This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/authorsrights
Transcript
This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
Works of art degrade due to the natural ageing of the materialscomposing them. In most cases, degradation processes occur at thesurface of the artefacts that is not only the locus where the artiststransferred their message and emotions, but also the place wheredifferent materials, with their own specific chemical composition andmechanical properties, coexist. If we think to awork of art in a “materialway” it is obvious to consider colloids and surface chemistry as thecorrect scientific framework to be used to understand and possibly todelay the ageing processes that depend on the particular artefact's
location and exposure to the environmental factors such as theexposure to light, temperature stresses, humidity cycles, insects andmicroorganisms.
However, for long time Colloids and Surface Science wasn't themajorplayer in the conservation arena and “classical” analytical and polymerchemistries were the privileged tools at the hands of conservators.
Several authors have described the principles and the approach topreserve cultural heritage, and also accounted for the efforts (and faults)in the restoration of artworks [1–4]. For example the use of polymers,such as painting consolidants and varnishes, that was and, unfortunately,is still very popular in the conservator community, produced severedegradation with detrimental effects [5] on the works of art surface thatultimately led to the disfiguration and loss of the objects.
During the years many systems belonging to the realm of colloidshave been specifically tailored for conservation issues, as nanoparticulate
Advances in Colloid and Interface Science 205 (2014) 361–371
j ourna l homepage: www.e lsev ie r .com/ locate /c i s
Author's personal copy
inorganic sols (nanosols) [6], colloidal silica [7], and alkoxysilane [8] playan important role in wall paintings, stone, paper, andwood conservation.Some of the most important advancements in the field have beenreviewed and the reader is referred to references [9–13].
In this review we will report on the application of micelles,microemulsions, and gels for the restoration of works of art. Theformulation of these systems would not be possible without thescientific advancements generated in the past forty years by manyscientists and in particular by Bjorn Lindman, to whom this paper isdedicated.
Complex fluids such as micelles and microemulsions are the mostadvanced systems used so far in the conservation field for their capabilityto remove soil coatings fromworks of art surfaces. In particular, swelling,solubilisation and selective removal of synthetic materials (acrylic andvinyl polymers), largely applied in past restorations and difficult to beremoved by classical cleaning methods, can be achieved by usingamphiphile-based systems.
The cleaning of wall and easel paintings presents several difficultiesdue to the physico-chemical properties of the substrate, which usuallyhas a very complex stratigraphic structure, both in terms of porosityand chemical composition. Therefore, the removal of soil, dirt/grime,and altered materials requires very high selectivity, with minimalinteractions with the layers beneath the dirt and coatings, i.e. the paintedlayer. For this reason the classical solvent technology, i.e. the use of pure(or mixed) organic liquids, is often unadvisable. In fact, the action oforganic solvents is usually scarcely controllable due to their surfacetension and to the common high wettability of the treated surface,leading to the solubilisation of soil material and to its spreading withinthe substrate porosity [14,15]. These processes are enhanced by thehigh evaporation rate of most of the solvents used for cleaning. Anadditional important issue associated with the use of neat organicsolvents is represented by the toxicity of most solvents used for cleaning.
In the cleaning, selectiveness is mandatory: the basic principle ‘likedissolves like’ implies that the removal of a soiling layer is very difficultsince it usually possesses physico-chemical properties similar to thesurface substrate. The use of solvents, in fact, may cause the partialswelling and solubilisation of the original artwork materials. The useof blend of different solvents could perfectly fit the solubility parametersof thematerials to be removed, but rarely the substrate results completelyinert.
An important improvement to “classical” cleaning procedures wasintroduced in the conservation field with the formulation of micro-emulsions and micellar solutions that, for specific applications, can beconfined into host systems like physical and chemical gels. The mostimportant systems used so far in conservation will be highlighted in thefollowing sections.
2. Microemulsions and micellar solutions as innovative low impact
cleaning tools for the conservation of wall paintings
2.1. Microemulsions
Lindman and Danielsson provided a useful definition of micro-emulsions described as “liquid, stable and homogeneous, opticallytransparent, isotropic and “spontaneously” formed systems, comprisingtwo liquids mutually insoluble; one dispersed in the other in formof micro-spheres stabilized by at least a monolayer of amphiphilicmolecules (surfactants)” [16]. The use of microemulsions in conservationdates back to the eighties and since then they are employed worldwide.Microemulsions are very versatile systems showing several advantagesin the field of artwork cleaning compared to conventional systems suchas neat solvents and solvent gels used by restorers:
• The continuous phase can be hydrophilic (o/w) or hydrophobic (w/o)allowing a control in the spreading of the continuous phase into theartefacts to be treated.
• The dispersed oil-in-water (o/w) or water-in-oil (w/o) nanodropletswith respect to simple emulsions develop a huge exchange surfacearea that enhances the interactions with soiling materials, facilitatingthe removal or the swelling of the materials to be removed.
• The spreading of the solubilised material into the porous matrixesmay be limited, because solubilisation or swelling occurs into thecore of nanodroplets and/or at the droplet interface. When dealingwithhydrophilic substrates (i.e.wall paintings) the aqueous continuousphase may act as a barrier, preventing the re-deposition of thehydrophobic coatings within the substrate porosity.
• Microemulsions are thermodynamically stable systems.• The formulation of o/w microemulsions requires small amounts ofsolvents with a consistent reduction of the toxicity and environmentalimpact.
• The cleaning process withmicroemulsions allows a controlled cleaningaction of the works of art surface.
The restoration of the Renaissance paintings by Masaccio, Masolino,and Lippi in the Brancacci Chapel in Florence (1984–1990) [13,17]represents the first case study where microemulsions were usedfor conservation purposes. Diagnostics on the wall paintings revealedthe presence of a large amount of wax-spots deposited over the surface.This unusual event was due to the blowing out of votive candles, keptclose to the paintings over centuries. The removal of this materialrequired the action of an apolar solvent to be applied over a hydrophilicsubstrate. The use of hydrocarbons (i.e. dodecane) allowed the solubi-lisation of wax, but the resulting solution was soaked by the wall and,after solvent evaporation, the wax was re-deposited within the pores.An oil-in-water microemulsion containing SDS/dodecane/n-butanol/H2Owas very effective in the removal of the apolar material. The positiveoutcome of this conservation paved the way for the use of micro-emulsions that become a standard method for cleaning.
Several microemulsive systems have been developed to solveconservation issues not manageable with conventional conservationmethods. Examples are reported in the following paragraphs.
In 2007 some wall painting decorations in the Oratory of San Nicola
al Ceppo, devastated by the 1966 flood of the Arno River in Florence,were restored by removing the patinas left after the flood [18]. Thepainted surfaces were covered by a crust of gypsum efflorescencesmixed with the residuals of oil fuel from the flooding water. Sixtyyears after the event, the ageing of this blend of hydrocarbons andothers materials resulted in an extensive cross-linking of the organicmaterials and further insolubilisation of the hydrophobic layer to mostorganic solvents. In this case, it was necessary to combine the actionof a solvent capable to swell and partially solubilise the coating and, atthe same time, to chemically attack the gypsum patina. This twofoldpurpose was successfully achieved by using an oil-in-water micro-emulsion based on 1% w/w xylene (for the swelling of the organiccomponents) dispersed in a ammonium carbonate solution (for thesolubilisation of gypsum) used as a continuous phase.
However, the most common use of microemulsions is representedby the removal of synthetic polymers largely used in the past to provideconsolidation and protection to the works of art surfaces [19]. In fact,acrylic and vinyl polymers (or co-polymers) become, upon naturalageing, especially in urban polluted environments, insoluble or hardlysoluble in solvents or solvent blends, mainly because of oxidation andcross-linking reactions [3,20–22]. Together with molecular changesdue to ageing, synthetic polymers concur to the degradation of wallpaintings because they change the physico-chemical properties of thesurfaces (i.e. vapour permeability), dramatically increasing the effects ofsalt crystallization over the painted surface (see Fig. 1) [23]. Wallpaintings from 16th century in San Salvador Church in Venice wererestored in 1970 by applying a nitro diluent solution of 10%w/w solutionof acrylic copolymer (Paraloid®; ethyl methacrylate–methyl acrylate70:30 w/w co-polymer). At the end of 2002, a new restoration wasnecessary to remove the compact, yellowed, and shiny coating that
362 P. Baglioni et al. / Advances in Colloid and Interface Science 205 (2014) 361–371
Author's personal copy
resulted upon thirty year ageing [18]. The ageing process strongly alteredthe appearance of the mural decorations. Together with aesthetic issuespaintings were seriously affected by the presence of salts, mainlysulphates and chlorides that are largely diffused on the architectonicsurfaces in the Venice lagoon area. The application of an oil-in-watermicroemulsion containing SDS/nitro diluent/1-pentanol/H2O [18]provided the complete removal of the coating, allowing the rescue ofthe original appearance.
Microemulsions specifically tailored to remove acrylic polymershave also been used to remove unwanted “graffiti” (i.e. paintings,writings, or drawings) from walls, stones and monuments made byusing common spray varnishes [24]. Traditionally, nitro diluent, chloroderivatives, acetone, or other organic solvents, either pure or inmixtureare used. As already described, these methods show some limitationsbecause of their toxicity and the spreading of the acrylic varnish intothe porous substrate that follows the solubilisation of the soilingmaterials. The same amphiphile-based system employed for paraloidremoval was successfully used to remove the graffiti from 18th centurysecco paintings conserved in the 16th centurymonumental villa, knownas Villa del Barone in Montemurlo (Prato, Italy).
The first microemulsions developed, and very often used inrestoration, are based on the use of sodium dodecyl sulphate (SDS)surfactant. In principle, the use of this anionic surfactant may presentsome limitations due to the high tendency to foam and to the relativelyhigh critical micellar concentration, (c.m.c., about 8.3 · 10−3 M) whencompared to the non-ionic surfactants, requiring a large amount ofsurfactant necessary to produce the microemulsion. The use of a lower
amount of surfactants may reduce the risk of leaving some residueswithin the porosity of the paint layer if paintings are not properly cleaned,for example by using cellulose poultice (see Fig. 2).
The optimal would be the use of surfactants that degrade withoutleaving residues on the treated surface. Several biodegradablesurfactants such as polyalkylglycosides have been investigated [18].Polyalkylglycosides are non-ionic surfactants produced by renewablesources, with a good ecotoxicological profile and high biodegradability.A microemulsion formulation based on a blend of AGE and AGESS,respectively a non-ionic and an anionic surfactants belonging to thisclass, with xylene as oil phase (water is 99% of the formulation) wasformulated and applied to effectively remove acrylics coatings from thewall paintings of Vecchietta in Santa Maria della Scala in Siena (Italy).
2.2. Micellar solutions
Xylene/nitro diluent-based o/w microemulsions, described above,were also tested, during the restoration of the façade of the Coneglianocathedral (Italy), to remove vinyl coatings [19]. This formulation, effectivefor the removal of acrylics, resulted not efficient for vinyl polymers due tothe too low xylene's polarity. By adding propylene carbonate up to a 20%w/w in a micellar solution of SDS and 1-pentanol in water, the completeremoval of the coatings was achieved. Interestingly, this formulation wasconstituted by a component soluble in water (up to 20% w/w) added tothe micellar solution in a concentration close to its solubility limit inwater. Pulsed gradient spin-echo NMR and SAXS investigations on thissystem showed that propylene carbonate acts partially as cosurfactant,
Fig. 1. SEMmicrographs of the sample reproducing awall paintedwith the fresco technique. SEM images show the appearance of awall paintingmodel sample before (a) and after (b) the
application of a Paraloid B72 coating.
Adapted from Ref. [24] with permission of Royal Society of Chemistry.
Fig. 2. The picture on the left illustrates the application procedure of an o/w microemulsion system. A cellulose poultice impregnated with microemulsion is applied over the painting
surface, previously protected with a Japanese paper tissue. On the right, the removal of the poultice, after the cleaning procedure, is shown.
Reproduced from Ref. [25] with permission from The Royal Society of Chemistry.
363P. Baglioni et al. / Advances in Colloid and Interface Science 205 (2014) 361–371
Author's personal copy
such as 1-pentanol, for the SDS micelles decreasing their size andaggregation number by increasing the mean head group area ofSDS [25,26]. Therefore micellar solutions, swollen with some organicmolecules, were for the first time successfully used for the cleaningof artwork surfaces.
Later, in the framework of the restorations performed in thearchaeological area of Mayapan in Mexico, some new formulationswere tested in order to remove the largely used copolymer knownas Mowilith DM5, a copolymer of vinyl acetate and n-butyl acrylate(65:35 w/w ratio) [5,27,28]. To this purpose ethyl acetate was added tothe propylene carbonate-based formulation described above, since it isa good solvent for the swelling and solubilisation of Mowilith familypolymers. The final formulation, known as EAPC, provided a completeswelling and removal of the coatings applied in the past restorations onthe Mayapan paintings.
EAPC systemhas also been successfully used onwall-paintings in thearchaeological area of Nazareth (Israel) to remove polysiloxane resincoatings, applied in the seventies (see Fig. 3), and can be consideredas one of the most versatile systems for polymer removal so farformulated. Small angle neutron scattering (SANS) measurements,performed to investigate the structure and to study the actionmechanism of this system, showed that ethylene acetate, EA, andpropylene carbonate, PC (the co-solvents), were partitioned betweenboth the disperse droplets and the continuous aqueous phase [29]. Theterm “swollen micelles” seems more appropriated to describe the EAPC
system, which represents a sort of borderline structure betweenmicellarsolutions and classic microemulsions.
EAPC system is more efficient and versatile than the “classical”xylene-based microemulsion system used in several conservationworkshops; probably the structural differences of the two systemsplay a role in the cleaning process. As a matter of fact, the interactionsbetween polymer coatings and cleaning fluid are enhanced. It washypothesized that EAPC system follows a mechanism where severalprocesses play concomitantly: i) the solvents dissolved in thecontinuous aqueous phase (PC and EA), that are in equilibrium withthe dispersed droplets, quickly interact with the polymer coating, ii)solventmigration occurs from the nanodroplets (that can be consideredas a dynamic nanocontainers) to the aqueous phase, and iii) furthermigration occurs from nanodroplets to the polymer. As a finalconsequence, the polymer coating “selects” an optimal compositionthat favours the chain disentanglement, the polymer layer swellingand the final detaching from the substrate. At the end of the process,the nanodroplets get smaller and re-organize their structure because ofthe outflow of the selected solvents (see Fig. 4) [28].
This description presents some elements that make this systemdifferent fromclassical binarymicellar solutions that proceed to completeremoval of oily soils through the typical detergency mechanisms ofrolling-up, emulsification and solubilisation [30].
It is also important to remark the application of simple surfactant-based solutions able to interact with oily soils forming microemulsions[31–34]. This is the case of the Brij solutions in nonane that have beenrecently shown as highly effective in the ‘emulsification’ of acrylicadhesives used for the relining of canvas paintings [35].
3. Gels: A cleaning tool for the restoration of paints
Complex systems such asmicelles andmicroemulsions are so far themost effective systems for the cleaning ofworks of art. However, the useof neat solvents is still quite diffuse in conservation. In the followingchapters we will report on the use of neat or mixture solvents confinedin physical and chemical gels. Microemulsion and micellar systems canbe confined as well to control and improve their cleaning action.
Unfortunately, as indicated by Michalski [36], many organicsolvents, pure or in mixtures, able in the solubilisation of surface dirt,mainly driven by capillary forces, can penetrate rapidly into the paintlayer (the average rate is 10mm/s) leading to swelling, softening andleaching ofmost of the bindingmedia commonly constituting it. Linseedoil is one of the essential fatty acids commonly used in the past as abinding medium for easel paintings and it can be used as a modelsubstrate. Based on the volume changes in aged and not aged linseedoil films induced by solvents having different polarities, it is possibleto predict the swelling effects on a painted film in terms of theHildebrand solubility parameters [37].
One of the main consequences of the swelling is that, once theresidual solvent evaporates, all the swelledmolecules reach a new spatialconformation, different from the original one. A meaningful decrease ofthe volume of the paint layer is observed and a not controllable loss ofthe strength of the paint layer's structure, that usually appears muchmore brittle and denser, occurs. A further effect is that it is very difficultto maintain the total control of the mechanical action usually necessaryto remove all the materials not appertaining to the original work of artfrom the swelled and softened paint layer. On the other hand if the filmis directly exposed a second time to the same solvent, the swelling effectis limited and no additional meaningful shrinkage or leaching is noted.
The technological approach purposed at the end of the eighties bythe community of the conservation scientists [38] allows to minimizeor completely avoid these drawbacks, providing for the use of highlyviscous systems like gels or gel-like materials as follows.
1. The slow release of the liquid phase at the interface between the gelor the gel-like system and the paint decreases the risk of swelling of
Fig. 3. Application of the EAPC system onwall paintings from the Annunciation Basilica in
Nazareth (Israel). Top: Before restoration. Bottom: After restoration. In the dashed box, an
area is highlighted where the polymer coating has been left untreated as a reference for
the evaluation of the cleaning result.
Reprinted with permission from Langmuir, 28, 15193–15202. Copyright 2012 American
Chemical Society.
364 P. Baglioni et al. / Advances in Colloid and Interface Science 205 (2014) 361–371
Author's personal copy
the paint layer. Furthermore, as indicated by the Lucas–Washburnequation (Eq. (1)), [39] which describes the penetration of a liquidinto a horizontal cylindrical capillary, the increase of the viscosity ofthe cleaning tool lowers the rate of its penetration into the porousmatrix:
L2¼
γDt
4ηcosθ ð1Þ
where: t is the time of the fluid to penetrate a distance L in the pore, γis the surface tension of the fluid, θ the contact angle between theliquid and the capillary surface, D is the diameter of the genericpore in which the fluid is penetrating and η is the viscosity of thefluid. For example, the embedding of a water solution (viscosity ofthe order of 0.9–1 mPa·s) into a 3D network of polyvinyl-acetate,80% hydrolysed, cross-linked by borate (viscosity is about 400Pa·s)allows the reduction of the penetration depth of more than threeorders of magnitude, drastically decreasing all the negative effectspreviously described.
2. The high viscosity reduces the diffusion of the solubilised moleculesinto the cleaning tool causing on one hand a decrease of thesolubilisation rate, but on the other allowing a perfect control of theinteraction between the viscous tool and the support [40]. Both thesefeatures increase the selectivity of the solvent action that canbe strictlylimited to the interface between the work of art and the cleaning tool.An additional consequence is the decrease of the evaporation rate ofthe solvents confined into the 3D matrix that further enhances thecontrol of the cleaning action and the reduction of the solvent toxicityon the health of restorers.
3. The diverse chemical nature of the materials to be removed makesnecessary the use of versatile highly viscous systems able to embeddifferent liquid media like organic solvents, micellar solutions, o/wmicroemulsions [41] or aqueous solutions containing enzymes or
chelates [2]. The solvent gels [42] together with highly innovativesystems like chemical gels [43,44] or highly viscous polymericdispersions (HVPDs) [45], recently developed by us, satisfy thisrequirements.
4. Even if gels and gel-like systems are mainly used for the cleaning ofeasel paintings, they can also be successfully applied for the cleaningof wall paintings [46], glasses [47], metals [48] and feathers too [49].
The importance of this approach, in terms of both selectivity andcontrol of the cleaning action, drove our research group toward the in-depth exploration of this frontier of the applied sciences [50].
3.1. Traditional physical gels: Advantages and limits
Apart from wax emulsion and resin soaps [51], the most commonhighly viscous systems used in cultural heritage conservation for thecleaning of painted surfaces usually contain water soluble macro-molecules (i.e. the gelator) as polymers derived from natural products(i.e. cellulose ethers) and/or synthetic polymers (i.e. polyacrylic acid,Carbopols®).
The selection of an appropriate solvent should follow two “golden
rules”: 1) it should be chemically inert toward all the materials con-stituting the work of art to be cleaned; 2) it should have high physico-chemical stability to avoid undesired interactions between residues ofgellant and the cleaned surface.
For the so-called Wolbers gels, introduced at the beginning of thenineties [52], the gelator is a polyacrylic acid which concentrationusually ranges around 1 wt.% (the molecular weights is about4 × 106 Da). Its solubility in water is usually very low; in fact thecarboxylic groups, at low pH, are in the protonated form and interactwith each other through hydrogen bonds giving the polyacrylic acidmacromolecules a folded conformation. When the acidic groups ofthe polymer are neutralized through the addition of a base, they are
Fig. 4. Schematic representation of the interaction mechanism between the detergent nanostructured systems (top: EAPC; bottom: XYL) and the polymer coating.
Reproduced from Ref. [25] with permission of Royal Society of Chemistry.
365P. Baglioni et al. / Advances in Colloid and Interface Science 205 (2014) 361–371
Author's personal copy
converted into carboxylate anions. Then, the repulsive inter- and intra-molecular interactions between the negatively charged groups force theunfolding of the chains and a random coil conformation. In this way thesolubilisation is maximised and the formation of hydrogen bondingbetween different chains favoured, leading to the 3D physical networkthat constitutes the skeleton of the gel [42]. The de-protonation of thepolyacrylic acid molecules is usually achieved through the addition of aweakly basic non-ionic surfactant appertaining to the class of coco-amines (Ethomeen C12 or C25®), which polar heads interact throughelectrostatic attractive forces with the negatively charged carboxylategroups. The gelation capabilities are related to the chemical structure ofthe surfactants (length of the alkyl chains): in Ethomeen C12, used forthe preparation of gel based on low polarity solvents (HLB = 10), Rcontains a number of carbon atoms between 8 and 20 (molecular weight217–385 Da). Moreover, the higher HLB value of C25 (HLB = 19,molecular weight 789–957 Da) makes this surfactant useful in thegelation of highly polar solvents and water solutions. In otherformulations, cellulose derivatives (hydroxyl ethyl cellulose andcarboxy methyl cellulose) are used as gelators for aqueous-basedsystems like enzymes and EDTA solutions to be applied under astrict control of pH and temperature.
The applicative procedures for the use of all these physical gels aremainly two. The first one is the direct application of the gel onto thepaint surfacewithout any further operation (contact time usually rangingbetween 1 and 5min); the second is the application carried out bymeansof a swab roll that involves a soft and continuous mechanical action onthe surface of the paint to maximise the interaction with the surface dirt.
The main limits of physical gels technology and, in particular, ofsolvent gels, are related to the procedures to be adopted to achievethe complete removal of all the not volatile components (mainly thegelator) from the treated area, once they have carried out their function[53].
The presence of gelator rests, that may be left behind as residueseven if the recommended clearance procedures are followed, canpotentially cause unpredictable degradation phenomena due to uncon-trollable interactions with the support. These dangers include anincrease of the solubility of the paint or a chemical alteration of thepaint layer, usually resulting in an undesired acceleration of the workof art deterioration. As extensively reported in the literature, residualsof these systems are detected mainly in correspondence with holes,craquelets and in all the highly rough regions of the treated work ofart [53,54]. Unfortunately, at present the most effective method toobtain the complete removal of the gel is the one based on theapplication of the same solvent mixtures confined into the gel phase.As a consequence, part of the problems related to the application ofneat liquids that suggested the use of confining gels, is still unsolved.
Further physical hydrogel formulations based on two natural poly-saccharides like gellan gum and agar have been recently introduced tominimise the impact of thewater, during the cleaning of highly sensitiveporous supports like gypsum, paper and parchment documents [55]. Infact, both agar and gellan, mainly used as thickeners, emulsifiers, andstabilizers in the food industry, can form transparent rigid hydrogelsthat, once in contact with a porous matrix, can retain high amounts ofwater, avoiding their penetration into the support. However, the waterretention in these systems is not enough to treat water sensitive workof arts and the excessive wetting of the painted surface can result in anundesired colour leaching [56].
3.2. Innovative nanostructured gels or gel-like systems
3.2.1. Responsive gels: Rheoreversible gels and nanomagnetic sponges
In order to overcome all the limits related to the use of traditionalsolvent gels, our research group pioneered conceptually new approachesfor the cleaning of painted surfaces based on the use of “responsive” gels.From the point of view of the surface cleaning, the word “responsive”means that the chemical architecture of these formulations allows a
rapid, complete and not invasive removal via a response to “chemical”or to “physical” stimuli; these systems canbe referred to as rheoreversiblegels or nanomagnetic sponges respectively, vide infra.
It is well-known that polyamine (i.e. polyallilamine, PAA or poly-ethyleneimine, PEI) solutions can be easily converted into a gel [57]directly applicable onto a paint surface to be cleaned [58], simply bybubbling CO2. In fact, the polyallylammonium carbamate (PAA•CO2),obtained through electrostatic attractive interactions between thepolymer chains, allows the formation of a 3D polymeric structure ableto support the liquid phase (the most stable systems are the ones inwhich the gelled phase is an alkyl alcohol). Once the gel carried out itscleaning action it can be completely removed from the treated surfaceby adding few drops of a weak acidic solution. As the carbamate isconverted in polyallylammonium (PAA+) ions, the gel changes to aliquid state (Fig. 5A) that can be easily wiped away using a dry cottonswab. The weak acidic solution acts as a molecular switch that chargingthe PAA chains destroys the 3D network via inter-chain electrostaticrepulsions. Unfortunately, this technology did not have further improve-ments because, due to the intrinsic thixotropy of these systems, evenif their cleaning ability is excellent (Fig. 5B) the possibility of a correctmanipulation and control is difficult.
Another approach that allows a safe and non-invasive removal ofgels is the use of polyacrylamide based chemical gels doped with ferritemagnetic nanoparticles chemically linked to the polymer [43]. Whilethe gel retains the magnetic response of ferrite and the structuralproperties typical of pure acrylamide gels (nanoscaled mesh sizes,inhomogeneous domain sizes of a few tens of nanometers, and micro-metric pores; see Fig. 6), the nanoparticles act as entanglement sites,enhancing the elasticity of the system that results in an increase of theshear modulus G [44].
These gels can be loaded with aqueous systems (i.e. aqueous solu-tions, micellar systems, o/w microemulsions, vide infra) and once put incontact with the painted surface to be cleaned, they can be completelyremoved simply by means of a permanent magnet without leavinganalytically detectable residues onto the treated surface. Moreover,these systems can be shaped as desired (during the synthesis orafterwards by means of a cutter) allowing a complete spatial controlof the area to be cleaned.
3.2.2. Peelable systems: Polyvinyl alcohol based highly viscous polymeric
dispersions (HVPDs) and semi interpenetrated networks
A further possible approach based on the use of soft matter providesfor the use of gels or gel-like systems characterised by high intrinsicelasticity (quantitatively expressed by the shear modulus G); this featureallows their easy, complete and safe removal by a peeling action withoutleaving instrumentally detectable residues onto the treated surface(Fig. 7) [59]. Then, the impact of the cleaning action given by theamount of gel residues left onto the treated surface and the mechanicalaction needed for its complete removal is minimised.
The attention was focused on two different formulations whichmechanical properties warrantee their peelability: in the first casethe gellant is poly vinyl alcohol (PVA) cross-linked by borax, in thesecond case the gelation is achieved thanks to a semi-interpenetratednetwork of poly vinyl-1-pyrrolidone (PVP) and Poly(2-hydroxyethyl-methacrylate) (pHEMA).
It is well-known that poly(vinyl alcohol) (PVA) trough a di-diolcondensation reaction with borax can form completely transparenthighly viscous aqueous fluids [60–63] characterised by the presence ofa 3D network as indicated in Fig. 8.
Even if all the PVA/borate based systems suitable for application ascleaning tools for painted surfaces, macroscopically look like gels [agel is “a coherent system of at least two components, which exhibits
mechanical properties characteristic of a solid, where both the dispersed
component and the dispersion medium extend themselves continuously
through the whole system”] [64], from the rheological point of view,they cannot rigorously considered appertaining to this class of materials.
366 P. Baglioni et al. / Advances in Colloid and Interface Science 205 (2014) 361–371
Author's personal copy
Fig. 5. A: appearance (top) and dynamic viscosity (bottom) of a sample of polyallylamine solution 4 w/w% in 1-pentanol before (A1) and after CO2 bubbling that gives the formation of a
white gel (A2) [54]. After the addition of 100ml of a 0.05M CH3COOH aqueous solution (A3). B: XIV century egg temperawood panel from the National Gallery of Siena, Italy™. The black
square indicates the region where the cleaning test (performed bymeans of PAA•CO2/pentanol based gel) for the removal of surface layer of a lacquer applied onto the paint in a previous
restoration, was carried out. On the right a grazing light image of the area interested by the treatment is also reported.
A: adapted with permission from Langmuir, 20, 8414–8418. Copyright 2004 American Chemical Society; B: reprinted with permission from Langmuir, 20, 8414–8418. Copyright 2004
American Chemical Society.
Fig. 6. SEM micrograph showing the porosity of the magnetic nanosponge. The brighter zones are the regions where the ferrite nanoparticles are embedded. (right) Removal of the gel
from the paint surface by means of an external magnet. Sequence from top left to lower right.
Left, reprintedwith permission from Langmuir, 24, 12644–12650. Copyright 2004 American Chemical Society; right, reprintedwith permission from Langmuir, 23, 8681–8685. Copyright
2007 American Chemical Society.
Fig. 7. (Left) “Les Voiles” by Marcel Burtin (1902–1979), oil on canvas, peeling of a PVA/borate HVPD from the paint surface by means of tweezers. (right) Cleaning of a water sensitive
Thang-Ka mock-up by a semi-interpenetrated network of PVP and pHEMA.
Reprinted with permission from Langmuir, 29, 2746–2755. Copyright 2013 American Chemical Society.
367P. Baglioni et al. / Advances in Colloid and Interface Science 205 (2014) 361–371
Author's personal copy
Fig. 8. Structure of the PVA/borate network and image of a PVA/borate based aqueous HVPD (top left).
Reprinted with permission from Langmuir, 27, 13226–13235. Copyright 2013 American Chemical Society.
Fig. 9. Top: Coronation of Virgin with Saints, XV Century egg tempera wood panel by Neri di Bicci, Galleria degli Uffizi, Florence, Italy. The square box indicates the area interested by the
cleaning test performed bymeans of a PVA/Borate based HVPD containing 15w/w% of acetone. Bottom: Image of the degraded surface coating constituted by aged shellac and awhite egg
based varnish before (a) and after (b) the cleaning test.
368 P. Baglioni et al. / Advances in Colloid and Interface Science 205 (2014) 361–371
Author's personal copy
In fact, being the frequency sweep curves characterised by a crossoverbetween the shearmoduli G′ (elasticmodulus) andG″ (viscousmodulus)[65], these fluids appertain to the class of HVPDs.
Apart from their mechanical properties that can be tuned by varyingthe PVAs' hydrolysis degree, concentration of both borax and PVA, thepH, the temperature and the composition of the continuous embeddedaqueous phase, a further property that makes PVA/borax HVPDs suitablefor the cleaning of work of art surfaces (see Fig. 9) is that theymaintain their thermodynamic stability also upon the addition ofmeaningful amounts of organic solvents (up to 50% w/w in the caseof 1-propanol, 2-propanol and N-methyl-2-pyrrolidinone, etc [66]).Moreover, depending on the chemical nature of the loaded solvent,it is possible to further regulate the visco-elastic properties of the
HVPDs. These features allow a rigorous modulation of the selectivity,the efficacy and the invasiveness of the cleaning action in terms bothof the Hildebrand parameters and of the shear modulus [66].
Chemical gels are some of the less invasive and innovative toolsdeveloped for the cleaning of delicate surfaces of historic and artisticinterest. They are bi-continuous systems where the skeleton thatsupports the liquid phase is usually formed through the polymerisationof monomer units. The modulation of the strength of the polymernetwork allows a perfect control of the release of the solutions ordispersed systems (i.e. micellar solutions and o/w microemulsions)embedded, and a safe, easy peeling from the treated surface withoutleaving residues.
Excellent systems, in terms of cleaning ability, peelability, opticaltransparency and water retention capability (the equilibrium watercontent (EWC) is above the 50%w/w), were set up taking enlightenmentfrom the formulations of soft contact lenses [67]. The best preparationmethod is based on the use of a prepolymerized N-vinyl-1-pyrrolidone(VP) chains (MW ~106 Da) that are entrapped by the poly(2-hydroxyethyl-methacrylate) (pHEMA) network obtained by covalentlybonding HEMA monomers [56].
Due to different chemical structure and polarity of HEMA and VP, itis possible to optimise the gel formulation for the support to be treatedin terms of retention power, porosity, visco-elastic behaviour andhydrophobic/hydrophilic character of the network. Above all, this lastfeature makes these chemical gels highly versatile containers able tosupport both aqueous detergent solutions (o/w microemulsions andmicellar solutions) and a wide number of organic solvents characterisedby different solubility parameters. As a consequence, it is possible to setup gels effective in the removal of materials with different physico-chemical properties, like hydrophobic hydrocarbon-based coatings(waxes), highly polar protein material (animal or vegetal glues), andnatural (dammar, mastic) or synthetic varnishes (acrylic and vinyl).
3.3. Confinement of microemulsions and micellar solutions in HVPD and
chemical gels
A further improvement of gel technology for cleaning of paintedsurfaces that explored the possibility to load aqueous dispersed systemslike microemulsions or micellar solutions in highly viscous matrixeswas the purpose of our research group. The idea that inspired thisapproach is to merge the properties that make suitable these classesof systems to obtain a synergistic improvement of their cleaningabilities.
Fig. 10. Region of the wall painting by Vecchietta (Santa Maria della Scala Sacristy, Siena,
Italy) where the microE/HMHEC system was tested. Bottom right: grazing-light image of
part of the painting after the application of the cleaning system (dashed line). Top: SEM
images before and after the cleaning test.
Adaptedwith permission fromAngewandte Chemie International Edition, 48, 8966–8969.
Fig. 11. Cleaning of the backside of a relined canvas painting by o/wmicroemulsion confined inside a acrylamide hydrogel: (left) cleaning step, (centre) image of the cleaned area after gel
removal, (right) SEM images of the cleaned canvas: the border between the cleaned (bottom) and not cleaned (top) areas is highlighted.
Adapted with permission from Langmuir, 28, 3952–3961. Copyright 2012 American Chemical Society.
369P. Baglioni et al. / Advances in Colloid and Interface Science 205 (2014) 361–371
Author's personal copy
The confinement of microemulsions and micellar solutions inchemical gels or HVPD allows first of all the maximum lowering of theenvironmental impact. In fact the choice to substitute a pure solventwith a system containing up to 99 w/w% of water [18] inside of a highlyretaining viscous matrix minimises the evaporation of the dispersedphase and allows the control of the diffusion of the nanodroplets intothe porous support through the modulation of both the mesh size andthe viscosity of the polymeric skeleton. Furthermore, a perfect control ofthe cleaning action is alsomaintained through the use of low penetratingand easily removable optically transparent tools.
It is well-known that liquid oil-in-water (o/w) microemulsions andmicellar solutions have been successfully used for the removal ofsynthetic polymers from the surface of inorganic porous matrixes [19].Fig. 10 shows a detail of the wall paintings by Vecchietta (Santa Maria
della Scala Sacristy, Siena, 15th century) affected by a 35 years oldpoly(ethylmetacrylate/methylacrylate) (p(EMA/MA), Paraloid B72®)surface film applied in a previous restoration treatment. The grazing-light image (bottom right)was collected after the application of a highlyviscous HVPD composed by an o/w microemulsion embedded in a 3Dmatrix of hydrophobically modified hydroxyethyl cellulose (HMHEC)(contact time 15min) [41]. The application of microemulsion/HMHECresults in the progressive softening of the acrylic coating used to protectthe wall surface. The picture shows the disappearance of the glossyeffect macroscopically indicating the removal of the aged coating; thisfeature was confirmed by scanning electron microscopy investigation(Fig. 10 bottom shows that after the cleaning test the surface appearshighly rough as typical of a plaster free from polymeric coatings) andFTIR reflectance analysis carried out onto the cleaned surface.
O/w microemulsions can also be loaded into acrylamide–bis-acrylamide chemical hydrogels and used to swell and remove syntheticpolymers (mainly acrylic and vinyl polymers and copolymers). Recentlyan o/w microemulsion confined inside this kind of chemical gel hasbeen successfully used to remove a vinyl-acrylate film from the back-side of relined canvas painting [68]. After the swelling of the polymericcoating, its mechanical removal can be achieved avoiding any strongmechanical action onto the fibres (Fig. 11).
4. Conclusions and perspectives
Colloids and soft matter science have generated, in the last twentyyears, several innovative and fairly inexpensive tools for the cleaning ofpainted surfaces. In particular, the design of these new systems has beenpossible mainly thanks to the theoretical background from detergencyand wetting. The nanostructured systems improved the traditionalmethods giving at the same time both the minimisation of theenvironmental impact and the optimisation of the cleaning performance.However, the applied research in science of cultural heritage is still farfrom being mature, and offers several possible future scenarios to beexplored. These may include (a) new “green” surfactant-based self-assembled systems, (b) water-in-oil and waterless cleaning micro-emulsions for the treatment of contemporary art materials (e.g.,acrylic paintings), (c) organogels to be used as support systems forthe above-mentioned fluids, and (d) gels that are responsive todifferent external stimuli.
The conservation and valorization of the cultural heritage legacymay provide conspicuous economical benefits (i.e., tourism) and, froma different point of view, improve the image and perception of appliedsciences.
Acknowledgements
We thank all of the conservators involved in the application ofthe conservation methodologies presented in this review article and, inparticular, Tiziana Dell'Omo and Lucia Di Paolo for the intervention inNazareth (Israel), Daniela Ristori, Paola Bracco, Kioko Nakahara(Fondazione UIA, Florence), Maria Sframeli, Cristina Masdea (Ministry of
Culture, MIBAC, Italy) for the experimentation on the Neri di Bicci's woodpanel. Alessandro Bagnoli (MIBAC) and Fabrizio Iacopini for theexperimentation on Santa Maria della Scala wall paintings (Siena, Italy),and Florence Gorel for the experiments on Thang-Ka paintings. Moreover,Lilia Rivero Weber, Claudia Garcia Solis, Yareli Jaidar Benavides, and Mariadel Carmen Castro Barrera (Coordinacioń Nacional de Conservacion delPatrimonio Cultural, CNCPC, Mexico), are acknowledged for theexperiments in the Mayapan archaeological area. CSGI, MIUR (PRIN2009,2009P2WEAT project), and European Union (project NANOFORART, FP7-ENV-NMP-2011/282816) are acknowledged for financial support.
References
[1] Cather S (ed.): The Conservation of Wall Paintings. Proceedings of a symposiumorganized by the Courtauld Institute of Art and the Getty Conservation Institute,London, July 13–16, 1987, Editor, The Getty Conservation Institute, 1991.
[2] Mora P, Mora L, Philipot P. Conservation of wall painting. Butterworths; 1984.[3] Horie CV. Materials for conservation. Butterworths; 1987.[4] Wolbers RC. Cleaning painted surfaces, aqueous methods. Archetype; 2000.[5] Giorgi R, Baglioni M, Berti D, Baglioni P. Acc Chem Res 2010;43:695.[6] Wheeler G, Mendez-Vivar J, Fleming S. J Sol–Gel Sci Technol 2003;26:1233.[7] Wheeler G. Alkoxysilanes and the consolidation of stones. Los Angeles: The Getty
Conservation Institute; 2005.[8] Mahltig B, Swaboda C, Roesslerc A, Böttcher H. J Mater Chem 2008;18:3180.[9] Nanotechnology for the conservation of works of art. In: Baglioni P, Chelazzi D,
editors. London: RSC Publishing; 2013.[10] Baglioni P, Dei L, Carretti E, Giorgi R. Langmuir 2009;25:8373.[11] Carretti E, Bonini M, Dei L, Berrie BH, Angelova LV, Baglioni P, et al. Acc Chem Res
2010;43:751.[12] Chelazzi D, Poggi G, Jaidar Y, Toccafondi N, Giorgi R, Baglioni P. J Colloid Interface Sci
2012;42:392.[13] Baglioni P, Chelazzi D, Giorgi R, Poggi G. Langmuir 2013;29:5110.[14] Ferroni E, Gabrielli G, Caminati G. La Cappella Brancacci. In: Zorzi R, editor. Olivetti:
La scienza per Masaccio, Masolino e Filippino Lippi; 1992. p. 162.[15] Borgioli L, Caminati G, Gabrielli G, Ferroni E. Sci Technol Cult Herit 1995;4:67.[16] Lindman B, Danielsson I. Colloids Surf A 1981;3:391.[17] De Gennes PG, Taupin C. J Phys Chem 1982;86:2294.[18] Carretti E, Giorgi R, Berti D, Baglioni P. Langmuir 2007;23:6396.[19] Carretti E, Dei L, Baglioni P. Langmuir 2003;19:7867.[20] Lazzari M, Chiantore O. Polymer 2000;41:6447.[21] Chiantore O, Lazzari M. Polymer 2001;42:17.[22] Favaro M, Mendichi R, Ossola F, Rosso U, Simon S, Tomasin P, et al. Polym Degrad
Stab 2006;91:3083.[23] Orea H, Magar VA. Preprints of the 13th triennial meeting ICOM Committee for
Conservation ICOM-CC, Rio de Janeiro, 22–27 September 2002. In: Vontobel R,Vontobel R, editors. James and James; 2002. p. 176.
[24] Carretti E, Salvadori B, Baglioni P, Dei L. Stud Conserv 2005;50:128.[25] Palazzo G, Fiorentino D, Colafemmina G, Ceglie A, Carretti E, Dei L, et al. Langmuir
2005;21:6717.[26] Colafemmina G, Fiorentino D, Ceglie A, Carretti E, Fratini E, Dei L, et al. J Phys ChemB
2007;111:7184.[27] Baglioni M, Rengstl D, Berti D, BoniniM, Giorgi R, Baglioni P. Nanoscale 2010;2:1723.[28] Baglioni M, Giorgi R, Berti D, Baglioni P. Nanoscale 2012;4:42.[29] Baglioni M, Berti D, Teixeira J, Giorgi R, Baglioni P. Langmuir 2012;28:15193.[30] Holmberg K, Jonsson B, Kronberg B, Lindman B. Surfactants and polymers in aqueous
solutions. Wiley; 2004.[31] Raney KH, Benton WJ, Miller CA. J Colloid Interface Sci 1987;117:282.[32] Tongcumpou C, Acosta EJ, Quencer LB, Joseph AF, Scamehorn JF, Sabatini DA, et al. J
Surfactant Deterg 2003;6:191.[33] Tongcumpou C, Acosta EJ, Quencer LB, Joseph AF, Scamehorn JF, Sabatini DA, et al. J
Surfactant Deterg 2005;8:147.[34] Langevin D. Reverse micelles. In: Luisi P, Straub BE, editors. Plenum Press; 1984. p. 287.[35] Arroyo MC. Development of innovative nanostructured systems for the cleaning of
works of art. University of Florence; 2012 (PhD dissertation).[36] Michalski S. Cleaning, retouching and coatings, Proc. of the contributions to the
Congress, Brussels, 3–7 September 1990. In: Mills JS, Smith P, Smith P, editors.International Institute for Conservation of Historic and Artistic Works; 1990. p. 85.
[37] Stolow N. Nature 1957;179:579.[38] Wolbers RC. Workshop on new methods in the cleaning of paintings and other
decorative surfaces: notes. Canadian Conservation Institute; 1990.[39] Xue HT, Fang ZN, Yang Y, Huang JP, Zhou LW. Chem Phys Lett 2006;432:326.[40] Burnstock A, White R. Cleaning, retouching and coatings, Proc. of the contributions
to the Congress, Brussels, 3–7 September 1990. In: Mills JS, Smith P, editors.International Institute for Conservation of Historic and Artistic Works; 1990. p. 111.
[41] Carretti E, Fratini E, Berti D, Dei L, Baglioni P. Angew Chem 2009;48:8966.[42] Cremonesi P, Curti A, Fallarini L, Raio S. Prog restaur 2000;7:25.[43] Bonini M, Lenz S, Giorgi R, Baglioni P. Langmuir 2007;23:8681.[44] Bonini M, Lenz S, Falletta E, Ridi F, Carretti E, Fratini E, et al. Langmuir 2008;24:12644.[45] Carretti E, Grassi S, Cossalter M, Natali I, Caminati G, Weiss RG, et al. Langmuir
2009;25:8656.[46] Curteis T. An investigation of the use of solvent gels for the removal of wax-based
coatings from wall paintings. The Getty Conservation Institute; 1991.
370 P. Baglioni et al. / Advances in Colloid and Interface Science 205 (2014) 361–371
Author's personal copy
[47] Valentin N, Sánchez A, Herraez I. ICOM Committee for Conservation, Prepr. 11thtriennial meeting, Edinburgh, 1–6 September 1996. In: Bridgland J, editor. James &James; 1996. p. 851.
[48] Tomozei M, Balta Z. Metal 98, Proc. of the International Conference on metalsconservation, Draguignan-Figanières, 27–29 May 1998. In: Mourey W, Robbiola L,editors. James & James; 1998. p. 188.
[49] Da Silveira L. Stud Conserv 1997;42:11.[50] Khandekar N. Rev Conserv 2000;1:10.[51] Von Gilsa B. Cleaning paintings with enzymes, resin soaps, and emulsions.
Department of the Secretary of State of Canada; 1991.[52] Wolbers RC. Restoration '92: conservation, training, materials and techniques: latest
developments, Prepr. conference held at the RAI, Amsterdam, 20–22 October 1992.IICIn: Todd, editor. ; 1992. p. 74.
[53] Stulik D, Miller D, Khanjian H, Khandekar N, Wolbers RC, Carlson J, et al. Solvent gelsfor the cleaning of works of art. In: Dorge V, Dorge V, editors. The residue question.Getty Publications; 2004. p. 18.
[54] Burnstock A, Kieslich T. ICOM Committee for Conservation, Prepr. of the, 11thtriennial meeting, Edinburgh, 1–6 September 1996. In: Bridgland J, editor. James &James; 1996. p. 253.
[55] Iannuccelli S, Sotgiu S. Choices in conservation practice versus research, PreprGDWG Interim Meeting ICOM-CC, Copenhagen, 6–8 October 2010. In: Watteeuw L,
Vest M, Palm J, Watteeuw L, Vest M, Palm J, Van der Reyden D, editors. ICOM-CC;2010. p. 47.
[56] Domingues JAL, Bonelli N, Giorgi R, Fratini E, Gorel F, Baglioni P. Langmuir 2013;29:2746.
[57] Carretti E, Dei L, Baglioni P, Weiss RG. J Am Chem Soc 2003;125:5121.[58] Carretti E, Macherelli A, Dei L, Weiss RG. Langmuir 2004;20:8414.[59] Carretti E, Natali I, Matarrese C, Bracco P, Weiss RG, Baglioni P, et al. J Cult Herit
2010;11:373.[60] Sakurada I. Polyvinyl alcohol fibers. Marcel Dekker; 1985.[61] ShibayamaM, Adachi M, Ikkai F, Kurokawa H, Sakurai S, Nomura S. Macromolecules
1993;26:623.[62] Tsujmoto M, Shibayama M. Macromolecules 2002;35:1342.[63] Ochiai H, Kurita Y, Muratami I. Macromol Chem 1984;185:167.[64] Hermans PH. In: Kruyt HR, editor. Colloid science, vol. 2. Elsevier; 1949. p. 483.[65] Almdal K, Dyre J, Hvidt S, Kramer O. Polym Gels Netw 1993;1:5.[66] Natali I, Carretti E, Angelova L, Baglioni P, Weiss RG, Dei L. Langmuir 2011;27:
13226.[67] Tighe B. Hydrogels in Medicine and Pharmacy. Properties and applications, vol. III.
CRC Press; 1987.[68] Pizzorusso G, Fratini E, Eiblmeier J, Giorgi R, Chelazzi D, Chevalier A, et al. Langmuir
2012;28:3952.
371P. Baglioni et al. / Advances in Colloid and Interface Science 205 (2014) 361–371
