Article pubs.acs.org/Langmuir
Adsorption and Surfactant-Mediated Desorption of Poly(vinylpyrrolidone) on Plasma- and Piranha-Cleaned Silica Surfaces Wiebe M. de Vos,*,†,‡ Beatrice Cattoz,†,§ Michael P. Avery,†,∥,# Terence Cosgrove,† and Stuart W. Prescott†,⊥ †
School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, U.K. Membrane Science and Technology, Mesa+ Institute for Nanotechnology, University of Twente, Enschede, The Netherlands § Department of Pharmaceutical, Chemical and Environmental Sciences, Faculty of Engineering and Science, University of Greenwich, Medway Campus, Central Avenue, Chatham Maritime, Kent ME4 4TB, U.K. ∥ School of Physics, University of Bristol, Tyndall Avenue, Bristol BS8 1TL, U.K. ⊥ School of Chemical Engineering, University of New South Wales, Sydney, NSW 2052, Australia # Bristol Centre for Functional Nanomaterials, Centre for NSQI, University of Bristol, Tyndall Avenue, Bristol, BS8 1FD, U.K. ‡
ABSTRACT: Optical flow cell reflectometry was used to study the adsorption of poly(vinylpyrrolidone) (PVP) to a silica surface and the subsequent surfactant adsorption and polymer desorption upon exposure to the anionic surfactant sodium dodecyl sulfate (SDS). We have studied these effects as a function of pH and surfactant concentration, but also for two different methods of silica preparation, O2 plasma and piranha cleaning. As a function of pH, a plateau in the amount adsorbed of ∼0.6 mg/m2 is observed below a critical pH, above which the adsorption decreases to zero within 2−3 pH units. An increase in pH leads to dissociation of surface OH groups and a decreased potential for hydrogen bonding between the polymer and surface. For the plasma- and piranha-cleaned silica, the critical pH differs by 1−2 pH units, a reflection of the much larger amount of surface OH groups on piranha-cleaned silica (for a given pH). Subsequent rinsing of the adsorbed layer of PVP with an SDS solution leads to total or partial desorption of the PVP layer. Any remaining adsorbed PVP then acts as an adsorption site for SDS. A large difference between plasma- and piranha-cleaned silica is observed, with the PVP layer adsorbed to plasma-cleaned silica being much more susceptible to desorption by SDS. For a plasma-cleaned surface at pH 5.5, only 30% of the originally adsorbed PVP is remaining, while for piranha-cleaned silica, the pH can be increased to 10 before a similar reduction in the amount of adsorbed PVP is seen. For a given pH, piranha-cleaned silica has a higher surface charge, leading to a smaller amount of adsorbed SDS per PVP chain on a piranha-cleaned surface compared to a plasma-cleaned surface under identical conditions. In that way, the high negative surface charge makes desorption by negatively charged SDS more difficult. The high surface charge thus protects the neutral polymer from surfactant-mediated desorption.
■
INTRODUCTION
the surfactant and the polymer have no attractive interactions but can both adsorb to the surface, a clear competitive mechanism is observed,4−6 similar to the competition for interfaces between different polymer chains.7,8 Here parameters such as the molecular weight of the polymer chain and the concentration and exact composition of the surfactant will determine which of the components dominates the adsorption to the interface.6 Much more focus has, however, been on systems in which the surfactant and the polymer have attractive interactions, for example, for nonionic polymers such as poly(ethylene oxide) (PEO) and poly(vinylpyrrolidone)
Many advanced colloidal systems, for their application in, for example, laundry detergents, paints, cosmetics, and drug formulations, are based on mixed stabilizers.1−3 These products combine stabilizers, such as polymers and surfactants, to produce a high colloidal stability in concentrated mixtures for long durations of time. However, for mixed stabilization to work, it is essential that the polymers and surfactants behave in a synergistic way. Understanding how these components interact with one another, but also with the colloidal interfaces and any other materials present in the solution, is essential for the development of these advanced stabilization systems.1 With such clear industrial interest, it is hardly surprising that extensive studies have been performed on the mixed interaction of surfactants and polymers at the solid−liquid interface. When © 2014 American Chemical Society
Received: May 15, 2014 Revised: June 23, 2014 Published: June 30, 2014 8425
dx.doi.org/10.1021/la501877v | Langmuir 2014, 30, 8425−8431
Langmuir
Article
(PVP), with the surfactant sodium dodecyl sulfate (SDS).3,9−16 In these systems, the components have often been found to adsorb to the surface in a cooperative fashion. The combined adsorption of polymers and surfactants can significantly enhance or diminish colloidal stability.17 However, the exact nature of the polymer surfactant complex depends strongly on the surfactant concentration. For the interaction of SDS with preadsorbed layers of PEO9 and PVP,11 an increase in the surfactant concentration led to a decrease in the adsorbed amount of polymer. This can again be seen as a competitive mechanism; however, in this case, it is a competition between the surface and the surfactant to bind to the polymer chain, as SDS does not adsorb to the silica surface. Cattoz et al.18 showed that these competitive interactions can also be influenced by the addition of the nonionic surfactant C13E7. Mixed micelles of SDS and C13E7 bind less strongly to PVP polymer chains, reducing the desorption of PVP from a silica nanoparticle interface. A parameter that would be expected to have a strong influence on the structure of PEO/SDS and PVP/SDS complexes at the silica−water interface is the solution pH. It is well-established that the adsorption of PEO and PVP strongly depends on the pH.19,20 The adsorption of these nonionic polymers is governed by hydrogen bonding with silica hydroxyl (Si-OH) groups. However, the availability of these hydroxyl groups is determined by the solution pH, as at higher pH values the hydroxyl groups dissociate. Indeed, simply increasing the pH to greater than 10−11 can lead to complete desorption of PEO and PVP from the silica interface. The pH thus has the potential to act as a very powerful control mechanism for tuning the exact interaction between silica and, for example, a PVP/SDS complex. However, to the best of our knowledge, it has not been studied previously. Another often overlooked parameter in the adsorption of polymers and surfactants to silica surfaces is the pretreatment given to the silica surface. For adsorption experiments, especially on planar interfaces, it is essential that the surface be well cleaned and that the oxidation of the surface be reproducible, but to achieve this, researchers employ a wide variety of cleaning and treatment methods. For example, silica can be exposed to a gas flame21 or treated with a piranha solution (70% H2SO4 and 30% H2O2),6,22 with UV ozone,23 or with different types of plasma such as air,24 nitrogen, argon, or oxygen.25 DeRosa et al.26 compared a number of these methods (UV ozone and oxygen and H2O plasma). They showed that different treatments can result in a different surface chemistry as demonstrated by Fourier transform infrared (FTIR) but also by simple contact angle measurements with water. An important parameter is the density of connected (vicinal) hydroxyl groups at the surface, with a higher density relating to a smaller contact angle with water. Of the investigated methods reported there, oxygen plasma treatment gave a relatively large contact angle (30°) and thus (confirmed by FTIR) a relatively small amount of vicinal hydroxyl groups.26 Frantz and Granick27 investigated the effect of a number of different pretreatments (including piranha, UV ozone, and oxygen plasma) on the adsorption of polystyrene from a cyclohexene solution. While they found significant differences in the adsorption kinetics between the differently treated surfaces, the plateau of the amount adsorbed was found to be nearly identical. For topics other than polymer adsorption, much more research has been performed to show the importance of the
method used for cleaning silicon/silica surfaces. For example, Cras et al.28 demonstrated that the quality of silanization reactions on silica surfaces depends strongly on the method chosen to clean and/or prepare the surface. Donose et al.29 showed that the friction between an atomic force microscopy tip and a silica surface in part depended on the cleaning procedure used to prepare the silica surface. In this work, we study, using optical reflectometry, the adsorption of PVP to a planar silica surface and subsequently the effect of exposure to a SDS solution on the PVP layer. A key parameter in this study is the pH of the solution as it determines the degree of dissociation of surface hydroxyl groups and thus the ability of the surface to form hydrogen bonds with the PVP chain. Indeed, we will show that with just the pH, we can switch between a state in which the PVP chain and surfactant co-adsorb onto the surface and a state in which the surfactant completely desorbs the polymer. Other parameters under investigation are the ionic strength and the concentration of SDS. All these experiments were performed for two different methods of silica pretreatment, exposure to a piranha solution and exposure to oxygen plasma. In this way, we can determine if slight differences in surface chemistry can affect the adsorption of PVP chains and the subsequent surface interactions between PVP and SDS.
■
MATERIALS AND METHODS
Materials. All chemicals were used as supplied. Sodium dodecyl sulfate was supplied by Sigma-Aldrich. Polyvinylpyrrolidone (Mw ∼ 40 kg mol−1; radius of gyration Rg ∼ 30 Å) was obtained from PolyScience. Stock solvent solutions were prepared for each experiment at a salt concentration of 10 mM using Milli-Q water. The pH was adjusted using 1 and 0.1 M solutions of HCl and NaOH, as required, and measured using a Mettler Toledo Five Easy pH meter. Polymer and surfactant solutions were prepared by dissolving the materials in a stock solvent solution at the required pH and salt concentration. At pH 6−8, values particularly susceptible to pH drift due to absorption of atmospheric CO2, the pH was checked before and after each experiment but was never found to drift more than 0.15 pH unit. Fixed Angle Optical Reflectometry. Reflectometry measurements were performed using a standard fixed angle optical reflectometer using an impinging jet flow cell.30 The reflectometer was kindly provided by the Laboratory of Physical Chemistry and Colloid Science of Wageningen University (Wageningen, The Netherlands). The reflectometer uses a polarized beam of coherent light from a 1 mW He−Ne laser to measure the intensity of the parallel (Rp) and perpendicular (Rs) components of the light after reflection from a silica surface. Dividing Rp by Rs produces a value for the measured signal, S. The silica surfaces were prepared by oxidizing silicon wafers in an oven at 1000 °C and subsequent cleaning (see below). The resulting silicon surface has a silicon oxide thickness of ∼75 nm (as determined by ellipsometry), and this layer acts as an optical spacer allowing very small changes in refractive index at the surface to have a large effect on the measured signal. Prior to measurements, the system was calibrated by being flushed with solvent to obtain a baseline signal (S0). Upon adsorption of the polymer and/ or surfactant on the substrate layer, the change in signal is measured (ΔS = S − S0). The adsorbed amount, Γ, can be calculated from
Γ=
ΔS Q S0
(1)
where Q is an instrument- and material-dependent sensitivity factor. Q depends on the angle of incidence of the laser light (θ), the refractive indices of the surface and adsorbed materials (n), the thickness of the oxide layer on the silicon wafer (d), and the refractive index increment of the adsorbant (dn/dc). The following values were used: θ = 71°, 8426
dx.doi.org/10.1021/la501877v | Langmuir 2014, 30, 8425−8431
Langmuir
Article
nSiO2 = 1.46, nSi = 3.85, nH2O = 1.3327, nPVP = 1.51, dSiO2 = 75 nm, dn/ dcPVP = 0.175 mL/g, and dn/dcSDS = 0.108 mL/g. The value for Q was calculated using the program “Professor Huygens” produced by Dullware and freely available online (http://www.dullware.nl/ products/huygens/huygens.htm). In some cases, when high concentrations of polymer or surfactant are employed, a variation in signal (ΔS) can also be the result of a change in the solution refractive index. In this experiment, a small, unwanted signal change was expected for changes to and from SDS solutions as these were relatively high in concentration (50 mM = 14.4 g/L). We measured this change in signal by exposing a clean silica surface to a 50 mM SDS solution. As SDS does not adsorb to silica,10,11 any change in signal could be directly related to a change in the bulk refractive index. We found the change in signal (ΔS/S0) to be 0.0164, close to the theoretical value (as calculated using “Professor Huygens”) of 0.020, when adjusting the bulk refractive index according to dn/dcSDS. This experimentally determined correction factor was applied to all measurements. For lower SDS concentrations, we assumed (in line with theory) a linear dependence between concentration and correction factor. The silicon wafers [boron-doped, orientation of (100), and resistance of 12−18 Ω cm] were obtained from WackerSiltronic, and a 75 nm oxide layer was grown on the surface by heating at 1000 °C for 105 min. The wafers were then cut to size (∼1 cm × ∼5 cm) using a diamond-tipped pen and cleaned using one of two methods as described below. Wafers were always handled using tweezers, holding a predetermined nonmeasurement location. Plasma Cleaning. Wafers were stored in ethanol and then sonicated to remove any particles from the surface. They were then cleaned in a plasma cleaner (Femto model from Henniker Plasma) using pure O2 feed gas for 10 min at 50% power output. After being removed from the plasma cleaner, the wafers were once again rinsed with water, Milli-Q water, and ethanol, in that order, and dried using N2. After the wafers had been cleaned, a contact angle with Milli-Q water of ∼30° was measured. Piranha Cleaning. Wafers were stored in ethanol and then sonicated to remove any particles from the surface. The wafers were then rinsed with Milli-Q water and dried with compressed air before being soaked in a piranha solution (a 2:1 mixture of 97% sulfuric acid and 30% hydrogen peroxide) for at least 20 min. Caution: Piranha solution reacts violently with most organic materials and must be handled with extreme care. Once the wafers had been removed from the piranha solution, they were once again rinsed with water, Milli-Q water, and ethanol, in that order, and dried using N2. After the wafers had been cleaned, a contact angle with Milli-Q water of
