In-Situ Comparative Study of Eucalyptus, Basil, Cloves, Thyme, Pine Tree, and Tea Tree Essential Oil Biocide Efficacy

Bio-colonization is a dynamic and multiphasic process headed by microorganisms. Conventional treatments to process affected stone materials include chemical biocides, whose formulations are mainly composed of quaternary ammonium salts(QAs), reported to be toxic for human health, dangerous for the environment, and not biodegradable. Accordingly, novel green and eco-friendly products are a promising alternative to treat stone materials deteriorated by microorganism colonization. In this study, the efficacy of pure essential oils (EOs) and a mix of EOs was assessed in situ and compared to a conventional biocide based on QAs, and two commercially green products based on EOs, which were taken as references, through application on a mosaic located at the Archaeological Park of Ostia Antica (Rome). The EO biocide efficacy was analyzed by ultraviolet induced luminescence, spectro-colorimetry and bio-luminometry analyses while the possibility of their permanence on simulated substrate was studied by FTIR spectroscopy. It was observed by FTIR analysis, that EOs considered volatile can leave a residue after the application; typical fingerprint bands at about 2926, 1510, and 1455 cm−1 were recorded in the EO spectra. Every tested oil was confirmed to have a biocide action although minimal in relation to the most conventional products based on QAs. The synergy of the essential oils revealed positive results, showing a stronger biocide efficacy. Further investigation should be carried out to develop the method of application and study of essential oils on cultural heritage.


Introduction
Bio-colonization is one of the main problems affecting the cultural heritage sector [1]. This phenomenon takes place on structures exposed to an environment characterized by specific local conditions, such as high moisture, high salinity, and abundance of organic nutrients [2,3]. According to the colonizing organisms and the characteristics of the interested area (surfaces, cracks, or pores), the entity of the damage changes, thus varying from exterior damage to irreversible disintegration of the inner substrate [4]. and α-terpineol) express a slightly less effective action against microorganisms, showing antiseptic, cytotoxic, antifungal, antiviral and anti-inflammatory action and mainly used in the pharmaceutical and cosmetic industry [16,22]. Geranium oil (geraniol and citronellol), peppermint oil (menthol), and lavender oil (linalool and linalooll acetate) are members of this group. Rakotonirainy et al. [35], demonstrated the high antifungal activity of linalool and linalool acetate, present in Lavandula angustifolia against fungal strains isolated from library and archive storage areas. Ketones are the main constituents of sage oil (thujone, camphor) and peppermint oils (menthone, carvone), while fennel, eucalyptus and rosemary oils are rich in ethers: an ethole and 1,8-cineole, respectively. Rosemary extract, moreover, contains carnosic acid, as carnosol and rosmarinic which present antimicrobial, antioxidant, and anti-inflammatory properties.
However, reports about the application of EOs in cultural heritage conservation, as a valid alternative to traditional biocides, are still very limited to in vitro applications and need an application "on-field" to test the effectiveness, the proper way of application, and a monitoring strategy to define the efficiency over time. In vitro analysis is limited as biofilms are known to have different biological properties on stone surfaces than in suspension [10]. In addition, other factors such as the surface of application, the environment, and the weather could influence the effectiveness of essential oils in the field.
The aim of this study was to test a combination of treatments, especially focusing on evaluating the effectiveness of selected EOs according to their biocidal activity as defined by the literature, through the application in a real case study [1,[38][39][40].
Pure EOs and mix of EOs were compared to a conventional product based on QACs and two commercially green products based on EOs, which were taken as references. The monitoring of their effectiveness was assessed when applied on a white and black mosaic in the XIX room of "Insula delle Muse" in the Archaeological Park of Ostia Antica (Rome).

Materials and Methods
Six EOs and their mixture were used as biocides and compared to three commercial products in the experimentation as reported in Table 1: Preventol ® RI50, based on QACs, and two commercially green products, Biotersus and Essenzio. Pure commercial EOs such as eucalyptus, basil, cloves, thyme, pine tree, and tea tree were tested and applied in June 2021 on the biodeteriorated mosaic surface by brushing on their prepared solutions. All the species were diluted with distilled water at a final concentration of 0.4% v/v as defined in the previously exhibited scientific literature [41]. The reported concentration used, by Amri and colleagues, was 0.4% v/v for Pinus pinea [39], while lower concentrations have been reported in literature for the other EOs. The oils were applied with the addition of 8.5 g of an agar thickener (NEVEK ® by CTS) to limit the evaporation of oil in the outdoor atmospheric conditions. To improve the homogeneity of water and oil dispersions, Tween ® 20 detergent was added at 0.3% v/v thus generating an emulsion. An innovative green product (BioTersus-Exentiae), a mixture of EOs (Cinnamonum zeylanicum (0.25% v/v), Eugenia caryophyllata (0.5% v/v), Corydo thymus capitatus (0.4% v/v), and tensioactive Tween20 (0.3% v/v) was applied diluted at 1.3% with distilled water, by brushing it on the surface, as provided by the manufacturer [42].
A new product (YOCOCOIL) was tested, formulated and proposed by YOCOCU APS, a mix based on the combination of the most promising oils in the literature, Eucalyptus globulus For the formulation of this product, the commercial product Biotersus, also composed of EOs, was taken as reference. The mix was applied adding to the pure essential oils 8.5 g of NEVEK ® and Tween ® 20 at 0.3% v/v and diluted with distilled water.
As a reference for the current treatments, a traditional biocide was used, Preventol ® RI50, containing quaternary ammonium salts in diluted concentration 3% v/v in distilled water. The product was applied with a soft brush.
Finally, the experimentation considered a natural product Essenzio, an extract of Thymus vulgaris and Origanum vulgare that was employed by spraying the pure product on the surface, as reported in the data sheet [43].
The products were applied on a white and black mosaic in the XIX room of "Insula delle Muse", in the Archaeological Park of Ostia Antica (Rome). The biofilm layer on the mosaic (Figure 1), observed under an optical microscope, produced by cyanobacteria, chlorophyta (photosynthetic prokaryotes), and green algae (Chlorella) represented pioneer phototrophic microorganisms on this kind of architecture [44]. A new product (YOCOCOIL) was tested, formulated and proposed by YOCOCU APS, a mix based on the combination of the most promising oils in the literature, Eucalyptus globulus For the formulation of this product, the commercial product Biotersus, also composed of EOs, was taken as reference. The mix was applied adding to the pure essential oils 8.5 g of NEVEK ® and Tween ® 20 at 0.3% v/v and diluted with distilled water.
As a reference for the current treatments, a traditional biocide was used, Preventol ® RI50, containing quaternary ammonium salts in diluted concentration 3% v/v in distilled water. The product was applied with a soft brush.
Finally, the experimentation considered a natural product Essenzio, an extract of Thymus vulgaris and Origanum vulgare that was employed by spraying the pure product on the surface, as reported in the data sheet [43].
The products were applied on a white and black mosaic in the XIX room of "Insula delle Muse", in the Archaeological Park of Ostia Antica (Rome). The biofilm layer on the mosaic (Figure 1), observed under an optical microscope, produced by cyanobacteria, chlorophyta (photosynthetic prokaryotes), and green algae (Chlorella) represented pioneer phototrophic microorganisms on this kind of architecture [44].  Two portions of the mosaic were selected and delimited, respectively in Area 1 and Area 2, characterized by a different level of colonization as shown in Figure 2. Area 1 presented by macroscopic and optical microscope observations, a minor entity of biodeterioration compared to Area 2, subject to a constant flux of water coming from the descending roof ( Figure 2). Each of those two areas was subdivided into other ten zones respectively called P, A, B, C, D, E, F, G, H, and I, where the different EOs or products were applied, as reported in Table 1.
After the product application, to enable the EOs to operate on the biodeteriogens, the areas were covered with transparent film for three days, as the EOs are considered highly volatile and alsoto prevent their oxidation and to limit environmental influences on the tested area. To be effective and kill the organisms involved, the tested agents need to penetrate into the target cells of the complex biofilms composed of different phototrophic and heterotrophic microbes (such as fungi, algae, cyanobacteria, and bacteria) and extracellular polymeric substances (EPS) [1,10]. After the above-mentioned time the protective film was removed. A week later from the application of the biocide, the two target portions were slightly brushed, then the biocidal treatment was repeated to strengthen the effect. Finally, after the second biocide treatment, the areas were cleaned mechanically with soft brushes and washed with distilled water to remove totally the biological patina. Several analytical investigations, such as ultraviolet induced luminescence (UVL), spectro colorimetric analysis, and adenosine triphosphate (ATP) test, were used to define the surface state before the treatment and after the second biocide application.
The effectiveness of the treatments was investigated by measuring the chlorophyll fluorescence, the measuring of the presence of ATP nucleotide, and through the definition of the colorimetric parameters of the mosaic surface.
Each experiment was repeated on average 3 times. Data are presented as means ± SD. Statistical evaluation was conducted by Student's t-test or a one-way ANOVA, followed by Tukey's multiple comparison test (homogeneous variances). Statistical significance was at p < 0.05. Area 1 presented by macroscopic and optical microscope observations, a minor entity of biodeterioration compared to Area 2, subject to a constant flux of water coming from the descending roof ( Figure 2). Each of those two areas was subdivided into other ten zones respectively called P, A, B, C, D, E, F, G, H, and I, where the different EOs or products were applied, as reported in Table 1.
After the product application, to enable the EOs to operate on the biodeteriogens, the areas were covered with transparent film for three days, as the EOs are considered highly volatile and alsoto prevent their oxidation and to limit environmental influences on the tested area. To be effective and kill the organisms involved, the tested agents need to penetrate into the target cells of the complex biofilms composed of different phototrophic and heterotrophic microbes (such as fungi, algae, cyanobacteria, and bacteria) and extracellular polymeric substances (EPS) [1,10]. After the above-mentioned time the protective film was removed. A week later from the application of the biocide, the two target portions were slightly brushed, then the biocidal treatment was repeated to strengthen the effect. Finally, after the second biocide treatment, the areas were cleaned mechanically with soft brushes and washed with distilled water to remove totally the biological patina. Several analytical investigations, such as ultraviolet induced luminescence (UVL), spectro colorimetric analysis, and adenosine triphosphate (ATP) test, were used to define the surface state before the treatment and after the second biocide application.
The effectiveness of the treatments was investigated by measuring the chlorophyll fluorescence, the measuring of the presence of ATP nucleotide, and through the definition of the colorimetric parameters of the mosaic surface.
Each experiment was repeated on average 3 times. Data are presented as means ± SD. Statistical evaluation was conducted by Student's t-test or a one-way ANOVA, followed by Tukey's multiple comparison test (homogeneous variances). Statistical significance was at p < 0.05.

FTIR Analysis
As a preliminary test, Fourier transform infrared spectroscopy (FTIR) was adopted to analyze the EO residues after drying them with a heater at 40 • C for 32 h to accelerate evaporation. The time of drying was selected on the basis of the weather conditions related to the real application on the mosaic (end of June in Rome). The EOs were applied on a glass specimen slide and analyzed after evaporation to define the possibility of permanence on the substrate. The aim of this activity was to define the presence of residue due to oil application and to limit secondary chimism or chromatic alterations in the real mosaic [1,45]. FTIR spectroscopy was applied in attenuated reflectance modality (ATR). The IR spectra were collected employing the spectrometer Nicolet Summit FTIR equipped with the accessory Everest Diamond ATR, which allows the analysis in total attenuated reflectance (ATR). This technique allows the direct examination of the liquid and solid samples without preparation of the materials analyzed. The resolution of the instrument is 8 cm −1 , the scans performed by the instrument on each sample were 32.

Ultraviolet Induced Luminescence (UVL)
Several pictures of the flooring were taken with visible light in order to evaluate the status of the mosaic, the extension of the biological patina, and the effectiveness of the biocides, before and after the treatments. Ultraviolet fluorescence induced by two Madatec spotlights at wavelengths of 365 nm (UV), was used to capture visible light-emitted by chlorophyll microorganisms still present on the mosaic surface after the cleaning procedures. Pictures of each area were taken before and after the biocide treatments using a Madatec multispectral system with the support of a Samsung NX50028.2 MPBSICMOS camera. To observe the fluorescence induced in the visible region by ultraviolet radiation, two filters were applied: HOYA UV-IR filter cut 52 and Yellow49552mm F-PROMRC022.

Spectro-Colorimetric Analysis
Spectro-colorimetric measurements, for each delimited area, were taken before and after the treatments in order to evaluate the chromatic variation of the parameters L*, a* and b* and to understand the surface color distance between the treated surface and untreated surface (with biological patina).
Spectro-colorimetric analysis (400-700 nm) was carried out by a portable spectrophotometer Y3060 (by 3nh), D65 illuminant, measurement aperture 8 mm, component SCI included. Data were analyzed using the CIELab color system.
The chromatic parameter a* corresponds to the change in redness-greenness of the surface analyzed, while b* is associated with the variation in yellowness-blueness and L represents the lightness-darkness. For each zone ∆L*, ∆a*, ∆b* and ∆E* were calculated taking into account the parameters L*, a*, and b* before the treatment with EOs and after the biocide procedure using the following formulas: where L* 1 , a* 1 , and b* 1 correspond to the measures before the application of the biocide and L* 2 , a* 2 , and b* 2 after the second treatment. If ∆a* results positive there is a variation of the surface analyzed versus red, vice-versa versus green; positive values of ∆b* indicate yellowing, instead of negative blueing. An increase of ∆L value corresponds to a brighter area vice-versa to a darker facade. A recolonization process can be observed when the values of ∆L* and ∆a* decrease and ∆b* increase indicating the surface analyzed is becoming darker and more greenyellowish due to the presence of phototrophic microorganisms [46]. Positive values of ∆L* and ∆a* with a decrease of ∆b*represent a decolonization process and therefore the successful biocidal action of the treatments. The total color change ∆E* gives information about the distance between two different points in the colorimetric space, for this research the color surface before and after the treatments. It indicates the total difference of the treated surface and how strong was the biocide action for each product tested.

Adenosine Triphosphate (ATP) Test
An average of three bio-luminometric measurements of adenosine triphosphate (ATP) were taken dabbing a 10 cm 2 surface using a portable bio-luminometer, Lumitester PD-30 (Kikkoman) and LuciPacPen AQUA, for each area (1 and 2). An external area was analyzed and taken as a reference of anuncontaminated zone. The ATP nucleotide (adenosinetriphosphate) is an energetic molecule contained in every cellular type. The instrument exploits the chemiluminescence reaction of the reagent luciferase, to emit light, in contact with ATP. The count of ATP detected from the surface, expressed in relative light units (RLU), reveals the metabolic activity of microorganisms or organic residues still present after the cleaning procedures.

Analysis of Residue Using FTIR Analysis
A 10 µL of each EO was applied on a glass slide for the analysis of FTIR in ATR spectroscopy, before and after being degraded for 32 h at about 40 • C in a heater [1].
The collected spectra reported a high variability of peaks due to multi-molecules that usually characterize the oils as terpenoids and derived hydrocarbons. In all the spectra EO residue was observed on the glass surface. Table 2 reports the main peaks identified in the spectra of the tested products in situ based on essential oils with the presence of the thickener, both before and after the drying process, and moreover the pure essential oils without the thickener analyzed after the drying process. The peaks detected at around 926, 911, and 763 cm −1 are related to the glass specimen slide. Peaks at around 3278, 1636, 1373, 1150, 1069, 1043, 966, 932, 713, and 687 cm −1 instead correspond to the thickener NEVEK. Peaks measured below 3064 (cis C=CH stretching), 2923 (CH2 asymmetric bending), 1744 (carbonyl stretching), 1635-1650 (RHC=CH2), 1511, 1450 and 1380 (CH2 bending), 1117 and 1097 cm −1 (C-O ether bending), 920-850 (isopropyl group), 810 (CH bending) can be considered the oil fingerprints [22,45]. To understand completely the obtained results (residues) of the tested reagents, an FTIR analysis was carried out using directly pure EOs, without the thickener. Additionally, in this case the spectra showed the presence of residues on the surface. The stability of EOs depends on several factors such as light, temperature, and oxygen. Around the temperature the stability decreases with a prolonged exposure to a value higher than 38 • C while degradation of terpenoids could be induced by temperatures above 100 • C [47].  Table 2. FTIR-ATR main peaks related to the tested reagents: as pure compound without thickener (after the evaporation process), as applied products with thickener (before and after drying). The collected peaks related to the thickener NEVEK and to the glass specimen were removed from each measure present in the table. In Figure 3 it is possible to observe thespectra difference between the residues detected by the solution of essential oils applied in situ and the pure EOs. The main bands of the tested products are related to the glass specimen slide and thickener, however a few bands with low intensity can be linked to some of the residues of the oils. This is detected especially in the spectra of thyme that shows bands at 2959, 2926, and 1457 cm −1 , pine tree at 2929, 1621, 1446, 1379, and 936, and basil at 1511 and 1455 cm −1 (Figure 3). Methods Protoc. 2022, 5, x FOR PEER REVIEW 9 of 16

Ultraviolet Induced Luminescence (UVL)
In visible reflectance imaging (Table 3), photographs were taken before, after the treatment, and after the cleaning procedure. The zones that looked more colonized by the green patina were in the order F, G, E, B, D, C, A, P, H, and I for Area 1. As for Area 2 the zones more infested were in the order H, D, E, C, A, B, F, I, P, and G. After the treatments the intensity of color biofilm decreased especially, for Area 1 in zone P followed by I, H, G, A, F, E, B, C, and D. Meanwhile, for Area 2, it decreased in this order P, H, I, G, F, E, A, B, and C. The best biocide efficacy both in Area 1 and Area 2 after the cleaning procedure was obtained by YOCOCOIL (applied in zone E) and Preventol ® RI50 (zone P) followed by the products Biotersus and thyme (zone F and G) for Area 1 and thyme and pine tree (zone G and H) for Area 2. A lower biocide efficacy was observed for Area 1 in comparison with Area 2. Macroscopically, the patina changed color from green to grey with the biocide treatments. After the cleaning process, the patina on the surface was totally removed.

Ultraviolet Induced Luminescence (UVL)
In visible reflectance imaging (Table 3), photographs were taken before, after the treatment, and after the cleaning procedure. The zones that looked more colonized by the green patina were in the order F, G, E, B, D, C, A, P, H, and I for Area 1. As for Area 2 the zones more infested were in the order H, D, E, C, A, B, F, I, P, and G. After the treatments the intensity of color biofilm decreased especially, for Area 1 in zone P followed by I, H, G, A, F, E, B, C, and D. Meanwhile, for Area 2, it decreased in this order P, H, I, G, F, E, A, B, and C. The best biocide efficacy both in Area 1 and Area 2 after the cleaning procedure was obtained by YOCOCOIL (applied in zone E) and Preventol ® RI50 (zone P) followed by the products Biotersus and thyme (zone F and G) for Area 1 and thyme and pine tree (zone G and H) for Area 2. A lower biocide efficacy was observed for Area 1 in comparison with Area 2. Macroscopically, the patina changed color from green to grey with the biocide treatments. After the cleaning process, the patina on the surface was totally removed. Table 3. Images of Areas 1 and 2 before, after the treatment with Eos, and after the cleaning procedure.

Area 1 Area 2
Before the treatment Table 3. Images of Areas 1 and 2 before, after the treatment with Eos, and after the cleaning procedure.

Area 1 Area 2
Before the treatment After the biocide treatment After the chemical and mechanical cleaning In Area 1, it was possible to distinguish a strong remitted light induced by chlorophyll molecules present in the microorganisms, particularly evident in zone F (Table 4). UV light confirmed the zones that seemed to have been more attacked by biodeteriogens in visible light, with a particular strong fluorescence being observed in the F, E, and G zones of Area 1. This fluorescence disappeared after the biocide application, as a result of the devitalization of the microorganisms. Such an effect can be observed especially in the E, C, and H zones of Area 1 where the product YOCOCOIL, Essenzio, and pine oil were applied, followed by a medium efficacy reported for the zones G, F, B, D, and P and a low action on I, B, and A. Table 4. Images of areas 1 and 2 before the treatment with EOs and after the cleaning procedure in UVL.

Before the Biocide Treatment
After the Biocide Treatment

Area 1
Area 2 In Area 2, the remitted light, before the treatment, was very reduced. After the biocide treatment the zones, except for the zone G, P, and E, were characterized by almost Table 3. Images of Areas 1 and 2 before, after the treatment with Eos, and after the cleaning procedure.

Area 1 Area 2
Before the treatment After the biocide treatment After the chemical and mechanical cleaning In Area 1, it was possible to distinguish a strong remitted light induced by chlorophyll molecules present in the microorganisms, particularly evident in zone F (Table 4). UV light confirmed the zones that seemed to have been more attacked by biodeteriogens in visible light, with a particular strong fluorescence being observed in the F, E, and G zones of Area 1. This fluorescence disappeared after the biocide application, as a result of the devitalization of the microorganisms. Such an effect can be observed especially in the E, C, and H zones of Area 1 where the product YOCOCOIL, Essenzio, and pine oil were applied, followed by a medium efficacy reported for the zones G, F, B, D, and P and a low action on I, B, and A. Table 4. Images of areas 1 and 2 before the treatment with EOs and after the cleaning procedure in UVL.

Before the Biocide Treatment
After the Biocide Treatment

Area 1
Area 2 In Area 2, the remitted light, before the treatment, was very reduced. After the biocide treatment the zones, except for the zone G, P, and E, were characterized by almost After the biocide treatment

Area 1 Area 2
Before the treatment After the biocide treatment After the chemical and mechanical cleaning In Area 1, it was possible to distinguish a strong remitted light induced by chlorophyll molecules present in the microorganisms, particularly evident in zone F (Table 4). UV light confirmed the zones that seemed to have been more attacked by biodeteriogens in visible light, with a particular strong fluorescence being observed in the F, E, and G zones of Area 1. This fluorescence disappeared after the biocide application, as a result of the devitalization of the microorganisms. Such an effect can be observed especially in the E, C, and H zones of Area 1 where the product YOCOCOIL, Essenzio, and pine oil were applied, followed by a medium efficacy reported for the zones G, F, B, D, and P and a low action on I, B, and A. Table 4. Images of areas 1 and 2 before the treatment with EOs and after the cleaning procedure in UVL.

Before the Biocide Treatment
After the Biocide Treatment

Area 1
Area 2 In Area 2, the remitted light, before the treatment, was very reduced. After the biocide treatment the zones, except for the zone G, P, and E, were characterized by almost

Area 1 Area 2
Before the treatment After the biocide treatment After the chemical and mechanical cleaning In Area 1, it was possible to distinguish a strong remitted light induced by chlorophyll molecules present in the microorganisms, particularly evident in zone F (Table 4). UV light confirmed the zones that seemed to have been more attacked by biodeteriogens in visible light, with a particular strong fluorescence being observed in the F, E, and G zones of Area 1. This fluorescence disappeared after the biocide application, as a result of the devitalization of the microorganisms. Such an effect can be observed especially in the E, C, and H zones of Area 1 where the product YOCOCOIL, Essenzio, and pine oil were applied, followed by a medium efficacy reported for the zones G, F, B, D, and P and a low action on I, B, and A. Table 4. Images of areas 1 and 2 before the treatment with EOs and after the cleaning procedure in UVL.

Before the Biocide Treatment
After the Biocide Treatment

Area 1
Area 2 In Area 2, the remitted light, before the treatment, was very reduced. After the biocide treatment the zones, except for the zone G, P, and E, were characterized by almost After the chemical and mechanical cleaning

Area 1 Area 2
Before the treatment After the biocide treatment After the chemical and mechanical cleaning In Area 1, it was possible to distinguish a strong remitted light induced by chlorophyll molecules present in the microorganisms, particularly evident in zone F (Table 4). UV light confirmed the zones that seemed to have been more attacked by biodeteriogens in visible light, with a particular strong fluorescence being observed in the F, E, and G zones of Area 1. This fluorescence disappeared after the biocide application, as a result of the devitalization of the microorganisms. Such an effect can be observed especially in the E, C, and H zones of Area 1 where the product YOCOCOIL, Essenzio, and pine oil were applied, followed by a medium efficacy reported for the zones G, F, B, D, and P and a low action on I, B, and A. Table 4. Images of areas 1 and 2 before the treatment with EOs and after the cleaning procedure in UVL.

Before the Biocide Treatment
After the Biocide Treatment

Area 1
Area 2 In Area 2, the remitted light, before the treatment, was very reduced. After the biocide treatment the zones, except for the zone G, P, and E, were characterized by almost

Area 1 Area 2
Before the treatment After the biocide treatment After the chemical and mechanical cleaning In Area 1, it was possible to distinguish a strong remitted light induced by chlorophyll molecules present in the microorganisms, particularly evident in zone F (Table 4). UV light confirmed the zones that seemed to have been more attacked by biodeteriogens in visible light, with a particular strong fluorescence being observed in the F, E, and G zones of Area 1. This fluorescence disappeared after the biocide application, as a result of the devitalization of the microorganisms. Such an effect can be observed especially in the E, C, and H zones of Area 1 where the product YOCOCOIL, Essenzio, and pine oil were applied, followed by a medium efficacy reported for the zones G, F, B, D, and P and a low action on I, B, and A. Table 4. Images of areas 1 and 2 before the treatment with EOs and after the cleaning procedure in UVL.

Before the Biocide Treatment
After the Biocide Treatment

Area 1
Area 2 In Area 2, the remitted light, before the treatment, was very reduced. After the biocide treatment the zones, except for the zone G, P, and E, were characterized by almost In Area 1, it was possible to distinguish a strong remitted light induced by chlorophyll molecules present in the microorganisms, particularly evident in zone F (Table 4). UV light confirmed the zones that seemed to have been more attacked by biodeteriogens in visible light, with a particular strong fluorescence being observed in the F, E, and G zones of Area 1. This fluorescence disappeared after the biocide application, as a result of the devitalization of the microorganisms. Such an effect can be observed especially in the E, C, and H zones of Area 1 where the product YOCOCOIL, Essenzio, and pine oil were applied, followed by a medium efficacy reported for the zones G, F, B, D, and P and a low action on I, B, and A. Table 4. Images of areas 1 and 2 before the treatment with EOs and after the cleaning procedure in UVL.

Before the Biocide Treatment
After the Biocide Treatment

Area 1
Before the treatment After the biocide treatment After the chemical and mechanical cleaning In Area 1, it was possible to distinguish a strong remitted light induced by chlorophyll molecules present in the microorganisms, particularly evident in zone F (Table 4). UV light confirmed the zones that seemed to have been more attacked by biodeteriogens in visible light, with a particular strong fluorescence being observed in the F, E, and G zones of Area 1. This fluorescence disappeared after the biocide application, as a result of the devitalization of the microorganisms. Such an effect can be observed especially in the E, C, and H zones of Area 1 where the product YOCOCOIL, Essenzio, and pine oil were applied, followed by a medium efficacy reported for the zones G, F, B, D, and P and a low action on I, B, and A. Table 4. Images of areas 1 and 2 before the treatment with EOs and after the cleaning procedure in UVL.

Before the Biocide Treatment
After the Biocide Treatment

Area 1
Area 2 In Area 2, the remitted light, before the treatment, was very reduced. After the biocide treatment the zones, except for the zone G, P, and E, were characterized by almost Before the treatment After the biocide treatment After the chemical and mechanical cleaning In Area 1, it was possible to distinguish a strong remitted light induced by chlorophyll molecules present in the microorganisms, particularly evident in zone F (Table 4). UV light confirmed the zones that seemed to have been more attacked by biodeteriogens in visible light, with a particular strong fluorescence being observed in the F, E, and G zones of Area 1. This fluorescence disappeared after the biocide application, as a result of the devitalization of the microorganisms. Such an effect can be observed especially in the E, C, and H zones of Area 1 where the product YOCOCOIL, Essenzio, and pine oil were applied, followed by a medium efficacy reported for the zones G, F, B, D, and P and a low action on I, B, and A. Table 4. Images of areas 1 and 2 before the treatment with EOs and after the cleaning procedure in UVL.

Before the Biocide Treatment
After the Biocide Treatment

Area 1
Area 2 In Area 2, the remitted light, before the treatment, was very reduced. After the biocide treatment the zones, except for the zone G, P, and E, were characterized by almost Area 2 Before the treatment After the biocide treatment After the chemical and mechanical cleaning In Area 1, it was possible to distinguish a strong remitted light induced by chlorophyll molecules present in the microorganisms, particularly evident in zone F (Table 4). UV light confirmed the zones that seemed to have been more attacked by biodeteriogens in visible light, with a particular strong fluorescence being observed in the F, E, and G zones of Area 1. This fluorescence disappeared after the biocide application, as a result of the devitalization of the microorganisms. Such an effect can be observed especially in the E, C, and H zones of Area 1 where the product YOCOCOIL, Essenzio, and pine oil were applied, followed by a medium efficacy reported for the zones G, F, B, D, and P and a low action on I, B, and A. Table 4. Images of areas 1 and 2 before the treatment with EOs and after the cleaning procedure in UVL.

Before the Biocide Treatment
After the Biocide Treatment

Area 1
Area 2 In Area 2, the remitted light, before the treatment, was very reduced. After the biocide treatment the zones, except for the zone G, P, and E, were characterized by almost Before the treatment After the biocide treatment After the chemical and mechanical cleaning In Area 1, it was possible to distinguish a strong remitted light induced by chlorophyll molecules present in the microorganisms, particularly evident in zone F (Table 4). UV light confirmed the zones that seemed to have been more attacked by biodeteriogens in visible light, with a particular strong fluorescence being observed in the F, E, and G zones of Area 1. This fluorescence disappeared after the biocide application, as a result of the devitalization of the microorganisms. Such an effect can be observed especially in the E, C, and H zones of Area 1 where the product YOCOCOIL, Essenzio, and pine oil were applied, followed by a medium efficacy reported for the zones G, F, B, D, and P and a low action on I, B, and A. Table 4. Images of areas 1 and 2 before the treatment with EOs and after the cleaning procedure in UVL.

Before the Biocide Treatment
After the Biocide Treatment

Area 1
Area 2 In Area 2, the remitted light, before the treatment, was very reduced. After the biocide treatment the zones, except for the zone G, P, and E, were characterized by almost In Area 2, the remitted light, before the treatment, was very reduced. After the biocide treatment the zones, except for the zone G, P, and E, were characterized by almost the same red-light intensity. For this Area the strongest biocide action was obtained certainly by Preventol ® RI50 and thyme oil, respectively in the region G and P. A moderate efficacy was achieved in order by the products applied in zone A, B, F, and E. Meanwhile Essenzio, pine tree, tea tree and clove oil showed a low antimicrobial activity. The best results among the oils were then detected by thyme oil for both Area 1 and Area 2 and pine tree for Area 1 as reported in Table 5. Area 1 ** * * *** ** *** ** *** *** * Area 2 *** *** ** * * ** ** *** * ** * = low efficacy, ** = medium efficacy, *** = high efficacy. Figure 4 reports the variation of the colorimetric parameters (L*, a*, b*) and the total color change ∆E* of the treated zones with the selected biocides, in Area 1 and Area 2, calculated as the difference between the surface after the second biocide treatment and the untreated surface.

Spectro Colorimetric Analysis
the same red-light intensity. For this Area the strongest biocide action was obtained certainly by Preventol ® RI50 and thyme oil, respectively in the region G and P. A moderate efficacy was achieved in order by the products applied in zone A, B, F, and E. Meanwhile Essenzio, pine tree, tea tree and clove oil showed a low antimicrobial activity. The best results among the oils were then detected by thyme oil for both Area 1 and Area 2 and pine tree for Area 1 as reported in Table 5. ** * * *** ** *** ** *** *** * Area 2 *** *** ** * * ** ** *** * ** * = low efficacy, ** = medium efficacy, *** = high efficacy. Figure 4 reports the variation of the colorimetric parameters (L*, a*, b*) and the total color change ΔE* of the treated zones with the selected biocides, in Area 1 and Area 2, calculated as the difference between the surface after the second biocide treatment and the untreated surface. The chromatic parameter, which changed the most for all the reagents used in both the areas, is the lightness L*, as the removal of the greenish biological patina enabled re-emergence of the white color of the mosaic tesserae [48]. For Area 1, ΔL* results in being positive almost for every zone. The two negative values reported for A (−9,18) and I (−12,26), treated respectively with eucalyptus and tea tree oil are not taken into account as related to measures recorded on several grey tesserae of the mosaic. The value of Δa* resulted positive for every zone while Δb* gave mainly negative values, except for zone G, treated with thyme oil, where Δb* is positive (3.02). Compared to Area 1, Area 2 is characterized by smaller chromatic variations. ΔL* and Δa* show positive values for every measure, while Δb* results in being negative. For both the areas, the best results are obtained by the application of YOCOCOIL in zone E, followed by Essenzio in zone C of Area 2, as the chromatic parameter L* increases the most, which indicates a lighting trend of the surface according to the cleaning procedure, while Δa* and Δb* remain almost unchanged. These results were confirmed by the highest values of ΔE* reached by YO-COCOIL in zone E both of Area 1 (30,34) and Area 2 (26,14), and therefore the highest The chromatic parameter, which changed the most for all the reagents used in both the areas, is the lightness L*, as the removal of the greenish biological patina enabled re-emergence of the white color of the mosaic tesserae [48]. For Area 1, ∆L* results in being positive almost for every zone. The two negative values reported for A (−9,18) and I (−12,26), treated respectively with eucalyptus and tea tree oil are not taken into account as related to measures recorded on several grey tesserae of the mosaic. The value of ∆a* resulted positive for every zone while ∆b* gave mainly negative values, except for zone G, treated with thyme oil, where ∆b* is positive (3.02). Compared to Area 1, Area 2 is characterized by smaller chromatic variations. ∆L* and ∆a* show positive values for every measure, while ∆b* results in being negative. For both the areas, the best results are obtained by the application of YOCOCOIL in zone E, followed by Essenzio in zone C of Area 2, as the chromatic parameter L* increases the most, which indicates a lighting trend of the surface according to the cleaning procedure, while ∆a* and ∆b* remain almost unchanged. These results were confirmed by the highest values of ∆E* reached by YOCOCOIL in zone E both of Area 1 (30,34) and Area 2 (26,14), and therefore the highest difference of the color surface detected before and after the application of the treatments. A positive trend was observed by the application of Essenzio and Biotersus in Area 2. Figure 5 reports the full spectrum color measurement of the treated zones in comparison with the untreated zone's color (plotted using a black curve). In all the spectra of the treated zones, the peak of chlorophyll absorption (near 670 nm) decreased and disappeared in Areas 1 and 2 only for the treatment with Preventol ® RI50. For the oils, the best results were obtained by I = Tea tree in both the areas and F = Biotersus, G = Thyme for Area 1. The spectra confirmed a higher efficacy of biocide treatments for Area 1 with respect to Area 2. Figure 5 reports the full spectrum color measurement of the treated zones in comparison with the untreated zone's color (plotted using a black curve). In all the spectra of the treated zones, the peak of chlorophyll absorption (near 670 nm) decreased and disappeared in Areas 1 and 2 only for the treatment with Preventol ® RI50. For the oils, the best results were obtained by I = Tea tree in both the areas and F = Biotersus, G = Thyme for Area 1. The spectra confirmed a higher efficacy of biocide treatments for Area 1 with respect to Area 2.  Table 6 reports the bioluminometric measures of adenosine triphosphate (ATP) before (14 June 2021) and after (24 June 2021) the application of the biocides for Area 1 and Area 2 and of a zone not treated. The best treatments seem to be reached by YOCOCOIL and Preventol ® RI50 respectively in zone E and P for Area 1 and Area 2, showing the lowest residue of microorganism metabolic activity, although good results were obtained for thyme, Biotersus, and tea tree oil. Taking into account the major difference in the values of ATP before and after the treatments, this indicates the strongest biocide action for Area 1 was instead reached in zone P where Preventol ® RI50 was applied followed by YOCOCOIL in zone E and Biotersus in zone F. As for Area 2 the best results considering the variation of ATP were achieved by Preventol ® RI50, basil, and Biotersus respectively in zone P, B, and F.   Table 6 reports the bioluminometric measures of adenosine triphosphate (ATP) before (14 June 2021) and after (24 June 2021) the application of the biocides for Area 1 and Area 2 and of a zone not treated. The best treatments seem to be reached by YOCOCOIL and Preventol ® RI50 respectively in zone E and P for Area 1 and Area 2, showing the lowest residue of microorganism metabolic activity, although good results were obtained for thyme, Biotersus, and tea tree oil. Taking into account the major difference in the values of ATP before and after the treatments, this indicates the strongest biocide action for Area 1 was instead reached in zone P where Preventol ® RI50 was applied followed by YOCOCOIL in zone E and Biotersus in zone F. As for Area 2 the best results considering the variation of ATP were achieved by Preventol ® RI50, basil, and Biotersus respectively in zone P, B, and F.  Figure 6 shows the values of ATP measured before and after the application of the biocides in each treated portion of Area 1 and Area 2. The best results with the highest difference and the lowest values of ATP were recorded both for Area 1 and Area 2 in zone P where Preventol ® RI50 was applied followed by zone E treated with YOCOCOIL in Area 1 in line with the previous findings.  Figure 6 shows the values of ATP measured before and after the application of the biocides in each treated portion of Area 1 and Area 2. The best results with the highest difference and the lowest values of ATP were recorded both for Area 1 and Area 2 in zone P where Preventol ® RI50 was applied followed by zone E treated with YOCOCOIL in Area 1 in line with the previous findings.  Table 7 are summarizes all the best results obtained from the different analytical analyses carried out on the treated surface of the mosaic. The best performances were achieved by Preventol ® RI50 in zone P for both Area 1 and Area 2, followed by YOCO-COIL in zone E and Biotersus in zone F with a minor efficacy. Amongst the oils, tea tree oil applied in zone I stands out.

Conclusions
In this study, on the base of scientific literature, the biocidal action of seven different products, mainly EOs, based on eucalyptus, basil, cloves, thyme, pine tree and tea tree oils and their mixture were tested in comparison with the traditional biocide Preventol ® RI50 and two commercial bio-nature products, Essenzio and Biotersus. The experimental research was carried out on a roman biodegraded mosaic located in the XIX room of "Insula delle Muse" at the Archaeological Park of Ostia Antica. The aim was to acquire data on the effectiveness of several EOs and natural products, applied in situ, in relation to one of the most diffuse biocides of which its effective antimicrobial action is more than known. Currently there is in fact a lack of information about their action on real case  Table 7 are summarizes all the best results obtained from the different analytical analyses carried out on the treated surface of the mosaic. The best performances were achieved by Preventol ® RI50 in zone P for both Area 1 and Area 2, followed by YOCOCOIL in zone E and Biotersus in zone F with a minor efficacy. Amongst the oils, tea tree oil applied in zone I stands out.

Conclusions
In this study, on the base of scientific literature, the biocidal action of seven different products, mainly EOs, based on eucalyptus, basil, cloves, thyme, pine tree and tea tree oils and their mixture were tested in comparison with the traditional biocide Preventol ® RI50 and two commercial bio-nature products, Essenzio and Biotersus. The experimental research was carried out on a roman biodegraded mosaic located in the XIX room of "Insula delle Muse" at the Archaeological Park of Ostia Antica. The aim was to acquire data on the effectiveness of several EOs and natural products, applied in situ, in relation to one of the most diffuse biocides of which its effective antimicrobial action is more than known. Currently there is in fact a lack of information about their action on real case studies. The effectiveness of promising natural derived biocides in a real application was analyzed, using the minimum biocide efficacy concentrations defined in the scientific literature and, for the commercial products using the data reported in the technical sheet of the customer.
The mosaic was treated twice with the selected biocides and the patina was then completely removed. To determine the best essential oil treatment several analytical investigations applied directly in situ were used: UVL imaging based on the UV fluorescence of chlorophyll, spectro-colorimetry, and bio-luminometry by adenosine triphosphate test. Finally, to evaluate the possible residues of the treatments, FTIR spectroscopy in ATR was used. The obtained results were encouraging, confirming the efficacy of the EOs which had been mainly observed and studied in vitro and their minor biocide ability in comparison with the conventional biocide Preventol ® RI50. All tested products demonstrated a biocide efficacy however the synergy of different EOs seemed to have a stronger biocidal action compared to the single oils. Among the oils, the best results were obtained by using tea tree, pine tree and thyme. This data was confirmed by the major efficacy of YOCO-COIL with respect to Biotersus and Essenzio. YOCOCOIL is a mixture of Pinuscembra L., Eucalyptus globulus, Thymus vulgaris, Eugenia caryophyllus, and Ocimum basilicum. FTIR analysis showed the possibility of the presence of essential oil residues on the tested substrates, after simulating an ageing process and adding the thickener in the formulation of the biocide products directly using the pure EOs. These results, depending on several factors such as light, temperature, and oxygen, could be a warning for the application of essential oils, especially using a thickener. A deeper investigation focusing on this topic should be carried out with the view of obtaining an understanding of what the presence of EOs might imply.
At the moment, there is no uniform bibliography on how EOs should be applied and studied over time, hindering the understanding of the dynamics and effectiveness. This research intended to give a first insight on how the antimicrobial activity of EOs should be analyzed when applied in situ.
New research should be focused on creating a standard system of EO application procedure on different materials in situ on stonework of cultural heritage, on the analysis and monitoring of their effective action during time, and on their concentration/mixture and of the way of their use.