Quantitative X-ray Fluorescence Analysis of Biomass (Switchgrass, Corn Stover, Eucalyptus, Beech, and Pine Wood) with a Typical Commercial Multi-Element Method on a WD-XRF Spectrometer

Quick and reliable inorganic elemental chemical analysis of biomass (including solid biofuels) is of importance in the increasing utilization and trade of biomass. In particular, it is important for the exploitation of contaminated/dirty biomass/biomass waste, and potentially also as a tool in ascertaining the type/origin of biomass. X-ray fluorescence (XRF) spectrometry performed directly on the raw biomass with limited prior sample preparation is an attractive method for performing such inorganic elemental analysis. In the present study, we therefore carefully investigate the performance of a commercial multi-element standardless XRF method by analyzing five common biomass types (switchgrass, corn stover, eucalyptus, beech, and pine wood). Sample preparation involves milling the raw biomass using cutter and rotor mills (avoiding ball-milling) and cold-pressing the powdered samples into pellets using wax binder. XRF users often rely on this type of commercial precalibrated or ‘standardless’ methods delivered with their XRF spectrometer. However, these methods are often sold without any guarantee on performance. We recently demonstrated the quite good performance of a typical commercial precalibrated/standardless method when analyzing biomass in the ideal form of certified reference material. In the present article, we report now on analysis of common raw biomass using the same method purchased with a 4 kW wavelength dispersive (WD) XRF spectrometer. The accuracy (trueness and precision) is determined by comparing the XRF data with the elemental composition obtained by standard elemental analysis (ICP-OES and ion-chromatography). The elements positively detected by the XRF are Na, Mg, Al, Si, P, S, Cl, K, Ca, Mn, Fe, Co, Ni, Cu, Zn, Sr, and maybe Mo. For elements above 25 ppm, the XRF data show a relative systematic error (bias, trueness) typically better than ±15% independent of the concentration. The elements present with >1000 ppm (Mg, Si, Cl, K, Ca) consistently show a positive bias of 3–18% relative. The relative precision (measured as the relative standard error) is better than ±5% (typically ±1%) for concentrations >25 ppm (obtained with 10–30 measurements). Quantifying elements below 25 ppm (Co, Ni, Cu, Zn, Sr, Mo) is possible in some cases, but it requires a more detailed study for each specific element. For example, Cu can be determined down to a few parts per million with an appropriate correction for the method bias. Occasionally, larger relative biases of up to 45–90% can occur for certain elements (Cl, Si) in certain samples, so care should be taken to carefully test the applied method for the particular samples and elements of interest. Quantification of silicon (Si) by XRF works well for concentrations >100 ppm. The XRF method can further be used to estimate the ash yield from biomass combustion with a relative bias better than ±10%. It is shown that the errors on the elemental composition are dominated by systematic errors (biases), and therefore, measuring the two sides of a single pellet combined with correction for any bias is the optimum approach. The five biomass types employed here, combined with the 13 certified reference materials employed in our previous study, span a broad range of biomass types with the XRF method generally producing reliable results (keeping in mind the limitations and needed bias corrections) with errors comparable to the standard reference methods. This suggests that typical standardless/precalibrated XRF methods work well in elemental analysis of raw biomass (keeping in mind the limitations) and therefore could be considered for general usage in, for example, industrial analytical laboratories requiring fast elemental analysis of biomass.