Direct Impact of Nonequilibrium Aggregates on the Structure and

Jul 2, 2014 - Department of Physical Chemistry, Lund University, P.O. Box 124, S-221 ... Institute of Chemistry, Eötvös Loránd University, Budapest 11...
0 downloads 0 Views 7MB Size
Download PDF
Article pubs.acs.org/Langmuir

Direct Impact of Nonequilibrium Aggregates on the Structure and Morphology of Pdadmac/SDS Layers at the Air/Water Interface Richard A. Campbell,*,† Marianna Yanez Arteta,†,‡ Anna Angus-Smyth,†,§ Tommy Nylander,‡ Boris A. Noskov,∥ and Imre Varga*,⊥ †

Institut Laue-Langevin, 6 rue Jules Horowitz, BP 156, 38042 Grenoble Cedex 9, France Department of Physical Chemistry, Lund University, P.O. Box 124, S-221 00 Lund, Sweden § Department of Chemistry, Durham University, South Road, DH1 3LE Durham, United Kingdom ∥ Chemical Faculty, St. Petersburg State University, Universitetsky pr. 2, 198904 St. Petersburg, Russia ⊥ Institute of Chemistry, Eötvös Loránd University, Budapest 112, P.O. Box 32, H-1518 Hungary ‡

ABSTRACT: We discuss different nonequilibrium mechanisms by which bulk aggregates directly modify, and can even control, the interfacial structure and morphology of an oppositely charged polyelectrolyte/surfactant (P/S) mixture. Samples are categorized at the air/water interface with respect to the dynamic changes in the bulk phase behavior, the bulk composition, and the sample history using complementary surface-sensitive techniques. First, we show that bulk aggregates can spontaneously interact with the adsorption layer and are retained in it and that this process occurs most readily for positively charged aggregates with an expanded structure. In this case, key nonequilibrium issues of aggregate dissociation and spreading of surface-active material at the interface have a marked influence on the macroscopic interfacial properties. In a second distinct mechanism, aggregates inherently become trapped at the interface during its creation and lateral flocculation occurs. This irreversible process is most pronounced for aggregates with the lowest charge. A third mechanism involves the deposition of aggregates at interfaces due to their transport under gravity. The specificity of this process at an interface depends on its location and is mediated by density effects in the bulk. The prevalence of each mechanism critically depends on a number of different factors, which are outlined systematically here for the first time. This study highlights the sheer complexity by which aggregates can directly impact the interfacial properties of a P/S mixture. Our findings offer scope for understanding seemingly mysterious irreproducible effects which can compromise the performance of formulations in wide-ranging applications from foams to emulsions and lubricants.

1. INTRODUCTION The interfacial properties of complex mixtures determine the performance of countless cleaning, cosmetic, and medical products as well as the stability of common food products.1,2 Polyelectrolyte/surfactant (P/S) mixtures are particularly important from both fundamental and applied perspectives as a result of their widespread use in formulations.3−6 A number of reviews have described the properties of synthetic P/S mixtures both in the bulk7,8 and at surfaces.9−13 Furthermore, the associative interactions of mixtures of amphiphiles and oppositely charged biological macromolecules such as proteins or DNA play an important role in biological processes14,15 and in biomedical applications such as drug16 or gene delivery.17 It is important therefore to understand the interactions between such species both in the bulk and at surfaces in order to predict their behavior and optimize performance. In recent years, an understanding has been developed of the strong bulk association between polyelectrolytes and surfactants in oppositely charged mixtures and the resultant rich phase behavior.18−22 Aggregation occurs in the bulk at compositions © 2014 American Chemical Society

close to charge neutralization of the P/S complexes due to their lack of colloidal stability, and samples become turbid as a result of the formation of large aggregates.23,24 Such samples have been shown to exhibit marked nonequilibrium effects where the size distribution of species generated in the bulk is determined by methodological considerations such as the order of addition of component polyelectrolyte and surfactant solutions during mixing.25−27 These nonequilibrium effects can determine the nature of aggregates produced, hence there is scope for tuning the properties of the samples simply by changing the way they are handled. The dynamic properties of P/S mixtures at the air/water interface have received some attention over the years.13,28,29 Most surface studies, however, have concerned properties of the static air/water interface, and while nonequilibrium aspects have been regularly commented upon, data are still typically Received: February 18, 2014 Revised: May 28, 2014 Published: July 2, 2014 8664

dx.doi.org/10.1021/la500621t | Langmuir 2014, 30, 8664−8674

Langmuir

Article

modeled using an equilibrium framework.30,31 Recently the effects of nonequilibrium interactions in the bulk on the interfacial properties of such systems have started to attract more attention. An indirect effect of bulk aggregation is the depletion of P/S adsorption layers at the air/water interface under both static32,33 and dynamic29 conditions, as recent studies have related a peak in surface tension isotherms to slow, comprehensive precipitation and sedimentation. Although much less well understood, aggregates from the bulk can also modify the interfacial behavior directly as a result of their penetration into adsorption layers.34 Pertinent observations have been made in a number of recent studies involving poly(diallyldimethylammonium chloride)/sodium dodecyl sulfate (Pdadmac/SDS) and poly(ethylene imine)/ SDS (PEI/SDS) mixtures. First, it was shown that the formation of microaggregates in Pdadmac/SDS layers at the air/water interface modifies the rheological properties in a nonlinear way.28 Second, PEI/SDS aggregates were demonstrated to penetrate the adsorption layer at the air/water interface spontaneously at high pH yet become irreversibly trapped at low pH.35 Third, it was shown that the application of a mild mechanical stress to a settled precipitate of Pdadmac/ SDS samples can change the interfacial properties.36 Fourth, nanostructured aggregates of Pdadmac/SDS were shown to be deposited at interfaces due to differences in their density with the bulk liquid.37 Fifth, the dissociation and spreading of material by PEI/SDS aggregates at the air/water interface was shown to modify the macroscopic surface properties.38 Sixth, it was shown at an expanding fluid interface that a convection/ spreading mechanism from PEI/SDS aggregates can even exceed the surface excess resulting from diffusion/adsorption at low pH.39 Together these studies show that P/S aggregates interact with interfaces via complex mechanisms that involve the trapping or spreading of particles delivered by adsorption, convection, or deposition under gravity. In spite of the recent observations concerning direct effects of bulk aggregates impacting the interfacial properties of oppositely charged P/S mixtures, to our knowledge no study has systematically characterized the types of interaction mechanism and their provenance. The scope of this article is to determine, through the use of complementary methodologies, the processes and conditions by which bulk Pdadmac/ SDS aggregates directly impact the static air/water interface. Since nonequilibrium effects in the two-phase region of the P/S aqueous system can be considerable, we have investigated three types of samples with the same composition but different preparation protocols: fresh−mixed, where in the phaseseparation region there is a dispersion of P/S aggregates, aged−settled where after 3 days of equilibration the aggregates have precipitated, leaving a clear supernatant, and aged− redispersed, where a mild mechanical stress is applied to the sediment of aged samples to liberate some particles back into the bulk liquid. The samples were subjected to precleaning by aspiration of the air/water interface prior to all measurements except in the last section where the issue of surface-trapped material is examined specifically. We have used surface tensiometry, ellipsometry, Brewster angle microscopy (BAM), and neutron reflectometry (NR) to examine the structure and morphology of the air/water interface. The interfacial composition has already been rigorously characterized for the three defined sample states of this system using NR,32 which is a technique of great strength but also limitations, and hence is strongly complemented by the

additional methodologies employed in the present work. Ellipsometry provides valuable additional information including (1) adsorption kinetics on shorter time scales, (2) higher sensitivity to the penetration of macroscopic aggregates of low surface coverage into the adsorption layer, and (3) probing of a much smaller area of the interface, meaning that temporal fluctuations in the optical signal can be interpreted in terms of aggregates embedded in the interfacial layer. In addition, the BAM imaging technique has allowed us to monitor effects of micrometer-sized aggregates at the interface for the different sample states, and NR has been used to identify the presence of nanostructured aggregates trapped at the air/water interface.

2. EXPERIMENTAL SECTION 2.1. Materials. Deionized water was passed through a purification system (Milli-Q, total organic content = 4 ppb, resistivity = 18 mΩ cm). Pdadmac (100k−200k g/mol, Aldrich) was purified by dialysis (regenerated cellulose, 12k−14k molecular weight cutoff, Medicell International Ltd.) in pure water to give a stock solution of ∼20 000 ppm free of low-molecular-weight impurities. The concentration of the polymer stock solution was determined by gravimetric analysis. Hydrogenous SDS (hSDS, Sigma, 99.9%) was recrystallized twice in ethanol, and each time the solutions were cooled over several hours. The high purity of the surfactant was verified with surface tension measurements. Deuterated SDS (dSDS, Cambridge Isotopes, 99%) was used as received after verifying the purity using surface tension and NR measurements. NaCl (Merck, 99.99%) was used as received. 2.2. Sample Preparation. The solutions comprised 100 ppm Pdadmac at various SDS concentrations in 0.1 M NaCl at 25 °C. A standard mixing approach was used,35 which minimized the formation of kinetically trapped aggregates as a result of local concentration gradients present during mixing.25 The Pdadmac stock solution obtained after the dialysis process was diluted to 200 ppm in 0.1 M NaCl, and a stock solution of SDS (∼20 mM) in 0.1 M NaCl was made fresh daily. For mixing, 25 mL of each of the Pdadmac/NaCl and SDS/NaCl solutions was poured together simultaneously. The mixed solution was then swirled gently for a few seconds, of which 35 mL was transferred by a glass pipet into the measurement vessel (fresh−mixed samples) or the entire solution was poured into a 100 mL volumetric flask for storage (aged samples). A 50 mL glass pipet was also used to transfer the supernatant of aged−settled samples to the measurement vessel to minimize the transfer of surface-trapped precipitate.36 Aged−redispersed samples were inverted and gently shaken once before the transfer of liquid by pipet to the measurement vessel. For all of the data presented in sections 3.1 to 3.3 of Results and Discussion, the surface of each solution was aspirated in the center of the dish continuously for 5 s using a clean pipet attached to a water suction pump immediately prior to measurement. This process removed surface-trapped particles and allowed measurements of material adsorbed to a fresh surface; checks have been carried out to confirm that this procedure does not result in a measurable depletion of surface-active material from the bulk of surfactant solutions. For the data in section 3.4, the ellipsometry measurements experienced only a light surface cleaning to remove obvious flocks or bubbles where the pipet tip was dabbed in various places across the surface and the BAM and NR measurements were not subjected to surface cleaning. 2.3. UV−Vis Spectroscopy. Pdadmac/SDS solutions in 0.1 M NaCl were measured using a UV−vis spectrometer (Jasco V-630 spectrophotometer) using a quartz Helma cell with a 1 mm path length. The turbidity was then evaluated according to the optical density at a wavelength of 450 nm. A spectrum of each solution was measured within 30 min of its preparation. 2.4. Surface Tensiometry. The surface tension measurements were recorded on an automated Krüss K10 balance. A sand-blasted platinum Wilhelmy plate (20 × 10 × 0.10 mm3) was immersed in liquid contained in a Petri dish, and then the plate was withdrawn until the bottom was parallel to the surface. The overall force F that acts on 8665

dx.doi.org/10.1021/la500621t | Langmuir 2014, 30, 8664−8674

Langmuir

Article

the plate was used to calculate the surface tension γ = F/2ppcos θ, where pp is the perimeter of the plate and θ is the contact angle between the liquid and plate. The full wetting criterion cos θ = 1 was assumed. The plate was cleaned with water and ethanol and was then heated with a Bunsen burner immediately before each measurement. The plate was hung on the surface pressure balance and then immersed in pure water to achieve a surface tension of γ = 71.8 ± 0.5 mN m−1 to ensure the cleanliness of the dish and the Wilhelmy plate. The plate was then used to measure the surface tension of Pdadmac/ SDS solutions with the same cleaning procedure as described repeated between each measurement. 2.5. Ellipsometry. Ellipsometry measurements are based on the change in polarization of light reflected at an interface. Upon reflection, the relative amplitude and the phase of the p and s components change by different amounts. The relative attenuation, Ψ, and the relative phase shift, Δ, depend on the optical properties of the surface and on the angle of incidence θ. The ellipsometric angles are related to the Fresnel reflectivity coefficients of the parallel and perpendicular components, rp and rs, respectively,40 rp rs

= tan Ψ exp iΔ

2.6. Brewster Angle Microscopy. A Nanofilm EP3 Brewster angle microscope was used with a 10× or 20× objective and automatic focusing. The 20× objective was used for some of the aged− redispersed samples to highlight the features in the lateral structure. The images were recorded at the Brewster angle for the air/water interface, and the background was subtracted from each image in the form of a corresponding image recorded with different polarizer and analyzer settings. In each case the scale bar represents 100 μm. 2.7. Neutron Reflectometry. NR measurements were performed on the FIGARO neutron reflectometer at the Institut Laue-Langevin (Grenoble, France).45 Neutron pulses with 4.0% dλ/λ were used in the wavelength range of λ = 2−30 Å. Data acquisitions were carried out at a fixed incident angle of θ = 3.8° for an acquisition time of 15 min. Samples were left to reach steady state for 24 h. The isotopic contrast dSDS in air-contrast-matched water (8.9% D2O in H2O by weight) was used.

3. RESULTS AND DISCUSSION We start this section with a reminder of the important dynamic features of the bulk solution behavior of the Pdadmac/SDS system. Figure 1 shows the optical density of aggregates for

(1)

A limitation of ellipsometry at the air/liquid interface is that Ψ is insensitive to the optical properties of a thin, transparent film. Therefore, in this work we present measurements of Δ. We define here the effect of the interfacial layer on the measured phase shift as Δsurf = ΔP/S − Δ0, where ΔP/S is the measured parameter for a P/S solution and Δ0 is the reference value for pure water. In the thin film limit, Δsurf is linearly proportional to the ellipsometric thickness η,41 Δsurf =

g(ϕ) η λ

(2)

where λ is the wavelength and g(ϕ) is the function41

g(ϕ) =

4π εr,amb εr,sub cos ϕ sin 2 ϕ [εr,sub − εr,amb][(εr,sub + εr,amb)cos2 ϕ − εr,amb]

(3)

and εr,amb is the relative permittivity of the ambient medium (air), εr,sub is the relative permittivity of the substrate (liquid), and εr = n2 (the square of the refractive index). For an optically isotropic interface, Drude showed that η can be written in terms of the relative permittivity profile across the interface εr(z),42

η=

∫

[εr(z) − εr,amb][εr(z) − εr,sub] εr(z)

dz

Figure 1. Optical density at 450 nm of fresh−mixed and aged−settled Pdadmac/SDS samples in 0.1 M NaCl (UV−vis spectroscopy); the point of charge neutrality of the bulk complexes is marked with a vertical dashed line (electrophoretic mobility); data are reproduced from ref 32. Seven bulk compositions measured using BAM are marked A−G, as described in Table 1; compositions B−F are also measured using NR.

(4)

where z is the Cartesian coordinate perpendicular to the interface. The effect on η from the change in scattering by thermal capillary waves between samples is neglected. For a uniform isotropic layer separated by sharp stratified interfaces, eqs 2 and 4 may be reduced to

Δsurf = ksurf d =

ksurf Γ ρ

fresh−mixed and aged−settled samples (UV−vis spectroscopy), and the point of charge neutrality is marked as a vertical dashed line (electrophoretic mobility); we note that this composition of stoichiometric binding (0.82 mM) is not equivalent to the composition of stoichiometric mixing or the charge equivalence point (0.62 mM).36 The fresh−mixed samples are turbid across the phase-separation region, at compositions close to charge neutralization of the bulk aggregates, as indicated by the high optical density. In contrast, the complexes have sufficient charge to avoid aggregation at bulk Pdadmac/SDS ratios far from stoichiometric binding. After 3 days, the optical density is much reduced for the supernatant of aged−settled samples in the phase-separation region due to precipitation and sedimentation. However, the optical density remains high at compositions close to the phase boundaries and even increases with time due to ongoing aggregation. These particles are formed as a result of local concentration gradients during the mixing process, and they

(5)

where d is the thickness, ρ is the density, Γ is the surface excess expressed in mass per unit area, and ksurf is a function of εr,amb, εr,sub, g(ϕ), and λ.35 Hence for the case of a uniform isotropic layer of constant density and zero roughness, Δsurf is linearly proportional to Γ. Such a model can be a reasonable approximation for an adsorption layer of nonionic surfactant molecules,43 but more sophisticated models are required to treat more complex interfaces.44 In this study we demonstrate the presence of surface aggregates under certain conditions; therefore, the application of a uniform layer model over the entire sample surface is not valid. We therefore consider the magnitude of Δsurf as an average measure of the total interfacial excess over the ∼1 mm2 area illuminated by the laser beam. A Beaglehole picometer light ellipsometer with a HeNe laser (λ = 632.8 nm) was employed at an angle of incidence of θ = 50°. The data acquisition rate was ∼0.2 Hz. 8666

dx.doi.org/10.1021/la500621t | Langmuir 2014, 30, 8664−8674

Langmuir

Article

remain dispersed in solution on experimentally accessible time scales. This study reports surface tension and ellipsometry measurements recorded on samples over the whole range of bulk compositions displayed in Figure 1. The BAM and NR measurements, however, are restricted to a selection of samples which exhibit distinct bulk behavior. They are marked with arrows in Figure 1, and their physical characteristics are described in Table 1. Work carried out previously on highTable 1. Seven Pdadmac/SDS Compositions Measured with BAM in Which the Middle Five Are Also Measured with NR label

SDS concentration (mM)

A

0.10

B C D E F G

0.54 0.62 0.82 1.3 2.0 4.0

observations far below stoichiometric binding; optically transparent kinetically trapped aggregates; positive charge phase-separation region; positive charge phase-separation region; neutral phase-separation region; negative charge kinetically trapped aggregates; negative charge far above stoichiometric binding; compacted particles

molecular-weight PEI/SDS mixtures at low pH showed that nonequilibrium aggregates formed during mixing became trapped at the interface but were irreversibly removed by surface aspiration.35 As a result, the data shown in sections 3.1−3.3 all involve samples where the surface was precleaned by aspiration, and the implications of this process are discussed in section 3.4. 3.1. Fresh−Mixed Samples with Surface Cleaning. The low-surface-tension data for fresh−mixed samples in Figure 2A show that over the whole range of bulk P/S compositions there is enough surface-active material at the air/water interface to lower its free energy on even short time scales. The corresponding ellipsometry data in Figure 2B reveal that the total interfacial excess is highest at low bulk SDS concentrations and deceases as the bulk surfactant concentration is increased. This trend is consistent with a gradual shift in interfacial composition, with decreasing amounts of polyelectrolyte at the interface at higher bulk SDS concentrations, as has been shown previously for samples in the same state using NR.32 The ellipsometry data for samples at the edges of the phaseseparation region show distinctive time-dependent features, which are emphasized in Figure 3. Close to the edge of the phase-separation region with the lowest bulk SDS concentration edge (0.46 mM, left side) there is a time-dependent depletion of the material at the interface on short time scales. After a reproducible small maximum in Δsurf after 2 min, the total interfacial excess decreases by over 30% after 30 min. While this depletion effect may be related to the ongoing coalescence of kinetically trapped aggregates in the bulk, an alternative explanation lies in the dynamic spreading of material delivered to the interface by a small number of kinetically trapped aggregates in a process which cannot be sustained over the whole measurement period due to their low abundance. Depletion may not be evident for samples in the phaseseparation region if a large abundance of such aggregates at the interface result in a sustained spreading of material to maintain a low surface tension for a longer period. Such a process is in line with the dynamic release of surface-active molecules from the interfacial dissociation of PEI/SDS aggregates investigated

Figure 2. (A) Surface tension and (B) ellipsometry data of fresh− mixed Pdadmac/SDS samples in 0.1 M NaCl with respect to the bulk composition after surface cleaning by aspiration; surface tension data with ages of 1 and 30 min are reproduced from ref 36; ellipsometry data of pure SDS solutions in 0.1 M NaCl are shown for reference as black squares; a dashed line marks the composition of charge neutrality.

Figure 3. Ellipsometry data of fresh−mixed Pdadmac/SDS samples in 0.1 M NaCl as a function of time. Surface cleaning defined the surface age of zero.

recently by Gao et al.38 and Angus-Smyth et al.39 On the edge of the phase-separation region with the highest bulk SDS concentration (2.0 mM, right side) there is an enhancement of the total interfacial excess from that corresponding to an SDS 8667

dx.doi.org/10.1021/la500621t | Langmuir 2014, 30, 8664−8674

Langmuir

Article

Figure 4. BAM images of fresh−mixed, aged−settled, and aged−redispersed Pdadmac/SDS samples in 0.1 M NaCl with respect to the bulk composition where the surfaces were cleaned by aspiration immediately prior to the measurements. The surface age is 5 min. The scale bars all represent 100 μm.

monolayer to a much larger value over a 30 min period. As the surface tension is not changing significantly during this period, the effects are attributed to the penetration of kinetically trapped aggregates into the interfacial layer. Corresponding BAM images of interfacial films formed from fresh−mixed samples after cleaning the surface by aspiration are shown in the top row of Figure 4. The samples in the phaseseparation region exhibit aggregates with a size of several micrometers present at the interface. The absence of observed surface structure for sample F at the edge of the high-SDSconcentration side of the phase-separation region indicates that the kinetically trapped aggregates which ellipsometry has shown penetrate the interfacial layer (Figures 2 and 3) must be smaller in diameter than the detection limit of several micrometers. 3.2. Aged−Settled Samples with Surface Cleaning. The surface tension of aged−settled Pdadmac/SDS samples is shown in Figure 5A where there is now a peak in the values due to comprehensive precipitation.32 A comparison of these data with the ellipsometry data in Figure 5B is revealing. The total interfacial excess in the phase-separation region is initially depleted, as expected for samples with high surface tension, yet for all of the samples measured which have residual positively charged aggregates in the supernatant (0.56−0.82 mM) there is a marked increase in Δsurf over the first 30 min, during which time the surface tension remains virtually constant. These trends are most pronounced close to the phase boundary. The effects are emphasized in Figure 6, where the kinetic data for the sample at the edge of the phase-separation region with the lowest bulk SDS concentration (0.59 mM, left side) are shown. As the total interfacial excess has increased steeply (by almost a factor of 3) while the surface tension has stayed virtually constant, the change in the interfacial properties is attributed to the penetration of aggregates into the adsorption layer. Previous NR measurements of aged−settled samples showed that the surfactant and polyelectrolyte surface excesses at the edge of the phase-separation region with the lowest bulk SDS concentration were depleted.32 It is a surprise, therefore, that the ellipsometry data show that at a surface age of 30 min the total interfacial excesses of the sample just outside the phaseseparation region (cSDS = 0.54 mM, before the surface tension peak, green arrows) and that just inside (cSDS = 0.59 mM, at the top of the peak, blue arrows) are almost identical. However, what seems like an anomaly when one considers the interfacial behavior from an equilibrium adsorption perspective can again be rationalized in terms of aggregate penetration. This process would have a large effect on the total interfacial excess

Figure 5. (A) Surface tension and (B) ellipsometry data of aged− settled Pdadmac/SDS samples in 0.1 M NaCl with respect to the bulk composition after surface cleaning by aspiration; surface tension data with ages of 1 and 30 min are reproduced from ref 36; ellipsometry data of pure SDS solutions in 0.1 M NaCl are shown as black squares; a dashed line marks the composition of charge neutrality. The green and blue arrows are reference points for discussion.

measured by ellipsometry but would not greatly affect the surface excesses from NR nor the surface tension as both techniques are insensitive to a low coverage of large particles embedded in a thin adsorption layer. The reasons behind the selective penetration and retention of positively charged aggregates at the air/water interface merit discussion. From dynamic light scattering measurements, Á brahám et al. described a distinction in the structure between positively and negatively charged Pdadmac/SDS aggregates.46 8668

dx.doi.org/10.1021/la500621t | Langmuir 2014, 30, 8664−8674

Langmuir

Article

surfactant solutions, an apparent inconsistency in our approach was the detection of residual polyelectrolyte bound to the headgroups of the surface monolayer of surfactant.32 The interfacial composition was derived by the fitting of a homogeneous layer model to neutron reflectivity data recorded in different isotopic contrasts. In this case the lowering of the reflectivity at high momentum-transfer values of data recorded from samples in D2O principally determined the polyelectrolyte surface excess. In light of the present results, we suggest that the residual penetration of aggregates into the adsorption layer of aged−settled samples may explain the anomaly as an increased effective surface roughness resulting from heterogeneities in the layer would have a largely similar effect on the data. The homogeneous layer model applied to the reflectivity data could not account for residual aggregates embedded in the interfacial layer, and as a result the surface excess of polyelectrolyte bound to the surfactant monolayer would be overestimated. Despite our concluding that bulk aggregates are present at the interface for aged−settled samples, a comparison of the BAM images in the middle row of Figure 4 shows the absence of micrometer-sized aggregates embedded in the interfacial layer, as there were for the fresh−mixed samples. We may infer that the residual positively charged aggregates that penetrate the interfacial layer of aged−settled must be smaller than the minimum length scale resolvable by the BAM measurements, i.e., below the micrometer size range. 3.3. Aged−Redispersed Samples with Surface Cleaning. For aged−redispersed samples, where a light mechanical stress had been applied to the settled precipitate of aged− settled samples, the surface tension in the phase-separation region falls to low values within the first half hour, as can be seen in Figure 7. On shorter time scales, the surface tension peak has changed position and shape: a transient maximum now occurs at the surfactant composition that corresponds to charge neutrality of the bulk complexes (cSDS = 0.82 mM). As with the aged−settled samples, these data are consistent with the interaction of aggregates originating from the bulk with the interfacial layer, but in this case, they are sufficiently numerous to spread material which modifies the free energy of the surface and hence lowers the surface tension. The faster changes in both the surface tension and total interfacial excess for samples with lower bulk SDS concentrations than the point of charge neutrality are consistent with the picture that that expanded, positively charged aggregates interact more readily than compacted, negatively charged aggregates with the interfacial layer. The BAM images in the bottom row of Figure 4 are qualitatively consistent with the kinetic dependencies of the surface tension and total interfacial excess as positively charged aggregates are embedded in the adsorption layer at the air/ water interface (samples B and C). These aggregates have a length scale on the order of tens to hundreds of micrometers, and hence they correspond to macroscopic particles of precipitate. It is these particles to which we attribute the spreading of material at the interface which lowers the surface tension. In contrast, the incorporation of negatively charged aggregates into the interfacial layer is either much slower or involves submicrometer particles below the detection limit of the technique. We note that in a recent paper by Bahramian et al. such nonequilibrium effects were dismissed as artifacts resulting from low-molecular-weight fractions in our samples.31 Here, we

Figure 6. Surface tension and ellipsometry data of aged−settled Pdadmac/SDS samples in 0.1 M NaCl as a function of time. Surface cleaning defined the surface age of zero.

The positively charged aggregates conformed to the accepted structure of surfactant aggregates wrapped by polymer chains, yet the negatively charged aggregates corresponded to compacted P/S particles coated by adsorbed surfactant molecules. They reasoned that the negatively charged particles represented a kinetically stable colloid dispersion rather than a thermodynamically stable system. This physical picture was later extended to the poly(sodium styrenesulfonate)/hexadecyltrimethylammonium bromide system but only at low ionic strength.47 This distinction seemingly had no marked influence on the interfacial properties measured previously using NR and surface tension, but the ellipsometry data in the present work reveal a clear change in the interaction of the bulk aggregates according to their charge (and corresponding change in structure) with the adsorption layer at the air/water interface. For aged−settled samples, the interfacial structure is principally a depleted surfactant monoloyer,32 hence we may reason that aggregates with exposed excess cationic polyelectrolyte segments may interact favorably with the anionic headgroups of the monolayer. Nevertheless, the corresponding change in structure of the aggregates may also play a role in the extent of the interaction. Furthermore, from the fact that the surface tension remains high for the samples with negatively charged aggregates, we may infer that the particles are either not numerous enough at the interface or are present in a kinetically arrested state to spread sufficient material to reduce its free energy. The observation that the charge and structure of P/S aggregates can have a significant influence on the properties at the air/water interface is in line with that in a recent paper on poly(amido amine) dendrimer/SDS mixtures.48 In this case, a peak in the surface tension was found for freshly mixed samples which contained negatively charged aggregates, and the feature was attributed to their less-favorable interactions with the adsorption layer. This comparison further emphasizes the possible importance of aggregate interactions in determining the interfacial properties of oppositely charged P/S mixtures. The new ellipsometry data presented here may help us to resolve an old mystery. While we have demonstrated previously that the surface tension of aged−settled samples in the phaseseparation region approximates well to that of depleted 8669

dx.doi.org/10.1021/la500621t | Langmuir 2014, 30, 8664−8674

Langmuir

Article

Figure 7. (A) Surface tension and (B) ellipsometry data of aged− redispersed Pdadmac/SDS samples in 0.1 M NaCl with respect to the bulk composition after surface cleaning by aspiration; surface tension data with ages of 1 and 30 min are reproduced from ref 36; ellipsometry data of pure SDS solutions in 0.1 M NaCl are shown for reference as black squares; a dashed line marks the composition of charge neutrality.

Figure 8. (A) Ellipsometry data of aged−settled Pdadmac/SDS samples in 0.1 M NaCl with respect to the bulk composition: after surface cleaning by aspiration (blue circles) and with only light surface cleaning (red squares). Data are averaged over a surface age of 15−30 min; error bars are equal to two standard deviations; the four colored arrows indicate the data shown in panel B; a dashed line marks the composition of charge neutrality. (B) Time-resolved data for bulk SDS concentrations of 0.49 mM (orange), 0.73 mM (cyan), 1.3 mM (green), and 2.0 mM (violet).

point out that in our previous studies29,32,36,37 and in our present work we have used dialyzed Pdadmac samples precisely to eliminate the influence of low-molecular-weight species on our data. As a consequence our results cannot be attributed to limitations in the definition of the samples. Instead, our data demonstrate that changes in the total interfacial excess and surface tension can be tuned by the liberation of macroscopic particles from the sediment, which in turn penetrate, dissociate, and spread material at the air/water interface in a dynamic nonequilibrium process. These findings emphasize the very different interfacial properties and morphologies that can be created depending on the sample handling method used, e.g., aged−settled vs aged−redispersed samples or indeed any state in between. The extreme nonequilibrium nature of the Pdadmac/SDS system is demonstrated by the fact that the small differences in the way the sample is handled result in large differences in interfacial properties. 3.4. Effects of Different Sample Histories: SurfaceTrapped Aggregates. The data presented in sections 3.1−3.3 were recorded on samples where the air/water interface was precleaned by aspiration. Here we discuss the mechanism by which nonequilibrium aggregates become trapped at the surface of untreated samples during the creation of the interface. First, ellipsometry measurements were carried out on aged−settled samples with only a light surface cleaning to remove obvious flocks or bubbles rather than efficiently extract the initial surface

film. Figure 8A shows the average values and standard deviations of the total interfacial excess, and Figure 8B shows examples of the time-resolved data. The lack of surface cleaning has the most pronounced effects on aged−settled samples closest to the edges of the phaseseparation region (orange and violet). In these cases, fluctuations in the data indicate a higher average total interfacial excess than for samples where the interface had been cleaned by aspiration. The irregular signal is explained by the mobility of aggregates embedded in the interfacial layer as they pass in and out of the area probed by the laser (∼1 mm2). We recall that the turbid samples at the edges of the phase separation have kinetically trapped aggregates formed during mixing, and while aggregation is ongoing, the particles remain in solution as they have sufficient surface charge (cf. Figure 1). We may infer that these particles become trapped in the adsorption layer during its creation, and the lack of fluctuations in the data 8670

dx.doi.org/10.1021/la500621t | Langmuir 2014, 30, 8664−8674

Langmuir

Article

Figure 9. BAM images of fresh−mixed, aged−settled, and aged−redispersed Pdadmac/SDS samples in 0.1 M NaCl with respect to the bulk composition where the surfaces were not cleaned by aspiration. The scale bars all represent 100 μm.

when in contact with the adsorption layer. Finally, it is noteworthy that there are some macroscopic aggregates trapped at the surface even for the sample well outside the phase-separation region (sample A), which is transparent to the naked eye. This observation is consistent with previous ellipsometry data on PEI/SDS mixtures at low pH where fluctuations in the data attributed to surface-trapped aggregates were observed for samples that are virtually transparent up to three orders of magnitude in composition outside the phaseseparation region.35 For the aged−settled samples there is a much reduced effect from surface-trapped aggregates in the phase-separation region as extensive precipitation and sedimentation of aggregates have already taken place. A supernatant results which contains only residual aggregates but which still become trapped in the interfacial layer during its creation. The number and size of surface-trapped aggregates is again highest for the sample at charge neutralization (sample D), which can again be explained in terms of their low charge. As the ellipsometry data of aged− settled samples showed the most pronounced effects from surface-trapped aggregates in samples at the edge of the phaseseparation region, we may reason that these effects must originate from aggregates smaller in size than a few micrometers, i.e., below the detection limit of BAM. The images from the aged−redispersed samples share several of the characteristics with those of the fresh−mixed samples. For example, surface-trapped aggregates are most extensive when the bulk aggregates have minimal surface charge (sample D) and are more extensive for positively charged aggregates (sample C) than for negatively charged aggregates (sample E). However, a clear difference is the turbid sample on the edge of the phase-separation region with the highest bulk SDS concentration (sample F) where large Newtonian ring patterns are observed. These interference fringes are typical of very large nearly spherical objects in contact with the planar interface, and their presence may be attributed here to the kinetically trapped particles or bubbles stabilized by the particles. As the only difference between the aged−settled and aged−redispersed samples is the way in which well-equilibrated samples were handled, the nonequilibrium nature of the interfacial morphology is clearly displayed. Finally, we consider the way in which nanostructured bulk aggregates can interact with adsorption layers at the air/water interface. It has been shown that Pdadmac/SDS aggregates formed as a result of bulk phase separation have a liquidcrystalline nanostructure of their hexagonal phase.49,50 As offspecular scattering in NR measurements is sensitive to an

measured from samples with surface cleaning shows that they had not adsorbed spontaneously. For the supernatant of aged−settled samples inside the phase-separation region, there are modest fluctuations in the data for the sample with positively charged aggregates (cyan) but no observable effect for the samples with negatively charged aggregates (green). The low amplitude of the fluctuations in the data in the former case may be explained by the fact that only residual aggregates in the depleted supernatant were transferred to the measurement dish. We may infer from the fact that the effect is observed only for the sample with positively charged residual aggregates that again these particles interact preferentially with the adsorption layer due to their different charge and/or structure. Also, the samples with compositions well outside the phase-separation region to the low-SDS-concentration side were unaffected by surface aspiration, which is linked to a greater homogeneity of the bulk. The finding that nonequilibrium aggregates become embedded in interfacial layers of Pdadmac/SDS mixtures for samples which are not aspirated prompted us to undertake a systematic investigation into the effects of surface cleaning on the interfacial morphology. Figure 9 shows BAM images of fresh−mixed, aged−settled, and aged−redispersed Pdadmac/ SDS samples which were not subjected to surface cleaning with respect to the bulk composition. These images may be compared to those taken from samples which had been aspirated (cf. Figure 4). The differences between Figures 4 and 9 are striking, as the features present only for the samples that had not been aspirated must all originate from the trapping of aggregates during the creation of the interface, which once removed do not adsorb to it spontaneously. For fresh−mixed samples not subjected to surface cleaning, the effects of surface-trapped aggregates are most pronounced when the bulk aggregates are neutral (sample D). In this case the surface structures are present on a length scale of hundreds of micrometers, the extent of which can be rationalized in terms of the minimal charge of hydrophobic aggregates formed at this bulk composition which flocculate laterally during the creation of the interface. This situation is in marked contrast to the specific interactions described in sections 3.1−3.3 where the extent of the spontaneous interaction is specifically related to differences in the charge and/or structure of the aggregates. Nevertheless, a comparison of the images from samples which contain positive (sample C) and negative (sample E) aggregates shows more extensive surface structure in the former case. This observation highlights the propensity of expanded, positively charged aggregates to flocculate laterally 8671

dx.doi.org/10.1021/la500621t | Langmuir 2014, 30, 8664−8674

Langmuir

Article

ordered morphology of repeating units of polyelectrolyte/ solvent and surfactant in the plane of the interface, we can use this technique to determine mechanisms by which nanostructured aggregates interact with the adsorption layer by performing measurements with respect to the sample history. We demonstrated previously using NR that these particles are transported to the air/liquid and solid/liquid interfaces under three strict conditions: (1) only for samples in the phaseseparation region, (2) only for fresh−mixed samples where there is a dispersion of P/S aggregates still present, and (3) only when the mean directionality of the aggregates, determined by their specific gravity, is toward the probed interface.37 As such, the particles were not detected at the air/ water interface for the samples of isotopic contrast Pdadmac/ dSDS/ACMW (dominated by H2O) because air is located above a planar liquid surface and with time the aggregates sink downward away from it. The question remained as to whether other interaction mechanisms were possible. To examine possible effects of surface-trapping in this context, as opposed to the delivery of particles under gravity, we performed parallel NR measurements on fresh−mixed samples without surface cleaning. Figure 10 shows the neutron detector images of the reflection angle with respect to the wavelength recorded after 1 day on samples with different bulk

compositions. The horizontal lines signify the specular reflection of neutrons, and any diagonal lines correspond to off-specular neutron scattering from repeating units inside hexagonal-phase nanostructured aggregates present at the interface. The liquid-crystalline particles are present at the air/water interface only for the sample at the edge of the phaseseparation region with the highest bulk SDS concentration (F). As a result, in addition to the gravity-driven process of transporting nanostructured aggregates to interfaces determined previously,33 we have demonstrated a second mechanism of surface trapping of particles at the interface during its creation, which themselves do not penetrate the surface spontaneously. This nonequilibrium process is of course highly dependent on the sample history.

4. CONCLUSIONS In the present study we have demonstrated that distinct nonequilibrium processes control the interfacial structure and morphology of a strongly interacting P/S mixture and that they depend on both the extent of bulk aggregation and the sample history. We have mapped out systematically the mechanisms by which aggregates interact with the air/liquid interface using a number of complementary techniques, which together have helped to unravel some of the mysteries of what is a staggeringly complex problem. Our key findings fall into three areas. First, nonequilibrium aggregates spontaneously penetrate Pdadmac/SDS adsorption layers at the air/water interface. The extent of the interaction is related to the quantity and nature of aggregates formed in the bulk solution. Shortly after mixing there is a fresh dispersion of small aggregates, which precipitate with time, but particles can be mechanically redispersed from the precipitate to modify the interfacial properties. On redispersion, particles reach the interface by convection, where they dissociate and spread surface-active material into the interfacial layer. The charge and structure of the particles significantly affect this interaction as expanded, positively charged aggregates penetrate the adsorption layer more readily than compacted, negatively charged aggregates. This selectivity may be related to preferential interactions of the exposed polyelectrolyte segments on the particles with the oppositely charged headgroups of the surface monolayer of surfactant. Second, nonequilibrium effects can be exploited to select the quantity of aggregates delivered to the interface and in turn to modify the macroscopic surface properties. Both the structure and morphology of the material at the air/water interface can be completely different even for chemically identical samples simply as a result of the sample handling method used. Consequently, the surface tension may be tuned to a value of choice by careful selection of the sample handling methods, as these determine the quantity and nature of the particles liberated from the sediment that interact with the adsorption layer. Third, nonequilibrium particles inherently become trapped at the air/water interface during its creation. The lateral flocculation of these trapped particles results in surface structures with a length scale on the order of hundreds of micrometers. This process occurs most extensively for aggregates which have the lowest charge and does not recur spontaneously if they are effectively removed by surface cleaning. We may now outline three distinct mechanisms by which bulk P/S aggregates modify the structure and morphology of

Figure 10. Off-specular NR images comprising color maps of scattering for reflection angles of ±1.5° (vertical axis) with respect to the wavelength over a range of 2−30 Å (horizontal axis) for fresh− mixed Pdadmac/SDS samples in 0.1 M NaCl without surface cleaning by aspiration. The diagonal streak in sample F indicates the presence of embedded nanostructured aggregates in the interfacial layers. 8672

dx.doi.org/10.1021/la500621t | Langmuir 2014, 30, 8664−8674

Langmuir

■

the air/water interface: (1) spontaneous penetration as a result of specific interactions with the adsorption layer (prevalent when the aggregates have exposed excess polyelectrolyte segments which are oppositely charged with respect to the surfactant headgroups), (2) trapping at the interface during its creation, resulting in lateral flocculation (prevalent for neutral aggregates), and (3) unspecific deposition as a result of transport under gravity (prevalent for a planar air/liquid interface if the aggregates are less dense than the bulk liquid).37 To our knowledge the effects of these distinct processes have not to date been clearly set out and discussed. Any one of them may affect the stability of a foam film at the air/water interface through the spreading of material or may even result in its rupture.51,52 Furthermore, equivalent mechanisms may affect the emulsion stability at the liquid/liquid interface,53 as well as lubrication properties at the solid/liquid interface.54 Also, the exploitation of the delivery of functional materials from nonequilibrium aggregates to surfaces could be relevant in areas such as drugs, antibacterial agent or fragrance targeting.55,56 As such, an appreciation of the mechanisms and conditions under which they have a dominant influence could lead to optimization in the performance of formulations. Taking these findings together, we conclude that the structure and morphology of interfacial layers formed from oppositely charged P/S systems can be dominated by nonequilibrium effects. Although previous experimental data on the investigated system indicated a depleted adsorption layer and high surface tension values for the supernatant of aged samples in the phase-separation region, such properties are likely to occur only in a controlled laboratory setting using delicate sample-handling methods. Such specific conditions are very unlikely to be relevant when formulations are used in real world applications because the state of bulk aggregation can still be evolving and the convection experienced by aggregates can lead to substantial modification of the macroscopic interfacial properties which depends on the sample history. In this case, the dissociation and spreading of surface-active material at the interface from aggregates that originated in the bulk can be critically important in determining the interfacial properties. The factors which affect the different nonequilibrium processes outlined in this work clearly need to be appreciated fully in order to model the interfacial properties of such systems under practically relevant conditions.

■

Article

REFERENCES

(1) Möbius, D.; Miller, R.; Fainerman, V. B. Surfactants: Chemistry, Interfacial Properties, Applications; Elsevier Science & Technology: Amsterdam, 2001. (2) Rodriguez Patino, J. M.; Sanchez, C. C.; Rodriguez Nino, Ma. R. Implications of interfacial characteristics of food foaming agents in foam formulations. Adv. Colloid Interface Sci. 2008, 140, 95−113. (3) Thalberg, K.; Lindman, B. Polymer−Surfactant Interactions Recent Developments. In Interactions of Surfactants with Polymers and Proteins, Goddard, E. D., Ananthapadmanabhan, K. P., Eds.; CRC Press: Boca Raton, FL, 1993. (4) Kwak, J. C. T., Ed. Polymer−Surfactant Systems; Marcel Dekker: New York, 1998; Vol. 77. (5) Kizilay, E.; Kayitmazer, B.; Dubin, P. L. Complexation and coacervation of polyelectrolytes with oppositely charged colloids. Adv. Colloid Interface Sci. 2011, 167, 24−37. (6) La Mesa, C. Polymer−surfactant and protein−surfactant interactions. J. Colloid Interface Sci. 2005, 286, 148−157. (7) Langevin, D. Complexation of oppositely charged polyelectrolytes and surfactants in aqueous solutions. A review. Adv. Colloid Interface Sci. 2009, 147, 170−177. (8) Kogej, K. Association and structure formation in oppositely charged polyelectrolyte−surfactant mixtures. Adv. Colloid Interface Sci. 2010, 158, 68−83. (9) Goddard, E. D. Polymer/surfactant interaction: interfacial aspects. J. Colloid Interface Sci. 2002, 256, 228−235. (10) Taylor, D. J. F.; Thomas, R. K.; Penfold, J. Polymer/surfactant interactions at the air/water interface. Adv. Colloid Interface 2007, 132, 69−110. (11) Langevin, D.; Monroy, F. Interfacial rheology of polyelectrolytes and polymer monolayers at the air−water interface. Curr. Opin. Colloid Interface Sci. 2010, 15, 283−293. (12) Bain, C. D.; Claesson, P. M.; Langevin, D.; Mészáros, R.; Nylander, T.; Stubenrauch, C.; Titmuss, S.; von Klitzing, R. Complexes of surfactants with oppositely charged polymers at surfaces and in bulk. Adv. Colloid Interface Sci. 2010, 155, 32−49. (13) Noskov, B. A.; Loglio, G.; Miller, R. Dilational surface viscoelasticity of polyelectrolyte/surfactant solutions: formation of heterogeneous adsorption layers. Adv. Colloid Interface Sci. 2011, 168, 179− 197. (14) Pérez-Gil, J. Structure of pulmonary surfactant membranes and films: the role of proteins and lipid−protein interactions. Biochim. Biophys. Acta 2008, 1778, 1676−1695. (15) Zhang, S. Biological and Biomedical Coatings Handbook; CRC Press: Boca Raton, FL, 2011. (16) Kawakami, K.; Ebara, M.; Izawa, H.; Sanchez-Ballester, N. M.; Hill, J. P.; Ariga, K. Supramolecular approaches for drug development. Curr. Med. Chem. 2012, 19, 2388−2398. (17) Donkuru, M.; Badea, I.; Wettig, S.; Verrall, R.; Elsabahy, M.; Foldvari, M. Advancing non-viral gene delivery: lipid- and surfactantbased nanoparticle design strategies. Nanomedicine 2010, 5, 1103− 1127. (18) Kogej, K.; Skerjanz, J. In Physical Chemistry of Polyelectrolytes; Radeva, T., Ed.; Surfactant Science Series; Marcell Dekker: New York, 2001; Vol. 99, p 793. (19) Holmberg, K.; Jö nsson, B.; Kronberg, B.; Lindman, B. Surfactants and Polymers in Aqueous Solution, 2nd ed.; John Wiley & Sons: Chichester, West Sussex, England, 2003. (20) Sitar, S.; Goderis, B.; Hansson, P.; Kogej, K. Phase diagram and structures in mixtures of poly(styrenesulfonate anion) and alkyltrimethylammonium cations in water: significance of specific hydrophobic interaction. J. Phys. Chem. B 2012, 116, 4634−4645. (21) Li, D.; Kelkar, M. S.; Wagner, N. J. Phase behavior and molecular thermodynamics of coacervation in oppositely charged polyelectrolyte/surfactant systems: a cationic polymer JR 400 and anionic surfactant SDS mixture. Langmuir 2012, 28, 10348−10362. (22) Piculell, L. Understanding and exploiting the phase behavior of mixtures of oppositely charged polymers and surfactants in water. Langmuir 2013, 29, 10313−10329.

AUTHOR INFORMATION

Corresponding Authors

*Tel: +33 476 207 097. E-mail: [email protected]. *Tel: +36 204 890 440. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank the ILL for beam time on FIGARO and the Partnership for Soft Condensed Matter for use of the ellipsometer and Brewster angle microscope. This research has received funding from the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme FP7/2007-2013 under REA grant agreement no. 290251 and from the Hungarian Scientific Research Fund (OTKA K100762 and K108646), which is gratefully acknowledged. 8673

dx.doi.org/10.1021/la500621t | Langmuir 2014, 30, 8664−8674

Langmuir

Article

(23) Hansson, P.; Lindman, B. Surfactant-polymer interactions. Curr. Opin. Colloid Interface Sci. 1996, 1, 604−613. (24) Bergfeldt, K.; Piculell, L.; Linse, P. Segregation and association in mixed polymer solutions from Flory-Huggins model calculations. J. Phys. Chem. 1996, 100, 3680−3687. (25) Mészáros, R.; Thompson, L.; Bos, M.; Varga, I.; Gilányi, T. Interaction of sodium dodecyl sulfate with polyethyleneimine: surfactant-induced polymer solution colloid dispersion transition. Langmuir 2003, 19, 609−615. (26) Naderi, A.; Claesson, P. M.; Bergstrom, M.; Dedinaite, A. Trapped non-equilibrium states in aqueous solutions of oppositely charged polyelectrolytes and surfactants: effects of mixing protocol and salt concentration. Colloids Surf. A 2005, 253, 83−93. (27) Mezei, A.; Mészáros, R.; Varga, I.; Gilanyi, T. Effect of mixing on the formation of complexes of hyperbranched cationic polyelectrolytes and anionic surfactants. Langmuir 2007, 23, 4237−4247. (28) Noskov, B. A.; Grigoriev, D. O.; Lin, S.-Y.; Loglio, G.; Miller, R. Dynamic surface properties of polyelectrolyte/surfactant adsorption films at the air/water interface: poly(diallyldimethylammonium chloride) and sodium dodecylsulfate. Langmuir 2007, 23, 9641−9651. (29) Campbell, R. A.; Ash, P. A.; Bain, C. D. Dynamics of adsorption of an oppositely charged polymer-surfactant mixture at the air-water interface: poly(dimethyldiallylammonium chloride) and sodium dodecyl sulfate. Langmuir 2007, 23, 3242−3253. (30) Bell, C. G.; Breward, C. J. W.; Howell, P. D.; Penfold, J.; Thomas, R. K. Macroscopic modeling of the surface tension of polymer−surfactant systems. Langmuir 2007, 23, 6042−6052. (31) Bahramian, A.; Thomas, R. K.; Penfold, J. The adsorption behavior of ionic surfactants and their mixtures with nonionic polymers and with polyelectrolytes of opposite charge at the airwater interface. J. Phys. Chem. B 2014, 118, 2769−2783. (32) Campbell, R. A.; Arteta, M. Y.; Angus-Smyth, A.; Nylander, T.; Varga, I. Effects of bulk colloidal stability on adsorption layers of poly(diallyldimethylammonium chloride)/sodium dodecyl sulfate at the air−water interface studied by neutron reflectometry. J. Phys. Chem. B 2011, 115, 15202−15213. (33) Á braham, Á .; Campbell, R. A.; Varga, I. New method to predict the surface tension of complex synthetic and biological polyelectrolyte/surfactant mixtures. Langmuir 2013, 29, 11554−11559. (34) McLoughlin, D.; Langevin, D. Surface complexation of DNA with a cationic surfactant. Colloids Surf., A 2004, 250, 79−87. (35) Tonigold, K.; Varga, I.; Nylander, T.; Campbell, R. A. Effects of aggregates on mixed adsorption layers of poly(ethylene imine) and sodium dodecyl sulfate at the air/liquid interface. Langmuir 2009, 25, 4036−4046. (36) Campbell, R. A.; Angus-Smyth, A.; Arteta, M. Y.; Tonigold, K.; Nylander, T.; Varga, I. New perspective on the cliff edge peak in the surface tension of oppositely charged polyelectrolyte/surfactant mixtures. J. Phys. Chem. Lett. 2010, 1, 3021−3026. (37) Campbell, R. A.; Arteta, M. Y.; Angus-Smyth, A.; Nylander, T.; Varga, I. Multilayers at interfaces of an oppositely charged polyelectrolyte/surfactant system resulting from the transport of bulk aggregates under gravity. J. Phys. Chem. B 2012, 116, 7981−7990. (38) Gao, Y.; Duc, L. T.; Ali, A.; Liang, B.; Liang, J.-T.; Dhar, P. Interface-induced disassembly of a self-assembled two-component nanoparticle system. Langmuir 2013, 29, 3654−3661. (39) Angus-Smyth, A.; Bain, C. D.; Varga, I.; Campbell, R. A. Effects of bulk aggregation on PEI−SDS monolayers at the dynamic air− liquid interface: depletion due to precipitation versus enrichment by a convection/spreading mechanism. Soft Matter 2013, 26, 6103−6117. (40) Azzam, R. M. A.; Bashara, N. M. Ellipsometry and Polarized Light; North-Holland: Amsterdam, 1977. (41) Motschmann, H.; Teppner, R. Ellipsometry in Interface Science. In Novel Methods to Study Interfacial Layers; Miller, R., Mœbius, D., Eds.; Elsevier: New York, 2001. (42) Drude, P. Ueber die Reflexion und Brechung ebener Lichtwellen beim Durchgang durch eine mit Oberflächenschichten behaftete planparallele Platte. Annalen der Physik 1891, 43, 126−157.

(43) Goates, S. R.; Schofield, D. A.; Bain, C. D. A Study of Nonionic Surfactants at the Air−Water Interface by Sum-Frequency Spectroscopy and Ellipsometry. Langmuir 1999, 15, 1400−1409. (44) Tyrode, E.; Johnson, C. M.; Rutland, M. W.; Day, J. P. R.; Bain, C. D. A Study of the Adsorption of Ammonium Perfluorononanoate at the Air−Liquid Interface by Vibrational Sum-Frequency Spectroscopy. J. Phys. Chem. C 2007, 111, 316−329. (45) Campbell, R. A.; Wacklin, H. P.; Sutton, I.; Cubitt, R.; Fragneto, G. FIGARO: the new horizontal neutron reflectometer at the ILL. Eur. Phys. J. Plus 2011, 126, 107. (46) Á brahám, A.; Mezei, A.; Mészáros, R. The effect of salt on the association between linear cationic polyelectrolytes and sodium dodecyl sulfate. Soft Matter 2009, 5, 3718−3726. (47) Pojják, K.; Bertalanits, E.; Mészáros, R. Effect of salt on the equilibrium and nonequilibrium features of polyelectrolyte/surfactant association. Langmuir 2011, 27, 9139−9147. (48) Yanez Arteta, M.; Campbell, R. A.; Nylander, T. Adsorption of mixtures of poly(amidoamine) dendrimers and sodium dodecyl sulfate at the air-water interface. Langmuir 2014, 30, 5871−5828. (49) Nizri, G.; Magdassi, S.; Schmidt, J.; Cohen, Y.; Talmon, Y. Microstructural characterization of micro- and nanoparticles formed by polymer−surfactant interactions. Langmuir 2004, 20, 4380−4385. (50) Nizri, G.; Lagerge, S.; Kamyshny, A.; Major, D. T.; Magdassi, S. Polymer−surfactant interactions: binding mechanism of sodium dodecyl sulfate to poly(diallyldimethylammonium chloride). J. Colloid Interface Sci. 2008, 320, 74−81. (51) von Klitzing, R.; Muller, H. J. Film stability control. Curr. Opin. Colloid Interface Sci. 2002, 7, 42−49. (52) Kristen, N.; von Klitzing, R. Effect of polyelectrolyte/surfactant combinations on the stability of foam films. Soft Matter 2010, 6, 849− 861. (53) Tadros, T. Polymeric surfactants in disperse systems. Adv. Colloid Interface Sci. 2009, 147−148, 281−299. (54) Bae, S. C.; Granick, S. Molecular motion at soft and hard interfaces: from phospholipid bilayers to polymers and lubricants. Annu. Rev. Phys. Chem. 2007, 58, 353−374. (55) Pavlukhina, S.; Sukhishvili, S. Polymer assemblies for controlled drug delivery of bioactive molecules from surfaces. Adv. Drug Delivery 2011, 63, 822−836. (56) Chiappisi, L.; Hoffmann, I.; Gradzielski, M. Complexes of oppositely charged polyelectrolytes and surfactants - recent developments in the field of biologically derived polyelectrolytes. Soft Matter 2013, 9, 3896−3909.

8674

dx.doi.org/10.1021/la500621t | Langmuir 2014, 30, 8664−8674