Final manuscript after referees Click here to download Manuscript: Lopez-Arce&Zornoza-Indart_JAppPhysA_2015_FINAL_AfterReferees.docx Click here to view linked References

1 Carbonation acceleration of calcium hydroxide nanoparticles 1 2 2 Induced by yeast fermentation 3 4 5 3 Paula Lopez-Arce*1,2  Ainara Zornoza-Indart2 6 7 4 1Géosciences et Environnement department, Université de Cergy-Pontoise, Neuville-sur-Oise, 8 9 5 Cergy-Pontoise cedex, France 10 11 6 2Instituto de Geociencias IGEO (CSIC,UCM), C/José Antonio Nováis 2, 28040 Madrid, Spain 12 13 14 7 *Corresponding author: [email protected] / [email protected] 15 16 8 Abstract 17 18 19 9 Carbonation of Ca(OH)2 nanoparticles and consolidation of limestone are accelerated 20 10 by high humidity and a yeast fermentation system that supplies a saturated 21 22 11 atmosphere on CO2, H2O vapor and ethanol during 28 days. Nanoparticles were 23 24 12 analyzed by X-ray diffraction and differential thermal analyses with thermogravimetry. 25 Spectrophotometry, scanning electron microscopy analyses, hydric and mechanical 26 13 27 14 tests were also performed in stones specimens. Samples exposed to the yeast 28 29 15 environment achieve 100% relative CaCO3 yield, whereas at high humidity but without 30 the yeast and under laboratory environment, relative yields of 95% CaCO and 15% 31 16 3 32 17 CaCO3 are respectively reached, with white crusts and glazing left on the stone 33 34 18 surfaces when the nanoparticles are applied at a concentration of 25g/l. 35 36 19 The largest increase in the drilling resistance and surface hardness values with slight 37 38 20 increase in the capillarity absorption and desorption coefficients and with lesser stone 39 40 21 color changes are produced at a concentration of 5g/l, in the yeast system 41 22 environment, in stone specimens initially with bimodal pore size distributions, more 42 43 23 amounts of pores with diameters between 0.1 and 1µm, higher open porosity values 44 45 24 and faster capillary coefficients. 46 47 25 An inexpensive and reliable method based on water and yeast-sugar solution is 48 49 26 presented to speed up carbonation of nanoparticles used as a consolidating product to 50 51 27 improve the mechanical properties of decayed limestone from archaeological and 52 28 architectural heritage. 53 54 55 29 Keywords: carbonation; yeast fermentation; Ca(OH)2 nanoparticles; environmental 56 57 30 conditions; humidity; CO2; limestone; consolidation 58 59 31 60 61 62 63 1 64 65 32 1 Introduction 1 2 33 When calcium hydroxide (Ca(OH)2) is exposed to atmospheric carbon dioxide (CO2) 3 4 34 under moist conditions, it reacts and converts into calcium carbonate (CaCO3) 5 6 35 releasing water as a result of carbonation. This reaction takes place in 7 36 biomineralization, medical and pharmacological processes [1,2], in concrete and 8 9 37 mortars [3-5] or as a product binder to improve the cohesion of highly porous materials 10 11 38 [6,7]. Achieving its stability is very important in architecture and decorative arts [8,9]. 12 13 39 An approach to improve conservative treatments based on the use of inorganic 14 15 40 compounds was to enhance and stimulate the growth of new calcite crystals inside the 16 17 41 porous system using biomineralization mechanisms [10]. Bacteria have been used to 18 42 create microbial deposited carbonate layers on ornamental stones. After the first 19 20 43 patent using microbial induced carbonate precipitation for the protection of ornamental 21 22 44 stone in the 90s, called Calcite Bioconcept technique, different approaches, bacterial 23 45 strains and methods, have been studied to solve the limitations of the technique . Even 24 25 46 though bacteria induced carbonate mineralization is considered a promising 26 27 47 consolidation treatment, compatible with carbonate stones and lime-based mortars and 28 48 cements, is not frequently used on-site by restorers. This is because it is a more 29 30 49 complex application process than traditional products, the incompatibility with previous 31 32 50 or subsequent treatments, their expensive price, need of knowledge on the bacteria 33 51 carbonation process and the interaction with the substrate with possible appearance of 34 35 52 secondary effects together with the lack of awareness on the long-term behavior [11- 36 37 53 13]. 38 54 Nanoscience deals with a large diversity of research fields and applications considering 39 40 55 that most of the properties are modified or improved by effect of the reduction in size of 41 42 56 nanomaterials. The use of nanoparticles is funded by the fact that the specific surface 43 57 area of a solid increases exponentially with the decrease of its volume, having a much 44 45 58 superior surface area available to react [14]. Colloidal nanoparticles are used for 46 47 59 cleaning, consolidation, and deacidification of materials from cultural heritage [15]. 48 60 Calcium hydroxide (Ca(OH)2) nanoparticles have been applied to improve the 49 50 61 petrophysical properties of lime mortars and carbonate stones, acting as a cementing 51 52 62 product to avoid decohesion of construction and building materials [16,17] or to 53 63 consolidate archaeological wall paintings [18]. 54 55 56 64 There are known factors and conditions that control the precipitation of CaCO3 57 58 65 polymorphs, such as pH, temperature, saturation, conductivity or impurities and 59 66 additives that determine the mineral precipitation and phase transformations together 60 61 62 63 2 64 65 67 with morphological and structural variations. The variables that play a role in the 1 68 carbonation of Ca(OH)2 nanoparticles, and the resulting mineral phases, are essential 2 3 69 for the final physical properties in a short- and longer- term [16]. It has been found that 4 5 70 these colloidal nano-dispersions are highly conditioned by the relative humidity (RH) 6 71 and exposure times [7,19] in the environment. These conditions directly influence the 7 8 72 speed of carbonation and the precipitation of CaCO3 polymorphs, as vaterite and 9 10 73 aragonite, which are less stable than calcite [20,21]. The alcohol present in the 11 74 environment also influences the stabilization of metastable CaCO3 polymorphic 12 13 75 phases, especially at high RH, where the rate of alcohol evaporation is slower, giving 14 15 76 rise to higher content of metastable phases, such as vaterite and aragonite [20]. 16 77 Moisture content and relative humidity during carbonation is a major factor in 17 18 78 determining the final mineral phases and pore structure of the resulting matrix in stone, 19 20 79 mortars or cement materials [16, 17], where the formation of nano-droplets of water is 21 80 necessary for the gaseous CO2 to react with Ca(OH)2 [22]. Due to the adsorption of 22 23 81 multilayers of water onto the solid Ca(OH)2 [23], high RH values (>75%) speed up the 24 25 82 carbonation process and the consolidation of lime mortars and carbonate stones [7,16, 26 17, 20, 21]. 27 83 28 29 84 Besides, even though CO2 is another key factor involved in the carbonation reaction, its 30 31 85 concentration and registration is not often considered in the studies of carbonation of 32 86 Ca(OH)2 nanoparticles. The influence of CO2 concentration on the acceleration of 33 34 87 carbonation process of Ca(OH)2 has been extensively debated. Even though, Van 35 36 88 Balen and Van Gemert [24] concluded that lime mortar carbonation depends on the 37 89 presence of water and occurs very quickly especially in a saturated CO2 atmosphere, 38 39 90 Van Balen [25] later shows that the reaction speed is not dependent upon the CO2 40 41 91 concentration, within the limits he tested. Dheilly et al. [26] demonstrate the importance 42 92 of CO2 concentration and RH during carbonation process. These authors observed 43 44 93 how a lime paste displays a rapid and complete reaction in a carbonic atmosphere, 45 46 94 while in a low CO2 environment carbonation takes twice as long. Cultrone et al. [5], 47 also demonstrate that high CO concentrations favor a faster and more complete 48 95 2 49 96 carbonation process, with 8 days being enough to convert portlandite (Ca(OH)2) into 50 51 97 90wt.% calcite (CaCO3) of lime-based mortars. They showed that under natural 52 conditions, carbonation is much slower and similar levels were not reached for 6 53 98 54 99 months, suggesting that the carbonation process is influenced by the amount of CO2. 55 56 100 Recently, Visser [27] showed that a change in CO2 concentration does not change the 57 58 101 carbonation process and that the amount of material that can carbonate can be 59 102 determined on the basis of the amount of calcium in the unreacted material. 60 61 62 63 3 64 65 103 Nevertheless, Cizer et al. [28] show that carbonation process is kinetically favored 1 104 under high RH and high pCO2 where supersaturation plays a critical role on the 2 3 105 nucleation density and size of CaCO3 crystals. 4 5 6 106 The control the CO2 concentration during carbonation process of lime based products, 7 107 for structures and building materials in the construction field, on-site under natural 8 9 108 conditions is a difficult task. It is hard to obtain CO2 sources that are reliable, durable, 10 11 109 cheap and easy to manage [29]. Saitoh et al. [30] developed an easy and cheap 12 110 method to produce CO2 for diseases purposes by using a yeast-sugar solution in 13 14 111 plastic bottles. Under anaerobic conditions, yeast (synonym for strains of 15 16 112 Saccharomyces cerevisiae or baker's yeast) converts sugar into CO2 and ethanol [31- 17 113 33]. 18 19 20 114 The aim of this research is to demonstrate the acceleration of carbonation of Ca(OH)2 21 22 115 nanoparticles by a yeast fermentation system that supplies a saturated atmosphere on 23 116 CO2, H2O vapor and ethanol, presenting an inexpensive and reliable method based on 24 25 117 water and yeast-sugar solution to improve the consolidation of decayed limestone. 26 27 118 2 Materials and methods 28 29 30 119 2.1 Ca(OH)2 nanoparticles 31 32 33 120 Two nanometric Ca(OH)2 colloidal dispersions employed as commercial consolidating 34 121 products were used in this study. Nanorestore®, developed at the University of 35 36 122 Florence (CSGI Consortium [34]), are hexagonal plate-like Ca(OH)2 nanocrystals (<100 37 38 123 nm in size [20]) with a concentration of 5g/l in isopropyl alcohol. CaLoSil® (Ziegenbald 39 124 [35]), was previously characterized [36] and consists of disaggregated hexagonal 40 41 125 nanoparticles with particle size from 60 to 130 nm. In this research we have used a 42 43 126 concentration of 25 g/l. Ultrasonic dispersion was performed introducing the products 44 127 during 5 min in an ultrasonic bath (Selecta model Ultrasounds-H). Both colloidal 45 46 128 dispersions were deposited on laboratory plastic dishes to leave them being absorbed 47 48 129 by capillarity by each stone specimen while these were exposed to three different 49 130 environmental conditions. Part of the concentration at 5 g/l that was not absorbed by 50 51 131 the specimens and remained on the dishes, was used to carry out X-ray diffraction 52 53 132 analyses and differential thermal analyses and thermogravimetry. 54 55 133 2.2 Stone specimens 56 57 58 134 Fresh stone detrital limestone samples, commercially called Leccese Stone (from 59 Southern Italy), and Bateig Stone (from Alicante, Eastern Spain), typically used in 60 135 61 62 63 4 64 65 136 historical and present-day buildings, were chosen for the application of Ca(OH)2 1 137 nanoparticles. 2 3 4 138 The Leccese Stone, also called Lecce Stone, selected for this study is a fine-grained 5 6 139 bio calcarenite, with a characteristic pale-yellow color, mainly composed of micritic 7 140 fraction (CaCO3 crystals <5 μm), mixed with fine clay minerals and poor 8 9 141 cryptocrystalline calcite cement; it contains fine microfossil fragments, grains of 10 11 142 glauconite, sporadic quartz grains and phosphate nodules. The total open porosity is 12 143 around 30-40% (with pore radius mainly below 5 μm). The rock has around 80% of 13 14 144 CaCO3 and the insoluble residue is essentially made of clay minerals and glauconite 15 16 145 [37,38]. The Bateig stone selected for this study, also called Novelda Stone, is the 17 146 commercial variety named Diamante (Diamond), a fine-grained bio calcarenite, with a 18 19 147 white-pale-yellow color, 75-80% of calcite (CaCO3), 5-10% dolomite CaMg(CO3)2, 5- 20 21 148 10% quartz and 5-10% phyllosilicates [39]. It contains fossils, such as foraminifers and 22 149 bryozoan, mollusk and echinoderm traces, as well as fragmentary quartz in different 23 24 150 amounts. There is hardly any cementing material (under 10%) and cement is 25 26 151 constituted by sparite (CaCO3 crystals >5 μm) and to smaller extent silica (SiO2). The 27 152 total open porosity is circa 22% [40] with pore radius mainly below 5 μm. 28 29 30 153 Two specimens for each stone type (Lecce (L) and Bateig (B), with the two selected 31 32 154 product’s concentration (5g/l and 25 g/l), were exposed to three different environmental 33 155 conditions, ventilate laboratory, closed container with high humidity and yeast 34 35 156 fermentation container with high humidity. A total of 24 cubic specimens 5x5x5 cm 36 37 157 were placed in individual laboratory plastic dishes with 32 ml and 20 ml of both colloidal 38 158 dispersions (quantity related to the capillary absorption of Lecce and Bateig stone, 39 40 159 respectively) to leave them being absorbed by capillarity through the porosity of each 41 42 160 stone specimen. Control specimens were used as a blank (with no nanoparticles 43 161 application) for each stone type, to evaluate the differences of mechanical properties of 44 45 162 the stones with and without nanoparticles exposed to the three environments during 28 46 47 163 days. 48 49 164 2.3 Environmental conditions 50 51 52 165 An air conditioning system inside the laboratory room stabilized the temperature (T) at 53 18±1 °C. The CO concentration inside the room and inside the container connected to 54 166 2 55 167 the yeast system was registered by a CO2 detector (Telaire® 7001, Goleta, CA, USA) 56 57 168 connected to a Hobo® Data Logger to register CO2 (0 -10,000 ± 1 ppm on display and 58 0-2,500 ppm on data logger, with accuracy of ±50 ppm or ±5% of reading up to 5,000 59 169 60 170 ppm), T range 0-50 ±1°C, and relative humidity (RH) range 0-95%. Environmental 61 62 63 5 64 65 171 control devices, ibuttons® model DS1923-F5 were introduced in each chamber and 1 172 placed in the laboratory room to measure T and RH during the 28 days experiment, 2 3 173 using the software OneWireViewer version 3.04. 4 5 6 174 The first group of samples were exposed to ventilate laboratory room conditions, 7 175 449±34 ppm CO2, 18 ± 1 °C and 59 ± 11 % RH. Water was placed underneath the 8 9 176 samples to increase a little bit the regular humidity of the room environment (45 ± 20 % 10 11 177 RH). The second group of samples was introduced into a closed desiccator, used as a 12 178 container. The equilibrium relative humidity (RHeq) of supersaturated MgSO4x7H2O 13 14 179 salt solution, placed at the bottom underneath the samples, was used to keep the 15 16 180 humidity constant, 90% RH at 20°C. The final average conditions registered by the 17 181 environmental sensors were 449±34 ppm CO2 (initial room air concentration), 18 ± 1 18 19 182 °C and 87±1% RH. Finally, a third group of samples were exposed to a yeast 20 21 183 fermentation system connected to a closed container with a supersaturated salt 22 184 solution of MgSO4x7H2O at the bottom. The final average environmental conditions 23 24 185 registered were above 2500 ppm CO2, T 20 ± 1 °C and 87±7% RH. To avoid the 25 26 186 contact of the samples with the water or magnesium sulfate salt solutions, the plastic 27 187 dishes and stone specimens were set on a grille placed several centimeters above the 28 29 188 bottom (Fig.1). 30 31 32 189 The yeast fermentation system was prepared using as a reference the data published 33 190 by Smallegange et al [29]. These authors calculated different carbon dioxide flow rates 34 35 191 (ml/min) produced by different yeast-sugar solutions. One of the mixtures used by them 36 37 192 to produce CO2 was 7g of yeast (Y) + 100 g of sugar (S) + 1L of water (W) with a rate 38 193 CO2 production of 3.5 ± 2.7 ml/min. In the present experimental work we use the 39 40 194 proportion 21g (Y) + 150g (S) + 1L (W), reaching 2,485ppm CO2 in 2 hours. This CO2 41 42 195 value is the maximum value registered by the data logger of the probe; however, on 43 196 display the value was much higher, showing values above 5,000 ppm. The value 44 45 197 recorded by the data logger of 2,500 ppm, was taken as a reference of the minimum 46 47 198 amount of CO2 produced by the yeast fermentation in a closed environment. A flask 48 199 filled with the prepared mixture was connected to a closed desiccator, used as a 49 50 200 container, through a lateral tube to supply the CO2 produced during the fermentation of 51 52 201 the yeast. The mixture was renewed 9 times, changing it every 3 days until the end of 53 202 the 28 days experiment, to ensure that the amount of CO2 inside the chamber were 54 55 203 always above 2,500 ppm. 56 57 58 204 2.4 Analyses of Ca(OH)2 nanoparticles 59 60 61 62 63 6 64 65 205 The samples of Ca(OH)2 nanoparticles at 5 g/l concentration were analyzed every 7 1 206 days during 28 days, using X-ray diffraction (XRD) to determine the mineralogy of the 2 3 207 different CaCO3 polymorphs precipitated during the carbonation reactions. Differential 4 5 208 thermal analyses and thermogravimetry (DTA-TG), was also carried out after the first 6 209 week (7days) and at the end of the last week (28 days) of the experiment, in order to 7 8 210 quantify the Ca(OH)2 and CaCO3 relative proportions in the samples for each 9 10 211 environment. 11 12 212 X-ray diffraction (XRD) was performed with a PHILIPS PW 1752 powder diffractometer 13 14 213 with CuK radiation. The samples were deposited in zero-background Si sample 15 16 214 holders. XRD patterns were obtained by step scanning from 2º to 68º2 with a scan 17 18 215 step size of 0.02º and 2º/min in a continuous mode. The working conditions were 40 kV 19 216 and 30 mA. The identification of the mineral phases was performed using the Bruker 20 21 217 AXS DiffracPlus EVA software. 22 23 218 Differential thermal analyses and thermogravimetry (DTA-TG), was performed using a 24 25 219 SDT Q600, TA Instruments analyzer, at a heating rate of 10 ºC/min over a range of 20- 26 27 220 1000 ºC. Approximately, samples of 10 ±1 mg were placed in alumina sample holders 28 and analyzed in N atmosphere. The total mass loss (%) during the thermal 29 221 2 30 222 decomposition of the samples was registered by the thermogravimetric analyses (TG). 31 32 223 The first derivative of the TG data (dTG curve) indicates the starting and ending 33 34 224 temperatures of mass loss during the TG analysis. So, the carbonation degree (yield) 35 225 has been calculated by means of the relative total proportion of Ca(OH)2 vs. CaCO3 36 37 226 present in the samples. These were recalculated for each temperature interval and 38 39 227 corresponding mass loss occurred during the dehydroxylation (loss of OH groups) and

40 228 decarboxylation (loss of CO2) processes. DTA curves indicate the range of 41 42 229 temperatures and maximum rates where these endothermic reactions take place. 43 44 230 2.5 Analyses and testing of stone samples 45 46 47 231 Spectrophotometry was used to measure the chromatic parameters before and after 48 49 232 the consolidation. Scanning electron microscopy (SEM) with energy dispersive X-ray 50 233 spectroscopy (EDS) analyses were performed to observe the surface morphology of 51 52 234 the stone samples. The elemental composition of some selected components was 53 54 235 qualitatively determined by means of EDS microanalyses. Mercury intrusion 55 236 porosimetry (MIP) was used to assess sample pore structure, i.e., total porosity (P) and 56 57 237 pore size distribution (PSD) of the blank stone specimens. Hydric properties (water 58 59 238 absorption by capillarity and water desorption) and mechanical properties (surface 60 239 hardness and drilling resistance) were measured in control specimens for each type of 61 62 63 7 64 65 240 stone used as blanks, and in all the specimens with the Ca(OH)2 nanoparticles after 1 241 the exposure during 28 days to the three different environmental conditions. Water 2 3 242 absorption under vacuum was performed in the blank and treated specimens to 4 5 243 determine changes on apparent density, real density, open porosity, and water 6 244 saturation. 7 8 9 245 Spectrophotometry was performed with a spectrophotometer MINOLTA CM-700d 10 11 246 using the CieLab color space; Standard illuminant was D65 and observer angle, 10º. 12 247 The measured parameters were L*, which accounts for luminosity, a* and b* 13 14 248 coordinates (a* being the red-green parameter and b* the blue-yellow), total chrome 15 2 2 1/2 16 249 difference ΔC* provided as a result of the formula ΔC*= (Δa*) + (Δb*) ) , total color 17 250 difference ΔE* provided as a result of the formula ΔE*= ((ΔL*)2 + (Δa*)2 + (Δb*)2)1/2, 18 19 251 white (WI) and yellow (YI) indexes were measured according to ASTM E313-73 [41], 20 21 252 brightness measured according to ISO 2470-2 [42] and reflectance. 22 23 253 The surface of stone blanks, with no nanoparticles, and the surface of the stone 24 25 254 specimens with nanoparticles at the end of the experiment, were analyzed with an 26 27 255 scanning electron microscope JEOL JSM 6400 and the analyses conditions were 0.2– 28 256 40 kV accelerating voltage, 6×10−1 A current, 10−5 Torr vacuum, 35 °A resolution, 8 mm 29 30 257 and 35 kV working distance and 20 kV accelerating voltage for image acquisition. The 31 32 258 spectrometer (EDS) was a microanalyzer Oxford instruments analytical Inca with a 133 33 259 eV–5.39 kV nominal resolution. Gold-sputtered samples were studied in secondary 34 35 260 electrons mode to observe the surface morphology and in retro-dispersed electrons 36 37 261 mode to analyze the chemical elementary composition. 38 39 262 Mercury intrusion porosimetry (MIP) readings were taken on three samples of each 40 41 263 stone type, at pore diameters of 0.005–400 m under measuring conditions ranging 42 43 264 from atmospheric pressure to 60,000 psia (228 MPa) on a Micromeritics Autopore IV 44 265 9500 MIP. Water absorption under vacuum was performed in the blank and treated 45 46 266 specimens to determine changes on apparent density, real density, open porosity and 47 48 267 to quantify the amount of water absorbed by the specimens once they reach saturation 49 268 as described in standard test UNE-EN 1936:2007 [43]. Water absorption by capillarity 50 51 269 and water desorption was carried out to calculate the capillary coefficients through the 52 2 1/2 53 270 stone according to the standard test UNE-EN 15801:2010 using g/(m ·s ) instead of 54 271 kg/(m2·s1/2) [44]. 55 56 57 272 Surface hardness was measured by means of a metal hardness tester Equotip 3 58 59 273 (Proceq), standardized according to the standard test ASTM A956-12 [45]. An impact 60 274 device “D” (tungsten carbide ball, 3mm diameter and 5.45 g weight) was used to obtain 61 62 63 8 64 65 275 HLD values with a range of measurements between 1-999 HLD, accuracy ± 4 HL 1 276 (0.5% at 800 HL), with impact energy of 11.5 N/mm2 and with automatic correction for 2 3 277 impact direction. Fifty measurements were performed on each blank specimen, and ten 4 5 278 measurements were performed on each specimen after treatment, over the 6 279 nanoparticles absorption face. Drilling resistance analyses were carried out with a 7 8 280 drilling Resistance Measurement System, DRMS Cordless by Sint Technology (Italy). 9 10 281 The drilling conditions were 15mm/min penetration speed for both types of stones and 11 282 400 rpm rotation speed for Bateig Stone and 200 rpm rotation speed for Lecce Stone. 12 13 283 Ten measurements with different drilling conditions, up to 25 mm penetration depth, 14 15 284 were performed on the blank specimens, to select the most appropriate and 16 285 homogeneous drilling conditions with less standard deviation. Then, three 17 18 286 measurements were done through the face surface that was directly in touch with the 19 20 287 colloidal dispersion. 21 22 288 3 Results and discussion 23 24 25 289 3.1 Characterization of nanoparticles 26 27 290 3.1.1 Samples exposed to ventilate laboratory room conditions 28 29 30 291 The XRD results show that after 7 and 14 days, portlandite (Ca(OH)2) is the main 31 32 292 mineral phase. Diffractograms display scarce peaks of calcite (C), only visible after 7 33 293 days (Fig.2a). After 21 days, the diffractogram pattern shows signs of amorphization 34 35 294 with the broadening of portlandite (P) peaks that are not well defined. After 28 days, 36 37 295 the crystallinity of portlandite seems to improve. However, there are only a few peaks 38 296 of calcite. These samples barely carbonated, displaying almost the total amount of 39 40 297 initial portlandite nanoparticles after 28 days under this environmental condition. 41 42 43 298 The TG analysis from 20°C up to 1000°C indicates a total mass loss of 27.46 % after 7 44 299 days (Fig.2b). DTA analysis shows three main endothermic peaks, which correspond 45 46 300 with the loss of adsorbed H2O with a maximum reaction rate at 140ºC, the loss of OH 47 48 301 groups from portlandite dehydroxylation at 436°C and loss of CO2 from calcite 49 302 decomposition at 682°C. The TG and dTG curves indicate that the mass loss during 50 51 303 dehydroxylation of portlandite (76%) occurs between 350° and 455°C, whereas 52 53 304 decarboxylation of calcite (24%) takes place between 516° and 722°C. After 28 days 54 305 (Fig.2c), there is a total mass loss of 29.94 % and two endothermic peaks with 55 56 306 maximum rates at 429°C and 648°C, which correspond to portlandite and calcite, 57 58 307 respectively. From 307°C up to 450°C there is a mass loss of 85% from 59 60 61 62 63 9 64 65 308 dehydroxylation of portlandite vs. 15% of mass loss from decarboxylation of calcite, 1 309 that takes place between 525° and 720°C. 2 3 4 310 3.1.2 Samples exposed into closed containers with high humidity 5 6 The XRD results show that after 7 days, calcite (C) and portlandite (P) are the main 7 311 8 312 mineral phases present in the sample (Fig.3a). Some small peaks could correspond to 9 10 313 ikaite phase (CaCO3*6H2O) that disappears in the diffractogram obtained after 14 days. 11 12 314 After 2 weeks, there are some aragonite (A) peaks, together with portlandite and 13 315 calcite. After 21 days, the peaks of the diffractogram are broadened and not so well 14 15 316 defined, showing signs of amorphization. Besides aragonite, portlandite and calcite, 16 17 317 there are some vaterite (V) peaks that disappear after 28 days. After one month, the 18 318 samples almost reached the total carbonation since there are lesser peaks of 19 20 319 portlandite and many more of aragonite and calcite showing higher intensities than 21 22 320 before. 23 24 321 The DTA analysis shows three main endothermic peaks, which correspond with the 25 26 322 loss of adsorbed H2O with a maximum reaction rate at 133ºC, the loss of OH groups 27 323 from portlandite at 429°C and loss of CO2 from calcite at 709°C, after 7 days. The TG 28 29 324 curve indicates a total mass loss of 36.24 % (Fig.3b). The TG and dTG curves indicate 30 31 325 that the mass loss during dehydroxylation of portlandite occurs between 359° and 32 466°C, which correspond to 31%, vs. 69% of mass loss due to decarboxylation of 33 326 34 327 calcite, between 513° and 732°C. After 28 days (Fig.3c), there is a total mass loss of 35 36 328 42.84 % and three endothermic peaks with maximum rates at 130°C, 399°C and 37 38 329 730°C, which correspond to adsorbed water, portlandite, and calcium carbonates, 39 330 respectively. From 350°C up to 420°C there is a mass loss of 5% from dehydroxylation 40 41 331 of portlandite, vs. 95% of mass loss from decarboxylation of calcium carbonates that 42 43 332 takes place between 530° and 756°C. In both dTG curves, 7 and 28 days, there is a 44 333 small bump between 150° and 300°C (Fig.3b) and between 180° and 347°C (Fig.3c) 45 46 334 that corresponds to 3% of the total amount of 95% considered as calcium carbonates 47 48 335 polymorphs. 49 50 336 3.1.3 Samples exposed to a yeast fermentation system with high humidity 51 52 53 337 The XRD results show that after 7 days of experiment, there is presence of portlandite 54 338 (P) and three calcium carbonate polymorphs, vaterite (V), aragonite (A) and calcite (C) 55 56 339 (Fig. 4a). Portlandite remains until the end of the second week of the experiment (14 57 58 340 days), while the number of peaks of vaterite, aragonite and calcite increase. After 21 59 341 days, there is no presence of portlandite, and the peaks of vaterite and calcite 60 61 62 63 10 64 65 342 decrease, whereas the peaks of aragonite increase. After 28 days, there is no 1 343 portlandite or vaterite, and the number and intensity of aragonite and calcite peaks 2 3 344 seems to decrease. 4 5 6 345 The DTA analysis shows three main endothermic peaks, which correspond with the 7 346 loss of adsorbed H2O with a maximum reaction rate at 140ºC, the loss of OH groups 8 9 347 from portlandite at 440°C and loss of CO2 from calcium carbonates at 733°C, after 7 10 11 348 days of exposure. The TG curve indicates a total mass loss of 40.06 % (Fig.4b). The 12 349 TG and dTG curves indicate that the mass loss during dehydroxylation of portlandite 13 14 350 occurs between 374° and 519°C, which correspond to 13%, vs. 87% of mass loss due 15 16 351 to decarboxylation of calcium carbonates, between 519° and 755°C. After 28 days 17 352 (Fig.4c), there is a total mass loss of 45.84 % and just two endothermic peaks with 18 19 353 maximum rates at 140°C, and 733°C, which correspond to adsorbed water and calcium 20 21 354 carbonates, respectively. From 371°C up to 429°C there is a 2% mass loss that may 22 355 correspond to aragonite-calcite phase transition, since there is no endothermic peak by 23 24 356 dehydroxylation of portlandite. Between 516° and 754°C there is a mass loss of 98% 25 26 357 from decarboxylation of calcium carbonates. 27 28 358 3.2 Interpretation of characterization of nanoparticles 29 30 31 359 From the XRD and DTA-TG results it can be clearly note that the yeast fermentation 32 system is the environment where the carbonation of Ca(OH) is faster compared to the 33 360 2 34 361 other environments, reaching 100% CaCO3 yield after 14 days. By the contrary, the 35 36 362 environment with the slowest carbonation rate is the one with ventilate laboratory room 37 38 363 conditions. In this latter case after 14 days, only portlandite Ca(OH)2 is present (which 39 364 corresponds with the original composition of the nanoparticles) in spite of few small 40 41 365 calcite peaks detected after 7 days exposure (Fig.2a). After 21 days, some calcite 42 43 366 peaks show up and portlandite peaks are broadened, like showing signs of 44 367 amorphization. This was previously observed when similar experiments were done 45 46 368 under a closed system with low RH conditions (33%RH), where the poor crystallinity 47 48 369 and the beginning of an amorphization process or loss of crystallinity of portlandite 49 370 nanoparticles, together with the presence of amorphous calcium carbonate (ACC) was 50 51 371 studied [21,36]. Under these conditions, there is a reduction in the size of the unit cells, 52 53 372 where the fast evaporation of the solvent (2-propanol) gives rise to smaller particles 54 373 with low crystallinity. The XRD data after 28 days reveals a better definition of the 55 56 374 maximum intensity peaks associated to portlandite and calcite, with no precipitation of 57 58 375 other CaCO3 polymorphs. However, by DTA-TG analyses the proportion of portlandite 59 376 seems to increase while the amount of calcite decreases, which is a rare behavior. The 60 61 62 63 11 64 65 377 pseudomorphic replacement of Ca(OH)2 particles by calcium alkoxides, which seem to 1 378 be poorly crystalline [46] together with ACC formation observed by SEM, could explain 2 3 379 this behavior and the signs of amorphization observed at 21 days. 4 5 6 380 At very high humidity (90% RH), calcite nucleation occurs at a very fast rate, because 7 381 high RH favors great concentration of structural defects. The presence of water around 8 9 382 the surface favors the introduction of point defects, enlarging the dislocation network 10 11 383 and generating other defects [47]. This allows speed up the carbonation reaction. The 12 384 high disorder degree contributes to the fast phase transformation from metastable 13 14 385 vaterite and aragonite to stable calcite [21]. However, the alcohol present in the 15 16 386 environment influences the stabilization of metastable CaCO3 polymorphic phases, 17 387 especially at high RH, where the rate of alcohol evaporation is slower, giving rise to 18 19 388 higher content of metastable phases, such as vaterite and aragonite [20]. Chen et al. 20 21 389 [48], obtained vaterite with traces of aragonite in a solution predominantly constituted 22 390 by ethanol, and pure aragonite when water predominated. In our closed system with 23 24 391 high RH (Fig.3a), vaterite only shows up after 21 days, and aragonite from 14 up to 28 25 26 392 days. Higher amount of vaterite, from 7 up to 21 days, and aragonite from 7 up to 28 27 393 days, is obtained when nanoparticles are exposed to the yeast system environment 28 29 394 (Fig.4a). In this latter case, there is a larger amount of alcohol that comes from the 30 31 395 nanoparticle’s product dispersion (2-propanol) plus the ethanol released from the yeast 32 396 fermentation. 33 34 35 397 Ethanol, isopropanol and diethylene glycol additives influence the morphology of the 36 37 398 formed vaterite crystals and stabilize this mineral phase by preventing the 38 399 transformation to calcite [49]. This could explain the worst definition of the maximum 39 40 400 intensity peaks associated to calcite at 21 days in the closed environment (Fig.3a), and 41 42 401 at 21 and 28 days in the yeast system (Fig.4a). Furthermore, the formation of ACC on 43 402 the faces of portlandite crystals hinders their further dissolution and the eventual 44 45 403 precipitation of crystalline calcite [50]. Once carbonation is completed, further exposure 46 47 404 to air or CO2 atmosphere will lead to the dissolution of newly formed CaCO3 in order to 48 405 neutralize the pore solution [28]. It is interesting to note that in this yeast system 49 50 406 environment there is an increase of 2°C (up to 20 ± 1 °C), due to the yeast 51 52 407 fermentation process, since the outdoor laboratory temperature condition was set up at 53 408 18 ± 1°C, which remains constant in the closed and open environments. Even though 54 55 409 calcite precipitation rates and total mineral precipitation increase with temperature, 56 57 410 these fall very quickly as calcite precipitates, being the proportion of calcite markedly 58 411 greater in a longer term in environments with lower temperatures [4]. 59 60 61 62 63 12 64 65 412 The abundance of metastable polymorphs (vaterite and aragonite) in both humid 1 413 systems can be related to an excess of alcohol in the environment (even higher in the 2 3 414 case of the yeast environment). Mono ethylene glycol also prolongs the transformation 4 5 415 time of metastable polymorphs by delaying the growth rate of more stable polymorphs 6 416 (calcite) [51]. Local fluctuations in the water/alcohol ratio also significantly affect the 7 8 417 precipitation/dissolution of anhydrous and hydrated polymorphs. Changes in lattice 9 10 418 parameters and particle size with exposure times, are related to surface tension 11 419 fluctuations, release of residual water and changes in the crystallinity of the particles 12 13 420 during carbonation, leading to differences in the nucleation and stability of each 14 15 421 polymorph [21]. 16 17 422 Rodriguez Navarro et al. [46] by combining FTIR and micro-Raman analyses found the 18 19 423 formation of calcium alkoxides ((Ca(OR)2) due to the reaction of ethanol and 2- 20 21 424 propanol with Ca(OH)2 particles. The hydrolysis of alkoxides in a humid environment 22 425 speeds up the carbonation process, increases the CaCO3 yield and favors the 23 24 426 formation of vaterite. It seems that these alkoxides are poorly crystalline or amorphous, 25 26 427 since these were undetectable under XRD. The yield depends on the reactivity of 27 428 Ca(OH)2 particles (surface area and lattice defects) and exposure time to the alcohol. 28 29 429 According to these authors, the contact of particles with alcohol (e.g. during storage of 30 31 430 dispersions) results in the replacement of Ca(OH)2 by calcium alcoxides. So, in freshly 32 431 made 2-propanol dispersions, the amount of newly formed calcium alcoxides should be 33 34 432 minimal, and after deposition of the particles carbonation takes place at a slower rate 35 36 433 compared to long contact periods (e.g. storage >> 2 months) where a fast and high 37 434 yield carbonation is produced. In our case, we used the same product of nanoparticles 38 39 435 dispersed in 2-propanol with more than 2 months storage, and hence we can compare 40 41 436 the differences produced in the three exposure environments. The high humidity 42 437 together with the extra alcohol (ethanol) and CO2 supplied by the yeast fermentation to 43 44 438 reach saturation could explain the most rapid carbonation yield in the yeast system. 45 46 439 The appearance of vaterite at 21 days in the closed environment and from 7 days in 47 the yeast system (with more amount of alcohol), and the signs of amorphization (by the 48 440 49 441 presence of ACC and alcoxides) observed in the XRD patterns of the three 50 51 442 environments indicate when the maximum reactions rates take place. 52 53 443 Calcium alcoxides have been used for consolidation of carbonate rocks and their 54 55 444 reactions with atmosphere produce CaCO3 where vaterite/calcite ratios generated from 56 57 445 isopropyl alcohol (2-propanol) solution were found to considerably vary with alkoxide 58 446 precursors [52]. Some alkoxides compounds after atmospheric exposure, give different 59 60 447 crystallographic results, ranging from amorphous deposits to crystalline systems with 61 62 63 13 64 65 448 vaterite only, vaterite-calcite mixtures, vaterite-portlandite mixtures, without great 1 449 reproducibility. This, together with alkoxides reactions are time-dependent, partially 2 3 450 explain why studies carried out by different authors using similar environmental 4 5 451 conditions, do not obtain the same mineralogical phases when the nanoparticles are 6 452 exposed to the atmospheric environment, since besides the type of alkoxide precursor 7 8 453 and storage time, there are several environmental variables that influence the final 9 10 454 precipitated product. 11 12 455 From the DTA-TG-dTG graphs (Fig. 2, Fig.3 and Fig.4) can be inferred that 13 14 456 carbonation degree from Ca(OH)2 nanoparticles (CaCO3 yield, from the relative amount 15 16 457 of Ca(OH)2 versus CaCO3) increases with the rise of RH and CO2 concentration of the 17 458 environment. After 28 days exposure, a rate of 85% Ca(OH)2 and 15% CaCO3 is 18 19 459 reached at laboratory conditions. Whereas, almost total carbonation is achieved under 20 21 460 the closed environment, with a rate of 5% Ca(OH)2 and 95 % CaCO3, and a total 22 461 carbonation, 100% CaCO3 yield is reached under the yeast fermentation system. 23 24 25 462 In the three environments, the thermal dehydroxylation reaction of portlandite at 28 26 27 463 days takes place at lower temperature ranges (between 307° and 450°C) compared to 28 464 the decomposition at 7 days of exposure (between 350° and 519°C). The 29 30 465 decarboxylation temperatures of calcite (with maximum endothermic peaks between 31 32 466 648ºC and 733ºC) are also lower than those usually reported (between 800 and 33 467 900°C). This can be related with the size of the nanocrystals, which thermal 34 35 468 decomposition is produced earlier compared to larger crystal sizes. 36 37 38 469 The DTA endothermic peak around 100°C is due to hygroscopic water (i.e. physically 39 470 adsorbed water), whereas the dTG peak appearing at about 250°C can be attributed to 40 41 471 bound water or to hydrated interlayer cations [53], since at around 241ºC the 42 43 472 transformation of ACC into CaCO3 takes place [28]. The thermal decomposition

44 473 associated to dehydroxylation of portlandite might be influenced by polymorphic CaCO3 45 46 474 phase transitions. The phase transformation of vaterite into calcite can takes place at a 47 48 475 maximum reaction rate ca. 457°C, whereas transformation of aragonite into calcite can 49 476 occurs ca. 430°C [54]. After 28 days in the closed system, portlandite was identified 50 51 477 under XRD and an endothermic peak with a mass loss of 5% was also detected under 52 53 478 DTA-TG. In the yeast system, no portlandite was identified under XRD and none 54 479 endothermic effect by dehydroxylation was found under DTA. The slight 2% mass loss 55 56 480 detected by TG, might corresponds to the loss of occluded water contained in mineral 57 58 481 aragonite due to the aragonite-calcite phase transition [54]. Therefore, the 59 482 decarboxylation in this case corresponds to 100% CaCO3 decomposition and 100% 60 61 62 63 14 64 65 483 CaCO3 yield in the samples exposed to the yeast fermentation system after 28 days, 1 484 due to the combination of CO2, high humidity and more amount of alcohol in the 2 3 485 environment, that speed up the carbonation reactions. 4 5 6 486 3.3 Characterization of stone specimens 7 8 487 3.3.1 Stone blank specimens 9 10 11 488 The physical properties of blanks specimens from both types of stones were firstly 12 489 characterized in order to compare the results between stone specimens with no 13 14 490 nanoparticles (Table 1) and specimens consolidated with the nanoparticles at a 15 16 491 concentration of 5g/l and 25 g/l and exposed to the three environmental conditions. 17 18 492 The surface of Lecce stone and Bateig stone blank specimens was observed under 19 20 493 SEM. The SEM images show a rough surface with pores of micrometric size. The 21 22 494 mercury intrusion porosimetry curves display the pore size distribution (PSD) of both 23 495 types of stones (Fig. 5). The mercury intrusion porosimetry results give a total open 24 25 496 porosity accessible to mercury of 31±5% in Lecce stone and 19±0.5% in Bateig stone. 26 27 497 The PSD of Lecce stone is mainly a bimodal distribution (Fig.5a). Circa half of the 28 498 amount of total porosity is bellow 1m (Table 1), and corresponds to pores in the range 29 30 499 between 0.1m and 1m (41±5%), between 0.01µm - 0.1µm (7±2%) and < 0.01 µm 31 32 500 (1%). The other half of amount of pores is above 1m, and is composed of pores in the 33 34 501 range between 1m and 10 m (49±2%) and pores >100 µm (1%). Whereas, PSD of 35 Bateig stone displays a polymodal distribution (Fig.5b) with most of the pores above 1 36 502 37 503 µm, with pores in the range between 1m and 10 m (58±3%), between 10 - 100 µm 38 39 504 (4±1%) and >100 µm (4±2%). Bellow 1m it displays pores in the range between 40 41 505 0.1m and 1m (22%), between 0.01µm - 0.1µm (12%) and < 0.01 µm (1%). 42 43 506 Spectrophotometry results and water absorption values obtained under vacuum are 44 45 507 also compiled in Table 1. The water absorption by capillarity and desorption of Lecce 46 2 0.5 47 508 stone show a capillarity coefficient of 98±4 g/(m s ) and water desorption rate of 24±2 48 509 g/(m2s0.5). Whereas Bateig stone display a capillarity coefficient of 41±0 g/(m2s0.5) and 49 50 510 water desorption rate of 15±0 g/(m2s0.5). Lecce stone shows a higher open porosity 51 52 511 values and a larger amount of absorption water by capillarity, which absorption and 53 512 desorption rates are produced at a faster rate compared to Bateig stone. 54 55 56 513 The results of surface hardness measurements indicate values of 234±21 HLD for 57 58 514 Lecce stone blank specimens and much higher values of 342±30 HLD for Bateig stone. 59 60 61 62 63 15 64 65 515 Microdrilling resistance (DRMS) measurements indicate values of 3.96±0.43 N for 1 516 Lecce stone and three times higher force values for Bateig stone (9.50±1.11 N). 2 3 4 517 3.3.2 Treated stone specimens exposed to ventilate laboratory room conditions 5 6 The main results from the spectrophotometry measurements show clear differences 7 518 8 519 between the two product´s concentration in both types of stones (Table 2). Figure 6 9 10 520 shows the macroscopic view of the stone surfaces after the application of Ca(OH)2 11 12 521 nanoparticles at 5g/l and 25 g/l, together with the SEM images obtained at 1000x and 13 522 6000x magnification. The macroscopic images of the stone surfaces after the 14 15 523 application of Ca(OH)2 nanoparticles show a white porous crust on the face where the 16 17 524 product was absorbed by capillarity, when the nanoparticles were applied at a 18 525 concentration of 25g/l in both types of stone (Fig. 6d and 6j). On the surface of Lecce 19 20 526 (Fig. 6e) stone a network of nano and micropores left by agglomerated flat 21 22 527 nanocrystals with sings of amorphization can be observed, together with some totally 23 528 amorphous particles that would correspond to amorphous calcium carbonate (Fig.6f). 24 25 529 On the surface of Bateig stone (Fig.6k), a similar crust shows smooth areas and cracks 26 27 530 in a network of agglomerated granular nanocrystals (Fig.6l) that also give rise to a 28 531 network of micropores. In the case of nanoparticles applied at a concentration of 5g/l, 29 30 532 small spots of white deposits can be observed in the Lecce stone (Fig. 6a) and the 31 32 533 microscopic SEM images show the nanoparticles covering the surface of the original 33 534 minerals of the stone (Fig. 6b and Fig. 6c). In the case of Bateig stone, the 34 35 535 macroscopic view apparently does not show white deposits (Fig. 6g). However, the 36 37 536 nanoparticles have covered the surface of the minerals with a smooth film (Fig. 6h) of 38 537 amorphous flat layers (Fig. 6i). The EDS analyses show that the chemical composition 39 40 538 of the nanoparticles is mostly made up by calcium, as was previously studied in other 41 42 539 research works [7, 20]. 43 44 540 There are almost no changes on the average density, open porosity, and water 45 46 541 saturation values. Regarding to the capillarity results, there is just a slight change on 47 48 542 the water absorption and desorption capillarity coefficients. These variations are a bit 49 543 higher in the Bateig stone specimens (Table 3). All the stone specimens display 50 51 544 changes in the surface hardness, with an increase of 5% and 8% for Lecce stone 52 53 545 consolidated at 5g/l and 25g/l, respectively. For Bateig stone, the values slightly 54 546 increase 1% and 4% when the nanoparticles were applied at a concentration of 5g/l 55 56 547 and 25g/l, respectively. The DRMS results show some changes in the drilling 57 58 548 resistance in all the stone specimens, with an increase of 19% and 11% in Lecce stone 59 549 and 1% and 5% in Bateig stone (Table 4). 60 61 62 63 16 64 65 550 3.3.3 Treated stone specimens exposed into closed containers with high humidity 1 2 551 The main results from the spectrophotometry measurements show slight differences 3 4 552 between the two product´s concentration in both types of stones (Table 2). Figure 7 5 6 553 shows the macroscopic view of the stone surfaces after the application of Ca(OH)2 7 554 nanoparticles, together with their SEM images. When the nanoparticles were applied at 8 9 555 a concentration of 5g/l, Lecce stone shows small spots of white deposits (Fig. 7a) that 10 11 556 cover the surface of the original minerals as an amorphous film (Fig. 7b) where flat 12 557 particles can barely be distinguished (Fig. 7c). In the case of Bateig stone, the 13 14 558 macroscopic view apparently does not show white deposits (Fig. 7g), but the 15 16 559 nanoparticles have covered the surface of some original minerals also with a smooth 17 560 film (Fig. 7h) of amorphous flat layers (Fig.7i). When the nanoparticles were applied at 18 19 561 a concentration of 25g/l, the stone specimens show white crust deposits on the edges 20 21 562 and corners of the faces where the product was absorbed by capillarity (Fig.7d and 7j). 22 563 In Lecce stone, the white deposits are formed by agglomerated granular nanocrystals 23 24 564 (Fig.7e) that give rise to a network of pores with nanometric sizes (Fig.7f). In Bateig 25 26 565 stone the nanoparticles are less agglomerated and these are covering the surface of 27 566 the stone (Fig. 7k and Fig. 7l). The EDS analyses show that the chemical composition 28 29 567 of the nanoparticles is mainly formed by calcium. 30 31 32 568 There are almost no changes on the average density, open porosity, and water 33 569 saturation values. There is just a slight change on the water absorption and desorption 34 35 570 coefficients, and these variations are a bit higher in the Bateig stone specimens (Table 36 37 571 3). All the stone specimens display a change in the surface hardness, with 8% and 7% 38 572 increase in Lecce stone consolidated at 5g/l and 25g/l respectively. In Bateig stone the 39 40 573 values barely change (3% and 5%, respectively). Both types of stone specimens show 41 42 574 a slight increase in the drilling resistance results, with 9% and 17% increase in Lecce 43 575 stone and 2% and 8% in Bateig stone (Table 4). 44 45 46 576 3.3.4 Treated stone specimens exposed to a yeast system with high humidity 47 48 577 The main results from the spectrophotometry measurements show very slight 49 50 578 differences between the two product´s concentration in both types of stones (Table 2). 51 52 579 Figure 8 shows the macroscopic view of the stone surfaces after the application of 53 Ca(OH) nanoparticles, together with their SEM images. At 5g/l Lecce stone shows 54 580 2 55 581 very few small white spots (Fig. 8a) and the SEM images display large amounts of 56 57 582 ACC together with nano and microcrystals of calcium carbonate (Fig. 8b and Fig.8c). In 58 Bateig stone, the macroscopic view apparently does not show white deposits (Fig. 8g), 59 583 60 584 but under SEM the surface of the stone appears covered by an amorphous film 61 62 63 17 64 65 585 (Fig.8h) where agglomerated flat particles can be distinguished (Fig. 8i). In general, no 1 586 white deposits are observed on the surface of the stones, except on Bateig specimens 2 3 587 that show a white glazing on the center of the face where the product was applied at 25 4 5 588 g/l (Fig. 8j). In this case, elongated calcium carbonate and some rhombohedra calcite 6 589 microcrystals give rise to a network of micropores, together with some small areas filled 7 8 590 with ACC (Fig. 8k and 8l). At this concentration, very small white spots can be 9 10 591 observed on the surface of Lecce stone (Fig. 8d). At 25g/l the nanoparticles have 11 592 covered the surface (Fig. 8e) and calcium carbonate nano and microcrystals (some 12 13 593 calcite rhombohedra) can be observed together with small areas filled with ACC (Fig. 14 15 594 8f). The chemical composition of these nano and microcrystals shows EDS spectra 16 595 composed by calcium. 17 18 19 596 There are almost no changes on the average density, open porosity, and water 20 21 597 saturation values. The water absorption and desorption coefficients of Lecce stone 22 598 specimens slightly increase, whereas these variations are a bit higher in Bateig stone 23 24 599 (Table 3). In this case, all the stone specimens display changes in the surface 25 26 600 hardness, with 10% and 11% increase in Lecce stone consolidated at 5g/l and 25g/l 27 601 respectively. In Bateig stone, these values slightly change (6% and 8% respectively). 28 29 602 Both types of stones show an increase in the drilling resistance results, 8% and 16% in 30 31 603 Lecce stone and 5% and 10% in Bateig stone (Table 4). 32 33 604 3.4 Interpretation of characterization of stone specimens 34 35 36 605 The results of color variations obtained by spectrophotometry on the surface of the 37 38 606 treated stone specimens show clear differences in the chromatic parameters 39 607 depending on the substrate, environment and concentration of the product (Table 2). In 40 41 608 Lecce stone specimens consolidated at 5 g/l, the chromatic differences decrease with 42 43 609 the rise of RH and CO2 in the environment. Lightness (ΔL*), white index (ΔWI*) and 44 610 brightness are the parameters with the highest increase due to the white color left by 45 46 611 the nanoparticles. It should be highlighted that the magnitude of the standard 47 48 612 deviations produced in the chromatic variations decrease with the increase of RH and 49 613 CO2 in the environment, giving rise to more homogeneous chromatic surfaces, where 50 51 614 the variations are produced in a similar way in the whole surface. However, in Bateig 52 53 615 stone specimens consolidated at 5 g/l, the total color difference (ΔE*) increases with 54 616 the increase of RH and CO2 of the environment. The differences on chromatic 55 56 617 variations after consolidation, besides the porosity, can be explained because the 57 58 618 application of nanoparticles produce an increase of WI, which rises with the 59 60 61 62 63 18 64 65 619 carbonation rate, being more remarkable in Bateig stone with an original lower WI 1 620 compared to Lecce stone. 2 3 4 621 Due to the suitability criteria used to asses conservation treatments, the variations 5 6 622 produced at 5 g/l would not be visually detectable by the human eye and these would 7 623 not significantly affect the colorimetric parameters of the substrate in any case, 8 9 624 because ΔE* is lower than 5 [55], or lower than 3 according to other authors [56,57]. In 10 11 625 the case of treated stones at a concentration of 25g/l, the chromatic variations are 12 626 greater than the suitability criteria in both type of substrates, producing detectable 13 14 627 visually changes. Both in Lecce and Bateig stone specimens, ΔE* is much greater in 15 16 628 the case of the specimens exposed to the laboratory conditions. ΔE* value decreases 17 629 in the closed environment and increase again in the yeast system environment, 18 19 630 although ΔE* is still lower than the difference produced in the laboratory environment. 20 21 631 In both stone types, an agglomeration of nanoparticles is produced on the surface of 22 632 the substrates giving rise to a homogeneous white crust that makes those parameters 23 24 633 to increase (Fig.6). In the closed environment, small agglomerations are produced, 25 26 634 giving rise to small crusts heterogeneously distributed on the surface (Fig.7). However, 27 635 in the yeast system environment there are no crusts, just a white glazing only observed 28 29 636 in Bateig stone at 25 g/l (Fig. 8). This effect is produced by initial bigger pores sizes 30 31 637 and lower open porosity values compared to Lecce, giving rise to higher chromatic 32 638 variations. 33 34 35 639 The SEM images (Fig. 6-8) show that the nanoparticles are aggregated and display 36 37 640 signs of amorphization. Moorehead [58], reported that the increase of CO2 38 641 concentration during carbonate cementation increase the rate of transformation from 39 40 642 portlandite into calcite, showing how the carbonation of hydrated lime (Ca(OH)2) gives 41 42 643 rise to mostly amorphous or very poor crystalline forms of CaCO3. This was also 43 644 observed in Ca(OH)2 nanoparticles showing ACC formation on the faces of portlandite 44 45 645 [20,21,50]. This can be seen in the SEM images, especially in the cases of 46 47 646 nanoparticles applied at 5g/l and exposed to the three environments, where nano and 48 647 microcrystals together with large amounts of ACC are observed. Colloidal particles may 49 50 648 undergo aggregation of plate-like Ca(OH)2 nanoparticles ranged from 30 nm up to 200 51 52 649 nm that aggregate into micron-sized clusters [20,21,59] 53 54 650 The initial higher open porosity of Lecce stone (Table 1) has favor a higher penetration 55 56 651 of the product, avoiding its saturation on the surface and the agglomeration of the 57 58 652 nanoparticles, with less white deposits left on the surface, especially in the yeast 59 653 system environment. Bulk and real density values barely change in both stone types 60 61 62 63 19 64 65 654 exposed to the three environments. The open porosity values slightly decrease in a 1 655 similar way (3% in Lecce stone and 5% in Bateig stone) in the closed and yeast system 2 3 656 environments, whereas, no changes are produced in the laboratory environment. 4 5 657 These results are in agreement with previous works, both in a short [7] and in a longer 6 658 term [16], where it was demonstrated that consolidation process with Ca(OH)2 7 8 659 nanoparticles improved the physical and hydric properties of limestone specimens 9 10 660 exposed to dry and humid environments. An initial high RH and a higher and larger 11 661 porosity favor the consolidation in a longer term. 12 13 14 662 After consolidation, water absorption and desorption capillarity coefficients barely 15 16 663 change (Table 3). Regarding to the quantity of absorbed water in Lecce and Bateig 17 664 stone specimens consolidated at 5g/l, a 3% and 6% increase is produced in the three 18 19 665 environments. Whereas, no changes are obtained at a concentration of 25g/l, with the 20 21 666 exception of specimens exposed to the laboratory environment (6% increase). 22 667 Although the values are variable, generally higher variations in the capillarity 23 24 668 coefficients (C) are produced in Bateig specimens, especially at 25 g/l. An increase in 25 26 669 the C coefficient means that specimens absorb water at a faster rate. Due to the 27 670 decrease of pore sizes, the suction force increases, so the higher the volume of 28 29 671 capillary pores the higher the C coefficient [60]. This could be related to the generation 30 31 672 of micropores with a diameter in the range between 0.1 and 1µm at the expense of 32 673 pores between 1 and 10µm and between 10 and 100µm. So, the sizes within these 33 34 674 latter ranges are reduced by the application of nanoparticles and give rise to higher C 35 36 675 coefficients, especially in the case of Bateig stone with higher amount of pores in these 37 676 ranges (Table 1). This supposition is in agreement with the MIP data obtained in 38 39 677 previous works before and after application of same type of nanoparticles under similar 40 41 678 conditions in dolostones specimens [16]. When comparing the PSD of the blanks of 42 679 both stones (Fig.5) and the C coefficients of both stones (Table 3), the main difference 43 44 680 can be found in the amount pores in the range between 0.1 and 1µm. In the case of 45 46 681 Lecce stone, this amount is approximately double than the obtained in Bateig stone for 47 this range, as well as the C coefficient and desorption rates. Therefore, it seems that 48 682 49 683 the ranges of pores that mostly influence the C coefficient are those comprised 50 51 684 between 0.1 and 1 µm. The results on capillary desorption rates of Lecce stone 52 specimens after consolidation, (Table 3) show a slight increase of this coefficient, i.e. 53 685 54 686 increase of evaporation rate of capillarity water, which supposes lower water retention 55 56 687 during drying. However, in Bateig specimens, the capillarity desorption rate is reduced 57 58 688 in the three environments, independently of the product´s concentration. Although, as

59 689 the same time the environment has a higher humidity and CO2 concentration, the water 60 61 62 63 20 64 65 690 evaporation is speed up, from 13% decrease in the laboratory environment down to 7% 1 691 decrease in the yeast system environment. Comparing the results obtained at both 2 3 692 concentrations of nanoparticles depending on the consolidation environment, the 4 5 693 improvement of hydric properties after consolidation (decrease of amount of absorbed 6 694 water and increase of capillarity and desorption coefficients) is obtained in the yeast 7 8 695 system environment. Nevertheless, the slight increase in the amount of absorbed water 9 10 696 (more pronounced in Bateig stone) and the decrease in the water evaporation rate (in 11 697 Bateig specimens) should be assessed. Although these are minor variations, they 12 13 698 could have adverse effects in the durability of the stone, because the presence of 14 15 699 water inside the rocks gives rise to a durability decrease since it can favor salt and ice 16 700 crystallization processes, dissolution, adsorption of pollution particles and 17 18 701 biodeterioration [61]. 19 20 21 702 In a similar way to the spectrophotometry results, the decrease of standard deviation 22 703 on the surface hardness values after consolidation (Table 1 and Table 4) suggests a 23 24 704 homogenization of the surface of the stone specimens and hence of the surface 25 26 705 hardness with almost none increase of this value in the specimens exposed to the 27 706 laboratory and closed environments. However, in the yeast system surface hardness 28 29 707 increases at both concentrations of the product, and this is higher in Lecce compared 30 31 708 to Bateig. The drilling resistance values also increase after consolidation of Lecce 32 709 specimens, especially at 25g/l. However, in Bateig specimens these values barely 33 34 710 change. The microdrilling results can be affected by the presence of microfossils that in 35 36 711 the case of Lecce stone provoke a higher heterogeneity of the measurements 37 712 compared to Bateig stone. This latter stone shows a homogeneous drilling resistance 38 39 713 increase with the rise of humidity and CO2 of the environment and with the increase of 40 41 714 the product’s concentration. 42 43 715 According to Moorehead [58], the main factors involved in controlling the carbonation 44 45 716 reaction, are CO2 gas concentration, moisture content and the permeability of the 46 47 717 substrate. From the SEM images, it can be inferred that porosity and permeability 48 718 values are more favorable and prone to the penetration of these nanoparticles in the 49 50 719 case of Lecce stone, especially when the concentration of the product is higher (25g/l) 51 52 720 and is exposed to the yeast system environment (Fig.6, Fig.7 and Fig.8). At 5g/l both 53 721 stone types apparently better absorb the product because no white deposits seem to 54 55 722 be left over the surface of the specimens in the three environments. However, at this 56 57 723 concentration the mechanical properties barely change in Bateig stone compared to 58 724 Lecce stone. Even though the changes in the mechanical properties are higher at 25g/l 59 60 725 for both stone types, this increase is not directly proportional to the product´s 61 62 63 21 64 65 726 concentration, being these properties just slightly increased. Besides, the white 1 727 deposits left by the nanoparticles are larger when the product is applied at 25g/l 2 3 728 concentration, especially in Bateig stone, because it has larger pore sizes but lower 4 5 729 total open porosity and C coefficients compared to Lecce stone. Eventually, the 6 730 appearance of these white deposits decrease from the laboratory to the closed 7 8 731 environment, and these are almost inexistent in the yeast system environment, 9 10 732 especially in Lecce stone. A second application at 5g/l (two applications instead of just 11 733 one) should be studied, after the carbonation of the first one, 28 days later, to know if is 12 13 734 there a similar increase on the mechanical properties, or even higher, that the obtained 14 15 735 with concentration at 25g/l but avoiding the formation of white deposits left on the 16 736 surface. 17 18 19 737 MIP results of Lecce stone show a higher total connected porosity and higher amount 20 21 738 of pores with smaller pore diameters (below 1 m, and mainly between 0.1 and 1m) 22 739 compared to Bateig. According to all the obtained results on the different physical 23 24 740 properties, it seems that the size of these nanoparticles (60-130 nm) better fit with the 25 26 741 bimodal PSD of Lecce stone compared to the polymodal PSD of Bateig stone (with 27 28 742 higher amount of pores above 1m). The size of this type of nanoparticles can increase 29 743 from nanometer sizes (<100 nm) up to micrometric size (2 m) at high RH [20]. 30 31 744 Besides, a higher open and more connected porosity improve the access of CO2 and 32 33 745 H2O to the interior of the stone, and therefore ease the carbonation process. It is 34 746 important not only to take into account the open porosity values of the substrate, but 35 36 747 the PSD as well, since depending on the size of the initial nanoparticles and the final 37 38 748 formed crystals, related to the environmental conditions during the crystallization 39 749 process, the filling of different pore sizes can be better or worse, increasing or 40 41 750 decreasing the consolidating/cementing efficacy of the nanoparticles. 42 43 44 751 The carbonation process and the improvement in the petrophysical properties seem to 45 752 be stopped (or slowed down) at low RH, whereas these processes keep going (or are 46 47 753 speed up) in a longer term at higher RH. So, the influence of the RH of the 48 49 754 environment during the first stage of consolidation is a key factor for the transformation 50 755 of Ca(OH)2 nanoparticles into CaCO3 and for the improvement of petrophysical 51 52 756 properties of the treated stones in a longer term [16]. Besides the high humidity, the 53 54 757 addition of CO2 and a source of ethanol also speed up the carbonation rate and the 55 758 mechanical properties of limestone. However, accelerated ageing tests should be the 56 57 759 next step in order to assessing the durability of the stones initially exposed to these 58 59 760 different environments. 60 61 62 63 22 64 65 761 4 Conclusions 1 2 762 A high humidity environment together with the CO2 and ethanol released by yeast 3 4 763 fermentation favors and accelerates the carbonation process of colloidal Ca(OH)2 5 6 764 nanoparticles (60-130 nm) dispersed in 2-propanol. These nanoparticles are 7 765 transformed into vaterite, aragonite and calcite (CaCO3 polymorphs), after 21 days of 8 9 766 exposure to this system, and achieve 100% CaCO3 yield (aragonite and calcite) after 10 11 767 28 days. Samples under high humidity but without the yeast system, display the same 12 768 mineral phases together with portlandite after 21 days, and reach 95% CaCO3 yield 13 14 769 (aragonite and calcite) vs. 5% Ca(OH)2 (portlandite) after 28 days. Whereas samples 15 16 770 exposed to laboratory room conditions barely carbonate, displaying almost the total 17 771 amount of initial portlandite nanoparticles, 85% Ca(OH)2 (portlandite) vs. 15% CaCO3 18 19 772 yield (calcite) after 28 days. 20 21 22 773 In the three environments all the limestone specimens consolidated with these 23 774 nanoparticles show an increase in the drilling resistance and surface hardness values 24 25 775 with variable results in the capillarity absorption and desorption coefficients, especially 26 27 776 when the nanoparticles were applied at a concentration of 25g/l. Besides, at this 28 777 concentration white crusts and glazing are left by the nanoparticles on the surface of 29 30 778 the stones, composed of a mixture of amorphous, nano and microcrystals of CaCO3. 31 32 The largest increase in these mechanical properties, with slight rise in the capillarity 33 779 34 780 absorption and desorption coefficients and with lesser stone color changes (with total 35 36 781 color differences lower than the established for stone conservation treatments) are 37 38 782 produced at a concentration of 5g/l, in the yeast system environment and especially in 39 783 stone specimens initially with bimodal pore size distributions (PSD), more amounts of 40 41 784 pores in the range between 0.1 and 1µm diameters, higher open porosity values and 42 43 785 faster capillary coefficients compared to Bateig stone with a polymodal PSD and more 44 786 amount of pores above 1µm. In both types of stone, density and open porosity values 45 46 787 barely change. 47 48 788 This research show the importance of controlling the environmental conditions during 49 50 789 the phase transformations of these nanoparticles used as cementing materials and 51 52 790 during their application as a consolidating product, both in laboratory and in the field. A 53 simple, inexpensive and reliable method that can be easily applied in the laboratory 54 791 55 792 and on-site, based on water and a yeast-sugar solution to create a microclimate 56 57 793 saturated on high humidity, CO2 and ethanol released to speed up carbonation of 58 Ca(OH) nanoparticles, is presented to improve the mechanical properties of decayed 59 794 2 60 795 limestone frequently used in archaeological and architectural heritage. 61 62 63 23 64 65 796 Acknowledgements 1 2 797 This work was carried out at Instituto de Geociencias (CSIC,UCM) and supported by Rafael 3 4 798 Fort and GEOMATERIALES (S2009/MAT-1629) Program and by a JAE-PreDoc CSIC 5 799 fellowship founded by the European Social Fund FSE 2007–2013. Thanks to Iván Serrano for 6 7 800 his help with the XRD analyses at the Department of Petrology and Geochemistry, Faculty of 8 801 Geology (UCM) and to Xabier Arroyo for the DTA-TG analyses at the CAI of Faculty of Geology 9 10 802 (UCM). Thanks also to Ana Vicente for the SEM-EDS analyses performed at the ICTS, Centro 11 803 Nacional de Microscopia Electronica (UCM) and Dr. Maria Jose Varas (UCM) for founding the 12 13 804 SEM analyses. Special thanks go to the technician Andres Lira from IGEO (CSIC, UCM) for his 14 805 useful ideas and valuable help with the set up to perform part of this research. 15 16 17 806 18 19 807 References 20 21 22 808 1. W. Suchaneka, M. Yoshimuraa, J Mater Res 13, 94–117 (1998) 23 24 809 2. M. Obst, J.J. Dynes, J.R. Lawrence, G.D.W. Swerhone, C. Karunakaran, K.V. Kaznatcheev, 25 D. Bertwistle, K. Benzerara, T. Tyliszczak, A.P. Hitchcock, Geochim Cosmochim Ac 73(14), 26 810 27 811 4180–4198 (2009) 28 29 812 3. D.W.S. Ho, R.K. Lewis, Cement Concrete Res 17, 489-504 (1987) 30 31 32 813 4. S. Sanchez-Moral, J. Garcia-Guinea, L. Luque, R. Gonzalez-Martin, P. Lopez-Arce, Mater 33 814 Construcc 54(275), 23-37 (2003) 34 35 36 815 5. G. Cultrone, E. Sebastian, M. Ortega Huertas, Cement Concrete Res 35, 2278– 2289 (2005) 37 38 816 6. E. Doehne, Clifford A. Price, Stone Conservation: An Overview of Current Research, 2nd edn. 39 40 817 (Getty Publications, Los Angeles, 2010), p. 176 41 42 818 7. P. López-Arce, L.S. Gomez-Villalba, L. Pinho, M.E. Fernández-Valle, M. Álvarez de Buergo, 43 44 819 R. Fort, Mater Charact 61(2),168–184 (2010) 45 46 820 8. J. Ashurst, in Conservation of Building and Decorative Stone: Methods of Repairing and 47 48 821 Consolidating Stone Buildings, eds. By J. Ashurst, F.G. Dimes (Butterworth-Heinemann, 49 822 London, 1990), pp. 1-54 50 51 823 9. K. Elert, C. Rodríguez-Navarro, E. Sebastián, E. Hansen, O. Cazalla, Stud Conserv 47(1), 52 53 824 62–75 (2002) 54 55 825 10. P. Tiano, Coalition 6(2), 4-8 (2003) 56 57 58 826 11. W. De Muynck, N. De Belie, W. Verstraete, Ecol Eng 36(2), 118-136 (2010) 59 60 61 62 63 24 64 65 827 12. J.P. Adolphe, J.F. Loubière, J. Paradas, F. Soleilhavoup, Procédé de traitement biologique 1 828 d’une surface artificielle, European patent 90400G97.0. (after French patent 8903517, 1989) 2 3 829 (1990) 4 5 830 13. C. Rodriguez-Navarro, M. Rodriguez-Gallego, K. Ben Chekroun, M.T. González-Muñoz, 6 7 831 Appl Environ Microbiol 69(4), 2182–2193 (2003) 8 9 832 14. S. Sequeira, C. Casanova, E.J. Cabrita, J Cult Herit 7(4), 264–272 (2006) 10 11 833 15. P. Baglioni, D. Chelazzi, R. Giorgi, G. Poggi, Langmuir 29, 5110-5122 (2013) 12 13 14 834 16. P. López-Arce, A. Zornoza-Indart, L. Gomez-Villalba, R. Fort, J Mater Civil Eng 25(11), 15 835 1655–1665 (2013) 16 17 18 836 17. A. Arizzi, L.S. Gomez-Villalba, P. Lopez-Arce, G. Cultrone, R. Fort, Eur J Mineral. 19 837 doi:10.1127/ejm/2015/0027-2437 (2015) 20 21 22 838 18. R. Giorgi, L. Dei, P. Baglioni, Stud Conserv 45, 154-161(2000) 23 24 839 19. V. Daniele, G. Taglieri, R. Quaresima, J Cult Herit 9, 294-301 (2008) 25 26 840 20. P. Lopez-Arce, L.S. Gomez-Villalba, S. Martinez-Ramirez, M. Alvarez de Buergo, R. Powder 27 28 841 Technol 205(1), 263–269 (2011) 29 30 842 21. L.S. Gomez-Villalba, P. Lopez-Arce, M. Alvarez de Buergo, R. Fort, Appl Phys A-Mater 104, 31 32 843 1249–1254 (2011) 33 34 844 22. T. Yang, B. Keller, E. Magyari, K. Hametner, D. Günther, J Mater Sci 38, 1909-1916 (2003) 35 36 37 845 23. D.T. Beruto, R. Botter, J Eur Ceram Soc 20, 497−503 (2000) 38 39 846 24. K. Van Balen, D. Van Gemert, Mater Struct 27, 393– 398 (1994) 40 41 42 847 25. K. Van Balen, Cement Concrete Res 35, 647– 657 (2005) 43 44 848 26. R.M. Dheilly, J. Tudo, Y. Sebai bi, M. Quéneudec, Constr Build Mater 16, 155–161 (2002) 45 46 849 27. J.H.M. Visser, Constr Build Mater 67, 8-13 (2014) 47 48 49 850 28. Ö. Cizer, C. Rodriguez-Navarro, E. Ruiz-Agudo, J. Elsen, D. Van Gemert, K. Van Balen, J 50 851 Mater Sci 47, 6151-6165 (2012) 51 52 53 852 29. R.C. Smallegange, W.H. Schmied, K.J. Van Roey, N.O. Verhulst, J. Spitzen, W.R. 54 853 Mukabana, W. Takken, Malaria J 9(292), 1-15 (2010) 55 56 57 854 30. Y. Saitoh, J. Hattori, S. Chinone, N. Nihei, Y. Tsuda, H. Kurahashi, M. Kobayashi, J Am 58 855 Mosquito Contr 20, 261-264 (2004) 59 60 61 62 63 25 64 65 856 31. J.P. Van Dijken, R.A. Weusthuis, J.T. Pronk, Anton Leeuw Int J G, 63(3-4), 343-352 (1993) 1 2 857 32. J.A. Barnett, Yeast 20(6), 509-543 (2003) 3 4 858 33. G. Walker, P. Dijck, in Yeasts in Food and Beverages: Physiological and molecular 5 6 859 responses of yeasts to the environment, eds. By A. Querol, G.H. Fleet (Springer-Verlag, 7 Berlin, 2006), pp 111-152 8 860 9 10 861 34. L. Dei, B. Salvadori, J Cult Herit 7(2), 110–115 (2006) 11 12 862 35. G. Ziegenbald, in Proceedings of 11th International Congress on Deterioration and 13 14 863 Conservation of Stone III (2008), p. 1109 15 16 864 36. L.S. Gomez-Villalba, P. Lopez-Arce, R. Fort, Appl Phys A-Mater 106(1), 213-217 (2012) 17 18 19 865 37. A. Calia, M. Laurenzi Tabasso, A.M. Mecchi, G. Quarta, in Stone in Historic Buildings: 20 866 Characterization and Performance, The study of stone for conservation purposes: Lecce 21 22 867 stone (southern Italy), eds. By Cassar et al., (Special Publications, Geological Society, 23 868 London, 2013), 391 24 25 26 869 38. E. Vasanelli, M. Sileo, A. Calia, A.M. Aiello, Procedia Chemistry 8, 35-44 (2013) 27 28 870 39. A. La Iglesia, M.A. Garcia del Cura, S. Ordonez, A. Bernabeu, Geogaceta 23, 79-82 (1998) 29 30 871 40. Bateig Piedra Natural SA, Determinacion de la densidad aparente y porosidad abierta UNE- 31 32 872 EN 1936:2007, Report 142208 (Aidico, Instituto Tecnologico de la Construccion, Alicante, 33 873 2014) 34 35 36 874 41. ASTM E313-73, Practice for calculating yellowness and whiteness indices from 37 875 instrumentally measured color coordinates (1993) 38 39 40 876 42. ISO 2470-2, Paper, board and pulps–Measurement of diffuse blue reflectance factor (ISO 41 877 brightness) (1999) 42 43 44 878 43. AENOR UNE-EN 1936, Natural stone test methods. Determination of real density and 45 879 apparent density, and of total and open porosity (2007) 46 47 48 880 44. AENOR UNE-EN 15801, Conservation of cultural property- Test methods- Determination of 49 881 water absorption by capillarity (2010) 50 51 52 882 45. ASTM A956-12, Standard Test Method for Leeb Hardness Testing of Steel Products (2006) 53 54 883 46. C. Rodriguez-Navarro, A. Suzuki, E. Ruiz-Agudo, Langmuir 29, 11457-11470 (2013) 55 56 884 47. L.S. Gomez-Villalba, P. Lopez-Arce, M. Alvarez de Buergo, R. Fort, Cryst Growth Des 57 58 885 12(10), 4844-4852 (2012) 59 60 886 48. S.F. Chen, S.H. Yu, J. Jiang, F. Li, Y. Liu, Chem Mater 18, 115–122 (2006) 61 62 63 26 64 65 887 49. F Manoli, E Dalas, J Cryst Growth 218, 359–364 (2000) 1 2 888 50. Ö. Cizer, K. Van Balen, J. Elsen, D. Van Gemert, Constr Build Mater 35, 741–751 (2012) 3 4 889 51. E.M. Flaten, M. Seiersten, J.P. Andreassen, J Cryst Growth 311(13), 3533–3538 (2009) 5 6 7 890 52. F. Ossola, P. Tomasin, C. De Zorzi, N. El Habra, M. Chiurato, M. Favaro, New J Chem 36, 8 891 2618–2624 (2012) 9 10 11 892 53. A. Moropoulou, A. Bakolas, K. Bisbikou, Thermochim. Acta. 269, 779-795. (1995) 12 13 893 54. J. Perić, M. Vučak, R. Krstulović, Lj. Brečević, D. Kralj, Thermochim. Acta. 277, 175–176 14 15 894 (1996) 16 17 895 55. NORMAL 20/85, Interventi conservativi: progettazione esecuzione e valutazione preventive 18 19 896 (1996) 20 21 897 56. D. Benavente, F. Martinez-Verdu, A. Bernabeu, V. Viqueira, R. Fort, M.A. Garcia del Cura, 22 23 898 C. Illueca, S. Ordoñez, Color Res Appl 28, 343-351 (2003) 24 25 899 57. J.D. Rodrigues, A. Grossi, J Cult Herit 8, 32-43 (2007) 26 27 900 58. D.R. Moorehead, Cement Concrete Res 16(5), 700–708 (1986) 28 29 30 901 59. C. Rodriguez-Navarro, E. Ruiz-Agudo, M. Ortega-Huertas, E. Hansen, Langmuir 21, 10948– 31 902 10957 (2005) 32 33 34 903 60. D Benavente, in Utilización de rocas y minerales industriales: Propiedades físicas y 35 904 utilización de rocas ornamentales, eds. By M.A .Garcia del Cura, J.C. Cañaveras ( 36 37 905 Universidad de Alicante, Alicante, 2006), pp. 123-153 38 39 906 61. F.G. Bell, Engineering Properties of Soils and Rocks, 4ª edn., (Blackwell, Oxford, 2000), p. 40 41 907 496 42 43 908 44 45 909 FIGURE CAPTIONS 46 47 48 910 Fig.1 Set up of the experiment at the laboratory showing the limestone specimens and the 49 911 consolidating product based on Ca(OH)2 nanoparticles, in the three environmental conditions. 50 51 912 a) to the left, stone specimens (5cm3) connected to the flask with the yeast fermentation; in the 52 913 back, specimens inside the closed container at high humidity; to the right and front, stone 53 54 914 specimens exposed to the ventilate laboratory conditions 55 56 915 Fig.2 XRD and DTA-TG analyses of Ca(OH)2 nanoparticles exposed to ventilate laboratory 57 58 916 room conditions (59 ± 11 %RH). a) XRD patterns of samples after 7, 14, 21 and 28 days 59 60 61 62 63 27 64 65 917 exposure; b) DTA, TG and dTG curves after 7 days exposure; c) DTA, TG and dTG curves after 1 918 28 days exposure 2 3 4 919 Fig.3 XRD and DTA-TG analyses of Ca(OH)2 nanoparticles exposed to a closed container with 5 920 high humidity (87±1% RH). a) XRD patterns of samples after 7, 14, 21 and 28 days exposure; 6 7 921 b) DTA, TG and dTG curves after 7 days exposure; c) DTA, TG and dTG curves after 28 days 8 922 exposure 9 10 Fig.4 XRD and DTA-TG analyses of Ca(OH)2 nanoparticles exposed to a yeast fermentation 11 923 12 924 system connected to a closed container with high humidity (87±7% RH). a) XRD patterns of 13 14 925 samples after 7, 14, 21 and 28 days exposure; b) DTA, TG and dTG curves after 7 days 15 926 exposure; c) DTA, TG and dTG curves after 28 days exposure 16 17 Fig.5 Pore size distribution, PSD (pore diameter), obtained with mercury intrusion porosimetry 18 927 19 928 of limestone specimens. a) PSD of Lecce stone; b) PSD of Bateig stone 20 21 929 Fig.6 Limestone specimen showing the product application face (from the bottom), after 28 22 23 930 days of the application of Ca(OH)2 nanoparticles at 25 g/l, exposed to ventilate laboratory room 24 931 conditions (59 ± 11 %RH). a) macroscopic view of Lecce stone with nanoparticles at 5g/l; b) 25 26 932 SEM image (1000x magnification) of former sample; c) SEM image of former sample (6000x 27 magnification); d) macroscopic view of Lecce stone with nanoparticles at 25g/l; e) SEM image 28 933 29 934 (1000x magnification) of former sample; f) SEM image (6000x magnification); g) macroscopic 30 31 935 view of Bateig stone at 5g/l; h) SEM image (1000x magnification) of former sample; i) SEM 32 936 image (6000x magnification); j) macroscopic view of Bateig stone at 25g/l; k) SEM image 33 34 937 (1000x magnification) of former sample; l) SEM image (6000x magnification) 35 36 938 Fig.7 Limestone specimens showing the product application face (from the bottom), after 28 37 38 939 days of the application of Ca(OH)2 nanoparticles exposed to a closed container with high 39 940 humidity (87±1% RH). a) macroscopic view of Lecce stone with nanoparticles at 5g/l; b) SEM 40 41 941 image (1000x magnification) of former sample; c) SEM image (6000x magnification); d) 42 942 macroscopic view of Lecce stone with nanoparticles at 25g/l; e) SEM image (1000x 43 44 943 magnification) of former sample; f) SEM image (6000x magnification); g) macroscopic view of 45 944 Bateig stone at 5g/l; h) SEM image (1000x magnification) of former sample; i) SEM image 46 47 945 (6000x magnification); j) macroscopic view of Bateig stone at 25g/l; k) SEM image (1000x 48 946 magnification) of former sample; l) SEM image (6000x magnification) 49 50 51 947 Fig.8 Limestone specimens showing the product application face (from the bottom), after 28 52 948 days of the application of Ca(OH)2 nanoparticles exposed to a yeast fermentation system 53 54 949 connected to a closed container with high humidity (87±7% RH). a) macroscopic view of Lecce 55 950 stone with nanoparticles at 5g/l; b) SEM image (1000x magnification) of former sample; c) SEM 56 57 951 image (6000x magnification); d) macroscopic view of Lecce stone with nanoparticles at 25g/l; e) 58 952 SEM image (1000x magnification) of former sample; f) SEM image (6000x magnification); g) 59 60 953 macroscopic view of Bateig stone at 5g/l; h) SEM image (1000x magnification) of former 61 62 63 28 64 65 954 sample; i) SEM image (6000x magnification); j) macroscopic view of Bateig stone at 25g/l; k) 1 955 SEM image (1000x magnification) of former sample; l) SEM image (6000x magnification) 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 29 64 65 Figure1 Click here to download Figure: Figure1.tif Figure2

P a

P P

P P P P P C P C P 28 days P P P P P C C P P P 21 days

P

P P P P P 14 days

P P P P P C P C P

Intensity (a.u) Intensity 7 days

2 10 20 30 40 50 60 2θ (º)

Ca(OH)2 b c + CaCO3 CaO Ca(OH) 2 CO2 DTA H2O + -OH -OH Endothermic 140 CaCO 3 648 Endothermic DTA

CO2 429 682

100 100 TG TG 436 80 CaO 80 76% 24% 85% 15% 60 60 Massloss (%) Massloss (%) dTG dTG 50 50 100 200 300 400 500 600 700 800 900 1000 100 200 300 400 500 600 700 800 900 1000 Temperature (ºC) Temperature (ºC) Figure3

a C

A A C C A C C p A A A p p C A A A C A A p C C C 28 days p V p AA pA C p VA C A C A A C A p C 21 days p p p

p C C C CC p p A p A p C A A A A C 14 days C I p p I C p p C C C p p I

Intensity (a.u) Intensity 7 days

2 10 20 30 40 50 60 2θ (º)

Ca(OH)2 b c + Ca(OH) 2 CaCO3 + H2O CaCO 3 H2O -OH -OH

DTA DTA Endothermic 133 130 399 Endothermic CO2

CO2 CaO

730 429

100 709 100 TG TG 5% 80 31% CaO 80 69% 60 60 95% Massloss (%) Massloss (%) dTG dTG 50 50 100 200 300 400 500 600 700 800 900 1000 100 200 300 400 500 600 700 800 900 1000 Temperature (ºC) Temperature (ºC) Figure4

C a

A C CA AC A A AC C A A C C A C C 28 days A A A V A AC CAA C V AC AC C A A V V V A A V C C A C 21 days C C P C V P C C P C A V A A A C V A V C V P C A VP A P C C C C CPC 14 days P P

P V C C CC C P V C V P VA A A A A V P C P PC V P A A A VC C C 7 days Intensity (a.u) Intensity

2 10 20 30 40 50 60 2θ (º)

Ca(OH)2 b c + CaCO3 CaCO 3 H2O H2O

-OH Endothermic DTA DTA 140 Endothermic 140

440 CO 2 CO 2

CaO

733 CaO 733

100 100 TG TG 13% 2% 80 80 87% 60 60 Mass lossMass(%) Massloss (%) 98% dTG dTG 50 50 100 200 300 400 500 600 700 800 900 1000 100 200 300 400 500 600 700 800 900 1000 Temperature (ºC) Temperature (ºC) Figure5 a

. . .

b

. . . Figure6 Click here to download Figure: Fig.6.tif Figure7 Click here to download Figure: Fig.7.tif Figure8 Click here to download Figure: Fig.8.tif Tables1-4

1 TABLES

2 3 Table 1 Physical properties of blank stone specimens of Lecce and Bateig without nanoparticles 4 Physical properties Lecce specimens Bateig specimens

Chromatic parameters 5 Lightness (L*) 78.93±4.87 76.55±1.63 6 a* and b* color coordinates 2.96±1.25 and 13.25±3.91 2.6±0.41 and 13.79±1.26 7 Chroma C* 13.64±3.88 14.04±1.31 8 Brightness 43.73±7.68 39.41±3.11 9 White index (WI) 5.98±3.82 1.79±4.94 Yellow index (YI) 22.70±6.60 24.23±2.33 10 Pore size distribution 11 Total open porosity (%) 31±5 19±0.5 12 < 0.01 µm 1 1 13 0.01 - 0.1 µm 7±2 12

0.1 - 1 µm 41±5 22 14 1-10 µm 49±2 58±3 15 10 - 100 µm - 4±1 16 >100 µm 1 4±2 17 Water absorption 18 Apparent density (kg/m3) 1,669±5 2,155±14 3 Real density (kg/m ) 2,714±0 2,710±7 19 Open porosity (%) 39 21 20 Water saturation (%) 23 10 21 Capillarity absorption 22 Total water absorption (g) 32 18 23 Capillarity coefficient g/(m2s0.5) 98±4 41±0 Desorption coefficient g/(m2s0.5) 24±2 15±0 24 Surface hardness 25 HDL values 234±21 342±30 26 Microdrilling resistance 27 Force values (N) 3.96±0.43 9.50±1.11 28 29 30 Table ·2 Average variations (Δ) promoted on chromatic parameters (L*, lightness; C*, Chroma; E*, total color; YI, yellow index; WI, white index) on the stone specimens after 28 days of consolidation with Ca(OH)2 nanoparticles exposed to different environments Environment Specimen ΔL* ΔC* ΔE* ΔYI ΔWI ΔBrightness Lecce 5g/l 4.34±1.65 -0.61±02.01 4.67±1.77 -1.47±3.31 3.81±8.88 6.65±4.45 Laboratory Lecce 25g/l 15.34±0.8 -13.09±0.1 20.85±0.6 -23.7±0.03 83.32±2 42.85±1.85 conditions Bateig 5g/l -0.08±0.65 -0.81±0.28 0.95±0.18 -1.27±0.6 2.41±1.3 0.42±1.12 Bateig 25g/l 14.69±0.1 -11.96±0.44 20.29±1.8 -24.03±4.5 78±11.7 40±2.1 Lecce 5g/l 3.85±0.85 0.15±1.48 4.13±0.69 -0.23±2.4 0.44±6.14 5.13±2.65 Closed Lecce 25g/l 4.27±0.11 -3.73±0.26 5.63±0.26 -6.3±0.42 15.8±1.11 9.21±0.42 container Bateig 5g/l -0.37±0.27 -1.19±0.91 0.32±0.75 -1.76±1.5 3.59±2.9 0.35±0.9 Bateig 25g/l 5.57±0.3 -7.52±0.81 9.36±0.83 -13±1.38 31.86±3.7 14.78±1.3 Lecce 5g/l 3.11±0.89 1.37±0.95 3.71±0.43 1.74±1.58 -4.46±3.79 2.96±2.13 Lecce 25g/l 6.63±1.73 -4.83±1.44 8.18±2.23 -8.26±2.47 22.8±8.12 14.38±4.52 Yeast system Bateig 5g/l 1.40±0.32 -2.42±0.9 2.85±0.9 -13.4±0.2 18.2±2.8 3.84±2.2 Bateig 25g/l 6.36±2.61 -9.35±1.96 11.37±3 -16.3±3.48 41.23±12 18.14±6.6 31 32 33 34 35

1 36 37 TaTable 3 Water absorption by capillarity and water desorption values and variations (∆) of stone specimens before and after

28 days of consolidation with Ca(OH)2 nanoparticles (NC: No change) Total water absorption by Capillarity coefficient Capillarity desorption capillarity (g) (g/(m2s0.5)) coefficient (g/(m2s0.5)) Environment Specimens Before After Average Average Average Before After Before After ∆ (%) ∆ (%) ∆ (%) Lecce 5g/l 33 3.0 107±0 9 25±1 4 32 98±4 24±2 Laboratory Lecce 25g/l 32 NC 98±0 NC 24±0 NC conditions Bateig 5g/l 19 5.5 38±1 -7 13±1 -13 18 41±0 15±0 Bateig 25g/l 19 5.5 47±1 15 13±0 -13 Lecce 5g/l 33 3.0 101±2 3 24±0 NC 32 98±4 24±2 Closed Lecce 25g/l 32 NC 88±6 -10 26±2 8 container Bateig 5g/l 19 5.5 43±2 5 13±0 -13 18 41±0 15±0 Bateig 25g/l 18 NC 43±1 5 14±1 -7 Lecce 5g/l 33 3.0 100±3 2 25±1 4 32 98±4 24±2 Lecce 25g/l 32 NC 100±2 2 25±0 4 Yeast system Bateig 5g/l 19 5.5 44±6 7 14±0 -7 18 41±0 15±0 Bateig 25g/l 18 NC 38±2 -7 14±0 -7 38 39 40 41 42 Table 4 Drilling resistance values (Force, N) and surface hardness (HLD) values and variations (∆) of stone specimens after 28 days of Ca(OH)2 nanoparticles application at a concentration of 5g/l and 25 g/l 5g/l concentration 25g/l concentration Surface Surface Environment Stone Force Force ∆ (%) hardnes ∆ (%) ∆ (%) hardnes ∆ (%) (N) (N) s (HLD) s (HLD) Lecce 4.71±0.1 19 245±0 5 4.41±0.1 11 254±7 8 Laboratory conditions Bateig 9.61±0.7 1 346±6 1 9.94±1 5 354±3 4

Lecce 4.3±0.1 9 253±3 8 4.62±0.2 17 251±4 7 Closed container Bateig 9.73±0.5 2 351±10 3 10.29±1 8 359±2 5

Lecce 4.26±0.3 8 257±13 10 4.61±0.2 16 259±7 11 Yeast system Bateig 9.93±1 5 361±15 6 10.46±1 10 368±13 8 43 44

2