A carbonate-fluoride defect model for carbonate-rich fluorapatite

We propose a microscopic model of the dominant carbonate for phosphate substitution in fluorapatite. A well-crystallized sedimentary fluorapatite sample containing similar to 2.3 +/- 0.8 wt% of carbonate was investigated using Fourier transform infrared spectroscopy (FTIR) and C-13 and F-19 magic angle spinning nuclear magnetic resonance (MAS NMR). About 75% of the carbonate groups replace the phosphate group ("B-site"), whereas a lesser contribution from carbonate groups located in the structural channels ("A-site") is observed. Beside the dominant F-19 NMR signal of channel ions at similar to-102 ppm, an additional signal corresponding to similar to 8% of fluoride ions is observed at -88 ppm. F-19 double quantum-single quantum (DQ-SQ) MAS NMR and C-13 {F-19} frequency-selective Rotational Echo DOuble Resonance (REDOR) experiments prove that this additional signal corresponds to isolated fluoride ions in the apatite structure, located in close proximity of substituted carbonate groups. Density functional theory (DFT) calculations allow us to propose a composite carbonate-fluoride tetrahedron defect model accounting for these experimental observations. The planar carbonate ion lies in the sloping face of the tetrahedron opposite a fluoride ion occupying the remaining vertex, together replacing the tetrahedral phosphate ion. This "francolite-type" defect leads to a diagnostic narrow IR absorption band at 864 cm(-1) that could be used as a guide to, e.g., detect the incipient transformation of fossil bone and teeth samples.