The Hakubi Project

In condensed matter physics, ‘strong electron correlation’ is a critically important concept and brings about many remarkable phenomena such as super conductivity. In contrast, it has been overlooked in chemistry until recently because most organic molecules are well described by mean-field picture and the molecular orbital theory has enjoyed great success over the decades. But now, the importance of strong correlation effect in chemical reactions has begun to be understood due to the connection with the exceptional catalytic ability of transition metal atoms in metalloenzymes. The biggest obstacle in studying strong correlation arises from non-trivial exponential complexity of strongly interacting electrons, which is similar to the difficulty in solving the Hubbard-model and called as ‘multireference’ problem in quantum chemistry. To solve this remaining problem in quantum chemistry, we utilized the density-matrix renormalization group (DMRG) theory, which can efficiently compress the exponentially growing complexity of the wavefunction by encoding for the special entanglement structure that is present in low-lying physical quantum states. As a result of our successful application of the DMRG theory to the multireference problem, much effort has been made to extend the methods of condensed matter physics, such as quantum Monte Carlo, dynamics mean-field theory, and tensor network theory, to the quantum chemical problems.