No abstract

We summarize progress with respect to (1) different approaches to isolate, extract, and quantify organo‐mineral compounds from soils, (2) types of mineral surfaces and associated interactions, (3) the distribution and function of soil biota at organo‐mineral surfaces, (4) the distribution and content of organo‐mineral associations, and (5) the factors controlling the turnover of organic matter (OM) in organo‐mineral associations from temperate soils. Physical fractionation achieves a rough separation between plant residues and mineral‐associated OM, which makes density or particle‐size fractionation a useful pretreatment for further differentiation of functional fractions. A part of the OM in organo‐mineral associations resists different chemical treatments, but the data obtained cannot readily be compared among each other, and more research is necessary on the processes underlying resistance to treatments for certain OM components. Studies using physical‐fractionation procedures followed by soil‐microbiological analyses revealed that organo‐mineral associations spatially isolate C sources from soil biota, making quantity and quality of OM in microhabitats an important factor controlling community composition. The distribution and activity of soil microorganisms at organo‐mineral surfaces can additionally be modified by faunal activities. Composition of OM in organo‐mineral associations is highly variable, with loamy soils having generally a higher contribution of polysaccharides, whereas mineral‐associated OM in sandy soils is often more aliphatic. Though highly reactive towards Fe oxide surfaces, lignin and phenolic components are usually depleted in organo‐mineral associations. Charred OM associated with the mineral surface contributes to a higher aromaticity in heavy fractions. The relative proportion of OC bound in organo‐mineral fractions increases with soil depth. Likewise does the strength of the bonding. Organic molecules sorbed to the mineral surfaces or precipitated by Al are effectively stabilized, indicated by reduced susceptibility towards oxidative attack, higher thermal stability, and lower bioavailability. At higher surface loading, organic C is much better bioavailable, also indicated by little 14C age. In the subsurface horizons of the soils investigated in this study, Fe oxides seem to be the most important sorbents, whereas phyllosilicate surfaces may be comparatively more important in topsoils. Specific surface area of soil minerals is not always a good predictor for C‐stabilization potentials because surface coverage is discontinuous. Recalcitrance and accessibility/aggregation seem to determine the turnover dynamics in fast and intermediate cycling OM pools, but for long‐term OC preservation the interactions with mineral surfaces, and especially with Fe oxide surfaces, are a major control in all soils investigated here.

No abstract

In soils and sediments ferrihydrite often precipitates from solutions containing dissolved organic matter, which affects its crystallinity. To simulate this process we prepared a series of 2-line ferrihydrite-organic matter coprecipitates using water extractable organic matter (OM) from a forest topsoil. The products were characterized byX-ray diffraction, Mössbauer spectroscopy, N2-gas adsorption and transmission electron microscopy. With increasing C/Fe ratios of the initial solution the d-spacings of the two major XRD peaks increased, while peak shoulders at 0.22 and 0.16 nm weakened. The asymmetry of the 0.26 nm peak decreased and disappeared at a C/Fe ratio of 0.78. The quadrupole splitting of the Mössbauer spectra at 300 K increased from 0.78 to 0.90 mm s(-1), the mean magnetic hyperfine field at 4.2 K dropped from 49.5 to 46.0 T, and the superparamagnetic collapse of the magnetic hyperfine splitting was shifted toward lower temperatures. These data reflect a strong interference of OM with crystal growth leading to smaller ferrihydrite crystals, increased lattice spacings, and more distorted Fe(O,OH)6 octahedra. Even small amounts of OM significantly change particle size and structural order of ferrihydrite. Crystallinity and reactivity of natural ferrihydrites will therefore often differ from their synthetic counterparts, formed in the absence of OM.

In soil and water, ferrihydrite frequently forms in the presence of dissolved organic matter. This disturbs crystal growth and gives rise to coprecipitation of ferrihydrite and organic matter. To compare the chemical fractionation of organic matter during coprecipitation with the fractionation involved in adsorption onto pristine ferrihydrite surfaces we prepared ferrihydrite-organic matter associations by adsorption and coprecipitation using (i) a forest-floor extract or (ii) a sulfonated lignin. The reaction products were studied by (13)C CPMAS NMR, FTIR, and analysis of hydrolyzable neutral polysaccharides. Relative to the original forest-floor extract, the ferrihydrite-associated organic matter was enriched in polysaccharides, especially when adsorption took place. Moreover, mannose and glucose were bound preferentially to ferrihydrite, while fucose, arabinose, xylose, and galactose accumulated in the supernatant. This fractionation of sugar monomers was more pronounced during coprecipitation and led to an enhanced ratio of (galactose + mannose)/(arabinose + xylose). Experiments with lignin revealed that the ferrihydrite-associated material was enriched in its aromatic components but had a lower ratio of phenolic C to aromatic C than the original lignin. A compositional difference between the adsorbed and coprecipitated lignin is obvious from a higher contribution of methoxy C in the coprecipitated material. Coprecipitated organic matter may thus differ in amount and composition from adsorbed organic matter.

Organo-mineral associations stabilize soil organic matter, though the mechanisms by which they do so are unclear. We used particle-size fractions < 6.3 mm of two soils to examine the importance of Fe oxides, short-range order Al silicates and the surface areas of minerals and micropores on the formation of organo-mineral associations. In the subsoil Fe oxides were most strongly statistically correlated with the mineral-bound organic carbon. We therefore assume that they are the most important substrates for the formation of organo-mineral associations. There is no indication that this is caused by physical protection of organic matter in their micropores (< 2 nm). In the Haplic Podzol, dithionite-citrate-bicarbonatesoluble short-range order Al silicates may also play a role. Fe oxide particles were calculated to offer specific surface areas of $ 200 m 2 g À1 (goethite) and $ 800 m 2 g À1 (ferrihydrite), corresponding to crystal diameters of only a few nm. We assume that the resulting large amount of oxide-specific reactive surface sites (conditionally charged hydroxyl groups) is responsible for their dominant role as sorbents. With maximum C loadings of 1.3 mg C per m 2 Fe oxide for the Dystric Cambisol and 1.1 mg C per m 2 Fe oxide þ short-range order Al silicates for the Haplic Podzol the subsoils of both soils seem to have reached saturation with respect to organic matter sorption. In contrast to subsoil horizons, organo-mineral associations from topsoils contain much larger amounts of organic matter. Here a larger C loading on Fe oxides or a greater importance of other sorbents in addition to the oxides must be assumed.

We tested the hypothesis whether organic matter in subsoils is a large contributor to organic carbon (OC) in terrestrial ecosystems and if survival of organic matter in subsoils is the result of an association with the soil mineral matrix. We approached this by analyzing two forest soil profiles, a Haplic Podzol and a Dystric Cambisol, for the depth distribution of OC, its distribution among density and particle‐size fractions, and the extractability of OC after destruction of the mineral phase by treatment with hydrofluoric acid (HF). The results were related to indicators of the soil mineralogy and the specific surface area. Finally, scanning electron microscopy combined with energy dispersive X‐ray spectroscopy (SEM‐EDX) was used to visualize the location of OC at mineral surfaces and associations with elements of mineral phases. The subsoils (B and C horizons) contained 40—50% of the soil OC including the organic forest floor layers. With increasing depth of soil profiles (1) the radiocarbon ages increased, and (2) increasing portions of OC were either HF‐soluble, or located in the density fraction d >1.6 g cm—3, or in the clay fraction. The proportions of OC in the density fraction d >1.6 g cm—3 were closely correlated to the contents of oxalate and dithionite‐citrate‐bicarbonate‐extractable Fe (r2 = 0.93 and 0.88, P <0.001). SEM‐EDX analyses suggested associations of OC with aluminum whereas silicon‐enriched regions were poor in OC. The specific surface area and the microporosity of the soil mineral matrix after destruction of organic matter were less closely correlated to OC than the extractable iron fractions. This is possibly due to variable surface loadings, depending on different OC inputs with depth. Our results imply that subsoils are important for the storage of OC in terrestrial ecosystems because of intimate association of organic matter with secondary hydrous aluminum and iron phases leading to stabilization against biological degradation.

No abstract