Seven magnesium-containing aluminoborosilicate glasses, with three to five oxides, have been studied through comprehensive multi-nuclear solid-state NMR ( 11 B, 27 Al, 29 Si, 23 Na, 17 O and 25 Mg) and Raman spectroscopy. The progressive addition of cations and the substitution of sodium and calcium by magnesium illuminate the impact of magnesium on the glass structure. The proportion of tri-coordinated boron drastically increased with magnesium addition, demonstrating the poor chargecompensating capabilities of magnesium in tetrahedral boron units. Oxygen-17 NMR showed the formation of mixing sites containing both Na and Mg near non-bridging oxygen sites. Furthermore, a high magnesium content appears to result in the formation of two sub-networks (boron and silicon rich) with different polymerisation degrees as well as to promote the formation of high-coordination aluminium sites (Al[V] and Al [VI]). Finally, magnesium coordination ranging from four to six, with a mean value shifting from five to six along the series, suggests that magnesium might endorse an intermediate role in these glasses.

Transitioning beyond a trial-and-error based approach for the compositional design of next-generation borosilicate-based bioactive glasses requires a fundamental understanding of the underlying compositional and structural drivers controlling their degradation and ion release in vitro and in vivo. Accordingly, the present work combines magic-angle spinning (MAS) NMR techniques, MD simulations, and DFT calculations based on GIPAW and PAW algorithms, to build a comprehensive model describing the short-to-medium-range structure of potentially bioactive glasses in the Na 2 O−P 2 O 5 −B 2 O 3 −SiO 2 system over a broad compositional space. P 2 O 5 preferentially tends to attract network modifier species, thus resulting in a repolymerization of the silicate network and a restructuring of the borate component. 11 B{ 31 P} and 31 P{ 11 B} dipolar recoupling experiments suggest that the ability of glasses to incorporate P 2 O 5 without phase separation is related to the formation of P−O−B(IV) linkages integrated into the borosilicate glass network. An analogous approach is used for elucidating the local environments of the Na + network modifiers. This work, along with future studies aimed at elucidating composition−structure−solubility/bioactivity relationships, will lay the foundation for the development of quantitative structure−property relationship (QSPR) models, thus representing a leap forward in the design of functional borosilicate bioactive glasses with controlled ionic release behavior.