Predicting the bioactivity of glasses using the network connectivity or split network models

Bioactive glasses (BG) are used as bone substitutes and re-mineralising additives in toothpastes. They work by precipitating apatite on their surface, and the network connectivity (NC) and split network models can be used to predict their bioactivity, i.e. their ability to form apatite. NC predicts glass degradation and has been used successfully to predict the bioactivity of BG, it does not take into account their phosphate content. Our experimental data confirm predictions using the split network model by Edén [Journal of Non-Crystalline Solids 357 (2011) 1595–1602], that “as long as P remains predominantly as QP0 tetrahedra and the average silicate network-polymerisation is ‘favourable’, the bioactivity enhances monotonically for increasing phosphorus content of the BG”. Results show that phosphate plays a key role in bioactivity and apatite formation of BG. This can be explained by the fact that phosphorus does not form part of the silicate network, but instead forms a separate orthophosphate phase. However, NC and split network models are still useful approaches for predicting BG bioactivity and apatite formation, if care is exercised when applying the models to glasses that contain more components than simple SiO2–P2O5–CaO–Na2O systems.