Boroaluminosilicate glasses – Effect of composition on mechanical properties

Boroaluminosilicate glasses are of significant interest for a variety of applications. More generally these glasses are interesting because both boron and alumina are in competition for charge compensating modifiers that allow them to adopt tetrahedral coordination within the glass network, which can be expected to affect overall glass properties Understanding composition-structure-property relationships in boroaluminosilicate glass is important to enabling property optimisation as a function of composition design. This project is aimed at increasing understanding of how composition affects the mechanical properties of boroaluminosilicate glasses with varying amounts of the three glass formers (SiO2, Al2O3 and B2O3) and different alkaline earth modifiers. Particular attention is focussed on the structural role of sodium as a charge compensator on the composition-structure-properties relationships in boroaluminosilicate glass. Essentially the network modifiers can be associated with SiO2, Al2O3 and B2O3 in different ways making it difficult to predict the mechanical properties of such glasses.
In this research, 10 different series of glasses have been synthesized. The series had varying silica/alumina ratios, boron/alumina ratios, alkaline earth/alumina ratios, alkaline earth/silica ratios or alkaline earth + alumina/silica ratios. The alkaline earths used were CaO or BaO. Variations in the hardness and fracture toughness of the glasses were measured using the indentation method and acoustic measurements were used to find their moduli. Differential thermal analysis was used to measure the glass transition temperature. Fourier Transform Infra-red spectroscopy (FTIR), Raman and Nuclear Magnetic Resonance (NMR) were used to investigate the base glass structure and the structural effects of adding different alkaline earth elements. It was found that the amount of Al2O3 used in the boroaluminosilicate glass significantly affects the properties: the glass transition temperature (Tg) and molar volume increased with increasing Al2O3 content in the series N7.6M1.1C3AXB7.6S80.7-X glasses. Meanwhile, observation for Tg to network charge balance for specific composition were notably in range of error bars when the network charge balance are > 0. However, Tg notably shows huge decreases when network charge balance between -21 to 0. The Raman peak at ~ 1000 cm–1 changes shape significantly for most of the glass series indicative of notable changes in the detailed structure of the glass including silica connectivity with composition. FTIR showed that the maximum peak shifted to lower wavenumbers (from 1047 to 1018 cm-1) with increasing amounts of Al2O3 and the intensity (absorption of IR) reduced which indicated that the polymerisation of glass is increased. Raman spectra were used to calculate a value of polymerisation index, specifically to see the connectivity of the glass structure in the boroaluminosilicate glass system. It was found that glasses that contain CaO had a higher polymerisation index compared to those that contained BaO and that higher Al2O3 contents gave higher polymerisation indices. This was also borne out by NMR studies on some of the glasses investigated.
In a composition-structure-properties relationship, from the overview of the correlations between the properties investigated, the network modifiers, formers and intermediates show there is a connection between most of the composition as shown in ternary diagrams and 3-D correlations. When the glass transition temperature increases the molar volume increases and the shear modulus decreases. Meanwhile, when correlating hardness, glass transition temperature and network former, it was found that increasing the network charge balance resulted in decreasing glass transition temperatures and increased hardness values. Overall it can be seen that increasing of network charge balance, the mechanical properties such as Young’s modulus, shear modulus, bulk modulus, Poisson’s ratio and hardness are increased and polymerisation index are decreased if the total network charge balance is > 0. Meanwhile, the value of Poisson’s ratio is ~0.24 for most of the compositions studied, and the hardness exhibited similar behaviour to Young’s modulus which increased with increasing network modifier. The composition-structure-mechanical properties can be controlled and designed from the various mixing ratio of different networks. However, the mixed network former in the composition leads to the non-linear trends in macroscopic properties as a function of composition design.