Langmuir 2008, 24, 7995-8000
7995
Effect of the Layer-by-Layer (LbL) Deposition Method on the Surface Morphology and Wetting Behavior of Hydrophobically Modified PEO and PAA LbL Films Jinhwa Seo,† Jodie L. Lutkenhaus,‡ Junoh Kim,† Paula T. Hammond,*,‡ and Kookheon Char*,† Center for Functional Polymer Thin Films and School of Chemical and Biological Engineering, Seoul National UniVersity, Seoul 151-744, Korea, and Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 ReceiVed March 24, 2008. ReVised Manuscript ReceiVed April 28, 2008 We demonstrate that the surface morphology and surface-wetting behavior of layer-by-layer (LbL) films can be controlled using different deposition methods. Multilayer films based upon hydrogen-bonding interactions between hydrophobically modified poly(ethylene oxide) (HM-PEO) and poly(acrylic acid) (PAA) have been prepared using the dip- and spin-assisted LbL methods. A three-dimensional surface structure in the dip-assisted multilayer films appeared above a critical number of layer pairs owing to the formation of micelles of HM-PEO in its aqueous dipping solution. In the case of spin-assisted HM-PEO/PAA multilayer films, no such surface morphology development was observed, regardless of the layer pair number, owing to the limited rearrangement and aggregation of HM-PEO micelles during spin deposition. The contrasting surface morphologies of the dip- and spin-assisted LbL films have a remarkable effect on the wetting behavior of water droplets. The water contact angle of the dip-assisted HMPEO/PAA LbL films reaches a maximum at an intermediate layer pair number, coinciding with the critical number of layer pairs for surface morphology development, and then decreases rapidly as the surface structure is evolved and amplified. In contrast, spin-assisted HM-PEO/PAA LbL films yield a nearly constant water contact angle due to the surface chemical composition and roughness that is uniform independent of layer pair number. We also demonstrate that the multilayer samples prepared using both the dip- and spin-assisted LbL methods were easily peeled away from any type of substrate to yield free-standing films; spin-assisted LbL films appeared transparent, while dip-assisted LbL films were translucent.
Introduction Layer-by-layer (LbL) deposition has been known to be one of the simple and versatile methods to prepare functional multilayered films.1,2 In conventional dip-assisted LbL assembly, a substrate is alternately dipped in aqueous solutions of oppositely charged polyelectrolytes1,2 (or hydrogen-bond donors and acceptors3–5), with rinsing steps between exposures; species diffuse and adsorb to the substrate from aqueous solution via molecular interactions, followed by molecular rearrangement along the surface. The process of diffusion, adsorption, and rearrangement may take from a few to twenty minutes, or longer, per exposure depending on macromolecular size, charge, and mobility. Over the past years, other LbL assembly techniques based on different assembly mechanisms have been introduced such as spin-assisted LbL assembly6,7 as well as spray-assisted LbL assembly.8–11 In contrast to the dip-assisted LbL method, multilayer films based on the spin-assisted LbL method are constructed by intermolecular interactions between two materials as well as by air shear force or centrifugal force, which influences the surface roughness and internal layer structure. The spin-assisted LbL assembly technique has received much attention for its own merits: minimizing solvent * To whom correspondence should be addressed. E-mail: khchar@ plaza.snu.ac.kr (K.C.);
[email protected] (P.T.H.). † Seoul National University. ‡ Massachusetts Institute of Technology. (1) Decher, G.; Maclennan, J.; Sohling, U.; Reibel, J. Thin Solid Films 1992, 210, 504. (2) Decher, G. Science 1997, 277, 1232. (3) Stockton, W. B.; Rubner, M. F. Macromolecules 1997, 30, 2717. (4) Wang, L.; Wang, Z.; Zhang, X.; Shen, J.; Chi, L.; Fuchs, H. Macromol. Rapid Commun. 1997, 18, 509. (5) Kharlampieva, E.; Sukhishvili, S. A. J. Macromol. Sci., Polym. ReV. 2006, 46, 377.
usage, significantly decreasing assembly time (seconds per layer), and obtaining an ordered internal structure owing to limited interpenetration between polymer layers.12–16 In this paper, we demonstrate the direct manipulation of LbL film morphologies via the choice of dip-assisted versus spin-assisted LbL assembly methods using systems containing an amphiphilic hydrogenbonding polymer. In previous work, we found that the dip-assisted hydrogenbonded LbL films consisting of hydrophobically modified poly(ethylene oxide) (HM-PEO, a triblock copolymer consisting of PEO endcapped with alkyl groups) and poly(acrylic acid) (PAA) yielded a unique surface morphology, developed above the critical number of layer pairs.17 This surface structure was related to the formation and aggregation of HM-PEO micelles within the film.17 Given the variety of LbL methods available, the surface structure and surface properties of HM-PEO/PAA (6) Cho, J.; Char, K.; Hong, J.; Lee, K. AdV. Mater. 2001, 13, 1076. (7) Chiarelli, P. A.; Johal, M. S.; Casson, J. L.; Roberts, J. B.; Robinson, J. M.; Wang, H. AdV. Mater. 2001, 13, 1167. (8) Schlenoff, J. B.; Dubas, S. T.; Farhat, T. Langmuir 2000, 16, 9968. (9) Porcel, C. H.; Izquierdo, A.; Ball, V.; Decher, G.; Voegel, J.; Schaaf, P. Langmuir 2005, 21, 800. (10) Izquierdo, A.; Ono, S. S.; Voegel, J.; Schaaf, P.; Decher, G. Langmuir 2005, 21, 7558. (11) Krogman, K. C.; Zacharia, N. S.; Schroeder, S.; Hammond, P. T. Langmuir 2007, 23, 3137. (12) Sohn, B.; Kim, T.; Char, K. Langmuir 2002, 18, 7770. (13) Jang, H.; Kim, S.; Char, K. Langmuir 2003, 19, 3094. (14) Jiang, C.; Markutsya, S.; Pikus, Y.; Tsukruk, V. V. Nat. Mater. 2004, 3, 721. (15) Park, J.; Fouche´, L. D.; Hammond, P. T. AdV. Mater. 2005, 17, 2575. (16) Cho, J.; Jang, H.; Yeom, B.; Kim, H.; Kim, R.; Kim, S.; Char, K. Langmuir 2006, 22, 1356. (17) Seo, J.; Lutkenhaus, J. L.; Kim, J.; Hammond, P. T.; Char, K. Macromolecules 2007, 40, 4028.
10.1021/la800906x CCC: $40.75 2008 American Chemical Society Published on Web 06/18/2008
7996 Langmuir, Vol. 24, No. 15, 2008
LbL films, created using either the dip-assisted or spin-assisted method, have yet to be studied and compared. Because the arrangement of amphiphilic copolymers at the interface is a kinetic process, it is expected that these properties would be highly dependent on the assembly technique, where the adsorption under shear at very short times (spin-assisted) may yield completely different growth profiles, surface roughness, and water contact angles relative to the processing based on the dip-assisted technique for several minutes per cycle. In the present study, the layer pair film thickness, surface roughness, and surface morphology of dip- and spin-assisted HM-PEO/PAA multilayer films are compared using profilometry, optical fluorescence microscopy, and confocal laser scanning microscopy. We also note that free-standing multilayer films17,18 were easily obtained using both LbL deposition methods and the film transparency changes with morphology and deposition technique. In addition, the free-standing films allowed us to analyze the composition of constituent polymers within both dip- and spin-assisted multilayered films using differential scanning calorimetry (DSC) and elemental analysis. Overall, HM-PEO/PAA LbL films created using the spin-assisted LbL method were thinner and less rough than their dip-assisted counterparts; spin-assisted LbL films yielded a smooth surface without visible micelle aggregation and yielded a water contact angle that remained constant with the layer pair number. In contrast, the water contact angle of dip-assisted HM-PEO/PAA LbL films ranged from 10 to 90°, depending on the layer number. These findings provide a basis for understanding the role of processing of micellar constituents in the LbL films.
Experimental Section Materials. Hydrophobically modified poly(ethylene oxide) (HMPEO), or poly(ethylene oxide) end-capped with alkyl groups using urethane linkages, was synthesized. Details of the synthesis of HMPEO are described elsewhere.19–21 Briefly, a PEO chain (Mw ) 35 000 g/mol) was end-capped with two alkyl chains containing 22 carbons. Poly(acrylic acid) (PAA, Polysciences Inc.; Mw ) 90 000 g/mol), poly(vinyl pyrrolidone) (PVPE, Aldrich; Mw ) 55 000 g/mol), poly(ethylene oxide) (PEO, Aldrich; Mw ) 4 000 000 g/mol), poly(styrene-b-acrylic acid) (PS-PAA, Polymer Source Inc.; Mw ) 4300 g/mol for PS and Mw ) 19 500 g/mol for PAA), and Nile-red dye (Aldrich) were used as received. Silicon wafers and Teflon were used as substrates for the LbL deposition. Prior to the multilayer deposition of a given polymer pair, the substrate was routinely cleaned. Silicon wafers were cleaned under oxygen plasma (100 W, 0.1-0.5 Torr, 5 min), and Teflon substrates were sonicated in Milli-Q water for 15 min prior to film deposition. Buildup of Dip- and Spin-Assisted Multilayer Films. Polymer solutions were prepared by dissolving polymers in 18 MΩ Milli-Q water. Polymer concentrations were adjusted to 0.02 M based on the repeat unit (alkyl chains of HM-PEO were neglected in calculating the concentration of polymer solutions). Since this concentration is much higher than the critical micelle concentration (CMC) of HMPEO,22 the HM-PEO chains in aqueous solution form micelles at equilibrium. HCl and NaOH were used for the pH adjustment of the solutions. The dip-assisted LbL films were prepared using a Carl Zeiss DS50 programmable slide stainer. Substrates were first exposed to (18) Lutkenhaus, J. L.; Hrabak, K. D.; McEnnis, K.; Hammond, P. T. J. Am. Chem. Soc. 2005, 127, 17228. (19) Lundberg, D. J.; Brown, R. G.; Glass, J. E.; Eley, R. R. Langmuir 1994, 10, 3027. (20) Vorobyova, O.; Yetka, A.; Winnik, M. A. Macromolecules 1998, 31, 8998. (21) Francois, J.; Maitre, S.; Rawiso, M.; Sarazin, D.; Beinert, G.; Isel, F. Colloids Surf., A 1996, 112, 251. (22) Vorobyova, O.; Lau, W.; Winnik, M. A. Langmuir 2001, 17, 1357.
Seo et al. HM-PEO solution for 15 min and then rinsed in three baths of Milli-Q water for 2, 1, and 1 min, respectively. The HM-PEOcoated substrates were then dipped into the PAA solution following the same procedure. The pH of all baths was maintained at pH 2.5 to suppress the ionization of PAA. These steps comprise one cycle, or “layer pair”, and can be repeated for “n” cycles to obtain (HMPEO/PAA)n multilayer films. All dip-assisted LbL films were assembled without an intermediate drying step. Following assembly, the films were dried under nitrogen stream. Spin-assisted LbL films were obtained using an automatic spin coater (Headway Research Inc.). Polymer solutions of HM-PEO and PAA were alternately spun onto silicon or Teflon, with washing steps between the layer deposition steps. As soon as a polymer solution was placed on a substrate, the substrate was spun at 2000 rpm for 20 s until a sufficiently dried film was obtained. The polymercoated substrate was washed at the same rpm on a spin coater with Milli-Q water with the pH adjusted at 2.5 for every deposition step. Nile-red dye, which preferentially associates with the hydrophobic core of HM-PEO micelles, was used to monitor the location of micelles via confocal laser scanning microscopy. In order to guarantee sufficient encapsulation of Nile-red, the dye was added to the HMPEO solution and stirred for 5 days in a dark environment prior to multilayer deposition. Free-standing films based on both LbL deposition methods were obtained by directly peeling the multilayer films off the Teflon substrates using tweezers or Scotch tape. Morphology Characterization. Thicknesses and roughnesses of dip-assembled multilayer films were analyzed using a Tencor P10 profilometer using a 2 mm stylus tip with 5 mg stylus force, and ellipsometry and atomic force microscopy (AFM) were used for spin-assisted multilayer films. Values reported here are the averages of three or more measurements (total span of error bars indicates the standard deviation value). Optical and fluorescence images of the assembled multilayer films were obtained using a Zeiss Axioplan 2 microscope (Carl Zeiss Inc.). Confocal laser scanning microscopy (CLSM) images were obtained using a Carl Zeiss-LSM510 microscope. Water contact angles were measured in both advancing and receding modes using a DE/DSA100 contact angle analyzer (Fru¨ss Inc.). Thermal Analysis. Free-standing multilayer films were further characterized using DSC (TA Instruments Q1000). DSC samples were heated from -90 to 110 °C at a rate of 10 °C/min under nitrogen purge, and the second heating scan values were reported. To minimize the effect of humidity, samples were dried for 30 min under nitrogen purge prior to the measurements. The compositions of free-standing films were also obtained using elemental analysis (EA1110, CE Instrument) for carbon, hydrogen, and oxygen.
Results and Discussion Characteristics of Dip- and Spin-Assisted HM-PEO/PAA Multilayer Films. HM-PEO/PAA multilayer films were constructed using both the dip- and spin-assisted LbL deposition methods to investigate the effect of the LbL deposition method. Figure 1a shows the thickness growth curve of HM-PEO/PAA multilayer films prepared using the two different deposition techniques. The film growth of dip-assisted HM-PEO/PAA films shows exponential growth at the initial growth stage, followed by linear growth after approximately 15 layer pairs. The transition from exponential to linear film growth has already been welldocumented by several groups.23,24 We also note that the spinassisted HM-PEO/PAA films, however, show linear growth regardless of the layer pair number tested in the present work. The average thickness per layer pair (21 and 210 nm/layer pair for spin- and dip-assisted LbL films, respectively) was calculated (23) Porcel, C.; Lavalle, P.; Ball, V.; Decher, G.; Senger, B.; Voegel, J.; Schaaf, P. Langmuir 2006, 22, 4376. (24) Yoo, P. J.; Zacharia, N. S.; Doh, J.; Nam, K. T.; Belcher, A. M.; Hammond, P. T. ACS Nano 2008, 2, 561.
HM-PEO/PAA LbL Films
Figure 1. (a) Thickness and (b) roughness growth curves of dip- and spin-assisted HM-PEO/PAA multilayer films measured using profilometry for dip-assisted multilayer films and ellipsometry and AFM for spin-assisted multilayer films (a magnified thickness and roughness growth curve of the spin-assisted multilayer film is also shown in the inset). Thickness and roughness growth curves of dip-assisted HMPEO/PAA multilayer films were taken from ref 17 for comparison.
from the slope of the linear portion of both spin- and dip-assisted films shown in Figure 1a. The surface roughness of LbL films created using different deposition techniques was also compared, as shown in Figure 1b. In previous work, it was found that the surface roughness of dip-assisted HM-PEO/PAA multilayers abruptly increases from tens of nanometers to several micrometers above a critical number of layer pairs (in this case, 26).17 In contrast, the present study finds that the average surface roughness of spin-assisted multilayer films remains almost constant, within 0.4-0.6 nm measured using AFM, for all the numbers of layer pairs tested. The observed difference in the average layer pair thickness of HM-PEO/PAA LbL films for the two processing methods is converse to that generally observed for the electrostatic systems. Here, the hydrogen-bonded LbL films are thinner when created using the spin-assisted technique (versus dip-assisted), whereas for the electrostatic systems the opposite trend has generally been observed.6 The contrary trends can be rationalized by considering the different interactions between electrostatic and hydrogen-bonding polymer chains. For example, electrostatic multilayer films consisting of poly(allylamine hydrochloride) (PAH)/poly(sodium 4-styrenesulfonate) (PSS) show that the amount of adsorbed polymer as well as the average layer pair thickness of spin-assisted multilayer films are significantly higher than those of dip-assisted multilayer films owing to the fast elimination of water during spinning.6 Water removal during spinning facilitates attractive interactions between polymer pairs and minimizes unfavorable long-range repulsive interactions among adsorbing polyelectrolytes, resulting in the greater amount
Langmuir, Vol. 24, No. 15, 2008 7997
of polyelectrolyte adsorption for each spin deposition. In the case of hydrogen-bonding LbL systems, however, the effect of water screening is less significant owing to the short range of the interactions between the hydrogen-bonding polymer pair. We also found that the layer pair thickness from the spin-assisted deposition when employing hydrogen-bonding pairs is always lower than the layer pair thickness obtained from the dip-based deposition. This trend also applies to the strong hydrogen-bonding system based on poly(vinyl pyrrolidone) (PVPE) hydrogenbonded with PAA (see the Supporting Information). This is to say that when the intermolecular interaction between an adsorbing pair is short-ranged, the shear force imposed during spinning is so dominant as to displace many adsorbing chains, thus lowering the layer pair thickness. In the case of the HM-PEO/PAA (micelle/ linear polymer) LbL system, it is likely that the spherical HMPEO micelles deform into the pancakelike structure under the strong action of air shear force during the spinning process, leading to a much lower average layer pair thickness of 21 nm compared with the case of the dip-assisted layers. The small surface roughness of the spin-assisted HM-PEO/PAA LbL films suggests that the temporary networking and merging of HMPEO micelles (in other words, aggregation) through bridging of hydrophobic alkyl chains, observed in the dip-assisted LbL method, is suppressed for the spin-assisted LbL deposition. Contrary to the dip-assisted LbL films, the spin-assisted LbL films did not show any “critical layer pair” behavior. Evidence of this proposed mechanism is investigated using optical microscopy, confocal laser scanning microscopy, and water contact angle measurements, which are discussed in detail in the following section. Morphology Comparison between Dip- and Spin-Assisted HM-PEO/PAA Multilayer Films. Optical microscopy (Figure 2a and b) and confocal laser scanning microscopy (Figure 2c and d) were performed on both dip- and spin-assisted HM-PEO/ PAA LbL films. Nile-red dye, which preferentially associates within the hydrophobic cores of micelles, was used to emphasize the presence of HM-PEO micelles. Figure 2a and c demonstrate that the spin-assisted HM-PEO/PAA multilayer films have a flat and smooth surface without any bumpy surface features. The surface morphology of spin-assisted HM-PEO/PAA multilayer films were also characterized using AFM (Figure 2e and f). In contrast, dip-assisted HM-PEO/PAA multilayer films show a unique surface morphology (Figure 2b and d), which originates from the aggregation and/or bridging of HM-PEO micelles within the film.17 The observed morphological difference between the two different LbL deposition methods suggests different modes of adsorption and rearrangement. For the spin-assisted LbL deposition with a hydrogen-bonding pair, the adsorption and further arrangement on the surface are quite limited during a short period of spinning (∼20 s). Because HM-PEO micelles are also easily deformed or disrupted by an imposed shear, we assume that it is difficult for HM-PEO micelles to retain their shape during spin-assisted LbL deposition, leading to the even distribution of Nile-red dye within the film (Figure 2c) as well as the lower average layer pair thickness. In contrast, dip-assisted LbL deposition allows enough time (15 min) for HM-PEO micelles to diffuse and adsorb onto the surface followed by surface rearrangement. During surface rearrangement, micelles further interact through bridging to form micelle clusters. The HM-PEO micelles, weakly hydrogen-bonded to PAA, are free to move and migrate within the film, leading to the development of a pronounced surface structure above a critical number of layer pairs.17
7998 Langmuir, Vol. 24, No. 15, 2008
Seo et al.
Figure 2. Optical microscopic images of (a) spin-assisted (HM-PEO/ PAA)30 and (b) dip-assisted (HM-PEO/PAA)25 multilayered films. Images (c) and (d) are confocal laser scanning microscopic images of dip- and spin-assisted HM-PEO/PAA multilayer films containing Nile-red dye within the cores of HM-PEO micelles. AFM images of (e) spin-assisted (HM-PEO/PAA)20 and (f) spin-assisted (HM-PEO/PAA)50 multilayered films.
Figure 4. (a) DSC heating curves (second scan) of spin-assisted HMPEO/PAA multilayer films and bulk HM-PEO. (b) Calculated HM-PEO weight fraction within the dip- and spin-assisted HM-PEO/PAA multilayered films.
Figure 3. Photo images of free-standing (a) spin-assisted (HM-PEO/ PAA)100 and (b) dip-assisted (HM-PEO/PAA)60 multilayer films (2 cm × 4 cm). The thicknesses of the spin- and dip-assisted multilayer films are 2 and 11.6 µm, respectively.
Both dip- and spin-assisted HM-PEO/PAA multilayer films were easily peeled off of their substrates (e.g., silicon, Teflon). When the two free-standing HM-PEO/PAA multilayer films were compared, the dip-assisted LbL films appear turbid (most probably, a result of the micrometer-scale surface structure) while the spin-assisted films are transparent, as evidenced in Figure 3. Obtaining free-standing multilayer films further allowed us to investigate the thermal properties of the spin-assisted multilayer films using DSC (Figure 4a). DSC results on the dip-assisted HM-PEO/PAA LbL films have been presented in a separate
study.17 HM-PEO exhibits a melting peak just below 60 °C, but when HM-PEO is incorporated into the spin-assisted multilayer film, the melting peak is completely suppressed, suggesting the disruption or prevention of HM-PEO crystallization. A similar phenomena was observed for the dip-assisted LbL films.17,18,25 Both dip- and spin-assisted HM-PEO/PAA LbL films exhibit a single glass transition temperature that lies between that of HMPEO and PAA alone (-57 °C for HM-PEO and 100 °C for PAA). In our previous work, we have demonstrated that the weight fraction of HM-PEO within the dip-assisted multilayer film increases with the number of layer pairs owing to the interdiffusion of HM-PEO micelles within the film.17 For reference, dip-assisted (HM-PEO/PAA)27 and (HM-PEO/PAA)41 multilayer films prepared at identical conditions yield HM-PEO weight fractions of 20 and 26 wt %, respectively, as determined by the Fox equation26 using the glass transition temperature measured by DSC. In contrast, the weight fractions of HM-PEO for the spin-assisted (HM-PEO/PAA)50 and (HM-PEO/PAA)100 (25) Lutkenhaus, J. L.; McEnnis, K.; Hammond, P. T. Macromolecules 2007, 40, 8367. (26) Fox, T. G. Bull. Am. Phys. Soc. 1956, 1, 123.
HM-PEO/PAA LbL Films
films are almost invariant (26 and 25 wt %, respectively) based on their glass transition temperatures (∼42 and 44 °C, respectively, Figure 4a). This result suggests that the composition of a spinassisted multilayer film is independent of the number of layer pairs, which is consistent with the highly linear growth curve. Of note, the Fox equation is used here as a qualitative estimation of composition because the presence of hydrogen-bonding interactions is known to cause deviations from the calculated glass transition temperature. To obtain more quantitative information on the composition of HM-PEO/PAA multilayered films, elemental analysis was performed. The expected ratios for carbon, hydrogen, and oxygen for pure HM-PEO and PAA are 55.2, 9.2, and 35.6%, and 50.0, 5.6, and 44.4%, respectively. In addition, the expected composition ratio of 25 wt % HM-PEO and 75 wt % PAA in the multilayered film obtained with the spin-assisted deposition is 51.3, 6.5, and 42.2% for carbon, hydrogen, and oxygen, respectively. Elemental analysis of the spin-assisted 30 and 50 layer pairs of HM-PEO/PAA multilayered films yielded compositions of 51.4 and 51.2% carbon, 7.1% hydrogen, and 41.5 and 41.7% oxygen (see the Supporting Information). This experimentally determined composition is in good agreement with the expected value of 25 wt % (or 35.3 mol %) HM-PEO within the multilayered film, confirming the composition determined by the Fox equation on DSC data. Surface Wetting Behavior of Dip- and Spin-Assisted HMPEO/PAA Multilayer Films. In order to characterize the surface properties of HM-PEO/PAA multilayer films, we investigated the wetting behavior of water on dip- and spin-assisted HMPEO/PAA multilayer films as a function of the layer pair number (Figure 5); all measurements were performed with PAA as the top layer. Dip-assisted LbL films exhibit varying water contact angles as a function of the layer pair number (Figure 5a). Initially, both the advancing and receding water contact angles increase with increasing number of layer pairs (e.g., from ∼60° for 20 layer pairs to 90° for 26 layer pairs in advancing contact angles). The water contact angle reaches the maximum at 26 layer pairs and then decreases at a higher number of layer pairs, eventually reaching an angle of θ for θ > π/2). The second case is related to the case where the contact liquid is readily wicked into the rough surface texture (i.e., θ < θc), leading to a lower apparent water contact angle (i.e., hemiwicking). In the dip-assisted HM-PEO/PAA multilayer system, we note that, by increasing the layer pair number above 26, the observed water contact angle decreases rapidly and then levels off (Figure 5a). The measured contact angle as a function of the layer pair number is explained as follows: when the layer pair number is increased above the critical number of layer pairs (∼26 in the present case), the surface structure develops and the surface roughness is amplified. According eq 2, the critical angle, θc, also increases as the roughness, r, increases. The increase in θc, in turn, enables the hemiwicking region to expand based on the θ versus θc criterion outlined above. The expansion of the hemiwicking region implies that the contact liquid is easily drained into the roughened surface structure, decreasing the measured contact angle. To the best of our knowledge, this is the first report on the simultaneous increase and decrease in the contact angle as a function of layer pair number by incorporating (aggregated) micelles within multilayer films. We also checked the possible influence of neutral water droplets (pH ∼ 7) on the surface morphology of hydrogen-bonded multilayer films during contact angle measurements because there is a possibility that the hydrogen-bonding between PEO and PAA can be altered by the ionization of carboxylic acid groups in PAA. In order to minimize the surface reconstruction during contact angle measurements, water droplets with pH ) 2.5 were employed for the measurements (see the Supporting Information). The tendency showing the maximum contact angle value (above 80°) at an intermediate bilayer number (∼30 bilayers) for the case of dipassisted HM-PEO/PAA multilayer films is again in good agreement with the results obtained with neutral water droplets, but we noted that the hysteresis in the contact angle is much more reduced in the case where water droplets at pH ) 2.5 were used to measure the contact angles. We believe that the HMPEO/PAA multilayer films are stable (i.e., maintaining the hydrogen-bonding) when the films are in contact with water at low pH and the possible surface rearrangement is thus minimized. In addition, we also noted that the contact angle behavior on spin-assisted HM-PEO/PAA multilayered films is almost similar between the two cases where water droplets with different pH values (neutral and pH ) 2.5) were used except for the reduced (29) Wenzel, R. N. Ind. Eng. Chem. 1936, 28, 988.
hysteresis in the contact angle in the case of water droplets at pH ) 2.5. The almost independence of the contact angle as a function of layer pair number in the case of spin-assisted HMPEO/PAA multilayered films implies that the films deposited by the spinning process remain quite dense each time polymer chains are deposited by hydrogen-bonding on the substrate.
Conclusions We demonstrated that the surface morphology and the resulting wetting behavior of HM-PEO/PAA multilayer films can be controlled using various LbL deposition methods (i.e., dip- and spin-assisted LbL assembly). Thin and uniform HM-PEO/PAA multilayer films with low surface roughness were obtained using the spin-assisted LbL deposition method; both the roughness and the water contact angle of spin-assisted films remain constant with the layer pair number, mainly due to the limited rearrangement and aggregation of HM-PEO micelles within the film during spin deposition. In comparison, dip-assisted HM-PEO/PAA multilayer films exhibit a three-dimensional surface texture above a critical number of layer pairs, which was attributed to the migration and aggregation of HM-PEO micelles within the film. The development of the surface structure in the dip-assisted multilayer films yielded turbid free-standing films, while transparent free-standing films were obtained using spin-assisted LbL assembly. Also, the contrasting surface morphologies have a significant effect on the wetting behavior of water droplets. Specifically, the water contact angle on the dip-assisted HMPEO/PAA LbL films gave the maximum value at an intermediate layer pair number; at a higher layer pair number, the contact angle decreases rapidly as the surface structure is increasingly evolved and amplified. This study demonstrates that the surface morphology and the resulting surface properties can be finely tuned by the LbL deposition method, the number of deposited layer pairs, and the mode of interaction. Acknowledgment. This work was financially supported by the Korea Science and Engineering Foundation (KOSEF) Grant funded by the Korea Government (MOST) (Acceleration Research Program (No. R17-2007-059-01000-0) and NANO Systems Institute-National Core Research Center (No. R15-2003032-05001-0)) and by the Brain Korea 21 Program endorsed by the Ministry of Education of Korea. We also acknowledge the Center for Materials Science and Engineering and the Institute for Soldier Nanotechnologies (ISN) at MIT for the use of their facilities. K.C. acknowledges financial support from the SBS Foundation for his sabbatical leave at ISN of MIT, and J.S. acknowledges the Seoul Science Fellowship for her graduate study at SNU. Supporting Information Available: Thickness growth curves of various hydrogen-bonded LbL films (Figures S1-S3), contact angle measurements with water droplets at pH ) 2.5 (Figures S4 and S5), and elemental analysis results on the spin-assisted HM-PEO/PAA LbL films (Table S1). This material is available free of charge via the Internet at http://pubs.acs.org. LA800906X
