Preferential Solubilization of Fragrances in Micelles with Different Geometric Shapes

Abstract

Surfactant self-assembled aggregates, the main components of consumer products, offer a solution for the solubilization of fragrances, which is crucial to the consumer’s product choice. The interactions between surfactant aggregates and fragrances are complex: surfactants influence fragrance release and performance, whereas fragrances affect the macroscopic properties of the formulation by changing the aggregate’s shape and size. The present chapter describes studies on the solubilization of some fragrance ingredients in spherical and cylindrical micelles for a better understanding of their influence on micellar structure, viscosity, conductivity, and solubilization capacity. Turbidity, conductivity, and viscosity measurements were performed simultaneously in order to monitor the solubilization of fragrance molecules and the geometric transition of the surfactant aggregate.

1. Introduction

Most home and personal care consumer products, such as detergents, all-purpose cleaners, dishwashing products, shower gels, shampoos, and liquid soaps, represent aqueous surfactant solutions with different surfactant concentrations that are, in general, composed of a mixture of anionic and nonionic surfactants. All of these products solubilize fragrances—complex olfactive mixtures of volatile ingredients—at different concentrations, depending on the application field. Consumers are highly sensitive to the olfactive profile, and often, the fragrance is the predominant factor in the choice of a given product. In many countries (in South America, for example), the olfactive profile and fragrance intensity are the main criteria for the perception of cleanliness and personal hygiene.

However, most perfumery compounds are strongly hydrophobic, hence possessing low solubility in water. Their hydrophobic character hampers their incorporation in water-based fragrance formulations. Amphiphilic association structures such as surfactant molecular aggregates offer a solution to the solubilization problem of oily synthetic perfumes and have been the subject of intensive investigation over the last decade [1, 2, 3, 4, 5, 6]. Scientific interest has been stimulated by subtle issues arising from the understanding of the solubilization phenomenon (e.g., the chemical nature and hydrophobicity of the perfume itself [1, 2, 4], the influence of the surfactant structure and its properties [1, 3, 7, 8], the perfume location in surfactant aggregates [5, 9, 10, 11, 12, 13, 14], or the interfacial curvature change of the surfactant assembly with the addition of perfume [8, 15, 16]). The fragrance ingredients represent small molecules (MW < 500 g/mol) that often combine two or more chemical functionalities, making them powerful “troublemakers” or efficient “helpers” in aqueous surfactant solutions. They can induce change in product properties such as turbidity, phase separation, viscosity, and others. The fragrance ingredients are distributed in the surfactant palisade layer (at the hydrophobic/hydrophilic interface) or the hydrocarbon part of the surfactant’s aggregates in different proportions, depending on the surface affinity and the level of hydrophobicity. In a previous study [17], we attempted to classify the fragrance ingredients in surface activity by using the parameter EACN_mix (equivalent alkane carbon number for mixture) to express the relative fragrance interfacial activity with respect to a reference oil. We established a rule according to which fragrance ingredients with an EACN_mix of <5.5 possess surface activity and preferential partitioning into the palisade surfactant layer of the droplets. Thus, fragrance ingredients having an EACN_mix of >5.5 are highly apolar and prefer to be localized in the micellar core.

The shape, size, charge, and diffusivity of the surfactant micelles also have an important role in successful fragrance solubilization. The most frequent practical problems are undesired product turbidity and viscosity changes induced by fragrance solubilization even at relatively low fragrance concentrations (0.25 to 1%wt in personal and home care products and up to 2%wt in some fragrance boosters). Fragrance content is limited not only by legislation but also by solubilization. Therefore, studies of fragrance solubilization mechanisms, rates, and limits, along with their consequences on the macroscopic properties of a product, are valuable for the consumer goods industry.

The solubilization limit, characterized by the so-called Maximum Additive Concentration (MAC), is a measure of the solubilizing capacity of surfactant aggregates for a given solute. The value of the MAC is affected by and mainly dependent on parameters such as the surfactant assembly itself and the hydrophobic character of the guest molecule, as well as temperature, pH, and electrolytes. Numerous studies have reported the maximum solubilization of fragrance ingredients in solutions containing different types of surfactants (nonionic, anionic, and cationic) and mixtures of [1, 2, 3, 4], focusing on the type of surfactant and not on the surfactant aggregate shape. Nonetheless, few studies have answered the following questions:

1. What is the solubilization capacity of surfactant aggregates with different geometries for fragrance molecules with different functionalities?

2. Which micelle geometry must be chosen for the specific solubilization of a fragrance with known hydrophobicity/surface activity?

The answers to these questions could help in the creation of compatible (surfactant self-assembly architecture–fragrance) combinations while avoiding technical issues.

The solubilization capacity of rod-like surfactant aggregates for different additives was investigated by Hoffmann and Ulbricht [18]. They studied the solubilization of hydrocarbons (linear and aromatic), linear alcohols, and esters (with different chain lengths) in aqueous surfactant solutions of tetradecyltrimethylammonium bromide (cationic surfactant) in the presence of equimolar quantities of sodium salicylate. They reported that linear alkanes at a low hydrocarbon-to-surfactant ratio induced a drastic decrease in light scattering, indicating a transformation from rod-like to spherical micelles. The aromatic hydrocarbons did not always show the same behavior at low concentrations (<10 mM). The authors mentioned that the effect was correlated to the chain length of the molecules. The alkanes could induce the formation of a coacervate phase, and the transition from rod to sphere could be observed at a higher solute concentration (>20 mM). The authors also studied the behavior of alcohols and demonstrated that those with a chain length less than that of n-pentanol reduced the rods to spheres, whereas the longer chain length alcohols stabilized the rods. The transformation of rods to spheres has been demonstrated at low concentrations (up to 15–20 mM) for both ionic (cationic) and nonionic micelles.

The esters behaved more as hydrocarbons than as alcohols. For example, ethylhexanoate had a similar effect as n-hexane. The esters can be solubilized in a relatively high concentration in the rod-like micelles before reducing the size of the rods. Thus, in contrast to that of alkanes, the solubilization of alcohols in surfactant self-assemblies is enhanced when the micelles are spherical, whereas cylindrical micelles allow the solubilization of alkanes better than they do of alcohols. The authors explain the observed phenomenon as an entropic gain due to geometric constraints: the required interfacial area per molecule in spherical micelles is higher than in rod-like micelles, but the hydrophobic core volume is smaller. Therefore, the spherical micelle allows better solubilization of surface-active molecules such as some alcohols and poorer solubilization of highly hydrophobic molecules such as hydrocarbons. Rod-like micelles can accommodate more hydrophobic molecules in the core, but the interfacial area is closely packed, which avoids almost any contact between the water and the hydrocarbon and lacks space for the cosurfactants.

At higher concentrations, the behavior of the short alcohols, solubilized in cylindrical micelles, was shown to be different: their solubilization caused the appearance of lamellar phases or the coexistence of a micellar solution and a lamellar phase. In this case, the authors distinguished the I-type of lamellar phase, which is formed mainly as a result of interactions between the surfactant and the cosurfactant and possesses a high solubilization capacity.

All of these micellar changes induced by the solubilization of solutes with different hydrophobicities and surface activities are important for industrial applications because they cause considerable viscosity and turbidity changes. The reduction in the size of the rod-like micelles would tend to decrease the solution viscosity, whereas the increase in the size of the micelles and the micelle-to-liquid-crystalline phase transition would increase the solution viscosity. For this reason, it is important to verify whether the same effects are observed with more complex molecules such as fragrance ingredients.

Perfumery ingredients cannot simply be classified in groups of hydrocarbons, alcohols, or esters. Often, they have a mixed structure composed, for example, of terpene and alcohol, or a combination of phenol and alcohol or aldehyde. In addition, in general, the concentrations used in consumer goods are higher than those applied in the previously mentioned studies (varying from 0.1 to 1%wt and exceptionally to 2%wt, which corresponds to a range of ∼4 to ∼150 mM). Therefore, it is not obvious that fragrance solubilization in spherical and cylindrical micelles would follow the rules established for linear hydrocarbons, alcohols, and esters.

Here, we report our studies on the solubilization of fragrance ingredients in spherical and cylindrical micelles, in which we aimed to determine the solubilization rate and solubilization capacity of micelles for these complex molecules. The effects of solubilization on the macroscopic properties, viscosity and conductivity, were also investigated. The correlation between the interfacial activity of perfumery ingredients and turbidity and viscosity is discussed. We chose to use a simple solubilizing system, composed of the anionic ethoxylated surfactant sodium lauryl ether sulfate (SLES, EO2), which is the main surfactant in rinse-off products [19]. The data were obtained with an experimental setup designed specifically for the investigation of solubilization capacity or the elaboration of phase diagrams [20]. Turbidity, conductivity, and viscosity measurements were performed to determine the solubility of the fragrances and the phase transition borders in case of surfactant aggregate shape evolution. Polarized microscopy was applied to confirm the existence of liquid crystalline phases.

We believe that this study would be useful in providing guiding principles for the formulation of successful fragranced surfactant-based products.

2. Experimental section

2.1 Materials

The basic formulation in many personal care products consists of a mixture of anionic and nonionic surfactants in water [21]. For the anionic surfactant, we used SLES with two ethoxylated groups on average (Texapon® N70, Cognis GmbH, Germany).

The perfumery ingredients phenethyl alcohol, eugenol, linalol, lilial®, hexyl salycilate, and limonene are high-purity products (>98%) supplied by Firmenich SA. Their chemical structures and hydrophilicity/hydrophobicity classification are shown in Table 1. As can be seen from the log P or EACN_mix values, the hydrophilic properties of the perfumery ingredients follow the order phenethyl alcohol > eugenol > linalol > lilial > limonene > hexyl salycilate.

Table 1.

Chemical functionality, log P, and EACN_mix of the different perfumery raw materials used in this study.

2.2 Experimental procedure

In this study, a multiparameter scan analysis method was used to obtain the maximum solubilization capacity of micellar surfactant phases for different fragrance molecules. This analysis method allows the simultaneous determination of different physicochemical parameters such as transient turbidity, viscosity, or conductivity after a concentration jump during titration. The experimental setup, “Multi Parameter Scanning Instrument v.2 (MPS-2),” supplied by Scanalys® (Sweden), is schematically presented in Figure 1. The working principle is to measure and record several parameters at a time, while controlling the temperature and composition. This setup combines several devices (controllers) to measure transient conductivity, pH, turbidity, and viscosity. The inspection window with a polarizer filter allows visualization of the formation of the liquid crystalline phase during the experiments. The precise description of the experimental setup is given in Ref [20].

Figure 1.

Schematic presentation of scanning instrument MPS-2. L1 and L2 indicate the automatic syringes; 1–4 correspond to the different measurement devices; 5 is the light beam for turbidity measurement; S means sample, and P represents the window with polarization filters.

Small quantities of a perfumery ingredient were added to the continuously stirred surfactant system. The titration was performed by steps of 0.25%wt fragrance ingredient concentrations with automatic syringes. The time between each fragrance ingredient addition was set to 10 min. This time was sufficient for the complete solubilization of each portion of 0.25%wt up to high concentrations. The experiments were performed at a temperature of 25°C. Turbidity and conductivity were measured continuously every 5 s. These two parameters measured at the same time during solubilization allowed us to know immediately and precisely the effect of fragrance on the surfactant system.

2.2.1 Conductivity

Conductivity, measured simultaneously with turbidity, was used to determine the interfacial curvature change of the surfactant structure or the presence of a two-phase solution. On the one hand, a phase transition of the surfactant aggregates from micellar to a more organized structure (such as hexagonal or lamellar phase) would reduce the mobility of the counterions of the charged surfactants and hence would tend to decrease the conductivity of the solubilizing surfactant structure. On the other hand, above the solubility limit, the water-based solution will be in equilibrium with an excess oil phase whose conductivity is lower than that of the water. As a result, a decrease in the overall solution conductivity in combination with a strong increase in turbidity would allow determination (confirmation) of the MAC.

2.2.2 Rheology

The viscosity measurement was used to detect the change in shape and size of the micelles, as well as the phase transitions, which can occur during the solubilization of a perfumery ingredient into surfactant aggregates. The viscosity of each fragrance/surfactant system was measured on a C-VOR 150 rheometer from Bohlin Instruments by using the cone-plate technique. The measurements were performed with an increasing shear rate from 0.01 to 100 s⁻¹. The plate temperature was controlled with a Peltier unit at 25°C.

The viscosity of the water/surfactant/perfumery ingredient is mainly governed by the viscosity of the continuous phase, that is, the surfactant phase in the aqueous solution [16, 22]. In addition, the rheological behavior of the different surfactant aggregates is variable [23].

3. Results and discussion

3.1 Formulation and identification of surfactant aggregates

From our previous studies and from the literature [24], we know that at low concentrations, SLES forms spherical (or globular) micelles. With increasing surfactant concentration, they transform into cylindrical micelles, followed by hexagonal phases and finally lamellar phases. In order to determine the phase boundaries precisely, we plotted the viscosity as a function of surfactant concentration at a fixed shear rate (52 s⁻¹) and temperature (25°C) (Figure 2). The slope change on the viscosity curve corresponds to the phase boundaries. When SLES forms spherical micelles, viscosity remains unchanged. Above 23%wt SLES, viscosity increases due to a phase transformation in rod-like micelles. Above 28%wt SLES, the cylindrical micelles are forced to pack into a hexagonal phase and the viscosity is superior to 10 Pa.s. Finally, above 45%wt SLES, lamellar phases are formed with a viscosity lower than that of the hexagonal phase.

Figure 2.

Phase behavior and viscosity for SLES.

For the purpose of our study, we chose to work with the concentrations 10%wt SLES (spherical micelles) and 23%wt SLES (cylindrical micelles) for the single surfactant system [25].

3.2 Solubilization of fragrance ingredients in spherical and cylindrical micelles: general trends

We investigated the solubilization of six fragrance ingredients (Table 1) with different hydrophobicities and surface activities into SLES aggregates with two different micellar geometries: spherical (or globular) and cylindrical micelles. Turbidity, viscosity, and conductivity as a function of the fragrance ingredient concentration are presented in Figures 3 and 4 for all ingredients solubilized in 10%wt SLES (left column) and 23%wt SLES (right column). For each ingredient, the changes in turbidity, viscosity, and conductivity could be compared as a function of the fragrance concentration.

Figure 3.

Turbidity (gold symbols), viscosity (black symbols), and conductivity (gray symbols) of surfactant solutions containing 0.25%wt perfumery ingredient, (A) eugenol, (B) linalol, (C) phenethylol, and 10%wt SLES (left column) and 23%wt SLES (right column).

Figure 4.

Turbidity (gold symbols), viscosity (black symbols), and conductivity (gray symbols) of surfactant solutions containing 0.25%wt perfumery ingredient, (A) lilial, (B) limonene, (C) hexyl salicylate, and 10%wt SLES (left column) and 23%wt SLES (right column).

The morphological transitions induced by the solubilization of the fragrance ingredients in the different micellar solutions are summarized in Table 2. As shown, the fragrance alcohols strongly influenced the micellar aggregates by changing their geometry and, consequently, the viscosity of the solutions.

Fragrance ingredient	10%wt SLES		23%wt SLES
	Cr. Conc (%)	Transition	Cr. Conc (%)	Transition
Eugenol	2.0	Sph - Cyl	2.0	Cyl - Hex
	5.5	Cyl - Lam
Linalol	3.0		2.5	Cyl - Hex
Phenylethyl alcohol	—	—	5.0	Cyl - Lam
	—	—	10.0	Lam - Cyl
Lilial	—	—	4.0	Cyl - Lam

Table 2.

Structural transitions of micellar phases induced by fragrance ingredient solubilization.

Eugenol induced two different aggregation changes at 2 and 5.5%wt in the 10%wt SLES solution. These changes could be observed as slope changes on the conductivity and viscosity curves, whereas turbidity increased continuously. The first transition should correspond to the spherical to cylindrical micelle transition and the second to the cylindrical to lamellar phase transition. The lamellar phase was identified from the texture on the microscopic images taken under polarized light. Solubilized in cylindrical micelles (23%wt SLES solutions), eugenol led to the formation of a hexagonal phase at 2%wt concentration. This fragrance ingredient is highly surface active with a very strong influence on the interfacial curvature of the microemulsion droplets in both spherical and cylindrical micelles.

Linalol solubilization led to a continuous increase in turbidity and viscosity in the solutions with spherical micelles. Conductivity at low linalol concentrations slightly increased and then started to decrease progressively. Phase transitions were not observed. The influence of linalol on the interfacial curvature seems less important than that with eugenol. In the solutions with cylindrical micelles, the effect of linalol solubilization was stronger. A clear phase transition from cylindrical micelles to the hexagonal phase was observed. The hexagonal phase was identified by microscopy in polarized light. Turbidity increased significantly, and viscosity maintained a constant high value, a characteristic property of the hexagonal phase.

The solubilization of phenylethyl alcohol in the solutions with spherical micelles induced a progressive, slight increase in viscosity and turbidity that was related to the change in micelle size. The conductivity curve went through a maximum value and then started to decrease similar to the curve for linalol. Linalol and phenylethyl alcohol could be solubilized up to very high concentrations in the spherical micelles without a significant perturbation of the macroscopic properties of the solutions. However, solubilization of phenylethyl alcohol in the cylindrical micelles (at 23% wt SLES) induced a couple of transitions at 5 and 10%wt fragrance concentrations. The turbidity and viscosity curves both underwent a maximum. At lower concentrations (<5%wt phenylethyl alcohol), a transparent viscous solution was obtained, which should contain cylindrical micelles. In the range from 5 to 10%wt phenylethyl alcohol, an intermediate lamellar phase was formed, and at concentrations >10%wt phenylethyl alcohol, an isotropic transparent gel was obtained until the critical concentration of 26%wt phenylethyl alcohol, at which point a phase separation was achieved.

The solubilization of lilial in the spherical micelles, observed on the turbidity and the viscosity curves, did not induce strong changes in the surfactant aggregate shape. However, the conductivity curve changed the shape strongly: from a constant value at the low concentrations of lilial, it started to decrease progressively above 1.5%wt lilial, indicating some charge or ion mobility reduction. The absence of a liquid-crystalline phase and the high solution turbidity, combined with the strongly decreasing conductivity, indicated a phase separation in microemulsion in equilibrium with an oil excess phase. Integrated into the cylindrical micelles (at 23%wt SLES), lilial induced different transitions. The conductivity curve (Figure 4A, right) is more complex than the curve corresponding to the spherical micelles, containing a minimum at low lilial concentrations and a maximum at higher lilial concentrations. The minimum at 3%wt lilial, indicating the branching of the cylindrical micelles, correlated very well with the viscosity peak. At above 5%wt lilial in the cylindrical micellar phase, the system turned into a lamellar phase. The value of the conductivity of the lamellar phase was higher compared with that of the hexagonal phase, obtained at 2.5%wt eugenol in cylindrical micelles. This difference between both types of liquid crystalline phases is well correlated to their viscosity and the mobility of the free electrolyte ions in the solution. The turbidity increased continuously without abrupt changes.

The solubilization of limonene and hexyl salicylate increased the turbidity continuously and the viscosity slightly without changing the morphology of the phase and the micelle shape at both surfactant concentrations. The conductivity kept an almost constant value in the whole range of fragrance ingredient concentrations. Limonene is an aromatic hydrocarbon and did not transform the cylindrical micelles into spherical ones, but rather stabilized the cylinders. This effect is in agreement with the studies of Hoffmann and Ulbricht [18] concerning aromatic alkanes. Hexyl salicylate is a phenyl ester and, as was found by Hoffmann and Ulbricht for the esters, it behaved similar to limonene without induction of any micelle shape transition or morphological change. We did not observe the reduction in turbidity (light scattering in [18]) reported by Hoffmann and Ulbricht due to a transformation of cylindrical to spherical micelles observed during the solubilization of the alkanes, short-chain alcohols, and esters in the range of their concentrations (up to 20 mM) and even at higher concentrations. Possible reasons for the disparity in results could be a) the mixed functionalities of the fragrance molecules or b) the starting high concentration (0.25% > 10 mM). To clarify the reasons for the different observations, we performed experiments in which hexane and decane were solubilized in the cylindrical micelles (23%wt SLES), expecting to reproduce the transformation effect. However, only a continuous increase in the turbidity and viscosity was observed (Figure 5).

Figure 5.

Turbidity as a function of the oil concentration of solutions at 23%wt SLES, solubilizing hexane (gray symbols), hexyl salicylate (gold symbols), and decane (black symbols). Inset: Turbidity as a function of oil concentration (<1%wt oil).

Comparison of the turbidity of the solutions containing hexane, decane, and hexyl salicylate in cylindrical micelles (23%wt SLES) demonstrated that hexyl salicylate solubilization is similar to that corresponding to decane solubilization. The inset in Figure 5 shows the behavior at low concentrations (<1%wt oil). Some reduction of the turbidity could be observed for the three oils, but the effect was negligibly weak and did not induce viscosity changes. Therefore, the transformation should happen only at very low concentrations or it should take longer than our experimental time, as was reported in [26] for the solubilization of triolein in cylindrical micelles of nonionic surfactant (3 to 16 days).

3.3 Solubilization rate and MAC

3.3.1 Turbidity (η)

The solubilization capacity, expressed by the MAC, was defined as the highest concentration of fragrance molecules that could be incorporated into the micellar structure at a given surfactant concentration before the two-phase separation. A typical experimental turbidity curve as a function of time and fragrance ingredient concentration is presented in Figure 6. The sequence of turbidity peaks corresponds to the stepwise addition of the fragrance ingredient [1, 3, 4]. When the solubilization limit is reached, a strong asymptotic and irreversible increase in turbidity takes place [20, 27], as shown in Figure 6, revealing the coexistence of two phases (i.e., excess oil phase in equilibrium with a microemulsion or liquid crystalline phase). The separation of the perfume-surfactant system in two phases is confirmed by centrifugation. This critical fragrance ingredient concentration is considered the MAC of the current surfactant system.

Figure 6.

Turbidity (η) experiments on a 10%wt SLES solution (η⁰) in which successive amounts of a fragrance ingredient (0.25%wt) are added. (inset) turbidity change during the solubilization of 0.25%wt fragrance.

The turbidity of the surfactant solution before the fragrance addition is denoted as η₀. As soon as the perfumery ingredient is injected, the turbidity jumps strongly because of the spontaneous formation of a two-phase solution. During the process of mixing and solubilization, the turbidity exponentially decreases and reaches a constant value B, as depicted in the inset for Figure 6 [27]. After each titration of a perfumery ingredient, the turbidity plateau B could theoretically recover the initial value η₀ of the surfactant solution (obtained before the turbidity peak), indicating the absence of any structural or size change in the solubilizing surfactant aggregate. In practice, oil solubilization always leads to swelling of the micelles and a consequent increase in their size, and as can be observed from Figure 6, the turbidity plateau B reaches a value higher than η₀, indicating an increase in micelle size or number.

The turbidity decay as a function of time (Figure 6 inset) was fitted by using the following model:

$$\mathrm{\eta }\left(t\right)=\mathit{\text{Aexp}}\left[-\tau \left(t-t_{\mathit{\text{peak}}}\right)\right]+B$$E1

where η is the turbidity (NTU), τ is the solubilization rate (s⁻¹), and A and B are constants. At t → ∞, η → B = η_s, and at t = t_peak, η_max = A + B.

If the turbidity results are presented in the form:

$$\left[\frac{\mathrm{\eta }-{\eta }_s}{{\eta }_{\mathit{max}}-{\eta }_s}\right]=\mathit{exp}\left[-\mathrm{\tau }\left(t-t_{\mathit{\text{peak}}}\right)\right],$$E2

the solubilization rate τ (s⁻¹) can be determined as the curve slope, assuming that η, η_s, η_max, and t_peak are experimental values for all molecules.

The constant η_s is the turbidity value at the end of the solubilization process of each injected fragrance quantity. If the solubilization does not induce any change in the micelles, the value of η_s should be equal to that before the injection of the fragrance (η₀). Increasing the solute concentration causes the micelles to swell and increase in size, which should lead to an increase in turbidity. Therefore, the plateau value η_s should be directly related to the droplet radius and density (interactions).

As can be observed from Figure 7, the solubilization rate depends not only on the type of micelles but also on the type of perfumery ingredients. The solubilization rate values (Table 3) required to solubilize 0.25%wt fragrance ingredient seem to be significantly sensitive to the perfume type in the spherical surfactant aggregate geometry. In contrast, in cylindrical micelles, the solubilization rate has similar and low values for all perfumery ingredients.

Figure 7.

Normalized turbidity as a function of time to solubilize 0.25%wt perfumery ingredient in (A) spherical (10%wt SLES) and (B) cylindrical (23%wt SLES) micelles: Phenethyl alcohol (black), eugenol (red), linalol (gray), lilial (green), limonene (violet), hexyl salicylate (blue).

Fragrance ingredient	τ (s^−1) (in spherical micelles)	τ (s^−1) (in cylindrical micelles)
Phenethyl alcohol	0.1020	0.003
Eugenol	0.0120	0.005
Linalol	0.0170	0.004
Lilial	0.0010	0.004
Limonene	0.0060	0.006
Hexyl salicylate	0.0004	0.004

Table 3.

Solubilization rate of the different perfumery raw materials studied in spherical and cylindrical micellar phases of SLES (10%wt SLES and 23%wt SLES, respectively).

According to the theory of solubilization [28], the solubilization rate should be dependent on the oil properties via the parameters V_oil (oil molecular volume), D_oil (diffusion coefficient in the surfactant solution), and C_eq (oil solubility in water): β = V_oil × D_oil × C_oil (Table 4).

Fragrance ingredient	Mvol (cm^3/mol)	D (cm^2/s)	Wsolubility (mg/L)	β (cm^2/s)
Phenylethyl alcohol	119.8	1.35E-05	21990.00	2.92E-07
Eugenol	154.9	1.05E-05	754.00	7.45E-09
Limonene	179.3	9.05E-06	683.70	7.20E-09
Linalol	217.3	7.47E-06	7.86	6.24E-11
Hexyl salicylate	162.1	1.00E-05	4.58	5.46E-11
Lilial	215.7	7.52E-06	6.01	4.39E-11
Decane	194.9	8.33E-06	0.05	5.93E-13

Table 4.

Maximum additive concentration of the different perfumery raw materials studied in spherical and cylindrical micellar phases of SLES (10%wt SLES and 23%wt SLES, respectively).

At the end of the solubilization process of each fragrance portion, the value of η_s was determined and presented as a function of the fragrance concentration. From this curve, the value of the MAC was found for each perfumery ingredient as the concentration at which the turbidity slope increases asymptotically. Therefore, this is the maximum fragrance concentration that can be incorporated in the current micellar system. The values of MAC are reported in Table 5. The results indicate that solubilization into spherical micelles is highest for phenethyl alcohol, followed by linalol and eugenol and then hexyl salicylate, lilial, and limonene. In the cylindrical micellar phase, the order of the MAC values changes: hexyl salicylate is the most solubilized, followed by lilial, phenethyl alcohol, and limonene and then linalol and eugenol. These results clearly indicate preferential solubilization of the fragranced molecules with surface activity in surfactant aggregates with spherical geometry. The highly apolar fragrance molecules are preferentially solubilized by the cylindrical surfactant aggregates. Our results confirmed the preferential solubilization effects observed in [18] for the perfumery ingredients.

Fragrance ingredient	MAC [%wt]
	10%wt SLES	23%wt SLES
Phenethyl alcohol	13.1	4.7
Eugenol	6.0	2.3
Linalol	10.7	3.2
Lilial	2.4	6.4
Limonene	2.2	3.7
Hexyl salicylate	3.1	6.7

Table 5.

Maximum additive concentration of the different perfumery raw materials studied in spherical and cylindrical micellar phases of SLES (10%wt SLES and 23%wt SLES, respectively).

Notably, increasing the surfactant concentration in the aqueous solution is not a sufficient factor to increase the MAC value, in particular for fragrance alcohols (such as eugenol and linalol).

In our previous studies [17], we introduced the parameter EACN_mix, a key indicator for the interfacial solubility of the perfumery ingredients and mixtures. We found that fragrance molecules leading to a low EACN_mix value, EACN_mix < 5.5, have an affinity for the interfacial domain within the surfactant aggregates, whereas those having a higher EACN_mix value, EACN_mix > 5.5, are solubilized into the apolar core of the surfactant structures. The correlation of the solubilization rate τ and MAC to the EACN_mix parameter for the perfumery molecules is presented in Figure 8A and B. The curves indicate that perfumes with high surface activity such as phenethyl alcohol, eugenol, and linalol (EACN_mix = 1.96, 2.99, and 4.33 respectively; see Table 1) are solubilized faster and at higher quantity than the highly hydrophobic perfumes in the spherical micelles. The cylindrical micelles are preferred by the three hydrophobic perfumery ingredients, lilial, limonene, and hexyl salicylate (EACN_mix = 5.7, 6.0, and 6.39, respectively; see Table 1). The observed phenomenon is closely related to the results of the previous studies concerning the correlation between the packing parameter, the interfacial concentration of the fragranced molecules, and the EACN_mix values [17, 29].

Figure 8.

Solubilization rate τ (A) and MAC (B) of perfumery ingredients in spherical (10%wt SLES) (♦) and cylindrical (23%wt SLES) (•) micelles as a function of EACN_mix.

For fragrance alcohols, it was commonly observed [8, 13, 15, 30] that these molecules (including eugenol and linalol) act as cosurfactants and penetrate into the palisade layer of the surfactant self-assembly. As a consequence, the larger the hydrophilic region of the micelle, the higher the solubilization/incorporation of the fragrance molecule with surface activity between the surfactant head groups. For geometric reasons, as was clearly explained by Hoffmann and Ulbricht [18], the spherical shape requires that the area per surfactant head group is $a_s\ge 3a_0$, where $a_0$ is the cross section of the hydrophobic surfactant chain. The rod-like shape requires that $a_r\ge 2a_0$. Therefore, the surface, accessible to the surface-active molecules, is higher in the spherical micelles. Moreover, the adsorption of fragrance molecules at the interface leads to a decrease in electrostatic repulsion between the SLES head groups and thus to closer packing of the palisade layer. This effect could also be observed as a decrease in conductivity. The consequence is transition from spherical to rod-like micelles [31]. For these reasons, phenethyl alcohol, eugenol, and linalol can be solubilized in higher quantities in spherical than in cylindrical micelles.

On the other hand, the highly hydrophobic perfumery ingredients such as hexyl salicylate and limonene, localized mostly in the micellar core, have a negligible influence on the packing parameter, mostly by oil penetration, between the hydrophobic surfactant chains. As is mentioned in [18], the solubilization of non-surface-active hydrophobic molecules in spherical micelles is limited to the micellar core and leads to swelling of the micelle without important curvature changes. The micelles swell until the limit of the packing surface area is reached, and the addition of more fragrance molecules induces a two-phase separation.

The relative turbidity, equal to the ratio Bi-η₀/η₀ after the solubilization of each portion i of fragrance and before reaching MAC, is presented in Figure 9 as a function of the volume fraction of fragrance in the surfactant solutions of spherical (A) and cylindrical (B) micelles. The curves can be sufficiently well fitted by a linear model:

Figure 9.

Relative turbidity as a function of the fragrance volume fractions in SLES aqueous solutions: (A) 10%wt SLES and (B) 23%wt SLES. Symbols are experimental points, and lines are fitting curves with linear dependence: Phenylethyl alcohol (black), eugenol (red), linalool (gray), lilial (green), limonene (violet), hexyl salicylate (blue).

$$\frac{{\eta }_s-{\mathrm{\eta }}_0}{{\mathrm{\eta }}_0}=\alpha {\mathrm{\Phi }}_o$$E3

where Φ_o is a perfumery ingredient volume fraction and α is a constant, determined by the curve’s slope. The turbidity values considered here correspond to solubilization in a single phase (spheres or cylinders). Any phase transitions are excluded. Approaching the phase boundaries, the turbidity changed exponentially as a function of the perfumery ingredient volume fraction.

Turbidity is a measure of the light scattered by the aggregates dispersed in the aqueous solutions. The relationship between turbidity and the properties of the scattering system is defined by the following equation [32]:

$$\eta =R_{\pi /2}\left[\frac{16}{3}\pi -\frac{128}{15}\frac{R^2{\pi }^3}{{\lambda }^2}\right]$$E4

where R is the droplet radius, λ is the light wavelength, and R_π/2 is:

$$R_{\pi /2}=\frac{2{\pi }^2{n}_w^2}{{\lambda }^4}{\left(\frac{\mathit{dn}}{\mathit{d\varphi }}\right)}^2{V}_{\mathit{\text{sphere}}}\mathit{\varphi S}\left(0\right)$$E5

where n is the refractive index, Φ is the volume fraction of the scattering objects (micelle + oil), S(0) is the structure factor, and V_sphere is the volume of droplets. These equations show that the turbidity is a function of micelle characteristics, as well as of oil characteristics directly via the refractive index increment $\frac{\mathit{dn}}{\mathit{d\varphi }}$ and indirectly via the droplet size.

$$\frac{\mathit{dn}}{\mathit{d\varphi }}=\frac{3}{2}{\epsilon }_w^{1/2}\left(A\frac{{\varphi }_o}{\varphi }+B\frac{{\varphi }_s}{\varphi }\right)$$E6

${\varphi }_o$ and ${\varphi }_s$ are the oil and surfactant volume fractions, respectively, and $\varphi$ is the total volume fraction of the scattering particles, $A=\left({\epsilon }_o-{\epsilon }_w\right)/\left({\epsilon }_o+2{\epsilon }_w\right)$ and $B=\left({\epsilon }_s-{\epsilon }_w\right)/\left({\epsilon }_s+2{\epsilon }_w\right),$ with ${\epsilon }_i$ being the dielectric constants of oil, water, and surfactant, assuming that n² = ε.

Thus, α should be dependent on the oil properties directly via the refractive index increment (dielectric constant of the oil) and indirectly via the droplet’s radius, both parameters closely linked to the hydrophobicity (polarity) and surface-active characteristics of the fragrance ingredients.

3.3.2 Viscosity (μ)

Figure 10 depicts the dependence of the specific viscosity on the oil volume fraction for all fragrance ingredients solubilized in (A) spherical micelles and (B) cylindrical micelles. The viscosity changes induced by fragrance solubilization depend on both the micellar system used and the type of fragrance ingredient, as can be observed from the graphs.

Figure 10.

Relative viscosity as a function of the fragrance volume fractions in SLES aqueous solutions: (A) 10%wt SLES and (B) 23%wt SLES. Phenylethyl alcohol (black), eugenol (red), linalool (gray), lilial (green), limonene (violet), hexyl salicylate (blue).

The correlation between the specific viscosity and the volume fraction of oil for the solutions at 10%wt SLES and at fragrance ingredient concentrations up to 2%wt (Figure 10C) follows a linear relation:

$$\frac{{\mu }_{o+s}}{{\mu }_s}-1=\gamma {\mathrm{\Phi }}_0$$E7

with coefficient γ being similar for all fragrance ingredients and having values in the range from 7 to 15. According to the slope coefficient value, the fragrance ingredients could be arranged in the following order: eugenol > linalol > phenylethyl alcohol> hexyl salicylate > lilial > limonene.

For the solubilization in the cylindrical micelles (at 23%wt SLES), the correlation between the specific viscosity and the volume fraction of oil (Figure 10D) cannot be fitted with linear relation but follows an exponential relationship:

$$\frac{{\mu }_{o+s}}{{\mu }_s}-1=exp\left(\delta {\mathrm{\Phi }}_0\right)$$E8

with coefficient δ sensitive to the fragrance ingredients properties and having values in a large range from 0 to 175, following the order eugenol > lilial > linalol > phenethyl alcohol > hexyl salicylate > limonene.

Therefore, the influence of solubilization on solution viscosity in spherical micelles for all of the fragrance ingredients studied is weaker than that on the viscosity of the cylindrical micellar solutions. In addition, the strong effect in the cylindrical micelles is observed mainly for surface-active molecules. Eugenol is the fragrance molecule with highest impact on the viscosity of the solutions, whereas limonene is the molecule with the lowest impact.

Reduction of the viscosity (not related to reduction in turbidity) was observed only with surface-active ingredients (phenethyl alcohol and lilial) at very high fragrance ingredient concentration. This finding is related to the melting of the lamellar phase and the transition to micelles. Eugenol (at 2%wt) and linalol (at 3%wt) induce the transition of cylindrical micelles to the hexagonal phase, and the viscosity maintains its high value.

4. Conclusion

The study of the maximum solubilization of various fragrance molecules covering the entire range of hydrophobicity into aqueous micellar solutions containing anionic surfactant revealed a preference of the perfumery ingredients for a given surfactant aggregate’s geometry. Correlation between the interfacial affinity of the fragrance molecules and the geometric shape of the surfactant aggregates appropriate for their solubilization is reported.

The perfume ingredients having a micellar core affinity, EACN_mix > 5.5, are better solubilized in the cylindrical micelles because the solubilization of these fragrance molecules is reduced to a hydrophobic core volume. Phase transition of the surfactant structure is not induced by the solubilization of the highly hydrophobic oils, and thus, the viscosity of the surfactant solution is not modified or slightly increased. Once the MAC of the micellar system is achieved, cloudiness indicates the coexistence of two phases: microemulsion in equilibrium with oil.

In contrast, the perfume ingredients having an interfacial solubility, EACN_mix < 5.5, are better solubilized in spherical micelles due to their ability to strongly affect the interfacial curvature. The incorporation of these types of fragrance molecules modifies the intermolecular interactions between the head groups of the surfactants themselves and the guest oil. Consequently, the dynamic surfactant-perfume aggregates tend to turn from spherical to cylindrical micelles to liquid crystalline phases. The mixed chemical functionalities of the fragranced molecules do not allow their classification simply into groups of alkanes, alcohols, and esters: the effects of the transition of rod-like to spherical micelles accompanied by a decrease in the viscosity observed by Ulbricht et al. with alcohols, esters, and alkanes were not observed. Moreover, the three molecules considered as alcohols, phenethyl alcohol, eugenol, and linalol, revealed very different behavior when solubilized in spherical and cylindrical micelles; the strongest interfacial influence was observed with eugenol. Phenylethyl alcohol expressed a nontypical alcohol behavior in the cylindrical micelles with a transition from lamellar to micellar phase at very high fragrance ingredient concentration.

Acknowledgments

The author would like to thank to Virginie Soulié, who performed important parts of the experiments during her training period at Firmenich SA.

Conflict of interest

The authors declare no conflict of interest.