The chemical composition of measurable soil organic matter pools Natacha Poirier a , Saran P. Sohi b , John L. Gaunt b , Nathalie Mahieu c , Edward W. Randall c , David S. Powlson b , Richard P. Evershed a, * a Organic Geochemistry Unit, Bristol Biogeochemistry Research Centre, School of Chemistry, University of Bristol, CantockÕs Close, Bristol BS8 1TS, UK b Agriculture and the Environment Division, Rothamsted Research, Harpenden AL5 2JQ, UK c Department of Chemistry, Queen Mary, University of London, E1 4NS, UK Received 8 March 2004; accepted 12 March 2005 (returned to author for revision 26 April 2004) Available online 2 June 2005 Abstract A range of spectroscopic, ÔwetÕ chemical, gas chromatographic (GC) and mass spectrometric (MS) techniques was applied to the characterisation of three soil organic matter (SOM) fractions that have been proposed as the basis of a new SOM turnover model based on measurable, physically defined fractions. The fractions were: the free light frac- tion (obtained by density separation in NaI solution at a density of 1.80 g cm 3 , without disruption of aggregates), the intra-aggregate light fraction (obtained using a second density separation after disrupting aggregates using ultrasonic dispersion) and the organomineral fraction corresponding to the residual heavy material. The techniques employed to investigate the composition of the organic constituents of each fraction were: 13 C nuclear magnetic resonance (NMR), Fourier transform infrared spectroscopy (FTIR), and pyrolysis-gas chromatography/mass spectrometry (py-GC/MS) to study bulk composition. Lipid, lignin and carbohydrate fractions were assessed using GC and GC/MS with appro- priate derivatisation, following oxidative and hydrolytic treatments, respectively, in the case of the latter two classes. Proteinaceous components were determined as amino acids using reversed-phase high performance liquid chromatog- raphy (HPLC) following 6 M HCl treatment and derivatisation. Each technique revealed marked differences in chem- ical composition between the organomineral and the two light fractions, with the results being consistent with the organomineral fraction having different biological sources or having undergone a greater degree of degradation or transformation. Several techniques detected differences between the composition of the free light fraction and the intra-aggregate light fraction. With the exception of carbohydrate composition, the results were consistent with the order of reactivity previously proposed from incubation studies with isotopically labelled substrates, namely: free > intra-aggregate > organomineral. The investigation illustrates the importance of using a range of different chemical characterisation techniques in studies of complex SOM fractions as each has limitations that could, if used alone, produce ambiguous findings or fail to detect differences between them. Ó 2005 Elsevier Ltd. All rights reserved. 0146-6380/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.orggeochem.2005.03.005 * Corresponding author. Tel.: +44 117 9287671; fax: +44 117 9291295/9293746. E-mail address: r.p.evershed@bristol.ac.uk (R.P. Evershed). Organic Geochemistry 36 (2005) 1174–1189 www.elsevier.com/locate/orggeochem Organic Geochemistry