ABSTRACT Magnesium and zinc are important additives or substituents in melt‐quenched Na 2 O–CaO–SiO 2 –P 2 O 5 bioactive glasses, due to their beneficial influence on glass stability and processability. They modify the dissolution kinetics of these glasses and impart favorable biological properties and functions to them. For rational bioactive glass design, it is essential to understand the local environments and distributions of these ions. While diffraction studies suggest MgO and ZnO to be close‐to‐four‐coordinate in silicate glasses, their structural roles in the network are still under debate. Traditionally these oxides have been viewed as network modifiers, converting Si–O–Si linkages to anionic non‐bridging oxygen atoms, whose charges are compensated by the divalent cations. It has been suggested that MgO 4/2 | 2– or ZnO 4/2 | 2– network‐forming units (NFUs) may be present, forming Mg–O–Si or Zn–O–Si linkages. If such units are formed, they would attract modifier cations for charge compensation, which would in turn result in an increased degree of polymerization of the silicate species, the main network‐forming component. This study explores the roles of MgO and ZnO in bioactive glasses with approximate composition 50SiO 2 –(50– x )MO,M’ 2 O– x P 2 O 5 in mol% (2 ≤ x ≤ 6), where M = Ca, Sr, Mg, Zn; M’ = Na, K. Quantitative estimates of the various silicate and phosphate NFUs were obtained from 29 Si and 31 P solid‐state nuclear magnetic resonance (NMR) spectroscopic techniques and molecular dynamics (MD) simulations, allowing the determination of the degree of network polymerization in terms of the average Si and P connectivities (‐values). Both NMR spectroscopy and MD studies consistently revealed that the extent of Si polymerization increases, as expected, with P 2 O 5 addition, reflecting the well‐documented preferential cation attraction by the phosphate species. On the other hand, data obtained from a set of comparative samples containing either Mg, Zn, or a mixture of both showed no significant changes in the degree of silicate network polymerization. This result strongly supports a network‐modifying role of both magnesium and zinc oxide, as suggested by previous works.
Bradtmüller et al. (Sun,) studied this question.