- Email: [email protected]
Study on the condensed state physics of poly(ε-caprolactone) nano-aggregates in aqueous dispersions
Journal of Colloid and Interface Science 450 (2015) 264–271
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Study on the condensed state physics of poly(e-caprolactone) nano-aggregates in aqueous dispersions Shuo Feng a, Yuenan Chen a, Chunfeng Meng a, BiYun Mai a, Qing Wu a,b, HaiYang Gao a,b, GuoDong Liang a,b, FangMing Zhu a,b,⇑ a b
GDHPPC Lab, School of Chemistry and Chemical Engineering, Sun Yat-Sen University, 510275, China Key Lab for Polymer Composite and Functional Materials of Ministry of Education, School of Chemistry and Chemical Engineering, Sun Yat-Sen University, Guangzhou 510275, China
g r a p h i c a l a b s t r a c t
a r t i c l e
i n f o
Article history: Received 21 January 2015 Accepted 11 March 2015 Available online 19 March 2015 Keywords: Poly(e-caprolactone) nano-aggregates Condensed state Condensed state transition Confined crystallization
a b s t r a c t Narrowly size distributed spherical, ellipsoid-like and lamellae stacked poly(e-caprolactone) (PCL) nanoaggregates in aqueous dispersions with a diameter ranging from about () 50 to 330 nm were prepared via nanoprecipitation method in the present study. Scanning electron microscopy (SEM), transmission electron microscopy (TEM) and atomic force microscopy (AFM) were used to characterize the morphology and size of the PCL nano-aggregates. We investigated the melt behaviors of the original (without any thermal treatment after preparation) PCL nano-aggregates in aqueous dispersions by nano differential scanning calorimetry (nano-DSC). In particular, the condensed state of the original 50 nm PCL nanospheres was demonstrated to be amorphous as a result of exhibiting no melting peak in the first nano-DSC heating scan. Furthermore, the rubbery M flow condensed state transition of the amorphous PCL nanospheres was explored by fluorescence measurements. Moreover, the confined crystallization of the 50 nm PCL nanospheres from rubbery state in aqueous dispersions was investigated via isothermal crystallization process. Enormous supercooling was observed during crystallization due to nanoconfinement effect. In addition, when the diameter of the original PCL aggregates was increased to more than 150 nm, PCL is in semi-crystalline state and the crystallinity increases with the diameter. Ó 2015 Elsevier Inc. All rights reserved.
⇑ Corresponding author at: Key Lab for Polymer Composite and Functional Materials of Ministry of Education, School of Chemistry and Chemical Engineering, Sun Yat-Sen University, Guangzhou 510275, China. Fax: +86 20 84114033. E-mail address: [email protected] (F. Zhu). http://dx.doi.org/10.1016/j.jcis.2015.03.029 0021-9797/Ó 2015 Elsevier Inc. All rights reserved.
1. Introductions Recently, condensed state and condensed state transition of crystalline polymer under nanoscale spatial confinement,
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
including one-dimensional (1D, ultrathin films), two-dimensional (2D, nanowires and nanocylinders) and three-dimensional (3D, nanoparticles and nanodroplets) confined systems, have attracted much attention because of the unique physical properties as well as potential applications [1–31]. To date, it is widely known that polymer chains confined geometrically and interfacially in nanoaggregates make the interaction among interchains and interface much different from that in the bulk [3,27,32]. For ultrathin polymer films on solid substrates or free-standing, crystal structure, crystallization kinetics and melt behaviors for ultrathin polymer films have been reported to depend upon not only the thickness of the film but also the identities of the substrate and polymer [4,6–8,33–36]. That is to say, the size effects and interfacial effects play a dominant role in the crystallization of 1D ultrathin polymer films. For crystalline polymer confined in three-dimensional (3D) nano-aggregates, such as nanodroplets [3,12,15,18,27] and selfassembled nanomicelles of block copolymers in bulk or selective solvent [1–3,17,26,29], previous work on the confined crystallization of polymer in nanodroplets and self-assembled nanomicellar cores has shown that nanoconfinement effect promotes a firstorder crystallization kinetics and very high supercooling [1–3]. Taden and Landfester [12] investigated the confined crystallization of PEO nanodroplets with a diameter of about 100 nm and narrow size distribution, reporting that each droplet was formed by 4–5 crystalline lamellae which were not interlamellarly connected but just loosely layered. In addition, the droplets displayed a larger supercooling as a result of exclusively homogeneous nucleation. Müller et al. [26] studied the confined crystallization of PEO nanospheres formed from the self-assembly of poly(ethylene oxide)-bpolybutadiene (PEO-b-PB) diblock copolymer in bulk. For the isothermal crystallization of PEO block at 25 °C, the Avrami index was around 1, representing confined crystallization with homogeneous nucleation. Recently, Carvalho and Dalnoki-Veress [27] described the confined crystallization of PEO nanodroplets on smooth amorphous isotactic polystyrene (ai-PS) or rough crystalline isotactic polystyrene (ci-PS) substrates. For the droplets on the smooth ai-PS substrates, PEO homogeneously nucleated at 4.5 °C. However, a remarkable increase in the crystallization temperature (12 °C) for the droplets on rough ci-PS substrate was observed, indicating that the surface/interface between nanodroplets and substrates is likely to lower the free energy for nucleation. Most studies concerning confined crystallization selected polymer nano-aggregates in bulk as research objects, whereas only a few reports focus their attention on the confined crystallization of polymer nano-aggregates in solution dispersion [13,16,20,37–40]. Ballauff et al. prepared the freely suspended nano-aggregates of PE which consist of a single crystalline lamella sandwiched between two thin amorphous polymer layers obtained by a water-soluble nickel catalyzed emulsion polymerization of ethylene in aqueous solution [16], and found the thickening of the lamellae from melting crystallization in confined nanospheres as thermal annealing at 125 °C [20]. Zhu et al. [38] demonstrated the confined crystallization of PE core in nanosized self-assembled polyethylene-b-poly(ethylene oxide) (PE-b-PEO) micelle dispersed in N,N-dimethylformamide (DMF), and found an enormous supercooling during PE crystallization from spherical melting core and a lower melting temperature (Tm) than that of bulk PE-b-PEO. Poly(e-caprolactone) (PCL) has received a great deal of strong scientific attention because of its excellent biocompatibility, biodegradability and permeation to drugs [41,42]. In addition, PCL is a semi-crystalline thermoplastic polyester with a Tm of 60 °C and a low glass transition temperature (Tg) of 60 °C. Accordingly, PCL nano-aggregates have been found broad applications in drug delivery due to its ability to incorporate lipophilic drugs in vitro and release the drugs in vivo at a later stage
265
[43,44]. For such applications, PCL nano-aggregates are usually loaded with water-insoluble drugs with a particle size of less than 200 nm in order to pass through cell membranes [45,46]. As far as we know, considerable efforts have been devoted to optimize the PCL’s drug loading capacity for individual drug through decoration on the chemical structure of PCL chains [47,48]. However, less attention has been paid to the aggregates structure of PCL, and how it affects drug loading and/or release properties. To date, different physical properties were observed for PCL nano-aggregates in comparison with that of bulk [13,49–54]. For example, Cho et al. [13] studied the effects of size and thermal treatment on the aggregation structure of PCL/PCL-b-PEO micellar cores in aqueous solutions using DSC. Xu et al. [14,50] thoroughly investigated the effects of the isothermal crystallization for the core-forming PCL blocks at different temperatures on micelle morphology of PEO-b-PCL in water. Nojima et al. [49] investigated the crystallization behavior of PCL chains (i.e., PCL block + PCL homopolymer systems) spatially confined in nanocylinders with a diameter of 14.9 and 17.2 nm as a function of the mole fraction of PCL homopolymers (fhomo). The half-time of crystallization decreased moderately with increasing fhomo for the PCL chains confined in the D = 14.9 nm nanocylinders, whereas it increased slightly for the PCL chains in the D = 17.2 nm nanocylinders. To sum up, melting crystallization of 3D polymer nano-aggregates exhibits a much higher supercooling than polymer bulk. Consequently, it is possible to explore the condensed state and condensed state transition of crystalline polymer from a flow state to a rubbery state and to a crystalline state. In the current study, narrowly size distributed PCL nano-aggregates in aqueous dispersions with a diameter ranging from 50 to 330 nm were prepared via nanoprecipitation method. Tween 80, a relatively hypotoxic surfactant was employed as the stabilizing agent. Scanning electron microscopy (SEM), transmission electron microscopy (TEM) and atomic force microscopy (AFM) were used to characterize the morphology and size of the nano-aggregates. Condensed state, condensed state transition and confined crystallization behaviors of the PCL nano-aggregates were further investigated by fluorescence measurements and nano differential scanning calorimetry (nano-DSC). 2. Materials and methods 2.1. Materials Molecular weight distribution PCL with number-average molecular weights (Mn) 45 kg/mol was purchased from J&K Scientific Ltd. The polymer was precipitated in excess methanol and reprecipitated twice from toluene/methanol and stored in drying cabinet at room temperature. Polyethylene glycol sorbitan monooleate (Tween 80, J&K Scientific Ltd.) and pyrene (Py, J&K Scientific Ltd.) were used as received. Deionized water was used for all experiments of samples preparation. All solvents and reagents if not specified were purchased from Sinopharm Chemical Reagent Co., Ltd. and used as received. 2.2. Preparation of PCL nano-aggregates in aqueous dispersions PCL nano-aggregates in aqueous dispersions were prepared via nanoprecipitation method which has been widely employed to prepare polymer nanoparticles [55–58]. Briefly, for the PCL nano-aggregates with a diameter of 50 nm, 16 mL of 0.75 mg/ mL PCL solution in tetrahydrofuran (THF) was dropped into 64 mL of 4 mg/mL Tween 80 aqueous solution under continuous magnetic stirring at room temperature. In particular, deionized water and Tween 80 were mixed by high cuts the isotropic mulser (Gongyiyuhua FA-20) to achieve an aqueous solution of stabilizing
266
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
agent before the nanoprecipitation. After 10 min, the THF was removed by vacuum evaporation at room temperature. The size of the PCL nano-aggregates was controlled by varying the ratio of Tween 80 to PCL in water. The resulting aqueous dispersions was dialyzed for 1 days (cutoff membrane: 8–12 kg/mol) to remove residual THF. Since no thermal treatment was carried out on the resulting aqueous dispersions, the PCL nano-aggregates were named the original PCL nano-aggregates. To obtain the aqueous dispersions of PCL nano-aggregates for fluorescence measurements, 0.5 wt.% Py was doped into PCL nano-aggregates via dissolving Py in THF together with PCL. Note that the resulting Py doped nano-aggregates in aqueous dispersions were only for the fluorescence measurements. 2.3. Characterizations SEM observation of PCL nano-aggregates was performed on a Hitachi S-4800 instrument with an accelerating voltage of 10.0 kV at 8.0 mm working distance. A drop from the previously prepared aqueous dispersions of PCL nano-aggregates was spun cast onto silicon wafers. The samples were dried at room temperature and atmospheric pressure for several hours before examination in SEM. TEM studies were conducted using a JEM100CX instrument operating at an accelerating voltage of 100 kV. The morphological observation of the PCL nano-aggregates was performed on TEM. One drop of dilute solution of the PCL nano-aggregates was deposited on a carbon-coated copper grid. Excess solution was swept away by touching the edge of the grid with a small piece of filter paper. For staining, a drop of 1 wt.% (weight percent) phosphotungstic acid was added to the sample on the grid. After 1 min, excess staining agent was swept away by filter paper, and the grid was dried under room temperature before measurement. AFM morphological observation of PCL nano-aggregates was performed on Bruker Nano Inc. Multimode 8 in ScanAsyst mode. 20 lL of the PCL nano-aggregates in aqueous dispersions was spun cast onto (1 1 1) single-crystal silicon wafers, which have previously been etched to remove the native oxide using hydrofluoric acid and dried at 25 °C for 72 h before the test. DLS measurements were conducted on a Brookhaven BI-200SM apparatus with a BI-9000AT digital correlator and a He–Ne laser at 532 nm. The samples were placed in an index-matching decaline bath with temperature controlled within ±0.2 K. Data were analyzed by CONTIN algorithm, while the average hydrodynamic diameter (hDhi) and size polydispersity of the particles were obtained by a cumulant analysis of the experimental correlation function. In DLS, the Laplase inversion of each measured intensity–intensity timecorrelated function in a dilute solution can result in a characteristic line width distribution G(C). For a purely diffusive relaxation, G(C) can be converted to a hydrodynamic diameter distribution f(Dh) by using the Stokes–Einstein equation. The calorimetry study of the original PCL nano-aggregates in aqueous dispersions was performed on a nano-DSC (TA Instruments). Equal volumes (0.5 mL) of aqueous dispersions of PCL nano-aggregates and reference solution of surfactant solution were injected into the sample and reference cells, respectively. Nano-DSC heating and cooling scans were performed at 2 °C/min over a temperature ranging from 25 to 70 °C at 3 atm. During each scan, the heat capacity difference between the sample cell and the reference cell was plotted as a function of temperature. The crystallinity (Xc) of the PCL nano-aggregates was figured by the formula: Xc = DHm/DH0m, where DHm was the melt enthalpy of the PCL nano-aggregates, and DH0m was the melt enthalpy of the fully crystallized PCL (135 J/g) [59]. Isothermal crystallization of PCL was carried out on a Perkin–Elmer Instruments DSC 8500 calorimeter. Calibration was
performed with indium and tin, and all tests were run employing ultrapure nitrogen as the purge gas. For the bulk sample, it was possible to extract the crystallization kinetics directly from the exothermic peak as a function of crystallization time, but for the aqueous dispersions, the crystallization process did not show a well-defined exothermic peak. Therefore, the sample was instead heated above the Tm of the PCL and quickly cooled to a preset crystalline temperature for a certain time. The Xc was obtained from the melting enthalpy during a subsequent heating process. Typically, for the PCL nano-aggregates in aqueous dispersions, high-pressure stainless-steel hermetic pans were utilized to avoid the evaporation of solvent during heating at high temperatures. About 30 mg aqueous dispersions of PCL nano-aggregates were encapsulated into the tightly sealed pan. The aqueous dispersions of PCL nano-aggregates was heated to 75 °C and kept for 10 min to ensure entirely melting of PCL chain. Thereafter, the heated sample was cooled to the designed isothermal crystallization temperature (Tic) in the DSC sampler with a cooling rate of 500 °C/min by means of liquid nitrogen accessory. The Xc was obtained from the melting enthalpy during a subsequent heating process performed at a heating rate of 5 °C/min. For PCL bulk, DSC heating and cooling scans were performed at 5 °C/min for measurements. Fluorescence measurements on the Py doped PCL nano-aggregates in aqueous dispersions were conducted on a Combined Fluorescence Lifetime and Steady State Spectrometer (Edinburgh Instrument Ltd., FLSP920) equipped with temperature control system (Oxford instruments). Excitation wavelength for Py was 322 nm. All measurements were made upon heating or cooling by increasing temperature in 5 °C increments and holding for 15 min at temperature before measuring the fluorescence spectrum. For the Py doped PCL nano-aggregates in aqueous dispersions, the emission wavelength was monitored in 0.5 nm increments from 360 to 405 nm. The flow temperature (Tf) of the Py doped PCL nano-aggregates in aqueous dispersions was determined via the temperature dependence of the ratio of fluorescence intensities at the first to third vibronic band peaks (I1/I3). Due to a slight shift in the position of the peaks with temperature, intensities were averaged over a 2 nm window, for example from 369 to 371 nm for the first peak and from 380 to 382 nm for the third peak [60]. Data were taken every 0.5 nm so that the averages consist of five data points. 3. Results and discussions 3.1. Preparation and thermal stability of PCL nano-aggregates in aqueous dispersions Polymer nano-aggregates can be conveniently prepared via emulsion polymerization of monomers [61], nanoprecipitation [57], solvent evaporation method [62] and so on [63,64]. Nanoprecipitation is a simple, fast and reproducible method which is widely used for the nano-aggregates preparation of the biodegradable polyesters, especially PCL [55,56] and polylactide (PLA) [65,66]. In the current study, PCL nano-aggregates with narrow size distribution and their aqueous dispersions were prepared by means of nanoprecipitation strategies. The tetrahydrofuran (THF) solution of PCL was added dropwise into an aqueous solution of Tween 80 under continuous magnetic stirring at room temperature and then THF was removed completely. The original aqueous dispersion of PCL nano-aggregates appeared as a bluish, translucent and stable colloidal solution. It is noteworthy that the PCL nano-aggragates were still wrapped by Tween 80 after dialysis, which was confirmed by elemental analysis (EA). The SEM, TEM and AFM images of the original PCL nano-aggregates with an average diameter hDi of 50 nm were shown in Fig. 1. It can be clearly seen from these images that spherical and
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
267
Fig. 1. The SEM (a), TEM (b) and AFM (c) images of the original 50 nm PCL nano-aggregates prepared via nanoprecipitation in aqueous dispersions.
Fig. 3. Py fluorescence emission spectrum of the original 50 nm PCL nanoaggregates (Py doped) in aqueous dispersions at various temperatures. Data was normalized by the maximum intensity of the spectrum at 25 °C. Inset shows the original 50 nm PCL nano-aggregates (Py doped) in aqueous dispersions.
Fig. 2. Size distributions of the 50 nm PCL nano-aggregates in aqueous dispersions (a: original, b: stored for 2 months under room temperature, and c: after DSC measurement).
narrowly size distributed PCL nano-aggregates were prepared. The thermal stability of the PCL nano-aggregates in aqueous dispersions was firstly investigated. As revealed in Fig. 2, the size
268
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
Fig. 4. Temperature dependence of the ratio of fluorescence intensities at the first to third peaks (I1/I3) of the original 50 nm PCL nano-aggregates (Py doped) in aqueous dispersions. (Data were obtained upon a: first heating scan, and b: first cooling scan.)
Fig. 6. Avrami plots for the 50 nm PCL nano-aggregates in aqueous dispersions and PCL bulk at different temperatures. Fig. 5. DSC heating curves of the original 50 nm PCL nano-aggregates and after isothermal crystallization for 3 h at different temperature in aqueous dispersions. The inset (a) presents the DSC second heating and first cooling curve of the PCL bulk; (b) shows the DSC cooling curve for the aqueous dispersions of PCL nanoaggregates.
distributions of the representative 50 nm PCL nanospheres in aqueous dispersions are not obviously altered whether stored for 2 months under room temperature or after DSC measurement, which indicates that the emulsified PCL nano-aggregates did not coalesce during storage or even the DSC measurement.
3.2. Rubbery M flow condensed state transition of the 50 nm PCL nanospheres in aqueous dispersions Nano-DSC was utilized to study the thermal behavior of the original 50 nm PCL nanospheres in aqueous dispersions. In this case, nano-DSC heating scans were performed at 2 °C/min over a temperature ranging from 25 to 70 °C. No melting peak was observed in the first nano-DSC heating trace even though the sensitivity of nano-DSC instrument was high enough to record 1 lJ thermal signal, which suggests that the condensed state of the original 50 nm PCL nanospheres should be amorphous. Fig. 3 shows Py fluorescence emission spectrum of the original
50 nm PCL nano-aggregates (Py doped) in aqueous dispersions at various temperatures. With increasing temperature, the total fluorescence intensity decreases as the pathways for non-radiative decay of the excited state dye (for example bond rotations and vibrations) are promoted [60,67]. Moreover, the ratio of the first peak (370 nm) to the third peak (381 nm) (I1/I3) decreases upon heating and the intersection of linear regressions was 45 °C (Fig. 4a). In addition, the ratio of I1/I3 increases upon cooling and the intersection of linear regressions was 48 °C (Fig. 4b). Note that the I1/I3 ratio was certificated to be sensitive to the local environment experienced by Py in the late 1970s [68], specifically the polarity, and has been used previously to determine glass transition for both polymer films and blends [60,67]. Therefore, the results of fluorescence measurements implied a condensed phase transition near 45 °C. For the condensed state of amorphous polymers, there include glass, rubbery and flow state according to the different molecular dynamics. When polymers are in glass state, both the segments and chains cannot move indicating that the molecular dynamics are frozen. When polymers enter in rubbery state from glass state with the rise of temperature, the segments are thawed and able to move. While as temperatures continue to rise, the polymers will enter in flow state and the permanent deformation of polymer is due to coordinated movement of segments.
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
Fig. 7. First nano-DSC heating curves for the original PCL nano-aggregates with different diameter (a: 50 nm, b: 150 nm, and c: 330 nm) in aqueous dispersions.
269
Consequently, there is reason to believe that the condensed state transition determined by fluorescence analysis can be attributed to the transition from rubbery state to flow state because the Tg of PCL is 60 °C [69] which is much lower than 45 °C. The Tf of the PCL bulk (Mn 45 kg/mol) is 66 °C which is measured via dynamic mechanical temperature spectrum, and is higher than that of the 50 nm PCL nano-aggregates. This may be related to the higher chain entanglement density in PCL bulk than in PCL nano-aggregates. It is noteworthy that, as shown in Fig. 3 inset, neither morphology nor average diameter of the PCL nano-aggregates (Py doped) is changed in comparison with that of the undoped one (shown in Fig. 1). And the condensed state of the original PCL nano-aggregates (Py doped) remained in amorphous state since no melting peak was observed in the first nano-DSC heating trace. Moreover, the PCL nano-aggregates (Py doped) were removed from the aqueous dispersions as centrifugated (Beckman-coulter Avanti J30I, USA) for 30 min at a speed of 25,000 rpm. The resultant supernate displays nearly no fluorescence emission, implying Py is almost completely encapsulated in the PCL nano-aggregates rather than diffusing into water.
Fig. 8. The SEM (a), TEM (b) and AFM (c) images of the original 150 nm PCL nano-aggregates prepared via nanoprecipitation in aqueous dispersions.
270
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
3.3. Confined crystallization of the 50 nm PCL nanospheres in aqueous dispersions Fig. 5 shows the DSC heating curves of the aqueous dispersions of 50 nm PCL nano-aggregates after isothermal crystallization at different temperatures ranging from 3 to 11 °C. When Tic > 0 °C, no melting peak was observed even the period of isothermal crystallization is prolonged from 3 to 12 h. Moreover, a crystalline melting peak attributed to crystallization of the PCL nano-aggregates is not apparently observed (Fig. 5) until the isothermal crystallization temperature dropped to 0 °C. Note that Tm and Xc remain constant as the crystallization time is more than 3 h, which indicates that the Xc of the PCL nano-aggregates has reached saturation within 3 h. Moreover, the Xc of the PCL nano-aggregates increased with the decrease of Tic. It is important to point out that a single melting peak for all the investigated systems is observed, which suggests that the system has only one kind of crystal structure [70,71]. Moreover, the 50 nm PCL nano-aggregates in aqueous dispersions exhibit a Tm of 52 °C, which is 8 °C lower than that of the PCL bulk (Fig. 5 inset (a)). This may be because of the crystal imperfections and/or the reduction of the lamellar thickness of PCL crystals due to the nanoconfinment [20,38,72,73]. Note that the aqueous dispersions of PCL nano-aggregates begun to freeze when the temperature drops below 13 °C as shown in Fig. 5 inset (b). In case of any impact (hard confinement caused by the freezing of the surrounding aqueous media) on the crystallization of PCL nano-aggregates from freezing of the aqueous dispersions, Tic was set above 13 °C. The isothermal crystallization kinetics of the 50 nm PCL nano-aggregates in aqueous dispersions was further investigated. As shown in Fig. 6, PCL exhibits an Avrami index of 3 in the bulk, indicating unconfined two-dimensional crystal growth after heterogeneous nucleation [1]. However, the Avrami index of the 50 nm PCL nanoaggregates in aqueous dispersions was close to 1, implying that this confined crystallization was dominated by nucleation [1,40]. It is generally recognized that when a polymer is divided into nanosized aggregates, and the number of nano-aggregates is much more than the number of active heterogeneous nuclei, the nucleation and crystallization of the polymer can experience dramatic changes, which usually occurs along with substantial supercooling [1,3,26,38,40]. As for the crystallization of heterogeneity-free nano-aggregates, the nucleation typically changes from heterogeneous nucleation to homogeneous nucleation [1,15,22,26,28,49]. However, as demonstrated by Carvalho and Dalnoki-Veress, the surface of the microdomains or the interface between the crystallizable nano-aggregates and the matrix surrounding them can induce the nucleation (i.e. surface nucleation) [3,18,27]. Furthermore, a first order or lower (n 6 1) crystallization kinetics can also be found for confined polymers undergoing both homogenous and surface nucleation [1,3,12,27]. Nevertheless, surface nucleation requires lower supercoolings than classic homogeneous nucleation, because it is a process characterized by a lower free energy than the more energetically costly process of creating new crystal surfaces inside an nano-aggregates [3,18,27]. In this case, the size of the nano-aggregates was 50 nm, the number of nano-aggregates greatly exceeded the number of heterogeneous nuclei, which giving rise to exclusively homogeneous or surface nucleation. As reported previously by Nojima et al., PCL blocks or homopolymers confined in spherical and cylindrical nano-aggregates exhibit a homogeneous nucleation and first-order crystallization kinetics at very low crystallization temperature of 45 °C which are much closer to their corresponding Tg values [29,74]. In this situation, however, the highest crystallization temperature of 0 °C is much higher than the reported value. We suggest that the surface/interface of the nano-aggregates probably play an important role in nucleation. Therefore, a surface nucleation
Fig. 9. The SEM image of the original 330 nm PCL nano-aggregates prepared via nanoprecipitation in aqueous dispersions.
mechanism was more plausible for the confined crystallization of the 50 nm PCL nano-aggregates in aqueous dispersions. 3.4. Condensed state of the PCL nano-aggregates with a diameter of more than 50 nm in aqueous dispersions Fig. 7 displays the first nano-DSC heating curves of the original PCL nano-aggregates in aqueous dispersions. Melting peaks are obviously observed for the original 150 and 330 nm PCL nanoaggregates, indicating that PCL is in semi-crystalline state as the diameter is increased more than 150 nm. Moreover, the Xc of the original PCL nano-aggregates increases with the diameter. Fig. 8 presents the typical SEM, TEM and AFM images of the original PCL nano-aggregates with a hDi of 150 nm and a height of 45 nm (Fig. 8c), indicating that ellipsoid-like nano-aggregates is formed in terms of two-dimensional growth of the PCL crystalline. Fig. 9 displays the SEM image of the original PCL nano-aggregates with a hDi of 330 nm which is consist of disc-shaped lamellae stacked each other. It was believed that the morphology of the original nano-aggregates in aqueous dispersions changes from spherical to ellipsoid-like even lamellae stacked with the increase of Xc. 4. Conclusions To our best knowledge, most studies on condensed state physics concerned polymer nano-aggregates in bulk, such as self-assembled nanomicelles of block copolymers in bulk [1–3]. Whereas only a few reports focus their attention on the condensed state physics of polymer nano-aggregates in solution dispersion [13,16,20,37–40]. In this work, we have demonstrated the condensed state physics of the narrowly distributed spherical or ellipsoid-like even lamellae stacked PCL nano-aggregates with a diameter ranging from 50 to 330 nm in aqueous dispersions which were prepared by nanoprecipitation strategies using Tween 80 as a surfactant and THF as a co-solvent at room temperature. The original PCL aggregates with a diameter of less than 100 nm, for representatively 50 nm, show spherical morphology and rubbery state. Fluorescence analysis indicated that the transition temperature from rubbery state to flow state was 45 °C. Meanwhile, the 50 nm PCL nanospheres show a supercooling point of <0 °C during crystallization from the rubbery state, which is much lower than that of PCL bulk. Furthermore, an investigation on the isothermal crystallization kinetics of PCL in the 50 nm nanospheres displays that the Avrami index n was
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
close to 1, indicating that the confined crystallization nucleation is dominated by nucleation. These findings about the large supercooling and first-order crystallization kinetics for crystallization of the 50 nm PCL nano-aggregates in aqueous dispersions are in line with previous studies carried on self-assembled nanomicelles of block copolymers in bulk [53,74]. However, the highest crystallization temperature of 0 °C is much higher than the reported value demonstrated in bulk system, implying that surface of the PCL nano-aggregates may play an important role in the crystallization process [3,27]. By contrast, the crystallization of PCL chain can promote the formation of ellipsoid-like nano-aggregates as the diameter is more than 150 nm. Interestingly, when the diameter is increased to 330 nm, spherical aggregates consisted of discshaped PCL crystalline lamellae stacked each other was formed. Briefly, our work thoroughly investigated the condensed state and condensed state transition of the poly(e-caprolactone) (PCL) nano-aggregates in aqueous dispersions which is of biomedical application. Significantly, it may contribute to understanding the relationship between the condensed state of PCL nano-aggregates and its drug loading capacity since most biomedical application of PCL-aggregates are carried out in aqueous dispersions rather than in dried state [13]. Acknowledgments This work was supported by the National Natural Science Foundation of China (21174167), the NSF of Guangdong Province (S2013030013474) and the Guangdong Province Higher School Science and Technology Innovation Key Project. References [1] A.J. Müller, V. Balsamo, M.L. Arnal, Adv. Polym. Sci. 190 (2005) 1. [2] M.C. Lin, B. Nandan, H.L. Chen, Soft Matter 8 (2012) 7306. [3] R.M. Michell, I. Blaszczyk-Lezak, C. Mijangos, A.J. Müller, Polymer 54 (2013) 4059. [4] Y.-X. Liu, E.-Q. Chen, Coord. Chem. Rev. 254 (2010) 1011. [5] J.M. Carr, D.S. Langhe, M.T. Ponting, A. Hiltner, E. Baer, J. Mater. Res. 27 (2012) 1326. [6] Y. Ma, Hu, G. Reiter, Macromolecules 39 (2006) 5159. [7] D.E. Martínez-Tong, B. Vanroy, M. Wübbenhorst, A. Nogales, S. Napolitano, Macromolecules 47 (2014) 2354. [8] P.C. Jukes, A. Das, M. Durell, D. Trolley, A.M. Higgins, M. Geoghegan, J.E. Macdonald, R.A.L. Jones, S. Brown, P. Thompson, Macromolecules 38 (2005) 2315. [9] E. Woo, J. Huh, Y.G. Jeong, K. Shin, Phys. Rev. Lett. 98 (2007) 136103. [10] Y. Zhou, S.-K. Ahn, R.K. Lakhman, M. Gopinadhan, C.O. Osuji, R.M. Kasi, Macromolecules 44 (2011) 3924. [11] Y.-L. Loo, R.A. Register, A.J. Ryan, Phys. Rev. Lett. 84 (2000) 4120. [12] A. Taden, K. Landfester, Macromolecules 36 (2003) 4037. [13] E.C. Cho, K. Cho, J.K. Ahn, J. Kim, I.S. Chang, Biomacromolecules 7 (2006) 1679. [14] Z.-X. Du, J.-T. Xu, Z.-Q. Fan, Macromolecules 40 (2007) 7633. [15] L. Kailas, C. Vasilev, J.-N. Audinot, H.-N. Migeon, J.K. Hobbs, Macromolecules 40 (2007) 7223. [16] C.H. Weber, A. Chiche, G. Krausch, S. Rosenfeldt, M. Ballauff, L. Harnau, I. Gottker-Schnetmann, Q. Tong, S. Mecking, Nano Lett. 7 (2007) 2024. [17] S. Nojima, Y. Ohguma, S. Namiki, T. Ishizone, K. Yamaguchi, Macromolecules 41 (2008) 1915. [18] J.L. Carvalho, K. Dalnoki-Veress, Eur. Phys. J. E, Soft Matter 34 (2011) 1. [19] J.L. Carvalho, M.E. Somers, K. Dalnoki-Veress, J. Polym. Sci., Part B: Polym. Phys. 49 (2011) 712. [20] C.N. Rochette, S. Rosenfeldt, K. Henzler, F. Polzer, M. Ballauff, Q. Tong, S. Mecking, M. Drechsler, T. Narayanan, L. Harnau, Macromolecules 44 (2011) 4845. [21] J.-C. Zhao, B. Xing, Z. Peng, J.-M. Zhang, Chin. J. Polym. Sci. 31 (2013) 1310.
271
[22] C.L. Huang, L. Jiao, J.B. Zeng, J.J. Zhang, K.K. Yang, Y.Z. Wang, J. Phys. Chem. B 117 (2013) 10665. [23] H.D. Huang, J.Z. Xu, Y. Fan, L. Xu, Z.M. Li, J. Phys. Chem. B 117 (2013) 10641. [24] A.J. Müller, V. Balsamo, M.L. Arnal, T. Jakob, H. Schmalz, V. Abetz, Macromolecules 35 (2002) 3048. [25] A.T. Lorenzo, M.L. Arnal, A.J. Müller, A. Boschetti-de-Fierro, V. Abetz, Macromolecules 40 (2007) 5023. [26] R.V. Castillo, M.L. Arnal, A.J. Müller, I.W. Hamley, V. Castelletto, H. Schmalz, V. Abetz, Macromolecules 41 (2008) 879. [27] J.L. Carvalho, K. Dalnoki-Veress, Phys. Rev. Lett. 105 (2010) 237801. [28] M. Massa, J. Carvalho, K. Dalnoki-Veress, Phys. Rev. Lett. 97 (2006) 247802. [29] S. Nojima, M. Toei, S. Hara, S. Tanimoto, S. Sasaki, Polymer 43 (2002) 4087. [30] L. Zhu, B.R. Mimnaugh, Q. Ge, R.P. Quirk, S.Z.D. Cheng, E.L. Thomas, B. Lotz, B.S. Hsiao, F. Yeh, L. Liu, Polymer 42 (2001) 9121. [31] H.-L. Chen, H.-C. Li, Y.-Y. Huang, F.-C. Chiu, Macromolecules 35 (2002) 2417. [32] M.D. Ediger, J.A. Forrest, Macromolecules 47 (2014) 471. [33] M.V. Massa, K. Dalnoki-Veress, J.A. Forrest, Eur. Phys. J. E, Soft Matter 11 (2003) 191. [34] H. Schönherr, C.W. Frank, Macromolecules 36 (2003) 1188. [35] Y.J. Ren, Z. Huang, W.B. Hu, J. Macromol. Sci. Part B – Phys. 51 (2012) 2341. [36] B. Vanroy, M. Wübbenhorst, S. Napolitano, ACS Macro Lett. 2 (2013) 168. [37] L. Yin, M.A. Hillmyer, Macromolecules 44 (2011) 3021. [38] Z.Y. Li, R. Liu, B.Y. Mai, W.J. Wang, Q. Wu, G.D. Liang, H.Y. Gao, F.M. Zhu, Polymer 54 (2013) 1663. [39] B.Y. Mai, Z.Y. Li, R. Liu, S. Feng, Q. Wu, G.D. Liang, H.Y. Gao, F.M. Zhu, J. Polym. Res. 20 (2013) 1. [40] R. Liu, Z.Y. Li, B.Y. Mai, Q. Wu, G.D. Liang, H.Y. Gao, F.M. Zhu, J. Polym. Res. 20 (2013) 64. [41] O. Coulembier, P. Degée, J.L. Hedrick, P. Dubois, Prog. Polym. Sci. 31 (2006) 723. [42] B. Amsden, Soft Matter 3 (2007) 1335. [43] L. Yu, J.D. Ding, Chem. Soc. Rev. 37 (2008) 1473. [44] M.A. Woodruff, D.W. Hutmacher, Prog. Polym. Sci. 35 (2010) 1217. [45] M.L. Gou, K. Men, H.S. Shi, M.L. Xiang, J.A. Zhang, J. Song, J.L. Long, Y. Wan, F. Luo, X. Zhao, Z.Y. Qian, Nanoscale 3 (2011) 1558. [46] X. Gao, B.L. Wang, X.W. Wei, K. Men, F.J. Zheng, Y.F. Zhou, Y. Zheng, M.L. Gou, M.J. Huang, G. Guo, N. Huang, Z.Y. Qian, Y.Q. Wei, Nanoscale 4 (2012) 7021. [47] S.M. Garg, X.B. Xiong, C. Lu, A. Lavasanifar, Macromolecules 44 (2011) 2058. [48] H.F. Sun, L. Mei, C.X. Song, X.M. Cui, P.Y. Wang, Biomaterials 27 (2006) 1735. [49] S. Nakagawa, T. Tanaka, T. Ishizone, S. Nojima, Y. Kakiuchi, K. Yamaguchi, S. Nakahama, Macromolecules 46 (2013) 2199. [50] Z.-X. Du, J.-T. Xu, Z.-Q. Fan, Macromol. Rapid Commun. 29 (2008) 467. [51] K. Schäler, A. Achilles, R. Bärenwald, C. Hackel, K. Saalwächter, Macromolecules 46 (2013) 7818. [52] Y. Suzuki, H. Duran, M. Steinhart, H.-J. Butt, G. Floudas, Macromolecules 47 (2014) 1793. [53] R.M. Michell, A.J. Müller, G. Deshayes, P. Dubois, Eur. Polym. J. 46 (2010) 1334. [54] R.-M. Ho, Y.-W. Chiang, C.-C. Lin, B.-H. Huang, Macromolecules 38 (2005) 4769. [55] J. Rieger, C. Passirani, J.P. Benoit, K. Van Butsele, R. Jerome, C. Jerome, Adv. Funct. Mater. 16 (2006) 1506. [56] F. Lince, D.L. Marchisio, A.A. Barresi, J. Colloid Interface Sci. 322 (2008) 505. [57] C. Zhang, J.W. Chung, R.D. Priestley, Macromol. Rapid Commun. 33 (2012) 1798. [58] C. Zhang, V.J. Pansare, R.K. Prud’homme, R.D. Priestley, Soft Matter 8 (2012) 86. [59] V. Crescenzi, G. Manzini, G. Calzolari, C. Borri, Eur. Polym. J. 8 (1972) 449. [60] S. Kim, C.B. Roth, J.M. Torkelson, J. Polym. Sci., Part B: Polym. Phys. 46 (2008) 2754. [61] K. Min, H.F. Gao, K. Matyjaszewski, J. Am. Chem. Soc. 127 (2005) 3825. [62] Y.Y. Yang, T.S. Chung, N.P. Ng, Biomaterials 22 (2001) 231. [63] K. Landfester, Angew. Chem. Int. Ed. Engl. 48 (2009) 4488. [64] J.P. Rao, K.E. Geckeler, Prog. Polym. Sci. 36 (2011) 887. [65] T. Govender, S. Stolnik, M.C. Garnett, L. Illum, S.S. Davis, J. Control. Release 57 (1999) 171. [66] Y.C. Dong, S.S. Feng, Biomaterials 25 (2004) 2843. [67] C.J. Ellison, J.M. Torkelson, Nat. Mater. 2 (2003) 695. [68] K. Kalyanasundaram, J.K. Thomas, J. Am. Chem. Soc. 99 (1977) 2039. [69] J.V. Koleske, R.D. Lundberg, J. Polym. Sci. Part A – 2: Polym. Phys. 7 (1969) 795. [70] E.S. Yoo, S.S. Im, J. Polym. Sci., Part B: Polym. Phys. 37 (1999) 1357. [71] Z.B. Qiu, M. Komura, T. Ikehara, T. Nishi, Polymer 44 (2003) 7781. [72] F.M. Bauers, R. Thomann, S. Mecking, J. Am. Chem. Soc. 125 (2003) 8838. [73] K. Shin, E. Woo, Y.G. Jeong, C. Kim, J. Huh, K.-W. Kim, Macromolecules 40 (2007) 6617. [74] S. Nojima, Y. Ohguma, K.-I. Kadena, T. Ishizone, Y. Iwasaki, K. Yamaguchi, Macromolecules 43 (2010) 3916.
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Study on the condensed state physics of poly(e-caprolactone) nano-aggregates in aqueous dispersions Shuo Feng a, Yuenan Chen a, Chunfeng Meng a, BiYun Mai a, Qing Wu a,b, HaiYang Gao a,b, GuoDong Liang a,b, FangMing Zhu a,b,⇑ a b
GDHPPC Lab, School of Chemistry and Chemical Engineering, Sun Yat-Sen University, 510275, China Key Lab for Polymer Composite and Functional Materials of Ministry of Education, School of Chemistry and Chemical Engineering, Sun Yat-Sen University, Guangzhou 510275, China
g r a p h i c a l a b s t r a c t
a r t i c l e
i n f o
Article history: Received 21 January 2015 Accepted 11 March 2015 Available online 19 March 2015 Keywords: Poly(e-caprolactone) nano-aggregates Condensed state Condensed state transition Confined crystallization
a b s t r a c t Narrowly size distributed spherical, ellipsoid-like and lamellae stacked poly(e-caprolactone) (PCL) nanoaggregates in aqueous dispersions with a diameter ranging from about () 50 to 330 nm were prepared via nanoprecipitation method in the present study. Scanning electron microscopy (SEM), transmission electron microscopy (TEM) and atomic force microscopy (AFM) were used to characterize the morphology and size of the PCL nano-aggregates. We investigated the melt behaviors of the original (without any thermal treatment after preparation) PCL nano-aggregates in aqueous dispersions by nano differential scanning calorimetry (nano-DSC). In particular, the condensed state of the original 50 nm PCL nanospheres was demonstrated to be amorphous as a result of exhibiting no melting peak in the first nano-DSC heating scan. Furthermore, the rubbery M flow condensed state transition of the amorphous PCL nanospheres was explored by fluorescence measurements. Moreover, the confined crystallization of the 50 nm PCL nanospheres from rubbery state in aqueous dispersions was investigated via isothermal crystallization process. Enormous supercooling was observed during crystallization due to nanoconfinement effect. In addition, when the diameter of the original PCL aggregates was increased to more than 150 nm, PCL is in semi-crystalline state and the crystallinity increases with the diameter. Ó 2015 Elsevier Inc. All rights reserved.
⇑ Corresponding author at: Key Lab for Polymer Composite and Functional Materials of Ministry of Education, School of Chemistry and Chemical Engineering, Sun Yat-Sen University, Guangzhou 510275, China. Fax: +86 20 84114033. E-mail address: [email protected] (F. Zhu). http://dx.doi.org/10.1016/j.jcis.2015.03.029 0021-9797/Ó 2015 Elsevier Inc. All rights reserved.
1. Introductions Recently, condensed state and condensed state transition of crystalline polymer under nanoscale spatial confinement,
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
including one-dimensional (1D, ultrathin films), two-dimensional (2D, nanowires and nanocylinders) and three-dimensional (3D, nanoparticles and nanodroplets) confined systems, have attracted much attention because of the unique physical properties as well as potential applications [1–31]. To date, it is widely known that polymer chains confined geometrically and interfacially in nanoaggregates make the interaction among interchains and interface much different from that in the bulk [3,27,32]. For ultrathin polymer films on solid substrates or free-standing, crystal structure, crystallization kinetics and melt behaviors for ultrathin polymer films have been reported to depend upon not only the thickness of the film but also the identities of the substrate and polymer [4,6–8,33–36]. That is to say, the size effects and interfacial effects play a dominant role in the crystallization of 1D ultrathin polymer films. For crystalline polymer confined in three-dimensional (3D) nano-aggregates, such as nanodroplets [3,12,15,18,27] and selfassembled nanomicelles of block copolymers in bulk or selective solvent [1–3,17,26,29], previous work on the confined crystallization of polymer in nanodroplets and self-assembled nanomicellar cores has shown that nanoconfinement effect promotes a firstorder crystallization kinetics and very high supercooling [1–3]. Taden and Landfester [12] investigated the confined crystallization of PEO nanodroplets with a diameter of about 100 nm and narrow size distribution, reporting that each droplet was formed by 4–5 crystalline lamellae which were not interlamellarly connected but just loosely layered. In addition, the droplets displayed a larger supercooling as a result of exclusively homogeneous nucleation. Müller et al. [26] studied the confined crystallization of PEO nanospheres formed from the self-assembly of poly(ethylene oxide)-bpolybutadiene (PEO-b-PB) diblock copolymer in bulk. For the isothermal crystallization of PEO block at 25 °C, the Avrami index was around 1, representing confined crystallization with homogeneous nucleation. Recently, Carvalho and Dalnoki-Veress [27] described the confined crystallization of PEO nanodroplets on smooth amorphous isotactic polystyrene (ai-PS) or rough crystalline isotactic polystyrene (ci-PS) substrates. For the droplets on the smooth ai-PS substrates, PEO homogeneously nucleated at 4.5 °C. However, a remarkable increase in the crystallization temperature (12 °C) for the droplets on rough ci-PS substrate was observed, indicating that the surface/interface between nanodroplets and substrates is likely to lower the free energy for nucleation. Most studies concerning confined crystallization selected polymer nano-aggregates in bulk as research objects, whereas only a few reports focus their attention on the confined crystallization of polymer nano-aggregates in solution dispersion [13,16,20,37–40]. Ballauff et al. prepared the freely suspended nano-aggregates of PE which consist of a single crystalline lamella sandwiched between two thin amorphous polymer layers obtained by a water-soluble nickel catalyzed emulsion polymerization of ethylene in aqueous solution [16], and found the thickening of the lamellae from melting crystallization in confined nanospheres as thermal annealing at 125 °C [20]. Zhu et al. [38] demonstrated the confined crystallization of PE core in nanosized self-assembled polyethylene-b-poly(ethylene oxide) (PE-b-PEO) micelle dispersed in N,N-dimethylformamide (DMF), and found an enormous supercooling during PE crystallization from spherical melting core and a lower melting temperature (Tm) than that of bulk PE-b-PEO. Poly(e-caprolactone) (PCL) has received a great deal of strong scientific attention because of its excellent biocompatibility, biodegradability and permeation to drugs [41,42]. In addition, PCL is a semi-crystalline thermoplastic polyester with a Tm of 60 °C and a low glass transition temperature (Tg) of 60 °C. Accordingly, PCL nano-aggregates have been found broad applications in drug delivery due to its ability to incorporate lipophilic drugs in vitro and release the drugs in vivo at a later stage
265
[43,44]. For such applications, PCL nano-aggregates are usually loaded with water-insoluble drugs with a particle size of less than 200 nm in order to pass through cell membranes [45,46]. As far as we know, considerable efforts have been devoted to optimize the PCL’s drug loading capacity for individual drug through decoration on the chemical structure of PCL chains [47,48]. However, less attention has been paid to the aggregates structure of PCL, and how it affects drug loading and/or release properties. To date, different physical properties were observed for PCL nano-aggregates in comparison with that of bulk [13,49–54]. For example, Cho et al. [13] studied the effects of size and thermal treatment on the aggregation structure of PCL/PCL-b-PEO micellar cores in aqueous solutions using DSC. Xu et al. [14,50] thoroughly investigated the effects of the isothermal crystallization for the core-forming PCL blocks at different temperatures on micelle morphology of PEO-b-PCL in water. Nojima et al. [49] investigated the crystallization behavior of PCL chains (i.e., PCL block + PCL homopolymer systems) spatially confined in nanocylinders with a diameter of 14.9 and 17.2 nm as a function of the mole fraction of PCL homopolymers (fhomo). The half-time of crystallization decreased moderately with increasing fhomo for the PCL chains confined in the D = 14.9 nm nanocylinders, whereas it increased slightly for the PCL chains in the D = 17.2 nm nanocylinders. To sum up, melting crystallization of 3D polymer nano-aggregates exhibits a much higher supercooling than polymer bulk. Consequently, it is possible to explore the condensed state and condensed state transition of crystalline polymer from a flow state to a rubbery state and to a crystalline state. In the current study, narrowly size distributed PCL nano-aggregates in aqueous dispersions with a diameter ranging from 50 to 330 nm were prepared via nanoprecipitation method. Tween 80, a relatively hypotoxic surfactant was employed as the stabilizing agent. Scanning electron microscopy (SEM), transmission electron microscopy (TEM) and atomic force microscopy (AFM) were used to characterize the morphology and size of the nano-aggregates. Condensed state, condensed state transition and confined crystallization behaviors of the PCL nano-aggregates were further investigated by fluorescence measurements and nano differential scanning calorimetry (nano-DSC). 2. Materials and methods 2.1. Materials Molecular weight distribution PCL with number-average molecular weights (Mn) 45 kg/mol was purchased from J&K Scientific Ltd. The polymer was precipitated in excess methanol and reprecipitated twice from toluene/methanol and stored in drying cabinet at room temperature. Polyethylene glycol sorbitan monooleate (Tween 80, J&K Scientific Ltd.) and pyrene (Py, J&K Scientific Ltd.) were used as received. Deionized water was used for all experiments of samples preparation. All solvents and reagents if not specified were purchased from Sinopharm Chemical Reagent Co., Ltd. and used as received. 2.2. Preparation of PCL nano-aggregates in aqueous dispersions PCL nano-aggregates in aqueous dispersions were prepared via nanoprecipitation method which has been widely employed to prepare polymer nanoparticles [55–58]. Briefly, for the PCL nano-aggregates with a diameter of 50 nm, 16 mL of 0.75 mg/ mL PCL solution in tetrahydrofuran (THF) was dropped into 64 mL of 4 mg/mL Tween 80 aqueous solution under continuous magnetic stirring at room temperature. In particular, deionized water and Tween 80 were mixed by high cuts the isotropic mulser (Gongyiyuhua FA-20) to achieve an aqueous solution of stabilizing
266
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
agent before the nanoprecipitation. After 10 min, the THF was removed by vacuum evaporation at room temperature. The size of the PCL nano-aggregates was controlled by varying the ratio of Tween 80 to PCL in water. The resulting aqueous dispersions was dialyzed for 1 days (cutoff membrane: 8–12 kg/mol) to remove residual THF. Since no thermal treatment was carried out on the resulting aqueous dispersions, the PCL nano-aggregates were named the original PCL nano-aggregates. To obtain the aqueous dispersions of PCL nano-aggregates for fluorescence measurements, 0.5 wt.% Py was doped into PCL nano-aggregates via dissolving Py in THF together with PCL. Note that the resulting Py doped nano-aggregates in aqueous dispersions were only for the fluorescence measurements. 2.3. Characterizations SEM observation of PCL nano-aggregates was performed on a Hitachi S-4800 instrument with an accelerating voltage of 10.0 kV at 8.0 mm working distance. A drop from the previously prepared aqueous dispersions of PCL nano-aggregates was spun cast onto silicon wafers. The samples were dried at room temperature and atmospheric pressure for several hours before examination in SEM. TEM studies were conducted using a JEM100CX instrument operating at an accelerating voltage of 100 kV. The morphological observation of the PCL nano-aggregates was performed on TEM. One drop of dilute solution of the PCL nano-aggregates was deposited on a carbon-coated copper grid. Excess solution was swept away by touching the edge of the grid with a small piece of filter paper. For staining, a drop of 1 wt.% (weight percent) phosphotungstic acid was added to the sample on the grid. After 1 min, excess staining agent was swept away by filter paper, and the grid was dried under room temperature before measurement. AFM morphological observation of PCL nano-aggregates was performed on Bruker Nano Inc. Multimode 8 in ScanAsyst mode. 20 lL of the PCL nano-aggregates in aqueous dispersions was spun cast onto (1 1 1) single-crystal silicon wafers, which have previously been etched to remove the native oxide using hydrofluoric acid and dried at 25 °C for 72 h before the test. DLS measurements were conducted on a Brookhaven BI-200SM apparatus with a BI-9000AT digital correlator and a He–Ne laser at 532 nm. The samples were placed in an index-matching decaline bath with temperature controlled within ±0.2 K. Data were analyzed by CONTIN algorithm, while the average hydrodynamic diameter (hDhi) and size polydispersity of the particles were obtained by a cumulant analysis of the experimental correlation function. In DLS, the Laplase inversion of each measured intensity–intensity timecorrelated function in a dilute solution can result in a characteristic line width distribution G(C). For a purely diffusive relaxation, G(C) can be converted to a hydrodynamic diameter distribution f(Dh) by using the Stokes–Einstein equation. The calorimetry study of the original PCL nano-aggregates in aqueous dispersions was performed on a nano-DSC (TA Instruments). Equal volumes (0.5 mL) of aqueous dispersions of PCL nano-aggregates and reference solution of surfactant solution were injected into the sample and reference cells, respectively. Nano-DSC heating and cooling scans were performed at 2 °C/min over a temperature ranging from 25 to 70 °C at 3 atm. During each scan, the heat capacity difference between the sample cell and the reference cell was plotted as a function of temperature. The crystallinity (Xc) of the PCL nano-aggregates was figured by the formula: Xc = DHm/DH0m, where DHm was the melt enthalpy of the PCL nano-aggregates, and DH0m was the melt enthalpy of the fully crystallized PCL (135 J/g) [59]. Isothermal crystallization of PCL was carried out on a Perkin–Elmer Instruments DSC 8500 calorimeter. Calibration was
performed with indium and tin, and all tests were run employing ultrapure nitrogen as the purge gas. For the bulk sample, it was possible to extract the crystallization kinetics directly from the exothermic peak as a function of crystallization time, but for the aqueous dispersions, the crystallization process did not show a well-defined exothermic peak. Therefore, the sample was instead heated above the Tm of the PCL and quickly cooled to a preset crystalline temperature for a certain time. The Xc was obtained from the melting enthalpy during a subsequent heating process. Typically, for the PCL nano-aggregates in aqueous dispersions, high-pressure stainless-steel hermetic pans were utilized to avoid the evaporation of solvent during heating at high temperatures. About 30 mg aqueous dispersions of PCL nano-aggregates were encapsulated into the tightly sealed pan. The aqueous dispersions of PCL nano-aggregates was heated to 75 °C and kept for 10 min to ensure entirely melting of PCL chain. Thereafter, the heated sample was cooled to the designed isothermal crystallization temperature (Tic) in the DSC sampler with a cooling rate of 500 °C/min by means of liquid nitrogen accessory. The Xc was obtained from the melting enthalpy during a subsequent heating process performed at a heating rate of 5 °C/min. For PCL bulk, DSC heating and cooling scans were performed at 5 °C/min for measurements. Fluorescence measurements on the Py doped PCL nano-aggregates in aqueous dispersions were conducted on a Combined Fluorescence Lifetime and Steady State Spectrometer (Edinburgh Instrument Ltd., FLSP920) equipped with temperature control system (Oxford instruments). Excitation wavelength for Py was 322 nm. All measurements were made upon heating or cooling by increasing temperature in 5 °C increments and holding for 15 min at temperature before measuring the fluorescence spectrum. For the Py doped PCL nano-aggregates in aqueous dispersions, the emission wavelength was monitored in 0.5 nm increments from 360 to 405 nm. The flow temperature (Tf) of the Py doped PCL nano-aggregates in aqueous dispersions was determined via the temperature dependence of the ratio of fluorescence intensities at the first to third vibronic band peaks (I1/I3). Due to a slight shift in the position of the peaks with temperature, intensities were averaged over a 2 nm window, for example from 369 to 371 nm for the first peak and from 380 to 382 nm for the third peak [60]. Data were taken every 0.5 nm so that the averages consist of five data points. 3. Results and discussions 3.1. Preparation and thermal stability of PCL nano-aggregates in aqueous dispersions Polymer nano-aggregates can be conveniently prepared via emulsion polymerization of monomers [61], nanoprecipitation [57], solvent evaporation method [62] and so on [63,64]. Nanoprecipitation is a simple, fast and reproducible method which is widely used for the nano-aggregates preparation of the biodegradable polyesters, especially PCL [55,56] and polylactide (PLA) [65,66]. In the current study, PCL nano-aggregates with narrow size distribution and their aqueous dispersions were prepared by means of nanoprecipitation strategies. The tetrahydrofuran (THF) solution of PCL was added dropwise into an aqueous solution of Tween 80 under continuous magnetic stirring at room temperature and then THF was removed completely. The original aqueous dispersion of PCL nano-aggregates appeared as a bluish, translucent and stable colloidal solution. It is noteworthy that the PCL nano-aggragates were still wrapped by Tween 80 after dialysis, which was confirmed by elemental analysis (EA). The SEM, TEM and AFM images of the original PCL nano-aggregates with an average diameter hDi of 50 nm were shown in Fig. 1. It can be clearly seen from these images that spherical and
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
267
Fig. 1. The SEM (a), TEM (b) and AFM (c) images of the original 50 nm PCL nano-aggregates prepared via nanoprecipitation in aqueous dispersions.
Fig. 3. Py fluorescence emission spectrum of the original 50 nm PCL nanoaggregates (Py doped) in aqueous dispersions at various temperatures. Data was normalized by the maximum intensity of the spectrum at 25 °C. Inset shows the original 50 nm PCL nano-aggregates (Py doped) in aqueous dispersions.
Fig. 2. Size distributions of the 50 nm PCL nano-aggregates in aqueous dispersions (a: original, b: stored for 2 months under room temperature, and c: after DSC measurement).
narrowly size distributed PCL nano-aggregates were prepared. The thermal stability of the PCL nano-aggregates in aqueous dispersions was firstly investigated. As revealed in Fig. 2, the size
268
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
Fig. 4. Temperature dependence of the ratio of fluorescence intensities at the first to third peaks (I1/I3) of the original 50 nm PCL nano-aggregates (Py doped) in aqueous dispersions. (Data were obtained upon a: first heating scan, and b: first cooling scan.)
Fig. 6. Avrami plots for the 50 nm PCL nano-aggregates in aqueous dispersions and PCL bulk at different temperatures. Fig. 5. DSC heating curves of the original 50 nm PCL nano-aggregates and after isothermal crystallization for 3 h at different temperature in aqueous dispersions. The inset (a) presents the DSC second heating and first cooling curve of the PCL bulk; (b) shows the DSC cooling curve for the aqueous dispersions of PCL nanoaggregates.
distributions of the representative 50 nm PCL nanospheres in aqueous dispersions are not obviously altered whether stored for 2 months under room temperature or after DSC measurement, which indicates that the emulsified PCL nano-aggregates did not coalesce during storage or even the DSC measurement.
3.2. Rubbery M flow condensed state transition of the 50 nm PCL nanospheres in aqueous dispersions Nano-DSC was utilized to study the thermal behavior of the original 50 nm PCL nanospheres in aqueous dispersions. In this case, nano-DSC heating scans were performed at 2 °C/min over a temperature ranging from 25 to 70 °C. No melting peak was observed in the first nano-DSC heating trace even though the sensitivity of nano-DSC instrument was high enough to record 1 lJ thermal signal, which suggests that the condensed state of the original 50 nm PCL nanospheres should be amorphous. Fig. 3 shows Py fluorescence emission spectrum of the original
50 nm PCL nano-aggregates (Py doped) in aqueous dispersions at various temperatures. With increasing temperature, the total fluorescence intensity decreases as the pathways for non-radiative decay of the excited state dye (for example bond rotations and vibrations) are promoted [60,67]. Moreover, the ratio of the first peak (370 nm) to the third peak (381 nm) (I1/I3) decreases upon heating and the intersection of linear regressions was 45 °C (Fig. 4a). In addition, the ratio of I1/I3 increases upon cooling and the intersection of linear regressions was 48 °C (Fig. 4b). Note that the I1/I3 ratio was certificated to be sensitive to the local environment experienced by Py in the late 1970s [68], specifically the polarity, and has been used previously to determine glass transition for both polymer films and blends [60,67]. Therefore, the results of fluorescence measurements implied a condensed phase transition near 45 °C. For the condensed state of amorphous polymers, there include glass, rubbery and flow state according to the different molecular dynamics. When polymers are in glass state, both the segments and chains cannot move indicating that the molecular dynamics are frozen. When polymers enter in rubbery state from glass state with the rise of temperature, the segments are thawed and able to move. While as temperatures continue to rise, the polymers will enter in flow state and the permanent deformation of polymer is due to coordinated movement of segments.
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
Fig. 7. First nano-DSC heating curves for the original PCL nano-aggregates with different diameter (a: 50 nm, b: 150 nm, and c: 330 nm) in aqueous dispersions.
269
Consequently, there is reason to believe that the condensed state transition determined by fluorescence analysis can be attributed to the transition from rubbery state to flow state because the Tg of PCL is 60 °C [69] which is much lower than 45 °C. The Tf of the PCL bulk (Mn 45 kg/mol) is 66 °C which is measured via dynamic mechanical temperature spectrum, and is higher than that of the 50 nm PCL nano-aggregates. This may be related to the higher chain entanglement density in PCL bulk than in PCL nano-aggregates. It is noteworthy that, as shown in Fig. 3 inset, neither morphology nor average diameter of the PCL nano-aggregates (Py doped) is changed in comparison with that of the undoped one (shown in Fig. 1). And the condensed state of the original PCL nano-aggregates (Py doped) remained in amorphous state since no melting peak was observed in the first nano-DSC heating trace. Moreover, the PCL nano-aggregates (Py doped) were removed from the aqueous dispersions as centrifugated (Beckman-coulter Avanti J30I, USA) for 30 min at a speed of 25,000 rpm. The resultant supernate displays nearly no fluorescence emission, implying Py is almost completely encapsulated in the PCL nano-aggregates rather than diffusing into water.
Fig. 8. The SEM (a), TEM (b) and AFM (c) images of the original 150 nm PCL nano-aggregates prepared via nanoprecipitation in aqueous dispersions.
270
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
3.3. Confined crystallization of the 50 nm PCL nanospheres in aqueous dispersions Fig. 5 shows the DSC heating curves of the aqueous dispersions of 50 nm PCL nano-aggregates after isothermal crystallization at different temperatures ranging from 3 to 11 °C. When Tic > 0 °C, no melting peak was observed even the period of isothermal crystallization is prolonged from 3 to 12 h. Moreover, a crystalline melting peak attributed to crystallization of the PCL nano-aggregates is not apparently observed (Fig. 5) until the isothermal crystallization temperature dropped to 0 °C. Note that Tm and Xc remain constant as the crystallization time is more than 3 h, which indicates that the Xc of the PCL nano-aggregates has reached saturation within 3 h. Moreover, the Xc of the PCL nano-aggregates increased with the decrease of Tic. It is important to point out that a single melting peak for all the investigated systems is observed, which suggests that the system has only one kind of crystal structure [70,71]. Moreover, the 50 nm PCL nano-aggregates in aqueous dispersions exhibit a Tm of 52 °C, which is 8 °C lower than that of the PCL bulk (Fig. 5 inset (a)). This may be because of the crystal imperfections and/or the reduction of the lamellar thickness of PCL crystals due to the nanoconfinment [20,38,72,73]. Note that the aqueous dispersions of PCL nano-aggregates begun to freeze when the temperature drops below 13 °C as shown in Fig. 5 inset (b). In case of any impact (hard confinement caused by the freezing of the surrounding aqueous media) on the crystallization of PCL nano-aggregates from freezing of the aqueous dispersions, Tic was set above 13 °C. The isothermal crystallization kinetics of the 50 nm PCL nano-aggregates in aqueous dispersions was further investigated. As shown in Fig. 6, PCL exhibits an Avrami index of 3 in the bulk, indicating unconfined two-dimensional crystal growth after heterogeneous nucleation [1]. However, the Avrami index of the 50 nm PCL nanoaggregates in aqueous dispersions was close to 1, implying that this confined crystallization was dominated by nucleation [1,40]. It is generally recognized that when a polymer is divided into nanosized aggregates, and the number of nano-aggregates is much more than the number of active heterogeneous nuclei, the nucleation and crystallization of the polymer can experience dramatic changes, which usually occurs along with substantial supercooling [1,3,26,38,40]. As for the crystallization of heterogeneity-free nano-aggregates, the nucleation typically changes from heterogeneous nucleation to homogeneous nucleation [1,15,22,26,28,49]. However, as demonstrated by Carvalho and Dalnoki-Veress, the surface of the microdomains or the interface between the crystallizable nano-aggregates and the matrix surrounding them can induce the nucleation (i.e. surface nucleation) [3,18,27]. Furthermore, a first order or lower (n 6 1) crystallization kinetics can also be found for confined polymers undergoing both homogenous and surface nucleation [1,3,12,27]. Nevertheless, surface nucleation requires lower supercoolings than classic homogeneous nucleation, because it is a process characterized by a lower free energy than the more energetically costly process of creating new crystal surfaces inside an nano-aggregates [3,18,27]. In this case, the size of the nano-aggregates was 50 nm, the number of nano-aggregates greatly exceeded the number of heterogeneous nuclei, which giving rise to exclusively homogeneous or surface nucleation. As reported previously by Nojima et al., PCL blocks or homopolymers confined in spherical and cylindrical nano-aggregates exhibit a homogeneous nucleation and first-order crystallization kinetics at very low crystallization temperature of 45 °C which are much closer to their corresponding Tg values [29,74]. In this situation, however, the highest crystallization temperature of 0 °C is much higher than the reported value. We suggest that the surface/interface of the nano-aggregates probably play an important role in nucleation. Therefore, a surface nucleation
Fig. 9. The SEM image of the original 330 nm PCL nano-aggregates prepared via nanoprecipitation in aqueous dispersions.
mechanism was more plausible for the confined crystallization of the 50 nm PCL nano-aggregates in aqueous dispersions. 3.4. Condensed state of the PCL nano-aggregates with a diameter of more than 50 nm in aqueous dispersions Fig. 7 displays the first nano-DSC heating curves of the original PCL nano-aggregates in aqueous dispersions. Melting peaks are obviously observed for the original 150 and 330 nm PCL nanoaggregates, indicating that PCL is in semi-crystalline state as the diameter is increased more than 150 nm. Moreover, the Xc of the original PCL nano-aggregates increases with the diameter. Fig. 8 presents the typical SEM, TEM and AFM images of the original PCL nano-aggregates with a hDi of 150 nm and a height of 45 nm (Fig. 8c), indicating that ellipsoid-like nano-aggregates is formed in terms of two-dimensional growth of the PCL crystalline. Fig. 9 displays the SEM image of the original PCL nano-aggregates with a hDi of 330 nm which is consist of disc-shaped lamellae stacked each other. It was believed that the morphology of the original nano-aggregates in aqueous dispersions changes from spherical to ellipsoid-like even lamellae stacked with the increase of Xc. 4. Conclusions To our best knowledge, most studies on condensed state physics concerned polymer nano-aggregates in bulk, such as self-assembled nanomicelles of block copolymers in bulk [1–3]. Whereas only a few reports focus their attention on the condensed state physics of polymer nano-aggregates in solution dispersion [13,16,20,37–40]. In this work, we have demonstrated the condensed state physics of the narrowly distributed spherical or ellipsoid-like even lamellae stacked PCL nano-aggregates with a diameter ranging from 50 to 330 nm in aqueous dispersions which were prepared by nanoprecipitation strategies using Tween 80 as a surfactant and THF as a co-solvent at room temperature. The original PCL aggregates with a diameter of less than 100 nm, for representatively 50 nm, show spherical morphology and rubbery state. Fluorescence analysis indicated that the transition temperature from rubbery state to flow state was 45 °C. Meanwhile, the 50 nm PCL nanospheres show a supercooling point of <0 °C during crystallization from the rubbery state, which is much lower than that of PCL bulk. Furthermore, an investigation on the isothermal crystallization kinetics of PCL in the 50 nm nanospheres displays that the Avrami index n was
S. Feng et al. / Journal of Colloid and Interface Science 450 (2015) 264–271
close to 1, indicating that the confined crystallization nucleation is dominated by nucleation. These findings about the large supercooling and first-order crystallization kinetics for crystallization of the 50 nm PCL nano-aggregates in aqueous dispersions are in line with previous studies carried on self-assembled nanomicelles of block copolymers in bulk [53,74]. However, the highest crystallization temperature of 0 °C is much higher than the reported value demonstrated in bulk system, implying that surface of the PCL nano-aggregates may play an important role in the crystallization process [3,27]. By contrast, the crystallization of PCL chain can promote the formation of ellipsoid-like nano-aggregates as the diameter is more than 150 nm. Interestingly, when the diameter is increased to 330 nm, spherical aggregates consisted of discshaped PCL crystalline lamellae stacked each other was formed. Briefly, our work thoroughly investigated the condensed state and condensed state transition of the poly(e-caprolactone) (PCL) nano-aggregates in aqueous dispersions which is of biomedical application. Significantly, it may contribute to understanding the relationship between the condensed state of PCL nano-aggregates and its drug loading capacity since most biomedical application of PCL-aggregates are carried out in aqueous dispersions rather than in dried state [13]. Acknowledgments This work was supported by the National Natural Science Foundation of China (21174167), the NSF of Guangdong Province (S2013030013474) and the Guangdong Province Higher School Science and Technology Innovation Key Project. References [1] A.J. Müller, V. Balsamo, M.L. Arnal, Adv. Polym. Sci. 190 (2005) 1. [2] M.C. Lin, B. Nandan, H.L. Chen, Soft Matter 8 (2012) 7306. [3] R.M. Michell, I. Blaszczyk-Lezak, C. Mijangos, A.J. Müller, Polymer 54 (2013) 4059. [4] Y.-X. Liu, E.-Q. Chen, Coord. Chem. Rev. 254 (2010) 1011. [5] J.M. Carr, D.S. Langhe, M.T. Ponting, A. Hiltner, E. Baer, J. Mater. Res. 27 (2012) 1326. [6] Y. Ma, Hu, G. Reiter, Macromolecules 39 (2006) 5159. [7] D.E. Martínez-Tong, B. Vanroy, M. Wübbenhorst, A. Nogales, S. Napolitano, Macromolecules 47 (2014) 2354. [8] P.C. Jukes, A. Das, M. Durell, D. Trolley, A.M. Higgins, M. Geoghegan, J.E. Macdonald, R.A.L. Jones, S. Brown, P. Thompson, Macromolecules 38 (2005) 2315. [9] E. Woo, J. Huh, Y.G. Jeong, K. Shin, Phys. Rev. Lett. 98 (2007) 136103. [10] Y. Zhou, S.-K. Ahn, R.K. Lakhman, M. Gopinadhan, C.O. Osuji, R.M. Kasi, Macromolecules 44 (2011) 3924. [11] Y.-L. Loo, R.A. Register, A.J. Ryan, Phys. Rev. Lett. 84 (2000) 4120. [12] A. Taden, K. Landfester, Macromolecules 36 (2003) 4037. [13] E.C. Cho, K. Cho, J.K. Ahn, J. Kim, I.S. Chang, Biomacromolecules 7 (2006) 1679. [14] Z.-X. Du, J.-T. Xu, Z.-Q. Fan, Macromolecules 40 (2007) 7633. [15] L. Kailas, C. Vasilev, J.-N. Audinot, H.-N. Migeon, J.K. Hobbs, Macromolecules 40 (2007) 7223. [16] C.H. Weber, A. Chiche, G. Krausch, S. Rosenfeldt, M. Ballauff, L. Harnau, I. Gottker-Schnetmann, Q. Tong, S. Mecking, Nano Lett. 7 (2007) 2024. [17] S. Nojima, Y. Ohguma, S. Namiki, T. Ishizone, K. Yamaguchi, Macromolecules 41 (2008) 1915. [18] J.L. Carvalho, K. Dalnoki-Veress, Eur. Phys. J. E, Soft Matter 34 (2011) 1. [19] J.L. Carvalho, M.E. Somers, K. Dalnoki-Veress, J. Polym. Sci., Part B: Polym. Phys. 49 (2011) 712. [20] C.N. Rochette, S. Rosenfeldt, K. Henzler, F. Polzer, M. Ballauff, Q. Tong, S. Mecking, M. Drechsler, T. Narayanan, L. Harnau, Macromolecules 44 (2011) 4845. [21] J.-C. Zhao, B. Xing, Z. Peng, J.-M. Zhang, Chin. J. Polym. Sci. 31 (2013) 1310.
271
[22] C.L. Huang, L. Jiao, J.B. Zeng, J.J. Zhang, K.K. Yang, Y.Z. Wang, J. Phys. Chem. B 117 (2013) 10665. [23] H.D. Huang, J.Z. Xu, Y. Fan, L. Xu, Z.M. Li, J. Phys. Chem. B 117 (2013) 10641. [24] A.J. Müller, V. Balsamo, M.L. Arnal, T. Jakob, H. Schmalz, V. Abetz, Macromolecules 35 (2002) 3048. [25] A.T. Lorenzo, M.L. Arnal, A.J. Müller, A. Boschetti-de-Fierro, V. Abetz, Macromolecules 40 (2007) 5023. [26] R.V. Castillo, M.L. Arnal, A.J. Müller, I.W. Hamley, V. Castelletto, H. Schmalz, V. Abetz, Macromolecules 41 (2008) 879. [27] J.L. Carvalho, K. Dalnoki-Veress, Phys. Rev. Lett. 105 (2010) 237801. [28] M. Massa, J. Carvalho, K. Dalnoki-Veress, Phys. Rev. Lett. 97 (2006) 247802. [29] S. Nojima, M. Toei, S. Hara, S. Tanimoto, S. Sasaki, Polymer 43 (2002) 4087. [30] L. Zhu, B.R. Mimnaugh, Q. Ge, R.P. Quirk, S.Z.D. Cheng, E.L. Thomas, B. Lotz, B.S. Hsiao, F. Yeh, L. Liu, Polymer 42 (2001) 9121. [31] H.-L. Chen, H.-C. Li, Y.-Y. Huang, F.-C. Chiu, Macromolecules 35 (2002) 2417. [32] M.D. Ediger, J.A. Forrest, Macromolecules 47 (2014) 471. [33] M.V. Massa, K. Dalnoki-Veress, J.A. Forrest, Eur. Phys. J. E, Soft Matter 11 (2003) 191. [34] H. Schönherr, C.W. Frank, Macromolecules 36 (2003) 1188. [35] Y.J. Ren, Z. Huang, W.B. Hu, J. Macromol. Sci. Part B – Phys. 51 (2012) 2341. [36] B. Vanroy, M. Wübbenhorst, S. Napolitano, ACS Macro Lett. 2 (2013) 168. [37] L. Yin, M.A. Hillmyer, Macromolecules 44 (2011) 3021. [38] Z.Y. Li, R. Liu, B.Y. Mai, W.J. Wang, Q. Wu, G.D. Liang, H.Y. Gao, F.M. Zhu, Polymer 54 (2013) 1663. [39] B.Y. Mai, Z.Y. Li, R. Liu, S. Feng, Q. Wu, G.D. Liang, H.Y. Gao, F.M. Zhu, J. Polym. Res. 20 (2013) 1. [40] R. Liu, Z.Y. Li, B.Y. Mai, Q. Wu, G.D. Liang, H.Y. Gao, F.M. Zhu, J. Polym. Res. 20 (2013) 64. [41] O. Coulembier, P. Degée, J.L. Hedrick, P. Dubois, Prog. Polym. Sci. 31 (2006) 723. [42] B. Amsden, Soft Matter 3 (2007) 1335. [43] L. Yu, J.D. Ding, Chem. Soc. Rev. 37 (2008) 1473. [44] M.A. Woodruff, D.W. Hutmacher, Prog. Polym. Sci. 35 (2010) 1217. [45] M.L. Gou, K. Men, H.S. Shi, M.L. Xiang, J.A. Zhang, J. Song, J.L. Long, Y. Wan, F. Luo, X. Zhao, Z.Y. Qian, Nanoscale 3 (2011) 1558. [46] X. Gao, B.L. Wang, X.W. Wei, K. Men, F.J. Zheng, Y.F. Zhou, Y. Zheng, M.L. Gou, M.J. Huang, G. Guo, N. Huang, Z.Y. Qian, Y.Q. Wei, Nanoscale 4 (2012) 7021. [47] S.M. Garg, X.B. Xiong, C. Lu, A. Lavasanifar, Macromolecules 44 (2011) 2058. [48] H.F. Sun, L. Mei, C.X. Song, X.M. Cui, P.Y. Wang, Biomaterials 27 (2006) 1735. [49] S. Nakagawa, T. Tanaka, T. Ishizone, S. Nojima, Y. Kakiuchi, K. Yamaguchi, S. Nakahama, Macromolecules 46 (2013) 2199. [50] Z.-X. Du, J.-T. Xu, Z.-Q. Fan, Macromol. Rapid Commun. 29 (2008) 467. [51] K. Schäler, A. Achilles, R. Bärenwald, C. Hackel, K. Saalwächter, Macromolecules 46 (2013) 7818. [52] Y. Suzuki, H. Duran, M. Steinhart, H.-J. Butt, G. Floudas, Macromolecules 47 (2014) 1793. [53] R.M. Michell, A.J. Müller, G. Deshayes, P. Dubois, Eur. Polym. J. 46 (2010) 1334. [54] R.-M. Ho, Y.-W. Chiang, C.-C. Lin, B.-H. Huang, Macromolecules 38 (2005) 4769. [55] J. Rieger, C. Passirani, J.P. Benoit, K. Van Butsele, R. Jerome, C. Jerome, Adv. Funct. Mater. 16 (2006) 1506. [56] F. Lince, D.L. Marchisio, A.A. Barresi, J. Colloid Interface Sci. 322 (2008) 505. [57] C. Zhang, J.W. Chung, R.D. Priestley, Macromol. Rapid Commun. 33 (2012) 1798. [58] C. Zhang, V.J. Pansare, R.K. Prud’homme, R.D. Priestley, Soft Matter 8 (2012) 86. [59] V. Crescenzi, G. Manzini, G. Calzolari, C. Borri, Eur. Polym. J. 8 (1972) 449. [60] S. Kim, C.B. Roth, J.M. Torkelson, J. Polym. Sci., Part B: Polym. Phys. 46 (2008) 2754. [61] K. Min, H.F. Gao, K. Matyjaszewski, J. Am. Chem. Soc. 127 (2005) 3825. [62] Y.Y. Yang, T.S. Chung, N.P. Ng, Biomaterials 22 (2001) 231. [63] K. Landfester, Angew. Chem. Int. Ed. Engl. 48 (2009) 4488. [64] J.P. Rao, K.E. Geckeler, Prog. Polym. Sci. 36 (2011) 887. [65] T. Govender, S. Stolnik, M.C. Garnett, L. Illum, S.S. Davis, J. Control. Release 57 (1999) 171. [66] Y.C. Dong, S.S. Feng, Biomaterials 25 (2004) 2843. [67] C.J. Ellison, J.M. Torkelson, Nat. Mater. 2 (2003) 695. [68] K. Kalyanasundaram, J.K. Thomas, J. Am. Chem. Soc. 99 (1977) 2039. [69] J.V. Koleske, R.D. Lundberg, J. Polym. Sci. Part A – 2: Polym. Phys. 7 (1969) 795. [70] E.S. Yoo, S.S. Im, J. Polym. Sci., Part B: Polym. Phys. 37 (1999) 1357. [71] Z.B. Qiu, M. Komura, T. Ikehara, T. Nishi, Polymer 44 (2003) 7781. [72] F.M. Bauers, R. Thomann, S. Mecking, J. Am. Chem. Soc. 125 (2003) 8838. [73] K. Shin, E. Woo, Y.G. Jeong, C. Kim, J. Huh, K.-W. Kim, Macromolecules 40 (2007) 6617. [74] S. Nojima, Y. Ohguma, K.-I. Kadena, T. Ishizone, Y. Iwasaki, K. Yamaguchi, Macromolecules 43 (2010) 3916.