**3.1 Morphology and microstructure of bioactive glass ceramics**

**Figure 2** shows the SEM image and EDS spectra of (a) 45S5 and (b) S53P4 3DOM-MBGCs. Bioactive glass with hierarchical porosity was formed through the PMMA and Pluronic P123 dual templating system. The SEM image of 53S4P (**Figure 2(b)**) shows more well ordered macroporous structure with spherical pores are around 300 nm. While, 45S5 shows distorted 3DOM structure (**Figure 2(a)**). The EDS spectrum of 45S5 and S53P4 in **Figure 2** shows the peaks corresponding to Si, Ca, Na and O that represent the preservation of the elements in the precursors without impurity elements. The FTIR spectra of 45S5 and S53P4 3DOM-MBGCs in **Figure 3** exhibits the characteristic peaks of Si-O-Si bending, symmetric and asymmetric stretching vibration at 467, 802 and 1086 cm−1, respectively. In the peak at 564 and 950 cm−1 corresponding to the P-O bending and stretching vibration, respectively. In addition, the peak at 1635 and 3450 cm−1 correlates to O-H bonds indicates the water trapped inside the sample. The narrow band near 1384 cm−1 indicates the characteristic of the carbonate group (CO3 2−) [35]. **Figure 4** indicates that all the BET curves of the 45S5 and S53P4 MBGCs presented a type IV isotherms

**Figure 2.** *SEM and EDS spectra of (a) 45S5 and (b) S53P4 3DOM-MBGCs.*

*Three-Dimensionally Ordered Macroporous-Mesoporous Bioactive Glass Ceramics for Drug… DOI: http://dx.doi.org/10.5772/intechopen.95290*

### **Figure 3.**

*FTIR spectra of (a) 45S5 and (b) S53P4 3DOM-MBGCs. Copyright [30].*

pattern with type H4 hysteresis loops, characterizing mesoporous materials with narrow slit-like pores, with internal voids of irregular shape and broad size distribution [36]. This is confirmed by the average data of specific surface area, pore volume and pore diameter of 45S5 and S53P4 3DOM-MBGCs listed in **Table 2**. The bioactive glasses show specific surface area in range of 96.54 to 116.76 m<sup>2</sup> /g. The 45S5 glasses shows a relatively wide pore size distribution calculated from the adsorption branch using the BJH model, and the average pore size is around 15.158 nm, while the S53P4 glasses gained average pore size around 11.230 nm.

### **3.2 Assessment of** *in vitro* **bioactivity test**

The *in vitro* bioactivity of 3DOM-MBGCs was tested at body temperature of 37°C by using the SBF solution whose composition and ionic concentration similar to human blood plasma. **Figure 5** shows the SEM images of 45S5 bioactive glasses having different soaking time in SBF solutions. Compared with the morphology of the prepared bioactive glasses in **Figure 5a**, the nucleation of hydroxyapatite occurred on the glass surfaces after soaking in the SBF solution for 2 days (**Figure 5c**). The surface of 45S5 glasses were covered by precipitation of apatite-like layer more than 3 days soaking in SBF solution (**Figure 5d**-**f**). **Figure 6**, the formation of hydroxyapatite-like on the surface of S53P4 glasses started after 3 days of immersion in SBF solution. Within 7 days, most of the glass surfaces were covered by the apatite-like layer (**Figure 6d**). 45S5 showed fast hydroxyapatite-like precipitation than S53P4. However, the hydroxyapatite formation depends on the incorporation of Ca2+ and PO4 3− on the MBG's glass surfaces during bioactivity test. Lower SiO2 and higher CaO, P2O5 content in 45S5 could amplify the rate of hydroxyapatite formation on MBG's glass surfaces. The chemical composition and the microstructural morphology of 3DOM-MBGCs directly related to their bioactivity. The SBF can easier penetrate the larger macropores in the 3DOM bioactive glass compared to the mesopores in bioactive glasses [37]. Therefore, the minor difference average surface area and pore size of 3DOM-MBGCs (**Table 3**) has no effect in determining apatite growth. Due to both 3DOM-MBGCs using the same size of PMMA spheres for macroporous and same surfactant for mesoporous.

### **Figure 4.**

*N2 adsorption–desorption isotherms and pore size distribution of (a) 45S5 and (b) S53P4 3DOM-MBGCs. Copyright [30].*


### **Table 2.**

*The average data of specific surface area, pore volume and pore diameter of 45S5 and S53P4 3DOM-MBGCs. Copyright [30].*

The FTIR spectra of 45S5 in **Figure 7**, at below spectrum, the sample before soaking in SBF solution exhibits the peaks at 467, 802 and 1086 cm−1 corresponding to the vibration of Si-O-Si bond, bending, symmetric and asymmetric stretching vibration, respectively. In vibrational peak at 564 and 950 cm−1 correlates to the P-O vibrational peak. In addition, the O-H bonds of the water trapped inside

*Three-Dimensionally Ordered Macroporous-Mesoporous Bioactive Glass Ceramics for Drug… DOI: http://dx.doi.org/10.5772/intechopen.95290*

### **Figure 5.**

*SEM images of 45S5 bioactive glass ceramics (a) before soaking in SBF solution and after soaking in SBF solution for (b) 1 day (c) 2 days (d) 3 days (e) 7 days and (f) 7 days with higher magnification. Copyright [30].*

### **Figure 6.**

*SEM images of S53P4 bioactive glass ceramics (a) before soaking in SBF solution and after soaking in SBF for (b) 1 day (c) 2 days (d) 3 days (e) 7 days and (f) 7 days with higher magnification. Copyright [30].*


### **Table 3.**

*The drug loading efficiency and content of 45S5 and S53P4.*

**Figure 7.** *FTIR spectra of 45S5 bioactive glass ceramics with different soaking time in SBF solution. Copyright [30].*

**Figure 8.** *FTIR spectra of S53P4 bioactive glass ceramics with different soaking time in SBF solution. Copyright [30].*

the sample was shown at 1635 and 3450 cm−1. The narrow band near 1384 cm−1 indicates the characteristic of the carbonate group (CO3 2−) [35]. After the soaking in SBF, all the characteristic peaks are still observed. The P-O peak at 564 splits into doublet peak at 586 and 564 cm−1 which normally appears after immersion of the bioactive glass in SBF solution15. All the bands corresponding to the P-O represent the formation of hydroxyapatite on the surface of MBGCs. **Figure 8** represents the FTIR spectra of S53P4 bioactive glasses. The sample before soaking in SBF solution shows peaks at 467, 1087 and the shoulder at 1087–1250 cm−1

*Three-Dimensionally Ordered Macroporous-Mesoporous Bioactive Glass Ceramics for Drug… DOI: http://dx.doi.org/10.5772/intechopen.95290*

correlates to the vibration of Si-O-Si bond. The peaks at 1385 and 1401 cm−1 indicates the characteristics of carbonate group (CO3 2−) [35]. In the peak around 576 and 966 cm−1 corresponding to the P-O bending and stretching vibration, respectively. In addition, the peak at 1631 and 3445 cm−1 correlated to O-H bonds. After S53P4 MBGCs were soaked in SBF for 1 day, the aforementioned vibrational peaks are still observed. The P-O peak at at 607 and 567 cm−1, which confirmed that the formation of amorphous phosphate phase on the glass surface [29]. Although, the splitting P-O peak of S53P4 appears in 1 day after soaking in SBF, while in the case of 45S5 after soaking for 2 days. However, the formation of hydroxyapatite-like on the surface of S53P4 glasses started after 3 days of soaking in SBF solution, slowly growth comparing with 45S5, indicated the better bioactivity of 45S5 than S53P4 bioactive glasses.

### **3.3** *In vitro* **study of drug release**

## *3.3.1 Drug loading*

The drug loading efficiency and drug loading content of both MBGCs are summarized in **Table 3**. The drug loading efficiency in 45S5 was 18.00 ± 3.16%, while S53P4 showed quite higher drug loading efficiency of about 22.14 ± 2.53%. However, the loading efficiency of S53P4 was not statistically different from that of 45S5 (independent t-test, p > 0.05). The drug loading content of the MBGCs was found to be 8.74 ± 1.09 wt% for 45S5 and 11.91 ± 2.09 wt% for S53P4 glasses. The significant difference was not observed for the drug loading content of 45S5 from S53P4 (independent t-test, p > 0.05). S53P4 provides high average drug-loading content compared with other inert carrier materials that generally have low drug-loading content (less than 10 wt%) [33]. The porous materials can be developed to fabricate high drug-loading carriers due to their promising intrinsic properties, such as large hollow interior, porous surface, high surface area and large pore volume [33]. The good drug-loading capacity obtained in this study could be related to high surface area of the carrier with porous structure as supported by the results obtained from N2 adsorption desorption analysis.

### *3.3.2 In vitro drug release*

The release profiles for gentamicin from the MBGCs to the PBS are represented in **Figure 9**. For both MBGCs, the release of gentamicin showed an initial fast release followed by a relatively slow subsequent release. An initial fast release of the antibiotic was observed during the first 24 hours of soaking, reaching the mean gentamicin release values of 34.53% (45S5) and 41.21% (S53P4). The subsequent release rate was quite low in comparison with the first period. However, the S53P4 bioactive glasses showed a higher initial drug release behavior than 45S5 glasses. But later, both bioactive glasses reached the same point after 96 hours of release values of 64.27% (45S5) and 64.53% (S53P4). Both S53P4 and 45S5 bioactive glasses showed a slowly continuous gentamicin release.

To study the mechanism of drug release from the MBGCs, the first 60% of gentamicin release profile was fitted in Peppas-Korsmeyer model. In this model, the value of n characterizes the release mechanism of drug as described in **Table 4**. As observed in **Table 4**, the n values of the release data of 45S5 and S53P4 glasses are 0.3992 and 0.3004, respectively. This indicated that the drug release from both systems can be described by Fickian diffusion [34]. Both MBGCs possessed porous structures with hollow interiors. The diffusion through channel might dominate the drug release from these mesoporous materials [13, 38].

### **Figure 9.**

*In vitro gentamicin release from 45S5 and S53P4 bioactive glass ceramics.*


### **Table 4.**

*Kinetic assessment of gentamicin release data of 45S5 and S53P4 in PBS (Peppas-Korsmeyer model).*
