Structures of crystalline materials play a crucial role in determining their properties, leading to advancements in various technologies. Crystal structure prediction is a key aspect of developing innovative functional materials. Researchers have devised efficient methods to identify structural minima on the potential energy surface. The University of Liverpool has introduced a mathematical approach that can accurately predict the structure of any material based on its atomic composition.
The algorithm, created by a collaborative team from the Departments of Chemistry and Computer Science at the University of Liverpool, evaluates entire sets of potential structures simultaneously, accelerating the process of finding the correct solution.
This breakthrough enables the discovery of producible materials and, in many cases, the anticipation of their characteristics. The new method has been tested on quantum computers, which have the capability to solve complex problems faster than classical computers, resulting in expedited calculations.
The demand for new materials to meet the net zero challenge, such as batteries, solar absorbers for clean energy, energy-efficient computing, and environmentally friendly catalysts for polymers and chemicals, is escalating for a sustainable future.
The search for new materials is challenging and time-consuming due to the multitude of ways atoms can bond to form different materials. Predicting the structure of unknown materials poses a significant scientific hurdle.
Professor Matt Rosseinsky, from the University’s Department of Chemistry and Materials Innovation Factory, stated: “The ability to accurately predict crystal structures now allows us to pinpoint exactly which materials can be synthesized and the structures they will assume, providing a foundation for future technologies.”
He added, “This new tool enables us to leverage widely available chemical elements to develop materials that can replace those based on scarce or toxic elements and surpass existing materials, addressing the challenges of a sustainable society.”
According to Professor Paul Spirakis of the University’s Department of Computer Science, the crystal structure prediction technique can be applied to various structures. The ongoing research aims to explore and implement more algorithmic concepts to discover novel and valuable materials in collaboration with chemists and computer scientists.
The potential energy surface landscape, with its peaks and valleys, corresponds to the atomistic arrangement of the garnet crystal. Advanced algorithms and quantum computers can pinpoint the lowest point on the surface, revealing the optimal garnet structure.
The study received funding from The Royal Society and Leverhulme Trust.
Journal Reference:
- Gusev, V.V., Adamson, D., Deligkas, A. et al. Optimality guarantees for crystal structure prediction. Nature. DOI: 10.1038/s41586-023-06071-y
