1 Materials & Design

2 Volume 120, 15 April 2017, Pages 10-21

3 https://doi.org/10.1016/j.matdes.2017.02.001

5 Artificial aging route for assessing the potential efficacy of

6 consolidation treatments applied to porous carbonate stones
7
8 B. Sena da Fonseca1, A. P. Ferreira Pinto2, S. Piçarra3,4, M.F. Montemor1
1
9 Centro de Química Estrutural-CQE, Instituto Superior Técnico, Universidade de Lisboa, 1049-001
10 Lisboa, Portugal
2
11 Department of Civil Engineering, Architecture and Georesources, CERIS, Instituto Superior Técnico,
12 Universidade de Lisboa, 1049-001 Lisboa, Portugal
3
13 Escola Superior de Tecnologia de Setúbal, Campus do IPS, Estefanilha, 2910-761 Setúbal, Portugal
4
14 Centro de Química-Física Molecular and Instituto de Nanociência e Nanotecnologia, Instituto Superior
15 Técnico, Universidade de Lisboa, 1049-001, Lisboa, Portugal
16
17 Abstract
18 Unsatisfactory behaviors of consolidation treatments applied to porous carbonate stones
19 in the field are frequent, although they had revealed promising performances when
20 applied to sound stones in laboratory. Thus, accurate assessment of the potential
21 efficacy of consolidation treatments in laboratory conditions is of outmost importance to
22 support conservation practice.
23 This paper compares heating and freeze-thaw as artificial aging procedures for testing
24 consolidants on a selected porous carbonate stone (CaCO3≈99.9%, porosity=16% and
25 average pore radius=0.4µm). The induced decay was assessed by physical tests and
26 drilling resistance measurements (DRMS), which is a well-establish method to
27 characterize materials and evaluate the efficacy of consolidation treatments.
28 Heating caused minor alterations, because the anisotropic thermal expansion of calcite
29 was compensated by the stone porosity, whereas the freeze-thaw cycles produced
30 important alterations.
31 Sound and pre-aged specimens by freeze-thaw were treated with different alkoxysilane-
32 based consolidants and their effect on the mechanical properties was assessed by DRMS
33 and analyzed using the difference drilling profile method (DDP).
34 The obtained results highlighted the interest of using pre-aged specimens by freeze-
35 thaw and the usefulness of the procedures adopted for the assessment of the potential
36 efficacy of consolidation treatments applied to porous carbonate stones.
37
38
39 Key-words: porous carbonate stones; artificial aging; consolidation; functional
40 treatments; micro-drilling; efficacy.
41 Corresponding Author
42 *E-mail: senadafonseca@gmail.com, Phone: (+351)925455230

1
43 1. Introduction

44 The consolidation of stone elements is one of the most risky actions in the conservation

45 field due to its non-reversibility and to the possibility of harmful side effects [1]. In

46 addition, successful consolidation has been very difficult to obtain with the available

47 products, especially on porous carbonate stones. Therefore, before any intervention

48 regarding stonework consolidation, it is crucial to guarantee that the treatment will meet

49 all the requirements for which it is being proposed, without any harmful side effect.

50 The most used, well-established and versatile consolidants are the ones based on

51 alkoxysilanes [2]. However, when used in porous carbonate stones, a few drawbacks

52 constraining their overall performance are commonly highlighted in literature [1, 3-5]:

53 lack of strong chemical bonds with the carbonate material and susceptibility to crack

54 during evaporation. Thus, several alternatives to the traditional alkoxysilane-based

55 products are being proposed such as nanolimes [2, 6], microbially induced carbonate

56 precipitation [7] or phosphate-based solutions [8, 9]. In parallel with these novel

57 alternatives, the versatility of alkoxysilane-based products and their good intrinsic

58 characteristics (capacity of penetration, stability, reduced influence on the vapor

59 characteristics, etc.) justify efforts to design new alkoxysilane-based products, in which

60 the mentioned limitations are suppressed or minimized. Examples are functional

61 alkoxysilanes containing organo-groups [10, 11], nanoadditives [12-14] or silicon-based

62 organic polymers [13-16].

63 The evaluation of the initial efficacy of consolidation treatments and potential

64 harmfulness is commonly studied under laboratorial conditions. However, despite the

65 use of the same or alike lithotypes, dissimilar results in terms of efficacy and

66 harmfulness between laboratory and in situ are often observed when applying

67 alkoxysilane-based treatments. Amongst other causes, one of the factors contributing to

2
68 those mismatched results is the use of sound stone specimens in laboratory, whereas in

69 situ stone to be consolidated is generally in an advanced state of decay. Consequently,

70 to better correlate laboratorial and field results, different artificial damaging procedures

71 were recently explored [17-19]. These procedures involve heating the stone specimens

72 (in dry or saturated conditions) during a certain period of time and are intended to

73 induce micro-cracks and grains de-cohesion, simulating decayed stones. In dry

74 conditions, the degradation mechanism varies according to the stone mineralogy [18]. In

75 dry stones composed by different minerals (e.g. granite or sandstones), the different

76 thermal expansion coefficients of stone constituents result into internal stress during

77 heating that causes the opening of micro-cracks, while in stones mainly composed by

78 calcite (e.g. marbles and limestones) the opening of micro-cracks occurs because of the

79 marked anisotropic thermal expansion behaviour of calcite. Besides, when saturated

80 specimens are heated, the pressure exerted by water on stone pore walls may be

81 responsible for further damaging [18].

82 Heating stones up to 400ºC showed to be an effective way to produce microstructural

83 changes and reproducible modifications on the mechanical strength of some lithotypes

84 [18, 19]. However, it was reported that in specific porous carbonate stone, the method

85 was less effective because larger volumes of voids allowed free deformation of calcite

86 crystals, without inducing internal stresses [20]. Furthermore, when the stones have

87 secondary minerals that are susceptible to “hot-melt effect” (e.g. clay minerals),

88 chemical-physical transformations may be responsible for a more effective bond of the

89 grains and consequently to strength increases [18]. These are noteworthy drawbacks of

90 the heating procedures, since porous carbonate stones have been quite used as

91 monumental stones, as they are fairly easy to carve due to their softness.

3
 92 Therefore, an alternative artificial degradation method for porous carbonate stones

93 based on physical decay and with less susceptibility of inducing chemical modifications

94 is required. Some possible ways can be considered, namely freeze-thaw cycles or salt

95 crystallization.

96 From the perspective of preparing artificially aged specimens to study consolidation

97 treatments, the freeze-thaw action seems to be more appropriate, as there is no risk of

98 having salts blocking the pores or interfering with the consolidation treatment under

99 study, while it is not predictable that it can produce significant and unexpected chemical

100 changes. Furthermore, it is a real decay mechanism that affects porous materials in

101 several countries. Depending on the tensile strength and porous characteristics of the

102 lithotypes, the major drawback of freeze-thaw cycles may be related to the long duration

103 of the procedure. However, a significant degradation degree is expected to occur with a

104 reduced number of cycles in porous soft carbonate stones (porosity >10%) with a well-

105 connected pore system [21]. The pore size distribution and pore shape also have a

106 prominent role on the decay susceptibility to ice crystallization [22]. In fact, it is known

107 that stones having a significant percentage of pores within the range 0.1 µm and 10 µm

108 are particularly vulnerable to damage due to ice and salt crystallization [21, 23].

109 Regarding the evaluation of the damage produced by aging procedures and the strength

110 gain due to subsequent consolidation treatments, several methods that can hardly

111 discriminate strength variations along the depth have been used (compressive and

112 tensile strength, dynamic and static elastic modulus) [8, 9, 17, 24, 25]. However, in

113 most field situations, the physical integrity tends to vary in-depth in naturally decayed

114 stones and in most of the treated stones. Due to this reason, it has been generally

115 recognized that the drilling resistance measurement system (DRMS) is one of the most

116 suitable methods for evaluating the consolidation efficacy, particularly in soft stones

4
117 [26-29]. The possibility of assessing the variation of material strength in depth in

118 laboratory and in situ is of outmost importance for the characterization of the decay

119 patterns and to assess the treatments efficacy and potential harmfulness [30-32].

120 According to the gaps that have been identified, the work disclosed henceforth intends

121 to investigate the interest of using artificially aged specimens to assess the potential

122 efficacy of consolidation treatments on porous carbonate stones and to compare heating

123 and freeze-thaw as artificial aging procedures for that purpose. The evaluation of the

124 alterations promoted by the tested aging procedures was based on the comparison of

125 physical and mechanical properties assessed by DRMS. Additional tests were carried

126 out to compare the initial efficacy of alkoxysilanes-based treatments when applied to

127 sound specimens and freeze-thaw pre-aged specimens.

128

5
129 2. Materials and Methods

130 2.1. Stone and consolidant products

131 The lithotype used in this study was a beige limestone with the commercial trade name

132 of Rosal AR. This stone is under current exploitation in Porto de Mós region (Portugal)

133 and is mainly used for wall cladding and flooring. Fairly similar lithotypes exploited

134 from the same massif were erstwhile used in the construction/reconstruction of some

135 important monumental constructions in Portugal [33, 34].

136 The tested stone can be considered a highly porous limestone, as its open porosity is

137 around 16.0%. It is composed by spar calcite cement as interstitial material and peloids,

138 being less abundant the oolits. The allochems vary in size from 0.1mm to 1mm, wherein

139 most of the particles have around 0.5 mm.

140 The pore size of Rosal AR ranges mainly between 0.02 and 1 µm, with higher

141 percentages of pores with dimensions between 0.1 and 0.6 µm (figure 1).

142 The pore size distribution of Rosal AR stone points out that this lithotype is susceptible

143 to be damaged by freeze-thaw cycles, because the majority of the pores display sizes

144 within the range 0.1 µm and 10 µm [21, 23].

145

146 Figure 1 - Pore size distribution of Rosal AR stone.

147

6
148 The artificial aging tests were performed on cubic specimens with the dimensions of

149 30x30x30mm3.

150 Five alkoxysilane-based products were used to treat the aged specimens: a commercial

151 product, Silres BS OH 100 (hereafter BS), and four modified alkoxysilane-based

152 products developed in our laboratory with the denominations of 2hR, 2hTUDA0.05,

153 6hTUDA0.01 and 2hSiDETA.The consolidants were selected in view of their distinct

154 chemical functionalities, which result into different behaviors in terms of consolidation,

155 when tested on other lithotypes [35-37].

156 A brief description on such functionalities is provided: The commercial product (BS) is

157 solvent-free and has close to 100% of ethyl-silicate as active component [38],

158 presumably pre-condensed oligomers from tetraethoxysilane (TEOS) [39]. The used

159 catalyst promotes the reaction between ethyl silicate and water from atmospheric

160 humidity [38].

161 All non-commercial products have ethanol as solvent, being formulated with 1

162 TEOS/3.8EtOH/2.1H2O molar ratio (water was added to the formulations to enable the

163 initiation of the sol-gel process).

164 2hTUDA0.05 and 6hTUDA0.01 also contain 3,6,9-trioxaundecanedioic acid (TUDA)

165 used in the molar ratio of 1TEOS/0.05TUDA and 1TEOS/0.01TUDA, respectively.

166 TUDA plays a twofold role: it acts as sol-gel catalyst and in basic medium, as flexible

167 spacer, condensing into the siloxane network [35-37].

168 2hSiDETA product have N1-(3-trimethoxysilylpropyl)diethylenetriamine (SiDETA) in

169 the molar ratio of 1TEOS/0.01SiDETA. SiDETA can act as coupling agent and a silica

170 structure modifier [37]. The pre-condensation was performed by stirring the

171 formulations 2hR, 2hTUDA0.05, 2hSiDETA for 2 hours and 6hTUDA0.01 for 6 hours.

172

7
173 2.2. Artificial aging

174 2.2.1. Heating aging procedures

175 The heating aging procedures used four different sets of dry limestone specimens, each

176 one with three specimens. Three sets of dry limestone specimens were exposed during

177 1h to three heating temperatures (200, 300 and 400◦C), by placing them in a kilt under

178 oxidizing environment.

179 The fourth set of specimens was immersed in distilled water during 48 hours and then

180 placed in a kilt at 200ºC for 1 hour. After cooling to room temperature, these specimens

181 were re-heated at 400ºC for 1 hour in dry conditions (hereafter WS specimens).

182 The described heating procedures for the study of stone consolidants were developed by

183 Franzoni et al. (2013) [18] and tested in other lithotypes [18, 20].

184

185 2.2.2. Freeze-thaw aging procedure

186 The specimens submitted to freeze-thaw cycles were firstly dried, weighed, and

187 afterwards immersed in distilled water for saturation during 48 hours, according to the

188 standardized procedure described in EN12371. Each freeze/thaw cycle had the duration

189 of 24 hours and comprised: a freezing period of 7±1h, performed in a freezing chamber

190 at constant temperature of -16±2ºC; and a thaw period of 17h±1h, where the specimens

191 were immersed in distilled water at 20±5ºC.

192 The freeze-thaw aging was performed on 15 specimens and lasted 25 cycles. At the end

193 of the 25th cycle, the stone specimens reached the 3rd degradation class as it is

194 established in EN 12371, as at least two specimens showed visible cracks (width higher

195 than 0.1mm). At the end of freeze-thaw aging procedure the dry mass of the specimens

196 was assessed.

197

8
198 2.3. Alterations induced by the aging procedures

199 2.3.1. Visual appearance

200 Firstly, all tested specimens were macroscopically analyzed and the damage evidences

201 registered. Subsequently, a binocular stereoscope (Nikon SMZ1000) was used to

202 perform a more detailed inspection of the superficial damages.

203 The color variation of the specimens submitted to the heating aging procedures was

204 evaluated by using the CIELab color system. The colorimetric coordinates of the

205 surfaces L*, a* e b* were registered by using a colorimeter KangGuang WSD-3A.

206 Alterations induced by heating were obtained by the difference between the average

207 values of twelve measurements in sound specimens (S) and the average values of twelve

208 measurements in heated specimens (H) (ΔL*=ΔL*H –ΔL*S; Δa*=Δa*H –Δa*S; Δb*=Δb*H –

209 Δb*S) by the total color variation (ΔE*), according to the following expression [40]:

210

211 ΔE* = (ΔL*2 + Δa*2 + Δb*2)1/2 (1)

212

213 To explain color variations, the stone mineralogical composition was also determined.

214 Thus, powders from sound stone specimens were analyzed by x-ray diffraction (XRD)

215 using a Bruker equipment, D8 Advance. To identify the presence of components in

216 trace amounts, the limestone powder was digested in a solution with 4 parts of distilled

217 water and 1 part of acetic acid (CH3COOH) to avoid the dissolution of clay minerals.

218 To avoid the precipitation of calcium acetate (Ca(CH3COO)2) and guaranty the

219 dissolution of calcite, for each 6 g of powder, 100ml of solution was used. When the

220 reaction stopped, the suspension was centrifuged at 3500 r.p.m. for 5 minutes (Hettich

221 Universal 320R) and the supernatant liquid removed. The insoluble residue was washed

222 by adding distilled water and the mixture was centrifuged again. This procedure was

9
223 repeated 5 times, to guarantee the absence of calcium acetate and for the whole removal

224 of acetic acid.

225 The residue was weighted to calculate the percentage of insoluble components and was

226 analyzed by XRD. Of this fraction, 0.100 g were suspended in 20 ml of 10% H2O2 at

227 70ºC for 24 hours, followed by treatment with 30% H2O2 at the same temperature for 48

228 hours, when the visible reaction ceased. Thereafter the suspension was dried and the

229 resulting decrease in mass determined gravimetrically. Similar procedures are

230 commonly used to determine the amount of organic matter in soils [41, 42].

231

232 2.3.2. Physical properties

233 The physical properties tested were: porosity accessible to water, water absorption by

234 immersion (48h), saturation coefficient and water absorption by capillarity. The tests to

235 determine the different physical properties were performed on the same specimens by

236 comparison before and after aging.

237 The porosity accessible to water (OP), the water absorption by immersion after 48 hours

238 (WA48h) and the saturation coefficient (SC) were determined by following the

239 procedures recommended by Commission 25-PEM [43].

240 The dried specimens (m1) (at 65ºC) were placed inside a container under vacuum for 24

241 hours (≈2.67Pa). Thereafter, the container was filled with distilled water and the

242 specimens remained immersed and under vacuum for another 24 hours. After this

243 period, the pressure was increased until atmospheric pressure and the specimens stayed

244 immersed for more 24 hours. The saturated specimens were weighed in water

245 (hydrostatic weighing) (m2) and then slightly wiped with a damp cloth and their mass

246 determined (m3). The porosity accessible to water is expressed as a percentage and was

247 calculated as follows:

10
248

249 (1)

250

251 For the determination of WA48h, dry specimens (m1) were placed inside a vessel filled

252 with distilled water. After 48 hours, the samples were slightly wiped with a dampened

253 cloth and their mass determined (m4). The water absorption by immersion is expressed

254 as a percentage and was calculated as follows:

255 (2)

256 The saturation coefficient is also expressed as a percentage and was calculated as

257 follows:

258 (3)

259 The water absorption by capillary test was performed according to standard EN 15801

260 [44]: the dry specimens were placed in a closed container on a pile of filter papers,

261 approximately 10 mm thick, partially immersed in water [45]; the evolution of the

262 specimens mass and capillary fringe position during the test were monitored by

263 successive weighing and by periodical measurements of the capillary fringe height,

264 respectively.

265 In addition, the weigh loss of the specimens submitted to 25 cycles of freeze-thaw was

266 determined by weight difference between the specimens initial and final mass.

267 In the case of heating aging, the results from physical tests are the average of the values

268 of three specimens, whereas the results of freeze-thaw aging specimens correspond to

269 the average of five determinations.

270

271 2.3.3. Drilling resistance

11
272 To assess the resistance decrease induced by the artificial aging procedures, the Drilling

273 Resistance Measurement System (DRMS) from SINT Technology srl. was used.

274 A guide hole with 3 mm of diameter was previously performed all across the 30 mm of

275 specimens. A 5 mm diameter diamond drill bit was used to drill over the guide hole and

276 to record the resistance along the depth. The resulting powder was vacuumed through

277 the end of the guide hole during drilling. All drilling tests were performed at a speed

278 rotation of 300 r.p.m. and with a penetration rate of 15 mm/min.

279 In each aged specimen, two drills were made in a defined position (according to step 1

280 in figure 2). The results achieved on the pre-aged specimens were compared with the

281 drilling profile of the sound stone calculated as the average of the results achieved on 10

282 holes.

283

284 2.4. Treatment procedure and initial efficacy

285 2.4.1. Treatment procedure

286 The application of the consolidants was performed on sound and pre-aged specimens.

287 Although, only the artificial aging procedure that induced significant alterations in the

288 physical properties and in the DRMS profiles of sound stone was used for the

289 assessment of the potential efficacy of consolidation treatments applied to porous

290 carbonate stones.

291 The consolidants were applied to stone by capillary suction; the stone specimens were

292 partially immersed (3 mm depth) in the liquid products for 3 hours. The application time

293 was determined by measuring the evolution of the capillary fringe (mm) on the lateral

294 surface of the specimens, being concluded that 3 hours was the time necessary for the

295 wet fringe of the majority of the consolidants to reach the top of the specimens (30mm).

12
296 The stone specimens were weighed before and after the consolidant application and the

297 difference between these two weighing steps was considered as the amount of absorbed

298 product. Thereafter, the treated specimens were placed in a chamber at 25ºC and

299 70±10% R.H.

300 After two months, their mass was taken to calculate the final residue left within the

301 stone pores by area (ρDR) (kg/m2) according the following expression:

302

303 (4)

304 Where mi is the initial mass of the non-treated specimen, mt is the mass of the treated

305 specimen after 2 months, and A is the area of the absorbing surface of the specimen.

306

307 2.4.2. Treatments initial efficacy

308 The initial efficacy of the tested treatments, i.e., the treatments ability to provide

309 cohesion to damaged stones, was quantified by DRMS.

310 To reduce the variability inherent to the stone and to the decay promoted by the aging

311 procedure and to guarantee an accurate knowledge about the initial efficacy of the

312 consolidation treatments, the products were applied to the same specimens that were

313 previously tested with DRMS, for the characterization of the aging procedure.

314 Before treatment, the aged specimens already containing 2 holes at a distance of 7.5mm

315 from the lateral surfaces (from the analysis of artificial aging procedures) were sawed,

316 as exhibited in step 2 - figure 2. Specimens with dimensions of 30x30x20 mm (without

317 holes) were obtained to be subsequently treated (step 3 - figure 2).

318 The DRMS holes used to evaluate the consolidation efficacy were made at a distance

319 equal to the one used in the first two drillings from the lateral surfaces (7.5 mm) (see

320 step 4 - figure 2).

13
321 The procedure used in the DRMS tests after treatment was the same used to assess the

322 cohesion decrease caused by the aging procedures, above described.

323 To simplify the analysis of the consolidation action promoted by the tested treatments,

324 the increase in the drilling resistance was analyzed through the difference drilling

325 profile (DDP), calculated by subtracting the average forces obtained from two drillings

326 after treatment and two drillings before the treatment (in the same specimen) [27]. To

327 better illustrate the procedure used, figure 3 depicts three exemplificative drilling

328 resistance plots: a drilling profile of the stone before (NTi) and after treatment (Ti); and

329 the curve that corresponds to the difference of the forces registered at each depth

330 between the first two drilling profiles (Ti – NTi), the difference drilling profile (DDP).

331

332

333 Figure 2 – Main steps of the procedure adopted for the assessment of the potential

334 efficacy of consolidation treatments on artificially pre-aged porous carbonate stone: 1)

335 Location of DRMS holes before treatment; 2) Specimen sawed to enable an

336 unidirectional consolidant application; 3) Consolidants applied by capillary suction

337 during 3 hours; 4) Location of DRMS holes after treatment.

338

14
339

340 Figure 3 – Exemplificative micro-drilling profiles of a stone before treatment (NTi), and

341 after treatment (Ti) (a) and the corresponding difference drilling profile (DDP) (b).

342 Figure 3 (b) also includes the standard deviation of the drilling resistance of the stone

343 before treatment.

344

345 3. Results and discussion

346 3.1. Alterations induced by the aging procedures

347 3.1.1. Visual appearance

348 The macroscopic inspection of the specimens submitted to heating allowed to identify

349 alterations on the visual appearance of the surfaces. The specimens aged at 200ºC

350 (figure 4 b) did not show evident color changes, but the specimens aged at 300ºC (figure

351 4 c), 400ºC (figure 4 d) and WS (figure 4 e) revealed noticeable color alterations. The

352 light-beige color of the natural stone changed to a reddish color in the specimens heated

353 at 300ºC and to a more grayish color in the specimens heated at 400ºC and WS. Table 1

354 exhibits the quantification of these color alterations in the CIELAB system. The

355 specimens submitted to 200ºC revealed minor color changes and the values of L*, a*

356 and b* were fairly similar to the ones obtained for sound specimens (∆E=0.78).

357 Specimens heated at 300ºC revealed an increase of the a* colorimetric coordinate, i.e.,

358 an increase of the red component. The specimens heated at 400ºC showed a decrease of

15
359 L* and b* coordinates, which means an overall darkening of the surfaces and a loss of

360 their original yellowish tonality. In the case of WS specimens, the same trend was

361 registered, but a higher a* value was obtained.

362

363 Table 1 – CIELAB parameters measured on the surfaces of the specimens before (sound

364 stone) and after the heating aging procedures.

L* a* b* ∆L* ∆a* ∆b* ∆E*

Sound stone 84.10±0.11 2.01±0.15 7.39±0.09 - - - -
200ºC 84.67±0.46 2.16±0.17 7.91±0.52 0.57 0.15 0.51 0.78
300ºC 83.82±0.46 3.22±0.20 7.55±0.30 -0.28 1.20 0.16 1.25
400ºC 80.71±0.92 2.21±0.21 6.09±0.44 -3.39 0.19 -1.30 3.64
WS 80.49±0.84 2.88±0.12 6.23±0.34 -3.61 0.87 -1.16 3.89
365
366 Since the unique component indentified through XRD of the sound stone was calcite

367 (figure 5), color modifications due mineralogical alterations at such temperatures were

368 not predictable: pure marbles do not experience significant color changes when

369 subjected to high temperatures and the calcined limestones have a dull/earthy

370 appearance [46]; however the calcination of calcium carbonate begins at higher

371 temperatures, typically around 600ºC. Still, it is widely recognized that limestone

372 varieties containing iron change their original color to red or to reddish-brown when

373 submitted to temperatures between 300-400ºC [46-48]. In several limestones, iron can

374 be distributed among a variety of different species, such as clay minerals or other

375 phyllosilicates. De-hydroxylation of these iron-bearing minerals, which starts at 200ºC,

376 is mostly associated to iron release. Much of this iron is oxidized and causes a red

377 discoloration of the limestone [49]. Even when present in very low amounts, when their

378 bearing-minerals are present in trace amounts and are too low to be traced by XRD or

379 microscope techniques, color modifications on limestone surfaces can be easily

380 observed in specimens submitted to temperatures above 250ºC [50]. Thus, the most

16
381 plausible and consensual explanation for the observed color changes is the existence of

382 residual amounts of iron bearing-minerals.

383

384

385 Figure 4 – Visual aspect of the surface of sound stone (a and g), heated at 200ºC (b),

386 300ºC (c), 400ºC (d), WS (e), and stone submitted to freeze/thaw cycles (f and h).

387

388 Due to the influence that iron-bearing minerals can have on the mechanical strength of

389 the limestone when heated [18], the presence of trace minerals was determined. The

390 percentage of insoluble residue (non-carbonate minerals) showed to be quite low, of

391 around 0.10%. From this fraction, at least 24% corresponds to organic carbon since it

392 was the corresponding percentage of material removed by the treatment with H2O2.

393 Therefore, trace minerals were determined as being 0.076% of the stone composition.

394 The spectra of the sound limestone present in figure 5 (in red) show the characteristic

395 sharp peaks of calcite and no other peak non-attributable to calcite was detected. The

396 XRD spectra in blue corresponds to the insoluble residue and exhibits two sharp peaks

397 at 12.4º and 24.9º and some minor peaks/broads between 20-22º and between 35º and

398 40º. The complete digestion of calcium carbonate by acetic acid was successfully
17
399 achieved since no traces of calcium carbonate were detected. The pattern was attributed

400 to minerals from kaolinite group (clay minerals) and the existence of such particular

401 broads are specific from disordered kaolinite [51]. In fact, despite not having iron as

402 major constituent, these broads can indicate defective structures, where Al ions may be

403 missing from normally occupied sites, being replaced by ferric or ferrous iron cations

404 [52]. As previously addressed, XRD confirms that the color changes could be caused by

405 the presence of clay minerals.

406 In any case, the detected thermally-induced modifications have a chemical nature, i.e.,

407 the heating procedure induced chemical alterations in the stone. In addition to the color

408 changes, it was not possible to detect other signs of degradation by visual observation,

409 namely, physical damages.

410

411 Figure 5 – X-ray diffraction spectra of sound limestone (red) and insoluble residue

412 (blue).

413

414 Regarding the specimens submitted to freeze-thaw cycles the macroscopic inspection

415 allows to identify further physical degradation in the form of pitting, in addition to the

416 cracks observed in some of them (figure 4 g and h). Under the binocular microscope

417 (figure 4 f), it was possible to unveil that the pitting corresponds to the detachment of
18
418 peloids/oolites. The registered mass loss of 0.18±0.02% derived essentially from the

419 superficial detachment of these elements.

420 The visible cracks grew at the interface of the sparry calcite cement and allochems and

421 resulted from the freezing of inter-particular water, which caused internal stress on these

422 structures and led to the fracture of stone. As a result, new voids in the form of fissures

423 appeared.

424 As expected, the observable cracks at naked eye were located at the corners, since lower

425 force is opposed to the internal expansive forces on these zones. Formation of both,

426 inter- and intra-granular micro-cracks is further expected, although undetected at naked

427 eye and hardly visible at binocular microscope [53].

428

429 3.1.2. Physical properties

430 The water absorption (WA48h) and open porosity (OP) are two useful parameters for

431 understanding some specificities of the stones internal structure, when analyzed in

432 conjunction. While the maximum water content obtained through water saturation under

433 vacuum reflects the nearly total void space, the water absorption by immersion during

434 48 hours at atmospheric pressure mostly reflects the water imbibition through capillary

435 pores and thus it is more influenced by pore space features, as the hardly accessible

436 pores are barely invaded by water at atmospheric pressure. The saturation coefficient

437 (SC), ratio between the water content after immersion during 48 hours and the

438 maximum water content, is useful to analyze the effects of the aging procedures on the

439 stones microstructure.

440 The values of OP, WA48h and SC of the tested stones before and after aging by heating

441 (table 2) indicated that the specimens underwent minimal changes. In fact, the variations

442 on the stone properties can be considered negligible in the case of 200ºC, 300ºC and

19
443 400ºC heating procedures; it is likely that the small changes observed lie on the stone

444 intrinsic variability. This data corroborates previous findings [20], where porous

445 limestones underwent rather limited alterations in microstructure due to heating up to

446 400ºC.

447 The group of specimens submitted to WS procedure revealed small, but more relevant

448 modifications in OP, approximately 3% in relation to sound stone, which can be

449 attributed to the effect of the pressure exerted by water on the pore walls. Despite of

450 this, it is interesting to note that the SC was similar to the SC of sound stone and

451 therefore the remaining features concerning the pores connectivity appeared to be

452 similar (connectivity, turtuosity, etc.).

453 In any case, the overall variations revealed to be minor, even the most extreme

454 procedure (WS) produced lower variations than the ones reported by Frazoni et al. [18]

455 in other porous limestones (containing 8-15% of other minerals).

456 The freeze-thaw cycles caused more relevant alterations. While the changes in OP were

457 around 3%, the alteration in WA48h value was much more relevant, it increased almost

458 9% relatively to sound stone. This denotes, as previously noted, the appearance of new

459 empty spaces within the stone structure. The specimens submitted to freeze-thaw cycles

460 underwent changes on these parameters with different magnitudes and thus the SC also

461 changed. As known, this coefficient can provide information about the extent of stone

462 decay [43] and depends on the porosity that is freely accessible at atmospheric pressure.

463 A porous material with a higher SC allows an easier and faster penetration of water

464 within the majority of accessible pores. Therefore, it is possible to state that the freeze-

465 thaw cycles enhanced the stone porosity and pores connectivity.

466

20
467 Table 2 – Physical properties of sound and artificially aged stone: open porosity (OP),

468 water absorption by immersion (WA48h), saturation coefficient (SC), coefficient of

469 water absorption by capillarity (A) and rate of water uptake (B).

OP (%) WA48h (%) SC (%) A (g/m2.sec0.5) B (mm/sec0.5)

Sound 16.0±0.21 6.0±0.07 85±1.1 72.3±6.2 0.658±0.026
Absolute values

200ºC 16.1±0.29 6.0±0.15 84±0.4 70.4±5.1 0.600±0.020

300ºC 15.9±0.10 5.8±0.06 83±0.4 75.8±1.1 0.625±0.003
400ºC 16.1±0.42 6.1±0.16 86±0.9 80.9±1.5 0.664±0.006
WS 16.5±0.56 6.1±0.10 84±2.3 84.0±9.8 0.669±0.015
Freeze-thaw 16.5±0.32 6.5±0.13 89±0.9 92.1±8.5 0.729±0.040
200ºC (%) +0.6 +0.0 -1.2 -2.6 -8.8
Variation in

sound stone
relation to

300ºC (%) -0.6 -3.3 -2.4 +4.8 -5.0

400ºC (%) +0.6 +1.7 +1.2 +11.9 +0.9
WS (%) +3.1 +1.7 -1.2 +16.2 +1.7
Freeze-thaw (%) +3.1 +8.3 +4.7 +27.4 +10.8
470 .

471
472 Figure 6 shows the water absorption and water uptake by capillary of the stone before

473 and after the tested aging procedures. The first phase of absorption occurs by a genuine

474 capillary imbibition mechanism, where the pores were freely invaded by water.

475 All cases showed a linear evolution with the square root of time, in the initial stage of

476 absorption, which is given by zone 1 in figure 6 and whose limit is determined when the

477 capillary front reached the top of the specimen. Therefore, the coefficient of water

478 absorption by capillarity (A) and the rate of water uptake (B) were determined and

479 presented in table 2.

21
480

481 Figure 6 – Water absorption (CWU) and capillary fringe registered on the stone before

482 (a) and after being submitted to 200ºC (b), 300ºC (c), 400ºC (d), WS (e) and freeze/thaw

483 cycles (f); water absorption (CWU) curves of sound and aged stone under all tested

484 aging procedures (g).

485

22
486 The initial rate of capillary absorption (A) of specimens submitted to 200ºC and 300ºC

487 showed to be within the dispersion of the rate of capillary absorption of sound

488 specimens, whereas the specimens submitted to 400ºC evidenced a slightly increase and

489 WS a more significant increase (see figure 6 g). Such trend is in accordance with the

490 results of OP and WA48h tests previously analyzed. Interestingly, the advance of the

491 capillary fringe (B) was slightly slower in the specimens submitted to 200 and 300ºC.

492 Although similar quantities of water entered into the pore network, a small delay in

493 water advance was registered. It is presumable that the slight delay in the advance of the

494 capillary fringe occurred due to the increase of volume as a result of the oxidation of

495 trace components caused by heating [54]. However, such increased size was not enough

496 to occlude the pores in a way that it could affect the other measured parameters, as only

497 around 0.076% of trace minerals were found.

498 When higher temperatures were achieved (400ºC), the anisotropy of calcite may have

499 induced more relevant faint cracks at micro and nano-scale [8] and the overall rate of

500 capillary absorption increased slightly, overlapping any possible effect owing to the

501 presence of oxidized trace minerals.

502 The porous network of the stone after being submitted to the freeze-thaw cycles had a

503 superior initial penetration rate (A), whereas the other kinetic parameter (B) also

504 reflected the same trend (table 2). The second stage of capillary imbibition (zone 2)

505 corresponded to the period between the moment when the capillary front reached the

506 top of the specimen and the mass uptake stabilized [55, 56]. The time lag between

507 visual saturation and stability of absorption by other capillary ways did not significantly

508 vary in the specimens aged by heating. However, this period (zone 2) decreased

509 unambiguously in specimens submitted to freeze-thaw cycles, which clearly indicated a

510 better connection between pores due to the generated cracks and an overall increase in

23
511 the average pore size. These results are in accordance with the results from water

512 absorption by immersion, where the freeze-thaw specimens also revealed a faster

513 absorption of water (higher SC value) compared to the sound and heated specimens.

514 Furthermore, it is possible to note a close relation between the OP results and the

515 maximum values of capillary water uptake (see figure 6 g), i.e., higher maximum values

516 were observed in WS and freeze-thaw specimens.

517 Summarizing, it is possible to state that the evaluated physical properties indicate that

518 freeze-thaw cycles produced noteworthy modifications on the internal structure of the

519 stone, whereas evident/important modifications on the physical properties of the

520 specimens submitted to the heating procedures were hardly disclosed.

521

522 3.1.3. Drilling resistance

523 The drilling measurements of the 10 holes made along the depth of the sound stone

524 revealed an average force of 7.58 N and a standard deviation of 1.24 N (table 3). Figure

525 7 a) and b) present the average profile of the sound stone (black continuous line) and the

526 standard deviation of the forces registered at each depth (black dashed lines).

527

528 Figure 7 – Drilling profiles of the tested stone before and after being heated (a) and

529 submitted to freeze/thaw cycles (b).

24
530

531 Table 3 – Drilling resistance forces of sound and artificially aged stone

Aging procedure Sound 200ºC 300ºC 400ºC WS Freeze-thaw

n – number of tests (n=10) (n=5) (n=5) (n=5) (n=5) (n=10)
Average force 7.58±1.24 7.39±1.14 7.37±1.24 7.87±1.10 6.79±1.08 4.19±1.36
±stand. deviation
532

533 The average force profiles obtained from 5 drillings on heated specimens are shown in

534 figure 7 a), from where the average values are given in table 3. With exception of the

535 specimens submitted to WS procedure, the forces obtained were fairly similar to the

536 forces obtained from sound stone. Figure 7 a) shows that the profiles of heated

537 specimens overlap the profile of sound stone, which revealed that the procedures did not

538 produce enough physical alterations that resulted into mechanical loss detectable

539 through the drilling profiles. Despite some minor alterations on the physical properties,

540 it has been reported by some authors that the intrinsic strength of limestone remained

541 unchanged when submitted to low temperatures, i.e. to temperatures inferior to 400ºC

542 [46, 47].

543 In fact, our results are comparable to the results obtained by Franzoni et al. [18] on other

544 porous limestone (Globigerina limestone): even containing relevant amounts of quartz,

545 the mechanical strength alterations determined by direct methods were never greater

546 than 13% for any heating procedure (uniaxial compression tests and Brazilian tests),

547 being the dynamic elastic modulus (estimated by ultrasonic pulse velocity) the unique

548 significantly affected parameter [18]. On the other hand, the authors also found that the

549 eventual presence of clay minerals could even have caused a strength gain (compressive

550 and tensile) due to chemical-physical transformations. Although there was a minor

551 increase in the overall resistance on the specimens submitted to 400ºC, there was a quite

25
552 reduced amount of kaolinite in the stones composition and therefore it was likely that

553 such small increase could be attributable to the stone natural heterogeneity.

554 The absence of significant disruption of the internal structure, caused by thermally

555 induced stress, can be explained by the stone high porosity which could compensate the

556 differential expansion of the calcite crystals [49].

557 Concerning the WS specimens, the profile revealed an average decrease of around 10%

558 to the initial resistance of sound stone. Such decrease indicated that the procedure could

559 have caused a strength loss detectable by the drilling measurements. However, the

560 decrease was still considered quite modest; the drilling profile of WS specimens

561 remained almost completely within the standard deviation of the sound stone and

562 therefore within the stone variability (see figure 7 a). In any case, pressure exerted by

563 the evaporating water on stone pore walls showed to be more damaging than eventual

564 internal stresses due to the anisotropic thermal expansion of calcite.

565 The drilling resistance of the specimens submitted to freeze-thaw cycles is depicted in

566 Figure 7 b (in blue). The cycles caused an almost uniform resistance decrease by half of

567 the original resistance value along the tested depth. Specifically the average force was

568 4.19N which corresponds to a 45% strength loss (table 3) and clearly results from the

569 cohesion loss due to the formation of a micro-cracks network at an inter-particular level

570 which affected the stone cementation, known to be the main controlling factor of stone

571 strength [57]. The absence of noticeable gradients in the stone resistance suggested that

572 this network of micro-cracks was generally homogenous across the specimens. The size

573 and the geometrical position of the holes also contributed to the uniform drilling

574 profiles.

575 Regarding the dispersion of results, the values of drilling force determined on the

576 specimens submitted to freeze-thaw cycles showed a standard deviation of 1.36N. The

26
577 similarity between this value and the standard deviation obtained for the sound stone

578 indicates that the dispersion of the aged specimens was not excessively high, making

579 the freeze-thaw cycles an interesting procedure for artificially damaging the tested stone

580 with reasonable dispersion. On the other hand, the similarity between the drilling

581 profiles from heated and sound stone specimens makes these procedures less attractive

582 for studies concerning the efficacy of consolidation treatments and therefore only the

583 specimens submitted to freeze-thaw cycles were treated and further studied.

584 These results highlighted the potential of using the tested freeze-thaw aging procedure

585 to pre-damage porous carbonate stones with similar properties to the tested stone and of

586 using DRMS for mechanical monitoring purposes. In the lithotype studied, the 3rd

587 degradation class was achieved after 25 freeze-thaw cycles, but depending on the stone

588 initial characteristics (microstructural, porosity, etc.) the number of cycles required to

589 obtain suitable degradation might need to be adjusted. The number of cycles to be used

590 should be based on the periodical visual inspection of the specimens at the end of each

591 cycle, as described in EN 15801 [44].

592

593 3.2. Assessment of the efficacy of the tested consolidation treatments

594 3.2.1 Tested treatments

595 Figure 8 a) and b) shows the evolution of the capillary fringe during the treatment of the

596 sound and pre-aged specimens, respectively, which are completely distinct. Concerning

597 sound specimens, only BS and 6hTUDA0.01 were able to reach the top of the

598 specimens after 3 hours, while all products reached the top of the specimens when

599 applied to pre-aged specimens.

600 Concerning the amount of absorbed product (figure 8 c and table 4), it varied between

601 2.7 and 4.1kg/m2 in sound specimens, and between 3.6 and 4.4kg/m2 in pre-aged

27
602 specimens. In both cases, the commercial product BS showed the highest final dry

603 residue per area (figure 8 c and table 4), of around 1.5 kg/m2, whereas the products

604 6hTUDA0.01 and 2hSiDETA presented the lowest final dry residue per area, between

605 0.6 and 0.7 kg/m3.

606 The discrepancies between sound and pre-aged specimens, especially on the kinetics of

607 the capillary fringe and on the amount of absorbed products, are a clear consequence of

608 the previously identified changes in the physical properties of the stones due to the

609 aging procedure.

610

611

612 Figure 8 – Capillary fringe evolution during the treatments performed in sound (a) and

613 in pre-aged specimens (b); amount of product absorbed and dry residue in sound and

614 pre-aged specimens (c).

615

28
616 Table 4 - Amount of product absorbed and dry residue in sound and pre-aged

617 specimens.

Product absorbed (kg/m2) Dry residue (kg/m2)

Product Sound Pre-aged Sound Pre-aged
BS 4.1 4.4 1.5 1.4
6hTUDA0.01 3.2 3.7 0.7 0.6
2hTUDA0.05 3.8 4.1 0.8 0.7
2hSiDETA 2.7 3.6 0.6 0.7
2hR 3.7 4.0 0.9 0.8
618

619

620

621 3.2.2. Efficacy of the tested consolidation treatments

622 The DDP profiles of tested treatments in the sound and pre-aged specimens are

623 exhibited in figure 9 a) and b), respectively.

624

625

626 Figure 9 – DDP (Ti-NTi) of sound (a) and pre-aged specimens (b) treated with the

627 different consolidants. The black dashed lines represent the standard deviation of the

628 drilling resistance of the sound stone.

29
629

630 All treatments applied to sound specimens revealed none or quite moderate strength

631 increase in depth. In fact, the treatments with 6hTUDA0.01, 2hTUDA0.05, 2hSiDETA

632 and 2hR only produced increments in a superficial zone of the specimens, until 2-5mm

633 depth, whereas inner consolidation was absent. BS also produced a significant strength

634 increase in a superficial thin layer, while a slight strength increase in-depth can be

635 unveiled.

636 The negative values in-depth obtained in some treatments are justified by the stone

637 variability, which is responsible for the fact that the average force of the stone after

638 treatment falls within the average force ± standard deviation of the untreated stone.

639 The capillary fringe on the lateral surface of the specimens showed that the majority of

640 the products have reached more than 2-5 mm height (figure 8 a); however, the drilling

641 resistance results suggested insufficient penetration of the active components from the

642 products. Other possible cause to explain this lack of strength gain in depth can be due

643 to the so-called reverse migration, where the consolidant becomes more concentrated on

644 the surface as a consequence of the evaporation of volatile components and capillary

645 action [58].

646 The information about the distribution of the consolidation action in depth could be

647 hardly acquired by other direct techniques commonly used to access the mechanical

648 properties of stones. For instance, a greater concentration of consolidant in the surface

649 will improve the surface hardness resulting in greater improvements on bending

650 strength because the surface layers contribute disproportionately to the resistance

651 against flexural loads [59]. However, especially in specimens treated with BS and

652 6hTUDA0.01 where marked strength peaks near the surface are evident, this

653 heterogeneous consolidation can be quite undesirable since it presents some potential

30
654 harmfulness. In practical terms, such more or less sharp boundary between the outer

655 harder layer and the inner less consolidated zone can increase the susceptibility of a thin

656 skin detachment due to further aging [1, 60, 61].

657 Contrarily to the treatments in sound specimens, all tested treatments when performed

658 in pre-aged specimens succeeded in promoting resistance increments in-depth, except

659 for 2hTUDA0.05. The product 6hTUDA0.01 was able to promote strength increase in-

660 depth due to its higher degree of pre-condensation (6 hours stirring) and the fact that its

661 consolidation ability is less sensitivity to the “carbonate environment” (less amount of

662 TUDA).

663 When comparing the strength gain provided by the reference product, 2hR, the use of

664 TUDA chains in low amounts caused additional strength increments up to

665 approximately 10 mm. The incorporation of SiDETA (2hSiDETA) also caused slight

666 additional strength increment up to 15 mm (in relation to 2hR).

667 Concerning the treatment using the commercial product BS the strength up to 5 mm

668 revealed to be inferior to the strength increment in further depth. The abrupt variation

669 seems to be typical of a stone fissure response with excessive thickness for the

670 consolidation capacity of the product (with the used treatment procedure) [30].

671 In any case, the BS product caused the most significant strength increase on the pre-

672 aged specimens throughout all the tested depth, mainly due to their higher final dry

673 residue (see table 4).

674 Interestingly, all the treated pre-aged specimens revealed a much more uniform

675 consolidation in-depth than the treated sound stone counterparts.

676 It is also worth to mention that the results of non-commercial solutions were obtained

677 by using much lower final dry residues than BS which indicates that more than the final

678 dry residue remaining within the stone pores (see table 4). Therefore, the different

31
679 efficacies lie on the different structures and amount of consolidation material inside the

680 pores.

681 Despite using the same products and the same treatment procedures, the consolidation

682 of the sound and pre-aged specimens showed to be quite distinct, either in the absolute

683 strength gains or in the consolidation behavior in terms of uniformity in-depth. The

684 higher consolidation action identified in the pre-aged treated specimens might result

685 from the increment in fine pores as a result of the artificial ageing by freeze-thaw

686 cycles. Therefore, the data emphasizes the importance of using pre-aged stones when

687 testing the efficacy of products and treatment procedures.

688 Furthermore, DDP method allowed to simultaneously analyzing the effect of different

689 consolidation treatments (enabling to distinguish different initial efficacies for the tested

690 treatments) and to detect potential harmful effects related to the formation of hardener

691 superficial layers.

692

32
693 5. Conclusions

694 Artificial aging by heating procedures causes none or minor modifications on the

695 physical and mechanical properties of the tested porous carbonate stone

696 (porosity=16%), although undesirable mineralogical/chemically modifications were

697 promptly detected through superficial color change even for a stone containing very low

698 amounts of other minerals (<0.1%). Such reduced alterations led to conclude that the

699 heating artificial aging procedures were not appropriate to damage alike porous

700 carbonate stones with the aim of being used to evaluate the potential initial efficacy of

701 subsequent consolidation treatments. On the other hand, freeze-thaw cycles produced

702 significant physical and mechanical alterations. Therefore, it was concluded that the

703 artificial aging by freeze-thaw cycles of porous stones composed essentially of calcium

704 carbonate and alike microstructural characteristics, is an alternative and suitable way to

705 induce “clean” physical degradation in the similar stones.

706 The consolidation action on sound and pre-aged stone specimens showed to be quite

707 distinct in terms of strength gain and uniformity in-depth. The DRMS besides having

708 characterized the loss of mechanical resistance due to the artificial aging allowed to

709 assess the initial efficacy of several alkoxysilane-based treatments, as well as to

710 discriminate their behavior in depth and to identify cases of potential harmful side

711 effects.

712 The achieved results highlighted the potential of using pre-aged specimens by freeze-

713 thaw and the usefulness of the procedures adopted for the assessment of the potential

714 efficacy of alkoxysilane-based consolidation treatments applied to porous carbonate

715 stones, which can be summarized by the following main steps: aging of the specimens

716 by freeze-thaw cycles until 3rd degradation class, as established in EN12371; assessment

717 and characterization of the degradation induced by the aging by means of visual

33
718 inspection, physical properties and DRMS; treatment of the pre-aged stone specimens;

719 assessment of the potential initial efficacy of the treatments by drilling resistance

720 measurements performed on the same pre-aged specimens before and after treatment,

721 which afterwards can be analyzed by using DDP method.

722

34
723 Acknowledgments

724 The authors acknowledge Fundação para a Ciência e Tecnologia (FCT) for the financial

725 support to CQE - UID/QUI/00100/2013, to Escola Superior de Tecnologia do

726 Barreiro/IPS for providing facilities to develop part of the experimental work and to

727 DIMPOMAR, LDA for providing the stone.

728 The author B. Sena da Fonseca also acknowledges Fundação para a Ciência e

729 Tecnologia (FCT) for the financial support through grant SFRH/BD/96226/2013.

730

35
731 5. References
732 1. Delgado Rodrigues, J., Consolidation of decayed stones. A delicate problem with few
733 practical solutions, in Proc. Int. Seminar on Historical Construction2001: Guimarães. p.
734 3-14.
735 2. Borsoi, G., et al., Understanding the transport of nanolime consolidants within
736 Maastricht limestone. Journal of Cultural Heritage, 2016. 18: p. 242-249.
737 3. Price, C.A. and E. Doehne, Stone Conservation: An Overview of Current Research2011:
738 Getty Conservation Institute.
739 4. Mosquera, M.J., J. Pozo, and L. Esquivias, Stress During Drying of Two Stone
740 Consolidants Applied in Monumental Conservation. Journal of Sol-Gel Science and
741 Technology, 2003. 26(1-3): p. 1227-1231.
742 5. Kim, E.K., et al., Effects of silica nanoparticle and GPTMS addition on TEOS-based stone
743 consolidants. Journal of Cultural Heritage, 2009. 10(2): p. 214-221.
744 6. Borsoi, G., et al., Effect of solvent on nanolime transport within limestone: How to
745 improve in-depth deposition. Colloids and Surfaces A: Physicochemical and Engineering
746 Aspects, 2016. 497: p. 171-181.
747 7. Rodriguez-Navarro, C., et al., Influence of Substrate Mineralogy on Bacterial
748 Mineralization of Calcium Carbonate: Implications for Stone Conservation. Applied and
749 Environmental Microbiology, 2012. 78(11): p. 4017-4029.
750 8. Sassoni, E., G. Graziani, and E. Franzoni, An innovative phosphate-based consolidant for
751 limestone. Part 1: Effectiveness and compatibility in comparison with ethyl silicate.
752 Construction and Building Materials, 2016. 102, Part 1: p. 918-930.
753 9. Sassoni, E., G. Graziani, and E. Franzoni, Repair of sugaring marble by ammonium
754 phosphate: Comparison with ethyl silicate and ammonium oxalate and pilot application
755 to historic artifact. Materials & Design, 2015. 88: p. 1145-1157.
756 10. Xu, F., et al., Use of coupling agents for increasing passivants and cohesion ability of
757 consolidant on limestone. Progress in Organic Coatings, 2014. 77(11): p. 1613-1618.
758 11. LI, D., et al., Effect of the addition of 3-glycidoxypropyltrimethoxysilane to
759 tetraethoxyorthosilicate-based stone protective coating using n-octylamine as a
760 catalyst. Bulletin of Materials Science, 2015. 38(1): p. 49-55.
761 12. Salazar-Hernández, C., et al., TEOS–colloidal silica–PDMS-OH hybrid formulation used
762 for stone consolidation. Applied Organometallic Chemistry, 2010. 24(6): p. 481-488.
763 13. Li, D., et al., The effect of adding PDMS-OH and silica nanoparticles on sol–gel
764 properties and effectiveness in stone protection. Applied Surface Science, 2013. 266(0):
765 p. 368-374.
766 14. Kapridaki, C. and N.-P. Maravelaki, TiO2–SiO2–PDMS nanocomposites with self-
767 cleaning properties for stone protection and consolidation. Geological Society, London,
768 Special Publications, 2015. 416.
769 15. Liu, Y. and J. Liu, Synthesis of TEOS/PDMS-OH/CTAB composite coating material as a
770 new stone consolidant formulation. Construction and Building Materials, 2016. 122: p.
771 90-94.
772 16. Xu, F., H. Liu, and D. Li, Synthesis of PDMS-SiO2 hybrids using different templates.
773 Journal of Porous Materials, 2016. 23(5): p. 1133-1141.
774 17. Sassoni, E., S. Naidu, and G.W. Scherer, The use of hydroxyapatite as a new inorganic
775 consolidant for damaged carbonate stones. Journal of Cultural Heritage, 2011. 12(4): p.
776 346-355.
777 18. Franzoni, E., et al., Artificial weathering of stone by heating. Journal of Cultural
778 Heritage, 2013. 14(3, Supplement): p. e85-e93.
779 19. Franzoni, E. and E. Sassoni, Comparison between different methodologies for artificial
780 deterioration of stone aimed at consolidants testing in 12th International Congress on

36
781 the Deterioration and Conservation of Stone2012: Columbia University, New York. p.
782 10.
783 20. Sassoni, E. and E. Franzoni, Influence of porosity on artificial deterioration of marble
784 and limestone by heating. Applied Physics A, 2014. 115(3): p. 809-816.
785 21. Martínez-Martínez, J., et al., Non-linear decay of building stones during freeze–thaw
786 weathering processes. Construction and Building Materials, 2013. 38: p. 443-454.
787 22. Ruedrich, J., D. Kirchner, and S. Siegesmund, Physical weathering of building stones
788 induced by freeze–thaw action: a laboratory long-term study. Environmental Earth
789 Sciences, 2011. 63(7): p. 1573-1586.
790 23. Benavente, D., Why pore size is important in the deterioration of porous stones used in
791 the built heritage Macla, 2011. 15: p. 41-42.
792 24. Sassoni, E., et al., Consolidation of calcareous and siliceous sandstones by
793 hydroxyapatite: Comparison with a TEOS-based consolidant. Journal of Cultural
794 Heritage, 2013. 14(3, Supplement): p. e103-e108.
795 25. Graziani, G., E. Sassoni, and E. Franzoni, Consolidation of porous carbonate stones by
796 an innovative phosphate treatment: mechanical strengthening and physical-
797 microstructural compatibility in comparison with TEOS-based treatments. Heritage
798 Science, 2015. 3(1): p. 1-6.
799 26. Ferreira Pinto, A.P. and J. Delgado Rodrigues, Stone consolidation: The role of
800 treatment procedures. Journal of Cultural Heritage, 2008. 9(1): p. 38-53.
801 27. Ferreira Pinto, A.P. and J. Delgado Rodrigues, Consolidation of carbonate stones:
802 Influence of treatment procedures on the strengthening action of consolidants. Journal
803 of Cultural Heritage, 2012. 13(2): p. 154-166.
804 28. Mosquera M.J, et al., New Nanomaterials for Protecting and Consolidating Stone.
805 Journal of Nano Research, 2009. 8: p. 1-12.
806 29. Ferreira Pinto, A.P., Conservação de Pedras Carbonatadas. Estudo e Selecção de
807 Tratamentos., 2002, U.T.L.: Lisboa.
808 30. Delgado Rodrigues, J., A.P. Ferreira Pinto, and D.R. Costa, Tracing of decay profiles and
809 evaluation of stone treatments by means of microdrilling techniques. Journal of
810 Cultural Heritage, 2002. 3(2): p. 117-125.
811 31. Svahn, H., Non-Destructive Field Tests in Stone Conservation: Final Report for the
812 Research and Development Project: Literature Study2006: Riksantikvarieämbetet.
813 32. Zarzuela, R., et al., CuO/SiO2 nanocomposites: A multifunctional coating for application
814 on building stone. Materials & Design, 2017. 114: p. 364-372.
815 33. Figueiredo, C., et al., Contribution to the technological characterization of two widely
816 used Portuguese dimension stones: the 'Semi-rijo' and 'Moca Creme' stones. Geological
817 Society, London, Special Publications, 2010. 333(1): p. 153-163.
818 34. Carvalho, J.M.F., et al., Maciço Calcário Estremenho - Caracterização da Situação de
819 Referência (Internal Report), 2011, LNEG: Lisboa. p. 42.
820 35. Sena da Fonseca, B., et al., Development of formulations based on TEOS-dicarboxylic
821 acids for consolidation of carbonate stones. New Journal of Chemistry, 2016. 40(9): p.
822 7493-7503.
823 36. Sena da Fonseca, B., et al., Polyethylene glycol oligomers as siloxane modificators in
824 consolidation of carbonate stones. Pure and Applied Chemistry, 2017. (Ahead of print).
825 37. Sena da Fonseca, B., et al., TEOS-based consolidants for carbonate stones: the role of
826 N1-(3-trimethoxysilylpropyl)diethylenetriamine New Journal of Chemistry (review
827 process), 2017.
828 38. Wacker, Technical data sheet for SILRES® BS OH 100, SILRES, Editor 2014. p. 3.
829 39. Snethlage, R., Stone Conservation, in Stone in Architecture, S. Siegesmund and R.
830 Snethlage, Editors. 2014, Springer Berlin Heidelberg. p. 415-550.

37
831 40. Aslanidou, D., I. Karapanagiotis, and C. Panayiotou, Tuning the wetting properties of
832 siloxane-nanoparticle coatings to induce superhydrophobicity and superoleophobicity
833 for stone protection. Materials & Design, 2016. 108: p. 736-744.
834 41. Leifeld, J. and I. Kögel-Knabner, Organic carbon and nitrogen in fine soil fractions after
835 treatment with hydrogen peroxide. Soil Biology and Biochemistry, 2001. 33(15): p.
836 2155-2158.
837 42. Mikutta, R., et al., Review. Soil science society of America journal, 2005. 69(1): p. 120-
838 135.
839 43. Commission:25-PEM, Tests defining the properties connected with the presence and
840 the movement of water. Matériaux et Construction, 1980. 13(3): p. 175-253.
841 44. CEN, Conservation of cultural property: Determination of water vapour permeability,
842 2010.
843 45. Sena da Fonseca, B., et al., Non-destructive and on site method to assess the air-
844 permeability in dimension stones and its relationship with other transport-related
845 properties. Materials and Structures, 2015. 48(11): p. 3795-3809.
846 46. Chakrabarti, B., T. Yates, and A. Lewry, Effect of fire damage on natural stonework in
847 buildings. Construction and Building Materials, 1996. 10(7): p. 539-544.
848 47. Ortiz, R., et al., Analysis of heat effects due to fires on calcareous stones, in Science and
849 Technology for the Conservation of Cultural Heritage2013, Informa UK Limited. p. 139-
850 142.
851 48. Borg, R.P., M. Hajpál, and Á. Torok, The fire performance of limestone, in Application of
852 Structural Fire Engineering2013: Prague, Czech Republic p. 6.
853 49. Sippel, J., et al., Decay of natural stones caused by fire damage. Geological Society,
854 London, Special Publications, 2007. 271(1): p. 139-151.
855 50. Dionísio, A., Stone decay induced by fire on historic buildings: the case of the cloister of
856 Lisbon Cathedral (Portugal). Geological Society, London, Special Publications, 2007.
857 271(1): p. 87-98.
858 51. Murray, H.H. and S. Lyons, Correlation of paper-coating quality with degree of crystal
859 perfection of kaolinite. Clays Clay Miner, 1956. 4: p. 31-40.
860 52. Swindale, L.D., The Crystallography of Minerals of the Kaolin Group, in Soil
861 Components: Vol. 2: Inorganic Components, J.E. Gieseking, Editor 1975, Springer Berlin
862 Heidelberg: Berlin, Heidelberg. p. 121-154.
863 53. Al-Omari, A., et al., Critical degree of saturation: A control factor of freeze–thaw
864 damage of porous limestones at Castle of Chambord, France. Engineering Geology,
865 2015. 185: p. 71-80.
866 54. Jain, S., Fundamentals of Physical Geology2013: Springer India.
867 55. Beck, K., et al., Characterization, water transfer properties and deterioration in tuffeau:
868 building material in the Loire valley—France. Building and Environment, 2003. 38(9–
869 10): p. 1151-1162.
870 56. Rozenbaum, O., et al., Significance of a combined approach for replacement stones in
871 the heritage buildings’ conservation frame. Comptes Rendus Geoscience, 2008. 340(6):
872 p. 345-355.
873 57. Török, Á., et al., Differences in texture, physical properties and microbiology of
874 weathering crust and host rock: a case study of the porous limestone of Budapest
875 (Hungary). Geological Society, London, Special Publications, 2007. 271(1): p. 261-276.
876 58. Hansen, E.F., R. Lowinger, and E. Sadoff, Consolidation of Porous Paint in a Vapor-
877 Saturated Atmosphere: A Technique for Minimizing Changes in the Appearance of
878 Powdering, Matte Paint. Journal of the American Institute for Conservation, 1993.
879 32(1): p. 1-14.
880 59. Schniewind, A.P. and P.Y. Eastman, Consolidant Distribution in Deteriorated Wood
881 Treated with Soluble Resins. Journal of the American Institute for Conservation, 1994.
882 33(3): p. 247-255.

38
883 60. Ferreira Pinto, A.P. and J. Delgado Rodrigues, Impacts of consolidation procedures on
884 colour and absorption kinetics of carbonate stones. Studies in Conservation, 2014.
885 59(2): p. 79-90.
886 61. Ferreira Pinto, A.P. and R.J. Delgado, Consolidation of carbonate stones: influence of
887 treating procedures on the strengthening action of consolidants. J Cult Herit, 2012. 13.
888
889

39