Wilmer G. Miller

THERMODYNAMICS AND DYNAMICS OF POLYPEPTIDE LIQUID CRYSTALS

ABSTRACT The prediction that molecular asymmetry alone is sufficient to force a phase transition from a disordered to an ordered phase in a system of rodlike particles has been investigated using α-helical synthetic polypeptides as the source of rodlike particles. To this end the temperature–composition phase diagrams for the two component systems polybenzylglutamate (PBLG)–dimethylformamide (DMF) and polycarbobenzoxylysine (PCBL)–dimethylformamide are compared with the Flory lattice model for rigid impenetrable rods. The experimental systems exhibit and the theory predicts three distinct regions in the phase diagram: a narrow biphasic region in which isotropic and liquid crystal phases differing only slightly in composition coexist, a transition region over which solvent is increasingly excluded from the coexisting ordered phase and rods excluded from the coexisting isotropic phase, and a region where almost pure solvent coexists with a highly concentrated liquid crystal phase. A detailed comparison of the theoretical phase diagram for rigid, impenetrable rods with the experimental ones reveals discrepancies which can be attributed to the facts that the experimental rods are neither completely rigid nor impenetrable. Analysis of thermal data on PBLG–DMF indicates that the latent heat for the isotropic to liquid crystal phase transition is small and endothermic. We conclude that molecular asymmetry alone is sufficient to produce a phase transition to an ordered phase. Dynamical data show that the bulk viscosity may be considerably lower in the ordered than in the disordered phase. Although rod motion in the liquid crystal phase is correlated, electron spin resonance data suggest that individual rod motion about its mean lattice position is greater than in the isotropic solution of equivalent concentration. Additional electron spin resonance studies show that the motion of small rods and of the polymer side-chains is little affected by the presence of a high concentration of long rods, whereas the tumbling of long rods is dramatically influenced by the presence of other long rods.

Superspreading of water—silicone surfactant on hydrophobic surfaces

Abstract We have studied the spreading of aqueous mixtures of six different silicone surfactants on Parafilm, a hydrophobic surface. Two of the surfactants were soluble in water and did not superspread. The other surfactants were insoluble and their aqueous dispersions superspread. Thus we found that a dispersed surfactant-rich phase is needed for superspreading and that the initial rate of spreading increases with decreasing size of the particles of the dispersed phase. We also found that the radius of the spreading drop varies as the square root of time during the initial spreading. This time dependence is consistent with but not unique to Marangoni flow. The initial spreading rate passes through a maximum with increasing concentration, but the equilibrium area, i.e. the area wetted by the drop after spreading has ceased, is proportional to the surfactant concentration. When the spreading experiment was conducted in a dry atmosphere the dispersions did not superspread. The dispersions were superspreaders at 100% humidity and spreading rates were even greater in supersaturated air. Thus we found that the presence of water vapor is necessary for superspreading. We speculate that the water vapor provides a thin high tension film at the leading edge of the spreading drop, and so spreading is driven by a Marangoni effect, but we do not know whether a pre-existing film is formed by the vapor or whether it is recruited during spreading. A water film mechanically deposited near a spreading drop accelerates the spreading in the direction of the mechanically deposited film. It was initially thought that a hammer-like geometry was responsible for superspreading. However, Hill et al. (R.M. Hill, M. He, H.T. Davis and L.E. Scriven, Langmuir, (10 1994) 1724) found that an analogous linear trisiloxane surfactant is a superspreader. The natural next question was whether surfactant-rich dispersions of the hydrocarbon polyoxyethylenes (C m E n ) are superspreaders. In separate work we studied a series of dispersions of both hammer-like and linear hydrocarbon polyoxyethylene surfactants and found that none were superspreaders. Thus, among the surfactants compared, we found that a silicone hydrophobe is required for superspreading, but the molecular geometry of the trisiloxane surfactant does not appear to be a critical parameter.

Direct imaging of surfactant micelles, vesicles, discs, and ripple phase structures by cryo-transmission electron microscopy

Abstract Surfactant microstructures in dilute aqueous solutions and dispersions—globular, swollen, and cylindrical or wormlike micelles, discoid and ripple phase structures, and uni- and multilamellar vesicles—can be seen at high resolution by cryo-transmission electron microscopy (cryo-TEM) of thin vitrified sample films. Sample films are prepared within a chamber where temperature and chemical activities of the surrounding vapor are controlled, thereby preventing evaporation and temperature changes that could alter the microstructure in the labile systems. The thin liquid films are quenched by rapidly plunging them into liquid ethane. The resulting vitrified samples are mounted into a cold-stage and transferred into a TEM for direct observation. Monophasic solutions of cetyltrimethylammonium bromide (CTAB) show globular micelles that swell with added toluene or styrene to form swollen micelles. Worm-like micelles form in CTABNaBr solutions. Dilute mixtures of dipalmitoylphosphatidylcholine (DPPC) and diheptanoylphosphatidylcholine (DHPC) show discoid structures above the main transition temperature of DPPC and A and A/2 ripple structures of the Pβ′ phase at temperatures below the main transition temperature. A new model is proposed for the A structure and the ripple structures are shown to exist as single bilayers. Biphasic dispersions of sodium 4-(1′-heptylnonyl)benzenesulfonate (SHBS) show spheroidal and tubular vesicles, and complex encapsulated vesicles and coiled tubules. Vesicle-like microstructures of SHBS persist at 90°C. At the relatively low SHBS concentrations studied there is no evidence of the constant spacing characteristic of the lamellar phase at higher concentrations, suggesting that the structures observed may result from unbinding fluctuations that disrupt lamellar phases.

Staining and drying-induced artifacts in electron microscopy of surfactant dispersions. II. Change in phase behavior produced by variation in ph modifiers, stain, and concentration

Abstract Uranyl acetate (UA) is routinely used to enhance contrast in electron microscopy of aqueous biological lipid and aqueous synthetic surfactant samples. In one common application, the surfactant-water sample is combined with a uranyl acetate solution, placed on a film-covered microscope grid, and allowed to dry before imaging. The structures observed in the dried state are assumed to be representative of the original, hydrated, unstained system. We report evidence from two ionic surfactant systems—sodium 4-(1′-heptylnonyl)benzenesulfonate and sodium octanoate—with water that crystalline and liquid crystalline structures upon drying can be qualitatively different from the original hydrated state, both in the presence and absence of UA. We also find that another heavy metal salt, BaCl 2 , produces effects similar to UA. Moreover, the state of the aqueous UA-surfactant solutions or dispersions can be a sensitive function of pH modifiers, a variable rarely controlled in previous work. Our evidence for large-scale structure is drawn from polarizing microscopy, and for microstructure from electron microscopy, NMR spectroscopy and differential scanning calorimetry. We show that under carefully controlled conditions, UA can be safely used as a structure-preserving contrast agent of the hydrated state. However, the air-dried state of the aqueous surfactant solutions appears to differ from the hydrated state, particularly in the presence of UA. In some instances, fringes observed by electron microscopy and believed by others to be evidence for the existence of bilayers in the original hydrated specimen are shown to result from surfactant-stain microstructures formed during drying. Structures observed in transmission electron micrographs of air-dried, stained surfactant preparations cannot, in general, be taken as a reliable indication of the surfactant microstructure in the original unstained, hydrated state.

Relation of phase behavior to interfacial tensions of mixed surfactant systems

Abstract We measured by the spinning-drop and occasionally by the sessile-drop method interfacial tensions against decane and hexane of aqueous preparations containing sodium 4-(1′-heptylnonyl)-benzenesulfonate (SHBS), sodium dodecylsulfate (SDS), and NaCl. We determined the phase behavior of these surfactants in water/NaCl mixtures by spectroturbidimetry, visual observations, polarizing microscopy, ultracentrifugation, 13C NMR spectroscopy, conductimetry, and differential scanning calorimetry. At 25°C and below 1 wt% NaCl in water, SHBS is less than 0.06 wt% soluble and forms micelles. SDS can solubilize into mixed micelles up to 0.35 ± 0.07 mole of SHBS per mole of SDS. SHBS can solubilize 0.25 ± 0.08 mole of SDS per mole of SHBS into mixed lamellar liquid crystals. The solubilized SDS causes the melting transition centered at −70°C of the hydrocarbon chains of SHBS liquid crystals to broaden. When only micelles of pure SDS or SDS mixed with SHBS are present, the interfacial tensions against hydrocarbons are higher than 1 mN/m. Lower tensions occur when liquid crystals consisting of either pure SHBS or SHBS mixed with SDS are present in the aqueous phase. Ultralow—less than 10−2 mN/m—tensions, which are important for enhancing oil recovery, require the presence of dispersed liquid crystals of composition that “suits” the hydrocarbon. This composition occurs at an optimal combination of salinity, SDS-to-SHBS weight ratio, and total surfactant concentration. We have concluded that ultralow tensions arise not from dissolved surfactant but from these dispersed microcrystallites which by taking up hydrocarbon form a surfactant-rich third phase at the hydrocarbon/aqueous interfacial region. On the basis of this mechanism, we can suggest an explanation of the pronounced tension-concentration minima observed with multicomponent petroleum sulfonate commercial surfactants, such as Witcos TRS 10–80, whose behavior can be well-mimicked by our SDS/SHBS mixture.

Vesicle formation and stability in the surfactant sodium 4-(1′-heptylnonyl)benzenesulfonate

Abstract The double-tail surfactant sodium 4-(1′-heptylnonyl)benzenesulfonate, which forms a smectic phase upon hydration, was shown to form vesicles upon sonication, analogous to those formed by phospholipids. The stability of vesicles was followed by turbidity, conductance, capacitance, and nuclear magnetic resonance spectroscopy measurements. It was observed that vesicle preparations slowly revert to bulk smectic phase, though much of the surfactant remained in vesicles after many weeks. Vesicles were visualized without chemical fixation, staining, or replication, by utilizing the new technique of viewing a fast-frozen hydrated sample between polyimide films in a transmission electron microscope cold stage. Dispersions of the smectic phase viewed by the same technique showed moire fringes revealing the ordered nature of the smectic surfactant phase. Conductance and capacitance measurements on liquid crystalline dispersions of both the sulfonate and a phospholipid surfactant revealed a strong frequency dependence over the range 80−20,000 Hz, a phenomenon previously unreported, which is attributed to polarization of the electric double-layer around the particles. It is suggested that the phenomenon can be used to determine size and size distribution in liquid crystalline dispersions of ionic surfactants.

Microstructure in n-alkane-water-electrolyte mixtures with small ethoxylated alcohol amphiphiles

The ability of an amphiphilic molecule to form topologically ordered, surfactant-like aggregates is probed by studying the ethoxylated alcohols 3-oxaheptanol (C4 E1) and 3,6-dioxadodecanol (C6 E2). Density, refractive index, and NMR measurements indicated C4 E1-water-(NaCl salt) solutions are ideal dilute solutions with no solute association up to a C4 E1 mole fraction of about 0.018 (ca. 11 wt%). Above this concentration, the amphiphile forms aggregates which appear to be at least weakly cooperative. At high C4E1 concentrations, 23Na NMR and quasielastic light scattering (QLS) measurements indicated the existence of small brine-rich domains (<10A˚) which though probably of short life (⩽1 μs) can interact and exchange with the amphiphile. Addition of n-decane to C4E1-water-salt and of n-dodecane to C6E2-water-salt mixtures leads to strong solution nonidealities evidenced by both brine-rich and oil-rich critical points. The spectroscopic and QLS data obtained from amphiphile-oil-brine mixtures at low brine content (1–13 wt% brine) indicated larger brine-rich domain sizes (25–40A˚) than those in the oil-free mixtures. The 23Na NMR data in the oil-containing mixtures were very sensitive to nearness to the binodal separating one- and two-phase samples, which indicates that the domain size was largest when there was just enough amphiphile present to completely solubilize oil and brine. 13C NMR measurements suggested that both amphiphiles were oriented in the oil-containing mixtures with ethylene oxide groups toward water and tail groups toward oil. From the magnitude of the chemical shifts observed, C6E2 appeared to be more oriented than C4E1. Hence, even with these simplest ethoxylated alcohols, there are indications of surfactant-like aggregation, i.e., topological ordering, between oil and brine domains.

Characterization and aging of aqueous vesicular dispersions of sodium 4-(1′-heptylnonyl)benzenesulfonate

Abstract Vesicles formed by sonication of aqueous dispersions of liquid crystals of the double-tailed surfactant sodium 4-(1′-heptylnonyl)benzenesulfonate (SHBS) are examined with several techniques. The average diameter of the vesicles prepared in water is about 450 A. The average size decreases when prepared in NaCl or at higher surfactant concentrations. The presence of a few large liquid crystallites in the dispersion, as detected by fast-freeze cold-stage transmission electron microscopy, is shown to severely bias the measurement of vesicle sizes by quasi-elastic light-scattering techniques. The commonly used techniques of gel-permeation chromatography and ultrafiltration are shown to be ineffective in separating liquid crystals from SHBS vesicle dispersions. Vesicle preparation in the presence of uranyl acetate is shown to dramatically reduce the vesicle size. The spontaneous, irreversible reversion of vesicles to liquid crystallites as the dispersions age is documented and proves that SHBS vesicles are not equilibrium structures in water or brine.

Vesicular dispersion delivery systems and surfactant waterflooding

Surfactant waterflooding processes that rely on ultralow interfacial tensions suffer from surfactant retention by reservoir rock and from the need to avoid injectivity problems. The findings reported open the possibility that by delivering the surfactant in vesicle form, more successful low-concentration, alcohol-free surfactant waterflooding processes can be designed. Recent work, including fast-freeze, cold-stage transmission electron microscopy (TEM), reveals that sonication both in the absence and the presence of sodium chloride converts particulate dispersions of Texas No. 1 into dispersions of vesicles, which are spheroidal bilayers or multilayers less than 0.1 /mu/m in diameter filled with aqueous phase. It is found that their stability depends on their preparation and salinity history, and that contact with oil can accelerate greatly the reversion of a vesiculated dispersion and enable it to produce low tensions between oil and water. 46 refs.

Interpreting the appearance of dispersed systems: I. Model dispersions of polymer latex microspheres

Measurements of total absorbance at wavelengths 350–780 nm of aqueous dispersions of polymer latex microspheres of diameters 0.091 μm, 0.254 gmm, 0.325 μm, and 1.10 μm were used to interpret systematic observations of them. Light scattering dissymmetries and scattering ratios of dispersions of the 0.091 μm microspheres were measured at varying concentration and path length at 546 nm and 436 nm. Spectroturbidimetry and observations were also made in binary mixtures of the above particle sizes and in dispersions of microspheres with added dye, the sodium salt of methyl red. For ab-sorbance due to scattering, Ascat, exceeding 0.04 but not 2, the ab-sorbance and its wavelength dependence yield reliable estimates of particle size, even though the dissymmetry and the scattering ratio do not. Observations of nonabsorbing systems under ordinary illum-ination are most reliably interpreted with 0.1 < Ascat < 1, i.e., when the systems look translucent to translucent-turbid, even though multiple scattering predominates in this range. That the Tyndall effect, or a variant of it when absorption is important, is visible im-plies that particles smaller than 0.1 μm are present. To estimate particle sizes in milky dispersions in which A > 2, it is necessary to decrease the path length — or the concentration, if tolerable — so that the absorbance falls in that optimal range. Outside this range, the literature rules are unreliable. Because observers and illumina-tion conditions vary among laboratories, it seems essential that model systems such as the microspheres and the dye employed here be used to simulate scattering and absorption features of dispersed systems. By direct comparisons of model systems to systems of interest, observations can be standardized and interpretation of appearances can become less subjective. Moreover, combining obser-vations with spectroturbidimetry provides a much more potent tool for estimating sizes simply and quickly.