Shifts in soil organic matter composition following treatment with sodium hypochlorite and hydrofluoric acid

A renewed interest in chemical fractionation of soil organic matter (SOM) originates from the premise that it enables to isolate labile SOM from SOM protected through mineral binding and recalcitrant SOM. Both selective removal of labile non-bound SOM through oxidation or hydrolysis as well as selective removal of minerals and attached SOM are often applied. Molecular-level SOM characterization by means of temperature resolved Pyrolysis-Field Ionization Mass Spectroscopy analysis (Py-FIMS) was used here as an approach to obtain insight into the fate of SOMuponwet chemical treatment with regard to composition and thermal stability. The applied sequential chemical treatment with 6% NaOCl and 10% HF yielded similar sizes in stable SOM fractions between sandy semi-native heathland and cultivated cropland soil pairs (i.e. NaOCl resistant OC: 12.3–15.0 g C kg−1 and NaOCl+HF resistant OC: 2.6–5.3 g C kg−1). Py-FIMS spectra of bulk SOM in both heathland–cropland soil pairs were dominated by signals assigned to lipids, alkylaromatics and sterols. Difference spectra and thermograms showed selective loss of signals from sterols, lignin dimers and thermolabile lipids. This matches advancing SOM decomposition as derived from previously reported gradients in SOM composition as decomposition proceeds fromplantmaterial over particulate organicmatter (OM) to SOMin silt and clay particle sizes. However, increased ion intensity attributed to carbohydrates, peptides and short-chained lipids after NaOCl treatment indicates that biologically labile SOM components were also enriched, and they may possibly have been protected through mineral binding or encapsulation in macromolecular OMstructures. Subsequent HF treatment yielded increased volatilization in the thermostable region for mass signals tentatively assigned to phenols and lignin monomers and of heterocyclic N-containing compounds and thermostable alkylaromatics. The resistance to chemical treatment of the latter two components matches with their hypothesized structural function in macro OM molecules. However, even for the sites investigated here, with a very similar soil texture, climate, land-use, drainage and contents of pedogenic oxides, contrasting and not readily explainable results were found for other SOMconstituents. Therefore, chemical fractionations seemto yield very site specific and less distinct patterns and this study demonstrated that derivation of useful information regarding SOM stabilization mechanisms from such experiments is by no means straightforward