Article

Gel Cleaning in Heritage: Comparison of the Water Release

among Gels and Traditional Pads
Antonio Sansonetti 1, *, Cristiano Riminesi 1 , Sónia Mironiouk 2 , Noemi Proietti 1 , Valeria Di Tullio 1 ,
Roberto Nisticò 3 , Barbara Sacchi 1 and Carmen Canevali 3

1 Institute for Heritage Science, National Research Council, ISPC—CNR Milan Unit, 20154 Milan, Italy;
[email protected] (C.R.); [email protected] (N.P.); [email protected] (V.D.T.);
[email protected] (B.S.)
2 Escola Superior de Conservació i Restauració de Béns Culturals de Catalunya (ESCRBCC),
08033 Barcelona, Spain; [email protected]
3 Department of Materials Science, University of Milano-Bicocca, INSTM, Via R. Cozzi 55, 20125 Milano, Italy;
[email protected] (R.N.); [email protected] (C.C.)
* Correspondence: [email protected]

Abstract: Water release is a crucial aspect when considering cleaning effects on water-sensitive
materials. In conservation practice, a water-based cleaning method which limits water release is very
often needed. Unfortunately, this is not accompanied by an appropriate measure of the effectively
released water. In this paper, water release has been measured by comparing traditional cleaning
formulations, such as paper pulp and sepiolite, with several gar gel formulations, used by both
Italian and European conservators. The assessment has been carried out by the gravimetric method,
using three different stone material specimens as reference: Noto calcarenite, Manciano sandstone
and Black Bergamo limestone, whose porosity values and distributions are known. Moreover, water
distribution has been evaluated by portable NMR tests. Different commercial agar gel products
(Bresciani, CTS, Sigma), having different concentrations (3, 4, and 5%), application modes (rigid
at room T or fluid warm gels, with and without inserting Japanese tissue paper), and geometry
(horizontal in gravity force direction or vertical), have been compared to obtain a full scenario among
different water release mechanisms present in real conservation works. The paper faces the important
Citation: Sansonetti, A.; Riminesi, C.; issue of preparing reproducible chemical or water pads as well, useful for further research aimed at
Mironiouk, S.; Proietti, N.; Di Tullio, comparing cleaning effects in heritage conservation. The most interesting quantitative results can
V.; Nisticò, R.; Sacchi, B.; Canevali, C.
be summarized as follows. The water release measured from paper pulp and sepiolite was found
Gel Cleaning in Heritage: Comparison
to be 2 to 4 times higher than from any tested agar gel. Water release decreases by increasing agar
of the Water Release among Gels and
concentration; an increase in the agar concentration by 1% induces a decrease in water release in
Traditional Pads. Gels 2024, 10, 708.
the range 16.98–66.88 g depending on the stone; the increase from 4% to 5% is more obvious with
https://doi.org/10.3390/
gels10110708
respect to that from 3% to 4%. It is possible to assess the effect of the presence of Japanese paper,
which is able to reduce the water release from 18 to 76%, depending on the stone and on the agar
Academic Editor: Avinash J. Patil
used. The gravimetric results were also used in the preliminary calibration tests of a contact probe
Received: 2 October 2024 named System Unit Salinity Index (SUSI), recently patented and useful in providing humidity and
Revised: 25 October 2024 salinity indexes in a given porous material.
Accepted: 28 October 2024
Published: 2 November 2024 Keywords: gel cleaning; water release; agar gel; gravimetry; heritage; water-sensitive materials

Copyright: © 2024 by the authors.

1. Introduction
Licensee MDPI, Basel, Switzerland.
This article is an open access article Cleaning the surface of a heritage object is a very delicate and irreversible work; it
distributed under the terms and is aimed at selectively removing only those substances which endanger the appropriate
conditions of the Creative Commons conservation of the object material while hampering the correct legibility of the object’s
Attribution (CC BY) license (https:// cultural values [1–3]. Conservation scientists should support conservators to implement
creativecommons.org/licenses/by/ cleaning methods to improve their effectiveness and to minimize the possible side effects.
4.0/). For these reasons cleaning operations should be cautious, gradual, selective, and safe for

Gels 2024, 10, 708. https://doi.org/10.3390/gels10110708 https://www.mdpi.com/journal/gels

Gels 2024, 10, 708 2 of 19

the operator, for the environment, and, of course, for the object itself. Cleaning is planned
for removing very different materials from a compositional point of view, considering also
the degree of adhesion and internal cohesion. Deposits, metal compound stains, biological
growth, and black crusts are only some of many examples. The soiling can be formed
directly on the surface, being produced by chemical reactions with some substances coming
from the environment (new formation products), or they can be of exogenic origin and
simply deposited on the surface. Many other possible mechanisms explain the presence
of substances partially or completely extraneous to the original object’s material, but a
complete discussion of the topic is out of the scope of the present paper [4]. Anyway, it is
worth remembering that solubility plays a crucial role in chemical mechanisms of soiling
removal. For this reason, water can be applied to the outer surface of heritage objects via
spray, nebulizing systems, pads, or gels. Chemical removal can be implemented by adding
organic solvents [5,6], chelant substances [7,8], or enzymes [9,10] mixed in the formulate
designed for cleaning. In some restoration works the use of free water-based formulation is
discouraged because of the sensitivity of the constituting materials. Hence water release
control, when cleaning with a water-based system on a heritage surface, is a crucial point.
The scientific literature has covered this issue [11,12], but a precise measure of the amount
of water released on reference heritage materials is still lacking.
Traditional pads are usually prepared by mixing water with a thickener: clays, cel-
lulose fibers, wood, or paper pulp. The chemical–physical characteristics of the system
allow the water, which is essentially free as to transport phenomena, to be released onto
the substrate as a function of its porous microstructure [13].
The phenomena of water release/water absorption are determined by:
(a) thickener’s chemical composition and hydrophilicity;
(b) microstructure of the specific materials used;
(c) procedure of pad making;
(d) porous system of the material receiving the water released. A set of three different
stone materials have been chosen and studied in previous published research [14,15].
Hence water release is proportional to the total porosity of the substrate and to its
pore size distribution. In other words, it is possible to say that water release and absorption
capacity are the two faces of the same coin. Water pads are frequently used as cleaning
materials when soluble salts are the main components of efflorescences or concretions.
These traditional pads are easy to apply even on vertical surfaces or in undercuts.
In the last two decades gel-forming natural and synthetic polymers have been used to
improve cleaning chemical effectiveness, while controlling the water release to the porous
substrate [16–18]. In fact, in gels the presence of water physically bound to the polymer
structure has been proved [19]; the state of physically bounded water should limit the
water release.
Agar is a gelling material composed of polysaccharides extracted from red seaweeds
(genera Gelidium and Gracilariales) [20]; it can form semi-rigid, physical, thermo-reversible
gels by dispersing agar raw powder in water, heating the blend, and cooling at room tem-
perature [21]. At the end of this procedure, the polymer chains form a three-dimensional
network containing liquid water in the cavities. Conservators are increasingly appreci-
ating this kind of material for its ability to control the water release, especially when
used on water-sensitive materials such as gypsum stuccoworks, mural paintings [22,23],
and paper [24].
The article aims to address the following research issues, having significant impli-
cations and impacts on the practical implementation of cleaning works and on the safe-
guarding of water-sensitive heritage surfaces: quantitative comparison of water release
among traditional pads and gels, among different commercial agar gels (Bresciani, CTS,
and Sigma), between rigid at room T and warm fluid gels, and between horizontal and
vertical geometry; evaluation of the effect of Japanese paper and of complex mixtures
including sand.
Gels 2024, 10, 708 3 of 19

These comparisons are made by very simple gravimetric measures and they have
been carried out for the first time to the authors’ knowledge; the study’s focus, the issue
of preparing traditional pads with high reproducibility, will be discussed together with
its limits. Moreover, stones with a different porous system are proposed as reference set
materials, useful to carry out an accurate comparison study and able to work as well in
similar future research.
The data obtained from gravimetric measurements were used for a preliminary com-
parison with other types of data obtained through innovative instrumental systems aimed
at studying the water distribution inside the stone porous system by means of:
1. the so-called System Unit Salinity Index (SUSI) probe useful for evaluating the mois-
ture content (MC) in a semi-sphere of about 2 cm in diameter;
2. a portable NMR device connected to a single-sided sensor.

2. Results and Discussion

2.1. Stone Porous System
The porous system of the three tested stones is given by the total open porosity P%
(total volume of the pores in %) and by the distribution of the pore volume, classified
according to the pore size [14,25,26].
Noto calcarenite is characterized by a high porosity (36.2%), due to a great amount of
macropores (86.8%); the most representative pore class corresponds to a diameter in the
range 0.5–5.0 µm [27] (see Tables 1 and 2). For these reasons the stone has a high water
absorption coefficient [14].

Table 1. Lithotype, % porosity, pore distribution, and pore diameter main range.

Pore Diameter
Lithotype Porosity (%) Pore Distribution
Main Range (µm)
Noto calcarenite 36.2 unimodal 0.5–5.0
Manciano sandstone 10.7 unimodal 0.01–5.0
Bergamo black limestone 0.5 unimodal 1.0–4.0

Table 2. Percentage porosity distribution in the three stones under test.

Lithotype Noto Manciano Bergamo

Megapores (%) 12.2 25.9 12.8
Macropores (%) 86.8 67.0 67.4
Mesopores (%) 1.0 7.1 19.7

Manciano sandstone porosity could be considered intermediate in the range of the

tested stones (10.7%) and it is composed of mega- (25.9) and macropores (67.0), having
size ranging from 0.01 and 5.0 µm [28]; this stone has a high water absorption coefficient
as well.
On the contrary, Bergamo Black limestone has a very low total open porosity (0.5%),
quite different from the other limestones; pore size ranges mainly from 1.0 to 4.0 µm, being
12.8% megapores and 67.4% macropores [29] (see Tables 1 and 2).
Thus, it is possible to observe that the three stones are characterized by a very different
total open porosity percentages; moreover, the larger class of pores is macropores for
each stone.

2.2. Water Release: Comparison among Traditional Pads and Agar Gels
The phenomenon under study could be observed as the water released by pads
and gels and the water absorbed by the stone capillary system. In the following series
 Gels 2024, 10, 708 4 of 19

Gels 2024, 10, x FOR PEER REVIEW of histograms, the water release, measured via gravimetry, has been displayed. 4 ofSome
20
considerations can be detailed observing the results.
The release of water from paper pulp and sepiolite is greater than from the 3% gels by
a factor varying from approximately 2 to 5. This ratio is evident when observing Figure 1a,b,
histograms, the water release, measured via gravimetry, has been displayed. Some con-
displaying the water releases measured on Noto calcarenite and Manciano sandstone. On
siderations can be detailed observing the results.
the contrary, on the black limestone the reading is heavily affected by experimental error
The release of water from paper pulp and sepiolite is greater than from the 3% gels
(Figure 1c).
by a factor varying from approximately 2 to 5. This ratio is evident when observing Figure
The two lithotypes with the greatest total open porosity (Noto and Manciano; see
1a,b, displaying the water releases measured on Noto calcarenite and Manciano sand-
Table 1) indicate the following trend for the amount of water released by gels:
stone. On the contrary, on the black limestone the reading is heavily affected by experi-
Noto:error
mental Sigma < CTS1c).
(Figure < Bresciani;
Manciano: Sigma ≈ CTS ≤ Bresciani.

(a)

(b)

Figure 1. Cont.
Gels 2024, 10, x FOR PEER REVIEW 5 of 20
Gels 2024, 10, 708 5 of 19

(c)
Figure 1. (a)1.Water
Figure released
(a) Water on Noto
released calcarenite.
on Noto (b) (b)
calcarenite. Water released
Water on on
released Manciano sandstone.
Manciano (c)(c)
sandstone. Wa-Water
ter released
releasedon onBergamo
BergamoBlack
Blacklimestone.
limestone.

TheThis
two trend
lithotypes with the greatest
is in agreement with thetotal open
better porosity
cleaning (Noto andobserved
effectiveness Manciano; when seegels
are1)used
Table for removing
indicate the following copper trendstains
for from marbleof
the amount [30,31].
water released by gels:
Noto:NotoSigmacalcarenite, due to its porous structure, absorbs approximately 5 times more
< CTS < Bresciani;
water
Manciano: Sigma ≈ CTS sandstone
than the Manciano ≤ Bresciani. and about 150 times more than black limestone. In any
tested set ofisspecimens,
This trend in agreement the with
blacktheBergamo limestone
better cleaning should not be
effectiveness considered
observed whena gels
reliable
type of
are used forstone
removingto be copper
used instains
such afromprocedure. This result is probably due to the very low
marble [30,31].
porosity, the small due
Noto calcarenite, amountto itsofporous
water structure,
absorbed, absorbs
and hence the high experimental
approximately 5 times more errors;
water than the Manciano sandstone and about 150 times more than black limestone. In for
these results are also confirmed by the trends of the water released vs. time, available
anythe agarset
tested CTS gel only. In the
of specimens, fact,black
theseBergamo
curves show that Noto
limestone shouldandnot
Manciano absorb athe
be considered water
reli-
released by the gel in a gradual and progressive manner (see
able type of stone to be used in such a procedure. This result is probably due to the very Figure 2), while the black
limestone absorbs in a less regular way and, indeed, in some
low porosity, the small amount of water absorbed, and hence the high experimental er- moments, it seems to release
rors;the absorbed
these resultswater.
are also Moreover,
confirmed forby
Noto and Manciano
the trends stones
of the water it was vs.
released possible to obtain the
time, available
absorption coefficient from the curve, while this was not
for the agar CTS gel only. In fact, these curves show that Noto and Manciano absorb possible for the black limestone
the
due to the non-linear trend (see Table 3).
water released by the gel in a gradual and progressive manner (see Figure 2), while the
Paper pulp
black limestone absorbsseems intoa release a slightly
less regular waygreater amount
and, indeed, inofsome
water with respect
moments, to sepiolite.
it seems to
release the absorbed water. Moreover, for Noto and Manciano stones it was possible to(see
For what concerns gels, the water release is a linear function of agar concentration
Figure
obtain 3a–c). By increasing
the absorption coefficient thefrom
agartheconcentration
curve, whilefrom this 3%
wastonot5%,possible
the water forrelease
the blacklowers
by a third; in particular, an increase
limestone due to the non-linear trend (see Table 3). of 1% in the agar concentration induces a decrease in
water release in the range 16.98–66.88 g depending on the stone: the increase from 4% to 5%
is 3.
Table more
Water obvious
release with
by CTSrespect
Agar 3%; to the oneabsorption
water from 3% to 4%; thiscoefficient
capillarity effect could
(N.D.be =due
not to the greater
detected).
hampering of the polymeric chains, influencing the water transport. This consideration
is evident when Lithotype looking at Noto and Manciano. Once again, CA [(Kg/m 2)/sec^1/2]
the results of the very low
porosity blackNoto Calcarenite
limestone should not be considered reliable. 3.80 × 10−3
Manciano Sandstone 2.10 × 10−3
Table 3. Water release by CTS Agar 3%; water absorption capillarity coefficient (N.D. = not detected).
Black Limestone N.D.
Lithotype CA [(Kg/m2 )/secˆ1/2 ]
Paper pulp seems
Noto to release a slightly greater amount of water
Calcarenite with
3.80 × 10respect
−3 to sepio-
lite.
Manciano Sandstone 2.10 × 10−3
For what concerns gels, the water release is a linear function of agar concentration
Black
(see Figure 3a–c). By Limestonethe agar concentration from 3% to 5%,
increasing N.D.the water release
lowers by a third; in particular, an increase of 1% in the agar concentration induces a
 decrease in water release in the range 16.98–66.88 g depending on the stone: the increase
fromdecrease
4% to 5% is more obvious with respect to the one from 3% to 4%; this effect could be
in water release in the range 16.98–66.88 g depending on the stone: the increase
due from
to the4% to 5% ishampering
greater of the
more obvious withpolymeric chains,
respect to the influencing
one from thethis
3% to 4%; water
effecttransport.
could be
Thisdue
consideration
to the greater hampering of the polymeric chains, influencing the water the
is evident when looking at Noto and Manciano. Once again, results
transport.
Gels 2024, 10, 708 of the very low porosity black limestone should not be considered reliable. 6 of 19
This consideration is evident when looking at Noto and Manciano. Once again, the results
of the very low porosity black limestone should not be considered reliable.

Figure 2. Amount
Figure of water absorbed forfor
surface unit vs.vs.
time ininmin
min(CTS
(CTSAgar
Agar3%
3%on
on Noto
Noto calcar-
Figure
enite, 2.2. Amount
Manciano Amount of water
of water
sandstone, and
absorbed
absorbed
Black
surface
for
limestone).
unit
surface unittime
vs. time in min (CTS Agar 3%calcar-
on Noto
enite, Manciano sandstone, and Black limestone).
calcarenite, Manciano sandstone, and Black limestone).

Gels 2024, 10, x FOR PEER REVIEW 7 of 20

(a)
(a)

(b)

Figure 3. Cont.
 Gels 2024, 10, 708 7 of 19

(b)

(c)
Figure 3. (a)
Figure Water
3. (a) released
Water on on
released Noto Calcarenite
Noto by by
Calcarenite gelsgels
as aasfunction of Bresciani
a function Agar
of Bresciani percentage
Agar percentage
(3%, 4%, 5%). (b) Water released on Manciano Sandstone by gels as a function of Bresciani
(3%, 4%, 5%). (b) Water released on Manciano Sandstone by gels as a function of Bresciani Agar
Agar
percentage (3%, 4%, 5%). (c) Water released on Black limestone by Gels as a function of Bresciani
percentage (3%, 4%, 5%). (c) Water released on Black limestone by Gels as a function of Bresciani
Agar percentage (3%, 4%, 5%).
Agar percentage (3%, 4%, 5%).
NoNo
visible effect
visible was
effect observed
was observedbetween
between thethe
horizontal geometry
horizontal to the
geometry vertical
to the oneone
vertical
(see Figure 4a,b). Hence, the overall experimentation results, more easily obtained
(see Figure 4a,b). Hence, the overall experimentation results, more easily obtained in the in
laboratory for the horizontal geometry, can be translated to conservation sites, where
the laboratory for the horizontal geometry, can be translated to conservation sites, where usu-
allyusually
the water
the is released
water via a vertical
is released geometry.
via a vertical geometry.
When agar is used as a warm fluid at around 40–45 ◦ C, it allows a greater water release
than as rigid gels at room temperature (see Figure 5a,b); once again the trend is evident in
Noto and Manciano, but not in black limestone.
For what concerns the role of Japanese paper, there is a sort of “barrier effect” by
the Japanese paper, more evident for Noto stone. In fact, looking at the histograms in
Gels 2024, 10, x FOR PEER REVIEW Figure 6a,b, it is possible to assess the effect of the presence of Japanese paper,8 which
of 20 brings

a heavy reduction of the water release from 18 to 76%, depending on the stone and on the
agar used. Bresciani agar is more sensitive to the presence of Japanese paper.

(a)

Figure 4. Cont.
Gels 2024, 10, 708 8 of 19

(a)

(b)
Gels 2024, 10, x FOR PEER REVIEW 9 of 20
Figure 4. (a) 4.
Figure Noto
(a) calcarenite water release
Noto calcarenite in vertical
water release and horizontal
in vertical geometrygeometry
and horizontal by CTS Agar 3% gel.
by CTS Agar 3% gel.
(b) Manciano sandstone
(b) Manciano water water
sandstone releaserelease
in vertical and horizontal
in vertical geometry
and horizontal by CTS Agar
geometry 3% Agar
by CTS gel. 3% gel.

When agar is used as a warm fluid at around 40–45 °C, it allows a greater water re-
lease than as rigid gels at room temperature (see Figure 5a,b); once again the trend is evi-
dent in Noto and Manciano, but not in black limestone.

(a)

(b)
Figure 5. (a)5.Noto
Figure calcarenite
(a) Noto water water
calcarenite releaserelease
by Bresciani Agar 4%,Agar
by Bresciani applied
4%,as warm fluid
applied and as
as warm rigidand as rigid
fluid
gel.gel.
(b) Manciano sandstone water release by Bresciani Agar 4%, applied as warm fluid and as rigid
(b) Manciano sandstone water release by Bresciani Agar 4%, applied as warm fluid and as rigid gel.
gel.

For what concerns the role of Japanese paper, there is a sort of “barrier effect” by the
Japanese paper, more evident for Noto stone. In fact, looking at the histograms in Figure
6a,b, it is possible to assess the effect of the presence of Japanese paper, which brings a
heavy reduction of the water release from 18 to 76%, depending on the stone and on the
Gels 2024, 10, 708 9 of 19

Gels 2024, 10, x FOR PEER REVIEW 10 of 20

Finally, black limestone does not allow any reliable consideration (Figure 6c).

(a)

(b)

(c)
Figure 6. (a)
Figure 6. Noto calcarenite.
(a) Noto Comparison
calcarenite. of waterofrelease
Comparison water with andwith
release without
andthe Japanese
without thepaper.
Japanese paper.
CTS Agar 3% and Bresciani Agar 3%. (b) Manciano sandstone. Comparison of water release with
CTS Agar 3% and Bresciani Agar 3%. (b) Manciano sandstone. Comparison of water release with and
without the Japanese paper. CTS Agar 3% and Bresciani Agar 3%. (c) Black limestone. Comparison
of water release with and without the Japanese paper. CTS Agar 3% and Bresciani Agar 3%.
Gels 2024, 10, x FOR PEER REVIEW 11 of 20

and without the Japanese paper. CTS Agar 3% and Bresciani Agar 3%. (c) Black limestone. Compar-
Gels 2024, 10, 708
ison of water release with and without the Japanese paper. CTS Agar 3% and Bresciani Agar103%.
of 19

2.3. Water Release and Distribution as Analyzed by Portable NMR

2.3. Water Release andof
The amplitude Distribution
the NMR as Analyzed
signal by Portable
corresponds to NMR
the concentration of water at dif-
ferentThe amplitude
depths. Higherof the NMR signal
amplitudes corresponds
indicate greaterto the concentration
water content, while oflower
water amplitudes
at different
depths.
suggestHigher
less or amplitudes
no water. By indicate greater
scanning water content,
at different depths,while
NMRlower amplitudes
can reveal suggest
how water is
less or no water.
distributed fromBythescanning
surface at different
into depths,
the interior NMR
of the can reveal
material. Thishow
helpswater is distributed
in understanding
from the surface
the extent intopenetration
of water the interiorand
of the
howmaterial.
it variesThis
withhelps in understanding
depth. The shape of the thedepth
extent of
pro-
water penetration and how it varies with depth. The shape of the depth profile
file can provide insights into the material’s absorption characteristics. For instance, a sharp can provide
insights
increaseinto the material’s
in signal near theabsorption characteristics.
surface followed For instance,
by a plateau suggestsasurface
sharp increase in signal
absorption with
near the surface followed by a plateau suggests surface absorption
limited deeper penetration. Conversely, a gradual increase in signal depth indicates with limited deeper
more
penetration.
uniform water Conversely, a gradual
distribution. increase
Furthermore, theinmaximum
signal depth indicates
depth at whichmore
wateruniform water
penetration
distribution. Furthermore, the maximum depth at which water penetration
is detected can be determined by analyzing where the NMR signal starts to diminish or is detected
can be determined
disappears. by analyzing
This provides whereofthe
an estimate how NMR signal
deeply starts
water canto diminish
infiltrate theormaterial
disappears.
over
This
a given time period. In Figure 7, all the 1H depth profiles acquired in Noto, given
provides an estimate of how deeply water can infiltrate the material over a time
Manciano,
period. In Figure
and Bergamo 7, all
black the 1H are
limestone depth profiles acquired in Noto, Manciano, and Bergamo
shown.
black limestone are shown.

Figure 7. ¹H NMR depth profiles for Noto stone (a), Manciano stone (b), and Black limestone (c) after
Figure
the 7. ¹H NMR
application depth profiles
of Sepiolite, for and
Agar CTS, NotoAgar
stone (a), Manciano
Bresciani stone (b), andforBlack
at 3% concentration limestone (c)
30 min.
after the application of Sepiolite, Agar CTS, and Agar Bresciani at 3% concentration for 30 min.
¹H depth profiles acquired in Noto stones (Figure 7a) reveal that Noto treated with
¹H depth
sepiolite exhibitsprofiles acquired
the highest NMR in Noto
signalstones (Figurewith
amplitude, 7a) reveal
a sharpthat riseNoto
neartreated with
the surface
sepiolite exhibits the highest NMR signal amplitude, with a sharp
(0–1000 µm) followed by a plateau as depth increases (up to 5000 µm). Noto stones treated rise near the surface (0–
1000
by CTS μm)
agarfollowed by a plateau
and Bresciani agar show as depth
overallincreases
lower NMR (up signal
to 5000amplitudes,
μm). Notowith stones treated
a gradual
by CTS agar
increase and
in the Bresciani
first 1000 µm agarandshow overall lower
a gradual declineNMR signalas
in signal amplitudes, with a grad-
depth increases. The
ual increase
maximum in the firstdepth
penetration 1000 μm and a into
of water gradual decline
the stone in signal
also shows as depth increases.
significant The
differences
maximum
among penetration
sepiolite, Bresciani depth
agar,ofand water
CTS into
agar.the stone
In CTS alsothe
agar, shows
¹H NMR significant differences
signal decreases to
among sepiolite, Bresciani agar, and CTS agar. In CTS agar, the
zero, indicating that water penetration does not reach a depth of 5 mm within the stone ¹H NMR signal decreases
to zero,
after 30 indicating that waterBresciani
min of application. penetration does
agar not reach
releases a depth
a larger of 5 mm
amount withinatthe
of water stone
deeper
after 30compared
depths min of application.
to CTS agar.Bresciani agar releases a larger amount of water at deeper
In Manciano
depths compared stone agar. 7b), 1 H depth profiles follow a similar pattern, though
to CTS(Figure
the NMR signal amplitudes
In Manciano stone (Figureare 7b),
lower1Hoverall compared
depth profiles to (a).
follow Sepiolite
a similar again though
pattern, shows thethe
highest
NMR signalsignal, but the profiles
amplitudes of CTS
are lower agarcompared
overall and Bresciani
to (a).agar are closer
Sepiolite againinshows
magnitude to
the high-
one another. Agar gels show a greater variation in the signal across
est signal, but the profiles of CTS agar and Bresciani agar are closer in magnitude to one the depth, particularly
in the firstAgar
another. 1000–2000
gels showµm, afollowed by a gradual
greater variation decline.
in the signal across the depth, particularly in
In black
the first limestone
1000–2000 μm,(Figure
followed7c),bythe depth profiles
a gradual show significantly lower NMR signal
decline.
amplitudes,
In blacksuggesting no water
limestone (Figure absorption
7c), the depthwithin
profilestheshowporous structure.lower
significantly All three
NMRgelssig-
exhibit similar profiles,
nal amplitudes, suggesting withnoinitial
waterpeaks in thewithin
absorption first 500theµm,
porousfollowed by aAll
structure. drop-off as
three gels
depth increases. This indicates that water molecules are distributed
exhibit similar profiles, with initial peaks in the first 500 μm, followed by a drop-off as in a thin layer on the
surface of the stone.
depth increases. ThisOverall,
indicates these
thatprofiles illustrate are
water molecules howdistributed
the different in agels
thininteract
layer onwith
the
various substrates in terms of water absorption and distribution, with sepiolite consistently
showing higher water release compared to agar gels, particularly in deeper layers.
The data acquired by means of NMR device allow the following further considerations:
 surface of the stone. Overall, these profiles illustrate how the different gels interact with
various substrates in terms of water absorption and distribution, with sepiolite consist-
ently showing higher water release compared to agar gels, particularly in deeper layers.
Gels 2024, 10, 708 The data acquired by means of NMR device allow the following further considera- 11 of 19
tions:
in Noto calcarenite water penetrates up to 5 mm with sepiolite, while agar pushes the
in Noto calcarenite
water penetration to 1–3 mm,water penetrates
ensuring up to 5 mm with
a non-superficial sepiolite,
cleaning while agar pushes the
action;
water penetration
a very to 1–3has
similar trend mm, ensuring
been showna by
non-superficial cleaning action;
Manciano sandstone: a smaller amount of
water isaabsorbed
very similar
withtrend hastobeen
respect Notoshown by Manciano
calcarenite, but aroundsandstone:
the same apenetration
smaller amount
depth of
water
is reached;is absorbed with respect to Noto calcarenite, but around the same penetration depth
is reached;
NMR data confirm the higher water release from Bresciani agar than from CTS agar;
forNMR
Blackdata confirm
Bergamo the higher
limestone, water
only release from
a superficial layerBresciani agar
of water is than from
released, withCTS agar;
almost
for Black Bergamo limestone, only a superficial layer of water is released, with almost
no penetration.
no penetration.
2.4. Water Absorbed by Stone Specimens Measured by Means of SUSI Probe
2.4. Water Absorbed by Stone Specimens Measured by Means of SUSI Probe
SUSI testing was carried out on stone specimens; to this aim a set of five specimens
SUSI testing was carried out on stone specimens; to this aim a set of five specimens for
for each stone was moistened with the procedure already described.
each stone was moistened with the procedure already described.
Then the MC % provided by the instrument was measured. The measures provided
Then the MC % provided by the instrument was measured. The measures provided
the following data shown in Figure 8a.
the following data shown in Figure 8a.

Noto Calcarenite
14
12
10
MC (%)

Paper pulp
8
Sepiolite
6
CTS Agar 3%
4
2
0
(a)

CTS Agar 3%
8
7
6 Noto Calcarenite
MC (%)

5
4 Manciano Sandstone
3
Black Limestone
2
1
0
(b)

Figure 8. Cont.
Gels 2024, 10, x FOR PEER REVIEW 13 of 20

Gels 2024, 10, 708 12 of 19

(c)
Figure 8. (a)
Figure 8. Noto calcarenite
(a) Noto Moisture
calcarenite content
Moisture (MC%)
content afterafter
(MC%) water release
water by traditional
release andand
by traditional CTSCTS
Agar 3%. (b) Comparison of the moisture content (MC%) released on the three different stones.
Agar 3%. (b) Comparison of the moisture content (MC%) released on the three different stones. (c)
Comparison of the water released by CTS Agar 3% on the three different stones.
(c) Comparison of the water released by CTS Agar 3% on the three different stones.

ForFor
water release
water by by
release traditional
traditionalpads
padsand gels,
and the
gels, thetrend
trendgiven
givenby bygravimetric
gravimetricmeas-
measure-
urements is confirmed, with release by CTS agar lower than
ments is confirmed, with release by CTS agar lower than that by traditional that by traditional
pads pads
(around
(around one third
one third less) less)
TheThe
SUSI datadata
SUSI indicate that,that,
indicate among the the
among traditional pads,
traditional the the
pads, oneone
releasing more
releasing water
more water
is sepiolite; these
is sepiolite; datadata
these are are
different from
different those
from from
those gravimetric
from gravimetricmeasures, where
measures, the the
where padpad
releasing more
releasing morewater is paper
water pulp.
is paper pulp.
TheThewater release trends measured
water release trends measured by gravimetry
by gravimetry andand
by SUSI
by SUSIare are
veryvery
similar, as as
similar,
shown
shownin Figure 8b,c.
in Figure 8b,c.

3. Conclusions
3. Conclusions
TheThe results
results obtained
obtained highlight
highlight thethe different
different performancesgiven
performances givenbybythe
thetested
testedclean-
cleaning
systems on the various tested stone substrates. The following points
ing systems on the various tested stone substrates. The following points summarize the summarize the main
scientific results.
main scientific results.
Traditional
Traditional pads
pads release
release a largeramount
a larger amountofofwater
waterthan
than gels,
gels, from
from 22 to
to55times,
times,depend-
de-
ing on the stone considered.
pending on the stone considered.
Paper
Paper pulppulp releases
releases the the greatest
greatest amount
amount of water
of water in any
in any case:case: in the
in the casecase of Noto
of Noto
calcarenite water released by agar gel (Bresciani 3%) is only 16.07% of
calcarenite water released by agar gel (Bresciani 3%) is only 16.07% of the water releasedthe water released
by paper
by paper pulp;pulp; incase
in the the case of Manciano
of Manciano sandstone
sandstone the water
the water released
released by Bresciani
by Bresciani agar 3%agar
3% rises to 31.09%. The comparison of sepiolite vs. Bresciani agar 3%
rises to 31.09%. The comparison of sepiolite vs. Bresciani agar 3% indicates that the latterindicates that the
latter releases only 25.86% (Noto) and
releases only 25.86% (Noto) and 39.73% (Manciano). 39.73% (Manciano).
Bresciani
Bresciani agaragar released
released a larger
a larger amount
amount of water
of water with with respect
respect to the
to the otherother tested
tested agaragar
gels (in the case of Noto stone, water release by Bresciani agar is 211% with respect to CTS
gels (in the case of Noto stone, water release by Bresciani agar is 211% with respect to CTS
agar 3%; in the case of Manciano sandstone the increase is around 127%).
agar 3%; in the case of Manciano sandstone the increase is around 127%).
The overall results advise conservators to choose any agar gel for cleaning very
The overall results advise conservators to choose any agar gel for cleaning very sen-
sensitive materials, such as murals, gypsum-based stuccoworks, or paper. The system can
sitive materials, such as murals, gypsum-based stuccoworks, or paper. The system can be
be improved through the use of Japanese paper, which can reduce water release by up
improved through the use of Japanese paper, which can reduce water release by up to
to 75%.
75%.
Gravity does not influence the measures which can be carried out both in horizontal
Gravity does not influence the measures which can be carried out both in horizontal
and in vertical geometry; hence, the results acquired in this study can be transferred to
and in vertical geometry; hence, the results acquired in this study can be transferred to
vertical surfaces on conservation sites.
vertical surfaces on conservation sites.
Regarding possible future research plans, lithotypes should be chosen with an appro-
priate porosity range; very low porosity stones, such as Black Bergamo limestone, proved
to be not reliable due to the high experimental errors.
Gels 2024, 10, 708 13 of 19

It is important to note the low values of standard deviation for Noto and Manciano
stones, assuring the data precision quality.
A possible list of errors includes the difficulties in correctly shaping a pad of the
desired sizes using sepiolite and paper pulp. Moreover, as the shaping operation was not
simple, it requires some time to be performed, during which the water release starts. The
need for squeezing these kinds of traditional pads is a problem in reproducibility as well.
Thus, experiments with both sepiolite and paper pulp are hardly reproducible.
On the contrary, agar gels can be cut in Petri dishes and allow a better reproducibility.
In the everyday life of conservators, a challenge is to match the pads’ correct range
of porosity in order to have the most effective water release, for example, for desalination.
In fact, to obtain the best desalination effect, the pads’ pores should compete with the
stones’ porous system to induce the best outward suction of the solutions formed inside
the pores. In attempting to do this, conservators prepare blends by mixing together paper
pulp, sepiolite, sand, and gels. As part of this research, the authors prepared three different
typologies of mixes (sepiolite, paper pulp, and sand, in various proportions), testing them
on the stones used in the experimentation. Unfortunately, no specific trends as concerns
water release have been individuated, hence the corresponding data are not presented.
The authors suppose that a great number of variables, not easily controlled, influence the
obtainment of reliable data; just to name a few, these mixes should be manually squeezed
in order to reach the right consistency to be applied on the stone surface. At the moment
the authors do not present a procedure to prepare a reproducible mix. To predict and to
explain the performance of these mixes is an open issue and a challenge for future works
for the scientific community.

4. Materials and Methods

4.1. Reference Stones
The lithotypes used are Noto calcarenite, Manciano sandstone, and Bergamo black
limestone [32]; these materials are classified as sedimentary rocks and they have been
chosen because of their different porous systems, both in terms of the total percentage
of porosity and pore size distribution. Moreover, they were part of a previous research
program on the common topic of gel water transfer phenomena on stone materials [14]; for
these reasons they have been considered as a reference.

4.2. Preparation and Application of Water-Releasing Materials

Traditional pads were made with paper pulp and sepiolite, two systems allowing a
different water distribution in the microstructure.
As paper pulp, Arbocel® BC 1000 has been used, chosen because of its widespread
presence in the market; the product is capable of mixing with an amount of water equal to
4/5 times its weight [33].
Sepiolite [34] can be blended with a large amount of water while avoiding exces-
sive swelling.
Very often and especially in the latter case, traditional pads are used on stone surfaces
by inserting a Japanese paper sheet in order to allow a faster and more complete removal
of the pads, avoiding any residues on the stone surface [35].
The optimization of each pad and gel application is part of the research plan; it has
been designed considering the following requirements: a good adhesion between pad (or
gel) and stone surface; reproducible size and thickness of the pads or gels; reproducible
water content in each pad or gel, used on a set of 5 identical 5 × 5 × 2 stone specimens;
minimization of any experimental error and optimization of a reproducible procedure.
Pads and gels of 5 × 5 × 1 cm were prepared, so that they could be perfectly super-
imposed on the stone specimen polished surface, each with an equal “reservoir” 1 cm
thick. Traditional pads were placed directly on the stone surface while gels were placed
inside a Petri dish, then cut and positioned on the stone surface. Each contact time between
traditional pads/gels and stones was 30 min. During the measuring test the pad (or the
Gels 2024, 10, 708 14 of 19

gel) was covered with a polyethylene film to avoid water evaporation. After this period,
the pad or gel was delicately removed, paying attention not to leave any residue, then the
stone specimen was immediately weighed.
To study the water release kinetics of gels during their application on the stone surface,
both gels and stone specimens were weighed at specified timings (every 5 min during the
first 30 min, then after 60 min).
In the following, each procedure is detailed.
Paper pulp pads were prepared with the ratio of paper:water reported in Table 4. The
paper pulp pads were applied on the surface and shaped till reaching a uniform 1 cm
thickness, then weighed. When required, Japanese paper was inserted. The first weight
was used as a reference for the whole series of five specimens, so that the following paper
pulp pads were prepared with exactly the same weight. After 30 min, the polyethylene
film and the pad were removed and the specimen was immediately weighed. At the end
of the procedure, the same specimens were put in an oven at 60 ◦ C to be prepared for the
next steps.

Table 4. Tested Mixes. Composition of the traditional pads used.

Pad Thickener Water (mL) Arbocel BC 1000 (g) Sepiolite (g) Sand (g)
Paper pulp Arbocel BC 1000 200 35 / /
Sepiolite Sepiolite CTS 200 / 120 /
Mix 1 Arbocel + sepiolite 200 26 30 /
Mix 2 Arbocel + sepiolite + sand 200 26 30 15
Mix 3 Arbocel + sepiolite + sand 200 17 60 15

Sepiolite pulp pads were prepared with the ratio of sepiolite:water reported in Table 4;
then the procedure was identical to the one used for paper pulp pads.
Following a previous experimentation attempting to remove soluble salts from a
plaster dated back to the XV century [19], the authors tested three different mixes composed
of sepiolite, paper pulp, and sand to match the different microstructures of the pads, hence
having different water transport phenomena and water releases. The three mixes tested
are described in Table 4. Conservators blend different materials in order to match the most
appropriate pad and optimize the soiling extraction capability and the water release [36].
Both of these physical phenomena mostly depend on the porous system present in the
pad. The operation attempts to optimize the cleaning effects especially during desalination.
Even in this case, water release has never been measured. Hence, the present study tested
three mixed materials, obtained by blending paper pulp, sepiolite, and sand (see Table 4).
It has been proven that the different agar products available in the market are slightly
different in their composition and properties, depending on natural variability but also on
different extraction techniques from the raw materials [37]. For what concerns the Italian
and, more generally, the European market, both CTS (CTS s.r.l., Monza, Italy) and Bresciani
(Bresciani s.r.l., Milano, Italy) products are the most widespread. Moreover, they were com-
pared with a Sigma product (A7002, Sigma-Aldrich, St. Louis, MO, USA). Agar powders
(CTS AgarArt; Bresciani; Sigma) were mixed with water in the specified concentration (3%);
for Bresciani powder, gels with concentrations of 4 and 5% were prepared as well to study
the rigid gels’ performance as a function of increasing concentration. Mixes were heated
in a microwave oven (700 W–2 min); then the sol produced was re-heated in the same
condition. Conservators heat the agar gels twice in order to improve their rheological and
chromatic features: agar gels heated twice loose the yellowish color obtained after the first
heating, and their transparency increases. Then, the sol was poured into a Petri dish up to a
thickness of 1 cm, cooled at room temperature, and sized to 5 × 5 cm. Agar gels were used
on three stones (Figure 9), either as warm fluid (around 40–45 ◦ C) or as rigid gels at room T.
 were heated in a microwave oven (700 W–2 min); then the sol produced was re-heated in
the same condition. Conservators heat the agar gels twice in order to improve their rheo-
the same condition. Conservators heat the agar gels twice in order to improve their rheo-
logical and chromatic features: agar gels heated twice loose the yellowish color obtained
logical and chromatic features: agar gels heated twice loose the yellowish color obtained
after the first heating, and their transparency increases. Then, the sol was poured into a
after the first heating, and their transparency increases. Then, the sol was poured into a
Petri dish up to a thickness of 1 cm, cooled at room temperature, and sized to 5 × 5 cm.
Gels 2024, 10, 708 Petri dish up to a thickness of 1 cm, cooled at room temperature, and sized to 5 ×155ofcm.
19
Agar gels were used on three stones (Figure 9), either as warm fluid (around 40–45 °C) or
Agar gels were used on three stones (Figure 9), either as warm fluid (around 40–45 °C) or
as rigid gels at room T.
as rigid gels at room T.

Figure 9. From left to right: Noto calcarenite, Manciano sandstone, Bergamo black limestone with
Figure9.9.From
Figure Fromleft
lefttotoright:
right:Noto
Notocalcarenite,
calcarenite,Manciano
Mancianosandstone,
sandstone,Bergamo
Bergamoblack
blacklimestone
limestonewith
with
rigid Bresciani agar gel 3%.
rigid
rigidBresciani
Brescianiagar
agargelgel3%.
3%.
The
The tested agar gels were studied and the comparisons are schematized in Figure 10.
Thetested
testedagar
agargels
gelswere
werestudied
studiedand
andthe
thecomparisons
comparisonsare
areschematized
schematizedin
inFigure
Figure10.
10.

Figure 10. Synoptic scheme of the overall comparison among gels.

Figure10.
Figure 10.Synoptic
Synopticscheme
schemeofofthe
theoverall
overallcomparison
comparisonamong
amonggels.
gels.
It is possible to see from Figure 10 that the testing procedure included another two
points:ItIt isis possible
possible to
to see
see from
fromFigure
Figure1010that
thatthethe
testing procedure
testing procedure included
includedanother two
another
points:
two points:
1. comparison between the water released by an agar gel in a warm fluid condition
1.1. (38/40
comparison between
comparison°C)
between the water
and the samethe
water released
gel rigid at
released byby an
an agar
room temperature;
agar gel
gel in
inaawarm
warmfluid
fluidcondition
condition
(38/40 °C)
(38/40 ◦ C) and
and the same gel rigid at room temperature;
2. to answer the research question regarding any possible differences among all the
2.2. to to answer
answer the
the research
research question
question regarding
regarding any any possible
possible differences
differences among
amongall allthe
the
tested materials presented here, with a horizontal or vertical position, so that the in-
tested materials presented here,
tested here, with
withaahorizontal
horizontalororvertical
verticalposition, soso
position, that thethe
that in-
fluence of gravity force can be induced and the geometry usually present in building
fluence ofofgravity
influence gravityforce
forcecan
canbebeinduced
induced and
and the
the geometry usually
usually present
presentininbuilding
building
conservation sites can be studied.
conservation sites
conservation sites can
can be
be studied.
studied.
4.3.
4.3. Study Methods
4.3.Study
StudyMethods
Methods
4.3.1. Procedure to Carry Out the Gravimetric Measures on Stone Specimens
The European standard EN 15801:2009 (Conservation of cultural property—Determination
of water absorption by capillarity) inspired the method used for the gravimetric measures
on the stone specimens [38]. The standard suggests the use of a pad made with filter
paper on which a stone specimen is placed. In the present paper, two variations have been
introduced to fulfil the needs of the gravimetric measures:
 4.3.1. Procedure to Carry Out the Gravimetric Measures on Stone Specimens
The European standard EN 15801:2009 (Conservation of cultural property—Deter-
mination of water absorption by capillarity) inspired the method used for the gravimetric
measures on the stone specimens [38]. The standard suggests the use of a pad made with
Gels 2024, 10, 708
filter paper on which a stone specimen is placed. In the present paper, two variations16have of 19

been introduced to fulfil the needs of the gravimetric measures:

(a) to substitute the filter paper with the specific material under testing (sepiolite, paper
(a) pulp,
to substitute
and gels);the filter paper with the specific material under testing (sepiolite, paper
pulp, and gels);
(b) to invert the geometry of the measure: here we used the pad positioned on the spec-
(b) imen
to invert the geometry
(see Figure 9). of the measure: here we used the pad positioned on the
specimen (see Figure 9).
Measures were carried out on a set of 5 specimens per lithotype, sized 5 × 5 × 2 cm.
Measures were
Specimens werecarried
preparedout byon polishing
a set of 5 specimens
one of theper lithotype,
two 5 × 55 ×
sizedsized
major faces × 25 cm.
cm
Specimens were prepared by polishing one of the two major faces sized
(Struers polishing equipment with the following parameters: 60 s, 5 N force, 150 rpm with 5 × 5 cm
(Struers polishing equipment with the following parameters: 60 s, 5 N
water, P32021 FEPA abrasive paper). Then, specimens were washed, brushed, immersed force, 150 rpm with
water,
in P32021water
deionized FEPAfor abrasive
30 min,paper).
and putThen, specimens
in an oven (60 were
°C forwashed, brushed,
24 h) for immersed
desiccation until
in deionized water weight. ◦
for 30 min, and put in an oven (60 C for 24 h) for desiccation until
reaching a constant
reaching a constant
Measures weight.
obtained from 5 specimens were used calculate the average weight and the
Measures obtained
standard deviation. from 5 specimens were used calculate the average weight and the
standard deviation.
4.3.2. Evanescent Field Dielectrometry (EFD)
4.3.2. Evanescent Field Dielectrometry (EFD)
In principle, the EFD technique uses the difference in the measures between the die-
In principle, the EFD technique uses the difference in the measures between the
lectric contrast of the water and the penetrated host material in the microwave frequency
dielectric contrast of the water and the penetrated host material in the microwave frequency
spectrum; the instrument measures the permittivity (dielectric constant and loss factor)
spectrum; the instrument measures the permittivity (dielectric constant and loss factor) by
by a resonant method [39]. The EFD system used here (called SUSI©) consists of a resonant
a resonant method [39]. The EFD system used here (called SUSI©) consists of a resonant
probe operating in the microwave frequency range (1–1.5 GHz) of a scalar network ana-
probe operating in the microwave frequency range (1–1.5 GHz) of a scalar network analyzer
lyzer (SNA) for measuring the probe frequency response and a laptop that drives the de-
(SNA) for measuring the probe frequency response and a laptop that drives the device by
vice by dedicated
dedicated software software and provides
and provides the diagnostics
the diagnostics results (humidity
results (humidity contentcontent and
and salinity
salinity index) from the electrical parameters (Figure 11a). Water has a dielectric
index) from the electrical parameters (Figure 11a). Water has a dielectric constant value constant
value of about
of about 80; stone
80; stone materials
materials usedused in architecture
in architecture (bricks,
(bricks, plaster,
plaster, mortar,
mortar, stone)have
stone) havea
adielectric
dielectricconstant
constantofofabout
about3–5;
3–5; this
this strong
strong difference
difference in in values
values generates
generates aa dielectric
dielectric
contrast
contrast between the water and the material, which is used to carry outaameasurement
between the water and the material, which is used to carry out measurementof of
the
themoisture
moisturecontent
contentand
andthe
thesoluble
solublesalts
saltscontained
containedin inthe
thematerial
materialup uptotoaadepth
depthof of22cm
cm
from
from the
the surface.
surface.

(a) (b)

Figure 11. (a) SUSI© system consisting of: scalar network analyzer (SNA), notebook (NB), and EFD-P
probe; (b) the solid line is the frequency response for the unloaded resonant probe, the dotted line
is the behavior for the probe in contact with the material under investigation (humidity and salt
quantity not specified).

The electrical parameters of interest of the resonant probe response are the resonance
frequency fres and the quality factor Q (or resonance bandwidth Lw) shown in Figure 11b.
Gels 2024, 10, 708 17 of 19

∆fr corresponds to the resonant frequency shift of the probe between the loaded behavior,
when in contact with the material under test (fres), and for an unloaded probe (in air) (fo).
These resonant parameters are related to the diagnostics parameters moisture content (MC)
and salinity index (SI) [40–42]. In the first instance, the frequency shift is related to the MC
while the spreading of the bandwidth is due to the combination of the dielectric losses and
conductivity losses in the microwave range, both related to the presence of electrolytes
in solution (i.e., presence of salts). The MC of the examined material may be related to
∆fr, according to the simplified equation shown in references [40–42], where a calibration
procedure is presented for the determination of the correlation constants between electrical
parameters and diagnostics parameters (MC and SI). The humidity content that can be
investigated is in the range 0% to 20% (on a dry basis), and for the salinity index it is from
1 to 10. The SI is a semi-empirical parameter, determined by an experimentation on plaster
samples with different salinity to obtain a calibration curve [43].

4.3.3. Portable Nuclear Magnetic Resonance

Gels obtained from different commercial agar powders show differences in porosity.
An investigation conducted using portable nuclear magnetic resonance (NMR) illustrates
the mobility of water within the polysaccharide network of agar gels and demonstrates
that the mobility of water molecules in the pores strongly depends on the concentration of
polysaccharide [37]. As the double helices and their interconnections increase, the mobility
and the values of free water in the gel decrease.
Additionally, another key factor in the absorption, distribution, and penetration of
water into the stone depends not only on the physico-chemical properties of the gel but is
also influenced by the porosimetric characteristics of the porous material itself. ¹H NMR
depth profiles are a powerful tool for assessing water absorption in porous materials.
This technique involves the use of nuclear magnetic resonance (NMR) to measure the
presence and distribution of hydrogen atoms (¹H) within a sample, which is indicative
of water content [44]. In this study, 1 H NMR depth profiles were acquired to examine
the amount of water released into three different stones with varying porous structures,
as well as the distribution and penetration depth of water absorbed by the stones. All
measurements were carried out 30 min after the application of gels at 13.62 MHz using a
portable NMR device from Bruker Biospin connected to a single-sided sensor developed
by RWTH Aachen University, Germany [45]. This sensor was mounted on a lift that moved
relative to the sample, allowing for precise micrometric positioning of the excited slice,
enabling us to acquire 1 H depth profiles. The intensity of the 1 H depth profiles recorded
on stone samples after the application of gels was determined by summing the first four
echoes collected using a Carr–Purcell–Meiboom–Gill (CPMG) pulse sequence (2t of 57 µs,
nominal resolution of 0.06 mm, with 0.2 mm steps). All profiles were acquired by scanning
the specimen to a maximum depth of 5 mm.

Author Contributions: Conceptualization, A.S.; methodology, A.S., N.P., V.D.T. and C.R.; validation,
A.S., B.S., C.C. and R.N.; investigation, V.D.T., N.P., C.R. and S.M.; data curation, A.S., R.N. and
B.S.; writing—original draft preparation, A.S. and C.C.; writing—review and editing, A.S., C.C.,
V.D.T., N.P. and C.R.; supervision, A.S. All authors have read and agreed to the published version of
the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Acknowledgments: The authors wish to acknowledge the Erasmus + program of the European
Commission for supporting the Student Mobility for Traineeship of Sonia Mironiouk Dobrovolskaia.
Gels 2024, 10, 708 18 of 19

Conflicts of Interest: The authors declare no conflict of interest.

References
1. Maxwell, I. Stone cleaning: For better or worse? An overview. In Stone Cleaning and the Nature, Soiling and Decay Mechanisms of
Stone, Proceedings of the International Conference, Edinburgh, UK, 14–16 April 1992; Webster Robin, G.M., Ed.; Routledge: London,
UK, 1992; pp. 3–49.
2. Maravelaki, P.N. Surface Cleaning: Implications from Choices & Future Perspectives. In Conserving Stone Heritage; Gherardi, F.,
Maravelaki, P.N., Eds.; Cultural Heritage Science; Springer: Berlin/Heidelberg, Germany, 2022.
3. Baglioni, M.; Poggi, G.; Chelazzi, D.; Baglioni, P. Advanced Materials in Cultural Heritage Conservation. Molecules 2021, 26, 3967.
[CrossRef] [PubMed]
4. Siegesmund, S.; Snethlage, R. Stone in Architecture. Properties and Durability; Springer: Berlin/Heidelberg, Germany, 2011.
5. Pianorsi, M.D.; Raudino, M.; Bonelli, N.; Chelazzi, D.; Giorgi, R.; Fratini, E.; Baglioni, P. Organogels for the cleaning of artifacts.
Pure Appl. Chem. 2017, 89, 3–17. [CrossRef]
6. Chelazzi, D.; Fratini, E.; Giorgi, R.; Mastrangelo, R.; Rossi, M.; Baglioni, P. Gels for the Cleaning of Works of Art. In Gels and Other
Soft Amorphous Solids; American Chemical Society: Washington, DC, USA, 2018; pp. 291–314.
7. Canevali, C.; Sansonetti, A.; Rampazzi, L.; Monticelli, D.; D’Arienzo, M.; Di Credico, B.; Ghezzi, E.; Mostoni, S.; Nisticò, R.;
Scotti, R. The Chemistry of Chelation for Built Heritage Cleaning: The Removal of Copper and Iron Stains. ChemPlusChem 2024,
89, e202300709. [CrossRef]
8. Giuliani, L.; Genova, C.; Stagno, V.; Paoletti, L.; Matulac, A.L.; Ciccola, A.; Favero, G. Multi-Technique Assessment of Chelators-
Loaded PVA-Borax Gel-like Systems Performance in Cleaning of Stone Contaminated with Copper Corrosion Products. Gels 2024,
10, 455. [CrossRef] [PubMed]
9. Morlotti, M.; Forlani, F.; Saccani, I.; Sansonetti, A. Evaluation of Enzyme Agarose Gels for Cleaning Complex Substrates in
Cultural Heritage. Gels 2023, 10, 14. [CrossRef]
10. Mazzuca, C.; Poggi, G.; Bonelli, N.; Micheli, L.; Baglioni, P.; Palleschi, A. Innovative chemical gels meet enzymes: A smart
combination for cleaning paper artworks. J. Colloid Interface Sci. 2017, 502, 153–164. [CrossRef] [PubMed]
11. Baglioni, M.; Giorgi, R.; Berti, D.; Baglioni, P. Smart cleaning of cultural heritage: A new challenge for soft nanoscience. Nanoscale
2012, 4, 42–53. [CrossRef]
12. Bonelli, N.; Chelazzi, D.; Baglioni, M.; Giorgi, R.; Baglioni, P. Confined Aqueous Media for the Cleaning of Cultural Heritage:
Innovative Gels and Amphiphile-Based Nanofluids. In Nanoscience and Cultural Heritage; Dillmann, P., Bellot-Gurlet, L., Nenner,
I., Eds.; Atlantis Press: Paris, France, 2016.
13. Domingues, J.A.; Bonelli, N.; Giorgi, R.; Fratini, E.; Gorel, F.; Baglioni, P. Innovative hydrogels based on semi-interpenetrating p
(HEMA)/PVP networks for the cleaning of water-sensitive cultural heritage artifacts. Langmuir 2013, 29, 2746–2755. [CrossRef]
14. Ludwig, N.; Rosina, E.; Sansonetti, A. Evaluation and monitoring of water diffusion into stone porous materials by means of
innovative IR thermography techniques. Measurement 2018, 118, 348–353. [CrossRef]
15. Vergès-Belmin, V.; Heritage, A.; Bourgès. A. Powdered Cellulose Poultices in Stone and Wall Painting Conservation Myths and
Realities. Stud. Conserv. 2011, 56, 281–297. [CrossRef]
16. Baglioni, P.; Berti, D.; Bonini, M.; Carretti, E.; Dei, L.; Fratini, E.; Giorgi, R. Micelle, microemulsions, and gels for the conservation
of cultural heritage. Adv. Colloid Interface Sci. 2014, 205, 361–371. [CrossRef]
17. Sansonetti, A.; Bertasa, M.; Canevali, C.; Rabbolini, A.; Anzani, M.; Scalarone, D. A review in using agar gels for cleaning art
surfaces. J. Cult. Herit. 2020, 44, 285–296. [CrossRef]
18. Giordano, A.; Cremonesi, P. New methods of applying rigid agar gels: From tiny to large-scale surface areas. Stud. Conserv. 2021,
66, 437–448. [CrossRef]
19. Bertasa, M.; Poli, T.; Riedo, C.; Di Tullio, V.; Capitani, D.; Proietti, N.; Canevali, C.; Sansonetti, A.; Scalarone, D. A study of
non-bounded/bounded water and water mobility in different agar gels. Microchem. J. 2018, 139, 306–314. [CrossRef]
20. Armisen, R.; Gaiatas, F. Agar. In Handbook of Hydrocolloids; Woodhead Publishing: Sawston, UK, 2009; pp. 82–107.
21. Lee, W.K.; Lim, Y.Y.; Leow, A.T.C.; Namasivayam, P.; Abdullah, J.O.; Ho, C.L. Factors affecting yield and gelling properties of
agar. J. Appl. Phycol. 2017, 29, 1527–1540. [CrossRef]
22. Carretti, E.; Dei, L.; Weiss, R.G.; Baglioni, P. A new class of gels for the conservation of painted surfaces. J. Cult. Herit. 2008,
9, 386–393. [CrossRef]
23. Baglioni, P.; Chelazzi, D.; Giorgi, R.; Baglioni, P.; Chelazzi, D.; Giorgi, R. Cleaning of wall paintings and stones. In Nanotechnologies
in the Conservation of Cultural Heritage: A Compendium of Materials and Techniques; Springer: Dordrecht, The Netherlands, 2015;
pp. 61–82.
24. Cremonesi, P.; Casoli, A. Thermo-reversible rigid agar hydrogels: Their properties and action in cleaning. In Gels in the Conservation
of Art; Archetype Publications Ltd.: London, UK, 2017; pp. 19–23.
25. Rossi-Doria, P. Porous Materials: Structure and Properties. In Encyclopedia of Materials Science and Engineering; Bever, M.B., Ed.;
MIT Press: Oxford, UK, 1986; Volume 5, pp. 3839–3847.
26. Everett, D.H. IUPAC Manual of Symbols and Terminology, Appendice 2, Pt. I, Colloid and Surface Chemistry. Pure Appl. Chem.
1972, 31, 587.
Gels 2024, 10, 708 19 of 19

27. Calia, A.; Lettieri, M.; Mecchi, A.; Quarta, G. The role of the petrophysical characteristics on the durability and conservation of
some porous calcarenites from Southern Italy. Spec. Publ. 2016, 416, 183–201. [CrossRef]
28. Rescic, S.; Fratini, F.; Sacchi, B.; Mattone, M. The Manciano Sandstone in Contemporary Architecture: Conservation Aspects in
Florence. Geoheritage 2024, 16, 114. [CrossRef]
29. Winkler, E.M. Stone in Architecture, Properties, Durability; Springer: Berlin/Heidelberg, Germany, 1997; p. 34.
30. Bertasa, M.; Chiantore, O.; Poli, T.; Riedo, C.; Di Tullio, V.; Canevali, C.; Sansonetti, A.; Scalarone, D. A study of commercial agar
gels as cleaning materials. In Gels in the Conservation of Art; Archetype Publications Ltd.: London, UK, 2017; pp. 11–18.
31. Bertasa, M.; Canevali, C.; Sansonetti, A.; Lazzari, M.; Malandrino, M.; Simonutti, R.; Scalarone, D. An in-depth study on the agar
gel effectiveness for built heritage cleaning. J. Cult. Herit. 2021, 47, 12–20. [CrossRef]
32. Marinoni, N.; Pavese, A.; Bugini, R.; Di Silvestro, G. Black limestone used in Lombard architecture. J. Cult. Herit. 2002, 3, 241–249.
[CrossRef]
33. Vergès-Belmin, V.; Heritage, A.; Bourgès, A. Powdered cellulose poultices in stone and wall conservation. In Desalinisation of
Historic Buildings, Stone and Wall Paintings; Archetype Publications Ltd.: London, UK, 2013; p. 63.
34. Alvarez, A. Sepiolite: Properties and use. Dev. Sedimentol. 1984, 37, 253–287.
35. EN 17138; Conservation of Cultural Heritage—Methods and Materials for Cleaning Porous Inorganic Materials. iTeh Inc.:
Newark, DE, USA, 2018.
36. Bertasa, M.; Bandini, F.; Felici, A.; Lanfranchi, M.R.; Negrotti, R.; Riminesi, C.; Scalarone, D.; Sansonetti, A. Soluble salts extraction
with different thickeners: Monitoring of the effects on plaster. In Proceedings of the IOP Conference Series: Materials Science and
Engineering, Kuala Lumpur, Malaysia, 13–14 August 2018; Volume 364, p. 012076.
37. Bertasa, M.; Botteon, A.; Brambilla, L.; Riedo, C.; Chiantore, O.; Poli, T.; Sansonetti, A.; Scalarone, D. Cleaning materials: A
compositional multi-analytical characterization of commercial agar powders. J. Anal. Appl. Pyrolysis 2017, 100, 310–317. [CrossRef]
38. EN 15801; Conservation of Cultural Properties—Test Methods—Determination of Water Absorption by Capillarity. iTeh Inc.:
Newark, DE, USA, 2009.
39. Olmi, R.; Priori, S.; Capitani, D.; Proietti, N.; Capineri, L.; Falorni, P.; Riminesi, C. Innovative techniques for sub-surface
investigations. Mater. Eval. 2011, 69, 89–96.
40. Olmi, R.; Riminesi, C. Study of water mass transfer dynamics in frescoes by dielectric spectroscopy. Nuovo C.-Soc. Ital. Di Fis. Sez.
C 2008, 31, 389–402.
41. Riminesi, C.; Olmi, R. Diagnostics and monitoring of moisture and salt in porous materials by evanescent field dielectrometry. In
Proceedings of the SWBSS 2017 Fourth International Conference on Salt Weathering of Buildings and Stone Sculptures, Potsdam,
Germany, 20–22 September 2017; pp. 49–55.
42. Tiano, P.; Riminesi, C. State of Arts of Monumental Stones Diagnosis and Monitoring. In Proceedings of the International Archives
of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 26th International CIPA Symposium, Ottawa, ON,
Canada, 28 August–1 September 2017; Volume XLII-2/W5.
43. Olmi, R.; Bini, M.; Ignesti, A.; Priori, S.; Riminesi, C.; Felici, A. Diagnostics and monitoring of frescoes using evanescent-field
dielectrometry. Meas. Sci. Technol. 2006, 17, 2281. [CrossRef]
44. Di Tullio, V.; Capitani, D.; Proietti, N. Unilateral NMR to study water diffusion and absorption in stone-hydrogel systems.
Microporous Mesoporous Mater 2018, 269, 180–185. [CrossRef]
45. Perlo, J.; Casanova, F.; Blümich, B. Profiles with microscopic resolution by single-sided NMR. J. Magn. Reson. 2005, 176, 64–70.
[CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.