Su questo sito si utilizzano cookies tecnici e di terze parti. Proseguendo la navigazioni accetti.

GRU>HOME>Gioventù Eleonora - Comparing the bioremoval of black crusts on colored artistic lithotypes of the Cathedral of Florence with chemical and laser treatment*

The external walls of the Cathedral of Florence are made of green serpentine, red marlstone and Carrara white marble, and intensive air pollution attack has led to their weathering, which caused black crust formation. 

A study was performed to evaluate the most appropriate cleaning treatment for black crust removal, adopting chemical (ammonium carbonate poultice), laser (1064 nm, Nd:YAG laser), and microbial (poultice embedding sulfate reducing bacteria) cleaning. The effects of the different procedures on the original surfaces were evaluated by scanning electron microscopy coupled with energy dispersive X-ray (SEM/EDS) spectroscopy, Fourier transform infrared (FTIR) spectroscopy, and color measurements. One year later further color measurements were made. It was found that chemical cleaning led to non-homogeneous crust removal and that for the extremely powdery and incoherent red substratum the preferred treatment was laser cleaning. Overall, the most satisfactory treatment was the microbial cleaning process. It was the most controllable process and the most efficient for sulfate removal. Its main drawback appears to be the time needed to remove thick black crusts since numerous applications were necessary.

Eleonora Gioventù a, Paola Franca Lorenzi a, Federica Villa b, Claudia Sorlini b, Maria Rizzi a, Andrea Cagnini a, Alessandra Griffo a, Francesca Cappitelli b *

a Opificio delle Pietre Dure, Via degli Alfani 78, 50121 Firenze, Italy
b Dipartimento di Scienze e Tecnologie Alimentari e Microbiologiche, Università degli Studi di Milano, Via Celoria 2, 20133 Milano, Italy

International Biodeterioration & Biodegradation 65 (2011) 832-839

One of the main conservation problems of outdoor stone artworks is surface sulphation and the consequent blackening in sheltered areas (Moropoulou et al., 1998). Interaction between a calcareous substrate and the sulfuric acid of air pollution leads to the chemical transformation of insoluble calcium carbonate (CaCO3) to the more soluble calcium sulfate dehydrate or gypsum (CaSO4•2H2O). In rain exposed areas, soluble salts are washed away, whereas in sheltered areas salts like gypsum embed airborne pollutants, such as carbonaceous particles, to form black crusts during crystallization/solubilisation cycles. In addition, this crust passively entraps anthropogenic compounds improving the microbial colonization of stones located in polluted environments (Saiz-Jimenez, 1997).

The cleaning of stone materials, an important procedure aimed at slowing down the decay process, is still a critical conservation step, particularly for historical buildings. Indeed, the cleaning treatment should remove all surface compounds dangerous for the original substratum, but it should, at the same time, be respectful of the chemicalephysical nature of the artwork materials and their historical and artistic value. The cleaning of crusts is essential, not only for the conservation of deteriorated areas but also for pre- venting further erosion phenomena (Kapsalas et al., 2008).

In the traditional conservation approach, the main methods for the removal of black crust are physical and chemical and more recently laser cleaning treatments, and a large body of liter- ature compares these methods (Lanterna and Matteini, 2000; Moropoulou and Kefalonitou, 2003; Bromblet et al., 2003; Gaspar et al., 2003). Chemical treatment generally produces good clean- ing results in a reasonable time, but because of its wide range of action it is not always selective in removing crusts and deposits (Lazzarini and Tabasso, 1986). Laser treatment is a more recent treatment that is spreading because of its high selectivity and its faster application time, though there is still uncertainty concerning its real interaction with the different substances in the crust (Salimbeni et al., 2001).

An alternative cleaning technology employs sulfate-reducing bacteria (Ranalli et al., 1997). Sulfate-reducing bacteria are of great ecological importance in the biodegradation of organic matter in all anaerobic environment rich in sulfates (Gu et al., 2008; Wolicka and Borkowski, 2009; Guerrero-Barajas et al., 2011). Recently, Cappitelli et al. (2006) optimized the system for the bioremoval of black crust from artistic stoneworks by using Desulfovibrio vulgaris subsp. vulgarisATCC 29579. The authors reported the successful application of this cleaning methodology for the removal of black crusts on white marble, comparing it to chemical treatment employing EDTA and ammonium carbonate (Cappitelli et al., 2007), and on porous lime- stone (Polo et al., 2010). The researchers demonstrated that chemical treatment cleaned the marble surface incompletely and non- uniformly. In addition, traces of the yellow-ocher layer were observed. The superficial crust was only partially removed; a sharp separation between cleaned and uncleaned areas was clearly visible. SEM observations of the loose uncleaned fragments revealed the superfi- cial morphology of residual crust areas with many cavities. FTIR analyses were done on residual crust, noncoherent material covering the cleaned surface, and cleaned marble. By contrast, the biocleaned marble surface had a rather homogeneous aspect and a pale yellow color. Visual inspection and microscope observations revealed that most of the crust was removed and there was not a well-defined separation between completely cleaned and incompletely cleaned areas. SEM observations revealed a more homogeneous surface texture with fewer noncoherent crystals. Finally, by using the bio- logical treatment, no noticeable breakdown in cohesion was observed.

The aim of the present work was to compare microbial cleaning with more traditional cleaning methods, and to this end we treated various lithotypes, namely Carrara marble, green serpentine and a red marlstone of the Cathedral of Florence. In the work ofLanterna and Matteini (2000), a comparison was made of mechanical, laser and chemical methods on the same lithotypes of the Cathedral of Florence. Although generally advantageous from an economical point of view, we did not consider the physical cleaning treatment as it can cause the detachment of grains andfissuring, leading to abrasion of stone surfaces (Moropoulou et al., 2002). Furthermore, Lanterna and Matteini (2000) found that this cleaning procedure was the least satisfactory one.

Our study presents comparative results of the three cleaning procedures adopted on three lithotypes of the Cathedral of Florence. We discuss the pros and cons of microbial cleaning, comparing it with the two traditional treatments, chemical and laser cleaning. To the best of our knowledge this is the first time microbial cleaning has been compared with laser cleaning, and that biocleaning has been performed on colored stone.

2. Materials and methods

2.1. Treated areas

The research was carried out on stone materials from the Cathedral of Florence, on a Carrara marble column taken from the Baccio d’Agnolo Balcony under the dome by Brunelleschi (its treatment was done in the laboratory), and an external pilaster (treated in situ) on the left side of the Cathedral and made of three different stone typologies: green serpentine, red marlstone and white Carrara marble. Green serpentine is a metamorphic rock coming from the Appennines, red marlstone a marlstone from Monsummano (Italy), and Carrara white marble is a metamorphic limestone from the Apuan Alps.

Three areas (≈15 x 15 cm2) of each of the surfaces were chosen, each being for one of the three different cleaning methodologies. Thus, there was the comparison of 12 areas: three on the column and nine on the external pilaster. Chemical treatment was performed on the areas 1C (column) and 1G, 1R, 1W (pilaster); laser treatment on the surfaces 2C (column) and 2G, 2R, 2W (pilaster) and microbial treatment on the substrata 3C (column) and 3G, 3R, 3W (pilaster) (Fig. 1a and b). Additional areas were treated with poultice without microorganisms and cleaning chemicals as controls (Fig. 1c).

2.2. Application protocols

Cleaning by the chemical and microbial procedures was performed using a poultice of Carbogel (CTS, Vicenza, Italy). On the column areas, four poultice applications (each of 10 h) for the chemical and microbial cleaning were carried out. Instead, on the external stone pilasters, there were two poultice applications (each of 10 h) for the chemical treatment and three for the microbial. On the poultice-only controls - column, white, green and red lithotypes four poultice applications (each of 10 h) were carried out. Laser cleaning was performed only once due to its working conditions, paying careful attention to keep the surface wet during the irradiation (Gaspar et al., 2003). The stone surfaces to be treated with poultices were moistened and covered with tissue paper before any poultice application began, the tissue paper enabling the easy removal of the poultice on treat-ment completion. After removing the tissue paper, a wet cotton swab was wiped with a soft mechanical action over the treated surfaces to remove any residual material softened by the effect of the treatments.

2.3. Chemical cleaning method

The poultice was prepared by mixing 100 g of ammonium carbonate and 10 mL of Tween 20 (a non-ionic detergent) with 1000 mL of distilled water, and suspending Carbogel (42 g in 1000 mL) in the solution.

2.4. Microbial cleaning method

For the biological treatment, the biomass entrapped in the delivery system was prepared as described by Cappitelli et al. (2006). Briefly, the bacteria employed were D. vulgaris subsp. vulgaris ATCC 29579 maintained in DSMZ 63 medium (Cappitelli et al., 2006). Before using them in the treatment, the cells were grown in DSMZ 63 medium, modified by eliminating any iron source. After centrifugation and prior to mixing with the Carbogel, the cell pellet was suspended in deaerated phosphate buffer supplemented with 0.599 g/L sodium lactate at pH 7.0. All the manipulations described above were done under anaerobic conditions in a glove box.

2.5. Laser cleaning method

The choice of the irradiation parameters was due to previous research, mainly done by the Opificio delle Pietre Dure and the IFAC CNR of Florence (Lanterna and Matteini, 2000; Giamello et al., 2004; Siano et al., 2005). The laser used was “Smart Clean II”(El.En. Spa), a short free running Nd:YAG laser (1064 nm, pulse duration 20e120 ms). The laser spot was 4.5 mm, and the repetition rate (5 Hz) was maintained constant throughout the test, while the range of fluences varied depending on the type of stone treated and the thickness of the crusts: from 5 J cm2 to 8.8 J cm2 for the white marble; from 5.7 J cm2 to 8.8 J cm2 for green serpentine; from 6.9 J cm2 to 7.5 J cmfor the red marlstone. The ablation test was performed in water assisted conditions (Bromblet et al., 2003).

2.6. Electron microscope observations

Microscope samples were collected before and after the three different surface cleaning procedures with a scalpel and embedded in a polyester resin, for analysis as cross sections. Scanning Electron Microscopy (SEM) was carried out by a Stereo scan 440® LEICA CAMBRIDGE microscope equipped with an elec-tron retro-diffuse detector (QBSD).The analysis of elements was performed with energy dispersive spectroscopy (EDS), both spot-mode analysis and obtaining element maps using INCA® system by Oxford.

2.7. Fourier transform infrared analyses

Samples for Fourier transform infrared (FTIR) analyses were collected before and after the three different surface cleaning procedures with a superficial scraping. FTIR spectroscopy analyses were carried out by a Nicolet Nexus spectrophotometer (software OMNICTM); spectra were recorded by KBr micropellets (Ø 1.5 mm) in transmission mode between 4000 and 400 cm-1, and 128 scans were collected for each spectrum. To avoid contamination, the sampled different layers were carefully collected under a STEMI 2000 C stereoscope (ZEISS) by means of a needle-sampler.

2.8. Color measurements

Color measurements were performed for each lithotype (green serpentine, red marlstone and white Carrara marble) and condition (freshly cut surfaces, altered surface, cleaned surface and surface 1 year after treatments). All measurements were made directly on the surface, without any sampling. The color was measured using a spectrocolorimeter Minolta CM 2006d with a sensitivity ranging from 360 nm to 750 nm and with an acqui- sition pitch of 10 nm. The measurements were performed with diffused lighting and observation at 8. The concerned area had a circular shape, 8 mm in diameter. A total of 5 readings were taken over different, randomly selected, zones in order to appreciate the color differences of the stone veins. Color measurements were analyzed considering the CIELAB color system (CIE Publication 15-2), which is organized with three axes in a spher- ical form: L*, a* and b*. The L* axis is associated with lightness of color, and moves from a top value (100: white) to a bottom one (0: black), whereas the a* and b* axes are associated with the change in redness-greenness (positive a* is red and negative a* is green), and in yellowness-blueness (positive b* is yellow and negative b* is blue); both move in the two axes that form a plane orthogonal to L*, and do not have specific numerical limits. Furthermore, the color parameters most closely related to the psychophysical characteristics of color, i.e. more related to color perception, and which correspond to the angular coordinates of chroma C*ab = [(a^ 2 + b^ 2)^ 1/2 and hue angle hab 1⁄4 [arctan (b*/a*)] were also calculated. C*ab is the relative strength of a color, chroma or saturation of color, and hab refers to the dominant wavelength, starts with 0 and increases counterclockwise (Wyszecki and Stiles, 1982). Color measurement controls were obtained measuring the CIELAB variables of freshly cut surfaces of each lithotype.

The partial color differences, ΔL*, Δa*, Δb*, ΔC*ab, ΔH*ab, and the total color difference,ΔE*ab , were also calculated (cleaned surface vs.altered surface; cleaned surface vs. reference surface; cleaned surface vs. 1 year after the treatment) for each lithotype and treatments (Table 1). ΔL*, Δa*, Δb*, and ΔC*ab represent the difference between both considered values of L*, a*, b* and C*ab respectively, whereas ΔH*ab is given by ΔH*ab = [(ΔE*ab )^ 2 - (ΔL*)^2 - (ΔC*ab)^2]^1/2 (CIE Publication 15:2004). According to the CIELAB system, the total difference of color between two samples is ΔE*ab= [(ΔL*)^ 2 - (Δa*)^ 2 - (Δb*)^ 2]^ 1/2, where ΔE*ab > 5 is perceived by the human eye (Palazzi, 1995; Reis-Menezes et al., 2011).

2.9. Statistical analysis

An explanatory examination of the color data was performed using Principal Component Analysis (PCA), a technique to extract, rationalize and visualize all the useful information from a data set. PCA is a multivariate method used to project the variables in a new space using a new matrix which shows the degree of similarity between the variables. If the information associated with the first 2 axes represents a sufficient percentage of the total variability of the scatter plot, the observations can be represented on a 2-dimensional chart, thus making interpretation much easier. The data obtained from the colorimetric measurements were arranged in a matrix (different treatments and lithotypes as rows, and average values of color parameters as columns) in order to compare the three cleaning methods tested, the data being plotted in two dimensions based on the scores in the first two principal components. Multivariate investigations were conducted with XLSTAT version 2009.4.07 (Addinsoft, NY) software using the Pearson correlation as similarity index. The significance of the PCA model was tested by a cross-validation procedure.

3. Results

3.1. Characterization of the surface alteration

SEM observations of cross sections showed that the stratification of the crusts before the treatments was quite similar on the three different lithotypes of the external pilaster, and it mainly consisted of sulfates, silicates and carbonaceous particles. On some of the column areas (2C and 3C, but not 1C) there was also an additional non-homogeneous and remarkably discontinuous layer between the black crust and the stone, of various thickness. On the 2C and 3C areas we found also gypsum under this discontinuous layer (Fig. 2a and b).

Fourier transform infrared analyses done before each treatment confirmed the presence of gypsum and silicates, and calcium oxalate (both weddellite and whewellite; 1,623, 1,320, and 780 cm1, areas 2C and 3C of the column). Oxalate was only present within the additional non-homogeneous and fragmentary layer. Instead on the 1C area there were only traces of calcium oxalate.

3.2. Visual inspection after the treatments

Visual inspection of the controls showed no difference with untreated surfaces (Fig. 1c). As a consequence, these surfaces were not subject to further analysis.

Visual inspection of the sample surfaces after the treatments showed that on the chemically treated area of the column we had not, despite numerous applications (four applications), reached complete crust removal, while on the pilaster we achieved this result with only two applications. Instead, four microbial applica- tions on the column and three applications on the pilaster had efficiently removed the crusted alteration layer. With regard to the laser, the green and red surfaces were cleaned satisfactorily but not the white marble of the pilaster or the column, where the laser treatment generated a thin yellow layer, easily seen by naked-eye (Fig. 1c).

3.3. Microscope observations after the treatments

After the microbial treatment, microscope observations of the column and pilaster revealed no gypsum residuals (Fig. 2c). The same appeared for the chemically treated areas of the pilaster, whereas on the chemically treated area of the column part of the crust remained. The laser cleaning left no residuals on green serpentine and red marlstone, but on the white Carrara marble, from both column and pilaster, there remained a layer, mainly constituted by gypsum and silicates. 

3.4. Fourier transform infrared analyses after the treatments

FTIR analyses after treatment were carried out only for the column areas. They confirmed the absence of black crust residuals on the microbial treated area, the rare presence of gypsum residuals on the chemically treated area and the presence of gypsum, silicates and calcium oxalate on the laser treated area (Fig. 2d).

3.5. Color measurements

Table 1 summarizes the color parameter changes. In comparison to the altered surfaces, all the cleaned areas had a remarkable increase in total color especially for the red and white stones. As regards the green serpentine, in most cases it was not possible to appreciate any differences as the original color was a dark green. In this comparison, ΔE*ab was largely due to the variation of L*, the only exception being the green stone subjected to all the cleaning methods and the white marble treated with laser, the Δb* being of the same order as ΔL*. Thus the laser cleaning applied to the white marble gave rise to an increase in the yellow color component.

In the case of the laser treated area, comparison between the cleaned surfaces and the reference substrata showed that ΔE*ab was always above 5, whereas color differences could be seen by the naked-eye on the white marble treated with the chemical cleaning and the red stone when the biological treatment was used.

Differences in color were appreciable one year after the treat- ments only in the green serpentine treated with the laser. The effectiveness of the three cleaning methods was investigated by visualizing the position of the samples relative to each other in a 2-dimensional space (2D) depicted in the PCA score plot (Fig. 3). The space defined by PC1 and PC2 explains 87.36% of the variance in the original data.

Pattern recognition revealed significant differences in the data, and indicated the existence of five distinct groups of objects, each called a cluster. Cluster I comprises all the surfaces presenting the black crust before treatment. Clusters II, III and V represent the three different lithotypes showing the marked differences of one from the other. Cluster II displays all treated and reference green serpentine surfaces. Cluster III groups white Carrara marble areas chemically and biologically-treated with the control ones. It is interesting to note that laser-treated white marble surfaces does not fall into Cluster III as they present a clear different position in a 2-dimensional space (2D) of the PCA score plot. Thus, Cluster IV represents the separate group of Carrara marble surfaces cleaned with the laser method. Finally, Cluster V shows all treated and reference red marlstone substrata.

The score plot of all the investigated lithotypes shows that the chemical and biological methodologies were the best cleaning methods for removing black crust. The spatial position of the objects in the PCA score plot highlights the similarity between the reference and treated surfaces, implying that chemical and bio- logical procedures reduced the color difference between the reference and the cleaned areas. The multivariate analysis showed that, in terms of color, the biological method proved the most suitable and the most effective in removing the black crust from white surfaces. In contrast, the presence of the separate cluster V clearly demonstrates that the highest change in color was caused by the laser cleaning method on Carrara marble. Finally, the PCA analysis results indicate that, after one year, all the cleaned surfaces fell into the same cluster group as that of the reference and cleaned lithotypes (see Clusters II, III and V). Interesting is the observation that, after one year, the white surface treated with laser falls into Cluster IV, that differed from the reference materials, as well as from the chemical and biological treated Carrara marble surfaces. 

4. Discussion

According to the results of the diagnostic study showing black crust as the main alteration of artwork on the Cathedral of Florence, alternative cleaning procedures were applied to the white, green and red stone of an external pilaster and the white marble of a column of the cathedral. Analyses of the sampling surfaces before treatment showed that the black crust consisted of different components and layers in various areas of the same monument, therefore it was important to remember that the differences in the results could be due not only to the cleaning methodologies employed but also to the non-homogeneous distribution of the alteration products. This is not surprising as climate, environmental conditions and surface exposure on a microscale level are more important in determining weathering conditions than local air pollution (Leysen et al., 1990; Lanterna and Matteini, 2000). We were interested not only in the final effects of the cleaning methods but also in evaluating the three methods during the treatments. The poultice-only controls were considered to isolate the effects of the poultice from the action of chemicals and microorganisms. After four applications visual inspection showed that these surfaces presented no difference with the untreated ones. As a consequence, the poultice alone was considered not effective in removing the black crust on the investigated stone surfaces. During the chemical cleaning the surface was cleaned unevenly, some areas being cleaned perfectly with the first application while others were left with thick black crust. Moreover, the chemicals removed the crust in a rough way, sometimes detaching fragments. In contrast, bacterial cleaning could be controlled better, probably because the method is more selective in removing specific compounds (e.g. sulfates) from the altered layer (Cappitelli et al., 2007). It is for this reason that the number of applications of this method was so closely related to crust thickness. During the biological application the black crust was thinned down uniformly.

On green serpentine, laser cleaning was performed without the complete ablation of the crust, the objective being to weaken it using lower ranges of fluences and then mechanically removing all the alteration layer residues with a wet cotton swab. The operations on the red stone were more difficult because of its bad state of conservation. Indeed, with aging, this kind of material is inclined to become powdery and extremely incoherent so the laser system seemed better than the poultice system, given its minor interaction with stone material.

With regard to the laser cleaning of the white marble of the pilaster and column, we were able to observe that the use of a lower range of fluences removed the crust but left a thin yellow layer easily seen by the naked-eye (data not shown). Raising the range offluences led to a thinner yellow layer but damage appeared on the stone surface, so it seems that in order not to damage the surface the yellow layer must be left. It is known that Nd:YAG laser on light materials leaves the substratum slightly yellowed (Delivré, 2003). Opinions regarding the yellowing are still open and controversial; some conservators consider the yellowing to be part of the genuine appearance of the object while others consider it to be further damage (Prasad and Siano, 2010).

Despite the variety of methods used for cleaning stone surfaces, one reason to explore new treatments it is to find a method able to remove, with great precision and efficiency, even very thin layers (down to a few micrometers) of materials, thus allowing the operator to stop the cleaning process at any selected level (Sabatini et al., 2000). In our case-study only the microbiological treatment seemed to fulfill this requirement. This is a big advantage of this technique over other methods, however uniformly removing the black crust layer by layer can certainly be a big disadvantage in terms of time, especially when the black crust is very thick.

Electron microscope observations, elemental analysis and infrared spectroscopy of the cleaned surfaces were performed to investigate the removal of crust compounds and the presence of residues. The esthetic impact of these treatments was quantified by color measurement, SEM and FTIR analyses, revealing that both microbial and chemical cleaning managed to remove all the gypsum residuals from the surfaces, except for the area of the column cleaned with chemicals. In contrast, the laser cleaning, particularly on the white marble, left a residual layer mainly composed of gypsum, with some calcium oxalate and impurities (silicon and phosphorus). On 3C areas of the column, microbial cleaning removed the discontinuous layer constituted also by calcium oxalate, removing the gypsum below it. Other studies (Cappitelli et al., 2007) have shown that bacteria are unable to remove calcium oxalate if this is a component of a well-stratified layer, the so-called patina noble. The lower reactivity of the oxalates, as a result of their poorer solubility (with respect to calcite), renders their sulphation slower, so that they partly protect the sound material for a certain time. However, when sulphation badly degrades the oxalate film, it ceases to act as a protective layer. Indeed, at this degradation stage, it is sometimes suggested to remove this heavily-deteriorated layer during the cleaning (Sabatini et al., 2000). Thus the activity of bacteria in removing the sulfates in the oxalate fragmentary layer, and underneath it, is not to be considered in a negative way. Unfortunately, we could not objectively investigate this situation with the chemical cleaning because in the chemically treated area the SEM analyses before the treatment did not reveal this kind of layer.

In terms of color, the multivariate analysis proved that the best performances were obtained with both chemical and biological treatments, whereas laser cleaning was the least satisfactory treatment. This is not surprising as when colorimetric data of laser cleaning are reported the change in the b* parameter in the CIELAB color scale, with respect to reference stone substrates, increases remarkably, up to +10 units (Klein et al., 2001), the same increase was found on the laser cleaned white marble. On comparing the chemical and biological treatments, the biological cleaning gave better results on the white marble while the green substratum benefited better from the chemical methodology.

The results of these tests confirm the efficacy of bacterial cleaning and highlight its potential competition with traditional cleaning methodologies. Although microbiological cleaning appears to be a valid tool for black crust removal, the feasibility of using this treatment needs to be assessed on a case by case basis. However, a conservator should remain aware that when the crust is thick a great number of microbiological applications might be needed, and this can result in a very time-consuming process. In addition, further aspects of the metabolism of Desulfovibrio should be considered. These microorganisms can not oxidize acetate so this compound resulting from the use of lactate as an energy source is for the most part excreted (Madigan et al., 2010). This disadvantage is partially compensated by the use of phosphate buffer. In contrast, a significant benefit of using Desulfovibrio is that calcium ions released by gypsum dissolution react with carbonate from bacterially produced CO2 and form calcite (Cappitelli et al., 2007). This conversion makes the biological treatment not only a cleaning procedure but also a consolidation treatment. Many recent studies have proposed carbonate mineralization as a method to protect artistic carbonate stone (Jimenez-Lopez et al., 2008).

The results also show that laser cleaning is a respectful method but it is not particularly efficient, it did not always completely remove the gypsum from the surface, and it must be remembered that gypsum can easily become the cause of further stone deteri- oration. Also, chemical treatment sometimes left gypsum, even after several applications, and was the least controllable of the cleaning methods. Note that one of the best properties of bacterial cleaning is its high efficiency and selectivity in the removal of gypsum.

For these reasons, it could be very interesting to pair laser and bacterial cleaning, first using laser cleaning with the lower range offluences to quickly thin the black crust, and then applying bacterial cleaning to remove the remaining gypsum layer with bacteria that are more selective and respectful to the surface.


Bromblet, P., Labouré, M., Orial, G., 2003. Diversity of the cleaning procedures including laser for the restoration of carved portals in France over the last 10 years. Journal of Cultural Heritage 4, 17-26.

Cappitelli, F., Toniolo, L., Sansonetti, A., Gulotta, D., Ranalli, G., Zanardini, E., Sorlini, C., 2007. Advantages of using microbial technology over traditional chemical technology in removal of black crust from stone surfaces of historical monuments. Applied and Environmental Microbiology 73, 5671-5675.

Cappitelli, F., Zanardini, E., Ranalli, G., Mello, E., Daffonchio, D., Sorlini, C., 2006. Improved methodology for bioremoval of black crusts on historical stone artworks by use of sulfate-reducing bacteria. Applied and Environmental Microbiology 72, 3733-3737.

CIE Publ. 15:2004, 2004. Colorimetry, third ed. CIE Central Bureau, Vienna.

CIE Publication 15-2, 1986. Colorimetry. CIE Central Bureau, Vienna.

Delivré, J., 2003. Laser cleaning: is there specific laser esthetics? Journal of Cultural Heritage 4, 245s-248s.

Gaspar, P., Hubbard, C., McPhail, D., Cummings, A., 2003. A topographical assessment and comparison of conservation cleaning treatments. Journal of Cultural Heritage 4, 294-302.

Giamello, M., Pinna, D., Porcina, S., Sabatini, G., Siano, S., 2004. Multidisciplinary study and laser clearing tests of marble surfaces of Porta della Mandorla, Florence. In: Proceedings of the 10th international congress on deterioration and conservation of stone, Stockholm, vol. 2. ICOMOS, Sweden, pp. 84-848.

Gu, J.G., Han, B., Duan, S., 2008. Initial transformation step of dicamba by a sulfate- reducing consortium enriched from sediment of the Pearl River of China. International Biodeterioration & Biodegradation 62, 455-459.

Guerrero-Barajas, C., Garibay-Orijel, C., Rosas-Rocha, L.E., 2011. Sulfate reduction and trichloroethylene biodegradation by a marine microbial community from hydrothermal vents sediments. International Biodeterioration & Biodegradation 65, 116-123.

Jimenez-Lopez, C., Jroundi, F., Pascolini, C., Rodriguez-Navarro, C., Piñar-Larrubia, G., Rodriguez-Gallego, M., González-Muñoz, M.T., 2008. Consolidation of quarry calcarenite by calcium carbonate precipitation induced by bacteria activated among the microbiota inhabiting the stone. International Biodeterioration & Biodegradation 62, 352-363.

Kapsalas, P., Maravelaki-Kalaitzaki, P., Zervakis, M., Delegou, E.T., Moropoulou, A., 2008. A morphological fusion algorithm for optical detection and quantification of decay patterns on stone surfaces. Construction and Building Materials 22, 228-238.

Klein, S., Fekrsanati, F., Hildenhagen, J., Dickmann, K., Uphoff, H., Marakis, Y., Zafiropulos, V., 2001. Discoloration of marble during laser cleaning by Nd:YAG laser wavelengths. Applied Surface Science 171, 242-251.

Lanterna, G., Matteini, M., 2000. Laser cleaning of stone artefacts: a substitute or alternative method? Journal of Cultural Heritage 1, 29-35.

Lazzarini, L., Tabasso, M.L., 1986. In: Il restauro della pietra. CEDAM, Padova. Leysen, L.A., Roekens, E.J., Van Grieken, R.E., De Geyter, G., 1990. Characterization of the weathering crust of various historical buildings in Belgium. Science of the Total Environment 90, 117-147.

Madigan, M.T., Martinko, J.M., Stahl, D.A., Clark, D.P., 2010. Brock biology of microorganisms, thirteenth ed. Benjamin Cummings.

Moropoulou, A., Bisbikou, K., Van Grieken, R., Zezza, F., Macri, F., 1998. Origin and growth of weathering crusts on ancient marbles in industrial atmosphere. Atmospheric Environment 32, 967-982.

Moropoulou, A., Kefalonitou, S., 2003. Efficiency and counter effects of cleaning treatment on limestone surfaces - investigation on the Corfu Venetian Fortress. Building and Environment 37, 1181-1191.

Moropoulou, A., Tsiourva, Th., Bisbikou, K., Tsantila, V., Biscontin, G., Longega, G.,Groggia, M., Dalaklis, E., Petritaki, A., 2002. Evaluation of cleaning procedures on the facades of the Bank of Greece historical building in the center of Athens. Building and Environment 37, 753-760.

Palazzi, S., 1995. Colorimetria. La scienza del colore nell’arte e nella tecnica. Nardini Editore, Florence, Italy, pp. 91.

Polo, A., Cappitelli, F., Brusetti, L., Principi, P., Villa, F., Giacomucci, L., Ranalli, G., Sorlini, C., 2010. Feasibility of removing surface deposits on stone using bio- logical and chemical remediation methods. Microbial Ecology 60, 1-14.

Prasad, M.B.S., Siano, S., 2010. Laser yellowing. Pramana 75, 1249-1252.

Ranalli, G., Chiavarini, M., Guidetti, V., Marsala, F., Matteini, M., Zanardini, E., Sorlini, C., 1997. The use of microorganisms for the removal of sulphates on artistic stoneworks. International Biodeterioration & Biodegradation 40, 255-261.

Reis-Menezes, A.A., Gambale, W., Giudice, M.C., Shirakawa, M.A., 2011. Accelerated testing of mold growth on traditional and recycled book paper. International Biodeterioration & Biodegradation 65, 423-428.

Sabatini, G., Giamello, M., Pini, R., Siano, S., Salimbeni, R., 2000. Laser cleaning methodologies for stone facades and monuments: laboratory analyses on lithotypes of Siena architecture. Journal of Cultural Heritage 1, S9-S19.

Saiz-Jimenez, C., 1997. Biodeterioration vs biodegradation: the role of microor- ganisms in the removal of pollutants deposited on historic buildings. International Biodeterioration & Biodegradation 40, 225-232.

Salimbeni, R., Pini, R., Siano, S., 2001. Achievement of optimum laser cleaning in the restoration of artworks: expected improvements by on-line optical diagnostics. Spectrochimica Acta Part B: Atomic Spectroscopy 56, 877-885.

Siano, S., Giamello, M., Bartoli, L., Mencaglia, A., Parfenov, V., Salimbeni, R., 2005. Phenomenological characterisation of stone cleaning by different laser pulse duration and wavelength. In: Lasers in the conservation of artworks, LACONA VI proceedings. New York Springer, Vienna, Austria, Berlin, pp. 87-96.

Wolicka, D., Borkowski, A., 2009. Phosphogypsum biotransformation in cultures of sulphate reducing bacteria in whey. International Biodeterioration & Biodegradation 63, 322-327.

Wyszecki, G., Stiles, W.S., 1982. Color science, concepts and methods, quantitative data and formulae, second ed. Wiley, New York.