Quantum embedding for condensed-phase reactivity

Computational modelling has a major and increasing role in modern chemistry. Grand challenges for the chemical sciences include cheap, clean and reliable synthesis of any molecule; harnessing the sun's energy for chemical change; and understanding the origins of biological complexity. For modellers the central challenge lies in elevating the status of existing methods so that they can cope with very large and complex molecules, and with processess that only take place over relatively long timescales.

Quantum embedding - the central idea in this proposed network - will allow us to focus highly accurate quantum mechanical descriptions of molecules at exactly the place where it matters: in the active sites of enzymes; at the point where photons are absorbed; or at the reactive sites on surfaces.

Bringing the required technologies together demands collaboration: and four leading research groups propose networking activities in the current proposal, three in the US and one in the UK.

Two primary deliverables have the potential to lead to impact: the theoretical methods developed, and software produced. Impact planning for the former is relatively straightforward, with the usual channels of dissemination serving well in this area. Delivering impact outside academia demands development and maintenance of high-quality software that (a) can utilize current and evolving computer architectures; and (b) can be used, and relied upon, by non-experts. The PI has an excellent track record in delivering on this, with successful projects being deployed in the Molpro software package. This is used in over 500 research institutions all over the world, and the PI is one of four lead developers (Werner, Knowles, Manby and Schütz). In this networking activity we will take significant steps towards disseminating such software tools, but more significant and lasting impact could be realized through a full EPSRC/NSF call in the area.