Monomer−Aggregate Exchange Rates in Dialkyl Chain Cationic

Jan 27, 2009 - ... Laboratory, Port Sunlight, Quarry Road East, Bebington, Wirral, U.K., ... Theoretical Chemistry Laboratory, Oxford UniVersity, Sout...
0 downloads 0 Views 651KB Size
Langmuir 2009, 25, 2661-2666

2661

Monomer-Aggregate Exchange Rates in Dialkyl Chain Cationic-Nonionic Surfactant Mixtures I. Tucker,*,† J. Penfold,‡,§ R. K. Thomas,§ and I. Grillo| UnileVer Research and DeVelopment Laboratory, Port Sunlight, Quarry Road East, Bebington, Wirral, U.K., ISIS Facility, STFC, Rutherford Appleton Laboratory, Chilton, Didcot, Oxon, U.K., Physical and Theoretical Chemistry Laboratory, Oxford UniVersity, South Parks Road, Oxford, U.K., and Institute Laue LangeVin, 6 rue Jules Horowitz, BP 156, 38042 Grenoble, Cedex 9, France ReceiVed October 8, 2008. ReVised Manuscript ReceiVed December 7, 2008 The monomer-aggregate exchange rate in self-assembled dialkyl chain cationic-nonionic mixed surfactant aggregates has been studied using small-angle neutron scattering, SANS, and a stopped-flow apparatus. SANS was used to follow the evolution of the structure with time of an equimolar mixture of the dialkyl chain cationic surfactant dihexadecyl dimethyl ammonium bromide, DHDAB, in D2O with the nonionic surfactant dodecaethylene monododecyl ether, C12E12, in D2O at a solution concentration of 1.5 mM. With increasing time, the bilamellar vesicle structure, blv, of DHDAB and the globular micellar structure, L1, of C12E12 evolved to a lamellar (Lβ or LR)/micellar (L1) coexistence. Measurements were made for the isotopically labeled combinations of hydrogeneous DHDAB (h-DHDAB) and alkyl chain deuterium-labeled C12E12 (d-C12E12) in D2O such that the lamellar contribution is the predominantly visible contribution to the scattering. From the variation (decrease) in the scattering intensity with time (measured at a scattering vector of ∼0.014 Å-1), a characteristic time was measured at 32 °C (T < Lβ/LR transition temperature) and at 46 °C (T > Lβ/LR). The characteristic time was ∼130 min and a few seconds respectively, indicating a dramatic change in the monomer/aggregate exchange rate between the solid-like Lβ and fluid-like LR phases. The characteristic time of ∼130 min in the Lβ phase is indicative of a slow monomer-aggregate exchange rate and is consistent with the slow kinetics of adsorption of DHDAB and DHDAB/nonionic surfactant mixtures observed at the air-water interface. This slow adsorption kinetics was assumed to arise from near-surface depletion effects associated with slow monomer/aggregate exchange rates, and these results support and reinforce that hypothesis.

Introduction Dialkyl chain cationic surfactants are a major constituent in a wide range of consumer products, such as hair and clothes care products and other associated lubricants. They are usually formulated with a range of other surfactants or cosurfactants and, notably, the polyoxyethylene glycol nonionic surfactants.4 Recent studies1-6 have shown that in solution the self-assembled structures evolve from predominantly planar structures (bilamellar or multilamellar vesicles or lamellar fragments, depending upon concentration, composition, and the exact nature of the nonionic cosurfactant) for solutions rich in the dialkyl chain cationic surfactant to more highly curved structures such as globular micelles for solutions rich in the nonionic cosurfactant. At intermediate compositions, lamellar/micellar coexistence exists. The resulting phase diagrams, mapped out in some detail for DHDAB/C12E3,3 DHDAB/C12E6, and DHDAB/C12E12,2 are complex and depend upon the relative curvatures associated with the aggregates of the pure components. The associated adsorption * Corresponding author. E-mail: [email protected]. † Unilever Research and Development Laboratory. ‡ Rutherford Appleton Laboratory. § Oxford University. | Institute Laue Langevin. (1) Tucker, I.; Penfold, J.; Thomas, R. K.; Barker, J. G.; Mildner, D. F. R. Langmuir 2008, 24, 6509. (2) Tucker, I.; Penfold, J.; Thomas, R. K.; Grillo, I.; Barker, J. G.; Mildner, D. F. R. Langmuir 2008, 24, 7674. (3) Tucker, I.; Penfold, J.; Thomas, R. K.; Grillo, I.; Barker, J. G.; Mildner, D. F. R. Langmuir 2008, 24, 10089. (4) Penfold, J.; Staples, E.; Ugazio, S.; Tucker, I.; Soubiran, L.; Hubbard, J.; Noro, M.; O’Malley, M.; Ferrante, A.; Ford, G.; Buron, H. J. Phys. Chem. B 2005, 109, 276. (5) Penfold, J.; Staples, E.; Tucker, I.; Thomas, R. K. Langmuir 2004, 20, 1269. (6) Penfold, J.; Tucker, I.; Staples, E.; Thomas, R. K. Langmuir 2004, 20, 2265.

at the air-water interface is also complex and illustrates the strong interplay between the surface and solution behaviors. For DHDAB/C12E3,3 the adsorption is close to ideal, whereas for DHDAB/C12E6 and DHDAB/C12E127 there is a marked departure from ideal mixing, and there is an abrupt change in the surface composition from one dominated by DHDAB as the solution becomes richer in the nonionic surfactant. A particular feature of the adsorption was that the kinetics of adsorption was relatively slow, >50-200 min to reach equilibrium. This is observed for a range of other systems8,9 and would normally be associated with a low critical micellar concentration, cmc, and hence a low monomer concentration in equilibrium with the surface. However, the measured cmc values1-3 are not consistent with that explanation for these surfactants: DHDAB, C12E3, C12E6, C12E12, and their mixtures. It was argued that the slow kinetics of adsorption was associated with near-surface depletion effects due to unusually slow monomer/aggregate exchange rates in these mixtures. However, there are no reported measurements for the exchange rates in these particular systems, and the focus of this article is to address that issue. Here we report measurements using a combination of SANS and stopped flow to evaluate the monomer/aggregate exchange rates in DHDAB/C12E12 surfactant mixtures. The choice of a nonionic compound, that is, C12E12 in preference to C12E6 or C12E3, was intended to provide optimal experimental conditions. Using C12E6 would introduce complications due to working at temperatures close to the cloud point of nonionic C12E6 in the LR phase. The DHDAB/C12E3 mixture (7) Tucker, I.; Penfold, J.; Thomas, R. K.; Tildesley, D. J. Langmuir 2008 ASAP Article; DOI: 10.1021/la801302z. (8) Bowers, J.; Danks, M. J.; Bruce, D. W.; Webster, J. R. P. Langmuir 2003, 19, 299. (9) Penfold, J.; Staples, E.; Tucker, I. J. Phys. Chem. B 2002, 106, 8891.

10.1021/la803329a CCC: $40.75  2009 American Chemical Society Published on Web 01/27/2009

2662 Langmuir, Vol. 25, No. 5, 2009

would not provide such a clear structural change on mixing because both DHDAB and C12E3 are vesicular. The dynamic nature of surfactant aggregates is well established and is characterized by two different time scales: τ1, the lifetime of a surfactant monomer in the aggregates, and τ2, the lifetime of the aggregate. In surfactant micelles, τ1 is typically ∼1 µs, and τ2 varies from milliseconds to seconds. The major results from a wide range of studies have been reviewed by Aniansson et al.10 These micelle relaxation processes have been probed by a range of different relaxation methods, such as T-jump, P-jump, and stopped flow, and the temporal changes have been monitored by different optical probe methods, such as fluorescence. More recently, stopped flow has been combined with scattering methods, such as SANS and small-angle X-ray scattering, SAXS, to follow in more detail the structural changes associated with the relaxation processes. Eastoe et al.11 combined stopped flow and SAXS to study the breakdown of micelles of the fluorocarbon surfactants, difluorocarbon sulfosuccinate, and sodium perfluorooctanoate, NaPFO, by dilution from above to below the cmc. The measured micelle lifetimes, τ2, were ∼10 and 13 s, respectively, which are larger than for a typical hydrocarbon surfactant. (τ2 is ∼0.34 s for sodium hexadecyl sulfate.10) In this case, the longer lifetimes were associated with the more hydrophobic fluorinated alkyl chains. The other notable examples in the recent literature all involved the transition from micellar to vesicle phases by the addition of electrolyte,12,13 by dilution,14,15 or by catanionic surfactant mixing.16-18 Grillo et al.12,13 used stopped flow and SANS to follow the electrolyte-induced transition of AOT from globular micelles to vesicles. An initial fast (200 s). Egelhaaf et al.14 used stopped flow and SANS to demonstrate the micelle-tovesicle transition on the dilution of lecithin/bile salt mixtures. In a very detailed study, they were able to identify the intermediate structures that were formed. The initial small globular micelles transferred to vesicles via an intermediate elongated flexible wormlike structure. The wormlike structure-to-vesicle transition is mediated by the formation of disklike structures. It was found that the micelle-to-intermediate transition is relatively fast and the intermediate-to-vesicle transition is relatively slow (∼150 min). Leng et al.15 have provided a theoretical treatment and justification of these structural changes. These recent studies clearly demonstrated the potential of the combination of scattering (SAS) and stopped flow to study self-assembled aggregate kinetics. (10) Anainsson, E. G.; Wall, S. N.; Almgren, M.; Hoffmann, H.; Keilmann, I.; Ulbricht, W.; Zana, R.; Lang, J.; Tondre, C. J. Phys. Chem. 1976, 80, 905. (11) Eastoe, J.; Dalton, J. S.; Downer, A.; Jones, G.; Clarke, D. Langmuir 1998, 14, 1937. (12) Grillo, I.; Kats, E. I.; Muratov, A. R. Langmuir 2003, 19, 4573. (13) Grillo, I. J Phys IV 2005, 130, 75. (14) Egelhaaf, S. U.; Schurtenberger, P. Phys. ReV. Lett. 1999, 82, 2804. (15) Leng, J.; Egelhaff, S. U.; Cates, M. E. Biophys. J. 2003, 85, 1624. (16) Schmolzer, S. T.; Grabner, D.; Gradzielski, M.; Narayanan, T. Phys. ReV. Lett. 2002, 88, 258301. (17) Weiss, T. H.; Narayanan, T.; Wolf, C.; Gradzielski, M.; Pannine, P.; Finet, S.; Helsby, W. I. Phys. ReV. Lett. 2005, 94, 038303. (18) Weiss, T. M.; Narayanan, T.; Gradzielski, M. Langmuir 2008, 24, 3759.

Tucker et al.

Figure 1. Evolution of surface excess with time for 3 × 10-4 M d-DHDAB in null reflecting water at 30 °C. The inset is a fit to the kinetics of adsorption for the data plotted as in eq 1.

Experimental Details The SANS measurements were made on the D22 diffractometer at the ILL, France,19 in the scattering vector range of 0.014-0.25 Å-1 (where the scattering vector is defined as Q ) 4π/λ sin θ/2 and θ is the scattering angle). The measurements were made with a neutron wavelength, λ, of 8 Å, a ∆λ/λ of 10%, a sample-to-detector distance of 3 m, 5.6 m incident collimation, and a detector offset of 400 mm to maximize the available Q range. A 7 × 10 mm2 sample aperture was used, and the sample path length in the Biologic Co SFM-3 stopped-flow apparatus was 1 mm. The SANS data were corrected for background scatter and detector response and converted to an absolute scattering cross-section, dσ/dΩ (in cm-1), using standard procedures.20 The measurements were made at a surfactant concentration of 1.5 mM at 32 and 46 °C (above the Krafft point and below and above the LR/Lβ transition temperature of the system). The two 1.5 mM solutions mixed in the stopped-flow apparatus were 1.5 mM h-DHDAB/D2O and 1.5 mM d-C12E12/D2O at an equimolar mixing ratio. Measurements were made with the following time resolutions: 0.2 to 30 s (log time scale), 30 s measurements intervals for 30 s to 10 min, 60 s measurement intervals for 10 to 30 min, 5 min measurement intervals for 30 to 130 min, and 10 min measurement intervals for 130 min to >400 min (∼7 h). h-DHDAB was obtained from Fluka and was recrystallized from ethyl acetate before use. d-C12E12 was synthesized by R. K. Thomas, Oxford University, according to previously published procedures,21 and its purity was verified by NMR and surface tension measurements. The time dependence of the evolution of the bilamellar vesicle structure for the pure DHDAB solution and the micellar structure of pure C12E12 in a mixed lamellar/micellar phase2 was evaluated from the normalized scattering intensity measured at Q ) 0.014 Å-1.

Results and Discussion Adsorption Kinetics. In the study of DHDAB and DHDAB/ nonionic surfactant mixture adsorption at the air-water interface1,5-7 using mixtures of protonated and deuterated surfactants in null reflecting water (NRW, where the solvent is neutron refractive index matched to air), we have previously reported the observation of relatively slow kinetics of adsorption. This is illustrated in Figure 1 for DHDAB, where the adsorption at the air-water interface is plotted as a function of time.1 (19) Neutron beam facilities at the high flux reactor available to users, ILL, Grenoble, France, 1994. (20) RE Ghosh, SU Egelhaaf, AR Rennie, ILL Internal Report, 1998, ILL98GH14T. (21) Lu, J. R.; Su, T. J.; Li, Z. X.; Thomas, R. K.; Staples, E. J.; Tucker, I.; Penfold, J. J. Phys. Chem B 1997, 101, 10332.

Monomer-Aggregate Exchange Rates

Langmuir, Vol. 25, No. 5, 2009 2663

Figure 3. Variation in the adsorbed amount, Γ, (× 10-10 mol cm-2) with time for 1.3 × 10-3 M 20/80 DHDAB/C12E12.

Figure 2. Variation in adsorbed amount, Γ, (× 10-10 mol cm-2) with time for (a) 1.3 × 10-4 M 15/85 DHDAB/C12E6 and (b) 1.3 mM 23/77 DHDAB/C12E6: (2) DHDAB and (b) C12E6.

Assuming a Langmuir-type isotherm, the characteristic time, τ, for adsorption was evaluated using

(

1-

)

t Γ ) e- τ Γs

(1)

where τ is the characteristic time for adsorption and Γ and Γs are the adsorption and adsorption at saturation, which provides a characteristic time of ∼50 min. Similarly slow kinetics of adsorption were reported for DHDAB/C12E6 mixtures,5 where the evolution of the surface composition occurred over a similar time scale, as illustrated in Figure 2 for a 1.3 × 10-4 M 15/85 mol/mol ratio of DHDAB/C12E6 and a 1.3 mM 23/77 mol/mol ratio of DHDAB/C12E6. The characteristic times associated with that data were 210 and 110 min, respectively. A similar evolution in the surface composition was also observed for DHDAB/C12E12 mixtures22 and is illustrated here in Figure 3 for a 1.3 mM 20/80 mol/mol ratio of DHDAB/C12E12. On the time scale of the neutron reflectivity measurements used to obtain the adsorption data, the adsorption is usually considered to be relatively instantaneous. Such slow adsorption kinetics is usually associated with low surfactant cmc values and the corresponding low surfactant monomer concentrations in equilibrium with the surface and has been reported in related systems.8,9 However, the cmc values for DHDAB1 and for the nonionic surfactants, C12E6 and C12E12,23 are not sufficiently low to account for the slow adsorption kinetics observed. In the more (22) Tucker, I. M. DPhil Thesis, Oxford University, Oxford, U.K., 2007. (23) Van Os, N. M.; Haak, J. R.; Rupert, L. A. M. Physicochemical Properties of Selected Anionic, Cationic and Nonionic Surfactants; Elsevier: New York, 1993.

recent papers on DHDAB adsorption1 and DHDAB/nonionic surfactant adsorption,7 it was proposed that the slow kinetics was due to depletion effects in the near-surface region resulting from the unusually slow monomer-aggregate exchange rates in the bulk solution in equilibrium with the surface. These slow exchange rates should hence be manifested in the form of correspondingly long aggregate lifetimes. Monomer-Aggregate Exchange Rates. To estimate the monomer-aggregate exchange rates in these systems, we have initiated an equimolar mixture of DHDAB and C12E12 at 1.5 mM and followed the evolution in the mixed microstructure with time using SANS. For the isotopically labeled combinations of h-DHDAB/D2O and d-C12E12/D2O, the scattering will be dominated by the DHDAB contribution to the scattering. Hence, it is expected that the initial bilamellar vesicle structure of h-DHDAB/D2O and the small globular micelles of d-C12E12/ D2O will evolve into a mixed Lβ(LR)/L1 phase. The scattering behavior of the pure components and their associated mixtures has been previously characterized by SANS,1,2 and the data for a surfactant concentration of 1.5 mM for the DHDAB/C12E12 mixture from ref 2 is reproduced in Figure 4. The evolution in time of the scattering measured here for the equimolar h-DHDAB/d-C12E12/D2O mixture at 1.5 mM is shown in Figure 5. The scattering in Figure 5 is broadly consistent with that reported earlier1,2 and shown in Figure 4 but measured over a more limited Q range. However, in detail the profiles are different because of the different contrasts used. The data in Figure 4 were measured for the isotopic combination of h-DHDAB/h-C12E12/ D2O, and the data in Figure 5 were measured for the isotopic combination of h-DHDAB/d-C12E12/D2O. In the latter data, the contribution to the scattering from C12E12 will be substantially reduced because it is effectively matched to the D2O solvent. Hence, for example, at early times the pure C12E12 micellar component should be essentially invisible, and only the pure DHDAB bilamellar vesicle component should be visible. With mixing, the intensity associated with the lamellar component will become less visible, and that associated with the micellar component will become more visible. Hence, the isotopic combination used here and the Q range probed provide an optimization of the sensitivity to the changes in the microstructure as the mixing between the two pure components evolves. This experiment was repeated several times, and the results were reproducible.

2664 Langmuir, Vol. 25, No. 5, 2009

Tucker et al.

Figure 4. Scattering data for 1.5 mM DHDAB/C12E12 mixtures at mole ratios of 100/0 (black) 90/10 (red), 80/20 (green), 70/30 (yellow), 60/40 (blue), 50/50 (white), 40/60 (pink), 30/70 (cyan), 20/80 (gray), 10/90 (olive), and 100/0 (purple). The scattering data are each displaced by a multiple of 4 (40/60 × 16) to avoid overlay with the 100/0 shown on the absolute scale. The errors are smaller than the data points used.

Figure 5. Time-dependent SANS scattering intensity, plotted as Q2I(Q), for 1.5 mM mixture of h-DHDAB and d-C12E12 in D2O, for (black) after 30 min, (red) 75 min, (blue) 155 min, (green) 215 min, and (light blue) 275 min. The inset is the low-angle region plotted on an expanded (linear) scale.

In this experiment, by mixing two solutions, each consisting of a pure single vesicular/micellar microstructure, the anticipated scattering at t ) 0 would be a mixture of the two contributions. As the system evolves into a different equilibrium mixed microstructure (as described earlier), subsequent changes as a result of intermixing between the two microstructures would simultaneously alter the power in the scattering of both components as the number density, size (and hence volume), and contrast would all be simultaneously varying. However, the use of deuterium labeling rendered the nonionic dominated microstructure invisible to the neutrons, and scattering would arise only from the microstructure dominated by the protonated component, in this case, the vesicular phase. Therefore, the evolution of the microstructure with time, following initial mixing, has been quantitatively evaluated by using the variation in the scattering intensity at Q ≈ 0.014 Å-1, where the scattering is dominated by the lamellar component and where the change in the intensity (reduction) with time will be directly related to the change in the structure from the initial pure vesicle and micellar phases to a mixed lamellar/L1 phase. The resulting variations in

the low-Q (Q ) 0.014Å-1) scattering intensity with time, measured for 1.5 mM h-DHDAB/d-C12E12/D2O at an equimolar composition, at 32 and 46 °C are shown in Figure 6. In our previous work,1,2 we postulated that the process of DHDAB adsorption at the air-water interface was diffusionlimited and that the supply of monomer was limited by the monomer diffusion rate into and out of the coexisting vesicles, and the change in adsorption with time can be described by a behavior represented by eq 1. The variation in intensity from the stopped-flow SANS data shows a similar functional form; consequently, the data in Figure 6 were replotted in a form equivalent to eq 1 (where Γ and Γs are replaced by I and I0) to enable the characteristic time, τ, associated with the intermixing to be evaluated. The variation of the exponent with time for the two data sets thus treated is plotted in Figure 7 for the two different temperatures. For the low-temperature data (T ) 32 °C), this analysis yields a characteristic time, τ, on the order of 130 min. The time scale of intermixing in the low-temperature state is comparable to the time scale on which the surface reaches equilibrium. Although this is subject to several assumptions, it is nevertheless evident that the time scale associated with the intermixing, related to the monomer-aggregate exchange rate, is relatively long and comparable to the time scale for adsorption at the air-water interface. In the high-temperature state, which is well above the LR/Lβ transition temperature, the time scales are very rapid and are comparable to time scales recorded at lower temperatures in more labile systems.10 Figure 7 shows that it is not possible to describe the intermixing using a single characteristic time scale. There is a markedly different rate of exchange depending upon whether the dialkyl chains of the DHDAB are in a solidlike or fluid phase. The variation in the scattering intensity (plotted as ln(1 (I/I0)) vs time) at the higher temperature of 46 °C is more complex than a simple linear dependence (which is consistent with a single exponential) observed at the lower temperature of 32 °C. However, the time scales are very short, spanning a range centered around 100-150 ms with a distribution width about 50 ms, and skew strongly toward short times. These are comparable to the time scales recorded at lower temperatures in more labile systems.10 The form of the time dependence

Monomer-Aggregate Exchange Rates

Langmuir, Vol. 25, No. 5, 2009 2665

Figure 6. Variation in scattering intensity (measured at Q ) 0.014 Å-1) for a 1.5 mM equimolar mixture of h-DHDAB/D2O and d-C12E12/D2O (blue) at 32 °C (blue) and at 46 °C (red). The inset is the 46 °C data plotted on a logarithmic scale covering the first 35 s only.

Figure 7. Variation in scattering intensity (measured at Q ) 0.014 Å-1) with time for a 1.5 mM equimolar mixture of h-DHDAB/D2O and d-C12E12/D2O (left) at 32 °C and (right) at 46 °C plotted in a form similar to eq 1, ln(1 - (I/I0)). The solid line in the left plot is a least-squares fit to eq 1.

shown in Figure 7b would imply a range of different time scales, but the origin of these is as yet uncertain. Schmolzer et al.,16 for example, observed different time scales in micelle-tovesicle transitions in cationic/anionic surfactant mixtures. These were attributed to initially fast micellar dissolution, followed by slower vesicle growth. However, although it is difficult at this low surfactant concentration (1.5 mM) to distinguish subtle structural changes at these very short times, a similar mechanism is unlikely to be the explanation here. A key aspect lies in the comparison of the LR and Lβ data. In the Lβ phase, a single time scale described the kinetics, whereas in the LR phase, multiple time scales exist. We postulate that in the Lβ phase the dissolution kinetics is so slow that it is essentially independent of the DHDAB aggregate size. In the LR phase, the corresponding dissolution is rapid and so will be strongly dependent upon the vesicle size. It is known that the DHDAB vesicles are highly polydisperse,1 and hence the broad distribution of (short) time scales is observed. Although the slow adsorption kinetics was originally observed and quantified for both the DHDAB/C12E6 and DHDAB/C12E12 mixtures,5,7 we have considered only the solution intermixing in DHDAB/C12E12 mixtures. As discussed in the Introduction, it was thought to be important to observe the changes associated with the LR/Lβ transition, and measurements with C12E6 were not

made in order to avoid the complicating effects due to the proximity of the C12E6 cloud point. Micelle-monomer exchange rates are normally relatively fast, and micelle lifetimes of