Quantum Dynamics in Phase space using the Biorthogonal von Neumann bases: Algorithmic Considerations

TL;DR

This paper introduces new iterative algorithms for quantum dynamics in phase space using biorthogonal von Neumann bases, enabling efficient, parallelizable solutions to Schrödinger equations with applications to tunneling and ionization.

Contribution

It presents novel, efficient, and parallelizable algorithms for quantum dynamics in phase space based on biorthogonal von Neumann bases, improving basis convergence and pruning methods.

Findings

Algorithms effectively solve time-dependent Schrödinger equations.

Demonstrated applications include double-well tunneling.

Showed efficiency in modeling double ionization of helium.

Abstract

The von Neumann lattice refers to a discrete basis of Gaussians located on a lattice in phase space. It provides an attractive approach for solving quantum mechanical problems, allowing the pruning of tensor-product basis sets using phase space considerations. In a series of recent articles Shimshovitz et al. [Phys. Rev. Lett. 109 7 (2012)], Takemoto et al. [Journal of Chemical Physics 137 1 (2012)] Machnes et al. [Journal of Chemical Physics, accepted (2016)]), we have introduced two key new elements into the method: a formalism for converging the basis and for efficient pruning by use of the biorthogonal basis. In this paper we review the key components of the theory and then present new, efficient and parallelizable iterative algorithms for solving the time-independent and time-dependent Schr\"odinger equations. The algorithms dynamically determine the active reduced basis iteratively without resorting to classical analogs. These algorithmic developments, combined with the previous formal developments, allow quantum dynamics to be performed directly and economically in phase space. We provide two illustrative examples: double-well tunneling and double ionization of helium.