Open susilehtola opened 9 months ago
This is indeed something we have contemplated doing for a long time. The point of XCFun was automatic differentiation to go to high order response. This has so far not yet happened. But in the meantime the development of XCFun has more or less stagnated whereas LibXC is the de-facto standard. I am also involved, with a minor role, in a ITN application where MRChem will be used for benchmarks. The need to use more modern fuctionals in MRChem is very rellevant for that too!
Hi,
referring to J. Chem. Phys. 154, 214302 (2021)
it sounds like MRChem would greatly benefit from an interface to Libxc which offers a consistent, reusable set of functionals across >40 programs including ORCA. Would you be interested in such an interface?