From chemical bonding to superconductor design

Superconductivity is one of the most exciting discoveries of the last century, with great interest in both, academic and technological environments.

These systems can conduct the electric current without energy losses and their commercial development would have an enormous societal impact, allowing large-scale production of important devices such as low-loss power cables, electric motors and high-speed quantum computers.Compressed hydrides bearing lanthanides and transition metals are especially promising materials to reach the goal of room-temperature superconductivity, especially after the recent discovery of superconductivity in LaH10 at 250 K.

Other relevant candidates are borides, which share many features with hydrides, as both elements are light metalloids that yield superconductivity at high-pressure and form superconducting compounds with characteristic cluster and ring patterns.

Unfortunately, the search is still based on extensive screenings along the periodic table and a framework that allows for the rational proposal of new candidates is missing. In this regard, theoretical chemistry is enormously helpful to make such predictions.

However, the calculations in the real (periodic) systems are very expensive as a consequence of the necessity of accounting for electron correlation and electron-phonon coupling, which are computationally demanding.In this context, we will build and study molecular models to bypass the aforementioned calculations and collect the features of superconductors.

This way, we will provide easy-to-compute bonding descriptors (e.g. the ELF, QTAIM …) for hydrides and borides.

Then, we will benchmark the models with periodic calculations to validate them, and we will propose new candidates for high-T superconductivity.

We will also focus on the role of anharmonic effects (crucial for superconductivity in LaH10).All in all, we expect to knock down the limitations in superconductors design by providing fast activity descriptors.