Molecular Dynamics Simulations of Disordered Materials by Carlo Massobrio Jincheng Du Marco Bernasconi & Philip S. Salmon

Author:Carlo Massobrio, Jincheng Du, Marco Bernasconi & Philip S. Salmon
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

The simple visual inspection of the models can provide some clues regarding these effects. For instance, Fig. 10.4 highlights a more uniform distribution of all species in the most bioactive composition (45S) compared to a higher-silica (bioinactive 65S) and a higher-phosphate (45S-P12) one [51, 86]. In particular, modifier cations are homogeneously spread across all the available space in 45S, whose silicate network does not contain large gaps, although some small gaps are visible that appear to be populated by small calcium phosphate aggregates. Larger gaps appear in the silicate network of 65S, that the figure suggests as associated with calcium phosphates: modifier cations also appear less uniformly spread in 65 S than in 45S. A higher-phosphate content (right panels in the figure) determines the appearance of large voids in the silicate network, mainly filled with calcium and phosphates, whereas again the sodium distribution appears more uniform compared to calcium [86].

These visual clues, albeit useful, clearly need to be assessed more quantitatively. The ratio R between the M(odifier)–M(odifier) coordination number extracted from the MD model and that expected from a uniform distribution of M cations through the available space represents a reliable measure of the extent of clustering [51, 77, 87]: R 1 denotes M–M clustering, more marked for larger deviations from the uniform R 1 distribution. Moreover, the ratio R calculated in the same way from the A–B coordination number is a useful measure of preferential aggregation between A and B species, for instance with A Na/Ca and B Si/P one can assess whether the Ca-P aggregation suggested by the figures is indeed stronger than Na-P. This analysis applied to the MD models shows that, in general, modifier cations (especially Ca) have a marked preference to associate with phosphate groups [51], which reflects the well-known repolymerization of the silicate network when modifier cations are stripped from it and shifted to the phosphates [88]. The R analysis of the MD models shows that the preference of Ca to coordinate phosphate tetrahedra is further enhanced moving to higher-silica and less bioactive compositions; the latter glasses are also characterised by a significantly less uniform distribution of Ca ions [86].

When the ions are confined into clusters spatially separated from each other, ion migration along channels connecting these clusters is inhibited; however, when the size of the clusters increases, their mutual separation will be reduced and fast-conducting migration channels may eventually be re-established, which may explain why in some cases phase separation of modifier-rich phases can lead to enhanced conductivity [89]. The size of the clusters, determined by the glass composition, is thus the key factor to take into account in order to establish whether nanosegregation may inhibit or enhance leaching of the ions in the surrounding environment, and, essentially, biodegradation [90].

Whereas the network connectivity reflects the average strength of the network of Si–O bonds building up the glass matrix, different structural descriptors can be devised to describe the strength of noncovalent interactions between modifier ions and the silicate matrix. Several studies have highlighted that

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.