Distribution of hydrous aluminium and iron over density fractions depends on organic matter load and ultrasonic dispersion

Abstract

Studies of organic–mineral interactions in soils often include physical fractionation according to particle size and density to separate organic debris and mineral-associated matter. For density fractions, it has been noticed that Fe hydrous compounds, although having a high absolute density, can be found in fractions supposed to contain only compounds of lower density. Probable reasons are incomplete wetting due to nanopores, thus reduced apparent density, the slow settlement during centrifugation because of the small particle size, and association with organic matter, reducing the absolute density and favouring dispersion. Allocation of associations of organic matter and hydrous mineral phases to light fractions may affect the estimation of mineral-associated organic matter in soils. We tested the distribution of goethite, 2-line ferrihydrite, and amorphous Al(OH) 3), associated with different amounts of organic matter, across density fractions obtained by floating in Na polytungstate solutions of 1.6, 2.0, and 2.4 g cm − 3 density, followed by centrifugation at 5000 g. Without organic matter, the two Fe minerals with absolute densities > 3.9 g cm − 3 , were recovered almost completely in the fraction > 2.4 g cm − 3 , irrespective the mode of dispersion (manual agitation or sonification with an energy input of 300 kJ ml − 1 ). Amorphous Al(OH) 3, having a density of 2.2 g cm − 3 , was recovered entirely in the fraction 2.0–2.4 g cm − 3 when agitated manually while little material (< 5%) was found in the fraction 1.6–2.0 g cm − 3 after ultrasonic dispersion. Thus density fractionation properly separates pure mineral particles according to their absolute density, neither affected by small particle size nor by incomplete wetting. Organic–mineral associations, resulting either from sorptive interactions or co-precipitation, had smaller absolute densities than the pure mineral phases, according to their content of organic matter. When agitated manually, > 95% of them were completely recovered in the appropriate density fraction, except for associations with extremely large organic contents (co-precipitates), of which small portions were found in the next lighter fraction. The application of ultrasonic energy, in contrast, resulted in up to 100% of the organic–mineral materials to be found in lighter fractions. The portion of material recovered in lighter fractions related positively to the content of mineral-associated organic matter. The results suggest that the application of ultrasonic energy to disrupt soil aggregates can cause dispersion of negatively charged organic–mineral associations in Na polytungstate solution, resulting in erroneous allocation to density fractions.