3.1. Oscillatory Rheological Characterization
The first series of experiments relates to the characterization of ternary systems using classic nano-emulsion ingredients, i.e., MCT, Kolliphor
® ELP, and water. A SOR of 60% was first fixed, and the SOWR was varied from 40% to 80%. Representative rheograms showing G′, G″, and phase angle
ϕ before the sol/gel transition (SOWR = 44%), at the gel point (SOWR = 52%), and at the gel state (SOWR = 60%) are shown in
Figure 1. The details of the composition are reported in
Table S1, in the Supplementary Information section. G′, G″, and phase angle
ϕ are represented as functions of the strain. This representation is generally proposed to show the linear regime clearly visible from around 0.1% to 5%.
In the linear regime, a phase angle from 45° to 90° indicates a prevalence of the viscous contribution G″ upon the elastic modulus G′ (
Figure 1a,d). During the sol/gel transition, elastic and viscous moduli show similar values, and the phase angle
ϕ presents a transitional value between a pure viscous state and a pure elastic state of around 50° (
Figure 1b,e). Finally, when the gel was formed (
Figure 1c,f), the elastic contribution G′ prevailed upon G″, and the phase angle dropped to a value close to 0°, indicating a purely elastic system. This first result demonstrated that the water content had a significant role in the viscoelastic behavior of the formulation. In general, nano-emulsion formulations are performed using SOWRs ranging from 20 to 30%, i.e., with a much higher water content. These results showed that the ternary system could be considered liquid up to a SOWR ~50% and transit to a gel state when lower amounts of water are involved in the nano-formulation.
Other sets of experiments were performed by increasing the proportion of surfactants in the mixture, using SORs = 40%, 60%, and 70%, and extended to other oils. Thus,
nanoemulgels were formulated with monoglycerides (Capmul
®), vitamin E acetate, and castor oil in the same conditions as described with Labrafac
®. Their respective results appear in
Figure 2 (for G′) and
Figure 3 (for the phase angle). The first observation is about the strength of the gels, in a comparable range > 10 MPa for Labrafac
®, castor oil, and vitamin E, when it is significantly weaker for Capmul
®. For this reason, the sol/gel transition is clearer when studying Labrafac
®, castor oil, and vitamin E acetate, as we clearly observed a dramatic drop in
ϕ to be close to zero, at low water content (SOWR = 40–50%). However, when the water amount was further decreased, the phase angle tended to rise again towards viscous values, making the gel weaker. The system tends to be a pure surfactant/oil phase, emphasizing the important role of water in the formation of the gel state. For most of the formulations, we observed a plateau that corresponded to the gel state of the formulation. Higher amounts of surfactants seem to shift the gel formation region towards lower SOWRs, i.e., toward a higher amount of water. On the other hand, the lower amount of surfactant (SOR = 40%) gave rise to significantly lower moduli and gel strength, compared to higher SOR = 60% and 70%. The stronger gel was obtained with Labrafac
® and castor oil, giving
Pa (at SORs = 70% and 60% for SOWR = 50–70%); vitamin E acetate also exhibits a similar strength in the range with a dependence on the water amount, i.e., SOWR = 40–55% for SOR = 70% and SOWR = 50–65% for SOR = 60%.
As a last observation, using a monoglyceride as the oil phase—Capmul
®—shows a behavior that was significantly different when compared with the other three oils, generating a weak gel with a maximum of
Pa (SOR = 70% and SOWR = 62%), along with a changed transition behavior. This different behavior, compared with other oils, can be attributed to the fact that Capmul
® MCM C8 is less hydrophobic and presents a structure close to a lipophilic surfactant, even having an HLB~4.7 [
23]. As a result, it can lead to a less stable water/oil interface and a weaker gel. Herein, it can be seen that if the SOWR is further decreased to below 40%, the water amount becomes too high and the ternary system becomes a classical emulsion, as seen with the other oils.
A correlation between
Figure 2 and
Figure 3 can be achieved; the phase angle variations clearly follow the rise of the storage modulus
G′, confirming the formation of a hydrogel. Hence, the gel strength appears to be strongly dependent not only on the nature of the oil but also on the SOR and SOWR. A general trend suggested that the higher the surfactant amount, the stronger the gel strength. A second observation was that as the surfactant amount increased, the sol/gel transition was shifted towards a lower SOWR, i.e.,
nanoemulgels were formed with more water. However, these observations were not followed with Capmul
® oil.
Interestingly, a comparison with classical hydrogels gives the
nanoemulgels between 3% agarose hydrogels [
24] and 2% calcium reticulated alginate [
25], giving
G′ around 100 kPa and below 10 kPa, respectively. The visual and digital evaluations of these gels evidenced thick and solid-like gels for all
nanoemulgels described, except for Capmul
® systems, which behave as a soft gel, in accordance with their lower measured modulus.
To summarize, the rheological characterization emphasizes the formation of a gel that is relatively strong and spontaneously generated when mixing water/oil/surfactants according to a specific protocol and order. Without water, the oil/surfactant system is an oily liquid. Upon the addition of water, and for the lower amounts of water (high SOWR), the ternary system is still a liquid, gradually going to the sol/gel transition. Increasing the water concentration results in the formation of a strong hybrid gel where aqueous and lipid phases are blocked together and their interface stabilized with surfactants. This ternary system probably behaves as a gel because of the structuration of the aqueous and oil domains, in which interfaces mechanically interact with each other on small scales with a high specific surface. In this hypothesis, the larger the interfacial area—due to a larger surfactant amount—the stronger the gel. Finally, the further addition of water results in the gel’s dislocation, along with the formation and release of lipid nano-emulsion droplets. The impact of increasing the surfactant amount appears to be increasing the gel’s strength and shifting the gelation domain to a lower SOWR. It might be due to the higher capability of the entrap water phase due to the higher water/oil interfacial area. When SOR = 40%, the low amount of surfactant may not permit an efficient surfactant/water-binding network to form, explaining why the gels formed appear less strong in comparable water content. The last point arising from this series of results is the influence of the nature of oil that likely modifies the interactions between the oil phase and surfactant, impacting the surfactant’s efficiency.
3.2. Polydispersity Index and Size Analysis of the Nano-Emulsion-Based Systems
When nanoemulgels are maintained in contact with the aqueous phase, nano-emulsions are produced according to the spontaneous nano-emulsification process, exactly when water is immediately added.
The results are reported in
Figure 4, which demonstrate, as is commonly known, that the higher the quantity of surfactants, the smaller the particle size and PDI. The water amount in the gel, SOWR, does not significantly impact the nano-emulsion particle size, giving, with Labrafac
®, the following mean values:
nm,
nm, and
nm (according to the conventional notation
d = mean (standard deviation)), for SORs of 40%, 60%, and 70%, respectively. The PDI follows the same trend, showing better monodispersity for larger surfactant amounts—a classical behavior of spontaneous emulsification [
6]. In addition, the nano-emulsion size obtained by this two-step process, i.e., (i) the fabrication of
nanoemulgels and (ii) their dissolution, is similar to those obtained by using a large water amount, for instance SOWR = 20%, generating nano-emulsions in a single step [
6]. Along with the fact that the SOWR does not impact the particle size, it also signifies that we can consider
nanoemulgels an intermediate stage in nano-emulsion formation.
A comparison of nano-emulsion size distribution with the different oils used above is compared with a representative formulation (SOR = 60% and SOWR = 50%), giving hydrodynamic radii of 44 nm (PDI = 0.10), 135 nm (PDI = 1.00), 27 nm (PDI = 0.17), and 34 nm (PDI = 0.10), for Labrafac®, Capmul®, castor oil, and vitamin E acetate, respectively. Labrafac®, castor oil, and vitamin E acetate also showed very small particle sizes in a close range, whereas Capmul® emulsions revealed a rough polydisperse dispersion that cannot be considered a nano-emulsion as its PDI is equal to one. The absence of nanostructuration can explain the absence of strong gel formation. Hence, it is clear that nanoemulgel formation is directly related to its ability to arrange in a nanoscale architecture.
3.3. Release of Hydrophilic Model Dye from Nanoemulgels
In the previous section, we described the formation and rheological properties of nanoemulgels based on the formulation parameters. In this section, we propose to evaluate the influence of the formulation parameters, i.e., the influence of the gel properties, on the release of a model dye encapsulated in aqueous phase. Here, we chose to study a model molecule to understand the behavior of the drug delivery system and the impact of the formulation parameters on the release of these molecules. In comparison with marketed products with the same delivery profiles, e.g., subcutaneous form with long release profile, implants, microspheres, etc., the strengths of nanoemulgels are the simplicity of the formulation and the simplicity of the composition, without polymers or without the use of organic solvents in the formulation process.
Figure 5 presents the MB release over time from Labrafac
®-based formulations. As expected, the dye was rapidly released in the external medium through direct diffusion from the water phase of the emulsion (release time < 30 min). However, in some conditions, we clearly observed a slowdown of the release profile, with a 100% release significantly delayed for SOWR = 70% and SOR = 60% and 70%. In general, the release properties of a molecule are related to its properties, Mw, solubility, pKa, and environmental conditions. However, in the present study, as the release molecule (MB) is the same over all the studies, we assumed that the different release profiles were fully comparable since the gels have been formulated with the same compositions. In general, the slowest release profiles correspond to a lower amount of aqueous phase. If we assume that these
nanoemulgels are formed with aqueous and oil domains mixed as a bicontinuous phase in a submicron range, these profiles show the kinetics for the dye to escape such a network without speculating on the release mechanism. The formulations for which the SOWR = 70% are the ones with minimized aqueous phases and thus with the smallest water domains compared to oil ones, which possibly explains the extended time necessary for the dyes to escape the gel compared with the lower SOWR. On the other hand, slower release profiles also appear in intermediate SOWR values, e.g., for SOR = 40%/SOWR = 60% and SOR = 60%/SOWR = 50% in
Figure 5a,b, values of SOWR that seem to correspond to the sol/gel transition, as can be seen in
Figure 2a. Considering the structure of the gels, where the aqueous and oil domains are mixed in a submicron-scale range, the behavior of the gels is changing at higher water amounts, which might be due to the smallest scale of the gel structuration. Increasing in the water proportion may enlarge the water channels and water network, making it easier for the leakage of encapsulated hydrophilic molecules. Additionally, the fact that for the SOR = 40% and SOWRs = 60% and 70%, a prolonged release behavior was observed for the solute molecule, despite the absence of the formation of a strong gel, evidenced the decorrelation between the thickening behavior and the release kinetics, likely due to some other considerations like the bi-phasic organization of water and oil.
Release profiles with SOR = 70% demonstrated delayed behavior as compared with formulations with fewer surfactants, but they gradually slowed down with decreasing water amounts. In order to understand the release mechanisms, these release profiles are analyzed with the Korsmeyer–Peppas model in our case of unidirectional diffusion of releasing compounds:
where
is the fraction of dye released at the time
;
is the release rate; and
is the release exponent. In our case, the release exponent is an indication of the type of diffusion: Fickian when
and non-Fickian for
. Values of the release exponent are reported in
Figure 5d, showing the notable threshold at
with a dotted line. The mean values and standard deviations of the release exponent
are obtained by a statistical analysis of the experimental data. As the experimental data can be considered as a set of N samples (x
i, y
i), with individual errors on the measurements y
i, and considering the fitting power law of the form
, the mean and standard deviation of the parameters k and
, corresponding to the best-fitting power law, are estimated using a simple statistical analysis performed with MATLAB (R2024a) based on Monte Carlo sampling. These results emphasize an increase in
, with both the SOR and SOWR inducing a transition between Fickian and non-Fickian regimes. Basically, the Fickian regime can be associated with the direct escape of the dyes from a communicating and porous gel network with straight pathways, while the release profiles are non-Fickian when the molecules need to escape from a convoluted pattern where molecules travel a longer distance—compared to Fickian regime—before escaping the gel network. According to this hypothesis, at a constant surfactant amount, for a SOR = 40% or SOR = 60%, the increase in
with the corresponding decrease in water amount results in reducing the connection between the water veins, thus increasing the distance for the dye to escape. It appears that at higher surfactant amounts (SOR = 70%), or when increasing the surfactant amount (SOR = 40–60%), the water–oil interfacial area is higher, the scale range of the morphology is reduced, and the samples are more rapid in the non-Fickian regime. The more complex gel morphology might exist at the sol/gel transition, for which the release exponents reach a maximum, as raised above. Herein, we show that
nanoemulgels can release hydrophilic molecules. In addition, we can also assume that these gels can release lipophilic molecules through the release of nano-emulsion droplets. As shown in
Figure 4,
nanoemulgels released nano-emulsions as one droplet population with a narrow monodispersity. On the other hand, it is well acknowledged in the literature that nano-emulsions are efficient and stable carriers of lipophilic molecules [
26,
27], thus making
nanoemulgels a potential system to also release lipophilic molecules.
3.4. Impact of the Nature of the Oil and Water Phases on the Release Profiles
The impact of several other conditions on the release profile, changing oil nature or the addition of a thickener in water, was also evaluated. To this end, representative
nanoemulgel formulations were selected (SOR = 60%/SOWR = 50%), and MB release profiles were determined with different oils (Labrafac
®, Capmul
®, castor oil, and vitamin E acetate) with and without a thickener in the aqueous phase. The idea was to first compare the impact of the oil nature of the profiles, as performed in the rheological characterization and, secondly, to emphasize whether the modification of the properties of the aqueous phase can potentially delay the dye release in the gel network. The results are reported in
Figure 6, showing slight differences in the release profiles between the different oils (
Figure 6a) and not necessarily correlated with the strength of the gel or the sol/gel state. All of the formulations presented a slight release retention, but castor oil revealed a relatively slower release, with a time of 50% between 20 and 30 min but a time of 100% around 60 min, while all other formulations released 90% or more of their dye within 20 min. It is important to note that this experimental setup can be considered the worst case that can lead to faster release kinetics, and, in other conditions, e.g., subcutaneous, with less aqueous drainage, the release kinetics could likely be slower. From this observation, we can conclude that the morphology of the formed gel also depended on the nature of the oil phase, probably impacting the morphology and size of the bicontinuous oil/water network. Adding hydroxylethylcellulose (HEC) to water results in significant modifications of the profiles (
Figure 6b). HEC exhibited a basic hydrophilic moiety that can help it adsorb onto the interface. As a main observation, the presence of HEC in water reduces the burst release of MB, either due to the slowing of MB diffusion in aqueous channels or modifying the gel morphology. Notably, the addition of HEC to castor oil
nanoemulgels did not significantly modify the release profile of this formulation, with a total release after the same period (60 min). Its presence in the water phase does not modify the gel structure and suggests that HEC does not directly interact with the dye. The differences observed for the other formulations can be more reasonably attributed to changes in the gel morphology.