Can a simple topological-constraints-based model predict the initial dissolution rate of borosilicate and aluminosilicate glasses?

Abstract

Tuning glass composition to obtain targeted properties generally relies on empirical approaches. However, a deep understanding of the physical and chemical mechanisms linking glass composition to its structure and properties would enable developing reliable predictive models. Indeed, although empirical models are usually able to interpolate composition-property relationships within a given compositional envelope, they often fail at extrapolating predictions far from their training domain. Here, as an alternative route to empirical models, we show that a structural descriptor based on the number of topological constraints per atom can be used to predict the initial dissolution rate of aluminosilicate and borosilicate glasses after being parameterized on different families of glasses (specific series of borosilicate glasses). Sixteen glasses belonging to these families were studied and their initial dissolution rates were determined at 90°C and pH 90°C = 9, covering rates spanning over 5 orders of magnitude. The model based on topological constraints was trained based on seven select borosilicate glasses (R 2 = 0.997) and used to predict the dissolution rate of nine additional borosilicate and aluminosilicate glasses. We show that, provided that corrections are made for high alkali content glasses that dissolve incongruently (preferential release of Na), the model gives reasonable predictions, even far from its training domain.npj Materials Degradation (2020) 4:6 ; https://doi.

Predicting the dissolution rate of borosilicate glasses using QSPR analysis based on molecular dynamics simulations

“…A group of 20 borosilicate and aluminosilicate glasses with compositions listed in Table 1 have been studied. Glass sample preparation and characterization details were reported in our previous paper 18 . A brief introduction is presented here.…”

“…Liu et al 16 proposed a topology‐informed ML approach, which can overcome the limitations (e.g., unreliable extrapolations and violation of the laws of physics) of traditional ML, offering a promising route to predict the properties of materials in a robust way. By using the topological constraint theory (TCT), we have demonstrated that simple topological constraints enumeration based on glass composition can be used to correlate the initial dissolution rate ( r 0 ) and, based on which, it was able to predict the dissolution rate of a series of aluminosilicate and borosilicate glasses 18 . It was found that a few aluminosilicate glasses were outliers in the correlation analysis with TCT due to preferential dissolution; nevertheless, when using corrected glass compositions accounting for the fast leaching of alkali ions, the TCT model provided reasonable predictions of r 0 values for aluminosilicate glasses with high alkali content glasses after modifying the compositions 18 …”

“…In this paper, we performed the QSPR analysis of a series of twenty borosilicate, aluminosilicate and boroaluminosilicate glasses with r 0 18 based on atomistic structural models generated from MD simulations. A variety of structural descriptors including, F net mentioned above, have been evaluated to decide on their ability to predict the r 0 .…”

Analytical model of the network topology and rigidity of calcium aluminosilicate glasses

“…First, the angular BB constraints associated with four-fold Al atoms have been suggested by many studies to be notably weaker than those associated with Si atoms. 27,28,[69][70][71][72] This has been attributed to the fact that the nearby presence of a charge-compensating cation tends to destabilize the angular environment of four-fold coordinated Al atoms. 28 Second, ionic Ca-FO constraints are expected to exhibit a low bond energy and, hence, to break at low temperature.…”

“…This ranking is based on the following observations. First, the angular BB constraints associated with four‐fold Al atoms have been suggested by many studies to be notably weaker than those associated with Si atoms 27,28,69–72 . This has been attributed to the fact that the nearby presence of a charge‐compensating cation tends to destabilize the angular environment of four‐fold coordinated Al atoms 28 .…”