Fertilization effects on organic matter in physically fractionated soils as studied by 13C NMR: Results from two long-term field experiments

^l3C–nuclear magnetic resonance (NMR) spectra taken using magic–angle spinning (MAS), cross polarization (CP) and with total suppression of side bands (TOSS) are reported for soils from two long–term field experiments. One set of soils was from the Broadbalk Experiment at Rothamsted, UK (monoculture of winter wheat since 1843) and the other was from the Lermarken site of the Askov Long–Term Experiment on Animal Manure and Mineral Fertilizers (arable rotation since 1894). At both sites soil samples were taken from three fertilizer treatments: nil, inorganic fertilizers, animal manure. Spectra were obtained from whole soil samples and from the size fractions clay (<2 μrn), silt (2–20 μm) and, in some cases, sand (20–2000 μm). Comparison of the total strengths of the ¹³C–NMR signal for each size separate in relation to its total organic C content shows that clay, particularly, contains large percentages of C not detected by NMR because of the large magnetic susceptibilities of the soil minerals. It is proposed that the observed signals come from the more labile pools of soil organic matter (SOM), on the presumption that these pools are less closely associated with soil minerals and iron oxides and are likely to be less protected from microbial or enzymic decomposition. For both Rothamsted and Askov, functional groups in the 45–110 ppm region (N– and O–alkyls) dominate in the spectra for whole soils, with aromatics (110–160 ppm) and alkyls (0–45 ppm) signals being the next prominent. In the Askov whole soil samples ¹³C–NMR revealed no differences between nil, inorganic fertilizer and animal manure treatments but in the Rothamsted whole soil there were some small differences. Clay and silt fractions from Askov contain more alkyls and less aromatics than those from Rothamsted. For both sites clay in enriched in alkyls and depleted in aromatics relative to silt. Clay from Askov, but not Rothamsted, contains more N–alkyls (45–65 ppm) and less acetals (90–110 ppm) than silt. O–alkyls (65–90 ppm) account for more than 20% of the total signal in clay and silt from both sites. Fertilization regimes have not significantly affected the chemical composition of SOM associated with clay– and silt–sized fractions in the soils at either site. We conclude that the chemical composition of SOM is determined primarily by the interaction between the organisms responsible for decomposition and the mineral soil matrix rather than the nature of substrate input.