Abstract
Carboxyl-functionalized polymer microspheres with a rosin moiety were prepared through dispersion polymerization using styrene, disproportionated rosin ester, and methylacrylic acid as raw materials. The effects of dispersion medium (ethanol/water) ratio, monomer mass proportion and initiator concentration on the polymer microspheres were studied. Scanning electron microscopy, laser particle size analysis, thermogravimetric analysis and Fourier transform infrared spectroscopy were used to characterize the microspheres, and their carboxyl contents were determined by the conductance titration method. The adsorption of methylene blue of the microspheres was also investigated. The results showed that rosin-based carboxyl-functionalized polymer microspheres were successfully synthesized. The microspheres exhibited smooth, spherical shapes with good monodispersity and high thermal stability. The carboxyl content of the microspheres prepared under optimum conditions was 0.089 mmol·g−1, with the average particle size approximately 950 nm. With increasing carboxyl contents of the polymer microspheres, their methylene blue adsorption capacities increased. The maximum methylene blue adsorption capacity of the microspheres was 59.55 mg·g−1 in the highest carboxyl content.
Keywords
Introduction
Polymer microspheres are new functional materials with some specific properties. Preparation and application of polymer microspheres have become an important research field in polymer science and have attracted the attentions of many researchers (Erdem and Bicak, 2016; Wei et al., 2016; Zhang et al., 2016) At present, primary methods of preparing polymer microspheres are suspension polymerization (Liu et al., 2016; Yu et al., 2014; Xie et al., 2017), emulsion polymerization (Fan et al., 2016; Yang et al., 2015; Zhang et al., 2012), dispersion polymerization (Beal et al., 2016; Zhang et al., 2011), and seed swelling polymerization (Srisopa, 2016; Xu et al., 2010). Dispersion polymerization is a relative new method for the preparation of monodispersed polymer microspheres with particle size ranging from 0.5 to 8.0 μm. Main factors influencing the dispersion polymerization are polymerization conditions, which include reaction temperature, dispersant level and stirring speed. The particle size and particle size distribution of the polymer microspheres are also strongly influenced by the polarity of the reaction medium. It was found that the particle size of the polymer microspheres decreased with increasing solvent polarity in the methanol/water system (Watanabe et al., 2010), and also with an increase of the ethanol content, the size of the particles increased, whereas no significant change was observed in particle size distribution when using the ionic comonomer benzene sulfonate as a raw material and AIBN as the initiator by dispersion polymerization in an ethanol/water medium (Adelnia et al., 2013).
Generally, functional polymer microspheres are synthesized by combining functional compounds with monomers then polymerization or by modification of common polymer microspheres. Carboxylated polymer microspheres have relatively large specific surface areas, strong adsorption capacities and other interesting properties. They are prepared using various polymerization methods (Guo et al., 2008; Wang et al., 2015). For example, magnetic composite microspheres with a high content of magnetic material and a carboxylic surface were prepared via emulsifier-free emulsion polymerization. The composite microspheres possessed a core-shell structure with a particle size of 422 nm and a narrow particle size distribution (Guo et al., 2008). Through the suspension emulsion polymerization method, carboxylated polymer microspheres with good resistance to high temperatures were obtained by Diels–Alder reaction (Grochowicz et al., 2015). Poly(lactic acid) and poly(ethylene glycol) diblock copolymer microspheres containing fluorescein grafted to the polymer chain were synthesized by a four-component condensation reaction (Icart et al., 2016). These microspheres have been widely used in targeted drug delivery (Gadalla et al., 2016).
As non-renewable oil resources are gradually exhausted, alleviating the energy crisis through the use of renewable resources instead of disposable one is becoming increasingly significant (Ahn et al., 2013; Brocas et al., 2014; Wilbon et al., 2013). For example, Yong et al.24 constructed biobased renewable copolymer microspheres with high-heat resistance by copolymerization of methyl isoeugenol and maleic anhydride. As one of the naturally occurring resource, rosin is particularly important in the application of polymeric materials. The highly rigid three-ring phenanthrene skeleton structure of the rosin endows the materials with excellent mechanical properties. As the reaction-active center, the conjugated double bond of the rosin can polymerize with other reactive monomer; in addition, the carboxyl group of the rosin can react with hydroxyl groups to form an ester (Atta et al., 2010; Li et al., 2015). Rosin-based polymer microspheres have shown great promise for development for sustained drug delivery (Nande et al., 2007). It was confirmed the adsorption capacity of the microspheres increased substantially by carboxylation treatment (Jie et al., 2010). On the basis of this, in the present work, we prepared carboxylated polymer microspheres using renewable-resource rosin as the raw material through the dispersion polymerization method. The effects of synthesis conditions on the properties of the microspheres were discussed, and the adsorption of methylene blue to the microspheres was studied. The carboxylic polymer microspheres containing a rosin moiety are expected to be applicable in the fields of cationic dye adsorption and environmental protection.
Experimental
Materials
Polyvinylpyrrolidone K-30 (PVP) was from Sinopharm Chemical Reagent Co., Ltd. 2,2-Azobisisobutyronitrile (AIBN, chemical purity (CP)), bought from Shanghai No.4 Reagent & H.V Chemical Co., Ltd. Divinylbenzene (DVB), hydroxyethyl methacrylate (HEMA), sodium hydroxide, absolute ethanol, sodium bicarbonate, sodium carbonate, all were analytical purity reagent (AR) and from Xilong Chemical Co., Ltd. Styrene (St), methacrylic acid (MAA, 80%), hydroquinone (HQ), hydrochloric acid, methylene blue (MB) were CP and from Xilong Chemical Co., Ltd. AIBN was recrystallized by ethanol. The polymerization inhibitors in the DVB and St were removed by passing through a column filled with potassium carbonate, silica oxide and aluminum oxide. MAA was treated by reducing pressure distillation. Disproportionated rosin (abietic acid <0.1%, composed of dehydroabietic acid, dihydroabietic acid and tetrahydroabietic acid, 99.9%, acid value 155 mg/KOH·g−1, without further purification) HEMA ester (DRH) was self-made using the method of Yu et al. (2012) with an acid value of 9.4 mg KOH·g−1.
Preparation of the carboxylic polymer microspheres
The typical procedure was as follows: 15 wt.% (based on mixed monomer mass, the same after) of dispersing agent PVP, absolute ethanol, and deionized water were introduced into a triple-necked flask; the mixture was then heated to 75°C under an N2 atmosphere while being stirred by an agitator for 30 min at 120 r/min. Subsequently, 17 wt.% of a monomer mixture composed of styrene (St), disproportionated rosin ester (DRH), and methylacrylic acid (MAA) [(St + DRH + MAA),
Adsorption of MB by the carboxyl polymer microspheres
Thirty milligrams of polymer microspheres were introduced into a 25-mL conical flask, and then 7.5 mL of MB solution (at an initial concentration of approximately 304.4 mg·L−1, a pH of 10.14, and solid-liquid ratio of approximately 1:250 g·mL−1) was added into the conical flask. The flask was subsequently placed at 40°C for 5 h in a thermostatic oscillator operated at a concussion rate of 150 r/min. The microsphere solutions were then centrifuged for 20 min. Finally, 1.0 mL of the supernatant was diluted to 20 mL, and the absorbance of the sample at 664 nm was measured using an UV-spectrophotometer. The concentration of the MB solution was obtained according to a work curve. The MB adsorption capacity of the microspheres was calculated using formula (1)
Characterization
FT-IR spectra of the microspheres were determined by Thermo-Scientific IS10 Fourier transformation infrared spectrometer (USA). Thermogravimetric analyses of the microspheres were carried out on a SDT-Q600 Simultaneous Thermal Analyzer (USA). Morphology of the microspheres was examined by JSM-6380 LV Scanning Electron Microscope (JEOL, Japan). Particle size of the microspheres was measured by Zetasizer Nano ZS90 Laser Particle Size Analyzer (Malvern Instruments Ltd. Britain). Absorbance of the solution was tested by UV-2450 Ultraviolet-Visible Light Detector (Shimadzu, Japan). Carboxyl content of the microspheres was determined according to Wang et al. (2013). A certain amount of polymer microspheres (
Results and discussion
The effects of preparation conditions on microspheres properties
Ratio of reaction medium
We investigated the effects of the reaction medium ratio on the properties of the microspheres while keeping the dispersant PVP and monomer concentrations at 15 and 17 wt.%, respectively, the AIBN concentration at 3 wt.%, and the mass ratio between St and MAA at 8:1. The SEM micrographs of the microspheres are shown in Figure 1, and the particle size distribution curves are shown in Figure 2.

SEM micrographs of the microspheres with different reaction medium ratio (

Particle size distribution of the microspheres with different reaction medium ratio (
As shown in Figures 1 and 2, when the volume ratio of ethanol and water was 36:4, the shape of the microspheres was relatively nonuniform and the microspheres appeared to adhere together. With increasing water content in the reaction medium, the microspheres’ adherence tended to be weaken, their monodispersion improved, and their shape became round; in addition, their average particle size decreased from 1 259 nm to 722 nm and their particle size distribution index (PDI) decreased from 0.023 to 0.003. The increase of water content led to an increase of polarity of the dispersion-medium system, which shortened the critical chain length of the oligomer and increased the amount of core in the system, resulting in a decrease of the particle size of the microspheres (Wan et al., 2013). Meanwhile, because water is a benign solvent for PVP, the solubility parameter of the system increased with increasing water content. The largest amount of the PVP segments was covered on the surface of microspheres, which inhibited monomer movement into the microspheres, thereby enhancing the stability of the formed microspheres and decreasing their particle size distribution (Zhu et al., 2012). Overall, the optimum ethanol-to-water volume ratio was 28:12.
Monomer ratio
The effects of monomer ratio on the properties of microspheres were investigated while maintaining the ethanol-to-water volume ratio at 28:12 and keeping the other conditions constant. The SEM micrographs of the obtained microspheres are shown in Figure 3, and the corresponding particle size distribution curves are shown in Figure 4.

SEM micrographs of microspheres with different monomer ratio (

Particle size distribution of the microspheres with different monomer ratio (
As shown in Figures 3 and 4, all of the microspheres exhibited uniform and spherical shapes under the various monomer ratios, with the exception of the
Initiator level
While keeping the ethanol-to-water volume ratio at 28:12, and the

SEM micrographs of microspheres with different initiator levels (AIBN wt.%, a: 1; b: 3; c: 5). AIBN: azobisisobutyronitrile.

Particle size distribution of the microspheres with different initiator levels.(AIBN wt.%, a: 1; b: 3; c: 5). AIBN: azobisisobutyronitrile.
As shown in Figures 5 and 6, the shapes of the microspheres prepared under different AIBN concentrations were relatively spherical. When the content of AIBN was 1 wt.%, the microspheres’ surfaces were concave and convex, and the particle size distribution was broad (PDI was 0.149). With increasing the amount of AIBN, the microspheres became smooth; the average particle size increased from 520 nm to 954 nm, and the PDI decreased from 0.149 to 0.083, indicating that the particle size distribution became narrow. However, when the amount of AIBN was increased from 3 to 5 wt.%, the particle size distribution of the microspheres exhibited little change. This lack of change in particle size distribution is attributed primarily to an increase in the free-radical generation rate with increasing concentration of the initiator agent, which resulted in a concomitant increase in the free-radical polymerization rate. The rate of formation of active chains increased during the initial stage of the reaction, which favored the formation of a large primary core. The capture efficiency of particles for oligomer free radicals and dead polymer chains increased, which led to a larger particle size and a more narrower particle size distribution (Ge et al., 2015). Overall, an AIBN concentration of 3 wt.% gave the best results.
Addition of crosslinking agent
The effects of adding 1 wt.% of DVB on the microspheres were investigated while keeping the ethanol-to-water volume ratio at 28:12 and an AIBN concentration of 3 wt.%. The results are presented in Table 1 and Figures 7 and 8.
Effect of monomer ratios on the properties of microspheres after adding DVB.
DVB: divinylbenzene; PDI: particle size distribution index.

SEM micrographs of the microspheres with different monomer ratios after adding DVB (

The particle size distributions of microspheres with different monomer ratios after adding DVB (
As shown in Table 1, and Figures 7 and 8, when the St-to-MAA ratio was 8:1, the carboxyl content and the average particle size of the microspheres, which exhibited a smooth surface, were 0.089 mmol·g−1 and 953 nm, respectively. As the amount of MAA was increased, the content of carboxyl groups increased; however, the microspheres’ shape became less spherical, the number of irregular particles increased substantially, and the average particle size increased from 953 nm to 1691 nm. No substantial change in the particle size distribution was observed after the DVB was added. The main result was caused by the free-radical crosslinking reaction between the residual monomer or dead polymer chains and the free radicals on the surface of the microspheres. In the absence of DVB, polymers comprising the microspheres exhibited linear structures. However, when the crosslinking agent was introduced, low-molecular-weight polymer chains were cross-linked into the large polymers with a spatial three-dimensional structure and closer cross-links. As a result, the particle size increased and the monodispersity became worse.
MB adsorption of the microspheres
Experiments involving the adsorption of MB onto the rosin-based carboxyl polymer microspheres were conducted with an initial MB solution concentration of 304.4 mg·L−1, pH = 10.14, and a solid-to-liquid ratio of 1:250 g·mL−1; the adsorption was carried out at 40°C for 5 h. The MB adsorption capacities of the microspheres are shown in Table 2.
MB adsorption of the microspheres with different carboxyl content.
MB: methylene blue.
As clearly shown in Table 2, the MB adsorption capacities of the microspheres increased with increasing carboxyl contents. This increase in adsorption capacity is attributed to MB being a cationic dye. Under alkaline conditions, the carboxyl groups on the polymer microspheres were in the form of carboxyl sodium salt, resulting in adsorption centers with negative charges. The electrostatic attraction between positive and negative charges promoted the adsorption of MB onto microspheres. In addition, microspheres with greater concentration of carboxyl groups had greater numbers of adsorption sites, leading to a corresponding increase in the MB adsorption capacity of the microspheres (Girgis et al., 2007). The maximum adsorption capacity of the carboxylated polymer microspheres on MB was 59.55 mg·g−1 in the present work, although this result was obtained under a certain condition (carboxyl content 0.290 mmol·g−1) that the shapes of the microspheres were not the best (Figure 7(c)), it pave a way to use natural biomass resource in preparation of polymer microspheres; furthermore, the adsorption capacity was higher than that of hollow poly(cyclotriphosphazene-co-phloroglucinol) microspheres (50.7 mg·g−1 at 25°C) (Fu et al., 2015).
FT-IR spectral analysis
The FT-IR spectra of DRH and the carboxylated polymer microspheres are shown in Figure 9. As shown in Figure 9(a), distinctive absorption peaks of saturated methyl (–CH3), (–CH2–), ester carbonyl, ether bonds (C–O–C), and double bonds (C = C) were observed at 2851, 2924, 1727, 1170 and 1242, 1637 cm−1, respectively; the deformation vibration absorption peak of the C–H bonds in the benzene ring was observed at 694 cm−1. All of the aforementioned peaks were attributed to the DRH. As shown in Figure 9(b), a distinctive absorption peak of the benzene ring with single displacement was observed at 757 cm−1; the peaks associated with the benzene-ring carbon skeleton were observed at 1452, 1493 and, 1601 cm−1, and the stretching vibration absorption peaks of C–H bonds in the benzene ring appeared at 3025, 3059 and, 3080 cm−1. These results show that St was introduced into the polymer microspheres. The stretching vibration absorption peak of the hydroxyl bond in the carboxyl groups was observed at 3443 cm−1. The appearance of this peak indicates that MAA had been introduced into the polymer microspheres. The disappearance of the absorption peak at 1637 cm−1 (double bond, C = C) indicates that each monomer participated in the polymerization.

FT-IR spectrum of DRH (a) and carboxyl polymer microspheres (b). DRH: disproportionated rosin ester.
Thermogravimetric analysis
Thermogravimetric analysis (TGA, Figure 10(a)) and differential thermogravimetry (DTG, Figure 10(b)) curves of the DRH and the microspheres are shown in Figure 10. The polymer microspheres lost weight slightly at approximately 100°C due to the loss of water and hydroxyl groups from the surface of the microspheres. The initial decomposition temperature of the microspheres was higher than that of the DRH, indicating that the microspheres are more thermally stable than the DRH. A comparison of the curves of the microspheres with DVB to these of the microspheres without DVB indicates that the adding of DVB to the microspheres did not substantially influence their thermal stability. As shown in Figure 10(b), the temperatures of the maximum decomposition rate of both the polymer microspheres with DVB and these without DVB were all approximately 420°C, indicating that the rosin-based carboxylated polymer microspheres exhibited excellent thermal stability.

TGA (a) and DTG (b) curves of DRH and microspheres with DVB or not. (a: DRH; b: without DVB; c: with DVB). DVB: divinylbenzene; DRH: disproportionated rosin ester; TGA: thermogravimetric analysis.
Conclusions
The carboxylated polymer microspheres were prepared by dispersion polymerization with St, DRH, and MAA as raw materials. The thermal stability of the microspheres was excellent and was not substantially affected by the adding of DVB. The optimal conditions for the synthesis of microspheres were as follows:
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: We are grateful for the financial support of the National Natural Science Foundation of China (31160148), the Guangxi Natural Science Foundation (2015GXNSFBA139231), the special funding for distinguished expert from Guangxi Zhuang Autonomous Region.
