Rate of desorption of eight heavy metals from goethite and its implications for understanding the pathways for penetration |
| |
| Authors: | N J Barrow G W Brümmer L Fischer |
| |
| Institution: | 1. Faculty of Natural and Agricultural Sciences, School of Plant Biology, University of Western Australia, 35 Stirling Highway, Crawley WA 6009, Australia;2. Institute of Crop Science and Resource Conservation, Division of Soil Science, University of Bonn, Nu?allee 13, 53115 Bonn, Germany |
| |
| Abstract: | We compared rates of desorption of heavy metals from goethite, an important soil constituent, in order to understand the mechanisms of sorption and desorption better. Samples of goethite were reacted with salt solutions of heavy metals for 2 hours or for 8 weeks. The metals were Cr, Mn, Co, Ni, Cu, Zn, Cd and Pb. Desorption was then induced by successive treatments with 0.7 m HNO3 for up to 360 hours followed by complete dissolution of the goethite particles. After brief sorption (2 hours), a large proportion of the sorbed metals was desorbed by brief treatment (15 minutes) with HNO3. This suggests that the metals were mainly sorbed on, or close to, external goethite surfaces. After longer sorption (8 weeks), even 360 hours of extraction with 0.7 m HNO3 did not recover all of the sorbed metals. Complete dissolution of the goethite particles was necessary. This suggests that the eight heavy metals had penetrated the goethite particles deeply. When desorption was summarized using a pragmatic equation, it was only for brief sorption followed by brief desorption that indices of the rate of desorption were well correlated with measures of the rate of sorption as obtained in an earlier work. When desorption was described with a mechanistic model, observed desorption was often faster than predicted and the discrepancy was greatest with the shorter periods of desorption. The discrepancy was marked for Ni, Cr and Co. We think this was because large portions of these metals were incorporated into the goethite structure by lattice diffusion and were therefore held close to the surface. When the acid dissolved some of the goethite these metals were released. For long‐term desorption and for most metals, the ratio of observed to predicted desorption decreased with increasing ionic radius. We think that this reflects a changing balance between lattice and pore diffusion, with lattice diffusion more important for metals with radii near that of Fe (Cr, Co, Ni), and pore diffusion more important for the larger cations of Cd and Pb. Manganese was an exception. Desorption was at first faster than predicted but then declined to be close to predicted values. This suggests that Mn mostly penetrated more deeply by pore diffusion. |
| |
| Keywords: | |
|
|
