Abstract
Polyoxometalates (POMs), also known as molecular metal oxides, have attracted continuous interest since 1980s due to their unique properties and structural diversity. In general, POMs are redox active, stable in solution, and functional at the nanoscale. Computational chemistry calculations using density functional theory (DFT) to reproduce redox potentials have often been reported without accounting for the electrolyte environment and counterions. These calculations typically use implicit solvation models and compare theoretical results with previously published experimental data, which may not be synchronous. Herein, we employed DFT calculations to assess the accuracy of various exchange-correlation (x-c) functionals in reproducing experimental redox potentials for X5[M(H2O)PW11O39] salts; where X= Li, Na or K, and M = Mn(III/II), Fe(III/II) or Co(III/II). Our aim was to ensure that the experimental conditions and theoretical results are closely fit enabling a higher accuracy when reproducing experimental redox processes. Optimal accuracy was accomplished with PBE/TZP demonstrated by the lowest score for mean absolute error (MAE) of 0.55 V across all compounds in our work. Further analysis of MAE revealed increasing contributions to HF-exchange coincided with larger discrepancies from experimental potentials, U0Exp. The challenge in attaining accurate potentials is effectively controlling the proximity of the ion-pairing. To address this, we have explicitly specified the heteroatom-counterion, dP-X, geometries at discrete intervals (6-10 Å). Our proposed method permits an economical route for attaining accurate potentials, as opposed to necessitating CPU-expensive optimisations with hybrid x-c functionals.
Supplementary materials
Title
Supporting Information for Ion-Pair Proximity in Cyclic Voltammetry and DFT studies of Aqueous Transition-Metal-Substituted Polyoxotungstates
Description
Raw data, synthesis details and computational details for the work.
Actions