Article pubs.acs.org/Langmuir
Layers over Layer-by-Layer Assemblies: Silanization of Polyelectrolyte Multilayers Ali Dirani, Antony E. Fernandes, Diana Ramirez Wong, Pascale Lipnik, Claude Poleunis, Bernard Nysten, Karine Glinel,* and Alain M. Jonas* Bio & Soft Matter, Institute of Condensed Matter and Nanosciences, Université Catholique de Louvain, Croix du Sud 1 L7.04.02, 1348 Louvain-la-Neuve, Belgium S Supporting Information *
ABSTRACT: The functionalization of poly(allylamine hydrochloride)/poly(acrylic acid) (PAH/PAA) polyelectrolyte multilayers by silanes reacted from the gas phase is studied depending on reaction time and temperature, pH of multilayer assembly, and nature of the reacting silane group. Whereas monochlorosilanes only diffuse in the multilayer and graft in limited amount, trichloro- and triethoxysilanes form rapidly a continuous gel layer on the surface of the multilayer, with a thickness of ca. 10−20 nm. The reactivity is lower in the strongly paired regime of the multilayers (neutral assembly conditions) but otherwise is not affected by the pH of multilayer assembly. Silanization considerably broadens the range of possible functionalities for (PAH/PAA) multilayers: hydrophobicity, surface-initiated polymerization, and grafting of fluorescent probes by the formation of disulfide bridges are demonstrated. Conversely, our results also broaden the range of substrates that can be functionalized by silanes, using (PAH/PAA) multilayers as ubiquitous anchoring layers.
■
INTRODUCTION Layer-by-layer assembly (LbL),1 which is the stepwise adsorption of polyelectrolytes in multilayers, and silanization,2 which is the assembly of silanes on hydroxylated surfaces such as silicon oxide, are two well-known methods for surface functionalization. Both share the trait of being excellent methods for the adaptation of inorganic surfaces to the requirements of organic or biological systems. However, they differ significantly in key aspects, making them complementary rather than competing solutions: for instance, while LbL films can be stratified in the vertical direction,3,4 silanization is essentially restricted to the formation of homogeneous (mono)layers (although exceptions exist5,6); while silane monolayers are covalently attached to their substrates, the adhesion of LbL assemblies is based on a multiplicity of weaker interactions; while silane monolayers are specific to some types of surfaces, LbL films can be deposited on virtually any substrate; finally, while the available range of chemical functions afforded by silanes is extremely large, functional waterprocessable polyelectrolytes are scarce. Combining the advantages of LbL and silanization would thus attractively open the application range of both techniques. It is obviously commonplace to grow LbL assemblies from surfaces modified by silanes, e.g., aminosilanes, in order to improve adhesion. However, depositing silanes over polyelectrolyte multilayers is much less frequent, although this could be an efficient way to introduce a large range of new chemical functions on the surface of a polyelectrolyte compartment. Here, we investigate this process for a series of silanes of different reactivity (monochloro-, trichloro- and triethoxysi© XXXX American Chemical Society
lanes), deposited from the gas phase to avoid the intricacies related to immersing polyelectrolyte multilayers in organic solvents; we also use the functional groups added by these silanes to continue building the films by different processes, such as surface-initiated polymerization or thiol−thiol coupling. As polyelectrolytic LbL components, we concentrate on the much studied poly(allylamine hydrochloride) (PAH)/poly(acrylic acid) (PAA) system. In addition to the wealth of data about the growth of this system depending on adsorption conditions,7 these multilayers offer the required chemical groups for the coupling of silanes. Silylation, the introduction of a substituted silyl group to a molecule, is a well-known practice in synthetic chemistry to protect sensitive functional groups such as alcohols, carboxylic acids, and amines.8 In analytical chemistry, silylation is a key derivatization strategy to improve detection of certain compounds by gas chromatography and mass spectrometry.9,10 The reaction mechanism consists of a SN2 nucleophilic attack on the electrophilic Si atom of the silylating donor (Scheme 1a); the reactivity is higher for hydroxyls, followed by carboxylic acids and then by amines. Protonated amines obviously do not react with silanes; hence, with a pKa of ca. 9, PAH is not expected to react significantly with chlorosilanes in LbL assemblies unless fabricated at high pH, which proves to be difficult.7 In contrast, carboxylic acids were reported to react with chlorosilanes in their protonated11−13 and ionized14 forms. Hence, PAA is Received: July 10, 2014 Revised: August 6, 2014
A
dx.doi.org/10.1021/la502729z | Langmuir XXXX, XXX, XXX−XXX
Langmuir
Article
synthesis of 2-bromo-2-methylpropionic acid 3-trichlorosilanylpropyl ester (ATRP-CS) and of 2-bromo-2-methylpropionic acid dimethylchlorosilanylpropyl ester (ATRP-DMCS) was adapted from previously published procedures.22 Di(ethylene glycol) methyl ether methacrylate (MEO2MA, 95%) and 2-hydroxyethyl methacrylate (HEMA, 97%) were from Sigma-Aldrich. CuCl (99.995%), CuCl2 (99.999%), and 2,2′-bipyridyl (bipy, 99%), were purchased from Sigma-Aldrich and used without further purification. Chloroamine T trihydrate (98%) from Sigma-Aldrich, 2-(N-morpholino)ethanesulfonic acid (MES) from Acros Organics, and rhodamine B PEG thiol from Interchim were used as received. Milli-Q water (resistivity 18.2 MΩ cm) was obtained from a Millipore Simplicity 185 system. Preparation of the Layers. Substrate Cleaning. The silicon substrates covered by a ∼1.5 nm thick native oxide layer were cleaned either by immersion in a hot freshly prepared piranha solution [H2SO4 (98%)/H2O2 (27%) 1/1 v/v] for 30 min (Caution: piranha solution is an extremely strong oxidant and should only be handled with proper equipment.) or by air plasma for 10 min at a power of 100 W, in an EMITECH K1050x plasma asher. The substrates were used immediately after cleaning in order to prevent surface contamination. Deposition of LbL Films. The polyelectrolyte multilayers were built by alternate deposition of PAH and PAA on the silicon wafers using a robotized dipping system. The concentration of polyelectrolyte solutions was 1 mg mL−1 in water (pH 5.5, unless otherwise noted). The substrates were first immersed for 3 min in the polycation solution and rinsed by three immersions in water at pH 5.5 for 1 s (15 times), 15 s (7 times) and 30 s (1 time). They were then immersed for 3 min in the polyanion solution and rinsed in the same fashion. The process was repeated nine times to obtain (PAH/PAA)9 multilayers. For a few samples, the pH of the polyelectrolyte and rinsing solutions was uniformly modified as stated in the text, and the number of PAH/ PAA layers was also varied. Silanization of the LbL Assemblies. Samples were placed on a Teflon sample holder, together with a reference silicon wafer, and introduced into a large Schlenk tube. The Schlenk was closed with a rubber septum and evacuated while being heated to 80 °C in an oil bath. Three to four vacuum/Ar cycles were then applied at 80 °C over a period of 1 h; 50 μL of silane was then injected and the closed system was left to react at 80 °C for a preset time, generally 2 h. After reaction, the samples were thoroughly rinsed with toluene and dried under a flow of nitrogen. For some samples, the reaction was performed at lower temperatures than 80 °C. Surface-Initiated ATRP. Polymer brushes were grown from ATRP initiator-silane modified LbL films or Si reference wafers according to a protocol adapted either from Jonas et al. (when led in a methanol/ water mixture)23,24 or Averick et al. (when led in water).25 The monomer (40 mmol), MEO2MA or HEMA, was dissolved either in a 22.5 mL water:MeOH 2:1 v:v mixture or in pure water, in a Schlenk sealed with a rubber septum. 2,2′-Bipyridine (2.18 mmol) and CuBr2 (0.9 mmol) were added to this solution, which was stirred and degassed with a stream of Ar for 1 h. CuCl (0.1 mmol) was then added quickly to the solution. The mixture was stirred and degassed (Ar) for 45 further min. Meanwhile, the substrates were sealed into Schlenk tubes and degassed (four vacuum/Ar cycles). The polymerization solution was then extracted with a syringe and quickly transferred to the Schlenk tubes. After various polymerization times at room temperature under inert atmosphere in the absence of stirring, the samples were removed, washed with water and then MeOH, and dried with a stream of N2. Grafting of Rhodamine B PEG Thiol over a MPTES-Modified LbL. The MPTES-modified LbL sample was immersed in a solution of rhodamine B PEG thiol (1 mg/mL) prepared in MES buffer (10 mM, pH 6.0) for 2 min, rinsed with water, and dried under N2. The sample was then immediately exposed to a solution of chloramine T (2 mM) in MES buffer (10 mM, pH 6.0) for 1 min. The sample was removed, washed with EtOH and then water, and dried with a stream of N2. The reference sample was prepared in the same fashion without immersion in chloramine T solution. Layer Characterization. X-ray Reflectometry (XRR). The measurements were carried out with a modified Siemens D5000
Scheme 1. (a) Reaction of Silanes with Carboxylic Acids and Aminesa and (b) Chemical Structures of the Silanes of This Study
a
X is a Cl or alkoxy moiety.
expected to be available for silylation irrespective of its assembly pH. The silanization of polyelectrolyte multilayers and related systems has been sparsely reported before, always with a practical perspective and no in-depth characterization of the resulting structures. The deposition of a fluorinated silane was performed on hybrid (PAH/PAA) multilayers roughened by the incorporation of silica nanoparticles, in order to generate superhydrophobic surfaces;15−17 this was done from the gas phase or in solution. A gel barrier of fluorinated silane was deposited on electrospun PAH/PAA fibers to create a diffusion barrier,18 and shells of silica gel were grown onto polyelectrolyte-coated micro- and nanoparticles by the hydrolysis of tetraethoxysilane from water:propanol solutions.19−21 Apart from these few studies, silanization of LbL assemblies was essentially ignored. Here, we mainly use X-ray reflectometry (XRR) and secondary ion mass spectrometry (SIMS), complemented by water contact angle measurements, ellipsometry, and fluorescence microscopy, to investigate in detail the silanization of (PAH/PAA) multilayers; we show that the method is able to provide access to a completely new range of systems combining the best of two worlds. As silanes (Scheme 1b), we use fluorinated alkylsilanes for hydrophobization purposes, atom transfer radical polymerization (ATRP)initiator silanes for the growth of polymer brushes (including temperature-responsive layers), and mercapto-substituted silanes, which offer opportunities for the coupling of functional macromolecules by the formation of disulfide bridges.
■
EXPERIMENTAL SECTION
Materials. Single-side-polished (100) silicon wafers were purchased from ACM and cut in rectangles of 1.5 × 3 cm2. PAH with an average molar mass of 58 000 g mol−1 and PAA with an average molar mass of 100 000 g mol−1 were supplied by Sigma-Aldrich. 3Mercaptopropyltriethoxysilane (MPTES, 95%), (heptadecafluoro1,1,2,2-tetrahydrodecyl) dimethylchlorosilane (FDMCS), (heptadecafluoro-1,1,2,2-tetrahydrodecyl)trichlorosilane (FCS), (heptadecafluoro-1,1,2,2-tetrahydrodecyl)triethoxysilane (FES), and tetrachlorosilane (TCS) were from ABCR Chemicals and used as received. The B
dx.doi.org/10.1021/la502729z | Langmuir XXXX, XXX, XXX−XXX
Langmuir
Article
Olympus) equipped with a CCD camera, a fluorescence illuminator, and various mirror units to detect different fluorochromes.
two-circle goniometer (0.002° positioning accuracy). X-rays of 0.15418 nm wavelength (Cu Kα) were obtained from a Rigaku rotating anode operated at 40 kV and 300 mA, fitted with a collimating mirror (Osmic) delivering a close-to-parallel beam of ∼0.0085° vertical angular divergence. The beam size was defined by a 40-μmwide slit placed 17.5 cm away from the focal spot. The sample was placed within 2 μm of the center of the goniometer, and the reflected beam was collected through a 200-μm-wide detector slit. Soller slits in the incident and reflected beam limited axial divergence to 0.02°. The data were corrected for spillover and normalized to unit incident intensity; they are reported as a function of kz0, the vertical component of the photon wavevector in a vacuum [kz0 = (2π/λ) sin θ, where θ is the angle between the incident ray and the sample surface and λ is the wavelength]. The XRR data were fitted using a model for the electron density profile of the films consisting of a superposition of layers of Gaussian roughness, except for the Si/multilayer interface, which was described by a succession of smooth slabs of 0.35 nm thickness. The method used the minimum number of layers and slabs required to represent properly the XRR data, and a regularization technique was employed to prevent spurious oscillations of the electron density, as described previously.26−28 Ellipsometry. Ellipsometry was performed with a spectroscopic ellipsometer Uvisel from Horiba-Jobin-Yvon at an incidence angle of ca. 70°, in a wavelength range from 400 to 850 nm. Ellipsometric data were fitted using the DeltaPsi 2 software with a three-layer model: silicon (bulk), native silicon oxide (1.5 nm thickness), and a single homogeneous film of adjustable thickness, including LbL, silane layer, and (if applicable) the supplementary top layer. The refractive index of this film was modeled as a transparent Cauchy layer, and the tabulated optical constants of Si and SiO2 were used. The measurements were carried out at least three times at different points on the substrate to obtain an average thickness. Time-of-Flight Secondary Ion Mass Spectrometry (ToF-SIMS). ToF-SIMS depth profiles were obtained with a time-of-flight secondary ion mass spectrometer ToF-SIMS V from IONTOF GmbH (Münster, Germany), operating in the dual-beam mode. The instrument was equipped with a Bin liquid metal ion gun, used as the analysis gun, and with an Ar cluster sputter ion gun that provided a mass-filtered (mass resolution of ca. 3.5) cluster beam with cluster sizes ranging from 500 to 10 000 atoms. Both ion columns are mounted at 45° with respect to the sample surface. The time-of-flight mass analyzer is positioned perpendicular to the sample surface. For this study, depth profiles were carried out in the “noninterlaced” mode, consisting of cycles of short pulses of Bi5+ ions for SIMS analysis followed by longer periods of Ar5000+ ion sputtering, during which the analyzer extraction potential was switched off. In the data reported in this paper, Ar5000+ ions had 10 keV impact energies. Beam currents were measured using a Faraday cup on the sample holder both before and after each experiment. For depth profiling, a focused beam of Ar5000+ primary ions was rastered over an area of typically 600 × 600 μm2. For analysis, the pulsed beam of 30 keV Bi5+ ions was rastered over a 200 × 200 μm2 area in the center of the sputter crater. To avoid the effects of uneven dose at the crater edges, the depth profiles were reconstructed by data processing from a smaller area at the center of the analyzed region (length ≤1/3 of the sputtering crater edge). Charge compensation with a low-energy (20 eV) electron flood gun was used during the sputtering and analysis cycles. A postacceleration voltage of 10 kV was applied in front of the detector for improved high mass sensitivity in the analysis stages. Water contact angle measurements. Water contact angle measurements were performed with an OCA-20 apparatus (Dataphysics Instruments GmbH) in the sessile drop configuration. The results are the average of three values each obtained with a droplet volume of 6 μL. Atomic Force Microscopy (AFM). AFM was performed with a Multimode AFM (Nanoscope IV, Bruker) in tapping mode using silicon cantilevers (with a force constant of ∼40 N m−1 and
