Fragment Molecular Orbital-based Variational Quantum Eigensolver for Quantum Chemistry in the Age of Quantum Computing

Abstract

Abstract Despite the potential of quantum computers in analyzing complex systems, their applications for larger systems are hindered by the limited qubit availability. Addressing this, our study presents the novel FMO/VQE algorithm that combines fragment molecular orbitals and variational quantum eigensolver methods, strategically designed for optimal qubit utilization in fragment-based quantum chemistry simulations. The FMO/VQE/UCCSD achieved an absolute error of just 0.053 mHa with 8 qubits in a H24 system using the STO-3G basis set, whereas traditional VQE methods require 48 qubits. In a H20 system with the 6-31G basis set, the FMO/VQE/UCCSD showed an error of 1.376 mHa with 16 qubits, as opposed to the 80 qubits required for the VQE/UCCSD. Consequently, the FMO/VQE not only outperforms in accuracy and scalability compared to conventional VQE, but also would facilitate quantum chemistry simulations of increasingly complex systems, as the field of quantum computing progresses and the availability of qubits expands.