Takaichi Watanabea,
Yukitaka Kimurab and
Tsutomu Ono*a
aDepartment of Applied Chemistry and Biotechnology, Graduate School of Natural Science and Technology, Okayama University, 3-1-1 Tsushima-naka, Kita-ku, Okayama 700-8530, Japan. E-mail: tono@cc.okayama-u.ac.jp; Fax: +81 86 251 8083; Tel: +81 86 251 8083
bDepartment of Material and Energy Science, Graduate School of Environmental and Life Science, Okayama University, 3-1-1 Tsushima-naka, Kita-ku, Okayama 700-8530, Japan. E-mail: yktkkimu@cc.okayama-u.ac.jp; Fax: +81 86 251 8909; Tel: +81 86 251 8909
First published on 10th December 2013
A simple approach to preparing monodisperse poly(D,L-lactide) (PDLLA) microcapsules with a single aqueous core is developed. The method is based on automatic water-in-oil-in-water double emulsion formation from oil-in-water single emulsion via spontaneous emulsification which voluntarily disperses part of continuous aqueous phase into the dispersed oil phase dissolving oil-soluble amphiphilic diblock copolymer, poly(D,L-lactide)-b-poly(2-dimethylaminoethyl methacylate)(PDLLA-b-PDMAEMA), followed by coalescence of tiny water droplets within the polymer droplets, coupled with quick precipitation of polymers by diluting the emulsion with water. In this study, we have investigated the effect of PDLLA to PDLLA-b-PDMAEMA ratios and flow rates of each solution during preparing the emulsion on the final morphology and the size of the microcapsules. It was found that the polymer blend ratio played a crucial role in determining internal structure of the microcapsules. The microcapsules size decreased with the increment of the flow rate ratios of the continuous phase to the dispersed phase and eventually reached 10 μm, while maintaining narrow size distribution. In addition, we have demonstrated that the microcapsules can encapsulate both hydrophilic and hydrophobic compounds during the formation.
Regarding preparation of poly(D,L-lactide) (PDLLA) microcapsules with aqueous cores, many studies have been demonstrated by W/O/W emulsion-solvent evaporation method.12 In this method, the microcapsules are produced via evaporation of volatile organic solvent from the prefabricated complex emulsion droplets. The complex emulsion droplets are typically produced in a two-step emulsification, including water-in-oil (W/O) emulsion formation by emulsifying the inner aqueous phase in the middle oil phase, and the second emulsification with the W/O emulsion and the external surfactant aqueous solution. However, each emulsification process results in polydisperse emulsion droplets, leading to microcapsules with poorly controlled internal structures and broader size distribution.
Recently, coaxial microcapillary devices have enabled to produce highly controlled double emulsions and the microcapsules in a continuous manner.13 However, the fabrication of the devices is difficult because it requires shaping and manual alignment of microcapillaries. In addition, to keep periodicity of the double emulsions formation, one must precisely control the flow rates of the innermost, middle, and outermost fluids, which gives rise to limitation on production of smaller microcapsules less than 50 μm. In addition, after the droplet formation, it takes several minutes to hours to solidify microcapsules via evaporation of organic solvent, which is a rate-determining step in the whole process. Therefore, simple preparation of monodisperse PDLLA microcapsules with tunable size and internal structure using a simple microfluidic device is still challenging.
In this report, we present a straightforward method for the preparation of monodisperse PDLLA microcapsules with a single aqueous core without any operation of complex fluids. This technique is based on “droplet-to-particle” technology consisting of microfluidic emulsification and solvent diffusion, with some modification, previously reported by our group.14 Monodisperse oil-in-water (O/W) emulsion droplets were used in this study and oil-soluble amphiphilic diblock copolymer was introduced into the organic phase to induce spontaneous emulsification,15 dispersing part of continuous aqueous phase into the dispersed oil phase. This process has enabled to convert O/W emulsion into W/O/W emulsion automatically. By using ethyl acetate (EA) as the organic solvent, solidification of the droplets was easily achieved by just diluting the emulsion with excess volume of pure water because of high solubility of EA in water (8.3 wt% at 20 °C) and microcapsules with a single aqueous core were eventually formed. This quick removal of EA from the droplets shortens the preparation time about several hundreds times compared with conventional techniques using halogenated solvents. In addition, all components that we use in the microcapsule formation are biocompatible, which is great advantageous to applications for biomedical field. Furthermore, easy flow control using a simple Y-shaped microfluidic device makes it possible to prepare monodisperse microcapsules as small as 10 μm in the diameter. To the best of our knowledge, this is the first report to fabricate monodisperse PDLLA microcapsules with a single aqueous core using spontaneous emulsification via amphiphilic diblock copolymer and subsequent solvent diffusion.
200, Mw/Mn = 1.28) was used as the dispersed organic phase. The PDLLA to PDLLA-b-PDMAEMA ratio was varied from 10/0 to 6/4 (w/w). The flow rates of the dispersed phase and the continuous phase were 120 μL h−1and 1.5–15 mL h−1, respectively. For fluorescence microscopy, a trace amount of Nile red and 1 mg mL−1 of calcein were added to the dispersed phase and the continuous phase, respectively. The obtained O/W emulsion was transferred to a bath filled with 100 mL of ultra pure water through Teflon tubing (Φ = 0.5 mm, L = 20 cm) whose exit tip was submerged in the water. The EA was then rapidly removed from the droplet to the large volume of pure water by solvent diffusion under gentle stirring at 120 rpm, leading to precipitation of PDLLA microcapsules. The microcapsules were washed with ultra pure water three times by centrifugation (himac CF 15R, Hitachi, Japan) (3000 rpm, 3 min) to remove the surfactant and unloaded fluorescence agents, followed by freeze-drying overnight, resulting in dried PDLLA microcapsules.
200, Mw/Mn = 1.28) were dissolved in EA, which was used as the dispersed phase. PDLLA-b-PDMAEMA was synthesized by atom transfer radical polymerisation of DMAEMA initiated from PDLLA macroinitiator (Scheme S1 and Fig. S1 in ESI†) and used as a co-surfactant to induce spontaneous emulsification within the organic droplets. For the continuous phase, an aqueous solution saturated with EA dissolving 1 wt% w-PEG-b-PDLLA (Mn = 4400, Mw/Mn = 1.04) was used as a surfactant solution.14 As shown in Fig. 1, monodisperse O/W emulsion droplets were formed in a Y-shaped microfluidic device (Fig. S2 in ESI†). The resultant polymer droplets travelled towards the downstream of the channel, where water molecules are extracted from the continuous aqueous phase into the polymer droplets through the EA/water interface via spontaneous emulsification, resulting in forming W/O/W emulsion with multiple tiny aqueous cores. Then the W/O/W emulsion droplets were diluted with 100 mL of pure water in the precipitation bath, which facilitated extraction of EA into the outer aqueous phase and coalescence of tiny water droplets within each emulsion droplet. Fig. 2 shows optical micrographs illustrating the ripening of inner aqueous droplets in an oil droplet during solvent diffusion. It should be noted that because of high volatility of EA and quick diffusion of EA into the aqueous phase within a few seconds, it was difficult to observe the ripening process using microfluidic approach in situ. Therefore, instead of this, the monitoring was carried out using the model emulsion prepared by hand shaking on a glass slide covered with a thin cover slip. We induced the EA diffusion by adding pure water into the system step-by-step. The result clearly shows that as the solvent diffusion proceeds, the volume of the emulsion droplet (coloured blue) gradually decreases and simultaneously, tiny water droplets (white) in the emulsion droplet become bigger and eventually form a single core. When using microfluidic approach, it is considered that monodisperse microcapsules would be formed in the collection bath under the same process, within a few seconds after onset of solvent diffusion. The detailed mechanism will be discussed later.
We investigated the effect of blend ratio of PDLLA/PDLLA-b-PDMAEMA on microcapsule structures. Fig. 3 shows optical micrographs and SEM images of the resultant microcapsules prepared by varying PDLLA/PDLLA-b-PDMAEMA ratios. When using only PDLLA, we obtained monodisperse microparticles with a compact internal structure (Fig. 3(a)), whilst monodisperse microcapsules with dark contrast were obtained at the ratio of 9/1 (Fig. 3(b)). The dark contrast means that there are many cores inside the microcapsules. With an increase of the ratio to 6/4, we obtained the microcapsules with a single aqueous core (Fig. 3(c)). In addition, the diameter of the microcapsules was increased with increasing the copolymer ratio, indicating that the polymer droplets extracted water molecules from the outer aqueous phase during the process.
![]() | ||
| Fig. 3 Effect of the blend ratio of PDLLA/PDLLA-b-PDMAEMA on microcapsule structures. The ratio was (a) 10/0, (b) 9/1 and (c) 6/4 (w/w). | ||
Using the ratio at 6/4, we tuned the diameter of the microcapsules by altering the flow rate of the continuous phase (Qc) whilst maintaining that of the dispersed phase (Qd) constant. As shown in Fig. 4, the diameter of the microcapsules decreased from 32.6 to 10.3 μm with increasing the Qc. It was also found that the coefficient of variation (CV) values were increased as increasing the Qc, which was due to unstable emulsification in the device when the Qc was increased. However, the CV values were kept at less than 10% regardless of the flow conditions, indicating that the microcapsules had relatively narrow size distribution.
![]() | ||
| Fig. 4 Effect of the Qc on the microcapsule diameter and the CV value. The Qc was (a) 1.5, (b) 3.0, (c) 6.0 and (d) 15 mL h−1 while the Qd was fixed at 120 μL h−1. | ||
Fig. 5 shows the optical micrographs and the SEM images of the microcapsules prepared by changing the flow rates upon emulsification. The results clearly revealed that the microcapsules had a single aqueous core regardless of the microcapsule size. These results indicate that our technique can tune microcapsule size in a controlled manner. It is also important to note that our technique has enabled to produce monodisperse microcapsules with less than 30 μm in the diameter, which is almost impossible when using microcapillary devices since the microcapsule diameter greatly depends on the diameter of the initial W/O emulsion droplets. From the point of view of the microcapsule size, our technique is especially useful for preparing monodisperse microcapsules with less than 50 μm.
![]() | ||
| Fig. 5 Optical and SEM images of the microcapsules prepared by varying Qc at a fixed Qd (120 μL h−1). Qc = (a) 1.5, (b) 3.0, (c) 6.0 and (d) 15 mL h−1. The inset scale bars represent 25 μm. | ||
Furthermore, we achieved encapsulating both hydrophilic compound and hydrophobic compounds in the microcapsules. As a hydrophilic compound, we dissolved calcein in the continuous phase and a hydrophobic compound, Nile red, was also dissolved in the dispersed phase. As shown in Fig. 6, the core of the microcapsules clearly displayed green fluorescence derived from calcein, implying that the water-uptake from the continuous phase would occur during which the polymer droplets travel towards the downstream of the channel. On the other hand, the shell of the microcapsules showed red fluorescence attributing to Nile red, indicating that hydrophobic compounds dissolved in the dispersed phase can be easily entrapped within the shell of the microcapsules. It was confirmed that the time scale for the release of these encapsulated species in pure water at pH = 7 was long and the fluorescence of both dyes was observed for more than 3 days, implying that the release rate of the substances from the microcapsules is very slow. We will investigate the detailed release profile of the encapsulated substances in the future.
![]() | ||
| Fig. 6 Fluorescence micrographs of the microcapsules encapsulating calcein in the core and Nile red in the shell. (a) Bright image and (b) overlay image of green and red fluorescence filtered images. | ||
Finally, we propose the microcapsule formation mechanism. Monodisperse O/W emulsion whose interface is stabilized by w-PEG-b-PDLLA is firstly produced at the Y-junction in the microfluidic device (Fig. 7(i)). Then water molecules in the continuous phase are extracted to the polymer droplets in the device through the O/W emulsion interface to form tiny water droplets stabilized by amphiphilic PDLLA-b-PDMAEMA within the emulsion droplets. Since the copolymer dissolves in EA as a reverse micellar structure with a hydrophilic PDMAEMA core and a hydrophobic PDLLA shell so as to minimize the interfacial energy, the hydrophilic PDMAEMA core is capable of entrapping aqueous molecules (Fig. 7(ii)). After that, when the droplets are brought into contact with a great amount of pure water, EA diffusion from emulsion droplets to the surrounding pure water phase quickly occurs due to its higher solubility in water, leading to the volume reduction of the emulsion droplets. Simultaneously, an increase in PDLLA-b-PDMAEMA concentration within EA droplets causes microemulsion instability because of its lower solubility in EA, facilitating ripening of inner water droplets within each EA droplet (Fig. 7(iii)). Further evolution of the inner water droplets results in bigger ones, and eventually forms double emulsion (Fig. 7(iv)). Continuous diffusion of EA into the outer aqueous phase leads to precipitation of the polymers from the interface between the emulsion droplets and the continuous phase, yielding monodisperse microcapsules with a single aqueous core (Fig. 7(v)). Due to rapid extraction of EA into the precipitation bath, the transition from the emulsion to the solid microcapsules has completed within a few seconds, which is great advantageous to continuous production of the microcapsules using microfluidics.
![]() | ||
| Fig. 7 Proposed microcapsule formation mechanism using spontaneous emulsification and solvent diffusion. | ||
Footnote |
| † Electronic supplementary information (ESI) available: NMR spectra, optical microphotograph of the emulsion and fluorescence microphotograph of the microcapsules. See DOI: 10.1039/c3ra44066d |
| This journal is © The Royal Society of Chemistry 2014 |