Cultural heritage assets are exposed to weather and submitted to influence of environmental parameters in a world where the climate is changing. Physical, chemical, and biological factors interact with constitutive materials inducing changes both in their compositional and structural characteristics [1
]. The great importance of water as a threat to heritage is acknowledged: in natural conditions atmospheric water is the main agent associated with stone degradation, acting mainly through capillary rising. Rainwater penetrating by absorption is a vehicle of airborne acidic pollutants interacting with stone through chemical reactions of dissolved CO2
, and SO2
. Moreover, water changes cohesion properties of the stone crystalline structure through physical/mechanical decay due to thermal excursions in wet conditions (freeze-thaw cycles) [4
]. Hence, the need to improve effectiveness and eco-sustainability of preventive conservation and maintenance solutions are grown hugely.
Different classes of polymers have been so far employed as protective coatings of stone heritage without adequate knowledge of the properties of both plain polymer and polymer/substrate system [5
]. As a result, insufficient efficacy and/or poor weatherability was usually observed. Such polymeric materials in most cases only provide short-term water repellency of the treated surfaces and are intrinsically unstable in photo-oxidative conditions typical of outdoor exposure. Notwithstanding that even polymers with partially fluorinated, side chains were ad hoc synthesized and tested to increase water repellency effectiveness and coating Ultraviolet (UV) radiation stability [8
], presently the scientific community is still far from the achievement of materials fulfilling all the fundamental requirements of protective coatings [9
While the last decade has seen several advancements in the field of polymer nanocomposites for a wide range of mechanical, electronic, magnetic, biological, and optical properties, fewer efforts have been focused on designing such a nanomaterial with optimal macroscale properties for protecting cultural heritage. A nanocomposite’s properties depend ultimately upon a myriad of variables that include the quality of dispersion, interfacial adhesion, extent of region between nanoparticles fillers and bulk polymer matrix, processing methods, loading of the particles, modification of the surfaces of nanoparticles, aspect ratio of particles, compatibility of particles and host polymer, size of particles, radius of gyration of the host polymer and the properties of the constituents. Even though in literature structure-property relationships are lacking, it is evident that the properties of polymer nanocomposites are highly sensitive to both the quality of dispersion and region between nanoparticles fillers and bulk polymer matrix and that small changes in processing conditions, particle size, or chemistry dramatically affects these two key factors [10
Recently results were achieved by matching a polymer with proper end properties, including eco-sustainable usage and non-toxicity, to create an inorganic photocatalytic nanocompound that was efficient in de-soiling and had biocide activities. A polymeric coating for protecting cultural heritage based on a water-dispersed TiO2
/poly (carbonate urethane) nanocomposite was prepared by a low impact procedure, i.e., cold mixing of the single components via sonication [11
]. By means of the polymeric nanocomposites technology, highly innovative and outstanding performances were also achieved in terms of stability and durability as compared with other treatments based on acrylic and vinylic polymers widely used in conservation and restoration [6
The next step of our investigation is concerned with applications of nanocomposite water dispersions on a porous degradable stone to demonstrate treatments’ aesthetical compatibility, and ability in reducing soiling and biocide properties [12
]. For a given nanocomposite concentration (w
%), water Absorption Coefficients (ACs) of untreated and treated stone samples were also evaluated according to NORmalizzazione MAteriali Lapidei (NORMAL) 11/85 [13
] as a function of the application procedure; i.e., air-brush until the stone surface was saturated, following a widespread practice in conservation, and full immersion in nanocomposite dispersions at room temperature for 1 h. The AC values achieved, pertaining to stone characterization, indicated that the treatments performed slowed the rate of water absorption of the stone. Hence, the nanocomposite homogeneous, transparent, colorless film formed by water casting at room temperature was proved to protect stone against water penetration.
The present work is focused, conversely, on water diffusion characteristics through TiO2
/poly (carbonate urethane) nanocomposite film samples, the novelty consisting of an in-depth analysis, on one hand, of nanocomposite diffusivity, and of the other hand, of effects of water uptake amount and nanoparticles/matrix interactions on glass transition temperature (Tg
) of Polyurethane (PU) soft and hard domains. As a matter of fact, Tg
has a deep influence on transport properties and, for applications of polymer-based materials in the field of cultural heritage, Tg
is, as well, a relevant requirement. Coatings with a Tg
value considerably higher than room temperature cannot be able to react to dimensional changes of treated items, whereas coatings with Tg
values conspicuously lower than room temperature are much too soft for working and moreover are inclined to pick up dirt. Nanocomposite water diffusion coefficients were determined by means of gravimetric techniques combined with on time-resolved Fourier Transform (FT)-Near Infrared (NIR) measurements and compared to that exhibited by pristine PU matrix. Moreover, vibrational spectroscopy was selected as one of the best-suited techniques for probing hydrogen-bonded molecular structures [14
] with the aim of gathering information on amount of water uptake, as well as, of species of water molecules absorbed within polymer matrix in presence of TiO2
nanoparticles. In particular, significant effects of the addition of 1% of TiO2
nanoparticles on amount of water free from, and strongly bonded through, H-bonding association absorbed/desorbed within the PU matrix, at environmental conditions under which weathering processes are most likely to occur, were highlighted. Correlations between adsorbed water amount and nanocomposite viscoelastic behavior were also established through Dynamic Mechanical Thermal Analysis (DMTA).
In order to counteract external degradation of monuments and buildings caused by the atmospheric pollution and meet the demands of cultural heritage with ecological, economic and social aspects, aqueous dispersions of different nanoparticles with photocatalytic capacity were used. Among them, nano-TiO2 is one of the most common owing to its versatility and green production, eco-compatibility and low-level impact on the chemical composition of materials. Notwithstanding this, relevant issues are still pending regarding the effectiveness and long-term stability of the coatings “in situ” and the impact of nanoparticles on human health and environment. As an alternative, and with outstanding advantages, the present paper shows that inorganic nanoparticles can be dispersed by means of low impact procedures into polymer matrices suitably selected and that modulation of relevant physical chemical properties such as water-repellency of a protective can be obtained.
The mechanism through which water diffuses into polymeric materials can be summarized as either infiltration into the free space or specific molecular interactions. The former is controlled by the free space available such as commonly occurring micro-voids and other morphological defects; an increase in the free space should result in an increase of both the water uptake and diffusivity. The diffusion of water by molecular interaction is, on the other hand, controlled by the available hydrogen bond at hydrophilic sites.
For the diffusion of water at room temperature through film samples of TiO2/poly (carbonate urethane) nanocomposite gravimetric sorption/desorption tests and FTIR spectroscopic analysis demonstrated that, for the composition investigated at least, the diffusion behavior is Fickian, and substantially linear, in so far as the diffusivity is independent of concentration. The mechanism expected when the diffusion rates are much slower than those of polymer relaxations (Fickian diffusion) can be summarized as follows. At temperatures below Tg, the polymer backbone is considered to be in a frozen state, segmental chain motions are drastically reduced, the number of free volume holes is fixed and no hole redistribution is likely. Mass transport is, therefore, assumed to take place via fixed (pre-existing) holes. A penetrant molecule must find its way from hole to hole along pathways involving only minor segmental rearrangements. This means that the diffusivity depends largely on the number of the holes with an appropriate size able to accommodate the diffusing molecule. In the rubbery state above Tg, the polymer chains are mobile and the free volume holes show a dynamic variation about size, shape, and position. The penetrant molecules diffuse within the fluctuating interstitial free volume with much greater mobility than in the glassy state.
Moreover, it was shown that the addition of only 1% (wt. %) of hydrophilic TiO2
nanoparticles to a poly (carbonate urethane) matrix strongly betters its barrier property. Water absorption in polymer nanocomposites containing impermeable anisotropic domains has been described in several publications. The most common nanocomposites investigated consist of a variety of polymers, both thermoplastic and thermoset, and nanoclay. Transport properties of PUs with soft segments consisting of polycaprolactone/organically modified montmorillonite nanocomposites have been investigated by Tortora et al. [49
]. Diffusivity of heterogeneous systems such as polymer nanocomposites is a complex phenomenon. Impermeable domains affect permeability not only by reducing the volume of material available for flow, but also by creating more sinuous pathways according to a tortuous model. Essentially impermeable nanoparticles act as obstacles forcing penetrant molecules to follow longer and complicated routes to diffuse through the material. At the same time, the incorporation of inorganic nano-fillers into the polymer matrix inevitably changes its morphological features and, consequently, its free volume properties. Effects of nanoparticles on polymer free volume to be expected are interfacial regions, interstitial cavities in the filler agglomerates, chain segmental motion immobilization, insufficient chain packaging, changes of the free volume hole size distribution, changes of the crystallinity of the matrix and change of the cross-linking density of the matrix. Which of them become dominant depends primarily on the degree of interaction between the components, the volume fraction of the filler and the geometrical features of the particles. Several studies carried out on reinforced epoxy nanocomposites showed that the maximum water absorption of a polymer system decreased due the presence of nano-filler [50
]. Such a phenomenon was generally ascribed to nano-fillers barrier properties together with a tortuous pathway for water molecules to diffuse. The achieved results indicate that the TiO2
/poly (carbonate urethane) nanocomposite defensive action against penetration of water free from, and bonded through, H-bonding association arises from a balance among TiO2
hydrophilicity, tortuosity effects and quality of nanoparticles dispersion and positive inter-facial interactions. Hence, the barrier property of such nanocomposite film is governed by a combination of physico-chemical properties including mode and state of dispersion of the minor component, the interaction between TiO2
nanophase and PU matrix, particle size and structure of TiO2
nanoparticles, PU morphology and structure, etc. Different analytical techniques, such as Thermo-Gravimetric Analysis–Differential Scanning Calorimetry (TGA-DSC), Field Emission Scanning Electron Microscopy (FESEM), Wide Angle X-ray Scattering (WAXS), DMTA and Attenuated Total Reflectance (ATR)-FTIR were, therefore, applied on both nanocomposite and pristine PU film samples to achieve a thorough characterization [11
]. The TiO2
/poly (carbonate urethane) nanocomposite is a multiphase system in which an inorganic phase with an average size of 31.08 nm was dispersed through sonication. Nanocomposite WAXS intensity profile shows a broad diffraction halo to be ascribed to the amorphous polyurethane phase [6
]; no Bragg reflection can be seen corresponding to both the TiO2
crystallographic forms Anatase and Rutile [38
]. Such a nanophase gives rise to superficial dissociation and/or adsorption and to specific interactions with the water molecules together with interactions with poly (carbonate urethane) hard segments. In turn, the poly (carbonate urethane) phase itself is to be considered as a two-phase amorphous-amorphous system, in which both hard and soft segments are permeable to the water molecules. The morphology of the hard and soft segments of the plain poly (carbonate urethane) was investigated through a careful examination of –NH and carbonyl peaks of ATR-FTIR spectra. It was found that the most of the amide groups are involved in hydrogen bonding [38
]. Work is in progress to investigate effects of the addition of TiO2
nanoparticles on PU phase separation by means of ATR-FTIR spectroscopy.
It is to be underlined that the amorphous structure of the TiO2/poly (carbonate urethane) nanocomposite confers material a certain degree of rubber elasticity essential for its applications on items with cultural value. In perspective of our final goal, i.e., showing that treatments based on water dispersions of TiO2/poly (carbonate urethane) nanocomposite successfully protect outdoor cultural assets stonework, it is to be pointed out that all the effects achieved by the addition of 1% (wt. %) of TiO2 nanoparticles are beneficial to combat both exposure/penetration of water and degradation agents conveyed by water and soiling and graffiti. Moreover, it is worthy to note that the nanocomposite Tg values, irrespective of water uptake amount, fulfill requirements for protective coatings. At environmental conditions under which weathering processes are most likely to occur, the PU soft phase remains above its Tg in an amorphous rubbery state, balanced by the PU hard phase in a glassy amorphous state below its Tg.
4. Materials and Methods
The raw materials used in this work are reported as follows: a linear aliphatic poly(carbonate urethane) (trade name Idrocap 994) was prepared by ICAP-SIRA (Parabiaco, Milano, Italy) in water dispersion with neutral pH to allow applications on substrates pH sensitive and organic solvents. The prepolymer mixing process followed was reported in a previous work [11
]. The Mw values of the poly(carbonate urethane) so achieved are in the range between 30,000 and 50,000 in Gel Permeation Chromatography (GPC) with standard Polystyrene (PS). Titanium dioxide (TiO2
) nanoparticles were synthesized and kindly supplied in water dispersion by the research center CE.RI.Col of Colorobbia Italia (Sovigliana, Vinci, Florence, Italy). [11
nanoparticles have an average size equal to 31.08 nm by Dynamic Light Scattering (DLS) technique with a polydispersity index of 0.241. All the reactants and solvents were used as received.
Plain poly (carbonate urethane) and nanocomposite film samples 0.60–1.00 mm thick were safely achieved using water-casting at room temperature. The preparation of a TiO2
/poly (carbonate urethane) nanocomposite containing 1% (wt. %) of TiO2
nanoparticles was performed by cold mixing the single components via sonication following the low impact method elsewhere reported [11
]. Also, the plain poly (carbonate urethane) was undergone identical sonication process.
Gravimetric sorption measurements were carried out by the so-called pat-and-weight technique. Film samples 0.60–1.00 mm thick were dried for 3 h at 100 °C under vacuum to achieve complete removal of absorbed water. The total absence of absorbed water was confirmed by means of FT-NIR spectroscopy. A Perkin-Elmer Spectrum 100 spectrophotometer (Perkin-Elmer, Beaconsfield, UK) was used. The instrumental parameters adopted for the FT-NIR monitored tests were as follows: resolution 4 cm−1, spectral range 8000–4000 cm−1.
FT-NIR spectra exhibited by the dried nanocomposite and plain poly (carbonate urethane) materials were also taken as a reference for spectral subtraction analysis.
To deeply investigate the effects related to the presence of TiO2 nanoparticles on water absorption and desorption kinetics of the poly (carbonate urethane) matrix the following procedures were carried out. Dried film specimens were introduced in an environmental climatic chamber SU250 Angelantoni Industries S.p.a (Cimacolle, Perugia, Italy) at the temperature of 20 °C and 50% of Relative Humidity (RH) simulating weathering. The samples, hereafter wet samples, were removed from the chamber at certain time intervals, weighted in a high precision analytical balance and FT-NIR transmission spectra were collected simultaneously. The amount of absorbed water was calculated by the weight difference. When the content of water remained invariable in the specimens then the kinetics were stopped.
Dried film samples were also immersed in a deionized water bath thermostatically controlled at 20 °C ± 1 °C until they adsorbed a water content constant in the time. The wet samples so achieved, hereafter wetI samples, were introduced in the chamber SU250 Angelantoni Industries (Angelantoni, Naples, Italy) setting the same conditions of temperature and relative humidity used for weathering simulation. Periodically, the samples were removed, blotted and reweighted, the desorption of water was so monitored. In such a procedure we could not apply FT-NIR technique as the high amounts of water absorbed.
Effects of water diffusion on the visco-elastic behavior of both nanocomposite and plain poly (carbonate urethane) were investigated through dynamic mechanical thermal analysis (DMTA) using a Perkin-Elmer Pyris Diamond DMA apparatus (Perkin-Elmer Italia S.p.A, Monza, Italy). Tests were performed in bending mode, applying a strain of 1%. Single-frequency measurements at 1 Hz were performed at a constant heating rate of 3 °C/min, in the temperature range from −100 °C up to 200 °C.
Mode and state of dispersion of the TiO2 nanoparticles into the poly (carbonate urethane) matrix were analyzed by means of a Fei Quanta 200 field emission Environmental Scanning Electron Microscope (ESEM, FEI, Hillsboro, OR, USA) operating in high vacuum mode.