The light organic matter fraction is known as an early indicator of changes in soil organic matter (SOM) as a result of different soil use and management. Several solutions are used in its separation, but little is known about the effect of these solutions on the quantity and quality of the extracted fraction, as well as the possible interferences of the separation solution on the quantification of organic carbon (C) content. Twenty soils with variable clay contents and under different uses were selected. Soil samples were collected in the 0-20 cm layer, with three replications. After air-drying, they were ground manually, passed through a 2 mm sieve and homogenized; sub-samples were taken for clay content determination and densimetric SOM fractionation. Based on the fractionation scheme using water or NaI (1.8 kg L-1) the SOM light fraction (LF) and the heavy fraction (HF) were obtained. The C concentration of LF was determined by dry combustion in an elemental analyzer (CHN) and the C content of HF by wet oxidation, according to Yeomans & Bremner (1988) (Y & B) and by dry combustion. Additionally, we determined the equivalent organic C concentration by wet oxidation of standard solutions containing increasing NaI concentration. It was found that the amount of NaI-extracted LF averaged across soils was 4.86 times greater than when extracted with distilled water. The C:N ratio of this fraction was, on average, 1.85 times greater than that extracted with water, and the variation among soils was high. With regard to HF, the C concentration determined by wet oxidation was similar to that separated with water, but 1.5 times larger than when determined by dry combustion if separated with NaI. These results indicate that the C content was overestimated, possibly due to the interference of NaI. This finding was confirmed by the linear increase in the consumption of Cr2O7-2 during organic C determination in the presence of increasing NaI concentration of organic-C free solutions. Therefore, the NaI extraction of LF recovers a larger amount of LF, which differs however in quality from that extracted with water. Additionally, any potential NaI remaining after the separation, especially in the HF, may interfere with the organic C quantification if wet oxidation with dichromate as oxidant agent is the chosen method. Results of a complementary study indicate that each mol of I- consumes approximately 1 mol of Cr2O7-2.
organic matter; densimetric fractionation; sodium iodide
