Technical
November / December 2017

# Improved Solubility of Vitamin E by Means of Free-surface Microemulsion Electrospinning

Jeremy Lewis
Ahn Lam
Keith M. Forward
Cuong M. Nguyen

Free-surface electrospinning of microemulsions increases API solubility and may offer an alternative to batch powder processes.

The findings in this paper are research oriented. They are not intended to provide a method for immediate application.

According to the US Food and Drug Administration Biopharmaceutics Classification System, 90% of active pharmaceutical ingredients (APIs) are partially or totally insoluble in water due to their hydrophobic characteristics.1 As a result, a majority of APIs pass through the gastrointestinal tract without being absorbed into the bloodstream. 2 While proven approaches exist to combat this, drug manufacturers are sometimes forced to introduce large doses of an API into the pharmaceutical to compensate for its low solubility.

There are other challenges as well: Most API powders are packed into tablets using a batch fill-and-pack process.3These powders exhibit variable flow and packing properties depending on their densities and coatings, which adds uncertainty to the final product composition. 4 Moreover, when these granular materials undergo friction they can become electrically charged by a process called “triboelectric charging.” This can have unwanted effects. 5,6

To address the challenge of API solubility, new strategies are being explored:

•     Salt formation has proven successful in converting acidic and basic APIs into ionic salts, increasing solubility in polar solvents such as water. 7
•     Micronization decreases API domain size; this increases contact surface area and results in more surface interaction.8
•     Adding a surfactant to an API has been found to increase solubility in both polar and nonpolar solvents. Surfactants contain both aqueous- and organic-soluble components, which increase intermolecular interaction between poorly soluble APIs and the surrounding environment. 9

A combination of these methods is expected to be more effective than any single method alone.

## ELECTROSPINNING

Electrospinning is a novel process that combines surfactants, a decreased domain size, and an amorphous microstructurea combination of the methods mentioned above. Traditional API electrospinning has involved needle-based electrospinning, a process in which a charged solution containing API, solvents, and a polymer is injected through a needle to form an electrohydrodynamic jet. A Taylor cone forms in the presence of applied voltage, and the jet travels down field towards a grounded plate. Before the jet reaches the plate, solvents evaporate, forming an amorphous API entangled in a nanofibrous polymer-based mat. 10
Numerous studies have investigated the solubility of electrospun mats containing API. Nagy showed that the nanofibrous mats are 40% more soluble than the API alone. 11 Yu and Taepaiboon had similar results using APIs such as ibuprofen, sodium salicylate, diclofenac sodium, naproxen, and indomethacin. 12,13

Although needle-based electrospinning has proven an effective technique to improve API solubility, the process has limited productivity. To achieve significant amounts of electrospun material, the process must be operated for several hours, as the injection rate from the needle is usually less than 30 milliliters per hour (mL/h), or 1 fluid ounce per hour (fl oz/h). 14,15,16 To increase productivity, researchers have used multiple-needle configurations, 17,18 which have shown higher productivity compared to single-needle electrospinning. This process, however, has produced inconsistent fiber diameters and less uniformity within the mat.

A method that has been introduced but not widely studied is free-surface electrospinning, sometimes called needle-less electrospinning. Like needle electrospinning, free-surface electrospinning uses the applied potential between a polymer solution and a grounded plate to produce Taylor cone jets. In free-surface electrospinning, however, electrohydrodynamic jets are produced in a greater density than in needle-based electrospinning and are created from a free liquid surface.

“Electrospinning is a novel process that combines surfactants, a decreased domain size, and an amorphous microstructure.”

In this study, we consider a wired electrode rotating in a bath that holds the polymer solution. Droplets of solution form on the wires and then jet toward the grounded plate. This method increases productivity because multiple drops are able to form and jet from any exposed wire surface. 19 In addition, because the jets come from a homogenous solution, the fibers that form on the plate have the same composition. This eliminates variability that may occur in multiple-needle electrospinning, and maintains higher productivity than using a single needle.

Although API electrospinning has been done before in an effort to decrease domain size and shift the API to an amorphous phase, 10,11,12,13,14 limited studies have been performed on free-surface electrospinning of API with the addition of a surfactant. The surfactant decreases the API domain size by emulsification. The polymer serves as the excipient of the microemulsion and produces an amorphous microstructure matrix. In addition, Lin et. al. have shown that including a surfactant in the electrospinning solution produces a uniform mat composition by reducing the undesired beads-on-a-string morphology, which occurs when solvents become entangled in the nanofibers. 20

In this paper, we consider free-surface electrospinning of a microemulsion—a poorly soluble API (vitamin E) and surfactant (Kolliphor EL)—to yield mats that exhibit high solubility and uniformity. Free-surface electrospinning creates a fibrous product at a higher rate compared to needle electrospinning. Furthermore, the process is continuous and liquid phase, offering an alternative to batch powder packing methods and a new tool to manage insoluble APIs.

## MATERIALS AND METHODS

#### Chemicals

Chemicals were: vitamin E, EL-35, ethyl butyrate, reagent-grade ethanol, polyvinylpyrrolidone (PVP), polyethylene glycol (PEG), high-performance liquid chromatography (HPLC)–grade methanol, acetonitrile, glacial acetic acid, sodium acetate, and polysorbate 80. Deionized (DI) water was dispensed at a conductivity of 18.02 megaohms per centimeter (cm) (1.150 × 107 parts per million total dissolved solids) from a water-purification system. Molecular weights of the PVP and PEG were 1,300,000 and 35,000 daltons, respectively.

#### Microemulsion preparation

Vitamin E, EL-35, ethyl butyrate, ethanol, and water were mixed at 3%, 10.8%, 3%, 76%, and 7.2% by weight, as previously done by Feng 21 in a 7-cm-/2.8-inch (in.)-diameter jar. With a 1-cm (0.39-in.) diameter and 2.54-cm (1-in.) long stirring rod, the solution was mixed using a magnetic stirring plate at 350 revolutions per minute (rpm) for varying amounts of time. The solution was then sonicated* for varying amounts of time at a maximum amplitude of 30%. Immediately after sonication, PEG and PVP were added in varying weight percentages (wt%) relative to the aqueous phase, and mixed on the stirrer for several hours until a homogenous solution was produced.

* Sonication uses sound waves to break the microemulsion into small drop sizes.

#### Turbidity

Turbidity of the prepared emulsions were measured after both stirring and sonicating using a turbidimeter. Samples were prepared by diluting 0.10 mL (0.0034 fl oz) of microemulsion in 14 mL (0.47 fl oz) of DI water.

#### Setup and procedure

The microemulsion solutions were electrospun as per the free-surface electrospinning apparatus described by Forward. 19 An American wire gauge 36-gauge stainless steel wire was wrapped around two Teflon disks of a 10-cm- (3.9-in.)-long spinneret six times and held submerged in a fluid bath by two bolts. One end of the spinneret was attached to a drive belt powered by a direct current (DC) motor. Two power supplies were attached to the apparatus. One provided voltage to the DC motor and was held constant at 9.6 volts (V). The other power supply was connected to the solution bath and collection plate to maintain a 56-kilovolt potential between them. The collection plate consisted of aluminum foil wrapped around a square piece of plexiglass. The working distance was held constant at 40 cm (16 in.) above the solution bath by a ring stand and clamp insulated with polyvinyl chloride pipe and styrofoam tubing. The entire apparatus was enclosed in a plexiglass box and fed with dry air to maintain a relative humidity less than 10% and temperature of 21 °C (70 °F). Solutions were spun for times ranging from 5 minutes to 1 hour. A simple schematic of the apparatus is shown in Figure 1.

• 1. Ku, M. S., and W. Dulin. “Biopharmaceutical Classification-Based Right-First-Time Formulation Approach to Reduce Human Pharmacokinetic Variability and Project Cycle Time from First-in-Human to Clinical Proof-of-Concept.” Pharmaceutical Development and Technology 17, no. 3 (May-June 2012): 1–18.
• 2. Yalkowsky, Samuel H., ed. Techniques of Solubilization of Drugs: Drugs and the Pharmaceutical Sciences, New York: Marcel Dekker, 1981.
• 3. Yu, Lawrence. “Pharmaceutical Quality by Design: Product and Process Development, Understanding, and Control.” Pharmaceutical Research 25, no. 4 (2008): 781–791
• 4. Freeman, Reg “Measuring the Flow Properties of Consolidated, Conditioned, and Aerated Powders—A Comparative Study Using a Powder Rheometer and a Rotational Shear Cell.” Powder Technology 1, no. 2, (May 2007): 25-33.
• 5. Felton, L. A. “Pharmaceutical Powder Compaction Technology,” book review. Drug Development and Industrial Pharmacy 38, no. 8, (2012): 1029.
• 6. Forward, Keith M., et al. “Triboelectric Charging of Granular Insulator Mixtures Due Solely to Particle-Particle Interactions.” Industrial and Engineering Chemistry Research, 48, no. 5 (2009): 2309–2314.
• 7. Serajuddin, Abu T. “Salt Formation to Improve Drug Solubility.” Advanced Drug Delivery Reviews 59, no. 7 (30 July 2007): 603–616.
• 8. Vasconcelos, Teofilo, et al. “Solid Dispersions as Strategy to Improve Oral Bioavailability of Poor Water-Soluble Drugs.” Drug Discovery Today 12, no. 23, (2007): 1068–1075.
• 9. Volkering, F., et al. “Influence of Nonionic Surfactants on Bioavailability and Biodegradation of Polysyslic Aromatic Hydrocarbons.” Applied and Environmental Microbiology 61, no. 5 (1995): 1699–1705.
• 10. a. b. Brettmann, Blair, et al. “Solid-State NMR Characterization of High-Loading Solid Solutions of API and Excipients Formed by Electrospinning.” Journal of Pharmaceutical Sciences 101, no. 4 (2012): 1538–1545.
• 11. a. b. Nagy, Zsombor K., et al. “Comparison of Electrospun and Extruded Soluplus-Based Solid Dosage Forms of Improved Dissolution.” Journal of Pharmaceutical Sciences 101, no. 1 (January 2012): 322-332.
• 12. a. b. Yu, D. G., et al., “Ultrafine Ibuprofen-Loaded Polyvinylpyrrolidone Fiber Mats Using Electrospinning.” Polymer International 58, no. 9 (30 June 2009): 1010–1013.
• 13. a. b. Taepaiboon, Pattama, et al., “Drug-Loaded Electrospun Mats of Poly(vinyl alcohol) Fibres and Their Release Characteristics of Four Model Drugs.” Nanotechnology 17, no. 9 (2006). http://iopscience.iop.org/article/10.1088/0957-4484/17/9/041/meta
• 14. a. b. Kenawy, El-Rafaie, et al., “Release of Tetracycline Hydrochloride from Electrospun Poly(ethylene-co-vinylacetate), Poly(lactic acid), and a Blend.” Journal of Controlled Release 81, no. 1, (May 2002): 57–64.
• 15. Wnek, Gary E., et al., “Electrospinning of Nanofiber Fibrinogen Structures.” Nano Letters 3, no. 2, (2003): 213–216.
• 16. Rasekh, Manoochehr, et al., “Electrospun PVP-Indomethacin Constituents for Transdermal Dressings and Drug Delivery Devices.” International Journal of Pharmaceutics 473, no. 1, (October 2014): 95–104.
• 17. Dosunmu, O. O. Chase, G. G., Kataphinan W. “Electrospinning of Polymer Nanofibres from Multiple Jets on a Porous Tubular Surface.” Nanotechnology 17, no. 4 (2006): 1123–1127.
• 18. Yang, Ying, et al. “A Shield Ring Enhanced Equilateral Hexagon Distributed Multi-NeedleElectrospinning Spinneret.” IEEE Transactions on Dielectrics and Electrical Insulation 17, no. 5 (2010): 1592–1601.
• 19. a. b. Forward Keith M., et al., “Free Surface Electrospinning from a Wire Electrode.” Chemical Engineering Journal 183 (February 2012): 492–503.
• 20. Lin, Tong., et al., “The Charge Effect on Cationic Surfactants on the Elimination of Fibre Beads in the Electrospinning of Polystyrene.” Nanotechnology 15, no. 9 (13 August 2004).
• 21. Feng, J. L., et al., “Study on Food-Grade Vitamin E Microemulsions Based on Nonionic Emulsifiers.” Colloids and Surfaces: A Physicochemical and Engineering Aspects 339, no. 1 (May 2009): 1–6.
Figure 1: Free-surface electrospinning apparatus

#### SEM procedure

The morphology of the electrospun mats was investigated using scanning electron microscopy (SEM). Small samples of the mats were coated with 20 nanometers (nm) (7.9 × 10-7 in.) of gold and analyzed using a scanning electron microscope.

#### Vitamin E release rate and HPLC analysis

One hundred milligrams (mg) (0.0035274 oz) of electrospun mats were dissolved in 50 mL (1.7 fl oz) of acetate buffer solution at 21°C (70°F). The buffer contained 13% sodium acetate and 1.3% glacial acetic acid by weight in DI water. For comparison, 100 mg of electrospun mats were also dissolved in a buffer solution according to Taepaiboon 22 containing 13% sodium acetate, 1.3% glacial acetic acid, 0.5% polysorbate 80, and DI water by weight. Polysorbate 80 is a surfactant that improves vitamin E solubility in the buffer solution.

The mats were placed on an orbital plate at rate of 100 rpm. Over a period of 36 hours, 0.5 mL (0.017 fl oz) of test solution was removed at selected times and replaced with 0.5 mL of buffer to maintain a constant volume.

An HPLC instrument with a 5-micrometer particle size and 150 × 4.6 mm column was utilized to determine the concentration of the collected samples. The mobile phase was composed of 48:48:4 parts by volume of acetonitrile/ methanol/ DI water. The elution rate was set to 1 mL/minute (0.034 fl oz/minute). Injection volume was set at 100 microliters (3.38 × 10-5 fl oz), with an ultraviolet light absorption at a wavelength of 295 nm. Peaks were shown at approximately 27 minutes. Calibration curves for the buffer solution with and without polysorbate 80 accounted for concentrations between 0 and 3.2 grams/mL (between 0 and 0.21 pounds/fl oz) of vitamin E, and were used to determine the concentration of vitamin E dissolved in solution.

#### Measuring productivity

To measure productivity, the collection plate was weighed before and after electrospinning at different spin times with constant parameters and a constant electrode length of 10 cm (3.9 in.). The mass difference was divided by the spin time and the electrode length to obtain productivity defined as mass per time per centimeter of electrode.

• 22. Taepaiboon, P., U. Rungsardthong, and P. Supaphol, “Vitamin-Loaded Electrospun Cellulose Acetate Nanofiber Mats as Transdermal and Dermal Therapeutic Agents of Vitamin A Acid and Vitamin E.” European Journal of Pharmaceutics and Biopharmaceutics 67, no. 2 (31 March 2007): 387–397.
Figure 2: Turbidity of microemulsion as a function of stir time
Figure 3: Emulsion turbidity as a function of sonication time. The highest turbidity occurred at 45 seconds, indicating the smallest possible drop size

## RESULTS AND DISCUSSION

#### Stir and sonication time

To obtain a stable microemulsion, solutions were mixed to obtain a similar Weber number of 1400, as defined by equation 1:

$$We = \frac{(pv^2L)}{σ}$$

In this equation, ⍴ is the solution density, v is the speed of the stir bar, L is the stir bar length, and σ is the surface tension between the organic and aqueous phases. This unitless number was held constant throughout this paper; it would be an important parameter if the process is up scaled-up in the future.

Reddy and Fogler have proposed that a microemulsion is stable when the turbidity of that solution remains constant.23 Figure 2 shows the turbidity stabilized at 60 minutes of stirring, indicating a stable microemulsion. After 60 minutes, the turbidity indicated periods of instability, likely a result of coalescence and separation among drops.

Because the solution concentration remained constant, turbidity served as a relative measure of drop size. To find the smallest possible drop size, which is thought to result in maximum dissolution, the microemulsion was sonicated for varying amounts of time. Figure 3 depicts a peak in turbidity at 45 seconds of sonication, indicating that the solution was stable and had the smallest drop size possible at that point. These conditions suggest high dissolution and uniformity in the final product.

#### Polymer concentration

Water, ethanol, and ethyl butyrate were used as solvents. They evaporate from the microemulsion during electrospinning, leaving the polymers, EL-35, and vitamin E to form a fibrous mat that can be rolled into a pill, 24 dissolved in fluid, or processed into a finished dosage form by thin-film techniques. 25 In this paper, the mat was simply removed from the foil and treated as a thin film without further processing.

Free-surface electrospinning of a microemulsion yields mats that exhibit high solubility and uniformity.

Electrospun microemulsions with effective polymer concentrations of 6 wt% PVP and 9 wt% PEG relative to the aqueous phase produced high-quality mats (Figure 4A). An SEM image of these mats confirmed uniform fiber thickness and desired amorphous structure (Figure 4B).

Higher polymer concentrations yielded highly viscous solutions; these produced relatively large Taylor cone drops that were unable to jet. Lower PVP and PEG concentrations produced low viscosities that yielded minimal drop formation on the wired electrodes. When compared to the solution at effective polymer concentration, solutions with both high and low viscosities resulted in (Figure 4C) solvent splattering and thin mats that failed to maintain mechanical integrity.

• 23. Reddy, S. R., and H. S. Fogler. “Emulsion Stability: Determination from Turbidity.” Journal of Colloid and Interface Science 79, no. 1 (1981): 101–104. https://deepblue.lib.umich.edu/bitstream/handle/2027.42/24510/0000787.pdf
• 24. Trout, Bernhardt Levy, et al. “Layer Processing for Pharmaceuticals.” US Patent US9205089 B2 (8 Dec. 2015). https://www.google.com/patents/US20120305174
• 25. Karki, Sandeep, et al. “Thin Films as an Emerging Platform for Drug Delivery.” Asian Journal of Pharmaceutical Sciences 11, no. 5 (October 2016): 559–574.
Figure 4: (a) Macroscopic view of desired mat (6% PVP, 9% PEG). All of the mat showed a thick layer of fiber formation. (b) SEM of desired mat (6% PVP, 9% PEG). (c)Macroscopic view of undesired mat example. Much of the mat showed areas of no fiber formati
THE MORPHOLOGY OF THE ELECTROSPUN MATS WAS INVESTIGATED USING SEM
Figure 5: HPLC calibration curve for vitamin E dissolved in three different buffers

#### HPLC calibration

HPLC calibration was used to quantify the vitamin E concentration in three different acetate buffers at 21°C (70°F). When vitamin E was introduced into the buffer solution without surfactant polysorbate 80 and allowed to mix for a substantial amount of time, vitamin E was almost undetectable in the sample solution. When vitamin E was introduced into a buffer containing either EL-35 or polysorbate 80, substantial amounts of vitamin E were detected. This identifies the importance of a surfactant in the dissolution process.

The HPLC calibration curve indicated the quantity of vitamin E in the mats (Figure 5). Release (dissolution) rates were based on the assumption that the mats contained only vitamin E, EL-35, and polymers at 10.5/38/51.5 wt%. These percentages were determined from the microemulsion composition without ethyl butyrate, ethanol, and water, which are expected to have evaporated during electrospinning. The mat was dissolved in buffers, both with and without polysorbate 80. The buffer with polysorbate 80 serves as a control, since it is known that vitamin E in the mat will dissolve completely in the presence of a surfactant.

The mats showed similar release characteristics in both solutions, reaching 100% dissolution within 16 minutes. This indicates that EL-35 was successfully incorporated into the mats and increased dissolution of vitamin E without the need for an additional surfactant.

These results were compared to cast-film microemulsion—in which the polymeric solution is left to dry into a film without being electrospun—and pure vitamin E, both of which were dissolved in the buffer without added polysorbate 80. The electrospun fibers showed much higher release rates than either the cast film or the pure vitamin E, indicating that electrospinning successfully increased the solubility of vitamin E (Figure 6).

Figure 6: Cumulative release of the nanofibrous mats in a buffer containing polysorbate 80, a buffer containing no added surfactants and cast film

#### Fickian diffusion

Fickian diffusion is a common mechanism used to describe the release characteristics of drugs in polymer carriers. The Higuchi equation is a simple but accepted way of verifying diffusion. The simplified Higuchi model is shown in equation 2:

$$\frac{M_t}{M_∞} = K√{t}$$

where M∞ is the cumulative absolute amount of drug released at infinite time, Mt is the cumulative absolute amount of drug released at time t, and K is a constant relating the system concentration and diffusivity. 26 Plotting release percentage versus the square root of time should yield a linear line with a slope of K.

The electrospun mat dissolved in a buffer with added polysorbate 80 had a K value of 37.15 min–0.5; the mat dissolved in the buffer without polysorbate 80 had a K value of 36.71 min–0.5. The cast film had a K value of 1.88 min–0.5. This verified that EL-35 improved the vitamin E release rate.

• 26. Siepmannm J., and N. A. Peppas. “Modeling of Drug Release from Delivery Systems Based on Hydroxypropyl Methylcellulose (HPMC).” Advanced Drug Delivery Reviews 48, no. 2–3 (June 2001): 139–157.
Figure 7: Fickian diffusion region for mat dissolution in buffers with polysorbate 80, without polysorbate 80, and cast-film emulsions
Figure 8: Production per length of electrode
This indicates freesurface electrospinning of microemulsions containing api and a surfactant is an effective method to increase api solubility.

#### Productivity

Figure 8 shows the production per length of electrode and productivity of free-surface electrospinning defined by equation 3, where:

$$Q m/tl$$

and m is the mass of the mat produced after time t per electrode length l. Maximum productivity of 0.29 mg/cm-min (1.6 × 10–6 lb/in.-min) occurred at 40 minutes of electrospinning.

After 40 minutes, the solution became highly viscous as the solvents evaporated from the exposed free surface—a phenomenon called “solution aging.” 19 When the solution aged, large drops formed on the wired electrodes but were unable to jet due to an increase in viscous forces. The productivity of 0.29 mg/cm-min (1.6x10–6lb/in.-min), however, was significantly higher than needle-based electrospinning, which typically produces this quantity of fibers on a scale of hours.

## CONCLUSION

An effective concentration of polymer mixture was determined based on macroscopic and microscopic characteristics of the nanofibrous electrospun mats that were produced. In addition, EL-35 was successfully incorporated into the mats and effectively increased the dissolution of vitamin E. The electrospun mats showed higher release characteristics compared to cast-film emulsions, and productivity was found to be higher than needle-based electrospinning. This indicates free-surface electrospinning of microemulsions containing API and a surfactant is an effective method to increase API solubility.

### Acknowledgments

This research was supported by the California State University Program for Education and Research in Biotechnology (CSUPERB). The SEM in the University of California, Los Angeles Department of Earth, Planetary, and Space Sciences used in these analyses was funded through the Army Research Office’s Defense University Research Instrumentation Program (DURIP).