401 THE EFFECT OF VEGETABLE OIL COMPOSITION and stearic acid and by comparing the results previously obtained in different oils, it is possible to see that the synergistic effect varies as a function of the oil (21). For example, only one optimum R = 8:2 was found in rapeseed oil, two optimum R = 3:7 and 7:3 in soybean oil, and two optimum R = 7:3 and 8:2 in canola oil (18). The variation of R as a function of the vegetable oil is still not explained in the literature. Recently, Shaink developed a model based on the Hildebrand equation to better understand these sys- tems and to predict the phase diagram of stearyl alcohol and stearic acid in sunflower oil (22). However, the model does not account for the possible formation of mixed crystals that have been observed experimentally (22). Stearyl alcohol is the only fatty alcohol registered as self-affirmed and generally is recognized as a safe (GRAS) ingredient for oleogelation and therefore, can be used as an oleogelator for the food industry (23–25). For other applications, such as cosmetic or pharmaceutical products, longer alkyl chain fatty acids and fatty alcohols [e.g., behenyl alcohol (BO) and behenic acid (BA)] are more appropriate because of their better gelling properties (1,11,26). In the pioneering work about fatty alcohol/fatty acid oleogelator system, Gandolfo et al. studied the mixture of BO and BA in sunflower oil at 5 wt.% of total structurant (18). There was only a slight effect of the weight ratio for this system in comparison to the stearyl alcohol/stearic acid oleogelator system. Indeed, R seemed to reach a maximum in terms of oleogel hardness around 140 g for R = 7:3 and R = 8:2, but the oleogel hardness was already around 125 g for BO alone (R = 10:0) (18). In order to clarify the effect of R for BO and BA in sunflower oil, we investigated recently this system at higher structurant concentration (10 wt.%) (27). In this previous work, the same trend was obtained as for stearic acid and stearyl alcohol-based oleogels, with a clear enhancement of oleogel properties for specific R (27). One R (7:3) gave oleogels with both the highest hardness and stability in terms of oil-binding capacity. It was defined as optimal. R is a key parameter for oleogels based on BO and BA (27). In the literature, it is known for specific organogelators systems that a change in solvent type can alter the network formation and the resulting oleogel proper- ties such as rheological properties (28–33). The interactions between the oleogelator and the solvent depend largely on the chemical composition and on the polarity of the solvent (28–32). The precise ratio of oleogelator-oleogelator interactions to oleogelator-solvent interactions is known to play a central role in the formation of an oleogel. However, the direct effects of solvent on the oleogel properties are still not well understood. For example, in ethylcellulose oleogels, the rheological properties depend on the polymer– solvent interactions, which are influenced by the molar volume of the solvent linked to the amount of unsaturation in oil (33). In oleogel based on the self-assembly of γ-oryzanol and β-sitosterol, the key parameter is the polarity of the oil (34). Not only the polarity affects oleogelator system, the viscosity of the oil phase which affects the gelation time and the final gel strength (35). In the same way, in oleogels based on monoglycerides, oils with different polarity and viscosity led to different gelation and crystallization behav- ior (36). One key parameter also is the polar minor components present inside the oils (37,38). For example, polyphenols in extra virgin olive oil decrease the oleogel hardness for ethylcellulose oleogels (38). The fatty acid chain length in the oil was also reported to modify the rheological behavior of oleogels (39). All these previous studies highlighted the complexity to make a clear link between oleogel properties and the nature of the oil used since there seem to be several factors contributing to the changes in gel formation and final gel properties (37). In the case of the BO and BA oleogelator system, the effect of the oil on the mixture and the resulting oleogel properties is not known. However, it is very important to determine this effect in terms of oleogels texture and stability, because

402 JOURNAL OF COSMETIC SCIENCE these oleogels have practical applications for the cosmetic industry since they are used to produce oil foams for skin and hair applications (11,40). In this study, our objective is to fill this gap in knowledge by systematically investigating the effect of the nature of the oil on these systems. Four vegetable oils, which are widely used in cosmetic applications, were studied: olive, apricot, camelina, and rapeseed oil. The oils were not purified to study the real industrial products as they are used in cos- metic industry (11). The new results were compared with the previous results obtained for sunflower oils (27). First of all, the oleogel properties as a function of R in terms of structuring potential (hardness, oil loss and gel stability) were determined. Then, struc- tural data was obtained using a multiscale approach. We correlated these observations with the fatty acid chain length composition of the oil. MATERIALS AND METHODS MATERIALS All the ingredients used were of cosmetic grade, and classically incorporated in cosmetic products. BO (1-docosanol), 77.0% purity, impurities include 17.2% arachidyl alcohol (1-icosanol), 5% stearyl alcohol (1-octodecanol), and 0.6% lignoceryl alcohol (1-tetraco- sanol)), was purchased from BASF (Ludwigshafen, Germany). BA (docosanoic acid), 88.2% purity, impurities include 1% palmitic acid (hexadecanoic acid), 3.8% stearic acid (octadecanoic acid), 5% arachidic acid (eicosanoic acid), and 2% lignoceric acid (tetraco- sanoic acid), was purchased from KLK OLEO (Emmerich am Rhein, Germany). Rape- seed and olive oils were purchased from HUILERIES DE LAPALISSE (Lapalisse, France). Apricot kernel oil and Camelina oil were purchased from Naturex (Avignon, France). The fatty acid chain length of the vegetable oils was determined by the different suppliers using capillary gas chromatography analysis after alkaline treatment in accordance to the European Official Methods of Analysis (Table I). All the raw materials were used without further purification. OIL PROPERTIES The viscosity of the oils was determined using a rheometer (MCR502, Anton Paar GmbH, Graz, Austria) with a double gap geometry. After sample loading, the oils were equilibrated 30 min at 25°C before measurements. The shear rate was increased from 0.1 to 100·s−1. All the oils behaved as Newtonian liquids. The surface tension was measured using a Krüss K100 tensiometer under ambient conditions, with a typical temperature of 25°C ± 0.5°C. The surface tension was measured using a du Nouÿ ring method. The density of the vegetable oils was determined with a digital densitometer (DMA-4500, Anton Paar Graz, Austria). All the measurements were performed in triplicate. SAMPLE PREPARATION All concentrations are expressed as weight percentage, w/w. The oleogelator concentration was kept constant at 10 wt.% in 90 wt.% of oil. This concentration was chosen in order to form oleogels with sufficient crystals to produce then oil foams by whipping, otherwise not enough crystals were present to stabilize the air bubbles (40). The weight ratio (R)