The role of DOM sorption to mineral surfaces in the preservation of organic matter in soils

Sorption of dissolved organic matter (DOM) is considered to be a major process in the preservation of organic matter (OM) in marine sediments. Evidence for this hypothesis includes the close relationship between sediment surface area (SA) and organic carbon (OC) concentrations and the strongly reduced biological degradability after DOM has sorbed to mineral surfaces. The aim of this study was to discuss the possibility of a similar process in the soil environment. We accomplished this by gathering information from the literature, and by an evaluation of our own studies on DOM sorption and accumulation of OM in soil. We found that in soil a close association of OM with the mineral matrix exists. Both the concentration of soil OM associated with the mineral matrix, and the sorption of DOM are related to reactive mineral phases such as Al and Fe oxyhydroxides. Sorption of DOM derived from the oxidative decomposition of lignocellulose to Al and Fe oxyhydroxides involves strong complexation bondings between surface metals and acidic organic ligands, particularly with those associated with aromatic structures. The strength of the sorption relates to the surface area but more importantly to the surface properties of the sorbing mineral phase. The sorption of a large part of DOM is hardly reversible under conditions similar to those during sorption (hysteresis). Because sorption of the more labile polysaccharide-derived DOM on mineral surfaces is weaker, adsorptive and desorptive processes strongly favour the accumulation of the more recalcitrant lignin-derived DOM. In addition, we found the soil OM in an alluvial B horizon and in the clay fraction of a topsoil strongly resembling lignin-derived DOM from the overlying forest ̄oors. Hence, it seems likely that sorption of DOM contributes considerably to the accumulation and preservation of OM in soil. However, this does not result in a signi®cant relationship between OC concentration and SA. Reasons for that ®nding may be the ''masking'' of mineral surfaces by adsorbed OM, the clustering of OM patches at highly reactive sites of metal hydroxides, and/or the absence of a relationship between SA and the concentration of surface-active Fe and Al oxyhydroxides in some soil types. Overall, we conclude that sorptive preservation of OM in soil is a€ected by the chemical structure of the sorbing DOM and the surface properties of the mineral matrix. Localisation and conformation of sorbed OM remains unclear and therefore should be subject of further research. # 2000 Elsevier Science Ltd. All rights reserved.