Competitive Surfactant Adsorption of AOT and TWEEN 20 on Gold

Aug 26, 2014 - ... Adsorb Sodium Dodecyl Sulfate (SDS). Zahra Rahmani , Mohammad Taghi Samadi. Avicenna Journal of Environmental Health Engineering ...
0 downloads 0 Views 914KB Size
Download PDF
Article pubs.acs.org/Langmuir

Competitive Surfactant Adsorption of AOT and TWEEN 20 on Gold Measured Using a Quartz Crystal Microbalance with Dissipation Jakkrit Thavorn,† Joshua J. Hamon,‡ Boonyarach Kitiyanan,† Alberto Striolo,*,‡ and Brian P. Grady‡ †

The Petroleum and Petrochemical College, Chulalongkorn University, Bangkok 10330, Thailand School of Chemical, Biological and Materials Engineering and Institute of Applied Surfactant Research, University of Oklahoma, Norman, Oklahoma 73019, United States

‡

ABSTRACT: Competitive surfactant adsorption of anionic surfactant AOT and nonionic surfactant Tween 20 on gold was investigated by using a quartz crystal microbalance with dissipation (QCM-D) at 25 °C. The adsorption isotherm of pure AOT did not reach a plateau at the CMC, but rather adsorption continued to increase gradually at concentrations higher than the CMC before reaching a plateau. This behavior is evidence of competitive adsorption between AOT and impurities. The adsorbed layer of AOT on gold became more viscoelastic as the concentration of AOT increased. Tween 20 reached the plateau adsorption on gold before its concentration reached the CMC, suggesting that the attraction between Tween 20 and gold is very strong. The Tween 20 adsorbed layer was rigid when compared to the AOT adsorbed layer, as indicated by low dissipation. The addition of Tween 20 to a surface covered by AOT resulted in an increase in adsorbed mass, suggestive of the insertion of Tween 20 into the AOT adsorbed layer as expected because Tween 20 is able to separate the repulsive headgroups of AOT. When AOT was added to a preformed Tween 20 layer, a drop in the adsorbed amount was found between 0 and 0.1 CMC, and then no change was observed until the CMC of AOT was reached; the adsorbed amount then increased, reaching a final adsorption greater than that of pure AOT. All data support the formation of mixed surfactant layers on the surface. Although a two-step model fit both AOT and Tween 20 adsorption kinetic data well, AOT was found to adsorb much more slowly than Tween 20.

■

acquisition.20 Studies using QCM on mixed surfactant adsorption have been detailed in the literature. Liu and Kim21 studied the adsorption of the anionic surfactant sodium lauryl sulfate (SLS), that of the amphoteric surfactant cocamidopropyl betaine (CAPB), and that of a mixture of the two on poly(ether sulfone)-coated quartz crystals. Surfactant adsorption was studied at the respective CMCs. When alone, both SLS and CAPB formed rigidly bound films. The mixed surfactants yield a film of mass in between those of the individual surfactants, and desorption did not occur after the injection of pure water. In contrast, either pure surfactant was removed when pure water was injected. Sakai et al.22 studied the adsorption of mixtures of cationic surfactants ((dodecyltrimethylammonium bromide (DTAB) and hexadecyltrimethylammonium bromide (HTAB)) and nonionic octaethylene glycol monododecyl ether (C12E8) with gemini surfactant 1,2-bis(dodecyldimethylammonio)ethane dibromide (12-2-12). Adsorption isotherms for the mixed surfactant systems were shifted to lower surfactant concentrations, i.e., more adsorption at less surfactant concentration, when the latter was normalized to the respective pure or mixed CMCs. The mixed adsorbed layer became more rigid when compared

INTRODUCTION The adsorption of surfactants at solid−liquid interfaces is a phenomenon of practical importance. Some relevant fields include detergency, surface treatment, protection of metal surfaces, and stabilization of solid dispersions.1 Surfactant adsorption has been studied using a variety of experimental instruments such as quartz crystal microbalance,2−4 ellipsometry,5−7 neutron reflection,8−10 and atomic force microscopy (AFM).11−13 This paper focuses on the adsorption of surfactant mixtures. Mixed surfactant systems show some promising advantages over single surfactant systems.14−18 In the case of surfactant mixtures where one surfactant is charged and the other is not, a synergistic effect, e.g., a reduction in the critical micelle concentration (CMC),19 is typically observed. This synergistic effect is due to nonideal mixing effects in micellar aggregates resulting from the more alternating character of individual surfactant placement as opposed to the random placement characteristic of ideal mixing. Because most surfactant solution characteristics scale with the CMC, less surfactant is typically necessary for a given application when mixtures are used rather than pure surfactants. The quartz crystal microbalance with dissipation (QCM-D) provides in situ measurement of both equilibrium and kinetics of adsorption from liquid systems onto solid surfaces. QCM-D is characterized by high sensitivity and real time data © 2014 American Chemical Society

Received: June 26, 2014 Revised: August 26, 2014 Published: August 26, 2014 11031

dx.doi.org/10.1021/la502513p | Langmuir 2014, 30, 11031−11039

Langmuir

Article

Figure 1. Chemical structures of (a) AOT and (b) Tween 20.

ethylene oxygens of Tween 20, and because of hydrophobic interactions between the hydrophobic part of the surface and Tween 20 tails. To our knowledge, no studies have examined adsorption from a mixture of AOT and Tween 20. In the present work, AOT and Tween 20 are studied with respect to their adsorption on gold, both individually and as mixtures, using the QCM-D. Gold was chosen as a solid substrate because our group has used this substrate for other surfactants, hence we can compare the present results to those reported previously, and because gold sensors are typically very reliable for conducting QCM-D experiments. The effect of the sequential addition of the two surfactants was investigated and compared with the results for the single surfactant systems to quantify competitive adsorption. As described previously,34 the treated gold surfaces used in this study yield a contact angle of 10 ± 2° for water. Prior XPS measurements showed that the oxygen content of the surface was 36.3%. In addition, because of surface roughness, the actual surface area of the gold surfaces used herein is larger than the nominal surface area. Following Gutig et al.,34 in the present work the actual surface area was estimated to be 1.39 cm2, compared to the nominal surface area of 0.79 cm2.

to the layers obtained using single surfactant systems in all cases. Other adsorption studies of surfactant mixtures not involving the QCM have been reported. Muherei and coworkers23 studied the adsorption of anionic surfactant sodium dodecyl sulfate (SDS) and nonionic surfactant poly(ethylene glycol) tert-octylphenyl ether (TX-100) on shale and sandstone. For the single surfactant systems, the amount of adsorbed TX-100 was higher than that of SDS on both surfaces whereas the adsorption of TX-100 was reduced in the presence of SDS, which was attributed to the lower monomer concentration in the mixed system. On polystyrene, SDS was allowed to adsorb first and then Triton X-405 was added; a small excess of Triton X-405 was able to replace SDS totally on the surface. When the experiment was reversed (SDS was added to preadsorbed Triton X-405), it was found that excess SDS in solution was able to replace the adsorbed Triton X-405. The mutual substitution of the adsorbed surfactant was interpreted as evidence of equilibrium behavior.24 In another experiment, SDS and nonionic pentaethylene glycol monodecyl ether (C10E5) were adsorbed onto silica. As pure components, SDS did not adsorb whereas C10E5 did adsorb. In the mixture, when increasing the SDS/C10E5 molar ratio, the amount adsorbed of C10E5 decreased because of the formation of mixed micelles in solution. This effect was also found when C10E5 was preadsorbed on silica, followed by exposure to SDS solution, i.e., C10E5 desorbed to form mixed micelles with SDS.25 Overall, in the case of one charged surfactant and the other uncharged where one surfactant adsorbs and the other does not, if the adsorbing surfactant is adsorbed followed by rinsing with a solution of the nonadsorbing surfactant, then the adsorption on the surface decreases because the adsorbed surfactant is pulled into micelles in solution.24−26 However, cooperative adsorption can occur, leading to a higher amount adsorbed.27 The surfactants used for this study are representative of mixed surfactant systems used as dispersants to clean up oil spills in water.28 COREXIT 9500, an oil dispersant produced by NALCO, was used during the BP Deepwater Horizon spill in the Gulf of Mexico in 2010.29 The two main surfactants used in COREXIT 9500 are dioctyl sodium sulfosuccinate (AOT) and polyoxyethylene (20) sorbitan monooleate (Tween 80).30 Li et al.31,32 found that AOT adsorbed on hydrophobic silica because of strong intermolecular interactions between the AOT hydrocarbon chains; AOT was also found to adsorb on hydrophilic silica as a lamellar aggregate at concentrations higher than its CMC. Polyoxyethylene sorbitan monolaurate (Tween 20) was used in this study. Tween 20 and Tween 80 have the same hydrophilic headgroup but different hydrophobic tails. Tween 20 and Tween 80 are derived from monolauric acid and monooleic acid, respectively. Seo and coworkers33 studied the adsorption of Tween 20 on lignocellulose. Tween 20 formed a monolayer on the surface because of hydrogen bonding between carboxyl and hydroxyl groups of lignin and

■

EXPERIMENTAL SECTION

Materials. Aerosol-OT, indicated as AOT (sodium bis(2-ethyl-1hexyl) sulfosuccinate, 98% purity, batch no. MKBJ6637 V) and Tween 20 (polyoxyethylene (20) sorbitan monolaurate, BioXtra, batch no. SLBD2600 V) were purchased from Sigma-Aldrich. The structures of these surfactants are shown in Figure 1. Neither surfactant has monodisperse alkyl or alkoxy moieties. All surfactants were used as received. A 25 mM stock solution of AOT and a 3.913 mM stock solution of Tween 20 were prepared in very pure water (resistivity = 18.2 MΩ· cm). Gold sensor crystals (fundamental oscillation frequency of 5 MHz, 100 nm (QSX 301)) were purchased from Q-Sense. Critical Micelle Concentration Measurements. The critical micelle concentrations (CMC) of AOT and Tween 20 were measured using a dynamic contact angle analyzer (DCA-322), purchased from Cahn Scientific, at ambient temperature using the Wilhelmy plate method with glass slides (25 × 25 cm2). The CMCs were measured three times for each pure surfactant and for the 1:1 molar mixture, and the average values are reported. QCM-D. The quartz crystal microbalance with dissipation (QCMD) measures adsorption characteristics at the solid−liquid interface through their impact on a piezo oscillation frequency. The piezo is a disk of single-crystal quartz accessed via metal electrodes. The disk is coated with various materials on which adsorption is measured.35 In our case, the coating was gold, although this gold shows hydrophilic properties, as discussed elsewhere.34 QCM-D measures the change in the resonance frequency of crystal quartz as well as energy dissipation. The change in resonance frequency relates to the amount of adsorbed mass according, as a first approximation, to the Sauerbrey relation:36 Δm =

− C Δf n

(1)

In eq 1, Δm is the adsorbed mass, C is a constant characteristic of the crystal (C = 17.7 ng·cm−2·Hz−1 for the crystals used in our 11032

dx.doi.org/10.1021/la502513p | Langmuir 2014, 30, 11031−11039

Langmuir

Article

experiments at 5 MHz), Δf is the difference between the resonance frequency of the crystal sensor exposed to a solution ( f) and its nominal value (f 0), Δf = f − f 0, and n is the overtone number (n = 1, 3, 5, ...). In this work the amount of adsorbed mass was calculated from the average of all data from overtones n = 3, 5, 7, ..., 13. Equation 1 is based on the assumption that the adsorbed mass forms a thin rigid film with uniform distribution over the whole surface. Equation 1 is appropriate when the dissipation is 2 × 10−6 or less.37 Energy dissipation reflects the viscoelasticity of the adsorbed films and is determined by measuring the characteristics of the decay in the crystal oscillation after the driving voltage is ceased. The dissipation factor is shown in eq 238

D=

In an attempt to quantify competitive surfactant adsorption, the effect of sequential addition was studied. Specifically, a 1.4(CMC) concentration of either AOT or Tween 20 was first injected into the QCM-D sensing element. Sufficient time was allowed for constant adsorption to be reached before further injection of a solution of either [Tween 20]/[AOT at 1.4(CMC)] or [AOT]/[Tween 20 at 1.4(CMC)] with the first surfactant in the set of pairs varying from 0.1(CMC) to 2.5(CMC). In these experiments, the concentration of the preadsorbed surfactant was maintained at 1.4(CMC) during the injection of the second surfactant to minimize the desorption of the preadsorbed surfactant due to lower bulk concentration.

■

RESULTS AND DISCUSSION CMC Determination. Critical micelle concentrations (CMC) of AOT and Tween 20, as measured by surface tension, are shown in Table 1. Acceptable agreement was

Edissipated 2πEstored

(2)

where Edissipated is the dissipated energy during one oscillation and Estored is the stored energy during the oscillation cycle. With the QCMD, we measured the change in the dissipation factor, ΔD = D − D0, where D is the dissipation factor at any given time during the experiment and D0 is the dissipation factor of a clean crystal immersed in the pure solvent. Because the dissipation changes with the solution viscosity, D0 should be measured when the crystals are exposed to a surfactant solution, before adsorption occurs, rather than with pure water. However, the CMCs for the surfactants used here are small enough that the viscosity change can be neglected.39 Cleaning Procedure for a Gold Surface. Before each crystal was inserted into the flow modules, the crystals were cleaned following the Q-Sense protocol40 to remove organic and biological material, and obtain reproducible measurements. Crystals were placed in a plasma cleaner (PCD-32G, purchased from Harrick Plasma) at a medium rf level for 10 min in air at 0.7 atm pressure using a vacuum pump (model 15 600, purchased from Robinair). Then the crystals were submerged in an ammonia−peroxide “base piranha” mixture (APM) (5:1:1 18.2 MΩ·cm water, 25% ammonium hydroxide, and 30% hydrogen peroxide, respectively) at 75 °C for 5 min. After that, the crystals were thoroughly washed with 18.2 MΩ·cm water, dried in a stream of N2, and again placed in the plasma cleaner in air at 0.7 atm at a low rf level for 5 min. Adsorption Isotherm Measurements and Dynamics of Adsorption. A QCM-D (model E4, purchased from Q-Sense) was used. The temperature was kept constant at 25 ± 0.05 °C. All data reported are the averages of three measurements. In each measurement, four sensing elements were used simultaneously, although in some cases a sensing element malfunctioned. The error bars reported represent one standard deviation. In brief, a QCM-D model E4 is composed of three parts: the chamber platform that contains the four flow modules, each containing a sensor crystal, the Ismatec peristaltic pump, and the electronics unit. Before starting a measurement, a stable baseline with pure water was obtained by recording the frequency that did not drift more than 0.5 Hz for at least 30 min. Once the baseline was obtained, a series of surfactant concentration increments from 0 to 2.5(CMC) flowed through the modules at a rate of 0.1 mL/min. The steps were ∼0.1(CMC), 0.2(CMC), and 0.5(CMC) for concentration ranges from 0 to 0.6(CMC), 0.6(CMC) to 2.0(CMC), and 2.0(CMC) to 2.5(CMC), respectively. For Tween 20, we needed smaller CMC increments from 0.004(CMC) to 0.08(CMC) to obtain the adsorption isotherm because Tween 20 adsorption reached a plateau at ∼0.1(CMC). At each surfactant concentration, sufficient time was allowed until the frequency reached a plateau before increasing the surfactant concentration further. The plateaus were identified as regions in which the frequency did not change more than 0.1 Hz per minute. This procedure required 30−60 min for each surfactant concentration, depending on the concentration and properties of the surfactants. To study adsorption dynamics, the frequency was recorded as a function of time when the bulk surfactant concentration was abruptly increased from 0 to 0.1(CMC) and from 0 to 1.4(CMC) on a clean surface (i.e., no surfactant preadsorbed).

Table 1. Critical Micelle Concentrations CMC (mM)

AOT Tween 20 1:1 molar mixture of AOT and Tween 20

current study

previous measurements

2.91 ± 0.12 0.08 ± 0.002 1.07 ± 0.03

2.50,41 3.1342 0.0643 N/A

achieved between our measurements and available literature data. Regarding AOT, the difference between literature data might be due to impurities in the samples. While Fragneto et al.41 purified AOT by liquid−liquid extraction and found a CMC of 2.50 mM, Grillo and Penfold42 measured the CMC of AOT (purity >98%) without additional purification and found that the CMC was 3.13 mM. As mentioned in the Experimental Section, the surfactants in our study were used as received. AOT Adsorption. The adsorption isotherm for AOT at 25 ± 0.05 °C, calculated using the Sauerbrey relation, is shown in Figure 2 (top panel, black circles). The isotherm follows the typical L2 designation.44 The amount of AOT adsorbed per unit of actual surface area increases continuously on the gold surface up to the bulk concentration of 1.7(CMC) and then reaches a plateau. The AOT surface density at the adsorption plateau in terms of mass and mole adsorption is reported in Table 2. The adsorbed mass was not expected to increase above the CMC, although some papers report such an increase. For example, Esumi et al.45 found an increase in adsorption above the CMC for C10E6 on silica and suggested that the hydrogen bonds between the ether oxygen atoms of the surfactant and the hydroxyl groups of silica plus the intermolecular hydrophobic interactions were two factors driving this behavior. Specifically, at low concentrations the authors postulated that C10E6 lies flat on silica and, as the concentration increases, surfactant tails tend to be squeezed out and become perpendicular to the surface to form a closely packed adsorbed layer. The authors suggest that this transformation does not complete until the concentration is higher than the CMC. In our case, because the amount of surface area relative to the mass of surfactant is quite low, and we used AOT without additional purification, the adsorption isotherm might be affected by impurities such as monoester, diester, ethylhexanol, and residues from the manufacturing process. Different chain lengths of hydrophobic tails can also cause such an effect. As micelles form, some of the coadsorbed minor component could partition to the micelles, which in turn affects the amount 11033

dx.doi.org/10.1021/la502513p | Langmuir 2014, 30, 11031−11039

Langmuir

Article

Although the dissipation is less than 2 × 10−6 over the entire adsorption isotherm and hence the Sauerbrey relation should hold, AOT adsorption data were also fitted to the Voigt model, using the Q-tool software, to determine whether the presented adsorbed mass values depend on the model chosen. AOT data were fit rather than Tween 20 ones because the dissipation of AOT is higher than that of Tween 20 (Figure 2). To implement the Voigt model values for the density of the adsorbed layer, the density of the supernatant and the viscosity of the supernatant must be assumed. These were assumed to be 1,050 kg·cm −3 , 1,000 kg·cm −3 , and 0.001 kg·m −1 ·s −1 , respectively. The mass adsorbed calculated by Voigt model (gray circles) are compared to those obtained by the Sauerbrey relation (black circles) in Figure 3. The two models do not differ within the experimental accuracy of the data. All data presented in this paper, except those in Figure 3, were obtained using the Sauerbrey relation.

Figure 2. Experimental adsorption isotherms (top) and measured energy dissipation (bottom) for AOT (black circles) and Tween 20 (gray squares) measured on a gold surface at 25 ± 0.05 °C as a function of bulk concentration. The adsorbed amount is estimated using the Sauerbrey relation. The bulk surfactant concentration is normalized by the respective surfactant CMC (Table 1).

Figure 3. Experimental adsorption isotherm for AOT on gold at 25 ± 0.05 °C calculated by either the Sauerbrey relation (●) or the Voigt model (gray circle).

Table 2. Surface Adsorption Density of AOT and Tween 20 on Gold above the CMC AOT Tween 20

adsorbed mass (ng·cm−2)

adsorbed mole (10−12 mol·cm−2)

146 ± 22 197 ± 23

328 ± 28 160 ± 10

Tween 20 Adsorption. The adsorption isotherm for Tween 20 is shown in Figure 2 (top panel). Also, this isotherm reproduces the typical L2 features.44 The maximum amount adsorbed is higher for Tween 20 than for AOT, indicating that packing is more efficient (efficiency being defined as more surfactant atoms adsorbed per unit surface area) for Tween 20. The amount of Tween 20 adsorbed above its CMC is reported in Table 2. Tween 20 is found to partition heavily on the surface even at low bulk concentrations; in fact, adsorption is complete already at about 0.15(CMC). Similar observations were reported by ́ ́ Martin-Rodri guez et al.,49 who studied the adsorption of Tween 20 onto latexes and found a saturation value at about 0.3(CMC). The dissipation measured for adsorbed Tween 20 aggregates is shown in Figure 2 (bottom panel, gray squares). The low dissipation indicates a rigid film. We note that the value of 0.4 × 10−6 for Tween 20 is lower than that for AOT and also lower than that reported previously for CTAB.34,47 This was not expected, given the molecular architecture of Tween 20. However, Tween 20 has been shown to adsorb more densely packed than Tween 40 on hydrophobic gold even though Tween 40 has a longer hydrophobic tail than Tween 20.50 On the basis of the fit of adsorption models, the authors postulated that at low surface coverage Tween 20 adsorbed by hydrophobic interaction with the surface and the tail group laid essentially flat on the surface. Once the surface coverage became high, Tween 20 could self-organize to allow other

adsorbed at the solid interface; such effects are well known in surface tension measurements. In a previous publication,46 we attributed an anomalous feature in the adsorption isotherm (a large peak in adsorption at the CMC) to impurities. In Figure 2 (bottom panel), changes in dissipation, ΔD, are shown for AOT as a function of bulk concentration (black circles). The change in dissipation remains below 2 × 10−6 within the entire range of concentrations considered, indicating that the AOT film is not very viscoelastic.37 However, the dissipation increases with concentration, especially close to and above the CMC. Our group previously suggested that morphology affects dissipation.47 Specifically, under certain conditions we have observed a decrease in the measured dissipation, which was attributed to a change in adsorbed CTAB aggregate morphology from cylindrical to flat bilayer structure. In the present experiment, an increase in dissipation occurs above the CMC as does the adsorbed amount. If impurities are responsible for these effects, as they partition from the admicelles to the bulk micelles, they render the admicelles less rigid. On the basis of recent simulation results,48 it is also possible that, because of heterogeneous surface properties, large flexible aggregates form on the surface under conditions near the surfactant CMC. 11034

dx.doi.org/10.1021/la502513p | Langmuir 2014, 30, 11031−11039

Langmuir

Article

SDS on silica precoated with nonionic surfactant C10E5. They found C10E5 desorbed and formed mixed micelles with SDS; the latter was determined via free energy calculations. Moreover, they observed an enthalpy increase when SDS was added due to the heat of mixing between desorbed C10E5 and the SDS due to mixed micelle formation. Such measurements were not attempted here. Dissipation data are shown in Figure 4 (bottom panel). When AOT is added to a Tween 20 film, ΔD scales with adsorbed amount in all the three regions discussed above. Specifically, a drop occurs in dissipation from 0 to 0.1(CMC), although the ΔD drop may not be outside of experimental error. From 0.1(CMC) to 1.0(CMC), the dissipation remains constant at 0.4 × 10−6, which corresponds to approximately the dissipation measured for pure Tween 20 around its CMC (Figure 2). Above the AOT CMC, the dissipation is lower than that of pure AOT but increases to values larger than those reported for Tween 20. Hence, both the dissipation and amount adsorbed indicate that the surface coverage is a mixture of Tween 20 and AOT when AOT is added to a preadsorbed Tween 20 film. Adsorption of Tween 20 on a Surface Covered with an Adsorbed AOT Film. The experiment in Figure 4 was repeated in reverse, i.e., preparing an AOT film and then gradually increasing the Tween 20 concentration. Adsorption isotherms and dissipation are shown in Figure 5 (top and bottom panels, respectively). When Tween 20 is added to the system, the results show that the adsorbed mass increases. The adsorbed amount increases quickly at Tween 20 concentrations

molecules to adsorb on the surface and form a high-density layer with cooperative interactions between Tween 20 molecules. For Tween 40, a simpler single-mechanism adsorption was found to be adequate to fit the adsorption data, suggesting that self-organization did not occur.50 Adsorption of AOT on a Surface Covered by an Adsorbed Tween 20 Film. We exposed a film of Tween 20 prepared on gold at 1.4(CMC) to increasing concentrations of aqueous AOT at 25 ± 0.05 °C. The results are shown in Figure 4. A small decrease in the adsorbed mass occurred when

Figure 4. Experimental adsorption data (top) and energy dissipation (bottom) as a function of bulk concentration at 25 ± 0.05 °C obtained when an aqueous system containing AOT (at increasing concentration) and Tween 20 (at 1.4 Tween 20 CMC) was allowed to flow over a gold surface covered by a preformed Tween 20 film (●). The results are compared to those obtained for pure AOT (gray circle), which are reproduced from Figure 2.

0.1(CMC) AOT was added, suggesting that some Tween 20 was removed from the surface. As the AOT concentration increased from 0.1(CMC) to 1.0(CMC), both the adsorbed amount and ΔD did not change, indicating that either (1) AOT displaces exactly the same weight of Tween 20 or (2) AOT does not displace additional Tween 20. Considering the differences in molecular properties (molecular weight, architecture, ionic features, etc., Figure 1) between the two surfactants, no displacement of Tween 20 by AOT likely occurred. When the AOT concentration was increased above its CMC, the amount adsorbed increased. Mass adsorbed at the highest AOT concentration considered in Figure 4 (top panel) corresponds to the mass of Tween 20 adsorbed when no AOT was present in the system and is higher than that obtained using only AOT. This result indicates that the final film is composed of both AOT and Tween 20. Unfortunately, QCM data are not sufficient to separate the two components. Other measurements can quantify the surface coverage of each component. Thibaut et al.25 observed the adsorption of anionic surfactant

Figure 5. Experimental adsorption data (top) and measured energy dissipation (bottom) as a function of bulk concentration at 25 ± 0.05 °C obtained when an aqueous system containing Tween 20 (at increasing concentration) and AOT (at 1.4 AOT CMC) was allowed to flow over a gold surface covered by a preformed AOT film (■). The results are compared to those obtained for pure Tween 20 (gray square), which are reproduced from Figure 2. 11035

dx.doi.org/10.1021/la502513p | Langmuir 2014, 30, 11031−11039

Langmuir

Article

Table 3. Parameters from Fitting the Adsorption Data on Gold to Kinetics Models [AOT] rate constant −1

k1 (min ) k2 (min−1)

[Tween 20]

0 → 0.1(CMC)

0 → 1.4(CMC)

0 → 0.1(CMC)

0 → 1.4(CMC)

0.157 ± 0.01 0.016 ± 0.003

0.216 ± 0.01 0.006 ± 0.0003

0.128 ± 0.01 0.041 ± 0.01

1.446 ± 0.04 0.108 ± 0.03

lower than ∼0.5(CMC) and then a little more slowly up to Tween 20 concentrations of ∼1.4(CMC), at which point a plateau is reached. The continuous increase in the adsorbed mass of Tween 20 to the AOT film is due to the fact that the nonionic surfactant can reduce the electrostatic repulsion between headgroups of the anionic surfactant so that more surfactants can adsorb on the surface.16,51 The adsorbed mass corresponding to the plateau obtained at high concentrations is similar, within experimental uncertainty, to that of pure Tween 20 adsorbed on gold above its CMC. This result is similar to that discussed in Figure 4 when the two surfactants were added in reverse order. The corresponding dissipation data are shown in Figure 5 (bottom panel). At all concentrations considered, ΔD for the mixed system is larger than that for Tween 20 alone. Furthermore, ΔD for the mixed system increases as the Tween 20 concentration increases. The rate of increase with concentration in the different regions follows the same qualitative trends discussed for the amount adsorbed. At high Tween 20 concentrations (above its CMC), ΔD results obtained for the mixed system are higher than those obtained for the mixed system prepared in reverse order (AOT added to a Tween 20 film, data in Figure 4) and are very similar, when statistical uncertainty is considered, to those obtained for AOT alone. In other words, the adsorbed amount for the mixed system at high Tween 20 concentration corresponds to that of Tween 20 alone, while the dissipation replicates that of AOT alone. We conclude that both procedures (adding AOT to a Tween 20 film or adding Tween 20 to an AOT film) yield an adsorbed film containing both AOT and Tween 20. Discussion of Equilibrium Adsorption. We can now determine whether the films obtained from the mixed surfactant systems, in the two approaches just described, have similar properties. Because the Tween 20 concentration in the supernatant was maintained at 1.4(CMC) during the experiments of Figure 4 while the AOT concentration was varied and because the AOT concentration in the supernatant was maintained at 1.4(CMC) during the experiments of Figure 5 while the Tween 20 concentration was varied, the results obtained at 1.4 AOT CMC in Figure 4 and those obtained at 1.4 Tween 20 CMC in Figure 5 were obtained at the same bulk conditions. The corresponding adsorbed masses (compare Figures 4 and 5, top panel) are similar in the two procedures (when errors are accounted for), while the film dissipations differ (compare Figures 4 and 5, bottom panel). This suggests that the films obtained during our experiments are not representative of equilibrated systems. This is not the only observation of nonequilibrium behavior; namely, the mass adsorbed for pure AOT in Figure 5 (concentration equals zero on this plot) does not match the value shown for AOT concentration = 1.4(CMC) in Figure 2. A similar observation holds for the dissipation of pure AOT at the same concentration in Figures 5 and 2. The adsorbed films used to start the experiment in Figure 5 were obtained after changing the surfactant concentration from 0 to 1.4(CMC), while that in Figure 2 was obtained by increasing step-by-step the

concentrations over small increments until 1.4(CMC) was reached. In our previous paper,34 we found that the amount adsorbed when adsorption occurred from a bulk concentration above the CMC can be much less than that from increasing the surfactant concentrations in a series of steps, although in that case the surfactant was a single component (if impurities are neglected). Atkin et al. also reported that different adsorbed amounts were obtained if adsorption was allowed to occur in an increasing step-by-step fashion or a decreasing step-by-step fashion.52 We do not have a good explanation for these aspects of nonequilibrium behavior, except to note that certain surfactant mixtures and impure surfactants take many hours or days to reach constant surface tensions. Perhaps our experiments indicate that equilibrium takes a very long time on a solid surface as well as at the air−liquid interface when the amount of surfactant adsorbed on a solid surface is small compared to the amount of surfactant in solution. Conducting long experiments with the QCM to test this possibility is not possible because of instrumental drift. Quantifying the presence of impurities in the adsorbed surfactant aggregates could be attempted by collecting FT-IR spectra.53 Such experiments were not attempted for this project. Dynamics of Adsorption. In our previous paper, we found adsorption rate constants by fitting adsorption models with either one-step rate (eq 3) or two-step rate models54 (eq 4). For CTAB on gold, we showed that the two-step rate model was necessary to fit the data. One- and two-step models are represented by the following equations: one‐step rate models: qt = qo + kot

(3)

two‐step rate models: qt = qe − A1e−k1t − A 2e−k 2t

(4)

In these equations, qo, qt, and qe are a constant, the amount of surfactant adsorbed at time t, and the amount of surfactant adsorbed at equilibrium, respectively. ki represents rate constants, and Ai represents pre-exponential terms. Rate constants k1 and k2 are related to the diffusion of monomer surfactants and micelles from the bulk to the surface and to the reorganization of the adsorbed surfactants. A two-step model was necessary to adequately fit the data obtained in the present experiments. The rate constants obtained for AOT and Tween 20 are shown in Table 3. The rate constants for both surfactants above the CMCs are larger than those obtained below the CMCs, which probably results from the large concentration gradient between the bulk and the surface above the CMCs. The results for AOT are discussed below. Kinetic measurements with surfactant concentration below and above the CMC were performed to help understand the mechanism of surfactant adsorption. The results are shown in Figure 6. Adsorption was not detected from 0 to 150 s after injecting the surfactant solution because of the time necessary for the fluid to flow through the tubing. Overall, AOT adsorption took a much longer time than Tween 20 to reach an approximately constant value (4 h vs less than 1 h). Under 11036

dx.doi.org/10.1021/la502513p | Langmuir 2014, 30, 11031−11039

Langmuir

Article

constant value is reached for the dissipation. Tween 20 shows much stronger adsorption in that the adsorbed amount reaches a plateau at ∼0.15(CMC) and the dissipation is indicative of a rigid film. In all cases, dissipation values are below the limit of applicability of the Sauerbrey model (2 × 10−6). For the addition of AOT to a Tween 20-covered surface, Tween 20 desorbs at very low AOT concentrations likely to form mixed micelles with AOT, and then the adsorbed mass does not change until the CMC is reached. As the AOT concentration increases above the CMC, the mass adsorbed increases until reaching that of pure Tween 20. Upon the addition of Tween 20 to an AOT-covered surface, the adsorbed mass increases monotonically until reaching an amount adsorbed comparable to that obtained from a pure Tween 20 solution. Evidence of nonequilibrium behavior was seen in experiments with both the pure surfactants and the surfactant mixtures. A two-step adsorption model was used to fit kinetics data for the adsorption of surfactants. The adsorption rate for AOT is slow both below and above the CMC, suggesting that the impurities in AOT solution can desorb from gold and lead to slow adsorption, or perhaps some very complicated surface rearrangement is responsible for the slow adsorption. Tween 20 adsorption is at a rate typically found for other surfactants; however, a peak in adsorption vs time for a concentration of 1.4(CMC) was found, which was attributed to impurities.

Figure 6. Fractional adsorption for AOT and Tween 20 as a function of time on the gold surface.

identical conditions, CTAB adsorption was complete within 1.5 h.34 Overall, the rate of Tween 20 adsorption is similar to that of most other surfactants measured using either QCM or ellipsometry. The peak in adsorption at short times for Tween 20 at 1.4(CMC) must be due to the fact that the Tween 20 solution is not purely one component, as we are not able to envision other explanations. Different mechanisms have been proposed for long equilibration times. Atkin et al.55,56 reported that CTAB adsorption on silica at 0.6 mM (0.67(CMC)) can take up to 3.5 h to reach equilibrium, and the kinetic models fitted to adsorption data obtained at this concentration yielded time constants much larger than those obtained at other concentrations. These authors explained this long time as being due to structural rearrangements on the surface. The concentration region at which the kinetics of adsorption was slow was termed the slow adsorption region. These authors also observed a slow adsorption region for cetylpyridinium bromide (CPBr) from 0.274 mM (0.3(CMC)) to 0.306 mM (0.34(CMC)) on silica. In our experiments, AOT adsorption was slow for both 0.1(CMC) and 1.4(CMC), suggesting that a specific morphological structure may not be responsible for the slow adsorption. One possibility for both the slow kinetics and the lack of change in rate with concentration is that the impurities in the AOT solution might cause slow adsorption: fast-adsorbing impurities may be deadsorbing to allow slowadsorbing species to adsorb because these species are energetically preferred at the surface. Finally, for Tween 20 as well as CTAB,55,57−60 the initial adsorption rate (less than 300 s) depends on the bulk surfactant concentration, with higher concentrations leading to faster adsorption. On the basis of concentration, the rate increase between 0.1(CMC) and 1.4(CMC) should be a factor of ∼10, and this increase is found for Tween 20. For AOT, the initial adsorption rates for AOT at the two different concentrations are basically the same, indicating that the number of adsorption sites for individual AOT molecules is limited.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Present Address

(A.S.) Department of Chemical Engineering, University College London, London WC1E 7JE, United Kingdom. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was supported by the Petroleum and Petrochemical College, Chulalongkorn University, Thailand, and the Institute of Applied Surfactant Research at the University of Oklahoma, Norman, Oklahoma, USA. The sponsoring companies for the Institute of Applied Surfactant Research are CESI Chemical, Church & Dwight, Clorox, ConocoPhillips, Cytec Industries, Ecolab, Halliburton Services, Huntsman, ITW Global Brands, MeadWestvaco, Novus International, Phillips 66, Procter and Gamble, Sasol, S. C. Johnson, Shell Chemical, and UK Abrasives. We thank Dr. Matthew Dixon from Biolin Scientific for fruitful help and discussions for using the Voigt Model in Q-tool software.

■

■

REFERENCES

(1) Clint, J. H. Surfactant Aggregation; Blackie: New York, 1992. (2) Knag, M.; Sjöblom, J.; Øye, G.; Gulbrandsen, E. A quartz crystal microbalance study of the adsorption of quaternary ammonium derivates on iron and cementite. Colloids Surf., A 2004, 250, 269−278. (3) Karlsson, P. M.; Palmqvist, A. E. C.; Holmberg, K. Adsorption of Sodium Dodecyl Sulfate and Sodium Dodecyl Phosphate on Aluminum, Studied by QCM-D, XPS, and AAS. Langmuir 2008, 24, 13414−13419. (4) Kou, J.; Tao, D.; Xu, G. A study of adsorption of dodecylamine on quartz surface using quartz crystal microbalance with dissipation. Colloids Surf., A 2010, 368, 75−83.

CONCLUSIONS Adsorption isotherms for AOT (anionic) and Tween 20 (nonionic) surfactants on gold are presented together with their competitive adsorption. In the case of AOT, the adsorbed mass increases above the CMC, a result thought to be due to impurities present in the AOT solution. The dissipation shows similar behavior, although the increase in dissipation above the CMC is larger than that below the CMC, and eventually a 11037

dx.doi.org/10.1021/la502513p | Langmuir 2014, 30, 11031−11039

Langmuir

Article

(5) Tiberg, F.; Ederth, T. Interfacial Properties of Nonionic Surfactants and Decane−Surfactant Microemulsions at the Silica− Water Interface. An Ellipsometry and Surface Force Study. J. Phys. Chem. B 2000, 104, 9689−9695. (6) Denoyel, R. Microcalorimetry and ellipsometry in surfactant adsorption studies. Colloids Surf., A 2002, 205, 61−71. (7) Poptoshev, E.; Claesson, P. M. Adsorption of dimethyldodecylamine-N-oxide at the mica-solution interface studied by ellipsometry. Colloids Surf., A 2006, 291, 45−50. (8) Vacklin, H. P.; Tiberg, F.; Fragneto, G.; Thomas, R. K. Composition of Supported Model Membranes Determined by Neutron Reflection. Langmuir 2005, 21, 2827−2837. (9) Yaseen, M.; Wang, Y.; Su, T. J.; Lu, J. R. Surface adsorption of zwitterionic surfactants: n-alkyl phosphocholines characterised by surface tensiometry and neutron reflection. J. Colloid Interface Sci. 2005, 288, 361−370. (10) Wang, X.; Lee, S. Y.; Miller, K.; Welbourn, R.; Stocker, I.; Clarke, S.; Casford, M.; Gutfreund, P.; Skoda, M. W. A. Cation Bridging Studied by Specular Neutron Reflection. Langmuir 2013, 29, 5520−5527. (11) Ducker, W. A.; Grant, L. M. Effect of Substrate Hydrophobicity on Surfactant Surface−Aggregate Geometry. J. Phys. Chem. 1996, 100, 11507−11511. (12) Wall, J. F.; Zukoski, C. F. Alcohol-Induced Structural Transformations of Surfactant Aggregates. Langmuir 1999, 15, 7432−7437. (13) Velegol, S. B.; Fleming, B. D.; Biggs, S.; Wanless, E. J.; Tilton, R. D. Counterion Effects on Hexadecyltrimethylammonium Surfactant Adsorption and Self-Assembly on Silica. Langmuir 2000, 16, 2548− 2556. (14) Esumi, K.; Sakamoto, Y.; Meguro, K. Mixed bilayers of anionic and nonionic surfactants on alumina. J. Colloid Interface Sci. 1990, 134, 283−288. (15) Huang, L.; Maltesh, C.; Somasundaran, P. Adsorption Behavior of Cationic and Nonionic Surfactant Mixtures at the Alumina−Water Interface. J. Colloid Interface Sci. 1996, 177, 222−228. (16) Sakai, K.; Yoshimura, T.; Esumi, K. Interaction Forces on αAlumina Surfaces with Coadsorption of Anionic/Nonionic Surfactant Mixtures. J. Phys. Chem. B 2003, 107, 12227−12232. (17) Ivanova, N. I.; Parfenova, A. M.; Amelina, E. A. Interactions between nonpolar surfaces in mixed solutions of cationic and nonionic surfactants. Moscow Univ. Chem. Bull. 2007, 62, 147−151. (18) Fainerman, V. B.; Aksenenko, E. V.; Petkov, J. T.; Miller, R. Equilibrium adsorption layer characteristics of mixed sodium dodecyl sulphate/Triton solutions. Colloids Surf., A 2011, 385, 139−143. (19) Zhang, Y.; Zhao, Y.; Zhu, Y.; Wu, H.; Wang, H.; Lu, W. Adsorption of mixed cationic-nonionic surfactant and its effect on bentonite structure. J. Environ. Sci. 2012, 24, 1525−1532. (20) Caruso, F.; Serizawa, T.; Furlong, D. N.; Okahata, Y. Quartz Crystal Microbalance and Surface Plasmon Resonance Study of Surfactant Adsorption onto Gold and Chromium Oxide Surfaces. Langmuir 1995, 11, 1546−1552. (21) Liu, S. X.; Kim, J.-T. Study of adsorption kinetics of surfactants onto polyethersulfone membrane surface using QCM-D. Desalination 2009, 247, 355−361. (22) Sakai, K.; Matsuhashi, K.; Honya, A.; Oguchi, T.; Sakai, H.; Abe, M. Adsorption Characteristics of Monomeric/Gemini Surfactant Mixtures at the Silica/Aqueous Solution Interface. Langmuir 2010, 26, 17119−17125. (23) Muherei, M. A.; Junin, R.; Bin Merdhah, A. B. Adsorption of sodium dodecyl sulfate, Triton X100 and their mixtures to shale and sandstone: A comparative study. J. Pet. Sci. Eng. 2009, 67, 149−154. (24) Bolze, J.; Hörner, K. D.; Ballauff, M. Competitive adsorption of an anionic and a nonionic surfactant on polystyrene latex particles as monitored by small x-ray scattering. Colloid Polym. Sci. 1996, 274, 1099−1108. (25) Thibaut, A.; Misselyn-Bauduin, A. M.; Grandjean, J.; Broze, G.; Jérôme, R. Adsorption of an Aqueous Mixture of Surfactants on Silica. Langmuir 2000, 16, 9192−9198.

(26) Kayes, J. B. The effect of surface active agents on the microelectrophoretic properties of a polystyrene latex dispersion. Microelectrophoretic studies. J. Colloid Interface Sci. 1976, 56, 426− 442. (27) Colombié, D.; Landfester, K.; Sudol, E. D.; El-Aasser, M. S. Competitive Adsorption of the Anionic Surfactant SLS and the Nonionic Surfactant Triton X-405 on Polystyrene Latex Particles. Langmuir 2000, 16, 7905−7913. (28) Lessard, R. R.; DeMarco, G. The Significance of Oil Spill Dispersants. Spill Sci. Technol. Bull. 2000, 6, 59−68. (29) Campo, P.; Venosa, A. D.; Suidan, M. T. Biodegradability of Corexit 9500 and Dispersed South Louisiana Crude Oil at 5 and 25 °C. Environ. Sci. Technol. 2013, 47, 1960−1967. (30) Place, B.; Anderson, B.; Mekebri, A.; Furlong, E. T.; Gray, J. L.; Tjeerdema, R.; Field, J. A Role for Analytical Chemistry in Advancing our Understanding of the Occurrence, Fate, and Effects of Corexit Oil Dispersants. Environ. Sci. Technol. 2010, 44, 6016−6018. (31) Li, Z. X.; Lu, J. R.; Fragneto, G.; Thomas, R. K.; Binks, B. P.; Fletcher, P. D. I.; Penfold, J. Neutron reflectivity studies of AerosolOT monolayers adsorbed at the oil/water, air/water and hydrophobic solid/wate interfaces. Colloids Surf., A 1998, 135, 277−281. (32) Li, Z. X.; Lu, J. R.; Thomas, R. K.; Weller, A.; Penfold, J.; Webster, J. R. P.; Sivia, D. S.; Rennie, A. R. Conformal Roughness in the Adsorbed Lamellar Phase of Aerosol-OT at the Air−Water and Liquid−Solid Interfaces. Langmuir 2001, 17, 5858−5864. (33) Seo, D.-J.; Fujita, H.; Sakoda, A. Effects of a non-ionic surfactant, Tween 20, on adsorption/desorption of saccharification enzymes onto/from lignocelluloses and saccharification rate. Adsorption 2011, 17, 813−822. (34) Gutig, C.; Grady, B. P.; Striolo, A. Experimental Studies on the Adsorption of Two Surfactants on Solid−Aqueous Interfaces: Adsorption Isotherms and Kinetics. Langmuir 2008, 24, 4806−4816. (35) Binks, B. P., Ed. Modern Characterization Methods of Surfactant Systems; Marcel Dekker: New York, 1999. (36) Merta, J.; Tammelin, T.; Stenius, P. Adsorption of complexes formed by cationic starch and anionic surfactants on quartz studied by QCM-D. Colloids Surf., A 2004, 250, 103−114. (37) Vogt, B. D.; Lin, E. K.; Wu, W.-l.; White, C. C. Effect of Film Thickness on the Validity of the Sauerbrey Equation for Hydrated Polyelectrolyte Films. J. Phys. Chem. B 2004, 108, 12685−12690. (38) Stålgren, J. J. R.; Eriksson, J.; Boschkova, K. A Comparative Study of Surfactant Adsorption on Model Surfaces Using the Quartz Crystal Microbalance and the Ellipsometer. J. Colloid Interface Sci. 2002, 253, 190−195. (39) Bordes, R.; Höök, F. Separation of Bulk Effects and Bound Mass during Adsorption of Surfactants Probed by Quartz Crystal Microbalance with Dissipation: Insight into Data Interpretation. Anal. Chem. 2010, 82, 9116−9121. (40) Q-Sense. Cleaning and Immobilization Protocols. (41) Fragneto, G.; Li, Z. X.; Thomas, R. K.; Rennie, A. R.; Penfold, J. A Neutron Reflectivity Study of the Adsorption of Aerosol-OT on Self-Assembled Monolayers on Silicon. J. Colloid Interface Sci. 1996, 178, 531−537. (42) Grillo, I.; Penfold, J. Self-Assembly of Mixed Anionic and Nonionic Surfactants in Aqueous Solution. Langmuir 2011, 27, 7453− 7463. (43) Mahmood, M. E.; Al-Koofee, D. A. F. Effect of Temperature Changes on Critical Micelle Concentration for Tween Series Surfactant. Global J. Sci. Front. Res. Chem. 2013, 13, 2268−2274. (44) Giles, C. H.; Smith, D.; Huitson, A. A general treatment and classification of the solute adsorption isotherm. I. Theoretical. J. Colloid Interface Sci. 1974, 47, 755−765. (45) Esumi, K.; Maedomari, N.; Torigoe, K. Adsolubilization of 2Naphthol by Binary Mixtures of Cationic and Nonionic Surfactants on Silica. Langmuir 2001, 17, 7350−7354. (46) Wu, S.; Shi, L.; Garfield, L. B.; Tabor, R. F.; Striolo, A.; Grady, B. P. Influence of Surface Roughness on Cetyltrimethylammonium Bromide Adsorption from Aqueous Solution. Langmuir 2011, 27, 6091−6098. 11038

dx.doi.org/10.1021/la502513p | Langmuir 2014, 30, 11031−11039

Langmuir

Article

(47) Shi, L.; Ghezzi, M.; Caminati, G.; Lo Nostro, P.; Grady, B. P.; Striolo, A. Adsorption Isotherms of Aqueous C12E6 and Cetyltrimethylammonium Bromide Surfactants on Solid Surfaces in the Presence of Low Molecular Weight Coadsorbents. Langmuir 2009, 25, 5536− 5544. (48) Suttipong, M.; Grady, B. P.; Striolo, A. Self-assembled surfactants on patterned surfaces: confinement and cooperative effects on aggregate morphology. Phys. Chem. Chem. Phys. 2014, 16, 16388− 16398. (49) Martín-Rodríguez, A.; Cabrerizo-Vílchez, M. A.; HidalgoÁ lvarez, R. A. Comparative Study on the Adsorption of Triton X100 and Tween 20 onto Latexes with Different Interfacial Properties. J. Colloid Interface Sci. 1997, 187, 139−147. (50) Shen, L.; Guo, A.; Zhu, X. Tween surfactants: Adsorption, selforganization, and protein resistance. Surf. Sci. 2011, 605, 494−499. (51) Somasundaran, P.; Fu, E.; Xu, Q. Coadsorption of anionic and nonionic surfactant mixtures at the alumina-water interface. Langmuir 1992, 8, 1065−1069. (52) Atkin, R.; Craig, V. S. J.; Wanless, E. J.; Biggs, S. Mechanism of cationic surfactant adsorption at the solid−aqueous interface. Adv. Colloid Interface Sci. 2003, 103, 219−304. (53) Kou, J.; T, D.; Sun, T.; Xu, G. Application of the quartz crystal microbalance with dissipation method to a study of oleate adsorption onto a hydroxyapatite surface. Miner. Metal. Process. 2012, 29, 47−55. (54) Biswas, S. C.; Chattoraj, D. K. Kinetics of Adsorption of Cationic Surfactants at Silica-Water Interface. J. Colloid Interface Sci. 1998, 205, 12−20. (55) Atkin, R.; Craig, V. S. J.; Biggs, S. Adsorption Kinetics and Structural Arrangements of Cationic Surfactants on Silica Surfaces. Langmuir 2000, 16, 9374−9380. (56) Atkin, R.; Craig, V. S. J.; Biggs, S. Adsorption Kinetics and Structural Arrangements of Cetylpyridinium Bromide at the Silica− Aqueous Interface. Langmuir 2001, 17, 6155−6163. (57) Pagac, E. S.; Prieve, D. C.; Tilton, R. D. Kinetics and Mechanism of Cationic Surfactant Adsorption and Coadsorption with Cationic Polyelectrolytes at the Silica−Water Interface. Langmuir 1998, 14, 2333−2342. (58) Paria, S.; Khilar, K. C. A review on experimental studies of surfactant adsorption at the hydrophilic solid−water interface. Adv. Colloid Interface Sci. 2004, 110, 75−95. (59) Howard, S. C.; Craig, V. S. J. Very slow surfactant adsorption at the solid-liquid interface is due to long lived surface aggregates. Soft Matter 2009, 5, 3061−3069. (60) Woods, D. A.; Petkov, J.; Bain, C. D. Surfactant Adsorption Kinetics by Total Internal Reflection Raman Spectroscopy. 1. Pure Surfactants on Silica. J. Phys. Chem. B 2011, 115, 7341−7352.

11039

dx.doi.org/10.1021/la502513p | Langmuir 2014, 30, 11031−11039