Standard Protocol and Quality Assessment of Soil Phosphorus Speciation by P K-Edge XANES Spectroscopy

Phosphorus (P) in soils is most often bound as phosphate to one or more of the following four elements or compounds: calcium, aluminum, iron, and soil organic matter. A promising method for direct P speciation in soils is synchrotron-based X-ray absorption near edge structure (XANES) spectroscopy at the K-edge of P. However, the quality of this method is debated controversially, partly because a standard protocol for reproducible spectrum deconvolution is lacking and minor modifications of the applied deconvolution procedure can lead to considerable changes in the P speciation results. On the basis of the observation that appropriate baseline correction and edge-step normalization are crucial for correct linear combination (LC) fitting results, we established a standard protocol for the deconvolution and LC fitting of P K-edge XANES spectra. We evaluated the quality of LC fits obtained according to this standard protocol with 16 defined dilute (2 mg P g(-1)) ternary mixtures of aluminum phosphate, iron phosphate, hydroxyapatite, and phytic acid in a quartz matrix. The LC fitting results were compared with the contribution of the different P compounds to total P in the various mixtures. Compared to using a traditional LC fitting procedure, our standard protocol reduced the fitting error by 6% (absolute). However, P portions smaller than 5% should be confirmed with other methods or excluded from the P speciation results. A publicly available database of P K-edge XANES reference spectra was initiated.