5.1. Thermal behavior of gel powders
Figure1 showstheTGA−DSC curvesoftheas-dried gel powders. The four specimens showed similar thermal behavior, indicating that the first weight loss of about 10% was found, followed by continuous weight loss from 100 to 450ºC in the TGA curve. The decomposition and oxidation of the decomposed products were accelerated at temperatures of 450−550ºC, as evidenced by a rapid weight loss of more than 30%. In the DSC traces, there were four endothermic peaks at about 100, 520, 540, and 650°C, and one exothermic peak at about 900°C.
5.2. Powder morphology and composition
The particle size of powders is an important characteristic, and therefore it was analyzed by SEM. In Figure 2, the SEM micrograph shows all of the milled particles before mixing with water to essentially be an assembly of irregular particles with sizes ranging from 0.5 to 5 µm. The more CaO contents the powders contained, the smaller the particle size was.
Additionally, the powers with higher SiO2contents presented a denser structure.
Figure 3A shows the XRD patterns of the 5 SiO2–CaO powders sintered at 800ºC and indicates that the phase evolution is dependent on the Si/Ca ratio of the precursors. The major diffraction peaksat2θ between 32o and 34o wereattributed to theβ-dicalcium silicate( β-Ca2SiO4)phase.Thepeak intensitiesofβ-Ca2SiO4 and CaO increased with increasing CaO content in the precursors. The trends in the FTIR spectra (Fig. 3B) are similar to those indicated by XRD. For the specimen with the greatest amount of silica (S70C30), the broad IR absorption band corresponding to SiO4asymmetric stretching extended over a wide wave-numberrangeof1300−950 cm-1. An obvious sharpening and shifting to lower frequency in Si–O–Si asymmetric stretching bands were detected as the silica content decreased.
5.3. Cement morphology and composition
When the powder solid was mixed with water, the products of the hydration process were C–S–H at 29.3o and incompletely reacted inorganic component phases (Fig. 4A). The lower Si/Ca ratio was in the precursor, the higher the C–S–H content was in the cement.
Except S70S30 cement, which had a loose and rough surface (Fig. 4B), all other specimens had a smooth appearance with entangled particles (Fig. 4C−4F). Moreover, it seemed that S50C50 had a denser structure than the other cements.
5.4. Setting time and compressive strength
Thesetting timesofthefiveCSC cementsranged from 12−42 minutes(Table1);these values were significantly different (P < 05). Table 1 also shows the CS values of cement specimens. One-way ANOVA analysis of the CS data shows that the variations in strength between specimens are significant (P < .05).
5.5. Characterization of immersed specimens
Broad face SEM micrographs and EDS profile of S40C60 cement specimens immersed in a simulated body fluid for 1 hour are shown in Figure 5. It is clear that, precipitation took place on the cement surface which was covered with clusters of precipitated spherulites (Fig.
5A). The Ca/P and Ca/Si ratios of the cement specimen after 1 hour of immersion were 1.98 and 6.19 (Fig. 5B), respectively.
When the powder solid is mixed with water, the products of the hydration process are C–
S–H at 29.3o and incompletely reacted inorganic component phases (Fig. 6A). Additionally, thediffraction peaksat2θ between 32.0o and 34.5o wereattributed to theβ-Ca2SiO4 phase.
Afterimmersion in simulated body fluid,itcan beeasily seen thattheintensitiesofthe β-Ca2SiO4 phase gradually reduced. Some broad and diffusepeaksat2θ = 25.9o as well as 31.8−32.9o appeared instead, which may be ascribed to the characteristic peaks of apatite.
Interestingly, the patterns also showed that the relative peak intensity of the C–S–H component of the day 1 specimen was higher than that of the day 0 specimen.
The changes in compressive strength of S40C60 cement specimens after immersion in simulated body fluid as a function of time are shown in Figure 6B. The values at 0-, 1-, 7-, and 30-day immersion in a simulated body fluid were 12.3 ± 2.3, 20.2 ± 2.8, 23.0 ± 3.4, and 22.3 ± 1.6 MPa, respectively. The immersed specimens were significantly (p < 0.05) higher than that obtained for the day 0 specimens.
5.6. Biocompatibility of S40C60 cement
In Figure 7A WST-1 assay shows that the cell viability increased 15% and 23% at 6-hour and 7-day incubation, respectively, when compared with the controls. The SEM image reveals
cells that appeared flat and exhibited intact, well-defined morphology (Fig. 7B).
5.7. Composition and morphology of gelatin-containing cement
XRD patterns ofallcementsrevealed an obviousdiffraction peak around 2θ = 29.4o, corresponding to the calcium silicate hydrate (C–S–H) gel, and incompletely reacted inorganiccomponentphasesofβ-Ca2SiO4(Fig. 8). It is clear that the peak intensity of C–S–H formed in the gelatin-containing cement specimen was lower, in particular for 10% gelatin, when compared with the corresponding control. The SEM micrographs of the cements with and without 10% gelatin indicated that the as-set CSC controls had an appearance with entangled particles and exhibited several pores. A dense structure appeared for the cement with larger CaO amounts in the sol. In contrast, when 10% gelatin was added, the organic-inorganic hybrid cements became more compact than the corresponding cements without gelatin.
5.8. Anti-washout properties
Washout resistance results indicated that the four controls appeared to degrade after immersion. It can clearly be seen that the gelatin hybrid cements resisted washout, showing no noticeable breakdown.
5.9. Setting time
After mixing with water, the control cements set in the range of 10 to 29 min, which were shortened with an increase in the concentration of the calcium component, as shown in Figure 6. These values were significantly (P < 0.05) different from each other. Scheffe’s multiple comparison testing revealed no significant difference (P > 0.05) between S30C70 and S40C60. It was also not significant different (P > 0.05) for the S50C50 and S60C40 control. The addition of 5% and 10% gelatin significantly (P < 0.05) prolonged the setting time of the hybrid cements by a factor of about 2 (25–69 min) and 8 (108–282 min), respectively. In the case of CSCs containing either 5% or 10% gelatin content, setting time had significant differences (P < 0.05) between various CSCs.
5.10. Diametral tensile strength and modulus
Figure 9 shows that the DTS values of the hardened control cements were 2.0, 2.6, 2.0, and 1.0 MPa with an increasing CaO content, indicating there was a significant difference (P
< 0.05). The incorporation of gelatin into the CSC did not significantly affect (P > 0.05) its strength, which was comparable to that of the control cement. In the case of S40C60 group, the DTS values were 2.0, 2.1, and 1.7 MPa for 0, 5, 10% gelatin content, respectively. The significant differences (P < 0.05) in DTS between S40C60 and S60C40 groups were not found. On the other hand, Scheffe’smultiplecomparison testingindicated that both 5% and 10% gelatin-containing S50C50 cement specimens had significantly (P < 0.05) higher DTS values than the other CSC systems with and without gelatin, with the exception of the S50C50 control.
The values of the mean elastic moduli obtained during the measurement of diametral tensile strength are also shown in Figure 11 along with the respective standard deviations. The elasticmodulioftheCSC systemswith and withoutgelatin in therangeof26.8−43.3 MPa presented significant differences (P < 0.05). It was found that the S50C50 control had an insignificantly higher modulus than the other three CSC controls (P > 0.05). The modulus decreased somewhat after the incorporation of either 5% or 10% gelatin to the CSC control, but there was not significantly different (P > 0.05) for the same CSC system after Scheffe’s
multiple comparison testing.