American Chemical Society
cm9023875_si_001.cif (1.47 kB)

Density Functional Theory Study of Ferrihydrite and Related Fe-Oxyhydroxides

Download (1.47 kB)
posted on 2009-12-22, 00:00 authored by Nathan Pinney, James D. Kubicki, Derek S. Middlemiss, Clare P. Grey, Dane Morgan
The atomic and magnetic structure and thermodynamic stability of ferrihydrite (Fe5O8H) are calculated based on the structure recently proposed by Michel et al. (Science 2007, 316, 1726). Ferrihydrite stability is compared with that of the Fe-oxyhydroxide polymorphs goethite (α-FeOOH) and lepidocrocite (γ-FeOOH) and the oxide hematite (α-Fe2O3). The GGA+U method is employed to correct known errors in treating correlated d-electrons in Fe atoms. GGA+U yields smaller errors in calculated thermodynamic quantities relative to experiment as compared with a standard GGA functional for all of the Fe-oxyhydroxides studied. Good agreement is obtained between the DFT-predicted and experimental ferrihydrite structure when the effects of varying crystallinity and particle size are taken into account. The magnetic properties of ferrihydrite are modeled using a Heisenberg model parametrized with DFT-based magnetic coupling constants. The groundstate magnetic ordering of bulk ferrihydrite is predicted to be ferrimagnetic, with the Fe-site spins ordering with alternating alignment in layers stacked along the c-direction in the crystallographic unit cell. The groundstate is predicted to disorder to a paramagnetic structure at TN = 250 K, undergoing a Néel transition. The enthalpy and Gibbs free energy of reaction of bulk crystalline ferrihydrite at 298.15 K are predicted to be ΔH298.15Krxn (Fe5O8H) = 6.4 kJ/(mol-Fe) and ΔG298.15Krxn (Fe5O8H) = 6.9 kJ/(mol-Fe), respectively, relative to bulk hematite and liquid water. The values demonstrate that fully crystalline ferrihydrite is metastable with respect to the formation of both hematite and goethite at 298.15 K but may be stabilized at small particle sizes due to favorably low surface formation energy. A simple surface energy model is used to predict the formation energy of ferrihydrite nanoparticles of arbitrary size, yielding results consistent with the observed formation energies for nanoparticle samples.