Quantitative Modelling of Phosphate Clustering in Bioactive Glasses Through Solid-State NMR

Bioactive glasses are at the forefront of modern glass research due to their unique ability to promote the regeneration of hard and soft tissues upon contact with human plasma. This property makes them highly promising for healthcare applications. Within the framework of our recently funded Young Investigator project (Jovem Pesquisador), we aim to advance beyond the empirical search for new bioactive compositions by investigating the fundamental (micro)structural origins of bioactivity.

Our approach combines macroscopic property evaluation with advanced structural analysis using solid-state NMR spectroscopy and simulation-based methods. A central focus of the present work is on elucidating the role of phosphorus oxide – an additive typically present in low concentrations in silicate-based bioactive glasses – that significantly influences the material’s ability to form apatite layers in physiological conditions. Recent studies have indicated that phosphate components tend to form clusters within the glass matrix, a phenomenon that can be probed and quantified by advanced solid-state NMR.^1,2

In this study, we present a quantitative modeling approach based on ³¹P spin-echo decay (SED) NMR combined with a dipolar transform, which links experimentally observed spin-echo decay curves to radial distribution functions specific to ³¹P nuclei. We explore a range of clustering scenarios defined by three parameters: cluster radius, cluster concentration, and matrix phosphorus concentration. Simulated decay profiles are compared to experimental ³¹P SED data, allowing us to identify the most likely structural configurations. This method provides a novel tool for validating molecular models of phosphate distribution derived from molecular dynamics and Monte Carlo simulations. Furthermore, the dipolar transform framework we propose is readily extendable to other dipolar-based NMR experiments and is currently under active development in our lab. Overall, this approach paves the way for experimentally anchored structural validation of digital glass models and deepens our understanding of the structure–property relationships in bioactive glasses.