The role of Al in the formation of secondary Ni precipitates on pyrophyllite, gibbsite, talc, and amorphous silica: A DRS study

Formation of secondary Ni precipitates is an important mechanism of Ni retention in neutral and alkaline clay/water systems. However, the structure and composition of these secondary phases, and their stability is still disputable. Using existing structure refinement data and new ab-initio FEFF 7 calculations we show that Ni-edge X-ray absorption fine structure spectroscopy alone may not be able to unequivocally discriminate four possible candidate compounds: a-Ni(OH)2, the isostructural but Al-substituted layered double hydroxide (Ni-Al LDH), and 1:1 and 2:1 Ni-containing phyllosilicates. Hence, we investigated the potential of diffuse reflectance spectroscopy (DRS) in determining in situ the Ni phase forming in the presence of four sorbents, pyrophyllite, talc, gibbsite, and amorphous silica. The A2g 3 T1g(F) band (n2) of octahedrally coordinated Ni could be reliably extracted from the reflectance spectra of wet pastes. In the presence of the Al-free talc and amorphous silica, the n2 band was at '14,900 cm, but shifted to 15,300 cm in the presence of Al-containing pyrophyllite and gibbsite. This shift suggests that Al is dissolved from the sorbent and substitutes for Ni in brucite-like hydroxide layers of the newly forming precipitate phase, causing a decrease of the Ni-O distances and, in turn, an increase of the crystal-field splitting energy. Comparison with Ni model compounds showed that the band at 14,900 cm is a unique fingerprint of a-Ni(OH)2, and the band at 15,300 cm 21 of Ni-Al LDH. Although the complete transformation of a-Ni(OH)2 into a Ni phyllosilicate causes a significant contraction of the Ni hydroxide sheet as indicated by band positions intermediate to those of a-Ni(OH)2 and Ni-Al LDH, incipient states of silication do not influence Ni-O distances and cannot be detected by DRS. The first evidence for the formation of a precipitate was obtained after 5 min (pyrophyllite), 7 hr (talc), 24 hr (gibbsite), and 3 days (amorphous silica). For both pyrophyllite and talc, where sufficiently long time series were available, the n2 energy slightly increased as long as the Ni uptake from solution continued (3 days for pyrophyllite, 30 days for talc). This may be explained by a relative decrease of relaxed surface sites due to the growth of crystallites. Our study shows that the formation of both a-Ni(OH)2 and Ni-Al LDH may effectively decrease aqueous Ni concentrations in soils and sediments. However, Ni-Al LDH seems to be thermodynamically favored when Al is available. Copyright © 1999 Elsevier Science Ltd