Surfactant molecules have the ability to self-assemble at concentrations greater than a critic micellar concentration (CMC) into several types of supra molecular aggregates. These aggregates include spheres (micellar phase), cylinders (hexagonal phase) or bilayers (lamellar phases) depending on the amount and types of surfactants that are used. Lamellar phases are not rigid, sheetlike bilayers but rather are elastic and can undulate. Upon shearing, lamellar phases can rearrange into soft colloidal objects of spherical shape consisting of a concentric stack of surfactant bilayers. These objects are called multilamellar vesicles (MLV).1
The MLV phase typically exists under shear and thus its structure is considered to be metastable. As a result, orientational phase diagrams frequently are used to describe the structure and phase space. Indeed, MLVs are not in thermodynamic equilibrium. Since they are formed by shear, they inevitably relax into a simple lamellar phase in time. This relaxation can take from hours to several months or years depending on the surfactant system used. Moreover, the stability of MLVs is sensitive to the conditions of process and the purity of the components.1 Specific mixtures of anionic, amphoteric and nonionic surfactants have proven to be effective for the formation of MLV systems with longtime stability, and numerous rinse-off products based on this technology have been commercialized; for example, sodium-trideceth sulfate and sodium-lauroamphoacetate in combination with lauric acid2 or alkanolamides3 have been used.
Considered to be soft solids, MLV systems obey the dynamics of soft glasses.4,5 The elasticity and yield stress originates from the close packing of the MLVs. MLV systems are used in body washes and shampoos because they impart the rheology necessary to suspend water-insoluble additives without stringy or tacky extensional properties. Upon dilution, the MLV phase breaks apart, allowing for foaming and efficient deposition.
In the present study, the MLV structure in a system under investigation is verified, the bulk rheology and MLV formation are studied, and a method to visualize individual MLVs is presented. The results are related to application-relevant properties such as texture and stabilization of performance additives.
The MLVs tested in the following study are comprised of a proprietary blend of amphoteric, anionic and nonionic surfactantsa. The blend is provided as a lamellar phase formed by 46% w/w active surfactant in water. Samples of MLVs are obtained through a simple cold process consisting of diluting the blend in water to the desired active concentration, adjusting the electrolyte level by adding between 2% and 4% w/w NaCl, adjusting the pH to 5.5 with citric acid, and shearing the resulting mixture at e.g. 300 rpm for several minutes. A schematic of the formulation process is shown in Figure 1. The resulting formulation is opaque and possesses a yield stress. From this basic platform formulation, fragrance, moisturizers, scrubbing beads, etc., can be added during or at the end of the process.
The pH adjustment and ionic strength are important for the formation of the MLV phase in the described system. The pH has an impact on the cationic to anionic charge ratio of the amphoteric surfactant. At a high pH level, the amphoteric surfactant is purely anionic but with decreasing pH, the molecules also develop cationic charges with the isoelectric point at a pH of 5.5. The increase of the ionic strength screens charges of the surfactant head groups.
Both the pH and ionic strength have an effect on the curvature of the self-assembled surfactant films, i.e., the structure of the surfactant aggregates. With these adjustments, bilayers can be observed in the system under investigation down to active concentrations as low as 10% w/w. More importantly, the elasticity of the surfactant bilayers as well as the interaction of the stacked bilayers is affected. The strength of the bilayer elasticity determines whether the formation of a planar lamellar phase, a bicontineous lamellar sponge phase, or an MLV phase is favored in a surfactant bilayer system.
Even though the phase behavior can be linked theoretically to the bilayer elasticity, it is difficult to predict the bilayer elasticity separately from the surfactant components and solution conditions. And even though the molecular structure can be used to estimate the packing parameter (a static property), it is not possible to predict bilayer motion because it is a dynamic behavior. The elasticity depends on many parameters that are difficult to list here because such a complex system includes charged and uncharged surfactants, where Coulomb interactions in a high ionic-strength aqueous solution impact behavior. Current theory cannot link the molecular structure of the surfactants used in the sample MLV system to these parameters.
The bilayer motion is on a length scale of around 1 nm or below, and on a time scale below 1 microsecond.
Scattering methods are used to determine the motion of the bilayers. Indirect methods such as small angle X-ray scattering (static) can be used to approximate the elasticity (Figure 2) but this determines only one component of the elasticity (kappa bar), multiplied with the bilayer interaction compressibility. Therefore, the development of MLV systems is based primarily on experience and screening work.
Indirect imaging via X-ray scattering: Although the resulting formulation is opaque, opacity does not fully prove the existence of MLVs. To this end, the structure of the formulations is directly and indirectly visualized. In one experiment, X-ray scattering is used to determine the internal lamellar structure and respective lamellar spacing of the MLVs at several active surfactant concentrations. In another experiment, a hydrophobic fluorescent dye is used to directly visualize individual MLVs in a formulation.
X-ray scattering is an established method used to characterize the structure of crystals and structured liquids.6 Such systems contain assemblies of repeating units with specific shapes and distances. The shapes and distances of the repeating units result in unique scattering profiles, allowing the identification of the structure in the sample. For example, cubic, hexagonal or lamellar phases can be identified by their specific scattering profiles. Because the internal structure of an MLV is a lamellar structure, it can be characterized by the small angle X-ray scattering intensity profile I(q), where I is the scattering intensity and q is the value of the scattering vector. The value of the scattering vector is related to the X-ray wavelength λ and the scattering angle θ by the expression: q = 4π/λ sin(θ/2)(see Figure 2).
The lamellar spacing d, the scattering angle θ, and the wavelength λ are related for a lamellar structure: nλ=2d(sinθ), where n is the order of Bragg-reflection (see Bragg-reflection). If this relation is fulfilled, the small angle scattering intensity shows a peak. The value of the corresponding scattering vector q at the intensity peak is usually denoted as q*n. When the ratio between the q- values of q*1- (first order) and q*2- (second order) is 2, a local lamellar structure is present. The lamellar spacing d is given by the relation d = n2π/q*n. The obtained intensity profiles (Figure 3a) show two peaks with a ratio q*1 to q*2 of 2, indicating the local lamellar structure of the MLV. The resulting lamellar spacing d is in the order of 10 nm. As expected for a lamellar phase, the bilayer spacing decreases with increasing active concentration c (data not shown). In addition to the lamellar spacing, the elasticity of the lamellar bilayer can be characterized by analyzing the line width of the peak at q*. It is well-known that the line shape obeys the expression:
where ηpeak is the Callé coefficient and given by:
where kB is the Boltzman constant. This expression links the rigidity product and the lamellar spacing d to the width of the peak ηpeak and the value of the wave-vector at this peak q*. As shown in Figure 3a, the peak positions give the lamellar spacing, and the line shape is a measure of the Callé coefficient.7
Direct imaging via fluorescence microscopy: To directly image individual MLVs, a hydrophobic dye can be incorporated into the surfactant bilayers. To the authors’ knowledge, this is the first time that a dye has been used to visualize MLVs. There is a wide range of dyes available with different structures, polarities and fluorochrome absorption-emission spectrums. In the present study, a specific boron dipyrromethene dyeb was used. This dye is hydrophobic, has good stability against light exposure, and its excitation and emission wavelength are compatible with the visualization technique. A solution of dye into chloroform at 10-3 mol/L was added to the unstructured blend (0.1% w/w CHCl3/dye solution in an aqueous solution of 25% w/w active surfactant). After adding salt and citric acid, the MLVs are formed, containing the dye inside the surfactant bilayers. Such a sample does not allow for the visualization of individual MLVs because the sample consists of closely packed fluorescing MLVs and, therefore, appears more or less homogeneous. However, a mixture of dyed and non-dyed MLVs generates a sample with individually dyed MLVs in a matrix of non-dyed MLVs. Images of such a sample are obtained with a fluorescent microscopec with a Hg-UV lamp for excitation and a standard green emission filter in combination with a CCD camera connected to a computer. An example of a resulting image is shown in Figure 3b.
The average diameter of the dyed MLVs is approximated 2.5 microns. Thus, one MLV in the sample consists of approximately 125 concentric bilayer shells (assuming a lamellar spacing of d ≈ 10 nm). The above described experiment allows for the visualization of individual MLVs, even in presence of dispersed oils, and also shows the capability of MLV formulations to encapsulate actives. The actives (here, a dye) are released upon use of the formulation; i.e., upon destruction of the MLVs by dilution with water.
Testing MLV Formation
To study the formation of 25% w/w fully structured MLVs, active surfactant samples were placed into an oven at 90°C for several hours. The temperature induces relaxation from an MLV phase to a planar lamellar phase. This planar lamellar phase persists even after the sample returns to ambient temperature. The lamellar phase is optically clear rather than opaque and appears highly birefringent when placed between cross-polarizers. The birefringence, as well as the observed flow properties of the sample, are indicative of a planar lamellar phase. The viscosity of the relaxed sample is significantly less than the structured MLV sample.
Applying shear: When a small amount of shear is placed on the sample, the sample returns to a thick opaque solution of MLVs. To quantify this structuring process, the viscosity of the relaxed sample was measured by a rheometerd with a cone and plate geometry as a function of increasing and decreasing shear rate. A shear rate was applied for 30 sec per point in Figure 4 to the 25% w/w active surfactant sample after its relaxation to a planar lamellar phase at 90°C; the images in Figure 4 show the clear planar lamellar phase before shearing and the opaque MLV phase after shearing.
Time and shear rate: The MLV formation time at a constant applied shear rate was investigated with the relaxed planar lamellar phase sample. The viscosity as a function of time was measured by rheometerd for several shear rates. The viscosity showed a strong increase with time and a plateau region at longer times. This strong increase in viscosity corresponds to the formation of MLVs, and the transition to a plateau at time tp corresponds to the formation of a fully structured sample. The formation time increased with decreasing applied shear rate. Figure 5 is a log-log plot of the formation time vs. shear rate.
A power-law dependence with an exponent of approximately -1.1 was observed; thus, , down to shear rates as low as 5 s-1. This result clearly showed that it is not necessary to use equipment with especially high shear rates to process MLV formulations. Common formulation equipment can be used and the process time can be adjusted according to the desired speed of mixing during formulation processing. The above described results on MLV formation are in accordance to findings on other MLV systems.8, 9
Bulk rheology: The bulk rheological properties of fully structured MLV samples also were investigated. Shear stress as a function of shear rate for an MLV formulation of 15% w/w active surfactant in water measured by rheometer is shown in Figure 6a.e The line on this graph corresponds to the equation . Here is the yield stress, is the characteristic time for stress-induced relaxation and is the viscosity at high shear rate. This equation differs from the standard Bingham fluid model typically applied for pseudoplastic flow. In this example, there is an anomalous ½ dependence on the strain rate. This dependence arises from the movement of randomly oriented slip planes in the close packed MLV sample. This behavior is also seen in compressed emulsions.10 Figure 6b shows the viscosity as a function of shear rate for the same data. In literature, the shear thinning exponent for MLV samples frequently is empirically stated as 0.8.11 From this data, it is seen that the previous equation divided by the shear rate (solid line) better explains the data both quantitatively and qualitatively.
The rheological behavior of MLV systems is key for its properties in personal care rinse-off formulations. These systems can be easily poured due to shear thinning behavior and their short stress-induced relaxation time allows for a quick recovery when poured. This property is important for a creamy, lotion-like, smooth texture rather than a formulation that spreads. This rheology is achieved without polymers that can leave a tacky texture. Furthermore, the existence of a yield stress allows for the suspension of water-insoluble additives. For example, up to 30% w/w of a vegetable oil or other moisturizing additive can be dispersed in an MLV formulation.
In conclusion, a blend of surfactants was used to form MLV formulations through cold processing. These MLV samples were tested and displayed a yield stress suitable to suspend water-insoluble additives.
In general, common methods were used to characterize the MLV phase. X-ray scattering was used to verify the internal lamellar structure and measure the lamellar spacing. Rheology was used to determine the yield stress of fully structured MLVs and the formation time of the MLV phase from a planar lamellar phase due to shear. In addition, the MLV formation time was shown to be inversely proportional to applied shear rate. The MLV phase was formed by small shear rates (1 to 5 s-1), implying that special equipment is not necessary for processing.
Finally, beyond the common methods described, the MLVs have been directly visualized with a fluorescent dye method. The uptake and encapsulation of the hydrophobic dye also shows that hydrophobic actives can be captured internally in MLVs and released during the destruction of MLVs upon dilution.
1. F Nettesheim, J Zipfel, U Olsson, F Renth, P Lindner and W Richtering, Pathway of the shear-induced transition between planar lamellae and multilamellar vesicles as studied by time-resolved scattering techniques, Langmuir, 19 3603-3618 (2003)
2. US20030180246, Stable surfactant compositions for suspending components, S Frantz, P Cotrell and S Warburton (Sep 25, 2003)
3. WO97/005857, Liquid cleansing composition comprising soluble, lamellar phase-inducing surfactant, S Puvvada, B Villa and R Kilidziej (May 23, 1996)
4. P Sollich, F Lequeux, P Hebraud and ME Cates, Rheology of soft glassy materials, Phys Rev Lett, 78(10) 2020–2023 (1997)
5. J Leng, F Nallet and D Roux, Anomalous elasticity of an ordered lamellar liquid foam, The European Physical Journal E, 4 337–341 (2001)
6. O Glatter and O Kratky, Small angle X-ray scattering, Academic Press (1982)
7. BA Coldren, H Warriner, R van Zanten, JA Zasadzinski and EB Sirota, Zero spontaneous curvature and its effect on lamellar phase morphology and vesicle size distributions, Langmuir 22 2474–2481 (2006)
8. JI Escalante, H Hoffmann, Non-Linear rheology of a flow-induced transition of a lamellar-to-vesicle phase in ternary systems of alkyldimethyl oxide/alchohol/water, Rheologica Acta 39(3) 209–214 (2000)
9. A Leon, Rhéologie des fluides complexes, transitions de texture et de phases induites par le cisaillement: Phases lamellaires et eponges de surfactan, PhD thesis, Universite Pierre Et Marie Curie—Paris VI (2001)
10. AJ Lie, S Ramaswamy, TG Mason, H Gang and DA Weitz, Anamalous viscous loss in emulsions, Physical Review Letters 76(16) 3017–3020 (1996)
11. S Fujii and W Richtering, Size and viscoelasticity of spatially confined multilamellar vesicles, The European Physical Journal E 19 139–148 (2006)