Quantum-DFT embedding algorithm for electronic structure calculations

Abstract

In the near future, material discovery and drug design will be aided by quantum computer assisted simulations. These have the potential to target chemical systems intractable for the most powerful classical computers. However, the resources offered by near term quantum computers are still limited. Thus, it is necessary to investigate hybrid quantum-classical computational schemes, like the embedding algorithm we recently proposed. In this work, we rapidly scaled up quantum- assisted simulations of molecular systems by means of embedding the quantum electronic structure calculation into a classically computed environment at the DFT level of theory. This allowed us to obtain significant energy corrections for a number of simple molecules, ranging from water to oxirane, in their strongly correlated limit (in the dissociation limit). Applications of this embedding method will have significant impact, as it paves the road towards quantum-assisted studies of more complex molecular systems in the near future. In this talk, I will present an extension of our work, where we interface the open source framework for quantum computing, Qiskit, with the highly parallelized classical code CP2K. The seamless integration of these two codes has enabled us to investigate significantly larger systems without being limited to comparatively small basis sets or lower quality exchange–correlation functionals. In particular, I will present the results of calculations of the ground state spin states of organo-metallic compounds, which are notoriously difficult for DFT.