Structural chemistry of the amorphous state
Amorphous materials – characterised by the absence of long-range structural order – are a current research frontier. We aim to create realistic models of amorphous structures and to understand the subtle rules that guide their formation.
Understanding amorphous structures with machine-learning-driven simulations
The emergence of ML-accelerated simulations has allowed for the study of amorphous materials in full atomistic detail and on realistic length scales. Our work covers a range of disordered systems, from “textbook” cases such as silicon or phosphorus, to chemically complex materials such as amorphous ices.
Recent highlights
- We revisited a long-standing debate on the structure of amorphous silicon, and showed that local signatures of crystallinity in the amorphous matrix are consistent with experimental observables (Nature Communications, 2025).
- Beyond elemental systems, we used a unified data-driven approach to study amorphous zeolitic imidazolate frameworks (a-ZIFs), invalidating the assumption that inorganic and hybrid glasses, like a-Si and a-ZIF, are topologically equivalent (arXiv, 2025).
- We examined the relevance of the traditional Zintl–Klemm concept in the diverse disordered Na–P phases that can form in sodium-ion battery anodes, providing atomic-scale insights into their structures and energetics (Angewandte Chemie, 2025).
What is next?
We envision the computationally-guided design of amorphous materials, wherein computational simulations will help inform experimental synthesis. Our future work will focus on elucidating the structure of other complex inorganic and hybrid glasses, and bridging the gap between computer simulations and experimental synthesis pathways.
