Exchange constants in molecule-based magnets derived from density functional methods

TL;DR

This study compares two density functional theory methods to calculate exchange constants in a molecule-based magnet, highlighting challenges in accurately determining small inter-chain couplings due to computational limitations.

Contribution

It demonstrates the limitations of current DFT approaches in accurately computing small exchange interactions in molecular magnets.

Findings

Intrachain coupling constants match experimental values.

Interchain couplings are below the resolution limit of the methods.

Computational limitations affect the accuracy of small exchange constant calculations.

Abstract

Cu(pyz)(NO3)2 is a quasi one-dimensional molecular antiferromagnet that exhibits three dimensional long-range magnetic order below TN=110 mK due to the presence of weak inter-chain exchange couplings. Here we compare calculations of the three largest exchange coupling constants in this system using two techniques based on plane-wave basis-set density functional theory: (i) a dimer fragment approach and (ii) an approach using periodic boundary conditions. The calculated values of the large intrachain coupling constant are found to be consistent with experiment, showing the expected level of variation between different techniques and implementations. However, the interchain coupling constants are found to be smaller than the current limits on the resolution of the calculations. This is due to the computational limitations on convergence of absolute energy differences with respect to basis set, which are larger than the inter-chain couplings themselves. Our results imply that errors resulting from such limitations are inherent in the evaluation of small exchange constants in systems of this sort, and that many previously reported results should therefore be treated with caution.