Effect of particle size on the in vitro bioactivity, hydrophilicity and mechanical properties of bioactive glass-reinforced polycaprolactone composites

ct ol i a zadi Ave di A Accepted 27 June 2011 Available online 1 July 2011 Keywords: Polymer-matrix composites Bioactive glass nanoparticles Bioactivity Mechanical properties posite films containing 5 wt.% bioactive glass (BG) particles of different sizes attractive materials for bone regeneration [11,12]. Therefore, the Materials Science and Engineering C 31 (2011) 1526–1533 Contents lists available at ScienceDirect Materials Science a .e l The scaffold-base tissue engineering approach is a widely used strategy for tissue repair and regeneration. Among various materials, polymeric scaffolds and their composites are commonly used to guide tissues and to operate as mechanical support [1]. Polymer based composites containing bioactive ceramic particles are promising materials for bone regenerative matrixes as these materials provide sufficient mechanical strength accompanied with excellent osteocon- ductivity andbioactivity [2,3]. So far,many systems composed of natural and synthetic biopolymers and inorganic particles (bioactive glasses, hydroxyapatite and tricalcium phosphate) have been designed and reinforcement of PCL by bioactive glass particles improves the mechanical strength while enhancing osteoconductivity. The bioactive glass particles also increase thehydrophilicity of PCLfilms that affect the hydrolysis rate of the polymer in physiological environment, which eventually improves the biodegradation in vivo [13,10]. Recently, there has been a great effort to fabricate nanocomposites for tissue engineering because thenanoscale organization of the ceramic component improves the level of structural integration and enhances the mechanical and biological properties [14]. Biomaterials containing nanoparticles exhibit better cell adhesion and enhanced cellular developed [4–9]. Relatively high mechanical dation rate of polycaprolactone (PCL) provi tissue engineering with long term implanta other hand, excellent osteoconductivity and ⁎ Corresponding authors. Tel.: +98 21 66165207; fax E-mail addresses: [email protected] (R. Bagheri), 0928-4931/$ – see front matter © 2011 Elsevier B.V. A doi:10.1016/j.msec.2011.06.013 degradability, and ability to deliver cells have made bioactive glasses 1. Introduction Hydrophilicity bioactive glass particles were studied by field-emission scanning electron microscopy (FE-SEM), transmission electron microscopy (TEM), dynamic light scattering (DLS), and X-ray diffraction (XRD) methods. In vitro bioactivity of the PCL/BG composite films was evaluated through immersion in the simulated body fluid (SBF). The films were analyzed by FE-SEM, energy dispersive spectrometry (EDS), XRD, and atomic force microscopy (AFM). The mechanical properties of highly-porous PCL/BG composites were examined on cylindrical specimens under quasi-static compression load. It was found that partial crystallization of amorphous BG particles during a prolonged mechanical milling occurred and calcium silicate (CaSiO3) and sodium calcium silicate (Na2CaSiO4) phases were formed. The introduction of submicron BG particles (250 nm) was shown to improve the bioactivity of PCL films. In contrast to BG microparticles, the submicron BG particles were distributed on the film surfaces, providing a high surface exposure to SBF with an improved nanotopography. A notable increase in the stiffness and elastic modulus of the composite was also obtained. As compared to submicron BG particles, lower bioactivity and elastic modulus were acquired for PCL/BG nanoparticles. It was also shown that in spite of high specific surface area of the nanoparticles, partial crystallization during mechanical milling and agglomeration of the nanoparticles during processing decrease the bioactivity, hydrophilicity and mechanical response of the BG-reinforced PCL composites. © 2011 Elsevier B.V. All rights reserved. strength and low degra- de advantages for bone tion period [10]. On the bioactivity, controllable behavior compar partly due to stim natural bone t nanocomposites lished [16]. Nano the polymer mat optical devices [1 enhanced apatit : +98 21 66005717. [email protected] (A. Simchi). ll rights reserved. ommercial 45S5 Bioglass® particles. The characteristics of Received 15 November 2010 Received in revised form 24 April 2011 (6 μm, 250 nm, b100 nm) were prepared by solvent casting methods. The ultra-fine BG particles were prepared by high-energy mechanical milling of c Article history: Polycaprolactone (PCL) com Effect of particle size on the in vitro bioa properties of bioactive glass-reinforced p E. Tamjid a, R. Bagheri b,⁎, M. Vossoughi a,c, A. Simch a Institute for Nanoscience and Nanotechnology (INST), Sharif University of Technology, A b Department of Materials Science and Engineering, Sharif University of Technology, Azadi c Department of Chemical and Petroleum Engineering, Sharif University of Technology, Aza a b s t r a c ta r t i c l e i n f o j ourna l homepage: www ivity, hydrophilicity and mechanical ycaprolactone composites ,b,⁎ Ave. P.O. Box: 11365-9466, 14588 Tehran, Iran . P.O. Box: 11365-9466, 14588 Tehran, Iran ve. P.O. Box: 11365-9466, 14588 Tehran, Iran nd Engineering C sev ie r.com/ locate /msec ed to conventional (micro-size) materials [8]. This is ulation of the nanometer surface roughness found in issue [15]. A review on polymer/bioactive glass for biomedical applications has recently been pub- composites with glass nanoparticles embedded into rix may also be promising for photoinduced nonlinear 7]. Rich et al. [18] and Jaakkola et al. [19] reported an e formation in PCL/PDLLA/bioglass composites by used, it was difficult to obtain a homogeneous distribution through out the films (as will be shown in the next section) if high concentrations (e.g. 20 wt.%) were utilized. In addition, the high bioactivity of BG particles could overshadow the effect of particle size at high concentrations, particularly when nano-scale BG particles are utilized. To prepare films (200–300 μm), the prepared suspensions were cast into glass Petri dishes (d=10 cm) and dried for 24 h. The cast filmswere cut into sliceswith 12 mmdiameter and used for the in vitro bioactivity analysis in SBF solution. In order to examine the effect of particle size on the mechanical response of the composite material, cylindrical specimens with a diameter of 10 mm and height of 14 mm were prepared. The prepared suspensions were cast in cylindrical aluminum molds (10 mm×14 mm) and hold in a refrigerator at ~80 °C for 3 h. The molds were inserted in a Martin–Christ GmbH (Germany) freeze drying and the samples were treated at−54 °C for 24 h. The prepared specimens had 90±1% porosity with pore size in 1527E. Tamjid et al. / Materials Science and Engineering C 31 (2011) 1526–1533 increasing the glass surface/volume ratio in the matrix. In contrast, Chatzistavrou et al. [20,21] showed that the bioactivity of BG particles in SBF is improved by increasing their size from b20 μm to 80 μm. Jo et al. [22] synthesized PCL scaffolds reinforcedwith 20 wt.% bioglass particles (4 μm) and bioglass nanofibers (450 nm). Their results indicated that the nanocomposite exhibits better bioactivity and higher mechanical stability compared to themicrocomposite. Similar resultswere reported byMisra et al. [23] andMohn et al. [24] for P(3HB)/Bioglass composites containing micro-size (b5 μm) and nano-size (35 nm) BG particles. The aim of the present work is to study the effect of bioglass particle size on in vitro bioactivity and mechanical properties of PCL/ BG composites. Although polycaprolactone was almost forgotten for two decades in the biomaterials-arena, there is a recent resurgence of interest on this polymer because it is relatively inexpensive and easy to manufacture andmanipulate PCL into a large range of implants and devices [25]. Due to its FDA approval, the material can be a promising candidate for longer term degradable implants. Since PCL has a low degradation rate, it should be manipulated physically, chemically or biologically to possess tailorable degradation kinetics to suit a specific anatomical site.While the bioactive glass particles are highly bioactive [2,3], the coupling of PCLwith BG particlesmakes it possible to control the degradation and bioactivity rates to fit the natural processes of repair [26]. In this work, bioactive glass particles with different size ranges of nano-scale (b100 nm), submicron (250 nm), and micro- scale (6 μm) were utilized. To produce the glass particles with different sizes, high-energy mechanical milling was performed. Although this process has extensively been used for several ceramic materials [27], its application for bioactive glass particles has not been reported, according to the best knowledge of the author. The bioactivity was examined by immersion in SBF. Since PCL is a hydrophobic polymer, it was also important to study the effect of BG particle size on its hydrophilicity. The mechanical properties were evaluated by quasi-static compression load on highly porous (90%) samples. 2. Experimental procedure 2.1. Preparation of bioactive glass particles Bioactive glass powder with chemical composition: 45SiO2– 24.5Na2O–24.5CaO–6P2O5 (wt.%) was utilized as the starting material. In order to decrease the size of the bioactive glass particles, high-energy mechanical milling in a SPEX 8000 M shaker/miller was performed for various times up to 20 h. Zirconia vial and balls (5 mm) were utilized. The process was performed at two ball-to-powder ratios (BPR) of 3:1 and 10:1. Absolute ethanol (Merck Inc., Germany) was used to prevent extensive agglomeration during the process. The milled powders were analyzed by a field-emission scanning electron microscopy (FE-SEM, LEO 435, Carl Zeiss, Germany), dynamic light scattering (MALVERN ZEN1600,UK), transmission electronmicroscopy (TEM, Philips CM200), and X-ray diffraction method (XRD, SIEMENS D5000, Germany). For XRD analysis, Cu Kα radiation (λ=1.5046 Å), 2θ-step of 0.02°, and scanningduration of 1 swere utilized. Thebioactive glass powderswere classified into three groups: b100 nm, 250 nm, and 6 μm and used for further examination. 2.2. Sample preparation A clear solution of 10 wt.% PCL (Aldrich) in DMC (Merck, Germany) was prepared. The bioactive glass particles were dispersed in DMC, sonicated for 15 min at 500 W, and then mixed up with the PCL solution and stirred for 30 min. The composition of the suspension was set to attain PCL/5 wt.% BG particles. It is important to mention that the amount of bioactive glass particles used in this work is relatively low to induce significant bioactivity and remarkably tailor the degradation rate of PCL. Nevertheless, as nano-scale particles were the range 30–150 μm. Fig. 1 shows that the pores are highly interconnected without having specific orientation. 2.3. In vitro bioactivity In vitro bioactivity studieswere carried outusing standard simulated body fluid (SBF) based on the formulation and method developed by Kokubo et al. [28], which contains inorganic ion concentrations similar to those of human blood plasma. The SBF solution was prepared by dissolving respective amounts of reagent chemicals (all purchased from Sigma, Steinheim, Germany) of NaCl, NaHCO3, KCl, K2HPO4.3H2O, MgCl2.6H2O, CaCl2.2H2O, and Na2SO4 into deionized water. The pH of the SBF solutionwas adjusted to 7.25 byHCl and tris(hydroxyl–methyl) aminomethane used as buffer. The composite films were placed in polyethylene jars (Kartell®, 50 mL) and subsequently 50 mLSBF (37 °C) was added to each jar. During the immersion period, thefilmswere kept at 37 °C in a humidified incubator, and the SBF solution was refreshed every week. The films were collected after 3, 7, 14, 21 and 28 days of incubation. The samples were rinsed in distilledwater three times, then dried using filter paper and stored in desiccators for further examina- tion. The pH value was found to be almost constant through the whole incubation period. The formation of Ca–P phases was examined by FE- SEM, energy-dispersive X-ray spectroscopy (EDS), and XRD. 2.4. Mechanical properties Room temperature compression test was performed on the cylindrical specimens by using a laboratory tensile/compression testing machine (H10KS, Hounsfield Test Equipment Ltd, UK) equipped with a 500 N load cell. The crosshead speedof 1 mm/minwas utilized. To avoid the plastic deformation of the porous structure, no preloadwas applied. The elastic modulus was calculated by the slope of the stress–strain curve before the plateau region. Five samples were examined for each Fig. 1. SEM micrograph showing the pore structure of the specimen prepared for mechanical testing. BG particle size, and the average results were recorded with the standard deviation. 2.5. Surface roughness and contact angle The surface roughnesswasmeasuredusingacommercial (Autoprobe, CP-Research, Veeco Instruments Inc., USA) atomic force microscope (AFM) operated in a non-contact mode. The patterns were scribed by using a 50×50 μm2piezoelectric scannerwhich candigitize the data into 1024×1024 pixels. The AFM tip was a pyramidal Si tip (NT-MDT NSG- 10) with tip radius of about 10 nm and aspect ratio of about 1.2. The speed of the tip was 1 μm/s. Wettability studies were performed with a OCA15plus video- based optical contact angle meter (Dataphysics Instruments GmbH, Filderstadt, Germany). An equal volume of deionizedwater (4 μL) was placed on every film by means of an electronic syringe unit forming a drop. Photos were taken to record the shape of the drops and the images were analyzed using the instrument software. 30 40 50 60 In te ns ity , a .u . 2θ, degree (a) (b) (c) 1 2 2 2 1 1: Calcium silicate2: Sodium calcium silicate Fig. 3. XRD patterns of BG particles: (a) Bioglass®; (b) milled for 10 h at BPR=3; (c) milled for 7 h at BPR=10. 10 100 1000 10000 0 10 20 30 40 50 60 Si ze d ist rib u tio n, % Diameter, nm b a Fig. 4. Volume size distribution of milled BG particles determined by dynamic light 1528 E. Tamjid et al. / Materials Science and Engineering C 31 (2011) 1526–1533 Fig. 2. FE-SEMmicrographs showing the used BG particles: (a) Bioglass®; (b) milled for 10 h at BPR=3; (c) milled for 7 h at BPR=10. 3. Results Fig. 2 shows FE-SEM images of the synthesized bioactive glass particles. The as-received powder is composed of corner-shaped particles with a typical particle size of about 6 μm (Fig. 2a). Fig. 2b shows the particles after mechanical milling for 10 h at BPR=3. It appears that mechanical milling refined the powder particles to submicron size. The formation of short rods and plate-like particles due to the impact of ceramic balls during the milling process is also noticeable. At thehigher ball-to-powder ratio (BPR=10),finer particles were achieved (Fig. 2c). XRD analysis revealed that calcium silicate (CaSiO3) and sodium calcium silicate (Na2CaSiO4) were formed at BPR=10 while the bioactive glass particles were amorphous (Fig. 3). The hydrodynamic size distribution of the milled bioactive glass particles is shown in Fig. 4. The results indicate that the particles have average sizes of 459 nm (for BPR=3) and 162 nm (for BPR=10). The volume size distribution (D90–D10, where Dn is the particle size at n percent of its cumulative curve) is 220 nm and 90 nm for BPR=3 and BPR=10, respectively. TEM images of the particles are shown in Fig. 5. The average size of particles is about 250 nm for BPR=3 (Fig. 5a). The particles are mostly below 100 nm for BPR=10 (Fig. 5b). It is pertinent to point out that the difference between the results of DLS and TEM is partly attributed to the powder agglomeration. It should also be mentioned that, experimental results (not shown here) revealed that continuation of the milling process slightly refine the particle size, however, significant crystallization occurred. Therefore, the powders prepared at BPR=3 and BPR=10 were used for the further experiments. scattering: (a) milled for 10 h at BPR=3; (b) milled for 7 h at BPR=10. Fig. 5. TEM images showing the BG particles milled for (a) 10 h at BPR=3 and (b) 7 h at BPR=10. Fig. 6. FE-SEM micrographs showing the surface of (a) PCL and the PCL-based composite films containing 5 wt.% BG particles with different sizes of (b) 6 μm, (c, d) 250 nm, and (e, f) b100 nm. 1529E. Tamjid et al. / Materials Science and Engineering C 31 (2011) 1526–1533 Fig. 6 shows FE-SEM images of the films before incubation in SBF. Fig. 6a illustrates that the surface of pure PCL film is smooth and dense. In contrast, the surface of PCL/BG microparticles is rough and porous (Fig. 6b). Almost no BGparticles could be seen on the surface, indicating that the bioactive glass particles were embedded throughout the polymer matrix. Similar surface features were observed for the submicron BG particles (Fig. 6c). However, study of the film surface at highermagnifications revealed that the BGparticleswere partly located on thefilm surface (Fig. 6d). Fig. 6e and f show that the BGnanoparticles are decorated on the surface of the polymer film. Agglomeration of the nanoparticles is also observed. Fig. 7 shows high-magnification images of the composite films after exposure to SBF for 28 days to show the formedprecipitates. Toexamine the phase formation, EDS was utilized. The results are summarized in Table 1. In agreement with previous report [29], hydroxyapatite (HA, Ca10(PO4)6(OH)2, Ca/P=1.67) particles are likely to form on the surface of composite film containing BG microparticles. As seen in Fig. 7a, HA particles have a granular and lamellar structure. HA particles were also detected on the PCL/submicron BG composite (Fig. 7c). In contrast to thesefilms, the ratio of Ca/P for thefilmcontainingBGnanoparticleswas 1.33 (Table 1), indicating the presence of octacalcium phosphate (OCP, Ca8H2(PO4)6·5H2O) that is a metastable precursor of hydroxyapatite. The OCP crystals have plate-shape morphology. This finding is in agreement with the previous investigation which confirmed the formation of non-stoichiometric HA in PDLLA/TiO2 nanocomposites upon immersion in SBF [7]. The XRD patterns of the composite films after immersion in SBF for 28 days are shown in Fig. 8. The formation of Ca–P phases is visible. The formation of OCP at the early stage is energetically and kinetically favored because of large activation energy of HA nucleation as described by reaction [30]: 8Ca2+ + 6HPO2−4 + 5H2O → Ca8H2 PO4ð Þ6 ⋅ 5H2O + 4Hþ ð1Þ e surface of PCL/BG composite films after exposure to SBF for 28 days: (a, b) 6 μm BG; (c, d) Table 1 Phase formation on the surface of PCL/BG composite films dependent on the BG particle size after 28 days immersion in SBF as determined by EDX analysis. Average BG particle size Ca P Ca/P Phase 6 μm 62.65 37.35 1.67 HA 250 nm 62.5 37.5 1.67 HA 100 nm 57.59 42.41 1.35 OCP 1530 E. Tamjid et al. / Materials Science and Engineering C 31 (2011) 1526–1533 Fig. 7. FE-SEMmicrographs and EDX results revealing the formation of precipitates on th 250 nm BG; (e, f) b100 nm BG. 1531E. Tamjid et al. / Materials Science and Engineering C 31 (2011) 1526–1533 With prolonged exposure to SBF solution, theOCP crystals transform to apatite by hydrolysis followed by partial dissolution and re- precipitation of HA in Ca2+ containing solution according to the below reaction [31]: Ca8H2 PO4ð Þ6 OHð Þ2 ⋅ 5H2O + Ca2+ → Ca10 PO4ð Þ6 OHð Þ2 + 3H2O + 4Hþ: ð2Þ Lei et al. [29] showed that this transformation occurs by gradual increase in Ca/P from 1.3 (day 17) to 1.5 (day 21) and finally to 1.67 Fig. 8. XRD patterns (background subtracted) of composite films containing BG particles wit characteristic peaks of HA, TCP and OCP are shown in (d) 1: HA; 2: TCP; 3: OCP; 4: CS. 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 100 200 300 400 St re ss , k Pa Strain 1 2 3 4 1: PCL 2: PCL/BG (6 μm) 3: PCL/BG (250 nm) 4: PCLBG ( found to be significant (Table 2). Therefore, more attention should be 1532 E. Tamjid et al. / Materials Science and Engineering C 31 (2011) 1526–1533 observed behavior could be attributed to three main effects including: (1) the effect of total surface area; (2) distribution of BG particles throughout the polymeric matrix; (3) phase constitution of the bioactive glass particles. It is suggestible that homogeneous distribu- tion of amorphous BG particles with high surface/volume ratio on the surface of PCL films induces a high exposure surface to SBF liquid, yielding to a high bioactivity response. The BG microparticles with relatively low specific surface area of b1 m2/g were embedded in the polymermatrix with a low surface exposure to SBF liquid (see Fig. 6b). The milled BG powder with an average size of 250 nm and specific surface area N8 m2/g was partially distributed on the surface of the polymeric film and provided a higher surface to volume ratio of the immersion liquid (see Fig. 6d). Additionally, the submicron BG particles induced nanotopography on the surface of the composite film by exposing directly the bioactive glass particles on the surface, which was not the case for the BG microparticles. AFM study (results are not shown here for brevity) showed a higher RMS value for the composite film containing the submicron BG particles compared to the film containing BG microparticle (see Table 2). The PCL film is hydrophobic and BGmicroparticles improve its wettability. Therefore, BG particles contribute to altering the degradation rate of PCL as evident in the surface roughness. On the other hand, FE-SEM images (Fig. 2) and XRD studies (Fig. 3) revealed that rods and plate-like particles were formed during high-energy mechanical milling. It seems that the collisions of balls provided sufficient energy for partial crystallization of the glass particles. It is well known that during mechanical milling, the collisions of balls exert significant energy to the system (about 30 kJ/mole deviation from equilibrium [27]) that can result in nanostructuring, amorphization, extended solid solution, reaction, and phase transformation. Thermal crystallization of amorphous BG particles has been reported by several authors, for example [32]. In the present work, we have shown solid-state room temperature crystallization of amorphous BG particles during high- energymechanical milling for the first time. It seems that the required energy for crystallization has been provided during the process. Although it is very difficult to study the micro-mechanisms involved in this dynamic process, temperature rise during high-energy mechanical milling for a prolonged time without affording cooling system, e.g. water circulation, could be the main reason of crystalli- zation. Therefore, partial crystallization of amorphous BG during high- energy mechanical milling contributed to the bioactivity of composite films because BG particles are highly bioactive in amorphous state. The results of compression test also showed that the elastic modulus and stiffness of PCL/BG composite were improved by decreasing the size of BG particles from micro-scale to submicron-scale (Fig. 9). This is mainly attributed to the higher interfacial surface area provided by the submicron particles, which in fact enhance the load transfer between the matrix and the stiff glass particles [33]. It is pertinent to point out that the interfacial bonding between the glass particles and polymer matrix significantly influences the mechanical response of the composite. For instance, Xie et al. [34] have shown that surface modification of glass beads highly improve the interfacial adhesion to poly(phenylene oxide) and affecting the mechanical properties of the composite. Therefore, partially crystallized bioactive glass particles could have different interfacial bonding strength with PCL as compared with the fully amorphous particles, which would affect the strength of the composite. XRD peak analysis by X'Pert HighScore software (PANalytical, Netherlands) also determined that the inten- sity ratio of the strongest peak of TCP (2θ=30.71°) to HA (2θ=31.76°) for the PCL/submicron BG composite is ~25% lower than that of PCL/BG micoparticles. Since the intensity of X-ray diffraction from any system of phases depends on the diffracting volume of each phase [35], it can be deduced that the submicron BG particles yield higher bioactivity compared to BG microparticles. Meanwhile, inferior bioactivity was obtained for the BG nanoparticles compared to the submicron BG particles in spite of their higher paved to decrease particle agglomeration during processing. 5. Conclusions Wehave studied the effect of BG particle size on in vitro bioactivity and mechanical properties of PCL/bioactive glass composites. The findings are summarized as follows. • High-energy mechanical milling of commercial Bioglass® 45S5 powder yields submicron and nanometric particles dependent on the BPR ratio. Solid-state room temperature crystallization of the amorphous phase and the formation of calcium silicate also occur. The amount of crystallization increases with increasing the milling time and intensifying the milling energy. • In contrast to BG microparticles, the milled BG particles are mostly distributed on the surface of PCL film and consequently providemore surface exposure to the SBF solution. • The submicron BG nanoparticles improve the nanotopography of the film surface and the composite bioactivity. The short rods and plate-like CS particles also contribute to the bioactivity. The elasticmodulus of the highly-porous composite (90% porosity) containing submicron BG particles is significantly higher than those of PCL/BG microparticles. • The introduction of submicron BG and BG microparticles improve the hydrophilicity of PCL. Consequently, in vitro bioactivity and degrada- tion rate of the polymer are enhanced. The composite film containing BG nanoparticles is more hydrophobic due to the presence of CS particles. • The bioactivity of PCL/BG nanoparticles is inferior to that of PCL/ submicron BG composite as a result of nanoparticle agglomeration. Lower elastic modulus is also obtained. References [1] Q. Yang, L. Chen, X. Shen, Z. Tan, Journal of Macromolecular Science 41 (2006) 1171–1181. [2] K. Rezwan, Q.Z. Chen, J.J. Blaker, A.R. Boccaccini, Biomaterials 27 (2006) 3413–3431. [3] V. Guarino, F. Causa, L. Ambrosio, Expert Review of Medical Devices 4 (2007) 405–418. [4] L.S. Nair, C.T. Laurencin, Progress in Polymer Science 32 (2007) 762–798. [5] P.A. Gunatillake, R. Adhikari, European Cells & Materials 5 (2003) 1–16. [6] S.-H. Rhee, Biomaterials 25 (2004) 1167–1175. [7] F.G. Torres, S.N. Nazhat, S.H. Shekh, M.D. Fadzollah, V. Maquet, A.R. Boccaccini, Composites Science and Technology 67 (2007) 1139–1147. [8] J. Wei, Q.Z. Chen, M.M. Stevens, J.A. Roether, A.R. Boccaccini, Materials Science and Engineering C 28 (2008) 1–10. [9] G. Chouzouri, M. Xanthos, Acta Biomaterialia 3 (2003) 745–756. [10] H. Sun, L. Mei, C. Song, X. Sui, P. Wang, Biomaterials 27 (2006) 1735–1740. [11] Q.Z. Chen, I.D. Thompson, A.R. Boccaccini, Biomaterials 27 (2006) 2414–2425. [12] G.E. Vargas, R.V. Mesones, O. Bretcanu, J.M.P. Lopez, A.R. Boccaccini, A. Gorustovich, Acta Biomaterialia 5 (2009) 374–380. [13] X. Teng, J. Ren, S. Gu, Journal of Biomedical Materials Research. Part B, Applied specific surface area (N30 m2/g) and nanotopography of the film surface (Fig. 6 and Table 1). The XRD peak analysis revealed that TCP/ HA intensity ratio is ~10% higher than that of the submicron BG particles while it is ~16% lower than that of the BG microparticles. Results showed that partial crystallization of the amorphous BG particles occurred during mechanical milling. Since silicates are hydrophobic [36], the hyrophobicity of the composite film containing BG/CS nanoparticles was higher than those containing submicron and micro-scale BG particles (Table 2). It is suggestible that agglomeration of the nanoparticles and higher hydrophobicity of the film provided lower effective surface area to the immersion liquid. Again, the bioactivity of the glass phase was reduced due to partial crystalliza- tion. Agglomeration of particles during mechanical milling is likely to occur [25] and the agglomerates are difficult to be disintegrated during the composite preparation (see Fig. 6f). The effect of particle agglomeration on the elastic modulus of PCL/BG nanoparticles was Biomaterials 81 (2007) 185–193. [14] A. Simchi, E. Tamjid, F. Pishbin, A.R. Boccaccini, Nanomedicine: Nanotechnology, Biology and Medicine 7 (2011) 22–39. [15] L.G. Gutwein, T.J. Webster, Journal of Nanoparticle Research 4 (4) (2002) 231–238. [16] A.R. Boccaccini, M. Erol, W.J. Stark, D. Mohn, Z. Hong, J.F. João, F. Mano, Composites Science and Technology 70 (2010) 1764–1776. [17] A. Majchrowski, J. Ebothe, E. Gondek, K. Ozga, I.V. Kityk, A.H. Reshak, T. Lukasiewicz, Journal Alloys and Compounds 485 (2009) 29–32. [18] J. Rich, T. Jaakkola, T. Tirri, T. Narhi, A. Yli-Urpo, J. Seppala, Biomaterials 23 (2002) 2143–2150. [19] T. Jaakkola, J. Rich, T. Tirri, T. Narhi, M. Jokinen, J. Seppala, A. Yli-Urpo, Biomaterials 25 (2004) 575–581. [20] X. Chatzistavrou, T. Zorba, K. Chrissafis, E. Kaimakamis, E. Kontanasaki, P. Koidis, K.M. Paraskevopoulos, Journal of Thermal Analysis and Calorimetry 85 (2006) 253–259. [21] X. Chatzistavrou, K. Chrissafis, E. Polychroniadis, E. Kontonasaki, P. Koidis, K.M. Paraskevopoulos, Journal of Thermal Analysis and Calorimetry 86 (2006) 255–259. [22] J.-H. Jo, E.-J. Lee, D.S. Shin, H.-E. Kim, H.-W. Kim, Y.-H. Koh, J.-H. Jang, Journal of Biomedical Materials Research Part B: Applied Biomaterials 91 (2009) 213–220. [23] S.K. Misra, D. Mohn, T.J. Brunner,W.J. Stark, S.E. Ipsita Roy, V. Salih, J.C. Knowles, A.R. Boccaccini, Biomaterials 29 (2008) 1750–1761. [24] D. Mohn, S.K. Misra, T.J. Brunner, A.R. Boccaccini, W.J. Stark, European Cells & Materials 16 (2008) 8. [25] M.A.Woodruff, D.W.Hutmacher, Progress in Polymer Science 35 (2010) 1217–1256. [26] V. Cannillo, F. Chiellini, P. Fabbri, A. Sola, Composite Structures 92 (2010) 1823–1832. [27] C. Suryanarayana, Progress in Materials Science 46 (2001) 1–184. [28] T. Kokubo, H. Takadama, Biomaterials 27 (2006) 2907–2915. [29] Y. Lei, B. Rai, K.H. Ho, S.H. Teoh, Materials Science and Engineering C 27 (2000) 293–298. [30] R. Horváthová, L. Müller, A. Helebrant, P. Greil, F.A. Müller, Materials Science and Engineering C 28 (2008) 1414–1419. [31] G.H. Nancollas, in: S. Mann, J. Webb, R.J.P. Williams (Eds.), Biomineralization: Chemical and Biochemical Perspectives, VCH Verlagsgesellschaft, Weinheim, 1989, p. 159. [32] L. Lefebvre, J. Chevalier, L. Gremillard, R. Zenati, G. Thollet, D. Bernache-Assolant, A. Govin, Acta Materialia 55 (2007) 3305–3313. [33] J. Cho, M.S. Joshi, C.T. Sun, Composites Science and Technology 66 (2006) 1941–1952. [34] X.-L. Xie, C.-Y. Tang, X.-P. Zhou, R.K.-Y. Li, Z.-Z. Yu, Q.-X. Zhang, Y.-W. Mai, Chemistry of Materials 16 (2004) 133–138. [35] B.E. Warren, X-ray Diffraction, Addison-Wesley, Reading, PA, 1968, p. 49. [36] A. Krysztafkiewicz, Journal of Materials Science 22 (1987) 478–482. 1533E. Tamjid et al. / Materials Science and Engineering C 31 (2011) 1526–1533 Effect of particle size on the in vitro bioactivity, hydrophilicity and mechanical properties of bioactive glass-reinforced polycaprolactone composites 1. Introduction 2. Experimental procedure 2.1. Preparation of bioactive glass particles 2.2. Sample preparation 2.3. In vitro bioactivity 2.4. Mechanical properties 2.5. Surface roughness and contact angle 3. Results 4. Discussion 5. Conclusions References