It has been well realized that environmental pollution has become significantly serious nowadays, and it is partially caused by the abandoning of traditional nondegradable materials. Degradable and recyclable polymers show increasing application potential and have become candidates to replace petroleum-based polymers. Poly(l
-lactide) (PLLA), a typical biocompatible, biodegradable, renewable, and nontoxic thermoplastic polymer, has attracted much attention for investigation [1
]. PLLA has been widely used in various fields such as packaging, medical devices, and biological engineering applications for its excellent thermal plasticity and easy processability [7
]. However, its poor thermal resistance and mechanical properties have somewhat limited its wider application [10
]. PLLA is a typical semicrystalline material, and its properties depend largely on its crystallinity and crystal structure [12
]. Therefore, similar to other semicrystalline polymers, including isotactic polypropylene (iPP) [13
], polycaprolactone (PCL) [17
], poly(vinylidene fluoride) (PVDF) [18
], and poly(cyclohexylene dimethylene cyclohexanedicarboxylate) (PCCE) [19
], controlling crystallization has gradually become known as an effective way to obtain the required performance of PLLA parts. These works can be roughly classified into three categories when considering large-scale industrial applications. The first is to incorporate a plasticizer agent into the PLLA matrix and form blends, such as poly(ethylene glycol) (PEG) [21
], glycerol [23
], poly(caprolactone) (PCL) [24
-lactide) (sPCLA) [25
-lactic acid)-grafted cellulose [26
], and poly(propylene glycol) (PPG) [27
]. However, the existence of these plasticizers cannot actually improve the crystallization rate of pure PLLA, and it is even the opposite in some cases. Another approach is to adjust the processing conditions [28
], which has been reported to have an obvious effect on the crystallization process and the ultimate crystallinity. It is well known that the amorphous phase will appear if the quench temperature is very low and the crystals have no time to form. However, improving the crystallinity is often not effective if only the temperature gradient and cooling rate are controlled, and the temperature cannot be controlled easily in real manufacturing. The final and maybe the most promising approach is to add some organic and/or inorganic particles as nucleating agents to the PLLA matrix, such as sodium stearate [30
], nano-clays [31
], graphite particles [32
], triclosan nanoparticles [33
], tungsten disulphide inorganic nanotubes (INT-WS2
], or magnesium oxide nanoparticles [35
]. Compounding with these particles by physical blending is a conventional method of fabricating PLLA composites and was proved to be convenient to improve the crystallization behavior of PLLA. However, the agglomeration of nanoparticles easily affects the efficiency and effects. How to improve the dispersion and compatibility of nanoparticles in the PLLA matrix during nanoparticle filling is a general concern.
Polyhedral oligomeric silsesquioxane (POSS), one member of the silsesquioxane family, is a type of intermediate cubic polyhedron that has a molecular weight up to 1000 and a nanoscale cage structure. The nanosized cage structure of POSS comprises both inorganic and organic substituents that form a framework of silicon and oxygen atoms and branched chains composed by a hydrocarbon group or a polar functional group [36
]. By adjusting the substituents of the POSS cage block at the molecule level in the processes of copolymerization, polycondensation, homopolymerization, and physical blending, the intermediate and composited materials with required functionalization can be fabricated and manufactured in a controlled way. In addition, an important feature of POSS is that its Si-O-Si bond, which has a diameter of 1.5 nm, can build an inorganic framework with the reactive and nonreactive groups when it is added to an appropriate polymer, which can easily control the compatibility between POSS and the polymer matrices. Hence, the POSS can be inserted into the polymer and function as a nanofiller in processes such as copolymerization, grafting reaction, and in situ polymerization, resulting in substantial improvements in thermal stability, crystallinity, and barrier properties of the polymeric materials [37
]. However, the bond–bond force in the polymer matrix generally stays at the molecular scale and is not effective enough, so the interface adhesion between the POSS and the polymer matrix still needs to be improved. Furthermore, the nanosized POSS particles are difficult to disperse uniformly into the polymer matrix in the melt and solution blending processes. As shown in Figure 1
a, it is easily observed that the POSS nanoparticles agglomerate after the solution blending.
In order to improve the compatibility between the POSS particles and the polymer matrices, many attempts have been made using a suitable process of polymerization other than simple physical blending. The POSS particle was first used as a precursor in the synthesis of some kinds of nanohybrid materials such as amines [40
], alkyls [41
], aminos, and bromophenyls [42
]. It is known that oxytrol and amino groups can accelerate the ring-opening polymerization of lactide monomers when they are used as initiators [44
]. Therefore, the POSS nanoparticles, which have a core-shell type containing amino groups, were directly dispersed in PLA-based systems using in situ polymerization, which was found to be feasible and relied on the functionalized groups connected with the octa-POSS [45
]. The developing hydrogen bond of the hydroxy group of PLLA resin can react with the amino group of POSS particles, and the graft structure between the POSS and the matrix is an interface interaction, leading to a more stable microstructure of the composites and extreme boosting of the glass transition temperature and improving melt flow behavior. As a result, it was reported that the elongation at break of PLLA/POSS composites increased approximately 10-fold compared to that of pure PLLA resin when other physical properties remained unchanged [46
]. Apart from the PLA matrix, POSS nanoparticles can also be introduced into other polyester polymers, such as polybutylene succinate (PBS) and poly(butyleneadipate-co-terephthalate) (PBAT), by triggering open-ring polymerization to influence the various properties of the composites. However, it must be noted that a negative correlation between the content of the POSS particles and the number of molecules of the polymerized materials was also found [47
]. This is attributed to the macromolecule chains of the polymerized materials being deeply influenced by the number of crosslinking points provided by the end group of silsesquioxane. This means that the content of the POSS particles in the composite cannot easily control the number of molecules of polymerized materials if the nanosized POSS particles are not dispersed uniformly.
Combining ring-opening polymerization and physical blending can be viewed as a revolutionary method to further improve the interfacial interactions between POSS nanoparticles and PLLA matrices. The representative work was performed by Liu and Lu et al. [48
]. They synthesized eight branched PLLA arms and a three-dimensional POSS structure with a cube-like core via ring-opening polymerization initiated by octa(3-hydroxypropyl)-POSS, and then compounded the obtained POSS-(PLLA)8 with the PLLA matrix. The obtained PLLA/POSS-(PLLA)8 composite was found to have dramatically increased mechanical properties compared with the neat PLLA resin. However, the requirement of anionic polymerization is relatively strict during the multi-arm ring-opening polymerization of the eight-branched POSS-PLLA composites. In addition, the eight arms were somewhat short, so the chain mobility of the obtained POSS-PLLA composites was probably limited, and hence the crystallization of the PLLA was influenced and blocked.
In this study, we propose a novel method to fabricate PLLA and POSS nanocomposites by combining in situ ring-opening polymerization and solution blending. For the first time, we applied nanoparticles of (3-amino) propylheptaisobutyl cage silsesquioxane POSS (AMPOSS) as an initiator and stannous (II) octoate (Sn(Oct)2) as a catalyst to trigger the ring-opening polymerization of l-lactide monomers. A large number of intermediates with different mass fractions were prepared and manufactured with the aim of producing a reactive functional group. The obtained PLLA-POSS was then incorporated into the pure PLLA homopolymers to form PLLA/PLLA-AMPOSS composites by means of solution blending. Moreover, to explore the reactive mechanism and modified effect, the microscopic chemical structures of the PLLA-POSS intermediate and PLLA/PLLA-AMPOSS composite were characterized by 1H-NMR spectroscopy and X-ray diffraction (XRD) analysis. The thermal properties, crystallization kinetics, crystallization morphology, and distribution of the dispersed phase of the PLLA/PLLA-AMPOSS composite were also investigated by thermogravimetric analysis (TGA), differential scanning calorimetry (DSC), and scanning electron microscopy (SEM), respectively. In addition, the mechanical properties of the obtained nanocomposite were tested and compared by the tensile test.
Poly (l-lactide) (PLLA, Mn = 1.0 × 105g/mol) used in this study was a commercial-grade product purchased from Jiuding Biological Engineering Co., Ltd. (Nantong, China), and l-lactide was self-made in our laboratory with a residual monomer content less than 0.5 wt%. Polyhedral oligomeric silsesquioxane (POSS) nanoparticles were obtained from Hybrid Plastics Co., Ltd., Hattiesburg, MI, USA. Stannous (II) octoate (Sn(Oct)2, 95%) was obtained from Macklin Biochemical Co., Ltd. (Shanghai, China). Dichloromethane (CH2CL2), dried methyl alcohol, and ethyl formate (CH3COOC2H5) were kindly supplied by Sinopharm Chemical Reagent Co., Ltd., Shanghai, China, and were of analytical grade, and there was no need for further purification.
2.2. Sample Preparation
The sample preparation was realized in two main stages, synthesis of PLLA-AMPOSS intermediates by single-arm ring-opening polymerization and fabrication of PLLA/PLLA-AMPOSS nanocomposites.
2.2.1. Synthesis of PLLA-AMPOSS Intermediates by Ring-Opening Polymerization
The fabrication of PLLA-AMPOSS intermediates was separated into the following steps. First, the l
-lactide (LLA) monomer was purified by recrystallization 3 times in a solution of ethyl acetate. Then, the purified LLA monomer was added to a flask ampoule that contained xylene solution. The ampoule was flame-dried and equipped with a stirring bar before use. Following that, the AMPOSS nanoparticles were added to the LLA solution, and its mole ratio to LLA was fixed at 0.02%, 0.05%, 0.10%, 0.20%, 0.50%, and 1.00%. Subsequently, a Sn(Oct)2
solution was injected into the ampoule, and its usage was 1/20,000 of the LLA monomer in mole fraction. After that, the reaction vessel was transferred to an oil bath at 120 °C under vigorous stirring until the mixture thoroughly melted. Then, the mixture was heated to 150 °C with the help of magnetic stirring under an atmosphere of dry nitrogen, and the LLA monomer and AMPOSS nanoparticles began to undergo ring-opening polymerization. After the reaction lasted for 6 h, the products were rapidly quenched using a refrigerator to terminate polymerization. The final reaction products, which were named PLLA-AMPOSSx
intermediate, were dissolved in chloroform, precipitated into excess methanol, and filtrated and dried in vacuum at 50 °C for 24 h. The subscript of PLLA-AMPOSSx
intermediate denotes the molar percentage of AMPOSS to LLA monomer. A series of intermediates were thus obtained: PLLA-AMPOSS0.02
, and PLLA-AMPOSS1.0
. The reaction scheme used in this study is shown in Figure 1
2.2.2. Fabrication of PLLA/PLLA-AMPOSS Nanocomposites
The obtained PLLA-AMPOSS0.50
intermediate was then used to fabricate PLLA/PLLA-AMPOSS nanocomposites. An optional synthesis procedure of the PLLA/PLLA-AMPOSS nanocomposites can be described as follows. Pure PLLA resin and PLLA-AMPOSS0.50
intermediate were dissolved in dichloromethane solution sequentially with vigorous magnetic stirring, and the content of PLLA-AMPOSS0.50
intermediate was 1%, 5%, 10%, 20%, and 30% in weight. Subsequently, the solution was precipitated with overdosed methanol, and the obtained nanocomposites were dried in vacuo at 50 °C for 24 h. The nanocomposites were named PLLA/PLLA-AMPOSSy
, where the superscript y represents the weight percentage of PLLA-AMPOSS0.50
intermediate. There were 5 kinds of PLLA/PLLA-AMPOSS nanocomposites used in this study: PLLA/PLLA-AMPOSS1
, and PLLA/PLLA-AMPOSS30
. For conventional comparison, simple solution blends of PLLA resin and AMPOSS nanoparticles with content of 0.1 wt% and 1.0 wt%, respectively, were also prepared using the traditional method, as shown in Figure 1
a, and named as PLLA/AMPOSS blends: PLLA/0.1wt%AMPOSS and PLLA/1.0wt%AMPOSS depending on the content of AMPOSS. The PLLA/1.0wt%AMPOSS and PLLA/0.1wt%AMPOSS blends were used to compare with the PLLA-AMPOSS intermediates and PLLA/PLLA-AMPOSS nanocomposites, respectively.
2.3. Sample Tests
Proton nuclear magnetic resonance (1H-NMR) spectra were probed by a 600 MHz NMR spectrometer (Bruker AVANCE II 600 MHz, Madison, WI, USA) in deuterated chloroform (CDCl3) at room temperature, with tetramethylsilane (TMS) used as the internal reference.
X-ray diffraction patterns were obtained with an XRD-6000 diffractometer (Shimadzu Co., Kyoto, Japan) equipped with Ni-filtered Cu Kα radiation (wavelength, λ = 0.154 nm), running at 40 kV and 200 mA at a scanning rate of 4.0° min−1 and with a scanning range 2θ = 5–40°. All the samples were heated to 200 °C, held for 5 min, and cooled down to 110 °C at a cooling rate of 100 °C/min until the crystallization of PLLA was completed.
The thermal properties of the sample, approximately 5 mg of the fabricated PLLA/PLLA-AMPOSS nanocomposites, was recorded by differential scanning calorimeter (DSC 204 F1 Phoenix, NETZSCH, Bavaria, Germany) under a nitrogen purge. The sample was heated to 220 °C at a rate of 10 °C/min and held for 5 min to eliminate the thermal history, and then cooled down to room temperature at a rate of 40 °C/min. The sample was heated from room temperature to 220 °C at a rate of 10 °C/min for the second time. In addition, to investigate the isothermal crystallization kinetics, after being heated to 220 °C at a rate of 20 °C/min and held for 5 min, the sample was cooled down to the desired crystallization temperature (Tc) at a rate of 40 °C min−1 until the isothermal crystallization process was completed. All exothermal traces were collected and analyzed.
A thermal gravimetric analyzer (TGA) (Pyris Diamond, Perkin Elmer, Waltham, MA, USA) was used to investigate the thermal stability of the sample under nitrogen and air atmosphere. The sample, with a mass of 20 mg, was heated from room temperature to 600 °C at a rate of 10 °C/min.
Spherulitic morphology image of the PLLA/PLLA-AMPOSS nanocomposite was observed directly using polarized optical microscopy (POM) (50iPOL, Nikon, Tokyo, Japan) with a hot stage. The selective samples were initially sandwiched in 2 cover glasses, then melted at 200 °C and compressed in a film for 3 min to remove the thermal-mechanical history. Following that, the samples were cooled down to 125 °C with a high cooling rate, and kept for the necessary time until the crystallization was completed. A comparison of the crystal’s morphology over 40 s and 200 s was recorded.
A scanning electron microscope (SEM, XL-30; Philips, Amsterdam, Holland) was used to examine the compatibility and dispersion status of the AMPOSS nanoparticles. The PLLA-AMPOSS blend and PLLA/PLLA-AMPOSS nanocomposites were immersed in liquid nitrogen for about 1 h and then freeze-fractured. The fractured surfaces were sputtered with gold before observation.
Tensile tests were carried out according to ASTM D638-10 standards, using a screw-driven universal testing instrument (CMT-4303, Chengde, China) under room conditions. A crosshead speed of 50 mm/min was used to study the stress and strain behavior of the molded samples. Seven tensile bars were tested for each formula, and the biggest and smallest values were excluded. Hence, 5 values were selected for analysis, and the mean and range of ultimate tensile strength and strain at break for each group of samples were calculated and reported.