Introduction

Bioactive glass and glass-ceramics are well-known reactive materials and can be integrated into human body where they can form biologically active apatite (Hydroxy apatite) layer at the bone/implant interface. Due to this reason they act as the fillers for bioactive composites (Bramhill et al. 2017; Greenspan et al. 1994, 1997, 1998). It is important to note that the bond between a glass material and the bone is a precipitation of an apatite layer on the surface of the glass material which is responsible for bioactivity. The mechanism for bone bonding is attributed to a hydroxy-carbonated apatite (HCA) layer on the glass surface, based on initial glass dissolution (Hench et al. 1971). Hydroxy carbonated apatite is similar to bone mineral and it interacts with collagen fibrils to integrate with the host bone (Hench and Polak 2002). For tissue engineering applications, it is essential that the nucleation and growth of hydroxy apatite (HA) layer be fast on the surface of the bioglass surface in a precise reaction time in body environment (Kokubo et al. 1990; Sooksaen et al 2015; Yousef et al. 2014).

The HA growth rate is strongly influenced by the glass synthesis processes such as sol–gel processes, melt quenching technique, etc. (Catauro et al. 2016). Sol–gel glasses can be produced at low temperature conditions with good homogeneity. The sol–gel method offers potential benefits for obtaining the powdered materials with good control of composition, microstructure, and wider range of bioactivity (Ahmed et al. 2012; Bellucci et al. 2012; Goller et al. 2004; Arcos and Vallet-Regíl 2010). Sol–gel process is a more convenient technique than melt quenching technique to improve HCA layer growth rate in simulated body fluid (SBF) solution based on dissolution property of synthesised glass–ceramics in SBF solution. SBF is a solution ionic concentration close to that of human blood plasma (Kokubo and Takadama 2006). Energy dispersive X-ray (EDS or EDX) analysis is an important characterisation technique for the identification of presented elements in the sample. Using EDX analysis it can be confirmed whether formed layer in dissolution process is HCA or not, by the identification of Ca, P, O and C elements (Goldstein 2003). Based on higher rate of dissolution in SBF solution these glass samples have implications for biomedical applications such as tissue engineering, bone bonding ability, cancer therapy, etc. (Hashmi et al. 2014; Peitl et al. 2001; Lukito et al. 2005; Sooraj Hussain et al. 2004; Laczka et al. 1999).

CaO/P2O5 ratio actually influences the HCA growth rate in SBF solution for calcium phospho silicate glasses (Sopcak et al. 2015; Rogar et al. 2017; Fateme and Saeed 2013). Network modifiers can break the glass structure leading to non-bridging oxygen [NBO] formation. The formation of non-bridging oxigen results in the softening of the glass-ceramic structure. In particular, the NBO/BO ratios affect its bio activity (Lucia et al. 2009). Raman spectroscopy is a powerful tool for non bridging oxygen identification and for finding the concentration of non bridging oxygens in glass matrix (Gonz_alez et al. 2003). Based on the sintering temperature, bio-activity of sol–gel SiO2–CaO–P2O5 bio glass can influence the wollastonite or psudo-wollastanite structure formation after SBF treatment (Ma et al. 2010). Li2O addition to the calcium phosphosilicate glass resists the HCA growth rate in dissolution process; the problem of HCA layer growth rate resistance in SBF-soaked calcium phosphosilicate glasses can be overcome by Na2O addition (Kavitha et al. 2014; Lusvardi et al. 2009). Catauro et al. (2016) reported that melt quenched 47S SiO2–CaO–P2O5–Na2O bio active glass form HCA crystalline layer on glass surface after 30 days of SBF soaking time for pelletized glass samples. HCA layer formation for sol–gel derived SiO2–CaO–P2O5–Na2O bio-glass strongly influenced by precursors such as tri ethyl phosphate (TEP) or phosphoric acid (H3PO3) (Siqueira and Edgar 2013). Based on CaO quantity, lower P2O5 content and higher molar SiO2 content sol–gel derived calcium phosphosilicate glasses exhibit good bioactivity. Higher CaO content leads to amorphous nature of glasses before SBF treatment (Kiran et al. 2017). Recent study shows that barium with low concentrations in the glasses acts as a muscle stimulant and is found in human teeth (Carta et al. 2007). Bone stimulating property can be improved by incorporation of bone stimulator ions into their chemical compositions. Moreover, there are limited studies on the role of CaO substitution on apatite layer formation of barium-doped phosphosilicate glasses (Sampath Kumar et al. 2015; Kokubo and Takadama 2006) and on the importance of higher BaO contents on HCA-forming ability for sol–gel derived soda lime phospho silicate glass system.

The present study is focussed on the sol–gel synthesis of [58SiO2–(32 − x) BaO–xCao–6Na2O–4P2O5 bioactive (x = 15, 20, 25 and 30 mol%)] composition. Thermal, structural properties of synthesised glass samples were studied using TGA/DTA analysis and XRD techniques for the confirmation glass-ceramic nature of the material. The non-bridging oxygens (NBO) and bridging oxygens (BO) in the synthesised composition were identified by FTIR and Raman spectroscopic techniques and further corresponding NBO/BO ratios were found using Raman spectroscopic analysis. The effect of NBO/BO ratio on early apatite layer formation for synthesised bioactive system has been investigated.

Materials and methods

Synthesis of barium doped soda lime Phosphosilicate glasses

58SiO2–(32 − x) BaO–xCaO–6Na2O–4P2O5 (x = 15, 20, 25 and 30 mol%) samples were synthesised by conventional sol–gel process and named as CPSNB1, CPSNB2, CPSNB3 and CPSNB4, respectively, as shown in Table 1. The selected precursors for preparation were tetraethylorthosilicate [Si(OC2H5)4] (≥ 99%), triethylphosphate (TEP) [(C2H5O)3PO] (≥ 99%), calcium nitrate tetra hydrate [Ca(NO3)24H2O] (98%), sodium nitrate (NaNO3) (99%), barium nitrate [Ba(NO3)2] (≥ 98%). Required chemicals for synthesis were purchased from Merck company (Mumbai, India). H2O and HNO3 (69%) were taken as [(mol of H2O)/(mol of TEOS + mol of TEP) = 10] and [(mol of HNO3)/(mol of TEOS + mol of TEP) = 0.05], respectively. In order to synthesise the glass sample, TEOS was mixed with H2O and HNO3 and stirred for 1 h. At an interval of 1 h, TEP, calcium nitrate, sodium nitrate and barium nitrate were added subsequently and stirred well. The prepared sols were poured into Teflon beakers, sealed and kept inside hot air oven at 60 °C temperature for 3 days for ageing purpose. Subsequently the aged gels were dried at 130 °C for 4 h. The dried gels were ground, made into powders and heated at the rate of 5 °C/min up to 700 °C and stabilised at that temperature for 4 h to obtain glass samples in the powder form. After getting powder samples, pellets have been prepared using hydraulic press by applying 5 tons of pressure for 5 min (Kiran et al. 2017).

Table 1 Batch composition of 58SiO2–(32 − x)BaO–xCaO–6Na2O–4P2O5 glasses
Full size table

Characterisation of bio-active glasses before and after soaking in SBF solution

The quantitative analysis is based on the glass transition (T g) and onset crystallisation (T x) temperatures of gel samples. T g and T x were identified by TGA and DTA analysis using SII EXTRAR 6000, Japan with flow rate of 10 °C/min in the temperature range 27–1000 °C. The crystalline nature of the samples was observed by JOEL, JDX-8P X-ray diffractometer (Japan) with CuKα radiation within the range of diffraction angles 20°–80°. SHIMADZU FTIR spectrometer (North America) and LABRAM-HR 800 Laser Raman spectrometer (Japan) were used for analysis of the assigned bands in glass samples. To study the HCA growth properties for the synthesised samples, SBF solution has been selected as physiological environment which has similar properties as those of human blood plasma. The SBF solution was prepared by dissolving reagent grade KH2PO4 (≥ 98%), CaCl2 (≥ 98%), NaHCO3 (99%), MgCl2·6H2O (98%), KCl (99%) and NaCl (99%) in the de-ionised water at pH 7.4 with tris-buffer (99.8%) while maintaining temperature at 37 °C (Table 2). The pH values of SBF solution were measured using pH meter (Eutech, pH 510, India) before and after soaking CPSNB samples. PO4 3−and Ca2+ ion concentrations were measured using UV/Vis absorption spectrometer (HITACHI PM & E 101, Canada) and Flame Photometer (ELICO CL378, Germany). The surface morphology of samples was observed by JOEL-JSM-6380LA, Japan, scanning electron microscope (SEM) and elements on surface were identified by EDX analyser after gold sputtering with JFC 1600 auto fine counter (JOEL, Japan). The presence of HCA layer was confirmed using XRD, SEM/EDX, TEM/SAED (JOEL Company JEM 2100 model, Japan), FTIR and Raman spectroscopic techniques.

Table 2 Ion concentration (mM) in SBF and in human blood plasma
Full size table

Results and discussion

TGA/DTA analysis

TGA/DTA curves were recorded to find the thermal behaviour of CPSNB glasses. Three weight losses were observed in TGA curves as shown in Fig. 1a–d. The first weight loss was at 108–123 °C related to the evaporation of organics. The second weight loss was at 497–535 °C related to the elimination of water caged in pores and third was at 598–627 °C related to the organic residues from precursors used for samples synthesis and to dehydration process (Table 3). The glass transition temperatures were observed using DTA curves. These results dictated that CaO incorporation increases all the three temperatures, namely (1) the nucleation temperature (T g ) at 294–340 °C (2) the onset stabilisation temperature (T x1) at 400–428 °C and (3) the crystallisation temperature (T c1) at 420–472 °C range. These three temperatures were known to be related to primary crystalline phase (Marczewska et al. 2017; Shelby 2005). Second, third, fourth and fifth crystalline temperatures were also observed for CPSNB1, CPSNB2, CPSNB3 and CPSNB4 samples as shown in Table 5 (Zhao et al. 2010). From these observations it was concluded that mixed alkaline earth oxide effect favours the crystalline nature of the glass samples.

Fig. 1
figure 1

TGA/DTA curves of a CPSNB1, b CPSNB2, c CPSNB3 and d CPSNB4 glasses

Full size image
Table 3 TGA/DTA-related temperature values of 58SiO2–(32 − x)BaO–xCaO–6Na2O–4P2O5 glasses
Full size table

XRD analysis

Figure 2a shows the XRD pattern of CPSNB samples. All samples show broad hump at 20°–30° with some sharp intense peaks, indicating that all CPSNB samples had glass-ceramic nature. Crystalline peak intensities decreased for CPSNB samples with increasing CaO content from 15 to 30 mol%. All glass samples were sintered at 700 °C. Sintering temperature was higher than the onset crystalline temperature (As discussed in DTA curves). Due to this reason all glass samples showed crystalline peaks. The difference between sintering and onset crystalline temperatures was in decreasing order from CPSNB1 to CPSNB4 glass and leading to decrease in crystalline nature with CaO addition.

Fig. 2
figure 2

XRD pattern of CPSNB samples a before and b after SBF treatment

Full size image

For SBF-treated CPSNB samples XRD pattern (Fig. 2b) showed HA crystalline peaks. The major diffraction peak was identified at 2θ ~ 32° [(hkl) = (211)]. d-spacing of crystalline peaks were matched with the standard JCPDS file with number 01-074-0565. Calcite phase was also observed at 29° (JCPDF NO 01-081-2027). In SBF treatment, a chemical reaction occurs on sample surface. In this process Ca2+ ions migrate into SBF solution and form silanol (Si–OH) groups. Due to poly-condensation process silica gel layer forms on sample surface. Phosphate and calcium ions migrate through silica gel layer and form apatite layer on the sample surface. Due to crystallization process between apatite layer and prevalent hydroxyl, calcium, phosphate ions in SBF, HA layer form on sample surface. CPSNB1 sample has more BaO content than CPSNB2. BaO addition decreases the glass strength, increases the dissolution in SBF and it leads to increased HA formation on sample surface. For SBF-treated samples crystallite sizes were calculated for (211) and (132) planes using following Scherrer equation:

$$D = k\lambda /\beta \cos \theta ,$$

where D is crystallite size, K is Scherrer constant, λ is wave length of CuKα radiaton, β is full width half maximum (FWHM) and 2θ is the angle of diffraction. The measured FWHM and D values are as shown in Table 4. HA crystallite sizes were decreased with increasing CaO content from 15 to 20 mol%. It corroborates that HA-forming ability is more for CPSNB1 compared to CPSNB2 sample. HA crystallite sizes were increased from 20 to 30 mol % of CaO. HA layer formation depends on Ca2+ ions’ dissolution in SBF solution. CaO quantity increases from CPSNB2 to CPSNB4 sample. Due to this reason Ca2+ ions’ dissolution increases from CPSNB2 to CPSNB4 sample, which led to increase in HA crystalline intensity from CPSNB2 to CPSNB4 sample. It corroborates that HA-forming ability increases from CPSNB2 to CPSNB4 sample. From all these observations it was concluded that both the conditions, namely (i) higher BaO content and (ii) increase in CaO content favour the HA formation.

Table 4 FWHM and crystallite size values of HA from XRD analysis for (211) and (132) planes
Full size table

Surface morphology

Surface morphology of CPSNB samples before and after SBF treatment is shown in Figs. 3 and 4. Before SBF treatment (Fig. 3a–d) all CPSNB samples had not shown any spherical shaped crystalline particles and corresponding EDX analysis confirmed that elements present in sample were Ba, Si, Ca, Na, P and O. After SBF treatment the SEM images (Fig. 4a–d) clearly exhibited the spherical shaped nuclei on sample surface. As discussed in the previous section the crystallization process between apatite layer and existing hydroxyl, calcium, phosphate ions led to the formation of HA on sample surface. In this process, with the presence of CO3 2−, HA gets converted into HCA. EDX analysis also supports the HCA layer formation on the CPSNB sample surfaces by identifying elements as Ca, P, C and O. EDX spectra show that Ca intensity decreased for CPSNB samples with increase in CaO content from 15 to 20 mol%. All SBF- treated CPSNB samples show good homogeneity in particle size and smooth spherical morphology with HCA particles (Fig. 5a–d). For CPSNB1 samples HCA average size is 1517 nm. For CPSNB2, CPSNB3 and CPSNB4 sample, average HCA particle sizes are 1355, 1587 and 1605 nm, respectively. This shows that average particle size decreases with increase in CaO content from 15 to 20 mol% and increases with increase in CaO content from 20 to 30 mol% (Yadav and Singh 2015). From these observations it could be concluded that HCA-forming ability is more for CPSNB1 compared to CPSNB2 sample and HA-forming ability is in increasing order from CPSNB2 to CPSNB4 sample as observed in XRD analysis.

Fig. 3
figure 3

SEM images of a CPSNB1, b CPSNB2, c CPSNB3 and d CPSNB4 glass samples and corresponding EDS spectra

Full size image
Fig. 4
figure 4

SEM images of a CPSNB1, b CPSNB2, c CPSNB3 and d CPSNB4 glass samples after SBF treatment and corresponding EDS spectra

Full size image
Fig. 5
figure 5

HCA Particle size distributions for SBF soaked a CPSNB1, b CPSNB2, c CPSNB3 and d CPSNB4 glass samples

Full size image

Raman analysis

Figure 6a shows Raman spectra of CPSNB samples. For all CPSNB samples assigned D2 modes were observed at the wave number range of 565–593 cm−1. Si–O–NBO, Si–O–Si asymmetric stretching modes were observed at 956–965 and 1058–1068 cm−1, respectively, as shown in Table 5. Using de-convolution process, non-bridging and bridging oxygen intensities were calculated (intensities were considered as corresponding fitted areas of Si–O–NBO, Si–O–Si asymmetric stretching modes) (Lucia et al. 2009). NBO/BO ratio varies with CaO as shown in Fig. 6b–e (Lucia et al. 2009). The obtained NBO/BO ratios for CPSNB1, CPSNB2, CPSNB3 and CPSNB4 samples were 0.36188, 0.22859, 0.26729 and 0.47349, respectively. It indicates that NBO/BO ratio decreased with increase in CaO from 15 to 20 mol% and increased with increase in CaO from 20 to 30 mol% (Yadav and Singh 2015). Wollstonite crystalline phase was also observed in the wave number region 1044–1051 cm−1 (Aguiar et al. 2009). For SBF-treated CPSNB samples, wollstanite crystalline phase has not been observed. It gave evidence to the fact that the wollstanite dissolved in SBF. Phosphate symmetric stretching, carbonate stretching vibrational modes were observed at 952–986 and 1076–1084 cm−1, respectively, as shown in Fig. 6f (Table 6). CaO addition led to the deformation of Si–O–Si asymmetric stretching mode and formation of non-bridging oxygen modes (Fig. 7a). Si–O–Si stretching deformation causes not only Si–O–NBO formation, but also D2 mode formation (Fig. 7b).

Fig. 6
figure 6

a Raman spectra of 58SiO2–(32 − x) BaO–xCaO–6Na2O–4P2O5 system and the deconvoluted results of the, b sample CPSNB1, c sample CPSNB2, d sample CPSNB3, e sample CPSNB4 f Raman spectra of SBF treated 58SiO2–(32 − x) BaO–xCaO–6Na2O–4P2O5 system

Full size image
Table 5 Raman absorption bands for 58SiO2–(32 − x)BaO–xCaO–6Na2O–4P2O5 glasses before soaking in SBF solution
Full size table
Table 6 Raman absorption bands for 58SiO2–(32 − x)BaO–xCaO–6Na2O–4P2O5 glasses after soaking in SBF solution for 7 days
Full size table
Fig. 7
figure 7

a Variation of Raman shift of Si–O–NBO (as s) versus Si–O–Si (as s) with CaO mol%. b Variation of Raman shift of Si–O–Si (as s) versus D2 with CaO mol%. c Variation of Raman Si–O–NBO (as s)/Si–O–Si (as s) intensity with CaO mol%

Full size image

For SBF-treated CPSNB samples XRD pattern confirmed the HA formation. Raman spectroscopic analysis confirmed that carbonate groups were present. From these observations it was concluded that HCA layer formation was complete. SEM/EDX analysis also confirmed the HCA formation. Raman spectra confirmed that NBO/BO ratio decreased with CaO from 15 to 20 mol% and increased from 20 to 30 mol% (Fig. 7c). SEM/EDX analysis confirmed that HCA particle sizes decreased with CaO in the range 15–20 mol% and increased in the range from 20 to 30 mol%. From all these observations, it was concluded that NBO/BO ratio was proportional to HCA particle sizes of SBF treated CPSNB samples.

FTIR analysis

Figure 8a shows FTIR spectra of CPSNB samples in the wave number range 400–3500 cm−1. Si–O–Si and Si–O–NBO asymmetric stretching modes related to silica matrix were observed at 1033–1103 and 925–948 cm−1, respectively. Si–O–Si rocking, bending modes were observed at 478–486, 771–779 cm−1, respectively. PO4 3− and CO3 2− bending modes were also observed at 594–624 and 1427 cm−1, respectively. The vibration mode due to deformation of –OH groups was assigned at 1650–1658 cm−1 (Table 7).

Fig. 8
figure 8

FTIR spectra of CPSNB samples a before and b after SBF treatment

Full size image
Table 7 FTIR band assignments of barium-doped soda lime phosphosilicate glasses before soaking in SBF solution
Full size table

For SBF-treated CPSNB samples (as shown in Table 8) Si–O–Si modes were related to Si–O–Si rocking, bending and asymmetric stretching modes are observed at 458–473, 729–786 and 1097–1243 cm−1, respectively (Fig. 8b). Calcium group was observed as Si–O–Ca (as Si–O–NBO) at 879–928 cm−1. CO3 2− bending mode was also observed at 1471–1491 cm−1. PO4 3− bending modes with amorphous and crystalline nature were observed at 599–661 and 679–692 cm−1, respectively. Si–O–Si asymmetric stretching mode broadness was increased for SBF-treated samples compared to synthesised samples, which shows that silica gel layer formed on the sample surface in poly condensation process. From these observations, it can be confirmed that HCA layer was completely formed in SBF treatment. OH groups and water molecules were observed at 1637–1718 and 2337–2363 cm−1, respectively.

Table 8 FTIR band assignments of barium-doped soda lime phosphosilicate glasses after soaking in SBF solution for 7 days
Full size table

TEM/SAED analysis

Figure 9a, c, e and g reports the TEM images of SBF-treated CPSNB samples. Spherical shaped particles have been observed in TEM images. Using SAED pattern (Fig. 9b, d, f and h) d-spacing of relevant TEM images were calculated using the following formula:

$$d_{\text{hkl}} = \frac{\lambda L}{R},$$

where d hkl = lattice spacing (nm), λ = wavelength of the TEM accelerating voltage (nm), L = camera length (mm), R = measured diffraction ring radius from a polycrystalline sample, or measured distance between two adjacent diffraction spots from a single crystal sample (mm) (John MC, Norrox Scientific Ltd, Ottawa, Canada, http://www.oocities.org/mag-i-cal@rogers.com/ABMcCaffrey-std-res.pdf).

Fig. 9
figure 9

TEM images of a CPSNB1, c CPSNB2, e CPSNB3 and g CPSNB4 glass samples and SAED pattern of b CPSNB1, d CPSNB2, f CPSNB3 and h CPSNB4 glass samples after SBF treatment

Full size image

The calculated d-spacings were matched with the d-spaces of HCA layer (Standard JCPDS card no.09-0432) (Table 9). Hence, it was concluded that the spherical shaped particles were HCA particles.

Table 9 Comparison of d-space for Hydroxyl apatite by SAED analysis and JCPDS file with Reference code: 01-086-0740
Full size table

pH Assessment, dissolution and weight loss studies

In the dissolution process calcium ions get released into SBF solution, which led to silanol groups’ (Si–OH) formation on sample surface. Due to the poly condensation process of silanol groups, silica gel layer was formed on sample surface. Ca2+ and PO4 3− ions of sample leach on the silica gel layer surface, leading to calcium phosphate layer formation. This changes the phosphate and calcium ion concentrations of SBF solution. Incorporation of OH−, PO4 3−, Ca2+ and carbonate ions from SBF solution into apatite layer led to HCA formation on the sample surface (in the crystallization process). pH values of SBF solution were measured before and after soaking CPSNB samples in SBF for 7 days. As shown in Fig. 10a the pH values were increased till 10 h; this was due to the fast release of Ca2+ ions into SBF and forming silanol groups (Introduction to Glass Science et al. 2005). After 10 h pH values were almost stabilised.

Fig. 10
figure 10

Variation of a pH, b Ca concentration c P concentration values with respect to SBF soaking time for CPSNB glass samples

Full size image

For CPSNB1 sample, NBO/BO ratio is high compared to CPSNB2 sample. NBOs are in the form of Si–O–Ca. HA layer formation on sample surface depends on calcium, phosphate ions’ dissolution of glass in SBF solution. NBO/BO ratio is also in the order CPSNB4 > CPSNB1 > CPSNB3 > CPSNB2. All CPSNB samples have 4 mol% P2O5. So, for SBF-treated samples CaO content plays major role for HA formation. Releasing Ca2+ ions into SBF for CPSNB samples are in the sample order of CPSNB4 > CPSNB1 > CPSNB3 > CPSNB2 and HA formation was also in the same (decreasing order) order Fig. 10b. From all these observations, it can be concluded that HCA formation is NBO/BO ratio dependent. Same observations were supported in XRD, SEM and EDX analyses also. Due to this, CPSNB samples’ weight loss also changed (decreased) in the order of CPSNB4 > CPSNB1 > CPSNB3 > CPSNB2 (Table 10) (Fateme and Saeed 2013). Ba2+ ionic radius is more than Ca2+. So, increase in BaO addition decreases the local field strength, leading to increase in dissolution of CPSNB sample. Compared to CPSNB2, CPSNB1 has more Ba2+ ions and leads to increase in Ca2+ and PO4 3− ion dissolution in SBF. The number of Ba2+ ions is less, and the number of Ca2+ ions is more in CPSNB3 compared to CPSNB2 sample, leading to increase in local field strength for CPSNB3 sample. Due to this reason PO4 3− ions’ release decreases in dissolution process for CPSNB3 compared to CPSNB2 sample. With low BaO, field strength is high for CPSNB4 compared to CPSNB3 sample. Due to this reason CPSNB4 sample releases less PO4 3− ions into SBF solution compared to CPSNB3 sample. From these observations it could be concluded that decrease in BaO reduces the PO4 3− dissolution. PO4 3− ions’ dissolution decreased from CPSNB1 to CPSNB4 sample. SBF solution needs to utilise PO4 3− ions for HCA formation in the increasing order from CPSNB1 to CPSNB4 sample. So, phosphate ion concentration of SBF solution decreased for CPSNB1 to CPSNB4 sample as shown in Fig. 10c.

Table 10 Weight loss % of SBF soaked 58SiO2–(32 − x)BaO–xCaO–6Na2O–4P2O5 glasses with NBO/BO ratio and HCA particle sizes
Full size table

Conclusions

  1. 1.

    SiO2–Na2O–BaO–CaO–P2O5 glasses were synthesised using sol–gel method. DTA analysis confirmed that increase in CaO content increases the nucleation and crystallizationtemperature. XRD pattern confirmed that increase in CaO content increased amorphous nature of the synthesised glass.

  2. 2.

    NBO/BO ratio is proportional to HA crystallite size. For synthesised glasses NBO/BO ratio decreased with increase in CaO content from 15 to 20 mol% and increased with increase in CaO content from 20 to 30 mol%.

  3. 3.

    After 7 days of SBF treatment HCA particle sizes were decreased with increase in CaO content from 15 to 20 mol% and increased from 20 to 30 mol% of CaO which has been observed by EDX, XRD analysis, respectively. FTIR, Raman and TEM/SAED analysis confirmed that formed crystals are HCA crystals

  4. 4.

    Dissolution studies also confirmed that weight loss of samples decreased with increase in CaO from 15 to 20 mol% and increased with increasing CaO from 20 to 30 mol%. Weight loss of SBF-treated samples was proportional to NBO/BO ratio. From these observations it was concluded that both network modifiers (CaO and BaO) play their important roles in the formation of HCA layer in SBF treatment.

  5. 5.

    This work adds valuable information for the improvement of HCA growth rates for SBF-treated samples based on CaO and BaO quantities, in tissue engineering applications.