Predictive computational methods for revealing the properties of hypothetical compounds will help determine whether developing preparation methods are worthwhile, say researchers in the US. 

Roald Hoffmann and colleagues at Cornell University aimed to develop new metal hydride superconducting materials that perform better than existing ones. They studied the currently unisolated ternary hydrides of ytterbium and thulium as possible candidates. 

Hoffmann used computations based on density functional theory to show that the series of lanthanide hydrides are likely to have superconducting properties when pressure is applied and/or chemical doping used. He also predicted the compounds' electronic and crystal structures; concluding that they will mostly likely be formed under very high pressure conditions, be energetically unstable, and decompose readily into their known binary hydride equivalents.

The group are already working on the high pressure synthesis of similar materials and think that it is only a matter of time before these compounds are made. According to Hoffmann, finding ways to quench the compounds under normal atmospheric pressure and effectively protecting them from oxygen and water to prevent their decomposition is a key factor in their development for real applications.

Other published reports on designing families of superconductors in such an explicit way are, said Hoffmann, extremely rare.

David J Parker