Nonionic Surfactant

Sep 3, 2005 - ISIS Facility, CCLRC, Rutherford Appleton Laboratory, Chilton, Didcot, OXON, UK, and UnileVer Research and DeVelopment, Port Sunlight, ...
1 downloads 0 Views 363KB Size
J. Phys. Chem. B 2005, 109, 18107-18116

18107

The Microstructure of Di-alkyl Chain Cationic/Nonionic Surfactant Mixtures: Observation of Coexisting Lamellar and Micellar Phases and Depletion Induced Phase Separation J. Penfold,*,† E. Staples,§ S. Ugazio,‡,| I. Tucker,‡ L. Soubiran,‡ J. Hubbard,‡ M. Noro,‡ B. O’Malley,‡ A. Ferrante,‡ G. Ford,‡ and H. Buron‡,⊥ ISIS Facility, CCLRC, Rutherford Appleton Laboratory, Chilton, Didcot, OXON, UK, and UnileVer Research and DeVelopment, Port Sunlight, Quarry Road East, Bebington, Wirral, UK ReceiVed: January 6, 2005; In Final Form: April 6, 2005

The evolution of the microstructure and composition occurring in the aqueous solutions of di-alkyl chain cationic/nonionic surfactant mixtures has been studied in detail using small angle neutron scattering, SANS. For all the systems studied we observe an evolution from a predominantly lamellar phase, for solutions rich in di-alkyl chain cationic surfactant, to mixed cationic/nonionic micelles, for solutions rich in the nonionic surfactant. At intermediate solution compositions there is a region of coexistence of lamellar and micellar phases, where the relative amounts change with solution composition. A number of different di-alkyl chain cationic surfactants, DHDAB, 2HT, DHTAC, DHTA methyl sulfate, and DISDA methyl sulfate, and nonionic surfactants, C12E12 and C12E23, are investigated. For these systems the differences in phase behavior is discussed, and for the mixture DHDAB/C12E12 a direct comparison with theoretical predictions of phase behavior is made. It is shown that the phase separation that can occur in these mixed systems is induced by a depletion force arising from the micellar component, and that the size and volume fraction of the micelles are critical factors.

Introduction The behavior of mixed surfactant solutions is of widespread interest because of the extensive use of mixed surfactants in a wide range of industrial, technological, and domestic applications.1 A key element in understanding their behavior is a detailed knowledge of the solution microstructure and associated phase behavior. In applications such as fabric conditioning, hair shampoos, and shower gels, di-alkyl chain cationic/nonionic surfactant mixtures are of particular interest and importance. The general phase behaviors of such formulations have been established.2 They are predominantly in the form of lamellar fragments, and with increasing nonionic surfactant content there is an evolution in the microstructure toward a mixed surfactant micellar phase and a region of a mixed lamellar and micellar microstructure. Theoretical treatments, which are capable of predicting phase behavior and the associated coexistence region, now exist.3,4 Although the detailed phase behavior of the cationic5-7 and nonionic8 components are know, there are no detailed studies on di-alkyl chain cationic-nonionic surfactants mixtures reported in the recent literature. However, there is a rich literature in a related but distinctly different area, in the use of surfactants in biomembrane studies to solubilize membranes by forming mixed micelles with membrane lipids and proteins,9,10 and in the use of longer ethylene oxide chain length nonionic surfactants, such as C12E8, as surfactant. Phase behavior studies11 have determined the extent of the micellar, vesicular, and coexistence regions, and some limited scattering studies12,13 * Corresponding author. † Rutherford Appleton Laboratory. ‡ Unilever Research and Development. § Affiliated to Rutherford Appleton Laboratory, formerly Unilever Research. | Presently affiliated with Rohn & Haas, France. ⊥ Presently affiliated with Dow-Corning, Belgium.

have characterized the general nature of the mixed micelles that are formed. However, there is limited detailed information on the structure of the mixed micelles that are formed, and particularly in the coexistence region of the phase diagram. Small angle neutron scattering, SANS, in combination with hydrogen/deuterium isotopic substitution, is a powerful technique for characterizing the structure and composition of mixed surfactant aggregates in solution.14,15 SANS is used here to characterize the nature of the lamellar and micellar phases and coexistence regions for a range of di-alkyl chain cationic/ nonionic surfactant mixtures. Measurements are reported for the mixtures dihexadecyl dimethylammonium bromide (DHDAB)/ dodecaethylene monododecyl ether, C12E12; dioctadecyl dimethylammonium chloride (2HT)/C12E12; 2,3-di-heptadecyl ester ethoxy-n-propyl-1, 1,1-trimethylammonium chloride (DHTAC)/ C12E12 (C12E23); DHTA methyl sulfate/C12E12, and distearyl oyloxy dimethylammonium methyl sulfate (DISDA) methyl sulfate/C12E12. In many colloidal mixtures it is well established that flocculation or phase separation occurs due to an attraction or depletion interaction. This was shown to arise from the addition of a nonadsorbing polymer to a dispersion of colloidal particles16,17 in binary mixtures of large and small colloidal spheres,16,18 and in mixtures of colloidal spheres and rods.19,20 It is demonstrated here that the charged mixed surfactant micelles that are in coexistence with lamellar fragments cause instability and ultimately phase separation in a way that can be explained by such a depletion interaction, which is sensitive to the size and volume fraction of the micellar component. Experimental Details The SANS measurements were made on the LOQ diffractometer21 at the ISIS pulsed neutron source at the Rutherford

10.1021/jp0500788 CCC: $30.25 © 2005 American Chemical Society Published on Web 09/03/2005

18108 J. Phys. Chem. B, Vol. 109, No. 38, 2005 Appleton Laboratory. The measurements were made using the white beam time-of-flight method, in the scattering vector range of 0.008 to 0.25 Å-1. The samples were contained in Hellma 1 mm path length quartz spectrophotometer cells and maintained at a temperature of 30 °C. The data were corrected for background scattering, detector response, and the spectral distribution of the incident neutron beam, and converted to an absolute scattering cross-section (I(Q) in cm-1) using standard procedures.22 Measurements were made for the combinations DHDAB/ C12E12, 2HT/C12E12, 2HT/C12E23, DHTAC/C12E12, DHTAC/ C12E23, DISDA methyl sulfate/C12E12, and DHTA methyl sulfate/C12E23. All the measurements were made in D2O to provide the optimal “contrast” between the micellar and lamellar aggregates and the solvent. The charged micelles in the mixed microstructure present in these systems can promote instability and phase separation by depletion into lamellar and micellar rich phases, as discussed in detail later. Ultracentrifugation at 13 000 rpm for 30 min is used to efficiently accelerate this phase separation. The resulting upper phase is predominantly lamellar and the lower phase micellar, and the two separated phases can then be measured separately and independently. The combinations 2HT (DHTAC)/C12E12 (C12E23) were measured at 5 wt % in D2O for the composition mole ratios of 1:2, 1:1.5, 1:1, 1:0.5, and 1:0.25. For the combinations 2HT/ C12E12 and DHTAC/C12E12 the isotopic combinations h-2HT (DHTAC)/h-C12E12 and h-2HT (DHTAC)/d-C12E12 were measured. These measurements were all made on samples that had not been phase separated. For the combination DHDAB/C12E12, measurements were made at constant surfactant concentrations (equivalent to 6 and 8 wt % at a mole ratio of 1:1) in D2O for the composition mole ratios of 1:1, 1:0.8, 1:0.7, 1:0.6, and 1:0.4, and for the isotopic combinations h-DHDAB/h-C12E12 and h-DHDAB/d-C12E12 (where d-C12E12 refers to the C12 alkyl chain deuterium labeled, and the combinations are abbreviated hh and hd). For these samples the mixed microstructure was separated by ultracentrifugation, and the upper and lower phases were measured separately. The di-alkyl chain cationic surfactants DHTAC and DISDAC were also measured with the methyl sulfate rather than chloride counterion. Measurements were made for DISDA methyl sulfate/C12E12 and DHTA methyl sulfate/C12E23 for the composition mole ratios of 1:2, 1:1.5, 1:1, 1:0.5 and 1:0.25, and for the isotopic combinations h-cationic/h-C12E12 and h-cationic/ d-C12E12 for the DISDA methyl sulfate cationic surfactant. For these samples the upper and lower phases following ultracentrifugation (as described for the DHDAB/C12E12 mixtures) were measured separately. The cationic surfactants featured in this study are shown schematically in Figure 1. Their different behavior originates from the structural organization of their headgroups. DHDAB is the simplest of all the molecules and consists of two 16carbon atom chains covalently bonded to a quaternary ammonium ion. 2HT is the commercially produced version of DHDAB and in terms of molecular structure is identical to DHDAB with the exception that the alkyl chains are a 2:1 mixture of stearyl and cetyl (C18 and C16 respectively) chain lengths and the anion is now a chloride. The DISDAC is a derivative of these molecules but has an ethoxide bond on each carbon chain to facilitate more rapid biodegradation. DHTAC also features this ethoxide bond on the alkyl chains but has a more complex headgroup structure. Of the four molecules

Penfold et al.

Figure 1. Schematic representation of the di-alkyl chain cationic surfactants: (a) 2HT, (b) DISDAC, and (c) DHTAC. For 2HT the C18 chains and for DISDAC and DHTAC the C17 chains are not shown in full. DHDAB is not shown and is similar to 2HT, except with C16 alkyl chains and bromide counterion.

featured, the headgroups are smallest and therefore presumed to possess the highest charge density on the DHDAB and 2HT molecules. The DHTAC headgroup consists of a mixture of permanent dipole and charged moieties, but as these are dispersed over a large volume, the headgroup surface specific charge can be considered to be quite low. By the same arguments, DISDAC is intermediate between DHDAB (2HT) and DHTAC due to the proximity of the ether oxygen links to the other water-soluble portion of the molecule. The protonated nonionic surfactant C12E23, was obtained from Sigma and used without further purification. DHDAB was sourced from Fluka and used as received. Protonated C12E12 was synthesized at Unilever Research and Development, Port Sunlight.23 The di-alkyl chain cationic surfactants, DISDAC, 2HT, and DHTAC, were purified at the same institution. Chain deuterated C12E12 was synthesized by Dr. R. K. Thomas, University of Oxford. Deuterium oxide, D2O, was obtained from Sigma-Aldrich, and high purity water (Elga Ultrapure) was used throughout. All glassware and sample cells were cleaned using alkali detergent (Decon 90), followed by copious washing in

Di-alkyl Chain Cationic/Nonionic Surfactant Mixtures

J. Phys. Chem. B, Vol. 109, No. 38, 2005 18109

high purity water. The samples were all prepared in D2O at a temperature of 70 °C, in the LR phase of the di-alkyl chain cationic surfactant. The solutions were equilibriated in an oven at 60 °C for several hours and cooled to room temperature before use. In SANS, the scattering cross section, or scattered intensity, for colloidal aggregates in solution can be written by the general expression24

I(Q) ) N|

∫V (Fp(r) - Fs) exp iQ‚rd3r|2

(1)

where Fp and Fs are the aggregate and solvent scattering length densities and N is the number of aggregates per unit volume. In the micellar phase, the micelle structure is determined by analyzing the scattering data using a standard and wellestablished model for globular micelles.19 For a solution of globular polydisperse interacting particles (micelles), the scattered intensity can be written, in the “decoupling approximation”24 as

I(Q) ) N[S(Q)|〈F(Q)〉Q|2 + 〈|F(Q)|2〉Q - |〈F(Q)〉Q|2]2

(2)

where the averages denoted by 〈 〉Q are averages over particles size and orientation, N is the particle number density, S(Q) the structure factor, and F(Q) the form factor. The micelle structure (form factor) is modeled using a standard “core and shell” model,24 where the form factor is

F(Q) ) V1(F1 - F2)F0(QR1) + V2(F2 - Fs)F0(QR2) (3) and Vi ) 4πR3i /3, F0(QRi) ) 3j1(QRi)/(QRi) ) 3[sin (QRi) QRicos(QRi)]/(QRi)3, F1, F2, and Fs are the scattering length densities of the micelle core and shell and of the solvent, and j1(QRi) is a first-order spherical Bessel function. The decoupling approximation assumes that there is no correlation between position and orientation and has been extensively used in the analysis of SANS data for globular micelles and colloids. The micelle core + shell form factor24 comprises an inner core made up of the alkyl chains only and is constrained to space fill a volume limited by a radius, R1, the fully extended chain length of the surfactant, lc, (in this case the length lc is taken as the largest of the C16, C17, or C18 of the di-alkyl chain cationic surfactant and the C12 of the nonionic surfactant in the mixture). For larger aggregation numbers, ν, volumes greater than that defined by R1 (as is found in this study) are accommodated by a prolate elliptical distortion with dimensions R1, R1, eR1 (where e is the elliptical ratio). The outer shell, of dimensions R2, R2, eR2, contains headgroups and the corresponding hydration. The volume occupied by the relatively bulky headgroups (especially the ethylene oxide chains and their associated hydration) ensures a significant contribution to the scattering from the outer shell, and a uniform globular structure rather than the core + shell model is not an adequate description. The packing constraints include the measured micelle composition and, for regions where it has not been measured, ideal mixing is assumed. For the micelle data in this study the packing constraints imposed by the di-alkyl chain requires an additional parameter, ex, for the cationic rich solution compositions, which allows R1 to be greater than lc. The interparticle interactions are included using the RMSA calculation25,26 for a repulsive screened Coulombic potential, defined by the surface charge, z, the micelle number density, N, the micelle diameter, and the Debye-Hu¨ckel inverse screening length, κ, which is defined in the usual way.25 The model parameters refined are then ν, z, and ex, and an acceptable model fit requires the shape of the

scattering to be reproduced and the absolute value of the scattered intensity to be predicted to within (10%. For the data where two isotopic combinations, h-cationic/hnonionic and h-cationic/d-nonionic (hh, hd) were measured, a ratio of the two scattered intensities extrapolated to small Q gives an estimate of the micelle composition (mole fraction of each surfactant component),9 as has been demonstrated for a range of micellar systems.14,15,27 In the limit of small Q, and for dilute systems (where S(Q), P(Q) ∼ 1.0), the scattered intensity is approximately

I(Q) ≈

∑i NiVi2(Fip - Fs)2

(4)

where N is the micelle number density, V the micelle “dry” volume, Fip the micelle scattering length density, Fs the solvent scattering length density, and ∑i is over all micelle compositions for a binary mixture of surfactants. Changing the isotopic labeling of one of the surfactant components changes Fip, and so a simple ratio of the intensities for the two different isotopic combinations provides directly an estimate of the volume (mole) fraction of the two components. Although eq 4 is strictly valid for dilute solutions, providing S(Q) and P(Q) are not substantially different with the different isotopic labeling, then it is applicable to more concentrated solutions,9,10 and has been applied here to both the micellar and lamellar phase component of the scattering. Using the approach developed by Nallet et al.28 the lamellar phase scattering pattern can be analyzed to estimate the Caille constant (which is related to the lamellar membrane rigidity) and the number of layers/lamellar fragment. Their analytical expression takes into account the lamellar form factor P(Q) and a structure factor S(Q), which accounts for the membrane fluctuations and assumes a powder average and a line shape width dominated by the instrumental resolution, such that

I(Q) ) 2π P(Q) ) N-1

Sh(Q) ) 1 + 2

∑1

V 1 P(Q)Sh(Q) d Q2

Qδ 4 ∆F2 sin2 2 2 Q

( )

( ) ( 1-

n

(5)

cos

(6)

Qdn

1 + 2 ∆Q2d2R(n) 2Q2d2R(n) + ∆Q2d2n2 1 N

2(1 + 2∆Q2d2R(n))

)

e-

x1 + 2∆Q d R(n)

(7)

2 2

where for small n R(n) ) 〈(un - u0)2〉/2d2 ) ηn2d2/8, N is the number of layers in a lamellar fragment, η is the Caille constant, which is related to the membrane rigidity, η ) Q20 kBT/8πxκB, Β and κ are the elastic constants, ∆Q the instrumental resolution, d the lamellar d spacing, δ the bilayer width, d ) δ/φ (where φ is the volume fraction), and Q0 ) 2π/d. This provides an adequate description of the lamellar phase in these systems and has been extensively used elsewhere.28 Results and Discussion (i) Evolution of Phase Behavior and Microstructure. Figure 2 shows the variation in scattering for 5 wt % 2HT/C12E12/D2O (Figure 2a) and for 5 wt % DHTAC/C12E12/D2O (Figure 2b) with solution composition. The scattering is consistent with mixed cationic/nonionic micelles, L1, for solutions rich in C12E12

18110 J. Phys. Chem. B, Vol. 109, No. 38, 2005

Figure 2. Scattering Intensity as a function of scattering vector for (a) 5 wt % h-2HT/h-C12E12/D2O for composition mole ratios 1:2 (red), 1:1.5 (blue), 1;1 (green), 1:0.5 (pink), and 1:0.25 (black); (b) as (a) but for 5 wt % DHTAC/C12E12/D2O.

and a Lβ lamellar phase for solutions rich in 2HT. At intermediate compositions there is evidence for a coexistence of micelles and lamellar phase. The transition from micellar to lamellar phase is less clearly defined for the DHTAC/C12E12 data than for the 2HT/C12E12 data. The data shown in Figure 2 are for the isotopic combination h-cationic/h-nonionic/D2O. The complementary data for the combination h-2HT/d-C12E12/D2O are shown in Figure 3a, and are broadly consistent with the data for h-2HT/h-C12E12/D2O. A similar observation is made for the combination of DHTAC/ C12E12 (shown in Figure 3b), where the scattering for nonionic rich compositions is typical of that for interacting globular micelles. Broadly similar results are also obtained for the corresponding mixtures with C12E23, namely 2HT/C12E23 and DHTAC/C12E23, and for the mixture DHDAB/C12E12. The DHDAB/C12E12 mixtures were studied at compositions close to the coexistence region in order to establish more accurately those boundaries, and are discussed in more detail in section (iv). In the di-alkyl chain cationic surfactants DISDAC and DHTAC hydrolysis is an important issue, which can modify the phase behavior and the resulting microstructure. To establish the extent of the role of hydrolysis in this study, measurements were made on the compounds DISDA methyl sulfate and DHTA methyl sulfate, where changing the nature of the counterion

Penfold et al.

Figure 3. As Figure 2, but for the isotopic combinations (a) h-2HT/ d-C12E12/D2O, and (b) h-DHTAC/d-C12E12/D2O, and for composition mole ratios of 1:2 (b), 1:1.5 (O), and 1:1 (4). The solid lines are model fits as described in the text.

eliminates the effects of hydrolysis. Measurements were made for direct comparison with DHTAC/C12E12 (C12E23) mixtures. The same broad phase behavior that was characteristic of the 2HT/C12E12 (C12E23) and DHTAC/C12E12 (C12E23) mixtures has also been observed in the DISDA methyl sulfate/C12E12 and DHTA methyl sulfate/C12E23 mixtures. For the DHTA methyl sulfate/C12E23 mixture (see Figure 4 in the Supporting Information), the upper phase shows scattering characteristic of a mixed L1/Lβ to a pure Lβ phase for solutions increasingly rich in cationic surfactant. In contrast, the lower phase is purely micellar for solutions rich in nonionic surfactant (see Figure 1 in the Supporting Information for details). A similar evolution in structure is obtained for DISDA methyl sulfate/C12E12, except that the Lβ phase is notable for its high degree of order (six ‘Bragg’ peaks are observed, see Figure 4a), and Figure 4a shows the entire evolution of the upper phase microstructure with composition. The lower phase is micellar for the entire composition range (see Figure 4b). Plotting the Q value corresponding to the peak in the scattering pattern as a function of the solution composition (see Figure 1 Supporting Information for 2HT/C12E12 (C12E23) and for DHTAC/C12E12 (C12E23)), which in the micellar phase is related to the inert-micellar spacing, and in the lamellar phase the lamellar spacing, shows a clear discontinuity for 2HT/C12E12 (C12E23) mixtures corresponding to the transition from micellar to lamellar microstructure at a composition ∼60 mol % cationic.

Di-alkyl Chain Cationic/Nonionic Surfactant Mixtures

J. Phys. Chem. B, Vol. 109, No. 38, 2005 18111 TABLE 1: Micelle Model Parameters for 5 wt % 2HT/ C12E12 and 2HT/C12E23 isotopic composition combination 1:2 1:1.5 1:1 1:0.75

1:2 1:1.5 1:1

ν

z

R1

(a) 2HT/C12E12 139 17 24.1 131 149 20 24.1 152 206 22 26.5 210 241 21 26.5 245

hh hd hh hd hh hd hh hd

R2

ex

ee

32.7 1.0 1.4 31.8 1.0 1.72 33.6 1.1 2.0 32.6 1.1 2.5

scale factor 1.05 1.11 0.96 1.09 1.07 1.10 1.02 1.10

(b) 2HT/C12E23 110 14 24.1 39.1 1.0 1.2 127 17 24.1 37.5 1.0 1.46 166 18 24.1 35.1 1.0 2.11

hh hh hh

TABLE 2: Micelle Model Parameters for 5 wt % DHTAC/ C12E12 and DHTAC/C12E23 isotopic composition combination 1:2 1:1.5 1:1 1:0.5

1:2 1:1.5 1:1 1:0.5

Figure 4. Scattering Intensity for DISDA methyl sulfate/C12E12 at 5 wt % for solution compositions 1:2 (red), 1:1.5 (blue), 1:1 (green), 1:0.5 (pink), and 1:0.25 (black) for (a) upper phases and (b) lower phases.

The DHTA methyl sulfate, DISDA methyl sulfate and DHDAB/ nonionic mixtures all show trends similar to 2HT. Of all the mixtures, the transition is less well-defined for DHTAC/C12E12 (C12E23), reflecting a more complex interplay between the Lβ and L1 phases. This results in a larger region of mixed microstructure, which is attributed to the scavenging of DHTAC from the Lβ into the L1 phase. In summary, for the whole range of di-alkyl chain cationic and nonionic surfactants studied here, 2HT/C12E12 (C12E23), DHTAC/C12E12 (C12E23), DHDAB/C12E12 (C12E23), DHTA methyl sulfate/C12E23, and DISDA methyl sulfate/C12E12, the same general phase behavior and evolution of the solution microstructure, as determined from the SANS measurements, is observed. Solutions rich in the cationic surfactant are in a predominantly Lβ phase, whereas solutions rich in the nonionic surfactant are predominantly in a mixed surfactant micellar phase. At intermediate compositions there is a region corresponding to the coexistence of lamellae and micelles. However, the extent of the coexistence region depends on the actual cationic/nonionic mixture and the solution concentration. (ii) Characterization of Micellar Phase. For the four mixtures 2HT/C12E12, 2HT/C12E23, DHTAC/C12E12, and DHTAC/ C12E23 (see Figures 2 and 3), the data from the nonionic rich micellar region have been analyzed as interacting elliptical micelles, as described in the Experimental Section (using eqs 3

hh hd hh hh hd hh hd hh hh hh hh

ν

z

R1

(a) DHTAC/C12E12 290 9 22.9 210 10 451 15 22.9 646 23 27.8 435 19 769 40 34.4 667 30 (b) DHTAC/C12E23 316 8 22.9 530 14 22.9 811 24 27.5 1063 28 32.1

ee

scale factor

32.6 1.0 3.2 2.4 32.0 1.0 5.2 36.7 1.2 4.8 3.2 42.3 1.5 3.5 3.0

1.0 0.9 0.94 1.0 0.8 1.0 0.96

R2

37.3 36.1 40.6 45.2

ex

1.0 1.0 1.2 1.4

3.9 6.8 6.6 5.8

and 4). The solid lines in Figure 3 are typical model fits, and the key model parameters are summarized in Tables 1 and 2. The variation of the micelle aggregation number with composition is shown in Figure 5, and for all four mixtures the micelle aggregation number increases with increasing cationic content. The micellar aggregates are in general substantially larger for DHTAC than for 2HT. The behaviors of the 2HT/ C12E12 and 2HT/C12E23 mixtures are very similar, whereas the DHTAC/C12E23 mixed micelles are significantly larger than the DHTAC/C12E12 micelles. Another noticeable difference between the 2HT and DHTAC based mixtures is that the fractional charge on the 2HT/nonionic micelles is greater than that for the DHTAC/nonionic mixed micelles. A similar analysis has been carried out in the predominantly L1 rich regions for the separated lower phases of the DHDAB/ C12E12, DHTA methyl sulfate/C12E23, and DISDA methyl sulfate/C12E12 mixtures, where the SANS data are broadly similar to that for the 2HT and DHTAC/nonionic mixtures. The key model parameters for the different mixed micelles are summarized in tables 3, 6, and 7 in the Supporting Information. The general trend of increased micelle size with increasing cationic content, as observed for 2HT and DHTAC, is also observed here for the DHTA methyl sulfate and for DISDA methyl sulfate/nonionic mixtures. The micellar growth that is observed in all these systems is consistent with the observed transition from globular micelles to lamellar fragments. The aggregation numbers for DISDA methyl sulfate/C12E12 and its variation with composition are broadly similar to that for DHTAC/C12E12. However, the fractional charge on the micelles is closer to that observed for 2HT/C12E12 than to that for DHTAC/C12E12. The replacement of the Cl- by the methyl sulfate also results in systematically smaller micelles. Comparing

18112 J. Phys. Chem. B, Vol. 109, No. 38, 2005

Penfold et al.

Figure 5. Variation of micelle aggregation number, ν, with solution composition for 5 wt % 2HT/C12E12 (b), 2HT/C12E23 (O), DHTAC/ C12E12 (2), and DHTAC/C12E23 (4).

TABLE 3: Summary of Phase Behavior for 6 and 8 wt % DHDAB/C12E12 and the Isotopic Combinations hh, hd, for the Upper and Lower Separated Phases upper phase

lower phase

composition

hh

hda

1:1 1:0.8 1:0.7 1:0.6 1:0.4

L1 LR LR LR LR

6 wt % LR LR* LR* LR* LR*

L1 L1 L1 L1 LR

L1* L1* LR LR LR*

1:1 1:0.8 1:0.6 1:0.4

L1 LR LR LR

8 wt % LR LR* LR* LR*

L1 L1 L1 LR

L1* L1* LR LR*

a

hh

hda

Asterisk (*) indicates a peak shift with isotopic content.

DHTA methyl sulfate/C12E23 with DHTAC/C12E23, the micelle aggregation numbers are substantially lower and the variation with composition is less marked (see Figure 5 in the Supporting Information). Methyl sulfate and chloride ions both possess a single negative charge; however, the MeSO4- ion has a considerably larger molecular volume than chloride ion. The consequential lowering of the diffusion coefficient leads to a three-fold reduction in the ionic mean free path, thereby reducing its activity and hence acidity in solution. The situation is somewhat different for the DHDAB/C12E12 mixture, albeit measured over a more limited range of compositions in the coexistence region. For the mixture at 6 wt % the aggregation number is roughly constant with composition. Whereas at 8 wt % a trend opposite to that observed for the other mixtures is observed, namely that the aggregation number decreases with increasing cationic content (see Table 3 in the Supporting Information). Previous measurements on the DHDAB/ C12E6 mixture27 showed the opposite trend for solutions rich in nonionic content, consistent with that observed for other C12E6/ ionic surfactant mixtures.10 However, in that study for the DHDAB/C12E6 mixture, between 40 and 50 mol % (data were only obtained up to 50 mol % DHDAB) there is the onset of an increase in the aggregation number with increasing cationic content reported (see Table 5 in ref 27). However, those measurements were made at a much lower concentration (1.0, constraining the parameter to be equal to 1.0 will not allow the model to converge to satisfactory fit. Other variants on the basic model, such as increasing the radius of the outer shell through increased hydration, or changing the effective concentration within acceptable limits, do not improve the model fits. The apparent additional constraint in the core alkyl chain packing has not been included or considered explicitly in recent the theoretical treatments.29,30 Model treatments in terms of short rods or oblate ellipses do not improve the model fitting. These packing arguments are reinforced by the observation that, in general the parameter, ex is substantially larger for the DISDAC/nonionic mixture over the entire composition range measured, where the molecular architecture of the cationic surfactant is more complex, than for the other mixtures (see schematic in Figure 1). For 2HT and DHTAC the parameter ex is ∼1.0 for the nonionic rich micelles and is only significantly greater than 1.0 close to the coexistence region (see Tables 1 and 2). Around the limited composition range measured for the DHDAB cationic (which has a structure not too dissimilar to 2HT, apart from differences in chain length) with C12E12, a value of ex (∼1.2 to 1.3), similar to that for 2HT and DHTAC, is required. From membrane solubilization studies, the structure of the mixed phospholipid/surfactant micelles are usually assumed to be elongated (cylindrical) or discoidal (oblate ellipses),11 and this has been argued on the basis of packing constraints. This is consistent with the analysis presented here, where the scattering from short cylinders and prolate ellipses is largely indistinguishable. In contrast, Funari et al.13 have interpreted X-ray scattering, NMR and calorimetry from DMPC/ C12E8 and DPPC/C12E8 mixed micelles on the basis of discoidal aggregates, but not on the basis of any detailed analysis of the scattering data. As described earlier, another difference between the 2HT and DHTAC/nonionic mixtures, apart from the micellar size, is the surface charge on the micelles. This is discussed in more detail in section (v) in the context of the depletion-induced creaming or phase separation in these mixtures. However, it requires some broader discussion here in the context of its structural implications. For the 2HT/nonionic mixtures (see Table 1) the fractional surface charge, δ, (where δ ) z/ν) is in the range 7 to 12%

Di-alkyl Chain Cationic/Nonionic Surfactant Mixtures

Figure 6. Scattering intensity for 6 wt % DHDAB/C12E12/D2O at a composition mole ratio of 1:0.6 (upper separated phase) (b). The solid lines are calculated curves using eqs 5-7 as described in the text and the parameters in Table 2 in the Supporting Information.

(similarly for the DHDAB/C12E12 mixtures δ is ∼10%), whereas for the DHTAC/nonionic mixtures it is ∼3%. It has been argued elsewhere that headgroup electrostatic interactions limits micellar growth through its larger contribution to the free energy of micellization,29 and this is consistent with the larger aggregation numbers observed for the DHTAC/nonionic mixtures. Apart from DHTAC, all cationic molecules studied feature a quaternary ammonium headgroup where two out of the four constituents are large molecules. The subsequent packing of these dialkyl chains confers some distortion of the tetrahedral conformation of the nitrogen atom from an optimally minimized state. DHDAB and 2HT have only one water-soluble group and have to be dissociated in order to be surface active. DISDAC and DHTAC, however, do not have to be so well dissociated in order to be water soluble, thanks to the presence of ether-oxygen groups. The difference in the headgroup substructure renders DHTAC even less dissociated than DISDAC. Unlike DHDAB, 2HT, and DISDAC, only one of the constituents in the tetramethylammonium ion is a large molecule, and this does not perturb the tetrahedral configuration of this ion as the rest of the quaternary ammonium ion is made up of very short molecules (methyl groups). Consequently, the DHTAC headgroup has a higher surface specific charge at the quaternary ammonium ion than the other molecules and, thanks to the presence of the ethoxide links, can bind the neutralizing anion more strongly than DISDAC, 2HT, or DHDAB without compromising water solubility. For a number of different cationic/nonionic surfactant mixtures, different isotopic combinations of the surfactants were measured, to provide access to not only the structure but also the composition of the aggregated structure. For the 2HT/C12E12 and DHTAC/C12E12 mixtures the micellar and lamellar phase compositions were evaluated from eq 4 using the isotopic combinations of hh and hd (see Figure 2 and Table 4 in the Supporting Information). The variation of the aggregate composition with solution composition is different for 2HT and for DHTAC. For the 2HT/C12E12 solutions rich in nonionic content, there is a departure from ideal mixing. The aggregates are rich in the cationic surfactant, and this corresponds primarily to the micellar rich region. For DHTAC, the variation is close to ideal mixing, and aggregate and solution compositions coincide. For the DISDA methyl sulfate/C12E12 mixture (see Figure 6 in the Supporting Information), the variation in micelle composition with solution composition is also close to ideal, with some

J. Phys. Chem. B, Vol. 109, No. 38, 2005 18113 deviations at the cationic and nonionic rich extremes of the measurement range. This variation in isotopic content was also exploited in the measurements of the DHDAB/C12E12 mixtures and is discussed more extensively in section (iv). Where measured, for the 2HT/C12E12, DHTAC/C12E12, and DISDA methyl sulfate/C12E12, the two isotopic combinations hh and hd give a broadly consistent description of the micelle structure (see tables 1a, 2a, and Table 7 in the Supporting Information). In detail for the mixtures 2HT/C12E12 and DISDA methyl sulfate/C12E12 the structural parameters obtained for the micelles from the two isotopic combinations, hh and hd, are remarkably consistent (see Table 1 and Table 7 in the Supporting Information). The equivalent data for DHTAC/C12E12 mixture (see Table 2) are not so consistent, and for the cationic-rich compositions some variation with isotopic content is observed. This may be associated with the micelle size, which is significantly larger here than for the other mixtures. However, as discussed in more detail in the context of the DHDAB/C12E12 measurements in section (iv), the result of changing isotope (from hh to hd) can be more clearly observed in the location of the phase boundaries and the location of the coexistence region (the transition form micellar to lamellar), which are sensitive to isotopic content. (iii) Characterization of Lamellar Phase. The mixtures 2HT/C12E12 (C12E23), DHDAB/C12E12 and DISDA methyl sulfate/C12E12 rich in cationic surfactant (at mole ratios 1:0.5 and 1:0.25) are predominantly in the Lβ phase. The well-defined lamellar scattering data observed for 2HT/C12E12, DISDA methyl sulfate/C12E12 and DHDAB/C12E12 mixtures rich in cationic surfactant were analyzed using the approach of Nallet et al.28 (eqs 5-7), to obtain a Caille parameter, η, (which is related to the lamellar bending modulus) and the number of layers per lamellar fragment. Details of the model parameters and the plotted data and fits can be found in the Supporting Information, and Figure 6 shows the data for 6 wt % DHDAB/ C12E12, which is typical of the other data in this region of the phase diagram. For the three different mixtures, the Caille parameter, η, which is related to the membrane rigidity, is in the range 0.02 to 0.08. The low value of η is indicative of a highly rigid membrane. For both mixtures the number of layers in the membrane, N, is low and typically