Polyelectrolyte Blend Multilayer Films: Surface Morphology, Wettability

Mar 31, 2007 - Centre for Nanoscience and Nanotechnology, Department of Chemical and Biomolecular Engineering, The University of Melbourne, Victoria 3...
0 downloads 22 Views 162KB Size
Download PDF
4944

Langmuir 2007, 23, 4944-4949

Polyelectrolyte Blend Multilayer Films: Surface Morphology, Wettability, and Protein Adsorption Characteristics Anthony Quinn,† Elvira Tjipto,† Aimin Yu,†,§ Thomas R. Gengenbach,‡ and Frank Caruso*,† Centre for Nanoscience and Nanotechnology, Department of Chemical and Biomolecular Engineering, The UniVersity of Melbourne, Victoria 3010, Australia, and CSIRO Molecular and Health Technologies, Bag 10, Clayton South, Victoria 3169, Australia ReceiVed NoVember 29, 2006. In Final Form: February 16, 2007 We report the influence of polyelectrolyte (PE) multilayer films prepared from poly(styrene sulfonate)-poly(acrylic acid) (PSS-PAA) blends, deposited in alternation with poly(allylamine hydrochloride) (PAH), on film wettability and the adsorption behavior of the protein immunoglobulin G (IgG). Variations in the chemical composition of the PAH/(PSS-PAA) multilayer films, controlled by the PSS/PAA blend ratio in the dipping solutions, were used to systematically control film thickness, surface morphology, surface wettability, and IgG adsorption. Spectroscopic ellipsometry measurements indicate that increasing the PSS content in the blend solutions results in a systematic decrease in film thickness. Increasing the PSS content in the blend solutions also leads to a reduction in film surface roughness (as measured by atomic force microscopy), with a corresponding increase in surface hydrophobicity. Advancing contact angles (θ) range from 7° for PAH/PAA films through to 53° for PAH/PSS films. X-ray photoelectron spectroscopy measurements indicate that the increase in film hydrophobicity is due to an increase in PSS concentration at the film surface. In addition, the influence of added electrolyte in the PE solutions was investigated. Adsorption from PE solutions containing added salt favors PSS adsorption and results in more hydrophobic films. The amount of IgG adsorbed on the multilayer films systematically increased on films assembled from blends with increasing PSS content, suggesting strong interactions between PSS in the multilayer films and IgG. Hence, multilayer films prepared from blended PE solutions can be used to tune film thickness and composition, as well as wetting and protein adsorption characteristics.

Introduction Control over the surface properties of ultrathin films, such as wettability and adsorption behavior, are important in a wide variety of applications, including surface self-cleaning, antiadhesives, biosensing, and separations.1,2 The layer-by-layer (LbL) assembly method offers a convenient, versatile, and reliable technique for the preparation of thin films with tailored composition, thickness, and function.3 The basic process involves the stepwise electrostatic assembly of oppositely charged components.4 Studies by McCarthy and co-workers demonstrate that the wettability of poly(allylamine hydrochloride) (PAH) and poly(styrene sulfonate) (PSS) multilayer films is controlled almost entirely by the outermost polyelectrolyte (PE) layer.5 Rubner et al. demonstrated that the wettability of weak PE multilayers comprising of poly(acrylic acid) (PAA) and PAH could be varied dramatically by controlling the pH of the dipping solutions.6 Similar to PAH/PSS multilayers, the wettability of PAA/PAH multilayers is controlled primarily by the outermost surface layer. Recently, the same group constructed stable superhydrophobic coatings based on PAH/PAA multilayers * To whom correspondence should be addressed. E-mail: [email protected]. † The University of Melbourne. ‡ CSIRO Molecular and Health Technologies. § Current address: ARC Centre for Functional Nanomaterials, The University of Queensland, St Lucia, 4073, Australia. (1) Horbett, T. A.; Brash, J. L., Eds. Proteins at Interfaces II; ACS Symposium Series 602; American Chemical Society: Washington DC, 1995; p 548. (2) Brash, J. L.; Wojciechowski, P. W., Eds. Interfacial Phenomena and Bioproducts; Marecel Dekker, Inc.: New York, 1996. (3) Decher, G.; Hong, J.-D. Ber. Bunsen-Ges. Phys. Chem Chem. Phys. 1991, 95, 1430. (4) Decher, G. Polyelectrolyte Multilayers, An Overview. In Multilayer Thin Films; Decher, G., Schlenoff, J. B., Eds.; Wiley-VCH Verlag: Weinheim, 2003. (5) Chen, W.; McCarthy, T. J. Macromolecules 1997, 30, 78. (6) Yoo, D.; Shiratori, S. S.; Rubner, M. F. Macromolecules 1998, 31, 4309.

coated with silica nanoparticles and a semifluorinated silane.7,8 These studies showed that by simply changing the nature of the outermost layer, it is possible to create surfaces with molecularly tunable wetting characteristics. The adsorption of proteins onto LbL assembled films has been extensively studied, largely because of the potential applications of these films as biosensors, bioreactors, and bioactive membranes.9-19 Proteins can adsorb onto such films via electrostatic, hydrogen-bonding, and/or hydrophobic interactions.20 As with the wettability of LbL films, the protein interactions and hence the amount of protein adsorbed are, among other factors, strongly dependent on the nature and charge of the outermost layer of the multilayer films. For example, Schaaf et al. investigated the interactions between human serum albumin (HSA) and PSS/PAH multilayer films.18 It was found that HSA adsorbs on multilayers terminated with either PSS (negatively (7) Zhai, L.; Cebecci, F. C.; Cohen, R. E.; Rubner, M. F. Nano. Lett. 2004, 4, 1349. (8) Zhai, L.; Berg, M. C.; Cebeci, F. C.; Kim, Y.; Milwid, J. M.; Rubner, M. F. Nano. Lett. 2006, 6, 1213. (9) Lvov, Y.; Ariga, K.; Ichinose, I.; Kunitake, T. J. Am. Chem. Soc. 1995, 117, 6117. (10) Onda, M.; Ariga, K.; Kunitake, T. Biosci. Bioeng. 1999, 87, 69. (11) Onda, M.; Lvov, Y.; Ariga, K.; Kunitake, T. Biotechnol. Bioeng. 1996, 51, 163. (12) Caruso, F.; Niikura, K.; Furlong, D. N.; Okahata, Y. Langmuir 1997, 13, 3427. (13) Caruso, F.; Furlong, D. N.; Ariga, K.; Ichinose, I.; Kunitake, T. Langmuir 1998, 1998, 4559. (14) Caruso, F.; Fiedler, H.; Haage, K. Colloids Surf., A 2000, 169, 287. (15) Caruso, F.; Schuler, C. Langmuir 2000, 16, 9595. (16) Schuler, C.; Caruso, F. Macromol. Rapid Commun. 2000, 21, 750. (17) Lvov, Y.; Caruso, F. Anal. Chem. 2001, 73, 4212. (18) Ladam, G.; Gergely, C.; Senger, B.; Decher, G.; Voegel, J.-C.; Schaaf, P.; Cuisinier, F. J. G. Biomacromolecules 2000, 1, 674. (19) Salloum, D. S.; Schlenoff, J. B. Biomacromolecules 2004, 5, 1089. (20) Lvov, Y.; Mohwald, H., Eds. Protein Architecture: Interfacing Molecular Assemblies and Immobilization Biotechnology; Marcel Dekker: New York, 2000.

10.1021/la0634746 CCC: $37.00 © 2007 American Chemical Society Published on Web 03/31/2007

Polyelectrolyte Blend Multilayer Films

charged) or PAH (positively charged) and that the adsorbed amount depends on the ionic strength of the adsorption solution.21 It was shown that proteins strongly interact with the PE films, irrespective of the sign of the charge of the multilayer film and the protein. However, when the charges on the multilayer (determined by the outermost layer) and the protein are similar, dense protein monolayers usually form, whereas adsorption of a protein on an oppositely charged multilayer film leads to larger adsorbed amounts. A recent study by Salloum and Schlenoff examined the effect of surface and protein charge, polymer hydrophobicity, and polymer hydrophilic repulsion on protein adsorption behavior on PE multilayer films.19 This study showed that the type and charge of the PE multilayer surface determines the amount of protein adsorbed, with an outermost layer of PAA providing an efficient protein-resistant surface. Recently, deposition from multicomponent PE solutions has been investigated as a means of imparting film and surface properties that would otherwise be unattainable from adsorption of single-component PE solutions.22-27 Leporatti et al. reported the coadsorption of chitosan and PSS, in alternation with PAH, to reduce the high stability of PAH and PSS multilayer films.22 In that study, the chitosan was incorporated into the multilayers, effectively reducing the number of electrostatic bonds between neighboring PSS and PAH, destabilizing the film, and facilitating thermal shrinkage. Conversely, Sui and Schlenoff, fabricated multilayers from PSS and a blended solution of poly(diallyldimethylammonium chloride), PDADMA, and a random copolymer, PDADMA-co-PAA, to enhance multilayer stability.24,25 By reducing the number of carboxylic acid groups in the films through the multicomponent adsorption, a net positive multilayer charge could be realized, thereby stabilizing the system and enabling electroosmotic flow through the microchannels. In 2003, Voegel et al. conducted a systematic study of multilayer films assembled from poly(L-glutamic acid) (PGA) and poly(L-aspartic acid) blends in alternation with poly(L-lysine) (PLL).23 Characterization of the multilayers revealed the incorporation of both anionic components, with an enhanced poly(L-aspartic acid) content in the films relative to the solution composition. Cooperative effects were also observed, giving rise to secondary structures and assemblies that were not attainable from single-component LbL assembly. More recently, we reported the fabrication of three-component multilayer films comprising PE ‘blends’ of DNA-PSS or PSSPAA, alternately deposited with the polycation PAH.26-28 In the case of the DNA-PSS blend, we demonstrated that the film composition could be regulated by varying the specific adsorption conditions used. For instance, adsorbing the blend layers from aqueous solutions containing 20 vol % ethanol led to an increase in the amount of DNA incorporated, while adsorption solutions containing 0.5 M NaCl increased the fraction of PSS in the film.28 The morphology and thickness of the film were also shown to correlate with changes in the film composition. Similarly, by assembling PAH in alternation with blend solutions comprising weak (PAA) and strong (PSS) polyanions, it was possible to (21) Ladam, G.; Schaaf, P.; Cuisinier, F. J.; Decher, G.; Voegel, J.-C. Langmuir 2001, 17, 878. (22) Leporatti, S.; Gao, C.; Voigt, A.; Donath, E.; Mohwald, H. Eur. Phys. J. E 2001, 5, 13. (23) Debreczeny, M.; Ball, V.; Boulmedais, F.; Szalontai, B.; Voegel, J.-C.; Schaaf, P. J. Phys. Chem. B 2003, 107, 12734. (24) Sui, Z.; Schlenoff, J. B. Langmuir 2003, 19, 7829. (25) Sui, Z.; Schlenoff, J. B. Langmuir 2004, 20, 6026. (26) Cho, J.; Quinn, J. F.; Caruso, F. J. Am. Chem. Soc. 2004, 126, 2270. (27) Yap, H. P.; Quinn, J. F.; Ng, S. M.; Cho, J.; Caruso, F. Langmuir 2005, 21, 4328. (28) Quinn, J. F.; Yeo, J. C. C.; Caruso, F. Macromolecules 2004, 37, 6537. (29) Zelikin, A. N.; Quinn, J. F.; Caruso, F. Biomacromolecules 2006, 7, 27.

Langmuir, Vol. 23, No. 9, 2007 4945

control multilayer film properties such as film thickness, composition, morphology, and pH stability by altering the proportion of PAA and PSS in the PE dipping solution.26 For example, the thickness of PAH/(PSS-PAA) multilayer films increases markedly (from 1 to >20 nm per layer) when the fraction of PAA (the weak polyanion) in the blend is increased. This trend was also observed at different salt concentrations and with varying molecular weight of PAA under the same pH conditions.26 The PSS-PAA blend ratios also have a significant influence on the colloidal stability of particles coated with PAH/(PSS-PAA) multilayers.27 Herein, we investigate the influence of changes in the blend composition (and hence film composition) on the surface morphology, wettability, and protein adsorption characteristics of PAH/(PSS-PAA) films. The films were characterized via spectroscopic ellipsometry, contact angle measurements, atomic force microscopy (AFM), quartz crystal microgravimetry (QCM), and X-ray photoelectron spectroscopy (XPS). We show that the composition of the outermost layer of the films can be readily altered by using PSS-PAA blend solutions with different contents of PSS and PAA. This makes it possible to create surfaces with molecularly tunable contact angles and protein adsorption characteristics. The results provide a basis for extending the current understanding of PE blend films, as well as designing multilayer films with tailored interfacial properties. Experimental Section Materials. Poly(acrylic acid) (PAA, Mw ) 30 000 g mol-1), poly(allylamine hydrochloride) (PAH, Mw ) 70 000 g mol-1), poly(styrene sulfonate) (PSS, Mw ) 70 000 g mol-1), polyethylenimine (PEI, Mw ) 10 000 g mol-1), and fluorescein isothiocyanate (FITC)labeled immunoglobulin G (IgG) were obtained from Sigma-Aldrich and used as received. PE dipping solutions of 1 mg mL-1 (with or without NaCl) were made from Milli-Q water, and the pH was adjusted with either HCl or NaOH. Glass microscope slides and silicon wafers (Silchem Handelgesellschaft GmbH, Germany) were cleaned in Piranha solution (7:3 v/v% H2SO4 (concd)/H2O2 (30%), Caution! Piranha solution is highly corrosiVe and extreme care should be taken when handling) for 30 min, followed by thorough water rinsing and drying with compressed nitrogen. Rhodamine isothiocyanate (RITC)-labeled PAA (PAA-RITC) was synthesized according to methods described elsewhere.29 High-purity water with a resistivity greater than 18 MΩ‚cm was obtained from an inline Millipore RiOs/Origin system. PE Multilayer Film Buildup. PEI (1 mg mL-1, 0.5 M NaCl) was used as a precursor layer to homogenously coat the substrates with positive charge. The anionic PE layers were deposited from 1 mg mL-1 solutions of PAA or PSS (pH 3.5; 0 or 0.5 M NaCl), or blend solutions of PSS-PAA with different compositions. The cationic PE layers were deposited from 1 mg mL-1 solutions of PAH (pH 7.5; 0 or 0.5 M NaCl). Glass, gold, and silicon substrates were coated with a precursor layer of PEI and then alternately immersed in anionic PE blend solutions and PAH solutions for 15 min, with water rinsing (3 × 1 min) after deposition of each layer. The films were dried after multilayer assembly by a gentle stream of nitrogen. To ascertain whether PE desorption occurred during specific solution posttreatments, selected experiments were conducted with PAA-RITC and analyzed via fluorescence spectroscopy. Multilayers are represented as (PAH/blend)x, where x is the number of bilayers and ‘blend’ refers to the PAA-PSS binary PE solution. Despite the nomenclature, the first layer deposited on all surfaces was a PEI layer. The polyanionic blend layer was always the terminating layer. Spectroscopic Ellipsometry. Thickness measurements were obtained with a UVISEL-NIR spectroscopic ellipsometer from HORIBA Jobin Yvon. Spectroscopic data was acquired over the range 340-825 nm with a 5 nm increment, with thicknesses calculated via the integrated software utilizing an amorphous wavelength dispersion model.

4946 Langmuir, Vol. 23, No. 9, 2007

Quinn et al.

Table 1. XPS Binding Energies and Peak Assignments for PSS and PAA binding energy (eV) S 2p C 1s(4) C 1s(5)

mean

standard deviation

peak assignments

168.11 288.27 289.31

0.03 0.1 0.16

SO3CdO, N-CdO O-CdO

Atomic Force Microscopy. The surface morphology and roughness of the films were examined with a MFP-3D Asylum Research instrument operated in AC mode. Multilayers were deposited onto silicon wafer substrates, dried with compressed nitrogen, and imaged in air with BS-Tap300 (Bulgaria) cantilevers. Contact Angle Measurements. Prior to contact angle measurements, samples were dried over phosphorus pentoxide for 48 h to remove water from the films. Advancing contact angles were measured using an OCA contact angle measurement device via the standard sessile drop technique.30 Four water droplets were deposited onto each surface with five independent measurements recorded per droplet. The contact angle values represent the average of 20 measurements. X-ray Photoelectron Spectroscopy (XPS). XPS was used to examine the surface composition of the blend films via detection of sulfonate and carboxyl groups. Relative compositions were determined from the atomic concentrations of each element relative to that of carbon, the most abundant element (atomic ratios X/C). A KRATOS Analytical AXIS-HSi spectroscopic instrument with a monochromated Al KR radiation source operated below 5 × 10-8 mbar with an analysis area of ∼0.8 mm2 was used to measure two locations per sample. Table 1 presents the binding energy values determined for the photoelectron peaks, which were obtained by correcting the measured peak position for sample charging, using the main C 1s signal as a reference (aliphatic hydrocarbon component at 285.0 eV31). Peak assignments are based on available reference data,32 with the C 1s(4) and C 1s(5) peaks used to determine the relative amounts of PAA, and the S 2p peak for PSS. Spectra were recorded for two separate emission angles (0° and 75°, with corresponding depths of penetration of up to 10 and 2-3 nm, respectively33), at two separate locations for each sample, with the averaged data presented. The error bars are therefore derived from the measured data rather than the absolute error of the instrument. Protein Adsorption. Protein adsorption on PE blend films was examined via fluorescence spectroscopy (Fluorolog-3, Jobin Yvon). After formation of the PE multilayers, the films were immersed into phosphate buffered saline (PBS) solution containing 40 µg mL-1 fluorescein isothiocyanate-labeled immunoglobulin G (IgG-FITC) for 40 min (pH 7.4), followed by water rinsing (3 × 1 min) and N2 drying. The films were excited at 480 nm, and the emission was monitored at 520 nm. Quartz Crystal Microgravimetry. Protein adsorption was monitored in situ via a quartz crystal microbalance with dissipation monitoring (QCM-D) from Q-Sense (Sweden). Adsorption induces a decrease of the crystal resonance frequency (∆F), which for rigid films can be converted to an adsorbed mass via the Sauerbrey equation using a mass sensitivity constant of 17.7 ng m-2 Hz-1 for a 5 MHz crystal.34

Results and Discussion Effect of Blend Composition on Film Thickness and Surface Morphology. Spectroscopic ellipsometry measurements show (30) Berg, J. C. Wettability; Surfactant Science Series; Schick M. J., Fowkes, F. M., Eds.; Marcel Dekker: New York, 1993; Vol. 49, p 451. (31) Beamson, G.; Briggs, D. High resolution XPS of Organic Polymers. The Scienta ESCA300 Database; John Wiley and Sons: New York, 1992. (32) Wagner, C. D.; Naumkin, A. V.; Kraut-Vass, A.; Allison, J. W.; Powell, C. J.; Rumble, J. R. NIST X-ray Photoelectron Spectroscopy Database; 2000. (33) Gengenbach, T. R.; Griesser, H. J. J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 2191. (34) Rodahl, M.; Hook, F.; Krozer, A.; Brzezinski, P.; Kasemo, B. ReV. Sci. Instrum. 1995, 66, 3924.

Figure 1. Thickness of (PAH/blend)8 films, as measured via spectroscopic ellipsometry. Films were assembled from PE solutions with no added salt (circles), and with 0.5 M NaCl (squares).

that blend multilayers assembled from PE solutions that contain 0.5 M NaCl are 70 ( 7% thicker than those prepared from solutions containing no added salt (Figure 1). This is attributed to intramolecular charge screening of the PEs, as the added electrolyte shields similar charges and reduces electrostatic repulsion, causing the PEs to assume a more coiled conformation and adsorb as thicker layers.35 In agreement with our earlier work,26 there is also a systematic trend in film thickness with variation of the PSS-PAA fraction in the blend solutions, indicating that both blend components are being adsorbed across the entire range of blend compositions. A decrease in film thickness was observed with increasing PSS fraction in the blend solutions, with the greatest change in film thickness occurring in films adsorbed from blend solutions containing between 0 and 20 wt % PSS (Figure 1). This trend is similar regardless of whether the films are adsorbed from solutions containing no added salt or from 0.5 M NaCl solutions. The relatively large thicknesses observed for the PAH/PAA films (220 nm (no salt), ∼300 nm (0.5 M NaCl) for 16-layer films) are attributed to the weakly charged nature of PAA under the deposition conditions of pH 3.5, where PAA adsorbs in a highly coiled conformation.36 In contrast, PSS, a strong polyanion, adopts a more linear (stretched) conformation due to the higher charge per molecule and intramolecular electrostatic repulsions. AFM examination of the multilayer films reveals a systematic variation of surface morphology with solution blend composition (Figure 2). The surface roughness systematically decreases for multilayers assembled in the absence of added salt and with increasing PSS content in the blend solutions (Figure 3). The (PAH/PAA)8 films exhibit an “open-framework” surface texture with a root-mean-square (rms) surface roughness of 23 nm. The presence of 10 wt % PSS in the blend solution results in a similar surface texture, but the film is considerably smoother (rms ) 8.8 nm). Higher fractions of PSS in the blends result in significantly smoother films. Similar to the film thickness data from ellipsometry (Figure 1), the largest change in rms roughness with blend composition occurs between 0 and 20 wt % PSS. We note that rinsing with water (pH ≈ 5.7) between successive PE depositions ionizes some of the carboxylic acid groups of PAA, resulting in morphological changes that could enhance the surface (35) Steitz, R.; Leiner, V.; Siebrecht, R.; v. Klitzing, R. Colloids Surf., A 2000, 163, 63. (36) Fery, A.; Schoeler, B.; Cassagneau, T.; Caruso, F. Langmuir 2001, 17, 3779.

Polyelectrolyte Blend Multilayer Films

Langmuir, Vol. 23, No. 9, 2007 4947

Figure 2. AFM surface topography images of (PAH/blend)8 films assembled from PE blend solutions containing (a) 0 wt % PSS, (b) 10 wt % PSS, (c) 25 wt % PSS, (d) 50 wt % PSS, (e) 75 wt % PSS, and (f) 100 wt % PSS. No salt was added to the PE solutions. The z scale for all images is 160 nm.

Figure 3. Advancing contact angles of (PAH/blend)8 films assembled from PE solutions with no added salt (circles) and with 0.5 M NaCl (squares). The secondary axis shows the RMS surface roughness of the (PAH/blend)8 multilayers assembled with no added salt, as determined by AFM. Roughness analysis was excluded for the multilayers assembled from 0.5 M NaCl containing PE solutions due to the scale of the roughness exceeding the 1 µm2 analysis area.

roughness.36 Consequently, higher PAA fractions in the blend solutions result in thicker and rougher multilayer films. Multilayer assembly in the presence of 0.5 M NaCl results in rougher films than in the absence of salt, with relatively largescale inhomogeneities (data not shown). Therefore, highresolution images of the surface morphology and roughness are not representative of the bulk films and were excluded from further quantitative analysis. Qualitatively, however, the surface roughness of these films is significantly larger than those assembled from solutions with no added salt. The large-scale inhomogeneities observed in the films diminished with increasing PSS content in the blend solutions. Effect of Blend Composition on Film Wettability and Composition. The wettability of adsorbed polymer films is typically determined by the outermost surface of the film. It is also influenced by various factors, including the chemical composition, surface roughness, and in the case of LbL films, the degree of interpenetration of the adsorbed PE layers.5,6 By varying the composition of the outermost layer through adsorption of PE layers from blend solutions, systematic control over the film wettability can be realized. Contact angles were measured on multilayers consisting of eight bilayers due to the observation of a systematic decrease in

contact angles with layer number up to approximately the sixth bilayer. This phenomenon is consistent with observations that PE film formation in the first few layers can be nonuniform37 and that several bilayers are required to yield a homogeneous film where substrate effects are no longer observed. First, contact angles were measured for films containing no added salt. The most wettable film is obtained when the aliphatic polyacid PAA forms the outermost layer (θ ) 7 ( 4°), while PSS-terminated films yield the highest contact angle, θ ) 53 ( 3°, which is largely attributed to the hydrophobic aromatic rings in PSS. The contact angles of the blend films range between these two values and increase with increasing proportion of PSS in the blend solution (Figure 3). The presence of only 10 wt % PSS in the blend solution dramatically increases the advancing water contact angle to 31 ( 3°. A film made from a 50:50 wt % PSS-PAA blend solution yields a contact angle of 41°, as compared to 53° for pure PSS, suggesting that either PSS is preferentially adsorbed over PAA and/or the surface is enriched with PSS. Our previous work showed that the amount of PSS in the film increases as the PSS content in the PSS-PAA blend increases.26 The contact angle data (Figure 3) follow this trend, in that the surface hydrophobicity increases with the PSS content of the blend solution, with the greatest increase occurring below ∼25 wt % PSS, while beyond 25 wt % a relatively minor increase is observed. Films deposited from PE solutions containing 0.5 M NaCl display the same trend of increasing hydrophobicity with PSS content; however, for the same blend ratios, these multilayers are more hydrophobic than those assembled from solutions containing no added salt (Figure 3). This reduced wettability is observed despite the fact that assembly from salt-containing PE solutions yields films with an enhanced surface roughness. Typically, an increase in roughness for surfaces exhibiting contact angles in this range (i.e., θ < 90°)38 would enhance the surface wettability. Therefore, the decreased wettability and enhanced roughness imply that the surface hydrophobicity is increased due to a greater PSS content at the interface. To further examine this, multi-angle XPS was conducted, which facilitates direct comparison of the surface composition of the assembled multilayers. PAA was identified via the C 1s (4) (carbonyl) and C 1s (5) (carboxylic acid or ester) peaks, and PSS via the S 2p (sulfonate) peak (Table 1). XPS analysis was performed at two separate emission angles (0° and 75° with (37) Bosio, V.; Dubreuil, F.; Bogdanovic, G.; Fery, A. Colloids Surf., A 2004, 243, 147. (38) Wenzel, R. N. Ind. Eng. Chem. 1936, 28, 988.

4948 Langmuir, Vol. 23, No. 9, 2007

Figure 4. XPS atomic ratios of PSS/PAA for (PAH/blend)8 films assembled from PE solutions with no added salt (circles), and with 0.5 M NaCl (squares). The emission angle was 0° (solid symbols) and 75° (open symbols).

respect to the surface normal) to probe film composition with different sampling depths. At 0°, the sampling depth is greater than at 75° (grazing angle);33 therefore, the latter is more sensitive to the composition of the outermost surface of the multilayer film. The ratios of PSS/PAA, as determined from the peak intensities relative to carbon,33 reveal consistent trends with the contact angle data. Films assembled with blend ratios below ∼25 wt % PSS indicate a significant incorporation of PSS, whereas above 25 wt % PSS, further increases in the blend ratio translate to only small increases in the adsorbed PSS/PAA ratio (Figure 4). Comparison of the data from the two emission angles reveals a gradient in the surface composition. The higher PSS/PAA ratios observed at 75° (compared with 0°) suggest an enhancement of PSS at the surface, as the XPS technique samples the top 2-3 nm of the film at 75° and up to 10 nm at 0°.33 This finding is in agreement with our previous study, where lower PSS/PAA ratios were observed for measurements of the bulk film composition.26 This indicates that while PAA is adsorbed to a greater extent in the bulk of the film, PSS is preferentially located at the interface. Multilayer assembly from PE solutions containing 0.5 M NaCl reveals an enhanced PSS/PAA ratio over those assembled with no added salt, Figure 4. This enhanced PSS surface composition translates to an enhanced hydrophobicity, which correlates with the decreased wettability observed from contact angle analysis (Figure 3). Furthermore, this enhanced PSS composition due to adsorption from 0.5 M NaCl solutions is similar to results from our previous study of PAH/(DNA-PSS) multilayer films. The presence of NaCl in the adsorption solutions also resulted in films containing a greater ratio of PSS relative to DNA.28 However, in the current work it is likely that contributions from adventitious carbon in XPS analyses result in overestimated PAA fractions at the surface of the films. This would be more pronounced for the thinner films assembled from solutions with no added salt. Hence, the actual PSS/PAA ratios at the surface of the films assembled from solutions with no added salt may be larger than those measured experimentally. This argument is supported by the nonzero PAA component in the PAH/PSS films (assembled from both no added salt and 0.5 M NaCl solutions) (Figure 4); however, it is not expected to influence trends in any set of samples, which were all assembled concurrently.

Quinn et al.

Figure 5. Fluorescence intensity of a (PAH/PAA-RITC)8 multilayer; after immersion in PBS for 40 min; and subsequent immersion in IgG for 40 min.

Protein Adsorption on Blend Multilayers. Protein adsorption is a complex process and depends on a number of parameters, such as electrostatic, hydrogen bonding, and the hydrophobic/ hydrophilic characteristics of both the protein and the surface for adsorption.39 The conformation of the PEs on the surface and the conformation of the proteins in solution will also affect protein adsorption. We sought to examine the influence of the blend films on protein adsorption. Adsorption of IgG on a (PAH/PSS)8 multilayer resulted in an adsorbed mass of 10 ( 0.6 mg m-2, as determined via QCM. This amount is approximately double that of a densely packed end-on-oriented monolayer (5.5 mg m-2).40 This could suggest interpenetration of the IgG into the multilayer; however, this behavior is typically not favored when both the multilayer and protein possess like charges.19 In accordance with this, the QCM frequency change, or adsorbed masses of IgG on both (PAH/ PSS)6 and (PAH/PSS)8 multilayers are identical, suggesting limited penetration of the protein into the bulk of the film. We found that QCM analysis of protein adsorption on PAH/ PAA multilayers is not reliable in high-ionic-strength media such as PBS. Competition with salt ions leads to reduced bond strengths between associated PEs and results in deconstruction and loss of mass from the multilayer. Utilizing fluorescently labeled PAA (rhodamine isothiocyanate labeled PAA (PAA-RITC)), we observed via fluorescence spectroscopy that as much as 60% of the PAA in the outermost layer of a (PAH/PAA)8 film was desorbed upon immersion in PBS (pH 7.4) (Figure 5). Fluorescence spectroscopy was thereby utilized for quantification of the protein adsorption. For blend multilayers prepared from 10 wt % PSS blends, the loss of PAA upon immersion in PBS drops to ∼10% of that in the outermost layer, with essentially no observable loss from higher PSS/PAA blend ratios. Correlating the fluorescence intensity of IgG-FITC adsorbed on the stable (PAH/PSS)8 film with the mass of IgG, as determined via QCM, enables quantification of the IgG-FITC adsorbed on the blend multilayers (Figure 6). The data reveal a systematic correlation between blend composition and protein adsorption (excluding the low PSS ratio blends of 0% and 10% where PAA desorption occurs). Salloum and Schlenoff,19 demonstrated that PAA-terminated multilayers are protein resistant due to their hydrophilic nature and possible excluded volume effects. (39) Malmsten, M. Biopolymers at interfaces, 2nd ed; Surfactant science series; Marcel Dekker: New York, 2003; Vol. 110, p 920. (40) Buijs, J.; Lichtenbelt, J. W. T.; Norde, W.; Lyklema, J. Colloids Surf., B 1995, 5, 11.

Polyelectrolyte Blend Multilayer Films

Figure 6. Adsorbed mass of FITC-labeled IgG on (PAH/blend)8 multilayers, as determined via fluorescence spectroscopy and QCM.

Therefore, blending PSS with PAA acts to reduce these hydrophilic repulsion effects and the film stability is enhanced. Furthermore, at the adsorption conditions of pH 7.4, both the multilayer surface and protein are like-charged, so minimal penetration of the protein into the multilayer is expected. Combined with the fact that surface roughness is essentially uniform for multilayers deposited from 25 wt % PSS and above (Figure 3), it is evident that hydrophobic interactions play a major role in IgG adsorption to these films. Interestingly, the correlation between contact angle and IgG adsorption is not linear, which may suggest more complex interactions are contributing to the total adsorbed mass; for example, the spatial distribution of hydrophobic domains associated with PSS at the interface may frustrate protein adsorption at low PSS blend ratios and become less inhibitive with further increases in the PSS composition.

Conclusions It has been demonstrated that the surface wettability of sequentially adsorbed PE multilayer films can be systematically varied by simply using blended dipping solutions of varying ratios (chemical composition). Film thickness measurements

Langmuir, Vol. 23, No. 9, 2007 4949

substantiate that both blend components are incorporated in the multilayer films throughout the entire blend spectrum despite their contrasting physical attributes at the assembly conditions of pH 3.5; PAA is hydrophilic and only weakly charged, whereas PSS contains hydrophobic aromatic groups and is highly charged. These dissimilar properties impart a decreasing surface roughness with increasing PSS content in the blend solutions (and hence multilayer films), while the film wettability systematically decreases from 7° to 53° with increasing PSS content in the blend solutions. The incorporation of salt into the PE dipping solutions results in an increase in the adsorbed film thickness due to enhanced charge screening and significantly enhances the surface roughness of the films. XPS analysis indicates that there is a surface compositional variation between the films constructed with and without salt in the PE solutions and that the interface is predominantly occupied by PSS. Films prepared from PAA blend solutions greater than 80 wt % are unstable in high-ionicstrength solutions, resulting in desorption of PAA. Stable blend films exist for films assembled from blend ratios containing in excess of 20 wt % PSS and reveal a systematic increase in protein adsorption with increasing PSS content in the blend solution. We have shown that the facile technique of LbL deposition is a convenient and accurate means of surface modification such that the surface morphology and film properties can be tailored from highly wettable to hydrophobic with a high protein affinity by simply altering the composition of the blend solutions. This work contributes to a better understanding of multilayer formation and control of surface characteristics for imparting surface specificity. Acknowledgment. Alison Mcgregor and Byoung-Suhk Kim are gratefully acknowledged for preliminary experiments, and Alexander N. Zelikin for synthesis of the RITC-labeled PAA. John F. Quinn is thanked for helpful discussions. This work was supported by the Australian Research Council under the Federation Fellowship and Discovery Project schemes, and the Victorian STI initiative. The Particulate Fluids Processing Centre (The University of Melbourne) is acknowledged for infrastructure support. Supporting Information Available: High-resolution XPS spectra of the C 1s and S 2p peaks showing the peak deconvolution. This material is available free of charge via the Internet at http://pubs.acs.org. LA0634746