Nanoemulsion‐Loaded Capsules for Controlled Delivery of Lipophilic Active Ingredients

Abstract Nanoemulsions have become ideal candidates for loading hydrophobic active ingredients and enhancing their bioavailability in the pharmaceutical, food, and cosmetic industries. However, the lack of versatile carrier platforms for nanoemulsions hinders advanced control over their release behavior. In this work, a method is developed to encapsulate nanoemulsions in alginate capsules for the controlled delivery of lipophilic active ingredients. Functional nanoemulsions loaded with active ingredients and calcium ions are first prepared, followed by encapsulation inside alginate shells. The intrinsically high viscosity of the nanoemulsions ensures the formation of spherical capsules and high encapsulation efficiency during the synthesis. Moreover, a facile approach is developed to measure the nanoemulsion release profile from capsules through UV–vis measurement without an additional extraction step. A quantitative analysis of the release profiles shows that the capsule systems possess a tunable, delayed‐burst release. The encapsulation methodology is generalized to other active ingredients, oil phases, nanodroplet sizes, and chemically crosslinked inner hydrogel cores. Overall, the capsule systems provide promising platforms for various functional nanoemulsion formulations.

where is the viscosity of the liquid, is the density of the liquid, is the diameter of the liquid droplet, and is the surface tension of the liquid.
For alginate bead formation (dripping an alginate solution into a CaCl 2 solution), prior work found that the ℎ of alginate solution had to exceed 0.24 (a critical value) to overcome the impact and drag forces for forming spherical beads. [1] However, for alginate capsule formation (dripping a calcium precursor into an alginate bath), the critical value of ℎ should be higher than 0.24 because of the large viscosity of the alginate bath. In this work, the calcium nanoemulsion has a high viscosity (301 mPa-s) and a low surface tension (35.7 mN/m). Both of these properties favor a large value of ℎ , and the ℎ for dripping the nanoemulsion with 18G and 22G dispensing tips are 0.79 and 0.89, respectively (Table S1).

S.4. Design of Capsule Dimensions
The radius of a capsule can be estimated by the sum of the inner core radius and the shell thickness. In this work, the capsule thickness and size are easily controlled by varying the calcium concentration and the dispensing tip size, respectively. As shown in Figure 2h, the shell thickness is linearly correlated with the calcium concentration. Further extrapolation of this linear correlation ( Figure S4) shows that it passes through the origin, which is reasonable because no shell can be formed without CaCl 2 being added. For the lower limit, the shell thickness should be further decreased by decreasing the 2 until the point where the capsules are too fragile to be collected. For the upper limit, the linear correlation is extrapolated up to the point where the continuous phase is saturated with CaCl 2 (star symbol).
The star symbol corresponds to the theoretically maximum CaCl 2 concentration (

,
) that can be achieved in the nanoemulsion system. With the solubility of CaCl 2 in water (74.5 g/100 mL at 20°C) [2] and the water volume in the continuous phase ( Figure S4. Linear correlation between shell thickness and 2 with extrapolation to both saturated conditions and zero 2 . Since we use a dripping method to generate droplets, the inner core radius ( ) can be described by Tate's law [1] : where is the outer diameter of the dispensing tip, is the surface tension of the nanoemulsion, is the density of the nanoemulsion, is the acceleration of gravity (9.8 m 2 /s). Figure S5 shows the linear correlation for the origin and the four data points (five points in total, R 2 = 0.99). This correlation can provide a criterion for designing the inner core radius of capsules. Alternatively, one could use centrifugal forces to produce even smaller droplets and hence inner cores. We have a previous publication which describes the centrifugal particle generation process in detail. [3]

S.5. Fitting the Bursting Events with Cumulative Distribution Functions
The bursting events occurring in the bursting regime (R2) should be statistically random, and the release profile should follow the behavior of a cumulative distribution function (CDF). To fit a CDF to the bursting regime (R2) of the release profile, we extracted the release profile R( ) from 10% to 100% (excluding the early diffusion release), and rescaled this 90% release into a CDF ( ) ranging from 0 to 100%. The ( ) represents the cumulative probability of capsule bursting. The ( ) is then transformed into ( ) with = − to shift the distribution mean to = 0. With the above setting, the only unknown parameter required to be fitted is the standard deviation in the following CDF ( ): The results of the CDF fitting are shown in Figure S8a-d. When the 2 or the dispensing tip size increases, a larger is observed. The fitted CDF can be further transformed into a probability density function ( ) (PDF) to represent the release rate in the bursting regime.
The ( ) is transformed back to ( ), which is further rescaled back to R( ). Figure S8e shows the release rate profiles of capsules for different preparation conditions. The PDF accurately follows the bursting release rates calculated from the experimental data.

S.6. Preparation of Alginate Beads for Nanoemulsion Encapsulation
Nanoemulsion-loaded alginate beads were prepared by dripping alginate nanoemulsions into a calcium gelation bath ( Figure S9a). The alginate nanoemulsions with three different alginate concentrations show droplet sizes between 50 to 55 nm ( Figure S9b). The droplet size slightly decreases as the alginate concentration increases, because the continuous phase becomes more viscous, which provides a larger shear to create smaller droplets. The nanoemulsion-laden alginate beads are shown in Figure S9c

S.7. Release Behavior of Nanoemulsion-Loaded Beads
With the linear calibration curve (Figure 3c), release tests are conducted to obtain the release profiles of beads for the three different alginate concentrations (Figure S10a). w/v, respectively. Finally, the release process enters the R3 as the absorbance signal reaches a saturated value indicating complete release of the cargo. Figure S10b shows the release rates for the three different alginate beads. For each alginate concentration, the release rate increases and develops a peak after the release mechanism transitions to the R2 regime. In the R2 regime, the alginate beads degrade significantly, which accelerates the nanoemulsion release. The sharpness of the peaks depends on the crosslinking density of the alginate beads.
The alginate beads prepared from a lower alginate concentration possess a lower crosslinking density, which leads to a sharper peak of the release rate. Figure S10c shows the retention of the nanoemulsion suspension in beads for different preparation conditions. The retention rate increases as the alginate concentration increases, and the best retention is about 58.8% for the 4% w/v alginate beads. To further analyze the release profiles, Equation S7 and Equation S8 based on Peppas power law are used to fit the R1 and R2 regimes, respectively. [4] (%) = (S7) where k is a geometric constant for a hydrogel system, and n is the diffusional exponent representing the release mechanism, (%) is the (%) at . For spherical matrices, the values of n = 0.43, 0.43 < n < 0.85, n = 0.85, and n > 0.85 represent Fickian diffusion, anomalous (non-Fickian) transport, Case II transport, and Super Case II transport, respectively. [4] The results of the power law fitting are shown in Figure S10d-f. The exponent n values are 0.43 < n < 0.85 and n > 0.85 for the R1 and R2 regimes, respectively. This indicates that the R1 and R2 regimes belong to anomalous (non-Fickian) transport and Super Case II transport, respectively. In the R1 regime, the release is mainly diffusion-controlled with some anomalous behavior due to relaxation and erosion. In contrast, the release is mainly erosion-controlled in the R2 regime. [5] Figure S10. a) Release profiles using a USP Dissolution Apparatus II of beads for different preparation conditions. The schematic images below the release profiles depict the release mechanisms for different regimes corresponding to the = 2% w/v (which also apply to other curves). R1, R2, and R3 are diffusion-controlled, erosion-controlled, and post-release regimes. b) Release rates (∆ ∆ ⁄ ) of beads for different preparation conditions. c) Retention of the nanoemulsion suspension in beads for different preparation conditions. d-f) Fitting the release profiles with time-dependent power law functions: d) = 1% w/v, e) = 2% w/v, and f) = 4% w/v.