- Email: [email protected]
Effectiveness of chemical, mechanical and laser cleaning methods of sulphated black crusts developed on granite
Construction and Building Materials 112 (2016) 682–690
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Effectiveness of chemical, mechanical and laser cleaning methods of sulphated black crusts developed on granite J.S. Pozo-Antonio a,⇑, A. Ramil b, T. Rivas a, A.J. López b, M.P. Fiorucci b a
Departamento de Enxeñaría de Recursos Naturais e Medio Ambiente, Escola de Minas, Universidade Vigo, 36310 Vigo, Spain Laboratorio de Aplicacións Industriais do Láser, Centro de Investigacións Tecnolóxicas (CIT), Departamento de Enxeñaría Industrial II, Escola Politécnica Superior, Universidade de Coruña (UDC), Campus Ferrol, 15403 Ferrol, Spain b
h i g h l i g h t s Mechanical, chemical and laser ablation procedures were applied to clean sulphated black crust on granite. None of the evaluated methods completely removed the studied crust. All methods produced unwanted effects on the rock. Papetta AB57 with Carbopol Ultrez 21 and Ethomeen C25 showed the highest cleaning effectiveness.
a r t i c l e
i n f o
Article history: Received 18 June 2015 Received in revised form 11 February 2016 Accepted 25 February 2016
Keywords: Cultural heritage Granite Sulphated black crust Stone cleaning Micro-sandblasting Laser Nd:YVO4 Poultices Papetta AB57
a b s t r a c t A study of the cleaning effectiveness of sulphated black crust developed on granite is presented. The sulphated black crust, previously characterized, was subjected to a cleaning by 1) a mechanical procedure –Hydrogommage– based on micro-sandblasting, 2) chemical procedures based on the application of poultices made on different mixes of thickening agents and cleaners and 3) laser cleaning using a 355 nm Nd: YVO4 nanosecond laser. Chemical, mineralogical and physical characterization of the cleaned surfaces were performed; the global effectiveness as well as the harmfulness were evaluated according to the level of black crust removal and the substrate damages. As result, none of the methods has been completely effective in removing the sulphated black crust and, also, all the methods produced undesirable effects on the stone. The crust nature, its degree of interaction with the stone and other factors related to the principle of the cleaning procedures were found as the main variables influencing the effectiveness of the cleaning procedures. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Black crusts are blackening originated from the interaction between the stone substrate and the atmospheric contamination, mainly SO2, NOX and CO2 [1]. Its impact on the historical building and monuments is considered so important that the Stone Weathering and Atmospheric Pollution Network (SWAPNET) was founded in 1993 to study the decay of the stones affected by the atmospheric pollution and to search for techniques to remedy it [2]. The final product of the reaction between sulphur oxide and Ca, from carbonate stones rich in calcium, is gypsum, which has different physical and chemical properties than the underlying stone [3,4]. ⇑ Corresponding author. E-mail address: [email protected] (J.S. Pozo-Antonio). http://dx.doi.org/10.1016/j.conbuildmat.2016.02.195 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
The processes that generate this kind of black crusts continue to raise much interest today, but almost exclusively those developed on carbonated sedimentary stones and marbles [5,6]. In [7], the authors pointed that stones, besides to their surface transformation, acquire a black discolouration because the incorporation of carbonaceous material from the contaminated atmosphere, that is originated from the combustion of diesel and gasoline by automobiles and industrial activity. There are few studies about the black crust formation in granite. During a study performed under artificial exposure atmospheres showed that granite sulphation is even possible at relatively poor SO2 atmospheres (10 ppm), as SO2 can react with Ca from the joint mortar [8]. A more recent study confirmed the contribution of two sulphur sources, i.e. anthropogenic SO2 and marine sulphate, in the development of black crust on coastal granitic constructions [4].
683
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
In general terms, crusts cleaning process must be understood as a soft and delicate extraction procedure intended to release the surface dirt without affecting the stone [9]. The particular characteristics of this crust make not advisable to extrapolate its cleaning results to other crust types [10]. There are many cleaning methods and the selection of the most adequate must be done taking into account a lot of theoretical and practical considerations that include (a) diagnosis of the stone surface (stone kind, extent and severity of decay) and (b) work accessibility and the need of scaffolding. Once the most adequate methods are chosen, they must be essayed on specific areas of the degraded surface, in order to estimate the effectiveness and side effects on the stone (risk evaluation) [11]. Since the 90s, the search for more effective cleaning methods that cause less damage to the stone substrate was increased. There are different cleaning methods, being the mechanical and chemical ones the most traditionally used. Mechanical methods, derived from the old sandblasting method, are based on the projection of different kind of particles. Due to their aggressiveness because the high pressure range used (above 50 bar), they are only used in certain circumstances [11,12]. Among chemical methods, the Papetta AB57 formulation (developed by the Istituto Superiore per la Conservazione ed il Restauro in Rome) has been hailed as an excellent cleaning agent for black crust developed on marble [13]. Laser ablation constitutes another cleaning procedure in heritage conservation which goes back to the 70s with the pioneering work by John Asmus about crust extraction in Venetian marble [14]. There are numerous bibliographic works based on the parameters of the laser systems, the mechanisms involved in the surface cleaning and the different uses of laser in restoration of Cultural heritage [15–17]. Concretely, works about its laboratory and in situ applications are very extensive, especially those related to carbonated stones [17–21]. Almost all the references regarding cleaning procedures to remove black crust are focused on carbonated stones (sedimentary or metamorphic) [18–21], therefore the studies related to granitic stones are scarce. This limited scientific production is centred particularly on laser cleaning [22–25]. Also, up to date, there are no works intending to compare effectiveness information between different cleaning procedures of black crust developed on granite. Therefore, the aim of this work is to enhance the knowledge about the effectiveness of different mechanical, chemical and laser ablation-cleaning procedures, in the removing of sulphated black crust developed on a granite widely used in the construction of architectural heritage in NW of Iberian Peninsula. Taking into account the deep influence of the textural and mineralogical peculiarities of the granites in their durability and in their response to conservation treatments [26,27], the access to specific results on the cleaning effectiveness of different methods of black crust in granites would be of great significance in the direct intervention in granitic cultural heritage.
biological structures. Chemically, besides calcium sulphate, alkane and carboxylictype organic compounds were identified; being probably these compounds the reason of the black colour of the crust [4].
2.2. Cleaning methods The cleaning methods used were the following: 1) Chemical procedures: Following previously reported studies [13,29], it was decided to apply nine different chemical cleaning treatments (see Table 1). All these treatments, except one, consisted on the application of a poultice made on a mixture of a cleaning agent and a thickening compound (Table 1). The selected cleaning agents were: - Standard Papetta AB57 composed of 25 g of EDTA, 50 g of sodium bicarbonate, 30 g of ammonium bicarbonate and 10 cm3 of Neodesogen (a biocide based on benzalkonium chloride with fungicide effect which was provided by BIC Materiales y Conservación S.L.), all dissolved in 1000 mL of distilled water. - Papetta AB57 without ammonium bicarbonate, prepared with 25 g EDTA, 50 g sodium bicarbonate and 10 cm3 of Neodesogen, dissolved in 1000 mL of distilled water. - Acidic mixture: 5% (vol.) HCl and 2.5% (vol.) (NH3)HF2 in aqueous solution. The selected thickeners supplied by CTS Slr (details in www.ctseurope.com) were: - Carbopol Ultrez21 prepared at 3% (wt.) in 10% (wt.) ammonium bicarbonate solution. As Table 1 shows, this thickener was prepared with Papetta AB57 without ammonium bicarbonate. - Carbopol Ultrez21 prepared at 3% (wt.) in 10% (wt.) Ethomeen C25 solution. - Laponite RD, a synthetic clay composed of sodium, lithium and magnesium silicates. - Carboxymethylcellulose as water soluble sodium salt of the cellulose glycolic acid. - Carbogel, a neutralized polyacrylic acid. Eight different poultice based treatments were prepared mixing the above mentioned cleaning compounds and thickeners at different proportions until achieving a gel with good workability. In Table 1, the proportions and acronyms for all these treatments are shown. The ninth chemical treatment consisted on the direct application of Amberlite 4400 OH which was also provided by CTS Slr. (www.ctseurope.com). It is a strong exchange resin. A poultice with a proper consistence was prepared mixing 50 g of Amberlite within 30.7 mL of distilled water (Table 1).
Table 1 Chemical cleaning treatments applied in this study. In the third column, the acronym and the cleaning compound-thickener proportion is indicated. Cleaning agent
Thickener
Treatment Cleaning-thickener proportion
PAPETTA AB57
Laponite RD
Papetta AB57 + Laponite (PL) 50 mL: 14.5 g Papetta AB57 + CMC (PCMC) 50 mL: 2 g Papetta AB57 + Carbogel (PC) 50 mL: 4.5 g Papetta AB57 + Carb + Eth (PCUE) 50 mL: 100 mL
Carboxymethylcellulose
Carbogel
2. Materials and methods
Carbopol Ultrez 21 and Ethomeen C25
2.1. Granite and sulphated black crust samples In order to carry out this study, an ashlar affected by an intense development of sulphated black crust was extracted from an ancient building in the city of Vigo (NW Iberian Peninsula). This building is constructed with a fine grained (2–0.3 mm) equigranular granite, composed of quartz (29%), potassium feldspar (25%), sodium plagioclase (24%), muscovite (13%) and biotite (4%) as main minerals [28]. A 100 cm 100 cm 2 cm plate where cut parallel to the crusted surface of the ashlar, in order to obtain more manageable smaller stone 29 slabs of 14 cm 7 cm. Previously to the cleaning process, the black crust was characterized following different techniques. Results of this characterization can be consulted in [4]: briefly, the sulphated black crust comprised a coating of 80–100 lm of thickness composed of acicular-shape crystals of calcium sulphate with occasionally
PAPETTA AB57 without NH4HCO3
Carbopol Ultrez 21 with NH4HCO3
Papetta AB57 + Carb + Bic (PCUB) 50 mL: 100 mL
ACIDIC MIXTURE
Laponite RD
Acidic + Laponite (AL) 50 mL: 19 g Acidic + CMC (ACMC) 50 mL: 3 g Acidic + Carbogel (AC) 50 mL: 4 g
Carboxymethylcellulose Carbogel AMBERLITE 4400 OH
(AM)
684
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
After mixing, 85 cm3 of each of the ninth poultice based treatments were applied on the surface of the stone slabs, completely covering the sulphated black crust and using three slabs for each chemical treatment. A portion of 2 7 cm2 of surface of one of the slabs was not treated therefore providing a not cleaned reference surface. After two hours (time required for a total drying of the poultice), the poultices were mechanically removed, the remains of the poultices dragged with the aid of an absorbent cellulose paper and the cleaned surfaces rinsed with 1:1 (vol.) waterethanol solution with the help of a short bristle brush. This procedure (the poultice application, the removal of the poultice after 2 h and the neutralization of the cleaners) was repeated three times more. After the fourth application and removal, the slab without the two-centimetre control area was subjected to a final neutralization with water vapour. 2) Mechanical cleaning was performed using the patented procedure called HydrogommageÒ. This is a procedure based on a low-pressure projection (0.5–1.5 bars) with a pneumatic rotator of an air-water-micro granule mixture. The abrasive, of grain size less than 80 lm, was composed of 99% of SiO2 and 1% of Al2O3, K2O, TiO2 and Fe2O3. 3) The laser cleaning was performed using a Nd:YVO4 laser (Coherent AVIA Ultra 355–2000) with 355 nm wavelength and 25 ns pulse duration. The laser output intensity profile was Gaussian TEM00 and the beam diameter at 1/e2 intensity level was about 2.2 mm. Pulse repetition rate can be changed from a single-shot to 100 kHz with energy per pulse around 0.1 mJ. The beam impinges perpendicularly onto the target sample surface, which was placed on a motorized XYZ-translation stage Newport ILS-CC. Different distances lens-sample had to be applied to avoid the influence of the high roughness, because the stone surface showed crests and valleys of around 1 mm. 2.3. Analytical techniques After the cleaning of the slabs by the different methods, the following analyses were applied. Stereoscopic zoom microscope (SMZ800 Nikon) was applied in order to get an initial qualitative evaluation of the effectiveness of the cleaning methods, considering for this the presence of crust remains and the recovery of the original colour of the granite. Also, the occurrence of residues of the cleaning products has been evaluated in this phase. Observation under ultraviolet light was performed in order to detect the presence of remaining residues. Cross-sections and superficial samples were observed using Scanning Electron Microscope (SEM) (Philips XL30 and JEOL JSM-6700) and also analysed by means of X-ray microanalysis (EDX). Xray diffraction (XRD) (SIEMENS D-5000) and Fourier Transformation Infrared Spectroscopy (FTIR) were used to evaluate the presence of by-products or cleaner remains on the surfaces. FTIR data were collected in reflection mode using a FTIR Nicolet Continuum (Thermo), analysing the cleaned surfaces and, for comparative purposes, the not-cleaned surface of the black crust and the granite without black crust (using the face of the slab which was opposite to that affected by the crust); data were transformed to absorption spectra. In order to assign the FTIR bands, different references were consulted [30–32]. Colour characterization before and after the cleaning treatments was made by means of a Minolta CM-700d spectrophotometer, expressing colour data in CIELAB colour space. Colourimetric differences DL⁄, Da⁄, Db⁄, DC⁄ab and DH⁄ and global colour change (DE⁄ab) were calculated taking the colour data for the sulphated black crust as reference colour. Analysis of variance was applied to colour data and Tukey post hoc test (or multiple comparison Tukey test) was used to determine the significant differences between the colour data measured after and before cleanings. Additionally, colour changes after cleaning were qualitatively analysed using scatter plots representing raw data for the granite with black crust, for the granite after the cleaning and also for the granite without crust (obtaining the colour data of the opposite surface to that affected by the crust). As chemical cleanings were performed in four applications, colour measurements were also taken after each application, in order to evaluate the effectiveness of each chemical procedure in terms of number of applications needed to achieve optimal results.
3. Results The first qualitative evaluation of the effectiveness by means of optical microscopy revealed none of the procedures applied (chemical, mechanical and laser ablation) was able to completely extract the sulphated black crust: after cleaning, the surface of the stone became lighter but with higher magnifications, gypsum crystals remains over the surface can be seen, in a greater or a lesser amount depending on the cleaning procedure (Fig. 1). Therefore, cleaning procedures seemed to be more effective in removing the carbonaceous fraction of the crust (which confers the dark colour) than in gypsum removal (which was the main compound of the crust).
For comparative proposes, in Fig. 1, the micrographs of the uncoated granite and the black crust are shown (Fig. 1A and B respectively).Considering this circumstance, it was possible to define a scale of effectiveness among the procedures applied (Table 2): the cleaning methods which showed a higher effectiveness were the chemical procedures PCUE, ACMC and AM (Fig. 1, C and D). Although, either of the chemical cleaning procedures has shown a higher gypsum removal effectiveness than that obtained for laser ablation or mechanical procedure, the cleaning levels obtained for chemical procedures were hampered by the occurrence of remains on surface associated with the chemical thickeners (compare Fig. 1, C–E with Fig. 1, F); these surface residues were found in smaller amounts when water vapour was used in order to neutralize the chemical cleanings (Fig. 1, G and H). Regarding chemical procedures and taking into account the different cleaning levels evaluated (high, medium and low) (Table 2), no relationship between a high cleaning level and the use of a specific thickener can be identified. Nevertheless, an apparently higher effectiveness in the use of acidic mixture, as cleaning agent, seemed to be noticed. Ultraviolet light observations confirmed the presence of organic residues on the surfaces subjected to chemical cleaning and also the fact that these residues were less abundant in the samples cleaned with chemical methods and subsequently neutralized with water vapour (Fig. 1, G and H). The advantage of the application of water vapour was especially evident in the case of Papetta AB57 with Laponite (Fig. 1, H). As commented above, either of the chemical cleaning procedures has shown a higher effectiveness than that obtained for laser ablation or mechanical procedure. After laser cleaning, it was observed that the black colour of the crust was eliminated, but not the gypsum which remained on the surfaces as a continuous and massive coating. Regarding Hydrogommage, after the cleaning, gypsum remains were observed all over the surface. SEM-EDX analysis confirmed the optical microscopy observations regarding the cleaning effectiveness of the different procedures: 1) None of the cleaning procedures achieved a completely gypsum extraction, because EDX detected the presence of gypsum through S and Ca-signals on all the cleaned surface. The scale of effectiveness using SEM-EDX observations was coincident with optical microscopy. Chemical procedures were more effective than Hydrogommage and laser in terms of gypsum extraction (compare Fig. 2, A–F with Fig. 2, G and H). Among the chemicals, the PCUE (Fig. 2, A) was the most effective in removing the black crust, the PCUB (Fig. 2, B) and AL were the procedures which gave the worst results and ACMC (Fig. 2, E) and AM (Fig. 2, F) achieved intermediate results. 2) All chemically cleaned surfaces showed residues and byproducts related to the chemical reagents all over the surfaces. On samples cleaned with the thickener CMC (PCMC treatment, Fig. 2, C and D), residues rich in C and Ca were identified, suggesting the precipitation of calcium carbonate derived from the gypsum dissolution and a later reaction with the sodium bicarbonate of the Papetta AB57 formulation. Also, deposits rich in Na, Cl, Mg and S were identified in the samples cleaned using the thickener Laponite. Ca, S and F rich deposits were found on the samples cleaned with acidic mixture adding Carbogel (treatment AC), being F associated with the acidic nature of the cleaning compound, rich in HF. Finally after cleaning with AM treatment, the difficult of removing the Amberlite from the surfaces was
685
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
A
B
C
D
2000 µm
2000 µm
F
E
2000 µm
H
G Not neutralized
Neutralized
Not neutralized
Neutralized
Fig. 1. A–B: Micrographs of the uncoated granite and the black crust respectively. C–F: Micrographs taken with optical microscope of the cleaned surfaces with different methods. C: cleaned with PCUE (Papetta AB57 with Carbopol Ultrez21 and Ethomeen C25). D: cleaned with AM (Ambelite 4400 OH). E: cleaned with PL (Papetta AB57 with Laponite). F: cleaned with 355 nm Nd:YVO4 laser (cleaned area pointed out with a rectangle). G: Photograph taken under UV light of the slabs cleaned with PCUE (Papetta AB57 wth Carbopol Ultrez21 and Ethomeen C25) treatment. H: Photograph taken under UV light of the slabs cleaned with PL (Papetta AB57 with Laponite) treatment. In these images (G and H), remains of chemical compounds are observed in bright colour. Also, on right side, the slab that was neutralized with water vapour after cleaning. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
corroborated by means of the SEM-EDX observations, due to the presence of a crust composed of this product (Fig. 2, F). Finally, cellulose fibres were found on almost all the surfaces cleaned with chemical methods (Fig. 2, D). 3) Laser and Hydrogommage didn’t achieve better results in terms of gypsum extraction (Fig. 2G and H). After cleaning with Hydrogommage, surfaces shown a large amount of gypsum crystals with acicular habit (Fig. 2, G). Also, SEMEDX observations of samples cleaned with laser allowed to determine that laser didn’t ablate the gypsum crust: surfaces appeared covered with massive gypsum granules; this fact
suggests a partial melting of this mineral (Fig. 2, H). It must also be noted that samples cleaned with laser showed a partial melting of the biotite crystals (Fig. 2, H). XRD (Table 3) confirmed the presence of gypsum remains after the cleanings in variable amounts (never above 10%), confirming that none of the cleaning procedures allowed a complete gypsum elimination. XRD results indicated that, in terms of the amount of gypsum remains after cleaning, the most effective treatments would be the chemical methods using either CMC or Laponite as thickeners. This finding, which didn’t match with the observations
686
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
Table 2 Cleaning level evaluation by means of optical microscopy of the different cleaning treatments. The occurrence of residues on the surfaces after the cleanings is also shown. Treatment
Clean level High
PL PCMC PC PCUE PCUB AL ACMC AC AM Nd:YVO4 laser Hydrogommage
Medium
Occurrence of residues Low
X X X
X X
X X X X
X X
X X
X X X
by means of optical microscopy and SEM-EDX, would indicate that, besides gypsum remains, the presence of residues of different nature influences on the qualitative evaluation of the cleaning level. Fig. 3 shows the FTIR spectra of the surfaces cleaned with chemical treatments (those that produced the best, the intermediate and the worst results by means of optical microscopy and SEM-EDX: PCUE, PCMC and AL treatments respectively), mechanical treatment (Hydrogommage) and laser cleaning. For comparative purposes, spectra of the sulphated black crust and of the granite without crust are also presented. The spectrum of the granite without crust is characterized by the functional SiAO group bands at around 785 cm 1 and between 1020 and 1168 cm 1, which correspond to the different chemical environments characteristic of the various minerals phases (quartz, muscovite, biotite, feldspar, etc.) [31,32]. In sulphated black crust FTIR spectrum, gypsum is identified through the absorbance bands occurring at around 667 cm 1, 1619 cm 1, 1682 cm 1, 3492 cm 1 and 3525 cm 1. The typical band that this mineral shows at 1122 cm 1 is overlapped by signals from the SiAO bond of the minerals in the stone. FTIR data for cleaned surfaces confirmed the results obtained using optical microscopy and SEM-EDX: gypsum bands were detected in almost all the samples; these bands were more evident in samples evaluated with a medium to low degree of cleaning (Table 2). In the spectra from surfaces cleaned with the chemical preparations, bands attributed to gypsum at around 667 cm 1, 1619 cm 1, 1682 cm 1, 3492 cm 1 and 3525 cm 1 were identified; these bands were more intense in surfaces cleaned with AL (one of the chemical procedures which gave worst results); they also appeared with less intensity in the spectrum of the surface cleaned with Hydrogommage but they were absent in laser cleaned surfaces. This indicates the low presence or absence of gypsum on the laser cleaned surfaces. In the surfaces cleaned with chemical procedures, bands which would indicate the presence of remains of the organic compounds of the cleaning agents and thickeners can be detected; in these cases, the crust extraction effectiveness was overshadowed by the band intensity corresponding to the chemical preparations. In the spectra of surfaces cleaned with PCMC and PCUE (Fig. 3), several bands that can be attributed to the cleaning agent residuals, were detected, i.e. bands at 680 cm 1 and 832 cm 1 were identified as from the benzene CAH functional group (corresponding to the Neodesogen benzil radicale). In the surfaces cleaned with PCMC also between 1400 and 1590 cm 1 there were bands associated with the CAO functional group; bands characteristic of the tertiary amines (EDTA) can be detected at the region at 1739 cm 1 and at 2300–2622 cm 1. Most notable was the finding of residuals from the CMC thickener, as demonstrated by the absorption band
detected at 1323 cm 1 – typical of the CAOAC functional group – and the shoulder-type peak at 3400 cm 1, assigned to the cellulose. Attending to these results, in the surface cleaned with PCUE, the effects attributed to the cleaning agents were lesser than those detected in surfaces cleaned with PCMC (Fig. 3). The crust extraction effectiveness was also evaluated by means of colour data. Table 4 shows the colourimetric differences obtained after the cleaning with each procedure. For chemical procedures, colour differences after each application are also shown. In Fig. 4, L⁄-C⁄ab scatter plots shows the raw colour data from the cleaned surfaces with different procedures and the colour of the granite with and without crust. It was established as a criterion for the best cleaning, the higher the global colour change (DE⁄ab), the more effective in removing the crust. So, in Table 4 it can be seen that the greater effectiveness was achieved by Hydrogommage (DE⁄ab: 24.98), laser (DE⁄ab: 20.90), AM (DE⁄ab: 20.62) and PCMC (DE⁄ab: 19.30). Recall that the latter was also considered as one of the most effective by microscopy and after FTIR analysis. In the opposite situation, preparations that produced the lower global colour change were the PCUB (DE⁄ab: 15.44) and AL (DE⁄ab: 8.46) In Table 4 it can be seen that lightness, L⁄, was the colourimetric coordinate that most contributed to the global colour change in all cases; coordinates a⁄ and b⁄ (especially the a⁄ coordinate) showed smaller changes. Concerning the chemical cleaning methods, it was clearly observed that the global colour change (DE⁄ab) increased with the number of applications, indicating a progressive extraction of the crust. When comparing among the different applications, it can be observed that with AM and AC, the maximum DE⁄ab value was almost achieved at the second application, suggesting a higher cleaning effectiveness. In Fig. 4, colour raw data of each cleaned surface, of the sulphated black crust and of the original colour of the granite are represented. It can be seen that none of the treatments allowed restore the original colour to the stone. Despite this fact, in the case of chemical treatments made with the acidic mixture (Fig. 4, A), those which were evaluated as the most effective in removing the crust by microscopy and FTIR (ACMC, AM) were those whose colours fell closer to the original colour of the stone. These same methods caused the highest DE⁄ab (Table 4). A same behaviour was not found, however, for the chemical treatments based on Pappeta AB57 (Fig. 4, B) or after cleaning with laser and Hydrogommage (Fig. 4, C). Thus, the colour after the cleaning with the PCUE treatment (evaluated as more effective) fell quite far from the original colour of the stone, especially as regards the C⁄ab. The cleaning with Hydrogommage showed a low efficacy in removing the crust; however it produced a very high DE⁄ab but that was not reflected in a higher recovery of the original colour of the stone. The cleaning with Laser and Hydrogommage produced a very high global colour change (DE⁄ab: 20.90 and 24.98 respectively). Nevertheless, it was observed that such high DE⁄ab values didn’t imply a closer approach of the colour of the cleaned surfaces to the original stone colour (Fig. 4, C).
4. Discussion The techniques applied to evaluate the cleaning effectiveness of the sulphated black crust with different methods (chemical, Hydrogommage and laser) in the studied granite confirmed the difficulty to successfully remove this crust. After the cleaning, gypsum was present on all the cleaned surfaces as corroborated by SEM-EDX, XRD and FTIR. Nevertheless, by the techniques applied, it was possible to identify those procedures which allowed a high cleaning level without producing damage to the stone (such as
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
687
Fig. 2. SEM micrographs of the granite surfaces after sulphated black crust cleaning. A: PCUE (Papetta AB57 wth Carbopol Ultrez21 and Ethomeen C25) treatment. B: PCUB (Papetta AB57 with Carbopol Ultrez21 and ammonium bicarbonate). C: PCMC (Papetta AB57 with CMC). D: PCMC with more detail. E: ACMC (Acidic mixture with CMC). F: AM (Ambelite 4400 OH). G: Hydrogommage. H: 355 nm Nd:YVO4 laser.
cleaning residues or mineralogical and textural alterations). These chemical methods were PCUE (using as cleaning agent Pappeta AB57) and ACMC (using as cleaning agent the acidic mixture). The treatment which gave the worst results were PCUB (using as cleaning agent Pappeta AB57). The cleaning effectiveness of chemical products was weakened by the presence of surface residuals associated to the chemical preparations.
SEM-EDX analyses confirmed that after the cleaning with all the chemical methods, residues of different nature remained on the stone surfaces, derived from the cleaning and thickener agents. Also, by-products of the reaction between the cleaning agents and the gypsum of the crust were observed, as was the case of calcium carbonate possibly formed by the reaction with sodium bicarbonate of the Pappeta AB57.
688
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
Table 3 Semiquantitative results of the XRD of the sulphated black crust and the surfaces after their cleaning with different procedures. Q: quartz; P: plagioclase; F-K: potassium feldspar; M: muscovite; B: biotite; G: gypsum. + + + +: >50%; + + +: 30–50%; + +: 10– 30%; +: 3–10%; tr (traces): <3%, N.D.: no detected. Treatment
Q
Black crust PL PCMC PC PCUE PCUB ACMC AL AC AM Nd:YVO4 laser Hydrogommage
+++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++
+ +
+ +
P
F-K
M
B
G
++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++
++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++
+ ++ ++ + ++ ++ + ++ N.D ++ ++ N.D.
N.D N.D N.D N.D N.D N.D N.D N.D + N.D N.D N.D.
+ tr tr + + + tr tr + + + +
Fig. 3. From bottom to top, FTIR spectra of granite without black crust, granite with black crust, surface cleaned with laser, surface cleaned with Hydrogommage and surfaces cleaned with AL (Acidic mixture with Laponite), PCMC (Papetta AB57 with CMC) and PCUE (Papetta AB57 wth Carbopol Ultrez21 and Ethomeen C25).
Table 4 Colourimetric differences DL⁄, Da⁄, Db⁄, DC⁄ab, DH⁄ and global colour change (DE⁄ab) after the cleaning of the sulphated black crust with the different procedures (chemicals, Hydrogommage and laser). Differences were computed relative to the crust colour, so the greater the difference the higher the cleaning effectiveness. For chemical methods, colour data for each of the four applications are shown. Ap. N°: application number. *: Significant differences (n.s. < 0.05) comparing to the crust colour data. Treatment
Ap. N°
DL ⁄
Da⁄
Db ⁄
DC⁄ab
DH⁄
DE⁄ab
PL
1 2 3 4
2.62⁄ 9.65⁄ 2.75⁄ 17.50
0.83⁄ 0.48⁄ 1.38⁄ 0.42⁄
1.88⁄ 1.15⁄ 4.11⁄ 1.70⁄
2.02⁄ 1.22⁄ 4.32⁄ 1.75⁄
2.43⁄ 1.69⁄ 1.92⁄ 0.19
3.33 9.73 5.13 17.58
PCMC
1 2 3 4
6.94⁄ 12.57 16.04 18.96
1.11⁄ 0.81⁄ 0.83⁄ 0.95⁄
3.49⁄ 2.82⁄ 3.15⁄ 3.52⁄
3.65⁄ 2.92⁄ 3.25⁄ 3.64⁄
1.61⁄ 1.06 0.79 0.84
7.85 12.90 16.37 19.30
PC
1 2 3 4
4.71⁄ 11.14 15.70 16.43
0.42⁄ 0.05 0.06 0.41⁄
2.28⁄ 0.90⁄ 0.38 1.03⁄
1.35⁄ 0.89⁄ 0.36 1.09⁄
2.82⁄ 2.14⁄ 1.85⁄ 1.51⁄
5.25 11.18 15.70 16.46
PCUE
1 2 3 4
3.22⁄ 6.44⁄ 8.35⁄ 14.41
1.20⁄ 1.71⁄ 1.53⁄ 1.98⁄
2.81⁄ 4.14⁄ 4.01⁄ 5.99⁄
3.02⁄ 4.45⁄ 4.27⁄ 6.29⁄
2.87⁄ 3.39⁄ 2.88⁄ 2.44⁄
4.44 7.84 9.39 15.73
PCUB
1 2 3 4
7.35⁄ 10.06 4.02⁄ 15.22
0.22 0.04 0.86⁄ 0.42⁄
0.25 0.64 3.15⁄ 2.53⁄
0.30 0.61 3.26⁄ 2.55⁄
0.97 1.10 0.77⁄ 0.82
7.36 10.08 5.18 15.44
AL
1 2 3 4
2.62⁄ 3.75⁄ 2.06⁄ 7.29⁄
0.77⁄ 1.05⁄ 1.71⁄ 1.36⁄
1.99⁄ 2.92⁄ 4.44⁄ 4.08⁄
2.11⁄ 3.09⁄ 4.74⁄ 4.29⁄
1.78⁄ 1.93⁄ 0.07 1.98⁄
3.38 4.87 5.19 8.46
ACMC
1 2 3 4
8.94⁄ 14.16 20.90 23.25
0.67⁄ 1.02⁄ 0.09⁄ 0.78⁄
1.64⁄ 2.62⁄ 0.02⁄ 2.53⁄
1.75⁄ 2.79⁄ 0.00⁄ 2.64⁄
2.04⁄ 2.42⁄ 0.03⁄ 1.47⁄
9.11 14.44 20.90 23.40
AC
1 2 3 4
11.03 11.96 13.51 15.38
0.40⁄ 0.41⁄ 0.59⁄ 0.56⁄
1.14⁄ 1.25⁄ 1.87⁄ 1.68⁄
1.20⁄ 1.31⁄ 1.95⁄ 1.76⁄
1.08 0.95 0.03⁄ 1.38⁄
8.49 12.03 13.65 15.49
AM
1 2 3 4
8.71⁄ 19.12 17.33 20.18
0.90⁄ 0.65⁄ 1.04⁄ 1.13⁄
1.65⁄ 1.36⁄ 3.10⁄ 4.04⁄
1.83⁄ 1.48⁄ 3.26⁄ 4.19⁄
3.28⁄ 2.61⁄ 0.04⁄ 1.20⁄
8.92 19.18 17.63 20.62
Nd:YVO4 laser Hydrogommage
1 1
20.82 23.90
0.92⁄ 2.31⁄
1.61⁄ 6.87⁄
1.79⁄ 7.23⁄
4.07⁄ 2.90⁄
20.90 24.98
FTIR spectra also helped to verify the permanence on the stone surfaces of residuals from the chemical preparations after the cleanings. They also were very useful to evaluate the extraction effectiveness: inexistent or low intensity bands assigned to gypsum were found on the surfaces cleaned with PCUE and ACMC, the two methods which were identified as the most effective by means of optical microscopy. The residues related to the chemical and thickener compounds in the case of the chemical treatments can be avoided by the application of water vapour after the cleaning. A fact to note is that, among the residues, many cellulose fibres have been detected and not only after the cleaning with the treatments based on CMC, but also after those which do not use it. These fibres come from the absorbent cellulose paper used to neutralize the cleaners. So, it is recommended the use of an alternative absorbent material for the mechanical elimination of the cleaner compounds. When using the colour data, it was done taking in consideration that the higher global colour changes (DE⁄ab) the better extraction effectiveness [33]. After cleaning, colour data indicated that all chemical treatments except AM needed four applications to reach
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
A AL ACMC AC AM
*
B PL PCMC PC PCUE PCUB
689
the PCUB, also in accordance with other evaluation techniques. Nevertheless, considering all the cleaning procedures (chemicals, Hydrogommage and laser ablation), the global colour change would point out the Hydrogommage as the most effective method, in disagreement with the other evaluation techniques. This lack of coincidence between the DE⁄ab, the evaluation under microscopy and FTIR and the degree of recovery of the original colour suggested that the cleaning treatments by themselves produced a colour modification of the rock, regardless of its effectiveness in removing the crust. This change may be related to the existence of remains on the surface of the stone after chemical cleaning, or mineralogical or textural changes caused by mechanical method and the laser. This leads to the need to consider the DE⁄ab variations and but also the evaluation of the degree of recovering the original colour for a correct effectiveness evaluation of the treatments. The very low effectiveness of the laser used in this work, a nanosecond Nd:YVO4 at 355 nm, in the removing of the sulphated black crust from the granitic surface must be highlighted. Previous works reported that the complete laser extraction of sulphated black crust of carbonated sedimentary stones and marble was not achieved [18,19,33], but a high effectiveness can be obtained with the third harmonic or in combination with [18]. In those works, Nd:YAG laser was applied to clean sulphated black crusts on homogeneous substrates from the mineralogical and textural point of view. In this work, the heterogeneous surface of the granite increases the gypsum removal difficulty as was concluded in Nd:YAG at 1064 nm cleaning of sulphated black crust [25].
5. Conclusions
*
C
* Fig. 4. L⁄ and C⁄ab scatter plots of representing raw colour data from A) surfaces cleaned with Acidic mixture with different thickeners and Amberlite. B: Surfaces cleaned with Papetta AB57 with different thickeners. C: Surfaces cleaned with laser and Hydrogommage. Plots also include colour values for the granite with and without sulphated black crust. For chemical procedures, data after the fourth application are shown. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
the maximum DE⁄ab value; however with AM the maximum DE⁄ab was reached with only two applications. Among the chemicals, the most effective cleanings, following colour data, were those performed with the PCMC and ACMC, being the latter considered, by means of the other techniques, as one of the most effective. According to the DE⁄ab the less effective chemical procedure was
From the results of this study, focused on the evaluation of the effectiveness of different cleaning procedures in removing sulphated black crust developed on granite, the following conclusions can be drawn: 1. None of the reported treatments completely removed the sulphated black crust. Nevertheless, a ranking of effectiveness have been defined, that may have immediate applications in conservation practice. So, all the techniques applied in this study, except colour data evaluation, agreed that the chemical methods were more effective than the mechanical method (Hydrogommage) and the use of a nanosecond Nd:YVO4 laser at 355 nm. The treatment that showed the higher effectiveness was that based on the Papetta AB57 with Carbopol Ultrez 21 and Ethomeen C25 as a thickeners (PCUE treatment). 2. All the evaluated chemical cleaning procedures left residues on the surface of the granite, which were identified as 1) residues of the cleaning and thickeners agents, 2) by-products from the reaction between the crust and chemical reagents and 3) remains of the absorbent paper used in the neutralization. Thus, to increase the global effectiveness of the granite cleaning interventions with chemical products, it is recommended the optimization of the neutralization process, avoiding the use of absorbent cellulose paper and applying water vapour on the surfaces after the treatment. 3. The cleaning with a 355 nm Nd:YVO4 laser led to biotite melting. Also, the mechanical method, Hydrogommage, increased the roughness of the stone. So, these methods need to be examined in order to optimize the cleaning parameters (in the case of laser) or the design (such as the spray pressure or the grain size of the abrasive, in the case of Hydrogommage), thereby increasing their efficiency and reducing the damage to the granite. 4. Because all cleaning methods caused in a greater or a lesser extent harmful effects on the stone (remains of chemical cleaners, textural changes in the mechanical method and textural
690
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
and mineralogical changes after the laser cleaning) that may contribute to a change in colour, the use of global colour change as a quantitative indicator of the removal degree of the crust should be undertaken with caution. A more appropriate way, in these cases, is also consider the qualitative assessment of the degree of recovery of the original colour of the stone. For these reason, besides the global colour change, it was necessary to consider the L⁄-C⁄ab scatter plots of representing raw colour data. 5. In order to achieve a complete removal of the sulphated crusts developed on granite, future works will be focused on optimizing the laser cleaning (combining different energy levels and different wavelengths or lasers) and other chemical products and mechanical procedures will be used.
Acknowledgements This work was supported by the Spanish project CTM201019584 funded by Ministerio de Ciencia y Tecnología. M.P. Fiorucci’s work was financed by the project: A-TEMPO Avances en tecnologías marinas: producción naval y Offshore. Plataforma robótica subacuática de inspección para el mantenimiento de estructuras Offshore. SEM-EDX, XRD and FTIR analysis were performed at the Centro de Apoyo Científico y Tecnológico a la Investigación – CACTI University of Vigo. The authors would like to thank Proliser S.L., ClinArte S.L. and Restoration and Conservation School of Galicia (Pontevedra), concretely Cristina Montojo and Fernando Carrera. J.S. Pozo-Antonio was supported by a postdoctoral contract with the University of Vigo within the framework of the 2011–2015 Galician Plan for Research, Innovation and Growth (Plan I2C) for 2014. References [1] P. Ausset, J.L. Crovisier, M. Del Monte, V. Furlan, F. Girardet, C. Hammecker, D. Jeannette, R.A. Lefevre, Experimental study of limestone and sandstone sulphation in polluted realistic conditions: the Lausanne Atmospheric Simulation Chamber (LASC), Atmos. Environ. 30 (1996) 3197–3207. [2] B.J. Smith, P.A. Warke, Processes of urban stone decay, in: Proceedings of SWAPNET ’95, Donhead Publishing, London, 1996, p. 274. [3] N. Schiavon, Biodeterioration of calcareous and granitic building stones in urban environments, in: S. Siegesmund, T. Weiss, A. Vollbrecht (Eds.), Natural Stone, Weathering Phenomena, Conservation Strategies and Case Studies, Geological Society of London, London, 2002, pp. 195–205. [4] T. Rivas, S. Pozo, M. Paz, Sulphur and oxygen isotope analysis to identify sources of sulphur in gypsum-rich black crusts developed on granites, Sci. Total Environ. s482–483 (2014) 137–147. [5] A.E. Charola, R. Ware, Acid deposition and the deterioration of stone: a brief review of a broad topic, Geol. Soc. Spec. Publ. 205 (2002) 393–406. [6] S. Siegesmund, J. Ruedrich, A. Koch, Marble bowing: comparative studies of three different public building facades, Environ. Geol. 56 (2008) 473–494. [7] J. Simão, E. Ruiz-Agudo, C. Rodríguez-Navarro, Effects of particulate matter from gasoline and diesel vehicle exhaust emissions on silicate stones sulfation, Atmos. Environ. 40 (2006) 6905–6917. [8] T. Rivas, B. Prieto, B. Silva, Gypsum formation in granitic rocks by dry deposition of sulphur dioxide, in: Proceedings 4th International Symposium on the Conservation of Monuments in the Mediterranean Basin, Rhodes, 1997, pp. 263–270. [9] E. Doehne, C.A. Price, Stone Conservation: An Overview of Current Research, second ed., The Getty Conservation Institute, Los Ángeles, USA, 2010. p. 158.
[10] B.J. Smith, M. Gómez-Heras, S. McCabe, Understanding the decay of stone-built cultural heritage, Prog. Phys. Geogr. 32 (2008) 439–461. [11] R. Bellmunt, A. Paricio, N. Vila, Reconocimiento, diagnosis e intervención en fachadas, Institut de Tecnología de la Construcció de Cataluña, Spain, 2002, p. 208. [12] P. Gaspar, C. Hubbard, D. McPhail, A. Cummings, A topographical assessment and comparison of conservation cleaning treatments, J. Cult. Heritage 4 (2003) 294–302. [13] L. Lazzarini, M.L. Tabasso, Il restauro della pietra, CEDAM-Casa Editrice Dott. A, Milani, Padova, 1986. p. 334. [14] J.F. Asmus, G. Guattari, L. Lazzarini, G. Musumeci, R.F. Wuerker, Holography in the conservation of statuary, Stud. Conserv. 18 (1973) 49–63. [15] M.I. Cooper, D.C. Emmony, J. Larson, Characterization of laser cleaning of limestone, Opt. Laser Technol. 27 (1995) 69–73. [16] C. Fotakis, D. Anglos, V. Zafiropoulos, S. Georgiou, V. Tornari, Lasers in the Preservation of Cultural Heritage. Principles and Applications, Taylor&Francis, London, 2006, p. 336. [17] I. Osticioli, M. Mascalchi, D. Pinna, S. Siano, Removal of Verrucaria nigrescens from Carrara marble artefacts using Nd:YAG lasers: comparison among different pulse durations and wavelengths, Appl. Phys. A: Mater. Sci. Process. 118 (2015) 1517–1526. [18] S.S. Potgieter-Vermaak, R.H.M. Godoi, R. Van Grieken, J.H. Potgieter, M. Ouija, M. Castillejo, Micro-structural characterization of black crust and laser cleaning of building stones by micro-Raman and SEM techniques, Spectrochim. Acta, A 61 (2005) 2460–2467. [19] M. Iglesias, J.L. Prada, N. Guash, Technique for cleaning Tarragona Miocene age dolomitized silty limestone, altered by urban pollution, Mater. Constr. 58 (2008) 247–262. [20] A. Khedr, V. Papadakis, P. Pouli, D. Anglos, M.A. Harith, The potential use of plume imaging for real-time monitoring of laser ablation cleaning of stonework, Appl. Phys. B 105 (2011) 485–492. [21] P. Ortiz, V. Antúnez, R. Ortiz, J.M. Martín, M.A. Gómez, A.R. Hortal, B. MartínezHaya, Comparative study of pulsed laser cleaning applied to weathered marble surfaces, Appl. Surf. Sci. 283 (2013) 193–201. [22] R.M. Esbert, C.M. Grossi, A. Rojo, F.J. Alonso, M. Montoto, J. Ordaz, M.C. Pérez de Andrés, C. Escudero, M. Barrera, E. Sebastián, C. Rodríguez-Navarro, Application limits of Q-switched Nd:YAG laser irradiation for stone cleaning based on colour measurements, J. Cult. Heritage 4 (2003) 50s–55s. [23] C.M. Grossi, F.J. Alonso, R.M. Esbert, A. Rojo, Effect of laser cleaning on granite color, Color Res. Appl. 32 (2007) 152–159. [24] A. Ramil, A.J. López, M.P. Mateo, C. Álvarez, A. Yáñez, Colour changes in Galician granitic stones induced by UV Nd:YAG laser irradiation, in: Lasers in the Conservation of Artworks – Proceedings of the International Conference LACONA 7, 2008, pp. 199–202. [25] S. Pozo, P. Barreiro, T. Rivas, P. Gonzalez, M.P. Fiorucci, Effectiveness and harmful effects of sulphated black crust removal from granite using Nd:YAG nanosecond pulsed laser, Appl. Surf. Sci. 302 (2014) 309–313. [26] C. Cardell, T. Rivas, M.J. Mosquera, B. Prieto, J.M. Birginie, B. Silva, A. Moropoulou, R. Van Grieken, Patterns of damage in igneous and sedimentary rocks under conditions simulating sea-salt weathering, Earth Surf. Proc. Land. 28 (2003) 1–14. [27] T. Rivas, B. Prieto, B. Silva, Permeability to water vapour in granitic rocks. Application to the study of deleterious effects on conservation treatments, Build. Environ. 36 (2000) 239–246. [28] IGME (Instituto Geológico y Minero de España), Mapa geológico de España. Serie Magna, E1:50.000, second ed. Sheet 223., 1985 (Vigo). [29] T. Rivas, J. Iglesias, J. Taboada, J.A. Vilán, Sulphide oxidation in ornamental slates: protective treatment with siloxanes, Mater. Constr. 61 (2011) 115–130. [30] G. Socrates, Infrared and Raman Characteristic Group Frequencies: Tables and Charts, third ed., John Wiley & Sons, 2001. p. 366. [31] P.A. Schroeder, Infrared spectroscopy in clay science, in: A. Rule, S. Guggenheim (Eds.), Teaching Clay Science, The Clay Mineral Society, Aurora, CO, 2002, pp. 181–206. [32] P. Makreski, G. Jovanovski, B. Kaitner, Minerals from Macedonia. XXIV. Spectra-structure characterization of tectosilicates, J. Mol. Struct. 924–926 (2009) 413–419. [33] E. Gioventù, P.F. Lorenzi, F. Villa, C. Sorlini, M. Rizzi, A. Cagnini, A. Griffo, F. Cappitelli, Comparing the bioremoval of black crusts on colored artistic lithotypes of the Cathedral of Florence with chemical and laser treatment, Int. Biodeterior. Biodegradation 65 (2011) 832–839.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Effectiveness of chemical, mechanical and laser cleaning methods of sulphated black crusts developed on granite J.S. Pozo-Antonio a,⇑, A. Ramil b, T. Rivas a, A.J. López b, M.P. Fiorucci b a
Departamento de Enxeñaría de Recursos Naturais e Medio Ambiente, Escola de Minas, Universidade Vigo, 36310 Vigo, Spain Laboratorio de Aplicacións Industriais do Láser, Centro de Investigacións Tecnolóxicas (CIT), Departamento de Enxeñaría Industrial II, Escola Politécnica Superior, Universidade de Coruña (UDC), Campus Ferrol, 15403 Ferrol, Spain b
h i g h l i g h t s Mechanical, chemical and laser ablation procedures were applied to clean sulphated black crust on granite. None of the evaluated methods completely removed the studied crust. All methods produced unwanted effects on the rock. Papetta AB57 with Carbopol Ultrez 21 and Ethomeen C25 showed the highest cleaning effectiveness.
a r t i c l e
i n f o
Article history: Received 18 June 2015 Received in revised form 11 February 2016 Accepted 25 February 2016
Keywords: Cultural heritage Granite Sulphated black crust Stone cleaning Micro-sandblasting Laser Nd:YVO4 Poultices Papetta AB57
a b s t r a c t A study of the cleaning effectiveness of sulphated black crust developed on granite is presented. The sulphated black crust, previously characterized, was subjected to a cleaning by 1) a mechanical procedure –Hydrogommage– based on micro-sandblasting, 2) chemical procedures based on the application of poultices made on different mixes of thickening agents and cleaners and 3) laser cleaning using a 355 nm Nd: YVO4 nanosecond laser. Chemical, mineralogical and physical characterization of the cleaned surfaces were performed; the global effectiveness as well as the harmfulness were evaluated according to the level of black crust removal and the substrate damages. As result, none of the methods has been completely effective in removing the sulphated black crust and, also, all the methods produced undesirable effects on the stone. The crust nature, its degree of interaction with the stone and other factors related to the principle of the cleaning procedures were found as the main variables influencing the effectiveness of the cleaning procedures. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Black crusts are blackening originated from the interaction between the stone substrate and the atmospheric contamination, mainly SO2, NOX and CO2 [1]. Its impact on the historical building and monuments is considered so important that the Stone Weathering and Atmospheric Pollution Network (SWAPNET) was founded in 1993 to study the decay of the stones affected by the atmospheric pollution and to search for techniques to remedy it [2]. The final product of the reaction between sulphur oxide and Ca, from carbonate stones rich in calcium, is gypsum, which has different physical and chemical properties than the underlying stone [3,4]. ⇑ Corresponding author. E-mail address: [email protected] (J.S. Pozo-Antonio). http://dx.doi.org/10.1016/j.conbuildmat.2016.02.195 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
The processes that generate this kind of black crusts continue to raise much interest today, but almost exclusively those developed on carbonated sedimentary stones and marbles [5,6]. In [7], the authors pointed that stones, besides to their surface transformation, acquire a black discolouration because the incorporation of carbonaceous material from the contaminated atmosphere, that is originated from the combustion of diesel and gasoline by automobiles and industrial activity. There are few studies about the black crust formation in granite. During a study performed under artificial exposure atmospheres showed that granite sulphation is even possible at relatively poor SO2 atmospheres (10 ppm), as SO2 can react with Ca from the joint mortar [8]. A more recent study confirmed the contribution of two sulphur sources, i.e. anthropogenic SO2 and marine sulphate, in the development of black crust on coastal granitic constructions [4].
683
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
In general terms, crusts cleaning process must be understood as a soft and delicate extraction procedure intended to release the surface dirt without affecting the stone [9]. The particular characteristics of this crust make not advisable to extrapolate its cleaning results to other crust types [10]. There are many cleaning methods and the selection of the most adequate must be done taking into account a lot of theoretical and practical considerations that include (a) diagnosis of the stone surface (stone kind, extent and severity of decay) and (b) work accessibility and the need of scaffolding. Once the most adequate methods are chosen, they must be essayed on specific areas of the degraded surface, in order to estimate the effectiveness and side effects on the stone (risk evaluation) [11]. Since the 90s, the search for more effective cleaning methods that cause less damage to the stone substrate was increased. There are different cleaning methods, being the mechanical and chemical ones the most traditionally used. Mechanical methods, derived from the old sandblasting method, are based on the projection of different kind of particles. Due to their aggressiveness because the high pressure range used (above 50 bar), they are only used in certain circumstances [11,12]. Among chemical methods, the Papetta AB57 formulation (developed by the Istituto Superiore per la Conservazione ed il Restauro in Rome) has been hailed as an excellent cleaning agent for black crust developed on marble [13]. Laser ablation constitutes another cleaning procedure in heritage conservation which goes back to the 70s with the pioneering work by John Asmus about crust extraction in Venetian marble [14]. There are numerous bibliographic works based on the parameters of the laser systems, the mechanisms involved in the surface cleaning and the different uses of laser in restoration of Cultural heritage [15–17]. Concretely, works about its laboratory and in situ applications are very extensive, especially those related to carbonated stones [17–21]. Almost all the references regarding cleaning procedures to remove black crust are focused on carbonated stones (sedimentary or metamorphic) [18–21], therefore the studies related to granitic stones are scarce. This limited scientific production is centred particularly on laser cleaning [22–25]. Also, up to date, there are no works intending to compare effectiveness information between different cleaning procedures of black crust developed on granite. Therefore, the aim of this work is to enhance the knowledge about the effectiveness of different mechanical, chemical and laser ablation-cleaning procedures, in the removing of sulphated black crust developed on a granite widely used in the construction of architectural heritage in NW of Iberian Peninsula. Taking into account the deep influence of the textural and mineralogical peculiarities of the granites in their durability and in their response to conservation treatments [26,27], the access to specific results on the cleaning effectiveness of different methods of black crust in granites would be of great significance in the direct intervention in granitic cultural heritage.
biological structures. Chemically, besides calcium sulphate, alkane and carboxylictype organic compounds were identified; being probably these compounds the reason of the black colour of the crust [4].
2.2. Cleaning methods The cleaning methods used were the following: 1) Chemical procedures: Following previously reported studies [13,29], it was decided to apply nine different chemical cleaning treatments (see Table 1). All these treatments, except one, consisted on the application of a poultice made on a mixture of a cleaning agent and a thickening compound (Table 1). The selected cleaning agents were: - Standard Papetta AB57 composed of 25 g of EDTA, 50 g of sodium bicarbonate, 30 g of ammonium bicarbonate and 10 cm3 of Neodesogen (a biocide based on benzalkonium chloride with fungicide effect which was provided by BIC Materiales y Conservación S.L.), all dissolved in 1000 mL of distilled water. - Papetta AB57 without ammonium bicarbonate, prepared with 25 g EDTA, 50 g sodium bicarbonate and 10 cm3 of Neodesogen, dissolved in 1000 mL of distilled water. - Acidic mixture: 5% (vol.) HCl and 2.5% (vol.) (NH3)HF2 in aqueous solution. The selected thickeners supplied by CTS Slr (details in www.ctseurope.com) were: - Carbopol Ultrez21 prepared at 3% (wt.) in 10% (wt.) ammonium bicarbonate solution. As Table 1 shows, this thickener was prepared with Papetta AB57 without ammonium bicarbonate. - Carbopol Ultrez21 prepared at 3% (wt.) in 10% (wt.) Ethomeen C25 solution. - Laponite RD, a synthetic clay composed of sodium, lithium and magnesium silicates. - Carboxymethylcellulose as water soluble sodium salt of the cellulose glycolic acid. - Carbogel, a neutralized polyacrylic acid. Eight different poultice based treatments were prepared mixing the above mentioned cleaning compounds and thickeners at different proportions until achieving a gel with good workability. In Table 1, the proportions and acronyms for all these treatments are shown. The ninth chemical treatment consisted on the direct application of Amberlite 4400 OH which was also provided by CTS Slr. (www.ctseurope.com). It is a strong exchange resin. A poultice with a proper consistence was prepared mixing 50 g of Amberlite within 30.7 mL of distilled water (Table 1).
Table 1 Chemical cleaning treatments applied in this study. In the third column, the acronym and the cleaning compound-thickener proportion is indicated. Cleaning agent
Thickener
Treatment Cleaning-thickener proportion
PAPETTA AB57
Laponite RD
Papetta AB57 + Laponite (PL) 50 mL: 14.5 g Papetta AB57 + CMC (PCMC) 50 mL: 2 g Papetta AB57 + Carbogel (PC) 50 mL: 4.5 g Papetta AB57 + Carb + Eth (PCUE) 50 mL: 100 mL
Carboxymethylcellulose
Carbogel
2. Materials and methods
Carbopol Ultrez 21 and Ethomeen C25
2.1. Granite and sulphated black crust samples In order to carry out this study, an ashlar affected by an intense development of sulphated black crust was extracted from an ancient building in the city of Vigo (NW Iberian Peninsula). This building is constructed with a fine grained (2–0.3 mm) equigranular granite, composed of quartz (29%), potassium feldspar (25%), sodium plagioclase (24%), muscovite (13%) and biotite (4%) as main minerals [28]. A 100 cm 100 cm 2 cm plate where cut parallel to the crusted surface of the ashlar, in order to obtain more manageable smaller stone 29 slabs of 14 cm 7 cm. Previously to the cleaning process, the black crust was characterized following different techniques. Results of this characterization can be consulted in [4]: briefly, the sulphated black crust comprised a coating of 80–100 lm of thickness composed of acicular-shape crystals of calcium sulphate with occasionally
PAPETTA AB57 without NH4HCO3
Carbopol Ultrez 21 with NH4HCO3
Papetta AB57 + Carb + Bic (PCUB) 50 mL: 100 mL
ACIDIC MIXTURE
Laponite RD
Acidic + Laponite (AL) 50 mL: 19 g Acidic + CMC (ACMC) 50 mL: 3 g Acidic + Carbogel (AC) 50 mL: 4 g
Carboxymethylcellulose Carbogel AMBERLITE 4400 OH
(AM)
684
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
After mixing, 85 cm3 of each of the ninth poultice based treatments were applied on the surface of the stone slabs, completely covering the sulphated black crust and using three slabs for each chemical treatment. A portion of 2 7 cm2 of surface of one of the slabs was not treated therefore providing a not cleaned reference surface. After two hours (time required for a total drying of the poultice), the poultices were mechanically removed, the remains of the poultices dragged with the aid of an absorbent cellulose paper and the cleaned surfaces rinsed with 1:1 (vol.) waterethanol solution with the help of a short bristle brush. This procedure (the poultice application, the removal of the poultice after 2 h and the neutralization of the cleaners) was repeated three times more. After the fourth application and removal, the slab without the two-centimetre control area was subjected to a final neutralization with water vapour. 2) Mechanical cleaning was performed using the patented procedure called HydrogommageÒ. This is a procedure based on a low-pressure projection (0.5–1.5 bars) with a pneumatic rotator of an air-water-micro granule mixture. The abrasive, of grain size less than 80 lm, was composed of 99% of SiO2 and 1% of Al2O3, K2O, TiO2 and Fe2O3. 3) The laser cleaning was performed using a Nd:YVO4 laser (Coherent AVIA Ultra 355–2000) with 355 nm wavelength and 25 ns pulse duration. The laser output intensity profile was Gaussian TEM00 and the beam diameter at 1/e2 intensity level was about 2.2 mm. Pulse repetition rate can be changed from a single-shot to 100 kHz with energy per pulse around 0.1 mJ. The beam impinges perpendicularly onto the target sample surface, which was placed on a motorized XYZ-translation stage Newport ILS-CC. Different distances lens-sample had to be applied to avoid the influence of the high roughness, because the stone surface showed crests and valleys of around 1 mm. 2.3. Analytical techniques After the cleaning of the slabs by the different methods, the following analyses were applied. Stereoscopic zoom microscope (SMZ800 Nikon) was applied in order to get an initial qualitative evaluation of the effectiveness of the cleaning methods, considering for this the presence of crust remains and the recovery of the original colour of the granite. Also, the occurrence of residues of the cleaning products has been evaluated in this phase. Observation under ultraviolet light was performed in order to detect the presence of remaining residues. Cross-sections and superficial samples were observed using Scanning Electron Microscope (SEM) (Philips XL30 and JEOL JSM-6700) and also analysed by means of X-ray microanalysis (EDX). Xray diffraction (XRD) (SIEMENS D-5000) and Fourier Transformation Infrared Spectroscopy (FTIR) were used to evaluate the presence of by-products or cleaner remains on the surfaces. FTIR data were collected in reflection mode using a FTIR Nicolet Continuum (Thermo), analysing the cleaned surfaces and, for comparative purposes, the not-cleaned surface of the black crust and the granite without black crust (using the face of the slab which was opposite to that affected by the crust); data were transformed to absorption spectra. In order to assign the FTIR bands, different references were consulted [30–32]. Colour characterization before and after the cleaning treatments was made by means of a Minolta CM-700d spectrophotometer, expressing colour data in CIELAB colour space. Colourimetric differences DL⁄, Da⁄, Db⁄, DC⁄ab and DH⁄ and global colour change (DE⁄ab) were calculated taking the colour data for the sulphated black crust as reference colour. Analysis of variance was applied to colour data and Tukey post hoc test (or multiple comparison Tukey test) was used to determine the significant differences between the colour data measured after and before cleanings. Additionally, colour changes after cleaning were qualitatively analysed using scatter plots representing raw data for the granite with black crust, for the granite after the cleaning and also for the granite without crust (obtaining the colour data of the opposite surface to that affected by the crust). As chemical cleanings were performed in four applications, colour measurements were also taken after each application, in order to evaluate the effectiveness of each chemical procedure in terms of number of applications needed to achieve optimal results.
3. Results The first qualitative evaluation of the effectiveness by means of optical microscopy revealed none of the procedures applied (chemical, mechanical and laser ablation) was able to completely extract the sulphated black crust: after cleaning, the surface of the stone became lighter but with higher magnifications, gypsum crystals remains over the surface can be seen, in a greater or a lesser amount depending on the cleaning procedure (Fig. 1). Therefore, cleaning procedures seemed to be more effective in removing the carbonaceous fraction of the crust (which confers the dark colour) than in gypsum removal (which was the main compound of the crust).
For comparative proposes, in Fig. 1, the micrographs of the uncoated granite and the black crust are shown (Fig. 1A and B respectively).Considering this circumstance, it was possible to define a scale of effectiveness among the procedures applied (Table 2): the cleaning methods which showed a higher effectiveness were the chemical procedures PCUE, ACMC and AM (Fig. 1, C and D). Although, either of the chemical cleaning procedures has shown a higher gypsum removal effectiveness than that obtained for laser ablation or mechanical procedure, the cleaning levels obtained for chemical procedures were hampered by the occurrence of remains on surface associated with the chemical thickeners (compare Fig. 1, C–E with Fig. 1, F); these surface residues were found in smaller amounts when water vapour was used in order to neutralize the chemical cleanings (Fig. 1, G and H). Regarding chemical procedures and taking into account the different cleaning levels evaluated (high, medium and low) (Table 2), no relationship between a high cleaning level and the use of a specific thickener can be identified. Nevertheless, an apparently higher effectiveness in the use of acidic mixture, as cleaning agent, seemed to be noticed. Ultraviolet light observations confirmed the presence of organic residues on the surfaces subjected to chemical cleaning and also the fact that these residues were less abundant in the samples cleaned with chemical methods and subsequently neutralized with water vapour (Fig. 1, G and H). The advantage of the application of water vapour was especially evident in the case of Papetta AB57 with Laponite (Fig. 1, H). As commented above, either of the chemical cleaning procedures has shown a higher effectiveness than that obtained for laser ablation or mechanical procedure. After laser cleaning, it was observed that the black colour of the crust was eliminated, but not the gypsum which remained on the surfaces as a continuous and massive coating. Regarding Hydrogommage, after the cleaning, gypsum remains were observed all over the surface. SEM-EDX analysis confirmed the optical microscopy observations regarding the cleaning effectiveness of the different procedures: 1) None of the cleaning procedures achieved a completely gypsum extraction, because EDX detected the presence of gypsum through S and Ca-signals on all the cleaned surface. The scale of effectiveness using SEM-EDX observations was coincident with optical microscopy. Chemical procedures were more effective than Hydrogommage and laser in terms of gypsum extraction (compare Fig. 2, A–F with Fig. 2, G and H). Among the chemicals, the PCUE (Fig. 2, A) was the most effective in removing the black crust, the PCUB (Fig. 2, B) and AL were the procedures which gave the worst results and ACMC (Fig. 2, E) and AM (Fig. 2, F) achieved intermediate results. 2) All chemically cleaned surfaces showed residues and byproducts related to the chemical reagents all over the surfaces. On samples cleaned with the thickener CMC (PCMC treatment, Fig. 2, C and D), residues rich in C and Ca were identified, suggesting the precipitation of calcium carbonate derived from the gypsum dissolution and a later reaction with the sodium bicarbonate of the Papetta AB57 formulation. Also, deposits rich in Na, Cl, Mg and S were identified in the samples cleaned using the thickener Laponite. Ca, S and F rich deposits were found on the samples cleaned with acidic mixture adding Carbogel (treatment AC), being F associated with the acidic nature of the cleaning compound, rich in HF. Finally after cleaning with AM treatment, the difficult of removing the Amberlite from the surfaces was
685
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
A
B
C
D
2000 µm
2000 µm
F
E
2000 µm
H
G Not neutralized
Neutralized
Not neutralized
Neutralized
Fig. 1. A–B: Micrographs of the uncoated granite and the black crust respectively. C–F: Micrographs taken with optical microscope of the cleaned surfaces with different methods. C: cleaned with PCUE (Papetta AB57 with Carbopol Ultrez21 and Ethomeen C25). D: cleaned with AM (Ambelite 4400 OH). E: cleaned with PL (Papetta AB57 with Laponite). F: cleaned with 355 nm Nd:YVO4 laser (cleaned area pointed out with a rectangle). G: Photograph taken under UV light of the slabs cleaned with PCUE (Papetta AB57 wth Carbopol Ultrez21 and Ethomeen C25) treatment. H: Photograph taken under UV light of the slabs cleaned with PL (Papetta AB57 with Laponite) treatment. In these images (G and H), remains of chemical compounds are observed in bright colour. Also, on right side, the slab that was neutralized with water vapour after cleaning. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
corroborated by means of the SEM-EDX observations, due to the presence of a crust composed of this product (Fig. 2, F). Finally, cellulose fibres were found on almost all the surfaces cleaned with chemical methods (Fig. 2, D). 3) Laser and Hydrogommage didn’t achieve better results in terms of gypsum extraction (Fig. 2G and H). After cleaning with Hydrogommage, surfaces shown a large amount of gypsum crystals with acicular habit (Fig. 2, G). Also, SEMEDX observations of samples cleaned with laser allowed to determine that laser didn’t ablate the gypsum crust: surfaces appeared covered with massive gypsum granules; this fact
suggests a partial melting of this mineral (Fig. 2, H). It must also be noted that samples cleaned with laser showed a partial melting of the biotite crystals (Fig. 2, H). XRD (Table 3) confirmed the presence of gypsum remains after the cleanings in variable amounts (never above 10%), confirming that none of the cleaning procedures allowed a complete gypsum elimination. XRD results indicated that, in terms of the amount of gypsum remains after cleaning, the most effective treatments would be the chemical methods using either CMC or Laponite as thickeners. This finding, which didn’t match with the observations
686
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
Table 2 Cleaning level evaluation by means of optical microscopy of the different cleaning treatments. The occurrence of residues on the surfaces after the cleanings is also shown. Treatment
Clean level High
PL PCMC PC PCUE PCUB AL ACMC AC AM Nd:YVO4 laser Hydrogommage
Medium
Occurrence of residues Low
X X X
X X
X X X X
X X
X X
X X X
by means of optical microscopy and SEM-EDX, would indicate that, besides gypsum remains, the presence of residues of different nature influences on the qualitative evaluation of the cleaning level. Fig. 3 shows the FTIR spectra of the surfaces cleaned with chemical treatments (those that produced the best, the intermediate and the worst results by means of optical microscopy and SEM-EDX: PCUE, PCMC and AL treatments respectively), mechanical treatment (Hydrogommage) and laser cleaning. For comparative purposes, spectra of the sulphated black crust and of the granite without crust are also presented. The spectrum of the granite without crust is characterized by the functional SiAO group bands at around 785 cm 1 and between 1020 and 1168 cm 1, which correspond to the different chemical environments characteristic of the various minerals phases (quartz, muscovite, biotite, feldspar, etc.) [31,32]. In sulphated black crust FTIR spectrum, gypsum is identified through the absorbance bands occurring at around 667 cm 1, 1619 cm 1, 1682 cm 1, 3492 cm 1 and 3525 cm 1. The typical band that this mineral shows at 1122 cm 1 is overlapped by signals from the SiAO bond of the minerals in the stone. FTIR data for cleaned surfaces confirmed the results obtained using optical microscopy and SEM-EDX: gypsum bands were detected in almost all the samples; these bands were more evident in samples evaluated with a medium to low degree of cleaning (Table 2). In the spectra from surfaces cleaned with the chemical preparations, bands attributed to gypsum at around 667 cm 1, 1619 cm 1, 1682 cm 1, 3492 cm 1 and 3525 cm 1 were identified; these bands were more intense in surfaces cleaned with AL (one of the chemical procedures which gave worst results); they also appeared with less intensity in the spectrum of the surface cleaned with Hydrogommage but they were absent in laser cleaned surfaces. This indicates the low presence or absence of gypsum on the laser cleaned surfaces. In the surfaces cleaned with chemical procedures, bands which would indicate the presence of remains of the organic compounds of the cleaning agents and thickeners can be detected; in these cases, the crust extraction effectiveness was overshadowed by the band intensity corresponding to the chemical preparations. In the spectra of surfaces cleaned with PCMC and PCUE (Fig. 3), several bands that can be attributed to the cleaning agent residuals, were detected, i.e. bands at 680 cm 1 and 832 cm 1 were identified as from the benzene CAH functional group (corresponding to the Neodesogen benzil radicale). In the surfaces cleaned with PCMC also between 1400 and 1590 cm 1 there were bands associated with the CAO functional group; bands characteristic of the tertiary amines (EDTA) can be detected at the region at 1739 cm 1 and at 2300–2622 cm 1. Most notable was the finding of residuals from the CMC thickener, as demonstrated by the absorption band
detected at 1323 cm 1 – typical of the CAOAC functional group – and the shoulder-type peak at 3400 cm 1, assigned to the cellulose. Attending to these results, in the surface cleaned with PCUE, the effects attributed to the cleaning agents were lesser than those detected in surfaces cleaned with PCMC (Fig. 3). The crust extraction effectiveness was also evaluated by means of colour data. Table 4 shows the colourimetric differences obtained after the cleaning with each procedure. For chemical procedures, colour differences after each application are also shown. In Fig. 4, L⁄-C⁄ab scatter plots shows the raw colour data from the cleaned surfaces with different procedures and the colour of the granite with and without crust. It was established as a criterion for the best cleaning, the higher the global colour change (DE⁄ab), the more effective in removing the crust. So, in Table 4 it can be seen that the greater effectiveness was achieved by Hydrogommage (DE⁄ab: 24.98), laser (DE⁄ab: 20.90), AM (DE⁄ab: 20.62) and PCMC (DE⁄ab: 19.30). Recall that the latter was also considered as one of the most effective by microscopy and after FTIR analysis. In the opposite situation, preparations that produced the lower global colour change were the PCUB (DE⁄ab: 15.44) and AL (DE⁄ab: 8.46) In Table 4 it can be seen that lightness, L⁄, was the colourimetric coordinate that most contributed to the global colour change in all cases; coordinates a⁄ and b⁄ (especially the a⁄ coordinate) showed smaller changes. Concerning the chemical cleaning methods, it was clearly observed that the global colour change (DE⁄ab) increased with the number of applications, indicating a progressive extraction of the crust. When comparing among the different applications, it can be observed that with AM and AC, the maximum DE⁄ab value was almost achieved at the second application, suggesting a higher cleaning effectiveness. In Fig. 4, colour raw data of each cleaned surface, of the sulphated black crust and of the original colour of the granite are represented. It can be seen that none of the treatments allowed restore the original colour to the stone. Despite this fact, in the case of chemical treatments made with the acidic mixture (Fig. 4, A), those which were evaluated as the most effective in removing the crust by microscopy and FTIR (ACMC, AM) were those whose colours fell closer to the original colour of the stone. These same methods caused the highest DE⁄ab (Table 4). A same behaviour was not found, however, for the chemical treatments based on Pappeta AB57 (Fig. 4, B) or after cleaning with laser and Hydrogommage (Fig. 4, C). Thus, the colour after the cleaning with the PCUE treatment (evaluated as more effective) fell quite far from the original colour of the stone, especially as regards the C⁄ab. The cleaning with Hydrogommage showed a low efficacy in removing the crust; however it produced a very high DE⁄ab but that was not reflected in a higher recovery of the original colour of the stone. The cleaning with Laser and Hydrogommage produced a very high global colour change (DE⁄ab: 20.90 and 24.98 respectively). Nevertheless, it was observed that such high DE⁄ab values didn’t imply a closer approach of the colour of the cleaned surfaces to the original stone colour (Fig. 4, C).
4. Discussion The techniques applied to evaluate the cleaning effectiveness of the sulphated black crust with different methods (chemical, Hydrogommage and laser) in the studied granite confirmed the difficulty to successfully remove this crust. After the cleaning, gypsum was present on all the cleaned surfaces as corroborated by SEM-EDX, XRD and FTIR. Nevertheless, by the techniques applied, it was possible to identify those procedures which allowed a high cleaning level without producing damage to the stone (such as
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
687
Fig. 2. SEM micrographs of the granite surfaces after sulphated black crust cleaning. A: PCUE (Papetta AB57 wth Carbopol Ultrez21 and Ethomeen C25) treatment. B: PCUB (Papetta AB57 with Carbopol Ultrez21 and ammonium bicarbonate). C: PCMC (Papetta AB57 with CMC). D: PCMC with more detail. E: ACMC (Acidic mixture with CMC). F: AM (Ambelite 4400 OH). G: Hydrogommage. H: 355 nm Nd:YVO4 laser.
cleaning residues or mineralogical and textural alterations). These chemical methods were PCUE (using as cleaning agent Pappeta AB57) and ACMC (using as cleaning agent the acidic mixture). The treatment which gave the worst results were PCUB (using as cleaning agent Pappeta AB57). The cleaning effectiveness of chemical products was weakened by the presence of surface residuals associated to the chemical preparations.
SEM-EDX analyses confirmed that after the cleaning with all the chemical methods, residues of different nature remained on the stone surfaces, derived from the cleaning and thickener agents. Also, by-products of the reaction between the cleaning agents and the gypsum of the crust were observed, as was the case of calcium carbonate possibly formed by the reaction with sodium bicarbonate of the Pappeta AB57.
688
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
Table 3 Semiquantitative results of the XRD of the sulphated black crust and the surfaces after their cleaning with different procedures. Q: quartz; P: plagioclase; F-K: potassium feldspar; M: muscovite; B: biotite; G: gypsum. + + + +: >50%; + + +: 30–50%; + +: 10– 30%; +: 3–10%; tr (traces): <3%, N.D.: no detected. Treatment
Q
Black crust PL PCMC PC PCUE PCUB ACMC AL AC AM Nd:YVO4 laser Hydrogommage
+++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++
+ +
+ +
P
F-K
M
B
G
++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++
++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++
+ ++ ++ + ++ ++ + ++ N.D ++ ++ N.D.
N.D N.D N.D N.D N.D N.D N.D N.D + N.D N.D N.D.
+ tr tr + + + tr tr + + + +
Fig. 3. From bottom to top, FTIR spectra of granite without black crust, granite with black crust, surface cleaned with laser, surface cleaned with Hydrogommage and surfaces cleaned with AL (Acidic mixture with Laponite), PCMC (Papetta AB57 with CMC) and PCUE (Papetta AB57 wth Carbopol Ultrez21 and Ethomeen C25).
Table 4 Colourimetric differences DL⁄, Da⁄, Db⁄, DC⁄ab, DH⁄ and global colour change (DE⁄ab) after the cleaning of the sulphated black crust with the different procedures (chemicals, Hydrogommage and laser). Differences were computed relative to the crust colour, so the greater the difference the higher the cleaning effectiveness. For chemical methods, colour data for each of the four applications are shown. Ap. N°: application number. *: Significant differences (n.s. < 0.05) comparing to the crust colour data. Treatment
Ap. N°
DL ⁄
Da⁄
Db ⁄
DC⁄ab
DH⁄
DE⁄ab
PL
1 2 3 4
2.62⁄ 9.65⁄ 2.75⁄ 17.50
0.83⁄ 0.48⁄ 1.38⁄ 0.42⁄
1.88⁄ 1.15⁄ 4.11⁄ 1.70⁄
2.02⁄ 1.22⁄ 4.32⁄ 1.75⁄
2.43⁄ 1.69⁄ 1.92⁄ 0.19
3.33 9.73 5.13 17.58
PCMC
1 2 3 4
6.94⁄ 12.57 16.04 18.96
1.11⁄ 0.81⁄ 0.83⁄ 0.95⁄
3.49⁄ 2.82⁄ 3.15⁄ 3.52⁄
3.65⁄ 2.92⁄ 3.25⁄ 3.64⁄
1.61⁄ 1.06 0.79 0.84
7.85 12.90 16.37 19.30
PC
1 2 3 4
4.71⁄ 11.14 15.70 16.43
0.42⁄ 0.05 0.06 0.41⁄
2.28⁄ 0.90⁄ 0.38 1.03⁄
1.35⁄ 0.89⁄ 0.36 1.09⁄
2.82⁄ 2.14⁄ 1.85⁄ 1.51⁄
5.25 11.18 15.70 16.46
PCUE
1 2 3 4
3.22⁄ 6.44⁄ 8.35⁄ 14.41
1.20⁄ 1.71⁄ 1.53⁄ 1.98⁄
2.81⁄ 4.14⁄ 4.01⁄ 5.99⁄
3.02⁄ 4.45⁄ 4.27⁄ 6.29⁄
2.87⁄ 3.39⁄ 2.88⁄ 2.44⁄
4.44 7.84 9.39 15.73
PCUB
1 2 3 4
7.35⁄ 10.06 4.02⁄ 15.22
0.22 0.04 0.86⁄ 0.42⁄
0.25 0.64 3.15⁄ 2.53⁄
0.30 0.61 3.26⁄ 2.55⁄
0.97 1.10 0.77⁄ 0.82
7.36 10.08 5.18 15.44
AL
1 2 3 4
2.62⁄ 3.75⁄ 2.06⁄ 7.29⁄
0.77⁄ 1.05⁄ 1.71⁄ 1.36⁄
1.99⁄ 2.92⁄ 4.44⁄ 4.08⁄
2.11⁄ 3.09⁄ 4.74⁄ 4.29⁄
1.78⁄ 1.93⁄ 0.07 1.98⁄
3.38 4.87 5.19 8.46
ACMC
1 2 3 4
8.94⁄ 14.16 20.90 23.25
0.67⁄ 1.02⁄ 0.09⁄ 0.78⁄
1.64⁄ 2.62⁄ 0.02⁄ 2.53⁄
1.75⁄ 2.79⁄ 0.00⁄ 2.64⁄
2.04⁄ 2.42⁄ 0.03⁄ 1.47⁄
9.11 14.44 20.90 23.40
AC
1 2 3 4
11.03 11.96 13.51 15.38
0.40⁄ 0.41⁄ 0.59⁄ 0.56⁄
1.14⁄ 1.25⁄ 1.87⁄ 1.68⁄
1.20⁄ 1.31⁄ 1.95⁄ 1.76⁄
1.08 0.95 0.03⁄ 1.38⁄
8.49 12.03 13.65 15.49
AM
1 2 3 4
8.71⁄ 19.12 17.33 20.18
0.90⁄ 0.65⁄ 1.04⁄ 1.13⁄
1.65⁄ 1.36⁄ 3.10⁄ 4.04⁄
1.83⁄ 1.48⁄ 3.26⁄ 4.19⁄
3.28⁄ 2.61⁄ 0.04⁄ 1.20⁄
8.92 19.18 17.63 20.62
Nd:YVO4 laser Hydrogommage
1 1
20.82 23.90
0.92⁄ 2.31⁄
1.61⁄ 6.87⁄
1.79⁄ 7.23⁄
4.07⁄ 2.90⁄
20.90 24.98
FTIR spectra also helped to verify the permanence on the stone surfaces of residuals from the chemical preparations after the cleanings. They also were very useful to evaluate the extraction effectiveness: inexistent or low intensity bands assigned to gypsum were found on the surfaces cleaned with PCUE and ACMC, the two methods which were identified as the most effective by means of optical microscopy. The residues related to the chemical and thickener compounds in the case of the chemical treatments can be avoided by the application of water vapour after the cleaning. A fact to note is that, among the residues, many cellulose fibres have been detected and not only after the cleaning with the treatments based on CMC, but also after those which do not use it. These fibres come from the absorbent cellulose paper used to neutralize the cleaners. So, it is recommended the use of an alternative absorbent material for the mechanical elimination of the cleaner compounds. When using the colour data, it was done taking in consideration that the higher global colour changes (DE⁄ab) the better extraction effectiveness [33]. After cleaning, colour data indicated that all chemical treatments except AM needed four applications to reach
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
A AL ACMC AC AM
*
B PL PCMC PC PCUE PCUB
689
the PCUB, also in accordance with other evaluation techniques. Nevertheless, considering all the cleaning procedures (chemicals, Hydrogommage and laser ablation), the global colour change would point out the Hydrogommage as the most effective method, in disagreement with the other evaluation techniques. This lack of coincidence between the DE⁄ab, the evaluation under microscopy and FTIR and the degree of recovery of the original colour suggested that the cleaning treatments by themselves produced a colour modification of the rock, regardless of its effectiveness in removing the crust. This change may be related to the existence of remains on the surface of the stone after chemical cleaning, or mineralogical or textural changes caused by mechanical method and the laser. This leads to the need to consider the DE⁄ab variations and but also the evaluation of the degree of recovering the original colour for a correct effectiveness evaluation of the treatments. The very low effectiveness of the laser used in this work, a nanosecond Nd:YVO4 at 355 nm, in the removing of the sulphated black crust from the granitic surface must be highlighted. Previous works reported that the complete laser extraction of sulphated black crust of carbonated sedimentary stones and marble was not achieved [18,19,33], but a high effectiveness can be obtained with the third harmonic or in combination with [18]. In those works, Nd:YAG laser was applied to clean sulphated black crusts on homogeneous substrates from the mineralogical and textural point of view. In this work, the heterogeneous surface of the granite increases the gypsum removal difficulty as was concluded in Nd:YAG at 1064 nm cleaning of sulphated black crust [25].
5. Conclusions
*
C
* Fig. 4. L⁄ and C⁄ab scatter plots of representing raw colour data from A) surfaces cleaned with Acidic mixture with different thickeners and Amberlite. B: Surfaces cleaned with Papetta AB57 with different thickeners. C: Surfaces cleaned with laser and Hydrogommage. Plots also include colour values for the granite with and without sulphated black crust. For chemical procedures, data after the fourth application are shown. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
the maximum DE⁄ab value; however with AM the maximum DE⁄ab was reached with only two applications. Among the chemicals, the most effective cleanings, following colour data, were those performed with the PCMC and ACMC, being the latter considered, by means of the other techniques, as one of the most effective. According to the DE⁄ab the less effective chemical procedure was
From the results of this study, focused on the evaluation of the effectiveness of different cleaning procedures in removing sulphated black crust developed on granite, the following conclusions can be drawn: 1. None of the reported treatments completely removed the sulphated black crust. Nevertheless, a ranking of effectiveness have been defined, that may have immediate applications in conservation practice. So, all the techniques applied in this study, except colour data evaluation, agreed that the chemical methods were more effective than the mechanical method (Hydrogommage) and the use of a nanosecond Nd:YVO4 laser at 355 nm. The treatment that showed the higher effectiveness was that based on the Papetta AB57 with Carbopol Ultrez 21 and Ethomeen C25 as a thickeners (PCUE treatment). 2. All the evaluated chemical cleaning procedures left residues on the surface of the granite, which were identified as 1) residues of the cleaning and thickeners agents, 2) by-products from the reaction between the crust and chemical reagents and 3) remains of the absorbent paper used in the neutralization. Thus, to increase the global effectiveness of the granite cleaning interventions with chemical products, it is recommended the optimization of the neutralization process, avoiding the use of absorbent cellulose paper and applying water vapour on the surfaces after the treatment. 3. The cleaning with a 355 nm Nd:YVO4 laser led to biotite melting. Also, the mechanical method, Hydrogommage, increased the roughness of the stone. So, these methods need to be examined in order to optimize the cleaning parameters (in the case of laser) or the design (such as the spray pressure or the grain size of the abrasive, in the case of Hydrogommage), thereby increasing their efficiency and reducing the damage to the granite. 4. Because all cleaning methods caused in a greater or a lesser extent harmful effects on the stone (remains of chemical cleaners, textural changes in the mechanical method and textural
690
J.S. Pozo-Antonio et al. / Construction and Building Materials 112 (2016) 682–690
and mineralogical changes after the laser cleaning) that may contribute to a change in colour, the use of global colour change as a quantitative indicator of the removal degree of the crust should be undertaken with caution. A more appropriate way, in these cases, is also consider the qualitative assessment of the degree of recovery of the original colour of the stone. For these reason, besides the global colour change, it was necessary to consider the L⁄-C⁄ab scatter plots of representing raw colour data. 5. In order to achieve a complete removal of the sulphated crusts developed on granite, future works will be focused on optimizing the laser cleaning (combining different energy levels and different wavelengths or lasers) and other chemical products and mechanical procedures will be used.
Acknowledgements This work was supported by the Spanish project CTM201019584 funded by Ministerio de Ciencia y Tecnología. M.P. Fiorucci’s work was financed by the project: A-TEMPO Avances en tecnologías marinas: producción naval y Offshore. Plataforma robótica subacuática de inspección para el mantenimiento de estructuras Offshore. SEM-EDX, XRD and FTIR analysis were performed at the Centro de Apoyo Científico y Tecnológico a la Investigación – CACTI University of Vigo. The authors would like to thank Proliser S.L., ClinArte S.L. and Restoration and Conservation School of Galicia (Pontevedra), concretely Cristina Montojo and Fernando Carrera. J.S. Pozo-Antonio was supported by a postdoctoral contract with the University of Vigo within the framework of the 2011–2015 Galician Plan for Research, Innovation and Growth (Plan I2C) for 2014. References [1] P. Ausset, J.L. Crovisier, M. Del Monte, V. Furlan, F. Girardet, C. Hammecker, D. Jeannette, R.A. Lefevre, Experimental study of limestone and sandstone sulphation in polluted realistic conditions: the Lausanne Atmospheric Simulation Chamber (LASC), Atmos. Environ. 30 (1996) 3197–3207. [2] B.J. Smith, P.A. Warke, Processes of urban stone decay, in: Proceedings of SWAPNET ’95, Donhead Publishing, London, 1996, p. 274. [3] N. Schiavon, Biodeterioration of calcareous and granitic building stones in urban environments, in: S. Siegesmund, T. Weiss, A. Vollbrecht (Eds.), Natural Stone, Weathering Phenomena, Conservation Strategies and Case Studies, Geological Society of London, London, 2002, pp. 195–205. [4] T. Rivas, S. Pozo, M. Paz, Sulphur and oxygen isotope analysis to identify sources of sulphur in gypsum-rich black crusts developed on granites, Sci. Total Environ. s482–483 (2014) 137–147. [5] A.E. Charola, R. Ware, Acid deposition and the deterioration of stone: a brief review of a broad topic, Geol. Soc. Spec. Publ. 205 (2002) 393–406. [6] S. Siegesmund, J. Ruedrich, A. Koch, Marble bowing: comparative studies of three different public building facades, Environ. Geol. 56 (2008) 473–494. [7] J. Simão, E. Ruiz-Agudo, C. Rodríguez-Navarro, Effects of particulate matter from gasoline and diesel vehicle exhaust emissions on silicate stones sulfation, Atmos. Environ. 40 (2006) 6905–6917. [8] T. Rivas, B. Prieto, B. Silva, Gypsum formation in granitic rocks by dry deposition of sulphur dioxide, in: Proceedings 4th International Symposium on the Conservation of Monuments in the Mediterranean Basin, Rhodes, 1997, pp. 263–270. [9] E. Doehne, C.A. Price, Stone Conservation: An Overview of Current Research, second ed., The Getty Conservation Institute, Los Ángeles, USA, 2010. p. 158.
[10] B.J. Smith, M. Gómez-Heras, S. McCabe, Understanding the decay of stone-built cultural heritage, Prog. Phys. Geogr. 32 (2008) 439–461. [11] R. Bellmunt, A. Paricio, N. Vila, Reconocimiento, diagnosis e intervención en fachadas, Institut de Tecnología de la Construcció de Cataluña, Spain, 2002, p. 208. [12] P. Gaspar, C. Hubbard, D. McPhail, A. Cummings, A topographical assessment and comparison of conservation cleaning treatments, J. Cult. Heritage 4 (2003) 294–302. [13] L. Lazzarini, M.L. Tabasso, Il restauro della pietra, CEDAM-Casa Editrice Dott. A, Milani, Padova, 1986. p. 334. [14] J.F. Asmus, G. Guattari, L. Lazzarini, G. Musumeci, R.F. Wuerker, Holography in the conservation of statuary, Stud. Conserv. 18 (1973) 49–63. [15] M.I. Cooper, D.C. Emmony, J. Larson, Characterization of laser cleaning of limestone, Opt. Laser Technol. 27 (1995) 69–73. [16] C. Fotakis, D. Anglos, V. Zafiropoulos, S. Georgiou, V. Tornari, Lasers in the Preservation of Cultural Heritage. Principles and Applications, Taylor&Francis, London, 2006, p. 336. [17] I. Osticioli, M. Mascalchi, D. Pinna, S. Siano, Removal of Verrucaria nigrescens from Carrara marble artefacts using Nd:YAG lasers: comparison among different pulse durations and wavelengths, Appl. Phys. A: Mater. Sci. Process. 118 (2015) 1517–1526. [18] S.S. Potgieter-Vermaak, R.H.M. Godoi, R. Van Grieken, J.H. Potgieter, M. Ouija, M. Castillejo, Micro-structural characterization of black crust and laser cleaning of building stones by micro-Raman and SEM techniques, Spectrochim. Acta, A 61 (2005) 2460–2467. [19] M. Iglesias, J.L. Prada, N. Guash, Technique for cleaning Tarragona Miocene age dolomitized silty limestone, altered by urban pollution, Mater. Constr. 58 (2008) 247–262. [20] A. Khedr, V. Papadakis, P. Pouli, D. Anglos, M.A. Harith, The potential use of plume imaging for real-time monitoring of laser ablation cleaning of stonework, Appl. Phys. B 105 (2011) 485–492. [21] P. Ortiz, V. Antúnez, R. Ortiz, J.M. Martín, M.A. Gómez, A.R. Hortal, B. MartínezHaya, Comparative study of pulsed laser cleaning applied to weathered marble surfaces, Appl. Surf. Sci. 283 (2013) 193–201. [22] R.M. Esbert, C.M. Grossi, A. Rojo, F.J. Alonso, M. Montoto, J. Ordaz, M.C. Pérez de Andrés, C. Escudero, M. Barrera, E. Sebastián, C. Rodríguez-Navarro, Application limits of Q-switched Nd:YAG laser irradiation for stone cleaning based on colour measurements, J. Cult. Heritage 4 (2003) 50s–55s. [23] C.M. Grossi, F.J. Alonso, R.M. Esbert, A. Rojo, Effect of laser cleaning on granite color, Color Res. Appl. 32 (2007) 152–159. [24] A. Ramil, A.J. López, M.P. Mateo, C. Álvarez, A. Yáñez, Colour changes in Galician granitic stones induced by UV Nd:YAG laser irradiation, in: Lasers in the Conservation of Artworks – Proceedings of the International Conference LACONA 7, 2008, pp. 199–202. [25] S. Pozo, P. Barreiro, T. Rivas, P. Gonzalez, M.P. Fiorucci, Effectiveness and harmful effects of sulphated black crust removal from granite using Nd:YAG nanosecond pulsed laser, Appl. Surf. Sci. 302 (2014) 309–313. [26] C. Cardell, T. Rivas, M.J. Mosquera, B. Prieto, J.M. Birginie, B. Silva, A. Moropoulou, R. Van Grieken, Patterns of damage in igneous and sedimentary rocks under conditions simulating sea-salt weathering, Earth Surf. Proc. Land. 28 (2003) 1–14. [27] T. Rivas, B. Prieto, B. Silva, Permeability to water vapour in granitic rocks. Application to the study of deleterious effects on conservation treatments, Build. Environ. 36 (2000) 239–246. [28] IGME (Instituto Geológico y Minero de España), Mapa geológico de España. Serie Magna, E1:50.000, second ed. Sheet 223., 1985 (Vigo). [29] T. Rivas, J. Iglesias, J. Taboada, J.A. Vilán, Sulphide oxidation in ornamental slates: protective treatment with siloxanes, Mater. Constr. 61 (2011) 115–130. [30] G. Socrates, Infrared and Raman Characteristic Group Frequencies: Tables and Charts, third ed., John Wiley & Sons, 2001. p. 366. [31] P.A. Schroeder, Infrared spectroscopy in clay science, in: A. Rule, S. Guggenheim (Eds.), Teaching Clay Science, The Clay Mineral Society, Aurora, CO, 2002, pp. 181–206. [32] P. Makreski, G. Jovanovski, B. Kaitner, Minerals from Macedonia. XXIV. Spectra-structure characterization of tectosilicates, J. Mol. Struct. 924–926 (2009) 413–419. [33] E. Gioventù, P.F. Lorenzi, F. Villa, C. Sorlini, M. Rizzi, A. Cagnini, A. Griffo, F. Cappitelli, Comparing the bioremoval of black crusts on colored artistic lithotypes of the Cathedral of Florence with chemical and laser treatment, Int. Biodeterior. Biodegradation 65 (2011) 832–839.