Street Art in the Rain: Evaluating the Durability of Protective Coatings for Contemporary Muralism Through Accelerated Rain Ageing

Abstract

Contemporary muralism has gained increasing cultural and social relevance in recent years, becoming a prominent form of urban artistic expression. However, its outdoor exposure makes it highly vulnerable to environmental degradation, raising significant challenges for long-term preservation. While solar radiation is widely recognized as a main agent of deterioration, the impact of rainfall has received comparatively little attention. This study addresses this gap by evaluating the durability of commercial protective coatings applied to modern paints (alkyd, acrylic, and styrene-acrylic) under simulated rain exposure. The ageing protocol replicates approximately 10 years of cumulative rainfall in Central-Southern Europe. A key innovation of this research is the use of a custom-built rain chamber, uniquely designed to expose a large number of samples simultaneously under highly uniform and controlled rain conditions. The system ensures reproducible exposure through a precision-controlled moving platform and programmable rain delivery. A comprehensive set of analytical techniques was employed to assess morphological, chemical, and functional changes in the coatings and paints before and after ageing. Results highlight the limited performance of current protective materials and the need for more effective solutions for the conservation of contemporary outdoor artworks.

1. Introduction

Contemporary muralism is an artistic phenomenon that has increased its social and cultural relevance in recent years. It represents a dynamic and visually powerful artistic form of expression, that unfortunately is highly vulnerable to environmental degradation, posing a major challenge for its long-term preservation [1,2,3]. The fading and degradation of painting surfaces exposed to outdoor conditions, caused by atmospheric agents (such as solar irradiation, temperature and relative humidity fluctuations, different and extreme rain regimes, and air pollution), require effective solutions, to preserve not only their material integrity, but also their historical, artistic and social value [4,5,6]. In fact, these degradation mechanisms can lead to discoloration, chalking, flaking, and loss of adhesion of the painting layers and finally to the rapid loss of readability of the artworks [7,8,9,10,11].

The application of transparent and durable protective coatings to the painted surfaces represents a possible solution, as explored in a recent review [4] and a previous experimental paper [12]. These coatings should act as a barrier against humidity, rainfall, dust deposition, pollution and vandalism and protect the painted surfaces from UV solar radiation. However, selecting a suitable and durable coating is a difficult challenge because the protective material should satisfy multiple requirements: effective protection against rainwater and humidity; chemical inertness towards the painting layers and the environment; and long-term durability in severe outdoor conditions [4,12,13]. Regarding compatibility and efficacy against liquid water and humidity, a previous study [12] led to some general conclusions, showing that the products currently available on the market are not completely satisfactory, because they offer only a modest barrier to water absorption; moreover, it was observed that the chemical nature of the paint and its adhesion to the substrate are crucial factors for the conservation of this type of mural paintings, while the substrate preparation and the use of a primer coating on the walls could be decisive measures for prolonging their durability.

Regarding the durability of the applied protective coatings, several studies [5,14,15,16,17] have investigated their stability to solar irradiation, while rainfall exposure is an environmental condition that is rarely considered. Studying the chemical and mechanical impact of rain on the painting layers with and without protective coatings is dramatically important to assess the possibility of protecting these artworks in outdoor conditions.

To overcome this gap, a novel experimental setup was introduced: a custom-built rain chamber capable of simulating 10 years of cumulative rainfall in Central-Southern Europe. The system enables uniform rain exposure, allowing comparative evaluation of the performance of commercial protective coatings applied to different types of paints, and providing new insights into the durability of those protective materials under rain exposure.

In an urban polluted context, as it is well known [8,18,19], rainwater contains chemicals and acid ions, which can react with painting materials, accelerating the deterioration of pigments, binders, and underlying substrates. Although the pH values of rainwater have significantly increased in the last decades [20], the presence of air pollutants (especially in urban or industrial areas) can still have a negative impact. Moreover, the effect of rain results from a combination of the mechanical action of water droplets and chemical reactions such as hydrolysis and oxidation. The resulting damage to the polymeric structure of paints and protective coatings can also lead to the release of compounds potentially hazardous to human health and the environment [21]. Additionally, when rainwater accumulates on walls, it can promote the formation of soluble salts which, as water evaporates, crystallize and cause structural damage, such as cracking or detachment of painting layers [22]. The adverse conditions generated in recent decades by climate changes [23,24] suggest that the increased and intense rainfall is a crucial factor in the conservation of buildings and heritage structures, including the external walls used by street artists.

The experiments in this study were motivated by the need to investigate how rainfall affects the durability of modern painting materials and commercially available protective coatings, and to evaluate whether these materials are sustainable and durable under outdoor urban conditions.

To accurately simulate rainfall conditions in the laboratory, it is essential to determine: (a) the geographical area of interest, (b) the average annual rainfall in that area, (c) the chemical quality of the rainwater, particularly its pH, and (d) the duration of the experiment. The first two points are interrelated. To define the appropriate amount of water for simulating a realistic rain regime, the central-southern Europe area was selected as the reference region. The cumulative annual precipitation values of this area were estimated through the open digital application Art-Risk 5 (using the satellite Persiann-Climate Data Record CDR), and the average value over the last 20 years was estimated at 720 mm/year [25,26], as shown in Figure 1.

Regarding rainwater quality under laboratory conditions, a highly influential and controllable parameter is its pH. Previous studies have used acid rain as a reference, typically with a pH ranging from 3.0 to 4.5 [19,27]. However, Timoncini et al. [20] have recently shown that rainwater can no longer be considered an acidic medium, showing in Central Europe an average pH value above 5.6 in atmospheric deposition. This shift is mainly attributed to the significant reduction in acidifying emissions (SO₂ and NO_x) due to European air quality policies, the relatively stable ammonia emissions, and the increasing frequency of alkaline events (such as Saharan dust outbreaks rich in CaCO₃), which contribute to the neutralization of acidity in atmosphere [20]. For this reason, demineralized water, with a pH value of 6.2 and a conductivity of 1.45 µS/cm, is deemed suitable for the ageing experiments. However, it is important to acknowledge that the chemical composition of real rainwater, especially in urban environments, can differ significantly due to the presence of pollutants, ions, and particulate matter. These components may influence degradation mechanisms through acidification, salt formation, or catalytic effects. Demineralized water was used in this study to ensure experimental reproducibility and to isolate the effects of water-induced dissolution and mechanical action under low-pollution conditions. As such, the potential impact of pH variation and other atmospheric pollutants was not reproduced in this setup. Regarding the duration of the experiment, the rain cycle was set up to simulate a total exposure to rainfall of about 10 years in Central Southern Europe, with a cumulative rainfall of 7320 mm. The core of our accelerated ageing strategy is an innovative, custom-built rain chamber, entirely developed and refined within our laboratory. It allows for precise control over rainfall intensity and water composition. The distinctive feature of our prototype lies in its ability to expose a large number of samples simultaneously to highly uniform rain conditions. This is achieved through a dynamic sample platform mounted on a precision-controlled Cartesian (X-Y) motion system, which ensures homogeneous exposure across all specimens.

Rain droplets are delivered through a dense network of hypodermic stainless-steel needles, while the samples are positioned on a perforated PMMA tray inclined at 45°, optimized for both water impact and drainage. The tray’s movement is driven by high-precision servomotors and controlled via a custom FISHINO-based system (an Italian Arduino-compatible board with integrated Wi-Fi), enabling programmable and repeatable motion patterns. This setup guarantees consistent and reproducible exposure conditions, making it particularly suitable for comparative studies on the durability of protective coatings. The accelerated rain cycle was optimized to last 15 daily cycles consisting of 1 h of softer rain, followed by 6 h of heavy rain, and then 17 h of drying, per day. Over the 15-day experiment, the samples were exposed to simulated rain for a total of 105 h, corresponding to approximately 10 years of cumulative rainfall in Central-Southern Europe. The device used for rain ageing is an experimental, recently improved, in-house prototype of a rain chamber which uses a network of very fine sanitary stainless-steel needles to deliver the rain droplets and a motorized plane that can host the specimens set with the painted surface inclined at 45° [28]. The plane moves during the ageing so that the rainfall is uniformly distributed on the mock-up surface of interest (in this case the painted surface).

The present study aims to investigate the stability of a selection of commercial protective coatings [12] under accelerated rain exposure. A total of 72 cement-based mortar mock-ups were prepared, painted with three different types of paint—alkyd, acrylic, and styrene-acrylic—applied both with and without a primer layer, and treated with five different commercial protective coatings. They were rain-aged in the above-mentioned conditions. A testing protocol, developed and implemented in a previous study [12] was carried out, before and after ageing. The selected testing methods allowed to investigate the aesthetic changes (surface morphology, roughness and colour) and the protective efficacy of the coatings (surface wettability and water absorption by capillarity), after rain ageing. Additionally, chemical changes in the coatings were investigated by ATR-FTIR spectroscopy.

It is important to highlight that the artificial ageing protocol adopted in this study does not fully replicate the complexity of natural weathering, which typically involves alternating more frequent wet and dry cycles, as well as simultaneous UV radiation and temperature fluctuations. While a more realistic simulation would ideally include these variables, such a setup would not allow for the isolation of the specific effects of rain. In contrast, the approach used here was designed to focus exclusively on the mechanical action of water droplets and the chemical effects of water as a solvent and reaction medium. This controlled strategy enabled us to investigate the degradation mechanisms induced by rain alone, excluding the influence of other synergistic environmental factors.

2. Materials and Methods

2.1. Painting Mock-Ups Preparation

Protective products were applied to a total of 72 cement-based mortar mock-ups (5 × 5 × 2.5 cm), whose preparation followed a protocol described in detail in a previous study [12]. Here, only the key features are summarized. The mock-ups were cast in silicone molds using a representative cement mortar mix and cured under controlled laboratory conditions. After curing, the surface was abraded to simulate outdoor wall textures. Half of the samples (36) were treated with an acrylic-impregnating primer (Alphatex SF, Sikkens, Novara, Italy) [29], applied by brush to improve paint adhesion and simulate common wall preparation practices.

Three types of magenta-coloured paints were selected to represent paints formulations in contemporary muralism: an acrylic emulsion spray paint (“Acr”, Montana Colours, Barcelona, Spain), an alkyd resin spray paint (“Alk”, Molotow, Lahr/Schwarzwald, Germany), and a styrene-acrylic emulsion applied by brush (“Sty”, Sikkens, Novara, Italy). The Acr and Alk paints were applied by spray in three successive layers using fine nozzles (super skinny, 0.5–2 mm), following the manufacturers’ instructions. The Sty paint was diluted and applied in two orthogonal coats using a 38 mm brush. The resulting mock-ups reproduce realistic stratigraphies found in street art, with and without primer, and were used to evaluate the performance of protective coatings under simulated rain exposure.

2.2. Protective Treatments

Five protective commercial coatings belonging to three different chemical macro-classes were selected based on the results of the interviews and questionnaires, the experiences of professionals in conservation, and the most recent literature in a previous study [12]. The list is reported in

Table 1. List of the selected protective coatings applied for experimental work with information regarding abbreviations and chemical composition.

Paint Abbreviation	Chemical Composition from Technical Datasheet	Chemical Identification *
Acr	Acrylic emulsion	Acrylic emulsion + PR122 + PR254 + PW6
Alk	Alkyd resin	Alkyd resin + PR122 + PW6 + Talc (filler)
Sty	Styrene-acrylic emulsion	Styrene-acrylic emulsion + Eosin B + CaCO3 (filler)
Protective Treatment Abbreviation	Product class from Technical Data Sheet	Chemical Identification *
A1	Acrylic	Acrylic polymer, polyurethane, Si-containing filler
A2	Acrylic	Copolymer MA-EMA
S1	Silane	Organic silane, acrylic polymer
S2	Silane/Siloxane	Dimethyl-siloxanes
SF3	Fluoro-silane	Undisclosed formulation (fluoro-silane from technical datasheet)

* Complete information is reported in the PRIN SuperStar project deliverables and in papers under preparation.

. The protective coatings were applied following the procedures described in detail in a previous study [12], which also reports the selection criteria, chemical composition, and performance evaluation of the products. Here we just briefly recall the application procedure and the main results. The coatings were applied to the painted surfaces of the mock-ups using a pipette to dispense the recommended amount of product, which was then evenly spread with a soft-bristle brush. Most formulations were used as they are sold, while A2 and S1 required specific dilution protocols. After application, the samples were left to dry under laboratory conditions for 15 days. The previous work [12] demonstrated that, although some coatings exhibited moderate water repellency and good aesthetic compatibility, their overall protective efficacy was modest and strongly influenced by the type of paint and the presence of a primer layer.

2.3. Rain Ageing

2.3.1. Chamber

The rain ageing was performed with a custom-built rain chamber, entirely developed and refined within the laboratory at the Politecnico of Milan. Like other systems, it allows for precise control over rainfall intensity and water composition. The distinctive feature of our prototype lies in its ability to expose a large number of samples simultaneously to highly uniform rain conditions. This is achieved through a dynamic sample platform mounted on a precision-controlled Cartesian (X-Y) motion system, which ensures homogeneous exposure across all specimens. The chamber consists of a stainless-steel structure shown in Figure 2a. A Peristatic pump 330 (Behr Labor-Technik GmbH, Düsseldorf, Germany) draws the solution from an external dedicated reservoir and delivers it, at variable flow rates, to a reservoir placed on the back of the rain chamber. This reservoir branches into nine segments, each controlled by a valve to regulate the water flow. Each segment consists of a tank positioned on the chamber’s plane roof, equipped on the bottom part with 130 medical grade stainless steel hypodermic needles (27G, Pikdare S.p.A, Como, Italy), for a total of 1170 needles (0.35 needles/cm²), that release water droplets due to the combined effects of pump pressure, hydrostatic pressure, and gravity (Figure 2a). The system has been designed so that the hydrostatic pressure is the same in each tank. This system offers high flexibility, enabling variable flow rates (17–66 mL/min) for diverse rain intensity simulations (10–90 mm/hour) and programmable cycle durations, allowing simultaneous analysis of numerous samples. Interconnected transparent silicone tubes above the valves ensure uniform water levels across all reservoirs, significantly increasing attainable flow rates and improving precipitation homogeneity. The water droplets fall onto a PMMA tray, which holds a PMMA sample holder platform designed to host the samples for the experiment, inclined by 45° for optimal wettability. The tray is mounted on metal supports controlled by four stepper motors, allowing the control of both directions (x, y) in the horizontal plane. This advanced movement system precisely translates the sample holder along independent X and Y axes, controlled electronically via a Tangible User Interface (TUI). Mechanical motion is achieved through trapezoidal lead screws converting the rotational input of precision-controlled stepper motors into linear translation. Electronic control of the stepper motors is operated by a Fishino Shark board. A synchronized second motor (Y2) prevents system rigidity loss on the Y-axis. Stepper motors were chosen for their suitability for precise, continuous, low-speed movement, ensuring uniform rain coverage. Motors are positioned externally to prevent rain damage, with motion transferred via direct couplers. To manage up to an approximate 25 kg load, a weight discharge system was integrated. This system uses support bars sliding through recirculating ball bushings, transferring the load to an external aluminum profile structure via linear guides, minimizing friction and ensuring smooth, stable sample movement. The sample tray is moved following a fixed scheme. The scheme consists of a repetition cycle of 14 steps of 0.15 cm along the Y axis and 12 steps of 0.8 cm along the X direction, to provide a uniform washing of the mock-up surface (Figure 2b) [30,31,32]. The drainage of water is collected in a separate tank in order, if needed, to analyse the water released during the rain ageing. The rain rate (in mm/h) is measured with a commercial pluviometer (rain gauge DROP, TFA Dostmann, Wertheim, Germany).

The rain rate distribution maps (Supplementary Figure S5), obtained by measuring rainfall intensity across the entire moving tray under two selected conditions, revealed a moderate variability, with values of 33.15 ± 3.00 mm/h and 19.42 ± 2.78 mm/h. This variability was further minimized by the continuous rotation of the mock-ups along the tray during the ageing experiment, ensuring more homogeneous exposure.

2.3.2. Accelerated Rain Ageing Conditions

The test was conducted for a total of 15 days, with 7 h of raining time per day: 1 h of softer rain at ~30 mm/h followed by 6 h of heavy rain (71 ± 8 mm/h); then, 17 h of drying at room temperature. At the end of each ageing day, the samples were rotated at 90° and shifted to 1 position along the Y-axis of the PMMA sample platform. Rain consisted of demineralized water (average pH 6.2 value, and conductivity of 1.45 µS/cm). The experiment run for a total time of 105 h, with a total rainfall of 7320 mm, corresponding to about 10 years of outdoor exposure in Central-Southern Europe (according to the estimation described in Section 1). In

Table 2. Legenda for the abbreviations that describe the mock-ups’ experimental preparation and ageing.

Mock-Ups Abbreviation	Meaning
untr	Painted mock-up without protective treatment (untreated)
PR	Painted mock-up with primer layer between mortar and paint
nra	Mock-up before rain ageing (not rain aged)
ra	Mock-up after rain ageing (rain aged)

, the abbreviation used in the paper to indicate the mock-ups’ experimental preparation and ageing conditions are reported.

2.4. Testing Methods and Measurements

Microscopy. The morphological changes observed after rain ageing on protective coatings were examined by using a Leica M205C stereomicroscope (Leica Microsystems, Milan, Italy) with an objective of 0.78× and a Leica DM6 3D optical microscope (Leica Microsystems, Milan, Italy) with an objective of 20×. The acquired images were processed by using LASX software (v. 5.1) according to the protocol explained in a previous paper [12,33].

SEM-EDX. A Zeiss EVO 50 EP (Carl Zeiss, Jena, Germany) environmental scanning electron microscope, equipped with an EDS Bruker Quantax 200 spectrometer, was used to perform SEM analyses. The operating conditions were the same as those described in a previous study [12].

Colorimetry. The aesthetic changes before and after rain ageing were studied by using a portable Spectro-colorimeter CM-2600d (Konica Minolta, Tokyo, Japan). Following the standard protocol EN15886 [34], the instrument used a D65 light source at 10°, with a spot size of around 8 mm. For each mock-up, 15 spots were measured, and the CIELab coordinates were averaged. The operating parameters followed the methodology described in a prior study [12].

Glossmetry. Gloss was evaluated with a multi-gloss 268 glossmeter (Konica Minolta, Tokyo, Japan), able to work with incidence angles of 20°, 60°, and 85°. Measurements were carried out on the planar sample surfaces according to the standard protocol UNI EN ISO 2813 [35]. For each mock-up, 10 spots were measured. The same operating parameters as detailed in a previous study [12] were applied.

Static contact angle. The protective effectiveness of the applied treatments after rain ageing was evaluated considering the wettability measurements, performed according to the standard protocol EN15802 [36]. The static contact angle measurements were carried out by using a designed prototype, measuring 20 contact angles on different spots for each mock-up. For more information about instrumental setup and data acquisition see [12].

Water absorption by capillarity. These measurements follow the indications reported in the standard protocol UNI 10859 [37] and the tests were conducted under the same conditions and protocol as described in an earlier publication [12]. Tests were performed in duplicate, with two sets of replicates conducted in parallel.

ATR-Fourier-Transform Infrared Spectroscopy. ATR-FTIR measurements were carried out by a Thermo Nicolet iZ10 MX spectrometer (ThermoFisher Scientific Inc., Waltham, MA, USA) equipped with a Smart iTX Diamond accessory, employing a DTGS detector. For each mock-up, 15 measurements were acquired, in the spectral range between 4000 and 480 cm⁻¹ with 64 scans and a 4 cm⁻¹ resolution. Spectra were acquired by using OMNIC software (v. 9.11, ThermoFisher Scientific Inc., Waltham, MA, USA) and elaborated by OPUS software (v. 8.5, Bruker, Billerica, MA, USA). Raw absorption spectra were normalized by the min-max function provided by OPUS software in the range of 2930 and 1750 cm⁻¹ with maximum peak intensity (aliphatic CH stretching at 2920 cm⁻¹) equal to 2. After the normalization, the 10 spectra were averaged and, from the resulting average spectrum, the intensity values of the peaks of interest were measured at different frequencies (

Table 3. List of peak ratios evaluated, their respective wavenumbers and integration band ranges used for the FTIR data evaluation of untreated and treated mock-ups.

Functional Groups	Peak Wavenumber (cm−1)	Peak Intensity Measurements Range (cm−1)
OH stretching	3440	3510–3085
C=O stretching	1720	1760–1700
Calcite	875	890–860
Out-of-plane bending (phenyl)	700	730–680
SiCH3 stretching	880	935–870
C-F stretching	1140	1220–1170
C=O stretching of β-diketones	1635	1650–1620
C-H stretching	2920	2930–2910

) and internal peak ratios were calculated (

Table 4. List of the selected peak ratios, used to evaluate organic matrices degradation during the accelerated ageing.

Selected Peak Intensity Ratios for Untreated Mock-Ups		Selected Peak Intensity Ratios for Treated Mock-Ups
C=O/OH	Calcite/C=O	C=O/CH
C=O/C=O β-diketones	Phenyl/C=O	Si-CH3/OH
		CF3/SiO

) before and after rain ageing to evaluate materials degradation.

3. Results and Discussion

3.1. Morphological Changes After Rain Ageing

From the comparison of the microscopic observations of the three untreated paints, before and after rain ageing (Figure 3), it is possible to notice that Acr and Sty paints do not show any visible degradation pattern, while Alk paint shows an evident blistering effect, i.e., the formation of bubbles, lifting the paint film from the substrate [38]. This effect on Alk paint is reduced when the primer is present between the paint layer and the substrate. It was observed that these bubbles tend to deflate during the drying phase and then swell again following subsequent rain exposure. This suggests that, during rain ageing, water infiltrates either between the paint layer and the substrate or between the paint layer and the primer and is also apparently absorbed by the paint.

This may be related to the fact that Alk paint is prone to lose adhesion when applied to a ceramic porous substrate due to the reaction between the oil in the paint and the alkalinity of the substrate. This phenomenon, commonly called saponification, is the hydrolysis reaction in which the ester bonds between the fatty acids and glycerol of the triglyceride, break. The saponification tends to reduce the adhesion of the paint film to the mortar substrate [39,40].

To confirm this hypothesis, FTIR analysis was performed on a fragment of the Alk paint in order to make a comparison between the absorption spectrum of the inner portion of the paint (closer to the substrate) and the surface of the paint (Figure S1). The following changes were observed: an increase in the intensity of the OH peak at 3346 cm⁻¹, a decrease of the carbonyl group absorption peak (C=O at 1721 cm⁻¹), and an increase of the β-diketones peak at 1635 cm⁻¹, that is an indicator of ageing of alkyd paints [41]. The C=O/β-diketones intensity ratio of the spectra collected on the paint surface and on the paint portion in contact with the mortar are 2.5 and 1.3, respectively confirming the saponification reaction and the consequent loss of adhesion of the alkyd paint. Interestingly, a new peak was observed at 1540 cm⁻¹, identified as metal carboxylates (more likely zinc or lead metal stearate) [42,43]. The formation of these products can be related to the presence of these metals in the substrate (or in the paint formulation) which can bind to free fatty acids forming the organic salt [44]. The identification of these metal soaps is important because they can lead to further degradation effects such as protrusions on the surface, the presence of efflorescence, delamination and cracking [45].

When the primer is present, the blistering phenomenon of Alk paint is reduced. A possible explanation of this evidence is that the primer is chemically unsaponifiable, as stated in its technical datasheet [29]. This means it does not react with alkaline components of the substrate (e.g., lime or cement), which are known to trigger saponification reactions in alkyd resins. By acting as a chemically stable barrier, the primer likely inhibits the migration of alkaline ions and prevents direct contact between the alkyd resin and the high-pH substrate. Therefore, it is concluded that the presence of the primer is a crucial point for a better resistance to rain.

Through optical microscopic, it was possible to point out some other signs of degradation after rain ageing. The Alk and Sty paints exhibit a reduction in surface gloss, and a different surface roughness (Figure 4). This effect is most probably connected to rain erosion, resulting from both the mechanical impact of water droplets and chemical interactions between rainwater and the surface materials, that can cause a partial washing away of the surface organic binder. From the evaluation of the roughness values (Figure 4), it is evident that for all three paints, there is an increase in surface roughness values, slightly greater for Sty paint followed by Alk paint and Acr paint.

The morphological analysis of the aged samples treated with protective coatings clearly reveals that the degradation patterns are strongly influenced by both the chemical nature of the paint and the type of protective product applied. Each paint-coating combination responds differently to rain exposure, underlining the importance of material compatibility in determining long-term performance. For instance, the alkyd paint treated with the A1 acrylic coating exhibited severe blistering and partial detachment of the protective film, particularly in the absence of a primer layer (an effect not observed with other paint types or coatings). This highlights how specific interactions between the coating, the paint matrix, and the substrate can critically affect the durability of the protective system. The following sections present in detail the experimental results that support these observations, highlighting the specific degradation patterns and protective behaviours associated with each paint-coating combination. A1 coating shows the greatest signs of degradation, mainly for Alk paint, followed by Acr and Sty paints (Figure 5). In the case of Alk paint, the rain caused the blistering of the paint film from the substrate, as well as the detachment of A1 coating from the paint (as evident in the grey areas in Figure 5 Alk_A1_ra and in Figure S2). Interestingly, this effect is particularly visible in the sample without primer showing a distribution mainly in the central part of the sample. In contrast, for samples with primer, paint blistering from the substrate is still observed (mainly for Alk paint), while coating detachment is practically absent (except in Acr paint). In Sty paint, only slightly more opaque areas are observed.

In Figure 6, the blistering effect is consistently observed across all Alk-painted mock-ups, with varying intensity depending on the protective coating applied. Moreover, the swelling of the paint layer often results in the rupture of the protective film, especially visible for the protective materials A2, S2 and SF3 (Figure S3). A further crucial point concerns the presence of the primer between the Alk paint and the substrate: even when the coating is applied over a complete stratigraphy (mortar/primer/alkyd paint), blistering still occurs in contrast to what is observed in the untreated PR_Alk_untr mock-ups, where the primer alone appears to prevent saponification and therefore blister formation). This phenomenon can be attributed to the presence of the protective coating, which introduces an additional barrier layer that hinders the rapid evaporation of water from the mock-up surface. As a result, moisture becomes trapped beneath the paint film, promoting the formation of blisters.

In the case of Acr paint, any significant degradation phenomenon was observed after rain ageing. However, in samples treated with A2 and S1 coatings, the formation of distinct agglomerates of protective material was detected, regardless of the presence of a primer layer. These accumulations suggest a possible incompatibility or poor film formation of the coating on the acrylic surface. In contrast, the treated aged Sty-painted mock-ups did not exhibit any evident morphological alterations following rain exposure, indicating a higher surface stability and better interaction with the applied protective coatings under the tested conditions.

In order to gain a deeper insight into the morphological alterations caused by rain exposure, a detailed Scanning Electron Microscopy (SEM) analysis was performed. Clear evidence of mechanical failure was observed across all samples, regardless of the chemical class of the applied protective product (Figure 7). In the case of acrylic-based coatings (A1 and A2), it is evident that both are able to form a continuous film over the painted mock-ups; the film is thicker and very smoother in the case of A2, while it is rougher for A1 due to the presence of SiO₂ particles as additive [12]. After ageing, cracks appear in the protective layer, associated with a subsequent micro-flaking phenomenon that exposes the morphology of the underlying paint. The difference between the two coatings is that A2 tends to fracture more sharply, while A1 crumbles more easily, especially at the edges of the crack, leaving visible coating residues on the surface. This behaviour could be attributed to the presence of inorganic additives.

At the stereomicroscopic level, silane-based and fluorinated coatings (S1, S2, and SF3) already exhibited signs of surface erosion; under SEM observation, distinct morphological differences emerged depending on the applied coating (Figure 7). S1 showed localized erosion, with partial loss of surface material, exposing isolated areas of the underlying painted layer painted layer. The protective coating S2 forms a thicker, smoother and more continuous layer on the mock-up surface, which remained largely intact after ageing, except for some localized cracking and lifting likely caused by the mechanical and physio-chemical action of water. In the case of SF3, a dense network of superficial micro-cracks was observed across the entire coating surface. After rain ageing, the coating layer exhibits widespread craquelure, which likely compromises its protective performance. Figure S4 provides the SEM images of the untreated and unaged Alk paint surface, allowing comparison with the aged and treated surfaces.

3.2. Colorimetric and Gloss Changes After Rain Ageing

The evaluation of colour and gloss variations plays a key role in assessing the aesthetic durability of painted surfaces after exposure to environmental stressors. In this study, chromatic and surface reflectance changes were monitored before and after the accelerated rain ageing to quantify potential visual alterations in the painted mock-ups. In

Table 5. Colorimetric evaluation according to the total colour variation (ΔE) and total gloss unit variation (ΔGU) on treated samples after rain ageing, with and without primer.

	ΔE	ΔGU		ΔE	ΔGU
Alk_untr_ra	1.6 ± 0.5	−0.05 ± 00.2	PR_Alk_untr_ra	0.7 ± 0.2	−0.02 ± 0.01
Alk_A1_ra	4.1 ± 1.1	−0.3 ± 0.05	PR_Alk_A1_ra	1.1 ± 0.3	−0.1 ± 0.02
Alk_A2_ra	2.2 ± 0.4	−0.1 ± 0.01	PR_Alk_A2_ra	2.5 ± 0.1	−0.08 ± 0.02
Alk_S1_ra	3.1 ± 0.6	−0.03 ± 0.06	PR_Alk_S1_ra	0.4 ± 0.2	−0.02 ± 0.02
Alk_S2_ra	2.9 ± 1.1	−0.3 ± 0.04	PR_Alk_S2_ra	0.6 ± 0.3	−0.3 ± 0.01
Alk_SF3_ra	2.9 ± 1.3	−0.4 ± 0.02	PR_Alk_SF3_ra	0.4 ± 0.1	−0.2 ± 0.01
Acr_untr_ra	3.6 ± 0.5	−0.2 ± 0.02	PR_Acr_untr_ra	1.2 ± 0.3	−0.1 ± 0.03
Acr_A1_ra	4.4 ± 1.0	−0.4 ± 0.05	PR_Acr_A1_ra	2.4 ± 0.2	−0.05 ± 0.02
Acr_A2_ra	2.6 ± 0.9	−0.06 ± 0.03	PR_Acr_A2_ra	2.8 ± 0.5	−0.06 ± 0.04
Acr_S1_ra	1.9 ± 0.7	−0.2 ± 0.08	PR_Acr_S1_ra	0.4 ± 0.5	−0.1 ± 0.03
Acr_S2_ra	1.9 ± 1.0	−0.2 ± 0.05	PR_Acr_S2_ra	0.2 ± 0.1	−0.2 ± 0.02
Acr_SF3_ra	4.2 ± 0.5	−0.6 ± 0.02	PR_Acr_SF3_ra	0.3 ± 0.2	−0.4 ± 0.03
Sty_untr_ra	2.0 ± 0.7	−0.1 ± 0.03	PR_Sty_untr_ra	3.1 ± 0.6	−0.08 ± 0.01
Sty_A1_ra	0.5 ± 0.5	−0.4 ± 0.02	PR_Sty_A1_ra	1.1 ± 0.7	−0.1 ± 0.01
Sty_A2_ra	0.5 ± 1.1	−0.3 ± 0.04	PR_Sty_A2_ra	1.1 ± 0.4	−0.3 ± 0.02
Sty_S1_ra	1.5 ± 0.8	−0.1 ± 0.03	PR_Sty_S1_ra	0.5 ± 0.6	−0.07 ± 0.01
Sty_S2_ra	3.0 ± 0.6	−0.5 ± 0.06	PR_Sty_S2_ra	0.2 ± 0.1	−0.4 ± 0.02
Sty_SF3_ra	1.1 ± 0.9	−0.4 ± 0.02	PR_Sty_SF3_ra	0.2 ± 0.1	−0.2 ± 0.01

and Figure 8, the colour change (ΔE) of all the mock-ups after ageing is reported and represented in the ΔC* and ΔL* planes. Considering the three types of untreated mock-ups after rain ageing, Alk and Sty are the two that show a negligible change of colour with ΔE of 1.6 and 2, respectively. Acr paint, on the other hand, shows a ΔE of 3.6. Alk and Sty paints show an increase in ΔL* value and a decrease in Δa* and Δb*, indicating the tendency of the paint to become paler, in the case of Acr the value of Δa* increases, while that of ΔL* and Δb* decreases indicating a darkening of the colour. When a primer is applied, the colour change is less pronounced.

Observing Figure 8, it is worth noting that the protective coatings do not affect much the surface colour of the paints. Only the silane-based treatments S1 and S2 saturate the colour of Alk paints. After ageing, the tendency is to slightly decrease ΔC* values, while ΔL* is maintained almost unchanged (ΔL* < 2). Overall, in the case of Alk paint, a desaturation of the colour occurs after rain ageing (Figure 8a). It should be also observed that S2 is the treatment that most saturates the colour, an effect that is completely lost after ageing (the blue stars are shifted towards negative values of ΔC*, for all paints). The presence of the primer reduces the overall colour change, particularly for Alk and Acr (Table 4). This effect may be due to the primer reducing water penetration, particularly by limiting water uptake from the mortar substrate.

Gloss measurements reveal that the exposure of mock-ups to rain action causes a decrease of gloss, that is negative ΔGU values (Table 5). This effect can be ascribed to the water mechanical erosion of the very surface [46,47] and to an increase of the surface roughness; in particular, it is highlighted that A1 and SF3 show the highest negative ΔGU values (i.e., a decrease of surface gloss after ageing) while S1 and A2 show the lowest.

To better understand the overall impact of rain ageing on the visual appearance of the painted surfaces, and the protective action of coatings, the main findings in terms of colour and gloss changes are summarized and discussed below.

Regarding aesthetic changes of paints after accelerated rain ageing, the measured values generally remained within tolerance thresholds typically accepted for outdoor painted surfaces. It is worth noting that, as discussed in a previous work [12], upon application, the protective coatings did not significantly alter the visual appearance of the paints in terms of colour or gloss, indicating good initial aesthetic compatibility.

After ageing, uncoated samples exhibited moderate changes, with colour variations (ΔE) generally below or around 3 units and limited gloss loss. These alterations, although perceptible in some cases, did not compromise the overall readability or aesthetic integrity of the painted surfaces.

The application of protective coatings, however, did not consistently improve the resistance to aesthetic alteration. In some cases, the coatings provided a slight reduction in colour change (particularly when combined with a primer layer) suggesting a partial barrier effect. Yet, in other instances, especially with A1 and SF3, higher ΔE and gloss losses were observed compared to the uncoated samples, likely due to coating degradation, surface roughening or film detachment.

These results underline that while coatings may play a role in modulating water interaction, they do not always enhance the aesthetic durability of painted surfaces under rain exposure. Their performance appears highly dependent on formulation and compatibility with the specific paint. Nonetheless, most observed changes remained within acceptable limits, and the overall visual perception of the painting layers was not severely compromised.

3.3. Protective Effectiveness After Rain Ageing

3.3.1. Static Contact Angles (WCA) Change After Rain Ageing

As discussed in a previous work [12], the static contact angles of untreated mock-ups showed rather low values (Acr 68° and Alk 69°, while Sty paint showed a θ equal to 92°, just above the hydrophobicity limit). After coating application, the surface water repellency slightly increases (higher WCA values), but only SF3 allows to reach θ values of 102° and 104°, without and with primer respectively.

After rain ageing, all coated samples showed a decrease in WCA values, as shown in Figure 9. Among them, only SF3, maintained an acceptable water repellency (WCA > 90°) across all three paints after rain ageing:

(1)

On Alk painted mock-ups, the most pronounced decrease in WCA was observed for A1 and S2, followed by A2 and SF3. S1 and the untreated samples showed the smallest WCA variation.

(2)

On Acr painted mock-ups, S2 showed the highest WCA reduction, followed by SF3 and the untreated sample. The smallest variation was observed for A2, A1 and S1.

(3)

On Sty paint, S2, A2 and A1 showed the greatest decrease in WCA values, followed by S1 and the untreated. The smallest variation was observed for SF3.

Overall, Figure 9a, highlights that, after rain ageing, the WCA values of most coated Alk-painted samples are even lower than those of the uncoated aged samples, indicating a worsening of surface water repellency due to the application of the protective treatments. The only exceptions to this trend are S2 and SF3, which show WCA values comparable to or slightly higher than the untreated Alk samples.

For what concerns Acr and Sty, the measured WCA values indicate that there is not a significant change in the surface water repellency after treatment and only a small decrease after rain ageing. SF3 coating, the fluorinated material, represents the only exception, although after rain ageing the WCA values dropped to the lower limit of hydrophobicity (around 90°).

Interestingly, Alk paint without any protective coating appears to better retain its inherently almost “water-repellent” character after rain exposure. As already noted, only SF3, the fluorinated material, is the only treatment that consistently preserves its characteristic water repellency after ageing. As expected, the presence of the primer does not substantially affect the WCA values, nor does it effectively mitigate the decrease induced by rain ageing (Figure 9b).

3.3.2. Water Absorption by Capillarity

The measurement of capillary water absorption is a key test to assess the protective behaviour of treatments, considering the prolonged contact of the painted surface with liquid water during the testing. In fact, it is well known that vertical surfaces exposed to rainfall can also absorb liquid water through capillarity and diffusion into the porous plaster layers and wall substrate. The ability of a protective treatment to reduce or prevent this water uptake is essential to ensure long-term durability [48,49]. The test evaluates whether degradation of the coatings due to prolonged rain exposure promotes increased water absorption into the painted mortar system.

In Figure 10a, a comparison of the water absorption curves for untreated painted mock-ups before and after rain ageing is presented. It is evident that rain exposure causes just a slight increase in water absorption for all the paints, with a more marked effect on Alk samples. Actually, the results show that Alk has higher IC_rel values than the other two paints, indicating a greater sensitivity to rain-induced degradation. The presence of the primer reduces this effect (Figure 10b), partially delaying the water uptake, although none of the three paints reaches the water absorption plateau (i.e., the maximum amount of water that could be absorbed by the sample during the remaining exposure time) as already discussed in a previous study [12].

After the application of the protective coatings, the absorption curves show a reduction in the amount of absorbed water for all paints and treatments. However, the extent of this reduction (ranging from 40% to 65%) is not fully satisfactory, as highlighted in previous findings [12]. After rain ageing, the protective coatings show slightly higher water absorption compared to the unaged condition, but they seem to maintain their moderate protective efficacy. Differences emerge depending on the type of applied coating: A1 shows the poorest performance after ageing, with an absorption curve that nearly overlaps with that of the untreated paint (Figure 11a). In contrast, S2 and SF3 perform better, showing only a limited increase in the curve trend after ageing (Figure 11b). This behaviour is consistent in all paint types (Alk, Acr and Sty).

The calculated IC_rel values (Figure 12a–c) clearly demonstrate that the protective efficacy of all treatments decreases after rain ageing, with IC_rel values higher than 1. The worst performance is observed for A2 applied to Acr-painted mock-ups. Interestingly, Sty-painted mock-ups show only a limited increase in capillary water absorption after ageing, (around 10%, Figure 12c), suggesting better interaction between paint and coating.

In summary, the commercial products tested offer limited protection against water absorption, both before and after ageing. Their action appears insufficient to ensure long-term water resistance in outdoor conditions. These results support the interpretation that the coatings function primarily as “sacrificial” layers, degrading under environmental stress while offering temporary protection to the underlying paint layers. Among the tested products, silane/siloxane-based coatings (S1 and S2) exhibit the most stable performance over time.

3.4. Chemical Changes After Rain Ageing

Chemical changes introduced by accelerated rain ageing were investigated by ATR-FTIR spectroscopy. For this analyses specific peak-ratios were selected for semi-quantitative evaluation, depending on the type of paint and protective coating (

Table 6. Peak ratios of ATR-FTIR spectra from untreated mock-ups.

		C=O/OH 1720/3440	C=O/C=O (β-Diketones Formation) 1720/1635
Alk_untr	nra	4.0	3.9
	ra	3.6	4.1
PR_Alk_untr	nra	4.5	4.3
	ra	4.4	4.6
		C=O/OH 1720/3440	C=O/CH 1720/2920
Acr_untr	nra	13.7	2.9
	ra	12.7	2.9
PR_Acr_untr	nra	17.3	2.9
	ra	16.4	3.0
		C=O/OH 1720/3440	Calcite/C=O 785/1720	Phenyl/C=O 700/1720
Sty_untr	nra	2.3	1.4	4.1
	ra	3.1	0.7	2.3
PR_Sty_untr	nra	2.5	1.2	3.8
	ra	2.9	0.7	2.5

).

In the case of untreated mock-ups, Alk and Acr paints exhibited only minor degradation after rain ageing; in particular, a slight increase in the intensity of OH stretching band [50,51] was observed. For the styrene-acrylic paint (Sty), the formulation appeared more affected by rain ageing: an increase in the C=O/OH ratio was evident, together with a marked decrease in the Phenyl/C=O ratio. A significant loss of the surface calcite filler was also observed as shown by a halving of the calcite/C=O ratio. The presence of the primer beneath the painting layers seems to slightly mitigate these spectroscopic changes after ageing, for all the analyzed paints.

The same evaluation was applied to the treated mock-ups.

Table 7. Peak ratios of ATR-FTIR spectra from treated mock-ups, divided per type of coating treatment (A1, A2, S1, S2, SF3), nature of paints (Alk, Acr, Sty) and ageing condition (not rain aged—nra, rain aged—ra). In the case of SF3, it was impossible to select suitable peak ratios since the composition of the coating was not disclosed.

Coating	Type of Paint		C=O/CH 1720/2920
A1	Alk	nra	1.67
		ra	1.67
	Acr	nra	1.70
		ra	1.68
	Sty	nra	1.62
		ra	1.61
A2	Alk	nra	6.23
		ra	5.77
	Acr	nra	4.66
		ra	4.36
	Sty	nra	4.99
		ra	4.48
			Si-CH3/OH 880/3440
S1	Alk	nra	4.5
		ra	3.3
	Acr	nra	4.6
		ra	3.3
	Sty	nra	7.6
		ra	5.4
S2	Alk	nra	11.0
		ra	10.0
	Acr	nra	10.4
		ra	8.1
	Sty	nra	9.5
		ra	7.5
			CF3/SiO 1140/1065
SF3	Alk	nra	2.1
		ra	2.8
	Acr	nra	2.3
		ra	1.9
	Sty	nra	2.1
		ra	1.3

reports the peak ratios for each coating-paint combination, before and after rain ageing.

For A1 and A2 coating, a reduction in the C=O/CH ratio suggests a modest deterioration of the acrylic matrix of the coatings, likely due to oxidative hydrolysis of carbonyl groups in the acrylic sidechain. Based on these results, A2 coating appears to undergo more substantial degradation than A1 during long-term rain exposure.

For S1 and S2 coatings, FTIR analysis revealed a small decrease in Si-CH₃/OH ratio following rain ageing, due to molecular structure degradation. The Si–CH₃ (methyl-silane) groups are inherently non-polar and play a crucial role in imparting hydrophobicity to the surface [52,53]. Their loss would lead to an increase in surface polarity and, consequently, higher surface energy. This trend was accompanied, in some cases, by a slight increase in OH bands, which may indicate hydrolysis and partial oxidation of the silane chains. It is important to note that these spectral changes are minor, indicating only limited chemical modifications.

Finally, the SF3 coating appeared rather stable under rain ageing, except for a notable reduction of the peaks at 1140 and 1065 cm⁻¹, associated with C-F stretching in CF₃ groups and Si-O stretching vibrations [54]. The reduction of the peak is most likely attributed to a degradation of –CF₃ groups, while the increase in the ratio observed on Alk may be associated with a loss of Si-based fillers whose presence has been confirmed by another ongoing characterization study. Nevertheless, the trend -CF₃/Si-O ratio (Table 7) is ambiguous and difficult to interpret as it increases when the coating is applied on Alk paint but decreases when the coating is applied on Acr and Sty paints.

In summary, chemical analyses show that rain ageing induces limited degradation in uncoated Alk and Acr paints, while Sty is more affected. In all cases, the presence of a primer reduces the extent of chemical changes. All coatings exhibit some molecular degradation, especially in acrylic and silane-based products. SF3 apparently shows the highest chemical stability after ageing. Although none of the coatings demonstrate full resistance to rain exposure, their presence may contribute to limiting the chemical alteration of the underlying paint layers. The chemical changes observed in the coatings are consistent with known degradation pathways triggered by hydrolysis and prolonged water exposure. These alterations were not detected in the unaged samples and appeared only after the rain ageing protocol, strongly suggesting a direct correlation with the simulated rain conditions. While external artifacts cannot be entirely ruled out, this supports a direct correlation between the observed chemical changes and the simulated rain exposure.

4. Final Remarks

This study assessed the durability of commercial protective coatings for contemporary muralism under simulated rain exposure, using a custom-built rain chamber replicating 10 years of rainfall in Central-Southern Europe.

Uncoated paints showed variable degradation, with alkyd paints being the most affected due to blistering and saponification. The use of a primer improved durability and reduced damage.

Selected protective coatings provided only partial and inconsistent protection. While some formulations (notably silane/siloxane-based and fluorinated coatings) reduced water absorption and maintained moderate water repellency, their performance declined after ageing. Acrylic-based coatings showed significant degradation, silane-based coatings lost hydrophobic Si–CH₃ groups, and SF3, despite its initial stability, exhibited ambiguous chemical changes depending on the paints on which it was applied. Overall, none of the tested products ensured long-term protection under rain exposure.

Rain ageing altered the coatings by reducing their water-repellent properties (WCA values) and increasing capillary water absorption. The coatings acted as sacrificial layers, degrading under environmental stress while offering temporary protection to the underlying paint. The primer layer contributed positively by mitigating water uptake and reducing mechanical damage but did not prevent coating degradation.

In conclusion, this study highlights the limited long-term efficacy of a selected group of commercial coatings in preserving contemporary mural paintings from rain exposure. Results demonstrated that even rain constituted of demineralized water can significantly degrade these commercial protective coatings, a finding not previously reported in the literature. It should also be noted that rain exposure alone does not account for the full complexity of outdoor weathering, and further studies are needed to assess the synergistic effects of UV radiation, temperature fluctuations, and pollutants. Even within the accelerated rain ageing, additional factors such as pH, water impact energy, droplet size, and more frequent wet-dry cycles may also play a role in the degradation mechanisms and should be considered in future experimental designs. Protective action of selected coatings revealed to be short-lasting and highly dependent on formulation and compatibility with the paint. Despite their limitations, application of protective coatings represents the only viable strategy to extend the durability of these artworks, acting as sacrificial layers that preserve the underlying paint. The custom rain chamber proved effective for realistic and reproducible ageing tests. Future research should focus on real outdoor conditions and multi factorial testing of coatings for developing more durable and chemically stable protective materials tailored to the specific needs of outdoor mural conservation. Possibly, protective materials should prioritize enhanced hydrophobicity, long-term resistance to environmental stressors, and ease of reapplication. In particular, the ability to maintain water repellency after prolonged rain exposure, resist UV-induced degradation, and provide a protective barrier without compromising the aesthetic qualities of the artwork are essential.