Dipole moments from atomic-number-dependent potentials in analytic density-functional theory
Brett I. Dunlap (1), Shashi P. Karna (2), and Rajendra R. Zope (3);, ((1) US Naval Research Laboratory, Washington DC, (2) U.S. Army Research, Laboratory, Weapons, Materials Research Directorate, Maryland, (3), University of Texas at El Paso)

TL;DR
This paper investigates how element-dependent exchange potentials in analytic density-functional theory affect molecular dipole moments, comparing results with quantum-chemical methods and experiments, and discusses optimal parameter choices for accurate predictions.
Contribution
It introduces a method to improve dipole moment calculations in density-functional theory by fitting element-dependent exchange potentials and analyzes the impact of parameter variations.
Findings
Fitting the Kohn-Sham potential yields less than 0.1 Debye mean absolute error.
Varying alpha values has less impact on dipole moments than on energies.
Short-range electrostatic effects dominate when varying alpha differences across atoms.
Abstract
Molecular dipole moments of analytic density-functional theory are investigated. The effect of element-dependent exchange potentials on these moments are examined by comparison with conventional quantum-chemical methods and experiment for the subset of the extended G2 set of molecules that have nonzero dipole moment. Fitting the Kohn-Sham [Phys. Rev. 140, A1133 (1965)] potential itself makes a mean absolute error of less than 0.1 Debye. Variation of alpha (Slater's [Phys. Rev. 81, 385 (1951)] exchange parameter) values has far less effect on dipole moments than on energies. It is argued that in variable alpha methods one should choose the smaller of the two rather than the geometric mean of the two alpha values for the heteroatomic part of the linear-combination-atomic-orbital density. Calculations on the dipole moment of NH2(CH)24NO2 are consistent with earlier calculations and show…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
