13C NMR studies of organic matter in whole soils: H. A case study of some Rothamsted soils |
| |
Authors: | P. KINCHESH D. S. POWLSON E. W. RANDALL |
| |
Affiliation: | Department of Chemistry, Queen Mary &Westfield College, London E1 4NS, UK;Department of Soil Science, Rothamsted Experimental Station, Harpenden, Herts AL5 2JQ, UK |
| |
Abstract: | Nuclear magnetic resonance (NMR) spectra were obtained for solid samples of whole soils from three long–term field sites at Rothamsted Experimental Station, UK. In all sites, soil organic matter content was either increasing or decreasing due to contrasted long–continued treatments. Two soils were from Highfield, one from under old grassland (47 g organic C kg?1) and one from an area kept as bare fallow following ploughing of grass 21 years previously (14 g organic C kg?1). Three soils were taken from Broadbalk, two from plots within the Broadbalk Continuous Wheat Experiment which had received no fertilizer or animal manure annually for 148 years (7 and 27 g organic C kg?1, respectively) and one from Broadbalk Wilderness, wooded section (38 g organic C kg?1). Broadbalk Wilderness was arable until 1881 and has reverted to deciduous woodland in the subsequent 110 years. Two soils were from Geescroft, one from an arable field (9 g organic C kg?1) and one from Geescroft Wilderness (35 g organic C kg?1) which began reversion to deciduous woodland at the same time as Broadbalk Wilderness but is now acid (pH = 4.2) in contrast to Broadbalk which is calcareous (pH = 7.3). Solid–state 13C NMR spectra were obtained on a 300–MHz instrument using cross polarization (CP) and magic angle spinning (MAS). All samples exhibited peaks in the following spectral regions: 0–45 ppm (alkyl), 45–60 ppm (methoxyl, carbohydrate and derivatives), 60–110 ppm (carbohydrates and derivatives, C–α of peptides), 110–160 ppm (aromatics) and 160–185 ppm (carboxyl groups and derivatives). Within the spectrum of a specific sample it was not possible to determine the relative proportions of soil organic carbon in the different forms identified because a range of factors can potentially alter the relative areas of peaks in different regions of the spectrum. However, from a comparison of relative peak areas within a set of soils from a given site, differing only in organic matter content, information can be deduced regarding the forms of C that are more or less subject to change in response to land use or management. At all sites carbohydrate C appears to be the form that is most subject to change, suggesting that it is an ‘active’ fraction compared with the other forms. It was greatest where organic matter inputs were greatest (due to inputs of farmyard manure or reversion to woodland) and declined relative to other forms following ploughing of old grassland. Alkyl C increased as total C accumulated but did not decline relative to other forms following ploughing of grass. One reason for the non–quantitative nature of the soil 13C CPMAS spectra was a short (approximately 1 ms) component of the rotating–frame TI relaxation time for H nuclei (T1pH). This problem was not overcome by acquiring data at – 60°C. In principle, solid–state spectra of soils obtained by direct polarization (i.e. without CP) might produce quantitative results, but the low C content of most mineral soils (10–50 g C kg?1) precludes this, given current instrumentation. |
| |
Keywords: | |
|
|