Aluminosilicate glasses are ubiquitous in high-performance displays due to their favorable thermal, mechanical, and optical properties. They also exhibit interesting structural features depending on the ratio of alumina to modifiers in the glass system. Excess modifiers exist in the metaluminous region, while the peraluminous region contains more negatively charged alumina structures than modifiers. As the composition switches from metaluminous to peraluminous, anomalous changes in properties such as the glass transition temperature, viscosity, and refractive index occur. This has been explained with two contrasting structural transformations to accommodate the lack of charge-balancing modifiers: either aluminum increases in coordination (forming five-coordinated or six-coordinated Al) and/or oxygens become three-coordinated (known as triclusters). The precise charge-balancing mechanism remains a subject of much debate in the community. This review highlights this structural debate by providing a chronological understanding of how these two theories evolved. We also summarize the state-of-the-art understanding of the aluminosilicate glass structure. By gaining a more comprehensive view of the two opposing structural views within the aluminosilicate glass system, we can gain insights on valuable future research from both experimental and modeling perspectives.

The mixed alkali effect, the deviation from expected linear property changes when alkali ions are mixed in a glass, remains a point of contention in the glass community. While several earlier models have been proposed to explain mixed alkali effects on ionic motion, models based on or containing discussion of structural aspects of mixed-alkali glasses remain rare by comparison. However, the transition-range viscosity depression effect is many orders in magnitude for mixed-alkali glasses, and the original observation of the effect (then known as the Thermometer Effect) concerned the highly anomalous temperature dependence of stress and structural relaxation time constants. With this in mind, a new structural model based on topological constraint theory is proposed herein which elucidates the origin of the mixed alkali effect as a consequence of network strain due to differing cation radii. Discussion of literature models and data alongside new molecular dynamics simulations and experimental data are presented in support of the model, with good agreement.

This is the author manuscript accepted for publication and has undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as

Calcium aluminosilicate (CAS) glass systems are industrially significant due to their favorable optical, mechanical, and thermal properties, as well as serving as a basis for many glasses in nuclear waste confinement and alkali-free display substrates. [1][2][3][4] The properties which make them industrially desirable are closely linked to their complex structure. In general, the glass system is comprised of network forming silica and alumina units, both ideally in tetrahedral arrangement with four bonded oxygens. Bonded oxygens which bridge to neighboring network formers are appropriately called bridging oxygens (BOs). An interesting feature of the CAS glass system occurs when varying the ratio, R, of [Al 2 O 3 ]/[CaO]. When R < 1 (i.e., percalcic regime), calcium cations act as charge compensators for the negatively charged (AlO 4/2 ) − tetrahedra. Excess calcium cations decrease network connectivity, resulting in non-bridging oxygens (NBOs). At R = 1, previous studies consider the network to be fully connected with no NBOs. 5 However, multiple studies have revealed that a finite concentration of NBOs exist at this ratio. 6,7 When R > 1 (i.e., peraluminous systems), the [AlO 4/2 ] − units are in excess, with insufficient calcium cations available for charge balancing. Two mechanisms have been proposed to compensate for the insufficient population of modifier cations: the formation of highly coordinated alumina (i.e., [5] Al and/or [6] Al), [8][9][10][11] or the formation of three-bonded oxygens (TBO), also known as triclusters. 12 While triclusters are present in crystalline polymorphs, 12,13 verification of triclusters in glassy systems has been limited spectroscopically, and hence molecular dynamics (MD) has been the primary

Ionic conductivity is a property of rapidly increasing interest. Various models attempting to explain ionic conductivity of glass systems have shown limited agreement with experimental results; however, none have been comprehensive. By using molecular dynamics simulations, the diffusion of ion species through a network can be directly observed, providing insights into the mechanisms and their relation to ionic conductivity models. In this report, a method of utilizing molecular dynamics simulations is proposed for the study of the ionic mobility of Na, Li, and K ions in binary silicate glasses. Values found for glasses with x = 0.1, x = 0.2, and x = 0.3 alkali content are between 10 −5 and 10 −4 cm 2 •s −1 •V −1 and did not change significantly with composition or temperature. This is in agreement with the interstitial pair and weak-electrolyte models used to explain ionic conductivity in glasses.