The chemical nature of nitrogen in native soil organic matter

Abstract

H. Knicker, R. Frtind and H.-D. Ltidemann Institut for Biophysik und physikalische Biochemie der Universit~it, W-8400 Regensburg, FRG Fossil fuels and soil organic matter (SOM) together contain approximately five times more carbon than the biota and the atmosphere. Of this, soil organic matter accounts for about 30 % of the carbon present. In addition, SOM has an average carbon/nitrogen ratio of 10/1 and contains a huge fraction of the total ni- trogen available for plant growth [1]. Taking into account that the abundance of nitrogen in the earth's crust is much lower than that of carbon, this is a signif- icant fraction of the total nitrogen acces- sible to the biosphere. Under natural soil conditions, without the addition of miner- al fertilizers, SOM provides the majority of the nitrogen necessary for plant growth. It is also thought to be respon- sible for the interaction between agricul- tural biocides and the soil [5-7]. The chemical structure of this ubiquitous material, SOM, and especially the chem- ical nature of the nitrogen are thus of great and general importance. The mo- lecular structure of the nitrogen- containing fraction is, however, still a matter of controversy [2-4]. Structural models based on partial chem- ical analysis claim that a significant part of the nitrogen is present in the form of heteroaromatic structures, while NMR- spectroscopic studies on lSN-enriched composts and recent humic material found approximately 85 % of the signal intensity in the amide/peptide region of chemical shift and no signals in the range typical for heteroaromatic nitrogen. A major fraction of the native soil organic matter has been in the soil for several hundred to several thousand years [8, 9]. Compared to these time spans, laboratory-produced material has been fermented for at most 1 year, and it could be argued that heteroaromatic structures are only produced after much longer fermentation periods. This criti- cism may be overcome by the study of lsN-CPMAS spectra of soil organic mat- ter with natural lSN levels. This has not been achieved hitherto, because the low natural abundance (0.4 %) and the small gyromagnetic ratio of the 15N nucleus and therefore its low sensitivity in NMR experiments appeared to make this experiment an impossible one. The most abundant 14N-isotope (99.6 %) cannot be studied by high-resolution NMR because its large nuclear quadrupole moment leads to very broad and unresolved signals, especially in solid-state NMR [101. In previous systematic studies on 15N- enriched composts and organic soil extracts [11] our group optimized all spectral parameters for the ~sN-CPMAS experiment. A crude estimate showed that it should be possible to obtain 15N spectra with a tolerable signal-to-noise ratio after the accumulation of approximately one million transients. In the present paper we report on the first successful results of such experiments. Six German soils were studied as detailed in Table 1. In Figs. 1 and 2 some of the spectra obtained are shown. They fully corrobo- rate the conclusions drawn from the stud- ies of short-term composting exper-