Open Access Article
Jared S.
Cobb
,
Chipo
Chapusha
,
Jui
Gaikwad
,
Joshua
Michael
and
Amol V.
Janorkar
*
Biomedical Materials Science, University of Mississippi Medical Center 2500 N State Street, Jackson, MS 39216, USA. E-mail: ajanorkar@umc.edu; Fax: 601-984-6087; Tel: 601-984-6170
First published on 19th April 2021
This work demonstrates the use of canola oil as a biologically compatible solvent for a thiol–ene driven suspension polymerization that effectively creates micron-sized particles. A 1
:
1 thiol to alkene functional group stoichiometry was maintained for all formulations. Thymol was incorporated into these particles during synthesis as a naturally derived, lipophilic drug with known anti-microbial properties and use in agriculture as a pesticide. UV-Vis spectroscopy demonstrated the removal of surfactant and any excess reagents to be 99% effective. Particle size was dependent on stirring speed with faster stirring resulting in smaller particles. The release of thymol from the selected ten formulations was measured with UV-Vis spectroscopy at a wavelength of 275 nm for ten days and was shown to be tailorable by altering the amount of crosslinker and surfactant used during the creation of the particles. A smaller molar ratio of 2-functional thiol monomer compared to 4-functional thiol monomer (0.1
:
0.9) crosslinker and a higher molar ratio of 2-functional alkene monomer compared to mono-functional alkene surfactant (0.9
:
0.1) resulted in the slower release of thymol. A higher molar ratio of 2-thiol to 4-thiol monomer (0.8
:
0.2) and a lower molar ratio of 2-alkene to surfactant (0.5
:
0.5) resulted in faster-releasing particles. The Peppas–Sahlin equation used to model the thymol release found that a super case II drug release mechanism that relies on physical interactions between the polymer chains determined the amount of thymol released. The utility of canola oil as a solvent for a polymer particle delivery system holds promise for their use in environmentally sensitive applications such as a pesticide carrier and anti-microbial products.
Thiol–ene based polymers show promise for drug delivery applications. They display distinct advantages such as delayed gel points with high monomer transformation, controlled mechanical and structural properties, and almost no oxygen inhibition.8 The advantages of the thiol–ene based polymers come from a step-growth free-radical mechanism. A thiyl radical is generated from the abstraction of a hydrogen by the photoinitiator. This radical then adds across the double bond of an alkene monomer generating a β-thioether carbon-centered radical; the latter radical rapidly undergoes chain transfer with an additional thiol to regenerate the thiyl radical species. The thio–ether radical, in the absence of the propagation step, chain transfers to another thiol monomer. The coupling of two radicals terminates the reaction. Oxygen does not inhibit the reaction; instead, the oxygen molecule produces a peroxy radical that then chain transfers with a thiol, regenerating the propagating species.8–10 The thiol–ene reaction is probably best known for the formation of near-perfect crosslinked polymer networks, wherein polymerization with high conversion is reached before gelation and results in a homogeneous medium to produce equally crosslinked networks with well-defined physical and mechanical properties.8–10 By varying the numerous combinations of different multifunctional thiol and alkene monomers, the mechanical properties and distance between crosslinks for these systems can be easily tuned. Recent work has shown the viability to use thiol–ene systems via either suspension or emulsion polymerizations to create particles with controllable sizes.11–16
Durham et al. were the first to show that a waterborne photopolymerization reaction could occur between a multifunctional alkene and thiol monomers to create particles. They used a suspension polymerization with sodium dodecyl sulfate (SDS) surfactant to react a 3-functional alkene to a 4-functional thiol with toluene or chloroform as the diluent. They were able to synthesize particles with a size range of 10 to 200 μm by altering the rate of stirring, the ratio of solvent to monomers, and the amount of surfactant.11 Zhang et al. used a similar synthesis route, but incorporated a linear poly(methylmethacrylate) molecule to disrupt the crosslinking of the thiol–ene network. The work resulted in particles that were larger than 200 μm in diameter and with controllable pore sizes.13 D. V. Amato et al. built on Durham et al.'s work to demonstrate the viability of using the mini-emulsion technique to create nanometer-sized particles with excess functional groups on the surface of the particles that could be used for subsequent functionalization.14 D. N. Amato et al. further expanded on this work to show that thiol–ene mini-emulsions could be used to encapsulate thymol dissolved in carvacrol. They further demonstrated that an alkene functionalized surfactant could be reacted into the particles to minimize the amount of surfactant needed to be removed.15 More recently, Shipp et al. demonstrated the viability to form linear and lightly crosslinked emulsions via thiol–ene reaction.16 While these examples are specific to the work presented in this paper, a more thorough review of particles formed via thiol-x chemistries can be found elsewhere.17
While the previous studies with photo-initiated waterborne thiol–ene polymerizations have established synthesis routes to form controllable particle sizes and tailorable pore sizes, the bulk of these studies either focused on micro and mini-emulsions, have not investigated the ability of the formed particles to carry a drug payload, use harsh solvents, or cannot remove the stabilizing surfactant. Our work builds on previous research while addressing these deficiencies by using canola oil as the diluent solvent, altering the monomer and surfactant ratio while maintaining 1
:
1 functional group stoichiometry to control particle size, efficiently removing excess reactants and surfactant, and demonstrating tailorable release profiles for thymol.
:
0.0 to 0.5
:
0.5 in 0.1 increments and by changing the ratio of 2-thiol to 4-thiol functional groups from 0.9
:
0.1 to 0.5
:
0.5 in 0.1 increments. A 1
:
1 stoichiometry of the SH and ene functional groups was maintained for all formulations by increasing the amount of 1-ene Hitenol KH as the amount 2-ene is lowered. The full list of the resulting 30 formulations used in this study can be seen in the ESI† (Fig. S1). These formulations allowed for the systematic alteration in both chemical structures, as well as the amount of crosslinking present in the particles. The water phase consisted of de-ionized (DI) water and Hitenol KH (25 wt% in water). After adding the organic phase to surfactant/water, the resulting suspension was stirred at either 600 rpm (hereafter referred as half-speed suspension, HSS) or 1200 rpm (hereafter referred as full-speed suspension, FSS) for 10 minutes on a stir plate (Corning PC-220, Corning, NY). The emulsified solution was then cured using UV light (365 nm, 40 mW cm−2, Omnicure S1500, Ecelitas, Waltham, MA) for 10 minutes.
:
0.7, 0.4
:
0.6 and 0.5
:
0.5 were too delicate and did not survive the centrifugation steps.
![]() | ||
| Fig. 3 Thymol release for the two formulations without the Hitenol KH using half speed suspension polymerization (HSS). Error bars represent 95% confidence intervals (C.I.). | ||
We used two purification processes to remove the excess reagents. The first was to centrifuge the reacted particles using seven centrifugation cycles described in the methods section. The second was to centrifuge for three cycles and soak the particles for 24 hours in DI water before performing one final centrifugation cycle. Both processes resulted in 99% removal of unreacted reagents (Fig. S3, ESI†). Due to the delicate nature of the particles, seven centrifugation cycles sometimes resulted in a portion of the particles becoming broken, so the fewer centrifugation cycles with soaking was preferred. Table 1 contains the formulations that survived the purification process. After purification, the control particles prepared using butyl acetate and no thymol were soaked in PBS for five days, and the absorbance was measured to determine if any reagents leached from the particles (Fig. S4a, ESI†). A small amount of butyl acetate was observed in the PBS, as indicated by the UV-Vis spectrum matching that of butyl acetate (Fig. S4b and S5, ESI†). During purification, it was found that the formulations with a 2-thiol to 4-thiol ratio of 0.3
:
0.7, 0.4
:
0.6, 0.5
:
0.5 (Fig. S1, ESI†) resulted in particles that did not survive the purification process (Fig. 2b) as the particles were too delicate, likely due to the lower amount of the 4-thiol crosslinker.
| # | Ene : Hitenol functional group ratio |
2-Ene (μL) | 2-Thiol (μL) | 4-Thiol (μL) | Hitenol (mL) | Water (mL) |
|---|---|---|---|---|---|---|
2-Thiol : 4-thiol functional group ratio 0.1 : 0.9 |
||||||
| 1 | 1.0 : 0.0 |
89.0 | 6.4 | 272.5 | 0.0 | 10 |
| 2 | 0.9 : 0.1 |
78.5 | 6.4 | 272.5 | 0.4 | 9.6 |
| 3 | 0.8 : 0.2 |
68.1 | 6.4 | 272.5 | 0.8 | 9.2 |
| 4 | 0.7 : 0.3 |
57.6 | 6.4 | 272.5 | 1.2 | 8.8 |
| 5 | 0.6 : 0.4 |
47.1 | 6.4 | 272.5 | 1.6 | 8.4 |
| 6 | 0.5 : 0.5 |
36.6 | 6.4 | 272.5 | 2.0 | 8 |
2-Thiol : 4-thiol functional group ratio 0.2 : 0.8 |
||||||
| 1 | 1.0 : 0.0 |
89.0 | 12.8 | 242.3 | 0.0 | 10 |
| 2 | 0.9 : 0.1 |
78.5 | 12.8 | 242.3 | 0.4 | 9.6 |
| 3 | 0.8 : 0.2 |
68.1 | 12.8 | 242.3 | 0.8 | 9.2 |
| 4 | 0.7 : 0.3 |
57.6 | 12.8 | 242.3 | 1.2 | 8.8 |
| 5 | 0.6 : 0.4 |
47.1 | 12.8 | 242.3 | 1.6 | 8.4 |
| 6 | 0.5 : 0.5 |
36.6 | 12.8 | 242.3 | 2.0 | 8 |
Fig. 2b shows that the broken particles formed a core–shell architecture where the polymer entraps the bulk of the thymol/canola oil. Butyl acetate was used to prepare control particles (Fig. 2c) as it has been successfully used previously in creating polymer particles and has a limited solubility in water at 6.8 mg mL−1. The control particles were used to show that the monomer combinations would result in the removal of unreacted reagents, and to show that the particles are not leaching reagents over time. The same formulations were made using thymol dissolved in canola oil as a diluent, which resulted in the successful creation of thymol-containing polymer particles (Fig. 2d). There have been several previous studies that have reacted thiols to the alkene in presence of fatty acids using UV light free radical reactions to create coatings. These studies have shown that the fatty acids have very low reactivity towards thiols and need very specific conditions to react; namely, a thiol to ene ratio greater than 3, negative temperatures to decrease the faster thiol–thiol coupling reaction, large amounts of photoinitiator (0.1
:
1 initiator molecule to alkene group), a nitrogen atmosphere to prevent competitive oxidation, hours to days to reach low to high conversions, and/or chemical modification of the fatty acid prior to the thiol–ene reaction. Given that our reactions happen in 10 minutes at ambient conditions, we suspect there to be negligible reaction between the thiol monomers and the alkene group of the canola oil.18–20
Fig. 4 shows the resulting particle sizes for the formulations varied with the thiol, ene, and surfactant composition. Overall, the higher amount of 4-thiol results in statistically larger particles. As shown in Fig. 4, all 0.1
:
0.9 2-thiol to 4-thiol formulations yield larger particles than the 0.2
:
0.8 formulations (p ≤ 0.05). For example, with the control particles, the 2-thiol to 4-thiol ratio of 0.1
:
0.9 uses the higher amount of 4-thiol and results in larger mean particle diameter (161 ± 3 μm, 2-ene0.9Hitenol0.12-thiol0.14-thiol0.9) compared to the 2-thiol to 4-thiol ratio of 0.2
:
0.8, which uses lower amount of 4-thiol and results in a smaller mean particle diameter (92 ± 2 μm, 2-ene0.9Hitenol0.12-thiol0.14-thiol0.9) (p ≤ 0.05). When the particles are synthesized with thymol and canola oil instead of butyl acetate at 1200 rpm (FSS) the same trends hold. When 2-ene0.5Hitenol0.52-thiol0.14-thiol0.9, the formulation results in a mean particle diameter of 149 ± 3 μm and when the amount of 4-thiol is decreased 2-ene0.5Hitenol0.52-thiol0.24-thiol0.8 the particle diameter decreases to 79 ± 4 μm. When the stirring speed is decreased for the thymol and canola oil particles to 600 rpm (HSS), the mean diameters increase. This can be seen by comparing the mean diameter for the 2-thiol to 4-thiol samples with a ratio of 0.1
:
0.9 and 0.2
:
0.8 for both HSS and FSS. An example is 2-ene0.8Hitenol0.22-thiol0.14-thiol0.9, where the HSS reaction resulted in a mean particle diameter of 384 ± 19 μm compared to the 114 ± 1 μm particles obtained in the FSS reaction. The slower stirring was expected to result in larger particles, as this has been demonstrated before by Durham et al.11
, which gives the percent of Fickian diffusion.21 The Fickian diffusion mechanism is driven by thermodynamic forces such as concentration gradients and chemical potential and typically exhibits near-constant diffusion of the eluents that follow Fick's first and second laws of diffusion.22 The value for m was chosen as 0.43 for a sphere.21 The calculated values for F were zero or very close to zero, indicating that the mechanism for drug release was not Fickian diffusion, but super case II (Table 2).
| Formulation | 2-Ene0.9Hitenol0.1 2-thiol0.14-thiol0.9 | 2-Ene0.7Hitenol0.3 2-thiol0.14-thiol0.9 | 2-Ene0.9Hitenol0.1 2-thiol0.24-thiol0.8 | 2-Ene0.7Hitenol0.3 2-thiol0.24-thiol0.8 | 2-Ene0.5Hitenol0.5 2-thiol0.24-thiol0.8 | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| Synthesis | HSS | FSS | HSS | FSS | HSS | FSS | HSS | FSS | HSS | FSS |
| k 1 | 0.0002 | 0 | 0.0004 | 0.0402 | 0 | 0.0034 | 0 | 0.0146 | 0.0299 | 0.0738 |
| k 2 | 0.0136 | 0.0002 | 0.0190 | 0.2006 | 0.0022 | 0.0586 | 0.0022 | 0.1210 | 0.1729 | 0.2717 |
| F at 1 day | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.01 |
| Slope | 0.0249 | 0.0007 | 0.0331 | 0.2512 | 0.0052 | 0.0872 | 0.0052 | 0.1626 | 0.2211 | 0.3261 |
| Intercept | −0.0006 | 0 | −0.0028 | −0.1 | 0 | 0.024 | 0 | −0.0036 | −0.0261 | 0.3386 |
| R 2 | 0.99 | 0.98 | 0.99 | 0.86 | 0.95 | 0.93 | 0.96 | 0.99 | 0.98 | 0.57 |
| Release at 10 days | 0.06 | 0.16 | 0.08 | 0.62 | 0.01 | 0.22 | 0.01 | 0.39 | 0.54 | 0.95 |
| Hixon-Crowell R2 | 0.61 | 0.61 | 0.65 | 0.84 | 0.52 | 0.48 | 0.57 | 0.62 | 0.67 | 0.31 |
The Hixon–Crowell cubic root equation and UV-Vis spectroscopy was used to determine if erosion of the particle was part of the release mechanism. This equation is specifically designed to evaluate the drug release for systems where the surface area and diameter is changing from erosion or swelling/shrinking.23 The low-moderate R2 values for the Hixon–Crowell equation indicate that erosion is not a primary mechanism for thymol release for these systems (Table 2). This is further confirmed by Nanodrop UV-Vis spectroscopy that indicates no change in the spectra for the control particles over five days (Fig. S5, ESI†). While erosion may not be a primary mechanism, the low-moderate R2 for Hixon–Crowell may indicate that a change in diameter likely plays a part in the diffusion mechanism of thymol release.
Footnote |
| † Electronic supplementary information (ESI) available. See DOI: 10.1039/d0ma00991a |
| This journal is © The Royal Society of Chemistry 2021 |