The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

THE USE OF HYDROXYAPATITE AS A NEW INORGANIC

CONSOLIDANT FOR DAMAGED CARBONATE STONES

Enrico Sassonia, Sonia Naidub, George W. Schererc,*

a Dipartimento di Ingegneria Civile, Ambientale e dei Materiali, Università di Bologna, via Terracini 28, 40131 Bologna, Italy,

Tel: +39 051 2090363, Fax: +39 051 2090322, e-mail: [email protected] Ph.D. Candidate

b Department of Chemical Engineering, Princeton University, Eng. Quad. E-226, Princeton, NJ 08544, USA,

Tel: +1 609 258 9089, Fax: +1 609 258 1563, e-mail: [email protected] Ph.D. Candidate

c Department of Civil and Environmental Engineering, Princeton University, Eng. Quad. E-319, Princeton, NJ 08544, USA,

Tel: +1 609 258 5680, Fax: +1 609 258 1563, e-mail: [email protected] Full Professor

* Corresponding Author

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

1

ABSTRACT

The feasibility and the effectiveness of using hydroxyapatite (HAP), formed by react-

ing limestone with a solution of diammonium hydrogen phosphate (DAP) in mild

conditions, as a consolidant for carbonate stones were investigated. Firstly, a novel

method for pre-damaging limestone was developed. Then, the effects of DAP solu-

tion concentration and reaction duration were evaluated to define the best treatment

conditions, and the strengthening effect was evaluated on artificially damaged Indi-

ana Limestone samples. Treated samples exhibit a significant increase in the dynamic

elastic modulus and tensile strength, which is attributed to microcrack reduction and

pore filling consequent to formation of calcium phosphate phases at grain boundaries,

as assessed by SEM/EDS and ESEM/EBSD. Consequent to a slight reduction of

coarser pores, as revealed by MIP, the sorptivity of treated samples is only slightly

reduced, so that water and water vapor exchanges with the environment are not sig-

nificantly blocked.

KEYWORDS

Hydroxyapatite; Consolidation; Carbonate stones; Dynamic elastic modulus; Tensile

strength

1. INTRODUCTION

Carbonate stones, such as limestone and marble, have been used since ancient

times in architecture and sculpture. As such lithotypes are subject to several weather-

ing mechanisms, whose harmfulness depends on stone tensile strength (e.g., salt crys-

tallization, freezing-thawing cycles, clay swelling, heating-cooling cycles), the devel-

opment of effective consolidants for carbonate stones is a key goal in cultural herit-

age conservation [1]. The immediate goal of consolidation is to restore the mechani-

cal integrity of weathered stone [2,3], and that is the focus of the present study. Of

course, it is ideal if the same treatment provides protection against further damage,

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

2

but that usually requires treatments in addition to consolidation [e.g., 1,4,5]. In fact,

the treatment to be described here is expected to provide corrosion resistance, but that

aspect will be explored in a future publication. The main consolidants currently used

for strengthening carbonate stones include lime-based, polymeric, and silicate

consolidants [6,7].

Lime-based consolidants, such as lime milk and lime water, aim to introduce

lime (Ca(OH)2) inside the pores of the stone. As Ca(OH)2 reacts with atmospheric

carbonic dioxide (CO2), calcium carbonate (CaCO3) is produced. Newly formed

CaCO3 is chemically compatible with the calcitic substrate and therefore can bond to

stone grains and strengthen the stone. In spite of the chemical compatibility, lime-

based treatments are characterized by some limitations, such as the reduced penetra-

tion depth, the extremely slow rate of conversion of Ca(OH)2 into CaCO3 and the

limited solubility of lime in water, causing chromatic alteration of stone surface [6,

8]. To avoid these limitations, studies on the use of nanolimes (i.e., lime particles

with submicrometric dimensions) have recently been undertaken [8,9].

Silicate consolidants, such as tetraethoxysilane (TEOS), aim to introduce sili-

con-based compounds (e.g., Si(OC2H5)4) into the pores [1,10]. Such compounds, in

contact with atmospheric moisture or liquid water, undergo a hydrolysis reaction, so

that ethoxy groups (OC2H5) are progressively replaced by hydroxyl groups (OH)

[11]. When hydroxyl groups of different molecules start to react, the molecules un-

dergo a condensation reaction and form a gel. In silicate stones, the deposited silica

gel can bond covalently to the grain surfaces, because the grains are covered with

silanol groups that can react with the silica gel, but in carbonate stones the bonding

between the silica gel and the grains is merely physical, as grain surfaces lack hy-

droxyl groups. As a consequence, silicate consolidants are much less effective on

carbonate stones than on silicate stones [1,12]. To improve the effectiveness of sili-

cate consolidants on carbonate stones, numerous coupling agents – having an anchor

group on one end that bonds to calcitic grains and a hydroxyl group on the other end

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

3

that bonds to silica gel – have been studied (e.g., tartaric acid, organoalkoxysilanes,

etc.) [1,10,13].

Polymeric consolidants, including acrylics, silicones, vinyls, polyesters, ure-

thanes and epoxies, aim to strengthen the stone by introducing macromolecules that

solidify when the solvent evaporates (thermoplastic polymers) or when curing agents

cross-link with the resin (thermosetting polymers), hence creating a solid network

that can bond the grains together. Polymeric consolidants have been widely used in

the past, as they produce consolidation and water repellency at the same time. How-

ever, the use of polymeric consolidants may lead to several problems, mainly related

to penetration depth, yellowing by ultraviolet rays, and biodeterioration owing to bac-

terial and fungal growth [6,14,15,16].

Considering the limitations of the above mentioned consolidants in strengthen-

ing carbonate stones, in this paper the effectiveness of hydroxyapatite (HAP) as an

inorganic consolidant for carbonate stones was investigated. The current study was

inspired by the work of Matteini et al. [17], who suggested that marble could be pro-

tected by coating it with a layer of calcium oxalate. A mild chemical reaction, where

calcite is exposed to a solution of ammonium oxalate in water, results in formation of

calcium oxalate monohydrate (whewellite, CaC2O4·H2O) on the surface of the stone.

Field studies [18] indicate that this treatment is helpful in delaying damage, but the

effect is not as dramatic as had been hoped. To understand why the oxalate treatment

is not more effective, we compared the crystal structures of calcite and oxalate, and

found that they are not compatible. That is, one cannot make a coherent (epitaxial)

layer of one on top of the other, because the atoms are arranged with different sym-

metries. This is important, because a layer that is porous or patchy can be undercut

and removed by the corrosion. Moreover, whewellite is not much less soluble than

calcite. In contrast, the mineral hydroxyapatite (HAP), Ca10(PO4)6(OH)2, which con-

stitutes our teeth and bones, is very durable, and there is a mild chemical reaction that

can convert calcite to hydroxyapatite. Moreover, the crystal structures of those two

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

4

minerals are quite similar, so it is likely that a coherent layer of apatite can be formed

on the surface of marble and limestone.

HAP has the formula Ca5(PO4)3(OH), but is usually written as

Ca10(PO4)6(OH)2, to denote that the crystal unit cell comprises two formula units.

HAP is expected to be an effective consolidating agent for carbonate stones since it:

(i) is notably less soluble than calcite (the solubility products at 25°C being Ksp =

1.6·10-117 for HAP [19] and Ksp = 3.4·10-9 for calcite [20]); since the formula

for HAP contains 18 ions and that of calcite contains 2, the solubility of the

“molecule” is cHAP = Ksp1/18 ≈ 3.25 x 10-7 M and cCalcite ≈ 5.83 x 10-5 M (or, in

terms of grams/liter or moles of Ca2+ dissolved in a liter of solution, HAP is

about 18 times less soluble than calcite)

(ii) has a dissolution rate [21,22,23] about 4 orders of magnitude lower than that

of calcite [24]

(iii) has a crystal structure similar to calcite (the unit cell being hexagonal for HAP

[25] and rhombohedral for calcite, but often described as hexagonal [26]);

(iv) has lattice parameters close to calcite (respectively, a = b = 9.43 Å and c =

6.88 Å for HAP [27] and a = b = 9.96 Å and c = 17.07 Å for calcite, consider-

ing two molecules per unit cell) [28]).

Thus, the lattice parameters of HAP and calcite differ by only ~5%, indicating com-

patibility of the structures sufficient to permit epitaxial growth. The good lattice

match between calcite and HAP favors nucleation of the phosphate layer on the sur-

face of marble or limestone, and encourages strong bonding of the layer to the sub-

strate.

Indeed, apatite coatings have been found on ancient monuments, apparently

formed by slow weathering of ancient treatments, such as milk-based coatings

[29,30]. The fact that these apatite layers still exist, and are not being deposited by

currently active processes (such as microbial activity), implies that they are durable

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

5

and coherent. One goal of our research is to develop a practical method for applying

such coatings, and to evaluate their ability to protect carbonate stones from corrosion.

A secondary benefit is to restore the mechanical integrity of weathered stone, and that

is the focus of the present paper.

Numerous techniques, divisible into solid-state and wet methods, have been

developed in bioceramic and biomineralisation studies for HAP preparation

[19,31,32,33,34,35]. Solution-deposited phosphates have also been used to consoli-

date loose sand [36,37] and to preserve frescoes [38]. Even among wet methods (such

as precipitation, hydrothermal, and hydrolysis processes), many are not applicable in

the case of monument restoration, as they involve high temperatures, hazardous

chemicals or extreme pH values. The most suitable method for an application in situ

for stone strengthening is the one [39] involving the reaction of calcite and

diammonium hydrogen phosphate (DAP, (NH4)2HPO4) to form carbonate-containing

HAP at temperature close to room temperature (40°C). According to Kamiya et al.,

HAP is formed by the following chemical reaction:

10CaCO3 + 5(NH4)2HPO4 →

Ca10(PO4,CO3)6(OH,CO3)2 + 5(NH4)2CO3 + 3CO2 + 2H2O,

where PO43- and OH- can be partially replaced by CO3

2-.

The resulting HAP is typically non-stoichiometric, as it contains carbonate

ions, and its precipitation is expected to be preceded by the formation of several in-

termediate metastable phases, such as monocalcium phosphate monohydrate and an-

hydrous (MCPM and MCPA, respectively), dicalcium phosphate dihydrate and anhy-

drous (DCPD and DCPA, respectively), octacalcium phosphate (OCP), amorphous

calcium phosphate (ACP) and/or calcium-deficient hydroxyapatite (CDHA). Such

precursor phases, whose formation depends on reaction conditions (e.g., degree of

supersaturation, temperature, pH, presence of foreign ions, etc.), are expected to

transform eventually into HAP by dissolution and reprecipitation processes [27,40].

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

6

Both cationic and anionic substitutions can be made in the HAP structure. The

substitution of fluoride ions into hydroxyapatite is known to improve its chemical du-

rability slightly [41], but the fluoride precursors are relatively dangerous to handle on

the scale that would be required for treating a monument. Replacing calcium cation

with strontium or barium cations in HAP improves the lattice match with calcite [42]

at the cost of a small increase in dissolution rate [43], so doping with those cations

will be investigated in the future. Improvement in the lattice mismatch would favor

heterogeneous nucleation of the HAP layer on calcite and thereby facilitate formation

of a hermetic coating.

2. EXPERIMENTAL PROCEDURE

2.1 Materials

Indiana Limestone (IL) was used to test the HAP consolidation effect. IL is a po-

rous carbonate stone (porosity ~14%), mainly made of calcite (> 97 wt%) and a small

amount of other components (e.g., Al2O3, SiO2 and MgCO3) [44]. Some of our sam-

ples showed traces of Mg by EDS, but no magnesite was detected by XRD. Indiana

Limestone mainly consists of calcite cemented oolites, even though a small amount

of sparry calcite crystals may be also present. Cubic samples (5 cm side) and cylin-

drical samples (2 cm diameter) were respectively cut and core-drilled from 5 cm thick

IL plates.

The DAP was purchased from Fisher Scientific (assay ≥ 98.0 %, reagent grade)

and used as received. All water used was de-ionized.

2.2 Artificial damaging

To test the consolidation effectiveness on weathered samples with uniform

characteristics, for reproducibility’s sake, IL samples were artificially damaged by

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

7

heating, following a suggestion by Flatt [45]. To assess the heating temperature ef-

fect, ten cubic samples (two for each temperature) were heated at 100°C, 200°C,

300°C, 400°C and 500°C for 1 hour. To assess the heating duration effect, nine cubic

samples (three for each duration) were heated for 1 hour, 4 hours and 16 hours at

300°C.

The damaging effect was evaluated by comparing the dynamic elastic modulus

(Ed) of the samples before and after heating. The modulus was calculated according

to the formula Ed = ρV2 where ρ is the density and V the pulse velocity, measured us-

ing a commercial instrument (PUNDIT) with 54 kHz transducers. For cubic samples,

the Ed was calculated as the average for the three values measured in the three direc-

tions.

The tensile strength (σt) variation was also evaluated by comparison between

three core-drilled samples heated to 300°C for 1 hour and three unheated core-drilled

samples. The tensile strength was measured by performing the Brazil test.

2.3 Definition of treatment conditions

The effects of DAP solution concentration and treatment duration were evalu-

ated on cubic samples artificially damaged by heating to 300°C for 1 hour.

2.3.1 Concentration effect

Four DAP solutions were prepared: 0.1 M, 0.5 M, 1.0 M and 4.4 M (the last

corresponding to the saturation concentration). All treatments were performed at

room temperature. In the following tests, the DAP was introduced by immersion of

the samples; however, other tests were performed by brushing on the solution until it

was rejected, and those samples performed at least as well.

The initial Ed was evaluated for eight cubic samples. The samples were then

water-saturated by immersing them in water up to one half of the height, letting them

soak up the water until they appeared completely wet (~60 minutes) and then adding

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

8

water up to 5 mm below the upper face of the cubes (the same procedure was fol-

lowed to solution-saturate all the samples described later in the paper). After 24

hours, the samples were taken out of the water, the surface gently dried with a towel

to remove excess water and the increase in the dynamic elastic modulus owing to wa-

ter filling the pores evaluated. The samples were then dried under a fan at room tem-

perature until constant weight (the same procedure was followed to dry all the sam-

ples described later in the paper). This procedure was used to avoid damage from

heating. Once completely dry, two samples for each concentration were immersed in

the DAP solutions. To avoid water evaporation and hence concentration changes, all

solution containers were sealed with parafilm.

The consolidation progress through time was monitored after 1, 2, 4, and 8

days by taking the samples out of the DAP solutions, removing the excess solution

from the surface, measuring Ed and then putting the samples back in the solutions.

After 8 days, all the samples were removed from the solutions and immersed in water

to remove the unreacted DAP from inside the pores. After washing for 3 days (the

water being completely renewed every day), the samples were finally dried and Ed

measured again. The washing procedure was used to insure that changes in Ed were

not caused by residual reactants or byproducts.

2.3.2 Time effect

The initial Ed was measured for eight cubic samples. The samples were then

immersed in a 1.0 M DAP solution and extracted, two at a time, after 1, 2, 4, and 8

days. Upon extraction, the samples were water-immersed for 3 days and then dried.

Then Ed was finally measured again.

2.4 Evaluation of consolidation effects

The effects of the consolidation treatment were evaluated on samples artifi-

cially damaged by heating to 300°C for 1 hour and then treated by immersion in a 1.0

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

9

M DAP solution for 2 days, then extracted, water-immersed for 3 days and dried. For

comparison, some samples were treated by brushing on the solution of DAP until it

was rejected.

2.4.1 Mechanical properties

The improvement of the mechanical properties of the stone was evaluated by

comparing (i) the dynamic elastic modulus of ten core-drilled samples before and af-

ter treating; (ii) the tensile strength of the same ten samples, after treating, with the

tensile strength of ten untreated samples. To further investigate the role of treatment

duration, the dynamic elastic modulus and the tensile strength of ten core-drilled

samples treated for 4 days were also measured.

2.4.2 Formation of calcium phosphate phases

The calcium phosphate phases formed after treatment were observed on a

scanning electron microscope (SEM, Philips XL30 Field-Emission-Gun). The sam-

ples for the observation were obtained by hammer fracturing of a 2 cm diameter, 5

cm length cylindrical sample which had been core-drilled from the center of a cubic

specimen. In this way, the penetration depth of the treatment could be also estimated.

The elemental composition of the observed phases was assessed by energy dispersive

X-ray spectrometry (EDS) performed on the scanning electron microscope. To identi-

fy the calcium phosphate phases, electron back-scattered diffraction (EBSD) was per-

formed in an environmental scanning electron microscope (FEI Quanta 200 ESEM).

2.4.3 Resistance to wetting/drying cycles

The resistance to wetting/drying cycles (w/d cycles) of the consolidating cal-

cium phosphate phases was evaluated by comparing the dynamic elastic modulus of

ten core-drilled samples, treated with the DAP solution, before and after each of 5

w/d cycles. The w/d cycles were performed by water-saturating the core-drilled sam-

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

10

ples for 24 hours (following the procedure described in §2.3.1) and then drying them

under a fan until constant weight.

After 5 w/d cycles, the tensile strength of the samples was measured and com-

pared to that of untreated samples and treated samples not subjected to w/d cycles.

2.4.4 Pore size distribution

The modification of the pore size distribution was evaluated by performing

mercury intrusion porosimetry (MIP, Micrometrics 9410) on fractured samples ob-

tained from various depths (0÷5 mm, 10÷15 mm and 20÷25 mm) of a treated cubic

specimen.

2.4.5 Sorptivity

The effect of HAP treatment on sorptivity (or, rate of capillary absorption)

was evaluated by comparing the sorption rates of a treated sample and an untreated

one. The sample (2.5 x 1.5 x 5 cm) was suspended below a balance, and a large dish

of water was raised until it just came into contact with the bottom of the sample. The

gain in weight was continuously recorded by a computer.

2.4.6 Color change

The color changes between untreated samples, treated samples and treated

samples subjected to w/d cycles were evaluated by measuring the CIE 1976 L*a*b*

color parameters and then calculating the total color difference ΔE* = (ΔL*2 + Δa*

2 +

Δb*2)1/2, using a commercial colorimeter (X-rite).

2.4.7 Effectiveness for different weathering levels

The effectiveness of the consolidating treatment on samples weathered to dif-

ferent extents was evaluated by measuring the variation of the dynamic elastic modu-

lus after consolidation for the same samples that had been used to evaluate the effect

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

11

of heating temperature on the artificial damage process (i.e., on samples heated to

100, 200, 300, 400, and 500°C for 1 hour).

3. RESULTS

3.1 Artificial damaging

The samples exhibited a decrease in Ed proportional to the heating temperature

(Figure 1). For samples heated at the same temperature for different periods (1, 4, 16

hours), no difference in the Ed decrease was found. For samples heated to 300°C for 1

hour, whose Ed was reduced by 43.4% after heating, the tensile strength was 27.1%

lower than that of the unheated samples (Table 1).

3.2 Definition of treatment conditions

3.2.1 Concentration effect

After water saturation for 24 hours, all the samples showed an increase in the

Ed, compared to dry samples, This is a result of the higher density of the wet sample,

and does not indicate stiffening. To demonstrate an increase in stiffness, the DAP-

saturated samples must show Ed higher than the water-saturated stone. Once dried

again and then solution-saturated, the samples exhibited an increase in Ed that rose

with solution concentration up to 1 M, but changed little between 1 and 4.4 M (Figure

2).

3.2.2 Time effect

Samples treated for increasing periods exhibited progressive Ed increases, as

indicated in Table 2. Notably, the increase in Ed measured after 2 days was already

91% of that measured after 8 days.

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

12

3.3 Evaluation of consolidation effects

3.3.1 Mechanical properties

Treated samples exhibited a significant Ed increase, amounting to +96.0% af-

ter 2 days and +96.8% after 4 days of treatment (Table 2). Analogously, the tensile

strength of samples treated for 2 days and 4 days was, respectively, 24.5% and 28.8%

higher than that of untreated samples (Table 2).

3.3.2 Calcium phosphate phase formation

As can be seen in the SEM images in Figure 3, calcium phosphate phases

formed on the surface of the pores and at grain boundaries. According to the EDS re-

sults (Figure 4), calcium, carbon, oxygen and phosphorus are present. No nitrogen

was detected, so there was no residual DAP. During the reaction, there was an odor of

ammonia, indicating that the ammonium carbonate by-product evaporated. Further

investigation of the calcium phosphate phases by EBSD identified:

(i) HAP, Ca10(PO4)6(OH)2 (Figure 5a);

(ii) octacalcium dihydrogen phosphate esa(phosphate), Ca8H2(PO4)6, which corre-

sponds to the general formula Ca10-x(HPO4)x(PO4)6-x(OH)2-x (with 0 ≤ x ≤ 2),

used in the literature to describe calcium-deficient hydroxyapatites [35], and

which, therefore, in the rest of the paper is referred to as calcium-deficient hy-

droxyapatite (CDHAP) (Figure 5b);

(iii) octacalcium dihydrogen esa(phosphate) monohydrate, Ca8H2(PO4)6·H2O,

which differs from OCP described in the literature [19] by being a monohy-

drate, rather than a pentahydrate, and which, therefore, in the rest of the paper

is called octacalcium phosphate monohydrate (OCPM) (Figure 5c);

(iv) calcium dihydrogen phosphate monohydrate, Ca(H2PO4)2·H2O, which is de-

scribed in the literature [19] as monocalcium phosphate monohydrate

(MCPM) (Figure 5d);

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

13

As the above mentioned phases were detected in samples coming from differ-

ent depths (0-5 mm and 15-20 mm) of a treated cubic specimen, the treatment pene-

tration depth can be assessed as greater than 2 cm.

3.3.3 Resistance to wetting/drying cycles

As illustrated in Figure 6, treated and dried samples that were subjected to

wetting/drying cycles experienced a 13.9% loss of the Ed improvement, which de-

creased from +93.8% (before the w/d cycles) to +80.7% (after w/d cycles) (Table 2).

The tensile strength comparison between untreated samples, treated samples and

treated samples subject to 5 w/d cycles shows a reduction of 31.4% of the σt im-

provement, which was +24.5% for samples not subject to w/d cycles and +16.8% for

samples subject to w/d cycles (Table 2).

3.3.4 Pore size distribution

The two samples that had been heated to 300ºC, but not treated with DAP (la-

beled “Untreated”) show good reproducibility (Figure 7). After the treatment, there is

a slight reduction of total open porosity that is more significant within 5 mm of the

surface. There is a decrease in the fraction of coarser pores (having radius larger than

1 μm) and an increase in the percentage of finer pores (having radius smaller than 0.1

μm). The latter effect is caused in part by the phosphate layer reducing the size of the

pores, and in part by the porosity within that layer.

3.3.5 Sorptivity

The sorptivity comparison between untreated and treated samples (Figure 8)

shows that the latter experienced a reduction of 44.0% of the sorptivity, which de-

creased from 0.0452 g/cm2·min1/2, before the treatment, to 0.0253 g/cm2·min1/2, after

the treatment.

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

14

3.3.6 Color change

Compared to untreated samples, treated samples and treated samples subjected

to w/d cycles exhibited a ΔE* = 7.65 and ΔE* = 6.45, respectively (values are averag-

es for 15 measurements). In most of the cases, the total color change is mainly due to

a change in the sample lightness (L* parameters), which generally increased after the

treatment.

3.3.7 Effectiveness for different weathering levels

Figure 9 shows the improvement in Ed for samples subjected to heat treatment

at increasing temperature. For treatment temperatures up to about 200˚C, the treated

samples have a higher Ed than the virgin stone. After treatment above 300˚C, the

modulus is not fully restored, but the fractional increase is greater. As indicated in

Figure 10, for a sample heated to 500˚C, the treatment produces an increase of more

than 120% in the modulus of the damaged stone.

4. DISCUSSION

4.1 Artificial damaging

The decreases in dynamic elastic modulus and tensile strength exhibited by the

heat-treated samples (Table 1) are to be ascribed to the anisotropic thermal expansion

of the calcite crystal, which expands parallel and contracts normal to the crystallo-

graphic c-axis [46,47]. As a consequence, stresses are generated at grain boundaries,

resulting in microcracks responsible for the decreases in Ed and σt. While the decrease

in Ed proved to be a linear function of heating temperature (Figure 1), consistent with

the thermal strain being a proportional to temperature, the heating duration proved to

have no effect on Ed. Evidently, as soon as the stone specimen reaches the heating

temperature, stresses arise and microcracks open; afterwards, sustaining the same

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

15

temperature for prolonged periods does not cause any further damage, and new

microcracks are generated only if the temperature is further increased.

The decrease in Ed assessed for actual stoneworks exposed to natural weather-

ing in the field was found to be ~36% [48], so a heating temperature of 300°C, re-

sponsible for an Ed decrease of ~43%, was chosen to artificially damage the IL sam-

ples for the prosecution of the study and the evaluation of treatment effectiveness.

Heating thus proved to be a very efficient and controllable method to artifi-

cially damage limestone samples to a desired “weathering” level, ensuring significant

constancy of the final specimen properties. Studies of the microstructural changes

caused by this treatment in limestone and marble are underway.

4.2 Definition of treatment conditions

The increase in Ed observed after water saturation for 24 hours (shown at Time

= -10 in Figure 2), is owing to water filling the pores, which allows the acoustic pulse

to travel faster inside the stone specimen, so that an increase in the elastic dynamic

modulus is recorded. After drying and resaturating with DAP solution for 24 hours,

the Ed increased compared to the sample containing only water. The difference be-

tween the modulus in solution and that in water rose in proportion to the solution

concentration (Time = 1 in Figure 2). Afterwards, as the reaction between calcite and

DAP proceeds, further Ed increases are achieved after 2, 4 and 8 days (Time = 2, 4

and 8 in Figure 2, respectively).

Unexpectedly, after 8 days of reaction both the samples immersed in the 4.4 M

DAP solution exhibited an Ed decrease (Time = 8 in Figure 2) and then a new in-

crease after drying (Time = 21 in Figure 2). Since all the other samples experienced a

reduction of the Ed after drying, as a consequence of the evaporation of water filling

the pores, it is reasonable to assume that the samples treated with 4.4 M DAP solution

also underwent such a reduction in Ed and that, therefore, the Ed value measured after

8 days of treatment was underestimated. The explanation for this underestimation

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

16

may be found considering that, during the treatment period, some superficial deposits

formed on the specimens treated with the 4.4 M DAP solution (whose concentration

corresponds to the saturation concentration). Since the pulse velocity measurement

highly depends on the contact between the specimen and the instrument, the surface

deposits presumably impeded a perfect contact with the transducers, resulting in an

underestimation of the elastic modulus. Once the samples were removed from the so-

lutions, the surface was gently washed and left in water for 3 days, so the superficial

deposits were mostly removed and thus the real Ed value could be estimated for the

dry samples.

Since treating the samples with a 1.0 M DAP concentration for 8 days pro-

duced an increase in Ed of 62% and since samples treated with a 4.4 M DAP concen-

tration, even though exhibiting a higher Ed increase, also exhibited surface deposits,

the 1.0 M DAP concentration was chosen for the prosecution of the study and the

evaluation of treatment effectiveness.

To confirm that the temporal trend of increase in Ed, exhibited by solution-

saturated samples, actually corresponds to the temporal trend experienced by dried

samples, the time effect was evaluated on dried samples treated with a 1.0 M DAP

concentration for different periods. Since the 91% of the 8-day increase in Ed was al-

ready achieved after 2 days of treatment, the latter duration – also corresponding to a

feasible treatment period in the field – was chosen for the prosecution of the study

and the evaluation of consolidation effectiveness.

4.3 Evaluation of consolidation effects

The core-drilled samples used to evaluate the mechanical properties exhibited

significant increases in dynamic elastic modulus and tensile strength, comparable to

the strengthening effect achieved by silicate consolidants applied on silicate stones

[49].

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

17

The core-drilled samples exhibited an Ed increase substantially higher than the

cubes treated in the same conditions (i.e., 1.0 M DAP concentration for 2 days): in-

deed, the average Ed increase amounted to 96.0% for the cylinders and 60.7% for the

cubes. To find the explanation for such a difference one should firstly consider that

the Ed increase calculated for each cube is the average for the three values calculated

for the three sides of the cube. Now, the Ed increase measured for the cube side that

corresponds to the cylinders’ height (and that also corresponds to the original thick-

ness of the plate the specimens were cut from) is actually much closer to the Ed in-

crease found for the cylinders (on the average, it amounts to 82.0%). As the Ed in-

crease measured for the other two sides of the cubes are lower (on the average,

53.0%), the overall Ed increase, calculated by averaging the values for the three sides,

is substantially lower. The reason for such anisotropic behavior may be found in

some “bedding plane” effects that have been described for Indiana Limestone: even if

no bedding plane is discernible and the change in physical properties with direction is

small, relatively large differences in the energy necessary to remove a unit volume of

rock have been assessed for different directions [50].

The observed improvement in mechanical properties is to be ascribed to the

deposition of calcium phosphate phases on the internal surfaces of the pores and at

grain boundaries (Figure 3). In fact, the deposition of calcium phosphate phases at the

contact points between grains led to a reduction of the microcracks that had opened

during heating, resulting in a greater resistance to crack propagation and new crack

formation.

The calcium phosphate phases include, alongside HAP, several intermediate

phases that are usually expected to precede stoichiometric HAP formation (Figure 5).

Such precursor phases are characterized by a Ca/P ratio lower than that of HAP:

whereas Ca/P = 1.67 for HAP, Ca/P = 0.5 for MCPM and Ca/P = 1.33 for OCPM and

CDHAP [19]. Thus, formation of MCPM, OCPM and CDHAP may be favored over

HAP, since a lower amount of Ca2+ ions, coming from calcite dissolution, is needed.

As a consequence of the calcium-deficiency, the intermediate phases are thermody-

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

18

namically less stable and more soluble than HAP: while the solubility of HAP at

25°C is ~0.0003 g/l, it is approximately ~0.0081-0.0094 g/l for OCPM and CDHAP

and ~18 g/l for MCPM [19].

Considering the high solubility of MCPM, potentially detrimental to the con-

solidation treatment effectiveness after exposure of the stone to water, the resistance

of the treatment to wetting/drying cycles was tested. As shown in Figure 6, after the

first w/d cycle the samples exhibited an Ed decrease of about 13.9% (from +93.8% to

+80.7%) and then, for the next 4 w/d cycles, Ed remained constant. The tensile

strength was then tested. Compared to that of samples treated but not subject to w/d

cycles, the tensile strength after 5 w/d cycles was 31.4% lower (on the average, 3.8

MPa, the average tensile strength of untreated and treated samples being 3.3 MPa and

4.1 MPa, respectively, Table 2).

The partial loss of the improvement in mechanical properties achieved after

the w/d cycle is related to the weight loss the samples experienced. Indeed, after the

treatment with the 1.0 M DAP solution, the core-drilled samples exhibited an average

weight gain of 0.1 g and an increase in Ed. Then, after the first w/d cycle, an average

weight loss of 0.03 g occurred and, correspondingly, a decrease in Ed was recorded.

Afterwards, the weight remained substantially stable and so did Ed.

As for the weight gain after the treatment, it is reasonable to ascribe it mainly

to the formation of calcium phosphate phases. Indeed, even though stoichiometric

HAP is supposed to form with no significant weight increase – as 10 moles of CaCO3

(totally weighing 1000 g) are expected to dissolve to form 1 mole of HAP (weighing

988 g) – the three intermediate calcium phosphate phases detected are expected to

form with a weight increase. Assuming that a number of CaCO3 moles corresponding

to the number of calcium atoms has to dissolve to form 1 mole of each of the inter-

mediate calcium phosphate phases, 1 mole of CaCO3 (weighing 100 g) is needed to

form 1 mole of MCPM (weighing 252 g) and 8 moles of CaCO3 (totally weighing

800 g) are needed to form 1 mole of MCPM (weighing 910 g) and 1 mole of CDHAP

(weighing 892 g). In addition, the HAP and the precursor phases formed by the reac-

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

19

tion of calcite with DAP may likely contain CO32- ions partially replacing PO4

3- and

OH- ions (cf. §1), which, in the latter case, would bring a further increase in weight.

Another possible reason for the weight increase might be the presence of some unre-

acted DAP inside the stone pores, notwithstanding the 3 day period of washing after

the treatment. However, the time required for unreacted ions to diffuse from the sam-

ples (assuming a diffusion coefficient D = 4·10-6 cm2/s for ions in water solutions in-

side of stone pores [51] and the radius of the samples being r = 1 cm) should amount

to t ≈ r2/D = 2.9 days, which actually corresponds to the time the samples were left in

D.I. water after treatment. In addition, in the EDS spectra no nitrogen peak, owing to

unreacted DAP still present inside the pores, was detected (Figure 4). Nevertheless,

the most likely cause of the weight increase seems to be the formation of the interme-

diate calcium phosphate phases, particularly the MCPM.

The main reason for the weight loss after the first w/d cycle may be found in

the high solubility of MCPM, that dissolves in contact with water, hence reducing the

mechanical property improvement achieved after treatment. However, after the first

w/d cycle, sample weight and Ed remained constant, suggesting that all the MCPM

formed was removed after the first contact with water.

The effects of a prolonged exposure to environmental weathering agents need

to be tested on some treated specimens exposed to outdoor conditions. Indeed, after

prolonged contact with rainwater, the intermediate calcium phosphate phases may

tend to dissolve and reprecipitate. This could even lead to the formation of more sta-

ble phases, such as HAP, and hence to a further strengthening of the stone; on the

other hand, the deposition of new phases may cause cracking and detachments, thus

damaging the stone. Therefore, field testing of the treated stones is essential.

The formation of HAP, rather than less stable precursor phases, during the re-

action between stone calcite and DAP might be promoted by providing calcium ions

from the external. Indeed, as the only calcium ions participating to the reaction come

from calcite dissolution in contact with the DAP solution [52], providing additional

calcium ions (for instance, adding calcium chloride to the DAP solution) could favor

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

20

the formation of phosphate phases with an higher calcium content, closer to HAP’s

one. Recent experiments show that this is indeed the case, resulting in faster growth

and better coverage of HAP on calcite [53]. Tests are underway to verify whether re-

duced porosity in the HAP deposit leads to enhanced consolidation and reduced solu-

bility.

As a consequence of the deposition of calcium phosphate phases on internal

surfaces of the pores, a slight reduction of the total open porosity was observed (Fig-

ure 7). Different variations of total open porosity and pore size distribution were de-

tected at different depths of a treated specimen, the greatest pore reduction being ob-

served for the most superficial sample (0-5 mm depth). As the samples were treated

by partial immersion and capillary rise, uniform solution penetration and reaction be-

tween the solution and the stone can be expected. During the drying phase, the solu-

tion moves towards the surface where evaporation occurs. As the solution may still

contain unreacted DAP and/or ions coming from dissolution of the most soluble

phase (i.e., MCMP), higher phosphate ion concentrations may occur near the stone

surface, leading to more abundant precipitation of calcium phosphate phases and

hence to a larger reduction of pore size. A relatively dense zone at the surface can

lead to damage, if salts accumulate behind it, but that is not expected in the present

case, owing to the small change in sorptivity caused by the treatment.

The average pore volume is 0.0624 cm3/g for the untreated samples and

0.0611 cm3/g for the treated samples, for an average pore volume reduction of 0.0013

cm3/g, or ~1%. This should correspond to the weight gain consequent to calcium

phosphate phase deposition, which amounts, on the average, to 0.0030 g/g. As the

densities of the calcium phosphate phases range from 2.23 g/cm3 (for the MCMP

[19]) to 3.16 g/cm3 (for the HAP [19]), the weight gain related to pore volume reduc-

tion is expected to amount to 0.0029÷0.0041 g/g. As the measured weight gain lies

close to the lower bound, an abundance of the phase with the lowest density (viz.,

MCMP) seems to be indicated, as suggested by thermodynamic considerations.

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

21

As a result of the reduction in coarser pores, which allow a large quantity of

water to quickly enter the stone, after the treatment a reduction of stone sorptivity

was found (Figure 8). In particular, the treated sample exhibited an initial sorption

rate markedly lower than the untreated sample, as a consequence of the higher pore

system modification that occurred near the surface. Afterwards, the treated sample’s

sorptivity was 44.0% lower than that of the untreated sample, until about 800 min-

utes, when the sorptivity of the two samples approaches the same value. The modest

reduction in sorptivity means that water and water vapor exchanges between the stone

and the environment are not significantly blocked.

As a consequence of the formation of calcium phosphate phases, color chang-

es were detected for treated samples (ΔE* = 7.65) that were slightly reduced follow-

ing w/d cycles (ΔE* = 6.45). Even after the w/d cycles, the change still remains

slightly higher than the threshold (ΔE* ≤ 5.00) generally accepted for stones subject-

ed to conservation treatments [49]. Nevertheless, it is noteworthy that a higher color

difference (amounting to ΔE* = 11.52, average for 10 measurements), was assessed

between different untreated samples, suggesting that color differences due to consoli-

dating treatment are much less pronounced than natural differences owing to stone

color variability.

For stone that has been weathered to the point that Ed decreases by half, the

DAP treatment can fully restore the modulus. For more seriously damaged stone, alt-

hough the original properties are not obtained following treatment, the relative im-

provement increases.

5. CONCLUSIONS

The use of HAP and HAP precursor phases, formed by reaction of a DAP so-

lution with calcite in limestone, seems to be a promising consolidating technique for

carbonate stones. Significant increases in the mechanical properties can be achieved

after the treatment, which has the advantage of consisting of a non-hazardous aque-

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

22

ous solution of DAP, able to penetrate deeply into the stone (> 2 cm) and bring sig-

nificant strengthening after just 2 days of reaction. This duration of exposure can be

achieved in the field by covering the treated surface to prevent evaporation. In the

lab, the performance of the treatment is equally good, whether the solution is brushed

onto the surface to the point of refusal, or introduced by immersion.

The treatment proved to be fairly resistant to wetting/drying cycles, and the

loss that did occur could be attributed to dissolution of an unstable phosphate com-

pound. The stability of the treatment is expected to increase if process modifications,

such as addition of calcium salts, result in formation of more stable phases.

ACKNOWLEDGEMENTS

The authors thank the Kress Foundation and the Getty Conservation Institute for fi-

nancial support. The authors also acknowledge the usage of PRISM Imaging and

Analysis Center, which is supported in part by the NSF MRSEC program through the

Princeton Center for Complex Materials (grant DMR-0819860). We are also indebted

to Dr. George Wheeler (Metropolitan Museum of Art) for the use of the colorimeter.

REFERENCES

[1] G.W. Scherer, G.S. Wheeler, Silicate consolidants for stone, Key Engineering Materials 391 (2009) 1-25

[2] G.G. Amoroso and V. Fassina, Stone decay and conservation, Elsevier, New York, 1983

[3] L. Lazzarini and M. Laurenzi Tabasso, Il restauro della pietra, CEDAM, Padua, 1986

[4] E. Wendler, D.D. Klemm, and R. Snethlage, "Consolidation and hydrohobic treatment of natural stone", pp. 203-212 in Proc. 5th Int. Conf. on Durability of Building Materials and

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

23

Components, eds. J.M. Baker, P.J. Nixon, A.J. Majumdar, and H. Davies (Chapman & Hall, London, 1991)

[5] J. Houck and G.W. Scherer, Controlling stress from salt crystallization, pp. 299-312 in Fracture and Failure of Natural Building Stones, ed. S.K. Kourkoulis, Springer, Dordrecht, 2006

[6] E.M. Winkler, Stone in architecture – Properties, durability, Springer, New York, 1997

[7] K.L. Gauri, J.K. Bandyopadhyay, Carbonate stone: chemical behavior, durability and conserva-tion, Wiley Interscience, New York, 1999

[8] V. Daniele, G. Taglieri, R. Quaresima, The nanolimes in Cultural Heritage conservation: Charac-terization and analysis of the carbonatation process, J Cult Herit 9 (2008) 294-301

[9] V. Daniele, G. Taglieri, Nanolime suspensions applied on natural lithotypes: The influ-ence of concentration and residual water content on carbonatation process and on treatment effectiveness, J Cult Herit 11 (2010) 102-106

[10] George Wheeler, Alkoxysilanes and the Consolidation of Stone, Getty Conservation In-stitute, Los Angeles, 2005

[11] C.J. Brinker and G.W. Scherer, Sol-Gel Science, Academic Press, New York, 1990, Ch. 3

[12] R. Zárraga, J. Cervantes, C. Salazar-Hernandez, G. Wheeler, Effect of the addition of hydroxyl-terminated polydimethylsiloxane to TEOS-based stone consolidants, J Cult Herit 11(2010) 138-144

[13] G. Wheeler, J. Mendez-Vivar, E.S. Goins, Fleming S.A., Brinker C.J., Evaluation of alkoxysilane coupling agents ain the consolidation of limestone, In: Proceedings of 9th Int. Con. on the Deterioration and Conservation of Stone, Elsevier, Amsterdam, 2000,541-545

[14] M. Favaro, R. Mendichi, F. Ossola, U. Russo, S. Simon, P. Tomasin, P.A. Vigato, Eval-uation of polymers for conservation treatments of outdoor exposed stone monuments. Part I: Photo-oxidative weathering, Polym Degrad Stabil 91 (2006) 3083-3096

[15] S. Vicini, S. Margutti, G. Moggi, E. Pedemonte, In situ copolymerisation of ethylmethacrylate and methylacrylate for the restoration of stone artefacts, J Cult Herit 2 (2001) 143-147

[16] F. Cappitelli, P. Principi, R. Pedrazzani, L, Toniolo, C, Sorlini, Bacterial and fungal de-terioration of the Milan Cathedral marble treated with protective synthetic resins, Sci Total Environ 385 (2007) 172-181

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

24

[17] M. Matteini, A. Moles, and S. Giovannoni, "Calcium oxalate as a protective mineral sys-tem for wall paintings: methodology and analyses", III Int. Symp. Conservation of Monu-ments in the Mediterranean Basin, ed. V. Fassina, H. Ott, F. Zezza, 1994

[18] B. Doherty, M. Pamplona, Costanza Miliani, M. Matteini, A. Sgamellotti, B. Brunetti, "Du-rability of the artificial calcium oxalate protective on two Florentine monuments", J. Cultural Heritage 8 (2007) 186-192

[19] S.V. Dorozhkin, Calcium orthophosphates in nature, biology and medicine, Materials 2 (2009) 399-498

[20] D.R. Lide (Ed.), CRC Handbook of Chemistry and Physics, The Chemical Rubber Co., 1999

[21] N. Harouiya, C. Chaïrat, S.J. Köhler, R. Gout, E.H. Oelkers, "The dissolution kinetics and apparent solubility of natural apatite in closed reactors at temperatures from 5 to 50 °C and pH from 1 to 6", Chemical Geology 244 (2007) 554–568

[22] D.G.A. Nelson, J.D.B. Featherstone, J.F. Duncan, and T.W. Cutress, "Effect of car-bonate and fluoride on the dissolution behavior of synthetic apatites", Caries Research 17 (1983) 200–211

[23]Y. Zhu, X. Zhanga, Y. Chena, Q. Xiea, J. Lana, M. Qiana and N He, "A comparative study on the dissolution and solubility of hydroxylapatite and fluorapatite at 25 °C and 45 °C", Chemical Geology 268 (2009) 89-96

[24] P.V. Brady, Ch. 4 in Physics and Chemistry of Mineral Surfaces, ed. P.V. Brady (CRC Press, Boca Raton, FL, 1996)

[25] G. Boivin, The hydroxyapatite crystal: a closer look, Medicographia 29 (2007) 126-132

[26] A.J. Skinner, J.P. LaFemina, H.J.F.Hansen, Structure and bonding of calcite: a theoreti-cal study, Am Mineral 79 (1994) 205-214

[27] M. Mathew, S. Takagi, Structures of biological minerals in dental research, J Res Natl Inst Stand Technol 106 (2001) 1035–1044

[28] E.N. Maslen, V.A.Streltsov, N.R. Streltsova, N. Ishizawa, Electron density and optical anisotropy in rhombohedral carbonates. III. Synchroton X-ray studies of CaCO3, MgCO3 and MnCO3, Acta Crystallogr B51 (1995) 929-939

[29] K. Polikreti, Y. Maniatis, "Micromorphology, composition and origin of the orange pat-ina on the marble surfaces of Propylaea (Acropolis, Athens)", The Science of the Total Envi-ronment 308 (2003) 111–119

[30] M. Alvarez de Buergo, R. Fort Gonzalez, "Protective patinas applied on stony facades of historical buildings in the past", Construction and Building Materials 17 (2003) 83–89

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

25

[31] C. Liu, Y. Huang, W. Shen, J. Cui, Kinetics of hydroxyapatite precipitation at pH 10 to 11, Biomaterials 22 (2001) 301-306

[32] A. Osaka, Miura Y., Takeuchi K., Asada M., Takahashi K., Calcium apatite prepared from calcium hydroxide and orthophosphoric acid, J Mater Sci: Mater Med 2 (1991) 51–55

[33] T. Brendel, A. Engel, C. Russel, Hydroxyapatite coatings by a polymeric route, J Mater Sci: Mater Med 3 (1992) 175-179

[34] M. Yoshimura, H. Suda, Okamoto K., Ioku K., Hydrothermal synthesis of biocompati-ble whiskers, J Mat Scie 29 (1994) 3399-3402

[35] A. Ślòsarczyk, E. Stobierska, Z. Paskiewicz, M. Gawlicki, Calcium Phosphate Materials Prepared from Precipitates with Various Calcium:Phosphorous Molar Ratios, J Am Ceram Soc 79 (1996) 2539-2544

[36] Christakis A. Paraskeva, Petros C. Charalambous, Lars-Erik Stokka, Pavlos G. Klepetsanis, Petros G. Koutsoukos, Peter Read, Terje Ostvold, and Alkiviades C. Payatakes, "Sandbed Consolidation with Mineral Precipitation", Journal of Colloid and Interface Science 232 (2000) 326–339

[37] Iosif T. Hafez, Christakis A. Paraskeva, Asimina Toliza, Pavlos G. Klepetsanis, Petros G. Koutsoukos, Øyvind Gustavsen, Terje Østvold, and Alkiviades C. Payatakes, "Calcium Phosphate Overgrowth on Silicate Sand", CRYSTAL GROWTH & DESIGN 6 [3] (2006) 675-683

[38] Rolf Snethlage; C.Gruber; V.Tucic; W.Wendler, "Transforming gypsum into calcium phosphate - a better way to preserve lime paint layers on natural stone?", Proc. Int. Symp. Stone consolidation in cultural heritage, eds. J. Delgado Rodrigues and J. Manuel Mimoso (Laboratório Nacional de Engenharia Civil, Lisbon, 2008) ISBN 978-972-49-2135-8

[39] M. Kamiya, J. Hatta, E. Shimida, Y. Ikuma, M. Yoshimura, H. Monma, AFM analysis of initial stage of reaction between calcite and phosphate, Mater Sci Eng B 111 (2004) 226-231

[40] M.S.A. Johnsson, G.H.Nancollas, The role of brushite and octacalcium phosphate in apatite formation, Crit Rev Oral Bio M 3 (1992) 61-82

[41] E.C. Moreno, M. Kresak, R.T. Zahradnik, "Fluoridated hydroxyapatite and caries for-mation", Nature 247 (January 4, 1974) 64-65

[42] A. Yasukawa, M. Kidokoro, K. Kandori, and T. Ishikawa, "Preparation and Characteri-zation of Barium–Strontium Hydroxyapatites", J. Colloid Interface Sci. 191, 407–415 (1997)

[43] J. Christoffersen, M.R. Christoffersen, N. Kolthoff, and O. Barenholdt, "Effects of Strontium Ions on Growth and Dissolution of Hydroxyapatite and on Bone Mineral Detec-tion", Bone 20 [1] (1997) 47-54

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

26

[44] T. Perry, N. M. Smith, W. J. Wayne, Salem limestone and associated formations in south-central Indiana, Indiana Department of Conservation, Geological Survey, Field Con-ference Guidebook 7, 1954

[45] Robert Flatt, ETH, 2003, private communication

[46] S. Siegesmund, K. Ullemeyer, T. Weiss, E.K. Tschegg, Physical weathering of marbles caused by anisotropic thermal expansion, Int J Earth Sci 89 (2000) 170-182

[47] T.Weiss, S. Siegesmund, E. R. Fuller Jr., Thermal degradation of marble: indications from finite-element modelling, Build Environ 38 (2003) 1251-1260

[48] T. Esaki, K. Jiang, Comprehensive study of the weathered condition of welded tuff from a historic bridge in Kagoshima, Japan, Eng Geol 55 (1999) 121-130

[49] C. Miliani, M.L. Velo-Simpson, G.W. Scherer, Particle-modified consolidants: A study on the effect of particles on sol–gel properties and consolidation effectiveness, J Cult Herit 8 (2007) 1-6

[50] R. Benjumea, D.L. Sikarkie, A note on the penetration of a rigid wedge into a nonisotropic brittle material, Int J Rock Mech Min Sci 6 (1969) 343-352

[51] A. Duguid, M. Radonjic, and G.W. Scherer, “Degradation at the Reservoir/Cement In-terface from Exposure to Carbonated Brine”, submitted to Int. J. Greenhouse Gas Control

[52] A. Kasioptas, C. Perdikouri, C.V. Putnis, and A. Putnis, "Pseudomorphic replacement of single calcium carbonate crystals by polycrystalline apatite", Mineralogical Magazine, 72 [1] (2008) 77–80

[53] Sonia Naidu, Enrico Sassoni, George W. Scherer, New Treatment for Corrosion-Resistant Coatings for Marble and Consolidation of Limestone, submitted to proceedings of Jardins de Pierres, Conservation of stone in Parks, Gardens and Cemeteries, Paris, 22-24 June 2011

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

Table 1. Dynamic elastic modulus and tensile strength of IL samples unheated and heated to 300°C for 1, 4 and 16 hours (values are averages for 3 samples)

Dynamic elastic modulus (Ed) Tensile strength (σt)

Samples Ed [GPa] Ed decrease [%] σt [MPa] σt decrease [%]

unheated 37.3 (±0.3) - 4.8 (±0.5) -

heated to 300°C for 1 hour

21.1 (±0.0) 43.4 3.5 (±0.1) 27.1

unheated 33.9 (±1.5) - - -

heated to 300°C for 4 hours

20.0 (±0.7) 40.9 - -

unheated 35.0 (±1.5) - - -

heated to 300°C for 16 hours

19.8 (±0.6) 43.4 - -

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

1

Table 2. Dynamic elastic modulus of IL cubic samples before and after treatment with 1.0 M DAP for 1, 2, 4 and 8 days (values are averages for 2 samples); dynamic elastic modulus and tensile strength of IL cylindrical samples before treatment and af-ter 2 days and 4 days treatment with 1.0 M DAP and after 2 days treatment with 1.0 M DAP and then 5 wetting/drying cycles (values are averages for 10 samples)

Dynamic elastic modulus [GPa] Tensile strength [MPa]

Treatment on cubic samples

Before treatment

After treatment

Increase [%]

Untreated Treated Increase [%]

1.0 M DAP for 1 day 23.1 (±0.8) 35.6 (±0.9) 54.5 - - -

1.0 M DAP for 2 days 22.3 (±0.4) 35.8 (±0.8) 60.7 - - -

1.0 M DAP for 4 days 22.6 (±0.2) 37.2 (±0.3) 64.4 - - -

1.0 M DAP for 8 days 21.3 (±1.1) 35.4 (±0.5) 66.8 - - -

Treatment on cylindrical samples

Before treatment

After treatment

Increase [%]

Untreated Treated Increase [%]

1.0 M DAP for 2 days 18.6 (±0.8) 36.5 (±1.2) 96.0 3.3 (±0.4) 4.1 (±0.7) 24.5

1.0 M DAP for 4 days 18.8 (±1.0) 37.0 (±1.4) 96.8 3.3 (±0.4) 4.2 (±0.4) 28.8

1.0 M DAP for 2 days + + 5 w/d cycles

18.8 (±0.8) 33.9 (±1.0) 80.7 3.3 (±0.4) 3.8 (±0.2) 16.8

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

2

Figure 1. Dynamic elastic modulus of IL cubic samples versus heating temperature (values are averages for 2 samples)

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

3

Figure 2. Dynamic elastic modulus of IL cubic samples treated with DAP solutions at vari-ous concentrations versus duration of treatment; open symbols represent the Ed of dry sam-ples (before and after the treatment) while solid symbols represent the Ed of water- and solu-tion-saturated samples (during the treatment); the final variation ΔEd,dry was calculated comparing the Ed values in the dry condition before and after treatment (all values are aver-ages for 2 samples)

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

4

Figure 3. SEM images of fracture surfaces of IL samples treated with 1.0 M DAP for 2 days; the arrows indicate the points where EDS analysis was performed (EDS results are reported in Figure 4.

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

5

Figure 4. EDS spectra of calcium phosphate phases observed in Figure 3; the EDS spectra, (a) and (b) respectively, were taken in the points indicated by arrows in Figure 3.

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

6

Figure 5. Unit cell, chemical formula, SEM image of the analyzed point and EBSD results for the four phosphate phases identified: (a) HAP; (b) calcium-deficient hydroxyapatite (CDHAP); (c) octacalcium phosphate monohydrate (OCPM); (d) calcium dihydrogen phos-phate monohydrate (MCPM).

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

7

Figure 6. Dynamic elastic modulus of IL core-drilled samples before treating (open square), after treating with 1.0 M DAP for 2 days and after w/d cycles (solid squares) (values are av-erages for 10 samples).

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

8

Figure 7. Pore size distribution of IL samples untreated (grey lines) and treated (black lines) with 1.0 M DAP for 2 days (ϕ = open porosity [%]; r > 1 μm = % of pores having radius > 1μm; r < 0.1μm = % of pores having radius < 0.1μm).

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

9

Figure 8. Sorptivity of IL samples untreated (gray line) and treated with 1.0 M DAP for 2 days (black line).

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

10

Figure 9. Dynamic elastic modulus of IL cubic samples unheated (open bars), heated (gray bars) and treated with 1.0 M DAP for 2 days (black bars) at different heating temperatures (values are averages for 2 samples).

Sassoni E., Naidu S., Scherer G.W., The use of hydroxyapatite as a new inorganic consolidant for damaged carbonate stones, Journal of Cultural Heritage 12 (2011) 346-355, doi:10.1016/j.culher.2011.02.005

11

Figure 10. Dynamic elastic modulus increase of IL cubic samples heated and then treated with 1.0 M DAP for 2 days versus heating temperature (values are averages for 2 samples).