New calcium alkoxides for consolidation of carbonate rocks. Influence of precursors' characteristics on morphology, crystalline phase and consolidation effects

Article abstract of DOI:10.1039/c2nj40708f

New ambient temperature liquid calcium alkoxides, [Ca(O(CH ₂CH₂O)₃CH₃)₂] (1) and [Ca(O(CH₂CH₂O)₃CH₂CH ₃)₂] (2), together with [Ca(OCH₂C(CH ₃)₃)₂] (3), were synthesised by reaction of ammonia-activated calcium with the appropriate alcohol. Their potentiality as stone consolidant products was investigated and compared with those of other alkoxides: [Ca(OCH₂CH₃)₂(CH₃CH ₂OH)₄] (4), [Ca(OCH₃)₂] (5), [{Ca(OCH₂CH₂OCH₃)₂}₉] (6), already described in the literature. Reaction of 1-3 with the atmosphere was studied, final products analysed and kinetic pathways investigated. The reaction produces CaCO₃ and the vaterite/calcite ratios observed in the coatings generated from isopropyl alcohol solutions of 1-6 were found to considerably vary with the alkoxide precursor, which has a strong influence also on the morphology of the produced films. Furthermore, their efficiency as stone consolidants was tested by ultrasound measurements. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012.

Full text of DOI:10.1039/c2nj40708f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2012, 36, 2618–2624
www.rsc.org/njc
PAPER
New calcium alkoxides for consolidation of carbonate rocks. Influence of
precursors’ characteristics on morphology, crystalline phase and
consolidation effects
Franco Ossola, Patrizia Tomasin,* Chiara De Zorzi, Naida El Habra,
Matteo Chiurato and Monica Favaro
Received (in Montpellier, France) 9th August 2012, Accepted 8th October 2012
DOI: 10.1039/c2nj40708f
New ambient temperature liquid calcium alkoxides, [Ca(O(CH2CH2O)3CH3)2] (1) and
[Ca(O(CH2CH2O)3CH2CH3)2] (2), together with [Ca(OCH2C(CH3)3)2] (3), were synthesised by
reaction of ammonia-activated calcium with the appropriate alcohol. Their potentiality as stone
consolidant products was investigated and compared with those of other alkoxides:
[Ca(OCH2CH3)2(CH3CH2OH)4] (4), [Ca(OCH3)2] (5), [{Ca(OCH2CH2OCH3)2}9] (6), already
described in the literature. Reaction of 1–3 with the atmosphere was studied, final products
analysed and kinetic pathways investigated. The reaction produces CaCO3 and the vaterite/calcite
ratios observed in the coatings generated from isopropyl alcohol solutions of 1–6 were found to
considerably vary with the alkoxide precursor, which has a strong influence also on the
morphology of the produced films. Furthermore, their efficiency as stone consolidants was tested
by ultrasound measurements.
due to their tendency to oligomerise,4–6 which minimize their
Introduction
utilization also in other fields of application, such as sol–gel and
chemical vapour deposition (CVD).7,8
Recently, we reported on the potential use of calcium alkoxides
as an alternative class of compounds for consolidation of
With the aim to increase solubility, starting alcohols with
carbonate rocks, most of them used for historical artifacts
increased steric hindrance (2,2-dimethyl-1-propanol) and
and buildings.1 We evaluated the possibility to use, as a
additional coordinating atoms (2-methoxyethanol, two
consolidating agent, the calcium carbonate produced from
oligoethers) were chosen for the synthesis of calcium alkoxides.
hydrolysis–carbonation of calcium alkoxides, in a similar way
Here we describe the new complexes synthesised with the
ammonia method,9 [Ca(O(CH2CH2O)3CH3)2] (1) and
as carbonation of calcium hydroxide leads to hardening of
deteriorated limestone, lime mortars and wall paintings; in fact,
[Ca(O(CH2CH2O)3CH2CH3)2] (2), the first calcium alkoxides
in liquid form. [Ca(OCH2C(CH3)3)2] (3), already known,6 was
the traditional lime method, whose main problem is the very
low solubility of Ca(OH)2, has recently been reconsidered with
newly synthesised with the same method.
the innovative use of nanolime dispersions.2,3 The calcium
Once fully characterized, they were applied on glass frits and
alkoxide, dissolved in an organic solvent, penetrates within
slides and the influence of precursors’ characteristics on the
the stone pore network where it reacts with moisture and
reaction with the atmosphere, on the morphology and phase
carbon dioxide of the atmosphere generating in situ calcium
of the resulting calcium carbonate coatings is discussed.
carbonate, which fills the micro-fissures and binds the stone
Furthermore, ultrasonic velocity measurements were used
fragments together. Up to now, their very low solubility has
to evaluate the consolidation efficacy after their application on
hampered their successful application as alternative consolida-
model samples of Carrara marble, metamorphic carbonate rock.
In fact, the velocity of the compressive wave (m s–1) is one of the
tion products to available commercial products, because the
low amount of CaCO3 deposited in each application would
non-destructive tests more widely used to assess strengthening
consequently determine too many disadvantageous repetitions
effects of stone treatments; the velocity of the ultrasonic wave
of the treatment. On the other hand, insolubility, together with
mainly depends on the porosity and the E-modulus of the
low volatility, are general characteristics of group 2 metal
material. Since consolidants are supposed to change stone
alkoxides with conventional monodentate alkoxide ligands,
porosity and mechanical properties, ultrasonic velocity is usually
considered a good diagnostic tool to trace these changes.10,11
Their decomposition behaviour and stone consolidation
Istituto di Chimica Inorganica e delle Superfici, C.N.R., Corso Stati
Uniti 4, 35127 Padova, Italy. E-mail: patrizia.tomasin@cnr.it
efficiency, after their application on marble specimens, were
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
2618 New J. Chem., 2012, 36, 2618–2624

View Article Online
compared with those of other calcium alkoxides:
[Ca(OCH2CH3)2(CH3CH2OH)4] (4), [Ca(OCH3)2] (5), [{Ca
(OCH2CH2OCH3)2}9] (6).12
heteronuclear bidimensional spectrum HMQC, the attribution
of the 13C resonances was also possible (see the Experimental
section).
In the 1H spectrum of 1 (C6D6), peaks were assigned on the
basis of chemical shift, integral values and by comparison with
the spectra of 2. The methoxy protons singlet at 3.20 ppm (6H)
Results and discussion
and the 3.44 ppm broad triplet of CH2OCH3 (4H) match well
À
Synthesis of calcium alkoxides
with the corresponding signals found for the analogous
barium complex [Ba[O(CH2CH2O)3CH3]2],13 whereas the
unassignable multiplet between 3.68 and 4.10 ppm and the
Synthesis of calcium alkoxides 4–6 involved direct reaction of
metallic calcium with the proper alcohol, whereas compounds
1–3 were prepared by using ammonia-activated calcium9 in
high yield and purity. It has to be noted that compound 3 was
newly synthesised with this method.
broad singlets at 3.61 (CH2O, 4H) and 4.37 ppm
À
(CH2CH2OCa, 4H) present a slight shift to higher frequencies
À
with respect to the barium congener. On the basis of these
NMR analogies, a monomeric structure, similar to that found
for the Ba complex, can be proposed for 1, the downfield shift
probably due to the shorter ionic radius of Ca with respect to
Ba, determining a slightly different rearrangement around
the metal.
In the synthesis of 1 and 2 the corresponding oligoether
alcohols were added to elemental calcium in slight defect with
respect to the stoichiometric amount, in order to avoid their
possible coordination (as unreacted reagents) to the calcium
metallic centre and unreacted calcium was filtered at the end of
the reactions. Differently from the features usually presented
by calcium alkoxides, 1 and 2 showed a peculiar room
temperature dense liquid state and were highly soluble in
organic solvents.
Calcium alkoxides reactions with the atmosphere: CaCO3
deposition
Reactions of complexes 1–6 with the atmosphere have been
studied, final products analysed and kinetic pathways
investigated.
The proposed formulas are made on the basis of elemental
analysis and NMR and FT-IR spectra. Since the main goal of
this paper was to verify the behaviour in marble consolidation,
no in depth structural analyses were carried out.
A preliminary evaluation of stability of alkoxides in i-PrOH
was run, resulting in a good stability and excluding alcohol
interchange reactions.
1
The interpretation of the H and 13C NMR spectra of 2 in
toluene-d8 was not straightforward; bidimensional experiments
2D COSY and HMQC were carried out to attempt a correct
attribution of the signals, most of which were overlapping.
It was possible to unambiguously identify the triplet related
Exposure of isopropyl alcohol solutions to the atmosphere
gave rise to spontaneous reactions of the complexes with air
and deposition of white coatings on the surfaces of the vessels.
Reflectance IR measurements have been performed on a
gold surface after deposition of a drop of alcoholic solution of
the alkoxides. IR spectra were collected at different times until
complete conversion of compounds to CaCO3. As an example,
spectra of compound 1 collected at different times are reported
in Fig. 1. In addition to the characteristic signals of the metal
alkoxides, an absorption at 1640–1655 cmÀ1 has been readily
observed for all complexes, corresponding to the CH3OCO2
to OCH2CH3 at 1.14 ppm (6H), the quartet related to
À
OCH2CH3 at 3.37 ppm (4H), the triplet assigned to
À
CH2OCH2CH3 at 3.49 ppm (4H) and the two broad signals
À
of the CH2CH2OCa protons at 3.77 (s, 4H) and 4.20 ppm (s,
À
À
4H). The assignment of the remaining peaks at 3.60 (sb, 4H),
3.69 (sb, 4H) and 3.81 ppm (sb, 4H), due to the other CH2O
protons, is less certain. Through the correlations in the
Fig. 1 FT-IR spectra of the products resulting from the reaction of 1 with the atmosphere. Time of collection: (a) 0; (b) 5 min; (c) 30 min; (d) 6 h;
(e) 4 days (f) six months.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 2618–2624 2619

View Article Online
group formed by insertion of CO2 in the Ca–O bond of the
calcium alkoxides and producing the corresponding calcium
alkyl carbonates.1,14 This insertion process is immediate for
compounds 4 and 5, giving rise to a strong band centred at
1653 cmÀ1(1634 cmÀ1 for 5), while it is much slower for the
other complexes. The different kinetic stability is probably due
to the reduced accessibility of the more sterically hindered
central calcium atom in 1, 2, and 3 with respect to 4 and 5.
Alkyl carbonate absorptions decrease until disappearance in
about 6 hours, while new signals at 1390–1450 and 873 cmÀ1
,
characteristic of CaCO3, appear and rise. These absorption
patterns indicate that reactions of 1–4 with the atmosphere
take place mainly through CO2 insertion into the Ca–O bond
of the calcium alkoxides with formation of alkylcarbonate
derivatives, subsequently transformed into CaCO3. Nevertheless,
a second kinetic pathway is suggested for 1, 4 and 6, as already
described for 5.1 In fact, the appearance of a sharp absorption at
3645 cmÀ1, typical of OH stretching, suggests that an hydrolysis
process occurs with formation of Ca(OH)2, which is successively
carbonated to CaCO3 through CO2 insertion into the Ca–O
bond. This process generates variously coordinated bicarbonate
groups, as evidenced by signals at 1598, 1475–1468 and
1395–1400 cmÀ1, overlapping CaCO3 absorption bands. The
hydrolysis process is competitive with one leading to insertion
of CO2 and it was observed in some compounds rather than in
others independently from their structure. It appears to be
influenced mostly by the humidity content in the atmosphere.
Morphological observations were performed on the calcium
carbonate films produced from reaction with the atmosphere
of complexes 1–6. With a view to a possible utilization as
stone consolidant products, porous glassy frits, simulating the
deteriorated stone, were impregnated with isopropyl alcohol
solutions of complexes 1–4, 6 as described in the Experimental
section and observations were carried out on the coatings
formed inside the frits. Coatings were homogeneously
deposited with good adhesion on the grains surface of the frit
inside its pore network. They appear to be built up by
aggregation of particles with nanoscale dimensions in one or
more directions. As shown in Fig. 2 while 1 and 5 (deposited
from solutions/dispersions with different solvents)1 produce
homogeneous coatings of round and platelet-like crystallites,
2 and 6 exhibit, respectively, isolated clusters of crystallization
grains and finely dispersed particles; the coating formed after
carbonation of 3 exhibits a fine-grained surface morphology,
while the film generated from 4 consists of an agglomeration
of spherical particles.
Fig. 2 Secondary electron images collected from (a) the untreated
surface and from the coatings obtained from isopropyl alcohol solu-
tions of (b) 1, (c) 2, (d) 3, (e) 4, (f) 5 and (g) 6. Magnification: 40.000Â
(a, b, c, e, f), 80.000Â (d, g).
Fig. 3 XRD patterns of the coatings obtained from isopropyl alcohol
solutions of complexes 1–4 (with p = portlandite, v = vaterite, c =
calcite).
01-084-1276) as the minority phase, as indicated by the
main Ca(OH)2 reflection at 2y = 17.81 corresponding to the
(001) Bragg reflection. It is noteworthy that the deposit
obtained from 2 is composed of vaterite as the only crystalline
phase. A semi quantitative estimation (RIR method) of the
different phases, present in the analysed samples, is reported in
Table 1, which clearly shows vaterite is the main carbonatic
phase. The vaterite prevalence among other phases can be
ascribed to the employed solvents: alcohols like ethanol,
isopropanol and diethylene glycol actually prevent the
transformation of vaterite aggregates to calcite by influencing
the surface charge of vaterite crystals.15,16 As previously
reported,1 depending on the solvent/dispersion agent and on
Microstructure of these coatings has been investigated
by XRD and the obtained patterns are reported in Fig. 3.
All the deposits from 1–5 were polycrystalline without
preferential orientation, as shown by the analysis relative
intensities of reflection, even if amorphous phases cannot be
excluded.
The deposits obtained from 1, 3 and 4 clearly showed the
co-presence of carbonatic phases (CaCO3), such as vaterite
with hexagonal structure (ICDD: 01-072-0506) and calcite
with rhombohedral structure (ICDD: 01-086-2341). Never-
theless, the XRD pattern of the deposit obtained from 4
highlighted also the presence of portlandite (i.e. calcium
hydroxide Ca(OH)2) with hexagonal structure (ICDD:
c
2620 New J. Chem., 2012, 36, 2618–2624
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
Table 1 Semiquantitative estimation (RIR method) of the phases
vaterite (hexagonal), calcite (rhombohedral) and portlandite (hexago-
nal) identified in the deposits analysed by XRD
high boiling point alcohols) could remain trapped in the
CaCO3 coatings.
Evaluation of Ca alkoxides as stone consolidants
Semiquantitative
(%)
Calcium alkoxide
Phase
Products applied as consolidants on disaggregated stone
should improve substrate intergranular cohesion, thus reducing
the porosity and restoring mechanical properties. Ultrasound
velocity measurements, related to these parameters, were
performed to evaluate consolidation efficacy.
[Ca(O(CH2CH2O)3CH3)2] (1)
Vaterite
Calcite
Vaterite
Vaterite
Calcite
Vaterite
Calcite
Portlandite
86
14
100
66
34
88
9
[Ca(O(CH2CH2O)3CH2CH3)2] (2)
[Ca(OCH2C(CH3)3)2] (3)
[Ca(OCH2CH3)2(CH3CH2OH)4]
(4)
As described in the Experimental section, marble specimens
were treated three times for two hours with alkoxides 1–6 in
isopropyl alcohol (Fig. 4). It was observed that during each
imbibition, the clear solutions suddenly turned cloudy for
alkoxides (3, 5) and a fine precipitate deposited on the bottom
of the vessel. This behaviour was not evidenced in 1, 2, 4 and 6,
whose solutions remained stable at least for 3–4 hours of
exposure. The identity of the precipitate (CaCO3) was
confirmed by IR measurements.
3
the substrate, the vaterite/calcite ratio in the deposit of 5 can
vary considerably. Compound 6 does not follow a defined
decomposition pathway; different preparation and samples of
the compound, after atmospheric exposure, gave different
crystallographic results, ranging from amorphous deposits to
crystalline systems with vaterite only, vaterite–calcite mixtures,
vaterite–portlandite mixtures, without great reproducibility.
Studies are currently in progress to determine the parameters
influencing the final product.
Three weeks after the treatment, the weight and ultrasonic
velocity changes were determined (Table 2). For 1, 2, 4 and 6
weight increases have been observed; while for Ca(OH)2, 3, 5
the increase is negligible. This can be related to the different
solubility and reactivity with the atmosphere, depending on
the different coordinative saturation of the calcium centre.
This explains the formation of CaCO3 precipitates for 3 and 5
and, above all, the observations of ultrasonic velocity variation
after treatment: Ca(OH)2, compounds 3 and 5 increased the
velocity by 5.5, 5.7 and 3.7%, respectively, in comparison with
the untreated stone, while compounds, 1, 2, 4, 6 induced a
positive variation that range from 9.7 to 28%.
In fact, the decomposition processes, carried out under
environmental conditions, might be influenced by the different
thermo-hygrometric conditions thus leading to the presence of
portlandite and the conversion of vaterite to calcite. This could
confirm what was already observed for 5, where the crystalline
phases of the final CaCO3 deposit varied in accordance with
different deposition conditions.1
Elemental analyses were also carried out on these coatings.
The results were: film obtained from 1 C 13.87, H 0.00%; from
2 C 13.08, H 0.15%; from 3 C 13.50, H 0.00 %; from 4 C
11.77, H 0.21%; calcd for CaCO3: C 12.00, H 0.00%. Results
are essentially in agreement with CaCO3 formulation and with
the XRD measurements. Higher percentage values found for
C in 1–3 suggest that small organic fractions (most likely the
According to the obtained results, it is confirmed that the
employment of more sterically hindered and/or electron donor
atom containing ligands produces alkoxides with a higher
solubility and stability. This allows a slower decomposition
of the product, so a longer staying inside the stone pore
network where the formation of CaCO3 can take place. This
Fig. 4 Images collected from Carrara marble samples sized 5 Â 5 Â 1 cm3: (a) untreated, un-aged, (b) untreated aged, (c) aged and treated with 5,
(d) aged and treated with 4, (e) aged and treated with 1, (f) aged and treated with 2. The reddish-orange spots on (d) and (f) are due to oxidation of
impurities of pyrite contained in the marble.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 2618–2624 2621

View Article Online
Table 2 Herein are reported the products applied on marble specimens, their molecular weights, the weight changes induced by the application,
the ultrasonic velocities measured before (bt) and after treatment (at) and the corresponding calculated percentage variation
Code
Applied product
Weight increase (mg)
US v-bt (m s–1
)
US v-at (m s–1
)
US v variation (%)
Untreated marble
Ca(OH)2
[Ca(MeO)2]
—
1307 190
1226 57
1191 10
975 68
1003 11
958 12
933 28
886 42
1308 87
1294 141
1235 151
1031 29
1095 35
1061 31
1070 0.0
1134 36
0.1
5.5
3.7
5.7
0.6 0.2
0.4 0.2
0.6 0.2
10.7 0.7
18.3 0.9
25.7 3.4
12.9 1.3
5
3
4
6
1
2
[Ca(NpO)2]
[Ca(EtO)2Á4EtOH]
9.3
[Ca9(MeOEtO)18
]
10.8
14.7
28.0
[Ca(TEGMME)2]
[Ca(TEGMEE)2
determines a better consolidating efficacy as evidenced by the
increase in US measurements.
A Nicolet microscope connected to a Nicolet 560 FT-IR
system, equipped with a MCT (Mercury Cadmium Telluride)
detector, has been used for spectra collection of calcium
alkoxides and related conversion products. Investigated micro
areas were about 50 Â 50 mm2 of size. IR spectra were recorded
in reflectance mode in the 4000–650 cmÀ1 range, with a
resolution of 4 cmÀ1. Recorded spectra have been expressed
by absorbance units and baseline corrected.
Conclusions
With the aim of finding new products for the re-aggregation of
carbonate stone, new calcium alkoxides were synthesised and
their properties compared with already known homologues.
It has to be noted that the first liquid calcium alkoxides were
synthesised and they were found to be very soluble in most
organic solvents.
The microstructure characterization was carried out by
X-ray diffraction (XRD) measurements by using a Philips
PW 3020 powder diffractometer with a Bragg–Brentano
y–2y geometry. The radiation used was the Cu Ka
(l=1.54056 A, 40 kV, 30 mA). All the patterns were collected
in the 151–701 2y range and peak positions were determined
with a statistical error d(2y) = 0.021. Phase identification was
performed with the support of standard patterns reported in
2002 ICDD database. The semi-quantitative estimation of the
different phases present in the single samples has been
obtained by the Reference Intensity Ratio (RIR) method.18
Observations of the inorganic coatings deposited on glassy
substrates have been performed by using a Fei Quanta 200
FEG-ESEM instrument to evaluate their morphology and the
distribution inside the pore network. The semi-quantitative
elemental compositions were obtained using an Energy Dispersive
X-ray Spectrometer EDAX Genesys, using an accelerating
voltage of 20 keV. The samples were analysed uncoated in low
vacuum mode.
Their decomposition pathways and the final formed CaCO3
coating were evaluated respectively by FT-IR, and by FEG-ESEM
and XRD; the complexes turn out to be possible candidates
for the consolidation purpose. The compounds are stable in
organic solvents and give a low interchange reaction with
i-PrOH. For this reason calcium alkoxides seem to be promising
as consolidating agents and this is confirmed by the good results
in US variation for the most soluble of them. The consolidation
properties seem to be related to solubility and stability of the
alkoxides in solution, these depending on the properties of the
used ligand.
Further studies are needed to ascertain the effects of the
length of the chain and of substituents or heteroatoms in the
chain on the oligomerisation degree and on the solubility of
the compounds. A better evaluation of the stability in different
solvents will be essential for storage and application of the
compounds in conservation.
Weight difference was determined using a SCALTEC
SBC22 balance (0.00001 g sensitivity).
Moreover, studies will be carried out to elucidate the
influence of the thermo-hygrometric parameters on the kinetic
pathway of alkoxide decomposition and on the final structure
of the coating.
The ultrasonic portable instrument employed for the
measurements is a USG 20 (Geotron Elektronik, Germany),
operated with a point-shaped 46 kHz-vibrator (UPG-T) and
receiver (UPE-T).
All the previous aspects are currently under investigation
within the European Project NANOMATCH.17
Calcium alkoxides syntheses
Experimental
Synthesis and manipulation of calcium alkoxides were carried
out in nitrogen filled glove-boxes with exclusion of moisture
and oxygen. Solvents were purified by standard procedures;19
before use, calcium granules (Aldrich Chemical Company, 99%)
were cleaned in dry ethyl ether by vigorous stirring, which allows
the removal of the oxide patina; other reagents were commercial
products used as supplied.
General instrumentation
NMR spectra were recorded on a Bruker AMX300 spectrometer,
equipped with a 5 mm broad-band multinuclear probe, operating
at a frequency of 300.13 MHz for 1H and 75.43 MHz for 13C. The
samples were dissolved in the appropriate deuterated solvents that
were also used as internal references.
[Ca(OCH2CH3)2(CH3CH2OH)4] (4), [Ca(OCH3)2] (5) and
[{Ca (OCH2CH2OCH3)2}9] (6) were synthesised according to
procedures described in the literature,1,20 their identity confirmed
by comparison with literature data and fully supported by
elemental analysis results.
Elemental analyses were performed with a Fisons EA 1108
(CHNS-O version) elemental analyser. The samples were
prepared by scratching the coatings from glass slides and
vessels.
c
2622 New J. Chem., 2012, 36, 2618–2624
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
Syntheses of [Ca(O(CH2CH2O)3CH3)2] (1), [Ca(O(CH2CH2O)3-
CH2CH3)2] (2) and [Ca(OCH2C(CH3)3)2] (3). Calcium granules
(10 mmol) and the suitable alcohol (triethylene glycol monomethyl
ether for (1), triethylene glycol monoethyl ether for (2) and 2,
2-dimethyl-1-propanol for (3)) (1.95 mmol) were added to toluene
(30 mL). The mixture was refrigerated just above the solvent
melting point and maintained at this temperature under stirring.
NH3 was bubbled into the mixture at intervals. Reaction took
place with formation of tiny bubbles evolution on the calcium
surface, which turned golden yellow, until total consumption of
calcium granules (48 h). The mixture was filtered and the solvent
evaporated under vacuum to give:
with MeOH and EtOH and the stability to i-PrOH alcoholysis.
Whereas the spectrum of the i-propyl solution recorded after
10 days, still evidence the sole presence of the alkoxide, only
the starting alcohol can be detected in the spectra of methanol
and ethanol solutions.
No systematic investigation of the stability of alkoxides was
performed: a detailed study is currently carried out within the
European Project Nanomatch.17
CaCO3 deposition from Ca alkoxides dispersed in alcohols
Suitable quantities of compounds 1–4, 6 (corresponding
to 2.5 g LÀ1 of Ca) were added to isopropyl alcohol (i-PrOH)
(5 was dissolved in methanol1) and the resulting solutions were
placed in glassy vessels and exposed to the atmosphere for
20 days. The white coatings produced on the glass surface
from reaction of the complexes with air were structurally
analysed by XRD.
(1) An amber-coloured viscous oil which was analysed and
identified as [Ca(O(CH2CH2O)3CH3)2] (found: C 45.98, H
8.09%; calcd. for C14H30O8Ca: C 45.90, H 8.20%; 84% yield).
1H NMR: C6D6, d, ppm: 3.20 (s, 6H, OCH3), 3.44 (tb, 4H,
À
CH2OCH3), 3.61 (sb, 4H, CH2O), 3.68–4.10 (mb, 12H,
À
À
CH2O), 4.37 (sb, 4H, CH2CH2OCa).
À
À
IR (cmÀ1): 2924 m, 2873 m (sh), 2815 m, 1454 w, 1351 w,
1301 w, 1247 w, 1200 w, 1107 s, 1030 m, 935 w, 884 w, 852 w,
732 w, 696 w.
Methanol and water dispersion of 5 were also left in contact
with air for 14 days, then the solid decomposition products
dried, filtered and analysed.1
(2) An amber-coloured viscous oil which was analysed and
identified as [Ca(O(CH2CH2O)3CH2CH3)2] (found: C 49.69,
H 8.77% ; calcd. for C16H34O8Ca: C 48.73, H 8.63; 88% yield).
Glass frits generally used for chemical filtration (Bibby
Scientific, diameter 30 mm, thickness 3.5 mm, porosity 4,
experimentally calculated pores diameter 5–15 mm) were
chosen as porous substrates, simulating the disaggregated
carbonate rocks, in order to study the behaviour of the calcium
alkoxides in their transformation to calcium carbonate inside
pores and cavities. This kind of porous substrate assures a
defined range of pore sizes, allowing a reliable comparison of
different series of experiments, difficult to perform in naturally
occurring stone substrates due to their great variability in pore
dimensions at the microscale. The frits were impregnated by
same solutions of complexes 1–6, through imbibition: they were
left in contact with the alkoxide solutions, which were drawn
into the frits by capillarity, by three different applications
(two hours each), performed at two days intervals. FEG-ESEM
observation of the CaCO3 films deposited onto the frits was
carried out.
1H NMR: d8-toluene, d ppm: 1.14 (tr, 6H, OCH2CH3), 3.37
À
(q, 4H, OCH2CH3), 3.49 (t, 4H, CH2OCH2CH3), 3.60 (sb, 4H,
À
À
CH2O), 3.69 (sb, 4H, CH2O), 3.77 (sb, 4H, CH2CH2OCa) 3.81
À
À
À
(sb, 4H, CH2O), 4.20 (sb, 4H, CH2CH2OCa).
13C NMR d8-toluene, d ppm: 15.6 ppm (OCH2CH3), 63.5
(CH2CH2OCa), 66.7 (OCH2CH3), 70.5 (CH2OCH2CH3), 70.7
À
À
À
À
À
À
(CH2O), 71.1 (CH2O), 76.8 (CH2CH2OCa).
IR (cmÀ1): 2974 m, 2927 m(sh), 2870 s, 1486 w, 1455 w,
À
À
À
1397 w, 1379 w, 1349 w, 1305 w, 1248 w, 1108 s, 1073 m(sh),
945 w, 885 w, 844 w, 700 w, 668 w.
(3) A white powder which was analysed and identified as
Ca(OCH2CMe3)2 (found: C 55.72, H 10.39% ; calcd. for
C10H22O2Ca: C 56.07, H 10.28; 81% yield).
1H NMR: pyr-d5, d ppm: 1.02 (s, 9H, (CH3)3C), 3.48 (s, 2H,
À
CH2O).
Consolidation and impregnation studies of 5 have been
previously reported.1
À
IR (cmÀ1): 2972 s, 2861 m, 2803 m, 1477 w, 1457 w, 1393 w,
1356 w, 1095 s, 1020 m, 928 w, 746 w, 718 w, 696 w.
Evaluation of consolidation effect
Preliminary studies of solution stability and alcoholic exchange
Ultrasonic measurements were carried out on artificially aged
Carrara marble specimens (dimension 5 Â 5 Â 1 cm) before
and after the application of alkoxides solution by imbibition.
The samples were impregnated by contact imbibition with
isopropyl alcohol solutions of complexes 1–6, prepared with a
compound quantity equivalent to 6.25 g LÀ1 Ca(OH)2 (2.5 g
LÀ1 of Ca), which was used by Baglioni et al. for Ca(OH)2
micro and nanoparticles dispersions in i-PrOH and tested for
consolidation of carbonate stones.21,22 To compare the
efficiency of the Ca alkoxides, a dispersion of Ca(OH)2 in
i-PrOH (6.25 g LÀ1) was applied and used as reference
consolidant product.
We already reported the possible interchange reaction for
calcium alkoxides in alcohols,1 with a complete formation of
methoxide after 24 h of stirring the ethoxide in methanol. For
this reason, (1), (2), (6) were dissolved in methyl alcohol
(MeOH), ethyl alcohol (EtOH) and isopropyl alcohol (i-PrOH)
to evaluate the possible exchange reaction. All solvents were
freshly distilled and the solution was maintained under an inert
(N2) atmosphere, to prevent hydrolytic processes.
In all cases the reaction with MeOH results in a suspension
with the precipitation of [Ca(MeO)2], the solution with EtOH
remains clear for 2 days before the formation of a white
precipitate [Ca(EtO)2], while the solution in i-PrOH remains
clear for at least 10 days.
Marble specimens set horizontally were soaked 3 times for
2 hours in alkoxides solutions, with the front side dipped by
2 mm. The amount of the deposited compound after application
was determined by weight difference (Fig. 4).
1H NMR spectra in benzene-d6 solution of the residue after
the evaporation of the alcohol confirm the exchange reaction
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 2618–2624 2623

View Article Online
7 D. C. Bradley, Chem. Rev., 1989, 89, 1317.
8 U. Schubert and N. Husing, Synthesis of Inorganic Materials,
Wiley-VCH, Weinheim, 2000.
9 S. R. Drake and D. J. Otway, J. Chem. Soc., Chem. Commun.,
1991, 517.
10 R. Snethlage and K. Sterflinger, Stone Conservation, in Stone in
Architecture, ed. S. Siegesmund and R. Snethlage, Springer-Verlag,
Berlin Heidelberg, 2011, p. 411.
11 A. P. Ferreira Pinto and J. Delgado Rodrigues, J. Cult. Heritage,
2008, 9, 38.
Acknowledgements
Part of this work has been supported by the bilateral project
GALILEO by both French and Italian Foreign Affairs Ministries,
which are kindly acknowledged. The authors wish to thank
Fila Industria Chimica Spa, San Martino di Lupari, Padova,
Italy, owner of the Fei Quanta 200 FEG-ESEM instrument,
for allowing its use for the research work described in this
article. A. Moresco, C. Valerio and A. Aguiari are also
acknowledged for the technical assistance.
12 Final report of Galileo Project (2008/2009) ‘‘Materiali innovativi
per il consolidamento della pietra calcarea e dei dipinti murali’’-
´
Nouveaux materiaux consolidants pour pierre calcaire et peintures
murales.
13 W. S. Rees Jr. and D. A. Moreno, J. Chem. Soc., Chem. Commun.,
1991, 1759.
References
1 M. Favaro, P. Tomasin, F. Ossola and P. A. Vigato, Appl.
Organomet. Chem., 2008, 22, 698.
2 J. Ashurst, The cleaning and treatment of limestone by the lime
method: part I, Monumentum, 1984, 3, 233.
14 V. C. Arunasalam, I. Baxter, J. A. Darr, S. R. Drake,
M. B. Hursthouse, K. M. A. Malik and D. M. P. Mingos,
Polyhedron, 1998, 17, 641.
15 H. Colfen, Curr. Opin. Coll. Int. Sci., 2003, 8, 23.
¨
3 M. Ambrosi, L. Dei, R. Giorgi, C. Neto and P. Baglioni, Langmuir,
2001, 17, 4251.
4 D. C. Bradley, R. C. Merhotra and D. P. Gaur, Metal Alkoxides,
Academic Press, London, 1978.
5 D. C. Bradley, R. C. Mehrotra, I. P. Rothwell and A. Singh, Alkoxo
and Aryloxo Derivatives of Metals, Academic Press, London, 2001.
6 N. Y. Turova, E. P. Turevskaya, V. G. Kessler and
M. I. Yanovskaya, The Chemistry of Metal Alkoxides, Kluwer
Academic Publishers, Dordrecht, 2002.
16 F. Manoli and E. Dalas, J. Cryst. Growth, 2000, 218, 359.
17 http://www.nanomatch-project.eu/nanomatch0.html.
18 C. R. Hubbard and R. L. Snyder, Powder Diffr., 1988, 3(2), 74.
19 D. D. Perrin, W. L. F. Armarego and D. R. Perrin, Purification of
Laboratory Chemicals, Pergamon Press, Oxford, 1966.
20 S. C. Goel, M. A. Matchett, M. Y. Chiang and W. E. Buhro,
J. Am. Chem. Soc., 1991, 113, 1844.
21 R. Giorgi, L. Dei and P. Baglioni, Stud. Conserv., 2000, 45, 154.
22 P. Baglioni and R. Giorgi, Soft Matter, 2006, 2, 293.
c
2624 New J. Chem., 2012, 36, 2618–2624
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012