New Horizons in Glass Structure Prediction and Mechanics

Project Information

NewGLASS

Grant agreement ID: 101044664

DOI

10.3030/101044664

EC signature date 22 April 2022

Start date 1 September 2022

End date 31 August 2027

Funded under

• European Research Council (ERC)

Total cost

€ 1 996 935,00

EU contribution

€ 1 996 935,00

Coordinated by

AALBORG UNIVERSITET
Denmark

Periodic Reporting for period 1 - NewGLASS (New Horizons in Glass Structure Prediction and Mechanics)

Reporting period: 2022-09-01 to 2025-02-28

Summary of the context and overall objectives of the project

Oxide glasses are one of the most important material families owing to their unique features, such as transparency, tunable properties, and formability. Emerging solutions to major global challenges related to energy, health, and electronics require new scientific breakthroughs in glass chemistry, mechanics, and processing. The realization of these goals is severely restricted by the main drawback of glass, namely high brittleness. Furthermore, new glass compositions are today developed through time-consuming trial-and-error experimentation due to their inherent non-equilibrium nature and disordered structure.

A major task is therefore to initiate a paradigm shift within the field of glass science and technology, going from empirical to model-based approaches for the design of new glass compositions and microstructures with improved fracture resistance. This requires the development of computational approaches, from ab initio calculations to artificial intelligence, to integrate structural descriptors and glass chemistry with advanced processing and mechanical properties into holistic tools.

NewGLASS challenges the current glass design strategies in order to create such tools. For this purpose, an interdisciplinary approach is proposed, in which structural descriptors at the short- and medium-range length scales are first identified and quantified based on emergent statistical mechanics and persistent homology techniques. Guided by these results, high-throughput simulations at various length scales are combined with machine learning algorithms to design novel glass compositions, tailored deformation mechanisms, and 3D-printed microstructures to achieve superior fracture resistance. By having experiments and modelling complement and advance each other reciprocally, NewGLASS will find order in disorder and provide the scientific breakthroughs for the accelerated design of glasses with outstanding mechanical performance, thus opening up for many new applications.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

1. Taking binary silicate glasses as an example, we have compared the medium-range order (MRO) structure as determined using persistent homology and classical ring analysis. While the latter only identifies chemically bonded rings, the former captures both chemically and non-chemically bonded ring/loop structures. We show that the covalently bonded loops can be directly extracted using persistent homology by ignoring the modifiers from the analysis. We also demonstrate that although the chemically bonded rings contribute to the MRO, non-bonded MRO features also need to be considered.

2. Zeolitic imidazolate frameworks (ZIFs) feature complex structures that influence their mechanical properties. We have trained a deep learning-based force field and use it to propose a new structural descriptor, namely, the ring orientation index, to capture the propensities for phase transitions. The outcomes of this work are useful for studying MRO structural changes in various metal-organic framework (MOFs) and may thus guide their property tuning.

3. Germanate glasses often feature non-monotonic variations in mechanical properties with varying chemical composition, temperature, and pressure. We have found that upon hot compression, shear modulus of sodium germanate glasses features a surprising non-monotonic variation upon increasing pressure. We clarify the structural origin of this observation through high-energy X-ray and neutron total scattering coupled with ab initio molecular dynamics simulations as input for Reverse Monte Carlo modeling. While only very minor changes in Ge-O coordination are observed, the shear modulus trend is attributed to decrease in edge-sharing with pressure.

4. We have discovered that upon hydrothermal treatment of ZIF-62 glass, both the melting and glass transition temperatures of ZIF-62 glass decrease remarkably (by >100 °C), and simultaneously, hardness and Young’s modulus increase by up to 100%. Structural analyses suggest water to partially coordinate to Zn in the form of a hydroxide ion by replacing a bridging imidazolate-based linker.

5. We have created interconnected structures in borosilicate glass through spinodal decomposition. Interestingly, this leads to improvements in Vickers hardness, crack initiation resistance, and fracture toughness. The interconnected glassy phases deflect the propagating cracks, causing the required energy for cracks to cross phase boundaries to increase when subjected to external stress.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

1. We have proposed and validated a strategy for enabling continuous composition-structure-property modification of MOF glasses. Specifically, we show that ZIFs can be co-melted with water or heterocycle-based halide salts to prepare modified ZIF glasses. The water/ZIF or salt/ZIF mixing ratio can be used to control the material properties, including viscosity (melting and glass transition temperatures) and mechanics (hardness and crack resistance), without the need to prepare stoichiometrically pure crystals. We elucidate the modification mechanism through detailed structural analyses, finding that upon co-melting, the imidazolate-based linkers are partially exchanged for hydroxide or halide ions, effectively driving the formation of ligands only bonding to one metal-center. In turn, this reduces the connectivity of the glass network (akin to the network modification in oxide glasses) and thus the monotonic decrease in melting and glass transition temperatures with increasing hydroxide or salt content. We envision that our proposed approach is not restricted to ZIFs but will enable continuous composition tuning of a wide range of MOF glasses. This work thus creates a new path to diversify MOF glass compositions.

2. We have developed methodologies for in-situ studying the indentation deformation response of oxide glasses using x-ray nano-diffraction at synchrotrons. Such characterization has typically been performed after indentation due to difficulties in high-resolution structural measurements under an indenter. Two-dimensional mapping of the diffraction pattern in the zone below a sharp wedge indenter reveals the local changes in the atomic structure and density, as well as cracking. The in-situ experiments provide information at different stages of the indentation process, showing the formation and evolution of the induced densification zone and different types of cracking with resolution on the nanoscale. From the information on the changes in the local atomic structure, we gain experimental insights into the indentation response of oxide glasses and the interplay between changes in local atomic structure, local densification, and cracking. Shedding light on this interplay provides fundamental knowledge on the underlying mechanisms causing crack initiation in oxide glasses and are crucial for bridging the current knowledge-gap between experiments, theories, and simulations of mechanics in glass materials.