Quantum Density Mechanics: Accurate, purely density-based \textit{ab initio} implementation of many-electron quantum mechanics

TL;DR

This paper introduces a novel purely density-based ab initio approach for calculating many-electron systems' energies and properties, eliminating the need for wave functions or orbitals, and demonstrating high accuracy and favorable scaling.

Contribution

It develops a density-mechanical framework that avoids orbitals and two-electron integrals, providing accurate energies and densities with linear scaling in electron number.

Findings

Achieves near-experimental energies for 2-electron atoms

Provides accurate energies for 3-5 electron atoms

Scales linearly with the number of electrons

Abstract

This paper derives and demonstrates a new, purely density-based ab initio approach for calculation of the energies and properties of many-electron systems. It is based upon the discovery of relationships that govern the "mechanics" of the electron density -- i.e., relations that connect its behaviors at different points in space. Unlike wave mechanics or prior electron-density-based implementations, such as DFT, this density-mechanical implementation of quantum mechanics involves no many-electron or one-electron wave functions (i.e., orbitals). Thus, there is no need to calculate exchange energies, because there are no orbitals to permute or "exchange" within two-electron integrals used to calculate electron-electron repulsion energies. In practice, exchange does not exist within quantum density mechanics. In fact, no two-electron integrals need be calculated at all, beyond a single coulomb integral for the 2-electron system. Instead, a "radius expansion method" is introduced that permits determination of the two-electron interaction for an N-electron system from one with (N-1)-electrons. Also, the method does not rely upon a Schrodinger-like equation or the variational method for determination of accurate energies and densities. Rather, the above-described results follow from the derivation and solution of a "governing equation" for each number of electrons to obtain a screening relation that connects the behavior at the "tail" of a one-electron density, to that at the Bohr radius. Solution of these equations produces simple expressions that deliver a total energy for a 2-electron atom that is nearly identical to the experimental value, plus accurate energies for neutral 3, 4, and 5-electron atoms, along with accurate one-electron densities of these atoms. Further, these methods scale in complexity only as N, not as a power of N, as do most other accurate many-electron methods.