While there is much interest in using fullerenes for medical applications (e.g. anti-cancer treatment) they have limitations, notably they cluster together and have poor water solubility. Heterofullerenes may make good replacements (fullerenes with one or more carbon atoms replaced by nitrogen or phosphorus). In this current work we calculated their response to oxidation, a crucial property for future medical use, as well as for other applications such as solar cells. While nitrogen doped fullerenes oxidize similarly to normal C60, phosphorus doping changes the oxidation behaviour (the phosphorus atom itself traps oxygen), and this ultimately changes the fundamental electronic behaviour of the fullerene.

Our original motivation for the work was curiosity; primarily to find out why it seems so difficult to produce phosphorus doped fullerenes. The oxidation behaviour seems to at least partly answer that question, namely that phosphofullerenes will likely become immediately oxidized in air. It remains a major open challenge to produce phosphofullerenes, or indeed their oxide forms as discussed in the paper. We hope that our work will stimulate the field and encourage the experimental testing of heterofullerenes for medical applications.