In vitro bioactivity of glass-ceramic / fibroin composites

Bioactive composite materials were prepared by mixing 20 wt.% of silk fibroin (SF) and 80 wt.% of glassceramics from CaO-SiO2-P2O5-MgO system. In vitro bioactivity of the prepared composites was evaluated in 1.5 simulated body fluid (1.5 SBF) in static conditions. The obtained samples before and after in vitro tests were characterized by X-ray diffraction (XRD) analysis, Fourier transform infrared spectroscopy (FTIR), and X-ray photoelectron spectroscopy (XPS). The changes in 1.5 SBF solutions after soaking the samples were evaluated by inductively coupled plasma atomic emission spectroscopy (ICP-AES). MG63 osteosarcoma cells were used for the biological experiments. The obtained experimental data proved that the synthesized composites exhibit excellent in vitro bioactivity.


I. Introduction
It is well known that natural bone is a typical inorganic/organic composite containing approximately 80% hydroxyapatite (HA) and 20% collagen matrix [1].In this context, HA has attracted much attention for large bone defects regeneration [2,3] due to its excellent osteoconductivity and similarity to the mineral phase of bone.Nevertheless, its low compressive strength limits its application to non low-load bearing bone repairs [4].
Among them SF is of a practical interest due to its excellent intrinsic properties utilizable in the biotechnological and biomedical fields.On the other hand, it is known that SF is composed of 17 aminoacids.It has been proved that SF is a good biocompatible material, which has been successfully used for various medical applications.Kong et at.[11] demonstrated that silk fibres could induce HA nucleation on the surfaces proteins in simulated body fluid (SBF) solution at pH = 8.The authors also proved that there are strong chemical bonds between SF and HA.For this reason, a coprecipitation method was often used to synthesize silk powder/HA nanocomposites [11,[17][18][19][20][21][22][23].For instance, Wang et al. [18] deposited HA on fibroin microspheres through the co-precipitation method.Furuzono et al. [24] developed a HA/F composite, where fibroin was modified with organic compound.In another article Kino et al. [25] reported that the regenerated fibroin films containing Ca 2+ ions were coated with HA, after soaking in 1.5 SBF for 6 h.Li et al. [26] proved that fibroin can induce HA deposition at 37 °C in 1.5 SBF.The effect of pH and initial Ca 2+ −H 2 PO 4 -concentration on the crystal growth of HA in the presence of SF is studied by Ren et al. [21].They evaluate the effect of pH and initial Ca 2+ and PO 4 3-concentration on the products of SF mineralization.They also demonstrated that pH = 7 promote the transition of dicalcium phosphate dihydrate (DCPD) to HA.On the other hand, Kong et al. [27] observed that SF can accelerate the phase trans-mittance of amorphous calcium phosphate (ACP) to the crystal hybrid between DCPD and HA.Li et al. [28] synthesized SF/calcium phosphate composite by adding the different amount of Na 2 SiO 3 to investigate the effect of silicon on the HA formation in the composite.Based on the obtained results they concluded that the addition of silicon could accelerate the HA formation.SF/wollastonite scaffolds were prepared by Zhu et al. [29] via freeze-drying method.They showed that the composite scaffold was in vitro bioactive, because wollastonite induced the formation of carbonate containing hydroxyapatite (CO 3 HA) on the surface of the samples, after soaking in SBF solution for 5 days [29].Four types of hybrid films were fabricated by Vachiraroj et al. [30].These materials include gelatine conjugated SF, HA/SF, HA/gelatin/SF, and HA/chitosan.The prepared films were investigated in terms of in vitro adhesion, proliferation and osteogenic differentiation with preosteoblast cell lines (MC3T3-E1) and rat bone marrow derived stem cells (MC3).The authors proved that MC3T3-E1 showed higher proliferation rate [30].
In our previous work [31], we have prepared some in vitro bioactive composites between glass-ceramics from CaO-SiO 2 -P 2 O 5 system and SF solution and evaluated apatite formation on their surface, after soaking in SBF solution.The purpose of the present article is to prepare novel glass-ceramics/SF bioactive composites, based on two types of glass-ceramics (from CaO-SiO 2 -P 2 O 5 -MgO system, analysed in our previous paper [32]), and to evaluate their in vitro bioactivity in 1.5 SBF solution after 14 days of soaking in static conditions.The novelty of the presented work is that the glass-ceramics were previously soaked in TRIS-HCl solution for 14 days, and in this form used for preparation of the bioactive composites.In this context, we try to discuss the main question -how the phase composition of the synthesized samples influences the structure and in vitro bioactivity?

Preparation of glass-ceramics/SF composites
Bioactive glass-ceramic/SF composites were prepared by mixing 20 wt.% of silk fibroin (SF) and 80 wt.% of glass-ceramics from CaO-SiO 2 -P 2 O 5 -MgO system.Silk fibroin (SF) solution was prepared by dissolving fibroin without sericin in 9 M LiBr at 60 °C for 4 h and then dialysed against distilled water at room temperature for 3 days using cellulose membrane (MWCO 12000 Da) for 72 h to remove the salts [30].The final concentration of the SF solution was 3 wt.%, in accordance with literature data [29].
After that two glass-ceramic powders from CaO-SiO 2 -P 2 O 5 -MgO system (denoted as C1 and C2), as inorganic part of the composites, were synthesized via multi step sol-gel method.The powders have different compositions, i.e.C1 contains 38 wt.% CaO, 29 wt.%SiO 2 , 31 wt.%P 2 O 5 and 2 wt.%MgO, whereas C2 con-tains 55 wt.% CaO, 20 wt.% SiO 2 , 22 wt.%P 2 O 5 and 3 wt.%MgO.In the first step SiO 2 sol was obtained from tetraethoxysilane (TEOS) by mixing with solvent (mixture of C 2 H 5 OH and H 2 O) and a catalyst (a small amount of HCl) in a volume ratio TEOS : C 2 H 5 OH : H 2 O : HCl = 1 : 1 : 1 : 0.01.After approximately 1 h and the formation of transparent solution, the magnesium salt (Mg(NO 3 ) 2 ×6 H 2 O) was dissolved in water and added to the prehydrolysed TEOS under stirring for 14 h.In the second step calcium phosphate (CP) solution was prepared by mixing Ca(OH) 2 and H 3 PO 4 at pH = 10-11.The calcium phosphate mixture was added dropwise into the modified silica sol under intensive stirring.The prepared mixed sol was stirred for 24 h, gelled at 120 °C for 12 h and thermally treated at 1200 °C for 2 h in a tubular furnace.The structure evolution of the thermally treated samples was presented in our previous paper [32].The obtained glass-ceramics was soaked in TRIS-HCl buffer for 14 days [32].
Organic/inorganic composites were prepared by mixing of the treated glass-ceramics and SF solution without binding agents.A certain amount of C1 or C2 glassceramic powder was added into the SF solution (using weight ratio C : SF = 80 : 20) under stirring for 4 h in order to disperse the glass-ceramic particles uniformly.The obtained mixture was poured into plastic cup.The mixture was frozen at −60 °C and then lyophilized.Finally, two samples were obtained and denoted as C1F and C2F.

In vitro bioactivity of the prepared composites
In vitro bioactivity of C1F and C2F samples was evaluated by examining the apatite formation on their surfaces in 1.5 SBF solutions.1.5 SBF solution was prepared by using the following reagents: NaCl = 11.9925g, NaHCO 3 = 0.5295 g, KCl = 0.3360 g, K 2 HPO 4 ×3 H 2 O = 0.3420 g, MgCl 2 ×6 H 2 O = 0.4575 g, CaCl 2 ×2 H 2 O = 0.5520 g, Na 2 SO 4 = 0.1065 g and buffering at pH = 7.25 at 36.5 °C with TRIS = 9.0075 g and 1 M HCl in distilled water.The synthesized composites were immersed in 1.5 SBF solution at human body temperature (36.6 °C) in polyethylene bottles in static conditions for 14 days.After soaking they were removed from the fluid, gently rinsed with distilled water, and dried at 37 °C for 12 h.
The structure and in vitro bioactivity of the synthesized C1F and C2F composites were studied by FTIR, XRD, ICP-AES and XPS.Powder X-ray diffraction spectra were collected within the 2θ range from 10°to 80°with a constant step of 0.04°and counting time of 1 s/step on a Bruker D8 Advance diffractometer with CuKα radiation and SolX detector.The spectra were evaluated with the Diffracplus EVA package.FTIR transmission spectra were recorded by using a Brucker Tensor 27 spectrometer with scanner velocity 10 kHz.Diluted pellets of KBr were prepared by mixing of ∼1 mg of the samples with 300 mg KBr.Transmission spectra were recorded using MCT detector with 64 scans and 1 cm -1 resolution.The ionic concentrations of Ca, P, Si, and Mg of these solutions were determined by inductively coupled plasma atomic emission spectroscopy (ICP-AES), Iris 1000, Thermo Elemental, USA.X-ray photoelectron spectroscopy (XPS) of the samples after in vitro test in 1.5 SBF solution for 14 days was carried out using ESCALAB MkII (VG Scientific) electron spectrometer at a base pressure in the analysis chamber of 5×10 -10 mbar (during the measurement 1×10 -8 mbar), using AlK X-ray source (excitation energy hν = 1486.6eV).The pass energy of the hemispherical analyser was 20 eV, 6 mm slit widths (entrance/exit).The instrumental resolution measured as the full width at half maximum of the Ag3d5/2, photoelectron peak is 1 eV [33].The energy scale is corrected to the C1s -peak maximum at 285.0 eV for electrostatic charging.The processing of the measured spectra includes a subtraction of X-ray satellites and Shirley-type background [34].The peak positions and areas are evaluated by a symmetrical Gaussian-Lorentzian curve fitting.The relative concentrations of the different chemical species are determined based on normalization of the peak areas to their photoionization cross-sections, calculated by Scofield [34].

Cell culture and treatment
Human osteosarcoma cells MG63 cells (MG63, CRL-1427) were used for these experiments.MG63 have been well characterized, and widely used for testing biomaterials.In order to limit the differences in phenotypic expression, the cells were used at a passage 10 to 15.The cells were maintained in Eagle's Minimum Essential Medium (MEM, Lonza, Switzerland) containing 10% heat-inactivated fetal bovine serum (FBS, Lonza, Switzerland), 100 units/ml penicillin, and 100 µg/ml streptomycin in a humidified CO 2 atmosphere at 37 °C.They were routinely checked for mycoplasma contamination by 4 ′ ,6-Diamidin-2-phenylindol staining (DAPI, Roche Diagnostics, Mannheim, Germany) and were found free of it.The microscopy was performed with a Carl Zeiss AM240 microscope equipped with Zeiss AxioCamM-Rmand Andor (iXon+) cameras.
In all experiments the initial cell number was 1×10 5 cells/ml.C1F and C2F samples were added in concentrations up to 10 µg/cm 2 .It was not possible to guarantee that the particles were evenly distributed in the well plates, the experiments were conducted in triplicate to minimize any gross errors caused by uneven distribution of the particles.
The cell growth and the cytotoxicity of C1F and C2F composites were determined by MTT assays [35].300 µl MTT solution (5 mg/ml) was added to each sample, where MG63 cells were growing for up to 72 h treated with different C1F and C2F concentrations.They were further incubated for 3 h at 37 °C in order to dissolve the formazan product of the MTT reduction by the cellular enzymes, the cell media was removed and 300 µl/well of 100% anhydrous isopropanol was added.Following the complete extraction of formazan the optical density of the obtained solutions was measured at 550/630 nm with a DTX880 spectrophotometer (Beckman Coulter, USA).The results were used to calculate the cell growth by the OriginLab program.

Detection and quantification of mineralization
The detection and quantification of the mineralization were performed as described by Gregory et al. [36] with some modifications.Monolayers of MG63 were plated in 48-well plates in osteoprogenitor media (Lonza) up to 21 days.Following different periods of treatment with 10 µg/cm 2 of C1F or C2F, the cells were washed with phosphate-buffered solution (PBS) and fixed in 10% (v/v) formaldehyde (Sigma-Aldrich) for 15 min on ice.The monolayers were washed twice with dH 2 O prior to addition of 0.5 ml of 40 mM Alizarin Red S (ARS) (pH 4.1 adjusted with ammonium hydroxide) per well.The plates were further incubated for 20 minutes at room temperature with gentle shaking.After complete aspiration of the unincorporated dye, the wells were washed five times with dH 2 O. Stained monolayers were visualized by phase microscopy.
Intracellular ALP activity was analysed by a commercial colorimetric ALP activity detection kit (Abcam, ab83369) following the manufacturer's instructions.ALP activity was further normalized by mg intracellular protein content.
The data were evaluated by analysis of variance (ANOVA) followed by Bonferroni's post-hock test.Differences in the results at the level of p < 0.05 were considered statistically significant.The statistical analysis was carried out using the PASW 18.0 statistical software package (IBM) for Windows.
Based on our preliminary results on the two glassceramics from CaO-SiO 2 -P 2 O 5 -MgO system, we can conclude that HA and hydromagnezite were formed on the surface of the soaked samples in TRIS-HCl buffer [32].If we compare XRD results for the C1 and C2 glass-ceramics, soaked in TRIS-HCl buffer, published in [32], with these in the present study (Figs. 1 and 2), it will be seen that HA in the C1F and C2F composites has low crystallinity.This fact could be related with the process of partial dissolution of HA under experimental conditions [37,38].
Structural changes for both parts (SF and glassceramics) of the synthesized composites were investigated by FTIR (Fig. 3).First of all, the FTIR spectra of the obtained composites, presented in Fig. 3, show very complicated character.Silk fibroin (SF), as the organic part of the two composites, displayed strong adsorption peaks, posited at 1620 cm -1 and 1538 cm -1 (for the C1F sample) and 1622 cm -1 and 1541 cm -1 (for the C2F sample).The observed peaks are characteristic for β-sheet conformation of fibroin [15,39].The peak at 1286 cm -1 (for the C2F sample) is a characteristic peak of amide III [14].Surprisingly, the bands for amide III disappeared in the C1F sample, as it can be seen from Fig. 3 (curve a).Xe and et al. [14] also observed that the bands, which are related to amide II and amide III, disappeared in the synthesized SF/HA composites.There is another explanation for the disappearance of amide III in our C1F sample.It is based on the research of chitosan/SF blends, studied by Ramya et al. [40].The authors have synthesized chitosan/SF blends by the mix-ing of chitosan and SF in a ratio 1 : 1 and they concluded that the disappearance of the band at 1628 cm -1 (amide I) could be related to the formation of intramolecular bond between chitosan and SF [40].In our case, the disappearance of amide III for the C1F sample could be explained by the formation of strong hydrogen bonds between OH groups of the Ca-deficient HA which is formed on the C1 glass-ceramic surface after soaking in TRIS-HCl buffer for 14 days in static conditions [32].On the other hand, in the FTIR spectrum of the C1F composite (Fig. 3, curve a), the slightly visible peak at 782 cm -1 and those at 560, 594, 606 and 853 cm -1 could be related to cristobalite (C), whitlockite (W) and hydromagnezite (HM) in accordance with our preliminary results [32].Furthermore, the bands posited at 654, 853, 960, 1009, 1024 and 1111 cm -1 can be assigned to the presence of HA in the C1F composite [41].In the FTIR spectrum of the C2F composite obtained (Fig. 3, curve b) we can observe some absorption bands at 520, 543, 606, and 840 cm -1 , which are indicative for crystalline HA structures.In addition, the peaks at 868, 939 and 1060 cm -1 can be assigned to the silicocarnotite (S) formed in the studied composite [41,42].The absorption maxima, visible at ∼720 cm -1 could be related to the presence of HM and amide V. From the FTIR presented results, we can conclude that our data are in accordance with our previous XRD results [32].

Samples after in vitro test in 1.5 SBF solution
XRD results obtained for the composites immersed in 1.5 SBF solutions for 14 days in static conditions are presented in Figs. 4 and 5 (and summarized in Table 1).
From the presented data it can be seen that XRD patterns are quite different.In the case of the C1F composite (Fig. 4) the obtained HA (PDF card 01-072-7532) has a high crystallinity.On the other hand, the crystallinity of W and HM was slightly decreased after immersion in 1.5 SBF solutions for 14 days of soaking.In the case of the C2F sample (Fig. 4) the crystallinity of  HA observed on the surface of the immersed sample has the lowest crystallinity.In the same composite, the crystallinity of HM and S phases has been decreased, i.e. they were partially dissolved in 1.5 SBF solutions during the immersion time.As can also be seen, in both synthesized and immersed composites, XRD detects the presence of fibroin amorphous halo.The obtained results are in a good agreement with some of our preliminary results [43].FTIR data of the C1F and C2F composites, after soaking in SBF solution for 14 days are presented in Fig. 6.From the FTIR data, it can be seen that the presented spectra are similar to those of HA/protein composites previously reported by Yang et al. [44].On the other hand, for two composites obtained after immersion process, the FTIR spectrum (Fig. 6, curve a) exhibits the presence of C at 795 cm -1 , HM at 591 and 851 cm -1 and HA at 545, 960, and 1111 cm -1 [32].Fur- thermore, for the C2F sample (Fig. 5, curve b), the band posited at 939 cm -1 can be ascribed to S, and a band at 795 cm -1 can be related to the HM, and those at 608 and 840 cm -1 are indicative for HA [32].These data are in a good agreement with XRD results (Figs. ) in HA lattice.It can also be seen that two FTIR spectra display a band posited at ∼1500 cm -1 which can be attributed to the presence of A-type carbonate substitution (CO 3 2-− −− → OH -) in HA lattice [31,48,58,59], as well as to amide II [66].This finding allows us to confirm the presence of CO 3 2-ions within the channels of the HA structure which is in accordance with Ma et al. [47].From the data presented in Table 1 we can conclude that B-type CO 3 HA (bioapatite) was preferentially formed on the surface of the soaked samples in 1.5 SBF solution [45,47,50,52,54,55,57,58,60,61], accompanied with traces of different CaCO 3 phases -vaterite, aragonite and calcite [53,63,64,67].
XPS investigations of the prepared samples, after immersion in 1.5 SBF solutions, were performed to identify chemical states and surface structures obtained.A survey scan XPS spectra for the C1F and C2F samples immersed in 1.5 SBF for 14 days in static conditions are shown in Figs.7 and 8.The selected scan areas of XPS spectra for both samples immersed for 14 days in 1.5 SBF solutions are given in Fig. 9.The C1s peak position due to the adventitious carbon was detected at the binding energy (BE) value of 285 eV.The peak positions of different chemical states are listed in Table 2.
In the pure HA the peaks at 345.5 and 347.5 eV (Ca2p) and 133.5, 133.7 eV (P2p) and 531.1, 531.3 eV (O1s) confirm the presence of apatite groups [69][70][71][72].In addition, the peaks for Ca2p with BE at 347.7 eV can also be described for Si-HA structures with different quantity of silicon, as has been published by Solla et al. [70].On the other hand, the O1s spectrum indicates the presence of two additional peaks with BE values at 532.6 eV (for both samples) and 533.6 eV (for the C1F sample) and 533.8 eV (for the C2F sample).The first one may correspond to the adsorbed H 2 O on  the deposits [72].The second one indicate the presence of H 2 O ads., in accordance with Taube et al. [73].The Si2p peak at 103.5 eV can be related to Si-O bond of SiO 2 probably from cristobalite [74] in accordance with our XRD results (Fig. 2).Finally, the well resolved peak at 285.0 eV (data are not shown here) can be related to the presence of C x H y in accordance with Ni and Ratner [75].
It is well known that, it should be possible to distinguish between the various calcium phosphate phases by measuring the atomic fractions of Ca and P from XPS.Thus, the surface compositions of the C1F and C2F composites are presented in Table 3. Chusuei et al. found that in practice, the measured Ca/P ratios were all consistently lower than the theoretical Ca/P value [76].In our case, Ca/P ratio for the C1F sample (1.52 at.%) is closely related to the CO 3 HA [76], while the Ca/P ratio for the C2F sample (1.79 at.%) is slightly higher than the expected for the stoichiometric HA (1.66) [77].The slightly elevated Ca/P ratio observed here in the XPS results may also be a consequence of calcium enrichment of the uppermost surface of the HA derived coatings, as it has been previously observed by other authors [71,76,78].In summary, the XPS study indicates the formation of HA in our deposits, and corroborates the XRD and FTIR results.
Figure 10 presents the ICP-AES data for the evaluation of the ionic concentrations of Ca, Si, P and Mg in   low crystallinity of HM in the composite in agreement with XRD data depicted in Fig. 1.The highest Mg content for the C2F sample (45 mg/l) can be related to the presence of HM with high crystallinity as can be seen from Fig. 2. Furthermore, the Si content increased from 8 mg/l (for the C1F sample) to 47 mg/l (for the C2F sample).This fact could be related to the phase composition of the samples (Figs. 2 and 4).On the other hand, cristobalite has a low dissolution rate in comparison with S in SBF solution.Furthermore, both prepared composites were tested in a cytotoxicity assay on MG63 cells.The results showed that there was no effect of toxicity in the concentration range from 1 to 10 µg/cm 2 of C1F or C2F from day 1 to day 7. Phase contrast microscopy showed that on the first day following the treatment, the cells were attached and spread on the culture dishes, exhibiting a polygonal morphology (Fig. 10).The aggregate formation at the C1F and C2F samples in cell culture media was also clearly visible (Fig. 11).Possessing a normal rate of cell division, the cells proliferated and reached confluence on the third day, immobilizing the granules in the cell layer.Thereafter, the cells piled up around the granules forming refringent multilayers (Fig. 11).The MG63 cells did not show any evidence of advanced differentiation, during the studied period, and no significant morphological differences were noted between the two substances used for treatment of cultures were noted.
A number of studies have shown that ionic products released from glass-ceramics by dissolution, creating a favourable environment for osteoblast proliferation and differentiation in vivo.However, the high bioactivity index could result in a rapid pH shift of the culture medium in vitro that may be unfavourable for cellular metabolism.To study this process we additionally examined the effect of the C1F and C2F on the MG63 mineralization in period up to 21 days, as an indicator of advanced cell differentiation.The formation of calcium nodules was detected by ARS staining.As illustrated in Fig. 12, differentiation induced by calcium nodules were observed in MG63 cells treated with both composites for 21 days.Cells were treated with 10 µg/cm 2 of C1F (Fig. 12A) or C2F (Fig. 12B).All cells were stained with Alizarin Red S to detect the calcium nodules and analysed at magnification of 40×.The cycles represent an area, where the cells form calcium nodules in contact with composites.
In order to determine the bioactivity of MG63 cells, cultured in the presence of C1F and C2F composites, we focused on their alkaline phosphatase (ALP) activity, as this enzyme is a common indicator of the expression of the osteoblastic phenotype.Figure 13 shows the ALP activity of MG63 cells cultured with C1F and C2F for 7 and 21 days.The ALP activity of the MG63 cells cultured for 7 days with C1F and C2F did not differ significantly.However, the ALP activity of the MG63 cells cultured with the C2F was slightly higher than those for the other samples cultured with the C1F composite.In particular, the ALP activity of MG63 cells cultured with the C1F and C2F composites was approximately 4-fold higher on day 21 than that on day 7.

IV. Conclusions
The two novel glass-ceramics in the CaO-SiO 2 -P 2 O 5 -MgO system, after soaking in TRIS-HCl solution, have been used for the synthesis of silk fibroin/glass-ceramic composites in 20 : 80 weight ratio by mixing method.After synthesizing the two samples, XRD proved the presence of different crystalline phases such as hydroxyapatite, whitlockite, cristobalite, silicocarnotite and hydromagnezite.FTIR of the prepared composites showed the presence of basic absorption bands corresponding to the organic and inorganic phases of the composites.
After vitro test in 1.5 SBF solutions for 14 days in static conditions, XRD revealed the presence of HA.On the other hand, the crystallinity of whitlockite, hydromagnezite and silicocarnotite slightly decreased, i.e. these phases partially dissolved in 1.5 SBF solutions during the soaking time.Furthermore, FTIR depicted that B-type CO 3 HA (bioapatite) was preferentially precipitated on the soaked surfaces.XPS of the immersed samples proved that HA was formed on the composites.ICP-AES measurements lead us to conclude that in C2F sample silicon substituted CO 3 HA can be formed on the soaked surface.The biological experiments with MG63 osteosarcoma cells led us to resume that the prepared composites create a favourable environment for osteoblast proliferation and differentiation in vivo.In this context, the ALP activity of the MG63 cells cultured for 7 days with C1F and C2F did not differ significantly.However, the ALP activity of the MG63 cells cultured with C2F was slightly higher than those for the other samples cultured with C1F.In particular, the ALP activity of MG63 cells cultured with C1F and C2F composites was approximately 4-fold higher on day 21 than that on day 7.

Figure 1 .
Figure 1.XRD pattern of C1F sample, before in vitro test

Figure 3 .
Figure 3. FTIR spectra of C1F (a) and C2F (b) composites before in vitro test

Figure 4 .
Figure 4. XRD pattern of C1F sample, after soaking in SBF solution for 14 days

Figure 5 .
Figure 5. XRD pattern of C2F sample, after soaking in SBF solution for 14 days

Figure 7 .Figure 8 .
Figure 7. XPS spectra of survey scan for C1F sample after in vitro test in SBF solution for 14 days

Figure 10 .
Figure 10.ICP-AES for C1F and C2F samples after immersion in SBF solution for 14 days the 1.5 SBF solutions after soaking of the obtained samples for 14 days in static conditions.The Ca rate for the C1F and C2F composites shows similar behaviour.The quantity of Ca increases to 330 and 394 mg/l compared with the Ca concentration of the initial 1.5 SBF solution (89 mg/l).The observed concentrations are in relation with the weight ratio of the C1 and C2 glass ceramics in the composites (80 wt.%), and the processes of partial dissolution of W, S, HA and HM under the experimental conditions are in agreement with XRD data (Figs.4and 5).In accordance with our preliminary results[31,32], the P concentration (6 and 0.5 mg/l) is probably the limiting factor for the growth of the CO 3 HA layer on the surface of the synthesized composites.The equal rate of Mg for the C1F sample (19 mg/l) with Mg content in 1.5 SBF solution (19 mg/l) can be explained with the

Figure 11 .
Figure 11.Phase contrast micrograph, showing MG63 cell proliferation following treatment with 10 µg/cm 2 of C1F or C2F at different time points (the arrow showed the formation of bioglasses aggregates and the arrowheads denote the formation of refringent matrix around the granules)

Table 1 .
IR vibration modes and their assignments observed for the C1F and C2F samples, after in vitro test in SBF

Table 2 .
XPS profile parameters for Ca2p, O1s, Si2p, P2p and C1s states for different samples obtained on the surface, after in vitro test in 1.5 SBF solutions for 14 days of soaking in static conditions

Table 3 .
Concentration of the elements on the surfaces of C1F and C2F samples immersed in SBF for 14 days