Abstract
Soil carbon (C) determinations are important for soil quality assessment. The standard method that is used in São Paulo state to determine soil organic C is a modified Walkely–Black method. This method has some disadvantages because of the time needed for implementation and the use of potassium dichromate. Therefore, the aims of this work were to test the efficiency and compare the following methods: (a) wet oxidation of organic C (OC-HO) followed by spectrophotometer determination, (b) dry incineration at 550 °C with a muffle furnace plus gravimetric determination (TC-I), and (c) dry combustion with a C–nitrogen (N)–sulfur (S) elemental analyzer (TC-EA). Sixty-one soil samples were collected at the 0- to 20-cm depth on agricultural soils of São Paulo state. Samples were dried at 65 °C, ground, and sieved through a 0.09-mm sieve. Carbon content was determined in triplicate. These sets of samples exhibited a wide range of C content: TC-EA was from 0.43 to 9.14 g 100 g−1. Mean C contents were 6.76 g 100 g−1, 2.31 g 100 g−1, and 2.52 g 100 g−1 for TC-I, OC-HO, and TC-EA, respectively. Moreover, in all the studied soil samples, the TC-I method overestimated the C content when compared with OC-HO and TC-EA methods. Losses of structural water and inorganic colloidal compounds, such as gibbsite, induced by high temperature during incineration probably were responsible for the greater C contents obtained by the TC-I method. Correlation coefficients between the three methods were rather similar: TC-I and TC-EA (r = 0.87), TC and OC-HO (r = 0.83), and TC-EA and OC-HO (r = 0.84). It was concluded that C extraction by the standard, routinely employed OC-HO method can be replaced by the TC-EA method, which has environmental advantages.