How useful is chemical oxidation with dichromate for the determination of “Black Carbon” in fire-affected soils?

Abstract

A common approach for the quantification of Black Carbon (BC) in soils and sediments represents the chemical oxidation with acid potassium dichromate. Because this method is still associated with some uncertainties, its applicability for the detection of BC produced during vegetation fires was examined by analyzing fire-affected and unaffected soil organic matter, including fresh and charred plant residues and a charcoal briquette. Comparably to the organic matter in a fire-unaffected soil sampled under pasture, grass and beech sawdust remains were completely oxidized after 4 to 6 h, whereas approximately 12% of the organic carbon (Corg) in pine needles resisted the chemical oxidation. Based on solid-state 13C nuclear magnetic resonance (NMR) spectroscopy, this Corg was assigned to plant waxes. In the fire-unaffected control soils located under pine and oak forests such acid-resistant paraffinic structures accounted for 6 to 22% of Corg of the untreated samples. Approximately half of them were removable by a subsequent Soxhlet extraction, clearly demonstrating that their survival is rather explainable by their hydrophobic nature than caused to their chemical recalcitrance. Chemical oxidation of charcoal briquette for 6 h resulted in a survival of 82% of its Corg. Beech sawdust BC and grass BC showed lower recalcitrance and only 66% and 40% were recovered. Dipolar dephasing NMR techniques used to elucidate the average protonation degree of the aromatic C of BC demonstrated that the variable reaction to chemical treatment cannot be explained by differences in the condensation degree. Further, after oxidation, the residual aromatic C of the reference chars shows no major decrease of the protonation degree, which contradicts a possible preferential preservation of polycondensed aromatic structures from the oxidative attack. Increasing the treatment time to 24 h, considerably lowered the amount of chemical oxidation resistant elemental carbon (COREC) to 18% for the grass BC and 60% for the charcoal briquette. This confirms that the amount of determined BC varies not only with source material but also relies strongly on the applied oxidation time. Considering further a contribution of non-BC derived paraffinic structures of up to 19 mg g−1 soil in the fire-unaffected soils, the reliability of BC data of soil organic matter obtained solely by chemical oxidation methods is questionable. Therefore, additional characterization of extraction residue is obligatory. Using solidstate 13C NMR spectroscopy, on the other hand, allowed the identification of BC by signal intensity assignable to aromatic C. However, using this signal as a means for BC-quantification, one has to encounter that chemical oxidation even attacks charcoal BC, which is most tentatively a major reason for BC underestimations in soils investigated by this method.