Open Access
Add to BibSonomy
Abstract
We explore the use of femtosecond laser pulses to clean a variety of colors of spray paint from the Moruya granite, a stone with high heritage value that is widely used for monuments and sculptures in Sydney and New South Wales (Australia). The efficiency of the cleaning treatment and the effects on the stone substrate are evaluated using optical microscopy, optical profilometry, Raman spectroscopy, energy-dispersive X-ray spectroscopy, and colorimetry. We demonstrate that femtosecond laser cleans granite without damaging it and without discoloration when the laser fluence is set below the damage threshold of the stone.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Graffiti is an increasing problem affecting many protected stone monuments and buildings across the world. The presence of graffiti in public and historic places may reduce their visual impact and enjoyment, and create irreparable damage, resulting in major resource expenditure to clean and protect these sites. Spray paint is a frequently used medium because it is readily available in urban areas, dries quickly and is easily applied to a wide range of substrates. Typical spray paints are made of a pigment, a binding medium in which the pigment is dispersed, a solvent that allows the mixture to flow on the substrate, and additives to improve paint performance, including its plasticity, thickness and UV resistance [1,2].
The removal of graffiti is a highly expensive endeavour, costing city councils and municipalities millions of dollars every year, even though complete cleaning is not always possible [1]. As an example, the Australian government spends annually over 1.5 billion dollars in remediating criminal damage, which includes the removal of spray-painted graffiti [3]. Conventional cleaning methods include sandblasting, pressure washing, and chemical cleaning, and each of these have potentially negative effects on the stone, such as causing abrasion or leaving residues on the surface and in the environment [2,4,5].
Pulsed laser cleaning has gained popularity in the conservation of cultural heritage in Europe and across the world for the treatment of historic stonework, ranging from individual sculptures to building facades. Cleaning heritage stonework with pulsed lasers has mainly focused on carbonate substrates [2], but cleaning other types of stone such as granite has gained interest in the last 15 years. To date, research in laser cleaning graffiti from granite surfaces has explored the capabilities of pulsed lasers emitting in the nanosecond range [5–11]. This type of laser relies on the vaporization of unwanted layers through heating, or explosive ejection of material via shockwaves [12]. Nanosecond pulse lasers have satisfactorily removed red, blue, and black paints from stone surfaces from a visual point of view, but silver paint is more challenging to remove because metallic particles frequently remain on the surface, potentially due to the different composition of silver paint compared to the other colors [5,9,11]. Moreover, nanosecond laser cleaning may induce discoloration of the stone after cleaning and damage the minerals within the granite, particularly biotite, which is most easily degraded, and sometimes also feldspars and quartz [5,7,10,11]. Nanosecond laser cleaning is therefore inadequate for removing silver paint and may damage the stone due to its principal removal mechanism: the generation of heat and shockwaves.
In recent years, ultrashort (femtosecond, 1 fs = 10−15 s) pulse lasers have emerged as powerful and precise tools to ablate a wide range of materials by breaking molecular bonds of the material to be removed, without inducing the propagation of heat or shockwave. Femtosecond laser pulses have a duration that is shorter than the electron-ion energy transfer time and also the time for heat diffusion from the skin layer (where the laser radiation is absorbed) to the bulk substrate [13,14]. The ablation therefore proceeds in non-equilibrium conditions from the skin layer, with only minimal heat transfer occurring into the bulk of the material. This is particularly interesting for the treatment of complex polymineralic substrates such as granite which contains minerals with different heat sensitivity.
A femtosecond pulse laser was used in 2013 to remove a biological crust from a Galician granite, and the results were compared to the results of a nanosecond pulse laser [15]. This study showed that the removal efficiency of femtosecond pulses was comparable to other cleaning methods including nanosecond pulse laser and demonstrated a better preservation of the morphology of minerals compared to the nanosecond pulse laser. Indeed, irradiating the stone with nanosecond pulses resulted in a decrease in surface roughness due to the changes induced to the minerals (melting and loss of cleavage planes of biotite, muscovite, and formation of fusion crust for feldspars), whilst femtosecond pulses allowed preserving the cleavage planes and relief of the minerals. This study illustrated the potential of femtosecond pulse lasers for improved conservation of heritage stonework.
In this work, we investigate the use of a femtosecond pulse laser to remove a range of spray paints from an Australian granite. The efficiency of the treatment and the effects on the stone substrate are evaluated using multiple analytical techniques including optical microscopy, optical profilometry, Raman spectroscopy, energy-dispersive X-ray spectroscopy, and colorimetry. We show that no damage and no color change occur when the laser fluence is below the damage threshold for the stone of 1.1 J·cm-2, and we demonstrate efficient and effective cleaning results. This study illustrates the significant advantages of femtosecond pulse lasers for the preservation of heritage stonework.
2. Materials and methods
2.1. Materials
The experiments were carried out on slabs of Moruya granite. This stone is a southeastern Australian granodiorite dominated by plagioclase (56.4%), quartz (22%), biotite (10.3%), hornblende (5.8%), potassium-feldspar (3.7%), and other minor phases [16], and has been used for the construction of buildings and monuments in Sydney and New South Wales. Samples were acquired from the Dorman Long Quarry site (New South Wales), best known as the source of material for the Sydney Harbour Bridge and prepared with a diamond-sawn finish.
Spray paints are available off the shelf from a variety of brands such as Montana Colors, Ironlak, Liquitex, Rust-Oleum, Krylon, Molotow, Loop, etc. We selected the Montana Colours Blood Red, Europe Blue, Cadmium Yellow, Natura Green, and Chrome Silver, providing a selection of common colors and finishes for testing. This brand of paint was selected to provide continuity with work that had previously been undertaken using nanosecond pulse lasers [5–11,17]. The paint was applied by placing the can at 30 cm from the surface of the granite with an angle of 45 degrees and spraying for 10 seconds over the entire surface. The samples were left to dry in ambient laboratory conditions, protected from the sun, for an average of 10 days. The paint was applied freehand, in ambient temperature and relative humidity and without standardisation of the thickness of paint deposited, to reflect the real-world conditions of spray paint use.
2.2. Laser setup and parameters
A Carbide 40W femtosecond laser (CB3-40W from Light Conversion) was used for the laser cleaning, operating at its fundamental wavelength of 1029 nm, with a set pulse duration of 275 fs at a repetition rate of 100 kHz, delivering up to 0.4 mJ in energy per pulse. The beam was scanned across the sample in the y-direction with a 10-facet polygon mirror (Precision Laser Scanning Inc.) and in the x-direction with a galvanometer scanner (Cambridge Technology Inc.). The maximum scan area was 280 mm x 280 mm, which allowed beam speeds of up to 880 m·s-1, high enough to operate in single shot per spot regime and avoid possible accumulation effects. The scanner speed was set to 1000 rpm (min-1). The Gaussian beam profile of the laser output was transformed using a flat top square-shaped beam homogeniser (ST-220-J-Y-A, Holo Or), resulting in a final beam size of 168 µm x 175 µm. The line-to-line displacement was set to 5 µm and the shot-to-shot spacing was 894.47 µm. The square flat top beam presented a more homogeneous fluence distribution over the entire spatial profile of the beam. The laser pulses were focused on the sample with a quasi-telecentric f-Theta scanning lens of 540 mm working distance (S4LFT1420/449, Sill Optics GmbH). Experimental work was undertaken on dry surfaces in ambient laboratory conditions (temperature of 21 °C ± 1°C and uncontrolled relative humidity typically around 20-30%).
2.3. Analytical methods
The cleaning efficiency and possible damage to the stone were visually evaluated with an optical microscope (LEICA DM 2700 M). Surface texture, roughness, and paint thicknesses were measured with an optical profilometer (Veeco Wyko NT9100, Bruker) in vertical scanning interferometry mode (VSI) using a 5× objective and 0.55× field-of-view multiplier.
The spray paint is largely made of organic molecules and so the presence of paint and paint residues on the surface were assessed by Raman Spectroscopy using a Renishaw InVia Reflex Raman spectrometer, with a 785 nm Near Infrared diode laser and a standard 20× lens (1.15 mm working distance). A 1200 lines·mm-1 grating was used for all measurements and the spectra were collected with a cooled CCD detector. On the paint, four scan accumulations with an acquisition time of 10s were used over a spectral range of 100 cm-1 to 2500 cm-1 at a laser power of less than 1 mW. On the cleaned granite surfaces, two scan accumulations with an acquisition time of 45 s were used over a spectral range of 100 cm-1 to 2500 cm-1 at a laser power of less than 6 mW. These parameters were selected after testing to optimize the signal-to-noise ratio of both the paint and granite surfaces. Raman reference spectra of minerals were taken on a slab of unpainted Moruya granite. The raw spectra were baseline-corrected with BSpline interpolation, smoothed with a Savitzky-Golay filter of polynomial order 3 and 25 points, and fitted with a mix of Gaussian and Lorentzian peak profiles.
The elemental composition of the paint was documented on carbon-coated samples by energy-dispersive X-ray spectroscopy using a Zeiss UltraPlus field-emission scanning electron microscope (FESEM) with a voltage of 10 kV, current of 1 nA, and a working distance of 11.5 mm.
Color changes of granite after the laser removal of paint were assessed by colorimetry using a Konica Minolta CM-2600d/2700d spectrophotometer emitting with the D65 illuminant and an observer at 10 degrees through a 3 mm mask diameter, with the specular component included. The results were expressed in the colors spaces defined by the International Commission on Illumination (abbreviated CIE) CIELab and CIELCh, where L* expresses the lightness and ranges from 0 (black) to 100 (white), a* the green (-a*) and red component (+a*), b* the blue (-b*) and yellow component (+b*), and the polar coordinates C* for chroma or relative saturation, and h the hue angle on the color wheel.
3. Results and discussion
3.1. Composition of the paint
The complete composition of the paint is proprietary information and not publicly available. Montana indicates that their spray paints contain modified alkyd binders with aromatic thinners. Raman spectroscopy allowed us to identify some pigments used in the formulations. The identified Raman peaks help to detect the presence of residual paint after laser cleaning, serving as potential in-situ control over the ablation process by indicating complete removal of paint from the granite surface, and to identify if there are any changes in molecular structures in granite as a result of laser irradiation.
Figure 1 presents the Raman spectra of different paints, with the best matching reference pigments [18]. The band positions and attributions are summarized in Table 1, along with EDS results. The spectrum of the blue paint is typical of phthalocyanine blue, a modern synthetic organic pigment of chemical formulae C32H16CuN8. The spectrum of green paint can be attributed to phthalocyanine green pigment, of chemical formulae C32Cl16CuN8, possibly mixed with another pigment or compound inducing a slightly different Raman spectrum compared to the pure phthalocyanine green pigment shown as reference. The red paint best matches naphthol red, another modern organic pigment, of chemical formulae C26H22N4O4, possibly with another compound inducing changes in intensity and position of Raman bands compared to the pure naphthol red. The spectrum of yellow paint was similar to arylide yellow, an organic pigment of chemical formulae C17H16N4O4. A very low signal-to-noise ratio was obtained for the spectrum of silver paint, which is due to the technique not being sensitive to the metal-organic composition.
Table 1. Raman bands and interpretations, and major and minor elements found in the different spray paints
The EDS results presented in Table 1 revealed that all paints except silver contained titanium, most likely titanium dioxide, as a white pigment, and silicon in varying amounts as a paint filler [19]. The green and blue paints contained copper, consistent with phthalocyanine pigments [20]. Small amounts of zirconium were detected in the blue and red paints, most likely as another paint filler. Iron was found in the red paint. Etxebarria et al. reported in [21] that a mixture of iron oxides was most likely used in the pigments of Montana browns and reds. The yellow paint had the lowest X-ray intensity and contained magnesium, titanium, and aluminium. The ‘Chrome’ effect of the silver paint may be achieved by using aluminium [19], which was the main element found in the silver paint. This paint also showed the highest amount of carbon.
3.2. Removal of spray paints
In a previous study focused on the Moruya granite [28], we determined the damage threshold of the four key minerals in the stone. As the purpose of this study is to show the capabilities of femtosecond pulse lasers to clean heritage stones, preserving the structural and visual integrity of the stone is of paramount importance. Biotite is the mineral most sensitive to the laser irradiation, and so its damage threshold of 1.1 J·cm-2 determined the damage threshold of the overall stone [28]. Therefore, the fluence applied to remove the paint was set at 1.0 J·cm-2, below the damage threshold of biotite.
3.2.1 Femtosecond pulse laser cleaning and visual assessments
The appearance of the spray paints on the granite surface before cleaning differed depending on the paint, as observed in the microscope pictures in Fig. 2. We attribute this mainly to the way the paint was applied on the surface, its thickness, and the drying process, and to the differing compositions of the paints, as they were from various series (MTN94, Hardcore and water-based).
Fig. 2. Optical microscopy image of the different spray paints on granitic surfaces, with various morphologies. (a) blue; (b) green; (c) yellow; (d) red; (e) silver.
Femtosecond pulse laser cleaning was performed by irradiating rectangular areas with an increasing number of laser scans until a satisfactory level of cleanliness was reached. Figure 3 shows the impact of increasing number of laser scans for a constant fluence of 1.0 J·cm-2. For all the tested paints, one laser scan removed the top layer of paint (patches marked (1)), resulting in a clear change in coloration to a greyer and darker appearance. This was confirmed by colorimetry measurements, with a general decrease in L* parameter (-10 CIELab units for red and green, -29 for yellow, -7 for blue), which corresponds to an increase in darkness.
Fig. 3. Optical microscope images of painted granite samples after laser cleaning with increasing numbers of scans (indicated by the number in the white square at side of the corresponding laser processed area). (a) blue paint; (b) green paint; (c) yellow paint (d) red paint and (e) silver paint.
Increasing the number of scans removed more paint as shown by the patch numbers in Fig. 3. The edges of the zones were sharp and well defined, demonstrating good cleaning accuracy which is important for removing paint from fine sculptural details and small architectural features.
The paint thicknesses were measured with optical profilometry, and the results are presented in Table 2. The red paint was the thinnest at 54.6 µm ± 2.3 µm, then the yellow paint at 69.3 µm ± 1.1 µm. The other colors were thicker, with 108.5 µm ± 1.3 µm for silver, 130.2 µm ± 2.0 µm for green, and 147.4 µm ± 0.8 µm for the blue paint. For those thicknesses, the surface was rendered partially clean for red and silver paints with 10 scans, and blue with 15 scans. Paint patches remained in the corners; most likely due to the samples not being completely flat, resulting in less energy being delivered to the surface in those corners. The surfaces were found to be completely cleaned with 15 scans for green, red, yellow, between 20 and 30 scans for blue, and between 20 and 40 scans for silver. A high number of scans were applied on the last blue and silver rectangles with a smaller beam step to remove the few paint particles that were still visible after 20 scans. The appearance of the cleaned silver surface seems rougher; however, the average surface roughness was not found to be significantly different before and after cleaning (4.03 µm ± 0.36 µm before, 4.8 µm ± 0.20 µm after).
Table 2. Thickness of the paint coatings and respective ablation rates (with 29.2 W of laser power for each) and efficiencies for each color (values with 15 scans for green, yellow, red paints and 20 for blue and silver paints)
The ablation efficiencies and rates for all the paints were in the same order of magnitude (see Table 2). Because we deliberately kept the laser fluence below the damage threshold for granite to avoid damaging the stone, these are not the maximum speed and power efficiencies that can be achieved, but the speed and power rates achieved for conditions that remove paint without damaging the stone.
Extrapolating for a surface of 10 cm x10 cm, and with the thicknesses in Table 2, it would take less than 8 minutes to clean the surface, independently of the spray paint’s color, with an applied laser power of 29.2 W. Extrapolating further for a laser power comparable to what is commercially available for nanosecond cleaning lasers, 100 W allows complete removal of paints on the 10 cm x 10 cm surface in less than 2 min 16 s, and 500 W in less than 26 seconds.
As the laser fluence was kept below the damage threshold of the stone, no mineral damage was observed, even on biotite, the most sensitive mineral of Moruya granite (Fig. 4). This demonstrates the potential of using femtosecond lasers for conservation treatment, to preserve the integrity of the substrates whilst achieving satisfactory cleaning results. In sum, the femtosecond pulse laser has the major advantage that the stone can be both satisfactorily cleaned and preserved, unlike nanosecond laser cleaning methods.
Fig. 4. Optical microscopy image of mineral grains after laser cleaning showing that there is no damage. Bt: Biotite; Qz: Quartz; Pl: Plagioclase. The mineral grains pictured are from (a) cleaned red paint surface; (b) cleaned blue surface with small paint patches remaining; (c) cleaned green surface; (d) cleaned silver surface; (e) cleaned yellow surface.
3.2.2 Colorimetry
The mineral grains in granite are typically smaller than the aperture of the spectrophotometer used for colorimetry measurements, resulting in values for the stone that vary from sample to sample depending on the contributions of the differently colored minerals. To determine a general measurement of the granite color before paint application, and by implication the color that should be recovered by successful cleaning, a slab of granite prepared at the same time as the test samples was used as a reference. Mineral grains close to the size of the measurement aperture were used to provide approximate color values for biotite and hornblende (black minerals), quartz (grey minerals) and feldspars (white minerals) in terms of (a*,b*) and (C*,L*) coordinates (Fig. 5). Based on the results for the minerals, we consider a surface ‘clean’ if the color measurements fall in a square delimited by the reference points: a* [-1.8;0.45], b* [-0.1;7.0], C* [0.6;7.1], L* [38.7;78.0].
Fig. 5. CIELab references for the minerals making up granite. Black: hornblende and biotite; White: feldspars; Grey: quartz. (a) (a*,b*) coordinates; (b) (C*,L*) coordinates.
The results for the paints and cleaned surfaces are presented in Fig. 6. Colorimetry results show that the treatment was very successful at recovering the initial color of the stone, with most cleaned points located within the red square defined by the reference points. In terms of lightness and chroma, it may be noted that ΔL* = -4.86 for removal of blue paint, ΔL* = -3.63 for removal of green, and ΔL* = -1.33 for removal of silver paint which suggest a darker surface post treatment. However, these differences measured with colorimetry are not striking to the naked eye.
Fig. 6. Colorimetry results expressed in CIELab color space. (a) general overview of (a*,b*) coordinates; (b) zoom into the cleaned points; (c) general overview of (C*,L*) coordinates; (d) zoom into the cleaned points.
The effectiveness of treatment may be described in terms of the Euclidian distance between the (L*a*b*) coordinates of the reference and treated surfaces. The smaller the distance, the more effective is the treatment at recovering the initial colors of the reference. It may be calculated easily using the following formulae [29]:
Table 3. ΔE* values in CIELab units for the granite surface after removal of the different paints
The results in Table 3 were obtained by comparing the average color of several cleaned points with the average of several reference points. We considered the treatment to be acceptable if $\Delta \textrm{E*}\mathop<\limits_\approx\,5$, given the high disparity in the colors of minerals in the granite. As a consequence, the cleaning treatment was successful for red, blue, yellow, and silver, which is confirmed by visual observation. We note that ΔE* = 6.04 for green paint, which falls outside of the numerical definition of success, but is nonetheless visually acceptable. The differences in the granite surfaces before and after cleaning with the femtosecond pulse laser are not perceptible to the naked eye.
3.2.3. Overexposure during laser cleaning
To determine the effects of a potential overexposure of the granitic surface to the laser, in terms of surface roughness, and morphological damage, laser irradiation experiments were undertaken on an unpainted and painted sample. Three overexposure conditions were applied on the unpainted sample at 1.0 J·cm-2: 4 minutes of scanning (20 scans) in zone 1, 15 minutes of scanning (50 scans) in zone 2, and 25 minutes of scanning (100 scans) in zone 3. For comparison, 100 scans were applied on the painted sample at 1.0 J·cm-2 (Zone 4). This overexposure represents a possible case of ‘overcleaning’, although it is worth noting that the scanning times selected are longer than what would be done in real-life situations.
Table 4 presents the average surface roughness measured for the four zones before and after irradiation with the laser.
Table 4. Average surface roughness Ra before and after overexposure to the laser beam for each unpainted zone
Those results show that the average surface roughness was not impacted by the excessive laser irradiation. Indeed, the surface roughness of Zone 4 increased from 5.66 µm ± 0.19 µm after 15 scans (needed to remove all the paint) to 6.07 µm ± 0.13 µm after 100 scans (85 extra scans where the stone was overexposed to the laser). This increase is minimal and not perceptible visually, as illustrated on Fig. 7.
Fig. 7. Microscopic image of the granite surface (a) Biotite (Bt) and plagioclase (Pl) before cleaning; (b) same spot after 100 laser scans; (c) hornblende (Hbl), biotite, plagioclase, and quartz (Qz) before cleaning (d) same spot after 100 laser scans.
It must be pointed out that some of the biotite’s gloss was lost after 100 scans, as seen in Fig. 7(a) and 7(b). No melting was detected but some damage was nonetheless observed at the minerals’ surface, perhaps due to oxidation or dehydration although this was not investigated further. The cleavage planes were preserved despite the high number of scans.
The other minerals did not present any sign of damage. No difference was observed for hornblende, quartz, and plagioclase, as seen on Fig. 7(c) and 7(d). These minerals were not impacted by the overexposure to the femtosecond pulse laser.
4. Vibrational spectroscopy
The spectra obtained for each paint and cleaned surfaces are shown on Fig. 8. The spectra collected for the cleaned areas are similar to plagioclase and quartz. No polymer features are visible on the spectra after irradiation with the laser, confirming the optical microscope observations and colorimetry measurements that complete removal of the paint was achieved in each case. Moreover, no shifts or broadenings are observed on the spectra of plagioclase and quartz in cleaned areas compared to the reference spectra, indicating that no damage (such as stress or melting) was induced in the crystal structure.
Fig. 8. Raman spectra of painted and cleaned granite surfaces. Paint spectra correspond to modified alkyd resin. Clean spectra show the typical features of plagioclase and quartz, highlighting the total removal of the paints.
5. Conclusion
In this paper we investigated the use of a femtosecond pulse laser emitting at its fundamental wavelength of 1029 nm to remove spray paints of different colors from Moruya granite. We demonstrated satisfactory removal of all the paints using a fixed laser fluence of 1.0 J·cm-2, which was determined to be below the damage threshold of the granite. The efficiency of the cleaning treatment was proved with optical microscopy, colorimetry, and Raman spectroscopy. These techniques revealed that no undesirable effects were induced on the stone, and that the femtosecond pulse laser was successful at retrieving the initial colors of the stone when the stone was not overexposed to the laser beam.
The ablation efficiencies and rates were calculated for the different paint thicknesses. The green paint was the most efficiently removed with 223.60 mm3.min-1, followed by blue with 189 mm3.min-1, silver with 139.75 mm3.min-1, yellow with 119.01 mm3.min-1 and finally red with 93.78 mm3.min-1. Those rates and efficiencies do not represent the highest ablation efficiencies and speed conditions achievable with femtosecond pulse lasers for cleaning purposes because we deliberately limited the maximum ablation fluence to avoid damaging the stone. The results demonstrate the significant advantages of using femtosecond pulse lasers for the preservation of heritage and cultural stones.
Funding
Australian Research Council (DP200100406, LP180100276).
Acknowledgements
The authors acknowledge Shane Paxton formerly from the Research School of Earth Sciences, and Michael Turner from the Department of Applied Mathematics in the Research School of Physics, for the preparation of stone samples. The authors also acknowledge the Department of Electronic Materials Engineering (EME) for the access to the Raman spectrometer, the facilities, and the scientific and technical assistance of the Australian Microscopy & Microanalysis Research Facility at the Centre of Advanced Microscopy, at the Australian National University. The research was supported by the Australian Government through the Australian Research Council's Linkage Project funding scheme (Project LP180100276) and funding provided by Transport for New South Wales. King was partially supported by DP200100406.
Disclosures
The authors declare no conflicts of interest.
L.R. and A.V.R. initiated the project. L.R. and S.M. designed and constructed the experimental setup. J.B. performed the experiments and analysis. A.W., A.V.R. and P.L.K. helped validating the analysis and reviewing the results. J.B. and L.R. wrote the manuscript with inputs from all authors. All authors discussed the results and contributed to the manuscript preparation.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. P. Sanmartín, F. Cappitelli, and R. Mitchell, “Current methods of graffiti removal: A review,” Construction and Building Materials 71, 363–374 (2014). [CrossRef]
2. V. Gomes, A. Dionísio, and J. S. Pozo-Antonio, “Conservation strategies against graffiti vandalism on Cultural Heritage stones: Protective coatings and cleaning methods,” Prog. Org. Coat. 113, 90–109 (2017). [CrossRef]
3. A. Morgan and E. Louis, Key issues in graffiti, in Research in practice.2009, Australian Institute of Criminology: Canberra.
4. C. Woolfitt and J. Fairchild, Graffiti on Historic Buildings Removal and Prevention, in Historic England.: Swindon2021.
5. J. S. Pozo-Antonio, T. Rivas, M. P. Fiorucci, A. J. López, and A. Ramil, “Effectiveness and harmfulness evaluation of graffiti cleaning by mechanical, chemical and laser procedures on granite,” Microchem. J. 125, 1–9 (2016). [CrossRef]
6. M. F. Costa, M. P. Fiorucci, J. Lamas, A. J. López, T. Rivas, and A. Ramil,. Laser cleaning of graffiti in Rosa Porrino granite. in International Conference on Applications of Optics and Photonics.2011 of Conference.
7. M. P. Fiorucci, A. J. López, A. Ramil, S. Pozo, and T. Rivas, “Optimization of graffiti removal on natural stone by means of high repetition rate UV laser,” Appl. Surf. Sci. 278, 268–272 (2013). [CrossRef]
8. V. Gomes, A. Dionísio, J. S. Pozo-Antonio, T. Rivas, and A. Ramil, “Mechanical and laser cleaning of spray graffiti paints on a granite subjected to a SO2-rich atmosphere,” Construction and Building Materials 188, 621–632 (2018). [CrossRef]
9. J. Pozo-Antonio, A. Papanikolaou, K. Melessanaki, T. Rivas, and P. Pouli, “Laser-Assisted Removal of Graffiti from Granite: Advantages of the Simultaneous Use of Two Wavelengths,” Coatings 8(4), 124–144 (2018). [CrossRef]
10. A. Ramil, J. S. Pozo-Antonio, M. P. Fiorucci, A. J. López, and T. Rivas, “Detection of the optimal laser fluence ranges to clean graffiti on silicates,” Construction and Building Materials 148, 122–130 (2017). [CrossRef]
11. T. Rivas, S. Pozo, M. P. Fiorucci, A. J. López, and A. Ramil, “Nd:YVO4 laser removal of graffiti from granite. Influence of paint and rock properties on cleaning efficacy,” Appl. Surf. Sci. 263, 563–572 (2012). [CrossRef]
12. M. Cooper, Laser cleaning in conservation : an introduction, ed. M. Cooper 1998, Butterworth-Heinemann.:Woburn, MA120 pages.
13. E. G. Gamaly and A. V. Rode, “Ultrafast ablation with high-pulse-rate lasers. Part I: theoretical considerations,” J. Appl. Phys. 85(8), 4213–4221 (1999). [CrossRef]
14. E. G. Gamaly and A. V. Rode, “Physics of ultra-short laser interaction with matter: From phonon excitation to ultimate transformations,” Prog. Quantum Electron. 37(5), 215–323 (2013). [CrossRef]
15. T. Rivas, A. J. Lopez, A. Ramil, S. Pozo, M. P. Fiorucci, M. E. L. d. Silanes, A. García, J. R. V. d. Aldana, C. Romero, and P. Moreno, “Comparative study of ornamental granite cleaning using femtosecond and nanosecond pulsed lasers,” Appl. Surf. Sci. . 278, 226–233 (2013). [CrossRef]
16. T. J. Griffin, A. J. R. White, and B. W. Chappell, “The Moruya Batholith and geochemical contrasts between the Moruya and Jindabyne suites,” J. Geol. Soc. Aust. 25(3-4), 235–247 (1978). [CrossRef]
17. C. Ricci, F. Gambino, M. Nervo, A. Piccirillo, A. Scarcella, F. Zenucchini, A. Ramil, and J. S. Pozo-Antonio, “Enhancement of graffiti removal from heritage stone by combining laser ablation and application of a solvent mixture,” Construction and Building Materials 262, 119934 (2020). [CrossRef]
18. M. C. Caggiani, A. Cosentino, and A. Mangone, “Pigments Checker version 3.0, a handy set for conservation scientists: A free online Raman spectra database,” Microchem. J. 129, 123–132 (2016). [CrossRef]
19. I. M. Cortea, L. Ratoiu, and R. Rădvan, “Characterization of spray paints used in street art graffiti by a non-destructive multi-analytical approach,” Color Res. Appl. 46(1), 183–194 (2021). [CrossRef]
20. P. Erk and H. Hengelsberg, Phthalocyanine Dyes and Pigments. The Porphyrin Handbook: Phthalocyanines: Spectroscopic and Electrochemical Characterization, 2003. 16: p. 117–170.
21. I. Etxebarria, N. Prieto-Taboada, E. Lama, G. Arana, M. D. Rodríguez-Laso, and J. M. Madariaga, “Graffiti Characterization Prior to Intervention in the Punta Begoña Galleries (Getxo, North of Spain): Raman and XRF Spectroscopy in the Service of Restoration,” Appl. Sci. 11(18), 8640 (2021). [CrossRef]
22. M. R. Lopez-Ramírez, N. Navas, L. R. Rodríguez Simon, J. C. Otero, and E. Manzano, “Study of modern artistic materials using combined spectroscopic and chromatographic techniques. Case study: painting with the signature “Picasso”,” Anal. Methods 7(4), 1499–1508 (2015). [CrossRef]
23. W. Fremout and S. Saverwyns, “Identification of synthetic organic pigments: the role of a comprehensive digital Raman spectral library,” J. Raman Spectrosc. 43(11), 1536–1544 (2012). [CrossRef]
24. A. Cesaratto, S. A. Centeno, J. R. Lombardi, N. Shibayama, and M. Leona, “A complete Raman study of common acid red dyes: application to the identification of artistic materials in polychrome prints,” J. Raman Spectrosc. 48(4), 601–609 (2017). [CrossRef]
25. P. Vandenabeele, L. Moens, H. G. M. Edwards, and R. Dams, “Raman spectroscopic database of azo pigments and application to modern art studies,” J. Raman Spectrosc. 31(6), 509–517 (2000). [CrossRef]
26. A. Colombini and D. Kaifas, “Characterization of some orange and yellow organic and fluorescent pigments by Raman spectroscopy,” E-Preservation Science7, 14–21 (2009).
27. S. Q. Lomax and J. F. Lomax, “The synthesis characterization of historical novel azo pigments: implications for conservation science,” Heritage Sci. . 7(1), 101 (2019). [CrossRef]
28. J. Brand, A. V. Rode, S. Madden, A. Wain, P. L. King, and L. Rapp, “Ultrashort pulsed laser ablation of granite for stone conservation,” Opt. Laser Technol. 151, 108057 (2022). [CrossRef]
29. International Commission on Illumination, Colorimetry Part 4: CIE 1976 L*A*B* Colour Space.2007.
30. G. Sharma, Digital Color Imaging Handbook. Electrical engineering and applied signal processing series, ed. CRC Press. 2003.
31. E. A. Early and M. E. Nadal, “Uncertainty analysis for reflectance colorimetry,” Color Res. Appl. 29(3), 205–216 (2004). [CrossRef]
32. G. M. Karma, “Determination and Measurement of Color Dissimilarity,” Int. J. Eng. Emerging Technol. 5(1), 67 (2020). [CrossRef]
