Surface tension and neutron reflection measurements have been used to study the surface composition of aqueous solutions of mixtures of sodium dodecyl sulfate (SDS) and n-dodecyl--D-maltoside (C 12 maltoside) and C 12 maltoside and n-dodecyl-N,N′-dimethylamino betaine (C 12 betaine). From measurements of surface tension and mixed critical micelle concentrations (cmc) the pseudo-phase separation model has been used to calculate values of the interaction parameters in the micelle, M , and at the surface, σ . SDS/C 12 maltoside mixtures behave nonideally and both M are negative, indicating attractive interactions between the two surfactants, but the C 12 maltoside/C 12 betaine mixtures are closer to being ideal. Direct measurements of the surface excess using neutron reflection on isotopic mixtures of the surfactants are shown to be consistent with the surface tension measurements using the integrated form of the Gibbs equation. These direct values of the surface excess were found to agree with predictions from the partial phase separation model in the case of the nonideal pair, SDS/C 12 maltoside, but were different for the ideal pair, C 12 maltoside/C 12 betaine. It is suggested that this may result from the relatively large errors in the partial phase separation model when the mixture is very unsymmetrical (very different cmc's coupled with weak interaction). The C 12 maltoside/ C 12 betaine mixture was found to behave regularly in that σ showed negligible variation with either surface pressure or composition. On the other hand for the nonideal pair, SDS/C 12 maltoside, the magnitude of σ decreased with surface pressure, consistent with reduced interaction as the molecules are more widely spaced, and also decreased with higher SDS fraction in the layer, consistent with an increasing contribution from electrostatic repulsion.

Aqueous solutions of the strongly interacting anionic/zwitterionic surfactant mixture of sodium dodecyl sulfate (SDS) and dodecyl dimethylamino acetate (C12betaine) have been studied by means of surface tension and neutron reflection. The mixed critical micelle concentrations (cmc) were used to derive the interaction parameter βM for micellization, which was found to be large and negative. It was also found to be composition dependent and therefore not to obey the first-order approximation for the activity coefficients in the pseudo-phase separation approximation. The temperature dependence of the mixed cmc gave the thermodynamic excess functions for micellization; S E was found to be positive over most of the composition range. Application of the pseudo-phase separation model to surface tension data was used to show that the interaction parameter βσ in the surface layer is similarly large and negative. However, direct measurement of the surface concentrations using neutron reflection disagree with the predictions of the pseudo-phase separation model and indicate that βσ, although negative, is much smaller in magnitude. The structure of the mixed layer was determined at three compositions and found to be significantly dehydrated in comparison with layers of the single surfactants, which may explain the positive excess entropies observed for both micellization and surface mixing. It is also suggested that changes of hydration on mixing invalidate the use of the pseudo-phase separation model and may be responsible for the deviations from the first-order model observed for both micellization and surface interaction.

Differential scanning calorimetry (d.s.c.) and nuclear magnetic resonance (n.m.r.) techniques were used to study the melting of ice in porous solids.At low water contents (less than about three monolayers) no freezing or melting was observed. As the water content was increased, a single melting peak was observed as the pores filled. The amplitude of this peak reached a constant value when the pores were completely filled, and then at higher water contents a second peak was observed increasing in amplitude as more water was added. The lower melting point was characteristic of the pore water and this melting point decreased with decreasing pore radius. The higher melting point (OOC) was that of the non-pore, bulk, water. The pore volume determined calorimetrically agreed with the quoted pore volume. A coefficient of 0.9 for the linear correlation between melting point depression (AT) and the reciprocal radius (1 / r ) confirmed the applicability of the Kelvin equation. It is concluded that the observed properties of water in pores are essentially bulk properties at distances of more than about 10 A from the surface, and influenced more by the Kelvin effect, i.e., capillarity, than by the surface of the silica. D.s.c. can be used as a rapid method to characterise these silicas. Two simple measurements on silica samples containing water allow an estimation of the surface area, pore volume and an average pore size to be made.

No abstract

The composition and structure of the interfacial layer formed at the air/water interface by a zwitterionic surfactant, n-dodecyl-N,N-dimethylamino acetate, has been determined by neutron reflection. The compositions determined by neutron reflection and by surface tension measurements in conjunction with the Gibbs equation are in excellent agreement if the prefactor in the Gibbs equation is one, i.e., the surfactant behaves as though it is completely uncharged. At the critical micelle concentration (cmc) the structure most consistent with the neutron data is one where the head group is vertically oriented with respect to the surface normal, but the chain is strongly tilted away from the surface normal with a value of 〈cos θ〉 ≈ 0.45. This is a larger angle of tilt than observed with other surfactants at a comparable surface density and is attributed to the vertical orientation of the head group forcing the chain to be tilted away from the surface normal. Unlike most other surfactants, there is little change in the tilt of the chain as the coverage is decreased, probably because the tilt is already nearly a maximum at the cmc. The coverage of the layer increases markedly with temperature, and analysis of the thermodynamic parameters of the adsorption suggests the possibility that, at higher coverages, this is driven by a large gain in entropy from loss of water of hydration from adsorbed surfactant.