Comparison of chemical fractionation methods for isolating stable soil organic carbon pools

Stable soil organic matter (SOM) is important for long‐term sequestration of soil organic carbon (SOC), but the usefulness of different fractionation methods to isolate stable SOM is open to question. We assessed the suitability of five chemical fractionation methods (stepwise hydrolysis, treatment with H₂O₂, Na₂S₂O₈, NaOCl, and demineralization of the NaOCl‐resistant fraction (NaOCl + HF)) to isolate stable SOM from soil samples of a loamy sand and a silty loam under different land use regimes (grassland, forest and arable crops). The apparent C turnover time and mean age of SOC before and after fractionation was determined by ¹³C and ¹⁴C analysis. Particulate organic matter was removed by density fractionation before soils were exposed to chemical fractionation. All chemical treatments induced large SOC losses of 62–95% of the mineral‐associated SOC fraction. The amounts of H₂O₂‐ and Na₂S₂O₈‐resistant SOC were independent from land use, while those of NaOCl‐ (NaOCl + HF)‐ and hydrolysis‐resistant SOC were not. All chemical treatments caused a preferential removal of young, maize‐derived SOC, with Na₂S₂O₈ and H₂O₂ being most efficient. The mean ¹⁴C age of SOC was 1000–10000 years greater after chemical fractionation than that of the initial, mineral‐associated SOC and mean ¹⁴C ages increased in the order: NaOCl < NaOCl + HF ≤ stepwise hydrolysis ? H₂O₂≈ Na₂S₂O₈. None of the methods appeared generally suitable for the determination of the inert organic matter pool of the Rothamsted Carbon Model. Nonetheless, our results indicate that all methods are able to isolate an older, more stable SOC fraction, but treatments with H₂O₂ and Na₂S₂O₈ were the most efficient ones in isolating stable SOM.