Portable x-ray fluorescence measurement viability in organic rich soils: pXRF response as a function of organic matter presence.

摘要

Portable X-Ray fluorescence provides a cost effective method for producing rapid geochemical data. With advancements in X-Ray generation and detection technology, pXRF has become feasible for use on pedologic materials. Factors affecting pXRF soil measurements such as heterogeneity, moisture content, object geometry, and matrix interferences are widely recognized. However, the influence of organic matter on pXRF soil measurements is poorly understood. This study examined the influence of organic matter fraction on pXRF trace metal measurements in a soil matrix. Incremental addition of three organic matter surrogates (cellulose, graphite powder, and confectioner's sugar) was used to investigate the influence of increasing organic matter content up to approximately 30%. Each organic matter surrogate was independently added to and homogenized with samples of Natural Resources Canada Till-1 standard reference material that was initially expunged of organic matter through combustion. Incremental addition was performed 20 times for each surrogate, and concentrations of arsenic, chromium, iron, manganese, lead, rubidium, strontium, thorium, titanium, vanadium, zinc and zirconium were measured as a function of varying organic matter fractions using a Thermo Scientific Niton XL3t GOLDD+ 950 XRF analyzer. Results demonstrate attenuation of the pXRF signal and elementally-dependent deviation from the expected concentration with increasing sample organic matter fraction. Based on the results, a pXRF organic matter fraction-dependent calibration method was developed and its performance was evaluated using four unmodified NRCan soil standards (Till-1, Till-2, Till-3, Till-4) with known organic matter content. In general, the new method improved measurement accuracy compared to conventional pXRF calibration methods. Because soils inherently contain varying amounts of organic matter, these results suggest more accurate geochemical data can be generated via pXRF if soil organic matter fraction is quantified and taken into account during measurement calibration.