With high-pressure pendant-drop tensiometry, the interfacial tension (γ) and surface excess (Γ∞) for a family of ionic surfactants with identical phosphate headgroups and varying fluorocarbon and hydrocarbon tail structures were examined at the water−CO2 interface. To compensate for the unusually weak CO2−surfactant tail interactions, we designed hydrocarbon tails with weak tail−tail interactions to achieve a more favorable hydrophilic−CO2-philic balance. Branching of hydrocarbon surfactant tails is shown to lead to more favorable adsorption at the interface, closer to that of fluorocarbon surfactants. γ for a double-tail hydrocarbon phosphate surfactant with a relatively high degree of tail branching was lowered from the water−CO2 binary interface value of about 20 mN/m at 25 °C and 340 bar to 3.7 mN/m. This reduction in γ is attributed to both a decrease in the free volume between tails at the interface and reduced tail−tail interactions. In addition to tail structure, the effects of surfactant counterion, salt concentration, temperature, and CO2 density on γ and Γ∞ were investigated. The hydrophilic−CO2-philic balances of these surfactants are mapped by investigating changes in interfacial tension with these formulation variables. Low-molecular-weight branched hydrocarbon ionic surfactants are shown to stabilize concentrated CO2-in-water emulsions for greater than 1 h.

Water-in-carbon dioxide (W/C) microemulsions stabilized by an ammonium carboxylate perfluoropolyether (PFPECOO -NH 4 + ) surfactant are studied with NMR diffusion and relaxation methods with the aim of obtaining information on the dynamics of this system, as well as aiding in the design of new surfactants that can form stable microemulsions in CO 2 . Short proton transverse relaxation times (3-10 ms) measured for water and ammonium ions are shown to agree with a simple proton exchange model. As the pressure is lowered below the phase boundary, the NMR spectra indicate that surfactant migrates to the new liquid phase along with the water. Diffusion coefficients are reported in the CO 2 density range of 0.88-1.00 g/mL at 25 °C. The fractional amounts of water diffusion in bulk CO 2 , within the droplets, and through the water channels are delineated quantitatively. In decreasing the density from 0.96 to 0.88 g/mL, the water diffusion coefficient increases by a factor of 2 while the diffusion coefficients for ammonium ions and PFPECOOremain approximately constant. The droplet clusters are formed with channels that permit water molecules to diffuse freely over distances on the order of microns. This detailed dynamic molecular description of these clusters complements, in a consistent manner, macroscopic studies of percolation by conductivity measurements and equilibrium measurements of correlation lengths by SANS.

Tracer diffusion coefficients have been measured for water in liquid and supercritical carbon dioxide (CO 2 ) from 10 to 35 °C in the pressure range from 130 to 300 bar. The measurements were performed by means of pulsed field gradient NMR (PFG-NMR) methods incorporating compensation for electrical eddy currents and mass convection. In the NMR active volume, the sample was contained in a 1.4 mm i.d. PEEK tube with provisions for recirculation and external sample loading. The diffusion coefficients are consistent with the Stokes-Einstein equation with "slip" boundary conditions and a hydrodynamic radius of 1.7 Å for water in the high temperature and low density region. In the low temperature and high density region, the diffusion coefficients indicate either a trend toward "stick" boundary conditions or the dynamic clustering of water molecules.

(1)H NMR relaxation and diffusion studies were performed on water-in-CO(2) (W/C) microemulsion systems formed with phosphorus fluorosurfactants of bis[2-(F-hexyl)ethyl] phosphate salts (DiF(8)), having different counterions (Na(+), NH(4)(+), N(CH(3))(4)(+)) by means of high-pressure in situ NMR. Water has a low solubility in CO(2) and is mainly solubilized by the microemulsion droplets formed with surfactants added to CO(2) and water mixtures. There is rapid exchange of water between the bulk CO(2) and the microemulsion droplets; however, NMR relaxation measurements show that the entrapped water has restricted motion, and there is little "free" water in the core. Counterions entrapped by the droplets are mostly associated with the surfactant headgroups: diffusion measurements show that counterions and the surfactant molecules move together with a diffusion coefficient that is associated with the droplet. The outer shell of the microemulsion droplets consists of the surfactant tails with some associated CO(2). For W/C microemulsions formed with the phosphate-based surfactant having the ammonia counterion (A-DiF(8)), the (1)H NMR signal for NH(4)(+) shows a much larger diffusion coefficient than that of the surfactant tails. This apparent paradox is explained on the basis of proton exchange between water and the ammonium ion. The observed dependence of the relaxation time (T(2)) on W(0) (mole ratio of water to surfactant in the droplets) for water and NH(4)(+) can also be explained by this exchange model. The average hydrodynamic radius of A-DiF(8) microemulsion droplets estimated from NMR diffusion measurements (25 degrees C, 206 bar, W(0) = 5) was R(h) = 2.0 nm. Assuming the theoretical ratio of R(g)/R(h) = 0.775 for a solid sphere, where R(g) is the radius of gyration, the equivalent hydrodynamic radius from SANS is R(h) = 1.87 nm. The radii measured by the two techniques are in reasonable agreement, as the two techniques are weighted to measure somewhat different parts of the micelle structure.