Probing the Cleaning of Polymeric Coatings by Nanostructured Fluids

May 24, 2017 - For a more comprehensive list of citations to this article, users are encouraged to perform ... Materials from Works of Art: SAXS Chara...
0 downloads 0 Views 2MB Size
Subscriber access provided by UNIV OF UTAH

Article

Probing the cleaning of polymeric coatings by nanostructured fluids: a QCM-D study Martina Raudino, Nicoletta Giamblanco, Costanza Montis, Debora Berti, Giovanni Marletta, and Piero Baglioni Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b00968 • Publication Date (Web): 24 May 2017 Downloaded from http://pubs.acs.org on June 5, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Probing the cleaning of polymeric coatings by nanostructured fluids: a QCM-D study Martina Raudinoa, Nicoletta Giamblancob, Costanza Montisa, Debora Bertia, Giovanni Marlettab*, Piero Baglionia* a

Department of Chemistry and CSGI, University of Florence, Via della Lastruccia 3, 50019 Sesto Fiorentino, Florence, Italy Department of Chemical Science, Laboratory for Molecular Surfaces and Nanotechnology (LAMSUN), University of Catania and CSGI, Viale A. Doria 6, 95129, Catania, Italy

b

KEYWORDS: Nanostructured fluids. Micelles. Polymer films. Detergency. De-wetting. Cleaning artifacts. Conservation Cultural Heritage. Confocal Microscopy. Quartz Crystal Microbalance.

ABSTRACT: Complex fluids composed of water, an organic solvent and a surfactant have been recently employed as cleaning systems to remove hydrophobic materials, such as polymeric coatings, from solid surfaces. The simultaneous presence of surfactants and an organic solvent with good affinity for the polymer was proven necessary for the polymer's removal, but the comprehension of the cleaning mechanism is poorly understood. In this article we investigated the mechanism of removal, highlighting the specific role of each component in the interaction with the polymer film. In particular, the results from quartz crystal microbalance with dissipation monitoring (QCM-D) were compared with those obtained by using confocal microscopy to follow in situ the effect of a nanostructured fluid, i.e., a ternary formulation containing water, 2-butanone (MEK) as a good solvent for the polymer, and a nonionic surfactant (C9-11 ethoxylated alcohol, BR) on acrylic co-polymer films (Paraloid B-72). The results indicate a two-step process: i) the penetration of the good solvent across the film causes the swelling of the polymer, the weakening of polymerpolymer interactions and an increase of molecular mobility, followed by ii) the slow adsorption of amphiphilic aggregates promoting the film detachment from the solid substrate. A different behavior is observed in the presence of similar formulations containing an anionic surfactant (sodium dodecyl sulfate, SDS), where the adsorption of SDS micelles on the surface of the polymeric film hinders solvent access into the polymer layer, rather than promoting its detachment from the solid substrate.

1. Introduction Synthetic polymers have been widely employed in conservative treatments on works of art, mainly on wall paintings, as consolidants to reinforce the adhesion between the damaged pictorial layer and the other layers of the wall painting itself. Among them, the acrylic co-polymer Paraloid B72 is one of the most common used in the field of conservation of cultural heritage. However, the application of a polymeric coating over a porous matrix like a wall painting presents several drawbacks, since the applied polymer modifies the physico-chemical properties of the treated surface1, eventually causing mechanical stresses and the degradation of the work of art. In addition, polymeric coatings degrade at fast pace altering the paintings' original aspect2,3, making necessary their removal4. Polymer removal is challenging with traditional cleaning media such as neat organic solvents, since it is difficult to control their removal (due to a lack of polymer solubility), the penetration inside the treated surface and the re-deposition of the removed polymer in the porous matrix of the treated substrate upon evaporation of the solvent is a common unwanted effect. In addition, both environmental effects and health consequences on restorers cannot be neglected5. These issues make necessary the implementation of new cleaning methods,

mainly based on microemulsions6,7 and gels8,9, where the amount of toxic organic solvent is a few percentage of the system while the continuous phase is water, which have been pioneered by ou5/7/2017 5:58:00 AMr group. While the dissolution process of polymer films in the presence of neat solvents has been widely investigated in the past10–12, the effect of a mixture of a good and a poor solvent for the polymer with a surfactant has been scarcely addressed. To fill the gap, in a recent work13 we reported the combined use of atomic force microscopy (AFM) and confocal laser scanning microscopy (CLSM) to study in situ the removal mechanism of Paraloid B72 films in contact with cleaning formulations containing water, an ethoxylated alcohol C9-11 as nonionic surfactant (called as BR) and 2-butanone (MEK) as a good solvent for the polymer. These amphiphilic media, that cannot be properly considered as microemulsions, due the large partition of the orgnic phase in both the dispersed and continuous phase, are particularly interesting thanks to their optimal cleaning efficacy and to the high biodegradability of the surfactant, which allows to reduce the problem related to the possible residues remaining on the treated artifact after the exposure to the amphiphilic formulation. The study was also extended to similar formulations containing an anionic surfactant (sodium dodecyl sulfate) to shed light on the effect of the chemical nature of the

1 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

amphiphile on the overall process efficiency. SDS-based formulations were included for comparative purposes since they have been the first microemulsions developed and used in restoration5,7. This study represented a step forward in the understanding of the removal process from a macroscopic point of view, but the removal mechanism and its time dependence have not been clarified. Quartz Crystal Microbalance (QCM) is a gravimetric technique allowing the detection of small amounts of substances (typically about 1 ng/cm2)14, either deposited on or removed from the sensor surface, by monitoring changes in the oscillating frequency of the resonator. Its high sensitivity is currently applied for the recognition and detection of volatile chemicals15–17, biomolecules18–20, pollutants21–23 and so on. In addition, QCM with dissipation monitoring (QCM-D) can simultaneously measure the variations of both frequency and dissipation parameters, the latter correlated to the viscoelastic properties of the adsorbed layer, providing insights on the structural features of the layer24–26. In particular, recent studies reported the use of quartz crystal microbalance on the removal efficacy of oily soil27,28 by different surfactants and proved the influence of different substrates on the removal process29. The study of the interaction mechanism between phospholipid oilin-water emulsions or microemulsions with polymeric films30 was investigated as well. In this paper, we used QCM-D to study the removal of acrylic films (Paraloid B72) in the presence of the same formulations previously employed to investigate this process by AFM and CLSM13. In particular, by monitoring frequency shifts as a function of exposure time to liquid phases of different compositions, we gained insight about the role of each component of the formulation in the interaction with the polymer film, highlighting the influence of the chemical nature of the surfactant on the efficiency of the cleaning process. The analysis of overtones was used to infer fundamental details on the mechanisms of the surfactant action throughout the thickness of the polymeric film. In order to compare the removal mechanism at the macro- and nano-level, the QCM-D analysis was combined with confocal microscopy observations. It is worth noting that the two techniques provide complementary information: QCM is able to detect the adsorption/desorption of mass from the sensor surface at the molecular level, thus allowing to monitor also the early stages of interaction between the polymer and the cleaning fluids, while confocal microscopy reveals morphological changes of the film occurring at the microscale. We should stress that we report here on simple, but enough representative model systems in which the polymer is deposited onto smooth and regular surfaces with a reproducible and homogeneous thickness. As a consequence, we cannot fully evaluate the effects of roughness and porosity of the substrate on the removal. To the best of our knowledge, QCM-D is employed for the first time to address the mechanism of interaction of complex fluids constituted by a good solvent, a non-solvent and a surfactant on a polymeric coating. These results provide a better comprehension of the mechanistic aspects of the cleaning process, beneficial for the design of detergent systems tailored for specific cleaning applications. In particular our findings are of outmost importance for the formulation of new cleaning systems in the field of Cultural Heritage Conservation and detergency.

Page 2 of 11

2. Experimental Section 2.1 Chemicals BR ethoxylated alcohol (purity 98%) was purchased from AkzoNobel; sodium dodecyl sulfate (SDS, purity 99%), 2butanone (MEK, purity 99%), ethyl acetate (purity ≥ 99.5%), and the fluorescent probes used during confocal laser scanning microscopy experiments (Rhodamine 110 chloride and Rhodamine B isothiocyanate, purity > 99%) were obtained from Sigma Aldrich and used as received without further purification. High-purity water (resistivity > 18 MΩ cm) was obtained with a Millipore Milli-Q gradient system. Paraloid B72 pellets were purchased from Zecchi, Florence. Figure 1 shows the chemical structure of Paraloid B72 and both the used surfactants, BR and SDS.

Figure 1. Chemical structure of A) BR surfactant, B) sodium dodecyl sulfate, C) Paraloid B72 (m:n = 70:30).

2.2 Samples preparation Binary water/MEK mixtures and water/BR micellar solutions were prepared mixing the components under stirring. BR and SDS-based ternary systems were prepared adding MEK to BR and SDS micellar solutions under constant stirring. Table 1 reports the chemical composition of all the investigated samples. Table 1. Chemical composition of the cleaning fluids. Sample

Composition (% w/w) Water Surfactant MEK

H2O/MEK

80

-

20

BR MEK 0%

95

5

-

BR MEK 20%

76

4

20

SDS MEK 20%

76

4

20

2 ACS Paragon Plus Environment

Page 3 of 11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

In order to obtain information on the interaction between the polymer and the cleaning fluids, the emission of both Rhodamine 110 chloride (green, 27 µM) staining the cleaning fluids and Rhodamine B isothiocyanate (red, 19 µM), co-dissolved in the polymeric film, was recorded by means of CLSM (Confocal Laser Scanning Microscopy) investigation. The dyes were chosen according to their different hydrophilic/hydrophobic nature and reliably mimic the behavior of the aqueous phase (Rhodamine 110 chloride) and of the polymer film (Rhodamine B isothiocyanate)13. For QCM-D analysis, water/MEK mixture and ternary fluids were diluted 10 times with water in order to slow down the removal process of the polymer film. It is worth noting that both BR and SDS surfactants in the nanostructured fluids are more than two orders of magnitude above the critical micellar concentration (0.9 mM and 8.3 mM for BR and SDS, respectively). 2.3 Quartz Crystal Microbalance with Dissipation monitoring (QCM-D) QCM-D measurements were performed with a Q-Sense E1 (Biolin Scientific, Sweden). Before use, the gold AT-cut quartz crystals were treated by UV-ozone for 20 minutes, rinsed with Milli-Q water and then dried under nitrogen flux. The polymer film was obtained by spin coating (1000 rpm, 120 s) on quartz crystals a droplet of 10% w/w Paraloid B72 solution in ethyl acetate. The coating film, with a uniform thickness of about 750 nm ± 100 nm measured by means of AFM analysis (non-contact mode, NCHR probes, radius of curvature at the tip about 5 nm), was allowed to stabilize in deionized water to establish the baseline. The Paraloid B72 coated sensors were placed in flow cell and the liquid phases, thermally equilibrated at 25 °C, were injected into the chamber (chamber volume = 500 µL) at flow rate of 100 µL/min using a cartridge pump (Ismatec-ISM 935). Simultaneous measurements of frequency and dissipation were taken for all the overtones (n = 3, 5, 7, 9, 11, 13). QCM experiments were repeated three times for each system. The acoustic response of the various systems has been analyzed by using the Voigt model, explicitly taking into account the presence of more or less viscous overlayers between the sensor and bulk liquid. The Voigt model, in particular, describes the viscoelastic behaviour in terms of a complex shear modulus (Eq. 1):

 =   +   =  + 2 

Eq. 1

where µf is the elastic shear modulus and ηf is the shear viscosity. Accordingly, the ∆f is given by the expression in Eq. 2:

Δf = Im(β)/2π 

Eq. 2

and the dissipation term is given by the expression in Eq. 3:

Δ = −  (β)/π  ρ

Eq. 3

where ρq is the density of quartz (2650 kg/m3) and β is a complex function of density and viscosity of the bulk liquid, the overlayers 1, 2, etc31... In the above relationship (Eq.1), the real part of G (i.e., the storage modulus) is independent on the frequency of oscilla-

tion of the sensor, while the imaginary part (loss modulus) depends upon the frequency. The detailed forms of the ∆f and ∆D response were employed to analyze the acoustic response of the various investigated systems. According to Eqs. 1 and 2, the introduction of fluids with different chemical composition in the measurement chamber could induce further changes in ∆F and ∆D due to viscosity and density changes. Thus, the effects of such fluids on ∆F and ∆D were taken into account using a bare gold sensor in contact with the fluids: the low ∆F and ∆D shifts measured in the calibration test (see Figure S1) suggested that liquid loading effects have a negligible role in the QCM-D response. The penetration depth δ of an acoustic wave is given by Eq. 332: '/(

 = "# % & !

$ !

Eq. 3

where ηf is the viscosity of the film, n is the overtone number, and ρf is the density of the film. Due to different penetration depths of the acoustic waves associated with different overtones, higher overtones (e.g., the 11th and 13th harmonics) are associated with properties of the film near the crystal surface, while lower overtones are more related to processes occurring relatively far from the sensor surface, but within the sampling depth of the technique (i.e., about 200 nm for dense and relatively viscous polymer films) and inside the film, well below the solution/polymer coating interface. As a consequence, recording ∆F and ∆D at multiple overtones provides information about the penetration gradients of the solvent solution and the related processes close to the sensor surface. It is important to note that QCM-D data will be used only to evaluate the relative mass loss and the related energy dissipation trends. 2.4 Confocal Laser Scanning Microscopy (CLSM) CLSM experiments were performed with a laser scanning confocal microscope Leica TCS SP2 (Leica Microsystems GmbH, Wetzlar, Germany) equipped with a 63X water immersion objective. 561 and 488 nm laser lines were used to acquire the fluorescent emission of Rhodamine B isothiocyanate co-dissolved with the polymer (red, fluorescence measured between 571−650 nm) and Rhodamine 110 chloride dissolved in the aqueous liquid phase (green, fluorescence measured between 498−530 nm). In order to follow the interaction between Paraloid B72 and the cleaning systems water/MEK, BR MEK 20%, and SDS MEK 20%, glass substrates were coated with a 30 µm thick polymer film obtained depositing a droplet of a 10% w/w Paraloid B72 solution in acetone and evaporating the solvent. Thinner films (thickness about 6 µm) were prepared by spin coating of a droplet of Paraloid B72 solution (1000 rpm, 120 s) and employed to monitor the milder morphological variations induced by water and BR micellar solution. The morphological variations of the polymeric film in presence of the different liquid phases were monitored as a function of time by putting into contact the coated glasses with 1 mL of the cleaning medium labeled with Rhodamine 110 chloride. 3. Results and Discussion Several works5,33 have demonstrated the good performances of ternary systems in the removal of polymer films, due to the

3 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

synergy between the amphiphile and a good solvent dispersed in aqueous media. However, not every surfactant shows the same effectiveness, highlighting a subtle role of the amphiphile in the overall process. In particular, laboratory tests have proved the excellent removal properties of the nonionic surfactant BR MEK 20% formulation (86 ± 2 wt% of polymer) within 1.5 h application13. The main purpose of this study is to provide fundamental understanding into this phenomenon and, in particular, if possibly specific interactions arise between the polymer and each component of the liquid phase in a precise multistep sequence. We will first address possible interactions between neat water and Paraloid B72 films; then, the complexity of the liquid phase is increased by adding the nonionic surfactant (BR MEK 0%) or the good solvent MEK (water/MEK). Finally the ternary system (BR MEK 20%) is considered and compared to a similar one containing SDS (SDS MEK 20%) to highlight the role of surfactant nature on the polymer removal. 3.1 Water - Paraloid B72 Paraloid B72 shows low affinity for water (static contact angle of water θ = 73° ± 3°)13. A CLSM scan reported in Figure 2 reveals the confinement of the green probe initially dissolved in the aqueous medium in the liquid layer, thus indicating that water, which is a non-solvent for the polymer, interacts only with the surface of the film without penetration inside the coating also for long exposure times.

Page 4 of 11

the quartz sensor surface. The slip motion, then, is “read” as a weight gain for the lower overtones (n = 3, 5), the related dissipation being in agreement with the expected more viscous behavior for a swollen polymer, while it is sensed as a small frequency increase for the higher overtones. It is worth to note that a relatively large dissipation, diagnostic of viscous behavior and mass gain, is observed also for the higher overtones. Overall, the QCM-D results for Paraloid B72 films in water suggested that the non-solvent penetrates only the shallower layers of the film, in good agreement with the CLSM data. 3.2. BR micellar solution - Paraloid B72 When the nonionic surfactant BR is added to water in a final concentration above its CMC, the confocal scan (Figure 4) reveals that the polymeric layer is partially lifted from the glass, suggesting that the presence of the surfactant induces a more evident swelling of the polymer with respect to neat water, even if the liquid phase does not penetrate through the entire film (as shown by the confinement of the green probe in the liquid phase).

Figure 2. Vertical confocal scan at t = 7 h of a Paraloid B72 film (thickness ~ 6 µm) stained with Rhodamine B isothiocyanate (red) incubated with water stained with Rhodamine 110 chloride (green).

The CLSM results (in part already reported in a previous study, see reference 13) have been compared with in situ QCM-D measurements, reported in Figure 3, where frequency and dissipation shifts for different overtones are monitored during exposure of the polymer film with neat water. For lower overtones (n = 3, 5) a peculiar behavior is observed, consisting in frequency decrease and dissipation increase, whereas for higher overtones (n = 9, 11, 13) frequency and dissipation increase simultaneously. According to the different sampling thickness of different overtones mentioned in paragraph 2.3, these findings suggest that water penetrates within the upper part of the region sampled by QCM-D, yielding two neighboring swollen and non-swollen regions with marked density and viscosity difference. The inhomogeneity of water penetration, in turn, prompts the slip motion of the upper polymeric part with respect to the underlying non-swollen regions closer to

Figure 3. Shifts in frequency (top) and dissipation (bottom) as a function of time (t = 3.5 h) of 3th, 5th, 7th, 9th, 11th, 13th overtones for the Paraloid B72 coated gold sensor (thickness = 750 nm) in contact with neat water.

We can conclude that the presence of the nonionic amphiphile enhances the wettability of the polymer by the aqueous medium, as confirmed by the strong decrease of the contact angle measured in the case of the micellar solution with

4 ACS Paragon Plus Environment

Page 5 of 11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

respect to neat water (θ = 8° ± 1° instead of 73° ± 3° for pure water).

Figure 4. Vertical confocal scan at t = 8 h of a Paraloid B72 film (thickness ~ 6 µm) stained with Rhodamine B isothiocyanate (red) incubated with BR MEK 0% mixture stained with Rhodamine 110 chloride (green).

The interaction between BR micellar solution and Paraloid B72 was also followed through QCM-D. Figure 5 shows a dramatic frequency decrease (up to -1300 Hz for n = 9) and a uniform decrease trend for all the overtones: this is consistent with a marked and homogeneous increase of the mass gained by the polymer film throughout its thickness. In addition, we observe that a higher frequency shifts is measured for higher overtones, suggesting that pores or holes in the polymer films act as initiating sites for the preferential penetration of the solvent mixture along the sensor/polymer interface, rather than within the “bulk” of the polymer layer. The dissipation closely follows the water/surfactant uptake that increases with the liquid penetration in the polymer layer. The whole process can be divided in two steps: 1) the first up to about 4 h, when dissipation values are lower than 90˙10-6 (about 4 times higher than the ones observed for the swelling induced by simple water), and 2) the second for t > 4 h, leading to very high dissipation values (∆D ∼ 220 - 600 ˙10-6 for the various overtones), consistent with the formation of a highly viscoelastic fluid, in agreement with a remarkable softening of the coating, following the penetration of surfactant and associated water prompting the detachment from the substrate27,34. During this latter phase, large frequency and dissipation jumps are observed for the higher overtones (n = 9, 11, 13), which account for the lift off of the region of the film closer to the sensor surface, accordingly to CLSM data revealing the formation of significant areas of the polymer lifted off from the glass substrate. Milder effects are observed for the lower overtones, probably due to the slighter modification of the upper regions. It is important to stress that confocal microscopy is able to detect only modifications of the film induced by the adsorption of surfactant aggregates occurring at the microscale, such as the deformation of the film visible after long incubation times. For this reason, differently from QCM-D, confocal microscopy cannot infer details about the interaction of micelles with the polymer during the early stages of the process. Overall, QCM-D and CLSM data suggest that micellar solutions may reduce chain/chain interactions, favoring a massive water penetration throughout the whole film, which in turn causes a remarkable softening of the coating.

Figure 5. Shifts in frequency (top) and dissipation (bottom) as a function of time (t = 8 h) of 3th, 5th, 7th, 9th, 11th, 13th overtones for the Paraloid B72 coated gold sensor (thickness = 750 nm) in contact with BR MEK 0% mixture. The inset in the top graph shows frequency shifts for all overtones during the first 150 seconds of incubation of the polymer with BR MEK 0% mixture.

3.3 Water/MEK binary system - Paraloid B72 When Paraloid B72 films are put in contact with a mixture of water and a good solvent, such as MEK, the confocal images reported in Figure 6 reveal strong morphological changes. In particular, we can observe the formation of regions at the interface polymer/glass containing only the liquid phase (see the presence of the green probe labeling the solvent inside the cavities) within a matrix of swollen polymer. The diameter of these round areas rapidly grows in the first 5-10 minutes of incubation, to remain almost constant for longer incubation times, even though the number of droplets per unit volume increases. This indicates a fast interaction between the polymer and the liquid during the first minutes of contact, followed by a slowdown of the process. The vertical confocal scan reported on the right side of Figure 6 shows the formation of round areas containing the liquid phase, where the polymer is lifted up from glass but remains substantially attached to the substrate even if in a swollen state. The fact that water/MEK mixture is not effective for polymer removal is also confirmed by lab

5 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

cleaning tests, which evidenced no detachment even for long incubation times (polymer mass removed from glass after 3 h of incubation with the liquid phase about 0 ± 10%) 33. Figure 7 reports the QCM-D measurements on the same system. The data are consistent with very fast penetration of the water/MEK mixture within the film, yielding a steady state in a very short time (∼ 10 minutes), which is stable up to long incubation times (t > 6 h). This very fast adsorption can be described with an exponential decay function, ∆F(t) = t0 + B exp (-t/τ), with a characteristic time τ = 193 ± 11 s. The amount of water/MEK adsorbed within the film is 4 times higher than for pure water (see Figure 3), as expected for the presence of a good solvent, but it is much less than the mass uptake measured for BR MEK 0% binary system. In particular, the exposure to water/MEK does not cause film de-

Page 6 of 11

tachment, but rather induces a mild swelling, as indicated by the dissipation data, similar to what found for the water/Paraloid B72 system. The overall stability of the swollen film is in agreement with CLSM observations, showing that a population of droplets are formed on the glass surface in the first 5-10 minutes, i.e., at the very same time when QCM-D monitors fast water/MEK uptake, to remain thereafter stable for long times in a swollen viscoelastic matrix (see above). The overtones show a large spread, consistent with a significant viscoelasticity gradient throughout the thickness, due to the gradient in water/MEK penetration. Accordingly, we can speculate that the higher ∆F and ∆D values observed for 11th and 13th overtones may point to a higher penetration of MEK in the region closer to the interface between the swollen and the more dense layer, at the quartz sensor surface.

Figure 6. Horizontal confocal scans (left) at t = 1.5, 7.5, 12.5 and 120 minutes of a Paraloid B72 film (thickness ~ 30 µm) stained with Rhodamine B isothiocyanate (red) incubated with water/MEK mixture stained with Rhodamine 110 chloride (green). Vertical confocal scan (right) after 40 minutes of incubation of the same polymer film. The panel at t = 12.5 minutes is adapted from Figure 6, panel C, ref. 13.

3.4. BR/MEK ternary system - Paraloid B72 Putting together water, surfactant and the good solvent MEK in ternary system, confocal imaging (Figure 8) reveals remarkable morphological variations within short incubation times. The round areas at the glass/polymer interface reach their maximum size within about 3 minutes of contact, instead of 10 minutes as in the case of water/MEK. Increasing the incubation time up to 30 minutes, differently from the case of MEK/water system, they coalesce until the film is removed from the glass surface, in agreement with the results of lab cleaning tests (86 wt% ± 2 of polymer removed within 1.5 h)13. Figure 9 shows ∆F and ∆D variations during the incubation of Paraloid B72 film with BR MEK 20% ternary system. The overall behavior of the system is similar to the one of BR micellar solution, the major difference being that no frequency/dissipation jumps are visible. This can be a confirmation of the significant morphological changes observed at the microscale by confocal imaging: the very large solvent uptake oc-

curring to the film when exposed to the ternary system is responsible for a change of the film properties too large to give a resonance with the employed QCM-D sensor, especially in the region of the film closer to the sensor surface where confocal imaging revealed the more important modifications of the film (see for example that 13th overtone goes in saturation after 2500 seconds). On the other hand, the frequency decrease trends are very similar for the two systems, with a continuous mass uptake without a steady state. The mass and the relative uptake rate are much larger for the ternary system, yielding, at the longest incubation time, ∆F values ranging between -1300 and - 2600 Hz, with a very large spread of dissipation values, in turn ranging between 20 and 700 ˙10-6. The data are consistent with a gradual but catastrophic softening of the polymer, supported by the very large overtone spread. Again, as for the BR MEK 0% binary system, higher frequency and dissipation values are found for the higher overtones, diagnostic of stronger changes in the region buried inside the polymer film.

6 ACS Paragon Plus Environment

Page 7 of 11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 8. Horizontal confocal scans at t = 45 s, 1.5, 3 and 30 minutes of a Paraloid B72 film (thickness ~ 30 µm) stained with Rhodamine B isothiocyanate (red) incubated with BR MEK 20% stained with Rhodamine 110 chloride (green).

Figure 7. Shifts in frequency (top) and dissipation (bottom) as a function of time (t = 6 h) of 3th, 5th, 7th, 9th, 11th, 13th overtones for the Paraloid B72 coated gold sensor (thickness = 750 nm) in contact with water/MEK. The inset in the top graph shows frequency shifts for all overtones during the first 150 seconds of incubation of the polymer with water/MEK mixture.

The frequency variation trend can be analyzed taking into account the fact that the ternary system contains two different components that interact with the polymer film at different times. Accordingly, the trend can be described by a double exponential function ∆F (t) = t0 + B1 exp (-t/τ1) + B2 exp (-t/τ2) with τ1 = 148 ± 5 s and τ2 = 2269 ± 56 s. The detailed data analysis shows in fact that these are consistent with a two-step interaction, where τ1 is comparable to the value found for the fitting of the water/MEK mixture (193 s), and the characteristic τ2 is consistent with the slow adsorption of BR surfactant aggregates.

3.5. SDS/MEK ternary system - Paraloid B72 In previous studies focusing on the application of these nanostructure fluids to the conservation of Cultural Heritage4,5,7,33, we evidenced that the chemical nature of the surfactant greatly influenced the cleaning process. This effect might be related to the different polymer-surfactant interaction that affects the polymer removal mechanism. To address this point, we compared the BR/MEK ternary system to a similar one, containing SDS. The confocal images reported in Figure 10 show milder morphological variations of the Paraloid film (i.e., the diameter of the rounded regions at the polymer/glass interface is smaller) than in the case of BR ternary system, suggesting a weaker affinity of SDS for the polymer, consistently with what observed in AFM (see SI, Figure S2) and cleaning tests (0 ± 10 wt% of polymer removed within 1.5 h13). In this case, the removal process is very slow. Indeed, the more hydrophilic character of SDS with respect to BR (HLB number 40 vs 13, respectively), and the electrostatic repulsion between the charged SDS molecules/micelles adsorbed on the polymeric surface and those moving from the bulk solution toward the outer layers of the polymeric film, can explain the scarce affinity of SDS for the hydrophobic polymer.

7 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 11

stained with Rhodamine 110 chloride (green). The panel at t = 90 minutes is adapted from Figure 8, panel C, ref. 13.

Figure 9. Frequency shifts (top) and dissipation (bottom) as a function of time (t = 1.5 h) of 3th, 5th, 7th, 9th, 11th, 13th overtones for the Paraloid B72 coated gold sensor (thickness = 750 nm) in contact with BR MEK 20%. The inset in the top graph shows frequency shifts for all overtones during the first 150 seconds of incubation of the polymer with BR MEK 20%.

In the case of the SDS ternary system, the film behavior could be measured with QCM-D only for the first few minutes of incubation, due to an inherent instability of the film. To compare this ternary system with the other ones, we measured the frequency shift for the 5th overtone (selected as yielding the best average response) during the first 150 s of incubation of a Paraloid B72 film with the four liquid phases (Figure 11). Since these early frequency variations are related to solvent uptake, the experimental data fitting yields a comparison of the swelling rates for the four liquid systems under examination. Figure 10 reports such comparison, while Table 2 shows the decay parameter obtained by fitting the frequency decrease profiles with a power law decay, which can be related to the swelling rate. When the film is challenged with the BR micellar solution (BR MEK 0%) and the SDS ternary system (SDS MEK 20%), no mass uptake occurs within the first 2 minutes of incubation (∆F is constant and equal to zero), pointing to a very slow penetration in the presence of SDS, in spite of the presence of the good solvent. Conversely, both water/MEK mixture and BR ternary system (BR MEK 20%) interact immediately with the film. The higher value of the power law exponent for BR MEK 20% confirm a faster stronger adsorption of the liquid phase, supporting an active and synergistic role of BR micelles even in these early stages. The high affinity of BR solutions for the polymer surface plays a crucial role in enhancing the water/MEK penetration within the film, while the lower affinity of SDS hinders good solvent penetration.

Figure 11. Shifts in frequency (t = 150 s) of 5th overtone for the Paraloid B72 coated gold sensor (thickness = 750 nm) in contact with BR MEK 0%, H2O/MEK, BR MEK 20% and SDS MEK 20%. Fitting curves are reported as continuous lines together with the experimental data.

Figure 10. Horizontal confocal scans at t = 4, 15, 90 and 130 minutes of a Paraloid B72 film (thickness ~ 30 µm) stained with Rhodamine B isothiocyanate (red) incubated with SDS MEK 20%

Table 2. Fitting values obtained from the fitting of frequency shifts (n = 5) for Paraloid B72 film in contact with BR MEK 0%, H2O/MEK, BR MEK 20% and SDS MEK 20% (t = 150 s). a is the power law decay exponent of water/MEK and SDS MEK 20% systems (dimensionless) and

8 ACS Paragon Plus Environment

Page 9 of 11

Langmuir

b

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

is the slope of the line interpolating the frequency shifts in presence of BR MEK 0% and SDS MEK 20% (Hz/s). Sample

Fitting value

H2O/MEK

1.671 ± 0.011a

BR MEK 0%

0b

BR MEK 20%

1.805 ± 0.007 a

SDS MEK 20%

0b

4. Conclusions This paper shows that the combination of quartz crystal microbalance with dissipation monitoring and confocal laser scanning microscopy provides a powerful strategy to get valuable insights on the mechanism involved during the removal of hydrophobic coatings from hard surfaces. In particular, we addressed the removal of an acrylic co-polymer (Paraloid B72) with amphiphile-based formulations containing a nonionic surfactant (BR), water (a non solvent for the hydrophobic polymer here studied) and MEK (a good solvent for the employed polymer). The mentioned binary and ternary systems are in fact nanostructured fluids, whose excellent cleaning performances have been verified but were not completely understood4,5,7,33. The results reported in the present study unravel important details on the cleaning process. QCM-D monitors the interaction between the liquid phase and the polymeric coating at the nanoscale, while confocal laser microscopy imaging provides hints on the cleaning process at the meso/micro scale. The combination of these techniques allowed to hypothesize for the BR micellar solution an active and specific role of the surfactant, showing that its action, possibly weakens intra- and inter-chain interactions, induces a softening of the polymer film and promotes water and MEK within its thickness. The synergistic action of the nonionic surfactant, enhancing the polymer chain mobility and decreasing the interfacial energy at the swollen/bulk interfaces, and a good solvent therefore yields a highly effective polymer removal. The proposed mechanism is sketched in Figure 12.

A completely different mechanism is observed in the presence of a ternary system containing an anionic surfactant (SDS) in contact with Paraloid B72 films. This surfactant is unable to promote MEK penetration, thus making the SDS ternary system poorly efficient for polymer removal, even in the presence of relatively high amounts of good solvent. This study represents a step towards a deeper comprehension of the removal process from a surface of hydrophobic polymers with amphiphilic nanostructured fluids important in several fields as cleaning works of art, detergency and the relevant area of coatings. Moreover the comprehension of the mechanistic behavior of the complex fluid components will allow the development of a database of cleaning fluids with a tailored chemical composition for the substance that has to be removed from surface/interface..

AUTHOR INFORMATION Corresponding Authors Piero Baglioni: Email: [email protected] - Internet: http://www.csgi.unifi.it/; phone: +390554573033. Giovanni Marletta: Email: [email protected] - Internet: http://www.csgi.unifi.it/; phone: +390957385130.

1

Nicoletta Gimblanco present address: Institut Européen des Membranes, Université de Montpellier II CC 047, Place Eugène Bataillon, 34095 Montpellier cedex 5.

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

ACKNOWLEDGMENT This project was supported by CSGI, the European Union, Project NANORESTART (H2020-NMP-21-2014/646063) and by the Project DELIAS (PON-MIUR, Rome). N.G. thanks professor Diethelm Johannsmann (Clausthal University of Technology) for valuable discussions and comments. We also thank Prof. Massimo Bonini (Department of Chemistry and CSGI, University of Florence) for the help during AFM experiments.

REFERENCES (1)

(2)

(3)

Figure 12. Schematic representation of the proposed mechanism of polymer removal: 0) Polymer film on gold sensor before interaction with the liquid phase. 1) Morphological re-organization of the film induced by amphiphilic aggregates. 2) The good solvent penetrates through the film inducing a significant swelling of the coating. 3) The combined action of both surfactant and solvent promotes the detachment of the film from the surface.

(4)

Horie, C. V. Materials for Conservation: Organic Consolidants, Adhesives and Coatings, 2nd ed.; ButterworthHeinemann: Amsterdam ; Boston, 2010. Favaro, M.; Mendichi, R.; Ossola, F.; Russo, U.; Simon, S.; Tomasin, P.; Vigato, P. A. Evaluation of Polymers for Conservation Treatments of Outdoor Exposed Stone Monuments. Part I: Photo-Oxidative Weathering. Polym. Degrad. Stab. 2006, 91 (12), 3083–3096. Favaro, M.; Mendichi, R.; Ossola, F.; Simon, S.; Tomasin, P.; Vigato, P. A. Evaluation of Polymers for Conservation Treatments of Outdoor Exposed Stone Monuments. Part II: Photo-Oxidative and Salt-Induced Weathering of Acrylic– silicone Mixtures. Polym. Degrad. Stab. 2007, 92 (3), 335– 351. Baglioni, P.; Chelazzi, D.; Giorgi, R. Nanotechnologies in the Conservation of Cultural Heritage: A Compendium of Materials and Techniques; Springer, 2014.

9 ACS Paragon Plus Environment

Langmuir (5)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(6) (7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

(18)

(19)

(20)

(21)

Baglioni, M.; Rengstl, D.; Berti, D.; Bonini, M.; Giorgi, R.; Baglioni, P. Removal of Acrylic Coatings from Works of Art by Means of Nanofluids: Understanding the Mechanism at the Nanoscale. Nanoscale 2010, 2 (9), 1723–32. Baglioni, P.; Carretti, E.; Chelazzi, D. Nanomaterials in Art Conservation. Nat. Nanotechnol. 2015, 10 (4), 287–290. Baglioni, M.; Berti, D.; Teixeira, J.; Giorgi, R.; Baglioni, P. Nanostructured Surfactant-Based Systems for the Removal of Polymers from Wall Paintings: A Small-Angle Neutron Scattering Study. Langmuir 2012, 28 (43), 15193–15202. Domingues, J.; Bonelli, N.; Giorgi, R.; Baglioni, P. Chemical Semi-IPN Hydrogels for the Removal of Adhesives from Canvas Paintings. Appl. Phys. A 2014, 114 (3), 705–710. Carretti, E.; Dei, L.; G. Weiss, R. Soft Matter and Art Conservation. Rheoreversible Gels and beyond. Soft Matter 2005, 1 (1), 17–22. Krasicky, P. D.; Groele, R. J.; Rodriguez, F. Measuring and Modelling the Transition Layer during the Dissolution of Glassy Polymer Films. J. Appl. Polym. Sci. 1988, 35 (3), 641–651. Manjkow, J.; Papanu, J. S.; Hess, D. W.; Soong, D. S. S.; Bell, A. T. Influence of Processing and Molecular Parameters on the Dissolution Rate of Poly- ( Methyl Methacrylate ) Thin Films. 2003, 134 (8), 3–7. Ueberreiter, K. The Solution Process. In Diffusion in polymers; Crank J, P. G., Ed.; Academic Press: New York, 1968; pp 219–257. Raudino, M.; Selvolini, G.; Montis, C.; Baglioni, M.; Bonini, M.; Berti, D.; Baglioni, P. Polymer Films Removed from Solid Surfaces by Nanostructured Fluids: Microscopic Mechanism and Implications for the Conservation of Cultural Heritage. ACS Appl. Mater. Interfaces 2015, 7 (11), 6244– 6253. Johannsmann, D. The Quartz Crystal Microbalance in Soft Matter Research; Springer International Publishing: Berlin, 2015. Lu, H.-H.; Rao, Y. K.; Wu, T.-Z.; Tzeng, Y.-M. Direct Characterization and Quantification of Volatile Organic Compounds by Piezoelectric Module Chips Sensor. Sens. Actuators B Chem. 2009, 137 (2), 741–746. Xu, X.; Cang, H.; Li, C.; Zhao, Z. K.; Li, H. Quartz Crystal Microbalance Sensor Array for the Detection of Volatile Organic Compounds. Talanta 2009, 78 (3), 711–716. Schneider, T. W.; Frye-Mason, G. C.; Martin, S. J.; Spates, J. J.; Bohuszewicz, T. V.; Osbourn, G. C.; Bartholomew, J. W. Chemically Selective Coated Quartz-CrystalMicrobalance (QCM) Array for Detection of Volatile Organic Chemicals; 1998; Vol. 3539, pp 85–94. Giamblanco, N.; Conoci, S.; Russo, D.; Marletta, G. SingleStep Label-Free Hepatitis B Virus Detection by a Piezoelectric Biosensor. RSC Adv 2015, 5 (48), 38152–38158. Mechler, A.; Praporski, S.; Atmuri, K.; Boland, M.; Separovic, F.; Martin, L. L. Specific and Selective PeptideMembrane Interactions Revealed Using Quartz Crystal Microbalance. Biophys. J. 2007, 93 (11), 3907–3916. Marx, K. A. Quartz Crystal Microbalance: A Useful Tool for Studying Thin Polymer Films and Complex Biomolecular Systems at the Solution-Surface Interface. Biomacromolecules 2005, 4 (5), 1099–1120. Rösler, S.; Lucklum, R.; Borngräber, R.; Hartmann, J.; Hauptmann, P. Sensor System for the Detection of Organic Pollutants in Water by Thickness Shear Mode Resonators. Sens. Actuators B Chem. 1998, 48 (1–3), 415–424.

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

(34)

Page 10 of 11 Harbeck, M.; Erbahar, D. D.; Gürol, I.; Musluoğlu, E.; Ahsen, V.; Öztürk, Z. Z. Phthalocyanines as Sensitive Coatings for QCM Sensors Operating in Liquids for the Detection of Organic Compounds. Sens. Actuators B Chem. 2010, 150 (1), 346–354. Lieberzeit, P. A.; Afzal, A.; Rehman, A.; Dickert, F. L. Nanoparticles for Detecting Pollutants and Degradation Processes with Mass-Sensitive Sensors. Sens. Actuators B Chem. 2007, 127 (1), 132–136. Feiler, A. A.; Sahlholm, A.; Sandberg, T.; Caldwell, K. D. Adsorption and Viscoelastic Properties of Fractionated Mucin (BSM) and Bovine Serum Albumin (BSA) Studied with Quartz Crystal Microbalance (QCM-D). J. Colloid Interface Sci. 2007, 315 (2), 475–481. Tammelin, T.; Merta, J.; Johansson, L.-S.; Stenius, P. Viscoelastic Properties of Cationic Starch Adsorbed on Quartz Studied by QCM-D. Langmuir 2004, 20 (25), 10900–10909. Rodahl, M.; Höök, F.; Fredriksson, C.; Keller, C. A.; Krozer, A.; Brzezinski, P.; Voinova, M.; Kasemo, B. Simultaneous Frequency and Dissipation Factor QCM Measurements of Biomolecular Adsorption and Cell Adhesion. Faraday Discuss. 1997, No. 107, 229–46. Weerawardena, A.; Drummond, C. J.; Caruso, F.; Mccormick, M. Real Time Monitoring of the Detergency Process by Using a Quartz Crystal Microbalance †. Science 1998, 7463 (14), 575–577. Weerawardena, A.; Boyd, B. J.; Drummond, C. J.; Furlong, D. N. Removal of a Solid Organic Soil from a Hard Surface by Glucose-Derived Surfactants: Effect of Surfactant Chain Length, Headgroup Polymerisation and Anomeric Configuration. Colloids Surf. Physicochem. Eng. Asp. 2000, 169 (1– 3), 317–328. Gotoh, K. The Role of Liquid Penetration in Detergency of Long-Chain Fatty Acid. J. Surfactants Deterg. 2005, 8 (4), 305–310. Stålgren, J. J. R.; Claesson, P. M.; Wärnheim, T. Adsorption of Liposomes and Emulsions Studied with a Quartz Crystal Microbalance. Adv. Colloid Interface Sci. 2001, 89–90, 383– 394. Höök, F.; Kasemo, B.; Nylander, T.; Fant, C.; Sott, K.; Elwing, H. Variations in Coupled Water, Viscoelastic Properties, and Film Thickness of a Mefp-1 Protein Film during Adsorption and Cross-Linking: A Quartz Crystal Microbalance with Dissipation Monitoring, Ellipsometry, and Surface Plasmon Resonance Study. Anal. Chem. 2001, 73 (24), 5796–5804. Voinova, M. V.; Rodahl, M.; Jonson, M.; Kasemo, B. Viscoelastic Acoustic Response of Layered Polymer Films at Fluid-Solid Interfaces: Continuum Mechanics Approach. Phys. Scr. 1999, 59 (5), 391. Baglioni, M.; Raudino, M.; Berti, D.; Keiderling, U.; Bordes, R.; Holmberg, K.; Baglioni, P. Nanostructured Fluids from Degradable Nonionic Surfactants for the Cleaning of Works of Art from Polymer Contaminants. Soft Matter 2014, 10 (35), 6798–809. Cox, M. F. Surfactants for Hard-Surface Cleaning: Mechanisms of Solid Soil Removal. J. Am. Oil Chem. Soc. 1986, 63 (4), 559–565.

10 ACS Paragon Plus Environment

Page 11 of 11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

SYNOPSIS TOC The mechanism and the kinetics of the removal process of polymeric coatings from solid surfaces by means of amphiphilic formulations are unraveled through the combination of QCM-D and confocal microscopy.

ACS Paragon Plus Environment

11