARTICLE

The effect of bird droppings on the corrosion of steel

and aluminum used in offshore applications

Fredrik Forr1 | Erlind Mysliu1 | Jan Halvor Nordlien1,2 | Andreas Erbe1

1
Department of Materials Science and
Engineering, NTNU, Norwegian Abstract
University of Science and Technology, Engineering materials in service are frequently exposed to bird droppings
Trondheim, Norway
(BDs). In this study, the interaction has been studied of BDs from chicken
2
Innovation & Technology ‐ Precision
Tubing, Hydro Extrusions, Håvik, with a carbon steel, AISI 316, 22% Cr duplex stainless steel (DSS), 25% Cr
Norway super duplex stainless steel (SDSS), carbon steel with a thermally sprayed
aluminum (TSA) coating, and aluminum alloy EN AW 6082. Samples with
Correspondence
Andreas Erbe, Department of Materials and without BDs are exposed to a cyclic salt spray test for up to 692 h
Science and Engineering, NTNU, (29 days). Electrochemical experiments in the presence of BDs were
Norwegian University of Science and
conducted in a syringe‐based droplet cell. Characterization of the corrosion
Technology, 7291 Trondheim, Norway.
Email: [email protected] products by infrared (IR) and Raman spectroscopy shows no build‐in of
organic components into the corrosion products in the samples exposed to
BDs. In general, BDs affect the anodic dissolution mechanism of metals and
increase the critical pitting potential. For carbon steel, a component of the
BDs acts as an anodic inhibitor, leading to passivation of the surface and
decreasing corrosion rates by 1–2 orders of magnitude. For 316, BDs increase
the critical pitting potential but may facilitate crevice‐like initiation of
localized corrosion. There are negligible effects of BDs on the corrosion of
SDSS, and no evidence either for adverse effects on DSS. TSA shows an
increased corrosion rate by half to one order of magnitude with BDs, but its
ability to galvanically protect a steel base material is not compromised.
AW 6082 shows decreased corrosion rate, no difference in mass loss, but
increased pitting in the presence of BDs. Overall, these model experiments
may only be seen as the start of a more systematic understanding of the
interaction of BDs with engineering materials.

KEYWORDS
aluminium, bird excrements, bird faeces, bird‐induced corrosion, electrochemical droplet
cell, metallic coatings, stainless steel, steel, vibrational spectroscopy, weight loss

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided
the original work is properly cited.
© 2022 The Authors. Materials and Corrosion published by Wiley‐VCH GmbH.

464 | www.matcorr.com Materials and Corrosion. 2023;74:464–481.

 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
FORR ET AL. | 465

1 | INTRODUCTION used for topside structures, piping, and vessels. Carbon

steel is also recommended to be coated to mitigate
Engineering materials in service are frequently exposed corrosion. For applications with higher demand for
to bird droppings (BDs). Studies exist, for example, of the corrosion resistance, duplex stainless steels are recom-
interaction of BDs with car paint,[1–3] copper and bronze mended. These applications could be piping and vessels
which are frequently encountered in cultural heritage with corrosive fluids. In addition, these may also be used
objects,[4–6] and highway structures consisting mainly of in pumps, and valve internals. Aluminium EN AW 6082
concrete.[7] The clear coat of car paint is degraded by the is recommended for “structural sections and compo-
BDs through enzymatic hydrolysis.[1–3] Copper and nents” in the NORSOK standard M‐121.[20] TSA is a
bronze suffered from significant chemical damage,[4,5] metallic coating that is often used on steel because of its
producing corrosion products not expected for atmo- excellent corrosion protection properties. In the event of
spheric corrosion.[4] Traces of deposits originating from a coating breakdown, exposing a small area of the base
birds were found in the patina on copper roofs.[8] No steel, the steel will be cathodically protected by the
effect on the degradation of bronze was found from aluminum.[21,22]
chloride‐containing uric acid in model experiments,[6] The purpose of this study was to investigate how BDs
whereas another study concluded there is damage of affect the corrosion behavior of materials as introduced
metals by uric acid.[9] Observation of increased rough- above. Corrosion has been assessed by salt‐spray testing
ness led to the conclusion that BDs degrade concrete and where samples were exposed with and without BDs.
stone.[7,10] Stones used, for example, in historic churches Following this test, the samples were visually examined,
are also broken down by the BDs.[11,12] Hence, BDs can and characterized by Fourier transform infrared (FTIR)
compromise the lifetime of several materials. and Raman spectroscopy to evaluate the presence of
The reported composition of BDs shows differences organic components on the metal surfaces. Mass loss
between different studies. Uric acid, proteins, ammonia, of the various samples was determined after removal
and nitrite have been shown to make up about 20% of the of corrosion products. Electrochemical experiments
seabird‐dropping composition.[13] BDs of urban pigeons were conducted in a syringe‐based electrochemical
proved to contain chemical elements not found in droplet cell containing BDs in direct contact with the
pigeons held at farms, without full quantitative analy- materials; polarisation resistance, and cathodic and
sis,[7] concluding that the diet plays a role in BD anodic polarisation curves were measured. The main
composition.[14,15] The number of soluble salts in BDs focus of this study is a comparison of the respective
from urban pigeons was found to be 4%, where the experiment in the presence and in the absence of BDs,
anions present are sulfate, chloride, phosphate, fluoride, and not a comparison of the different materials with each
bromide, and nitrate, ranging from highest concentration other. Likewise, a detailed interpretation of the corrosion
to lowest.[12] The term “guano” is used to describe dried mechanisms of the plain materials—as interesting as that
and solidified excrement, also in combination with may be—shall not be the focus of this study.
several species, for example, bird‐ and bat guano.[16]
Fresh bat excrement has a pH of 5.1–7.3.[17] However,
when the excrement dries and solidifies, as shown for bat 2 | MATERIALS AND METHODS
guano, the pH drops to 2.7–4.1.[17] How the effect is on
BDs is not known, but others have assumed similar The raw data used in this study are available at
behavior.[18] Bacterial respiration occurring in the BDs DataverseNO.[23]
can produce products that lower the pH of the droppings
additionally.[16]
A current trend in the offshore industries is the 2.1 | BDs
creation of unmanned or normally unmanned installa-
tions, which most likely introduces a greater presence of The BDs used were provided by chickens from a local
birds on the installations.[18] In this study, the interaction owner and were stored in an air‐tight container in a
of BDs with metallic structural materials important for fridge. Feathers and straws were removed with the best
offshore applications is investigated; the investigated effort before use. For both the salt spray test and the
materials are carbon steel, stainless steel, and aluminum, electrochemical experiments, the BDs were mixed with
along with a metallic coating on steel, and thermal water to create a slurry. The amount of water used to
sprayed aluminum (TSA). Based on NORSOK standard create the slurry was not constant as the consistency of
M‐001,[19] which describes the materials selection of the BDs varied. The pH of slurries of BDs in artificial
steels, carbon steel, and 316 stainless steel are typically seawater up to 60 g/L was between 8 and 8.25.
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
466 | FORR ET AL.

2.2 | Materials supported by a plastic piece. This gave the samples an

angle of 23∘ to the horizontal plane. Images of the samples
Commercial materials were used for this study, as before testing and at different stages of the test are shown
typically used in the shipyards building offshore struc- in Supporting Information: Section 1.
tures. Table 1 lists the standardized composition of the The testing was inspired by the ASTM G85‐19
different materials used in this study. The TSA sprayed standard,[25] which is a cyclic test made to simulate rain,
on carbon steel had a thickness of 200–300 with a 25 μm high humidity, and sunny weather. Cycle steps were
silicone resin sealer. The wire used for the TSA used in 45 min spraying, 2 h dry air purge, and 3:15 h soaking at
the experiments was of the grade EN AW 1050A (>99.5% high relative humidity, all at 49∘C.
Al). The composition of the coating is assumed to be A salt solution was made according to section 6.3 in the
close to that of the wire. None of the materials had a G85 standard (five parts NaCl dissolved in 95 parts
special pretreatment; in particular, AW 6082 was not drinking water). The initial pH of the solution was not
anodized, while it is frequently anodized in applications. adjusted to 2.8–3.0 as the standard suggests. The low pH
Samples were used as received, without polishing or would make the carbon steel corrode very fast, and
grinding to have a surface that corresponds to the surface therefore the initial pH was 6.5 for the salt solution. After
in‐service conditions.[24] 2 weeks of running (334 h), the most corrosion‐resistant
materials showed almost no corrosion. Thus, the solution
pH was adjusted to 3.0 by concentrated acetic acid, as
2.3 | Salt spray test suggested by the G85 standard. Acidification was done to
accelerate the corrosion process. The experiment proceeded
Three samples of each material, in total of 18 samples, under these conditions for an additional 358 h, resulting in
were prepared for a salt spray test. The samples were cut a total exposure time of 692 h for the materials that were
into 4 × 8cm2 rectangular pieces. These pieces were then left in the chamber for the whole experiment. Supporting
rinsed with distilled water, acetone, and ethanol. All Information: Section 1 contains an overview of the time
samples were then weighed individually before clear nail each sample was held in the salt spray chamber; the
polish was applied on the edges of the samples to avoid time was decided after a visual inspection of the samples at
edge effects (The backside of the sample was not the given opening times. After 692 h the experiment was
protected, which leads to a different situation than in an concluded and the remaining samples were removed from
application scenario). A BD slurry was spread on half the the chamber. When samples were removed from the
sample surface; this design was preferred over a design chamber, they were rinsed in water to remove salt deposits
where twice as many samples with the full area exposed to and left to dry before storing them in individual plastic
BDs are used because the latter does not include effects bags, following the ASTM G85‐19 standard.
from the practically important contact between a BD‐
containing side and a BD free side. Then the samples were
placed in an Ascott CC450IP cyclic chamber. Because of 2.4 | Raman and FTIR spectroscopy
the sample thickness, the standard sample holders could
not be utilized. To ensure that water would run off, and Both the areas exposed to BDs and areas without BDs
not accumulate on the surface, the samples were exposed longest in the salt spray chamber were subjected

TABLE 1 Composition of materials used here; except for thermally sprayed aluminum (TSA), these are standard values
Steel Cr Ni C Mn Si P S Mo Fe PREN
Carbon steel – – 0.4 – – – – – Balance –
AISI 316 16–18 10–14 0.08 2.0 1.0 0.045 0.030 2–3 Balance 25
22% Cr DSS 21–23 4.5–6.5 0.03 2.0 1.0 0.030 0.020 2.5–3.5 Balance 33
25% Cr SDSS 24–26 6–8 0.03 1.0 1.0 0.030 0.010 3–4 Balance 40
Aluminium Mn Fe Mg Si Cu Zn Ti Cr Al
EN AW 6082 0.40–1.00 0.5 0.60–1.20 0.70–1.30 0.10 0.20 0.10 0.025 Balance
EN AW 1050A (TSA) 0.05 0.40 0.05 0.25 0.05 0.07 0.05 – >99.5
Note: Values are given as wt% and single values are maximum values. Abbreviations: DSS, duplex stainless steel; PREN, pitting resistance equivalent number;
SDSS, super duplex stainless steel.
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
FORR ET AL. | 467

to post‐mortem FTIR and Raman spectroscopic analysis. 2.6 | Method of electrochemical

After removal from the salt spray chamber, samples were experiments
rinsed with water to remove soluble deposits and
subsequently dried. No other cleaning step was con- A syringe was modified to fit a reference electrode and a
ducted before measurements; in particular, the nail platinum wire counter electrode through the rubber gasket
polish covering the edges and the corrosion products of the syringe, based on a design published in the
were only removed after spectroscopic experiments. The literature,[27] which in turn is based on previous scanning
instruments used were Bruker Vertex 80v vacuum FTIR droplet cell designs.[28,29] The reference electrode was an
spectrometer and WITech Alpha 300R confocal Raman Ag/AgCl/KCl (sat.); all electrode potentials in this study are
microscope. reported with respect to this reference. The platinum wire
The Raman spectra were recorded using a 50 × was turned around the reference electrode. Approximately
magnification and a working distance of 9.1 mm. The 115 mm of platinum wire with a diameter of 0.5 mm was
laser (532 nm) was set to a power of (0.15 ± 0.05) mW. submerged in the electrolyte. With an average area of
The accumulation time was at least 60 s, with 6–10 0.2 mm2 for the working electrode (WE), this yields a CE to
accumulations. WE area ratio of 900. The shell of the syringe was then
In FTIR, samples were analyzed at grazing incidence. filled with a modified synthetic seawater electrolyte, based
The light‐exposed area was 1.78 cm2 . The spectral on ASTM standard D1141,[30] but without some of the
resolution was set to 4 cm−1, with an accumulation of minor components. The solution contained 24.53 g/L NaCl,
100 scans for each measurement. Absorbance A was 5.2 g/L MgCl 2, 4.09 g/L Na2SO4 , 1.16 g/L CaCl2, and 0.695
calculated in two different variants. First, a gold g/L KCl. All used salts had a purity of at least 98%. The pH
mirror was used as background to make corrosion was adjusted from 0.1 M NaOH to 8.9.
product peaks visible against a noncorroding reference; The upper part of the syringe, consisting of a gasket,
the absorbance against this background is calculated as reference electrode, and counter electrode, was then
A = −log10 (Isample ∕Igold ), where I is the intensity of light inserted into the shell to create a seal, as shown in
recorded in the presence of the species indicated in the Figure 1. While more sophisticated versions of movable,
subscript. Second, spectra in which the corrosion localized electrochemical cells exist that permit localized
products with BDs were used as samples against a measurements,[31–34] the nature of the BDs made such an
reference of the side without BDs are also displayed to easy‐to‐handle set‐up desirable. This setup allows the tip
easily visualize the differences between the two surfaces. of the syringe to be filled with BDs to perform
To that end, difference absorbance was calculated measurements on the metal when covered by BDs. The
as A = −log10 (IBD ∕Iplain ). BDs must have the right consistency which is adjusted
using distilled water; distilled water was used as there is

2.5 | Mass loss

(a) (b)
After the spectroscopic analysis, the samples were
weighed again to analyze the mass loss during
exposure in the salt‐spray chamber. For comparison
of the corrosion rates between the side with and
without BDs, the samples needed to be cut. They were
cut across the middle of the sample, creating samples
close to 4 × 4cm2, yielding an area exposed to BDs and
an area without. These pieces were then weighed
before and after the removal of the corrosion products.
The corrosion products were removed chemically
according to the ASTM G01 standard.[26] Supporting
Information: Section 5 lists the details. Using this
procedure, the backside of the samples where the front
side is covered with BDs also contributes to the mass F I G U R E 1 (a) Setup used in electrochemical experiments,
loss; it is difficult to fully quantitatively account for this including an Ag/AgCl/KCl (sat.) reference electrode, platinum wire
effect, as the front sides were exposed directly to the counter electrode, and electrolyte inside the syringe; (b) picture of
salt spray to a large extend by the design of the the syringe with bird droppings (BDs) inside the syringe tip [Color
chamber. figure can be viewed at wileyonlinelibrary.com]
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
468 | FORR ET AL.

a certain salt content in the BDs of ≈4 %. Too much water between experiments, for example, through heterogene-
will lead to a flowing mass that will run out of the ity of the materials surface.
syringe; too little water results in the droppings plugging
the syringe and contact loss during the experiment. The
consistency of the BDs was varying, resulting in different 3 | R ES U L T S
amounts of water in the mixture. Overall, a cement‐like
consistency should be pursued. The setup was always 3.1 | Salt spray test
installed inside a fume hood, regardless of whether the
experiment was run with or without BDs. No other Images of the samples in the salt spray chamber after 49,
preparation of the BDs was performed. 334, 592, and 692 h of exposure are shown in Supporting
The syringe setup was also tested with BDs deposited Information: Section 1. The first two samples, carbon
on top of the metal sample rather than inside the syringe. steel and 316 were removed after 49 h. At this stage, the
However, the BDs showed to act as a sponge in contact stainless steel samples showed only some small discolor-
with the electrolyte and absorb a large amount of water, ing on their surfaces, while the TSA and AW 6082 did not
resulting in an exposed area that was both inconsistent show any discoloring. The carbon steel had discoloring
and difficult to determine. There were also some on the whole surface visible in the picture.
challenges related to filling the tip of the syringe, caused After 334 h (14 days), some corrosion of the carbon
by the consistency of the droppings. steel was observed, indicated by the darker color around
the edges and corrosion product on the sample holder.
There was a slight increase in the discoloring for DSS
2.7 | Procedure of electrochemical sample 1. Salt deposits had accumulated around the
experiments edges of the TSA, along with some salt deposits on
the 316.
Metal samples were rinsed with ethanol, acetone, and Acidified solution led to high amounts of corrosion
distilled water before drying with pressurized air. The on the carbon steel, resulting in a short stint from 334 to
samples were not polished. The top of the syringe must 592 h. After 592 h (25 days), not much difference was
be properly sealed so there is no leakage from the observed in the condition for the carbon steel other than
syringe. For the measurement including BDs, the BDs the darker color on the sample holder. 316 showed some
would be added to the syringe at this point by using a more discoloring, both on the plain side and in the
spatula. To avoid evaporation of the electrolyte during bottom right corner. AW 6082 had become more matt on
experiments, the syringe tip needed to be as close to the the plain side, indicating some corrosion. The TSA had
metal surface as possible without the risk of touching. even more salt deposits on the surface.
The potentiostat used for the measurements was a After 692 h (29 days), TSA had more salt deposits on
Gamry 750 potentiostat. the surface, and AW 6082 showed some corrosion
Polarisation resistance, cathodic polarisation curves, product on the side without BDs. The DSS and SDSS,
and anodic polarisation curves were measured with this however, showed almost no signs of corrosion, only
setup. Before any of these measurements were done, the minor discoloring on the surface. 316 showed more or
open circuit potential (OCP) was measured for 15 min. less the same amount of discoloring as after 592 h.
The polarization resistance was measured by applying a Figure 2 shows pictures of the different samples after
potential of ±20 mV around the OCP with a rate of exposure; the BDs were on the right‐hand side. For
1.25 mV/s. The anodic polarisation curves were mea- carbon steel, corrosion craters were visible on the side of
sured by increasing the potential with a rate of 2.5 mV/s the sample without BDs. The side with BDs had a more
from OCP until the potential was in the range of homogeneous topography. There was also a color
1.0–1.3 V above the OCP, depending on the material to difference between the two sides. For 316, there was a
be measured. The cathodic curve was measured with the spotted pattern on the side with BDs. The left side
same parameters as the anodic curve but ended up at a showed a gradual increase in discoloring with increased
potential of 1.0 V below the OCP. The sequence of the exposure time. For DSS, the first and the last image
measurements could vary, but to ensure similar condi- showed more uniform corrosion on the side without BDs
tions, all measurements were made on an unused, clean, during exposure as well as a spotted pattern where BDs
spot of the sample. Because all measurements were were present. Figure 2h showed less discoloring than the
started at OCP, the terms OCP and corrosion potential other two samples and no spotted pattern for the right
Ecorr are synonymous in this study. Differences in OCP side. For SDSS samples, there was very little corrosion
between repeats should thus reflect the true variation visible for any exposure time. Most corrosion was visible
15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
| 469

(o)
(c)

(r)
(f)

(i)

(l)
(e)
(b)

(h)

(n)

(q)
(k)

(See caption on next page)

(m)
(a)

(g)
(d)

(p)
(j)

FIGURE 2
ET AL.
FORR
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
470 | FORR ET AL.

on the plain side after 692 h. TSA samples showed a electrochemical cell using dilute dispersions of BDs of
deposit around the edge of all samples. The first image up to 60 g/L in the same electrolyte. In these experi-
(334 h) showed more corrosion on the side with BDs, ments, the corrosion current density i corr increased up to
later there was corrosion on both sides. For AW 6082 concentrations of 45 g/L, however, dropped to the same
there was consistently at all times more pitting corrosion value as in the absence of BDs for 60 g/L.
on the side with BDs; the differences increased with Polarisation curves for 316 stainless steel are pre-
exposure time. sented in Figure 4. The critical pitting potential Epit
Sample masses before salt spray tests were converted increased in the presence of BDs. The curves with BDs
into area density to be able to compare the samples also showed little to no metastable pitting before Epit . The
before and after the salt spray test. The area density after polarization resistance was not that easy to comprehend,
removal of corrosion products is listed in Table 2. The as there were no clear trends in the curves, but already a
raw data used to calculate the area density, and the data significant variation in the individual materials.
before removal of corrosion products are shown in DSS proved to be a difficult material to work with, as
Supporting Information: Section 5. there were a lot of fluctuations in the OCP. Conse-
The area densities of the AW 6082 samples were quently, many attempts were needed to get reproducible
nearly equal. For carbon steel, the presence of BDs measurements. Figure 5 shows the polarisation curves
decreased the rate of mass loss. For the three stainless with and without BDs. The DSS showed less metastable
steels, there was no clear trend of an effect of the BDs, pitting when there were BDs present in the system. There
but overall the mass loss was rather low. For TSA, the was little difference in the OCP between the systems with
BDs increased slightly the rate of mass loss. and without BDs. The polarization resistance (Support-
After salt spray testing, the samples have been ing Information: Section 2) without BDs had bigger
characterized by FTIR spectroscopy (Supporting Infor- deviations than the measurements where BDs were
mation: Section 3) and by Raman spectroscopy (Support- present. Qualitatively, the system with BDs was more
ing Information: Section 4). While this characterization stable as the OCP and curves were easier to reproduce.
yielded relevant data for understanding the corrosion The observation of metastable pitting is also a possible
process of the materials themselves, the spectra did not explanation for the large fluctuations of the OCP.
reveal characteristic consistent differences between the Figure 6 shows the polarisation curves for SDSS; these
samples with and without BDs. Therefore, these spectra curves show little differences in the Ecorr range that was
are not discussed in detail here; a summary is given in measured. SDSS is the material that shows the least change
Supporting Information: Sections 3 and 4. in the presence of BDs. There were more differences
between the samples in the polarisation resistance measure-
ments, both in terms of OCP and noise in the measurements.
3.2 | Electrochemical experiments The polarization curves for TSA are presented in
Figure 7. There was an increase in the potential where
An overview of the parameters derived from polarisation the pitting initiates when BDs were present. The
curves is shown in Table 3. The data from polarisation polarization resistance for the TSA was lower with BDs
resistance measurements is shown in Supporting Infor- compared to the plain surface. There were some
mation: Section 2. variations in the measurements.
Polarisation curves for carbon steel are shown in AW 6082 also showed large variations of OCP,
Figure 3. In the presence of BDs, a passive region making it hard to find stable starting conditions. Stable
occurred in the anodic polarisation curves, with a polarisation curves obtained (Figure 8) showed a distinct
significant increase in current around 250 mV. This difference in Epit . With an increase of ≈500 mV, this is
increase can be attributed to pitting. There was also an the biggest difference seen in any of the materials studied
increase in polarisation resistance and Ecorr in the here. Because of the difficulties to find a stable OCP, the
presence of BDs. In a preliminary study to this one, measurements of the polarisation resistance proved to be
polarisation curves were measured in a classical hard in the absence of BDs, which is why only two curves

F I G U R E 2 Samples after different exposure times in salt spray test; sample with bird droppings (BDs) on the right in the respective panel.
Carbon steel after (a) 49 h, (b) 334 h, (c) 592 h; 316 after (d) 49 h, (e) 334 h, (f) 692 h; duplex stainless steel (DSS) after (g) 334 h, (h) 692 h, (i) 692 h
(repeat); super duplex stainless steel (SDSS) samples after (j) 334 h, (k) 592 h, (l) 692 h; thermally sprayed aluminum (TSA) samples after (m)
334 h, (n) 592 h, (o) 692 h; AW 6082 after (p) 334 h, (q) 692 h, (r) 692 h (repeat). [Color figure can be viewed at wileyonlinelibrary.com]
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
FORR ET AL. | 471

T A B L E 2 Area density of the samples ( ρA,0 , area density before were obtained (Supporting Information: Section 2). The
experiments; ρA,plain , area density after removal of corrosion polarization resistance was much higher when BDs were
products of the side without bird droppings [BDs]; ρA,BD area present compared to the plain surface. Qualitatively, the
density after removal of corrosion products for the side with BDs) OCP was more stable, for example, because of inhibition
Exp. of metastable pitting, when BDs were present which
Material time ρA,0 ρA,plain ρA,BD makes it easier to get reproducible measurements.
[h] [g/cm2 ] [g/cm2 ] [g/cm2 ] From the polarization curves presented above, Tafel
Carbon steel 49 4.47 4.35 4.44 extrapolation was utilized to find i corr and Ecorr . Obtained
334 4.42 4.16 4.20 parameters are summarised in Table 3. The polarisation
resistance of most of the materials were fluctuating, and
592 4.40 3.74 4.08
for some materials up to the order of several MΩcm2 .
316 49 4.47 4.47 4.45 Large polarisation resistance is expected for passive
334 4.48 4.44 4.46 materials. Also, the fact that there are several orders of
692 4.47 4.46 4.41 magnitude difference for some materials between differ-
ent measurements is not uncommon for passive materi-
Duplex stainless 334 4.63 4.59 4.59
als, leading to large statistical uncertainty. These results
steel (DSS)
are thus not discussed quantitatively.
692 4.67 4.65 4.65
692 4.61 4.60 4.57
Super duplex stainless 334 5.18 5.14 5.14 4 | DISCUSSION
steel (SDSS)
592 5.23 5.18 5.21
4.1 | General aspects
692 5.18 5.13 5.13
There is no evidence for the occurrence of new, different
Thermally sprayed 334 4.62 4.60 4.57 corrosion products under the influence of BDs, as
aluminum (TSA) evidenced by the absence of spectral features in the IR or
592 4.64 4.62 4.57 Raman spectra which occur only in the areas covered by
692 4.55 4.48 4.45
BDs. This absence of bird‐specific corrosion products
applies to all investigated materials. Consequently, no
AW 6082 334 0.40 0.39 0.39
organic substances are to a significant extent built into the
692 0.39 0.39 0.40 corrosion products that form. The absence of different
692 0.39 0.39 0.39 corrosion products does, however, not rule out adsorption
Note: Raw data and data before removal of corrosion products in Supporting
of compounds to the surface during and after corrosion
Information: Section 5. processes, and an effect on the kinetics in situ. The fact that

T A B L E 3 Different quantities obtained from electrochemical experiments (CS, carbon steel; CR, corrosion rate from i corr ; Ecorr ,
corrosion potential from polarisation curves; Epit , critical pitting potential [where clearly visible]; i corr , corrosion current density)
Without bird droppings (BDs) With BDs
i corr Ecorr CR E pit i corr Ecorr CR E pit
Material [mA/cm2 ] [mV] [mm/year] [mV] [mA/cm2 ] [mV] [mm/year] [mV]
CS (6.7 ± 0.5) ⋅ 10−1 −580 ± 28 7.82 – (2.2 ± 1.5) ⋅ 10−2 −233 ± 76 0.25 150 ± 50
316 (4.0 ± 0.5) ⋅ 10−2 −255 ± 10 0.47 −50 (7.3 ± 3.2) ⋅ 10−3 −165 ± 50 0.085 350
Duplex stainless steel (1.1 ± 0.4) ⋅ 10−1 −437 ± 81 1.28 – (7.3 ± 0.4) ⋅ 10−4 −185 ± 10 0.085 –
Super duplex stainless (3.7 ± 2.3) ⋅ 10−4 90 ± 42 0.0043 – (8.0 ± 2.8) ⋅ 10−4 −155 ± 36 0.0093 –
steel
Thermally sprayed (2.5 ± 2.1) ⋅ 10−6 −800 ± 283 0.00003 −200 (3.3 ± 0.4) ⋅ 10−5 −656 ± 50 0.00036 125 ± 125
aluminum
AW 6082 (2.0 ± 1.4) ⋅ 10−4 −663 ± 53 0.0022 −550 ± 50 (1.0 ± 0.1) ⋅ 10−4 −685 ± 10 0.0011 0
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
472 | FORR ET AL.

(a) F I G U R E 3 Polarisation curves for carbon

steel without (a) and with (b) bird droppings
[Color figure can be viewed at
wileyonlinelibrary.com]

(b)

some Raman spectra show differences, for example, in observed on the passive materials are indeed related to the
fluorescence background and in inorganic corrosion materials' properties, and not to problems with the cell.
products does not contradict this conclusion, because There is an enrichment of dissolution products in the
Raman spectra are measured highly localized, and the porous BDs, evidenced by the rather large differences in
corrosion product distribution is in any case heterogeneous. starting potential of the different polarisation curves with
Electrochemical measurements are not adversely BDs, which are not present without BDs. Since all
affected by the presence of the BDs on the surface. The measurements are done on a new spot on the surface, the
porous, poorly defined BDs are expected to add an surface should be in the same condition. Typical variation
additional resistance in the electrolyte, which (a) would between different regions of the surfaces should be
affect polarisation resistance measurements if it is too high, accounted for in the measurements without BDs, but these
and (ii) currents in general by an Ohmic potential drop. differences are much smaller. The only possibility for aging
The fact that for materials such as carbon steel with low of a component is thus the BDs in the syringes. Ecorr with
polarisation resistance the measurement of appropriate BDs will therefore to a larger degree vary with time than
curves is possible shows that the variation and differences potentials under free corrosion in free solution.
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
FORR ET AL. | 473

F I G U R E 4 Polarisation curves for 316 (a)

without (a) and with (b) bird droppings [Color
figure can be viewed at wileyonlinelibrary.com]

(b)

The electrochemical data has been interpreted based homogeneous topography, the appearance of a passive
on the premise that the BDs themselves do not contribute region in the anodic polarisation curves in the presence of
significantly via redox‐active compounds to the overall BDs, the increase of Ecorr accompanied by a decrease by
currents. The three orders of magnitude differences one order of magnitude in i corr in the presence of BDs, and
between corrosion currents detected on steel and TSA the increased polarisation resistance with BDs. Already
justify this assumption. In such slow polarisation curves, after 49 h salt spray test, there are differences in
one would also expect only limited release and diffusion appearance of the surfaces. After longer exposure times
rates of species to the surface. the plain side shows bigger craters and the BDs side
appears more homogeneous.
The protective effect of BDs becomes more pro-
4.2 | Carbon steel nounced with decreasing pH, shown by the ca. 50%
reduced mass loss in the period between 334 and 592 h of
The BDs slow down the corrosion of carbon steel by exposure when the solution was acidified. At the same
adding an additional barrier and by passivating the time, the mass loss per hour for plain carbon steel is two
surface, in line with decreased mass loss, more orders of magnitude higher from 334 to 592 h than from
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
474 | FORR ET AL.

(a) F I G U R E 5 Polarisation curves for duplex

stainless steel without (a) and with (b) bird
droppings. Line in (a) highlights the open circuit
potential for experiments 1 and 2 [Color figure
can be viewed at wileyonlinelibrary.com]

(b)

0 to 334 h. With BDs, however, the mass loss per hour is Raman for what causes the passivation. Possible reasons
in the same order of magnitude throughout the whole are (i) formation of a dense barrier oxide layer on the
experiment. Thus, lower pH accelerated the corrosion on surface, for example, from an oxidizing compound, or (ii)
the plain side of the sample, but not on the side with adsorption, for example, on monolayer or few‐layer level,
BDs. The BDs' pH when dry is believed to be somewhat of a component from the BD. In the case of adsorption,
similar to the solution at the end of the experiment.[17] the surface structure is likely significantly different after
The BDs positive effects are therefore more beneficial removal from the solution, which requires thus sophisti-
than the detrimental effects of the low pH and anions in cated experiments to prove.[35–37] There is no hint here
the dry BDs. that an increase in pH would have happened, which
The anodic polarization curves show a passive region could also explain the formation of a passive oxide. Such
in the presence of BDs which is absent in the plain an increase is also not likely based on the literature of
material. At higher potentials, the polarisation curves BDs.[17] Inhibition effects can also be caused by a rather
strongly resemble the curve of a system with pitting. intricate influence on the defect structure of the forming
There is no direct evidence, for example, from IR or oxide.[38]
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
FORR ET AL. | 475

F I G U R E 6 Polarisation curves for super (a)

duplex stainless steel without (a) and with (b)
bird droppings [Color figure can be viewed at
wileyonlinelibrary.com]

(b)

The cathodic processes are not affected by the BDs, as There were, however, no signs of pitting on the surface of
evidenced by the good agreement of the cathodic the samples exposed to the salt‐spray chamber. In
polarisation curves in the presence and absence of BDs. applications, the galvanic coupling may shift the poten-
Therefore, the decreased corrosion rate must be caused tial of a steel component to above Epit , which must be
by an effect on the anodic processes, which implies that a avoided.
component from the BDs acts like an anodic corrosion
inhibitor. This interpretation is in line with the decrease
in i corr above a certain BD concentration in dispersion as 4.3 | 316
observed in preliminary experiments, as these often
accelerate corrosion in low concentrations before they Epit of 316 is increased by nearly 500 mV in the presence
passivate materials.[39] of BDs. The 316 also shows less metastable pitting in the
The passivation of the material in the presence of BDs passive area in the presence of BDs. Up to Epit , there is no
could lead to localized corrosion, such as pitting. Indeed, difference in the dissolution kinetics between the system
an Epit was observed in the anodic polarisation curves. in the presence and absence of BDs. Thus, the BDs make
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
476 | FORR ET AL.

(a) F I G U R E 7 Polarisation curves for

thermally sprayed aluminum without (a) and
with (b) bird droppings [Color figure can be
viewed at wileyonlinelibrary.com]

(b)

the 316 less prone to pitting initiation. A possible exposed in the salt spray tests over a much longer time
explanation of increased pitting resistance could be the and also during wet‐dry cycles in the cyclic test compared
decreased amount of anions near the surface due to poor to electrochemical experiments, we reason that the
ion transport through the BDs, or pH buffering effects apparent contradiction between these two observations
near the surface. must be related to the barrier properties of the BDs and
The BDs do, however, initiate localized rust forma- the consequent possibility for differential aeration in the
tion, possibly by crevice‐like differential aeration effects. inhomogeneously covered surface. This phenomenon
Visual inspection of the samples exposed for 334 and requires a more detailed investigation.
692 h shows a spotted rust pattern on the surface with There is no difference in uniform corrosion kinetics
BDs, which is absent in the absence of BDs. As pit between the systems with and without BDs, even though
initiation is retarded by BDs (c.f. preceding paragraph), this mode of corrosion is less relevant for this kind of
other, long‐term effects must lead to this spotted, material. Corrosion current densities agree well between
localized rust pattern observed. As the samples are the system with and without BDs, and the mass losses are
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
FORR ET AL. | 477

F I G U R E 8 Polarisation curves for AW 6082 (a)

without (a) and with (b) bird droppings [Color
figure can be viewed at wileyonlinelibrary.com]

(b)

relatively small in both cases. Before acidifying the test exposure than for one sample after 692 h, large differ-
solution, the mass loss in the presence of BDs was ences in appearance between two samples exposed for
slightly lower, and after acidifying slightly higher, than in 692 h), (ii) large variations in Ecorr over several 100 s of
their absence. mV, (iii) variations in the shapes of the anodic
polarisation curves, and (iv) a higher uniform i corr than
316. The differences are not supposed to be caused by
4.4 | DSS the presence of both the austenite and ferrite phases. The
size of these phases is usually in the μm region, while the
The stock from which the samples investigated here were opening of the syringe is in the mm region. The area that
taken may have suffered from a too heterogeneous is exposed to the electrolyte is therefore orders of
surface quality, which manifests itself (i) in the large magnitude larger than the austenitic and ferritic phases
differences between two salt spray samples investigated in the steel. Hence, the difference must be caused by
for the same time and a rather nonintuitive trend with differences in the chemical composition or micro-
time without BDs (larger mass loss after 334 h of structure of the steel across the samples. Alternatively,
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
478 | FORR ET AL.

a surface impurity that could not be removed by the systems, SDSS also shows the least effect of BDs in the
comparably mild cleaning procedure may have been shape of the polarisation curves. While the mass loss was
present. The peculiar features in the different repeats of ca. 50% lower with BDs than without, i corr was on the
the polarisation curves in the absence of BDs are not same order of magnitude but approximately twice as high
discussed here for this reason. with BDs; both changes, however, on an overall very low
Similar crevice type of effects as discussed for 316 corrosion rate. Despite the ≈250 mV lower Ecorr in the
may also be observed here, with a rust spot appearing on presence of BDs, SDSS is not significantly affected by
one of the samples exposed for 692 h. However, here the BDs.
effect is only observed for one sample and may thus be BDs do not affect the mechanism of metal dissolution
less common than for 316. or passivation in SDSS to a significant extent. The anodic
BDs do affect the electrochemical properties of DSS. branches with and without BDs have a similar shape.
The polarization curves with BDs display a different With BDs, there is a passive region when the potential is
behavior than without. With BDs, the steel no longer kept < ≈100 mV. At higher potential, slowly propagating
shows a textbook‐like passive domain in the polarisation pitting in the ferrite may occur up to 750 mV; the final
curve, with potential independent current. There is a increase above 1000 mV may be caused by pitting in both
kink in the anodic branches of the polarisation curves the austenite and ferrite phases,[42] or by the onset of
around 300 mV which could indicate enhanced dissolu- oxygen evolution. Oxygen evolution is thermodynamically
tion. This feature could originate from pitting initiated at expected at so high potentials.
an austenite/ferrite grain boundary which propagates As for DSS, the BDs decrease the cathodic activity for
slowly into the ferrite phase, as observed previously oxygen reduction, as discussed in Section 4.4 most likely
despite the fact that the ferrite phase is the phase richest because of a decrease in the available amount of cathodic
in corrosion‐resisting elements Cr and Mo.[40] sites which are active for oxygen reduction.
The presence of the BDs deactivates cathodic sites on
the surface of DSS. There are clear differences in the
cathodic polarisation curves comparing the system with 4.6 | Thermally sprayed aluminum
and without BDs; the plateau typically attributed to the
diffusion‐limited oxygen reduction is almost one order of The presence of BDs increases the self‐corrosion rate of
magnitude lower with BDs than without. This difference TSA, evidenced by 2–3 times increased mass loss in the
could be related to differences in the spontaneous early phase of salt spray exposure, much stronger
convection patterns in the system,[41] however, as the prominence of corrosion products on the surface, and a
current here is in between those observed for carbon steel one order of magnitude higher i corr in the electrochemi-
and 316, it is more reasonable to attribute these cal experiments. The increase could be related to the
differences to a different amount of cathodically active ability of components in the BDs to complex Al3+,
sites. increasing oxide solubility. Remarkably, the differences
There is no evidence of a strong negative effect of the between salt spray samples were negligible in the second
BDs in this system, however, for DSS this is not so easy to part of the exposure where the sample was acidic. The
determine conclusively from the results obtained here. low pH in the second half of the experiments accelerated
Again the rate of uniform corrosion, given by i corr , the corrosion of the plain TSA, without an additional
decreases significantly (by two orders of magnitude) in effect from BDs. The higher mass loss may thus be
the presence of BDs, though part of this decrease is explained by a decrease in pH during the drying cycles in
because of the unreasonably high i corr in the absence of the salt spray chamber; the low pH at high humidity may
BDs. Pit initiation may, however, be affected by the BDs not represent a realistic use scenario, though. However,
adversely. in the electrochemical experiments, the pH decrease is
not expected to be so high. Therefore, a pH effect alone is
not the only aspect. Likewise, both localized (judged by
4.5 | SDSS appearance) and uniform (judged by mass loss and i corr )
corrosion rates increase. In the areas where BDs are
BDs have no significant effect on the corrosion resistance present, these may act like a further organic coating,
of SDSS. In the salt spray tests after 692 h, there is only which is not desirable for TSA.[43]
very limited corrosion on both sides. Unlike 316 and one The BDs do not compromise the ability of TSA to
of the DSS samples, the SDSS does not show any spotted galvanically protect underlying steel, because its Ecorr is
pattern on the side with BDs for any samples, that is, no unchanged by the presence of BDs and because of the
evidence for localized rust formation. For all investigated large i corr . For the samples exposed to the salt spray test
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
FORR ET AL. | 479

for the longest time, corrosion down to the substrate has The BDs also decrease the cathodic activity of
been observed in some areas, with relatively little cathodic inclusions, as shown by the lower cathodic
difference between the sample with and without BDs. currents. As opposed to the stainless steels above, the
The BDs also change the anodic reaction mechanism cathodic activity both of the diffusion‐limited oxygen
for TSA, as evidenced by the differences in the anodic reduction and the hydrogen evolution may be decreased,
polarisation curves; whereas the current‐potential rela- though a full assignment of the different features in the
tion in the absence of BDs is nearly potential indepen- curves is challenging.
dent in the passive region, there is a current increase in
the presence of BDs. This behavior implies that the BDs
compromise passivity in this system. On the other hand, 5 | CONCLUSIONS
Epit increases when the BDs are present by 250‐500 mV.
Pitting itself is most likely not a big issue in a metallic The corrosion behavior of different materials used in
coating that is supposed to protect the base material offshore applications has been investigated in the
galvanically. presence of BDs by salt spray tests and electrochemical
experiments in a syringe‐based droplet cell. In general,
BDs affect the anodic dissolution mechanism, and
4.7 | AW 6082 decrease the rate of dissolution, except for the stainless
steels they promote the passivity of materials, and where
There is an increased susceptibility to pitting corrosion applicable also increase Epit . At the same time, under
when exposed to acidic solution, even though the uniform acidic test conditions, BDs promote pitting initiation. At
corrosion rate in neutral or slightly alkaline solution least for DSS, AW 6082, and possibly SDSS, they also
decreases in the presence of BDs. Pits are clearly visible in reduce cathodic activity, possibly by blocking active sites
the two samples with BDs exposed in the salt spray for cathodic processes. Specifically, for the different
chamber for 692 h, however, only to a very limited extent materials, the following has been found.
in the sample exposed for 334 h; as the test solution was
acidified after 334 h, this acidification in combination with • Carbon steel benefits from the presence of BDs with a
the BDs is likely the reason for the observed pitting. Some lower corrosion rate, denser corrosion products, and
pitting is also observed for the side without BDs. It must passivation.
be noted that acidification in the accelerated tests may • 316 shows an increase in Epit but at the same time a
often be far from actual application conditions, as also larger amount of rust spots in the salt spray tests in the
noted for intergranual corrosion testing.[44] At the same presence of BDs.
time, i corr decreases in the presence of BDs, and the mass • DSS showed conflicting results between different
loss is not detectable even in the samples exposed for surface areas, but no evidence was found for a
692 h. The latter finding is clear evidence for the absence significant effect of BDs.
of enhanced uniform corrosion in the presence of BDs. • SDSS shows a behavior that is nearly unaffected by
The BDs introduce an environment that may be harmful the BDs.
to non‐anodized AW 6082. • TSA suffers from localized corrosion and an increased
The presence of BDs increases the passive region in self‐corrosion rate under the BDs. While TSA is the
the slightly alkaline chloride solution, shown by the material that is affected most by BDs, its function as a
polarisation curves. BDs thus increase Epit by almost sacrificial coating is still maintained.
500 mV; in general, in chloride solution, many aluminum • AW 6082 suffers from localized corrosion underneath
alloys show Epit ≈ Ecorr .[45] Thus, the trend that the BDs the BDs in acidic tests but also shows increased Epit in
stabilize the oxide at least in the short term is not only near‐neutral solution.
true for steel, but also for aluminum. This phenomenon
may therefore either relate to a specific chemical Overall, in practice, a lot will depend on details that
interaction that can take place in several metals and could not be addressed in a study with a limited scope
metal oxides or is a pH buffering effect from a such as this. For instance, the diet of birds affects the
component in the BDs. Note, however, that the passive composition of the BDs, and thus could have an effect on
current density is increased by the BDs by half an order the corrosion rates and mechanisms. It remains an open
of magnitude. This increase may indicate an interaction question how far the experiments with chicken drop-
of a component in the BDs with the oxide, for pings conducted here can be transferred to other birds,
example, with point defects, similar to what has been and many mechanistic questions remain open for several
shown for zinc.[38] of the investigated materials.
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
480 | FORR ET AL.

AUTHOR CONTRIBUTIONS [7] C. P. Huang, G. Lavenburg, Impacts of Bird Droppings and

Fredrik Forr was involved in the conceptualization, Deicing Salts on Highway Structures: Monitoring, Diagnosis,
conducted all the investigations, and wrote the original Prevention, Tech. Rep. Delaware Center for Transportation,
University of Delaware. https://cpb-us-w2.wpmucdn.com/
draft. Erlind Mysliu developed the measurement meth-
sites.udel.edu/dist/1/1139/files/2013/10/Rpt-221-Bird-
odology used for the electrochemical experiments and
Dropping-Impacts-Huang-DCTR422185-xk4gdv.pdf
was involved in writing‐review and editing. Jan Halvor [8] A. Stoch, J. Stoch, J. Gurbiel, M. Cichocińska,
Nordlien was involved in supervising, analysis, and M. Mikołajczyk, M. Timler, J. Mol. Struct. 2001, 596, 201.
writing‐review and editing. Andreas Erbe was involved [9] A. Vasiliu, D. Buruiana, Int. J. Conserv. Sci. 2010, 1, 83.
in conceptualization, analysis, project management, [10] B. Cámara, M. Á. De Buergo, R. Fort, V. Souza‐Egipsy,
supervision, and writing‐review and editing. S. Perez‐Ortega, A. De Los Ríos, J. Wierzchos, C. Ascaso,
Period. Mineral. 2015, 84, 539.
[11] P. Tiano, Biodegradation of Cultural Heritage: Decay Mecha-
ACKNOWLEDGMENTS
nisms and Control Methods, Tech. Rep. CNR ‐ Centro di
We thank Ørjan Ellingsen and Vegard Schøyen, Shell
studio sulle "Cause Deperimento e Metodi Conservazione
Norway, for bringing this topic to our attention and for Opere d'Arte”. https://www.researchgate.net/publication/
providing sample material, Kurt Ekrem and Darko 240635027_Biodegradation_of_Cultural_Heritage_Decay_
Petkovic, Randaberg Coating AS, for providing a TSA‐ Mechanisms_and_Control_Methods
coated sample, and Espen Strandheim, Aker Solutions [12] M. Gömez‐Heras, D. Benavente, M. Á. De Buergo, R. Fort,
ASA, for helpful discussions. Håvard Wilson, MainTech Eur. J. Mineral. 2004, 16, 505.
AS, is acknowledged for leading the industrial side of this [13] P. Szpak, J.‐F. Millaire, C. D. White, F. J. Longstaffe,
J. Archaeol. Sci. 2012, 39, 3721.
project. For technical support, we thank Andrey Kosins-
[14] S. De La Peña‐Lastra, Sci. Total Environ. 2021, 754, 142148.
kiy, Anita Storsve, and Marthe Folstad, for helpful [15] D. H. R. Spennemann, M. J. Watson, APT Bull. J. Preserv.
discussions Roald Lilletvedt and Iman Taji. Erlind Mysliu Technol. 2018, 49, 19.
acknowledges a PhD scholarship in the frame of the [16] R. Shahack‐Gross, F. Berna, P. Karkanas, S. Weiner,
project “Coated Recycled Aluminium ‐ developing J. Archaeol. Sci. 2004, 31, 1259.
surfaces for well‐adhering and corrosion resistant coating [17] C. M. Wurster, N. Munksgaard, C. Zwart, M. Bird,
systems” supported by the Research Council of Norway Biogeochemistry 2015, 124, 163.
[18] S. Christensen‐Dalsgaard, N. Dehnhard, B. Moe,
(No. 309875), Hydro and Speira.
G. H. R. Systad, A. Follestad. Unmanned Installations and
Birds. A Desktop Study on How to Minimize Area of Conflict,
CONFLI CT OF I NTER EST Nina Report. Norwegian Institute for Nature Research.
The authors declare no conflict of interest. http://hdl.handle.net/11250/2630793
[19] NORSOK Standard M‐001. Material Selection, Rev. 3, Tech.
DATA A VAILABILITY S TATEMENT Rep. Norwegian Technology Centre. https://www.standard.
The data that support the findings of this study are no/pagefiles/1176/m-001.pdf.
openly available in NTNU Open Research Data (https:// [20] NORSOK Standard M‐121. Aluminium Structural Material,
Tech. Rep. Norwegian Technology Centre.
dataverse.no/dataverse/ntnu) at https://doi.org/10.
[21] C. Holager, R. Johnsen, K. Nisancioglu, NACE Int. Corros.
18710/VSZVHL.
Conf. Ser. 2015, 5570.
[22] H. Wilson, R. Johnsen, C. T. Rodriguez, S. M. Hesjevik,
ORCID Mater. Corros. 2019, 70, 293.
Andreas Erbe http://orcid.org/0000-0003-0225-2204 [23] F. Oftedal Forr, E. Mysliu, J. H. Nordlien, A. Erbe,
Replication Data For: "The Effect of Bird Droppings on the
REFERENCES Corrosion of Steels and Aluminium for Offshore Applications”,
[1] H. Yari, M. Mohseni, B. Ramezanzadeh, J. Coat. Technol. Res. NTNU Open Research Data, 10.18710/VSZVHL 2022.
2011, 8, 83. [24] A. J. Sendriks, Corrosion of Stainless Steels, 2nd ed., John
Wiley & Sons, New York 1996.
[2] H. Yari, M. Mohseni, B. Ramezanzadeh, N. Naderi, Prog. Org.
[25] ASTM G85‐19. Standard Practice for Modified Salt Spray
Coat. 2009, 66, 281.
(Fog)testing, Tech. Rep. ASTM International, West Consho-
[3] B. Ramezanzadeh, M. Mohseni, H. Yari, S. Sabbaghian, Prog.
hocken, 2019.
Org. Coat. 2009, 66, 149.
[26] ASTM G1‐03(2017)e1. Standard Practice for Preparing,
[4] E. Bernardi, D. J. Bowden, P. Brimblecombe, H. Kenneally,
Cleaning, and Evaluating Corrosion Test Specimens, Tech.
L. Morselli, Sci. Total Environ. 2009, 407, 2383.
Rep. ASTM International, West Conshohocken, 2017.
[5] K. Balogh, Z. Slívz, K. Kreislová, e‐Monumentica 2017,
[27] A. M. Panindre, K. H. Chang, T. Weirich, G. S. Frankel,
2017, 5.
Corrosion 2018, 74, 847.
[6] M. A. Ginez, A. Espinoza‐Vázquez, F. J. Rodríguez‐Gómez, [28] A. W. Hassel, M. M. Lohrengel, Electrochim. Acta 1997, 42,
ECS Trans. 2018, 84, 125. 3327.
 15214176, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/maco.202213533, Wiley Online Library on [19/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
FORR ET AL. | 481

[29] M. M. Lohrengel, A. Moehring, M. Pilaski, Fresenius J. Anal. [40] K. S. Siow, T. Y. Song, J. H. Qiu, Anti‐Corros. Methods Mater.
Chem. 2000, 367, 334. 2001, 48, 31.
[30] ASTM D1141‐98. Standard Practice for the Preparation of [41] H. Wilson, A. Erbe, Electrochem. Commun. 2019, 106, 106513.
Substitute Ocean Water, Tech. Rep. ASTM International, West [42] H. Luo, X. G. Li, C. F. Dong, K. Xiao, Arabian J. Chem. 2017,
Conshohocken, 1998. 10, S90.
[31] K. J. Jenewein, A. Kormányos, J. Knöppel, K. J. J. Mayrhofer, [43] O. Ø. Knudsen, A. Forsgren, Corrosion Control Through
S. Cherevko, ACS Meas. Sci. Au. 2021, 1, 74. Organic Coatings, CRC Press, Boca Rayton, USA. 2017.
[32] V. Shkirskiy, F. D. Speck, N. Kulyk, S. Cherevko, [44] S. Kumari, S. Wenner, J. C. Walmsley, O. Lunder,
J. Electrochem. Soc. 2019, 166, H866. K. Nisancioglu, J. Electrochem. Soc. 2019, 166, C3114.
[33] N. Jadhav, V. J. Gelling, J. Electrochem. Soc. 2019, 166, C3461. [45] D. Cicolin, M. Trueba, S. P. Trasatti, Electrochim. Acta 2014,
[34] J. P. Kollender, A. I. Mardare, A. W. Hassel, Electrochim. Acta 124, 27.
2015, 179, 32.
[35] Y.‐H. Chen, A. Erbe, Corros. Sci. 2018, 145, 232.
SU PPOR T ING INF OR MATI ON
[36] A. Erbe, A. Sarfraz, C. Toparli, K. Schwenzfeier, F. Niu, Soft
Additional supporting information can be found online
matter at aqueous interfaces, Vol. 917 of Lect. Notes Phys.,
Springer, Cham, Switzerland 2016, p. 459. in the Supporting Information section at the end of this
[37] A. Erbe, S. Nayak, Y.‐H. Chen, F. Niu, M. Pander, article.
S. Tecklenburg, C. Toparli, Encyclopedia of interfacial
chemistry, Elsevier, Oxford 2018, p. 199.
[38] A. Altin, M. Krzywiecki, A. Sarfraz, C. Toparli, C. Laska,
P. Kerger, A. Zeradjanin, K. J. J. Mayrhofer, M. Rohwerder, How to cite this article: F. Forr, E. Mysliu, J. H.
A. Erbe, Beilstein J. Nanotechnol. 2018, 9, 936. Nordlien, A. Erbe, Mater. Corros. 2023;74:464–481.
[39] I. Jevremović, Y.‐H. Chen, A. Altin, A. Erbe, Corrosion
https://doi.org/10.1002/maco.202213533
Inhibitors in the Oil and Gas Industry, Wiley‐VCH, Wein-
heim, Germany 2020, pp. 359‐382.