pubs.acs.org/Langmuir © 2010 American Chemical Society
Contribution Toward Comprehension of Contact Angle Values on Single Polydimethylsiloxane and Poly(ethylene oxide) Polymer Networks Murielle Bouteau, Sophie Cantin,* Odile Fichet, Franc-oise Perrot, and Dominique Teyssie Laboratoire de Physico-Chimie des Polym eres et des Interfaces (LPPI, EA 2528), Institut des Mat eriaux, Universit e de Cergy-Pontoise, 5 mail Gay-Lussac Neuville/Oise, 95000 Cergy-Pontoise Cedex, France Received June 11, 2010. Revised Manuscript Received August 31, 2010 The large application ranges of polydimethylsiloxane (PDMS) and poly(ethylene oxide) (PEO) based materials justify the importance of controlling polymer surface properties including morphology and wettability behavior. However, it appears that the reported contact angle values of PDMS surfaces show significant scattering which cannot always be interpreted in terms of sole chemical data. In addition, few values are reported concerning pure PEO surfaces, since the polymer generally swells in the presence of water. Thus, in order to correlate surface properties with sample preparation, several single PDMS and PEO polymer networks were synthesized with varying cross-linkers and different cross-linking densities. First, the sample surface topography was systematically analyzed by atomic force microscopy (AFM). It was proven that the removal process of the polymer film from the mold plays a significant role in surface topography according to the vitreous or rubbery state of the given polymer network at room temperature irrespective of mold surface treatment. AFM-scale smooth surfaces can be obtained for all the samples by removing them systematically from the mold at a temperature below the R-relaxation temperature. Dynamic water contact angles were then measured and the values analyzed as a function of cross-linker nature and cross-linking density.
I. Introduction Surface properties of polymers are important for many applications such as adhesion, coating, friction, and biocompatibility. Static contact angle measurements are widely used to characterize polymer surfaces. Indeed, the material surface properties, such as surface roughness, can considerably emphasize hydrophobic or hydrophilic character.1 On such nonideal surfaces, some scattering of static contact angle values can often be observed. Thus, dynamic contact angle measurements are more suitable for assessing wetting behavior when surfaces display chemical or physical heterogeneities.1 Indeed, the contact angle hysteresis value defined as the difference between advancing (θa) and receding (θr) contact angles can provide more information than the static contact angle, notably the roughness characteristics. Due to an attractive combination of properties including biocompatibility and low surface energy, polydimethylsiloxane (PDMS) elastomers are among the most widely used classes of coating polymers. In particular, intensive research is carried out for the development of PDMS-based microfluidic devices incorporating surface micropatterning for guiding fluids. Despite many advantageous surface properties, tuning of the PDMS surface behavior is frequently necessary to fulfill some requirements allowing successful applications.2 The design of sensitive PDMS surfaces bearing the desired functionality is thus the goal of many research investigations. For example, the coating of poly(ethylene oxide) (PEO) polymers onto PDMS is extensively studied as a strategy to make the channels more hydrophilic and foulingresistant3 in microfluidic devices. Indeed, PEO is well-known to *Corresponding author. E-mail:
[email protected]. (1) Callies, M.; Quere, D. Soft Matter 2005, 1, 55. (2) Efimenko, K.; Wallace, W. E.; Genzer, J. J. Colloid Interface Sci. 2002, 254, 306. (3) Klasner, S. A.; Metto, E. C.; Roman, G. T.; Culbertson, C. T. Langmuir 2009, 25, 10390. (4) Dalsin, J.; Lin, L.; Tosatti, S.; Voros, J.; Textor, M.; Messersmith, P. B. Langmuir 2005, 21, 640.
Langmuir 2010, 26(22), 17427–17434
allow resistance to protein adsorption and bacterial adhesion.4,5 Moreover, as PEO-based materials swell significantly in water, they have gained much attention as fouling-resistant coatings for water purification membranes.6 Amphiphilic PEO polymer networks are often combined with hydrophobic polymers such as fluoropolymers.7 The swelling in polar solvents makes them infrequently studied as pure material, and very few contact angle measurements are reported on crude PEO. This large range of applications for PDMS- and PEO-based materials justifies the importance of controlling polymer surface properties including morphology and wettability behavior. PDMS samples used for water contact angle measurements can be broadly classified as follows: (a) PDMS fluids adsorbed onto solids such as glass or metals,8,9 (b) cross-linked PDMS coatings on substrates,10 and (c) PDMS elastomer surfaces synthesized inside a mold.11 The range of contact angle data reported in the different studies is considerable: on the order of 20° for water (95-113°), for example. When a PDMS film is adsorbed onto a rigid substrate, the maximum hydrophobicity effect is not reached. Indeed, a thermal baking treatment is required to develop the familiar highly waterrepellent character. For example, when films of 500 centistokes PDMS are first formed on glass by dipping in benzene solution, water contact angles between 50° and 60° are initially measured, whereas values greater than 100° are obtained by curing samples to 200 °C.12 The exact causes of this behavior are still not fully (5) Groll, J.; Amiregoulova, E. V.; Ameringer, T.; Heyes, C. D.; Rocker, C.; Nienhaus, U.; Moller, M. J. Am. Chem. Soc. 2004, 126, 4234. (6) Sagle, A. C.; Ju, H.; Freeman, B. D.; Shara, M. M. Polymer 2009, 50, 756. (7) Hu, Z.; Chen, L.; Betts, D. E.; Pandya, A.; Hillmyer, M. A.; DeSimone, J. M. J. Am. Chem. Soc. 2008, 130, 14244. (8) Zisman, W. A. In Adhesion and Cohesion; Weiss, P., Ed.; Elsevier Publishing Company: New York, 1962; p 201. (9) Owens, D. K.; Wendt, R. C. J. Appl. Polym. Sci. 1969, 13, 1741. (10) Gordon, D. J.; Colquhoun, J. A. Adhesives Age 1976, 19, 21. (11) Chaudhury, M. K.; Whitesides, G. M. Langmuir 1991, 7, 1013. (12) Hunter, M. J.; Gordon, M. S.; Barry, A. J.; Hyde, J. F.; Heidenreich, R. D. Ind. Eng. Chem. 1947, 39, 1389.
Published on Web 10/04/2010
DOI: 10.1021/la102384s
17427
Article
understood. In the case of PDMS-containing block and graft copolymers, a large enrichment of the surface in PDMS with respect to the bulk concentration is expected owing to the low surface energy of PDMS and the good flexibility of the chains. Studies carried out on cast films led to the conclusion that the topmost airside layer of copolymer containing 40% PDMS or more in bulk is made up of practically pure PDMS.13 Two cross-linked PDMS coatings are mainly studied and understood: tin-catalyzed systems based on the condensation of silanol and alkoxysilyl functionalities, and platinum-catalyzed systems based on hydrosilylation addition of SiH to vinyl functional siloxanes. The surface characterization of these last coatings shows a large range of contact angle values. For example, on one hand, the surface properties of Sylgard 184, a Dow Corning silicone elastomer, were widely studied. The water static contact angle measurements carried out by different authors with this precursor are quite dispersed and seem to depend on the thermal treatment chosen for the synthesis. Indeed, the static contact angle values vary from 109° for a cure at 70 °C for 1 h14 to 117° ( 2° for a cure at 85 °C for 4.5 h,15 respectively. On the other hand, single PDMS networks were also synthesized starting from divinylterminated PDMS macromonomers via a hydrosilylation reaction catalyzed by platinum with methyltris(dimethylsiloxy)silane as a cross-linker, at room temperature. The advancing (θa) and receding (θr) water contact angles measured on these PDMS networks are θa = 118° ( 2° and θr = 83° ( 2°.16 Perutz et al. obtained θa = 118° ( 2° and θr = 90° ( 2° on similar PDMS networks extracted with toluene.17,18 Divinyl-terminated PDMS macromonomers were also cross-linked with tetrakis(dimethylsiloxy)silane leading to θa and θr values of 107° and 102°, respectively.19 Finally, with respect to coatings on substrates, cross-linking of polymers inside a mold should ensure the stability of both bulk and surface morphologies. Moreover, no contamination of the liquid drop by polymer chains should occur.16 However, the chemical composition of the material surface depends on the mold chemical nature20 that further reduces confidence in the data. Single PDMS tin-catalyzed networks were formed by dibutyltin dilaurate catalyzed addition between the hydroxy end groups of R,ω-(3-hydroxypropyl)polydimethylsiloxane and a pluri-isocyanate cross-linker.21 These networks show a static water contact angle of 114° ( 2°, which is comparable to that measured on PDMS networks prepared via a sol-gel method from R,ω-dihydroxy PDMS cured with tetraethoxysilane.22 The several examples reported here show that advancing and receding water contact angles depend on the cross-linker chemical nature. It appears that, although the values are within the same order of magnitude, they can be considered as relatively dispersed, i.e., between 107° and 118°. This is particularly clear for receding (13) Belorgey, G.; Sauvet, G. Organosiloxane block and graft copolymer. In Silicon-containing polymers, Jones, R. G., Ando, W, Chojnowski, J, Eds.; Kluwer Academic Publisher: Dordrecht, 2000. (14) Hu, S.; Ren, X.; Bachman, M.; Sims, C. E.; Li, G. P.; Allbritton, N. Anal. Chem. 2002, 74, 4117. (15) Hellmich, W.; Regtmeier, J.; Duong, T. T.; Ros, R.; Anselmetti, D.; Ros, A. Langmuir 2005, 21, 7551. (16) Uilk, J. M.; Mera, A. E.; Fox, R. B.; Wynne, K. J. Macromolecules 2003, 36, 3689. (17) Perutz, S.; Kramer, E. J.; Baney, J.; Hui, C. Y. Macromolecules 1997, 30, 7964. (18) Perutz, S.; Kramer, E. J.; Baney, J.; Hui, C. Y.; Cohen, C. J.Polym. Sci., Part B: Polym. Phys. 1998, 36, 2129. (19) Li, Z.; Han, W.; Kozodaev, D.; Brokken-Zijp, J. C. M.; de With, G.; Th€une, P. C. Polymer 2006, 47, 1150. (20) Lipatov, Y. S. Prog. Polym. Sci. 2002, 27, 1721. (21) Darras, V.; Fichet, O.; Perrot, F.; Boileau, S.; Teyssie, D. Polymer 2007, 48, 687. (22) Uilk, J.; Bullock, S.; Johnston, E.; Myers, S. A.; Merwin, L.; Wynne, K. J. Macromolecules 2000, 33, 8791.
17428 DOI: 10.1021/la102384s
Bouteau et al.
contact angle values.16,17,19,23 This dispersion can be due to the use of different chemical parameters such as cross-link density or cross-linker nature, in the various reported studies. In particular, Wynne et al.24 proved some possible effects of the cross-link density. Indeed, the synthesis conditions may, in particular, modify both surface topography and bulk properties. A correlation between surface properties and sample preparation, which has not been systematically demonstrated to our knowledge, is necessary for the rational design of materials showing controlled surface properties. AFM provides a direct display of the surface topography of the samples and may thus be essential to analyze the results of dynamic contact angle measurements. To our knowledge, even though contact angle measurements are frequently used to characterize single polymer networks, quite a few studies simultaneously have shown an investigation of the polymer network surfaces by AFM.7,15,19,25-27 In this work, two series of polymer network surfaces, widely used for their attractive surface properties, were investigated. Indeed, single polyethylene oxide (PEO) networks and single polydimethylsiloxane (PDMS) networks were synthesized in order to identify which parameters may affect water contact angle values measured on such materials. In particular, cross-linker nature and cross-link density effects on the surface properties were studied. Thus, the material bulk properties were systematically characterized by dynamic mechanical thermal analysis (DMA), while the surface topography was examined by atomic force microscopy (AFM) and correlated to advancing and receding water contact angle measurements.
II. Experimental Setup and Materials Materials. Poly(ethylene glycol) dimethacrylate oligomers with various molar weights (PEGDM, Mn = 875, 550, and 330 g 3 mol-1) and poly(ethylene glycol) methacrylate (PEGM, Mn = 300 g 3 mol-1) were provided by Acros. R,ω-(3-Hydroxypropyl)poly(dimethylsiloxane) (dihydroxyPDMS) (Mn = 1150 g 3 mol-1 determined by 1H NMR) and R,ωacrylate poly(dimethylsiloxane) (diacrylate-PDMS) (Mn = 1680 g 3 mol-1 determined by 1H NMR) were kindly provided by Rhodia and Tego, respectively. Dihydroxy-PDMS was dried under vacuum before use. R,ω-Divinyl poly(dimethylsiloxane) (divinyl-PDMS, Mn =1150 g 3 mol-1 determined by 1H NMR, ABCR) was used as received. Dibutyltin dilaurate (DBTDL) (Aldrich), pentaerythritol tetrakis(3-mercaptopropionate) (Tetrathiol in the text, Aldrich), and Desmodur N3300 (Bayer) (NCO content by weight: 21.8 ( 0.3% according to the supplier) were used as received. This last compound is described as an isocyanurate mixture resulting from the condensation of three to several hexamethylene diisocyanate molecules and mainly composed of mono-, di-, and triisocyanurates with a global functionality higher than 2.28 Thus, mere “tri(6-isocyanatohexyl)isocyanurate” is not a proper description, and the compound is referred to as an isocyanate cross-linker. Dicyclohexylperoxydicarbonate (DCPD, Groupe Arnaud) was used as received. Azobis isobutyronitrile (AIBN, Acros) was recrystallized in methanol before use. Toluene, dichloromethane, and chloroform (puro, Carlo Erba) were distilled and dried before use. (23) Extrand, C. W.; Kumagai, Y. J. Colloid Interface Sci. 1996, 184, 191. (24) Wynne, K. J.; Ho, T.; Johnston, E. E.; Myers, S. A. Appl. Organomet. Chem. 1998, 12, 763. (25) Dou, Q.; Wang, C.; Cheng, C.; Han, W.; Th€une, P. C.; Ming, W. Macromol. Chem. Phys. 2006, 207, 2170. (26) Liu, L.; Sheardown, H. Biomaterials 2005, 26, 233. (27) Bullock, S.; Johnston, E. E.; Willson, T.; Gatenholm, P.; Wynne, K. J. J. Colloid Interface Sci. 1999, 210, 18. (28) Nia, H; Aaserudb, D. J.; Simonsick, W. J., Jr.; Souceka, M. D. Polymer 2000, 41, 57.
Langmuir 2010, 26(22), 17427–17434
Bouteau et al.
Article
Single Network Synthesis. Single network starting from diacrylate-PDMS oligomer (PDMS-A network in the text) was prepared as follows: 1 g diacrylate-PDMS, 30 mg AIBN (3% by weight with respect to diacrylate-PDMS oligomer), and few drops of chloroform were stirred together under argon atmosphere for 30 min at room temperature. The mixture was then poured into a mold made from two glass plates clamped together and sealed with a 1 mm thick Teflon gasket. The mold was then kept at 50 °C for 16 h. After 16 h, the sample was removed from the mold, kept for 1 h at room temperature, and then dried for 2 h at 50 °C under vacuum. After complete solvent evaporation, the sample was postcured for 3 h at 110 °C. PDMS networks starting from divinyl-PDMS29 and dihydroxy-PDMS30 oligomers have been synthesized with procedures previously described. Single polyethylene oxide (PEO) network preparation consists of dissolving 0.03 g DCPD (3% by weight with respect to PEGDM oligomer) in 1 g PEGDM oligomer with various molar weights (Mn: 875, 550, 330 g 3 mol-1). The mixture was degassed under argon and then poured into a mold as described in the PDMS-A network synthesis. The mold was then kept at 50 °C for 16 h. The postcure was carried out for 1 h at different temperatures following the PEGDM (at 80 °C for PEGDM 550 and 875 and at 110 °C for PEGDM 330). All PEO networks are reported as PEOm, the number following PEO corresponding to the molar weight of the PEGDM oligomer used for the network synthesis. The cross-link densities of all the networks shown in this study are given as the weight proportion of the cross-linker to the total weight of the network components. Indeed, the final cross-link densities in the PDMS sample series are quite close as far as the molar proportions of cross-links to polymer are concerned. However, in these cases the molecular weights of the different cross-linking agents differ widely, i.e., Mn Desmodur is ∼1100 g 3 mol-1, whereas the acrylate- and methacrylate-terminated oligomers do not need any added cross-linking agent and the molar mass of the cross-linking group is about 71 to 85 g 3 mol-1. In order to vary the cross-link density, side chains were included into the PEO330 network. These networks are also reported as PEGMx, where x corresponds to the weight ratio of the oligomers, i.e., the side chain content in the network. Thus, a PEGM network obtained from a mixture of 0.75 g of PEGDM and 0.25 g of PEGM will be noted PEGM0.25. All PEGM networks were synthesized by dissolving 0.03 g DCPD (3% by weight with respect to PEGDM and PEGM oligomers) in a mixture of x g PEGDM oligomer (Mn: 330 g 3 mol-1) and (1 - x) g PEGM. The material was then synthesized following the same procedure as the PEO network. Molds. The mold used for each network synthesis was formed with two new glass plates (RS France). Prior to use, these plates were carefully cleaned in a beaker with ultrapure water (Millipore -18 MΩ.cm) placed in an ultrasonic bath for at least 20 min at room temperature and were then dried with argon. Soluble Fraction. In order to determine the soluble fraction amount contained in materials and thus the extent of covalent bond formation in the network, single networks were extracted in a Soxhlet with dichloromethane which is a common solvent of all the precursors for 48 h. After extraction, the sample was dried under vacuum and then weighed. The soluble fraction (SF) is given as a weight percentage: SFð%Þ ¼
ðW0 - WE Þ 100 W0
where W0 and WE are the sample weights before and after extraction, respectively. Dynamic Mechanical Thermal Analysis (DMA). Dynamic mechanical thermal analysis (DMA) measurements were carried (29) Fichet, O.; Vidal, F.; Laskar, J.; Teyssie, D. Polymer 2005, 46, 37. (30) Darras, V.; Fichet, O.; Teyssie, D.; Boileau, S. Polym. Prep. 2004, 45, 680.
Langmuir 2010, 26(22), 17427–17434
out on a film sample with a Q800 apparatus (TA Instruments) operating in tension mode. Experiments were performed at a 1 Hz frequency and a heating rate of 3 °C min-1 from -140 to 150 °C. Typical dimensions of the samples were 30 mm 8 mm 1 mm. The setup provides the storage and loss modulus (E 0 and E 00 ) and the damping parameter or loss factor (tan δ). All storage modulus values were normalized at 3000 MPa on the vitreous plateau in order to compare the relative evolutions with temperature. AFM. The atomic force microscopy (AFM) experiments were performed with a Nanoscope IIIA Dimension 3100 microscope from Digital Instruments. Measurements were carried out in air at room temperature. Standard silicon cantilevers (Digital Instruments) were used to acquire images in “tapping” mode near the resonant frequency close to 300 kHz. The images were obtained with the height mode and a 256 256 dots resolution. For each sample, AFM images were recorded on at least four samples and in several places in order to check the reproducibility of the displayed images. The vertical heights were measured by cross-section analysis. The root-mean-square roughness Rq was measured on 50 μm 50 μm images, and the results are indicated as an average over the values obtained on four images. Phase mode images were also recorded by systematically using two free amplitudes of the oscillator (10 and 50 nm) and several force levels corresponding to set-point ratios between 0.4 and 0.9. Contact Angle Measurements. Advancing and receding contact angles were measured using the drop shape analysis profile device equipped with a tiltable plane (DSA-P, Kruss, Germany). An ultrapure water drop (Millipore, resistivity: 18 MΩ.cm) was first deposited on the sample using a variablevolume micropipet. The drop volume was set to 25 μL for all experiments. The static corresponding contact angle was measured by means of a Young-Laplace drop profile fitting. The water surface tension γ was also deduced in order to check systematically that no water contamination by polymer residues occurs during polymer network exposure. It was also checked that the polymer network surfaces are not modified by the drop deposition. Indeed, after surface drying, new contact angle measurements lead to the same results. In order to perform dynamic contact angle measurements, the polymer network surface sustaining the water drop was tilted at a constant speed (1 deg.s-1) and the images of the drop simultaneously recorded. The advancing and receding water contact angles were measured at the front and rear edges of the drop, respectively, just before the triple line starts moving. The angles were obtained using the tangent of the drop profile at the triple line. For each network composition, contact angles were measured on about ten samples resulting from different syntheses; six drops per network were analyzed. The reported contact angle values correspond to the average of all measurements with an error bar corresponding to twice the standard deviation.
III. Results and Discussion The polyethylene oxide (PEO) based network series were synthesized under similar experimental conditions in order to identify which parameters may affect the surface morphology and the resulting water contact angle values. This study was also supplemented by the characterization of three types of poly(dimethylsiloxane) networks. In order to make the samples comparable, they were all elaborated inside a glass mold. Particular attention was brought to the release process of the samples from the mold. In order to avoid a potential contamination of the network surfaces by a release agent, the glass plates making up the mold were not treated with trichloromethylsilane as often reported. Indeed, this treatment turns out to be of no effect in modifying the material surface in this study, as shown below. In addition, crosslinker nature and cross-link density effects were studied. Thus, the material surfaces were systematically characterized by AFM and DOI: 10.1021/la102384s
17429
Article
the observations correlated to advancing and receding water contact angle measurements (DCA). In parallel, bulk properties of the networks were mainly determined by DMA (no glass transition temperatures were detected by DSC measurements). This systematic study would hopefully provide the connection between sample composition and wetting properties. In the first part, the role of experimental conditions in elastomer sample elaboration in view of surface analysis is highlighted. Then, the surface properties of PEO and PDMS networks are displayed. Role of Experimental Conditions in Sample Preparation for Surface Analysis. In a first step, we investigated the surface properties of a series of PEO networks synthesized by free-radical polymerization of poly(ethylene glycol) dimethacrylate (PEGDM) oligomers with two different molar weights, 875 and 550 g 3 mol-1, corresponding to 16 and 9 ethylene glycol units between both methacrylate end chain groups. With PEGDM being difunctional, the cross-linking reaction occurs naturally and only DCPD is required as initiator to the exclusion of added cross-linking agent. The proportion of cross-links can be estimated at 19.4 and 30.9 wt % for PEO875 and PEO550, respectively. In order to determine the soluble fraction content and the quality of these single networks, the samples were extracted in a Soxhlet leading to a very low soluble fraction amount, i.e., about 2%. These insoluble transparent materials can be thus considered correctly cross-linked. Figure 1 displays AFM images obtained on both PEO-based networks. The PEO875 network shows a surface with many topographical defects (Figure 1a). The lateral diameter can exceed 2 μm while the height can reach 350 nm. In contrast, the surface of PEO550 networks (Figure 1b) appears very smooth. Indeed, the measured root-mean-square roughness Rq is only 4 nm, whereas it reaches 58 nm for PEO875 networks.
Figure 1. AFM images on a 50 μm 50 μm scale obtained on two different PEO-based networks: (a) PEO875, (b) PEO550. The z-scale is 100 nm.
Bouteau et al.
In order to identify the origin and the chemical nature of the defects appearing on the elastomer surface, AFM images were performed in phase mode. Independent of the imaging parameters used, no contrast between the aggregates and the smooth surface was detected on the PEO875 network surface. This means that the topographical defects possess the same chemical nature as the surrounding phase and do not result from surface contamination. In addition, it should be emphasized that, with the extracted content being very low, these defects are not created by adsorbed un-cross-linked polymer chains. Then, since a change in the chain length of the PEGDM oligomer modifies the cross-link density of the PEO network and thus the mechanical properties, the mechanical R-relaxation temperature TR was measured by means of DMA. It was determined at -21 and þ21 °C for the PEO875 and PEO550 networks, respectively (Figure 2a). In addition, the storage modulus E0 at 20 °C provided by the DMA setup was shown to be about six times higher for PEO550 networks (close to 190 MPa) than for PEO875 samples (close to 30 MPa) (Figure 2b). At room temperature, the PEO875 network is thus as an elastomer, whereas the PEO550 network, although with a modulus higher than 100 MPa, is within the mechanical relaxation process. The topographical defects evidenced by means of AFM could thus result from sticking points of the soft polymers to the mold, leading to a local stretching when the polymer network is turned out after the synthesis. To confirm this hypothesis, the glass plate surfaces were observed by AFM after turning the networks out. Figure 3 shows an image of a glass plate used as a mold for a PEO875 elastomer network: many polymer residues are detected, in agreement with the presence of sticking points. In addition, to check this hypothesis, PEO875 elastomer networks were removed from the mold after ten minutes immersion in liquid nitrogen. The polymer networks were thus cooled down below their R-relaxation temperature. As shown in Figure 3b, a significantly lower proportion of topographical defects is observed on the network surface, while the images of the glass plates no longer display polymer residues. These results clearly show the care needed to prepare soft materials into molds for surface characterizations. To confirm the effect of the temperature at which the polymer networks are removed from the glass molds on the surface topography, the well-known and well-characterized elastomer polymer, i.e., PDMS, was investigated. PDMS networks were synthesized starting from telechelic PDMS oligomers with two different end chain functions (acrylate, vinyl). Single PDMS-A networks were obtained by free-radical cross-linking reaction of R,ω-acrylate poly(dimethylsiloxane) (Mn = 1680 g 3 mol-1)
Figure 2. Loss tangent (a) and storage modulus (b) of PEO875 (dotted line) and PEO550 (solid line) networks. 17430 DOI: 10.1021/la102384s
Langmuir 2010, 26(22), 17427–17434
Bouteau et al.
Article
Figure 3. AFM images on a 50 μm 50 μm scale obtained on a glass plate used as a mold for a PEO875 network turned out from the mold (a) at room temperature and (b) after ten minutes immersion in liquid nitrogen. The z-scale is 100 nm.
without any cross-linking agent. Indeed, with this oligomer being difunctional, the cross-linking reaction occurs naturally and only AIBN is required as an initiator to the exclusion of any added cross-linking agent. Cross-links are ester groups, and their proportion in the network can be estimated to 8.4 wt %. R,ω-Divinyl poly(dimethylsiloxane) (Mn = 1150 g 3 mol-1) does not react by free-radical polymerization; thus, single PDMS-V networks were synthesized by the thiol-ene addition of thiol groups of a tetrathiol cross-linker on PDMS vinyl end chain functions. The cross-link is a S-C bond, while the cross-linker proportion is close to 18.7 wt %. With the molar weight of these different PDMS oligomers being similar, the molar cross-link density of the resulting networks can be considered constant. However, due to the molecular weight of the different cross-linkers used, the weight proportions in the network are different. These samples contain a very low soluble fraction amount, below 5%, mainly composed of PDMS chains as checked by 1H NMR. These single PDMS networks can thus be considered correctly cross-linked. DMA measurements were then performed on these two PDMS networks. The resulting R-relaxation temperatures (TR) are equal to -105 and -70 °C, while the measured storage moduli E0 at 20 °C are close to 6 and 1 MPa for PDMS-A and PDMS-V networks, respectively. This is in agreement with the fact that the nature and molecular weight of the two cross-linkers are very different, i.e., single acrylate function in the first case or tetrathiol cross-linker in the second case. Indeed, as already described by Clarson et al., the R-relaxation temperature of PDMS networks clearly depends on the cross-linker nature, while it hardly varies with the PDMS oligomer weight.31 Finally, since both TR values are below the room temperature, the synthesized PDMS networks are in an elastomer state at room temperature regardless of the cross-linker nature. Figure 4 shows AFM images of the two PDMS network surfaces for samples removed from the glass mold either at room temperature, i.e., in an elastomer state, or after ten minutes immersion in liquid nitrogen, i.e., in a vitreous state. For samples removed from the mold at room temperature (Figure 4a,c), the whole surfaces are smooth but covered with many topographical defects whose average height is close to 200 nm for PDMS-A and 280 nm for PDMS-V. In contrast, samples turned out from the mold at a temperature below the R-relaxation temperature display smooth surfaces without any defects due to the presence of sticking points of the soft polymer to the mold (Figure 4b,d). Indeed, the measured root-mean-square roughness Rq decreases significantly, from about 70 to 3 nm for PDMS-V networks and from about 47 to 4 nm for PDMS-A samples. These observations (31) Clarson, S. J.; Mark, J. E.; Dodgson, K. Polym. Comm. 1988, 29, 208.
Langmuir 2010, 26(22), 17427–17434
Figure 4. AFM images on a 50 μm 50 μm scale obtained on PDMS-A (a,b) and PDMS-V (c,d) polymer networks, for samples removed from the glass mold at room temperature (a,c) or after ten minutes immersion in liquid nitrogen (b,d). The z-scale is 100 nm.
Figure 5. AFM image on a 10 μm 10 μm scale obtained on a PDMS-V polymer network, synthesized inside a glass mold treated with trichloromethylsilane and removed from the mold at room temperature. The z-scale is 100 nm.
are in good agreement with the low roughness of PDMS-V networks cross-linked with tetrakis(dimethylsiloxy)silane and coated on a surface instead of being poured into a mold19 or Sylgard 184 membranes exposed to air during preparation.26 These results confirm that the way elastomers are removed from the mold is crucial in view of surface topography analysis and thus wettability studies. In order to check that this phenomenon cannot be avoided by pretreatment of the glass mold with trichloromethylsilane, a PDMS-V network was synthesized into a treated glass mold and turned out from it at room temperature. The image of the network surface shown in Figure 5 highlights the same defects as on the samples synthesized into nontreated glass molds. Thus, the pretreatment of the mold does not appear efficient in this case. Then, the influence of surface topography on wetting properties was investigated. Advancing and receding contact angles were thus measured on these PDMS samples (Table 1). Considering the advancing contact angles θa, similar values were measured on the two different PDMS networks, independently of the temperature at which the polymers are turned out. No significant effect of the cross-linker or PDMS end-chain function is obvious. The measured values are in good agreement with those reported for PDMS networks synthesized starting from dihydroxy-PDMS cross-linked with tetraethyl orthosilicate27,19 or metyltriethoxysilane24 or with a poly(isocyanate),25 as well as from divinyl-PDMS cross-linked with a tetrakis(dimethylsiloxy)silane19 or a methyl hydrosiloxane.2 DOI: 10.1021/la102384s
17431
Article
Bouteau et al.
In contrast, the receding contact angle θr is significantly affected by the presence of topographic defects. Indeed, receding contact angles are close to each other for the two PDMS samples but strongly depend on the way the polymer is removed from the mold. Significantly lower θr values are measured on PDMS samples turned out at room temperature than on those turned out in liquid nitrogen. The contact angle hysteresis (θa - θr), which depends here on surface roughness, is thus lower, in agreement with the observation of smoother surfaces by means of AFM. This confirms the strong influence of surface topography on contact angle hysteresis. Thus, the way the elastomer samples are removed from the mold clearly influences surface topography as well as contact angle hysteresis. In the following, the entire elastomer networks were turned out after immersion in liquid nitrogen in order to avoid local pulling out of the material. Characterization and Wetting Properties of a Series of PEO and PDMS Polymer Networks. The surface analysis of single polymer networks was carried on with the study of, on one hand, PDMS networks synthesized starting from telechelic PDMS oligomers with hydroxy end chain functions (PDMSOH network) and, on the other hand, a series of PEO networks differing by the cross-linking density. Table 1. Advancing and Receding Contact Angles Recorded on PDMS-A and PDMS-V Networks, Removed from the Mold at Room Temperature (RT) or below the r-Relaxation Temperature (
