Cluster (spacecraft)

Chemistry

Metallurgy

Materials science

Programming language

Computer science

Thorium

Metal

Uranium

Nature

Metal–metal bonding is a widely studied area of chemistry1–3, and has become a mature field spanning numerous d transition metal and main group complexes4–7. By contrast, actinide–actinide bonding, which is predicted to be weak8, is currently restricted to spectroscopically detected gas-phase U2 and Th2 (refs. 9,10), U2H2 and U2H4 in frozen matrices at 6–7&nbsp;K (refs. 11,12), or fullerene-encapsulated U2 (ref. 13). Furthermore, attempts to prepare thorium–thorium bonds in frozen matrices have produced only ThHn (n = 1–4)14. Thus, there are no isolable actinide–actinide bonds under normal conditions. Computational investigations have explored the probable nature of actinide–actinide bonding15, concentrating on localized σ-, π-, and δ-bonding models paralleling d&nbsp;transition metal analogues, but predictions in relativistic regimes are challenging and have remained experimentally unverified. Here, we report thorium–thorium bonding in a crystalline cluster, prepared and isolated under normal experimental conditions. The cluster exhibits a diamagnetic, closed-shell singlet ground&nbsp;state with a valence-delocalized three-centre-two-electron σ-aromatic bond16,17 that is counter to the focus of previous theoretical predictions. The experimental discovery of actinide σ-aromatic bonding adds to main group and d&nbsp;transition metal analogues, extending delocalized σ-aromatic bonding to the heaviest elements in the periodic table and to principal quantum number six, and constitutes a new approach to elaborate actinide–actinide bonding. A crystalline cluster exhibits thorium–thorium bonding, adding to our knowledge of actinide–actinide bonding.

A crystalline tri-thorium cluster with σ-aromatic metal–metal bonding