Extension of Dynamical Variational Monte Carlo and its application for Fermi arcs

The theoretical and numerical treatment of strongly-correlated many-electron systems requires the continued development of high-accuracy many-body approaches. One recently developed method is dynamical Variational Monte Carlo (dVMC) [1], which enables the computation of the Green's function for large systems. In this work [2] we present a generalized dVMC technique that can serve as an impurity solver in quantum cluster methods. Using this new approach we perform a systematic study of the doped t-t'-t'' Hubbard model using cluster perturbation theory with a dVMC impurity solver. We reach system sizes unattainable by exact diagonalization solvers, and find robust evidence of the existence of Fermi arcs, a result of direct relevance to the cuprate superconductors.

[1] M. Charlebois and M. Imada, Physical Review X 10, 041023 (2020).

[2] P. Rosenberg et al. arXiv:2209.08092