Composites of poly(L-lactide-trimethylene carbonate-glycolide) and surface modified calcium carbonate whiskers as a potential bone substitute material

Abstract

Calcium carbonate whisker (CCW) particles were surface modified by grafting of poly(L-lactide) (PLLA) chains in order to improve their affinity to a poly(L-lactide-trimethylene carbonate-glycolide) (PLTG) terpolymer matrix. Composites of the PLTG matrix with CCW and PLLA-g-CCW of various contents were prepared by mixing in solution followed by solvent evaporation. The structure and properties of pure CCW, surface modified PLLA-g-CCW and PLTG/PLLA-g-CCW composites were investigated using Fourier transfer-infrared spectrometry (FT-IR), mechanical testing, scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDX), X-ray diffraction (XRD), contact angle measurement, thermal gravity analysis (TGA) and differential scanning calorimetry (DSC). Data show that PLLA chains were successfully grafted on the CCW surface. The PLTG/PLLA-g-CCW composites exhibit a higher tensile strength and elongation at break than neat PLTG. Optimal values are obtained with a PLLA-g-CCW content of 2 wt%. It is assumed that PLLA-g-CCW particles present both reinforcing and toughening effects on the PLTG matrix. The cytocompatibility of the materials was evaluated from cell morphology and MTT assay using the L929 mouse fibroblast cell line. The results indicate that the composite presents very low cytotoxicity. Therefore, PLTG/PLLA-g-CCW composites with improved mechanical properties and good cytocompatibility could be promising as a potential bone substitute material.

---

1 Introduction

The increasing need for bone fixation and bone repair accelerated the development of novel polymeric materials with better mechanical properties and biocompatibility.^1–3 In the past decades, bioresorbable synthetic polymers such as polylactide (PLA), polyglycolide (PGA), poly(trimethylene carbonate) (PTMC), and various copolymers have received growing attention because of their biocompatibility and degradability.^4–13 Among them, poly(L-lactide) (PLLA) appears most promising because PLLA has high strength and degrades into nontoxic degradation products in vivo. In fact, PLLA has already reached the stage of clinical utilization as surgical implants for internal fixation of bone fractures.^14–17 But its high rigidity and crystallinity, acidic degradation products and slow degradation constitute the main disadvantages for larger applications.¹⁸ PTMC is an amorphous polymer with a glass transition at about −12 °C, and is thus a rubbery material at room temperature. It hardly degrades by pure hydrolysis, but rapidly degrades in vivo yielding neutral products, i.e., diols and carbon dioxide.¹⁹ PTMC exhibits good flexibility, and is widely used as a softening component together with other polymers to prepare bioresorbable sutures known as Maxon®, Monosyn® or Biosyn®.^20,21 On the other hand, PGA is a fast degrading polyester with high strength and low elongation at break. Meanwhile, the GA monomer shows much higher reactivity than the LLA and TMC ones.^22,23 Therefore, incorporation of GA units into PLLA–TMC chains should allow to modulate the properties of materials, especially the degradation rate.

The first terpolymers of LLA, TMC and GA were synthesized by Zini et al.²⁴ The authors described the shape memory properties of the resulting terpolymers for minimal invasive applications. It is well known that the degradation rate and mechanical properties in body environment can be tailored by varying the copolymer molar mass and composition. Thus in our previous work, PLLA-TMC-GA (PLTG) terpolymers with a broad range of compositions were synthesized to evaluate their potential as cardiovascular stent material.^25–27 It was observed that PLTG terpolymers with high LLA contents exhibit greatly improved toughness as compared to PLLA homopolymer in spite of slight loss of tensile strength. The in vitro and in vivo degradation studies show that degradation of PLTG terpolymers is rather complex and depends on many factors such as copolymer composition, molar mass, morphology, and chain structure. Amorphous areas and GA component are more susceptible to hydrolysis, leading to compositional and morphological changes. In particular, the mechanical strength of polymers can be conserved beyond 3 months.^26,27

The mechanical properties of an ideal bone scaffold should match those of the host bone which greatly vary from cancellous to cortical bone. The Young's modulus of cortical bone is between 15 and 20 GPa, and that of cancellous bone is between 0.1 and 2 GPa. The compressive strength varies between 100 and 200 MPa for cortical bone, and between 2 and 20 MPa for cancellous bone.²⁸ In the past decade, novel composites of bioresorbable polymers and bioceramics with improved mechanical properties and biocompatibility have been studied for bone repair applications.^29–31

The most representative bioactive ceramics are bioglass, hydroxyapatite (HA) and glass-ceramics containing HA or its components, such as CaO and P₂O₅. While the excellent biological performance of HA and related ceramics have been well documented, their relatively slow biodegradation and low mechanical strength limit applications in bone engineering, especially at load-bearing sites.^32,33 Other materials also present various bioactivity after a simple chemical treatment.^34,35 As a bioactive inorganic material with high mechanical strength, CaCO₃ has been extensively studied for biomedical applications in combination with biopolymers.^36–39 More importantly, addition of slightly basic CaCO₃ in a polyester matrix should help to neutralize acidic products produced by degradation of polyesters, thus minimizing the risk of inflammation.^40–43 Nevertheless, only physical adsorption exists between bioceramic and polymeric matrix in these composites, and the improvement of mechanical properties is rather limited. It is thus of major importance to enhance the compatibility between the filler and the matrix.

This work aims to develop novel composites from PLTG terpolymer and surfaced modified calcium carbonate whiskers (CCW). Aragonite CCW particles were first modified with grafting of PLLA chains. Composites of PLTG and modified CCW were then prepared, and the physic-mechanical properties were determined to evaluate their potential for bone repair applications.

2 Materials and methods

2.1 Materials

L-Lactide was purchased from Daigang Biomaterial Co., Ltd and re-crystallized from distilled ethyl acetate five times. Sn(Oct)₂, CaCl₂, Na₂CO₃ and NH₄Cl were obtained from SCRC (China), and used as received. Solvents were of analytical grade and used without further purification.

PLTG terpolymer with LLA/TMC/GA feed molar ratio of 90/5/5 was synthesized as previously reported.²⁵ The number average molar mass (Mn) and polydispersity index (PDI) of PLTG is 150 000 and 2.0, respectively.

2.2 Synthesis of calcium carbonate whisker (CCW)

3.33 g calcium chloride and 3.18 g sodium carbonate were dissolved in 150 mL and 100 mL distilled water at room temperature, respectively. Then the two solutions were added dropwise to a three-necked round bottomed flask under magnetic stirring. The pH of the mixture was maintained at 8–9 with a 0.1 M ammonium chloride buffer solution. The solution was stirred at 120 rpm for 4 h at 80 °C, followed by aging for 1 h. The precipitate was washed three times with distilled water and then washed with ethanol. The obtained CCW particles were vacuum dried at 120 °C for 24 h.

2.3 Grafting of PLLA on the surface of CCW particles

A typical procedure for CCW surface grafting is briefly described as follows:⁴⁴ in an intensively flame dried glass ampoule, 2 g L-lactide was dissolved in 10 mL xylene at 120 °C. 2 g CCW was previously vacuum dried at 120 °C for 24 h and suspended in 20 mL of dried xylene together with 0.002 mL Sn(Oct)₂ in a flask. The suspension was heated to 90 °C, then dropped into the L-lactide solution under nitrogen protection and stirring. The reaction proceeded at 120 °C for 18 h. Then the reaction mixture was cooled down to room temperature. The PLLA-g-CCW particles were separated by centrifugation at 5000 rpm, and washed with excessive chloroform five times to completely remove non-grafted PLLA chains. Finally, the separated sediment was vacuum dried at 50 °C for 24 h to remove residual chloroform. Non-grafted PLLA was precipitated in an excess of ether, and vacuum dried at 45 °C for 24 h to remove residual solvent. The Mn of non-grafted PLLA was 5000 from GPC measurement.

2.4 Preparation of PLTG/PLLA-g-CCW composites

PLTG/PLLA-g-CCW composites were prepared as follows. Pre-weighed PLLA-g-CCW was uniformly suspended in chloroform under magnetic stirring and ultrasonic treatment. Then the suspension was added into a 10% PLTG chloroform solution. The final PLLA-g-CCW content ranges from 1 to 5 wt% in the composite. The mixture was precipitated in an excess of ethanol and the composite was vacuum dried at 45 °C for 24 h to remove residual solvents. For the sake of comparison, PLTG/CCW composites were prepared from PLTG and non modified CCW using the same procedure.

Films were prepared by solution casting for tensile tests. Pure PLTG and various composites were dissolved in dichloromethane at a concentration of 10 w/v%. The solutions were poured onto a quartz plate, and air dried for 48 h. Then the films were vacuum dried at 45 °C for one week up to constant weight. Dumbbell-shaped specimens with dimensions of ASTM D882-02 standard (4 × 75 × 0.3 mm³) were finally cut from the films.

2.5 Characterization

Fourier-transform infrared (FTIR) spectra were registered using Bio-Rad Win-IR spectrometer. Measurements were performed using KBr disks.

Gel permeation chromatography (GPC) measurements were performed on a Shimadzu apparatus equipped with a refractive index detector. Tetrahydrofuran was used as eluent at a flow rate of 1.0 mL min⁻¹. 80 μL of 1.0 w/v% solution were injected for each analysis. Calibration was achieved with polystyrene standards (Polysciences, Warrington, PA).

Differential scanning calorimetry (DSC) was registered with a TA Q2000 instrument. All samples were first heated at 10 °C min⁻¹ to 200 °C to erase the thermal history, followed by a rapid cooling at 50 °C min⁻¹ to room temperature. Finally, a second heating scan was realized at 10 °C min⁻¹ up to 200 °C. The glass transition temperature (T_g) was taken at the midpoint of the heat capacity change from the second run.

Thermogravimetric analysis (TGA) was investigated with a TA Q500 instrument. All the samples were heated from room temperature to 800 °C at a heating rate of 10 °C min⁻¹. Measurements were made under nitrogen atmosphere with a flow rate of 40 mL min⁻¹.

X-ray diffraction (XRD) spectra were registered with a Philips diffractometer composed of a Cu K_α (λ = 1.54 Å) source, a quartz monochromator and a goniometric plate.

Contact angle was determined on solvent casting films using a SL200KS instrument at room temperature. Every measurement was repeated three times.

Scanning electron microscopy (SEM) was performed by using a Hitachi S-4800 microscope operating at 2.0 kV and 10 μA after sputter coating with 5 nm thick gold. Energy-dispersive X-ray spectroscopy (EDX) was performed using Hitachi S4500 microscope.

Tensile tests were carried out at room temperature on a DXLL10000 universal tensile testing machine operating at a crosshead speed of 10 mm min⁻¹, and a grip-to-grip separation of 25 mm. The tensile strength, Young's modulus and strain at break were obtained from the stress–strain curves. All data were obtained from at least five duplicate measurements.

2.6 Cytocompatibility

Extracts of PLTG and PLTG/PLLA-g-CCW2% films were obtained by immersion at 37 °C for 72 h in Dulbecco's Modified Eagle's Medium (DMEM, Hyclone products) with a surface area/volume ratio of 6 cm² mL⁻¹ as recommended by ISO 10993.

L929 mouse fibroblast cells in logarithmic growth phase were harvested and diluted with DMEM (10% calf serum, 100 μg mL⁻¹ penicillin, 100 μg mL⁻¹ streptomycin) to a concentration of 1 × 10⁴ cells per mL, and were seeded in 96-well plates (Corning Costar, USA) with 100 μL per well. The cells were then placed in a CO₂ incubator (NU-4850; NuAire, USA) at 37 °C under humidified atmosphere containing 5% CO₂. After 24 h, the medium was replaced by 50 μL of fresh medium and 50 μL of extract with 10% calf serum. 100 μL of fresh medium were used as the negative control, and 100 μL of 5% phenol solution as the positive control. After 24, 48, and 72 h, 20 μL of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, 5 mg mL⁻¹) were added. The medium was removed after 6 h incubation, and 150 μL of dimethylsulfoxide (DMSO) were added. The plates were then shaken for 10 min. The optical density (OD) was measured by using microplate reader (Elx800; BioTek, USA) at 570 nm. The cell relative growth ratio (RGR) was calculated from the OD value of the test sample and the negative control using the following equation:

RGR (%) = (OD_{test sample}/OD_{negative control}) × 100

The cytotoxicity is generally noted in 0–5 levels according to the RGR value as shown in Table 1.

Table 1 Relationship between the RGR value and cytotoxicity level

RGR (%)	≥100	75–99	50–74	25–49	1–24	0
Level	0	1	2	3	4	5

---

The L929 cell morphology was also examined during cell culture using an Olympus CKX41 inverted microscope.

3 Results and discussion

3.1 Surface modified PLLA-g-CCW particles

Aragonite CCW was obtained by reaction of calcium chloride and sodium carbonate, and surface modified PLLA-g-CCW was prepared by ring opening polymerization of L-lactide in the presence of CCW. Fig. 1 shows the SEM images of as prepared CCW and PLLA-g-CCW particles. CCW appears as hexagonal cylinders with sizes from 1 to 5 μm and lengths from 5 to 50 μm, in agreement with literature.^36–39 After surface modification with grafting of PLLA chains, no apparent changes are observed at the surface of PLLA-g-CCW.

Fig. 1 SEM images of CCW (a) and PLLA-g-CCW (b) particles.

Fig. 2a shows the FT-IR spectra of the raw CCW, PLLA and PLLA-g-CCW. On the spectrum of CCW, the broad peak at 3420 cm⁻¹ belongs to free hydroxyl group vibration in crystal lattice, and the peak at 1790 cm⁻¹ corresponds to the stretching vibration ν(CO₃²⁻, C=O). In addition, the formation of aragonite CCW (CO₃²⁻) framework is evidenced by the asymmetric stretching vibration band located at 1493 cm⁻¹ (ν_as, O–C–O), symmetric stretching vibration band at 1084 cm⁻¹ (ν_s, O–C–O), and out-plane flexural vibration band at 853 cm⁻¹ (γ, O–C–O), in-plane flexural vibration band at 705 cm⁻¹ (β, O–C–O). These bands are similar to those of natural bone aragonite CCW.⁴⁵ After surface grafting, a new absorption band appears at 1127 cm⁻¹ on the spectrum of PLLA-g-CCW belonging to the ether bond stretching vibration of the ester groups. Three small peaks are observed at 2987 cm⁻¹, 2922 cm⁻¹, 2852 cm⁻¹ which are assigned to C–H stretching vibration of the CH₃– and CH– groups of PLLA. Therefore, FTIR data suggest that PLLA was successfully grafted on the surface of CCW.

Fig. 2 FT-IR spectra of (a) CCW, PLLA and PLLA-g-CCW, and (b) PLLA–CCW mixture, LLA–CCW mixture and PLLA-g-CCW.

In order to further evidence the grafting of PLLA on the surface of CCW, CCW particles were treated with LLA/xylene solution and PLLA/xylene solution, respectively. After the same treatment process as for PLLA-g-CCW, no LLA or PLLA remained in the sample, as demonstrated by the FT-IR spectra in Fig. 2b. This means that PLLA was effectively grafted at the surface of CCW in PLLA-g-CCW.

Fig. 3a shows the XRD patterns of CCW prepared by the double decomposition method and surface modified PLLA-g-CCW. The two patterns present exactly the same diffraction peaks. Comparison with the data from the JCPDS file no. 5-453 indicates formation of single-phase aragonite.⁴⁶ Fig. 3b presents the TGA thermograms of CCW and PLLA-g-CCW. The TGA curve of CCW shows a small weight loss of 2% in the 300–450 °C range, and a strong weight loss up to 45% in the 600–750 °C range. Compared with CCW, PLLA-g-CCW displays a similar weight loss profile in two stages. The initial stage ranges from 250 to 450 °C with about 5% weight loss, and a second stage similar to that of CCW. The difference in the weight loss data in the first stage suggests that PLLA-g-CCW contains nearly 3% PLLA chains chemically bonded at the CCW surface.

Fig. 3 (a) XRD patterns, and (b) TGA thermograms of pure CCW and surface modified PLLA-g-CCW.

3.2 Characterization of PLTG/PLLA-g-CCW composites

3.2.1 Mechanical properties. PLLA presents good mechanical strength for bone repair applications. But its uses are limited due to its high rigidity and crystallinity, acidic degradation products and slow degradation rate.¹⁸ PLTG terpolymers with predominant LLA contents exhibit greatly improved toughness as compared to PLLA homopolymer, with only slight loss of tensile strength.²⁵ In the present work, PLTG/PLLA-g-CCW composites were prepared to increase the interfacial affinity between the PLLA-g-CCW filler and the PLTG matrix, and thus to improve the mechanical properties of the resulting composites. PLTG/CCW composites containing non modified CCW were also prepared for the sake of comparison.

The stress–strain curves of pure PLTG, PLTG/CCW2% and PLTG/PLLA-g-CCW2% composites are shown in Fig. 4. It is noticed that the stress–strain behaviors are characteristic of tough materials. All the three materials exhibit a necking phenomenon, and are broken after yielding. Both composites exhibit improved toughness and tensile strength, although PLTG/PLLA-g-CCW2% has a more obvious effect due to the better affinity of PLLA-grafted whiskers to the matrix as compared to non modified CCW.

Fig. 4 Stress–strain curves of PLTG, PLTG/CCW2% and PLTG/PLLA-g-CCW2% composites (2% is the mass fraction of CCW and PLLA-g-CCW).

The mechanical properties of neat PLTG and PLTG/PLLA-g-CCW composites with various PLLA-g-CCW contents were evaluated. Fig. 5 shows the changes of the tensile strength, Young's modulus, and elongation at break as a function of the filler content. Neat PLTG exhibits a tensile strength of 23.8 MPa. The tensile strength of the composites rapidly increases with a filler content of 1 and 2% (Fig. 5a). When the content of PLLA-g-CCW increases to 2%, the tensile strength reaches a maximum of 32.3 MPa. Beyond 2%, the tensile strength of the composites decreases almost linearly with the filler content to reach 25.2 MPa at 5%.

Fig. 5 Effect of the filler content on: (a) the tensile strength, (b) Young's modulus, and (c) elongation at break for neat PLTG and PLTG/PLLA-g-CCW composites.

The Young modulus changes of PLTG and PLTG/PLLA-g-CCW composites are shown in Fig. 5b. From 1.18 GPa for neat PLTG, the Young modulus of the composites continually increases with increasing filler content. A maximum of 1.92 GPa is obtained for the composite with 5% of PLLA-g-CCW loading.

The effect of the filler content on the elongation at break of the PLTG/PLLA-g-CCW composites is illustrated in Fig. 5c. Neat PLTG and PLTG/PLLA-g-CCW composites (1–3%) behave as a tough material since they exhibit the necking phenomenon after yielding and are finally broken. The fracture elongation ratio of PLTG/PLLA-g-CCW2% is 255%, in comparison with that of 212% for PLTG. Beyond 3%, however, the fracture strain of the composites rapidly decreases. At the highest PLLA-g-CCW content of 5%, the elongation at break of the composite is only 29%.

It is well known that the yield of a polymer material usually starts with the movement of polymer chain segments. In the PLTG/PLLA-g-CCW composites, polymer chains would adsorb on the surface of whisker particles owing to their large specific surface area. Thus, the whisker particles dispersed in a polymer matrix could form entanglement crosslinks of polymer chains. The movement of polymer chains should drive the cooperative movement of whisker particles in the composite network. Therefore, the yield of PLTG/PLLA-g-CCW composites needs more stress than that of neat PLTG material.

When a stress is loaded on one of the polymer chains in a composite, it can be evenly distributed to other chains through the whisker particles. Even if some polymer chains in the network are broken up, the remaining chains still could share the loading stress and are continuously oriented along the direction of stress under the traction of whisker particles, thus avoiding fracture of the whole material. Hence, the fracture elongation of the composites increases at low whisker contents. Nevertheless, when the content is above 2%, the excess PLLA-g-CCW whisker particles could result in dense physical crosslinking and short free chain segments in the composite network. Thus polymer chains could hardly move out of the crosslinks. Moreover, excess PLLA-g-CCW whisker particles tend to agglomerate, and to form lacunae in the composite. Therefore, composites with high whisker contents could break up before the yield under tensile stress.

It is of interest to compare our PLTG/PLLA-g-CCW composites with PLLA/PLLA-g-HA ones. The latter were prepared by reinforcing a PLLA matrix with surface grafted HA particles with up to 6% grafting ratio.⁴⁴ Addition of PLLA-g-HA ones allows to slightly increase the tensile strength of the composite. A maximum of 8% increase was obtained with 10% filler content. In the case of PLTG/PLLA-g-CCW composites, 36% increase of tensile strength was obtained with addition of 2% PLLA-g-CCW, indicating a much better reinforcing effect of PLLA-g-CCW as compared to PLLA-g-HA particles.

3.2.2 Morphology of fracture surface. The morphology of the fracture surface of pure PLTG, PLTG/CCW2%, PLTG/PLLA-g-CCW2% composites was examined using SEM, as shown in Fig. 6. The fracture microstructure of pure PLTG revealed a homogeneous phase structure (Fig. 6a and b). In the case of the PLTG/PLLA-g-CCW composite, good adhesion is observed at the interface between the polymer matrix and PLLA-g-CCW particles. The average length of the particles is 20–30 μm. No agglomeration of particles is visible (Fig. 6c). No distinct boundary is observed between the dispersed phase and the matrix (Fig. 6d), which could be attributed to the good affinity of PLLA grafted whiskers to the matrix. In contrast, the PLTG/CCW composite reveals the presence of some holes (Fig. 6e). Agglomeration of CCW whiskers and a gap between the whiskers and the matrix are observed (Fig. 6f). It is thus suggested that PLLA grafted CCW enhances the interface compatibility, thus improving the mechanical properties of PLTG/PLLA-g-CCW as compared to PLTG/CCW composite with non-modified CCW.

Fig. 6 SEM images of the fracture surface of (a) and (b) PLTG, (c) and (d) PLTG/PLLA-g-CCW2%, (e) and (f) PLTG/CCW2%.

Fig. 7 shows the EDX patterns of the fracture surface of PLTG/CCW2% and PLTG/PLLA-g-CCW2%. It appears that the Ca distribution of PLTG/CCW2% is not homogeneous (Fig. 7a). CCW particles seem to concentrate at the lower face of the films. This finding can be attributed to the preparation process. In fact, CCW particles tend to go down toward to bottom in spite of stirring during solvent evaporation. On the other hand, the Ca distribution of PLTG/PLLA-g-CCW2% appears homogeneous due to the affinity of PLLA chains in PLLA-g-CCW particles to the PLTG solution which prevents them from going down (Fig. 7b).

Fig. 7 EDX patterns of the fracture surface of (a) PLTG/CCW2%, and (b) PLTG/PLLA-g-CCW2%.

3.2.3 XRD analysis. Fig. 8a shows the X-ray diffraction patterns of neat PLTG, non-modified CCW and PLTG/CCW2%, PLTG/PLLA-g-CCW2% composites. Neat PLTG exhibits two diffraction peaks at around 2θ = 16.6° and 19.0° which are assigned to the (110)/(200) planes, and the (203) plane of the α form of PLLA crystal, respectively.⁴⁷ The main peaks of non-modified CCW particles are observed at 2θ = 26.2°, 27.2°, 36.2°, 38.4°, 43.0° and 45.9° which result from the (111), (021), (200), (130), (220) and (221) planes of the aragonite phase of CCW. Peaks with a relatively high intensity at 2θ = 36.2°, 38.4° and 43.0° are observed for both composites with preferred-orientation corresponding to the (200), (130) and (220) reflection planes.⁴⁸ This finding well agrees with the SEM images where CCW particles in the PLTG matrix are preferably oriented along one direction and paralleled to the surface (Fig. 6c and e). Compared with the PLTG/CCW composite with non-modified CCW, the peak intensity of PLTG/PLLA-g-CCW composite peak is significantly lower due to the decrease of the aragonite phase structure after PLLA grafting at the CCW surface.

Fig. 8 XRD patterns of (a) neat PLTG, non-modified CCW, PLTG/CCW and PLTG/PLLA-g-CCW composites, and (b) PLTG/PLLA-g-CCW composites with various filler contents.

Fig. 8b shows the X-ray diffraction patterns of PLTG and PLTG/PLLA-g-CCW composites with various filler contents. It appears that all the composites exhibit the diffraction peaks of neat PLTG and preferred-orientation reflection planes of CCW. Moreover, the peak intensity of PLTG diffraction peaks decreases and that of CCW ones increases with increase of the filler content.

3.2.4 Contact angle measurement. Contact angle measurement is a standard method used to determine the hydrophilicity of a biomaterial surface, allowing to evaluate the possible effect of the biomaterial on cell adhesion and growth. Fig. 9 presents the contact angle values of PLTG and various PLTG/PLLA-g-CCW composites determined by static contact angle measurement. Neat PLTG exhibits a contact angle of 80.2°, in agreement with the hydrophobic nature of the polymer. With the increase of the PLLA-g-CCW content, the contact angle gradually decreases from 79.2° for PLTG/PLLA-g-CCW1% to 67.5° for PLTG/PLLA-g-CCW5%. This is because with the increase of PLLA-g-CCW content, there are more and more hydrophilic CCW whiskers and hydroxyl end groups of PLLA short chains. Besides, the contact angle of PLTG/CCW2% is 78.6°, i.e. slightly higher than the value of 77.8° obtained for PLTG/PLLA-g-CCW2%. This difference can be attributed to the good affinity of PLLA-g-CCW toward the PLTG matrix. It is well known that the hydrophobic/hydrophilic balance is an important factor affecting the biocompatibility and platelet adhesion of materials. Good biocompatibility and strong platelet adhesion would favor wound healing applications as hemostatic materials.^49,50

Fig. 9 Contact angle changes of neat PLTG and the composites with various PLLA-g-CCW contents.

3.2.5 Thermal stability of PLTG/PLLA-g-CCW composites. In general, bone scaffold materials should be able to maintain the original mechanical properties within the first 3–6 months of implantation. It is known that the molar mass of biodegradable polymers decreases during processing, leading to deterioration of the mechanical properties. Therefore, it is necessary to evaluate the thermal stability and the degradation mechanism because these polymers are usually melt-processed.

Fig. 10a shows the TGA curves of PLTG and various PLTG/PLLA-g-CCW composites at a heating rate of 10° min⁻¹. The main thermal parameters are summarized in Table 2. It is found that with the increase of the filler content, T_max (maximum weight-loss temperature) gradually increases from 325.8 °C for PLTG to a maximum of 334.1 °C for the composite with 2% filler, and then decreases slightly to 330.1 °C for 5% filler composite. Similar trends are also observed for T_ei (extension initial temperature), T_5% (5% weight-loss temperature), and T_ef (extension terminal temperature) for the composites with maximal values obtained with the 2% composite.

Fig. 10 (a) TGA curves, and (b) ln β/(T_max²) vs. 1000/T_max plots used for the determination of the activation energy according to Kissinger's method for PLTG and various PLTG/PLLA-g-CCW composites.

Table 2 Thermal analysis and activation energy data of PLTG and composites with various PLLA-g-CCW contents

Sample	Tei^a (°C)	T5%^a (°C)	Tmax^a (°C)	Tef^a (°C)	Ea (kJ mol^−1)	Tg^b (°C)	Tm^b (°C)
a The values of Tei, T5%, Tmax, and Tef are obtained from the TGA curves.b The values of Tg and Tm are obtained from the DSC curves at the second heating scan.
PLTG	300.6	299.2	325.8	339.1	98.6	57.4	157.7
1%	318.2	317.4	333.2	345.2	99.9	59.1	162.0
2%	320.1	318.2	334.1	348.7	123.9	59.3	162.1
3%	316.3	298.1	332.9	344.5	120.8	57.9	161.9
4%	315.4	290.5	332.7	343.6	116.3	57.1	161.9
5%	312.2	289.1	330.1	342.7	112.2	55.9	158.2

---

The activation energy characterizes the thermal stability, allowing to understand the thermal decomposition behavior of materials. It is thus important to determine the activation energy of the composites for potential applications as bone scaffold. The Kissinger equation is a widely used differential method for determining the activation energy (E_a) as shown in the following:⁵¹

where β is the heating rate, A is the pre-exponential factor, R is the universal gas constant, α_max is the maximum conversion, and n is the reaction order. E_a can be calculated from the slope of ln β/(T_max²) as a function of 1000/T_max as shown in Fig. 10b. The E_a values are listed in Table 2. With the increase of the PLLA-g-CCW content, E_a gradually increases from 98.6 kJ mol⁻¹ for PLTG to a maximum 123.9 kJ mol⁻¹ for the composite with 2% PLLA-g-CCW content, and then slightly decreases. The values of all the composites are larger than that of pure PLTG. These results indicate that the reinforcing effect depends not only on the PLLA-g-CCW content, but also on the intermolecular hydrogen bonding between PLLA-g-CCW and PLTG.⁵²

3.2.6 DSC measurement. DSC was used to determine the glass transition temperature (T_g) and the melting temperature (T_m) of PLTG and the composites (Table 2). Neat PLTG exhibit T_g and T_m at 57.4 and 157.7 °C, respectively. With the increase of the PLLA-g-CCW content, T_g increases to a maximum value of 59.3 °C, and T_m to a maximum value of 162.1 °C for 2%. Beyond, both T_g and T_m slightly decreases to 55.9 and 158.2 °C for the composite with 5% filler. These findings suggest that the T_g and T_m of composites with low filler contents tend to increase due to homogeneous dispersion of PLLA-g-CCW in the PLTG matrix.

3.3 Cytocompatibility

The cytocompatibility of the materials was investigated in order to evaluate their potential for bone tissue engineering. Cytocompatibility generally includes cell morphology and cell toxicity evaluation on the basis of cell–material interactions. In such tests, cytotoxicity rating of biomaterials is defined according to visual evidence of morphological cellular changes, including cell lysis, rounding, spreading, and proliferation. Quantitative determination of cytotoxicity is realized using MTT assay.⁵³

Fibroblasts are widely distributed in many types of tissues such as tendon, ligament and skin, and are defined as the cells that produce collagens. They are the primary source of most extracellular matrix components, playing a critical role in regulating the turnover of extracellular matrix and in wound healing. Fig. 11 shows the morphology of L929 mouse fibroblast cells during 3 days culture with extracts of PLTG, PLTG/PLLA-g-CCW2%, in comparison with DMEM medium and 5% phenol solution taken as negative and positive controls, respectively. Large number of cells died in the positive control due to the toxicity of phenol medium. And most cells appeared round in shape, especially after 48 and 72 h. On the contrary, cells adhesion and proliferation are observed on PLTG and PLTG/PLLA-g-CCW2% substrates as well as in the negative control. Cells exhibit spindle, polygon or oval shapes, and the pseudopodium of cells stretches out. Little difference of morphology or relative density of cells is detected between PLTG and the negative control. Nevertheless, cell density seems to be slightly lower in the PLTG/PLLA-g-CCW2% medium as compared to PLTG and the negative control.

Fig. 11 Microscopic images of L929 cells after 24, 48 and 72 h culture in different media: PLTG, PLTG/PLLA-g-CCW2%, negative, and positive control.

MTT assay is widely used to evaluate the cytotoxicity in vitro. It is based on the reaction between MTT and mitochondrial succinate dehydrogenases in living cells to form a purple formazan which is not soluble in water but soluble in DMSO. The OD value of formazan–DMSO solution is considered to be proportional to the number of living cells. Table 3 shows the RGR data derived from the OD values of the test and negative control groups, and corresponding cytotoxicity levels. During the 72 h cell culture period, all the RGR values of PLTG are well above 95%, corresponding to cytotoxicity levels of 0 and 1. The RGR values of PLTG/PLLA-g-CCW2% are above 83% during 72 h incubation with L929 cells, corresponding to cytotoxicity levels of 1. Therefore, cell morphology and cell toxicity results indicate that both materials present very low cytotoxicity, although the presence of 2% PLLA-g-CCW in the composite slightly affects its cytocompatibility.

Table 3 RGR values of PLTG and PLTG/PLLA-g-CCW2% during 72 h incubation with L929 cells in comparison with negative and positive control

Sample	24 h		48 h		72 h
	RGR (%)	Level	RGR (%)	Level	RGR (%)	Level
PLTG	101.6 ± 5.46	0	97.7 ± 8.0	1	96.1 ± 12.7	1
PLTG/PLLA-g-CCW	91.7 ± 7.0	1	87.2 ± 7.7	1	83.8 ± 12.0	1
Negative control	100	0	100	0	100	0
Positive control	35.6 ± 13.8	3	45.7 ± 1.0	3	48.2 ± 1.6	3

---

4 Conclusions

In this work, the surface of calcium carbonate whisker is successfully modified by grafting PLLA chains in the absence of catalyst. The grafting is evidenced by FTIR and TGA analyses. The grafting ratio of PLLA is estimated to be ca. 3 wt% from TGA data. The modified PLLA-g-CCW particles are more uniformly dispersed in the PLTG matrix than pure CCW ones because of the improved compatibility between the PLLA-g-CCW filler and the matrix. Consequently, the PLTG/PLLA-g-CCW composites exhibit improved mechanical properties compared to corresponding PLTG/CCW ones. Especially at PLLA-g-CCW content of 2 wt%, the PLTG/PLLA-g-CCW composite exhibits optimal tensile strength and elongation at break, indicating that the PLLA-g-CCW particles have both reinforcing and toughening effects in the composites. L929 cell morphology and toxicity results indicate that neat PLTG and PLTG/PLLA-g-CCW2% materials present very low cytotoxicity, although the presence of 2% PLLA-g-CCW in the composite slightly affects its cytocompatibility. Therefore, PLTG/PLLA-g-CCW composites with improved mechanical properties and good cytocompatibility could be promising as potential bone substitute material.

Acknowledgements

The authors acknowledge the financial support from the National Natural Science Foundation of China (51073041 and 51373041) and the Science and Technology Commission of Shanghai Municipality (124419003103).

References

1. N. Huebsch and D. J. Mooney, Nature, 2009, 462, 426–432.
2. J. Hrkach, D. V. Hoff, M. M. Ali, E. Andrianova, J. Auer and T. Campbell, Sci. Transl. Med., 2012, 4, 128–166.
3. L. J. Bonderer, A. R. Studart and L. J. Gauckler, Science, 2008, 319, 1069–1073.
4. Y. Shikinami and M. Okuno, Biomaterials, 1999, 20, 859–877.
5. S. P. Ju, W. C. Huang and C. C. Chen, RSC Adv., 2014, 4, 35862–35867.
6. N. Goonoo, A. Bhaw-Luximon and D. Jhurry, Polym. Int., 2015, 64, 1289–1302.
7. C. M. Agrawal, K. A. Athanasiou and J. D. Heckman, Mater. Sci. Forum, 1997, 250, 115–128.
8. K. Fukushima, Biomater. Sci., 2016, 4, 9–24.
9. Y. Takeoka, M. Hayashi, N. Sugiyama, M. Yoshizawa-Fujita, M. Aizawa and M. Rikukawa, Polym. J., 2015, 47, 164–170.
10. T. Kokubo, H. Kim and M. Kawashita, Biomaterials, 2003, 24, 2161–2175.
11. R. Orefice, J. West, G. LaTorre, L. Hench and A. Brennan, Biomacromolecules, 2010, 11, 657–665.
12. T. Gong, J. Xie, J. F. Liao, T. Zhang, S. Y. Lin and Y. F. Lin, Bone. Res., 2015, 3, 15029–15035.
13. N. Goonoo, A. Bhaw-Luximon, G. L. Bowlin and D. Jhurry, Polym. Int., 2013, 62, 523–533.
14. Y. Matsusue, S. Hanafusa, T. Yamamuro, Y. Shikinami and Y. Ikada, Clin. Orthop. Relat. Res., 1995, 317, 246–253.
15. A. Joukainen, E. Partio, P. Waris, J. Joukainen, H. Kroger and P. Tormala, J. Orthop. Sci., 2007, 12, 28–34.
16. A. Backstrom, J. Raiha, T. Valimaa and R. Tulamo, Vet. Surg., 2005, 34, 11–17.
17. P. Zhao, D. W. Li, F. Y. Y. Z. Ma, T. T. W. S. Duan, H. S. Q. Cai, D. C. Wu, X. Q. Yang and S. G. Wang, J. Mater. Chem. B, 2015, 3, 6885–6896.
18. A. U. Daniels, M. K. O. Chang, K. P. Andriano and J. Heller, J. Appl. Biomater., 1990, 1, 57–78.
19. A. P. Pego, M. J. A. Van Luyn, L. A. Brouwer, P. B. van Wachem, A. A. Poot and D. W. Grijpma, J. Biomed. Mater. Res., Part A, 2003, 67, 1044–1097.
20. E. Díaz-Celorio, L. Franco, A. Rodríguez-Galán and J. Puiggalí, Polym. Eng. Sci., 2010, 95, 2376–2387.
21. Y. Márquez, L. Franco, P. Turon, A. Rodríguez-Galán and J. Puiggalí, Polym. Eng. Sci., 2013, 98, 2709–2721.
22. K. J. Zhu, R. W. Hendren, K. Jensen and C. G. Pitt, Macromolecules, 1991, 24, 1736–1740.
23. P. Dobrzynski, J. Kasperczyk, H. Janeczek and M. Bero, Macromolecules, 2001, 34, 5090–5098.
24. E. Zini and M. Scandola, Biomacromolecules, 2007, 8, 3661–3667.
25. J. T. Dong, L. Liao, L. Shi, Z. S. Tan, Z. H. Fan and S. M. Li, Polym. Eng. Sci., 2014, 54, 1418–1444.
26. L. Liao, J. T. Dong, L. Shi, Z. Y. Fan and S. M. Li, Polym. Degrad. Stab., 2015, 111, 203–210.
27. Y. Yuan, X. Jin, Z. Fan, S. Li and Z. Lu, J. Mater. Sci.: Mater. Med., 2015, 26, 139,  DOI:10.1007/s10856-015-5384-8.
28. M. J. Olsta, X. Cheng, S. S. Jee, R. Kumar, Y. Y. Kim and M. J. Kaufman, Mater. Sci. Eng., R, 2007, 58, 77–116.
29. C. D. Li, L. Gao, F. P. Chen and C. S. Liu, J. Mater. Sci., 2015, 50, 7182–7191.
30. X. M. Luo, D. Barbieri, R. Q. Duan, H. P. Yuan and J. D. Bruijn, Acta Biomater., 2015, 26, 331–337.
31. W. Yang, E. Fortunati, F. Dominici, J. M. Kenny and D. Puglia, Eur. Polym. J., 2015, 71, 126–139.
32. J. O. Hollinger and G. C. Battistone, Clin. Orthop. Relat. Res., 1986, 207, 290–305.
33. M. Bongio, J. Beucken, S. Leeuwenburgh and J. Jansen, J. Mater. Chem., 2010, 20, 8747–8759.
34. C. T. Wu, Y. H. Zhou, W. Fan, P. P. Han and J. Chang, Biomaterials, 2012, 33, 2076–2085.
35. S. R. Chowdhury, Polym. Int., 2008, 57, 1326–1332.
36. C. Schiller, C. Rasche, M. Wehmoller, F. Beckmann, H. Eufinger and M. Epple, Biomaterials, 2004, 25, 1239–1247.
37. S. Camprasse and G. Camprasse, Clin. Mater., 1990, 5, 235–250.
38. K. Fujihara, M. Kotaki and S. Ramakrishna, Biomaterials, 2005, 26, 4139–4147.
39. J. Abert, A. Amella, S. Weigelt and H. Fischer, J. Mech. Behav. Biomed. Mater., 2016, 54, 82–92.
40. S. Backhaus, T. Annen and M. Epple, Materialwiss. Werkstofftech., 2013, 44, 107–111.
41. Z. N. Cao, D. D. Wang, L. W. Lyu, Y. H. Gong and Y. Li, RSC Adv., 2016, 6, 10641–10649.
42. K. L. Gerlach, Clin. Mater., 1993, 13, 21–28.
43. M. Ara, M. Watanabe and Y. Imai, Biomaterials, 2002, 23, 2479–2483.
44. Z. K. Hong, X. Y. Qiu, J. R. Sun, M. X. Deng, X. S. Chen and X. B. Jing, Polymer, 2004, 45, 6699–6706.
45. J. Chen and L. Xiang, Powder Technol., 2009, 189, 64–69.
46. W. Z. Wang, G. H. Wang, Y. K. Liu, C. L. Zheng and Y. J. Zhan, J. Mater. Chem., 2001, 11, 1752–1754.
47. T. Kawai, N. Rahman, G. Matsuba, K. Nishida, T. Kanaya and M. Nakano, Macromolecules, 2007, 40, 9463–9471.
48. D. Jadsadapattarakul, C. Thanachayanont, J. Nukeaw and T. Sooknoi, Sens. Actuators, B, 2010, 144, 73–80.
49. Z. Chen, S. Cheng, Z. Li, K. Xu and G. Q. Chen, J. Biomater. Sci., Polym. Ed., 2009, 20, 1451–1471.
50. Z. Chen, S. Cheng and K. Xu, Biomaterials, 2009, 30, 2219–2230.
51. S. Y. Yeo, W. L. Tan, A. M. Bakar and J. Ismail, Polym. Degrad. Stab., 2010, 95, 1299–1304.
52. H. Y. Yu, Z. Y. Qin, Y. N. Liu, L. Chen, N. Liu and Z. Zhou, Carbohydr. Polym., 2012, 89, 971–978.
53. S. K. Bhatia and B. A. Yetter, Cell Biol. Toxicol., 2008, 24, 315–319.

---