Spiking Improved Solution Phosphorus‐31 Nuclear Magnetic Resonance Identification of Soil Phosphorus Compounds

Chemical characterization of soil organic P using 31 P nuclear magnetic resonance (NMR) spectroscopy relies on the correct assignment of resonances. We examined eight Australian soils and identified the main orthophosphate monoester peaks by spiking model organic P compounds (α‐ and β‐glycerophosphate, ethanolamine phosphate, phytate, scyllo‐inositol hexakisphosphate, and choline phosphate) into NaOH–ethylenediaminetetraacetic acid (EDTA) soil extracts. For five of the soils, the strongest monoester resonances were identified as being due to phytate, while for the other three soils, the strongest resonances were identified as being due to α‐ and β‐glycerophosphate. Importantly, the appearance of spectra dominated by phytate and those dominated by glycerophosphate were deceptively similar because the separation between the two strongest phytate resonances was very similar to the separation between the α‐ and β‐glycerophosphate resonances. We believe this may have resulted in the misidentification of these species in some previous studies. Identification of these species is hindered by the sensitivity of their chemical shifts to pH and electrolyte concentration. Our spiking methodology, in which we add only enough of the compound to approximately double the native concentration, overcomes this problem. We also investigated the alkaline hydrolysis of phospholipids as a likely source of α‐ and β‐glycerophosphate. We found that the rate of phospholipid hydrolysis was dependant on NaOH concentration and that the presence of glycerophosphate resonances in soil extracts probably results from the extraction and redissolution procedures rather than the presence of native glycerophosphate in the soils. We also identified an intermediate diester product that may have previously been misidentified as the phospholipid itself.