Aqueous Interface and

Feb 28, 2013 - YKI, Institute for Surface Chemistry, Box 5607, SE-114 86 Stockholm, ... The compactness of the PEO star polymers (molecular weight 1.2...
0 downloads 0 Views 873KB Size
Download PDF
Article pubs.acs.org/Langmuir

Poly(Ethylene Oxide) Star Polymer Adsorption at the Silica/Aqueous Interface and Displacement by Linear Poly(Ethylene Oxide) Trishna Saigal,† John K. Riley,† Patricia Lynn Golas,‡ Rasmus Bodvik,∥ Per M. Claesson,∥,⊥ Krzysztof Matyjaszewski,‡ and Robert D. Tilton†,§,* Center for Complex Fluids Engineering, †Department of Chemical Engineering, ‡Department of Chemistry and §Department of Biomedical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, United States ∥ KTH Royal Institute of Technology, School of Chemical Science and Engineering, Department of Chemistry, Surface and Corrosion Science, Drottning Kristinas väg 51, SE-100 44 Stockholm, Sweden ⊥ YKI, Institute for Surface Chemistry, Box 5607, SE-114 86 Stockholm, Sweden S Supporting Information *

ABSTRACT: Multiarm star copolymers with approximately 460 poly(ethylene oxide) (PEO) arms that have a degree of polymerization N = 45 were synthesized via atom transfer radical polymerization (ATRP) of PEO-methacrylate macromonomers in the presence of divinyl benzene cross-linkers. These are an example of molecular or nanoparticulate brushes that are of interest as steric stabilizers or boundary lubrication agents when adsorbed from solution to a solid/aqueous interface. We use ellipsometry to measure adsorption isotherms at the silica/aqueous interface for PEO star polymers and linear PEO chains having molecular weights comparable either to the star polymer or to the individual arms. The compactness of the PEO star polymers (molecular weight 1.2 × 106) yields a saturation surface excess concentration that is approximately 3.5 times greater than that of the high molecular weight (1 × 106) linear PEO. Adsorption of low molecular weight (6000) linear PEO was below the detection limit. Competitive adsorption experiments were conducted with ellipsometry, complemented by independent quartz crystal microbalance with dissipation (QCM-D) measurements. Linear PEO (high molecular weight) displaced preadsorbed PEO star polymers over the course of approximately 1.5 h, to form a mixed adsorbed layer having not only a significantly lower overall polymer surface excess concentration, but also a significantly greater amount of hydrodynamically entrapped water. Challenging a preadsorbed linear PEO (high molecular weight) layer with PEO star polymers produced no measurable change in the overall polymer surface excess concentration, but changes in the QCM-D energy dissipation and resonance frequency suggested that the introduction of PEO star polymers caused a slight swelling of the layer with a correspondingly small increase in entrapped water content.

■

INTRODUCTION

lyophobic anchor blocks and lyophilic buoy blocks. Alternatively, an electrostatic anchor block can be combined with a nonionic buoy block.22 Besides its simplicity, the adsorption approach is advantageous for its applicability to a wide variety of surface chemistries. Nevertheless, it does have limitations. Block copolymer adsorption often fails to deliver high brush densities, because steric repulsions among buoy blocks on neighboring chains limit the number of adsorbed chains per unit area that can be achieved.11,23,24 Furthermore, when the lyophilic buoy block has a thermodynamic affinity for the surface, it may be able to compete effectively with the intended anchor block for adsorption. When this occurs, adsorption produces a loop-train-tail conformation typically expected of an adsorbed homopolymer rather than the desired brush

Star polymers are compact, high molecular weight structures with multiple linear chains emanating as “arms” from a central core.1−5 If linked to the core at high density, then the chains are stretched so that the structure is a nanoscale brush. Interest in polymer brushes is motivated by numerous surface treatment applications. Under good solvent conditions, brushes provide steric repulsive forces that stabilize colloidal suspensions6−9 and also provide friction reduction between surfaces in sliding contact.10−19 PEO brushes are of particular interest because of the nonadhesive interaction of PEO with proteins and cells.20 PEO is well-known to reduce protein adsorption to surfaces, and PEO brushes offer superior passivation performance compared to adsorbed linear PEO chains.21 To achieve the most robust steric repulsion, a brush should have a high grafting density that produces a high degree of chain stretching. The simplest strategy that can be used to form a brush is to adsorb amphiphilic block copolymers that have © 2013 American Chemical Society

Received: December 21, 2012 Revised: February 28, 2013 Published: February 28, 2013 3999

dx.doi.org/10.1021/la305085a | Langmuir 2013, 29, 3999−4007

Langmuir

Article

Scheme 1. Synthetic Scheme for Copolymerization of PEOMA with DVB Initiated by Ethyl-2-Bromoisobutyrate to Produce PEO Star Polymers

structure.25 Similar limitations can apply when covalently endgrafting preformed polymer chains to a surface. To achieve a high brush grafting density, a “grafting from” approach is more effective, whereby polymer chains are grown by polymerization of monomers from surface-bound initiators.26,27 Densely grafted brush layers can also be prepared by Langmuir−Blodgett transfer of compressed block copolymer monolayers from a liquid/vapor interface to a solid substrate.6,14,16 These latter methods can be quite effective, but compared to adsorption-based methods, they are procedurally more complex and can be limited in terms of the surface chemistries or shapes of surfaces to which they can be applied. A different approach is to adsorb “pre-assembled” brushes. For example, adsorption of poly(tert-butylstyrene-block-sodium 4-styrenesulfonate) micelles to octadecyltrichlorosilane surfaces created a brush structure by postadsorption rearrangement of the adsorbed micelles.28 A similar approach that is more closely related to the current work is to adsorb linear bottlebrush polymers, a form of molecular brush, to a surface.18,29−33 Side chain extension is ensured by the steric constraints imposed by their close proximity along the polymer backbone. The degree of side chain extension can be tuned by the proportion of side chain-bearing repeat units in the polymer backbone. When the bottlebrush polymer adsorbs, some fraction of the side chains extend from the surface as desired. The balance between side chains extending into solution and side chains orienting along the surface depends on the relative surface affinities of the side chains and the backbone. Side chain extension into solution is particularly strong when only the backbone is surface active.32 Recently, extended layers were shown to form when a linear cationic anchor block was combined with a nonionic bottlebrush block built from a similar PEOMA macromonomer as the one used in the current study, to form a “brush of molecular brushes”.22 The application effectiveness of a brush-like layer prepared by adsorbing molecular brushes depends on the density of the layer that can be produced by adsorption. We extend the concept of adsorbing molecular brushes to examine structures of higher compactness than linear bottlebrush polymers, namely star polymers, in an attempt to obtain even denser adsorbed layers. Adsorbed star polymers can adopt different conformations, depending on the number of arms, solvent quality and adsorption strength of arm or core segments.34−37 High arm adsorption strength can produce a flattened conformation, but as the number of arms increases, stars adopt a “sombrero” structure34,37 with a broad flattened region and a central segment with a curved structure more similar to the star structure in solution. Decreasing arm adsorption strength favors adsorbed star conformations more like deformed droplets. When stars are adsorbed via surface active cores with nonadsorbing arms, they create a brush with arms extending into solution but the extent of adsorption decreases

as the number of arms increases. Structures that adsorb densely and extend numerous arms into solution may prove to be effective steric stabilizers or boundary lubrication agents. In the current study, poly(ethylene oxide) (PEO) star polymers were created by atom transfer radical polymerization38−40 of PEO-methacrylate macromonomers (PEOMA) in the presence of divinyl benzene, a bifunctional cross-linker that leads to the formation of a core−corona, or multiarm star polymer, structure. Ellipsometry measurements indicate that PEO star polymers produce significantly larger saturation surface excess concentrations than linear PEO chains of comparable molecular weight, due to greater efficiency of packing mass on the surface. While higher surface excess concentrations are achieved by adsorbing PEO stars, competitive adsorption experiments conducted by ellipsometry and by quartz crystal microbalance with dissipation (QCM-D) indicate that linear PEO adsorbs preferentially in competition with PEO stars, consistent with previously published theory.37

■

EXPERIMENTAL SECTION

Materials. PEO star polymer synthesis by atom transfer radical polymerization of macromonomer PEO methacrylate (PEOMA, molecular weight 2000), with divinylbenzene (DVB) comonomer as a cross-linker (Scheme 1) was described in detail previously.41 Gel permeation chromatography indicated 75% conversion of PEOMA. Gas chromatography analysis indicated that DVB conversion exceeded 90%. Thus, the resulting polymer was approximately 20 mol % PEOMA (corresponding to approximately 80 wt % PEOMA, given the relative molecular weights of DVB and PEOMA). The copper catalyst was removed by passing the solution through a column filled with neutral alumina. Polymer was purified by dialysis against methanol for one day, followed by dialysis against deionized water for 7 days (membrane molecular weight cut off = 25 000) with once daily water changes. The resulting solution was filtered through a 0.2 μm Nylon syringe filter, and the final polymer concentration was determined gravimetrically. Linear PEO samples with molecular weight 1 × 106 (Polysciences) or 6000 (Fluka) were purchased and used as received. All water was first deionized by reverse osmosis and purified to 18.2 MΩcm resistivity using a Barnstead Nanopure Diamond system. Characterization. The size and molecular weight of the PEO star polymers were measured by aqueous size exclusion chromatography− multiangle light scattering (SEC-MALS) using a DAWN HELEOS-II MALS + DLS system (Wyatt Technology) plus Optilab T-rEX refractive index detector, fed by an Agilent HPLC 1100 Series HPLC system (Agilent Technologies) with Superdex 75 10/300 GL (GE Healthcare) SEC column. Sample injection volume was 15 μL with an eluent flow rate of 0.7 mL/min. Adsorption Substrates. Ellipsometry adsorption experiments were carried out on polished silicon wafers (International Wafer Service Inc.), oxidized in air at 1000 °C for 15 min to create an approximately 40 − 50 nm thick oxide film on the surface. Surfaces were cleaned with No-Chromix (Sigma-Aldrich Company) for 20 min, followed by 20 min in 6 N HCl (Fisher Scientific) and 30 min in 10 mM NaOH (Fisher Scientific). In between treatment solutions, the wafers were rinsed with copious amounts of deionized water. This left 4000

dx.doi.org/10.1021/la305085a | Langmuir 2013, 29, 3999−4007

Langmuir

Article

the surfaces wetted by deionized water. The oxide layer thickness was measured in water prior to adsorption experiments using phase modulation ellipsometry (Picometer Ellipsometer, Beaglehole Instruments).42,43 QCM-D experiments were performed with quartz crystals having 5 MHz fundamental frequency coated with a 50 nm silica film (Q-Sense). These were cleaned with a 2 wt % Hellmanex solution (Hellma), rinsed copiously with deionized water and stored in ethanol until use. Adsorption Measurements. Surface excess concentrations were measured in situ using phase modulation ellipsometry with a Picometer Ellipsometer (Beaglehole Instruments).42,43 It detects adsorption by the change in ellipticity ρ̃ of a polarized 632 nm laser beam upon reflection from a surface. The angle of incidence is set to 72°, near the Brewster angle for silicon/water, in order to increase sensitivity. Further details are provided in the Supporting Information. Analyzing ellipticity changes according to a homogeneous two-layer surface model (bulk silicon + oxide layer + adsorbed layer + bulk solution) yields the optical average thickness d1 and refractive index n1 of the adsorbed layer, from which the surface excess concentration Γ is calculated as follows:44 Γ=

d1(n1 − no) dn/dc

observed. All results were replicated in at least two independent experiments. All experiments were conducted in 18.2 mΩcm resistivity deionized water at a temperature of 23 ± 0.1 °C. QCM-D measurements were conducted with a Q-Sense E4 microbalance. QCM-D principles are detailed elsewhere.46 QCM-D detects adsorbed mass primarily by the decrease in the quartz crystal resonance frequency. Both adsorbed polymer and hydrodynamically trapped water are detected by this method. The frequency shift and energy dissipation signals contained in the fifth, seventh and ninth overtones were analyzed simultaneously according to the extended viscoelastic film model47,48 using Q-Tools software (Q-Sense). Comparison of polymer mass detected by ellipsometry with total sensed mass detected by QCM-D allowed an estimate of the trapped water content in the adsorbed layer.49−51

■

RESULTS AND DISCUSSION PEO Star Polymer Characterization. MALS + dynamic light scattering analysis indicated a weight average molecular weight Mw = 1.2 × 106 (Mw/Mn = 1.08), and intensity-weighted average RG = 14 ± 0.3 nm and RH = 13 ± 0.5 nm (number average RG = 12 ± 0.2 nm and RH = 11 ± 0.4 nm). The material is 80 wt % PEOMA, so the overall molecular weight of 1.2 × 106 indicates there are approximately 460 PEO arms per star. Although the detailed internal distribution of mass and refractive index in the star polymers is not known, RG being slightly greater than RH would indicate these star polymers are not spherical but somewhat elongated, assuming they behaved as rigid bodies.52 In situ AFM images of the star polymers adsorbed at low coverage on an oxidized Si wafer (SI Figure S1) show prolate ellipsoidal objects having dimensions consistent with the light scattering measurements, although of course it must be noted that these images show deformable objects on a solid substrate. We interpret the structure of the star polymers using the starlike micelle model of Vagberg and co-workers,53 originally developed to describe the dimensions of block copolymer micelles that have a corona of polymer chains surrounding a core of finite size. The radius of such a particle is predicted as follows:

(1)

where no is the bulk refractive index. The refractive index increment of the polymer solution dn/dc was measured by differential refractometry (Phoenix Precision Instrument Co.) for both linear PEO and PEO star polymers, finding dn/dc = 0.13 cm3/g for each. TF Companion software (Version 3.0, Semicon Software Inc.) was used for ellipsometry analysis. The optical average thickness of the adsorbed layer was determined after the ellipsometry signal had reached a steady value. The adsorbed layer is the only one with an unknown refractive index. Thickness values were determined assuming one of three possible adsorbed layer refractive indices, n1 equals either 1.37, 1.40, or 1.44, spanning a reasonable range of values between pure water (1.333) and pure PEO (1.454). The thickness and refractive index of the adsorbed layer are highly coupled so that their individual values are highly model-dependent for very thin films, but their errors are mutually compensating and the quantity d1(n1 − no) is nearly invariant, with the result that the surface excess concentration calculated by eq 1 is nearly independent of the optical assumptions. The assumed value of n1 had only a minor effect on Γ, as noted below in the Results and Discussion. Time-dependent surface excess concentration values were plotted based on the proportionality between Γ and the Beaglehole ellipsometric “y-parameter”42,43 defined in the Supporting Information, SI. Ellipsometry experiments were conducted in a custom liquid cell, with inlet and outlet ports, an open top, and the wafer resting on the bottom of the cell. The incident beam was passed to the wafer, and the reflected beam passed to the detector, via optical light guides that transit the air/water interface, similar to the system described by Benjamins and co-workers.45 The 12 mL liquid cell was always filled first with deionized water containing no polymer. Solution contents were changed by continuously pumping new solution through the cell at 1.5 mL/min, until at least 5 cell volumes had passed through the cell. The remainder of the experiment was conducted with a quiescent solution to conserve material. The liquid cell and tubing were cleaned with RBS detergent (Thermo Scientific) and rinsed extensively with deionized water before and after each experiment. For a given bulk concentration, surface concentrations were measured by one of two different procedures in separate experimentseither single-shot or sequential adsorption experiments. Single-shot adsorption experiments began with a bare surface, and the polymer was adsorbed directly from a solution of the desired concentration over a period of 6−10 h, until a constant surface excess concentration was achieved. Sequential adsorption experiments proceeded via stepwise increases in bulk concentration. In one sequential adsorption series, 30 min were allowed for each concentration increment. One hour was allowed per increment in another series of measurements, enabling some dynamic features to be

⎛ 8 Nf (1 − ν)2ν ⎞ν 1/ ν 1/ ν ⎟ ⎜ Rp = ⎜ a + Rc ⎟ 1/ ν ⎝3 4 ν ⎠

(2)

where Rp and Rc are the radii of the particle and core respectively, N = 45 is the number of segments in each arm, a = 0.36 nm is the statistical length of an EO segment,54 f = 460 is the number of arms attached to the core, and ν = 3/5 is the Flory exponent for a good solvent. The first term in parentheses in eq 2 accounts for the spherical corona of polymer arms and is derived from the blob model of Daoud and Cotton.55 Assuming a dense spherical core composed entirely of pDVB, the core radius is as follows: 1/3 ⎛ M pDVB ⎞ 3 ⎟ R c = ⎜⎜ ⎟ ⎝ 4π NAρpDVB ⎠

(3)

where MpDVB = 240 000 is the molecular weight of pDVB in the core determined from conversion data, NA is Avogadro’s number, and ρpDVB = 1.02 g/cm3 is the bulk density of pDVB. Equation 3 indicates Rc = 4.5 nm. This confirms that the core has a non-negligible size, in keeping with the premise of this model. With this core radius, eq 2 predicts Rp = 14 nm, consistent with the light scattering measurements. Using the measured RG and the calculated core radius, the PEO arms in 4001

dx.doi.org/10.1021/la305085a | Langmuir 2013, 29, 3999−4007

Langmuir

Article

the star polymers have an estimated length Rarm ≈ RG − Rc ≈ 8 nm. This is compared to the unperturbed root mean squared radius of gyration that the side chains would have in aqueous solution, given in nm for a PEO chain as56 R g2⟩1/2 = 0.445N 0.58

optical properties of the layer. The refractive index and thickness of the adsorbed layers will be discussed further below, after considering QCM-D results. PEO star polymers adsorbed in single-shot experiments showed a high affinity adsorption isotherm on silica, with a plateau surface excess concentration of approximately 2.7 ± 0.15 mg/m2, whereas the high molecular weight linear PEO adsorption maximum was approximately 0.7 ± 0.04 mg/m2. The latter is consistent with literature values for high molecular weight PEO adsorption to silica.57−59 The PEO arms account for approximately 80 wt % of the star structure, so 2.7 mg/m2 corresponds to a surface excess concentration of approximately 2.2 mg/m2 of PEO, which is still three times the amount of PEO adsorption achieved by the linear polymer. This enhancement of PEO adsorption does not happen with amphiphilic, PEO-containing nonionic block copolymers on silica in water, as noted above.25 PEO star polymer adsorption was also compared to the adsorption that would be displayed by small chains that are comparable in size to the PEO side chains. Linear 6000 molecular weight PEO was adsorbed to silica in a similar bulk concentration range, but its surface excess concentration was consistently less than 0.07 mg/m2 and difficult to distinguish from noise in the ellipsometry signal. A previous study by Naderi and co-workers32 considered linear PEO bottlebrush polymers that were also based on 2000 molecular weight PEO side chains. These had a molecular weight of 4 × 105 and reached a plateau surface excess concentration of 1.17 mg/m2 on silica. This falls between the surface concentrations achieved by the linear PEO and the PEO star polymers measured here. The bottlebrush polymer yielded higher surface concentrations than 5 × 105 molecular weight linear PEO in that study, consistent with the current observation of significantly increased adsorption provided by the branched polymer structure. Their PEO bottlebrush layer was significantly denser than the linear PEO layer: viscous energy dissipation (QCM-D) and steric force measurements showed that the linear PEO layer was thicker than the bottlebrush layer. The bottlebrush polymer displayed a steeper, shorter-ranged steric repulsion between opposing surfaces. The thickness of the linear PEO layer was established by long loop and tail segments, whereas the thickness of the bottlebrush was established mainly by the extension of the low molecular weight PEO side chains as individual tails protruding into solution. Olanya and co-workers60 studied statistical copolymers of PEOMA and cationic methacryloxyethyl trimethylammonium chloride (PMETAC) adsorbing on silica. Those copolymers had a cationic backbone and the same 2000 molecular weight PEO side chains, and the charge density was varied by changing the monomer ratio. The extent of adsorption was pH- and ionic strength-dependent, and the maximum surface concentration occurred at a moderate degree of charge incorporation. In all cases, the maximum surface concentration was less than that produced by the PEO star polymer studied here. At larger backbone charge densities, the chains tended to lay flatter on the surface, leading to lower surface concentrations. The observation that PEO star polymers gave still larger surface concentrations than the linear PEO bottlebrush polymers shows that increasing the compactness of the preformed PEO brush structure significantly favors denser packing on the surface. The intramolecular cross-links in the star polymer core inhibit its spreading on the surface and allow

(4)

For N = 45, this is 4.0 nm. Thus, the arms are estimated to be stretched to a length approximately twice that of free chains. Adsorption Isotherms for PEO Star Polymers and Linear PEO on Silica. The quasi-equilibrium, single shot adsorption isotherm and the apparent isotherms obtained by stepwise sequential adsorption for PEO star polymers and for 1.0 × 106 molecular weight linear PEO on silica are shown in Figure 1. The significant dependence of the star polymer

Figure 1. Surface excess concentrations of 1.2 × 106 molecular weight PEO star polymers (unfilled symbols) or 1.0 × 106 molecular weight linear PEO (filled symbols) on silica attained via different adsorption procedures: sequential adsorption experiments where 30 min (○ and ●) or 1 h (Δ and ▲) were allowed for adsorption after each of several stepwise increases in bulk concentration, or single-shot experiments where polymers adsorbed to an initially bare substrate for 6−10 h (□ and ■). Data points are means of at least two replicate experiments, with error among replicates of approximately ±6%. Lines are drawn to guide the eye.

surface excess concentrations on the adsorption protocol (stepwise or single-shot) indicates that slow relaxation processes affect the conformation and packing of PEO stars on the silica surface. At each stage of the sequential adsorption experiments, the surface excess concentration had reached a steady value before increasing the bulk concentration for the next stage. This history-dependent extent of adsorption is typical of macromolecules that experience significant conformational relaxation on the surface on time scales that are comparable to the overall adsorption time scale. The high molecular weight linear PEO chains exhibited no such dependence on adsorption procedure. A small uncertainty in surface excess concentration arises based on the assumed layer refractive index used to interpret ellipsometry data, but the uncertainty produced by assuming n1 = 1.37, 1.40, or 1.45 when calculating the adsorbed layer optical average thickness d1 was only ∼5−6% of the reported surface excess concentration. This broad range of assumed refractive indices was chosen to verify that the difference in adsorbed amount between the PEO star polymers and the linear PEO was outside any error that could be associated with the assumed 4002

dx.doi.org/10.1021/la305085a | Langmuir 2013, 29, 3999−4007

Langmuir

Article

it to maintain a compact structure that packs mass more efficiently than the linear bottlebrush polymer. Examination of the ∼2.7 mg/m2 plateau surface concentration for PEO star polymers in light of the 1.2 × 106 molecular weight and 12 nm number-average RG, suggests that approximately 60% of the surface was occupied by PEO stars. This area fraction estimate is sensitive to the star polymer size distribution since these are not monodisperse objects, and it is based on the simple assumption that the stars occupy an excluded area equal to a circle of radius RG (assumes star polymers are in the droplet conformation, rather than the sombrero conformation) so the 60% value calculated here should not be overinterpreted. Yet, it is noteworthy that it comparable to the approximately 52% jamming limit for random sequential adsorption of low aspect ratio objects.61 While PEO star adsorption produced a surface excess concentration that significantly surpassed that of a linear PEO bottlebrush polymer on silica, it produced the same extent of adsorption as a diblock copolymer consisting of a cationic PMETAC anchor block and poly(PEOMA) bottlebrush buoy block.22 That cationic anchor/PEO bottlebrush diblock copolymer adsorbed to 2.75 mg/m2. The strong electrostatic surface affinity of the PMETAC anchor block allowed the bottlebrush block to extend from the silica surface, achieving a high surface excess concentration by forming a thick, heavily hydrated (94.5% water content) layer. The quantitative similarity of surface excess concentrations (2.7 and 2.75 mg/ m2) is a coincidence, since even higher surface excess concentrations would be expected for the PMETAC-poly(PEOMA) block copolymer if the poly(PEOMA) bottlebrush block had been longer, as it appears to be strongly excluded from the surface by the anchor block. Competitive Adsorption of PEO Star Polymers and Linear PEO. Ellipsometry results in Figure 2 demonstrate that high molecular weight linear PEO can displace preadsorbed PEO star polymers from the silica surface. Competitive adsorption was investigated by allowing either a PEO star polymer layer or a linear PEO layer to reach a steady surface concentration, and then switching to a solution of the other polymer type. When linear PEO was preadsorbed and then challenged by a solution of PEO star polymers, there was no detectable change in the ellipsometry signal, and the surface excess concentration remained constant at 0.7 mg/m2 for more than an hour of competitive adsorption, despite the fact that PEO star polymers produced 3-fold higher surface concentrations than linear PEO in a conventional single-component adsorption experiment. In contrast, when a preadsorbed PEO star polymer layer was challenged with a linear PEO solution, the surface excess concentration decreased significantly. Although it had not yet decreased all the way to the 0.7 mg/m2 surface excess concentration characteristic of linear PEO adsorption, the surface concentration was still decreasing steadily after 80 min of linear PEO exposure. Separate experiments were conducted wherein the PEO star polymer concentration was again 0.01 wt % but the linear PEO concentration was 0.1 wt %. Similar results were obtained (SI), but the surface excess concentration appeared to have leveled off at approximately 1.5 mg/m2 after approximately 25 min, still exceeding the normal linear PEO surface concentration. This suggests the final layer contained some residual PEO star polymers. Independent experiments were also conducted to measure the degree to which either PEO star polymers or linear PEO chains desorb during a rinse

Figure 2. In situ competitive adsorption of 1.2 × 106 molecular weight PEO star polymers and 1.0 × 106 molecular weight linear PEO from 0.01 wt % solutions of either polymer. The red (upper) curve illustrates PEO star polymer preadsorption followed by switching the solution to linear PEO after 60 min. The lag in response after the PEO star polymer to linear PEO switch is due in part to a lag time of ∼15 min associated with the changeover in flowcell solution contents. The blue (lower) curve illustrates linear PEO adsorption followed by switching the solution to PEO star polymer at 60 min. Simultaneous coadsorption from a solution of 0.01 wt % linear PEO and 0.01 wt % PEO star polymer is shown in black and overlays the blue curve for linear PEO followed by PEO star polymer. Each experiment began with a four minute recording of a stable baseline before starting to flow the polymer solution.

with polymer-free water. As is normally observed with high molecular weight macromolecules, neither polymer showed any detectable desorption over approximately one hour of rinsing (SI). Thus, since PEO star polymers do not desorb spontaneously, the decrease in surface excess concentration during the competitive adsorption experiment must be due to PEO star displacement by adsorbing linear PEO chains and the formation of a mixed layer. The results of simultaneous coadsorption from a mixed solution containing equal concentrations (0.01 wt %) of PEO star polymers and linear PEO were consistent with the competitive adsorption experiments. The adsorption kinetics and final surface excess concentration achieved by coadsorption were nearly indistinguishable from linear PEO adsorption. Since the hydrodynamics are not well controlled in this ellipsometry cell, the initial adsorption kinetics should not be interpreted quantitatively, but simple qualitative comparisons can be made. The diffusion coefficient of 1.2 × 106 molecular weight PEO star polymers is larger than that of 106 molecular weight linear PEO (compare RH = 13 nm for PEO star polymers vs 38 nm for 106 molecular weight linear PEO in water reported elsewhere62). Empirically, we observe that initial adsorption rates were similar for PEO star polymers and linear PEO and for the coadsorption experiments. When considering the fact that PEO star polymers have larger diffusion coefficients, this observation indicates that some PEO stars must have been able to adsorb, at least transiently, during the early stages of the coadsorption experiment, but the thermodynamically preferential adsorption of linear PEO ensured that the final layer was dominated by the linear polymer. Complementary QCM-D experiments, Figures 3 and 4, support the finding that high molecular weight linear PEO 4003

dx.doi.org/10.1021/la305085a | Langmuir 2013, 29, 3999−4007

Langmuir

Article

Simultaneous analysis of ΔF and ΔD for overtones 5, 7, and 9 by the extended viscoelastic film model yielded the total sensed mass ΓQCM for each stage of the QCM-D experiments, as well as the apparent thickness of the layer.50,51 The analysis assumed a density of 1050 kg/m3 for the adsorbed layer. The sensed mass calculated by the extended viscoelastic model was within 10% of the value calculated by the simpler Voigt analysis that assumes frequency independent viscoelastic moduli.46,51 Furthermore, the trapped water content of the adsorbed layer was estimated on a mass basis by comparing the polymer surface excess concentration determined by ellipsometry and the total sensed mass detected by QCM-D,50 as ϕw = 1 − ΓEllips/ΓQCM (see Table 1), bearing in mind that the

Figure 3. 1 ×106 molecular weight linear PEO challenged by PEO star polymers. QCM-D results obtained when first adsorbing from a 0.01 wt % linear PEO solution from t = 25 min to t = 80 min, then switching to a 0.1 wt % PEO star solution at t = 80 min. Curves for frequency shift (ΔF) or dissipation change (ΔD) are indicated by their corresponding overtone number. Spikes in the data near the switch in solution contents are due to transient pressure disturbances.

Table 1. Summary and Interpretation of QCM-D Results after the First Stage of Adsorption and after the Second Stage of Challenging the Pre-Adsorbed Layera

ellipsometry Γ (mg/m2) QCM-D Γ (mg/m2) dissipation ×106 trapped water content QCM-D thickness (nm) effective ellipsometric thickness (nm)

Figure 4. PEO stars challenged by 1 × 106 molecular weight linear PEO. QCM-D results obtained when first adsorbing from a 0.1 wt % PEO star solution from t = 20 min to t = 45 min, then switching to a 0.01 wt % linear PEO solution at t = 45 min. Curves for frequency shift (ΔF) or dissipation change (ΔD) are indicated by their corresponding overtone number.

PEO star pre-adsorption

linear PEO pre-adsorption

1st stage: PEO star

2nd stage: linear PEO

1st stage: linear PEO

2nd stage: PEO star

2.7 ± 0.2

1.5 ± 0.1

0.7 ± 0.04

0.7 ± 0.04

7.9 ± 0.03

8.3 ± 0.2

5.4 ± 0.5

6.4 ± 0.2

1.75 to 2.25 ∼66%

3.0 to 3.4 ∼82%

2.75 to 3.25 ∼87%

2.8 to 2.9 ∼89%

7.5 ± 0.03

7.9 ± 0.2

5.2 ± 0.5

6.1 ± 0.2

8.8

8.9

5.4

5.6

a

These data correspond to 0.01 wt % PEO star solutions and 0.1 wt % linear PEO solution (see SI for ellipsometry data for these competitive adsorption conditions).

ellipsometry and QCM-D measurements were conducted separately in different experiments. Linear PEO produced a layer that had lower total sensed mass, contained a significantly higher fraction of trapped water and was more dissipative than the layer produced by PEO star polymers. The larger dissipation combined with larger trapped water content indicates that the linear PEO layer had a lower density and was less rigid than the PEO star layer. This is consistent with Naderi and co-workers’ observations32 regarding the relative densities of adsorbed linear PEO and PEO bottlebrush polymer layers on silica. After switching solution contents to initiate competitive adsorption, challenging a preadsorbed PEO star polymer layer with a linear PEO solution (Figure 4) produced a slow, modest decrease in the magnitude of ΔF and a more rapid and significant increase in dissipation from the values established by the adsorbed PEO star layer. ΔD was restored to a value similar to that achieved by linear PEO adsorption to a bare surface. Data analysis indicated that the total sensed mass increased slightly when linear PEO challenged the preadsorbed PEO star layer, whereas ellipsometry indicated a significant decrease in polymer surface concentration. Neither method can determine the relative amounts of PEO star polymer and linear PEO in the layer, but comparing the QCM-D total sensed mass and the ellipsometric polymer surface excess concentration provides a useful estimate of the hydrodynamically entrapped water content of the layer.49−51 This comparison indicates that the competitive adsorption process significantly increased the

chains displace preadsorbed PEO star polymers, but they also reveal a richness in both competitive adsorption scenarios that goes beyond what can be determined by ellipsometry alone. In QCM-D experiments, larger amounts of adsorbed mass produce larger negative changes ΔF in the crystal resonance frequency, although the relationship between ΔF and total sensed mass is not linear in systems that produce significant energy dissipation. Whereas ellipsometry detects only the adsorbed polymer, QCM-D detects the mass of adsorbed polymer as well as water of hydration and water that is otherwise hydrodynamically trapped within the layer. QCM-D also provides a measure of energy dissipation to the surrounding solution. The dissipation ΔD tends to be larger for thicker or less rigid layers. Before discussing the competitive adsorption portions of each QCM-D experiment, consider the values attained for ΔF and ΔD (each normalized by overtone number) for PEO star polymers or linear PEO each adsorbing on its own. Examining the data immediately before the change in solution contents, one observes that linear PEO (Figure 3) produced ΔF of approximately −15 Hz and ΔD ranging from 2.75 to 3.25 × 10−6 for overtones 5, 7, and 9, while PEO star polymers (Figure 4) produced ΔF of approximately −40 Hz and ΔD of approximately 1.75 to 2.25 × 10−6 for the same overtones. This is consistent with the significantly larger PEO star adsorbed mass detected by ellipsometry. 4004

dx.doi.org/10.1021/la305085a | Langmuir 2013, 29, 3999−4007

Langmuir

Article

polymers may remain adsorbed by forming loops when the segmental adsorption energy is weakened or when the surface becomes crowded, star polymer arms tend to desorb. Simulations of isolated star polymers predict that weakening adsorption strength is associated with a change in conformation from the sombrero structure to a truncated spherical structure resembling a sessile droplet.34 At higher surface coverage excluded volume interactions lead to arm desorption, and the star polymer tends to lift up from the surface in the simulations. This transition to a more drop-like structure has been observed by atomic force microscopy, although that experiment was performed after quenching to poor solvent quality conditions (i.e., in air).35 The only situation where star polymers should adsorb preferentially is when segmental adsorption energies are very weak, and the smaller entropic penalty for adsorption allows star polymers to adsorb when linear polymers would not.63 Observations on the Compactness of the Adsorbed PEO Star Layer. While the polymer mass adsorbed per unit area on the silica surface is the same for PEO stars as it was for the diblock copolymers consisting of a cationic PMETAC anchor block and a PEOMA bottlebrush block noted above, these materials produce significantly different layer structures. Consistent with the greater compactness of the PEO star polymers in solution, their layers are significantly denser than the diblock copolymer layers. The former contain ∼66% water and occupy a thickness of ∼8 to 9 nm, while the latter contain 95% water in a layer that is nearly 50 nm thick. These layers can therefore be expected to have significantly different loadbearing and frictional properties that would affect their performance in applications as steric stabilizers or boundary lubrication agents.

fraction of trapped water in the layer as linear PEO chains adsorbed and displaced a fraction of the PEO star polymers. The trapped water content became comparable to that established by linear PEO adsorption to a bare surface (see Table 1). Taken together, the decreasing ellipsometry surface concentration, the trend of increasing trapped water content and the restoration of the dissipation to the level established by linear PEO adsorption to a bare surface indicate that linear PEO was able to displace preadsorbed PEO star polymers at the silica/aqueous interface. Since, as noted above, the ellipsometry and QCM-D experiments were conducted separately, the percentage water content in the layer is an estimate, but the qualitative trends indicate with certainty that linear PEO displacement of PEO stars increases the trapped water content, since the ellipsometry-determined polymer mass decreased, while the QCM-D sensed mass of polymer plus water increased during this process. When a solution of PEO star polymers challenged a preadsorbed linear PEO layer, ellipsometry indicated no change in polymer surface excess concentration, while analysis of the QCM-D data (Figure 3) indicated a 20% increase in total sensed mass, corresponding to a slight increase in trapped water content from ∼87% to 89%. After an initial transient due to the disturbance caused by PEO star sample injection, the dissipation returned to approximately the same value it had before PEO star injection, although it is noted that the spread in the normalized dissipation values for each overtone became significantly smaller after PEO star introduction. The very small change in trapped water content and the preservation of a large dissipation level close to that established by linear PEO adsorption, indicate that there was little PEO star adsorption into the preadsorbed linear PEO layer, even though PEO star polymers produced significantly larger surface excess concentrations than linear PEO when adsorbing to a bare surface. The calculated water contents were used to estimate the average refractive indices of the adsorbed layers as neff 1 ≈ ϕwnw + (1 − ϕw)nPEO, where the refractive indices of pure water and pure PEO are nw = 1.333 and nPEO = 1.454. These refractive index estimates were then used to reinterpret the ellipsometry isotherm data to estimate the effective ellipsometric thickness of the adsorbed layers. These are given in Table 1. Changes in these layer thickness estimates followed the same trends as the estimates based on QCM-D model analysis in the competitive adsorption experiments. As a consistency check, it is noteworthy that both the QCM-D and the ellipsometric thickness estimates correspond to ∼65−75% of the PEO star polymer RG measured in solution. Preferential adsorption of linear PEO relative to PEO star polymers is consistent with prior theoretical modeling. Basing their analysis on the blob model, Joanny and Halperin37 predicted that linear polymers that have the same number of monomers as the total number of monomers in the star polymer, f N, should always adsorb preferentially and should displace preadsorbed star polymers. This is due to greater elastic constraints on the star polymer. For constant f N under good solvent conditions, the fraction of arm segments bound to the surface decreases as the number of arms increases, and linear polymers should always have a higher bound segment fraction than the equivalent star polymer (a schematic illustration is provided in SI). This prediction is supported by molecular simulations.34,63 Relative to linear polymers, star polymer adsorption is hindered by intramolecular excluded volume interactions. Whereas linear

■

CONCLUSIONS

■

ASSOCIATED CONTENT

Nonionic star polymers that consist of a highly cross-linked core surrounded by a corona of stretched PEO chains adsorb to the silica/aqueous interface at significantly higher surface excess concentrations than do linear PEO chains of an equally large molecular weight. The saturation surface concentration for the PEO star polymers is approximately three times greater than it is for the comparable linear PEO. The compact star polymers pack efficiently on the surface and provide a route to establishing very dense layers with a large number of chain ends oriented toward solution. Such layers may be expected to be effective agents for steric stabilization or boundary lubrication. Despite the high packing efficiency, elasticity constraints affect the strength of PEO star polymer adsorption in competitive adsorption situations. PEO star polymers can be displaced from the interface by linear PEO, suggesting that care would need to be taken when formulating a multicomponent stabilizing or lubricating system in order to decrease the likelihood of star polymer displacement from the surface.

S Supporting Information *

In situ AFM image of PEO stars adsorbed at low coverage on silica, irreversible adsorption of PEO stars and linear PEO upon rinsing in polymer-free solution, displacement of PEO stars by linear PEO at a higher linear PEO concentration, elaboration of ellipsometry analysis terms, schematic illustration of star and linear polymer adsorption.This material is available free of charge via the Internet at http://pubs.acs.org. 4005

dx.doi.org/10.1021/la305085a | Langmuir 2013, 29, 3999−4007

Langmuir

■

Article

(19) Dedinaite, A. P., T.; Mohanty, B.; Claesson, P. M. Lubrication by organized soft matter. Soft Matter 2010, 6, 1520−1526. (20) Alexandre, E.; Schmitt, B.; Boudjema, K.; Merrill, E.; Lutz, P. Hydrogel Networks of poly(ethylene oxide) star-molecules supported by expanded polytetrafluoroethylene membranes: Characterization, biocompatibility evaluation and glucose diffusion characteristics. Macromol. Biosci. 2004, 4, 639−648. (21) Olanya, G.; Thormann, E.; Varga, I.; Makuska, R.; Claesson, P. Protein interactions with bottle-brush polymer layers: Effect of side chain and charge density ratio probed by QCM-D and AFM. J. Colloid Interface Sci. 2010, 349, 265−274. (22) Liu, X.; Dedinaite, A.; Rutland, M.; Thormann, E.; Visnevskij, C.; Makuska, R.; Claesson, P. M. Electrostatically anchored branched brush layers. Langmuir 2012, 28, 15537−15547. (23) Dunlop, I. E.; Briscoe, W. H.; Titmuss, S.; Sakellariou, G.; Hadjichristidis, N.; Klein, J. Interactions between polymer brushes: varying the number of end-attaching groups. Chem. Phys. 2004, 205 (18), 2443−2450. (24) Ligoure, C.; Liebler, L. Thermodynamics and kinetics of grafting end-functionalized polymers to an interface. J. Phys. (Paris) 1990, 51 (12), 1313−1328. (25) Pagac, E. S.; Prieve, D. C.; Solomentsev, Y.; Tilton, R. D. Comparison of polystyrene-poly(ethylene oxide) diblock copolymer and poly(ethylene oxide) homopolymer adsorption from aqueous solutions. Langmuir 1997, 13 (11), 2993−3001. (26) Sofia, S.; Premnath, V.; Merrill, E. Poly(ethylene oxide) grafted to silicon surfaces: Grafting density and protein adsorption. Macromolecules 1998, 31, 5059−5070. (27) Holmberg, K.; Tiberg, F.; Malmsten, M.; Brink, C. Grafting with hydrophilic polymer chains to prepare protein-resistant surfaces. . Colloids Surf. A: Physicochem. Eng. Aspects 1997, 123−124, 297−306. (28) Toomey, R.; Mays, J.; Yang, J.; Tirrell, M. Postadsorption rearrangements of block copolymer micelles at the solid/liquid interface. Macromolecules 2006, 39, 2262−2267. (29) Geoffrey, O.; Iruthayaraj, J.; Poptoshev, E.; Makuska, R.; Vareikis, A.; Claesson, P. Adsorption characteristics of bottle-brush polymers on silica: Effect of side chain and charge density. Langmuir 2008, 24, 5341−5349. (30) Linse, P.; Claesson, P. Modelling of bottle-brush polymer adsorption onto mica and silica surfaces: effect of side-chain length. Macromolecules 2009, 43, 2076−2083. (31) Müller, M.; Lee, S.; Spikes, H. A.; Spencer, N. D. The influence of molecular architecture on the macroscopic lubrication properties of the brush-like co-polyelectrolyte poly(L-lysine)-g-poly(ethylene glycol) (PLL-g-PEG) adsorbed on oxide surfaces. Tribol. Lett. 2003, 15, 395− 405. (32) Naderi, A.; Iruthayaraj, J.; Pettersson, T.; Makuska, R.; Claesson, P. Effect of polymer architecture on the adsorption properties of nonionic polymer. Langmuir 2008, 24, 6676−6682. (33) Torbjom, P.; Naderi, A.; Makuska, R.; Claesson, P. Lubrication properties of bottle-brush polyelectrolytes: An AFM study on the effect of side chain and charge density. Langmuir 2008, 24, 3336− 3347. (34) Chremos, A.; Camp, P. J.; Glynos, E.; Koutsos, V. Adsorption of star polymers: Computer simulations. Soft Matter 2010, 6, 1483−1493. (35) Glynos, E.; Chremos, A.; Petekidis, G.; Camp, P. J.; Koutsos, V. Polymer-like to soft colloid-like behavior of regular star polymers adsorbed on surfaces. Macromolecules 2007, 40, 6947−6958. (36) Hiotelis, I.; Koutsioubas, A. G.; Spiliopoulos, N.; Anastassopoulos, D. L.; Vradis, A. A.; Toprakcioglu, C.; Menelle, A.; Sakellariou, G.; Hadjichristidis, N. Neutron reflectivity and computer simulation studies of self-assembled brushes formed by centrally adsorbed star polymers. Macromolecules 2008, 41, 7648−7655. (37) Joanny, J. F.; Halperin, A. Adsorption of star polymers. J. Phys. II 1991, 1, 623−636. (38) Gao, H.; Matyjaszewski, K. Synthesis of functional polymers with controlled architecture by CRP of monomers in the presence of cross-linkers: From stars to gels. Prog. Polym. Sci. 2009, 34, 317−350.

AUTHOR INFORMATION

Corresponding Author

*Tel:1-412-268-1159; e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This material is based on work supported in part by the National Science Foundation under Grants CBET 0729967 and CBET-1133175. Acknowledgment is made to the donors of the American Chemical Society Petroleum Research Fund for partial support of this research.

■

REFERENCES

(1) Blencowe, A.; Tan, J. F.; Goh, T. K.; Qiao, G. G. Core crosslinked star polymers via controlled radical polymerisation. Polymer 2009, 50, 5−32. (2) Gao, H.; Matyjaszewski, K. Low-polydispersity star polymers with core functionality by cross-linking macromonomers using functional ATRP initiators. Macromolecules 2007, 40, 399−401. (3) Gao, H.; Matyjaszewski, K. Synthesis of low polydispersity miktoarm star copolymers via a simple “arm-first” method: Macromonomers as arm precursors. Macromolecules 2008, 41, 4250−4257. (4) Hadjichristidis, N. Synthesis of miktoarm star (μ-star) polymers. J. Polym. Sci. Part A: Polym. Chem. 1999, 37, 857−871. (5) Hadjichristidis, N.; Pitsikalis, M.; Pispas, S.; Iatrou, H. Polymers with complex architecture by living anionic polymerization. Chem. Rev. 2001, 101, 3747−3792. (6) Vos, W.; Keizer, A.; Kleijn, M.; Cohen Stuart, M. The production of PEO polymer brushes via Langmuir-Blodgett and LangmuirSchaeffer methods: Incomplete transfer and its consequences. Langmuir 2009, 25 (8), 4490−4497. (7) Jusufi, A.; Likos, C. N.; Ballauff, M. Counterion distributions and effective interactions of spherical polyelectrolyte brushes. Colloid Polym. Sci. 2004, 282, 910−917. (8) Striolo, A.; Egorov, A. S. Steric stabilization of spherical colloidal particles: Implicit and explicit solvent. J. Chem. Phys. 2007, 126. (9) Napper, D. Polymeric Stabilization of Colloidal Dispersions; Academic Press: London, 1989. (10) Polotsky, A.; Gillich, T.; Borisov, O.; Leermakers, F.; Textor, M.; Birshtein, T. Dendritic versus linear polymer brushes: Selfconsistent field modeling scaling theory, and experiments. Macromolecules 2010, 43, 9555−9566. (11) Dunlop, I. E. B., W. H.; Titmuss, S.; Jacobs, R. M. J.; Osborne, V. L.; Edmondson, S.; Huck, W. T. S.; Klein, J. Direct measurement of normal and shear forces between surface-grown polyelectrolyte layers. J. Phys. Chem. B 2009, 113, 3947−3956. (12) Klein, J. K., E.; Mahalu, D.; Perahla, D.; Fetters, L. J. Reduction of frictional forces between solid surfaces bearing polymer brushes. Nature 1994, 370, 634−636. (13) Kobayashi, M. T., A. Tribological properties of hydrophilic polymer brushes under wet conditions. The Chemical Record 2010 2010, 10, 208−216. (14) Liberelle, B. G., S. Friction and normal interaction forces between irreversibly attached weakly charged polymer brushes. Langmuir 2008, 24, 1550−1559. (15) Nordgren, N.; Rutland, M. W. Tunable nanolubrication between dual-responsive polyionic grafts. Nano Lett. 2009, 9, 2984−2990. (16) Raviv, U. G., S.; Kampf, N.; Gohy, J.-F.; Jerome, R.; Klein, J., Normal and frictional forces between surfaces bearing polyelectrolyte brushes. 24 2008, 8678-8687. (17) Grest, G. S. Interfacial sliding of polymer brushes: A molecular dynamics simulation. Phys. Rev. Lett. 1996, 76, 4979−4982. (18) Lee, S. M., M.; Ratoi-Salagean, M.; Vörös, J.; Pasche, S.; DePaul, S. M.; Spikes, H. A.; Textor, M.; Spencer, N. D. Boundary lubrication of oxide surfaces by poly(L-lysine)-g-poly(ethylene glycol) (PLL-gPEG) in aqueous media. Tribol. Lett. 2003, 15, 231−239. 4006

dx.doi.org/10.1021/la305085a | Langmuir 2013, 29, 3999−4007

Langmuir

Article

(39) Matyjaszewski, K.; Xia, J. Atom transfer radical polymerization. Chem. Rev. (Washington, D. C.) 2001, 101, 2921−2990. (40) Matyjaszewski, K. Atom transfer radical polymerization (ATRP): Current status and future perspectives. Macromolecules 2012, 45, 4015−4039. (41) Saigal, T.; Yoshikawa, A.; Kloss, D.; Kato, M.; Golas, P. L.; Matyjaszewski, K.; Tilton, R. D. Stable emulsions with thermally responsive microstructure and rheology using poly(ethylene oxide) star polymers as emulsifiers. J. Colloid Interface Sci. 2013, 394, 284− 292. (42) Beaglehole, D. Ellipsometric study of simple liquids. Physica B +C 1980, 100, 163−174. (43) Beaglehole, D. Ellipsometry of liquid surfaces. J. Phys. Colloque 1983, C10, 147−154. (44) DeFeijter, J. A.; Benjamins, J.; Veer, F. A. Ellipsometry as a tool to study the adsorption of synthetic and biopolymers at the air−water interface. Biopolymers 1978, 17, 1759−1772. (45) Benjamins, J.; Jönsson, B.; Thuresson, K.; Nylander, T. New experimental setup to use ellipsometry to study liquid−liquid and liquid−solid interfaces. Langmuir 2002, 18, 6437−6444. (46) Voinova, M. V.; Rodahl, M.; Jonson, M.; Kasemo, B. Viscoelastic acoustic response of layered polymer films at fluid-solid interfaces: continuum mechanics approach. Phys. Scr. 1999, 59, 391−396. (47) Eisele, N. B.; Andersson, F. I.; Frey, S.; Richter, R. P. Viscoelasticity of thin biomolecular films: a case study on nucleoporin phenylalanine-glycine repeats grafted to a histidine-tag capturing QCM-D sensor. Biomacromolecules 2012, 13, 2322−2332. (48) Reviakine, I.; Johannsmann, D.; Richter, R. P. Hearing what you cannot see and visualizing what you hear: interpreting quartz crystal microbalance data from solvated interfaces. Anal. Chem. 2011, 83, 8838−8848. (49) Bittrich, E.; Rodenhausen, K. B.; Eichhorn, K.-J.; Hofmann, T.; Schubert, M.; Stamm, M.; Uhlman, P. Protein adsorption on and swelling of polyelectrolyte brushes: A simultaneous ellipsometryquartz crystal microbalance study. Biointerphases 2010, 5, 159−167. (50) Bodvik, R.; Thormann, E.; Karlson, L.; Claesson, P. M. Temperature responsive surface layers of modified celluloses. Phys. Chem. Chem. Phys. 2011, 13, 4260−4268. (51) Domack, A.; Prucker, O.; Rühe, J.; Johannsmann, D. Swelling of a polymer brush probed with a quartz crystal resonator. Phys. Rev. E 1997, 56, 680−689. (52) Ortega, A.; García de la Torre, J. Hydrodynamic properties of rodlike and disklike particles in dilute solution. J. Chem. Phys. 2003, 119, 9914. (53) Vagberg, L. J. M.; Cogan, K. A.; Gast, A. P. Light-scattering study of starlike polymeric micelles. Macromolecules 1991, 24, 1670− 1677. (54) Smith, G. D.; Yoon, D. Y.; Jaffe, R. L.; Colby, R. H.; Krishnamoorti, R.; Fetters, L. J. Conformations and structures of poly(oxyethylene) melts from molecular dynamics simulations and small-angle neutron scattering experiments. Macromolecules 1996, 29, 3462−3469. (55) Daoud, M.; Cotton, J. P. Star-shaped polymers: a model for the conformation and its concentration dependence. J. Phys. (Paris) 1982, 43, 531−538. (56) Kawaguchi, S.; Imait, G.; Suzuki, J.; Miyahara, A.; Kitanoll, T.; Ito, K. Aqueous solution properties of oligo and poly(ethylene oxide) by static light scattering and intrinsic viscosity. Polymer 1997, 38, 2885−2891. (57) Kapilashrami, A.; Malmsten, M.; Eskilsson, K.; Benjamins, J.-W.; Nylander, T. Ellipsometric studies of nonionic block copolymers adsorbed at the solid/water and oil/water interfaces. Colloids Surf. A: Physicochem. Eng. Aspects 2003, 225, 181−192. (58) Flood, C.; Cosgrove, T.; Howell, I.; Revell, P. Effects of electrolytes on adsorbed polymer layers: Poly(ethylene oxide)-silica system. Langmuir 2006, 22 (16), 6923−6930. (59) Trens, P.; Denoyl, R. Conformation of poly(ethylene glycol) polymers at the silica/water interface: A microcalorimetric study. Langmuir 1993, 9, 519−522.

(60) Olanya, G.; Iruthayaraj, J.; Poptoshev, E.; Makuska, R.; Vareikis, A.; Claesson, P. Adsorption characteristics of bottle-brush polymers on silica: Effect of side chain and charge density. Langmuir 2008, 24, 5341−5349. (61) Vigil, R. D.; Ziff, R. M. Random sequential adsorption of unoriented rectangles onto a plane. J. Chem. Phys. 1989, 91, 2599− 2602. (62) Devanand, K.; Selser, J. C. Asymptotic behavior and long-range interactions in aqueous solutions of poly(ethylene oxide). Macromolecules 1991, 24, 5943−5947. (63) Striolo, A.; Prausnitz, J. M. Adsorption of branched homopolymers on a solid surface. J. Chem. Phys. 2001, 114, 8565− 8572.

4007

dx.doi.org/10.1021/la305085a | Langmuir 2013, 29, 3999−4007