Can the electron density be interpreted information-theoretically? A critical analysis using quantum information theory

TL;DR

This paper critically examines the validity of interpreting electron density as a probability distribution within quantum information theory, revealing fundamental inconsistencies and prompting a reevaluation of information-based measures in quantum chemistry.

Contribution

It demonstrates that common assumptions about electron density as a probability distribution lead to inconsistencies and proposes a new perspective using quantum information theory for conceptual quantum chemistry.

Findings

Electron densities cannot be reliably interpreted as probability distributions.

Normalization of electron densities introduces significant theoretical inconsistencies.

Quantum information theory reveals the need to rethink information measures in quantum chemistry.

Abstract

Many quantum chemical similarity measures have been derived and substantiated by applying concepts and quantities from information theory to the electron density. To justify the use of information theory, the electron density is usually equated to a probability distribution, despite the fact that such an assumption can give rise to inconsistencies such as negative Kullback-Leibler divergences. In this work we show, using quantum information theory, that both interpreting electron densities as probability distributions as well as any pragmatic normalization thereof gives rise to far-reaching information theoretical inconsistencies. By applying the theory of open quantum subsystems to quantum chemical partitions, we show using thought experiments that many statements concerning the information content of atoms in molecules need to be reconsidered. This study represents a step towards generalizing and reformulating the information theoretical approach to conceptual quantum chemistry beyond its classical perspective.