Our research focuses on finding new solid-state inorganic materials, and techniques to discover and analyse them. Our goal is to combine experimental synthesis, structure and physical property characterisation, and data-driven computational approaches to make new materials with useful properties. A particular focus is on mixed-anion materials (particularly metal oxyfluorides) and the impact anionic composition can have on physical properties such as magnetism and electronic or ionic conductivity.
We use a range of approaches to discover new materials; in addition to traditional solid state chemistry, we incorporate methods from condensed matter physics, materials science, computer science and mathematics. Some of our current research interests are shown below.
New Oxyfluoride Materials
Key to our research is the synthesis of new materials. We use high-temperature ceramic methods, often within sealed quartz or metal ampoules to maintain precise oxidation states and stoichiometry. High temperatures (up to 1600 °C) are achieved using a range of furnaces, and syntheses can be performed under oxidising, reducing or inert gases.
In addition, we have the facilities and expertise to perform direct fluorination using F2 gas at elevated temperatures. High pressure syntheses are performed in collaboration with Prof. J. Paul Attfield in the Centre for Science at Extreme Conditions (CSEC).
Effects of Local Structure on Physical Properties
Increasingly it is being found that bulk (average) atomic structure (obtained through neutron or X-ray diffraction) does not fully explain physical properties; local structural correlations must also be considered. Particularly in the case of mixed-anion materials, local ordering of different anions gives an extra degree of freedom in controlling physical properties. As an example, single crystal X-ray diffraction from TiOF (see image and paper) shows diffuse streaks, due to short-range correlations of Ti position with O/F ordering. This influences the physical properties, suppressing any magnetic ordering or metal-insulator transitions.
The number of known crystal structures is continually increasing, and are gathered in databases such as the ICSD. We use this body of information to explore general structural features across inorganic materials, and to harness data science techniques to produce predictive tools. A key interest is in developing methods for describing structural motifs (such as quantifying distortion in coordination environments using ellipsoids) for use as data features in a machine-learning approach.
Our research is generously supported by: